DeBusserolles Fanny 2013

Myctophid vision: seeing
and being seen in the

mesopelagic zone
By
Fanny de Busserolles, BSc, MSc
Supervised by:
Winthrop Professor Shaun P. Collin
Professor N. Justin Marshall
This thesis is presented for the degree of Doctor of Philosophy of the

University of Western Australia
School of Animal Biology

The Oceans Institute
2013
ஒளி கூட நிழல் பார்க்க தேவை
Sri Ramana Maharshi
(1879-1950, Tamil Nadu)
Light is necessary, even to see the shadow.

La lumière est nécessaire, même pour voir l’ombre.
iii
Abstract
The visual system of different organisms strongly reflects each species’ ecology and
behaviour by being adapted to their life style and ambient environmental constraints. In
the mesopelagic zone of the oceans (200-1000m), only very low intensities of sunlight
remain, creating a relatively dark environment accompanied by a multitude of
bioluminescent flashes emitted by its inhabitants. As a result, the visual system of
mesopelagic organisms has been under selective pressure to adapt to the ambient light
and behavioural tasks of this extreme environment.
This thesis aims to better understand mesopelagic fishes visual adaptations in relation to
their environment and evolution history. Using a multidisciplinary approach, the thesis
focuses on the visual system of one family of deep-sea fishes, the Myctophidae, taking
into consideration 61 species, representatives of more than 50% of the recognised
genera. Myctophids or lanternfishes are one of the most abundant groups of
mesopelagic fishes in the world’s oceans and represent a good model for visual
adaptation studies due to their diversity (~250 sp) and broad intra- and interspecific
variation in the depths they occupy, as well as their migration patterns and the location
of their luminous organs.
The variability in relative eye size was first assessed within the family and the factors
influencing interspecific differences investigated using phylogenetic comparative
analyses. A great variability in relative eye size was found at all taxonomic levels,
suggesting that this character is relatively labile. Variability in eye size within the
family could not be explained by any of the ecological variables tested
(bioluminescence and depth patterns), and appears to be driven instead by phylogenetic
relationships.
Since assessing visual capabilities using only eye size is limiting, the general
morphology of the myctophid eye was investigated in detail. Myctophids possess
several other visual adaptations for dim-light conditions, an aphakic gap, a tapetum
lucidum and a pure rod retina with high densities of long photoreceptors, all of which
were found to be highly variable between species. Two novel retinal specialisations, a
fundal pigmentation and microtubular-like structures, were also described.
v
Special attention was given to the first stage of retinal processing, the photoreceptors,
by examining their size, arrangement, topographic distribution and contribution to
optical sensitivity. Clear interspecific differences in visual specialisations, photoreceptor
designs (length and diameter), density and sensitivity to downwelling light and
bioluminescent emissions were found. Moreover, phylogenetic comparative analyses
revealed several relationships between photoreceptor characteristics and the ecological
variables tested.
A new short-wavelength intra-ocular filter, a yellow pigmentation, present in the outer

nuclear layer of the retina was found in several species and described. This
specialisation appears to be species-specific, varying in location, shape, size and was
found to be sexually dimorphic in two different species. The presence of at least two
visual pigments, resulting from a duplication of the Rh1 opsin gene was also found in
species presenting the retinal yellow pigment which is thought to improve the detection
of a more extensive range of bioluminescent emissions in a specific part of each
species’ visual field.
Finally, ganglion cell distribution and spatial resolving power was investigated in a
range of species. Again, interspecific differences were present in terms of acuity and
visual specialisations. Three different types of visual specialisations were observed;
streak-like elongated areae, areae temporalis and areae centralis reflecting different
visual demands for survival.
These findings show a surprising interspecific variability in the visual system of

myctophids and indicate the influence of both phylogeny and ecology on the evolution
of different visual specialisations. The presence of new visual specialisations in some
species and the extreme interspecific variability observed, most likely matches the
ecological and behavioural variability seen within the family indicating the presence of
different strategies for survival in the mesopelagic zone.
vi
Acknowledgements
First and foremost, I would like to sincerely thank my supervisors, Professor Shaun
Collin and Professor Justin Marshall, for giving me the chance and the means to
conduct this fantastic project and for their guidance, support, encouragements and
enthusiasm. It has been a challenging but wonderful journey full of travels, discoveries,
surprises and encounters. I have learned so much, and for that I extend my thanks to
both of them.
I am grateful to The University of Western Australia for financially supporting my

candidature through the Scholarship for International Research Fees (SIRF), the
University Postgraduate Award for International Student (UPAIS) and the UWA
Safety-Net Top-Up Scholarship.
For the success of this study, I am indebted to all the people involved in collecting
lanternfishes. I would like to especially thank Mike Hall and the crew and master from
the RV Cape Ferguson, and Hans-Joachim Wagner and the crew and master from the
FS Sonne for numerous sea time opportunities. Thank you also to Adrian Flynn,
Lynnath Beckley, Pilar Olivar, Anna Bozzano, Brigitte Guillaumont and John Denton
for providing additional specimens.
I would like to sincerely thank John Paxton, John Fitzpatrick, Nathan Hart, David Hunt,
Wayne Davies, Michael Clarke and Dorothee Hahne for their contribution to the
scientific research in this thesis and for sharing their knowledge with me. My greatest
thanks to Joao Paolo Coimbra for taking the time to teach me all the basics of visual
neuroscience lab work and for sharing his knowledge with such passion. I am grateful to
Eduardo Garza Gisholt for making my life much easier by inventing such a great
mapping program and teaching me how to use it.
A huge thank you to Caroline Kerr, Michael Archer and Alan Goldizen for their
amazing support and help with the field trip preparations, and the laboratory and
administrative work throughout my PhD. I would also like to acknowledge Jan Poulsen
for allowing me to use his data before publication and Julian Partridge, Julien Claes and
Kara Yopak for providing advice and comments on various aspects of my thesis.
vii
Many thanks to all the members of the Neuroecology and Earn groups, and especially to
the girls in my office (Maria Needhamsen, Kristyn Bates, Tenelle Wilks, Sophie Payne,
Paula Fuller, Audrey Bester and Natalie Morellini) for providing support, making me
laughs and for all the great times shared.
Finally but not least, I am very grateful to my family for their unconditional love,
understanding and support, and especially my parents and stepfather, for giving me the
means to achieve my dreams and to come this far. Thank you to all my friends both here
and back in France for their great support and understanding all along this journey.
More particularly, thanks to Kat Markey and Tyrone Ridgeway for proofreading the
final version of my thesis, to Lucille Chapuis for the many French nights and to Rob
Williams for his everyday support, confidence and love.
viii
Statement of candidate contribution
Thesis format and authorship

“The regulations of The University of Western Australia provide the option for
candidates for the Degree of Doctor of Philosophy to present their thesis as a series of
papers which have been published in refereed journals, manuscripts that have been
submitted for publication but not yet accepted, or manuscripts that could be submitted.”
In accordance with the University of Western Australia’s regulations regarding

Research Higher Degrees, this thesis is presented as a series of papers. The estimated
contribution of the candidate for the papers comprising chapters 2, 3, 4, 5 and 6 are
hereby set forth.
Chapter 2
Published in PLoS ONE as:
de Busserolles F, Fitzpatrick JL, Paxton JR, Marshall NJ, Collin SP. 2013. Eye-size
variability in deep-sea lanternfishes (Myctophidae): an ecological and phylogenetic
study. PLoS One 8(3): e58519. doi:10.1371/journal.pone.0058519.
All specimens were collected by the applicant, Shaun Collin and Justin Marshall. The
data was acquired by the candidate. John Paxton provided a critical review of the
myctophid’s ecological dataset. The design of the statistical analyses was done by the
candidate and John Fitzpatrick. The candidate analysed and interpreted the results, and
drafted the paper. A critical review of the manuscript was provided by the co-authors.
The overall estimated contribution of the candidate is 85%.
Chapter 3
Accepted in the Journal of Comparative neurology as:
de Busserolles F, Marshall NJ, Collin SP. The eyes of lanternfishes (Myctophidae,

teleostei): novel ocular specialisation for vision in dim light.
candidate acquired the data, analysed and interpreted the results, and drafted the paper.
Shaun Collin contributed to the interpretation of the fundal pigmentation. A critical
review of the manuscript was provided by the co-authors.
ix
Chapter 4
To be submitted to PLoS ONE as:
de Busserolles F, Fitzpatrick JL, Marshall NJ, Collin SP. The influence of photoreceptor
size and distribution on optical sensitivity in the eyes of lanternfishes (Myctophidae).
data was acquired by the candidate. The design of the statistical analyses was done by
the candidate and John Fitzpatrick. The candidate analysed and interpreted the results,
and drafted the paper. A critical review of the manuscript was provided by the co-
authors.
Chapter 5
de Busserolles F, Hart NS, Hunt DM, Davies WL, Clarke MW, Hahne D, Marshall NJ,
Collin SP. Spectral tuning in the eyes of lanternfishes (Myctophidae): description of a
novel sexually dimorphic intra-ocular filter.
data was acquired by the candidate. David Hunt and Wayne Davies provided
supervision for the molecular work, Nathan Hart provided supervision for the
microspectrophotometry and spectrophotometry, and Michael Clarke and Dorothee
Hahne provided supervision for the pigment extraction and preparation for HPLC
analysis. The HPLC analysis was conducted by Dorothee Hahne. The candidate
analysed and interpreted the results, and drafted the paper. A critical review of the
manuscript was provided by the co-authors.
Chapter 6
de Busserolles F, Marshall NJ, Collin SP. Retinal ganglion cell distribution and spatial
resolving power in lanternfishes (Myctophidae).
candidate acquired the data, analysed and interpreted the results, and drafted the paper.
A critical review of the manuscript was provided by the co-authors.
Peer reviewed journal articles
1. de Busserolles, F., Fitzpatrick, J.L., Paxton, J.R., Marshall, N.J., Collin, S.P. 2013.
Eye-size variability in deep-sea lanternfishes (Myctophidae): an ecological and
phylogenetic study. PLoS One 8(3): e58519. doi:10.1371/journal.pone.0058519.
x
Conference presentations
1. de Busserolles, F., Hart N. S., Hunt D. M., Marshall, N. J., and Collin, S. P. 2012.
Spectral tuning in the eyes of lanternfishes (Myctophidae). 13th Deep-Sea Biology
Symposium, Wellington, New Zealand, 3rd-7th December 2012. Oral presentation
(Oral presentation award).
2. de Busserolles F., Yopak K. E., Archer M., Partridge J. C., Marshall N. J., Collin S.
P. 2012. Approaching the extremes of vision: Retinal specializations in lanternfishes
(Myctophidae). Australian Marine Sciences Association and New Zealand Marine
Sciences Society Joint Conference (AMSA-NZMSS), Hobart, Australia, 1st-5th July
2012. Oral presentation.
3. de Busserolles F., Yopak K. E., Archer M., Partridge J. C., Marshall N. J., Collin S.
P. 2012. Visual adaptations to low light conditions and bioluminescence in
lanternfishes (Myctophidae). The Australasian Society for the Study of Animal
Behaviour (ASSAB), Geelong, Australia, 26th-28th June 2012. Oral presentation.
4. de Busserolles F., Yopak K. E., Marshall N. J., Hart N. S., Collin S. P. 2011. Visual
adaptations to the deep-sea environment in lanternfishes (Myctophidae). 3rd
Australian National Network in Marine Science (ANNiMS) conference, Perth,
Australia, 29th November – 1st Decembre 2011. Oral presentation.
5. de Busserolles F., Marshall N. J., Collin S. P. 2010. Deep-sea Myctophid vision:

measure of spatial resolving power and sensitivity. 13th International Behavioural
Ecology Congress (ISBE), Perth, Australia, 27th September – 1st October 2010.
Poster presentation.
6. de Busserolles F., Marshall N. J., Collin S. P. 2010. Spatial resolving power and
sensitivity in lanternfish: adaptations to the deep-sea environment. 20th Annual
Combined Biological Science Meeting, The University of Western Australia, 27th
August 2010. Poster presentation (Poster award).
7. de Busserolles F., Marshall N. J., Collin S. P. 2010. Lanternfish vision:

morphological measure of spatial resolving power and sensitivity. 12th Deep-Sea
Biology Symposium, Reykjavik, Iceland, 7th-13th June 2010. Oral Presentation.
8. de Busserolles F. 2010. Myctophid vision: seeing and being seen in the mesopelagic
zone. WAMSI-AMSA WA Marine science in Western Australia – Show and tell
symposium, Fremantle, Australia. 9th February 2010. Oral presentation.
Fanny de Busserolles Shaun P. Collin

Candidate Co-ordinating supervisor
xi
Table of contents
Citation ............................................................................................................................iii
Abstract ............................................................................................................................ v
Acknowledgements ........................................................................................................vii
Statement of candidate contribution ............................................................................ ix
Table of contents ..........................................................................................................xiii
List of figures ...............................................................................................................xvii
List of tables .................................................................................................................. xxi
List of abbreviations ..................................................................................................xxiii
CHAPTER 1: General introduction .............................................................................. 1

1.1. Context: the mesopelagic zone .............................................................................. 2
1.2. Fish model: the Myctophidae ................................................................................ 5
1.2.1. Distribution ............................................................................................... 5
1.2.2. Vertical migration behaviour .................................................................... 5
1.2.3. Ecology ..................................................................................................... 6
1.2.4. Taxonomy ................................................................................................. 8
1.3. Being seen in the mesopelagic zone: bioluminescence and its role in vision ....... 9
1.3.1. Mechanism .............................................................................................. 10
1.3.2. Function .................................................................................................. 12
1.3.2.1. Interaction with prey by attraction or illumination..................... 12
1.3.2.2. Interaction with predator for the purpose of defence. ................ 14
1.3.2.3. Interaction with congeners: intraspecific communication .......... 14
1.3.3. Bioluminescence in myctophids ............................................................. 15
1.4. Seeing in the mesopelagic zone........................................................................... 18
1.4.1. Teleost vision .......................................................................................... 18
1.4.1.1. Eye .............................................................................................. 18
1.4.1.2. Retina .......................................................................................... 20
1.4.1.3. Image processing by the central nervous system ....................... 22
1.4.1.4. Visual Specialisations................................................................. 24
1.4.2. Visual adaptations to the deep-sea .......................................................... 25
1.4.3. Myctophid vision .................................................................................... 27
1.5. Visual Ecology .................................................................................................... 30
1.6. Aim of this thesis ................................................................................................. 31
1.7. References ........................................................................................................... 31
CHAPTER 2: Eye-size variability in deep-sea lanternfishes (Myctophidae): an

ecological and phylogenetic study ........................................................................ 45
2.1. Abstract ............................................................................................................... 46
2.2. Introduction ......................................................................................................... 46
2.3. Materials and Methods ........................................................................................ 49
2.3.1. Ethics statement ...................................................................................... 49
2.3.2. Data collection and morphometric measurements .................................. 50
2.3.3. Taxonomic remarks ................................................................................ 51
2.3.4. Ecological data ........................................................................................ 52
xiii
2.3.5. Phylogenetic analyses ............................................................................. 56
2.3.6. Estimating phylogenetic signal ............................................................... 58
2.3.7. Phylogenetic linear models ..................................................................... 58
2.3.8. Phylogenetic ANOVAs........................................................................... 59
2.4. Results ................................................................................................................. 59
2.4.1. Morphometric measurements ................................................................. 59
2.4.2. Estimating phylogenetic signal ............................................................... 60
2.4.3. Relationship among morphometric traits ................................................ 62
2.4.4. Morphometric comparisons among tribes and clades ............................. 63
2.4.5. Relationship between morphometric and ecological traits ..................... 66
2.5. Discussion ........................................................................................................... 68
2.5.1. Eye-size variation and ecology ............................................................... 68
2.5.2. Eye-size variation and phylogeny ........................................................... 71
2.5.3. Limits of the study .................................................................................. 72
2.5.3.1. Sampling methods ...................................................................... 72
2.5.3.2. Phylogeny ................................................................................... 73
2.5.4. Conclusion .............................................................................................. 73
2.6. Acknowledgments ............................................................................................... 74
2.7. References ........................................................................................................... 74
CHAPTER 3: The eyes of lanternfishes (Myctophidae, Teleostei): novel ocular

specialisations for vision in dim light ................................................................... 81
3.1. Abstract ............................................................................................................... 82
3.2. Introduction ......................................................................................................... 82
3.3. Material and methods .......................................................................................... 85
3.3.1. Collection of species and preservation of ocular tissue .......................... 85
3.3.2. Preparation of eyecup and retinal wholemounts ..................................... 86
3.3.3. Light and electron microscopy ............................................................... 87
3.3.4. Phylogenetic comparison of visual characteristics ................................. 88
3.4. Results ................................................................................................................. 89
3.4.1. General morphology of the eye and retina .............................................. 89
3.4.2. Anatomy of the inner retina .................................................................... 92
3.4.3. Anatomy of the outer retina .................................................................... 95
3.4.4. Specialisations of the retinal pigment epithelium ................................... 98
3.4.5. Tapetum lucidum .................................................................................. 101
3.4.6. Interspecific variability within a phylogenetic context......................... 104
3.5. Discussion ......................................................................................................... 106
3.5.1. The function of the aphakic gap ........................................................... 106
3.5.2. Tapetum lucidum .................................................................................. 107
3.5.3. Interspecific variation in photoreceptor size ......................................... 108
3.5.4. Novel retinal specialisations and their putative function ...................... 109
3.5.4.1. The fundal pigmentation .......................................................... 109
3.5.4.2. The microtubule-like structures................................................ 112
3.5.5. Interspecific variability and the influence of phylogeny ...................... 113
3.5.6. Conclusions ........................................................................................... 115
3.6. Acknowledgements ........................................................................................... 115
3.7. References ......................................................................................................... 116
xiv
CHAPTER 4: The influence of photoreceptor size and distribution on optical
sensitivity in the eyes of lanternfishes (Myctophidae) ...................................... 123
4.1. Abstract ............................................................................................................. 124
4.2. Introduction ....................................................................................................... 124
4.3. Material and methods ........................................................................................ 127
4.3.1. Ethics statement .................................................................................... 127
4.3.2. Samples ................................................................................................. 128
4.3.4. Preparation of retinal wholemounts ...................................................... 128
4.3.5. Stereological analysis and the construction of topographic maps ........ 129
4.3.6. Morphometric measurement of the photoreceptors .............................. 131
4.3.7. Optical sensitivity estimations .............................................................. 132
4.3.8. Phylogenetic analyses ........................................................................... 133
4.3.9. Ecological data ...................................................................................... 135
4.3.10. Estimating phylogenetic signal ........................................................... 135
4.3.11. Phylogenetic linear models ................................................................. 136
4.4. Results ............................................................................................................... 136
4.4.1. Topographic distribution of photoreceptors.......................................... 136
4.4.2. Morphometric analyses of the photoreceptors ...................................... 142
4.4.3. Optical sensitivity ................................................................................. 143
4.4.4. Estimating phylogenetic signal ............................................................. 149
4.4.5. Relationship among morphometric traits .............................................. 149
4.4.6. Relationship between morphometric and ecological traits ................... 150
4.5. Discussion ......................................................................................................... 152
4.5.1. Topographic variations in sampling the ambient light environment .... 152
4.5.2. Different strategies for optimizing light capture by retinal photoreceptors
............................................................................................................... 155
4.5.3. The influence(s) of the photoreceptors on ecological variation in visual
behaviour ............................................................................................... 157
4.5.3.1. Rod diameter and sensitivity .................................................... 157
4.5.3.2. Outer segment length and density ............................................ 159
4.5.3.3. Inner segment length ................................................................ 159
4.5.4. Conclusion ............................................................................................ 160
4.7. References ......................................................................................................... 161
CHAPTER 5: Spectral tuning in the eyes of lanternfishes (Myctophidae):

description of a novel sexually dimorphic intra-ocular filter .......................... 169
5.1. Abstract ............................................................................................................. 170
5.2. Introduction ....................................................................................................... 170
5.3.1 Samples .................................................................................................. 173
5.3.2. Retinal wholemounts and spectrophotometry ....................................... 173
5.3.3. Cryosections .......................................................................................... 176
5.3.4. HPLC analysis ...................................................................................... 176
5.3.5. Microspectrophotometry ....................................................................... 178
5.3.6. Molecular analyses................................................................................ 179
5.3.7. Modelling of the association of the yellow pigment and the visual
pigments ................................................................................................ 180
5.4. Results ............................................................................................................... 181
5.4.1. Description of the yellow pigment ........................................................ 181
5.4.2. Sexual dimorphism ............................................................................... 185
xv
5.4.3. Spectrophotometry ................................................................................ 187
5.4.4. HPLC analysis ...................................................................................... 190
5.4.5. Microspectrophotometry ....................................................................... 190
5.4.6. Molecular analyses ............................................................................... 192
5.4.7. Modelling of the association of the yellow pigment and the visual
pigments ................................................................................................ 194
5.5. Discussion ......................................................................................................... 197
5.5.1. The unique yellow pigment in the retina of myctophids ...................... 196
5.5.2. Visual pigments and spectral tuning in myctophids ............................. 199
5.5.3. Putative functions of the yellow pigment in myctophids ..................... 201
5.5.4. Conclusion ............................................................................................ 203
5.6. Acknowledgment .............................................................................................. 204
5.7. References ......................................................................................................... 204
CHAPTER 6: Retinal ganglion cell distribution and spatial resolving power in

lanternfishes (Myctophidae) ............................................................................... 215
6.1. Abstract ............................................................................................................. 216
6.2. Introduction ....................................................................................................... 216
6.3.1. Collection of species and preservation of ocular tissue ........................ 219
6.3.2. Preparation of retinal wholemounts and Nissl staining ........................ 219
6.3.3. Amacrine cell labelling using immunohistochemistry ......................... 220
6.3.4. Stereological analyses and the construction of topographic maps ....... 221
6.3.5. Phylogenetic comparison of visual characteristics ............................... 224
6.3.6. Calculation of spatial resolving power ................................................. 225
6.4. Results ............................................................................................................... 225
6.4.1. Immunolabeling of amacrine cells in the lanternfish retina ................. 225
6.4.2. Neural cell distribution in the ganglion cell layer of lanternfishes ....... 226
6.4.3. Interspecific differences in ganglion cell distribution .......................... 234
6.4.4. Spatial resolving power ........................................................................ 238
6.5. Discussion ......................................................................................................... 239
6.5.1. Ecological significance of the retinal specialisations in lanternfishes .. 239
6.5.2. Acuity in lanternfishes .......................................................................... 242
6.5.3. Putative function of the amacrine cell population ................................ 243
6.5.4. Interspecific differences in retinal specialisations in lanternfishes:
ecology or phylogeny? .......................................................................... 244
6.7. References ......................................................................................................... 246
CHAPTER 7: General discussion.............................................................................. 253

7.1. Myctophid visual system ................................................................................... 254
7.2. Interspecific variability ..................................................................................... 255
7.3. Conclusions and future directions ..................................................................... 256
7.4. References ......................................................................................................... 259
APPENDIX I ............................................................................................................... 263
xvi
List of figures
CHAPTER 1
1.1. Sunlight spectrum profile at the earth's surface ......................................................... 2

1.2. Some descriptive features of the oceanic environment .............................................. 3
1.3. The distribution of bioluminescence emission maxima ............................................. 4
1.4. Diel vertical migration patterns of myctophid fishes ................................................. 6
1.5. Distribution and abbreviated terminology of luminous organs.................................. 9
1.6. The chemical structures of the four best-known luciferins ...................................... 11
1.7. Schematic diagram showing the functions of bioluminescence .............................. 13
1.8. Cross section of a ventral photophore ...................................................................... 16
1.9. Vertical cross-section through a teleost eye ............................................................. 19
1.10. Diagram of a vertebrate retina showing the cell types ........................................... 20
1.11. Brain of the rainbow trout Salmo gairdneri. .......................................................... 23
1.12. Topographic maps of the ganglion cells showing different specialisations. .......... 24
CHAPTER 2
2.1. Phylogenies of Myctophidae reconstructed ............................................................. 57

2.2. Difference in eye size compared to body size in two species of lanternfish............ 60
2.3. Relationship between lens diameter and eye diameter ............................................ 63
2.4. Residuals eye size corrected for body size of Myctophidae .................................... 64
2.5. Residuals eye size corrected for body size by genera of Myctophidae.................... 65
2.6. Variation in the size and location of the Dn, Vn and So luminous organs within the
genus Diaphus ......................................................................................................... 67
CHAPTER 3
3.1. Position of the eyes in Myctophum spinosum and Diaphus danae .......................... 89
3.2. Transverse section through the eye of Ceratoscopelus warmingii .......................... 90
3.3. Variation of the aphakic gap within the lanternfish family...................................... 91
3.4. Light micrograph of a transverse section through the eye cup of two lanternfish
species, showing differences of retinal thickness. ................................................... 92
3.5. Light microscopy picture of a transversal section through the retina of Diogenicthys
laternatus ................................................................................................................. 94
3.6. Light microscopy and transmission electron microscopy of the microtubular like
structure (MLS) in Myctophum nitidulum ............................................................... 96
3.7. Transmission electron microscopy of the microtubular-like structure (MLS)......... 97
3.8. Transmission electron micrographs of the rod photoreceptors ................................ 98
3.9. Different pigmentation patterns in lanternfishes. ..................................................... 99
3.10. Macroscopic pictures of the fundal pigmentation ................................................ 100
3.11. Fundal pigmentation in Nannobrachium cf. nigrum. ........................................... 101
3.12. Variation in tapetum lucidum pattern and color .................................................. 102
3.13. Tapetum lucidum variation in three species of lanternfishes ............................... 103
3.14. Tapetum lucidum structure in myctophids ........................................................... 103
3.15. Presence-absence of specific visual characteristics ............................................. 105
xvii
CHAPTER 4
4.1. Wholemount view of the rod photoreceptors ......................................................... 130

4.2. Phylogenetic trees of the Myctophidae family. ..................................................... 134
4.3. Topographic maps of photoreceptor densities of Bolinichthys longipes ............... 138
4.4. Topographic maps of photoreceptor densities for Diaphus brachycephalus and
Nannobrachium idostigma. ................................................................................... 139
4.5. Topographic maps of photoreceptor densities of Lampanyctus parvicauda. ........ 140
4.6. Topographic maps of photoreceptor densities of Myctophum brachygnathum ..... 141
4.7. Aphakic gap position, tapetum lucidum pattern and topographic maps of
photoreceptor densities .......................................................................................... 146
4.8. Transverse light microscopy sections though the retina ........................................ 147
4.9. Relative sensitivity to bioluminescence and downwelling sunlight ...................... 148
CHAPTER 5
5.1. Phylogenetic trees of Myctophidae reconstructed from Paxton et al. (1984). ....... 182
5.2. Retinal wholemounts showing the yellow pigment diversity ................................ 183
5.3. Retinal wholemounts highlighting the yellow pigment diversity .......................... 184
5.4. Light micrograph of a transverse cryosection through the retina of Gonichthys
tenuiculus ............................................................................................................... 185
5.5. Retinal wholemounts showing the sexual dimorphism in the yellow pigment...... 186
5.6. Normalised corrected absorbance spectra of the yellow pigment.......................... 187
5.7. Normalised corrected absorbance spectra of the yellow pigment.......................... 188
5.8. Map of corrected absorbance at the λYPmax for six species of lanternfish .............. 189
5.9. Mean bleaching difference absorbance spectra with wavelength of maximum
absorbance (λmax) of the rod visual pigments ........................................................ 191
5.10. Phylogenetic tree of opsin gene nucleotide sequences of the zebrafish, Danio rerio,
and the two rod opsins of the lanternfish Symbolophorus evermanni .................. 193
5.11. Amino acids sequences of the two rod opsins ..................................................... 193
5.12. Phylogenetic tree of opsin gene nucleotide sequences showing the origin of the
rod opsin duplication ............................................................................................. 194
5.13. Modelling of the quantal spectral sensitivity ....................................................... 195
5.14. Modelling of the effect the yellow pigmentation on the quantal spectral sensitivity
of Symbolophorus evermanni. ............................................................................... 196
CHAPTER 6
6.1. Light micrograph of the Nissl stained ganglion cell layer of Ceratoscopelus
warmingii............................................................................................................... 224
6.2. Light micrographs from a transverse section and a wholemount of the amacrine
cells, immunolabeled with anti-parvalbumin. ....................................................... 226
6.3. Topographic map showing the amacrine cell distribution within the retina of two
lanternfish species ................................................................................................. 227
6.4. Topographic distribution of the total neural cells, amacrine cells, and ganglion cells
of Ceratoscopelus warmingii ................................................................................ 228
of Myctophum asperum ......................................................................................... 229
of Bolinichthys nikolayi ......................................................................................... 230
xviii
of Diogenichthys laternatus .................................................................................. 231
of Notoscopelus kroeyerii ...................................................................................... 232
6.9. Topographic distribution of Nissl stained retinal ganglion cells in Myctophum
asperum, using two different counting frame sizes. .............................................. 233
6.10. Topographic distribution of retinal ganglion cells ............................................... 235
6.11. Type of retinal specialisation in each of the 18 species of lanternfishes analysed
and plotted against the phylogeny of Paxton et al. 1984 ....................................... 238
xix
List of tables
CHAPTER 1
1.1. Presence or absence of photophores and luminous organs for each genus of the
Myctophidae family. ................................................................................................ 17
1.2. Type of opsins expressed in the different photoreceptors of the vertebrate retina and
their spectral sensitivities......................................................................................... 21
1.3. Summary of the visual characteristics available in the literature for the Myctophidae
family ....................................................................................................................... 29
CHAPTER 2
2.1. Summary of the research cruises.............................................................................. 50

2.2. Analyses of covariance (ANCOVA) of the eye diameter versus standard length
between juveniles and adults in Myctophidae ......................................................... 51
2.3. Dataset used in the phylogenetic comparative analysesa ......................................... 53
2.4. Summary of the juvenile-adults night and day depth ranges. .................................. 54
2.5. Range of standard length, eye diameter and lens diameter ...................................... 61
2.6. Estimates of the phylogenetic signal. ....................................................................... 62
2.7. Regression model of eye diameter with different predictor variables when
controlling for phylogeny (PGLS). .......................................................................... 66
CHAPTER 3
3.1. Summary of retinal measurements for 53 species of lanternfishes. ......................... 93
CHAPTER 4
4.1. Summary of the stereological parameters .............................................................. 130

4.2. Summary of the quantitative data obtained from the optical fractionator. ............ 137
4.3. Summary of eye and retinal measurements. .......................................................... 144
4.4. Estimates of the phylogenetic signal ...................................................................... 149
4.5. Regression models of several visual traits with different predictor variables when
controlling for phylogeny (PGLS) ......................................................................... 151
CHAPTER 5
5.1. Summary of the individual analysed ...................................................................... 174

5.2. List of primer sets................................................................................................... 180
5.3. Spectral characteristics of rod visual pigments in the retina of three myctophids
species measured using microspectrophotometry (MSP)...................................... 190
xxi
CHAPTER 6
6.1. Summary of the parameters used for the analysis of the neural cell ...................... 222
6.2. Summary of the parameters used for the analysis of the ganglion cell .................. 222
6.3. Summary of the neural cell quantitative data......................................................... 227
6.4. Summary of the ganglion cell quantitative data..................................................... 239
xxii
List of abbreviations
α Combined attenuation coefficient of bioluminescence (Chapter 4)

Angle subtending 1 mm on the retina (Chapter 6)
β Partial regression slope
λ Pagel’s lambda
λmax Maximum absorbance
λYPmax Maximum absorbance of the yellow pigment
ø Diameter
A Pupil diameter (lens diameter)
A Adult
ABC Avidin-biotin complex
AIMS Australian Institute of Marine Science
Ala Alanine
AM Amacrine cell
ANCOVA Analysis of covariance
ANOVA Analysis of variance
Ant Antorbital organ
AO Anal organ
AOAS All-Opsins, All Species
APE Analyses of phylogenetics and evolution
ARC Australian Research Council
Asp Aspartic acid
ASIP Agouti signalling protein
Aus Australia
C Cornea
Cb Cerebellum
cDNA Complementary DNA
Ce Cervical photophore
CE Coefficient of error
ChiS China Sea
CMCA Centre for Microscopy, Characterisation and Analysis
cNAtl Central North Atlantic
cNPac Central North Pacific
CorS Coral Sea
Cp Cheek photophore
cPac Central pacific
CT Connective tissue
d Photoreceptor diameter
D Dorsal
DAB 3,3 diaminobenzidine
Di Diencephalon
DMSO Dimethyl sulfoxide
DVM Diel vertical migration
Dn Dorsal nasal organ
E Number of photon emitted at the source
eAtl Eastern Atlantic
eAus Eastern Australia
xxiii
eNAtl Eastern North Atlantic
eNPac Eastern North Pacific
eSPac Eastern South Pacific
ePac Eastern Pacific
EZ Multiple plankton net system
f Focal length
F F-number
Female
FWHM Full-width at half maximum
g Glia cells
GBRPMA Great Barrier Reef Marine Park Authority
GC Ganglion cell
GCal Gulf of California
GCL Ganglion cell layer
gDNA genomic DNA
GEIGER Analysis of evolutionary diversification
Gln Glutamine
GMex Gulf of Mexico
HPLC High-performance liquid chromatography
ICO Infra-caudal organ
IKMT Isaacs-Kidd Midwater Trawl
INL Outer nuclear layer
IPL Inner plexiform layer
IS Inner segment of the photoreceptor
J Juvenile
k Absorption coefficient of the photoreceptor
l Photoreceptor outer segment length
L Lens
LC Liquid chromatography
LW-S Long wave-shifted
LWS Long-wave sensitive
M Male
Md Medulla
Med Mediterranean Sea
MEGA Molecular Evolutionary Genetics Analysis
MLS Microtubular-like structure
MRM Multiple reaction monitoring
MS Mass spectrometry
MSP Microspectrophotometry
MTBE Methyl tertiary-butyl ether
MW Molecular weight
MWS Middle-wave sensitive
n Sample size
N Nasal
Sensitivity to point-like sources / bioluminescence (Chapter 4)
n.a. Not available
NH&MRC National Health and Medical Research Council
Ob Olfactory bulb
OCT Optimal cutting temperature
ON Optic nerve
ONL Outer nuclear layer
xxiv
OPL Outer plexiform layer
OS Outer segment of the photoreceptor
Ot Optic tectum
P Photocyte
Pac Pacific
PB Phosphate buffer
PBS Phosphate buffer saline
PCR Polymerase chain reaction
PCT Peru-Chile Trench
PFA Paraformaldehyde
PGC Peak density of ganglion cells
PGLS Phylogenetic generalised least squares regressions
Phe Phenylalanine
Phi Philippines
PHYTOOLS Phylogenetic tools for comparative biology
PRL Photoreceptor layer
PLO Pectoral lateral organ
PO Pectoral organ
Pol Posterior lateral organ
PVO Pectoral ventral organ
r Distance between the light source and the eye
R Reflector
RACE Rapid amplification of cDNA ends
Rh1 Rhodopsin
Rh2 Rhodopsin-like or middle-wave sensitive
RMT Rectangular Midwater Trawl
RPE Retinal pigment epithelium
S Sensitivity to extended sources (downwelling light)
SAfr South Africa
Sc Sclera
SCO Supra-orbital organ
Ser Serine
Sex. C. Sexual dimorphism in caudal luminous organs
Sex. D. Sexual dimorphism in Dv/Dn luminous organs
Sex. dim. Sexual dimorphism in luminous tissue
Sex. P. Sexual dimorphism in luminous tissue patches
SL Standard length
So Suborbital organ
SPR Spatial resolving power
Suo Supraobital organ
svg Scalable vector graphics
SWS Short-wave sensitive
T Temporal
Tas Tasmania
TC Total neural cells
Te Telencephalon
TEM Transmission electron microscopy
Tyr Tyrosine
V Ventral
VLO Ventral lateral organ
Vn Ventral nasal organ
xxv
VO Ventral organ
wAus Western Australia
wMed Western Mediterranean Sea
wNAtl Western North Atlantic
xxvi
Chapter 1 Introduction
CHAPTER 1
General introduction
1
1.1. Context: the mesopelagic zone

The properties of light in water are very different from those in the air. In sea
water, even in the clearest part of the world’s ocean, downwardly directed illumination
by the sun (Figure 1.1A) is selectively absorbed and scattered by the particles present in
the water column, leading to a rapid decrease in light intensity with depth. The water
column also acts as a filter regarding the light spectral absorbance, with short and long
wavelengths rapidly attenuated with depth and only blue and green wavelengths
penetrating to deeper levels (Figure 1.1C-B).
Figure 1.1. Sunlight spectrum profile at the earth's surface (A) compared to the irradiance of an
artificial light bulb (B). Typical underwater absorption of oceanic (C) and coastal or continental
water (D). From Villamizar et al. (2011).
Oceans can be divided vertically into three different zones according to the
intensity of downwelling sunlight (Figure 1.2, Herring, 2002). The epipelagic zone,
from the surface to 200 m, marks the limit of the photic zone, where the amount of light
is sufficient to enable photosynthesis. From 200 to 1000 m, some light still penetrates
but the amount is not enough to allow photosynthesis. This zone, called the mesopelagic
or twilight zone can then be seen as a transitional zone where daylight is still present but
its intensity progressively diminished with depth. In the mesopelagic zone, light has a
2
relatively constant colour and direction but exponentially diminishing intensity.

Beyond 1000 m, lies the bathypelagic zone where no daylight penetrates; this
environment provides a featureless, dark background.
Figure 1.2. Some descriptive features of the oceanic environment. Indicated are the extent of
diel vertical migration (DVM), the relative biomass of zooplankton, the light regime, and the
temperature profile of a warm ocean. Redrawn and modified from Herring (2002).
Daylight is not the only source of light in the ocean. There is another very
common source of light, emitted by the animals themselves, called bioluminescence
(See section 1.3). Bioluminescent light is found at all depths but is particularly prevalent
in the mesopelagic zone and in some parts of the world up to 70% of the fishes and 90%
of all other organisms living below 500 m are luminous (Herring, 2002). Due to the
physical properties of sea water mentioned above, most organisms have adapted to this
environment by producing bioluminescence in the blue-green part of the visible
spectrum (Figure 1.3, Denton and Locket, 1989; Widder, 2010). Bioluminescent light
can vary in intensity, duration, frequency and angular distribution but has an effective
visual range limited to about 100-150 m (Denton, 1990).
The mesopelagic zone is also characterised by unique physical and chemical

properties. In addition to daylight, temperature falls progressively with depth, nutrients
become less available and pressure increases steadily (Locket, 1977). Despite all these
conditions, the greatest biomass and diversity is concentrated in this part of the ocean
(Figure 1.2).
3
Figure 1.3. The distribution of bioluminescence emission maxima varies by marine

environment and organism type. Bioluminescent emissions extend over the full visible range
and beyond. [Photo credits: J. Cohen for the photograph of S. crassicornus; P. Herring, P.
Bifrons; and P. Batson (DeepSeaPhotography.com), C. faurei]. From Widder (2010).
Deep-sea organisms rely on different sensory systems (vision, olfaction, hearing,

taste, electroreception, lateral line) to survive within such an inhospitable environment.
Depending on the depth and type of habitat in which they live, organisms mainly rely
on one or more of these sensory systems. In the mesopelagic zone, where the amount of
light declines exponentially and where bioluminescent animals predominate, vision
becomes very important and seems to be the dominant sense used by its inhabitants
(Wagner, 2001). However, to be able to see in dim conditions and for viewing
bioluminescence, the visual system of mesopelagic organisms must have been under
tremendous selective pressure(s). Myctophids are one of the most abundant families of
mesopelagic fishes in the world’s ocean and reveal high interspecific variability in
habitat and behaviour, i.e. the diversity of their vertical migration patterns. Therefore,
they appear to be an important model to better understand fish visual adaptations in
relation to their environment.
4
1.2. Fish model: the Myctophidae

1.2.1. Distribution
The Myctophidae is one of the most abundant families of mesopelagic fishes in
the world’s ocean (Hulley, 1981). However, the biology, ecology and physiology of this
family are still poorly understood compared to many shallow water species. As a result,
most studies on myctophids deal principally with species’ description and their
distribution around the world (adults: Paxton, 1967; Kawaguchi et al., 1972; Wisner,
1976; Karnella, 1987; Hulley, 1992; Wang and Chen, 2001; Tsarin, 2002; Collins et al.,
2008; Koubbi et al., 2011; Flynn and Williams, 2012; larvae: Olivar and Beckley, 1994;
Olivar et al., 1999; Evseenko, 2006). Myctophids are distributed worldwide in all major
oceans and seas and most species exhibit extensive diel vertical migration. Therefore,
they are found from the surface to depths exceeding 2000 m (Hulley, 1994). However,
vertical migration behaviour can vary within and between species. During the day,
myctophids can be found in the mesopelagic zone but during the night most of them can
be observed in the epipelagic zone.
1.2.2. Vertical migration behaviour

Myctophids undertake diel vertical migration for the purposes of feeding. They
migrate at night to the rich surface layers to feed and then return to the deep during the
day to avoid predators (Holton, 1969). A great interspecific variability in the migration
patterns is observed within the Myctophidae and species can be classified into four
major groups based upon these migration patterns (Figure 1.4, Watanabe et al., 1999).
Species can be classified as 1. Migrant when day time and night time habitats are clearly
separated; 2. Semi-migrant when only part of the population migrate at night towards
the surface and the other part remain in the day time habitat; 3. Passive migrant when
the population does not show any clear migration pattern and the upper night time
distribution could be attributed to predation of vertical migratory prey; and finally 4.
Non-migrant. Interspecific variability in migration pattern is also seen within each of
these four groups. For example, Symbolophorus californiensis and Diaphus theta can be
classified as migrants as both species are present at different depths during the day and
night, although the former species could also be classified as a surface migrant and the
latter as a mid-water migrant (Figure 1.4, Watanabe et al., 1999).
Intraspecific variability in migration pattern also exists and may depend on

latitude, season, sex, and stage in the life cycle (Hulley, 1994). For example, juveniles
5
of several species of myctophids appear to be non-migratory (i.e. Benthosema

suborbitale, Diaphus holti, Diogenychthys atlanticus, Lampanyctus photonotus, Hulley,
1984). More rarely, some specimens become non-migratory when they reach a
particularly large size (Bolinychthys supralateralis, Diaphus rafinesquii, Hulley, 1984).
These differences in the migration pattern between and within species might be
explained by a combination of ecological factors such as differences in life style (Moku
et al., 2000), distribution of their prey and potential predators (Collins et al., 2008),
differences in temperature preferences (steno- or eurythermal species, Watanabe et al.,
1999) and/or the influence of the lunar cycle (Linkowski, 1996).
Figure 1.4. Diel vertical migration patterns of myctophid fishes in transitional waters of the
western North Pacific off Japan. The layers in which more than 75% of the total catch was
recorded are indicated as the habitat depth. * Surface migrants according to Ogawa (1961),
Hattori (1964), and Kawamura and Fujii (1988). From Watanabe et al. (1999).
1.2.3. Ecology
Myctophids are secondary and tertiary consumers in the pelagic environment
and play a major role in marine ecosystems by transferring energy to the deeper and
depleted layers of the ocean through their vertical migration behaviours and the
production of fast sinking faeces (Moku and Kawaguchi, 2008; Catul et al., 2011). They
are zooplankton (meso- and macro-zooplankton) consumers and produce food for
higher level organisms, like other fishes (Goldsworthy et al., 2002; Esposito et al.,
2009; Battaglia et al., 2013; Horn et al., 2013), squids (Watanabe et al., 2004; Parry,
2006; Watanabe et al., 2008), seabirds (Jackson, 1988; Falk et al., 1992; Green et al.,
6
1998), and mammals (Green et al., 1991; Carey, 1992; Pusineri et al., 2007; Pereira et
al., 2011).
Myctophids are mainly carnivorous, feeding mainly on zooplankton (copepods,

euphausiids, amphipods, Young and Blaber, 1986; Tyler and Pearcy, 1975;
Podrazhanskaya, 1993; Kinzer et al., 1993; Kozlov, 1995; Ishihara and Kubota, 1997;
Pusch et al., 2004; Shreeve et al., 2009; Tanaka et al., 2013) and the differences
observed in the diet of the different species are thought to be mainly due to regional and
seasonal variability in zooplankton composition and availability (Kozlov, 1995). A few
species seem to have expanded their diet by including food sources from much lower or
equal trophic levels. For example, occasional herbivory has been observed in
Ceratoscopelus warmingii (Robison, 1984) and several myctophids have been found to
feed on fish including other myctophids species (Young and Blaber, 1986;
Podrazhanskaya, 1993; Kinzer et al., 1993).
Stomach contents and isotope studies of myctophids indicate a high level of

trophic partitioning, which is correlated with migration behaviour and body size (Moku
et al., 2000; Cherel et al., 2010; Flynn and Kloser, 2012). Therefore, different species
possess different life style strategies: high, intermediate and low energy. High energy
life-style species exhibit extensive diel vertical migration to the surface layers, where
temperature and food availability are high (Moku et al., 2000) and their feeding
intensity seems to be directly correlated to the zooplankton biomass available at the
time (Uchikawa et al., 2008). Low energy life style species are non-migrators; they live
in an environment with low temperature and low prey densities (Moku et al., 2000).
Growth rate may also be linked to temperature preferences and may vary
between species (Takagi et al., 2006). Most species of myctophid appear to have daily
growth (Gartner, 1991) and an annual life cycle (Clarke, 1973; Gjosaeter, 1973;
Gartner, 1993). Some species are thought to live longer (up to 8 years, Childress et al.,
1980) than others, but the mortality rate for these species appears to be high (Gjosaeter,
1973).
Little is known about myctophid reproductive biology and strategies. Both sexes
are non-guarding pelagic spawners and seem to reproduce in the upper layers of the
water column (Tsarin, 2002) in aggregations (Flynn and Paxton, 2012). The fecundity
7
rate appears low in all species and two major reproductive strategies have been
observed within the family (Gartner, 1993). Some species seem to have a protracted
spawning season of four to six months with individuals spawning every one to four
days, while other species seem to have a more restricted spawning period occurring only
once or twice a year. In warm tropical waters, females appear to spawn seasonally, in
spring and summer, which coincides with the seasonal peak in zooplankton numbers
(Clarke, 1973).
1.2.4. Taxonomy
The Myctophidae is a teleostean family belonging to the order Myctophiformes,
which has been recognized as a monophyletic group by several authors (Paxton, 1972;
Moser et al., 1984; Stiassny, 1996). A total of 33 genera have been classified with
approximately 250 species (Hulley and Paxton, in press). The taxonomy of this group is
rather complicated and myctophid intrarelationships remain under discussion (Stiassny,
1996). The most comprehensive phylogenetic study to date is by Paxton et al. (1984)
which classified genera using derived character states of adult osteology and photophore
patterns, as described by Paxton (1972), and of larvae as described by Moser and
Ahlstrom (1970, 1972, 1974). The phylogeny divides the family into two subfamilies
(Myctophinae and Lampanyctinae) and seven tribes (Electronini, Myctophini,
Gonichthyini, Diaphini, Gymnoscopelini, Lampanyctini and Notolychnini). The first
molecular phylogeny for the family was recently constructed using mitogenomic results
from DNA sequences and unique gene orders from 38 lanternfish species (Poulsen et
al., 2013). This new phylogeny confirmed the presence of the two subfamilies
(Myctophinae and Lampanyctinae) and identified 10 monophyletic lineages or clades.
Since 1949, several taxonomic studies were done and keys to genera and species
from different regions are available (Fraser-Brunner, 1949; Wisner, 1976; Hulley, 1981;
Hulley, 1984; Nafpaktitis et al., 1977; Paxton and Hulley, 1999). Some general
morphological characteristics of myctophids include their small size (ranging from 3 cm
in Diogenichthys panurgus to 35 cm in Gymnoscopelus bolini, Hulley, 1994), a
compressed head and body, very large and laterally placed eyes (dorsolateral in
Protomyctophum), a large to very large mouth which is usually terminal (sub-terminal
in Centrobranchus, Gonichthys, Loweina) and extending to or far beyond the eye,
numerous small teeth, one dorsal adipose fin and the origin of the anal fin being close
behind the base of dorsal fin. All species are luminous, with large primary photophores
8
always present (except Taaningichthys paurolychnus) and arranged in distinct groups

on the head and body, which constitutes the main criterion for species identification
(Figure 1.5, Hulley, 1990). Small secondary photophores are present on the head, body
and median fins of some species and luminous tissues of various shapes and sizes are
found on the head, caudal peduncle and/or at the bases of various fins in most species.
Most deep-water species are black or brown in colour, shallow water species are mainly
silvery and some species possess metallic green and blue scales.
Figure 1.5. Distribution and abbreviated terminology of luminous organs in myctophid fishes
(based on Nafpaktitis et al. (1977). Cp-Cheek photophores; Ce-Cervical photophore; Suo-
Supraorbital organ; Ant-Antorbital organ; Dn-Dorsal nasal organ; Vn-Ventral nasal organ; So-
Suborbital organ; ICO-Infracaudal organ(s); SCO-Supracaudal organ(s); PLO-Pectoral lateral
organ; PVO-Pectoral ventral organs; VLO Ventral lateral organ; SAO-Supra anal organs; Pol-
Posterior lateral organ; PO-Pectoral organs; VO-Ventral organs; AO-Anal organs, anterior &
posterior; Prc-Precaudals. From Hulley (1990).
1.3. Being seen in the mesopelagic zone: bioluminescence and its role in
vision
Bioluminescence has been observed from a broad range of organisms belonging
to the main taxonomic groups: bacteria, archea, fungi, plants, animals, all of which
inhabit a broad range of environments. The greatest bioluminescent diversity is found in
the marine environment and because of its widespread occurrence it is thought to be an
important form of communication. Bioluminescence is predominantly found in the
deep-sea environment, where it acts as the main source of light. Below 500 m, 70% of
the fishes are bioluminescent (Herring, 2002). Among bony fishes, bioluminescent
species belong to at least 42 families across 11 orders (Haddock et al., 2010).
9
1.3.1. Mechanism
Bioluminescence is the production and emission of visible light by an organism
resulting from a chemical reaction. The reaction involves the oxidation of a light
emitting organic molecule, called luciferin, catalysed by an enzyme called luciferase.
The luciferin molecule is highly conserved between phyla and four main forms
are used by marine organisms (Figure 1.6, Haddock et al., 2010; Widder, 2010).
However, one particular type, coelenterazine, predominates and is found in organisms
from at least seven phyla (Herring, 2002). In contrast, the enzyme catalysing the
reaction, luciferase, is very diverse and each species may have its own “luciferase”
(Herring, 2002).
Among deep-sea fishes, two types of bioluminescence can be observed:

symbiotic bacterial luminescence or self-luminescence. Usually organisms possess one
or the other system but in a few cases both types of bioluminescence have been
observed in a single species. This is the case for the female anglerfish from the genus
Linophryne which possesses one luminous organ containing symbiotic bacteria and
another one with intrinsic, intracellular luminescence (Hansen and Herring, 1977).
Symbiotic bacterial luminescence is the less common type of bioluminescence in

the ocean and occurs principally in fishes and squid. In fishes, the symbiotic bacteria,
which are from the genera Photobacterium or Vibrio and seem to be species-specific to
a particular host (Nealson and Hastings, 1979), are cultured in special organs. Both host
and symbiont benefit from this association. The host provides nutrients and a protected
environment for the symbiont which, in turn, supplies continuous light to the host. As
symbiotic bacteria produce light continuously, the light emission has to be controlled by
the host through specific mechanisms such as shutters, organ rotation or
chromatophores (Herring, 2002). The most well known mesopelagic example of this
type of bioluminescence is seen in the female anglerfishes, which culture luminous
bacteria in a light organ, called an esca, situated on the top of their head (Hansen and
Herring, 1977).
10
Figure 1.6. The chemical structures of the four best-known luciferins are as diverse as their
phylogenetic distribution. Bacterial luciferin may occur in free-living or symbiont bacteria (e.g.,
in squid such as Heteroteuthis dispar) or in fish such as Melanocetous johnsoni. Dinoflagellate
luciferin occurs not only in dinoflagellates (e.g., Pyrocystis fusiformis) but also in euphausiids
(e.g., Meganyctiphanes norvegica). Some of those using coelenterazine as luciferin include
radiolarians (e.g., unidentified polycystine radiolarians), cnidarians (e.g., scyphozoan Periphylla
periphylla, as seen in the light and photographed by its own light), ctenophores (e.g.,
Bathocyroe fosteri, with bioluminescence display shown in inset), vampire squid (e.g.,
Vampyroteuthis infernalis), ostracods (e.g., Orthoconchoecia agassizi), copepods (e.g., Gaussia
princeps releasing its bioluminescent chemicals from glands on its tail, shown in inset),
decapods (e.g., Acanthephyra purpurea spewing luciferin and luciferase out of its mouth),
chaetognaths (e.g., Caecosagitta macrocephala), and fish (e.g., the myctophid Diaphus sp. has a
large preorbital light organ). Cypridina luciferin, which is an imidazopyrazinone like
coelenterazine, is found in ostracods such as Vargula hilgendorfii and is the dietary source of
luciferin for the midshipman fish Porichthys notatus. [Photo credits: S. Haddock, radiolarians
and chaetognath; K. Reisenbichler, V. infernalis; J. Case, copepod luminescent glands and
midshipman fish photophores]. From Widder (2010).
11
Self luminous bioluminescence is the most common mechanism used in oceanic

species. The reaction is induced by the individuals themselves through nervous (Case
and Strause, 1978) or hormonal (Claes and Mallefet, 2009) control using their own
luciferin and luciferase. The light produced by the organism is emitted through more or
less complex structures. These structures range from a simple cell called a photocyte to
complex photophores composed of accessory optical structures or tissues (Denton et al.,
1970; Herring, 2002). These accessory structures (pigment cup, reflectors, lens,
pigmented absorption filters) limit the aperture of the photophore, increase the
efficiency and/or define the spatial and spectral characteristics of the emitted light
(Herring, 1985; Herring, 2002).
1.3.2. Function
Emission of bioluminescent light is thought to have many functions in the deep-
sea. Since behavioural studies on deep-sea specimens are not yet possible, the different
functions of deep-sea bioluminescence have been interpreted using comparisons with
well-known shallow water examples and most functions are still very hypothetical.
Three main biological interactions using bioluminescent signals have been identified:
interactions with prey, predators and congeners (Figure 1.7, Herring, 2002; Haddock et
al., 2010; Widder, 2010).
1.3.2.1. Interaction with prey by attraction or illumination.

Attraction. Several deep-sea species possess luminous lures close to their jaws to attract
prey. The best example is seen among female anglerfishes who wave their esca, placed
on the top of their head, to mimic something edible and attract potential prey (Marshall,
1954). A similar attraction behaviour was also observed in some stomiatoid fishes that
possess luminous chin barbels (Marshall, 1954) and siphonophores from the genus
Erenna (Haddock et al., 2005).
Illumination. Some fishes possess luminous head organs (dragonfishes, lanternfishes).

These headlights may either be used to stun or confuse prey or to search for prey by
illuminating their surrounding environment. The most striking example is seen in
Malacosteus, one of the few marine organisms able to emit and see red (long
wavelength) light (Widder et al., 1984; Douglas et al., 1998). By producing its own long
wavelength emission, Malacosteus can then illuminate potential prey passing by
12
without the prey noticing. This is a very powerful predation technique, as most other
animals in the deep-sea cannot see red light.
Figure 1.7. Schematic diagram showing the functions of bioluminescence. Marine

luminescence can be used for defence (blue), offence (magenta), and intraspecific
communication (gray). The organisms thought to benefit from these functions are listed on the
right. Some animals are thought to use their luminescence in two, three, or even four different
roles. From Haddock (2010).
13
1.3.2.2. Interaction with predator for the purpose of defence.

Counterillumination. This form of camouflage is used by several mesopelagic
organisms (decapods, Latz, 1995; squids and fishes, Young et al., 1980; sharks, Claes et
al., 2010) and involves the use of ventral photophores to match the downwelling light
and potentially make their silhouettes disappear against the lighter upper mesopelagic
zone when viewed from below. Counterillumination can be achieved by either a
complete obliteration of the ventral silhouette or by sufficiently disrupting the ventral
surface (Young et al., 1980; Denton et al., 1985; Johnsen et al., 2004).
Flashes or squirts. Intimidation, distraction or disorientation are common uses of flashes

or squirts of lights. In the case of mobile organisms, these types of flashes will
supposedly confuse predators leaving enough time for the prey to escape. In the case of
sessile or fragile (yet deadly) organisms, bioluminescence could be interpreted as a
signal of an animal’s location or as a warning signal.
Indirect effect. Several organisms emit bioluminescent signals to distract and mislead
predators disguising their real position in order to escape. These emissions can take
different forms like bioluminescent clouds, secreted by many organisms including
copepods (Herring, 1988), decapods (Herring, 1976) and squid (Robison et al., 2003),
or flashes associated with the shedding of specific body parts. Some species may even
lose these bioluminescent body parts, also called sacrificial tags, to a predator in order
to escape (Robison, 1992; Robison et al., 2003). These sacrificial tags can glow several
hours once detached from the main body part and in some cases, even continue to glow
once ingested by a predator. A predator ingesting a bioluminescent sacrificial tag would
then become conspicuous and vulnerable to higher order predators; an example of
another defensive strategy called burglar alarm (Abrahams and Townsend, 1993;
Mesinger and Case, 1992; Robison, 1992).
1.3.2.3. Interaction with congeners: intraspecific communication

Communication between individuals of the same species using bioluminescence
is a well known concept in the terrestrial environment (fireflies, Lloyd, 1971; Lloyd,
1983, for review). In the deep-sea, bioluminescence may play an important role in
maintaining aggregation and in intraspecific communication. For example, species-
specific photophore patterns (lanternfishes), specific wavelength emissions
(Malacosteus) and sexual dimorphism in luminous organs (dragonfishes, lanternfishes)
14
may be used for species recognition and/or mating communication (Herring 2000,
2007).
1.3.3. Bioluminescence in myctophids

The Myctophidae owes its common name (lanternfish) to its conspicuous ability
to produce bioluminescent light. All the representatives of the family with the exception
of one species (Taaningichthys paurolychnus) possess photophores, which were often
referred to as “pearly spots” in the early descriptions (Beebe, 1934).
Myctophids produce bioluminescence using a luciferin-luciferase reaction (Tsuji

and Haneda, 1971; Haygood et al., 1994) and possess two kinds of photophores or
bioluminescent organs that light up independently. The ventral and ventrolateral
photophores also called primary photophores are minute photophores approximately
spherical and 0.5 mm in diameter. Each individual may have between 50 and 80
photophores present on the ventral part of the body and head (Nafpaktitis et al., 1977)
arranged in a species-specific pattern. The luminous organs and tissue patches are of
different size and shape, located on the caudal peduncle, head and body and are
frequently sexually dimorphic.
The arrangement of the luminous structures differs from species to species

(Edwards and Herring, 1977). However, there seems to be only one type of photogenic
tissue in the Myctophidae (Anctil, 1972). Luminous tissue structure can be separated
into two groups (Edwards and Herring, 1977). The first and most common group
possesses structures composed of regularly-arranged lamellae. The second group,
observed in the tribe Lampanyctini, possesses the same structure but the lamellae are
arranged in a less regular way. Primary photophores have a very similar structure as the
luminous organs but possess less photogenic tissue and lack any filter pigments (Denton
et al., 1985). Denton et al. (1985) give a description of the function of the ventral
photophores in Diaphus rafinesquii (Figure 1.8). The light produced by the photocyte is
prevented from being emitted directly from the photophore by a dense band of melanin.
The light is then directed inward, first on to the disc-shaped reflector which then reflects
the light out of the aperture. The light emitted through the photophore will then be
influenced by the reflective characteristics of the reflectors which, in the case of D.
rafinesquii, appears to be similar in colour (deep-blue) to the tapetum lucidum found in
the eye (Denton et al., 1985).
15
Figure 1.8. Cross section of a ventral photophore (A) in Diaphus rafinesqueii from Denton et
al. (1985); (B) in Electrona risso stained with Eosin-HTX from Kronstrom and Mallefet,
(2010); CT, connective tissue; R, reflector; P, photocyte; L, lens; bars is 50 μm.
Even if the primary photophores and luminous patches appear to comprise the
same structures, they most likely differ from each other in function as photophores can
produce light continuously over longs period of time, while luminous organs only
briefly produce light (Edwards and Herring, 1977). Primary photophores seem to be
used in counterillumination to camouflage the ventral outline of the fish from predators
situated below them. Denton et al. (1985) suggested that the blue reflectors present in
the ventral photophores may improve the spectral match between the light emitted and
the ambient daylight. Luminous caudal organs and tissue patches are thought to play a
role in communication within and between species. For example, Lampanyctus niger
and L. tenuiformis have similar photophore distributions and overlapping habitats but
different flash patterns, which can potentially be used to differentiate the two species
(Mensinger and Case, 1990). Several functions have been hypothesized and it is
probable that these organs serve more than one purposes especially because no clear
inter-specific pattern was identified with respect to the presence and/or absence of these
luminous organs.
Bioluminescent sexual dimorphism is very common and diverse among

myctophid species (Table 1.1). These differences are seen in the presence and/or size of
the luminous tissue patches on the head and/or tail. Herring (2007) gives a detailed
review of the type of sexual dimorphism encountered within the family. The main
sexual character is the caudal organ at the base of the tail (infra- or supra-caudal), which
is present in 30 of the 33 genera and is different between sexes in 22 of the 30 genera
16
(Herring, 2007). One of the genera that does not possess any caudal organs, Diaphus,
usually exhibits a very pronounced sexual dimorphism of at least one of the head organs
(Herring, 2007). Due to the very common occurrence and diversity of sexual
dimorphism in the luminous organs among myctophids, it is thought to play a very
important role in sexual selection (Herring, 2007).
Table 1.1. Presence or absence of photophores and luminous organs for each genus of the
Myctophidae family and the occurrence of sexual dimorphism associated with these luminous
patches. + present in all the species, + present in at least one species, - absent. Data from
Herring (2007), Hulley (1981), Nafpaktitis et al. (1977) and Paxton (1972).
Caudal Head Gland Sexual

Genus Photophores
organs organs patches dimorphism
Benthosema + + - - +
Bolinichthys + + - + +
Centrobranchus + + - - +
Cerastocopelus + + - - -
Diaphus + - + - +
Diogenichthys + + + - +
Electrona + + - - +
Gonichthys + + - - +
Gymnocospelus + - - + +
Hintonia + - - - +
Hygophum + + - - +
Idiolychnus + + - - +
Krefftichthys + + - - +
Lampadena + + - - -
Lampanyctodes + + - - +
Lampanyctus + + - + +
Lampichthys + + - - -
Lepidophanes + + - - -
Lobianchia + + - - +
Loweina + + - - +
Metelectrona + + - - -
Myctophum + + - - +
Nannobrachium + + - - +
Notolychnus + + - - +
Notoscopelus + + - + +
Parvilux + + - - -
Promyctophum + + - - +
Scopelopsis + + - - +
Stenobrachius + + - - -
Symbolophorus + + - - +
Taaningichthys + + - - +
Tarletonbeania + + - - +
Triphotorus + + - - -
Unfortunately, it is very hard to study live myctophids and even harder to

observed them in their natural environment. Studies on live animals under laboratory
conditions are extremely rare and challenging to conduct due to the difficulty of
capturing live specimens in good enough condition and in maintaining them in aquaria.
To date, the longest time a myctophid has been kept in an aquarium is 72 h
17
(Tarletonbeania crenularis Robison, 1973) but usually most of them die after a couple
of hours (Anctil, 1972; Barnes and Case, 1974). During the few successful attempts,
photophores and luminous glands were electrically and/or chemically stimulated and the
duration and intensity of light produced was recorded (Anctil, 1972; Barnes and Case,
1974; Christophe and Baguet, 1982; Mensinger and Case, 1990). The results showed
that both photophores and luminous tissues respond in different ways to the stimuli and
that photophore light emission seems to be sufficient to match the downwelling light
environment. These results also confirmed that the light emission is under nervous
control as previously described by Ray (1950). A few studies have measured the
emission spectra of the photophore of theses fishes and revealed that light emissions
extend from 400 nm to 600 nm with a maximal emission at about 470 nm (Myctophum
punctatun, Nicol, 1960) and 456 nm (Diaphus elucens, Tsuji and Haneda, 1971).
Vision seems to play a major role in counterillumination. Using different

background irradiance levels, Case et al. (1977) showed that myctophids may use their
visual sensitivity to control the output of the photophores, thereby being useful in
countershading. Young et al. (1979), later confirmed that the visual system of
Myctophum spinosum was involved in counterillumination and showed that both the
eyes and the extraocular photoreceptors (within the pineal organ) are required to detect
and mimic the intensity of downwelling light by comparing the difference between the
downwelling sunlight and the bioluminescent emissions. The efficiency of camouflage
mediated by counterillumination is not only dependent on the visual sensitivity of the
producer but is also dependant on the predator’s visual acuity. In fact, predators with an
acute eye (0.11 degrees resolution) would easily be able to break myctophid camouflage
at distances greater than 1m (Johnsen et al., 2004).
1.4. Seeing in the mesopelagic zone

1.4.1. Teleost vision
1.4.1.1. Eye
A full description of the teleost eye and retina is given by Walls, (1942). The eye
structure and properties are very conservative among vertebrates. Teleosts, like any
other vertebrates, possess a camera type eye in which a single lens focuses an image
onto the retina (Figure 1.9). However, unlike birds and mammals, the fish eye continues
to grow throughout life (Walls, 1942).
18
Figure 1.9. Vertical cross-section through a teleost eye. From Walls (1942).
The main difference between a terrestrial and an aquatic vertebrate eye is the
ability to cope with the refractive index of the surrounding media (1.000 for air,
between 1.333 and 1.339 for water, Land, 1990). Terrestrial eyes relies on the cornea
(principally) and the lens for light refraction and to focus an image but for aquatic eyes
the lens does all the focusing. Since the cornea is not refractive underwater, the lens of
most teleosts is spherical and its radius and the distance from the lens centre to the
retina (focal length) are related by a “constant” defined over 100 years ago as
Matthiessen’s ratio (Matthiessen, 1882, 1886; Sadler, 1973), regardless of the size of
the eye (Hulley, 1981). Thus, the acuity (quality of the image) of the teleost eye is
directly related to eye size and lens diameter but is also influenced by other factors such
as the quality of the optics, the amount of light received (aperture or pupil size), the
degree of retinal summation and the degree of overlap between the dendritic fields of
neighbouring retinal receptors.
Depending on their life style and particularly their feeding strategy, different
organisms will possess eyes specialised for high acuity, high sensitivity or both (by
19
different parts of the eye playing different roles). Although the ocular media (cornea,
lens, and vitreous) play a role and may filter or tune the incident light, it is the retina,
sitting at the back of the eye, that ultimately samples the visual environment and sends
the information to the visual regions of the central nervous system (Collin and Shand,
2003).
1.4.1.2. Retina
The retina is a thin sheet of neural tissue composed of different layers containing
five classes of neurons (photoreceptors, bipolar cells, horizontal cells, amacrine cells
and ganglion cells) that receive information from the surrounding light environment,
convert this energy into electrical impulses and transmit this information to the visual
centres of the brain (Figure 1.10).
Figure 1.10. Diagram of a vertebrate retina showing the cell types (A) and the various layers as
seen using light microscopy from a histological section of the wild-type zebrafish retina (B).
Modified from (A) Molday and Zhang (2010) and (B) Schonthaler et al. (2010). Scale bar in
(B): 25 μm.
The retinal pigment epithelium (RPE) is composed of a single layer of cuboidal

cells, each containing pigment granules (melanosomes) and a number of organelles and
inclusions that form a close relationship with the photoreceptor outer segments to
20
maintain visual function. In addition to capturing the remaining light that has not been
absorbed by the photoreceptors, the RPE also provides the necessary nutrients and
allows the recycling of the photoreceptor tips and visual pigments in order to maintain
photoreceptor excitability (Strauss, 2005).
Teleost fishes possess two types of photoreceptors, cones and rods, implicated in
image-forming vision. While cones are associated with diurnal (photopic), low
sensitivity, high acuity vision, and are involved in colour processing, rods are associated
with nocturnal (scotopic), high sensitivity and low acuity vision. Cone and rod
photoreceptors are composed of four main parts, the outer segment (OS) containing the
visual pigment of the cell, the inner segment (IS) comprising the organelles of the cell,
the nucleus (forming the outer nuclear layer or ONL) and the receptor terminal, often
containing multiple synapses (Figure 1.10). Within the outer segment of each
photoreceptor, a visual pigment, either a rhodopsin or a porphyropsin, will absorb
photons of light and transform this energy to an electrical signal. Visual pigments are
composed of a protein (opsin) and a chromophore derived from either vitamin A1 or A2.
Vertebrates possess different opsins which have different spectral sensitivities (Table
1.2; Bowmaker, 2008). When visual pigments (opsin protein + chromophore) are
exposed to light they undergo a conformational change, leading to rod and cone
hyperpolarisation (transduction) and the creation of an electrical signal. This new
electrical information will travel from the photoreceptors to the ganglion cells via
several interneurons (horizontal cells, bipolar cells, amacrine cells, i.e. interplexiform
cells, (Wagner, 1990).
Table 1.2. Type of opsins expressed in the different photoreceptors of the vertebrate retina and
their spectral sensitivities. Four types of cone opsins exist: SWS1, SWS2, RH2, LWS (SWS =
short-wave sensitive ; LWS = long-wave sensitive ; Rh2 = middle-wave sensitive). Only one
type of opsin is expressed in the rod: Rh1 or rod opsin. Information from Bowmaker (2008).
Opsins SWS1 SWS2 LWS Rh2 Rh1

Green light
Spectral sensitivity UV-violet Blue-violet Red-green green
absorbing pigment
Wavelength (nm) 355-440 410-490 490-570 480-535 460-530
The transmission of the electrical signal from the photoreceptors to the

interneurons (i.e. bipolar cells, horizontal cells and amacrine cells) is through synapses,
which form the outer plexiform layer of the retina (OPL, Figure 1.10). The cell bodies
21
of the interneurons form the inner nuclear layer (INL, Figure 1.10). Bipolar cells
transmit the information directly or indirectly from the photoreceptors to the ganglion
cells via complex dendritic interactions within the inner plexiform layer (IPL, Figure
1.10). Horizontal cells and amacrine cells regulate the information of the rods and
cones, mainly at the level of the bipolar cells but also at the level of the ganglion cells in
the case of the amacrine cells. More specifically, horizontal cells interconnect
photoreceptors and bipolar cells in the IPL and help to integrate and regulate the
information by a process called lateral inhibition, which acts at the level of the bipolar
cells to enhance spatial differences in photoreceptor activation. Amacrine cells in
general are in contact with every class of neuron with the exception of the
photoreceptors, and many different types of amacrine cells exist (around 40 in teleosts,
(Wagner and Wagner, 1988). However, the functional role of many of these amacrine
cells is still unknown.
The nucleus of another class of cells, the Müller cells (a type of glial cells), is
also situated in the INL (Figure 1.10). Müller cells nearly span the entire thickness of
the retina and possess a symbiotic relationship with all the neurons, serving as support
cells. Several other functions have recently been discovered for Müller cells
(Reichenbach and Bringmann, 2013). For example, due to their funnel shape, high
refractive index and orientation the along the direction of light propagation, Müller cells
have recently been assigned the additional function of living optical fibre by mediating
the image transfer (Franze et al., 2007).
Ganglion cells are the only cells that possess an axon connecting to the visual
centres of the central nervous system and are thus the only cells sending information of
the perceived environment to the brain. The information is sent to the central nervous
system in the form of action potentials with different rates influenced by the activity of
the different interneurones (Rodieck 1998). There are several types of ganglion cells,
each specialised for coding a particular aspect of the visual environment (i.e. color,
movement, contrast) and projecting to different areas of the brain.
1.4.1.3. Image processing by the central nervous system

The brain of teleost fishes, like all vertebrates, is composed of five main areas;
the telencephalon, diencephalon, mesencephalon, cerebellum and medulla oblongata
(Figure 1.11, Nieuwenhuys, 1982; Meek and Nieuwenhuys, 1998; Lisney and Collin,
22
2006; Yopak, 2012). The telencephalon (Te), diencephalon (Di) and cerebellum (Cb)
are all integration areas implicated in the processing of multi-sensory information. The
olfactory bulbs (Ob), optic tectum (Ot, part of the mesencephalon) and medulla (Md)
are sensory areas of the brain shown to possess olfactory, visual, and auditory, gustatory
and lateral lines inputs. The diencephalon also possesses the pineal gland, a non-image-
forming photosensory organ similar in organisation to the retina and containing
photoreceptors and neurons. Since the pineal organ does not possess any focussing
mechanism, possess a high convergence of photoreceptors to neurons and a slow
response to stimuli, its main function is the detection of slowly changing ambient light
levels, implicated in the control of circadian and seasonal rhythms (Ekström and Meissl,
1997).
Figure 1.11. Brain of the rainbow trout Salmo gairdneri. Ob, olfactory bulb; Te, telencephalon;
Ot, optic tectum (part of the mesencephalon); Di, diencephalon; Cb, cerebellum, Md, medulla.
* indicate the location of the pineal gland. Modified from Meek and Nieuwenhuys (1998).
The structure and size of the different brain areas clearly reflect sensory
specialisations in teleost fishes. Since the optic tectum is the main centre for image
processing by receiving visual input directly from the retinal ganglion cells via the optic
nerve, analysis of its relative proportion can provide information about the importance
of the visual system of a particular species. Pelagic teleosts, for example, possess a
23
relatively large optic tectum, which appears to be the dominant brain area, an indication
of the importance of the visual system for survival in these species (Wagner, 2001a,b;
Lisney and Collin, 2006).
1.4.1.4. Visual Specialisations

The density of the retinal cells across the retina is not uniform and their
distribution reflects the symmetry of each species’ perceived visual environment (Collin
and Shand, 2003). Using the wholemount technique (Stone, 1981; Coimbra et al., 2006;
Ullmann et al., 2011), the general topographic distribution of the different neural cells
in the retina can be analysed. These topographic analyses together with knowledge of
the size of the eye, position of the eye in the head, degree of eye movement and size of
the visual field provide information on visual perception (Collin, 2008).
Two main specialisations, characterised by high density cell regions, can be

identified from topographic analyses; an area centralis and a horizontal streak (Figure
1.12).
Figure 1.12. Topographic maps of the ganglion cells showing different specialisations, (A) an
area centralis in the coral cod Cephalopholis miniatus (<10.0 to 47.0 x 103 ganglion cells per
mm2, redrawn from Collin and Pettigrew (1998a); (B) a horizontal streak in the small-spotted
dogfish Scyliorhinus canicula (<0.5 to 2.4 x 103 ganglion cells per mm2), redrawn from
Bozzano and Collin (2000).
An area centralis is characterised by a concentric increase of neural cell density

(Figure 1.12A). It is identified as an area of acute vision and is most commonly found in
species that live in a three dimensional and often enclosed environment (i.e. reef fishes)
(Collin and Shand, 2003). In contrast, an elongated increase in neural cells across the
24
retinal meridian is called a horizontal streak (Figure 1.12B). A horizontal streak allows
an animal to scan a broad horizon with maximum acuity (panoramic field) without
using distinctive eye movements. This type of specialisation is found in species
inhabiting an open area where they perceive their environment with an uninterrupted
view of a horizon. In the case of marine species the horizon is either defined by the
sand-water or air-water interfaces (Collin and Shand, 2003). Teleosts possess either one
of these specialisations or a combination of both, also called a dual visual specialisation.
In fact, several reef teleosts living in open water areas often possess a temporal area
centralis and a horizontal visual streak, the former most probably used in feeding
behaviour and the latter used for movement detection and/or predator surveillance
(Collin and Pettigrew, 1988b).
Some teleosts possess what is called a fovea or retinal pit, which is an

indentation of a particular region of the retina in which there are high densities of retinal
receptors. The fovea is a retinal region of high acuity, which may concomitantly
magnify the image due to the difference in refractive index between the vitreous humor
and the retinal tissue within the foveal depression. In humans and some other vertebrate
predators, the foveal region is devoid of rods and therefore is usually associated with
diurnal, bright light vision. In teleosts, four types of fovea have been identified based on
structural criteria (Collin and Collin, 1999).
1.4.2. Visual adaptations to the deep-sea

Among deep-sea teleosts, three types of eyes can be found: normal (similar to
shallow water species), tubular and degenerated. As most teleost species, including
lanternfishes, possess normal shaped eyes, visual adaptations in this type of eye will be
only considered here.
The basic structure and retina of a normal shaped teleost eye is similar to that of
most other vertebrates but variations occur depending on species, habitat diversity,
visibility, size and mobility of the prey, size and feeding strategies of potential predators
and the symmetry of each species’ ecological niche (Wagner et al., 1998). Because of
the low light environment and the predominance of small bioluminescent flashes, the
eyes of deep-sea fishes require enhanced sensitivity rather than high acuity. To enhance
sensitivity, the eyes of deep-sea species are adapted to optimize light collection and
extend the visual field (Wagner et al., 1998). The most obvious adaptation is the
25
presence of enlarged eyes, relative to body size, with a large pupil area. However,
several other visual adaptations are found in deep-sea teleosts and a brief description of
each specialisation is given below.
Aphakic gap. An aphakic gap is a gap between the lens and iris that allows the light
from oblique angles to reach the retina, often without being focussed or refracted by the
lens. These gaps increase the chance of photon capture in specific parts of the visual
field but often to the detriment of resolution. Some species possess a rostral aphakic gap
that allows the image of a frontally located light stimulus to be focussed on the temporal
part of the retina. Other species have a circumlental aphakic gap, a gap all around the
lens, which maximises light collection from every direction (Locket, 1985; Warrant and
Locket, 2004).
Yellow lens. Yellow lenses act as short-wavelength absorbing filters and may help to
break the camouflage of counterilluminating species by enhancing the contrast between
bioluminescent and downwelling light (Somiya, 1982; Douglas et al. 1998).
Tapetum lucidum. Specific to organisms living in dim light environments, the tapetum
lucidum is very common amongst deep-sea fishes. It is a reflective structure or mirror
sitting at the back of the eye, which reflects light back through the retina, thereby
increasing the light availability to the photoreceptors. From an anatomical point of
view, two types of tapetum lucidum are known according to their location, within either
the choroid or the retinal pigment epithelium. Amongst deep-sea teleosts, the retinal
pigment epithelium tapetum is the most common type and can be composed of several
substances (guanine, lipid, uric acid, pteridine, melanoid, caroteinoid, Somiya, 1980;
Douglas et al., 1998). The tapetum usually appears silver, white or coloured due to the
level of constructive and destructive interference, the biochemical composition of the
reflecting material and the shape of the structures (eliciting either a diffuse or specular
reflection). The colour of the reflex (leaving the eye) depends on the spacing of the
layered structures and the available light (Douglas et al., 1998).
Fovea. Two types of foveas, shallow and deep, have been observed in deep-sea teleosts
(Wagner et al., 1998). These foveas, often associated with rostral aphakic gaps, were
found in at least 30 species of deep-sea teleosts, most of them from the family
Alepocephalidae (Collin et al., 2000). In the deep-sea environment, foveas are thought
26
to increase resolution and especially enhance the ability to detect movement (Herring,
2002). They might also play a role in depth perception and bioluminescent camouflage
breaking by providing a skewed image as an object passes across the visual field
(Locket 1985; Collin et al. 2000).
Pure rod retina. The majority of deep-sea fishes have developed a pure-rod retina in
response to the low light levels and therefore are truly scotopic (Munk, 1966). By only
possessing this type of photoreceptor, deep-sea teleosts increase the chance of photon
capture, making the eye more sensitive.
Increases in rod length, high photoreceptor density, grouped photoreceptors and

multibanks of rods. These four adaptations all enhance sensitivity by increasing the
chance of photon capture (Locket, 1985; Wagner et al., 1998; Collin et al. 1998).
Multibanked retina may also allow colour vision in single pigment species (Denton and
Locket, 1989; Denton 1990) or at least hue discrimination in the blue to greenish-yellow
part of the spectrum, which could be useful for breaking camouflage by
counterillumination.
Single visual pigment. In the deep-sea environment, fishes appear to have evolved
pigments with a maximum absorbance that matches more or less the wavelengths of
light that penetrate to these depths. As a result, most deep-sea fishes (more than 80%)
possess a single visual pigment within their photoreceptors with a λmax between 468-494
nm (Douglas and Partridge, 1997; Douglas et al., 1998).
1.4.3. Myctophid vision

Our knowledge of myctophid sensory systems is very poor despite the fact that
they play a major role in the deep-sea ecosystem and are one of the most numerous
fishes in the world’s oceans. The sensory modalities of olfaction (Lawry, 1972; Lawry,
1973a) and lateral line (Lawry, 1973b; Marshall, 1954) have been described very briefly
but very few studies have examined the visual system in myctophids, which, like most
mesopelagic fishes, principally rely on vision (Wagner, 2001a). To date, the most
comprehensive studies on myctophid vision are Bozzano et al. (2007) and Turner et al.
(2009), which focus mainly on photoreceptors and the sensitivities of their visual
pigments. Bozzano et al. (2007) studied the eye and retina of myctophid larvae at
different stages. They found both cones and rods in the retina of different larval stages,
27
with a predominance of rods, and found a drastic decrease in cone numbers and an
increase in rod outer segment length during development. Moreover, cones were almost
totally absent in post-larval stages (Bozzano et al., 2007) and one can ask the question if
cones are totally absent in adults or are just found in very low numbers. Turner at al.
(2009) studied the spectral sensitivity of the pigment present in the rod photoreceptors
of 58 different species of myctophids. Most of the Myctophidae were single pigment
species and only four species (Cerastocopelus warmingii, Hygophum proximum,
Myctophum aurolaternatum, M. nitidulum), belonging to both sub-families, possessed
two visual pigments. The wavelength of maximum absorption (λmax) of myctophids
visual pigments falls between 480 and 492 nm (Turner et al., 2009; Douglas and
Partridge, 1997; Douglas et al., 1998; Hasegawa et al., 2008; Partridge et al., 1992) and
are therefore well adapted to the visualisation of blue-green bioluminescent light, which
is the most common and emitted by most deep-sea organisms. However, using a
mathematical model, Turner et al. (2009) suggested that some of the myctophid species
possessing two visual pigment, one being long-wave shifted, may be sensitive to the far-
red bioluminescence produced by some of their predators, the stomiid dragonfishes.
Very few data are available on myctophid visual perception and how they
sample their visual environment. These data have been summarised in Table 1.3.
Myctophids appear to have evolved eyes designed to enhance sensitivity. They possess
aphakic gaps (Lawry, 1974), a pure rod retina (Vilter, 1951; Pankhurst, 1987; O'Day
and Fernandez, 1976), a tapetum lucidum (O'Day and Fernandez, 1976), a high
photoreceptor density (Vilter, 1951; Pankhurst, 1987; O'Day and Fernandez, 1976), and
a rather unspecialised retina with poor acuity (Collin and Partridge, 1996; Wagner et al.,
1998). However, these data have been compiled from very few species and most of the
studies only examined one or a few of these characteristics.
To summarise, myctophid eyes seem to be highly specialised, probably to

efficiently detect bioluminescent flashes and downwelling light. However, important
characteristics of their visual system are still poorly known especially in regards to the
great variability between species with respect to the immense range of ecological niches
they occupy and their visual demands.
28
Table 1.3. Summary of the visual characteristics available in the literature for the Myctophidae family. * Results including all the neural cells present in the ganglion
Chapter 1
cell layer (ganglion cells + amacrine cells). Cell densities in x103 cells/mm2. Spatial resolving power (SPR) in cycles per degree of arc.
Species Peak GC Total PR density Retinal specialisation SPR Other observations References
density number GC
Diaphus effulgens 16.4* 830, 000 Concentric increase toward periphery* Wagner et al., 1998
Diaphus rafinesquii 33.3* Concentric increase toward periphery* Wagner et al., 1998
Lampanyctus ater 26.3* / 7.4 268,000 Concentric increase toward periphery* 2.0 Wagner et al., 1998
Slight area centralis centrally
L. crocodilus 500 Pure rod retina, large Vilter, 1951

number amacrine cells
L. festivus 20.4* / 5.7 415,000 Concentric increase toward periphery* 1.3 Wagner et al., 1998
Slight area centralis centrally
L. hectoris 2.4 to 31 177 to 4872 Single bank, Pankhurst, 1987

29
pure rod retina
L. macdonaldi 17.1* / 3.6 931,000 Concentric increase toward periphery* Collin and Partridge, 1996
Slight area centralis centrally Wagner et al., 1998
Lobianchia gemellarii 17.5 Concentric increase toward periphery* Wagner et al., 1998
Myctophum punctatum 19.4* 905,000 Concentric increase toward periphery* Collin and Partridge, 1996
Wagner et al., 1998
Stenobrachius leucopsarus 500 Pure rod retina, O'Day and Fernandez, 1976
choroidal tapetum
Tarletonbeania crenularis Aphakic gap Lawry, 1974
Introduction
1.5. Visual Ecology

The visual system of different organisms strongly reflects each species’ ecology
and behaviour by being adapted to their life style and ambient environmental
constraints. This can be seen from the range of adaptations between species from
different environments.
In the marine environment, visual conditions are influenced by the light

intensity, turbidity (amount of particulate or dissolved organic matter), the type of
habitat (benthic, pelagic, deep-sea), prey and predator relationships (size, color), season,
and size and development of the fish (Sandström, 1999). For example, deep-sea species
have evolved sensitive eyes adapted for viewing in dim conditions and for viewing
bioluminescent point sources of light (Wagner et al., 1998). Similarly, reef fishes have
developed different retinal regions for acute vision depending on the type of
environment they live in (well lit, enclosed or open water environments, Collin and
Pettigrew, 1988a; Collin and Pettigrew, 1988b).
Myctophids show a great variability in their ecology and behaviour by inhabiting

different environments, possessing different migration patterns, photophore patterns,
sexual dimorphism in luminous organs, and by using different life style strategies. As a
result, one would expect to find, in addition to the visual adaptations to the deep-sea
environment, different adaptations in the visual system between species with different
life styles (migratory vs non-migratory), different luminous organ patterns (luminous
organs species vs non luminous organ species), and sexual dimorphism in luminous
organs (male vs female, sexually dimorphic species vs non-sexually dimorphic species).
Unfortunately, most studies have only addressed a specific part of the visual system of
these fishes and almost none of these studies have referred to the great variability
observed in their behaviour and ecology. Turner et al. (2009) found no significant
differences in spectral sensitivity between different species of myctophid. These authors
concluded that the lack of variation in the wavelength of maximum absorption between
species might be more a consequence of a shared evolutionary history than adaptive
variation (Turner et al., 2009). However, several other visual characteristics could be
compared between and within species. Species presenting sexual dimorphism in the
caudal organs appear to have bigger eyes than those without caudal dimorphism (Tsarin,
2001). If this is true, we might also reveal different retinal adaptations between species
with and without sexual dimorphism and between males and females of sexually
30
dimorphic species. These different adaptations, if present, could provide a solid

argument in favour of the hypothetical use of the luminous organs for intraspecific
communication. Similarly, most myctophids possess a tapetum lucidum, which is
thought to enhance sensitivity of the eye by increasing the chance of photon capture by
the photoreceptors. Some of these tapeta are composed of guanine arranged in parallel
plates within the retinal pigment epithelial cells (Locket, 1977). However, the presence
of another type of tapetum, the choroidal tapetum, has also been described in
myctophids (Somiya, 1980; Nicol, 1989). If the presence of two different types of
tapetum is confirmed, could these different types be related to the ecology and/or
behaviour of these fishes and their potential need for different sensitivity levels?
1.6. Aim of this thesis

The main aim of this study is to better understand deep-sea fish visual
adaptations in relation to their environment and possibly shed some light on the function
of bioluminescence in the deep-sea. Using anatomical, ecological and phylogenetic
approaches, this PhD thesis investigates the visual system of a large number of
representatives within the Myctophidae family. The visual perception of these fishes is
investigated in detail at macro- micro- and ultrastructural levels from the optical
detection of a signal to its reception within the retinal ganglion cells and compared to
each species’ ecological and behavioural characteristics taking into account the
evolutionary history of the family.
This PhD thesis is presented as a series of papers, each paper addressing a

specific component of the visual system of myctophids: relative eye size (Chapter 2),
general morphology of the myctophid eye (Chapter 3), photoreceptor and optical
sensitivity (Chapter 4), spectral tuning (Chapter 5), and ganglion cells and spatial
resolving power (Chapter 6).
1.7. References
Abrahams MV, Townsend LD. 1993. Bioluminescence in dinoflagellates: a test of the
burgular alarm hypothesis. Ecology 74(1):258-260.
Anctil M. 1972. Stimulation of bioluminescence in lanternfishes (Myctophidae). II.

Canadian Journal of Zoology 50(2):233-237.
31
Barnes AT, Case JF. 1974. Luminescence of lanternfish (Myctophidae) - Spontaneous

activity and responses to mechanical, electrical, and chemical stimulation.
Journal of Experimental Marine Biology and Ecology 15(2):203-221.
Battaglia P, Andaloro F, Consoli P, Esposito V, Malara D, Musolino S, Peda C, Romeo

T. 2013. Feeding habits of the Atlantic bluefin tuna, Thunnus thynnus (L. 1758),
in the central Mediterranean Sea (Strait of Messina). Helgoland Marine
Research 67(1):97-107.
Beebe W. 1934. Half mile down. New York: Hardcourt, Brace and Company. 344 p.
Bowmaker JK. 2008. Evolution of vertebrate visual pigments. Vision Research

48(20):2022-2041.
Bozzano A, Collin SP. 2000. Retinal Ganglion Cell Topography in Elasmobranchs.

Brain, Behavior and Evolution 55(4):191-208.
Bozzano A, Pankhurst PM, Sabatés A. 2007. Early development of eye and retina in
lanternfish larvae. Visual Neuroscience 24:423-436.
Carey PW. 1992. Fish prey species of the New Zealand fur seal (Arctocephalus forsteri,
Lesson). New Zealand Journal of Ecology 16(1):41-46.
Case JF, Warner J, Barnes AT, Lowenstine M. 1977. Bioluminescence of lantern fish
(Myctophidae) in response to changes in light intensity. Nature 265(5590):179-
181.
Case JF, Strause L. 1978. Neurally controlled luminescent systems. In: Herring PJ,
editor. Bioluminescence in action. London: Academic Press. p 331-366.
Catul V, Gauns M, Karuppasamy PK. 2011. A review on mesopelagic fishes belonging

to family Myctophidae. Reviews in Fish Biology and Fisheries 21(3):339-354.
Cherel Y, Fontaine C, Richard P, Labat JP. 2010. Isotopic niches and trophic levels of
myctophid fishes and their predators in the Southern Ocean. Limnology and
Oceanography 55(1):324-332.
Childress JJ, Taylor SM, Cailliet GM, Price MH. 1980. Patterns of growth, energy
utilization and reproduction in some mesopelagic and bathypelagic fishes off
Southern California. Marine Biology 61(1):27-40.
Christophe B, Baguet F. 1982. Luminescence of isolated photophores and supracaudal

gland from Myctophum punctatum: Electrical stimulation. Comparative
Biochemistry and Physiology Part A: Physiology 71(1):131-136.
Claes JM, Mallefet J. 2009. Hormonal control of luminescence from lantern shark
(Etmopterus spinax) photophores. Journal of Experimental Biology
212(22):3684-3692.
Claes JM, Aksnes DL, Mallefet J. 2010. Phantom hunter of the fjords: Camouflage by
counterillumination in a shark (Etmopterus spinax). Journal of Experimental
Marine Biology and Ecology 388(1-2):28-32.
32
Clarke TA. 1973. Some aspects of the ecology of lanternfishes (Myctophidae) in the
Pacific Ocean near Hawaii. Fishery Bulletin 71(2):401-434.
Coimbra JP, Marceliano MLV, Andrade-da-Costa BLS, Yamada ES. 2006. The retina
of tyrant flycatchers: topographic organization of neuronal density and size in
the ganglion cell layer of the Great Kiskadee Pitangus sulphuratus and the
Rusty Margined Flycatcher Myiozetetes cayanensis (Aves: Tyrannidae). Brain,
Behavior and Evolution 68(1):15-25.
Collin SP, Pettigrew JD. 1988a. Retinal topography in reef teleosts. I. Some species
with well-developed areae but poorly-developed streaks. Brain, Behavior and
Evolution 31:269-282.
Collin SP, Pettigrew JD. 1988b. Retinal topography in reef teleosts. II. Some species
with prominent horizontal streaks and high-density areae. Brain, Behavavior and
Collin SP, Partridge JC. 1996. Retinal specializations in the eyes of deep-sea teleosts.
Journal of Fish Biology 49:157-174.
Collin SP, Collin H. 1999. The foveal photoreceptor mosaic in the pipefish,
Corythoichthyes paxtoni (Syngnathidae, Teleostei). Histology and
Histopathology 14(2):369.
Collin SP, Lloyd DJ, Wagner H-J. 2000. Foveate vision in deep-sea teleosts: a
comparison of primary visual and olfactory inputs. Philosophical Transactions
of the Royal Society of London Series B, Biological Sciences 355:1315-1320.
Collin SP, Shand J. 2003. Retinal sampling and the visual field in fishes. In: Collin SP,
Marshall NJ, editors. Sensory Processing in Aquatic Environments: Springer
New York. p 139-169.
Collin SP. 2008. A web-based archive for topographic maps of retinal cell distribution
in vertebrates. Clinical and Experimental Optometry 91:85-95.
Collins MA, Xavier JC, Johnston NM, North AW, Enderlein P, Tarling GA, Waluda
CM, Hawker EJ, Cunningham NJ. 2008. Patterns in the distribution of
myctophid fish in the northern Scotia Sea ecosystem. Polar Biology 31(7):837-
851.
Denton EJ, Gilpin-Brown JB, Wright PG. 1970. On the 'filters' in the photophores of
mesopelagic fish and on a fish emitting red light and especially sensitive to red
light. Journal of Physiology 208:72-73.
Denton EJ, Herring PJ, Widder EA, Latz MF, Case JF. 1985. The roles of filters in the
photophores of oceanic animals and their relation to vision in the oceanic
environment. Philosophical Transactions of the Royal Society of London Series
B, Biological Sciences 225(1238):63-97.
Denton EJ, Locket NA. 1989. Possible wavelength discrimination by multibank retinae
in deep-sea fishes. Journal of the Marine Biological Association of the United
Kingdom 69(2):409-435.
33
Denton EJ. 1990. Light and vision at depths greater than 200 metres. In: Herring PJ,
Campbell AK, Whitfield M, Maddock L, editors. Light and life in the sea:
Cambridge University Press. p 127-148.
Douglas RH, Partridge JC. 1997. On the visual pigments of deep-sea fish. Journal of
Fish Biology 50(1):68-85.
Douglas RH, Partridge JC, Marshall NJ. 1998. The eyes of deep-sea fish I: Lens
pigmentation, tapeta and visual pigments. Progress in Retinal and Eye Research
17(4):597-636.
Edwards AS, Herring PJ. 1977. Observations on the comparative morphology and
operation of the photogenic tissues of myctophid fishes. Marine Biology
41(1):59-70.
Ekström P, Meissl H. 1997. The pineal organ of teleost fishes. Reviews in Fish Biology
and Fisheries 7(2):199-284.
Esposito V, Battaglia P, Castriota L, Finoia MG, Scotti G, Andaloro F. 2009. Diet of

Atlantic lizardfish, Synodus saurus (Linnaeus, 1758) (Pisces: Synodontidae) in
the central Mediterranean Sea. Scientia Marina 73(2):369-376.
Evseenko S. 2006. On species composition and distribution of lanternfish larvae

(Myctophidae) in the eastern South Pacific. Journal of Ichthyology 46:110-115.
Falk K, Jensen J-K, Kampp K. 1992. Winter diet of Atlantic puffins (Fratercula
arctica) in the northeast Atlantic. Colonial Waterbirds 15(2):230-235.
Flynn A, Kloser R. 2012. Cross-basin heterogeneity in lanternfish (family Myctophidae)

assemblages and isotopic niches (δ13C and δ15N) in the southern Tasman Sea
abyssal basin. Deep Sea Research Part I: Oceanographic Research Papers
69:113-127.
Flynn AJ, Paxton JR. 2012. Spawning aggregation of the lanternfish Diaphus danae
(family Myctophidae) in the north-western Coral Sea and associations with tuna
aggregations. Marine and Freshwater Research 63(12):1255-1271.
Flynn AJ, Williams A. 2012. Lanternfish (Pisces: Myctophidae) biomass distribution

and oceanographic-topographic associations at Macquarie Island, Southern
Ocean. Marine and Freshwater Research 63(3):251-263.
Franze K, Grosche J, Skatchkov SN, Schinkinger S, Foja C, Schlid D, Uckermann O,

Travis K, Reichenbach A, Guck J. 2007. Müller cells are living optical fibers in
the vertebrate retina. Proceedings of the National Academy of Sciences of the
United States of America 104(20):8287-8292.
Fraser-Brunner A. 1949. A classification of the fishes of the family Myctophidae.

Proceedings of the Zoological Society of London 118(4):1019-1060.
Gartner JV. 1991. Life histories of three species of lanternfishes (Pisces: Myctophidae)
from the eastern Gulf of Mexico. Marine Biology 111(1):11-20.
34
Gartner JV. 1993. Patterns of reproduction in the dominant lanternfish species (Pisces:
Myctophidae) of the eastern Gulf of Mexico, with a review of reproduction
among tropical-subtropical Myctophidae. Bulletin of Marine Science 52(2):721-
750.
Gjosaeter J. 1973. Age, growth, and mortality of the myctophid fish, Benthosema
glaciale (Reinhardt), from western Norway. Sarsia 52:1-14.
Goldsworthy SD, Lewis M, Williams R, He X, Young JW, van den Hoff J. 2002. Diet
of Patagonian toothfish (Dissostichus eleginoides) around Macquarie Island,
South Pacific Ocean. Marine and Freshwater Research 53(1):49-57.
Green K, Williams R, Burton HR. 1991. The diet of Antactic fur seals during the late
autumn and early winter around Heard Island. Antarctic Science 3(4):359-361.
Green K, Williams R, Green MG. 1998. Foraging ecology and diving behaviour of
Macaroni Penguins Eudyptes chrysolophus at Heard Island. Marine Ornithology
26(1-2):27-34.
Haddock SHD, Dunn CW, Pugh PR, Schnitzler CE. 2005. Bioluminescent and red-
fluorescent lures in a deep-sea siphonophore. Science 309(5732):263.
Haddock SHD, Moline MA, Case JF. 2010. Bioluminescence in the sea. Annual Review
of Marine Science 2(1):443-493.
Hansen K, Herring PJ. 1977. Dual bioluminescent systems in the anglerfish genus
Linophryne (Pisces: Ceratioidea). Journal of Zoology 182(1):103-124.
Hasegawa EI, Sawada K, Abe K, Watanabe K, Uchikawa K, Okazaki K, Toyama M,

Douglas RH. 2008. The visual pigments of a deep-sea myctophid fish
Myctophum nitidulum Garman; an HPLC and spectroscopic description of a
non-paired rhodopsin–porphyropsin system. Journal of Fish Biology 72:937-
945.
Hattori S. 1964. Studies on fish larvae in the Kuroshio and adjacent waters. Bulletin of
Tokai Regional Fisheries Research Laboratory 40:1-111.
Haygood MG, Edwards DB, Mowlds G, Rosenblatt RH. 1994. Bioluminescence of

myctophid and stomiiform fishes is not due to bacterial luciferase. Journal of
Experimental Zoology 270(2):225-231.
Herring PJ. 1976. Bioluminescence in decapod Crustacea. Journal of the Marine

Biological Association of the United Kingdom 56:1029-1047.
Herring PJ. 1985. How to survive in the dark: bioluminescence in the deep sea.
Symposia of the Society for Experimental Biology 39:323-350.
Herring PJ. 1988. Copepod luminescence. Hydrobiologia 167-168(1):183-195.
Herring PJ. 2002. The biology of the deep-sea: Oxford University Press. 314 p.
Herring PJ. 2007. Sex with the lights on? A review of bioluminescent sexual
dimorphism in the sea. Journal of the Marine Biological Association of the
United Kingdom 87(04):829-842.
35
Holton AA. 1969. Feeding behavior of a vertically migrating lanterfish. Pacific Science
23(3):325-331.
Horn PL, Forman JS, Dunn MR. 2013. Moon phase influences the diet of southern
Ray's bream Brama australis. Journal of Fish Biology 82(4):1376-1389.
Hulley PA. 1981. Results of the research cruises of the FRV "Walther Herwig" to south
America, LVII: family Myctophidae (Osteichthyes, Myctophiformes). Archiv
für Fischereiwissenschaft 31:1-300.
Hulley PA. 1984. Myctophidae. In: Whitehead PJP, Bauchot M-L, Hureau J-C, Nielsen
J, Tortonese E, editors. Fishes of the North-eastern Atlantic and the
Mediterranean: Unesco. p 429-483.
Hulley PA. 1990. Family Myctophidae. In: Gon O, Heemstra PC, editors. Fishes of the
Southern Ocean. Grahamstown: J.L.B. Smith Institute of Ichthyology. p 146-
178.
Hulley PA. 1992. Upper-slope distributions of oceanic lanternfishes (family:

Myctophidae). Marine Biology 114(3):365-383.
Hulley PA. 1994. Lanternfishes. In: Paxton JR, Eschmeyer WN, editors. Encyclopedia
of fishes: UNSW Press. p 127-128.
Hulley PA, Paxton JR. In press. Myctophidae, lanternfishes. In: Carpenter KE, editor.
The Living Marine Resources of the Eastern Central Atlantic. Rome: FAO:In
press.
Ishihara S, Kubota T. 1997. Food habits of the lanternfish Benthosema pterotum in the
East China Sea and the Yellow Sea. Nippon Suisan Gakkaishi 63(4):522-530.
Jackson S. 1988. Diets of the White-chinned Petrel and Sooty Shearwater in the
southern Benguela region, South Africa. The Condor 90(1):20-28.
Johnsen S, Widder EA, Mobley CD. 2004. Propagation and perception of

bioluminescence: Factors affecting counterillumination as a cryptic strategy.
Biological Bulletin 207(1):1-16.
Karnella C. 1987. Family Myctophidae, lanternfishes. In: Gibbs RH, Krueger WH,
editors. Biology of midwater fishes of the bermuda Ocean Acre. Washington,
D.C.: Smithsonian Institution Press. p 51-168.
Kawaguchi K, Ikeda H, Tamura M, Ueyanagi S. 1972. Geographical distribution of

surface-migrating myctophid fishes (genus Myctophum) in the tropical and
subtropical Pacific and Indian Oceans. Bulletin of the Far Seas Fisheries
Research Laboratory 6:23-37.
Kawamura A, Fujii F, Nemoto T. 1988. Forage of the mesopelagic fishes,

Symbolophorus californiensis (Eigenmann and Eigenmann) and Tarletonbeania
taylori Mead caught in gill nets in the northwestern North Pacific Ocean.
Bulletin of the Ocean Research Institute, University of Tokyo 26(Part II): 143-
159.
36
Kinzer J, Böttger-Schnack R, Schulz K. 1993. Aspects of horizontal distribution and

diet of myctophid fish in the Arabian Sea with reference to the deep water
oxygen deficiency. Deep Sea Research Part II: Topical Studies in Oceanography
40(3):783-800.
Koubbi P, Moteki M, Duhamel G, Goarant A, Hulley PA, O'Driscoll R, Ishimaru T,

Pruvost P, Tavernier E, Hosie G. 2011. Ecoregionalization of myctophid fish in
the Indian sector of the Southern Ocean: Results from generalized dissimilarity
models. Deep-Sea Research Part II: Topical Studies in Oceanography 58(1-
2):170-180.
Kozlov AN. 1995. A review of the trophic role of mesopelagic fish of the family
Myctophidae in the Southern Ocean ecosystem. CCAMLR Science 2:71-77.
Kronstrom J, Mallefet J. 2010. Evidence for a widespread involvement of NO in control

of photogenesis in bioluminescent fish. Acta Zoologica 91(4):474-483.
Land MF. 1990. Optics of the eyes of marine animals. In: Herring PJ, Campbell AK,
Whitfield M, Maddock L, editors. Light and life in the sea: Cambridge
University Press. p 149-166.
Latz MI. 1995. Physiological mechanisms in the control of bioluminescent

countershading in a midwater shrimp. Marine and Freshwater Behaviour and
Physiology 26(2-4):207-218.
Lawry JV. 1972. Olfactory receptors of lantern fishes. Anatomical Record 172(2):352.
Lawry JV. 1973a. Olfactory epithelium of lantern fish, Tarletonbeania crenularis

(Myctophidae). Zeitschrift für Zellforschung und Mikroskopische Anatomie
138(1):31-39.
Lawry JV. 1973b. The trigeminofacial innervation of the cephalic lateral line organs and
photophores of the lantern fish Tarletonbeania crenularis (Myctophidae).
Marine and Freshwater Behaviour and Physiology 1(1):285-293.
Lawry JV. 1974. Lantern fish compare downwelling light and bioluminescence. Nature
247(5437):155-157.
Linkowski TB. 1996. Lunar rhythms of vertical migrations coded in otolith

microstructure of North Atlantic lanternfishes, genus Hygophum (Myctophidae).
Marine Biology 124(4):495-508.
Lisney TJ, Collin SP. 2006. Brain morphology in large pelagic fishes: a comparison
between sharks and teleosts. Journal of Fish Biology 68(2):532-554.
Lloyd JE. 1971. Bioluminescent communication in insects. Annual Review of

Entomology 16(1):97-122.
Lloyd JE. 1983. Bioluminescence and communication in insects. Annual Review of

Entomology 28(1):131-160.
Locket NA. 1977. Adaptations to the deep-sea environment. In: Crescitelli F, editor.
Handbook of sensory physiology, vol II/5 Berlin: Springer-Verlag. p 67-192.
37
Locket NA. 1985. The multiple bank rod fovea of Bajacalifornia drakei, an
alepocephalid deep-sea teleost. Proceedings of the Royal Society of London.
Series B: Biological Sciences 224(1234):7-22.
Marshall NB. 1954. Aspects of deep sea biology. London: Hutchinson. 380 p.
Matthiessen L. 1882. Über die Beziehungen, welche zwischen dem Brechungsindex des
Kernzentrums der Krystalllinse und den Dimensionen des Auges bestehen.
Pflügers Archiv für die gesamate Physiologie des Menschen und der Tiere
27:510-523.
Matthiessen L. 1886. Über den physikalisch-optischen Bau des Auges der Cetacean und
der Fische. Pflügers Archiv für die gesamate Physiologie des Menschen und der
Tiere 38:521-528.
Meek J, Nieuwenhuys R. 1998. Holosteans and Teleosteans. In: Nieuwenhuys R, Ten
Donkelaar HJ, Nicholson C, editors. The Central Nervous System of Vertebrates
Vol 2. Berlin: Springer. p 759-937.
Mensinger AF, Case JF. 1990. Luminescent properties of deep-sea fish. Journal of
Experimental Marine Biology and Ecology 144(1):1-15.
Mesinger AF, Case JF. 1992. Dinoflagellate luminescence increases susceptibility of

zooplankton to teleost predation. Marine Biology 112(2):207-210.
Moku M, Kawaguchi K, Watanabe H, Ohno A. 2000. Feeding habits of three dominant

myctophid fishes, Diaphus theta, Stenobrachius leucopsarus and S. nannochir,
in the subarctic and transitional waters of the western North Pacific. Marine
Ecology Progress Series 207:129-140.
Moku M, Kawaguchi K. 2008. Chemical composition of three dominant myctophid

fish, Diaphus theta, Stenobrachius leucopsarus, and S. nannochir, in the
subarctic and transition waters of the western North Pacific. Journal of the
Marine Biological Association of the United Kingdom 88(4):843-846.
Molday RS, Zhang K. 2010. Defective lipid transport and biosynthesis in recessive and
dominant Stargardt macular degeneration. Progress in Lipid Research 49(4):476-
492.
Moser HG, Ahlstrom EH. 1970. Development of lanternfishes (family myctophidae) in

the California Current. Part I. Species with narrow-eyed larvae. Science
Bulletin, Los Angeles County Natural History Museum 7:1-145.
Moser HG, Ahlstrom EH. 1972. Development of lanternfish, Scopelopsis

multipunctatus Brauer 1906, with a discussion of its phylogenetic position in the
family Myctophidae and its role in a proposed mechanism for the evolution of
photophore patterns in lanternfishes. Fishery Bulletin 70(3):541–564.
Moser HG, Ahlstrom EH. 1974. Role of larval stages in systematic investigations of
marine teleosts: the Myctophidae, a case study. Fishery Bulletin 72(2):391-413.
Moser HG, Ahlstrom EH, Paxton JR. 1984. Myctophidae: development. American
Society of Ichthyologists and Herpetologists Special Publication:218-239.
38
Munk O. 1966. Ocular anatomy of some deep-sea teleosts. Dana Report 70:5-62.
Nafpaktitis BG, Backus RH, Craddock JE, Haedrich RL, Robison BH, Karnella C.
1977. Family Myctophidae. In: Gibbs Jr RH, Berry FH, Böhlke JE, Cohen DM,
Collette BB, Eschmeyer WN, Mead GW, Merriman D, Pietsch TW, editors.
Fishes of the Western North Atlantic: Memoir, Sears Foundation for Marine
Research. p 13-265.
Nealson KH, Hastings JW. 1979. Bacterial bioluminescence: its control and ecological
significance. Microbiology and Molecular Biology Reviews 43(4):496-518.
Nicol JAC. 1960. Spectral composition of the light of the lantern-fish, Myctophum
punctatum. Journal of the Marine Biological Association of the United Kingdom
39(1):27-32.
Nicol JAC. 1989. The eyes of fishes: Oxford University Press. 308 p.
Nieuwenhuys R. 1982. An overview of the organization of the brain of Actinopterygian

fishes. American Zoologist 22(2):287-310.
O'Day WT, Fernandez HR. 1976. Vision in lanternfish Stenobrachius leucopsarus

(Myctophidae). Marine Biology 37(2):187-195.
Ogawa T. 1961. Study on the fishes of the family Myctophidae in the northeastern sea
area along the Pacific coast of Japan. Part 1. Species and distribution. Bulletin of
Tohoku Regional Fisheries Research Laboratory 19:81-89.
Olivar MP, Beckley LE. 1994. Influence of the Agulhas Current on the distribution of
lanternfish larvae off the southeast coast of Africa. Journal of Plankton Research
16(12):1759-1780.
Olivar MP, Moser HG, Beckley LE. 1999. Lanternfish larvae from the Agulhas current
(SW Indian Ocean). Scientia Marina 63(2):101-120.
Pankhurst NW. 1987. Intra- and interspecific changes in retinal morphology among
mesopelagic and demersal teleost from the slope waters of New-Zealand.
Environmental Biology of Fishes 19(4):269-280.
Parry M. 2006. Feeding behavior of two ommastrephid squids Ommastrephes bartramii

and Sthenoteuthis oualaniensis off Hawaii. Marine Ecology Progress Series
318:229-235.
Partridge JC, Archer SN, Vanoostrum J. 1992. Single and multiple visual pigments in
deep-sea fishes. Journal of the Marine Biological Association of the United
Kingdom 72(1):113-130.
Paxton JR. 1967. A distributional analysis for the lantern fishes (family Myctophidae)
of the San Pedro Basin, California. Copeia 1967:422-440.
Paxton JR. 1972. Oeteology and relationships of the lanternfishes (family

Myctophidae). Science Bulletin, Los Angeles County Natural History Museum
13:1-81.
39
Paxton JR, Ahlstrom EH, Moser HG. 1984. Myctophidae: relationships. In: Moser HG,
Richards WJ, Cohen DM, Fahay MP, Kendall AW, Richerdson SL, editors.
Ontogeny and Systematic of Fishes: American Society of Ichthyologist and
Herpetologist Special Publication. p 239-244.
Paxton JR, Hulley PA. 1999. Families Neoscopelidae, Myctophidae. In: Carpenter KE,
Niem VH, editors. The living marine resourses of the Western Central Pacific.
Rome: FAO Species Identification Guide for Fisheries Purposes. p 1955-1965.
Pereira JN, Neves VC, Prieto R, Silva MA, Cascão I, Oliveira C, Cruz MJ, Medeiros
JV, Barreiros JP, Porteiro FM, Clarke D. 2011. Diet of mid-Atlantic Sowerby's
beaked whales Mesoplondon bidens. Deep Sea Research Part I: Oceanographic
Research Papers 58(11): 1084-1090.
Podrazhanskaya SG. 1993. Feeding habits of mesopelagic species of fish and estimation
of plankton graze in the northwest Atlantic. NAFO Scientific Council Studies
19:79-85.
Poulsen JY, Byrkjedal I, Willassen E, Rees D, Takeshima H, Satoh TP, Shinohara G,

Nishida M, Miya M. 2013. Mitogenomic sequenses and evidence from unique
gene rearrangements cooroborate evolutionary relationships of Myctophiformes
(Neoteleostei). BMC Evolutionary Biology 13: 111.
Pusch C, Schnack-Schiel S, Mizdalski E, von Westernhagen H. 2004. Feeding ecology

of three myctophid species at the Great Meteor Seamount (North-east Atlantic).
Archive of Fishery and Marine Research 51(1-3):251-271.
Pusineri C, Magnin V, Meynier L, Spitz J, Hassani S, Ridoux V. 2007. Food and

feeding ecology of the common dolphin (Delphinus delphis) in the oceanic
northeast Atlantic and comparison with its diet in Neritic areas. Marine Mammal
Science 23(1):30-47.
Ray DL. 1950. The peripheral nervous system of lampanyctus leucopsarus. Journal of
Morphology 87(1):61-178.
Reichenbach A, Bringmann A. 2013. New functions of Müller cells. Glia 61(5):651-

678.
Robison B. 1973. A system for maintaining midwater fishes in captivity. Journal of the
Fisheries Research Board of Canada 30(1):126-128.
Robison BH. 1984. Herbivory by the myctophid fish Ceratoscopelus warmingii. Marine
Biology 84(2):119-123.
Robison BH. 1992. Bioluminescence in the benthopelagic holothurian Enypniastes

eximia. ournal of the Marine Biological Association of the United Kingdom
72(2):463-472.
Robison BH, Reisenbichler KR, Hunt JC, Haddock SHD. 2003. Light production by the
arm tips of the deep-sea cephalopod Vampyroteuthis infernalis. The Biological
Bulletin 205(2):102-109.
Sadler JD. 1973. The focal length of the fish eye lens and visual acuity. Vision Research
13(2):417-423.
40
Sandström A. 1999. Visual ecology of fish–a review with special reference to percids.
Fiskeriverket Rapport 2:45-80.
Schonthaler HB, Franz-Odendaal TA, Hodel C, Gehring I, Geisler R, Schwarz H,

Neuhauss SCF, Dahm R. 2010. The zebrafish mutant bumper shows a
hyperproliferation of lens epithelial cells and fibre cell degeneration leading to
functional blindness. Mechanisms of Development 127:203-219.
Shreeve RS, Collins MA, Tarling GA, Main CE, Ward P, Johnston NM. 2009. Feeding
ecology of myctophid fishes in the northern Scotia Sea. Marine Ecology
Progress Series 386:221-236.
Somiya H. 1980. Fishes with eye shine: functional morphology of guanine type tapetum
lucidum. Marine Ecology Progress Series 2:9-26.
Somiya H. 1982. `Yellow lens' eyes of a stomiatoid deep-sea fish, Malacosteus niger.
Proceedings of the Royal Society of London Series B, Biological Sciences
215(1201):481-489.
Stiassny MLJ. 1996. Basal ctenosquamate relationships and the interrelationships of the
myctophiform (scopelomorph) fishes. In: Stiassny MLJ, Parenti LR, Johnson
GD, editors. Interrelationships of fishes: Academic Press. p 405-426.
Stone J. 1981. The whole mount handbook: a guide to the preparation and analysis of
retinal whole mounts. Sydney: Maitland Publications. 128 p.
Strauss O. 2005. The retinal pigment epithelium in visual function. Physiological

Reviews 85(3):845-881.
Takagi K, Yatsu A, Moku M, Sassa C. 2006. Age and growth of lanternfishes,

Symbolophorus californiensis and Ceratoscopelus warmingii (Myctophidae), in
the Kuroshio-Oyashio Transition Zone. Ichthyological Research 53(3):281-289.
Tanaka H, Sassa C, Ohshimo S, Aoki I. 2013. Feeding ecology of two lanternfishes

Diaphus garmani and Diaphus chrysorhynchus. Journal of Fish Biology
82(3):1011-1031.
Tsarin SA. 2001. Morphology of luminous organs of mesopelagic fishes in one group of
Myctophidae. Journal of Morphology (Abstract, Sixth International Congress of
Vertebrate Morphology) 248(3):292.
Tsarin SA. 2002. Vertical migrations of Myctophidae and peculiarities of their

behaviour. Advances in Ethology (Abstarct, 4th International Symposium on
Physiology and Behaviour of Wild and Zoo Animals) 37:79.
Tsuji FI, Haneda Y. 1971. Luminescent system in a myctophid fish, Diaphus elucens
Brauer. Nature 233(5322):623-624.
Turner JR, White EM, Collins MA, Partridge JC, Douglas RH. 2009. Vision in
lanternfish (Myctophidae): Adaptations for viewing bioluminescence in the
deep-sea. Deep-Sea Research Part I: Oceanographic Research Papers
56(6):1003-1017.
41
Tyler HR, Pearcy WG. 1975. The feeding habits of three species of lanternfishes
(family myctophidae) off Oregon, USA. Marine Biology 32(1):7-11.
Uchikawa K, Yamamura O, Hattori T. 2008. Feeding habits of mesopelagic fish

Lampanyctus jordani (Family : Myctophidae) over the continental slope off
Tohoku area, northern Japan. Fisheries Science 74(1):69-76.
Ullmann JFP, Moore BA, Temple SE, Fernandez-Juricic E, Collin SP. 2011. The retinal
wholemount technique: a window to understanding the brain and behaviour.
Brain Behavior and Evolution 79(1):26-44.
Villamizar N, Blanco-Vives B, Migaud H, Davie A, Carboni S, Sánchez-Vázquez FJ.

2011. Effects of light during early larval development of some aquacultured
teleosts: A review. Aquaculture 315:86-94.
Vilter V. 1951. Bases cyto-architectoniques de l'acuité visuelle chez un poisson abyssal,

le Lampanyctus crocodilus. Comptes Rendus des Séances de la Sociète de
Biologie et de ses Filiales 145(1-2):52-54.
Wagner H, Wagner E. 1988. Amacrine cells in the retina of a teleost fish, the roach
(Rutilus rutilus): a Golgi study on differentiation and layering. Philosophical
Transactions of the Royal Society of London Series B, Biological Sciences
321:263-324.
Wagner HJ. 1990. Retinal structure of fishes. In: Douglas RH, Djamgoz MBA, editors.
The visual system of fish: Chapman and Hall. p 109-157.
Wagner H-J, Fröhlich E, Negishi K, Collin SP. 1998. The eyes of deep-sea fish II.
Functional morphology of the retina. Progress in Retinal and Eye Research
17(4):637-685.
Wagner H-J. 2001a. Sensory brain areas in mesopelagic fishes. Brain, Behavior and
Evolution 57(3):117-133.
Wagner H-J. 2001b. Brain areas in abyssal demersal fishes. Brain, Behavior and
Evolution 57(6):301-316.
Walls GL. 1942. The vertebrate eye and its adaptive radiation. Bloomfield Hills: The
Cranbrook Institute of Science. 785 p.
Wang JTM, Chen CT. 2001. A review of lanternfishes (Families : Myctophidae and
Neoscopelidae) and their distributions around taiwan and the Tungsha Islands
with notes on seventeen new records. Zoological Studies 40(2):103-126.
Warrant EJ, Locket NA. 2004. Vision in the deep sea. Biological Reviews 79(3):671-
712.
Watanabe H, Moku M, Kawaguchi K, Ishimaru K, Ohno A. 1999. Diel vertical

migration of myctophid fishes (Family Myctophidae) in the transitional waters
of the western North Pacific. Fisheries Oceanography 8(2):115-127.
Watanabe H, Kubodera T, Ichii T, Kawahara S. 2004. Feeding habits of neon flying

squid Ommastrephes bartramii in the transitional region of the central North
Pacific. Marine Ecology Progress Series 266:173-184.
42
Watanabe H, Kubodera T, Ichii T, Sakai M, Moku M, Seitou M. 2008. Diet and sexual
maturation of the neon flying squid Ommastrephes bartramii during autumn and
spring in the Kuroshio-Oyashio transition region. Journal of the Marine
Biological Association of the United Kingdom 88(2):381-389.
Widder EA, Latz MI, Herring PJ, Case JF. 1984. Far Red Bioluminescence from Two
Deep-Sea Fishes. Science 225(4661):512-514.
Widder EA. 2010. Bioluminescence in the Ocean: Origins of Biological, Chemical, and
Ecological Diversity. Science 328(5979):704-708.
Wisner RL. 1976. The taxonomy and distribution of lanternfishes (family Myctophidae)
of the eastern Pacific Ocean: Navy Ocean Research and Development Activity.
NORDA Report. 3. 229 p.
Yopak KE. 2012. Neuroecology of cartilaginous fishes: the functional implications of

brain scaling. Journal of Fish Biology 80(5):1968-2023.
Young RE, Kampa EM, Maynard SD, Mencher FM, Roper CFE. 1980.
Counterillumination and the upper depth limits of midwater animals. Deep-Sea
Research Part I: Oceanographic Research Papers 27(9):671-691.
Young JW, Blaber SJM. 1986. Feeding ecology of three species of midwater fishes
associated with the continental slope of eastern Tasmania, Australia. Marine
Biology 93(1):147-156.
Young RE, Roper CFE, Walters JF. 1979. Eyes and extra-ocular photoreceptors in
midwater cephalopods and fishes - Their roles in detecting downwelling light for
counter-illumination. Marine Biology 51(4):371-380.
43
Chapter 2 Eye size
CHAPTER 2
Eye-size variability in deep-sea lanternfishes

(Myctophidae): an ecological and phylogenetic study
Fanny de Busserolles1, John L. Fitzpatrick2,3, John R. Paxton4, N. Justin Marshall5,

Shaun P. Collin1.
1. Neuroecology Group, School of Animal Biology and the Oceans Institute, The University of Western
Australia, Crawley, WA, Australia.
2. Centre for Evolutionary Biology, School of Animal Biology, The University of Western Australia,
Crawley, WA, Australia.
3. Computational and Evolutionary Biology, Faculty of Life Sciences, University of Manchester,
Manchester M13 9PT, UK
4. Ichthyology, Australian Museum, Sydney, NSW, Australia.
5. Sensory Neurobiology Group, Queensland Brain Institute, University of Queensland, Brisbane, QLD,
Australia.
Chapter published in PLoS ONE (Appendix 1):

variability in deep-sea lanternfishes (Myctophidae): an ecological and phylogenetic
study. PLoS One 8(3): e58519. doi:10.1371/journal.pone.0058519.
45
Chapter 2 Eye size
2.1. Abstract
One of the most common visual adaptations seen in the mesopelagic zone (200-
1000 m), where the amount of light diminishes exponentially with depth and where
bioluminescent organisms predominate, is the enlargement of the eye and pupil area.
However, it remains unclear how eye size is influenced by depth, other environmental
conditions and phylogeny. In this study, we determine the factors influencing variability
in eye size and assess whether this variability is explained by ecological differences in
habitat and life style within a family of mesopelagic fishes characterized by broad intra-
and interspecific variance in depth range and luminous patterns. We focus our study on
the lanternfish family (Myctophidae) and hypothesise that lanternfishes with a deeper
distribution and/or a reduction of bioluminescent emissions have smaller eyes and that
ecological factors rather than phylogenetic relationships will drive the evolution of the
visual system. Eye diameter and standard length were measured in 237 individuals from
61 species of lanternfishes representing all the recognised tribes within the family, in
addition to compiling an ecological dataset including depth distribution during night and
day and the location and sexual dimorphism of luminous organs. Hypotheses were
tested by investigating the relationship between the relative size of the eye (corrected
for body size) and variations in depth and/or patterns of luminous-organs using
phylogenetic comparative analyses. Results show a great variability in relative eye size
within the Myctophidae at all taxonomic levels (from subfamily to genus), suggesting
that this character may have evolved several times. However, variability in eye size
within the family could not be explained by any of our ecological variables
(bioluminescence and depth patterns), and appears to be driven solely by phylogenetic
relationships.
2.2. Introduction
The detection of light signals is one of the most important means by which
organisms perceive and react to their surroundings. In bony fishes (infraclass Teleostei),
like all other vertebrates, light detection is achieved by a highly conserved structure, the
eye, in which a single lens focuses an image on to the neural retina. Since the cornea is
not refractive underwater (Land, 1990), the lens of most teleosts is spherical and its
radius and the distance from the lens centre to the retina (focal length) are related by a
“constant” defined, over 100 years ago, as Matthiessen’s ratio (Matthiessen, 1882,
1886), regardless of the size of the eye. Thus, the acuity (quality of the image) of the
46
Chapter 2 Eye size
teleost eye is directly related to eye size and lens diameter but is also influenced by
other factors such as the quality of the optics, the amount of light received (governed by
the aperture or pupil size) and the degree of overlap between the dendritic fields of
neighbouring retinal receptors. As visual acuity is limited by the amount of light
available, the evolution of the visual system in teleosts is also expected to be influenced
by the depth at which an individual lives.
The mesopelagic zone (200-1000 m), also referred to as the twilight zone, is
characterised by exponentially diminishing levels of downwelling sunlight. Although
the low amount of sunlight in this zone negates the process of photosynthesis, enough
light is present in the water column to create an extended visual scene both vertically
and horizontally (Warrant, 2004), thereby allowing animals to detect the silhouettes of
potential prey items against a lighter background when viewed from below (Denton,
1990). The mesopelagic zone also contains the greatest biomass and diversity of
animals in the mid-waters of the ocean (Locket, 1977), many of which are
bioluminescent, producing light signals using either symbiotic bioluminescent bacteria
(i.e. anglerfish, Hansen and Herring, 1977) or their own enzymatic complex (luciferin-
luciferase, i.e. viperfish, Mallefet and Shimomura, 1995). Due to the low levels of
sunlight and the predominance of small bioluminescent flashes, vision is considered
very important in the twilight zone and seems to be the dominant sense used by its
inhabitants (Wagner, 2001). To be able to see in dim conditions and for viewing
bioluminescence, the visual system of mesopelagic fishes requires higher sensitivity
than acuity (Warrant and Locket, 2004). To enhance sensitivity, species have adapted to
optimize light collection and extend the visual field (Wagner et al., 1998). One of the
most extreme ocular adaptations at these depths (200-1000 m) is the tubular shape of
the eyes of a number of species (i.e. Argyropelecus sp., Winteria sp.), which are directed
upwards (and rostrally in some species) to optimise light capture of the downwelling
sunlight (Munk, 1966; Locket, 1977; Collin et al., 1997; Collin et al., 1998). However,
the most important morphological adaptation of the visual system of mesopelagic fishes
is arguably the enlargement of their eyes compared to body size (Marshall, 1954); a
larger eye will increase the chance of photon capture (greater pupillary aperture),
thereby allowing improved detection of body silhouettes and bioluminescent flashes
against an increasingly dim background (Denton, 1990; Land, 1990; Warrant, 2000).
47
Chapter 2 Eye size
Deeper in the ocean, in the bathypelagic zone (1000-4000 m), where no

downwelling sunlight penetrates, eye size tends to decrease with some species of
teleosts having such small eyes they were considered “degenerate” (Munk, 1965). The
reduction in eye size in inhabitants of the bathypelagic zone could be explained by the
predominant use of other sensory systems and/or the brighter appearance of the
bioluminescent signals in the absence of residual downwelling sunlight. The
bathypelagic zone represents a visual scene composed only of bright, point sources of
light viewed against a completely dark background. To detect these intermittent sources
of light, the eyes do not need to be very large as the flashes will appear extremely bright
compared to the background (Warrant, 2000). In addition to the absence of residual
daylight, the bathypelagic zone is also characterised by a drastic diminution in animal
biomass and biodiversity (Locket, 1977). An individual living in this zone will
encounter fewer animals and therefore less bioluminescent signals (the only type of
light present in the bathypelagic zone). In this zone, vision may be used secondarily
after a potential prey or mate has been detected using other sensory modalities.
More than half a century ago, Marshall (Marshall, 1954) compared the eye size
of three species of Gonostoma living in different oceanic zones, the upper mesopelagic
(G. denudatum), lower mesopelagic (G. elongatum) and the bathypelagic
(G. bathyphilum) zones and observed that the deeper the species the smaller the size of
the eye. For species that are restricted to a particular zone, this inverse relationship
between depth and eye size appears intuitive. However, some species may frequent all
three zones (epipelagic, mesopelagic, bathypelagic) during their lifetime or even within
the same day (diel vertical migrations). How does eye size vary in these species?
Despite the importance of considering depth, in terms of light availability, when
evaluating the evolution of the visual system in deep-sea fishes, it remains unclear how
eye size is shaped by depth and other environmental conditions.
Consequently, in this study we aim to assess to what extent variability in eye

size can be explained by ecological differences in habitat and life style within a single
family of fishes characterized by broad intra- and interspecific variance in depth range
and luminous patterns. Specifically, we focus our investigation on lanternfishes (family
Myctophidae), one of the most abundant group of mesopelagic fishes in the world’s
oceans (Hulley, 1981), with some 250 species in 33 genera currently recognised (Hulley
and Paxton, in press). Most species exhibit extensive diel vertical migration toward the
48
Chapter 2 Eye size
surface at night in order to feed and a great interspecific variability in their migration
patterns has been observed (Watanabe et al., 1999). Individuals can migrate from the
epipelagic zone at night to the upper part of the bathypelagic zone during the day.
Myctophids produce bioluminescence using a luciferin-luciferase reaction (Tsuji and
Haneda, 1971; Haygood et al., 1994). They possess two kinds of luminous structures,
with highly variable patterns, that light up independently. Those are the ventral and
ventrolateral photophores which are thought to play a role in counter-illumination (Case
et al., 1977) and species recognition (Beebe and Vander Pyl, 1944), and the luminous
organs and tissue patches, which are frequently sexually dimorphic, are located on the
caudal peduncle and/or head and/or body and may play several different roles including
intra- and interspecific communication, prey illumination and distraction (Edwards and
Herring, 1977).
Consequently, we hypothesised that lanternfishes with a deeper distribution

and/or a reduction of bioluminescent emissions will have smaller eyes and that
ecological factors rather than phylogenetic relationships will drive the evolution of their
visual system. We test these predictions by investigating the relationship between the
relative size of the eye (corrected for body size) and variations in depth and/or
luminous-organ patterns in the Myctophidae using phylogenetic comparative analyses.
2.3. Materials and Methods

2.3.1. Ethics statement
For cruises 1-4 (Table 2.1), sampling was carried out under the following
collection permits: Coral Sea waters (CSCZ-SR-20091001-01), Commonwealth waters
(AU-COM2009051), GBRMPA (G09/32237.1) and Queensland Fisheries (133805),
(Marshall, AEC # SNG/080/09/ARC). For cruises 5-6, sampling permits were obtained
by the Chief Scientist of the respective cruises (AIMS, University of Tübingen) for their
target species. We obtained our samples as by-catch and therefore no collection permits
were required.
Most individuals caught were already deceased; however, moribund animals

were humanely euthanized following the guidelines of the NH&MRC Australian Code
of Practice, under our University of Western Australia Animal Ethic protocol
49
Chapter 2 Eye size
(RA/3/100/917). Tissue samples obtained from collaborators (cruises 7-9) did not
require any UWA collection or animal ethics permits.
Table 2.1. Summary of the research cruises from which the samples were collected, together
with their geographic location, fishing equipment, time of sampling and fixative used. 1 Sourced
from Olivar et al. (2012), 2 sourced from Mahé & Poulard (2005). EZ = multiple plankton net
system; IKMT = Isaacs-Kidd Midwater Trawl; RMT = Rectangular Midwater Trawl. *
Indicates the use of an opening-closing device.
Cruise Location Date Equipment Time of Fixation

sampling
1 Coral Sea 11/2009 RMT* / plankton net / Night 4% PFA, Karnovsky
neuston net
2 Off Osprey reef, Coral Sea 05/2010 IKMT Night 4% PFA, Karnovsky
3 Coral Sea 12/2010 RMT* Night 4% PFA, Karnovsky
6 Peru-Chile Trench 09/2010 RMT* / neuston net Day, Night 5% formalin, 2%
glutaraldehyde
7 Bay of Biscay, North East 10/2009 Bottom trawl GV1001 Day Karnovsky
Atlantic
8 Balearic Islands, Western 07/2010 Double-warp modified Day, Night Karnovsky
Mediterranean Sea commercial mid-water
trawl / IKMT2
9 Western Australia 07/2010 EZ net* Night 5% formalin
2.3.2. Data collection and morphometric measurements

A total of 237 lanternfishes from 61 species and 19 genera were analysed in this
study. Samples were obtained from different research cruises in different parts of the
world using different methods of sampling (Table 2.1). For each individual, the standard
length, rostro-caudal eye diameter (measured in situ) and lens diameter were measured
with digital calipers to 0.1 mm. For most of the lanternfish, measurements were
performed on fresh specimens on board ship prior to fixation. However, when samples
were acquired from collaborators (Cruises 7, 8, 9, see Table 2.1), the measurements
were made after fixation. The fixatives used in those cases were 5% buffered formalin
and Karnovsky’s fixative (2.5% paraformaldehyde and 2% glutaraldehyde in 0.1M
cacodylate buffer). Since we were interested in the eye size: body size relationship and
because formaldehyde has previously been found to affect body length by only 0.8%
(Kristoffersen and Salvanes, 1998), no measurement correction was used between fresh
and fixed specimens and all the morphometric data for each species were pooled.
The life stage of each specimen was estimated by length measurements

published in the literature. With the exception of sexually dimorphic features, juvenile
and adult lanternfishes are identical in appearance (i.e. their body proportions, pigment
50
Chapter 2 Eye size
and photophore patterns, Karnella, 1987) and since the regression slopes of eye
diameter versus standard length were not significantly different between the two stages
(Table 2.2), these two groups were analysed together. One specimen of Scopelengys
tristis of the family Neoscopelidae, sister family of the Myctophidae, was also measured
for comparison.
Table 2.2. Analyses of covariance (ANCOVA) of the eye diameter versus standard length
between juveniles and adults in Myctophidae. n = sample size, β = partial regression slope. Only
species with at least three observations for each stage (juvenile, adult) were analysed.
Significant differences are shown in bold. The slopes are not significantly different between
juveniles and adults.
Species n juveniles n adults Predictors β T-value P

Ceratoscopelus warmingii 3 8 SL 0.92 6.93 <0.001
stage -0.03 -1.30 0.23
Lampanyctus alatus 5 4 SL 1.47 4.34 0.005

stage 0.05 1.00 0.35
Lampanyctus parvicauda 3 6 SL 0.91 9.53 <0.001

stage -0.05 -1.21 0.27
2.3.3. Taxonomic remarks

Most of the samples from the Coral Sea (Cruises 1-5) and Chile-Peru Trench
(Cruise 6) are registered as voucher specimens at the Australian Museum in Sydney,
Australia. However, further taxonomic analyses need to be carried out for six of our
study species to confirm identification. A comment for each of these six species is given
below.
Lampanyctus vadulus: requires confirmation of some northeastern Australian

variants of L. nobilis currently under study. Myctophum spinosum / M. lychnobium: the
characters distinguishing these two species appear to form a continuum in the western
South Pacific; the specimens studied here are the two extremes of the continuum that
might be all M. spinosum. Symbolophorus cf. boops: Symbolophorus from the eastern
South Pacific require more study to allow species identification. Nannobrachium cf.
nigrum: the specimens used match the description in Zahuranec (Zahuranec, 2000), but
not the figure and brief description of the holotype in Nafpaktitis et al. (Nafpaktitis et
al., 1977). Triphoturus oculeus: detailed examination of the specimens from the eastern
South Pacific to distinguish any possible T. mexicanus has not been completed to date.
51
Chapter 2 Eye size
2.3.4. Ecological data

A dataset of the location and sexual dimorphism of the luminous tissue was
created using information found in the literature (Table 2.3). Presence-absence of
luminous organs (head, caudal), additional luminous patches and sexual dimorphism in
luminous tissues were noted (Nafpaktitis and Nafpaktitis, 1969; Paxton, 1972;
Nafpaktitis et al., 1977; Nafpaktitis, 1978; Zahuranec, 2000; Herring, 2007). All
lanternfish genera possess a dorsal nasal organ (Dn) and/or a ventral nasal organ (Vn)
on their head associated with the eye. Only species presenting an enlarged Dn and/or Vn
were given a special category in this study. Depending on the species, sexual
dimorphism in luminous tissues can be seen in the Dn/Vn, caudal luminous organs
and/or luminous patches (Table 2.3). Differentiation of the type of sexual dimorphism
in our analyses did not show significant differences. As a consequence, only results for
the presence/absence of sexually dimorphic features are presented in this study (Table
2.3).
For each species, the juvenile/adult depth distributions during night and day
were recorded, taking into consideration the individual size and area of sampling when
possible (Table 2.4.). Published studies using opening-closing sampling devices (Kinzer
and Schulz, 1985; Karnella, 1987; Gartner et al., 1987; Robison, 1972; Williams and
Koslow, 1997; Ross et al., 2010; Flynn and Kloser, 2012) and differentiating life stages
(Ahlstrom and Stevens, 1976; Karnella, 1987; Gartner et al., 1987; Ross et al., 2010)
were given priority when making the dataset.
Categorised depth ranges were created for the statistical analyses based on the
published data and our own capture-depth data. The same number of depth classes was
used for both diurnal and nocturnal depth distribution in order to statistically compare
the results. The group’s cut off, during day and night, was chosen at the most
meaningful depth in terms of vision and accuracy of the depth data. Three group
categories were created in terms of the amount of downwelling light present: moderate
light level (0-5 m at night, 200-500 m during the day), low light level (5-100 m at night,
500-900 m during the day), no light (<100 m at night, <900 m during the day). For the
night-depth range, species (excluding larvae) were classified according to their
shallowest depth recorded. In the surface category at night, only species dip-netted or
sampled with a neuston net (surface) were included. For the day-depth range, species
(excluding larvae) were classified according to their deepest depth recorded.
52
Chapter 2 Eye size
Table 2.3. Dataset used in the phylogenetic comparative analyses of the location of the
luminous tissue (Dn/Vn, Caudal, Body patches), whether there was any level of sexual
dimorphism present (Sex. dim., Sex. D., Sex. C., Sex. P.) and the depth categories during night
and day. Sex. dim. = sexual dimorphism in luminous tissue, Sex. D. = sexual dimorphism in
Dn/Vn luminous organs, Sex. C. = sexual dimorphism in caudal luminous organs, Sex. P. =
sexual dimorphism in luminous tissue patches, 0 = character absent, 1 = character present. For
the night depth range, 0 = 0-5 m, 1 = 5-100 m, 2 = <100 m. For the day depth range, 0 = 200-
500 m, 1 = 500-900 m, 2 = >900 m, n.a. = missing values.
Species Dn/Vn Caudal Patches Sex. dim. Sex. D. Sex. C. Sex. P. Night Day
Benthosema glaciale 0 1 0 1 0 1 0 1 1
B. suborbitale 0 1 0 1 0 1 0 1 1
Bolinichthys longipes 0 1 1 1 0 0 1 1 1
B. nikolayi 0 1 1 0 0 0 0 2 1
B. supralateralis 0 1 1 0 0 0 0 1 1
Centrobranchus andreae 0 1 0 1 0 1 0 2 1
Ceratoscopelus maderensis 0 1 1 0 0 0 0 1 2
C. warmingii 0 1 1 0 0 0 0 1 2
Diaphus brachycephalus 1 0 0 1 1 0 0 1 1
D. danae 1 0 1 0 0 0 0 1 1
D. fulgens 1 0 1 1 1 0 1 1 2
D. garmani 1 0 1 1 1 0 0 1 1
D. holti 1 0 1 1 1 0 0 1 1
D. luetkeni 1 0 1 1 1 0 0 1 1
D. meadi 1 0 1 1 1 0 1 1 2
D. mollis 1 0 1 1 1 0 0 1 1
D. parri 1 0 1 1 1 0 0 1 1
D. phillipsi 1 0 1 0 0 0 0 1 1
D. regani 1 0 1 0 0 0 0 1 2
D. splendidus 1 0 1 1 1 0 0 1 1
D. termophilus 1 0 1 1 1 0 0 1 1
D. whitleyi 1 0 1 1 1 0 0 2 1
Diogenichthys atlanticus 1 1 0 1 1 1 0 1 2
D. laternatus 1 1 0 1 1 1 0 0 0
Electrona risso 0 1 0 1 0 1 0 1 1
Gonichthys tenuiculus 0 1 0 1 0 1 0 0 n.a.
Hygophum benoiti 0 1 0 1 0 1 0 1 2
H. hygomii 0 1 0 1 0 1 0 1 2
H. proximum 0 1 0 1 0 1 0 0 1
Lampadena luminosa 0 1 0 0 0 0 0 1 2
L. urophaos 0 1 0 0 0 0 0 1 1
Lampanyctus alatus 0 1 1 0 0 0 0 1 1
L. crocodilus 0 1 1 0 0 0 0 1 2
L. iselinoides 0 1 1 0 0 0 0 1 1
L. nobilis 0 1 0 0 0 0 0 1 2
L. omostigma 0 1 0 0 0 0 0 1 0
L. parvicauda 0 1 0 0 0 0 0 1 2
L. pusillus 0 1 0 0 0 0 0 1 2
L. vadulus 0 1 0 0 0 0 0 1 2
Lobianchia dolfleini 0 1 0 1 0 1 0 1 1
L. gemellari 0 1 0 1 0 1 0 1 1
Loweina interrupta 0 1 0 1 0 1 0 1 n.a.
Myctophum asperum 0 1 0 1 0 1 0 0 1
M. aurolaternatum 0 1 0 1 0 1 0 0 2
M. brachygnathum 0 1 0 1 0 1 0 0 0
M. lychnobium 0 1 0 1 0 1 0 0 0
M. nitidulum 0 1 0 1 0 1 0 0 1
M. obtusirostre 0 1 0 1 0 1 0 0 1
M. spinosum 0 1 0 1 0 1 0 0 1
Nannobrachium cf. nigrum 0 1 0 0 0 0 0 2 1
N. idostigma 0 1 0 0 0 0 0 1 0
N. phyllisae 0 1 0 0 0 0 0 2 n.a.
Notolychnus valdiviae 0 1 0 1 0 1 0 1 2
Notoscopelus elongatus 1 1 1 1 0 1 0 1 2
N. kroeyerii 1 1 1 1 0 1 0 1 2
Symbolophorus cf. boops 0 1 0 1 0 1 0 0 1
S. evermanni 0 1 0 1 0 1 0 0 1
S. rufinus 0 1 0 1 0 1 0 0 1
S. veranyi 0 1 0 1 0 1 0 0 1
Triphoturus nigrescens 0 1 0 0 0 0 0 2 2
T. oculeus 0 1 0 0 0 0 0 1 1
53
Table 2.4. Summary of the juvenile-adults night and day depth ranges found in the literature for the 61 species of Myctophidae studied. The sampling area and depth
Chapter 2
range of our study samples is also given (N=night, D=day). CorS = Coral Sea, PCT = Peru-Chile Trench, Med = Mediterranean Sea, wMed = western Mediterranean
Sea, Aus = Australia, wAus = western Australia, eAus = eastern Australia, Tas = Tasmania, ChiS = China Sea, Pac = Pacific, cPac = central Pacific, ePac = eastern
Pacific, cNPac = central North Pacific, eNPac = eastern North Pacific, eSPac = eastern South Pacific, eAtl = eastern Atlantic, eNAtl = eastern North Atlantic, wNAtl
= western North Atlantic, cNAtl = central North Atlantic, Phi = Philippines, GMex = Gulf of Mexico, GCal = Gulf of California, SAfr = South Africa.
Species Sampling depth Sampling area Night range Day range Area References
Benthosema glaciale 100-200 N wMed 750-1250 / 12-800 750-1050 / 375-800 wNAtl / Med (Karnella, 1987)*/ (Hulley, 1984)
B. suborbitale 0-200 N wAus 0-100 600-700 wNAtl (Karnella, 1987)*
Bolinichthys longipes 87-107 N CorS 50-150 525-725 cNPac (Clarke, 1973)
B. nikolayi 202-429 N CorS <300 <500 cEPac (Hulley and Duhamel, 2009)
B. supralateralis 500 N CorS 200-700 / 60-700 600-650 / 250-600 wNAtl / GMex (Karnella, 1987)*/ (Gartner et al., 1987)*
Centrobranchus andreae 100 N CorS 100-165 640-650 cNPac (Backus et al., 1968)
Ceratoscopelus maderensis 65 N wMed 33-1000 / 12/800 751-1000 / 100-1000 wNAtl / Med (Karnella, 1987)*/ (Hulley, 1984)
C. warmingii 83-300 N CorS 0-500 800-1550 wNAtl (Karnella, 1987)*
Diaphus brachycephalus 202-329 N CorS 150-225 / 50-300 250-450 / 300-600 wNAtl / GMex (Karnella, 1987) */ (Gartner et al., 1987)*
D. danae 253-377 N CorS 0-300 300-650 Tas (Williams and Koslow, 1997)*
D. fulgens 500 N CorS 85 1000 ChiS (Yang et al., 1996)
D. garmani 100-150 N CorS / wAus surface / 0-125 <225 / 325-750 ePac / SAfr (Wisner, 1976) / (Hulley, 1986a)
D. holti - wMed 80-235 500-675 Med (Hulley, 1984)
54
D. luetkeni 202-325 N CorS 400-800 / 60-300 500-800 / 300-600 wNAtl / GMex (Karnella, 1987)* / (Gartner et al., 1987)*
D. meadi 50-100 N wMed 0 1250 Tas (Flynn and Kloser, 2012)*
D. mollis 202-429 N CorS / wAus 0-700 / 33-350 100-800 / 300-500 wNAtl (Karnella, 1987) */ (Hulley, 1984)
D. parri 200-329 N CorS 0 900 Aus (Flynn, 2012)
D. phillipsi 107 N CorS 50-200 600 SAfr (Hulley, 1986a)
D. regani 50-400 N CorS 0 / <50 1000 Aus (Flynn, 2012)/ (Hartmann and Clarke, 1975)
D. splendidus 107 N CorS 51-250 / 30-550 501-650 / 300-600 wNAtl / GMex (Karnella, 1987) / (Gartner et al., 1987)
D. termophilus 200-477 N CorS 40-225 / 75-150 325-850 eNAtl / GMex (Hulley, 1984) / (Gartner et al., 1987)*
D. whitleyi 329 N CorS 136-152 170-710 Phi (Bourret, 1985)
Diogenichthys atlanticus 50-150 N wAus 20-1000 / 50-700 500-1000 / 350-700 wNAtl / GMex (Karnella, 1987)* / (Gartner et al., 1987)*
D. laternatus 36-146 N PCT Surface / >50 100-500 / 200-400 eNPac / eSPac (Ahlstrom and Stevens, 1976)/ (Cornejo and Koppelmann, 2006)
Electrona risso 330-1000 D wMed 400-500 / 0-200 700-750 / 600-700 Med / cAtl (Nafpaktitis et al., 1977) / (Kinzer and Schulz, 1985)*
Gonichthys tenuiculus Surface N CPT surface - cPac (Ahlstrom and Stevens, 1976)
Hygophum benoiti 410 D wMed 18-1050 450-1100 wNAtl (Karnella, 1987)*
H. hygomii 0-202 N wAus 10-300 / 0-1000 400-800 / 500-1000 eNAtl / wNAtl (Hulley, 1984)/ (Karnella, 1987)*
H. proximum Surface N CorS / PCT 25-150 500-700 cNPac (Clarke, 1973)
Eye size
Lampadena luminosa 325 N CorS 75-250 / 65-600 525-725 / 500-1000 cNPac / GMex (Clarke, 1973) / (Gartner et al., 1987)*
L. urophaos 200 N CorS 50-600 600-750 wNAtl (Karnella, 1987)*
Table 2.4. (Continued)
Chapter 2
Lampanyctus alatus 202-286 N CorS / wAus 50-300 700-850 wNAtl (Karnella, 1987)*
L. crocodilus 600 D wMed 50-950 / 1200 750-1000 / 1200 wNAtl / Med (Karnella, 1987)*/ (Stefanescu and Cartes, 1992)
L. iselinoides 450-707 D PCT <70 - ePac (Wisner, 1976)
L. nobilis 107 N CorS 40-600 / 100-200 800-1000 / 300-<900 GMex / wNAtl (Gartner et al., 1987) / (Karnella, 1987)
L. omostigma 36-146 N PCT <50 200-400 eSPac (Cornejo and Koppelmann, 2006)
L. parvicauda 61-302 N / 610-1200 D PCT 0-150 - GCal (Robison, 1972)*
L. pusillus 100-200 N wMed 50-325 / 50-1000 500-1000 / 550-850 Med / wNAtl (Hulley, 1984) / (Karnella, 1987)*
L. vadulus 100 N CorS 0 1000 Aus (Flynn, 2012)
Lobianchia dolfleini 65 N wMed 25-400 375-600 Med (Hulley, 1984)
L. gemellari 107-329 N CorS 20-210 / 50-600 300-450 / 400-850 GMex / wNAtl (Gartner et al., 1987)* / (Karnella, 1987)*
Loweina interrupta 100-150 N wAus 60-800 - wNAtl (Hulley, 1984)
Myctophum asperum Surface N CorS Surface-150 / 0-125 400 / 425-750 GMex / SAfr (Gartner et al., 1987)* / (Hulley, 1986a)
M. aurolaternatum Surface N CorS 0 1000 eAus (Flynn, 2012)
55
M. brachygnathum Surface N CorS - 280-340 Phi (Bourret, 1985)

M. lychnobium Surface N CorS 0 100 eAus (Flynn, 2012)
M. nitidulum Surface N PCT 0-15 / surface 600-800 cNPac / eNpac (Clarke, 1973) / (Ahlstrom and Stevens, 1976)
M. obtusirostre Surface N CorS Surface-150 / 0-15 500-600 / 500-700 GMex / cNPac (Gartner et al., 1987)* / (Clarke, 1973)
M. spinosum Surface N CorS 0-15 600 cNPac (Clarke, 1973)
Nannobrachium cf. nigrum 200-300 N CorS 100-310 640-900 cNPac (Clarke, 1973)
N. idostigma <61 N PCT >50 250 eSPac (Cornejo and Koppelmann, 2006)
N. phyllisae <108 N PCT <300 - ePac (Wisner, 1976)
Notolychnus valdiviae 211-348 N CorS 50-250 / 30-1050 350-1050 / 400-850 GMex / wNAtl (Ross et al., 2010)* / (Karnella, 1987)*
Notoscopelus elongatus 248 N, 414 D wMed 45-150 375-1000 Med (Hulley, 1984)
N. kroeyerii 366-500 D eNAtl 0-200 325->1000 eNAtl (Hulley, 1984)
Symbolophorus cf. boops Surface N CorS 0 900 eAtl (Hulley, 1992)
S. evermanni Surface N CorS 0-125 600-900 cNPac (Clarke, 1973)
S. rufinus Surface N CorS 0-125 / 0-900 425-850 / 750-900 SAfr / wNAtl (Hulley, 1986a) / (Karnella, 1987)*
S. veranyi Surface N wMed 0-150 100-700 Med (Hulley, 1984)
Triphoturus nigrescens 321 N PCT 200-1000 400-1000 cNPac (Hulley, 1986b)
T. oculeus 5-290 N <450 D CorS >50 >100 eSPac (Cornejo and Koppelmann, 2006)
Eye size
Chapter 2 Eye size
2.3.5. Phylogenetic analyses

Standard statistical analyses assume independence of the samples. This
assumption is unfortunately not met when comparing different species as more closely
related species are expected to be more similar to one another due to the sharing of a
common ancestor. Therefore, all data analyses were performed using phylogenetic
comparative analyses to account for the shared history among species (Harvey and
Pagel, 1991).
Unfortunately, no fully resolved phylogeny is available for the family

Myctophidae to date. Consequently, two different phylogenies, A and B (Figure 2.1),
were built in the Mesquite program v. 2.75 (Madison and Madison, 2011) based on two
published phylogenies (Paxton et al., 1984; Poulsen et al., 2013).
Paxton et al.’s phylogeny classified genera using derived character states of

adult osteology and photophore patterns, as described by Paxton (Paxton et al., 1984),
and of larvae as described by Moser and Ahlstrom (1970, 1972, 1974). The phylogeny
divided the family into two subfamilies (Myctophinae and Lampanyctinae) and seven
tribes (Electronini, Myctophini, Gonichthyini, Diaphini, Gymnoscopelini, Lampanyctini
and Notolychnini, Figure 2.1A). The only difference between Paxton’s originally
described phylogeny and the one used in the present analysis (phylogeny A) is the
inclusion of the genus Nannobrachium, which was added to Paxton’s phylogeny after
Zahuranec (2000).
The phylogeny of Poulsen et al. (2013) was the first molecular phylogeny for the
family and used mitogenomic results from DNA sequences and unique gene orders
from 38 lanternfish species. Poulsen et al. (2013) confirmed the presence of the two
subfamilies (Myctophinae and Lampanyctinae) and identified 10 monophyletic lineages
or clades (Figure 2.1B). The genus Hygophum was added to Poulsen et al.’s originally
described phylogeny in our study (phylogeny B), although no clade was assigned, and
its position in the phylogeny was kept identical to Paxton et al.’s phylogeny. The main
differences seen in Poulsen et al.’s phylogeny compared to Paxton et al’s are the taxon
Notolychnus, which became a sister taxon of all the remaining myctophids, and the tribe
Diaphini, which became a sister tribe of the Lampanyctini.
56
Chapter 2 Eye size
Figure 2.1. Phylogenies of Myctophidae reconstructed from (A) Paxton et al. (1984), (B)
Poulsen et al. (2013). The red branches indicate the main differences between the two
phylogenies. Branch lengths are arbitrarily ultrametricized on the figure. In A, the numbers
identify the different tribes of Paxton et al. (1984), 1. Electronini, 2. Myctophini,
3. Gonichthyini, 4. Notolychnini, 5. Lampanyctini, 6. Gymnoscopelini, 7. Diaphini. In B, the
letters identify the different clades of Poulsen et al. (2013), A. Notolychnini, B. Diaphini,
C. Notoscopelini, D-E-F. Lampanyctini, G. Electronini, H. Myctophini, I. Myctophini (cycloid-
species-group), J. Myctophini (ctenoid-species group) + Gonichthyini.
Due to the lack of resolution, both phylogenies are only resolved to generic
level, resulting in several polytomies (i.e. unresolved relationship among species).
Unfortunately, the presence of polytomies prevents the application of many
phylogenetic analyses that require a fully resolved phylogeny. Therefore, to bypass this
problem, 100 alternative phylogenies were generated with polytomies randomly
resolved to infinitesimally small (10-6) branch lengths using the Mesquite program v.
57
Chapter 2 Eye size
2.75 (Madison and Madison, 2011). Ten of these phylogenies with randomly resolved
polytomies were selected at random to perform the different analyses and the results
between each of the 10 phylogenies compared for consistency. Moreover, to fit the
statistical requirements for the phylogenetic linear models described below, branch
lengths were transformed using Grafen’s method (Grafen, 1989) with rho
transformation set at 2.5 before all analyses.
All statistical analyses were performed, using both phylogenies separately, on

Log10-transformed species averages with the statistical program R v.2.15.0 (R
Foundation for Statistical Computing 2012).
2.3.6. Estimating phylogenetic signal

The phylogenetic signal for continuous and discrete traits was estimated with
Pagel’s lambda (λ) using the package GEIGER in R (Harmon et al., 2008). Pagel’s λ is
a measure of the degree of phylogenetic dependence in the data (Pagel, 1999), meaning
to which degree closely related species are more similar to each other than what is
expected by random evolutionary processes.
Pagel’s λ varies from 0 to 1, with a λ value of 1 indicating that traits gradually

accumulate changes over time in a Brownian motion process (i.e. random change in any
direction) and λ values of 0 indicating that no phylogenetic signal is present and that
traits have evolved in response to selective processes. The observed λ value for each
trait was compared to λ values of 0 and 1 using likelihood ratio tests with df = 1.
2.3.7. Phylogenetic linear models

Relationships between morphological traits (eye diameter, lens diameter, body
size) and the relationship between morphological and ecological traits were assessed
using phylogenetic generalised least squares regressions (PGLS, Freckleton et al., 2002)
with the package APE in R (Paradis et al., 2004). PGLS are classic generalised least
squares regressions that additionally take into account the shared history of the different
species by incorporating phylogenetic information into the analyses. PGLS regressions
estimate a phylogenetic scaling parameter, λ, using maximum likelihood methods to
determine the degree of covariance in the residuals of the model, while controlling for
phylogenetic effects. This approach also examines if the scaling parameter λ
significantly differs from 0 or 1 using likelihood ratio tests, where λ = 0 indicates no
58
Chapter 2 Eye size
phylogenetic dependence in the data and λ = 1 indicates strong phylogenetic association

in the data (Pagel, 1999; Freckleton et al., 2002). PGLS models were used to assess the
relationship between morphological traits, to identify if eye size differs between the two
subfamilies (Myctophinae, Lampanyctinae) when correcting for the effects of body size,
and to assess if eye diameter was related with various ecological parameters. Standard
length was added as a covariate in all models.
2.3.8. Phylogenetic ANOVAs

To identify differences at the tribal (phylogeny A), cladal (phylogeny B) and
generic (phylogenies A and B) levels, phylogenetically corrected residuals of the eye
diameter were calculated from eye diameter - standard length regression fit lines using
PGLS. Statistical analyses on residuals are usually not recommended as they often lead
to biased results, especially if the variables tested are colinear with the controlled
variable (Freckleton, 2009). However, when too many groups are present, phylogenetic
ANCOVA (i.e. analysis of covariance incorporating phylogenetic information) cannot
sort out the differences using the PGLS approach. Consequently, residuals were used in
this particular case to estimate differences between groups (tribes, clades, genera) using
phylogenetic ANOVA (i.e. a classic ANOVA incorporating phylogenetic information,
Garland et al., 1993) followed by a sequential Bonferroni post-hoc test using the
GEIGER (Harmon et al., 2008) and PHYTOOLS (Revell, 2012) packages in R. At the
generic level comparison, both phylogenies A and B were used separately and the
results compared. Only groups with at least three observations were included in those
analyses. The genus Hygophum was excluded from the analyses at the cladal level due
to its hypothesized position in Poulsen et al.’s phylogeny.
2.4. Results
2.4.1. Morphometric measurements
Eye size varied greatly within the lanternfish family (Figure 2.2). The range of
values for standard length, eye diameter and lens diameter for each species is given in
Table 2.5 in addition to the life stages and the origin of the specimens. In our dataset,
standard length varies from 17.6 mm (Diogenichthys laternatus) to 126.4 mm (Diaphus
danae); eye and lens diameter range from 1.5 mm to 11.1 mm and from 0.6 mm to 5.0
mm in Nannobrachium idostigma and Myctophum lychnobium, respectively.
59
Chapter 2 Eye size
Figure 2.2. Difference in eye size compared to body size in two species of lanternfish. (A)
Myctophum brachygnathum, (B) Nannobrachium phyllisae. Scale bar = 10 mm.

Estimation of the phylogenetic signal using Pagel’s lambda gives relatively
similar results with both phylogenies (Table 2.6). Results show that Dn/Vn and caudal
luminous organs have a strong phylogenetic signal, suggesting that they gradually
accumulate changes over time in a random evolutionary process. On the contrary, no
phylogenetic signal is observed for the standard length and the day-depth distribution
(phylogeny A only) variables. The other variables (eye diameter, lens diameter,
residuals eye diameter, luminous patches, sexual dimorphism in luminous tissue and
night depth distribution) show intermediate values of Pagel’s lambda, which, although
significantly different from 0 or 1, are generally closer to 1 depending on the phylogeny
used. However, independent of the phylogeny used, the eye size corrected for body size
(residuals eye diameter), the luminous patches and the luminous tissue sexual
dimorphism variables show a strong phylogenetic signal very close to 1, again
indicating that these traits changed randomly over time during lanternfish evolution
(Table 2.6).
60
Chapter 2 Eye size
Table 2.5. Range of standard length, eye diameter and lens diameter for each of the 61 species
of Myctophidae analysed in this study. For each species, the sample size (n), the life stage (A =
adult, J = juvenile) and the sample origin (Cruise) is given. For cruise number refer to Table 2.1.
The superscript number for each Diaphus species indicates the group number made from the
absence (1) or presence (2) of the So.
Species n SL Eye ø Lens ø stage Cruise

Benthosema glaciale 1 42 4.6 1.8 A 8
B. suborbitale 2 28.2 – 29.0 3.0 – 3.1 1.3 A 9
Bolinichthys longipes 5 36.8 – 45.4 3.3 – 4.4 1.5 – 1.8 A 3
B. nikolayi 4 23.3 – 37.7 2.5 – 4.1 1.0 – 1.8 A+J 3
B. supralateralis 1 41.2 4.2 2.0 J 2
Centrobranchus andreae 1 31.5 2.2 0.8 ? 1
Ceratoscopelus maderensis 1 52 5.1 2.1 A 8
C. warmingii 11 36.3 – 71.9 3.5 – 6.6 1.4 – 2.8 A+J 1, 3
Diaphus brachycephalus2 2 33.9 – 35.2 4.1 – 4.3 1.6 – 1.8 A 3
D. danae1 16 81.2 – 126.4 7.2 – 9.0 2.9 – 3.8 A 3
D. fulgens2 1 41.2 3.9 1.8 ? 2
D. garmani1 3 32.8 – 42.1 2.4 – 3.5 1.0 – 1.5 A+J 4, 9
D. holti2 1 40 5.1 2.1 A 8
D. luetkeni1 5 35.2 – 40.8 2.3 – 2.9 2.4 – 3.1 A+J 1, 3
D. meadi2 1 28.3 3.3 1.2 J 8
D. mollis2 3 36.6 – 66.2 3.9 – 6.5 1.6 – 3.2 A 3, 9
D. parri2 5 24.9 – 51.0 2.9 – 5.7 1.7 – 2.4 A+J 1, 2, 3, 4
D. phillipsi1 4 26.2 – 29.5 2.1 – 2.4 1.0 J 3
D. regani1 2 40.8 – 42.3 2.5 – 2.8 1.2 – 1.2 J 4
D. splendidus1 4 26.7 – 48.1 1.5 – 2.8 0.6- 1.3 A+J 3, 4
D. termophilus1 6 48.3 – 77.3 4.2 – 6.5 1.9 – 3.2 A+J 1, 3, 4
D. whitleyi1 2 59.5 – 90.3 4.0 – 5.5 1.7 – 2.6 A+? 3
Diogenichthys atlanticus 7 18.0 – 21.4 1.9 – 2.7 0.8 – 1.0 A 9
D. laternatus 11 17.6 – 31.1 2.0 – 3.2 0.7 – 1.3 A 6
Electrona risso 1 46 7.0 3.0 A 8
Gonichthys tenuiculus 3 40.6 – 49.4 2.7 – 3.5 1.3 – 1.6 A 6
Hygophum benoiti 1 45 6.2 2.3 A 8
H. hygomii 2 22.2 – 57.3 3.0 – 7.5 1.1 – 3.1 A+J 9
H. proximum 4 25.8 – 38.2 3.1 – 4.8 1.3 – 2.1 A+J 3, 4, 5, 6
Lampadena luminosa 1 104.3 8.4 3.7 A 3
L. urophaos 1 40 2.8 1.3 A 3
Lampanyctus alatus 9 30.3 – 49.8 1.7 – 3.3 ? – 1.3 A+J 3, 4, 9
L. crocodilus 1 31 1.7 0.6 J 8
L. iselinoides 1 34.3 2.0 0.8 J 6
L. nobilis 2 36.4 – 50.4 1.7 – 2.7 1.0 J 3
L. omostigma 1 27.8 1.9 0.7 ? 6
L. parvicauda 9 28.4 – 106.0 1.7 – 6.4 0.9 – 2.8 A +J 6
L. pusillus 1 37 2.1 0.8 A 8
L. vadulus 4 37.4 – 81.8 2.2 – 5.2 0.9 – 2.12 A+J 1, 3, 4
Lobianchia dolfleini 2 27 – 33.3 1.9 – 2.2 1.0 A 8
L. gemellari 4 32.3 – 52.9 2.2 – 3.4 0.8 – 1.3 A+J 3
Loweina interrupta 1 25.9 2.0 0.9 J 9
Myctophum asperum 1 76.4 8.5 3.8 A 1
M. aurolaternatum 1 57.78 5.3 2.3 J 4
M. brachygnathum 7 65.7 – 69.7 6.8 – 7.6 1.7 – 3.6 A+J 2, 5
M. lychnobium 4 51.0 – 106.9 4.8 – 11.1 2.3 – 5.0 A+J 2, 4
M. nitidulum 11 25.5 – 85.4 2.3 – 7.1 0.9 – 3.2 A+J 6
M. obtusirostre 3 90.9 – 97.6 10.1 – 10.5 4.2 – 4.7 A 2, 4, 5
M. spinosum 5 32.7 – 87.5 3.2 – 8.7 1.4 – 3.9 A+J 1, 2, 4
Nannobrachium cf. nigrum 7 36.3 – 90.0 1.8 - 4 0.8 – 2.1 A 1, 3
N. idostigma 8 31.1 – 72.2 1.5 – 3.9 0.6 – 2.0 A+? 6
N. phyllisae 3 48.9 – 72.5 2.3 – 3.4 0.8 – 1.4 A+? 6
Notolychnus valdiviae 1 21.4 1.4 0.6 A 3
Notoscopelus elongatus 1 42 3.2 1.2 A 8
N. kroeyerii 4 95.2 – 105.1 6.1 – 6.7 2.7 – 2.9 A 7
Symbolophorus cf. boops 1 74.0 6.0 2.3 J 6
S. evermanni 13 33.6 – 65.8 2.5 – 6.2 1.0 – 2.8 J 2, 4
S. rufinus 8 30.5 – 73.7 2.1 – 6.2 1.2 – 2.9 J 2, 4
S. veranyi 1 85 6.7 2.9 A 8
Triphoturus nigrescens 1 35.4 2.0 0.9 A 3
T. oculeus 8 30 – 58.3 1.7 – 4.0 0.6 – 1.6 ? 6
61
Chapter 2 Eye size
Table 2.6. Estimates of the phylogenetic signal for each variable using Pagel’s Lambda. The
results are presented for one of the ten randomly selected polytomy resolved phylogenies for the
two different phylogenies. A λ value of 1 indicates that the trait gradually accumulates changes
over time in a Brownian motion process. A λ values of 0 indicates that no phylogenetic signal is
present and that traits have evolved in response to selective processes. The superscript values
are likelihood ratio tests different from 0 and 1. Sample size is 61 for all variables except day
depth (58).
Variables λ (Phylogeny A) λ (Phylogeny B)

Eye diameter 0.94<0.001, <0.001 0.750.02, <0.001
Lens diameter 0.95<0.001, <0.001 0.790.02, <0.001
Standard length <0.0011.0, <0.001 <0.0011.0, <0.001
Residuals (eye/SL) 0.96<0.001, <0.001 0.94<0.001, <0.001
Dn/Vn organs 1<0.001, 1.0 1<0.001,1.0
Caudal luminous organs 1<0.001, 1.0 1<0.001,1.0
Luminous patches 0.99<0.001, <0.001 0.99<0.001, <0.001
Luminous tissue sexual dimorphism 0.93<0.001, <0.001 0.97<0.001, <0.001
Day depth <0.0011.0, 0.04 0.790.113, <0.001
Night depth 0.75<0.001, <0.001 0.67<0.001, <0.001
2.4.3. Relationship among morphometric traits

Phylogenetic linear regression shows that lens diameter is strongly (positively)
correlated with eye diameter (PGLS, n = 61, R2 = 0.98, t-value = 53.48, P ≤ 0.001;
Figure 2.3A). Due to this close relationship between eye and lens diameter, we focused
all subsequent analyses solely on the eye diameter – standard length relationship.
Phylogenetic linear regression reveals that the eye diameter is positively correlated with
standard length (PGLS, n = 61, R2 = 0.74, t-value = 13.05, P ≤ 0.001, Figure 2.3B).
However, variations in eye size are observed between the different myctophid species,
both at the level of subfamilies and tribes with representatives of the same subfamily or
tribe having smaller or larger eyes (Figure 2.3B). Probably due to this great variability,
no significant difference was found between the two subfamilies (Myctophinae and
Lampanyctinae) in terms of eye size (PGLS, n = 61, standard length effect: tA = 12.95,
pA ≤ 0.001, tB = 13.85, pB ≤ 0.001; subfamily effect: tA = 0.41, pA = 0.68, tB = 0.62, pB =
0.54). The representative of the lanternfish sister family Neoscopelidae, Scopelengis
tristis, showed a relatively small eye compared to all the Myctophidae analysed (Figure
2.3B).
2.4.4. Morphometric comparisons among tribes and clades

Phylogenetic ANOVAs reveal differences at both tribal (phylogeny A, n = 4, F
= 12.4, P = 0.001, Figure 2.4A) and cladal (phylogeny B, n = 6, F = 14.33, P = 0.001;
62
Chapter 2 Eye size
Figure 2.4B) levels. At the tribal level, post-hoc analyses revealed that the Myctophini
possess significantly larger eyes than the other tribes analysed statistically and probably
larger eyes than the tribes Notolychnini and Gymnoscopelini, which were excluded
from the analysis due to insufficient sampling size (Figure 2.4A). Although also
excluded from the statistical analysis for the same reasons, the tribe Electronini seems
to possess the largest eyes. The tribes Diaphini and Lampanyctini are significantly
different from each other, with the latter showing smaller eyes, but not significantly
different from the Gonichthyini. At the cladal level (Figure 2.4B), most of the clades
statistically analysed present similar eye sizes except clade E (Lampanyctini), which has
significantly smaller eyes than all the other clades analysed. The three clades from the
tribe Lampanyctini (D, E, F) do not overlap and show very different eye sizes, with
clade D showing the largest eyes and clade E the smallest eyes. To the contrary, within
the tribe Myctophini (H, I, J), all the clades possess similar eye sizes. As for the tribal
level, Clade G (Electronini) did not possess sufficient observations to be included in the
statistical analysis, but appears to be the clade showing the largest eye size.
Figure 2.3. (A) Relationship between lens diameter and eye diameter after correcting for
phylogeny (PGLS). (B) Relationship between eye diameter and standard length in 61 species of
Myctophidae. Each point represents the mean for the species; individual details are in Table 2.5.
Shapes represent the subfamilies, circles = Myctophinae, triangles = Lampanyctinae. Colors
represent the tribes of Paxton et al. (1984), brown = Electronini, red = Myctophini, blue =
Lampanyctini, green = Diaphini, yellow = Gonichthyini, pink = Gymnoscopelini, purple =
Notolychnini. The black square represents one individual, Scopelengys tristis, from the sister
family Neoscopelidae. The fitted lines are the linear regressions corrected for phylogeny
(PGLS) using the phylogeny of Paxton et al. (1984).
63
Chapter 2 Eye size
Figure 2.4. Residuals eye size corrected for body size of Myctophidae after correcting for
phylogeny: (A) by tribes (phylogeny A, Paxton et al., 1984) and (B) by clades (phylogeny B,
Poulsen et al., 2013). Colours represent the tribes of Paxton et al. (1984) for comparison, as in
Figure 2.3. Groups sharing the same superscript letter are not significantly different from one
another based on the post-hoc analyses. Groups with no superscript letters were not included in
the analysis due to the low samples size (n < 3). #, $ indicates genera that were significantly
different for one of the ten randomly resolved polytomy phylogenies (P# = 0.04, P$ = 0.03).
64
Chapter 2 Eye size
Figure 2.5. Residuals eye size corrected for body size by genera of Myctophidae, corrected for
phylogeny (phylogeny A, Paxton et al., 1984). Colours represent the tribes of Paxton et al.
(1984), as in Figure 2.3. The genus Diaphus was divided into two groups based on the
presence/absence of the So. Groups with the same superscript letters are not significantly
different from one another, independent of the phylogeny used based on the post-hoc analyses.
Groups with no superscript letters were not included in the analyses due to the low samples size
(n < 3). # indicates genera that were not significantly different when analysed with phylogeny B
(P = 1) and for four of the ten randomly resolved polytomy phylogenies of phylogeny A (P =
0.06 – 0.08).
Eye-size variation is also observed between and within genera independently of

the phylogeny used (n = 8, phylogeny A, F = 39.15, P = 0.001, Figure 2.5; phylogeny
B, F = 41.54, P = 0.001). The results are similar between the two phylogenies except for
two genera. When analysed with phylogeny A, Diaphus 2 is found to possess larger
eyes than Bolinichthys (Post-hoc test, P = 0.028). However, no significant difference
was found between the two groups when analysed with phylogeny B (Post-hoc test, P =
1). The genera Lampanyctus and Nannobrachium are not significantly different from
each other, but possess significantly smaller eyes than the rest of the genera analysed.
Despite belonging to different tribes/clades, several genera share similar eye sizes (i.e.
65
Chapter 2 Eye size
Myctophum, Bolinichthys, Diaphus 2). Conversely, some genera belonging to the same
tribe or clade show significantly different eye sizes (i.e. Hygophum and Myctophum
from the tribe Myctophini and Symbolophorus and Gonichthys from clade I). Genera
from the tribe Gonichthyini (Loweina, Gonichthys, Centrobranchus) appear to have
smaller eyes than all the genera present in the tribes Electronini and Myctophini and
similar eye sizes to the other tribes. A great variation in eye size is also observed within
the same genus, i.e. Diaphus. This genus possesses great variability in Dn/Vn luminous
organs (Figure 2.6) and can be further divided into two groups based on the presence or
absence of one light organ, the So just below the eye. The two groups were found to be
significantly different in term of eye size, with Diaphus species possessing an So
(Diaphus 2) having larger eyes than Diaphus species without an So (Diaphus 1).
2.4.5. Relationship between morphometric and ecological traits

Phylogenetically controlled multiple linear regression models do not reveal any
significant relationships or trends between eye diameter, corrected for standard length,
and any of the ecological variables examined in this study (Table 2.7). This lack of
significant relationships between eye diameter and ecological traits persists regardless
of the phylogeny used in the analysis (Table 2.7).
Table 2.7. Regression model of eye diameter with different predictor variables when controlling
for phylogeny (PGLS). The results are presented for one of the ten randomly selected polytomy
resolved phylogenies for the two different phylogenies. Standard length was added as a
covariate in the models. n=sample size, λ= phylogenetic scaling parameter, the superscript *
after the parameter λ indicates whether the parameter was significantly different from 0 (first
position) and from 1 (second position) in the likelihood tests, β=partial regression slope. The
significance levels are shown in bold.
Phylogeny n λ Predictor variables β t-value P

Phylogeny A 58 0.939*,* Standard length 0.925 11.927 <0.001
Dn/Vn luminous organs -0.012 -0.211 0.834
Caudal luminous organs -0.120 -1.485 0.144
Luminous patches 0.017 0.343 0.733
Luminous tissue sexual dimorphism 0.052 1.259 0.214
Day depth 0.008 0.408 0.685
Night depth -0.013 -0.434 0.666
Phylogeny B 58 0.935*,* Standard length 0.922 12.123 <0.001

Dn/Vn luminous organs -0.043 -0.717 0.477
Luminous patches -0.025 -0.636 0.528
Day depth 0.008 0.419 0.677
Night depth -0.039 -1.451 0.153
66
Chapter 2 Eye size
Figure 2.6. Variation in the size and location of the Dn, Vn and So luminous organs within the
genus Diaphus. A = D. luetkeni, B = D. brachycephalus, C = D. danae, D = D. mollis, E = D.
phillipsi, F = D. parri, G = D. termophilus, H = D. holti. A, C, E, G = Diaphus group 1 (So
absent); B, D, F, H = Diaphus group 2 (So present). The yellow arrows indicate the position of
the So photophore.
67
Chapter 2 Eye size
2.5. Discussion
The aim of this study was to examine the variability in eye size within a range of
species of lanternfishes from different depths to assess the influences of both ecology
and phylogeny. To our knowledge, this is the first study to examine eye-size variability
within the same family of vertebrates using representatives of more than 50% of the
recognised genera.
2.5.1. Eye-size variation and ecology

With respect to assessing whether the variation in eye size observed within the
lanternfishes could be explained by ecological differences using phylogenetic
comparative analyses, no significant relationship was found between relative eye size
and any of our predictor variables. Our hypothesis that lanternfish species with a deeper
distribution and/or less luminous tissues will have smaller eyes could not be validated
with our dataset. This hypothesis was posed under the assumption than the gradual
change in the visual scene in the mesopelagic zone will lead to a great diversity in eye
size (Denton, 1990; Warrant, 2000; Warrant, 2004). At depths where both
bioluminescence and downwelling sunlight are present, as in the mesopelagic zone,
adaptations of the eye toward the detection of one or the other will depend on the
ecological task required. Adaptations to assess the intensity of downwelling light are
essential to aid a species to hold a depth station during the day, to camouflage
themselves by counter illumination, to vertically migrate, to set their circadian rhythm
and/or to detect the presence of other animals above them. Adaptations for viewing
bioluminescent signals will be an advantage in detecting other individuals (prey,
predator, mate) in deeper zones, where bioluminescent cues predominate.
In addition to the ventral photophores, lanternfishes possess a number of

luminous tissues that are thought to have several functions. Caudal luminous organs
play a role in either escape responses by producing a blinding flash (Haddock et al.,
2010) or in sexual communication, where these organs are sexually dimorphic. Several
hypotheses have been proposed for the function of the Dn/Vn organs in myctophids.
They may be used as a head torch to search for prey (Herring, 1985; Haddock et al.,
2010), to compare the intensity of their own photophore emissions with the
downwelling sunlight (Lawry, 1974) and/or for intraspecific communication in sexually
dimorphic species. The function(s) of the other luminous patches remains unclear.
However, the occurrence of sexual dimorphism in some of those patches suggests a role
68
Chapter 2 Eye size
in sexual communication. Herring (2007) concluded that the variety and complexity of
sexual dimorphism in luminous organs were most likely due to their role in sexual
signalling. The fact that some lanternfishes possess additional luminous patches and/or
dimorphic luminous organs indicates than some species may rely more on
bioluminescent signals than others. However, this hypothesis was not supported by our
dataset in terms of eye size.
The visual capabilities of an eye are influenced by its size (Walls, 1942). A
larger eye will provide an advantage in the mesopelagic zone as it will increase the
chance of photon capture, since a large eye normally has a greater pupillary aperture.
However, the larger the eye the more energetically costly it will be (Laughlin et al.,
1998). Smaller eyes are less energetic and can act as a “distance filter” by reducing the
visibility of a bioluminescent signal against a completely dark background (viewed
within the bathypelagic zone and deeper, Warrant et al., 2003). However, a small eye
will be a disadvantage for species adapted for survival higher up in the water column,
where high levels of background illumination are present and increased sensitivity is
required. The results from this study indicate that lanternfishes show a great variability
in eye size, independent of their depth distribution. While some species show the
expected pattern by being strictly mesopelagic with relatively large eyes (i.e.
Myctophum nitidulum), or venturing down into the upper parts of the bathypelagic zone
during the day and possessing relatively small eyes (i.e. Lampanyctus crocodilus),
others present an inverse pattern. For example, Hygophum benoiti possesses large eyes
and frequents upper bathypelagic depths, while some Lampanyctus species with small
eyes are restricted to the mesopelagic zone, with L. omostigma recorded at no deeper
than 400 m (Table 2.4). These results suggest that some species may be clearly
disadvantaged in term of vision by their small eye size in the mesopelagic zone (i.e. L.
omostigma). After sampling very different eye types/sizes in the abyss, Murray & Hjort
(1912) wrote “Nothing has appeared more hopeless in biological oceanography than the
attempt to explain the connection between the development of the eyes and the intensity
of light at different depths in the ocean”. The conclusion of the present study seems to
agree with this statement. However, the same analyses using light intensities rather than
depth distribution might be more relevant here and will need to be considered in any
future analyses, in addition to a more detailed analysis of the lanternfish visual system.
69
Chapter 2 Eye size
Even though the relative enlargement of the eye is an important adaptation for
vision in dim-light conditions and for viewing bioluminescence, it is not the only visual
adaptation found in the mesopelagic zone. Deep-sea fishes possess several other visual
specialisations to enhance the sensitivity of the eye, especially at the level of the retina,
that will need to be considered. Moreover, two types of light can be seen in the
mesopelagic zone, bioluminescence and downwelling light, which do not require the
same adaptations in terms of sensitivity for their detection. In fact, while sensitivity to
bioluminescence (point-like light sources) is directly influenced by the size of the eye
(Warrant, 2000; Warrant and Locket, 2004), this is not the case for downwelling light
(extended light source), where sensitivity is independent of eye size and is directly
proportional to the size of the visual pixel (photoreceptor diameter, Land, 1981). As a
result, small-eyed species could potentially have a greater sensitivity to downwelling
light than larger-eyed species depending of the diameter of their photoreceptors. This
stresses the point that the visual capabilities of a species cannot be solely assessed by
the size of the eye and that a number of other physical factors in addition to the type of
visual stimulus might need to be taken into consideration in order to assess relationships
with environmental variables.
The question that then remains is: Do small-eyed lanternfishes really have a
limited visual system or do they compensate for their small eye size with other visual
specialisations? Very few studies have examined the visual system of myctophids.
However, the group appears to have evolved eyes designed to enhance sensitivity, i.e.
with an aphakic gap (Tarletonbeania crenularis, Lawry, 1974), a pure-rod retina
(Lampanyctus crocodilus, Vilter, 1951; Lampanyctodes sp, Pankhurst, 1987;
Stenobrachius leucopsarus, O'Day and Fernandez, 1976), a tapetum lucidum
(Stenobrachius leucopsarus, O'Day and Fernandez, 1976), a high photoreceptor density
(Lampanyctus crocodiles, Vilter, 1951; Lampanyctodes sp, Pankhurst, 1987;
Stenobrachius leucopsarus, O'Day and Fernandez, 1976), visual pigments tuned to view
bioluminescence (58 species, Turner et al., 2009), and a rather unspecialised retina with
poor acuity (Lampanyctus macdonaldi, Myctophum punctatum, Collin and Partridge,
1996; Wagner et al., 1998). However, these data have been compiled from very few
species and most of the studies have only examined one or a few of these
characteristics, which negated their inclusion in our analysis. It is also possible that
small-eyed species have adapted to the mesopelagic zone in other ways by relying less
on vision and more on other sensory systems. This hypothesis seems plausible
70
Chapter 2 Eye size
considering the extremely high number of myctophid species present in the mesopelagic
zone (~250) and the quantitative differences in the size of the optic tecta (Wagner,
2001).
2.5.2. Eye-size variation and phylogeny

Results from this study showed great differences in eye sizes within the
Myctophidae at all phylogenetic levels. At the subfamilial level both small and large
eyes are present in the two different subfamilies, indicating that eye-size variations may
have evolved several times within the family. Paxton (1972) discussed the evolution of
the Myctophidae and its two subfamilies and considered the Neoscopelidae the more
generalised of the two families and the Myctophinae closer to the ancestral state than
the Lampanyctinae, while admitting the larval characters indicated the Myctophinae
were more specialised (Moser and Ahlstrom, 1970). Poulsen et al. (2013) presented the
first molecular phylogeny for the family, but failed to resolve this question, with the
exception of Notolychnus, which was used as the ancestral species of the two
subfamilies. Notolychnus valvidiae is a small eyed species, suggesting that small eyes
represent the ancestral condition and that larger eyes might have evolved several times
within the lanternfish family.
The estimation of the phylogenetic signal by Pagel’s lambda revealed that

relative eye size within the family has a strong phylogenetic signal independent of the
phylogeny used. By definition, members of a taxon are more closely related to one
another than to any members of another taxon. Based on this definition, and if relative
eye size is a relatively conserved variable over time, as Pagel’s lambda seems to
indicate, then eye size within a taxon (tribe or clade) would be more similar than
between taxa and as a result, eye size of a species might be estimated simply based on
the phylogenetic position of this species.
In this sense, the sub-division of Paxton’s tribe Lampanyctini into three different
clades by Poulsen et al. (2013) appears to be supported by our relative eye-size data,
since clades D, E and F present very different eye-size ranges that do not overlap.
However, great variation in eye size is also observed within the single genus Diaphus in
our study. Kawaguchi and Shimizu (1978) presented a taxonomic key for Diaphus,
which divided the genus into four groups (SuO-group, So group, Ant-group and Dn-Vn
group) based on the presence or absence of different luminous organs associated with
71
Chapter 2 Eye size
the eye. The presence of the So below the eye is one of the first identifying characters
used by several taxonomic keys to identify Diaphus species (Nafpaktitis, 1978; Paxton
and Williams, unpublished data). Further division of this genus based on the absence or
presence of the So in our study (Diaphus 1, Diaphus 2) shows that species possessing
an So photophore have significantly larger eyes than species without this photophore,
indicating a possible subdivision of this taxon based on the So photophore. Future
molecular, morphological phylogenetic analyses will hopefully shed more light on this
large complex group, which comprises some 75 species.
The subdivision of the tribe Myctophini and inclusion of the tribe Gonichthyini
within the Myctophini by Poulsen et al. (2013) seems less obvious from our results.
Relative eye sizes between Myctophini and Gonichthyini using Paxton et al.’s
phylogeny are significantly different and do not overlap, indicating that members of the
Gonichthyini have systematically smaller eyes than members of the Myctophini. We
realise that eye size is only one character in the evolution of the group and our data set
does not include all genera of either tribe.
2.5.3. Limits of the study

2.5.3.1. Sampling methods
In this study, an attempt was made to categorize species according to their
diurnal and nocturnal depth ranges. This task was complicated because of the lack of
accurate depth-distribution data in the literature. Unfortunately, very few studies
accurately estimate the depth at which a specimen is sampled, with the minority of
sampling using opening-closing devices. Moreover, even fewer studies report depth
distribution by life stages. Karnella (1987) is to date the most comprehensive lanternfish
depth-distribution study, where fishes were sampled seasonally from the Ocean Acre in
the Northern Sargasso Sea at day and night using predominantly discrete-depth
sampling gear every 50 m. Results from this study present accurate depth-distribution
data by life stages for 20 of our 61 species.
In addition to the lack of accurate depth-distribution data in the literature, several

other issues make the task of depth categorisation of species challenging. In addition to
the high level of interspecific variability in depth pattern, there is observed intraspecific
variability in depth distribution depending on the season (Karnella, 1987), moon cycle
(i.e. Hygophum hygomii, Linkowski, 1996), changes with size/age where larger/older
72
Chapter 2 Eye size
specimens become non-migrant (i.e. Benthosema glaciale, Karnella, 1987) and ocean
physics (i.e. Lampanyctus crocodiles, Stefanescu and Cartes, 1992). Studies will often
record the depth of a species at a specific location, season and time, which might not
reveal the general pattern for the species. Finally, lanternfishes are able to efficiently
avoid the net during the day (Kaartvedt et al., 2012), increasing the chances of biased
data for those species. As a consequence, our depth categorisation may not represent the
true distribution of several species, especially in cases where information in the
literature was from a different ocean than the species analysed in our study.
2.5.3.2. Phylogeny
Phylogenetic comparative analyses are dependent on the phylogeny used. To
allow the most accurate analyses, a fully resolved phylogeny with branch lengths is
usually required. Unfortunately, the phylogeny of the lanternfish family remains poorly
resolved, even though they represent one of the most important groups of mesopelagic
fishes in the ocean in term of biomass and energy transfer (Cherel et al., 2010).
Nevertheless, the first molecular phylogeny for the family presented by Poulsen et al.
(2013) appears to support the basic morphological phylogeny of Paxton et al. (1984),
even though some differences in tribal arrangement are recognised. A better resolved
phylogeny will undoubtedly improve our understanding of the relationship between the
different tribes/clades and the variation in eye size, especially as more ecological data is
accumulated (i.e. diet: luminescent prey vs non-luminescent) for this group.
2.5.4. Conclusion
A great variability in relative eye size within the Myctophidae was observed at
all taxonomic levels. However, variability in eye size within the family could not be
explained by ecological variables (bioluminescence and depth patterns) in this study and
seems instead to be driven by phylogenetic parameters. Further analyses including other
environmental variables (i.e. diet, prey, light intensities) and a more complete
phylogeny are needed to understand the great variability in eye size within myctophids.
Moreover, examination of the visual system in more depth will be an essential step in
assessing the visual capabilities of each species and to shed light on the visual
adaptations of the lanternfish family in relation to their environment.
73
Chapter 2 Eye size
2.6. Acknowledgments
We wish to thank Dr. Mike Hall (AIMS) and the Masters and crews of the RV Cape
Ferguson and, Prof. Hans-Joachim Wagner (University of Tübingen) and the Masters
and crews of the FS Sonne for sea time opportunities. We thank Prof. Lynnath Beckley
(Murdoch University), Dr. Pilar Olivar (CSIC), Dr. Anna Bozzano (CSIC) and Dr.
Brigitte Guillaumont (Ifremer), for providing additional samples and, Adrain Flynn
(UQ), Caroline Kerr (UWA) and Alan Goldizen (UQ) for their precious help during
field trips. Finally, we wish to thank Dr Kara Yopak (UWA) for her help and for
initially suggesting the use of phylogenetic comparative analyses for this study and Jan
Poulsen for access to his phylogeny prior to publication.
2.7. References
Ahlstrom EH, Stevens E. 1976. Report of neuston (surface) collections made on an
extended CalCOFI cruise during May 1972. California Cooperative Oceanic
Fisheries Investigations. Reports 18:167-180.
Backus RH, Craddock JE, Haedrich RL, Shores DL, Teal JM, Wing AS, Mead GW,
Clarke WD. 1968. Ceratoscopelus maderensis: Peculiar sound-scattering layer
identified with this myctophid fish. Science 160:991-993.
Beebe W, Vander Pyl M. 1944. Eastern Pacific expeditions of the New York Zoological
Society. XXXIII. Pacific Myctophidae (fishes). Zoologica 29:59-95.
Bourret P. 1985. Poissoins teleosteens: Gonostomatidae, Sternoptychidae, et

Myctophidae (MUSORSTOM II). Memoires du Museum National d'Histoire
Naturelle 133:55-82.
181.
Collin SP, Hoskins R, Partridge J. 1997. Tubular eyes of deep-sea fishes: a comparative
study of retinal topography (Part 2 of 2). Brain, Behavior and Evolution
50(6):347-357.
74
Chapter 2 Eye size
Collin SP, Hoskins RV, Partridge JC. 1998. Seven retinal specializations in the tubular
eye of the deep-sea pearleye, Scopelarchus michaelsarsi: a case study in visual
optimization. Brain, Behavior and Evolution 51(6):291-314.
Cornejo R, Koppelmann R. 2006. Distribution patterns of mesopelagic fishes with

special reference to Vinciguerria lucetia Garman 1899 (Phosichthyidae: Pisces)
in the Humboldt Current Region off Peru. Marine Biology 149(6):1519-1537.
41(1):59-70.
Flynn AJ. 2012. Ecology and Zoogeography of Lanternfishes (family Myctophidae).

PhD thesis.: The University of Queensland.
Flynn AJ, Kloser R. 2012. Cross-basin heterogeneity in lanternfish (family

Myctophidae) assemblages and isotopic niches (δ13C and δ15N) in the southern
Tasman Sea abyssal basin. Deep Sea Research Part I: Oceanographic Research
Papers 69:113-127.
Freckleton RP, Harvey PH, Pagel M. 2002. Phylogenetic analysis and comparative data:
a test and review of evidence. The American Naturalist 160(6):712-726.
Freckleton RP. 2009. The seven deadly sins of comparative analysis. Journal of
Evolutionary Biology 22(7):1367-1375.
Garland T, Dickerman AW, Janis CM, Jones JA. 1993. Phylogenetic analysis of
covariance by computer simulation. Systematic Biology 42(3):265-292.
Gartner JV, Hopkins TL, Baird RC, Milliken DM. 1987. The lanternfishes (Pisces:
Myctophidae) of the eastern Gulf of Mexico. Fishery Bulletin 85(1):81-98.
Grafen A. 1989. The phylogenetic regression. Philosophical Transactions of the Royal

Society of London Series B, Biological sciences 326(1233):119-157.
Hansen K, Herring PJ. 1977. Dual bioluminescent systems in the anglerfish genus
Linophryne (Pisces: Ceratioidea). Journal of Zoology 182(1):103-124.
Harmon LJ, Weir JT, Brock CD, Glor RE, Challenger W. 2008. GEIGER: investigating
evolutionary radiations. Bioinformatics 24(1):129-131.
Hartmann AR, Clarke A. 1975. The distribution of myctophid fishes across the central
equatorial Pacific. Fishery Bulletin 73(3):633-641.
75
Chapter 2 Eye size
Harvey PH, Pagel MD. 1991. The comparative method in evolutionary biology. Oxford:
Oxford University Press. 239 p.

Experimental Zoololy 270(2):225-231.
Hulley PA. 1986a. Order Myctophiformes. Myctophidae. In: Smith MM, Heemstra PC,
editors. Smiths' Sea Fishes. Heidelberg: Springer-Verlag. p 282-322.
Hulley PA. 1986b. A taxonomic review of the lanternfish genus Triphoturus Fraser-
Brunner, 1949 (Myctophidae, Osteichthyes). Annals of the South African
Museum 97(4):71-95.
Hulley PA. 1992. Upper-slope distributions of oceanic lanternfishes (family:

Myctophidae). Marine Biology 114(3):365-383.
Hulley PA, Duhamel G. 2009. A review of the lanternfish genus Bolinichthys Paxton,
1972 (Myctophidae). Cybium 33(4):259-304.
The Living Marine Resources of the Eastern Central Atlantic. Rome: FAO: In
press.
Kaartvedt S, Staby A, Aksnes DL. 2012. Efficient trawl avoidance by mesopelagic

fishes causes large underestimation of their biomass. Marine Ecology Progress
Series 456:1-6.
Kawaguchi K, Shimizu H. 1978. Taxonomy and distribution of the lanternfishes, genus

Diaphus (Pisces, Myctophidae) in the western Pacific, eastern Indian Oceans
and the southeast Asian Seas. Bulletin of the Ocean Research Institute,
University of Tokyo 10:1-145.
76
Chapter 2 Eye size
Kinzer J, Schulz K. 1985. Vertical distribution and feeding patterns of midwater fish in
the central equatorial Atlantic. I. Myctophidae. Marine Biology 85(3):313-322.
Kristoffersen JB, Salvanes AGV. 1998. Effects of formaldehyde and ethanol

preservation on body and otoliths of Maurolicus muelleri and Benthosema
glaciale. Sarsia 83:95-102.
Land MF. 1981. Optics and vision in invertebrates. In: Autrum H, editor. Handbook of
Sensory Physiology, Vol VII/6B. Berlin, Heidelberg, New York: Springer.
p 471-492.
Land MF. 1990. Optics of the eyes of marine animals. In: Herring PJ, Campbell AK,
Whitfield M, Maddock L, editors. Light and life in the sea: Cambridge
University Press. p 149-166.
Laughlin SB, van Steveninck RRR, Anderson JC. 1998. The metabolic cost of neural
information. Nature neuroscience 1(1):36-41.
247(5437):155-157.
Linkowski TB. 1996. Lunar rhythms of vertical migrations coded in otolith

microstructure of North Atlantic lanternfishes, genus Hygophum (Myctophidae).
Handbook of sensory physiology. Berlin: Springer-Verlag. p 67-192.
Madison WP, Madison DR. 2011. Mesquite: a modular system for evolutionary
analysis. Version 2.75. http://mesquiteproject.org
Mahé JC, Poulard JC. 2005. Manuel des protocoles de campagne halieutique.
Campagnes EVHOE (EValuation des ressources Halieutiques de l’Ouest
Europe). http://archimer.ifremer.fr/doc/00036/14707/.
Mallefet J, Shimomura O. 1995. Presence of coelenterazine in mesopelagic fishes from

the Strait of Messina. Marine Biology 124(3):381-385.
27:510-523.
Tiere 38:521-528.

77
Chapter 2 Eye size

Munk O. 1965. Ocular degeneration in deep-sea fishes. Galathea-Rep 8:21-31.
Murray J, Hjort J. 1912. The Depths of the Ocean. London: Macmillan. 821 p.
Nafpaktitis BG, Nafpaktitis M. 1969. Lanternfishes (family Myctophidae)

collectedduring cruises 3 and 6 of the R/V ANTON BRUUN in the Indian
Ocean: Science Bulletin, Los Angeles County Natural History Museum. 1-79 p.
Research. p 13-265.
Nafpaktitis BG. 1978. Systematics and distribution of lanternfishes of the genera

Lobianchia and Diaphus (Myctophidae) in the Indian Ocean. Science Bulletin,
Los Angeles County Natural History Museum 30:1-92.

Olivar MP, Bernal A, Moli B, Pena M, Balbin R, Castellon A, Miquel J, Massuti E.

2012. Vertical distribution, diversity and assemblages of mesopelagic fishes in
the western Mediterranean. Deep Sea Research Part I: Oceanographic Research
Papers 62(0):53-69.
Pagel M. 1999. Inferring the historical patterns of biological evolution. Nature

401(6756):877-884.
Paradis E, Claude J, Strimmer K. 2004. APE: Analyses of phylogenetics and evolution

in R language. Bioinformatics 20(2):289-290.

13:1-81.
78
Chapter 2 Eye size

gene rearrangements cooroborate evolutionary relationships of myctophiformes
(Neoteleostei). BMC Evolutionary Biology 13:111.
Revell LJ. 2012. phytools: an R package for phylogenetic comparative biology (and
other things). Methods in Ecology and Evolution 3(2):217-223.
Robison BH. 1972. Distribution of the midwater fishes of the Gulf of California. Copeia
1972(3):448-461.
Ross SW, Quattrini AM, Roa-Varon AY, McClain JP. 2010. Species composition and
distributions of mesopelagic fishes over the slope of the north-central Gulf of
Mexico. Deep-Sea Research Part II: Topical Studies in Oceanography 57(21-
23):1926-1956.
Stefanescu C, Cartes J. 1992. Benthopelagic habits of adult specimens of Lampanyctus

crocodilus (Risso, 1810) (Osteichthyes, Myctophidae) in the western
Mediterranean deep slope. Scientia Marina 56(1):69-74.
Brauer. Nature 233(5322):623-624.
deep-sea. Deep-Sea Research I: Oceanographic Research Papers 56(6):1003-
1017.

17(4):637-685.
Wagner H-J. 2001. Sensory brain areas in mesopelagic fishes. Brain, Behavior and
Evolution 57(3):117-133.
Warrant EJ. 2000. The eyes of deep-sea fishes and the changing nature of visual scenes
with depth. Philosophical Transactions of the Royal Society of London Series B,
Biological Sciences 355(1401):1155-1159.
79
Chapter 2 Eye size
Warrant EJ, Collin SP, Locket NA. 2003. Eye disign and vision in deep-sea fishes. In:
Collin SP, Marshall NJ, editors. Sensory processing in aquatic environments.
New York: Springer-Verlag. p 303-322.
Warrant EJ. 2004. Vision in the dimmest habitats on Earth. Journal of Comparative
Physiology A, Sensory Neural and Behavioral Physiology 190(10):765-789.
712.

Williams A, Koslow JA. 1997. Species composition, biomass and vertical distribution
of micronekton over the mid-slope region off southern Tasmania, Australia.
Wisner RL. 1976. The taxonomy and distribution of lanternfishes (family Myctophidae)
of the eastern Pacific Ocean: Navy Ocean Research and Development Activity.
NORDA Report. 3. 229 p.
Yang J, Huang Z, Chen S, Lee C. 1996. The deep-water pelagic fishes in the area from
Nansha Islands to the northeast part of South China Sea. Beijing: Science Press.
p 58-141.
Zahuranec BJ. 2000. Zoogeography and systematics of the lanternfishes of the genus
Nannobrachium (Myctophidae: Lampanyctini). Smithsonian Contributions to
Zoology 607:1-69.
80
Chapter 3 The eyes of lanternfishes
CHAPTER 3
The eyes of lanternfishes (Myctophidae, Teleostei):

novel ocular specialisations for vision in dim light
Fanny de Busserolles1, N. Justin Marshall2, Shaun P. Collin1
Australia, Crawley, WA, 6012, Australia.
2. Sensory Neurobiology Group, Queensland Brain Institute, University of Queensland, St Lucia, QLD,
4072, Australia.
81
3.1. Abstract
Lanternfishes are one of the most abundant groups of mesopelagic fishes in the
world’s oceans and play a critical role in biomass vertical turnover. Despite their
importance, very little is known about their physiology and how they use their sensory
systems to survive in the extreme conditions of the deep-sea. In this study we provide a
comprehensive description of the general morphology of the myctophid eye, based on
the analysis of 53 different species in order to better understand their visual capabilities.
Results confirm that myctophids possess several visual adaptations for dim-light
conditions, including enlarged eyes, an aphakic gap, a tapetum lucidum, and a pure rod
retina with high densities of long photoreceptors. Two novel retinal specialisations were
also discovered. The first specialisation is a fundal pigmentation in adult eyes, found
within an isolated retinal region (typically central retina) composed of modified pigment
epithelial cells, which we hypothesize to be the remnant of a more pronounced visual
specialisation important in larval stages. The second specialisation is an aggregation of
extracellular microtubular-like structures found within the sclerad region of the inner
nuclear layer of the retina. We hypothesize that the marked interspecific differences in
the hypertrophy of these microtubular-like structures may be related to inherent
differences in visual function. A general interspecific variability in other parts of the eye
is also revealed and examined in this study. The contribution of both ecology and
phylogeny to the evolution of ocular specialisations and vision in dim light are
discussed.
3.2. Introduction
The lanternfishes belong to the Myctophidae, a monophyletic family of deep-sea
teleost fishes belonging to the order Myctophiformes (Paxton, 1972; Stiassny, 1996).
The family is divided into two sub-families, the Myctophinae and Lampanyctinae,
seven tribes, 33 genera and is represented by around 250 species (Hulley and Paxton, In
press). All species are luminous and possess ventral and ventro-lateral photophores
(except Taaningichthys paurolychnus) arranged in species-specific patterns and often
possess additional luminous organs and tissue over the head and body (Paxton, 1972).
Myctophids are one of the most abundant groups of mesopelagic fishes in the world’s
oceans and play a major role in the marine ecosystem by transferring energy to the
deeper layers of the ocean through their vertical migration behaviours (Moku et al.,
2000; Cherel et al., 2010). Despite their critical role, very little is known about their
82
physiology and how they use their sensory systems to survive in the extreme conditions
of the mesopelagic zone of the deep-sea (200-1000 m).
The mesopelagic zone is characterised by very low levels of downwelling

residual sunlight and consequently is often referred to as the twilight zone. In the
twilight zone, the intensity of the downwelling sunlight diminishes exponentially with
depth creating a semi-extended visual scene, where shadows (prey, predator or
conspecific) can be detected against the lighter background when viewed from below.
The mesopelagic zone is also characterised by the presence of bioluminescence flashes,
produced by the animals themselves, which are mainly concentrated in the blue-green
part of the visible light spectrum (Widder, 2010). Those bioluminescent signals are
thought to play important roles in the deep-sea, subserving behavioural interactions with
prey, predators and congeners (Herring, 2002; Haddock et al., 2010).
Due to the dim light conditions and the production of large numbers of
intermittent, high contrast, bioluminescent flashes, vision is an important sensory
system used by many inhabitants of the mesopelagic zone (Wagner, 2001). However, to
see in dim light conditions and to view these brief bioluminescence emissions, the eyes
of mesopelagic fishes have evolved specialisations to increase sensitivity and optimise
photon capture. With the exception of tubular-eyed species, which have eyes directed
upwardly to maximise the capture of the downwelling sunlight (Munk, 1966; Locket,
1977; Collin et al., 1998), most mesopelagic fishes possess hemispherical, camera-type
eyes placed laterally in the head. Many species show specialisations for optimizing
photon capture, extending the visual field and tuning their spectral sensitivity to the
ambient light environment.
The most common adaptation for photon capture is the enlargement of the eyes
and pupil size (Marshall, 1954). In numerous species, the amount of light entering the
eye is also increased, by virtue of an aphakic gap, where the lens fails to fill the
pupillary aperture increasing the field of view, generally forward (Munk, 1966; Munk
and Frederiksen, 1974). Other species (i.e. Scopelarchus sp.) possess a yellow lens,
which absorbs short-wavelengths and may help to break the camouflage of counter-
illuminating species by enhancing the contrast between their bioluminescent emissions
and the downwelling sunlight (Somiya, 1982; Douglas et al., 1998). Visual adaptations
enhancing sensitivity include a tapetum lucidum, a reflective structure or mirror, sitting
83
at the back of the eye, which reflects light back through the retina for a second time,
thereby increasing photon capture by the photoreceptors (Arnott et al., 1970; Somiya,
1980). Most mesopelagic teleosts possess only rod photoreceptors, which are
specialised for dark-adapted (scotopic) vision in dim light, where sensitivity is enhanced
by increasing the length of the rod outer segment and surveying the visual scene with a
dense array of closely-packed receptors (Locket, 1977). In some species, rod
photoreceptors are arranged in multibanks (Munk, 1966; Fröhlich et al., 1995; Locket,
1977; Fröhlich and Wagner, 1998), which, in addition to enhancing the chance of
photon capture, may also allow colour vision in single (visual) pigment species or at
least provide hue discrimination to potentially break camouflage by counter-
illumination (Denton and Locket, 1989). A single visual pigment within the rod
photoreceptors seems to be typical for most deep-sea fishes (over 80%), where the peak
spectral sensitivity (λmax) ranges from 468 to 494 nm (Partridge et al., 1989; Crescitelli,
1990; Douglas and Partridge, 1997) and matches approximately the spectral
composition of the downwelling sunlight and the bioluminescent flashes.
Despite their abundance and role in vertical turnover in biomass, very little is
known about the visual system of myctophids. To date, the most comprehensive studies
on lanternfish vision are from Bozzano et al. (2007) and Turner et al. (2009), which
focus mainly on the photoreceptors and the spectral sensitivity of their visual pigments.
Bozzano et al. (2007) studied several larval stages in three different lanternfish species
and found a predominance of rods with increases in outer segment length during
development but, surprisingly, also cones, although cone numbers steadily decreased,
being lost in all post-larval stages (Bozzano et al., 2007). Turner at al. (2009) studied
the spectral sensitivity of the rod visual pigment in 58 different species of myctophids
using visual pigment extract spectrophotometry. Most of the Myctophidae were single
pigment species with only four species (Cerastocopelus warmingii, Hygophum
proximum, Myctophum aurolaternatum, M. nitidulum), belonging to both sub-families,
possessing two visual pigments. The wavelength of maximum absorption (λmax) of these
pigments falls between 480 and 492 nm (Turner et al., 2009; Douglas and Partridge,
1997; Douglas et al., 1998; Hasegawa et al., 2008; Partridge et al., 1992) and are,
therefore, well-adapted to visualisation of blue-green bioluminescent light, the most
commonly emitted light by deep-sea organisms.
84
Some older studies on myctophids have investigated retinal structure and the
sampling of the visual environment. However, these data have been compiled from very
few species, which all reveal that lanternfishes appear to have evolved eyes designed to
enhance sensitivity with the presence of an aphakic gap (Tarletonbeania crenularis,
Lawry, 1974), a pure rod retina (Lampanyctus crocodilus, Vilter, 1951; Lampanyctodes
sp., Pankhurst, 1987; Stenobrachius leucopsarus, O'Day and Fernandez, 1976), a
tapetum lucidum (Stenobrachius leucopsarus, O'Day and Fernandez, 1976), high
photoreceptor densities (Lampanyctus crocodilus, Vilter, 1951; Lampanyctodes sp.,
Pankhurst, 1987; Stenobrachius leucopsarus, O'Day and Fernandez, 1976), and a rather
unspecialised retina with poor acuity (Lampanyctus macdonaldi, Myctophum
punctatum, Collin and Partridge, 1996; Wagner et al., 1998).
In this study, we provide a comprehensive description of the general

morphology of the myctophid eye, based on the analysis of several species in order to
better understand their visual capabilities. One of the characteristics of the Myctophidae
is their great interspecific variability in ecology and behaviour (depth distribution,
Karnella, 1987; migration pattern, Watanabe et al., 1999; luminous organs, Edwards
and Herring, 1977) and so a diverse range of ocular specialisations may also be
expected.
3.3. Material and methods

3.3.1. Collection of species and preservation of ocular tissue
The eyes of 53 different species from 18 genera were analysed in this study.
Samples were obtained from several research cruises in the Coral Sea (RV Cape
Ferguson) under the following collection permits: Coral Sea waters (CSCZ-SR-
20091001-01), Commonwealth waters (AU-COM2009051), GBRMPA (G09/32237.1)
and Queensland Fisheries (133805), (Marshall, AEC # SNG/080/09/ARC), and in the
Peru-Chile trench (FS Sonne, sampling permits obtained by the Chief Scientist,
University of Tübingen). For all specimens, sampling was carried out following the
guidelines of the NH&MRC Australian Code of Practice, under a University of Western
Australia Animal Ethics protocol (RA/3/100/917). Additional specimens from Western
Australian waters, the Western Mediterranean Sea and the Bay of Biscay were acquired
through collaborators (de Busserolles et al., 2013; Chapter 2). Only large juveniles and
adults were analysed in this study. While most of the samples are registered as voucher
85
specimens at the Australian Museum in Sydney, Australia, further taxonomic analyses

need to be carried out for five of our study species to positively confirm identification
(Lampanyctus vadulus, Myctophum spinosum, Nannobrachium cf. nigrum,
Symbolophorus cf. boops, Triphoturus oculeus, de Busserolles et al. 2013; Chapter 2).
Two species were photographed using a digital camera and/or a

stereomicroscope to show the position and relative size of the eyes in the head, the
extent of the binocular overlap and the location of both photophores and luminous
organs. Observations, dissection and morphometric analyses were performed on board
ship and on fresh specimens, wherever possible. For each individual, the standard length
and rostro-caudal eye diameter were measured with digital callipers (to a precision of
0.1 mm) prior to dissection and fixation. The position of the aphakic gap, when present,
was noted (and ascribed to falling within the dorsal, nasal, ventral or temporal quadrants
of the eye) and photographed using an Olympus stereomicroscope SZX10 mounted with
a Canon digital camera. Once the eyes were enucleated, the cornea and lens were
dissected free of the eyecup and the colour, extent and integrity of the tapetum lucidum
assessed before photographing the fundus with an Olympus stereomicroscope SZX10
mounted with a Canon digital camera. The appearance of the tapetum can change
dramatically depending on the light conditions and freshness of the specimen. As a
consequence, only the presence/absence of the tapetum (defined as a colored and often
silvered reflex emanating from the fundus) for each species was noted, although we also
have identified species where the tapetum was observed to be covering the whole retina.
Eyes, lens and cornea were then fixed in 4% paraformaldehyde in 0.1M phosphate
buffer (PFA, pH 7.4) or Karnovsky’s fixative (2% paraformaldehyde, 2.5glutaraldehyde
in 0.1M cacodylate buffer, pH 7.4) for at least 1h and then stored in 0.1M phosphate
buffer. When we encountered small specimens (with small eyes), the entire body was
immersion fixed onboard and the dissections performed within the Neuroecology
laboratory at The University of Western Australia. When specimens were acquired from
collaborators, observations, measurements and dissections were all performed on
previously fixed tissue, which were preserved in 5% buffered formalin or Karnovsky’s
fixative.
3.3.2. Preparation of eyecup and retinal wholemounts

The eyes of several species (fixed in either 4% PFA or Karnovsky’s fixative)
were dissected further for analyses of the topographic arrangement of novel
86
specialisations, the orientation of the tapetum lucidum, comparisons of retinal thickness

and ultrastructural analysis (see below). For wholemounts, radial cuts were performed
in order to flatten the eye cup. The orientation of the eye was kept by making a small
additional cut in the nasal or dorso-nasal part of the eye. Retinal wholemounts were
made according to standard protocols (Stone, 1981; Coimbra et al., 2006; Ullmann et
al., 2011). However, the retinal pigment epithelium was left intact and remained
attached to the back of the retina in order to observe the size and appearance of peculiar
pigmentation (defined here as “fundal pigmentation”) situated in the centre of the retina.
Moreover, both the retinal pigment epithelium and underlying tapetum lucidum were
left intact for the wholemounts of the eyecup to enable the assessment of the coverage
of the tapetal reflex. A schematic of a lanternfish eye (redrawn from the schematic of a
teleosts eye) was used to approximate the size of the visual field affected by the fundal
pigmentation in relation to the size of the fish.
3.3.3. Light and electron microscopy

Samples fixed in Karnovsky’s fixative (2 % paraformaldehyde, 2.5 %
glutaraldehyde in 0.1 M sodium cacodylate buffer, pH 7.4) were preferentially used to
prepare tissue samples for examination using light and electron microscopy. However,
when no Karnovsky’s-fixed samples were available, tissue fixed in 4%
paraformaldehyde in 0.1 M phosphate buffer or in 5 % buffered formalin were also
used. One eye per species was analysed and characterised using both light and
transmission electron microscopy. Depending of the size of the specimens, either the
whole eye or a part of the eye (half or quadrant) was processed for histology. Following
initial preservation (up to 1 month at 4 deg C), the ocular tissue was then postfixed for
1hr with 1 % osmium tetroxide in 0.15 M phosphate buffer, dehydrated through an
alcohol and propylene oxide series and infiltrated with procure/araldite (ProSciTech).
For light microscopy, semithin sections (1 μm) were cut with a glass knife using a LKB
Bromma Ultratome NOVA. Sections were stained with an aqueous mixture of 0.5 %
Toluidine blue and 0.5 % borax, examined using an Olympus BX50 compound light
microscope and photographed using an Olympus DP70 digital camera. For transmission
electron microscopy, thin sections (110 nm) were cut using a diamond knife, mounted
on a 200 mesh copper grid and stained with Reynold’s lead citrate. Examination of the
sections was done at the UWA Centre for Microscopy, characterisation and Analysis
(CMCA) using a JEOL 2100 transmission electron microscope operated at 120 kV and
digital images taken using an 11 megapixel Gatan Orius digital camera. All
87
measurements were made on pictures using ImageJ 1.45 (National Institutes of Health,
USA). To allow comparison between species, all measurements were taken in the
central part of the retina. Eight measurements were taken for each eye component from
a single section; the average measure is reported here.
All colored (non-histological) photographs presented in this study show the

colours present in either fresh (Figures 3.1, 3.3, 3.11) or fixed (Figures 3.9, 3.10, 3.12,
3.13) material, where only slight adjustments to brightness and contrast were performed
(using Photoshop CS4). However, slight changes in hue may have occurred post-
mortem or after exposure to the low concentrations of fixative used.
3.3.4. Phylogenetic comparison of visual characteristics

In order to assess the influence of phylogeny on the occurrence and relative
development of some of the ocular and retinal specialisations observed in lanternfishes,
we mapped each trait (fundal pigmentation, microtubular-like structure, tapetum
lucidum and aphakic gap) onto an existing phylogenetic tree (Paxton et al., 1984)
including all of the species examined here. Paxton et al. (1984) have published a
phylogeny classifying genera using derived character states of adult osteology and
photophore patterns (Paxton et al., 1984), and of larvae, as described by Moser and
Ahlstrom (1970, 1972, 1974) and is currently the most up to date morphological
phylogeny. The only difference between Paxton et al’s originally described phylogeny
and the one used in the present study is the inclusion of the genus Nannobrachium,
which was added to Paxton et al’s phylogeny after Zahuranec (2000). For each visual
trait, the presence or absence of the character was noted for every species on the tree.
For the aphakic gap trait, the presence/absence of the aperture was noted for each eye
quadrant (dorsal, nasal, ventral, temporal) separately to provide additional information
about interspecific differences in the location of the aphakic gap. Additional information
about the microtubular-like structure (MLS) and the tapetum lucidum was also provided
by indicating the degree of development of the MLS (abundance) and by highlighting
the species, which posssess a tapetum lucidum covering the whole retina.
88
3.4. Results
3.4.1. General morphology of the eye and retina
All lanternfish species examined in this study possessed laterally-oriented eyes.
The extent of the visual field of view was not investigated in detail but basic
observations indicated a large monocular visual field and binocular overlap frontally,
dorsally and ventrally (Figure 3.1).
Figure 3.1. Position of the eyes in Myctophum spinosum (A) and Diaphus danae (B, C, D). (A)
lateral view, (B) dorsal view, (C) ventral view, (D) frontal view. Scale bar 10 mm (A), 0.5 mm
(B, C, D).
Lanternfishes possess a classic, camera-type eye (Figure 3.2). The shape of the
eye is relatively hemispherical in all species, however, there is a great interspecific
variation in pupil size and shape and irideal pigmentation (Figure 3.3), where the
aphakic gap (or region of the pupil that is not occupied by the spherical lens) accounts
for the greatest variation in the size of the pupillary aperture (Figure 3.3). Some species
do not possess an aphakic gap (i.e. Diaphus danae, Figure 3.3A), while others have an
aphakic gap in a specific part of the eye (i.e. ventral in Hygophum proximum, Figure
3.3F), or a circumlental gap that extends all the way around the lens (i.e. Lampanyctus
parvicauda, Figure 3.3K). Pigmentation also varies across the iris, which can be entirely
silver (i.e. Myctophum nitidulum, Figure 3.3C), completely dark (i.e. Nannobrachium
cf. nigrum, Figure 3.3D), of a specific colour (i.e. blue in Nannobrachium idostigma,
Figure 3.3I) or a combination of colours and dark pigmentation (i.e. Nannobrachium
phyllisae, Figure 3.3L).
89
Figure 3.2. Transverse section through the eye of Ceratoscopelus warmingii. C = cornea,
L = lens, ON = optic nerve, R = retina, Sc = sclera. Scale bar = 1 mm.
The myctophid eye possesses six extra-ocular muscles presenting the regular
arrangement for vertebrates. The optic nerve is non-pleated and its position at the back
of the eye is slightly temporal of centre in most species. All the species analysed in this
study possess a single (primary), transparent cornea and a colourless vitreous humour.
The lens is also colourless, spherical in shape and there is a positive relationship
between lens size and eye size (de Busserolles et al., 2013). Although, often difficult to
assess (especially in species with small eyes), the lens was supported by a ventrally-
located retractor lentis muscle and a dorsal suspensory ligament. No attempt was made
to investigate accommodative lens movement in this study, which remains unknown for
this family of deep-sea teleosts. No falciform process was observed in any of the species
examined, where the vascular supply is presumably achieved by the network of vitreal
vessels overlying the inner limiting membrane (providing nutrition to the inner retina)
and the dense network of blood vessels comprising the pigmented choriocapillaris
(providing nutrition to the outer retina). There does not appear to be a distinct choroidal
gland.
90
Figure 3.3. Variation of the aphakic gap within the lanternfish family from no aphakic gap (A)
to circumlental aphakic gap (L). Diaphus Danae (A), Gonichthys tenuiculus (B), Myctophum
nitidulum (C), Nannobrachium cf. nigrum (D), Diaphus brachycephalus (E), Hygophum
proximum (F), Diogenichthys laternatus (G), Triphoturus oculeus (H), Lampanyctus omostigma
(I), Lampanyctus parvicauda (J), Nannobrachium idostigma (K), Nannobrachium phyllisae (L).
The white arrows indicate the orientation of the eyes, T = temporal, V = ventral.
91
The fixed retina is typically whitish in colour, with the exception of a small
number of species, which possess isolated patches of yellow pigmentation in different
parts of the retina. Retinal thickness appears to be relatively uniform, except for one
species, Benthosema suborbitale, where the retina was found to be significantly thicker
in the temporo-ventral region (~180 μm) compared to the rest of the retina (~95 μm,
Figure 3.4). Retinal thickness in the central part of the retina also shows significant
interspecific variation (from about 65 μm to 220 μm, Table 3.1) but the retinal cell
layers are typical of other vertebrate retinas (Figure 3.5).
Figure 3.4. Light micrograph of a transverse section through the eye cup of two lanternfish
species, showing differences of retinal thickness. The ventral part of the retina is thicker in
Benthosema suborbital (A, C, D) in comparison to the uniform retinal thickness in Lampanyctus
parvicauda (B, E). Scale bars = 0.5 mm (A, B), 50 μm (C, D, E).
3.4.2. Anatomy of the inner retina

The ganglion cell layer is comprised of a single layer of ganglion cell somata
and a population of presumed amacrine cell somata (smaller and more darkly-staining,
Bozzano et al., 2007). The inner plexiform layer (IPL) is usually quite thick (between
8.4 and 64.7 μm, Table 3.1) and is composed of the axonal connections between the
ganglion cells and the cells within the inner nuclear layer.
92
Table 3.1. Summary of retinal measurements for 53 species of lanternfishes. The standard
length (SL), eye diameter (eye ø), lens diameter (ø) and life stage (S, adult = A, juvenile = J,
? = unknown) of each specimen is also given. All measurements were done in the central part of
the retina. IPL = inner plexiform layer, INL = inner nuclear layer, OPL = outer plexiform layer,
ONL = outer nuclear layer, PRL = photoreceptor layer, OS = proportion of the retina occupied
by the outer segments of the photoreceptors, n.a. = values not available due to the poor
condition of the sample.
Species S SL Eye ø Lens ø Retina IPL INL OPL ONL PRL OS

(mm) (mm) (mm) (μm) (μm) (μm) (μm) (μm) (μm) (%)
Benthosema glaciale A 42.0 4.6 1.8 146.6 40.0 15.6 5.9 30.0 50.0 27.0
B. suborbitale A 28.2 3.0 1.3 n.a. n.a. n.a. n.a. n.a. n.a. n.a.
Bolinichthys longipes ? 35.0 3.8 1.5 151.6 33.6 13.2 4.8 27.9 68.0 38.7
B. nikolayi J 27.5 2.9 1.2 152.7 37.5 15.5 4.9 18.9 92.9 57.2
B. supralateralis J 41.2 4.2 2.0 181.1 26.5 16.8 3.2 33.0 96.3 49.2
Ceratoscopelus maderensis A 52.0 5.1 2.1 n.a. n.a. n.a. n.a. n.a. 40.8 n.a.
C. warmingii A 60.8 5.9 2.5 140.7 29.3 11.5 4.8 29.8 61.1 34.9
Diaphus brachycephalus A 35.2 4.3 1.8 166.2 46.1 15.2 8.8 20.4 67.0 30.2
D. danae A 92.2 8.2 3.3 187.1 52.5 22.5 9.9 41.4 51.4 32.3
D. fulgens ? 41.2 3.4 1.8 161.4 34.0 14.4 8.9 43.2 54.7 26.5
D. gamani J 33.1 2.5 1.0 203.6 45.4 21.8 8.6 53.8 65.7 24.6
D. holti J 40.0 5.1 2.1 n.a. 40.0 n.a. 9.7 33.5 58.4 n.a.
D. luetkeni A 38.3 2.4 1.1 153.1 30.9 19.6 7.1 30.2 60.6 35.0
D. meadi J 28.3 3.3 1.2 179.2 38.4 23.1 6.5 46.2 46.4 20.2
D. mollis A 39.5 4.0 1.8 n.a. 48.3 19.8 11.7 n.a. 58.3 n.a.
D. parri A 51.0 5.6 2.1 181.7 47.0 16.5 12.6 38.5 60.5 25.0
D. phillipsi J 27.4 2.2 1.0 153.2 45.4 17.5 5.2 36.1 41.7 18.1
D. regani J 40.8 2.5 1.2 154.1 25.4 16.7 5.6 31.8 67.4 36.3
D. splendidus J 34.1 2.1 0.8 167.2 32.0 15.3 6.4 34.8 69.2 36.2
D. termophilus J 48.3 4.2 1.8 159.3 39.0 19.3 7.2 30.2 53.7 25.4
Diogenichthys atlanticus A 21.4 2.7 1.02 104.3 19.9 12.1 3.2 28.3 38.4 30.2
D. laternatus A 22.1 2.2 0.9 139.5 29.6 13.1 4.9 36.9 47.8 27.3
Electrona risso A 46.0 7.0 3.0 n.a n.a n.a n.a n.a 37.2 n.a
Gonichthys tenuiculus A 41.3 2.9 1.4 200.0 27.0 18.8 10.8 77.4 62.0 20.0
Hygophum benoiti A 45.0 6.2 2.3 112.3 20.9 12.2 3.5 17.4 56.3 40.1
H. hygomii A 57.3 7.5 3.1 115.2 21.6 14.2 2.1 16.2 54.2 38.8
H. proximum J 26.2 3.2 1.2 166.4 48.5 16.8 3.4 24.1 55.1 26.9
Lampadena. urophaos A 40.0 2.8 1.3 162.2 32.6 15.5 6.4 42.2 55.6 26.2
Lampanyctus alatus A 43.2 2.1 0.9 118.4 26.4 9.5 4.4 16.3 56.7 43.6
L. crocodilus J 31.0 1.7 0.6 64.8 8.4 7.9 2.1 12.4 33.3 45.7
L. iselinoides J 34.3 2.0 0.8 127.0 22.2 13.6 3.1 15.3 64.6 46.8
L. nobilis J 37.5 2.1 1.0 112.6 25.9 11.8 2.7 15.0 50.3 41.1
L. omostigma ? 27.8 1.9 0.6 155.7 33.3 23.3 5.0 33.9 55.4 30.8
L. parvicauda A 28.4 1.8 0.9 84.8 20.0 10.7 2.3 16.2 42.0 42.9
L. pusillus A 37.0 2.1 0.8 n.a. n.a. n.a. n.a. n.a. 67.3 n.a.
L. vadulus J 37.4 2.2 1.0 158.2 29.4 20.1 6.3 28.5 66.8 37.0
L. gemellari J 32.7 2.1 1.2 n.a. 32.6 n.a. 7.7 38.0 57.2 n.a.
Loweina interrupta J 25.0 2.1 0.9 157.7 30.6 30.0 2.8 23.7 58.2 26.5
Myctophum brachygnatum A 67.7 7.6 3.2 185.8 36.7 23.5 11.7 46.8 60.2 24.5
M. nitidulum A 85.4 7.1 2.5 204.2 48.4 26.4 10.9 47.5 65.6 22.0
M. spinosum J 39.6 3.9 1.6 171.8 38.2 16.6 7.6 45.5 54.1 22.7
Nannobrachium cf. nigrum A 52.3 2.5 1.0 120.6 25.1 11.2 3.8 15.8 57.1 43.7
N. idostigma ? 72.2 3.9 1.2 n.a. n.a. n.a. n.a. n.a. 51.0 n.a.
N. phyllisae ? 50.4 2.2 1.0 106.0 24.5 8.0 3.0 11.5 57.7 49.4
Notolychnus valdiviae A 19.5 1.5 0.6 143.8 22.4 12.3 5.0 32.4 62.7 38.8
Notoscopelus elongatus ? 47.0 3.2 1.2 176.1 64.7 22.6 3.5 30.8 37.9 13.4
N. kroeyeri A 101.2 6.1 2.7 188.2 58.7 21.0 n.a n.a 51.4 19.4
Symbolophorus cf. boops J 72.0 6.0 2.1 218.9 61.7 29.6 7.9 41.1 66.0 22.2
S. evermanni J 59.0 5.1 2.8 155.4 50.6 20.3 3.9 32.2 51.8 24.2
S. rufinus J 69.0 6.5 2.3 193.7 31.6 24.16 7.1 67.6 51.7 20.1
S. verany A 85.0 6.7 2.9 n.a. n.a. n.a. n.a. n.a. 54.8 n.a.
Triphoturus nigrescens A 35.4 2.0 0.9 160.6 40.9 16.2 4.9 23.9 61.6 31.7
T. oculeus ? 33.9 2.1 0.8 94.2 22.4 10.10 2.7 13.2 48.1 45.1
93
Figure 3.5. Light microscopy picture of a transversal section through the retina of Diogenicthys
laternatus. OS = outer segment, IS = inner segment, ONL = outer nuclear layer, OPL = outer
plexiform layer, INL = inner nuclear layer, IPL = inner plexiform layer, GCL = ganglion cell
layer. Scale bar = 20 μm.
The inner nuclear layer is relatively thin compared to the other layers (7.9 to
30.0 μm, Table 3.1) and contains four different types of cells: bipolar cells
(distinguished by their small size, sclerad location and darkly-stained nucleus),
amacrine cells (identified by their larger size and more vitread position), horizontal cells
(identified by their larger size, horizontal shape and most sclerad location within the
inner nuclear layer) and Müller cells (distinguished by their long descending processes
extending from the inner limiting membrane to the outer limiting membrane, Collin et
al., 1996). Horizontal cells are arranged within a single layer and occur in variable
densities across species with some species having very few horizontal cells.
The vitread part of the inner nuclear layer is filled with microtubular-like
structures (MLS, Figure 3.6). The MLS are extracellular and randomly-arranged into
bundles and are often in contact with the membranes of inner nuclear cells (bipolar
94
cells, horizontal cells, Müller cell processes). Each microtubule is between 26 and 28
nm in (outermost) diameter, which is greater than the inter-tubule space. All the species
analysed in this study appear to possess MLS but the density is highly variable (Figure
3.7). Most of the species possess bundles of MLS, arranged into a hexagonal array,
which are only visible at high power using transmission electron microscopy (TEM)
(i.e. Lampancytis nobilis, Figure 3.7C, F). Several species possess well-developed
systems of MLS, which are not clearly visible in light microscopy but are easily
observed at the level of the TEM (i.e. Diogenichthys laternatus, Figure 3.7B). Finally,
several species possess a well-developed MLS system forming bundles of several
micrometers and are very easily seen at the level of the light microscope (i.e.
Myctophum brachygnathum, Figure 3.7A). When the MLS are well-developed
(abundant), they seem to be predominantly surrounding the horizontal cells, which, in
these species, appear to be more abundant.
3.4.3. Anatomy of the outer retina

Axonal connections between the inner nuclear layer cells and the photoreceptor
cells are evident within the outer plexiform layer (2.1 μm to 12.6 μm, Table 3.1) of all
myctophid retinas examined. The outer nuclear layer (ONL) is relatively thick and, in
central retina, varies in thickness from 11.5 µm to 77.4 μm (Table 3.1) and is composed
of three to twelve sublaminae of rod nuclei. In the species with the highest
photoreceptor densities (Myctophum brachygnathum), the retinal thickness in the region
of highest photoreceptor density (temporo-ventral retina) is up to 130 μm, consisting of
around eighteen layers of rod nuclei (Chapter 4).
All the species of myctophids examined possessed a pure rod retina arranged in
a single bank. The rod photoreceptors are thin (0.94 μm to 2.74 μm in the central
retina), long and cylindrical, varying in length from 33 μm to 96 μm in Lampanyctus
crocodilus and Bolinichthys supralateralis, respectively. The outer segments, which
account for between 13.4 and 57.2 % of the total retinal thickness (Table 3.1), are
composed of a stack of membranous discs (21 μm to 24 μm in thickness with an inter-
disc space of between 12 μm and 14 μm in Myctophum brachygnathum) enclosed in a
cell membrane (Figure 3.8). In transverse sections, several connecting cilia can be
distinguished (Figure 3.8), characterised by a relatively long basal cilium (~ 1.9 μm in
length), containing highly osmiophilic material.
95
Figure 3.6. Light microscopy and transmission electron microscopy of the microtubular like
structure (MLS) in Myctophum nitidulum. PRL = Photoreceptor layer, ONL = outer nuclear
layer, OPL = outer plexiform layer, INL = inner nuclear layer, GCL = ganglion cell layer, HC =
horizontal cell . The black arrow indicates the location of the MLS. Scale bars = 200 μm (A),
50 μm (B), 2 μm (C), 1 μm (D), 200 nm (E), 100 nm (F).
96
Figure 3.7. Transmission electron microscopy of the microtubular-like structure (MLS) in three
species of lanternfishes, showing the variation in the degree of development of the MLS with
higher power ultrastructural detail depicted in the right hand panels. (A, D) MLS very well
developed, Myctophum brachygnatum; (B, E) MLS developed, Diogenichthys laternatus; (C, F)
MLS present, Lampanyctus nobilis. The white boxes on the left panels are magnified as the
right panels. INL = inner nuclear layer, OPL = outer plexiform layer, ONL = outer nuclear
layer, HC = horizontal cell, the white starts indicate the presence of MLS. Scale bars = 5 μm (A,
B, C), 0.5 μm (D, E, F).
97
Figure 3.8. Transmission electron micrographs of the rod photoreceptors in the lanternfish
Myctophum brachygnathum. OS = rod outer segment, IS = rod inner segment, c = connecting
cilium. Scale bars = 0.5 μm (A and C), 100 μm (B), 200 μm (D).
3.4.4. Specialisations of the retinal pigment epithelium

The retinal pigment epithelium (RPE) layer is very thin and typically depleted of
melanin in most of the species examined. However, in several species, the central retina
is pigmented but only within a discrete retinal region surrounding the optic nerve head,
where isolated and discontinuous patches of bundles of melanosomes can easily be
observed in retinal wholemount (Figure 3.9). This fundal pigmentation is composed of
modified pigment epithelial cells, which form membrane-bound groups of melanosomes
often appearing as a triangle-shaped wedge (up to 17 μm in width and up to 26 μm in
length in Diaphus splendidus) interleaved between the densely-arranged rod outer
segments. The density of the melanosomes varies between species, as does the spacing
of the pigmented wedges (i.e. Bolinichthys longipes, Figure 3.9), which reduce in size
towards their outer limit (Figure 3.10). In a few species, the melanosomes almost form a
continuum within the RPE, but again the limits of the fundal pigmentation are finite (i.e.
Lampanyctus alatus).
98
Figure 3.9. Different pigmentation patterns in lanternfishes. Top row: Bolinichthys longipes,
middle row: Lampanyctus alatus, bottom row: Diogenichthys atlanticus. Left column:
wholemounted retina, scale bar = 0.5 mm; middle column: light micrograph of a transverse
section of the retina, scale bar = 25 μm; right column: transmission electron micrograph of a
transverse section of the retina, scale bar = 5 μm. The black arrows indicate the position of the
pigmentation. The white arrows indicate the orientation of the wholemount retinas
(T = temporal, N = nasal). PRL = photoreceptor layer, ONL = outer nuclear layer, INL = inner
nuclear layer, GCL = ganglion cell layer, N = nucleus of retinal pigment epithelium cell.
The size of the fundal pigmentation appears to be constant throughout life. For
example, in Nannobrachium cf. nigrum, the diameter of the pigment array was found to
be constant (around 1.25 mm) in three adults with very different standard lengths
(44.26 mm, 57.97 mm, 85.49 mm, Figure 3.11A, B), indicating that the relative size of
the fundal pigmentation (with respect to the total retinal area) decreases with standard
length (and therefore age). This means that, in small individuals, the fundal
pigmentation subtends a much greater proportion of the visual field than in large
individuals (Figure 3.11B). In several species, the fundal pigmentation and the tapetum
lucidum were coincident (i.e. Lampanyctus parvicauda, Triphoturus oculeus).
99
In a single species (Diogenichthys atlanticus), the retinal pigment epithelial cells

are modified to form a dense series of pigmented wedges in an elliptically-shaped patch
in the retinal periphery (Figure 3.9). This peripheral pigmentation is found in the nasal
area of the retina in both males and females and is composed of a thick layer of
melanosomes (up to 20 μm in length and 14 μm in width).
Figure 3.10. Macroscopic pictures of the fundal pigmentation in Bolinichthys longipes (A and
B) and Ceratoscopelus warmingii (C), showing the arrangement and spacing of the
melanosomes which reduce in size toward the outer limit of the pigmentation. In (A) and (B),
the silver materials are residuals of the tapetum lucidum. The white arrows indicate the
orientation of the wholemount retina (T = temporal, N = nasal). Scale bars = 0.5 mm (A and B),
50 μm (C).
100
Figure 3.11. Fundal pigmentation in Nannobrachium cf. nigrum for 3 different size individuals.
(I) SL = 44.26 mm, (II) SL = 57.97 mm, (III) SL = 85.49 mm. (A) Wholemount retinas showing
the location and size of the fundal pigmentation. The dotted line in wholemount retina III
represents the original location of the pigmentation which was damaged during dissection. (B)
Schematic representation of the approximate size of the visual field affected by the fundal
pigmentation in relation to the size of the fish. The fundal pigmentation is represented in black
on the schematic. (C) Schematic superposition of the three wholemounts retinas emphasising
the fact that the retina keep growing as the fish grow but not the fundal pigmentation. The black
arrows indicate the orientation of the wholemount retinas (T = temporal, N = nasal).Scale bars =
1 mm.
3.4.5. Tapetum lucidum

Several species of lanternfishes possess a tapetum lucidum, which appears to
vary in colour and pattern and gives a specular reflexion (Figure 3.12). The colour of
the tapetum was recorded (digitally) but not investigated further due to the major
changes in colours depending on the freshness of the sample and consequent effects of
fixation. However, there appears to be a great interspecific and intraretinal variation in
the colour of the tapetal reflex even in live specimens sacrificed and preserved
immediately (see the range of colours in Figure 3.12). Under all conditions, the tapetal
colours were restricted to variations of blue, green, white and silver. Interestingly, the
orange/yellow retinal colours shown in Figure 3.12 are not produced by the tapetum
lucidum but are isolated patches of retinal pigment that remain even when the tapetal
tissue has been surgically removed (Chapter 5).
While in some species, the tapetum extends across the entire retina (i.e.
Lampanyctus parvicauda, Figure 3.13A), other species possess a tapetum that is
restricted to only a specific retinal region (i.e. dorso-temporal retina in Bolinichthys
longipes, Figure 3.13B) or is completely absent (i.e. Nannobrachium cf. nigrum, Figure
101
3.13C). Sexual dimorphism in the coverage and colour of the tapetum occurs in
Diaphus danae, with females having a greenish tapetum covering the entire retina and
males having a bluish tapetum only present in the temporal part of the eye (Figure
3.12A, B). The tapetum lucidum is situated in the vitread part of the choroid, adjacent to
Bruch’s membrane (Figure 3.14B-D) and is classified as a choroidal tapetum. This
choroidal tapetum is composed of several horizontal plates (Figure 3.14A) filled with a
highly osmiophilic component, arranged perpendicularly to the photoreceptor outer
segments and tiled into layers in a concentric pattern around the optic nerve head
(Figure 3.14).
Figure 3.12. Variation in tapetum lucidum pattern and color in nine species of lanternfishes:
Diaphus danae female (A), Diaphus danae male (B), Gonichthys tenuiculus (C), Lampanyctus
parvicauda (D), Myctophum nitidulum (E), Myctophum brachygnatum (F), Myctophum
lychnobium (G), Symbolophorus rufinus (H), Myctophum obtusirostre (I). The white arrows
indicate the orientation of the eye cup (T = temporal, V = ventral).
102
Figure 3.13. Tapetum lucidum variation in three species of lanternfishes. (A) tapetum covering
the entire retina in Lampanyctus parvicauda, (B) tapetum only present in the dorso-temporal
part of the retina in Bolinichthys longipes, (C) no tapetum in Nannobrachium cf. nigrum. N =
nasal, V = ventral. Scale bar = 1 mm.
Figure 3.14. Tapetum lucidum structure in myctophids. (A) close-up picture of the tapetum
lucidum in Myctophum spinosum. (B, C, D) transmission electron microscopy pictures of the
tapetum lucidum structure of Myctophum nitidulum. PR = photoreceptors, T = tapetum,
BV = blood vessel, RPE = retinal pigment epithelium. The white triangles indicate the position
the Bruch’s membrane. Scale bars = 1 μm (B), 0.5 μm (C), 100 nm (D).
103
3.4.6. Interspecific variability within a phylogenetic context

As revealed above, there is a marked interspecific variability in the range of
ocular specialisations within the large number of lanternfishes examined, many of
which have not been previously described. We consider it important to assess the
influence of phylogeny on the occurrence and relative development of these
specialisations, especially given the paucity of information regarding their behaviour
and life history traits. Therefore, we have mapped these traits onto an existing
phylogenetic tree (according to Paxton et al. 1984), that includes all of the species
examined.
Depending on the specialisation (fundal pigmentation, microtubule-like

structures, tapetum lucidum and aphakic gap), variations are observed at different
taxonomic levels (Figure 3.15). For the aphakic gap (with respect to both
presence/absence and location), a great variability is observed at all taxonomic levels,
where, at the level of subfamily, the Lampanyctinae reveal a greater diversity than the
Myctophinae. While the majority of the Myctophinae possess a ventral aphakic gap, the
Lampanyctinae possess a greater diversity in gap location with all possible
combinations observed in this subfamily. Variation in aphakic gap morphology is even
observed within the same genus with the most extreme variations seen within the genus
Lampanyctus. While some Lampanyctus species have no aphakic gap (i.e. L. pusillus),
others possess a gap just in one part of the eye (i.e. nasal in L. festivus) with a few
species possessing a circumlental aphakic gap (i.e. L. idostigma). With regard to the
fundal pigmentation (presence or absence) and the microtubule-like structures (MLS,
relative development), variability is mainly seen at the subfamily level. In fact, the
fundal pigmentation was found in 45 out of the 53 species analysed and was
systematically found in species from the subfamily Lampanyctinae. Only a few species
did not possess any fundal pigmentation and these were restricted to the Myctophinae
subfamily. Similarly, even if all the species analysed in this study possessed MLS, only
species from the subfamily Myctophinae presented hypertrophied MLS systems.
Variation in the tapetum lucidum pattern (presence/absence, and total retinal coverage)
is also revealed at the subfamily level but also at the generic level. The Myctophinae
consistently possesses a tapetum, which frequently covers the entire retina. Within the
Lampanyctinae, variation in the presence or absence of the tapetum is observed between
and within genera (Figure 3.15).
104
Figure 3.15. Presence-absence of specific visual characteristics in 53 species of lanternfishes

plotted against the phylogeny from Paxton et al. (1984). FP = fundal pigmentation, MLS =
microtubular-like structure, T = tapetum lucidum, AD = aphakic gap dorsal, AN = aphakic gap
nasal, AV = aphakic gap ventral, AT = aphakic gap temporal. Presence of the character (+),
absence of the character (-), information missing (?). For the tapetum lucidum, no tapetum (-),
tapetum present (+), tapetum observed everywhere (+++). For the MLS, MLS poorly developed
(+), MLS well developed (++), MLS very well developed (+++). A circumlental aphakic gap is
indicated when a (+) is present for all four areas of the aphakic gap (AD, AN, AV, AT).
105
3.5. Discussion
The large number of representative species examined from the Myctophidae
reveals that lanternfishes possess a typical vertebrate eye with a retina composed of cell
layers, comprising two interplexiform layers and three neuronal layers. The myctophid
retina is relatively thin (<200 μm in the central part) compared to what is usually found
in shallow water teleosts e.g. 200-300 μm (Wagner, 1990), and some species of deep-
sea teleosts with multiple banks of photoreceptors (up to 900 μm, Munk, 1980). As
predicted and described from previous studies, myctophids also possess what may be
considered typical visual adaptations for dim-light conditions (Vilter, 1951; Lawry,
1974; O'Day and Fernandez, 1976; Locket, 1977; Pankhurst, 1987) in order to enhance
the sensitivity of the eye, including enlarged eyes (Marshall, 1954; de Busserolles et al.,
2013), an aphakic gap, a tapetum lucidum and a pure rod retina with a high density of
long photoreceptors. However, these adaptations have not been previously examined
systematically within any family of deep-sea teleosts. Moreover, the high level of
variability and the fact that new ocular specialisations (e.g. fundal pigmentation and
microtubule-like structures) have been discovered, suggests that there is a wealth of new
information to be gleaned on the influences of ecology and phylogeny on the success of
this abundant group, which is obviously under intense selection pressure with respect to
the visual system.
3.5.1. The function of the aphakic gap

Most of the lanternfish species analysed in this study (46 out of 53) possess an
aphakic gap. Munk and Frederiksen (1974) described two types of aphakic gap in deep-
sea teleosts, crescent-shaped and circumlental, which are thought to functionally
enhance the relative retinal illumination in one particular zone of the eye (crescent-
shaped gap) or over the entire eye (circumlental gap). Rostrally located crescent-shaped
apertures are often associated with a temporal area of high acuity (i.e. area centralis or
fovea, Walls, 1942; Munk, 1966; Munk and Frederiksen, 1974) and are thought to
enhance visual sensitivity in this specific acute zone. However, this gap will only be
beneficial in the case of a visual stimulus located in front of the eye, where more of the
frontal visual field will be viewed by the temporal retina (potentially increasing the
binocular overlap) and where light falling on temporal retina will be refracted by the
lens, thereby providing a focussed image. Light entering the eye through the gap from
other visual axes will not be focussed by the lens, only providing course resolution of
any visual stimulus in a specific region of the visual field. Crescent-shaped aphakic
106
gaps are common in lanternfishes (Figure 3.3 and 3.15) but these extensions of the
pupillary aperture are not restricted to the nasal part of the iris with some gaps located
naso-ventrally (i.e. Diogenichthys laternatus), temporo-ventrally (i.e. Lampanyctus
omostigma) or only ventrally (i.e. Myctophum), thereby providing an increase in the
visual field along specific visual axes.
In order to test the relationship between the position of the aphakic gap and the
location of retinal regions for acute vision (Munk, 1966; Munk and Frederiksen, 1974),
a topographic analysis of the distribution of photoreceptors and/or ganglion cells in
lanternfishes would have to be performed. Although topographic maps of ganglion cell
distribution and calculations of spatial resolving power have been made for selected
deep-sea species (Collin and Partridge, 1996; Wagner et al., 1998), few have been
examined in myctophids and unfortunately the size and location of any aphakic gaps
was not noted. Since the diversity in the shape and location of aphakic gaps is variable
in myctophids and without the associated topographic analyses, we unfortunately cannot
make any firm conclusions in terms of the function of these crescent-shaped gaps.
Circumlental aphakic gaps are also found in myctophids (i.e. Nannobrachium

phyllisae). Circumlental apertures are thought to either increase the sensitivity of the
whole eye by increasing the amount of available light entering the eye (large gap) or to
facilitate the detection of weak visual signals on the optical axis by increasing the size
of a centrally illuminated retinal area (small gap, Munk and Frederiksen, 1974).
However, a circumlental gap could potentially be a disadvantage, allowing unfocussed
light from any direction to enter the eye, thereby not providing any directional
information about potential prey or predators and increasing the signal to noise ratio.
The distribution of light in the mytophid habitat is however very different to many
animals in that the visual field horizontally and downward is likely dominated with
discrete flashes or point sources of bioluminescent light. The lack of an extended field
of complex shapes and intensities, as we see for instance, would reduce this over all
noise. In other words, the benefits of having an aphakic aperture in such a light habitat
outweigh the possible disadvantages.
3.5.2. Tapetum lucidum

The presence of a tapetum lucidum is a common feature in the eyes of deep-sea
fishes. It is thought to increase the sensitivity of the eye by reflecting the light incident
107
on the retina that was not absorbed, back through the photoreceptors, again resulting in
an increased chance of photon capture (estimated to increase photon capture by 22% in
lanternfishes, Temple, 2011). Out of the 53 species analysed in this study, 37 species
possessed a tapetum lucidum. As found in several other studies (O’Day & Fernandez,
1976; Somiya, 1980; Nicol, 1989), the tapetum in lanternfishes is choroidal in contrast
to the more common tapetum located within the retinal pigment epithelium (RPE
tapetum, Nicol, 1989; Nicol et al., 1973). Interestingly, Locket (1977) described the
presence of a RPE tapetum in myctophids giving the example of Diaphus holti.
However, after examining freshly fixed specimens of D. holti, we found this tapetum to
be also choroidal. Differentiation between the two tapetal types could be difficult in
suboptimally-fixed tissue, especially given the lack of pigment granules within the RPE
in myctophids (with the exception of the fundal pigmentation). However, the clear
presence of Bruch’s membrane between the tapetum and the retinal pigment epithelial
cells clearly indicates a choroidal tapetum (Figure 3.14). The chemical composition of
the tapetum was not investigated in this study but another study has identified the
tapetal stacks to contain guanine (Nicol, 1989).
It is not uncommon to find both a tapetum lucidum and an aphakic gap in the
eyes of deep-sea fishes. In their study of Howella sp (Teleostei), Best and Nicol (1978)
described the presence of a nasal aphakic gap, which would allow more light coming
from the frontal visual field to be focussed on the temporal part of the retina, where a
tapetum lucidum was also present. In this species, sensitivity is enhanced by the aphakic
gap increasing the relative retinal illumination and by the tapetum lucidum increasing
the chance of photon capture. In our study, a relationship between the location of the
aphakic gap and tapetal coverage could also be identified. Bolinichthys longipes for
example, possesses an aphakic gap located naso-ventrally (Figure 3.15), which would
increase both the amount of light and the size of the visual field subtended by the dorso-
temporal part of the retina, where a tapetum is located (Figure 3.13B). The presence and
location of these two features in B. longipes suggests that it is important for this species
to closely survey the area directly below and in front for bioluminescent signals.
3.5.3. Interspecific variation in photoreceptor size

Every lanternfish species analysed in this study possessed a pure-rod retina.
Although cone photoreceptors have been indentified in larval stages (Bozzano et al.,
2007), our study confirmed their absence in adult myctophids. The contribution of rod
108
outer segments to the entire retinal thickness ranged from 13 to 57 % within the central
part of the retina in this study. As pointed out by Lockett (1970, 1977), having the rod
outer segments contributing a large proportion of the thickness of the retina is quite
common in deep-sea fishes compared to their shallow water counterparts. However,
although outer segment length can be considerable in lanternfishes, the total length of
the rod photoreceptor is not exceptionally long in comparison to other deep-sea teleosts,
which can often reach 170 μm in Sternoptix sp (Nicol, 1989) and even 600 μm in
Diretmus argentus (Munk, 1980).
Rod inner/outer segment diameter in myctophids can be extremely small

(<1 μm, Myctophum brachygnathum, Figure 3.8) making them one of the smallest
photoreceptors found in both vertebrates and invertebrates (at least insects, Land and
Nilsson, 2002). This small rod diameter also equates to a very high photoreceptor
density (up to 120 x 104 mm-2) in the central part of the retina. Very high rod
photoreceptor densities usually mean a high level of summation between photoreceptors
and ganglion cells resulting in high sensitivity. Peak densities of ganglion cells was not
investigated in this study, although if one considers the highest recorded density of
ganglion cells found in the literature for a lanternfish (7.4 x 103 mm-2, Lampanyctus
ater, Wagner et al., 1998) and the highest photoreceptor density reported in this study
(119 x 104 mm-2, Symbolophorus evermanni in central retina, Chapter 4), then
summation/convergence ratio could potentially be as high as 160 photoreceptors for one
ganglion cell. A similar level is predicted for many of the species examined here, where
most species possessed a relatively thick outer nuclear layer (high numbers of rod
nuclei) and a thin inner nuclear layer (low number of bipolar cells). However, one may
expect intraretinal variation since we restricted our analyses to central retina and retinal
thickness varies markedly not only in myctophids i.e. Benthosema suborbitale (Figure
3.4A) but also in other deep-sea species (i.e. Scopelarchus sagax, Locket, 1971;
Diretmus argenteus, Munk, 1980).
3.5.4. Novel retinal specialisations and their putative function

3.5.4.1. The fundal pigmentation
In lanternfishes, the retinal pigment epithelium (RPE) is extremely thin and
depleted of melanin, except in the central part of the retina of most species where the
RPE cells are modified and contain a series of localised aggregations of melanosomes
that form a radial arrangement of triangular-shaped wedges interleaved between the
109
densely-packed rod outer segments either surrounding the optic nerve head or within
discrete patches of the retinal periphery (Figures 3.9 and 3.10). To our knowledge, this
is the first description of such a localised fundal pigmentation. O’Day and Fernandez
(1976) made a brief mention about the occurrence of a discontinuous pigmentation in
the central fundus surrounding the optic nerve head in their description of the visual
system of the myctophid Stenobrachius leucopsarus, but offered no further description.
The function of this unique pigmentation remains uncertain but we would like to
present some putative suggestions based on our knowledge of its development and the
movement patterns of myctophids as they perform circadian vertical migration up and
down the water column in search of food. Being composed of dense aggregations of
melanin granules, the function of the fundal pigmentation is most likely to absorb light
within this localised retinal region, thereby reducing sensitivity, although pigmentation
is not continuous, where light not absorbed in some regions of the retina will be
reflected by the underlying choroidal tapetum lucidum to provide limited enhancement
of sensitivity or more complete absorption depending on the direction of the incident
light on this specular reflector. However, why is it important to absorb light in these
retinal regions and/or reduce the amount of reflected light leaving the eye?
The optic nerve, which does not possess any pigment (visual or melanin), is
composed of the retinal ganglion cell axons. Since retinal axons in lanternfishes are
mostly surrounded by myelin, a material possessing a large refractive index due to its
high percentage of lipids (Van der Zee, 1992), the optic nerve head (central fundus)
could contribute to a point source of reflectance, that might appear different to a tapetal
reflex, although it is difficult to reconcile how this signal may be detected. Similarly,
any light entering the head via the pineal window (present in myctophids, McNulty and
Nafpaktitis, 1976; McNulty and Nafpaktitis, 1977) may exit the eye at the optic nerve
head and also alert potential predators (N. Michiels, pers. comm). The presence of the
fundal pigmentation might reduce both of these effects by absorbing the light scattered
from the optic nerve head on the optical axis. However, this does not explain why some
species do not possess this pigmentation, why fundal pigmentation has not been
observed in any other groups of deep-sea fishes and the fact that pigmentation is not
always restricted to a concentric arrangement surrounding the optic nerve i.e.
Diogenichthys atlanticus (Figure 3.9). Therefore, we consider that this pigmentation is a
110
vestige of a visual specialisation important in larval stages of development but may not
have any function in adults.
In three specimens of Nannobrachium cf. nigrum of different standard length

and therefore of different age (Figure 3.11), the fundal pigmentation remains a constant
size. This indicates that as the eye (and retina) continues to grow throughout life (as has
been found to occur for presumably all teleosts, Fernald, 1985), the size of the visual
field subtended by this pigmentation will be different and increase in size as the fish
continues to grow. Myctophids are well known for having very different day and night
depth distributions depending on their life stage (Clarke, 1973; Hulley, 1984; Hulley,
1994). For example, while adults are found in the mesopelagic zone during the day
(Karnella, 1987), all myctophid larvae are found exclusively in the shallower epipelagic
zone (upper 200m, Sassa et al., 2002; Sassa et al., 2004). Myctophid larvae are known
to have fully-functional eyes with a well differentiated retina composed mainly of rods
but also of cones, and a retinal pigment epithelium (RPE) containing dense (continuous)
aggregations of melanin granules (Sabates et al., 2003; Bozzano et al., 2007). Since
lanternfish larvae are visual predators feeding mostly during daytime in surface waters
(Sassa et al., 2002), a melanin-rich RPE would protect the rod photoreceptors from
excessive (over) stimulation across the whole retina. The presence of a RPE containing
absorbing pigment would also enhance resolution, especially during the transient period
and when cones occupied a part of the photoreceptor array.
As the eye grows, the size of the continuous fundal pigmentation, which may
have filled the entire retinal area in the larval stages, becomes increasingly smaller as
new retinal tissue is added to the periphery. Since newly-differentiated rods are added
throughout the retina (Johns and Fernald, 1981), this would explain the regular
interruptions in the fundal pigmentation, where new rods have penetrated the melanin-
rich RPE cells, giving the peculiar appearance of an array of triangular wedges of
melanosomes (Figure 3.10). The presence of a nasal fundal pigmentation in
Diogenichthys atlanticus (Figure 3.9) could indicate that, in contrast to many of the
other species with fundal pigmentation surrounding the optic nerve head, this species
may undergo asymmetric retinal growth as has been shown to occur in several species
of shallow water teleosts (Easter, 1992; Cameron, 1995; Kwan et al., 1996). Several
other species of deep-sea fishes lack melanin pigment in the RPE, a sign of high visual
sensitivity (Somiya, 1980) but little is known of their early life history stages (O'Day
111
and Fernandez, 1976). Further analysis of this fundal pigmentation at all life stages will
be necessary to better understand its function.
3.5.4.2. The microtubule-like structures

The presence of microtubular-like structures (MLS) in the outer part of the inner
nuclear layer have been reported and/or described in a small number of teleost species
i.e. shallow water species such as the common snook, Centropomus undecimalis and the
swordspine snook, Centropomus ensiferus (Villegas & Villegas 1963), the carp,
Cyprinus carpio (Witkovsky and Dowling, 1969), the rainbow trout, Salmo gairdneri
(Kurz-Isler and Wolburg, 1978; Wolburg and Kurz-Isler, 1977), the zebrafish, Danio
rerio (Tarboush et al., 2012), and one species of deep-sea pearleye, Scopelarchus
guntheri (Locket, 1971). The aforementioned studies and the findings described here all
confirm that the MLS are different from the classic microtubules occurring in all cell
types of the retina. Contrary to classic microtubules, the MLS are always extracellular,
in close association to membranes, are larger in size (with a diameter >20 nm versus 18-
20 nm for microtubules, Wolburg and Kurz-Isler, 1977) and are more closely-packed
than microtubules. MLS are mostly arranged in bundles, that often appear interlaced
with one another i.e. are oriented differently (Locket, 1971) although in zebrafish they
are also organised into linear arrays (Tarboush et al., 2012), and range in size from
<1 µm (Villegas and Villegas, 1963; this study) to several micrometers thick (Locket,
1971; Stell, 1972; this study).
In lanternfishes, the MLS seem to be systematically organised into bundles,

which vary greatly in size. In some species, the MLS are so well-developed, they can
easily be distinguished by light microscopy, typically surrounding the horizontal cells
within the inner nuclear layer. Since species with very well-developed MLS also appear
to possess higher densities of horizontal cells, there may be a possible relationship
between MLS and the presence of horizontal cell soma and their processes. This finding
is in agreement with previous studies that all found the MLS to be associated with the
horizontal cells and/or horizontal cell processes (Villegas and Villegas, 1963; Locket,
1971; Stell, 1972; Tarboush et al., 2012). Locket (1971) even suggests the MLS
originate from the horizontal cells in Scopelarchus guntheri.
The function of the MLS is still unknown. In terms of optical properties, it is

very unlikely that the MLS would have any structural/spectral transmission or spectral
112
reflexion properties due to their random organisation and small diameter, which is much
less than the wavelength of light (N. Roberts and T. Jordan, personal communication).
Few studies have suggested a function for these extracellular inclusions but Villegas
and Villegas (1963) proposed that the MLS may form a bridge-like structural network
that interconnects the horizontal cells. In the rainbow trout, Salmo gairdneri, Kurz-Isler
and Wolburg (1978) found some morphological evidences for an interconnection of
different cell types (between bipolar cells and horizontal cells) via the MLS, reinforcing
the idea of a bridge-like structure. Stell (1972) suggested that the MLS may help to
maintain a certain volume of extracellular fluid within the retina and even suggested the
network acts as a means of ionic composition control around the horizontal cells. The
MLS start developing at an early stage of eye development in the rainbow trout (at
around 30 - 40 days post hatching), which also coincides with the start of their active
feeding phase, where there is an increase in visual acuity (Kurz-Isler and Wolburg,
1978). The marked interspecific differences in the hypertrophy of the MLS in
lanternfishes described here may be related to inherent differences in visual functions
and/or the different ages of the specimens examined.
3.5.5. Interspecific variability and the influence of phylogeny

Results from this study show a great interspecific variability in the visual system
of lanternfishes at many levels (iris colour, aphakic gap, tapetum lucidum, fundal
pigmentation, microtubule-like structures, retinal thickness and photoreceptor size).
After examining the retinal structure of several taxonomic and ecological groups of
deep-sea fishes, Wagner (1990) concluded that the retinal morphology of a species was
more a consequence of ecological and behavioural factors than of phylogeny. To
explore this hypothesis, we mapped these different visual characteristics onto an
existing phylogenetic tree of myctophids (Figure 3.15). While our results appear to
agree with Wagner’s conclusion for some of the lanternfish specialisations examined
here (aphakic gap, tapetum lucidum), it is not always the case and some specialisations
appear to closely follow the phylogeny (fundal pigmentation, MLS). de Busserolles et
al. (2013) investigated variations in eye size in a large number of lanternfishes with
respect to phylogenetic and ecological factors (depth distribution, luminous organ
patterns, sexual dimorphism). The results of their study did not reveal any relationships
between eye size and any of these ecological variables, where the size of the eye in
different tribes of lanternfishes was strongly influenced by phylogenetic relatedness.
Similarly, Turner et al. (2009) concluded than even though visual pigments in
113
lanternfishes were well-adapted to their environment, the variation of the maximum

absorbance (λmax) of the visual pigments in 58 species of myctophids was more the
consequence of shared evolutionary history than environmental constraints. The fact
that some of the specialisations observed in myctophids were found well partitioned
against the phylogeny, like the MLS and the fundal pigmentation (Figure 3.15), also
indicates a strong phylogenetic influence.
However, other adaptations like the position of the aphakic gap and the
presence/absence/coverage of the tapetum lucidum appear randomly distributed against
the phylogeny, suggesting the influence of other factors. The effect of the environment
(illumination) and specific visual tasks have previously been related to the variation
seen in such structures. For example, variation in the size of the circumlental aphakic
gap within the genus Gonostoma was found to be related to an increase in depth
distribution (Marshall, 1954). Similarly, regionalisation of the tapetum lucidum in the
eye of several species has been related to the amount of light irradiance present in the
environment, either to enhance the sensitivity in a specific part of the eye (Nicol, 1989)
or to potentially reduce conspicuousness in relation to predators (Shelton et al., 1992).
Moreover, variation in tapetum colour also seems to be related to the

environmental conditions and the specific needs of a species. For example, species
living in fresh water, which often appears brownish in colour due to the large amount of
particulate and plant matter, often possess a yellow to red tapetum, which transmits light
of longer wavelengths (Nicol et al., 1973; Wang et al., 1980). In the mesopelagic zone,
where most of the light present (downwelling sunlight and bioluminescence) falls
within the short wavelength range of the visible spectrum, tapetum colour tends to be in
the blue-green range, optimally matching the peak spectral absorption of the visual
pigments of most deep-sea fishes (~480 nm, Partridge et al., 1988; Partridge et al.,
1992; Douglas and Partridge, 1997). In lanternfishes, the dual tapetum colour seen in
Notoscopelus resplendens was found to match the spectral distribution of the ambient
light environment at the depth at which this species is usually found (Douglas et al.,
1998). It was also proposed that the reflectance characteristics of the tapetum lucidum
may contribute to the complex array of camouflage mechanisms seen in lanternfishes
(Douglas et al., 1998). In fact, a blue tapetum, in addition to increasing light capture in
the right spectral zone, may provide a better camouflage than a silvery tapetum by
blending the eye-shine into the generally blue background. A similar function may be
114
attributed to the interspecific variation in iris colour also observed within the family.
Body colour in lanternfishes varies from dark brown to silvery with some species also
possessing metallic green or blue scales (Hulley, 1990; Paxton and Hulley, 1999). Iris
colour may, in this case, match the body colour of the fish in an attempt to be less
conspicuous (Walls, 1942).
3.5.6. Conclusions
The results from this study confirm that Myctophids possess what may be
considered typical visual adaptations for dim-light conditions in order to enhance the
sensitivity of the eye, including enlarged eyes, an aphakic gap, a tapetum lucidum, a
pure rod retina with a high density of long photoreceptors. Moreover, new
specialisations such as the fundal pigmentation and the microtubular-like structure
emphasize the fact that we still know little about these important group of fishes, which
occupy such a large part of the deep ocean. Although we propose some putative
functions to both specialisations, additional analyses, especially at several different life
stages of development, will be necessary to confirm our hypotheses. There is an almost
unprecedented interspecific variability in the visual system of myctophids. Our results
indicate the influence of both phylogeny and ecology on the evolution of different
visual specialisations. The great interspecific variability observed in some of the ocular
specialisations (i.e. aphakic gap, tapetum lucidum) most likely matches the ecological
and behavioural variability seen within this large and diverse group of lanternfishes (i.e.
depth distribution, migration pattern, luminous organs pattern). However, a more
detailed comparative analysis including phylogenetic information will have to be
conducted to fully understand the factors driving the evolution of the visual system for
overcoming such a challenging light environment.
3.6. Acknowledgements
Brigitte Guillaumont (Ifremer), for providing additional samples. We gratefully
acknowledge John Paxton (Australian Museum) for providing most of the fish
identification and, Caroline Kerr and Alan Goldizen for their precious help during field
115
trips. We also wish to thank Michael Archer for his precious help with the preparation
of light and electron microscopy sections and Prof. Julian Partridge (University of
Bristol), Nick Roberts and Tom Jordan for helpful discussions regarding the
interpretation of the structure and function of the MLS. Finally, we acknowledge the
facilities, and the scientific and technical assistance of the Australian Microscopy &
Microanalysis Research Facility at the Centre for Microscopy, Characterisation &
Analysis, The University of Western Australia, a facility funded by the University, State
and Commonwealth Governments. This work was funded by the Australian Research
Council (Discovery grant to SPC and the Deep-Australia Linkage Grant to NJM and
SPC) and the West Australian State Government. FdB was supported by Scholarship for
International Research Fees (SIRF) and University International Stipend (UIS) at the
University of Western Australia.
3.7. References
Arnott HJ, Maciolek NJ, Nicol JAC. 1970. Retinal tapetum lucidum: a novel reflecting
system in the eye of teleosts. Science 169(3944):478-480.
Best A, Nicol J. 1978. Notes on the retina and tapetum lucidum of Howella (Teleostei:
Cheilodipteridae). Journal of the Marine Biological Association of the United
Kingdom 58(03):735-738.
Cameron DA. 1995. Asymmetric retinal growth in the adult teleost green sunfish
(Lepomis cyanellus). Visual Neuroscience 12(01):95-102.
Collin SP, Collin HB, Ali MA. 1996. Fine structure of the retina and pigment
epithelium in the creek chub, Semotilus atromaculatus (Cyprinidae, Teleostei).
Histology and Histopathology 11(1): 55-69.
116
Crescitelli F. 1990. Adaptations of visual pigments to the photic environment of the

deep sea. Journal of Experimental Zoology 256(S5):66-75.
variability in deep-sea lanternfishes (Myctophidae): an ecological and
Kingdom 69(2):409-435.
17(4):597-636.
Easter SS, Jr. 1992. Retinal growth in foveated teleosts: nasotemporal asymmetry keeps
the fovea in temporal retina. The Journal of Neuroscience 12(6):2381-2392.
41(1):59-70.
Fernald RD. 1985. Growth of the teleost eye: novel solutions to complex constraints.
Fröhlich E, Negishi K, Wagner HJ. 1995. Patterns of rod proliferation in deep-sea fish
retinae. Vision Research 35(13):1799-1811.
Fröhlich E, Wagner HJ. 1998. Development of multibank rod retinae in deep-sea fishes.
Visual Neuroscience 15(3):477-483.

945.
117
J, Tortonese E, editors. Fishes of the north-eastern Atlantic and the
Mediterranean. Unesco. p 429-483.
178.
Hulley PA. 1994. Lanternfishes. In: Paxton JR, Eschmeyer WN, editors. Encyclopedia
of fishes. UNSW Press. p 127-128.
press.
Johns PR, Fernald RD. 1981. Genesis of rods in teleost fish retina. Nature
293(5828):141-142.
Kurz-Isler G, Wolburg H. 1978. Extracellular microtubule-like structures in the retina of

the rainbow trout. Cell and Tissue Research 191(1):15-26.
Kwan JW, Lee MJ, Mack AF, Chiu JF, Fernald RD. 1996. Nonuniform distribution of
cell proliferation in the adult teleost retina. Brain Research 712(1):40-44.
Land MF, Nilsson D-E. 2002. Animals eyes. Oxford University Press. 221 p.
247(5437):155-157.
Locket NA. 1970. Deep-sea fish retinas. British Medical Bulletin 26(2):107-111.
Locket NA. 1971. Retinal anatomy in some scopelarchid deep-sea fishes. Proceedings
of the Royal Society of London Series B, Biological Sciences 178(1051):161-
184.
McNulty JA, Nafpaktitis BG. 1976. The structure and development of the pineal
complex in the lanternfish Triphoturus mexicanus (family Myctophidae).
Journal of Morphology 150(2):579-605.
McNulty JA, Nafpaktitis BG. 1977. Morphology of pineal complex in 7 species of

lanternfishes (pisces Myctophidae). American Journal of Anatomy 150(4):509-
529.
118

myctophid fishes, Diaphus theta, Stenobrachius leucopsarus and S-nannochir,

Bulletin, Los Angeles County Natural History Museum 7: 1-145.

photophore patterns in lanternfishes. Fishery Bulletin 70: 541–564.
marine teleosts: the Myctophidae, a case study. Fishery Bulletin 72: 391-413.
Munk O, Frederiksen RD. 1974. On the function of aphakic apertures in teleosts.

Videnskabelige meddelelser fra Dansk naturhistorisk forening 137:65-94.
Munk O. 1980. Retinal tapetum of the deep-sea teleost Diretmus argenteus.

Videnskabelige Meddelelser fra Dansk Naturhistorisk Forening 142:179-186.
Nicol JAC, Arnott HJ, Best ACG. 1973. Tapeta lucida in bony fishes (Actinopterygii): a
survey. Canadian Journal of Zoology 51(1):69-81.
Nicol JAC. 1989. The eyes of fishes. Oxford University Press. 308 p.

Partridge JC, Archer SN, Lythgoe JN. 1988. Visual pigments in the individual rods of
deep-sea fishes. Journal of Comparative Physiology A, Sensory Neural and
Behavioral Physiology 162(4):543-550.
Partridge JC, Shand J, Archer SN, Lythgoe JN, Vangroningenluyben W. 1989.

Interspecific variation in the visual pigments of deep-sea fishes. Journal of
Comparative Physiology A, Sensory Neural and Behavioral Physiology
164(4):513-529.
Kingdom 72(1):113-130.
119

13:1-81.
Ontogeny and Systematic of Fishes. American Society of Ichthyologist and
Paxton JR, Hulley PA. 1999. Families Neoscopelidae, Myctophidae. In: Carpenter KE,
Niem VH, editors. The living marine resourses of the Western Central Pacific.
Rome: FAO Species Identification Guide for Fisheries Purposes. p 1955-1965.
Sabates A, Bozzano A, Vallvey I. 2003. Feeding pattern and the visual light
environment in myctophid fish larvae. Journal of Fish Biology 63(6):1476-1490.
Sassa C, Moser HG, Kawaguchi K. 2002. Horizontal and vertical distribution patterns
of larval myctophid fishes in the Kuroshio Current region. Fisheries
Sassa C, Kawaguchi K, Hirota Y, Ishida M. 2004. Distribution patterns of larval

myctophid fish assemblages in the subtropical–tropical waters of the western
North Pacific. Fisheries Oceanography 13(4):267-282.
Shelton PMJ, Gaten E, Herring PJ. 1992. Adaptations of tapeta in the eyes of
mesopelagic decapod shrimps to match the oceanic irradiance distribution.
Journal of the Marine Biological Association of the United Kingdom 72(1):77-
88.
215(1201):481-489.
Stell WK. 1972. The morphological organization of the vertebrate retina. In: Fuortes
MGF, editor. Handbook of sensory physiology VII/2 Physiology of
photoreceptor organs. Springer-Verlag. p 111-213.
Stiassny MLJ. 1996. Basal ctenosquamate relationships and the interrelationships of the
myctophiform (scopelomorph) fishes. In: Stiassny MLJ, Parenti LR, Johnson
GD, editors. Interrelationships of fishes. Academic Press. p 405-426.
Tarboush R, Chapman GB, Connaughton VP. 2012. Ultrastructure of the distal retina of
the adult zebrafish, Danio rerio. Tissue and Cell 44(4):264-279.
120
Temple SE. 2011. Why different regions of the retina have different spectral
sensitivities: A review of mechanisms and functional significance of intraretinal
variability in spectral sensitivity in vertebrates. Visual Neuroscience 28(04):281-
293.
56(6):1003-1017.
Van der Zee P. 1992. Measurement and modelling of the optical properties of human
tissue in the near infrared. PhD thesis University College London.
Villegas GM, Villegas R. 1963. Neuron-glia relationship in the bipolar cell layer of the
fish retina. Journal of Ultrastructure Research 8:89-106.

Wagner H-J. 1990. Retinal structure of fishes. In: Douglas RH, Djamgoz MBA, editors.
The visual system of fish. Chapman and Hall. p 109-157
17(4):637-685.
Wagner H-J. 2001. Sensory brain areas in mesopelagic fishes. Brain, Behavior and
Evolution 57(3):117-133.
Wang RT, Nicol JAC, Thurston EL, McCants M. 1980. Studies on the eyes of bigeyes
(Teleostei Priacanthidae) with special reference to the tapetum lucidum.
210(1181):499-512.

Widder EA. 2010. Bioluminescence in the ocean: Origins of Biological, Chemical, and
Witkovsky P, Dowling J. 1969. Synaptic relationships in the plexiform layers of carp

retina. Zeitschrift für Zellforschung und mikroskopische Anatomie 100(1):60-
82.
121
Wolburg H, Kurz-Isler G. 1977. Microtubules and extracellular microtubule-like

structures in the retina of the rainbow trout. Cell and Tissue Research
177(1):127-140.
Zoology 607:1-69.
122
Chapter 4 Photoreceptors and optical sensitivity
CHAPTER 4
The influence of photoreceptor size and distribution on

optical sensitivity in the eyes of lanternfishes
(Myctophidae)
Fanny de Busserolles1, John L. Fitzpatrick2,3, N. Justin Marshall4, Shaun P. Collin1.
Australia, Crawley, WA, Australia.
2. Centre for Evolutionary Biology, School of Animal Biology, The University of Western Australia,
Crawley, WA, Australia.
3. Computational and Evolutionary Biology, Faculty of Life Sciences, University of Manchester,
Manchester M13 9PT, UK
Australia.
123
4.1. Abstract
The mesopelagic zone of the deep-sea (200-1000 m) is characterised by
exponentially diminishing levels of downwelling sunlight and by the predominance of
bioluminescence emission produced by the animals themselves. The ability of
mesopelagic organisms to detect and behaviourally react to downwelling sunlight
and/or bioluminescent emissions will depend on the visual task and ultimately on the
eyes and their capacity for detecting low levels of illumination and intermittent point
sources of bioluminescent light. In this study, we investigate the diversity of the
myctophid visual system, concentrating specifically on the photoreceptor cells by
examining their size, arrangement, topographic distribution and contribution to optical
sensitivity in 53 different species from 18 genera. We also examine the influence(s) of
both phylogeny and ecology on these photoreceptor variables using phylogenetic
comparative analyses in order to understand the constraints placed on the visual systems
of this large group of mesopelagic fishes, at least at the first stage of retinal processing.
Results show a great diversity in the visual system of the Myctophidae at the level of
the photoreceptors. Photoreceptor distribution reveals clear interspecific differences in
visual specialisations (areas of high rod photoreceptor density), indicating potential
interspecific differences in interactions with prey, predators and/or mates. A great
diversity in photoreceptor design (length and diameter) and density is also present.
Overall, the myctophid eye is very sensitive compared to other teleosts and each species
seems to be specialised for the detection of a specific signal (downwelling light or
bioluminescence), potentially reflecting different visual demands for survival. Results
from the phylogenetic comparative analyses highlights several relationships between
photoreceptor characteristics and the ecological variables tested (depth distribution and
luminous tissue patterns). Depth distribution at night was found to be a significant
factor in most of the models tested, indicating that vision at night is of great importance
in lanternfishes.
4.2. Introduction
As sunlight penetrates the water column of the ocean, it is absorbed and
scattered leading to a rapid decrease in intensity with depth, and in the mesopelagic
zone (200 to 1000 m) very low levels of residual downwelling sunlight remain.
Residual light in the mesopelagic zone is of constant colour and direction but of
exponentially diminishing intensity, creating an extended visual scene (vertically and
124
horizontally), where silhouettes of organisms can be distinguished, against a light

background, when viewed from below. The intensity gradient of downwelling sunlight
might be also used to maintain a specific depth during the day (Denton, 1990), be used
for orientation for vertical migration (Frank and Widder, 1997), to camouflage the
outline of the body by counter-illumination (Young et al., 1979; Claes et al., 2010),
and/or to detect the presence of other animals above, since they would cast a detectable
shadow. However, most mesopelagic organisms also produce and/or emit
bioluminescent signals, which can be of different spectral composition, intensity and
duration (Widder, 2010) and are used in the mediation of a number of important
behaviours including the detection of prey and predator, the attraction and/or avoidance
of other organisms and the communication with conspecifics (Herring, 2002; Haddock
et al., 2010). Therefore, the ability of mesopelagic organisms to detect and
behaviourally react to bioluminescent emissions and/or downwelling sunlight will
depend on the visual task and ultimately on the eyes and their capacity for detecting low
levels of illumination and intermittent point sources of bioluminescent light.
Due to the low light environment and the predominance of small bioluminescent
flashes, the eyes of mesopelagic fishes are more sensitive than their shallow water
counterparts, relying less on acuity to resolve fine visual detail. Specialisations to
increase the sensitivity of the eye include extending the visual field to allow the capture
of as many photons as possible. This can be achieved with the help of tubular-shaped
eyes, which are directed upwards to maximise the capture of the downwelling sunlight
(Munk, 1966; Locket, 1977; Collin et al., 1997, 1998), with an increased size of both
the eye and the pupillary aperture (Marshall, 1954) and/or the presence of an aphakic
gap, a region of the pupillary aperture that allows light from a specific part of the visual
field to reach the retina without necessarily passing through the lens (Munk, 1966;
Munk and Frederiksen, 1974). Visual adaptations for increasing sensitivity at the level
of the retina include 1. the tapetum lucidum, a mirror-like structure which sits at the
back of the eye and increases light absorption (Arnott et al., 1970; Somiya, 1980), 2.
high numbers of long rod photoreceptors, which are specialised for scotopic vision
(Locket, 1977) often arranged in multibanks (Munk, 1966; Locket, 1977) which, in
addition to enhancing the chance of photon capture, may also allow colour vision in
single pigment species or at least provide hue discrimination to potentially break
camouflage by counterillumination (Denton and Locket, 1989) and 3. most mesopelagic
fishes possess a single visual pigment within their photoreceptors that closely matches
125
both the predominant wavelengths of the downwelling sunlight and the spectral
emission of most bioluminescence (Partridge et al., 1989; Crescitelli, 1990; Douglas
and Partridge, 1997).
The lanternfish family (Myctophidae) is one of the most abundant groups of

mesopelagic fishes in the world (Hulley, 1981) with more than 250 representative
species from 33 genera (Hulley and Paxton, in press). They are present worldwide and
live at the surface down to depths exceeding 1000 m, thereby playing a major role in
oceanic ecosystems by transferring energy to deeper levels through their daily vertical
migrations. These vertical migrations are for the purposes of feeding but there is large
inter- and intra-specific variability (Watanabe et al., 1999) depending on life stage and
season (Karnella, 1987). Like most mesopelagic organisms, lanternfishes are
bioluminescent and produce their own light through a luciferin-luciferase reaction (Tsuji
and Haneda, 1971; Haygood et al., 1994), which takes place within two kinds of
bioluminescent structures; the photophores found on the ventral and ventrolateral parts
of the body and the luminous organs and tissue patches present on the head, body and/or
tail. While the photophores are thought to play a major role in camouflage by counter-
illuminating the underside of the body (Case et al., 1977), luminous organs are thought
to play several different roles in intra- and interspecific communication, distraction or
illumination (Edwards and Herring, 1977). Luminous tissue patterns are highly variable
between species and, in some cases, are sexually dimorphic, indicating that the visual
system must play an important role in finding reproductive partners. Overall, the
abundance of lanternfishes and the high level of variability in their depth distribution,
vertical migration patterns and luminous tissue dimorphisms, make this group an
important model for visual adaptation studies. However, very little information is
available about their visual capabilities with respect to their photic environment.
Like other mesopelagic organisms, lanternfishes possess well-developed eyes

(Chapter 3), which appear to be adapted to enhance sensitivity due to specialisations
such as an aphakic gap (Lawry, 1974; Chapter 3), a tapetum lucidum (O'Day and
Fernandez, 1976; Chapter 3), a pure rod retina (Vilter, 1951; Pankhurst, 1987; O'Day
and Fernandez, 1976; Chapter 3), a high density of photoreceptors (Vilter, 1951;
Pankhurst, 1987; O'Day and Fernandez, 1976), and visual pigments tuned to their
ambient light environment of downwelling sunlight and bioluminescent flashes (Turner
et al., 2009; Douglas and Partridge, 1997; Douglas et al., 1998; Hasegawa et al., 2008;
126
Partridge et al., 1992). Recently, de Busserolles et al. (2013) highlighted the great
interspecific variability in eye size within the Myctophidae by investigating the
relationship between eye size (corrected for body size) and ecological variables (i.e.
depth distribution and luminous tissue pattern). They hypothesized that species living at
greater depths and/or relying less on bioluminescent emissions for vision will
consequently have smaller eyes. However, they did not find any relationship between
eye size and any of the ecological variables tested but instead indentified the presence of
a strong phylogenetic signal (de Busserolles et al., 2013; Chapter 2). It is therefore
apparent that the visual capabilities of a species may not be solely assessed by the size
of the eye and that a number of other physical factors (i.e. biophysical properties of the
photoreceptors) in addition to visual stimuli need to be considered in order to assess the
relationship between vision and ecology. In this study, we investigate the diversity of
the myctophid visual system, concentrating specifically on the photoreceptor cells by
examining their size, arrangement, topographic distribution and contribution to optical
sensitivity in 53 different species from 18 genera. We also examine the influence(s) of
both phylogeny and ecology on these photoreceptor variables in order to understand the
constraints placed on the visual systems of this large group of mesopelagic fishes, at
least at the first stage of retinal processing.

4.3.1. Ethics statement
Australian waters, the Western Mediterranean Sea and the Bay of Biscay were acquired
through collaborators (de Busserolles et al., 2013; Chapter 2) and did not require UWA
collection or animal ethics permits.
127
4.3.2. Samples
The eyes of 53 different species of lanternfishes from 18 genera were analysed
in this study. While most of the samples are registered as voucher specimens at the
Australian Museum in Sydney, Australia, further taxonomic analyses need to be carried
out for five of our study species to positively confirm identification (Lampanyctus
vadulus, Myctophum spinosum, Nannobrachium cf. nigrum, Symbolophorus cf. boops,
Triphoturus oculeus, de Busserolles et al. 2013; Chapter 2).
Observations were performed on board and on fresh specimens when possible.

For each individual, the standard length and rostro-caudal eye diameter were measured
with digital callipers to an accuracy of 0.1 mm prior to dissection. The position of the
aphakic gap (dorsal, nasal, ventral, temporal), when present, was noted and
photographed onboard using a Canon digital camera mounted on an Olympus
stereomicroscope (Model SZX10). Eyes were then enucleated, the cornea and lens
dissected free of the posterior chamber and the lens diameter measured using digital
callipers (to 0.1 mm). The appearance and colour of the tapetum lucidum was noted and
the fundus of the eye was also photographed. The eyes, lens and cornea were all fixed in
4% paraformaldehyde in 0.1M phosphate buffer (PFA, pH 7.4) or Karnovsky’s fixative
(2% paraformaldehyde, 2.5 glutaraldehyde in 0.1 M cacodylate buffer, pH 7.4) for at
least 1h and then stored in 0.1 M phosphate buffer. When specimens and their eyes were
very small, the entire individual was fixed (as above) and the eyes dissected (post
fixation) back in the mainland laboratory. When samples were acquired from
collaborators, observations, measurements and dissections were all performed on
postfixed tissue, with the eyes and bodies preserved in 5% buffered formalin or
Karnovsky’s fixative.
4.3.4. Preparation of retinal wholemounts

Several eyes fixed in 4% PFA or Karnovsky’s fixative were dissected in order to
analyse the topographic distribution of photoreceptors across the retina. Wholemounts
of the retina were prepared according to standard protocols (Stone, 1981; Coimbra et al.,
2006; Ullmann et al., 2011). Radial cuts were performed in order to flatten the eye and
subsequently the retina in toto onto a glass slide, where the orientation was confirmed
by making a small additional cut in the nasal or dorso-nasal part of the eye. The sclera,
retinal pigment epithelium and tapetum (when present), were gently removed with the
help of No. 3 watchmaker’s forceps and a kolinsky hair paint brush. Each retinal
128
wholemount was then processed following the protocol of Curcio et al. (1987) to allow
better visualisation of the photoreceptors. Briefly, the retina was quickly rinsed in
distilled water and flat-mounted on a microscope slide, photoreceptor side facing up.
The retinal wholemount was initially mounted in 100% dimethyl sulfoxide (DMSO) for
24 h to clear the tissue and re-mounted in 100% glycerol for topographic analysis.
4.3.5. Stereological analysis and the construction of topographic maps

The topographic distribution of the photoreceptors was assessed using the
optical fractionator technique (West et al., 1991) modified by Coimbra et al. (2009,
2012). Briefly, the retinal wholemount was considered as a single section and
consequently, the thickness sampling fraction (tsf) was fixed at 1. The outline of the
retinal wholemount was then digitised using a x4 objective (numerical aperture 0.13)
mounted on a compound microscope (Olympus BX50) equipped with a motorized stage
(MAC5000, Ludl Electronics Products, USA), a digital video camera (MicroFIRE,
OPTRONICS), and a computer running Stereo Investigator software (Microbrightfield,
USA). Using a x100 oil immersion objective (numerical aperture 1.40), rod
photoreceptors were randomly and systematically counted using the parameters listed in
Table 4.1.
Photoreceptor counts were found very challenging in lanternfishes due to the

small size of their rod photoreceptors, which were often beyond the optical limits of the
light microscope. As a result, only five species were analysed in this study. The number
of individuals per species analysed ranged from one to three due to the availability of
samples and the quality of fixation, which was often outside of our control depending
on the delay in preserving specimens once they reached the deck and/or the depth of
capture. In order to maintain the same high level of sampling and the comparison of
small and large individuals, the grid size was modified to allow the sampling of around
200 sites per retina and to achieve an acceptable Schaeffer coefficient of error (CE). The
CE is a measure of the accuracy of the density estimates and is considered acceptable
below 0.1 (Glaser and Wilson, 1998; Slomianka and West, 2005). For one species,
Myctophum brachygnathum, photoreceptor counts was particularly challenging, even at
the highest magnification offered by the stereology setup, due to their extremely small
size (~ 1 μm, Figure 4.1B). Consequently, a larger grid was used for this species,
sampling only 100 sites per retina, but still achieving a CE <0.1, although sub-sampling
129
in the area of high cell density was carried out for two individuals in order to
compensate for the use of a larger counting frame.
Table 4.1. Summary of the stereological parameters used for the topographic analyses of the
photoreceptor cells in five lanternfish species.
Species Individual SL Eye ø Counting frame Grid Site

(mm) (mm) (μm x μm) (μm x μm) numbers
Bolinichthys longipes A 35.0 3.2 10 x10 320 x 320 200
B 36.8 3.4 10 x 10 340 x 340 204
C 45.4 4.4 10 x 10 370 x 370 199
Lampanyctus parvicauda A 54.2 3.3 10 x 10 270 x 270 198

B 60.3 3.7 10 x 10 300 x 300 211
C 65.2 4.5 10 x 10 350 x 350 209
Myctophum brachygnathum A 65.7 7.1 5x5 850 x 850 107

B 67.7 7.6 5x5 850 x 850 100
C 68.3 7.7 5x5 850 x 850 97
Diaphus barchycephalus A ? ? 15 x 15 270 x 270 205
Nannobrachium idostigma B 72.2 3.9 15 x 15 280 x 280 197
Figure 4.1. Wholemount view of the rod photoreceptors in Diaphus brachycephalus (A) and
Myctophum brachygnatum (B). Scale bar = 10 μm.
Topographic maps of photoreceptor density were constructed using the

statistical program R v.2.15.0 (R Foundation for Statistical Computing 2012) with the
results exported from the Stereo Investigator Software according to Garza Gisholt et al.
(in press). Garza Gisholt et al. (in press) proposed several smoothing models to
construct the iso-density maps and, for this study, we chose to use the Gaussian Kernel
Smoother from the Spatstat package (Baddeley and Turner, 2005). For each map, the
sigma value was adjusted to the distance between points (i.e. grid size).
130
4.3.6. Morphometric measurement of the photoreceptors

Outer and inner segment length and diameter were measured for a subset of the
rod photoreceptors from transverse sections of the retina using light and electron
microscopy. Samples fixed in Karnovsky’s fixative (2 % paraformaldehyde, 2.5 %
glutaraldehyde in 0.1 M sodium cacodylate buffer, pH 7.4) were preferentially used,
although when no samples preserved in Karnovsky’s fixative were available, samples
fixed in 4% paraformaldehyde in 0.1 M phosphate buffer or in 5 % buffered formalin
were used. One eye per species was analysed. Depending on the size of the samples, the
whole, half or a quadrant of the eye was processed. The tissue was postfixed for an hour
with 1 % osmium tetroxide in 0.15 M phosphate buffer, dehydrated through an alcohol
and propylene oxide series and infiltrated with procure/araldite (ProSciTech). For light
microscopy, semi-thin sections (1 μm) were cut with a glass knife using a LKB
Bromma Ultratome NOVA. Sections were stained with an aqueous mixture of 0.5 %
Toluidine Blue and 0.5 % borax, viewed with an Olympus BX50 compound light
microscope and photographed using an Olympus DP70 digital camera. For transmission
electron microscopy, thin sections (110nm) were cut using a diamond knife, mounted
on a 200 mesh copper grid and stained with Reynold’s lead citrate. Examination of the
sections was done using a JEOL 2100 transmission electron microscope operating at
120 kV and images taken using an 11 megapixel Gatan Orius digital camera.
All measurements were done from digital images using ImageJ 1.45 (National
Institutes of Health, USA). To allow comparison between species, all measurements
were taken in the central part of the retina, except for Benthosema suborbitale. In the
case of B. suborbitale, which possesses a great variability in retinal thickness across the
retina, with the ventral part being nearly double the thickness of the remaining retina
(Chapter 3), measurements were made in both the ventral and dorsal retinal regions. For
all species, eight measurements were performed for each of the photoreceptor
parameters and the average measure was reported. Please note that the term inner
segment in this study refers only to the ellipsoid region of the rod photoreceptor and
does not include cell bodies, axons and synapses. Also, the diameter of the
photoreceptor was estimated by measuring the diameter of the outer segment only.
In addition to the labour-intensive assessment of the density of rod

photoreceptors using the stereological optical fractionator method in the retinal
wholemounts of five different species, photoreceptor density was also measured in
131
transverse section in a defined region of central retina for all other species for
interspecific comparison. To confirm, that the densities calculated from retinal
wholemounted material align with those using sections, we performed a direct
comparison in the five species for which both data were available and reveal that there
was close agreement.
4.3.7. Optical sensitivity estimations

Optical sensitivity to downwelling light and to bioluminescent flashes was
estimated separately for each species using direct measurements and data from the
literature.
Sensitivity to downwelling light (extended sources, in units of mm2 sr) was

estimated using the formula from Land (1981):
In the equation, A is the diameter of the pupil, f the focal length of the eye (lens radius x
a Matthiessen ratio of 2.55), d, l and k are the diameter, outer segment length and
absorption coefficient of the photoreceptors, respectively. Due to the difficulty in
measuring the pupil aperture on board ship and the great variability in the location of
the aphakic gap within lanternfishes (Chapter 3), we used the lens diameter as a
measure for A in this study. The absorption coefficient k is fixed at 0.035 μm-1, which is
the average value for vertebrates (Warrant and Nilsson, 1998).
The previous equation can now be expressed in a different way;
where F = F-number = f/A. Since Matthiessen’s ratio states that the focal length in
fishes is about 2.55 the radius of the lens (Matthiessen, 1882; Matthiessen, 1886), then f
= 1.275A , which means that F = 1.275 for all the species irrespective of eye size.
The equation can then be written as follows;
132
As a result, sensitivity to downwelling light in fishes is completely independent of the

size of the lens and is directly proportional to the diameter of the photoreceptors and to
a lesser extent to the length of the photoreceptor outer segments.
Sensitivity to bioluminescent flashes (point-like sources, in photons) was

estimated using the following formula from Warrant (2000) and Warrant and Locket
(2004):
In the equation, E is the number of photon emitted at the source, A is the pupil diameter,
r is the distance in meters between the light source and the eye, α is the combined
attenuation coefficient of bioluminescence (due to scattering and absorption of light by
water) and k is the photoreceptor absorption coefficient. As for the previous equation,
the pupil diameter was replaced by the lens diameter for A and k was fixed at 0.035 μm-1
(Warrant and Nilsson, 1998). E was fixed at 1010 photons as in Warrant and Locket
(2004), r was set at 1 m and α fixed at 0.05 m-1 (Denton, 1990). This equation shows
that sensitivity to bioluminescence is directly proportional to the size of the lens and to a
lesser extent to the length of the outer segments.
4.3.8. Phylogenetic analyses

While standard statistical analyses assume independence of the samples, this
was not true when comparing different species, as more closely related species are
expected to be more similar to one another due to sharing a common ancestor.
Therefore, all data analyses were performed using phylogenetic comparative analyses to
account for the shared history among species (Harvey and Pagel, 1991).
The analyses were conducted as described in de Busserolles et al. (2013).

Briefly, as no fully resolved phylogeny is currently available for the family
Myctophidae, two different phylogenies, A and B (Figure 4.2), were built using the
Mesquite program v. 2.75 (Madison and Madison, 2011) based on two different
published phylogenies (Paxton et al., 1984, Poulsen et al., 2013).
133
Figure 4.2. Phylogenetic trees of the Myctophidae family reconstructed from (A) Paxton et al.
(1984), (B) Poulsen et al. (2013). The red branches indicate the main differences between the
two trees. Branch lengths are arbitrarily ultrametricized on the figure. Modified from de
Busserolles et al. (2013).
The main differences between the two phylogenies are the position of the taxon
Notolychnus and the position of the tribe Diaphini. In fact, in phylogeny B, Notolychnus
became a sister taxon of all the remaining myctophids, and the Diaphini became a sister
tribe of the Lampanyctini. Due to the lack of resolution, both phylogenies are only
resolved to the generic level, resulting in several polytomies (i.e. unresolved
relationship among species). Unfortunately, the presence of polytomies prevents the
application of many phylogenetic analyses. Therefore, to bypass this problem, 100
alternative phylogenies were generated with polytomies randomly resolved to
infinitesimally small (10-6) branch lengths using the Mesquite program v. 2.75 (Madison
and Madison, 2011). Ten of these phylogenies with randomly resolved polytomies were
selected at random to perform the different analyses and the result between each of the
10 phylogenies was compared for consistency. Moreover, to fit the statistical
requirements for the phylogenetic linear models described below, branch lengths were
134
transformed using Grafen’s method (Grafen, 1989) with rho transformation set at 2.5
before all analyses. All statistical analyses were performed, using both phylogenies
separately, on Log10-transformed data with the statistical program R v.2.15.0 (R
Foundation for Statistical Computing 2012).
4.3.9. Ecological data

For the purpose of the statistical analyses, we used the same ecological dataset
as in de Busserolles et al. (2013, Chapter 2), which includes information about the
luminous organs and depth distribution patterns of each species examined here. More
specifically, the presence-absence of luminous organs (enlarged Dn/Vn, caudal),
additional luminous patches, sexual dimorphism in luminous tissues and the type of
sexual dimorphism in luminous tissues (Dn/Vn, caudal luminous organs and/or
luminous patches) was noted for each species. However, as in de Busserolles et al.
(2013), the differentiation of the type of sexual dimorphism in our analyses did not
show significant differences. As a consequence, only results for the presence/absence of
sexually dimorphic features are presented in this study. In terms of depth distribution,
each species was assigned a categorised depth range during night and day (de
Busserolles et al. 2013). The group categories were created in terms of the amount of
downwelling light present and resulted in three groups for each time of the day:
moderate light level (0-5 m at night, 200-500 m during the day), low light level (5-
100 m at night, 500-900 m during the day) and no light (<100 m at night, <900 m
during the day).

The phylogenetic signal for continuous and discrete traits was estimated with
Pagel’s lambda (λ) using the package GEIGER in R (Harmon et al., 2008). Pagel’s λ is
a measure of the degree of phylogenetic dependence in the data (Pagel, 1999), meaning
to which degree closely related species are more similar to each other than what is
expected by random evolutionary processes. Pagel’s λ varies from 0 to 1, with λ value
of 1 indicating that traits gradually accumulate changes over time in a Brownian motion
process (i.e. random change in any direction) and λ values of 0 indicating that no
phylogenetic signal is present and that traits have evolved in response to selective
processes. The observed λ value for each trait was compared to λ values of zero and one
using likelihood ratio tests with df = 1.
135
4.3.11. Phylogenetic linear models

The relationships between each of the morphological traits (eye size,
photoreceptor size) and the relationships between the morphological and ecological
traits (luminous organs, depth distribution) were all assessed using phylogenetic
generalised least squares regressions (PGLS, Freckleton et al., 2002) with the package
APE in R (Paradis et al., 2004). PGLS regressions estimate a phylogenetic scaling
parameter, λ, using maximum likelihood methods to determine the degree of covariance
in the residuals of the model, while controlling for phylogenetic effects. This approach
also examines whether the scaling parameter λ significantly differs from 0 or 1 using
likelihood ratio tests, where λ = 0 indicates no phylogenetic dependence in the data and
λ = 1 indicates strong phylogenetic association in the data (Pagel, 1999; Freckleton et
al., 2002). PGLS models were first used to assess relationships between eye size and
photoreceptor traits (rod diameter, outer and inner segment length). Since eye diameter
is strongly correlated with standard length within the family (de Busserolles et al.,
2013), standard length was added as a covariate in all analyses. Finally,
phylogenetically controlled multiple regression models were used to assess if
photoreceptor length and diameter were related with various ecological parameters
when correcting for the effect of eye size and standard length.
4.4. Results
4.4.1. Topographic distribution of photoreceptors
Topographic maps of photoreceptor density were constructed for five different
species of lanternfishes from wholemounted retinae. Only a single population of rod
photoreceptors is present (Figure 4.1). Rods are densely packed and individually
arranged into an hexagonal array (Figure 4.1), and their density across the retina is
heterogeneous, varying greatly between species (Figures 4.3 to 4.6). Although some
intra-specific variation exists (i.e. Lampanyctus parvicauda, Figure 4.5), consistent
patterns can easily be discerned and different specialisations or areae of high density
can be outlined for each species. Our results reveal the presence of at least three
different types of specialisations in lanternfishes: an arch, a ring and a streak-like
elongated area. An arch specialisation is observed in two different species, Bolinichthys
longipes and Diaphus brachycephalus (Figures 4.3 and 4.4A, respectively). In B.
longipes, the arch specialisation is present spanning the dorsal-temporal-ventral part of
the retina with a peak density of rod photoreceptor cells situated in temporal retina with
136
densities ranging from 650 to 760 x 103 rods mm-2 (Table 4.2). In D. brachycephalus,
the arch specialisation is present spanning the nasal-dorsal-temporal part of the retina. A
ring specialisation is observed in two species; Nannobrachium idostigma and
L. parvicauda (Figures 4.4B and 4.5, respectively) with a peak density of rod
photoreceptors ranging from 480 to 570 x103 rods mm-2 (Table 4.2). Finally, a streak-
like specialisation is observed in the ventral-temporal part of the retina in one species,
Myctophum brachygnathum with an elongated decrease of rod photoreceptor density
from peak cell densities ranging from 1840 to 2280 x 103 rods mm-2 in the three
individuals examined (Figure 4.6, Table 4.2).
In terms of topographic changes across the retina, the gradient of rod density is
quite shallow for each species. However, M. brachygnathum shows a much higher
photoreceptor density than the other species with a peak density of 2280 x 103 rods
mm-2 compared to peak densities ranging from 418 to 760 x 103 rods mm-2 for the other
species (D. brachycephalus and B. longipes, respectively, Table 4.2). Intraspecific
differences in peak photoreceptor densities are also present and due mainly to
differences in the size of the individuals examined, with larger individuals having
higher densities across the retina and higher numbers of rods per retina. For a few of the
individuals (B. longipes C, M. brachygnathum A, N. idostigma, Table 4.2), the total
number of photoreceptors was most likely under-estimated due to the inability to count
cells for around 30% of the sample sites randomly chosen by the StereoInvestigator
system.
Table 4.2. Summary of the quantitative data obtained from the optical fractionator method on
the wholemount retina. Densities and total cell number are given for each species in addition to
the coefficient of error (Schaeffer CE). The peak cell densities in brackets for
M. brachygnathum were found by sub-sampling. Asterisks indicate possible under-estimation
due to a reduced number of sites countable.
Species Individual Peak cell density Mean cell density Total cell Schaeffer
(rods x 103.mm-2) (rods x 103.mm-2) number CE
Bolinichthys longipes A 750 522 9,185,280 0.030
B 650 455 9,827,156 0.024
C 760 575 12,041,724* 0.040
Lampanyctus parvicauda A 520 370 4,348,485 0.035

B 480 374 5,930,100 0.032
C 570 448 10,218,950 0.026
Myctophum brachygnathum A 1840 1139 61,701,500* 0.036

B 1960 (2280) 1268 83,376,496 0.038
C 1800 (2120) 1190 85,139,400 0.060
Diaphus barchycephalus A 418 255 3,177,144 0.034
Nannobrachium idostigma B 480 358 3,763,200* 0.050
137
Figure 4.3. Topographic maps of photoreceptor densities (cells x 103.mm-2) for three different
individuals of Bolinichthys longipes. The black arrows indicate the orientation of the retina. T =
temporal, V = ventral. Scale bar = 1 mm. Information about the size of each individual, the
stereological parameters used and the quantitative results from the analyses can be found in
Table 4.2 and Table 4.3.
138
Figure 4.4. Topographic maps of photoreceptor densities (cells x 103.mm-2) for Diaphus
brachycephalus (A) and Nannobrachium idostigma (B). The black arrows indicate the
orientation of the retina. T = temporal, V = ventral. Scale bar = 1 mm. Information about the
size of each individual, the stereological parameters used and the quantitative results from the
analyses can be found in Table 4.2 and Table 4.3.
139
individuals of Lampanyctus parvicauda. The black arrows indicate the orientation of the retina.
T = temporal, V = ventral. Scale bar = 1 mm. Information about the size of each individual, the
stereological parameters used and the quantitative results from the analyses can be found in
Table 4.2 and Table 4.3.
140
individuals of Myctophum brachygnathum. The black arrows indicate the orientation of the
retina. T = temporal, V = ventral. Scale bar = 1 mm. Information about the size of each
individual, the stereological parameters used and the quantitative results from the analyses can
be found in Table 4.2 and Table 4.3.
141
The relationship between the location of the aphakic gap, the presence of a
tapetum lucidum and its regional coverage across the retina, and the photoreceptor
distribution was visually investigated (Figure 4.7). Apart from D. brachycephalus,
which possesses a tapetum lucidum underlying the area of high photoreceptor density,
there was no clear relationship between the pattern of tapetal coverage and
photoreceptor density. However, the areas of highest photoreceptor density closely align
with the position of the aphakic gap, so that the region of the visual field subtended (and
therefore sampled) would be increased. However, this was not the case for
M. brachygnathum, which possesses a peak photoreceptor density directly beneath the
aphakic gap, where light striking this retinal region of increased sampling would not be
focussed by the lens. Similarly, the two species which possess a ring specialisation of
high densities of rod photoreceptors (i.e. L. parvicauda, N. idostigma, Figures 4.7 C and
D) also have a circumlental aphakic gap, suggesting that large regions of seemingly
“specialised” retina are illuminated by unfocussed light, at least on the optical axis. It is
therefore clear that the location and function of the aphakic gap may differ between
species (Figure 4.7).
4.4.2. Morphometric analyses of the photoreceptors

All the lanternfish species analysed in this study seemed to possess a pure rod
retina. A summary of each species’ rod morphometric data (length and diameter/width
of the inner and outer segments) and rod density estimates in the central part of the
retina, in addition to eye and lens diameter and standard length is given in table 4.3. A
great variation in rod photoreceptor size is revealed (Figure 4.8) with species possessing
wide rods with small inner segments (i.e. Lampanyctus parvicauda, Figure 4.8A), wide
rods with long inner segments (i.e. Diaphus brachycephalus, Figure 4.8B), thin rods
with short inner segments (Bolinichthys supralateralis, Figure 4.8C) and thin rods with
long inner segments (i.e. Myctophum brachygnathum, Figure 4.8D). In the central part
of the retina, rod length (i.e. outer segment plus ellipsoid region of inner segment)
varied from 33.3 to 92.9 μm (Lampanyctus crocodilus and Bolinichthys nikolayi,
respectively) with outer segment length ranging from 23.6 μm (Notoscopelus elongatus)
to 89.0 μm (Bolinichthys supralateralis) and inner segment length varying from 3.7 μm
(Lampanyctus crocodilus) to 22.0 μm (Gonichthys tenuiculus). Rod (outer segment)
diameter varied from 0.9 to 2.7 μm in Symbolophorus verany and Nannobrachium
phyllisae, respectively, and rod density estimates from sections ranged from 19 to 119 x
104 mm-2 in Nannobrachium phyllisae and Symbolophorus evermanni, respectively.
142
Density estimates in the central part of the retina from wholemounts and sections give
similar results with the exception of Lampanyctus parvicauda for which estimation
from sections underestimated the rod density compared to the rod density estimated in
wholemounts. However, this underestimation could be due to the smaller size of the
individual used for sectioning (28.4 mm) compared to the individuals used for
wholemounts (54.2 to 65.2 mm).
4.4.3. Optical sensitivity

Optical sensitivity to downwelling light and bioluminescent flashes was
estimated for each species using the lens diameter and the rod measurements taken in
the central part of the eye. Sensitivity measures varied greatly between species (Table
4.3, Figure 4.9). While some species appear to be particularly sensitive to
bioluminescent flashes (Myctophum sp, Symbolophorus sp, Figure 4.9), others are more
sensitive to downwelling light (Lampanyctus sp, Nannobrachium sp, Figure 4.9). A
great variation in optical sensitivity was also observed within the same genus, with the
genus Diaphus representing the most extreme example.
However, these results have to be interpreted carefully. While sensitivity to

downwelling light is independent of the size of the eye and principally depends on the
diameter of the photoreceptor, this is not the case for sensitivity to bioluminescence,
which is mainly influenced by the size of the eye. As a result, fishes with larger eyes
will automatically have a greater sensitivity to bioluminescent emissions. For this
reason, the sensitivity estimations for viewing bioluminescence presented here may not
easily be compared without some standardised method of comparing similar-sized
individuals and their comparative reliance on each of these two light sources.
143
Chapter 4
Table 4.3. Summary of eye and retinal measurements for 53 species of lanternfishes. Sensitivities to downwelling light (S) and bioluminescence (N) and rod
photoreceptor density estimations are also given. IS = inner segment (ellipsoid region only), OS = outer segment.
Species SL Eye ø Lens ø OS length IS length Rod ø Density estimate Sensitivity S Sensitivity N
(mm) (mm) (mm) (μm) (μm) (μm) (x103.mm-2) (μm2.sr ) (photons)
Benthosema glaciale 42.0 4.6 1.8 40.3 9.6 1.6 433 ± 82 0.75 1377
B. suborbitale long rods 28.2 3.0 1.3 75.1 7.8 1.3 643 ± 122 0.55 990
short rods 34.3 3.7 1.9 346 ± 66 0.92 746
Bolinichthys longipes 35.0 3.8 1.5 58.7 9.2 1.3 430 ± 82 0.55 1182
B. nikolayi 27.5 2.9 1.2 87.4 5.5 1.6 313 ± 59 0.96 871
B. supralateralis 41.2 4.2 2.0 89.0 7.3 1.1 666 ± 127 0.46 2160
Ceratoscopelus maderensis 52.0 5.1 2.1 34.3 6.5 1.3 520 ± 99 0.41 1763
C. warmingii 60.8 5.9 2.5 49.2 12.0 1.4 457 ± 87 0.58 3026
Diaphus brachycephalus 35.2 4.3 1.8 50.3 16.7 2.3 233 ± 44 1.60 1595
D. danae 92.2 8.2 3.3 34.7 16.6 1.4 379 ± 72 0.49 4609
D. fulgens 41.2 3.4 1.8 42.7 12.0 1.2 659 ± 125 0.40 1478
D. gamani 33.1 2.5 1.0 50.2 15.6 1.0 732 ± 139 0.33 532
D. holti 40.0 5.1 2.1 45.0 13.4 1.0 840 ± 160 0.31 2099
D. luetkeni 38.3 2.4 1.1 53.6 6.9 1.5 437 ± 83 0.61 552
144
D. meadi 28.3 3.3 1.2 36.2 10.2 1.0 873 ± 166 0.26 657
D. mollis 39.5 4.0 1.8 46.0 12.9 1.2 767 ± 146 0.45 1474
D. parri 51.0 5.6 2.1 45.5 15.0 1.1 530 ± 683 0.34 2050
D. phillipsi 27.4 2.2 1.0 27.7 14.0 1.3 425 ± 55 0.42 362
Photoreceptors and optical sensitivity

D. regani 40.8 2.5 1.2 55.9 11.5 1.1 731 ± 79 0.39 711
D. splendidus 34.1 2.1 0.8 60.5 8.7 1.5 532 ± 55 0.72 335
D. termophilus 48.3 4.2 1.8 40.5 13.2 1.4 540 ± 65 0.54 1525
Diogenichthys atlanticus ♀ 20.1 2.1 0.8 43.3 6.9 1.1 879 ± 184 0.35 312
Diogenichthys atlanticus ♂ 21.4 2.7 1.0 31.5 6.9 1.2 843 ± 146 0.37 413
D. laternatus ♀ 31.1 3.2 1.3 42.8 8.9 1.2 478 ± 62 0.43 780
D. laternatus ♂ 22.1 2.2 0.9 38.1 9.7 1.2 552 ± 54 0.38 347
Electrona risso 46.0 7.0 3.0 31.8 5.4 1.0 805 ± 111 0.28 3666
Gonichthys tenuiculus 41.3 2.9 1.4 40.0 22.0 1.3 973 ± 228 0.50 890
Hygophum benoiti 45.0 6.2 2.3 45.1 11.3 1.8 271 ± 43 0.96 2561
H. hygomii 57.3 7.5 3.1 44.7 9.4 1.4 362 ± 54 0.60 4607
H. proximum 26.2 3.2 1.2 44.7 10.3 2.1 195 ± 29 1.32 700
Lampadena urophaos 40.0 2.8 1.3 42.5 13.1 1.2 498 ± 64 0.40 790
Lampanyctus alatus 43.2 2.1 0.9 51.5 5.2 1.5 452 ± 58 0.73 376
L. crocodilus 31.0 1.7 0.6 29.6 3.7 1.5 294 ± 47 0.53 138
L. iselinoides 34.3 2.0 0.8 59.4 5.2 1.7 264 ± 27 0.99 341
L. nobilis 37.5 2.1 1.0 46.3 4.1 1.7 405 ± 56 0.89 516
Table 4.3. (Continued).
Chapter 4
Species SL Eye ø Lens ø OS length IS length Rod ø Density estimate Sensitivity S Sensitivity N
(mm) (mm) (mm) (μm) (μm) (μm) (x103.mm-2) (μm2.sr ) (photons)
L. omostigma 27.8 1.9 0.6 48.0 7.4 1.6 362 ± 50 0.74 198
L. parvicauda 28.4 1.8 0.9 36.4 5.6 2.3 280 ± 29 1.40 324
L. pusillus 37.0 2.1 0.82 53.1 14.2 1.4 259 ± 54 0.63 337
L. vadulus 37.4 2.2 1.02 58.5 8.3 1.4 225 ± 39 0.64 539
L. gemellari 32.7 2.1 1.24 45.9 11.3 1.4 595 ± 64 0.59 728
Loweina interrupta 25.0 2.1 0.88 41.8 16.3 1.5 381 ± 46 0.61 354
Myctophum brachygnathum 67.7 7.6 3.15 45.5 14.7 1.0 1078 ± 226 0.32 4699
M. nitidulum 85.4 7.1 2.47 44.9 20.7 1.1 1050 ± 168 0.34 2874
M. spinosum 39.6 3.9 1.63 39.1 15.1 1.1 866 ± 105 0.31 1177
145
Nannobrachium cf. nigrum 52.3 2.5 0.97 52.7 4.4 2.1 246 ± 34 1.37 471
N. idostigma 72.2 3.9 1.16 45.6 5.4 1.7 312 ± 40 0.89 638
N. phyllisae 50.4 2.2 0.95 52.1 5.7 2.7 194 ± 27 2.39 450
Notolychnus valdiviae 19.5 1.5 0.60 55.9 13.4 1.4 484 ± 92 0.64 184
Notoscopelus elongatus 47.0 3.2 1.23 23.6 14.3 1.1 548 ± 104 0.26 505
N. kroeyerii 101.2 6.1 2.74 36.5 14.8 1.0 852 ± 148 0.26 3221
Symbolophorus cf. boops 72.0 6.0 2.12 37.6 14.2 1.1 782 ± 116 0.34 1954
S. evermanni 59.0 5.1 2.81 39.0 12.7 1.1 1186 ± 206 0.32 3496
S. rufinus 69.0 6.5 2.30 48.5 17.4 1.0 871 ± 231 0.33 2569
S. veranyi 85.0 6.7 2.94 41.1 13.7 0.9 751 ± 120 0.26 3917
Photoreceptors and optical sensitivity

Triphoturus nigrescens 35.4 2.0 0.85 50.9 10.7 1.5 303 ± 39 0.74 357
T. oculeus 33.9 2.1 0.80 42.5 5.6 1.9 318 ± 59 1.11 295
Figure 4.7. Aphakic gap position (left), tapetum lucidum pattern (middle) and topographic
maps of photoreceptor densities (right) for five species of lanternfish. (A) Bolonichthys
longipes, (B) Diaphus brachycephalus, (C) Lampanyctus parvicauda, (D) Myctophum
brachygnathum, (E) Nannobrachium idostigma. T = temporal, V = nasal. The aphakic gap is
represented in white. The tapetum lucidum, when present, is represented in grey. Scale bar for
the maps = 1 mm.
146
Figure 4.8. Transverse light microscopy sections through the retina of four species of
lanternfish showing the variability in rod length and diameter. (A) Lampanyctus parvicauda,
(B) Diaphus brachycephalus, (C) Bolinichthys supralateralis, (D) Myctophum brachygnathum.
OS = outer segment, IS = inner segment (ellipsoid region only), scale bar = 10 μm.
147
Figure 4.9. Relative sensitivity to bioluminescence (black bars) and downwelling sunlight (grey
bars) for each species of lanternfish analysed in this study. The species are ranked in
phylogenetic order following phylogeny A.
148

Estimation of the phylogenetic signal using Pagel’s lambda gives relatively
similar results with both phylogenies (Table 4.4). Results show that Dn-Vn and caudal
luminous organs have a strong phylogenetic signal and that they gradually accumulated
changes over time in a Brownian motion process. On the contrary, no phylogenetic
signal was observed for the standard length and the outer segment length variables. An
intermediate value of Pagel’s lambda was found for the density variable, which,
although significantly different from 0 and 1, was closer to 0. All the remaining
variables (eye diameter, rod diameter, inner segment length, sensitivity S and N,
luminous patches, sexual dimorphism in luminous tissue and depth distributions) show
intermediate values of Pagel’s lambda, which, although significantly different from 0 or
1 (except day depth distribution), were generally closer to 1 depending on the
phylogenetic tree used.
Table 4.4. Estimates of the phylogenetic signal for each variable using Pagel’s Lambda. The
results are presented for one of the ten randomly selected polytomy resolved trees for the two
different phylogenies. A λ value of 1 indicates that the trait gradually accumulates changes over
time in a Brownian motion process. A λ values of 0 indicates that no phylogenetic signal is
present and that traits have evolved in response to selective processes. The superscript values
are likelihood ratio tests different from 0 and 1. Sample size is 53 for all variables except day
depth (50).
Variables λ (Tree A) λ (Tree B)

Eye diameter 0.88<0.001, <0.001 0.830.002, <0.001
Standard length <0.0011, <0.001 <0.0011, <0.001
Residuals eye/SL 0.95<0.001, <0.001 0.93<0.001, <0.001
Rod diameter 0.900.003, <0.001 0.920.01, <0.001
IS length 0.78<0.001, <0.001 0.92<0.001, <0.001
OS length 0.030.72, <0.001 <0.0011, <0.001
Sensitivity S 0.920.001, <0.001 0.950.003, <0.001
Sensitivity N 0.88<0.001, <0.001 0.850.002, <0.001
Rod density 0.11<0.001, <0.001 0.09<0.001, <0.001
Dn/Vn organs 1<0.001, 1 1<0.001, 1
Caudal luminous organs 1<0.001, 1 1<0.001, 1
Luminous patches 0.97<0.001, <0.001 0.98<0.001, <0.001
Luminous tissue sexual dimorphism 0.92<0.001, <0.001 0.96<0.001, <0.001
Day depth 0.950.53, <0.001 0.790.38, <0.001
Night depth 0.79<0.001, <0.001 0.74<0.001, <0.001
4.4.5. Relationship among morphometric traits

The results examining the relationships among morphometric traits are similar
and all independent of the phylogeny used and so only the results obtained with
phylogeny A are presented here. A phylogenetic linear regression shows that rod
diameter is negatively correlated with eye diameter (PGLS, n = 53, R2 = 0.18, t-value =
149
-2.842, P = 0.006), indicating that species with small eyes have larger rods and vice-
versa. No other relationships between eye diameter and the other photoreceptor traits
(inner and outer segment length) could be identified as being statistically significant
using PGLS.
4.4.6. Relationship between morphometric and ecological traits

The phylogenetically controlled multiple linear regression models did reveal
some relationships between the photoreceptor traits and some of the ecological variables
analysed in this study (Table 4.5). Again, the results are all similar and independent of
the phylogeny used and only the results obtained with phylogeny A are presented here.
Our results show that rod diameter is negatively correlated with the presence of
luminous patches (Table 4.5) and positively correlated with night depth distribution
(Table 4.5). These results indicate that species that do not possess any additional
luminous patches and that venture to deeper depths at night have larger rod
photoreceptors and vice-versa. Similar results are found for the sensitivity S (to
downwelling light) with species without luminous patches (Table 4.5) and species with
a deeper distribution profile at night (Table 4.5) having a greater sensitivity to
downwelling light.
Phylogenetically controlled multiple linear regression models also reveal some

relationships between photoreceptor length and ecological traits. The results indicate
that the inner segment length is negatively correlated with the night depth distribution
(Table 4.5) meaning that species living deeper at night have smaller inner segments and
vice-versa. Finally, outer segment length was found to be negatively correlated with the
presence/absence of head luminous organs (Table 4.5) and positively correlated with
night depth distribution (Table 4.5) indicating that species having Dn/Vn luminous
organs and a shallow distribution at night have smaller outer segments and vice-versa.
A positive relationship was also found between rod photoreceptor density and
the presence of sexual dimorphism (Table 4.5) where species with sexually dimorphic
luminous tissues have greater rod densities.
Finally, no statistically-significant relationships were found between relative eye

size and sensitivity N (to bioluminescence) and any of the ecological variables using
PGLS (Table 4.5).
150
Table 4.5. Regression models of several visual traits with different predictor variables when
controlling for phylogeny (PGLS). The results are identical and independent of the phylogeny
used. Standard length was added as a covariate in the models. λ = phylogenetic scaling
parameter, the superscript * after the parameter λ indicates whether the parameter was
significantly different from 0 (first position) and from 1 (second position) in the likelihood tests,
β = partial regression slope. In bold are the significant results. The sampling size was 50.
Trait λ Predictor variables β t P

Rod diameter <0.001ns, * Eye diameter -0.073 -0.677 0.502
Standard length -0.070 -0.562 0.577
Dn/Vn Luminous organs -0.055 -1.131 0.264
Luminous tissue sexual dimorphism -0.040 -1.139 0.261
Day depth -0.011 -0.539 0.592
Night depth 0.071 2.242 0.030
Sensitivity S <0.001ns, * Eye diameter -0.179 -0.809 0.423

Luminous tissue sexual dimorphism -0.062 -0.860 0.395
Day depth -0.039 -0.917 0.365
Night depth 0.170 2.624 0.012
Inner segment length <0.001ns, * Eye diameter 0.120 0.629 0.533

Standard length 0.219 1.008 0.319
Dn/Vn Luminous organs 0.066 0.775 0.443
Day depth 0.044 1.193 0.240
Night depth -0.117 -2.102 0.042
Outer segment length <0.001ns, * Eye diameter -0.070 -0.532 0.598

Day depth -0.047 -1.855 0.071
Night depth 0.078 2.024 0.049
Rod density <0.001ns, * Eye diameter -75.59 -0.294 0.770

Standard length -407.40 -1.392 0.171
Dn/Vn Luminous organs 70.76 0.616 0.541
Caudal luminous organs 63.84 0.541 0.592
Day depth -50.38 -1.009 0.319
Night depth -128.87 -1.713 0.094
Eye diameter 0.909*, * Standard length 0.932 10.708 <0.001

Day depth 0.003 0.154 0.878
Night depth 0.043 1.133 0.264
Sensitivity N 0.933*,* Standard length 1.766 9.379 <0.001

Day depth 0.028 0.606 0.548
Night depth 0.099 1.215 0.231
151
4.5. Discussion
The aim of this study was to assess the variability in photoreceptor
characteristics within a large range of species of lanternfishes with different ecological
traits to assess the influence(s) of both ecology and phylogeny on the evolution of their
visual system. This study follows on from a previous investigation looking at eye size
variation in lanternfishes in relation to their ecology (de Busserolles et al., 2013). The
aim of the de Busserolles et al. (2013) study was borne from the assumption that there is
a gradual change in the visual scene in the mesopelagic zone with depth and that this
will ultimately result in a great diversity in eye size (Denton, 1990; Warrant, 2000;
Warrant, 2004). They hypothesized that lanternfishes with a deeper distribution range
and/or that have less reliance on bioluminescence (i.e. having less luminous tissues) will
have smaller eye sizes. This relationship was found not to exist for lanternfishes, where
relative eye size could not be linked to any of the ecological variables tested but was
instead strongly influenced by phylogeny (de Busserolles et al., 2013). The present
investigation was initiated to explore what other visual characteristics may account for
the high levels of variability in behaviour in an environment, which would appear to be
heavily reliant on visual signals for survival.
4.5.1. Topographic variations in sampling the ambient light environment

Topographic analyses of photoreceptor and ganglion cell distributions are very
useful in providing information about the visual ecology of a species by identifying
areas of the visual field of high importance (i.e. area of high cell densities, Hughes
1977; Collin and Pettigrew, 1988a; Collin and Pettigrew, 1988b; Collin, 2008). Despite
photoreceptors playing a major role in the process of vision by collecting light
information and initiating phototransduction, topographic analyses of photoreceptor
densities are quite sparse in teleosts compared to ganglion cell analyses and are non-
existent in deep-sea teleosts. To our knowledge, this is the first analysis of
photoreceptor distribution across the retina in any deep-sea species of bony fishes.
Since the acuity is limited by the amount of light available, areas of high
photoreceptor density usually match the peak(s) in ganglion cell density (Fritsches et
al., 2003; Litherland and Collin, 2008). Topographic analyses of ganglion cell
distribution (not including amacrine cells) have been examined in three species of
lanternfishes from the genus Lampanyctus, and reveal a poorly specialised retina,
showing a nearly uniform distribution of cells within the ganglion cell layer (Collin and
152
Hoskins, 1997; Wagner et al., 1998). Conversely, results from this study, on different
species from a range of genera, show a great diversity in visual specialisations with
different species having distinct areas of high photoreceptor density (with respect to
both peak density and the shape of the specialised acute zone) i.e., an arch in
Bolinichthys longipes and Diaphus brachycephalus, a ring in Lampanyctus parvicauda
and Nannobrachium idostigma and a streak-like elongated area in Myctophum
brachygnathum. Although a topographic analysis of ganglion cells was not performed
for these species, it is likely that their distribution would match that of the
photoreceptors, thereby indicating possible interspecific differences in visual capability.
Moreover, the fact that the position of the aphakic gap matches the photoreceptor
distribution in most species examined emphasises the importance of a specific area of
the visual field in any visually-guided behaviours in each species.
The retinal morphology of teleost fishes is highly diverse and is found to

correlate very well with habitat complexity and behavioural ecology (Collin and
Pettigrew, 1988a; Collin and Pettigrew, 1988b, Collin, 1997). In the mesopelagic zone,
different ecological tasks are more likely to influence each species’ photoreceptor
topography. Two types of light stimuli can be detected in the mesopelagic zone,
downwelling sunlight and bioluminescence. Assessing the intensity of downwelling
light is essential to a species’ ability to maintain a particular depth during the day,
camouflage its silhouette by counter illumination, trigger vertical migration, set
circadian rhythms and/or detect the presence of prey or predator from below. In
contrast, assessing the intensity and frequency of bioluminescent signals will be crucial
for detecting other individuals (prey, predator, mate) at deeper depths, where
bioluminescent cues predominate (i.e. in the North Atlantic, 90% of the individuals
below 500 m produce bioluminescence, Herring, 2002). As most lanternfish vertically
migrate (Hulley, 1984) and possess photophores used for counterillumination (Case et
al., 1977), interspecific differences in photoreceptor topography will most likely be due
to differences in how each species interact with prey, predators and/or mates. Several
lanternfish species possess sexually dimorphic luminous tissues that are thought to play
a role in sexual communication (Herring, 2007). However, topographic analyses of the
retinae of both males and females will have to be conducted to reveal any sexual
dimorphism with respect to the location of these retinal “acute” zones and how they are
used in visually-guided behaviour(s) underlying reproduction.
153
Diaphus brachycephalus possesses a dorsal arch, a dorsal tapetum and a ventral

aphakic gap. All these specialisations may work together to enhance the capture of
photons of light emanating from below, most likely to detect bioluminescence signals in
the lower part of the visual field. A similar scenario may apply to Bolinichthys longipes,
which possesses a temporal arch extending dorsally and ventrally, a dorso-temporal
tapetum and a ventro-nasal aphakic gap. In this species, the eye is specialised to light
capture in the frontal and ventral visual fields. In Myctophum brachygnathum, the area
of peak photoreceptor density is situated in the ventral part of the retina, providing
higher sampling of light signals emanating from above. If the distribution of both the
photoreceptor and ganglion cell populations are in register, this ventral-temporal acute
zone will enhance the detection of a silhouette against the lighter background of the
upper mesopelagic zone. Although not optimised for receiving a focussed image, the
ventral aphakic gap in this species might facilitate the detection of bioluminescent
signals (prey, predator or mate) situated below the fish, within the increased visual field
produced by this ventral extension of the pupillary aperture.
Lampanyctus parvicauda and Nannobrachium idostigma do not possess any

isolated specialisations enhancing visual capabilities within a particular part of the
visual field. Instead, both species possess a ring specialisation in addition to a
circumlental aphakic gap, which would enhance the chance of photon capture in all
directions. The lack of a specialisation mediating acute vision within a specific part of
the visual field may indicate that these two species do not rely on vision as much as
other species and rely more on other sensory systems. It is also possible that those
species are visual generalists, interacting with other individuals to avoid predation but
feeding opportunistically and targeting a wide range of prey items (Tyler and Pearcy,
1975; Kozlov, 1995). Myctophids are mainly zooplankton consumers (i.e. copepods,
euphausids, amphipods) with a high diversity of organisms comprising their diet
(Young and Blaber, 1986; Tyler and Pearcy, 1975; Podrazhanskaya, 1993; Kozlov,
1995; Ishihara and Kubota, 1997; Pusch et al., 2004; Shreeve et al., 2009).
Lanternfishes also provide food for a wide range of higher level organisms like teleost
fishes (Goldsworthy et al., 2002; Esposito et al., 2009), cephalopods i.e. squid (Parry,
2006; Watanabe et al., 2008), seabirds (Jackson, 1988; Green et al., 1998), and
mammals (Green et al., 1991; Carey, 1992; Kozlov, 1995). Interspecific differences in
the diet and methods of predation could explain differences in the topographic
distribution of photoreceptor cells. Unfortunately, these data are not yet available for the
154
species analysed in this study and would have to be considered in any future
interpretations.
4.5.2. Different strategies for optimizing light capture by retinal photoreceptors

Our results highlight a great diversity in photoreceptor design within the
Myctophidae at all levels. Differences are evident in terms of photoreceptor distribution
(as discussed above), photoreceptor dimensions (length and diameter) and density.
Overall, rod outer segments in lanternfishes are not particularly long compared
to other deep-sea species with similar retinal organisation (i.e. a single bank of
photoreceptors, Wagner et al., 1998), with a maximum length of 89 μm recorded for
Bolinichthys longipes (this study) compared to others species which possess rod outer
segments over 100 μm in length i.e.150 μm in Platytroctes apus (Locket, 1971) and 170
μm in Sternoptix sp. (Nicol, 1989). Some species of myctophids possess relatively small
outer segments i.e. Diaphus phillipsi, comparable to the rod photoreceptors in
goldfishes (Bassi and Powers, 1990).
Rod outer segment diameter in myctophids is also relatively small compared to

what has been recorded for other deep-sea fishes (Locket, 1970, Locket 1977). In some
myctophid species, rods are extremely small i.e. <1 μm, Symbolophorus rufinus, making
them one of the narrowest photoreceptors found in both vertebrates and invertebrates,
including insects (Land and Nilsson, 2002) and approaching the optical limits for
photon capture. These minute rod diameters also equate to very high photoreceptor
densities, reaching peaks of 1186 x 103 mm-2 in Symbolophorus evermanni and 2280 x
103 mm-2 in Myctophum brachygnathum. These high rod densities far exceed what has
previously been recorded for any other deep-sea fishes, including those species that
possess a deep convexiclivate fovea (Wagner et al., 1998). Even at the lowest end of the
rod densities recorded for the myctophid species examined in this study, i.e. 194 x 103
mm-2, Nannobrachium phillisae, the values are still higher than those found in bottom
dwelling deep-sea fish (Wagner et al., 1998), sharks (Schieber et al., 2012) and shallow
water teleosts (Fernald 1990). Furthermore, around half of the species analysed in this
study showed peak rod densities higher than those recorded for nocturnal birds and
mammals, i.e. 341 x 103 rods mm-2 in the great horned owl (Fite, 1973) and 500 x 103
rods mm-2 in the cat (Steinberg et al., 1973).
155
Very high photoreceptor densities usually denote a high level of summation

between photoreceptors, interneurons and ganglion cells resulting in high sensitivity
(Warrant and Locket, 2004). Peak densities of ganglion cells were not investigated in
this study, although if one considers the highest recorded density of ganglion cells
found for a lanternfish (7.4 x 103 rods mm-2 in Lampanyctus ater, Wagner et al., 1998)
and the highest photoreceptor density reported in this study (2286 x 103 mm-2,
Myctophum brachygnathum), the summation/convergence ratio could potentially be as
high as 309 photoreceptors to one ganglion cell. In broad terms, high levels of
summation are indicated by the presence of a relatively thick outer nuclear layer (high
numbers of rod nuclei) and a thin inner nuclear/ganglion cell layer (low number of
bipolar cells and ganglion cells), which has been revealed for a large number of
lanternfish species (Chapter 3).
In terms of optical sensitivity, each species seems to be specialised for the

detection of a specific signal (downwelling light or bioluminescence), which might
reflect different behaviours (figure 4.9). While the determination of sensitivity estimates
to downwelling light is straightforward, the sensitivity estimations to bioluminescence
are biased due to the influence of eye size (as outlined in the materials and methods and
results sections). In fact, fishes with larger eyes will automatically have a greater
sensitivity to bioluminescent emissions. For this reason, the sensitivity estimations for
viewing bioluminescence presented here may not easily be compared without some
standardised method of comparing similar-sized individuals (i.e. relative eye size).
Moreover, comparisons of optical sensitivities between species have to be made
cautiously since the influence of several specialisations were not accounted for in the
calculations i.e. the presence of a tapetum lucidum and an aphakic gap in some species.
The presence of a tapetum lucidum and an aphakic gap would augment photon capture
by indirectly increasing both the outer segment length (given the reflection of light rays
incident on the tapetal plates) and the size of the pupillary aperture, respectively. One or
both of these ocular specialisations occur in a number of species and will undoubtedly
increase sensitivity to both downwelling light and bioluminescent light flashes, in a
specific part of the visual field. For example, some species like Lampanyctus sp and
Nannobrachium sp possess very small eyes and large photoreceptors. The large size of
these photoreceptors means that they are particularly sensitive to downwelling light, but
the size of the eye is limiting in terms of sensitivity to bioluminescence. However, both
156
species possess specialisations, such as a circumlental aphakic gap and a tapetum

lucidum to overcome these issues.
The high level of interspecific variability in strategies for optimizing light

capture by the photoreceptors within the Myctophidae make the task of assessing visual
capabilities and sensitivity quite challenging, especially if one wants to use a
standardised method.
4.5.3. The influence(s) of the photoreceptors on ecological variation in visual

behaviour
Results from the phylogenetic comparative analyses highlighted several
relationships between photoreceptor characteristics and the ecological variables tested
(depth distribution and luminous tissue patterns). The results of these models are
discussed in detail below.
4.5.3.1. Rod diameter and sensitivity

Our results reveal that species with no luminous patches and a deeper depth
distribution at night possess larger rods. Although a negative relationship between rod
diameter and relative eye size was found in a previous model with species with smaller
eyes having larger rods (this study), this relationship disappears when all the ecological
traits are added to the model. This shows that even though rod diameter is influenced by
the relative eye size of a species and that both characters are strongly influenced by
phylogeny (Pagel’s lambda, Table 4.5), it is the species’ ecology that drives this
component of the visual system in lanternfishes.
Since rod diameter and, to a lesser extent, outer segment length both determine
sensitivity to downwelling light in lanternfishes, it is not surprising to find similar
relationships between sensitivity to downwelling light and ecological traits to the ones
found with rod diameter. Results indicate that species without any luminous patches and
with a deeper distribution profile at night possess higher sensitivity to downwelling
light. A positive relationship between sensitivity to downwelling light and depth
distribution at night is surprising. In fact, even though some downwelling light is
present in the ocean at night, its intensity is considerably dimmer (10-6 - 10-7, Denton,
1990 ; Clarke and Wertheim 1956) than sunlight, only reaching the first few meters of
water (downwelling irradiance from the moon and stars at 5 m depth similar to that
157
found during the middle of the day at 300-500 m, Johnsen et al., 2004). Nevertheless,
our results indicate that species living at depths, where potentially no downwelling
sunlight is present possess a greater sensitivity to such a light. However, if we look
more closely at eye design in relation to the visual environment during the day, this
relationship may be more easily interpreted. During the day, the deep-sea can be divided
into two different visual environments; the mesopelagic zone, where both dim extended
(downwelling light) and point-like (bioluminescence) sources of light are present, and
the bathypelagic zone and beyond, where only point-like sources of light are visualized
against a completely dark background. If we compare optimised eye designs of species
living in each environment (as described by Warrant, 2000), mesopelagic species
possess large eyes with a large pupil and high summation ratios and bathypelagic
species possess small eyes with a large pupil and small summation ratios. While spatial
summation greatly improves photon catch by providing visual channels that view large
solid angles of space, it occurs at the expense of spatial and temporal resolution. Since
bioluminescent light flashes will create a point-like image on the retina, large visual
channels are not necessary and lower levels of summation will be sufficient to view the
signal and keep image resolution optimized (Warrant, 1999, 2000). At night, the two
visual environments (extended and point-like), can only be differentiated within the first
100 meters of the water column. Below 100m at night, the light environment might be
comparable to that of the bathypelagic zone during the day (< 1000 m), i.e. an
absolutely dark environment dominated by point-like sources of bioluminescent light.
As discussed previously, rod size in lanternfishes is a good predictor of density;

larger rods denote lower rod densities and therefore lower levels of spatial summation,
making the eyes of deeper living species at night better adapted to visualise their
environment. We therefore consider that the relationship between rod size and each
species’ depth distribution at night to be the most significative in terms of visual
ecology, where the relationship between sensitivity and downwelling light is harder to
interpret and most probably biased due to the influence of rod size on estimates of
sensitivity. This hypothesis will have to be verified in future studies with the analysis of
ganglion cell density and distribution for the same species examined in this study.
Moreover, further analyses are needed to understand the role of the luminous patches in
the visual behavior of myctophids. Although there is a possible role in intraspecific
communication given the presence of sexual dimorphism, the function(s) of the
luminous patches in lanternfishes remains unclear.
158
4.5.3.2. Outer segment length and density

Mesopelagic fishes possess several specialisations to enhance sensitivity
compared to their shallow water counterparts. Within the Myctophidae, these
specialisations include an increase in outer segment length and high densities of rod
photoreceptors. A large variability in both of these parameters also indicates different
levels of sensitivity. As per the formula of Land (1981) and Warrant and Nilsson
(1998), an increase in outer segment length will augment both sensitivity to
downwelling light and bioluminescence. PGLS results indicate that species with no
Dn/Vn luminous organs and with a deeper depth distribution at night have longer outer
segments. Several hypotheses have been proposed for the function of the Dn/Vn organs
in myctophids. They may be used as 1. a head torch, creating an extended scene of light
in their frontal visual field to search for prey (Herring, 1985, Haddock et al., 2010;
Warrant, 2000), 2. may be used to compare the intensity of their own photophore
emissions with the levels of downwelling sunlight in order to camouflage the silhouette
of their body when viewed from below (Lawry, 1974) and/or 3. for intraspecific
communication in sexually dimorphic species (Herring, 2007). At deeper depths at
night, sensitivity to downwelling light is not that useful and therefore increases in outer
segment length is more likely an adaptation to better visualise bioluminescent signals.
PGLS results also reveal that species with a sexual dimorphism in luminous
tissues possess higher rod densities. Higher photoreceptor density usually denotes the
presence of thin, tightly-packed rods and high levels of spatial summation, all
adaptations for achieving very high levels of sensitivity. The fact that species with
sexually dimorphic luminous tissues possess more sensitive eyes (as revealed here)
strengthens a long standing hypothesis proposed by a range of authors (Barnes and
Case, 1974; Edwards and Herring, 1977; Herring, 2007) that bioluminescence is used in
intraspecific communication in lanternfishes.
4.5.3.3. Inner segment length

Our results show that species with a deeper depth distribution at night possess
shorter rod inner segments. Photoreceptor inner segments contain mitochondria, the
metabolic drivers of the cell. In addition to their clear metabolic function, mitochondria
may also have an optical function by guiding the light toward the outer segments
(Hoang et al., 2002). A shorter inner segment could therefore indicate lower energetic
requirements (by the presence of less mitochondria) or less reliance on light guiding. In
159
both cases, the reason as to why depth would influence rod inner segment length is
currently unknown. However, since mitochondria could be packed in different ways
depending on the width of the inner segment, variability in mitochondrial density
between lanternfish species could be investigated in future analyses to help in
understanding interspecific variation. There may also be differential effects of pressure
on metabolic function that could be investigated.
4.5.4. Conclusion
A great diversity in the visual system of the Myctophidae is observed at the level
of the photoreceptors, the first stage of retinal processing. This study provides the first
analyses of photoreceptor distribution in any deep-sea teleost and reveals clear
interspecific differences in visual specialisations (areas of high rod photoreceptor
density), indicating potential interspecific differences in interactions with prey,
predators and/or mates. A great diversity in photoreceptor design (length and diameter)
and density is also present. Overall, the myctophid eye is very sensitive compared to
other teleosts and each species seem to be specialised for the detection of a specific
signal (downwelling light or bioluminescence), potentially reflecting different visual
demands for survival. Differences in photoreceptor characteristics could be related to
differences in ecological variables i.e. depth distribution at night, which was found to be
a significant factor in most of the models tested, indicating that vision at night is of
great importance in lanternfishes.
Brigitte Guillaumont (Ifremer), for providing additional samples. We gratefully
identification and, Adrian Flynn (UQ), Caroline Kerr (UWA) and Alan Goldizen (UQ)
for their help during field trips. We also wish to thank Michael Archer (UWA) for his
help with the preparation of light and electron microscopy sections, Prof. Julian
Partridge (University of Bristol) for helpful discussions regarding sensitivity estimations
and Jan Poulsen for access to his phylogeny prior to publication. We are indebted to
160
Joao Paolo Coimbra (UWA) for his assistance with retinal wholemount preparation and
stereology analyses and to Eduardo Garza-Gisholt for his help in using his R scripts for
the creation of the topographic maps. This work was funded by the Australian Research
Council (Discovery grant to SPC, the Deep-Australia Linkage Grant to NJM and SPC,
and a Postdoctoral Fellowship to JLF) and the West Australian State Government
(SPC). FdB was supported by a Scholarship for International Research Fees (SIRF) and
a University International Stipend (UIS) at the University of Western Australia.
4.7. References
Arnott HJ, Maciolek NJ, Nicol JAC. 1970. Retinal tapetum lucidum: a novel reflecting
system in the eye of teleosts. Science 169(3944):478-480.
Baddeley A, Turner R. 2005. Spatstat: an R package for analyzing spatial point patterns.
Journal of Statistical Software 12:1-42.

Bassi CJ, Powers MK. 1990. Rod outer segment length and visual sensitivity.
Investigative Ophthalmology and Visual Science 31(11):2320-2325.
Carey PW. 1992. Fish prey species of the New Zealand fur seal (Arctocephalus forsteri,
Lesson). New Zealand Journal of Ecology 16(1):41-46.
181.
Claes JM, Aksnes DL, Mallefet J. 2010. Phantom hunter of the fjords: camouflage by
counterillumination in a shark (Etmopterus spinax). Journal of Experimental
Marine Biology and Ecology 388(1-2):28-32.
Clarke GL, Wertheim GK. 1956. Measurements of illumination at great depths and at
night in the Atlantic Ocean by means of a new bathyphotometer. Deep Sea
Research (1953) 3(3):189-205.
161
Coimbra JP, Trévia N, Marceliano MLV, Andrade-Da-Costa BLS, Picanço-Diniz CW,

Yamada ES. 2009. Number and distribution of neurons in the retinal ganglion
cell layer in relation to foraging behaviors of tyrant flycatchers. The Journal of
Comparative Neurology 514(1):66-73.
Coimbra JP, Nolan PM, Collin SP, Hart NS. 2012. Retinal ganglion cell topography and
spatial resolving power in penguins. Brain, Behavior and Evolution 80(4):254-
268.
Collin SP, Pettigrew JD. 1988a. Retinal topography in reef teleosts. I. Some species
Collin SP, Pettigrew JD. 1988b. Retinal topography in reef teleosts. II. Some species
with prominent horizontal streaks and high-density areae. Brain, Behavior and
Collin SP. 1997. Specialisations of the teleost visual system: adaptive diversity from
shallow-water to deep-sea. Acta Physiologica Scandinavica Supplementum
638:5-24.
Collin SP, Hoskins RV. 1997. Topographic mismatching of the displaced amacrine and
ganglion cell populations in the retina of deep-sea lanternfish. Society for
Neuroscience Abstracts 23(1-2):729.
Collin SP, Hoskins RV, Partridge JC. 1997. Tubular eyes of deepsea fishes: a
comparative study of retinal topography. Brain, Behavior and Evolution 50:
335-357.

deep sea. Journal of Experimental Zoololy 256(S5):66-75.
Curcio CA, Packer O, Kalina RE. 1987. A whole mount method for sequential analysis
of photoreceptor and ganglion cell topography in a single retina. Vision
Research 27(1):9-15.
Kingdom 69(2):409-435.
162
17(4):597-636.
41(1):59-70.
Esposito V, Battaglia P, Castriota L, Finoia MG, Scotti G, Andaloro F. 2009. Diet of

Atlantic lizardfish, Synodus saurus (Linnaeus, 1758) (Pisces: Synodontidae) in
the central Mediterranean Sea. Scientia Marina 73(2):369-376.
Fernald RD. 1990. Teleost vision: seeing while growing. J Exp Zool 256(S5):167-180.
Fite KV. 1973. Anatomical and behavioral correlates of visual acuity in the great horned
owl. Vision Research 13(2):219-230.
Frank TM, Widder EA. 1997. The correlation of downwelling irradiance and staggered
vertical migration patterns of zooplankton in Wilkinson Basin, Gulf of Maine.
Journal of Plankton Research 19(12):1975-1991.
Freckleton RP, Harvey PH, Pagel M. 2002. Phylogenetic analysis and comparative data:
a test and review of evidence. The American Naturalist 160(6):712-726.
Fritsches KA, Marshall NJ, Warrant EJ. 2003. Retinal specializations in the blue marlin:
eyes designed for sensitivity to low light levels. Marine and Freshwater
Research 54(4):333-341.
Garza Gisholt E, Hemmi JM, Hart NS, Collin SP. In press. A comparison of spatial
methods for the construction of topographic maps of retinal cell density. PLoS
ONE: In press.
Glaser EM, Wilson PD. 1998. The coefficient of error of optical fractionator population
size estimates: a computer simulation comparing three estimators. Journal of
Microscopy 192(2):163-171.
Goldsworthy SD, Lewis M, Williams R, He X, Young JW, van den Hoff J. 2002. Diet
of Patagonian toothfish (Dissostichus eleginoides) around Macquarie Island,
South Pacific Ocean. Marine and Freshwater Research 53(1):49-57.
Grafen A. 1989. The phylogenetic regression. Philosophical Transactions of the Royal

Society of London Series B, Biological Sciences 326(1233):119-157.
Green K, Williams R, Burton HR. 1991. The diet of Antactic fur seals during the late
autumn and early winter around Heard Island. Antarctic Science 3(4):359-361.
163
Green K, Williams R, Green MG. 1998. Foraging ecology and diving behaviour of
Macaroni Penguins Eudyptes chrysolophus at Heard Island. Marine Ornithology
26(1-2):27-34.
Harmon LJ, Weir JT, Brock CD, Glor RE, Challenger W. 2008. GEIGER: investigating
evolutionary radiations. Bioinformatics 24(1):129-131.
Harvey PH, Pagel MD. 1991. The comparative method in evolutionary biology. Oxford:
Oxford University Press. 239 p.

945.

Experimental Zoology 270(2):225-231.
Hoang QV, Linsenmeier RA, Chung CK, Curcio CA. 2002. Photoreceptor inner
segments in monkey and human retina: Mitochondrial density, optics, and
regional variation. Visual Neuroscience 19(04):395-407.
Hughes A. 1977. The topography of vision in mammals of contrasting life style:

comparative optics and retinal organisation. In: Crescitelli F, editor. Handbook
of sensory physiology, vol VII/5. Berlin: Springer. p 613-756.
press.
164
Jackson S. 1988. Diets of the White-chinned Petrel and Sooty Shearwater in the
southern Benguela region, South Africa. The Condor 90(1):20-28.

Land MF. 1981. Optics and vision in invertebrates. In: Autrum H, editor. Handbook of
Sensory Physiology, Vol VII/6B. Berlin, Heidelberg, New York: Springer. p
471-492.
Land MF, Nilsson D-E. 2002. Animals eyes: Oxford University Press.
247(5437):155-157.
Litherland L, Collin SP. 2008. Comparative visual function in elasmobranchs: Spatial

arrangement and ecological correlates of photoreceptor and ganglion cell
distributions. Visual Neuroscience 25(04):549-561.
Locket NA. 1970. Retinal structure in a deep-sea fish, Sternoptyx diaphana, Hermann.
Experimental Eye Research 9(1): 22-27.
Locket NA. 1971. Retinal structure in Platytroctes apus, a deep-sea fish with a pure rod
fovea. Journal of the Marine Biological Association of the United Kingdom
51:79-91.
Madison WP, Madison DR. 2011. Mesquite: a modular system for evolutionary
analysis. Version 2.75. http://mesquiteproject.org.
Marshall NB. 1954. Aspects of deep sea biology. London: Hutchinson 380 p.
27:510-523.
165
Tiere 38:521-528.
Munk O, Frederiksen RD. 1974. On the function of aphakic apertures in teleosts.

Videnskabelige meddelelser fra Dansk naturhistorisk forening 137:65-94.
Nicol JAC. 1989. The eyes of fishes: Oxford University Press. 308p.

Pagel M. 1999. Inferring the historical patterns of biological evolution. Nature

401(6756):877-884.
Paradis E, Claude J, Strimmer K. 2004. APE: Analyses of phylogenetics and evolution

in R language. Bioinformatics 20(2):289-290.
Parry M. 2006. Feeding behavior of two ommastrephid squids Ommastrephes bartramii

and Sthenoteuthis oualaniensis off Hawaii. Marine Ecology Progress Series
318:229-235.
Partridge JC, Shand J, Archer SN, Lythgoe JN, Vangroningenluyben W. 1989.

Interspecific variation in the visual pigments of deep-sea fishes. Journal of
Comparative Physiology A, Sensory Neural and Behavioral Physiology
164(4):513-529.
Kingdom 72(1):113-130.
Podrazhanskaya SG. 1993. Feeding habits of Mesopelagic species of fish and estimation
19:79-85.

166

Schieber NL, Collin SP, Hart NS. 2012. Comparative retinal anatomy in four species of
elasmobranch. Journal of Morphology 273(4):423-440.
Slomianka L, West MJ. 2005. Estimators of the precision of stereological estimates: An

example based on the CA1 pyramidal cell layer of rats. Neuroscience
136(3):757-767.
Steinberg RH, Reid M, Lacy PL. 1973. The distribution of rods and cones in the retina
of the cat (Felis domesticus). The Journal of Comparative Neurology
148(2):229-248.
Brauer. Nature 233(5322):623-624.
56(6):1003-1017.
(family myctophidae) off Oregon, USA. Marine Biology 32(1):7-11.

17(4):637-685.
Warrant EJ, Nilsson D-E. 1998. Absorption of white light in photoreceptors. Vision
Research 38(2):195-207.
167
Warrant EJ. 1999. Seeing better at night: life style, eye design and the optimum strategy
of spatial and temporal summation. Vision Research 39(9):1611-1630.
Warrant EJ. 2000. The eyes of deep-sea fishes and the changing nature of visual scenes
with depth. Philosophical Transactions of the Royal Society of London Series B,
Biological Sciences 355(1401):1155-1159.
Warrant EJ. 2004. Vision in the dimmest habitats on Earth. Journal of Comparative
Physiology A, Sensory Neural and Behavioral Physiology 190(10):765-789.
712.

Watanabe H, Kubodera T, Ichii T, Sakai M, Moku M, Seitou M. 2008. Diet and sexual
maturation of the neon flying squid Ommastrephes bartramii during autumn and
spring in the Kuroshio-Oyashio transition region. Journal of the Marine
Biological Association of the United Kingdom 88(2):381-389.
West MJ, Slomianka L, Gundersen HJG. 1991. Unbiased stereological estimation of the
total number of neurons in the subdivisions of the rat hippocampus using the
optical fractionator. The Anatomical Record 231(4):482-497.
Widder EA. 2010. Bioluminescence in the ocean: Origins of Biological, Chemical, and
Biology 93(1):147-156.
168
Chapter 5 Novel intra-ocular filter
CHAPTER 5
Spectral tuning in the eyes of lanternfishes (Myctophidae):

description of a novel sexually dimorphic intra-ocular filter
Fanny de Busserolles1, Nathan S. Hart1, David M. Hunt1,2, Wayne L. Davies1, N. Justin

Marshall3, Michael W. Clark4, Dorothee Hahne4 and Shaun P. Collin1,2.
Australia, Perth, WA, Australia.
2. Lions Eye Institute, The University of Western Australia, Perth, WA, Australia
Australia.
4. UWA Centre for Metabolomics, Metabolomics Australia, The University of Western Australia, Perth,
WA, Australia.
169
5.1. Abstract
Patches of yellow pigment were found in the retina of several lanternfish species
(Myctophidae), all belonging to a single subfamily, the Myctophinae. Using a
multidisciplinary approach (light microscopy, spectrophotometry,
microspectrophotometry, HPLC, molecular biology, modelling), we describe and
investigate the function of this new specialisation. The yellow pigmentation spans the
entire thickness of the outer nuclear layer and occupies a well defined area of the retina.
This specialisation appears to be species-specific, varying in location, shape, size and is
found to be sexually dimorphic in two different species. To our knowledge, this is the
first record of sexual dimorphism in the visual system of any non-primate vertebrate.
This yellow pigmentation acts as a short-wavelength absorbing (long-pass) filter,
maximally absorbing between 356 nm and 443 nm. Species containing the yellow
pigmentation also appear to possess at least two rod visual pigments, one shifted toward
longer wavelengths, from the result of a duplication of the Rh1 opsin gene. We suggest
that this unique pigmentation is associated with the long wave-shifted rod and acts as a
filter to enhance contrast, thereby improving the detection of a more extensive range of
bioluminescent emissions in a specific part of each species’ visual field. The ecological
significance of this yellow pigmentation in myctophids is discussed in relation to its
interspecific variability and its role in the detection of bioluminescent signals in the
deep-sea.
5.2. Introduction
The evolution of visual sensitivity appears to be driven primarily by the spectral
range and intensity of available light within a species’ visual environment. As a result,
different visual pigments are found in the retina of different species, with spectral
sensitivities broadly matching the surrounding light conditions (Lythgoe and Partridge,
1989). Spectral adaptation to the photic environment is achieved by two types of tuning
mechanisms: variation in the number of spectral classes of cone photoreceptors through
loss or duplication of opsin genes, and variation in the type of chromophore within the
photoreceptor outer segment. Visuals pigments in vertebrates are composed of a
chromophore (a derivative of vitamin A1 or A2) attached to an opsin protein, the
combination of the two determining the absorption spectrum of the visual pigment.
There are five different classes of pigment with different spectral sensitivities each
coded by five distinct opsin genes families: rod opsin or Rh1 (maximally sensitive in the
170
green from about 460 nm - 530 nm), long wave sensitive opsin or LWS (red-green, 490
nm - 570 nm), middle-wave sensitive opsin or Rh2 (green, 480 nm - 535 nm) and two
short-wave sensitive opsins or SWS1 and SWS2 (UV-violet, 355 nm - 440 nm and blue-
violet, 410 nm - 490 nm, respectively, (Bowmaker, 2008). At the level of the
photoreceptor, a spectral shift can result from one or several amino acid substitutions in
the opsin protein or from a switch in chromophore (Bowmaker and Hunt, 2006;
Bowmaker, 2008; Davies et al., 2012, for reviews).
Due to the physical properties of water, aquatic animals can be found in a wide
range of spectral environments, from clear blue oceanic water to turbid brown river
water, and as a result possess different sets of cone classes, ranging from a single class
(several species of sharks, Hart et al., 2011) to at least one of each class (Australian
lungfish, Bailes et al., 2007). In the open ocean, light conditions vary greatly with
depth, decreasing in intensity and spectral range as the short and long wavelengths are
attenuated. In the deep-sea, only very low intensities of sunlight in the blue-green range
remain below 200 m, creating a relatively dark environment accompanied by a
multitude of bioluminescent emissions. Bioluminescent emissions, which are also
mainly concentrated in the blue-green part of the visible light spectrum (Widder, 2002,
2010), are thought to play important roles, subserving behavioural interactions with
prey, predators and congeners (Herring, 2002; Haddock et al., 2010).
Deep-sea teleost fishes are a good example of spectral sensitivity adaptation.

Most species have adapted to their dim-light environment by losing their cones (photic
vision) in favour of a single type of rod photoreceptor (involved in scotopic vision)
containing a single visual pigment, encoded by the rod opsin gene Rh1. While most
terrestrial and shallow water vertebrates usually have an Rh1 visual pigment maximally
absorbing around 500 nm, deep-sea teleosts have shifted the spectral sensitivity of their
rod pigment towards shorter wavelengths to spectrally match their ambient photic
environment, i.e. maximal absorbance around 480 nm (Crescitelli, 1990; Douglas and
Partridge, 1997; Douglas et al., 2003). The tuning mechanism responsible for the
spectral shift toward shorter wavelengths in the Rh1 gene has been indentified at the
molecular level and occurs as a result of different combinations of substitutions at eight
different amino acid sites depending on the species (Hunt et al., 2001). An additional
tuning site has also been identified, in species possessing an additional visual pigment,
171
shifting the peak spectral sensitivity toward longer wavelengths i.e. 520 nm in
dragonfishes (Hunt et al., 2001).
Members of the lanternfish family (Myctophidae), one of the most abundant

groups of mesopelagic fishes in the world’s oceans (Hulley, 1981), tend to possess a
single visual rod pigment with a peak spectral absorption tuned to the blue-green region
of the visible spectrum (Partridge et al., 1992; Douglas and Partridge, 1997; Douglas et
al., 1998; Hasegawa et al., 2008; Turner et al., 2009). However, the presence of two
visual pigments was found in a few species based on the extract spectrophotometry
analysis of the photoreceptors present in adults (Hasegawa et al., 2008; Turner et al.,
2009).
Myctophids are present worldwide and play a major role in oceanic ecosystems
by transferring energy to deeper levels through their daily vertical migrations. Like most
mesopelagic organisms, lanternfishes are bioluminescent and possess two kinds of
bioluminescent structures: the photophores found on the ventral and ventrolateral parts
of the body, and the luminous organs and tissue patches present on the head, body
and/or tail. While the photophores are thought to play a major role in camouflage by
counter-illuminating the underside of the body (Case et al., 1977), luminous organs are
thought to play several different roles in intra- and interspecific communication,
distraction or illumination (Edwards and Herring, 1977). The Myctophidae is
characterized by broad intra- and interspecific variation in the depths they occupy
(Karnella, 1987), their migration patterns (Watanabe et al., 1999) and the location of
their luminous organs (Nafpaktitis and Nafpaktitis, 1969; Paxton, 1972; Nafpaktitis et
al., 1977; Nafpaktitis, 1978; Zahuranec, 2000; Herring, 2007; de Busserolles et al.,
2013). A broad intra- and interspecific diversity in the visual system of this large group
of fishes has also been revealed recently (de Busserolles et al., 2013; Chapters 2 - 4).
Like most mesopelagic fishes, myctophids possess several visual adaptations for
life in the mesopelagic zone that serve to increase the sensitivity of the eye and optimise
photon capture. These include well-developed eyes (Chapter 3), an aphakic gap (Lawry,
1974; Chapter 3), a tapetum lucidum (O'Day and Fernandez, 1976; Chapter 3), a high
density of photoreceptors (Vilter, 1951; Pankhurst, 1987; O'Day and Fernandez, 1976;
Chapter 4) and a pure rod retina (Vilter, 1951; Pankhurst, 1987; O'Day and Fernandez,
1976; Chapter 3). In this study, we describe a new specialisation, a yellow pigment,
172
found in the retina of several species of lanternfishes and explore its possible roles in
the visual system.

5.3.1 Samples
Samples were collected in the Coral Sea during three cruises on the RV Cape
Ferguson (AIMS, Townsville, Australia) in Autumn-Winter 2010-2012. Sampling was
performed at the surface (>5 m) during night time using an Isaacs-Kid Midwater Trawl
fitted with buoys. For the first two cruises, sampling was carried out under the
following collection permits: Coral Sea waters (CSCZ-SR-20091001-01),
Commonwealth waters (AU-COM2009051), GBRMPA (G09/32237.1) and Queensland
Fisheries (133805), (Marshall, AEC # SNG/080/09/ARC). For the third cruise,
sampling permits was obtained by the Chief Scientist (AIMS) for their target species.
For this particular Cruise, samples were donated and therefore no UWA collection
permits were required. Most individuals caught were dead; however, moribund or live
animals were humanely euthanized following the guidelines of the NH&MRC
Australian Code of Practice, under our University of Western Australia Animal Ethic
protocol (RA/3/100/917). For each individual, the standard length and rostro-caudal eye
diameter were measured with digital callipers (to a precision of 0.1 mm) prior to
dissection and fixation (Table 5.1). Eyes were then enucleated, the cornea and lens were
dissected free from the eye cup and tissue fixed specifically for different analyses (4%
paraformaldehyde, RNALater, liquid nitrogen). One individual of Myctophum
brachygnathum, sampled in Hawaii in May 2011 and fixed in 5% formalin, was
provided by the American Museum of Natural History, New York, USA (Table 5.1).
5.3.2. Retinal wholemounts and spectrophotometry

For each species, eyes preserved in 4% paraformaldehyde in 0.1M sodium
phosphate buffer (pH 7.4, PFA) were dissected to reveal the location and shape of the
yellow pigmented patches of retinal tissue. Retinal wholemounts were made according
to standard protocols (Stone, 1981; Coimbra et al., 2006; Ullmann et al., 2011). Briefly,
radial cuts were performed in order to flatten the eye cup. The orientation of the eye was
maintained by making a small additional cut in the nasal or dorso-nasal part of the eye.
The sclera, pigment epithelium and tapetum were gently removed with the help of
forceps and paint brushes, and the retina wholemounted photoreceptor side up in 0.1M
173
phosphate buffer (pH 7.4) between two No. 1 glass coverslips. Prior to analysis, images
of the retinal wholemounts illuminated from above with white light were taken using a
Nikon SMZ 745T or SMZ800 stereomicroscope (Nikon, Tokyo, Japan) fitted with a
Leica EC3 digital camera (Leica Microsystems, Wetzlar, Germany). Images presented
here show the real colours where only brightness, contrast and exposure were adjusted
(for the entire image) using Adobe Photoshop CS4 (Adobe System Inc., USA). Images
of the wholemounts were also taken while transilluminated with either 420 nm or 450
nm light generated by placing narrowband interference filters (Edmund Industrial
Optics, Barrington, NJ, USA) over a broadband tungsten source in the base of the
stereomicroscope in order to better visualise the extent of the yellow pigment by
increasing the contrast. For display purposes, the image background of each
wholemount was deleted using the ‘extract tool’ in Adobe Photoshop CS4. For one
specimen of Symbolophorus rufinus, the eye was dissected on board ship directly after
capture and the retina wholemounted fresh in between two No. 1 coverslips and kept at -
20°C in a light-tight container until analysis was performed at The University of
Western Australia.
Table 5.1. Summary of the individual analysed in this study that presented a yellow patch of
retinal tissue. For each individual, the sex (F = female, M = male, J = juvenile), standard length
(SL, in mm), eye diameter (eye ø, in mm) and location of sampling is given.
Species Sex SL Eye ø Sampling location

Myctophum brachygnathum F 65.7 6.8 Coral Sea
F 67.7 7.2 Coral Sea
M 66.6 7.3 Coral Sea
M 58.1 7.3 Hawaii
M. nitidulum F 85.4 7.1 Peru-Chile Trench
F 80.0 6.9 Peru-Chile Trench
M 77.9 6.9 Peru-Chile Trench
M. Obtusirostre M 92.4 10.6 Coral Sea
M. lychnobium F 106.1 11.1 Coral Sea
M. spinosum F 87.5 8.7 Coral Sea
M. aurolaternatum J 57.8 5.3 Coral Sea
Symbolophorus rufinus J 73.7 6.6 Coral Sea
J 62.5 6.1 Coral Sea
S. evermanni J 65.8 6.7 Coral Sea
Gonichthys tenuiculus M 49.4 3.5 Peru-Chile Trench
? 43.1 2.9 Peru-Chile Trench
? 40.6 2.7 Peru-Chile Trench
Hygophum proximum J 43.3 5.2 Coral Sea
174
Retinal wholemount preparations were then placed onto the stage of a modified
Zeiss Universal microscope and the transmission of light (300-800 nm) through the
retina was measured systematically every 0.5 mm or 1 mm (depending on the size of the
retina) across the wholemount. Light produced by a 175 W Xenon Arc lamp (Spectral
Products, Putnam, CT, USA) was delivered to the retina via a 200 μm diameter quartz
fibre optic (Ocean Optics, Dunedin, FL, USA) and focused into a beam approximately
1 mm in diameter using a fused silica 74-UV collimating lens (Ocean Optics). Light
transmitted through the retina was collected by a second lens situated immediately
below the stage (which was covered with a Teflon diffuser to capture off-axis scattered
light) and delivered via a 1000 μm diameter quartz fibre optic to a S2000 CCD (charge-
coupled device) Spectrometer (Ocean Optics) connected to a Toshiba PC laptop running
Windows 98. At the start and periodically throughout, a ‘dark’ scan was made with all
light blocked from entering the collecting lens connected to the spectroradiometer. This
was done to correct for drift and electrical noise in the S2000. At the start of a row of
sampling points, a baseline or ‘reference’ scan was made outside the retina, which was
subtracted by the S2000 operating software (OOIbase32, Ocean Optics) from the
sample transmission scan recorded at each sampling point within the retina along the
row. Each scan (reference or sample) was the average of 50 individual scans (averaging
performed by the S2000 acquisition software). Percentage transmission spectra were
converted to decadic absorbance spectra offline and further analysed in Microsoft Excel
2007. A ‘corrected’ absorbance spectrum was calculated by subtracting, from each
sample spectrum, a mean reference spectrum made from five ‘white’ absorbance sample
scans measured in the far periphery of the retina.
Each of the corrected absorbance spectra were fitted with an 11-point (approx.
equivalent to 3 nm) unweighted running average to smooth the data. To account for
minor variations in baseline absorbance, the corrected absorbance at the wavelength of
maximum absorbance of the yellow pigment (λYPmax) for each sampling site was
obtained by subtracting the absorbance at 750 nm (where absorbance by the yellow
pigment is negligible) from the absorbance at the wavelength of λYPmax for that species.
The λYPmax value varied between species and was defined as the wavelength of peak
absorbance of the corrected sample scan with the highest absorbance in that species.
The λYPmax values were then mapped onto an outline of the retinal wholemount traced
from the calibrated digital image in Adobe Illustrator CS4. The whole file was saved as
a scalable vector graphics (.svg) file and imported into R v.2.15.0 (R Foundation for
175
Statistical Computing 2012) to construct the maps using custom scripts written by
Garza-Gisholt et al. (in press). Garza Gisholt et al. (in press) proposed several
smoothing models to construct the iso-density maps and, for this study, we chose to use
the Gaussian Kernel Smoother from the Spatstat package (Baddeley and Turner, 2005).
For each map, the sigma value was adjusted to 30.
5.3.3. Cryosections
The retina of one individual of Gonichthys tenuiculus, preserved in 4% PFA,
was processed for cryosectioning in order to visualise the position of the yellow
pigmentation within the retina. A piece of retina containing the yellow pigment was
cryoprotected by sequential incubations (5 mins each) in a graded series of sucrose
solutions (10%, 20% and 30% in 0.1M PB). After a 24 h incubation at 4°C in 30%
sucrose, the sample was embedded in OCT Tissue-Tek medium (Sakura Finetek, CA,
USA) and 40 μm sections were cut using a Leica CM1900 cryostat-microtome. Sections
were then mounted in VectaMount Aqueous medium (Vector Laboratories, CA, USA)
and observed under an Olympus BX50 compound light microscope and pictures taken
with an Olympus DP70 digital camera.
5.3.4. HPLC analysis

High performance liquid chromatography (HPLC) coupled with mass
spectrometry was used in an attempt to identify the composition of the yellow
pigmentation in the retina of the lanternfishes. Yellow pigments have previously been
observed in the eyes of vertebrates, either as a diffuse pigment within the lens and
cornea or within the inner segment of the photoreceptors i.e. in oil droplets or ellipsoids
(Muntz, 1976; Appleby and Muntz, 1979; Goldsmith et al., 1984; Ohtsuka, 1985; Collin
et al., 2003; Siebeck et al. 2003; Bailes et al. 2006). The yellow pigments have been
identified as tryptophan derivatives (van Heyningen, 1971a, b; Thorpe et al., 1992) or
mycosporine-like amino acids (Thorpe et al., 1993) in the lens and as carotenoids
(Goldsmith et al., 1984) in the retina. As a result, we tried to extract the yellow pigment
found in the retina of our lanternfishes following classic extraction protocols using
standards to these previously described compounds.
Retinae of Myctophum aurolaternatun and Hygophum proximum, both possess a

yellow pigmentation and were therefore dissected while onboard ship, deep-frozen in
liquid nitrogen in small eppendorf tubes and stored at -20°C until further analyses back
176
in the laboratory. Extraction of possible tryptophan derivatives and amino acids was
performed as described by Thorpe et al. (1992, 1993). The retina of M. aurolaternatum
was thawed at room temperature and homogenised in a glass tube with 1 ml of distilled
water (3 mins at 30Hz). The process was repeated several times without success and the
yellow pigmentation could not be extracted from the tissue with water. Extraction of
possible carotenoids was also performed as described by Toomey and McGraw (2007,
2009). The retina of H. proximum was thawed and 10 μl of Trans-β-Apo-8-Carotenal
(1ng/μl, Sigma Aldrich) and 50 μl of Retinyl Acetate (1 ng/μl, Sigma Aldrich) were
added to the sample as internal standards. The sample was then homogenised with 1ml
of methanol for 3 mins at 30 Hz, centrifuged (3 mins at 13000 rpm) and the supernatant
transferred into a glass tube. The procedure was repeated twice with 1 ml of
1:1 hexane:tert-butyl methyl ether (MTBE) and the supernatants combined. Since the
yellow pigment was still present in the tissue and not in the supernatant, we repeated the
procedure two more times with 1 ml of dimethyl sulfoxide (DMSO) as described by
Sedmak et al. (1990). Although we were doubtful of the success of the extraction at this
stage, we decided to proceed with the analysis.
Following the protocol described by Toomey and McGraw (2009), the combined
supernatants were evaporated to dryness under a stream of nitrogen and saponified for
4 h at room temperature with 2 ml of 0.2M potassium hydroxide in methanol under
gentle agitation. Volumes of 1 ml of saturated sodium chloride in water, 2 ml of water
and 2 ml of 1:2 hexane:MTBE were then added and the sample was vortexed for 1 min
and subsequently centrifuged for 10 mins at 3000 rpm. Finally, 1.5 ml of the organic
layer was dried down in a brown high recovery HPLC vial under a gentle stream of
nitrogen and reconstituted in 100 μl of 90% methanol with 0.1% formic acid.
HPLC-MS analysis was carried out on an Agilent 1290 Infinity LC System

coupled with an Agilent 6460 Triple Quad LC-MS System using electrospray ionisation
in positive mode and multiple reaction monitoring (MRM). System control and data
processing was performed using Agilent MassHunter Workstation Software. Shortly
after extraction, 10 μl of our reconstituted extract was injected onto a GraceSmart RP 18
column (100 mm length, 4.6 mm i.d., 3 μm particle size) in the x skip. An isocratic
gradient of 90% solvent A (90% methanol, 10% water) and 10% solvent B (MTBE) was
maintained from time 0 to 14 mins, 50% solvent A / 50% solvent B from 15 to 25 mins
and 90% solvent A /10% solvent B until 35 mins. The HPLC flow rate was 0.5 ml/min
177
and the column temperature was maintained at 20°C. Data was collected in full scan
mode from m/z 50-600 and in MRM mode. Pure standards of beta-Carotene (Sigma-
Aldrich), astaxanthin (AK Scientific), xanthophylls (AK Scientific) and retinol (Sigma-
Aldrich) were injected to confirm retention times and MRM transitions of the
compounds.
5.3.5. Microspectrophotometry
The eyes of three species of lanternfish, Myctophum obtusirostre, M. spinosum
and Symbolophorus evermanni, were preserved for microspectrophotometry (MSP) as
described by Hart et al. (2011). Specimens were collected alive and were dark adapted
for 15 mins before being euthanized following the guidelines of the NH&MRC
Australian Code of Practice, under the University of Western Australia Animal Ethics
protocol (RA/3/100/917). Eyes were then enucleated in the dark under dim red light, the
cornea removed and the whole eye lightly fixed for 30 seconds in 2.5% glutaraldehyde
in phosphate buffered saline (PBS, 340 mOsmol kg-1, pH 7.0), washed in PBS for 30 s
and stored in PBS at 4°C in a light-tight container until they were analysed back in the
laboratory, 3 to 4 weeks later.
In all cases, small pieces (ca. 1x1 mm to 3x3 mm) of retina were mounted
between No. 1 glass coverslips in a drop of 425 mOsmol kg-1 PBS containing 10%
dextran (MW 282,000; Sigma D-7265). Absorbance spectra (330–800 nm) were made
using a computer-controlled single-beam wavelength-scanning microspectrophotometer
(Hart et al., 2004, 2011, 2012). Due to the very small diameter of the rod photoreceptors
of the lanternfish species analysed (<1 μm, Chapter 4), longitudinal “end-on”
absorbance spectra were measured from patches of rod outer segments. A sample scan
was made by aligning the measuring beam (typical dimensions 10 ×10 μm) within a
patch of rods in a specific part of the retina and recording the amount of light
transmitted at each wavelength across the spectrum. Baseline scans were made in an
identical way to sample scans but from a cell-free area of the preparation. The
transmittance (ratio of sample to baseline signal) of the outer segments was calculated at
each wavelength and converted to absorbance to give a prebleach spectrum. Each patch
of rods was then bleached with broadband white light for 2 min and subsequent sample
and baseline scans were made to create a postbleach spectrum. Because we were
measuring the absorbance through a patch of retina that contains other absorbing
material (i.e. other retinal layers, yellow pigment) and not through a single rod outer
178
segment, a difference spectrum was calculated by subtracting the postbleach scan from
the prebleach scan. Only spectra that satisfied established selection criteria (Hart et al.,
1998; Levine and MacNichol Jr, 1985) were retained for further analysis. Difference
absorbance spectra were analysed as described elsewhere (Govardovskii et al., 2000;
Hart, 2002; MacNichol Jr, 1986) to provide an estimate of the wavelength of maximum
absorbance (λmax) of the retinal pigment contained in the rods, assuming that only one
type of rod was present within the patch analysed. The mean λmax of a given patch was
then calculated from these individual λmax values. For display purposes, a mean
difference absorbance spectrum was calculated by averaging acceptable individual
(non-normalised) absorbance spectra.
To investigate the possibility that the photoreceptors of some species contained a

mixture of visual pigment molecules utilising both the A1 and A2 chromophores, mean
prebleach absorbance spectra (smoothed with a variable point unweighted running
average) were fitted iteratively using the Excel Solver function in a custom macro
written by NSH with mixed-chromophore pigment templates to find the combination of
visual pigment λmax (pure A1) and A1/A2 ratio that gave the smallest sums of squares
deviation between the real and modelled spectra between 0.8 and 0.2 normalised
absorbance on the long-wavelength limb of the real spectrum (approximately the same
region used to estimate λmax; Temple et al. 2010). We made the assumption that only a
single type of visual pigment opsin protein was expressed in each rod within the patch
of rods and used established A1 and A2 visual pigment templates (Govardovskii et al.
2000) and a known relationship between the λmax values of A1 and A2 visual pigment
pairs (Parry and Bowmaker 2000).
5.3.6. Molecular analyses

Eye and muscle tissues from Symbolophorus evermanni were preserved for
molecular analyses by immersion in RNALater and 100% alcohol, respectively,
followed by storage at -20oC. Genomic DNA (gDNA) was extracted from the muscle
tissue using the DNeasy Blood & Tissue Kit (Qiagen) and following the protocol
provided by the manufacturer. First-round PCR amplifications utilised the “All-Opsins,
All-Species” (AOAS) degenerate PCR primer set F1/R2 designed by Davies et al.
(2009). A second hemi-nested PCR was carried out with primer sets F1/R1 and F2/R2
(Davies et al. 2009). The sequences of the primers used in this study along with the
annealing temperature used and expected PCR product are listed in Table 5.2.
179
Table 5.2. List of primer sets used in this study along with their annealing temperature (Ann.
temp.) and the expected size of PR product. bp = base paire.
Primer sets Ann. temp. °C PCR product (pb)

AOAS F1 + AOAS R2 45 535
3RACE1 55 635
3RACE2 55 325
Primer sequences (5’ - 3’)

AOAS F1 CGCGAGAGATACATNGTNRTNTGYAARCC
AOAS F2 ATTTTAGAAGGTCTGCCRGWSNTCNTGYGG
AOAS R1 ATTGGTCACCTCCTTYTCNGCYYTYTGNGT
AOAS R2 CCCGGAAGACGTAGATGANNGGRTTRWANA
3RACE1 CTCCTGGGTAATGGCGACCAC
3RACE2 GTCTGCTGGGTGCCGTACGCT
RNA was extracted from eye tissue using the PureLink RNA Mini Kit
(Invitrogen) and cDNA synthesised using the 5’/3’ RACE Kit 2nd generation (Roche)
according to the manufacturer’s instructions. The 3’ ends of the two coding sequences
obtained from gDNA as described above were amplified by RACE using primers
designed from these sequences (Table 5.2).
All PCR amplifications were performed using a HotStarTaq DNA Polymerase

Kit (Qiagen). Following a heat-activation step of 15 mins at 95°C, 45 cycles were
performed with a denaturation step of 30 s at 94°C, an annealing step of 60 s at 45-50°C
(Table 5.2), an extension step of 90 s at 72°C and a final extension of 10 mins at 72°.
Gel-purified (Wizard SV Gel and PCR Clean-up System, Promega) PCR products were
sequenced by the Sanger method.
Two phylogenetic trees, including the Symbolophorus evermanni opsin genes

sequenced in this study, were constructed by neighbour-joining (Saitou and Nei, 1987)
using nucleotide sequences. All sequences were aligned in Clustal Omega (Sievers et
al., 2011) and refined by eye. The parameters for the phylogenetic analysis, carried out
using the MEGA phylogenetic package (Tamura et al., 2007), were pair-wise deletion
and Kimura 2-parameter correction. The degree of support for internal branching was
assessed by bootstrapping with 1000 replicates.
5.3.7. Modelling of the association of the yellow pigment and the visual pigments
The effect of the yellow pigment on the spectral sensitivity of the rods was
modelled for two species, Myctophum obtusirostre and Symbolophorus evermanni. The
180
quantal spectral sensitivity of the rods associated with the yellow pigment was modelled
by multiplying rod visual pigment absorptance by yellow pigment transmittance.
Instead of the actual measured spectra, we used visual pigment templates (Govardovskii
et al., 2000) of appropriate max for the modelling. Normalized absorbance templates
were converted to axial outer segment absorptance by assuming a visual pigment
specific absorbance of 0.013 μm-1 (rhodopsin, Partridge et al., 1988; Turner et al., 2009)
and a rod outer segment length of 39 μm for S. evermanni (Chapter 4) and 45 μm for M.
obtusirostre as per a similar species (M. brachygnathum, Chapter 4). In all cases, we
assumed that the yellow pigment was associated with the long wave-shifted rod visual
pigment only. Measurements of ocular media transmittance could not be made from
fresh specimens at sea and so were excluded from the modelling.
5.4. Results
5.4.1. Description of the yellow pigment
A yellow pigmentation within the retinal tissue was noted in 10 species of
lanternfishes from 4 different genera all belonging to a single subfamily, the
Myctophinae (Figure 5.1). After detailed observation of 6 species, this yellow
pigmentation appears to be species-specific, varying in location, shape, size and even
number of patches (Figure 5.2). Observation of the retinal wholemount under a 420 nm
or 450 nm filter, also suggest that the pigmentation is restricted to well-defined areas of
the retina (Figure 5.3). Two of the six species analysed showed two patches of yellow
retinal tissue; Gonichthys tenuiculus possesses two ovoid-like patches, a small one in
the dorsal part of the retina and a larger one in the ventral retina (Figure 5.3A), whereas
Myctophum lychnobium possesses extensive pigment patches covering the entire dorsal
part of the retina and the centro-ventral retina in a streak-like shape (Figure 5.3E). The
four other species all possess a single patch of yellow retinal tissue. While Hygophum
proximum, Symbolophorus rufinus and S. evermanni all have an ovoid-like patch
situated in the dorsal or dorso-nasal part of the retina (Figure 5.3B-D, respectively),
Myctophum obtusirostre possesses a streak-like patch in the temporo-ventral area
(Figure 5.3F). The intensity of the yellow pigmentation appears to vary between species
with some species showing very intense yellow pigmentation (i.e. G. tenuiculus, H.
proximum, S. rufinus, S. evermanni, Figure 5.2A-D) and others paler pigmentation (i.e.
M. lychnobium, M. obtusirostre, Figure 5.2E-F). This is also confirmed by the use of
181
filters to observe the wholemounts since strong yellow pigments are better discerned
with the 450 nm filter and the paler yellow pigmentation is more clearly visible with the
420 nm filter.
Figure 5.1. Phylogenetic trees of Myctophidae reconstructed from Paxton et al. (1984). Ten
species (in red) out of 61 possess yellow patches of retinal tissue. Those 10 species are from 4
different genera and are all from the Myctophinae subfamily. Branch lengths are arbitrarily
ultrametricized on the figure.
182
Figure 5.2. Retinal wholemounts showing the yellow pigment diversity in six species of
lanternfishes, (A) Gonichthys tenuiculus, (B) Hygophum proximum, (C) Symbolophorus rufinus,
(D) S. evermanni, (E) Myctophum lychnobium, (F) M. obtusirostre. Yellow patches of retinal
tissue vary in number, size, location and intensity between species. The dotted lines outlined the
paler yellow patches. The white arrows indicate the orientation of the retina wholemount (T =
temporal, V = ventral). Scale bars: 1 mm.
183
Figure 5.3. Retinal wholemounts highlighting the yellow pigment diversity in six species of
lanternfishes, (A) Gonichthys tenuiculus, (B) Hygophum proximum, (C) Symbolophorus rufinus,
(D) S. evermanni, (E) Myctophum lychnobium, (F) M. obtusirostre, viewed under a 450 nm
filter (A-D) or a 420 nm filter (E-F). Yellow patches of retinal tissue vary in number, size,
location and intensity between species. The black arrows indicate the orientation of the retina
wholemount (T = temporal, V = ventral). Scale bars: 1 mm.
184
Sections through the retina of Gonicthys tenuiculus reveal that this pigmentation
is only situated within the outer nuclear layer (Figure 5.4). The pigmentation appears to
be distributed evenly within the outer nuclear layer and does not appear to be
compartmentalised within any organelles (i.e. oil droplets, ellipsosomes).
Figure 5.4. Light micrograph of a transverse cryosection through the retina of Gonichthys
tenuiculus showing the location and distribution of the yellow pigmentation. OS = outer
segment, IS = inner segment, ONL = outer nuclear layer, INL = inner nuclear layer, GCL =
ganglion cell layer. Scale bar = 50 μm.
5.4.2. Sexual dimorphism

A sexual dimorphism in the size, shape and location of the yellow pigment was
observed in two species, Myctophum brachygnathum and Myctophum nitidulum, out of
the 10 analysed (Figure 5.5). In M. brachygnathum, this sexual dimorphism was
observed in three females and three males. The females possess a small yellow patch in
the central part of the retina just ventral to the optic nerve head (Figure 5.5B). The
males possess a larger yellow patch at the periphery of the temporo-ventral retina
(Figure 5.5B). In Myctophum nitidulum, the sexual dimorphism was observed in two
185
females and one male. In the females, the yellow patch takes the shape of a band
extending from the temporal retina to the ventral retina ventral to the optic nerve head
(Figure 5.5D). In the male, the patch is approximately circular and located in the nasal
periphery (Figure 5.5C). At this stage, it is unknown if the other species with a retinal
yellow pigmentation also possess a sexual dimorphism due to the inability to catch both
sexes in our sampling and/or the presence of immature specimens (juveniles, i.e.
Symbolophorus evermanni, S. rufinus, Table 5.1).
Figure 5.5. Retinal wholemounts showing the sexual dimorphism in the yellow pigment in two
species of lanternfish, Myctophum brachyhnathum (A) male, (B) female, and M. nitidulum
(C) male, (D) female. The yellow patch in the female M. brachygnathum is highlighted by a
dotted line. The white arrows indicate the orientation of the retina wholemount (T = temporal, V
= ventral). Scale bars: 1 mm.
186
5.4.3. Spectrophotometry
Spectrophotometric analysis of the retinal wholemounts reveals that the yellow
pigment acts as a short-wavelength absorbing (long-pass) filter. The maximum
absorbance of the yellow pigment (λYPmax) varies between species from 356 nm
(Myctophum obtusirostre, Figure 5.6F) to 443 nm (Gonichthys tenuiculus, Figure 5.6A).
Figure 5.6. Normalised corrected absorbance spectra of the yellow pigment along with its
maximum absorbance λYPmax and the full-width at half maximum bandwidth (FWHM) in six
lanternfish species, (A) Gonichthys tenuiculus, (B) Hygophum proximun, (C) Symbolophorus
rufinus, (D) Symbolophorus evermanni, (E) Myctophum lychnobium, (F) Myctophum
obtusirostre.
187
Spectrophotometric analysis of the fresh retinal wholemount of Symbolophorus

rufinus also reveals that the fixation process did not alter the absorbance spectra of the
yellow pigment with a difference in the λYPmax between the fresh and fixed samples of
only 3 nm (result not shown).
The full-width at half maximum (FWHM) bandwidth also varies between

species from 89 nm (M. lychnobium, Figure 5.6E) to 146 nm (S. rufinus, Figure 5.6C).
Comparison of the spectra of the yellow pigments from different species, when
overlaid, would suggest that within each species the total absorbance spectrum is given
by an additive mixture of at least two compounds having peak absorbance at
approximately 380 nm and 440 nm (Figure 5.7). M. obtusirostre presents a different
spectrum from the other species, possibly indicating the presence of an additional
compound (Figure 5.7F).
Figure 5.7. Normalised corrected absorbance spectra of the yellow pigment in six lanternfish
species. (A) Gonichthys tenuiculus, (B) Hygophum proximun, (C) Symbolophorus rufinus, (D)
Symbolophorus evermanni, (E) Myctophum lychnobium, (F) Myctophum obtusirostre.
Maps of the corrected absorbance at λYPmax for each species (Figure 5.8)
highlight the yellow pigmented patch and correlate very well with direct visual
observations. The maximum corrected absorbance at λYPmax varied between species
from 0.78 (Symbolophorus evermanni, Figure 5.8D) to 2.02 (Gonichthys tenuiculus,
Figure 5.8A).
188
Figure 5.8. Map of corrected absorbance at the λYPmax for six species of lanternfish, (A)
Gonichthys tenuiculus, (B) Hygophum proximum, (C) Symbolophorus rufinus, (D)
Symbolophorus evermanni, (E) Myctophum lychnobium, (F) Myctophum obtusirostre. The black
arrows indicate the orientation of the retina wholemounts (T = temporal, V = ventral). Scale
bars: 1 mm.
189
5.4.4. HPLC analysis
The extraction of the yellow pigment was very difficult and probably
unsuccessful at this stage. The pigment is very resistant and could not be extracted from
the tissue using standard techniques. In contrast to yellow lenses, we could not extract
the pigment with water, which eliminates the possibility of amino acids and tryptophan
derivatives as the main constituents of the yellow pigment in the retina. The yellow
pigment could not be extracted with non-polar solvents either and HPLC analysis did
not reveal the presence of any carotenoids.
5.4.5. Microspectrophotometry
Microspectrophotometric results for the three species analysed are summarised
in Table 5.3. The mean difference spectrum for each type of rod in each species is
displayed in Figure 5.9. Except for Myctophum spisosum, in which only a single visual
pigment was identified (492 nm, Figure 5.9F), several visual pigments were found for
the two other species.
Table 5.3. Spectral characteristics of rod visual pigments in the retina of three myctophids
species measured using microspectrophotometry (MSP). 1 Maximum corrected absorbance
measured at the max of the bleaching difference spectrum. 2 Variable-point unweighted running
average maximum of the data, which is a measure of the wavelength of peak absorbance of the
pigment that is independent of any assumptions as to the type of chromophore (A1 or A2)
present. Numbers in brackets represent the standard deviation (SD). max values in nm.
M. spinosum M. obtusirostre S.evermanni

1 1 2 1 2 3
Number scan analysed 7 4 9 4 10 6
Chromophore A1 A1 A2 A1 A1 A2
Mean λmax (SD) 492.11 473.35 527.15 475.88 503.41 512.70
(4.31) (1.72) (1.11) (1.23) (1.37) (7.42)
λmax of the mean spectrum 492.37 473.18 526.53 476.02 503.51 511.92
Maximum corrected absorbance1 0.0197 0.0576 0.0553 0.0233 0.1262 0.0617
Running average2 λmax 494 477 528 477 505 510
For Myctophum obtusirostre, two visual pigments were identified with mean
λmax values of 473 nm in the non-yellow retinal area and 527 nm in the area containing
the yellow pigmentation (Figure 5.9 A-B). On the basis of goodness-of-fit to established
A1 and A2 visual pigment templates (Govardovskii et al., 2000), the pigment in the 473
nm rod was considered to represent 100% A1 and the 527 nm pigment 100% A2.
Moreover, given the known relationship between the λmax values of A1 and A2 visual
190
pigment pairs (Parry and Bowmaker 2000) it is probable that these two pigments
originate from two different rod opsins, as the 100% A2 chromophore-constituted
pigment pair of an opsin that has a 100% A1 max of 473 nm would be 484 nm (Parry
and Bowmaker, 2000).
Figure 5.9. Mean bleaching difference absorbance spectra (black lines) with wavelength of
maximum absorbance (λmax) of the rod visual pigments in three species of lanternfish, (A, B)
Myctophum Obtusirostre (A non yellow pigmented area and B yellow pigmented area), (C, D,
E) Symbolophorus evermanni, (E) M. spinosum. Spectra are fitted with visual pigment templates
(grey lines, Govardovskii et al., 2000) of appropriate λmax value (Table 5.3).
191
For Symbolophorus evermanni, three visual pigments were identified with a

mean λmax of 476 nm, 503 nm and 512 nm (Figure 5.9C-E). While the 476 nm and 503
nm pigments were found to best fit vitamin A1-based templates, the 512 nm pigment
seemed to present a mix of A1 and A2 with 47% A1 and 53 % A2. The fact that both 476
nm and 503 nm pigments represent 100% A1 indicates the presence of two different
opsins. It is unlikely that the 512 nm pigment originates from a third opsin although this
cannot be ruled out at this stage. Most probably, the 512 nm pigment comes from the
same opsin as the 503 nm pigment, the difference in wavelength arising from the
presence of an A1 (47%) and A2 (53%) chromophore in the outer segment.
5.4.6. Molecular analyses

Two opsin sequences were PCR-amplified from Symbolophorus evermanni
retinal cDNA. They were aligned with the Rh1 rod and cone opsin coding sequences of
the zebrafish, Danio rerio using Clustal Omaga (Sievers et al., 2011), and the
alignments used to generate a phylogenetic tree (Figure 5.10). Both Symbolophorus
evermanni sequences fall into the same clade as the zebrafish Rh1 sequences, thereby
desmonstrating that both are Rh1 opsins. The two sequences were therefore designated
as Rh1A and Rh1B. Rh1A and Rh1B differ at certain important sites that have previously
been shown to be important for spectral tuning (Hunt et al., 2001, Figure 5.11); Rh1B
possesses Tyr rather than Phe at site 261 and Ala rather than Ser at site 292. These two
substitutions in Rh1B would be expected to result in a long-wavelength shift in the
absorbance peak of the pigment compared to Rh1A, and extrapolation from pigments
with identical substitutions would indicate a red shift of around 20-25 nm (Asenjo et al.,
1994; Yokoyama et al., 1995; Hunt et al., 1996; Sun et al., 1997; Fasick and Robinson,
1998). This is consistent with the different λmax values found by MSP for rod
photoreceptors in this species.
Despite several attempts to complete the sequence of Rh1A, the 5’ end remains
truncated; the missing portion includes two potential tuning sites (sites 83 and 122).
However, since Rh1A reveals a similar spectral sensitivity to the other myctophid rod
opsins (~ 470 nm) which possess Asp83 and Gln122 (Yokoyama et al., 2008), as does
Rh1B (the long wave-shifted rod), it is unlikely that either of these sites in Rh1A are
substituted.
192
Figure 5.10. Phylogenetic tree of opsin gene nucleotide sequences of the zebrafish, Danio rerio,
and the two rod opsins of the lanternfish Symbolophorus evermanni. The tree was constructed
by neighbour-joining (Saitou and Nei, 1987). The bootstrap confidence values are shown for
each branch. The scale bar if calibrated at 0.1 substitutions per site. GenBank accession
numbers for D. rerio: Rh1-A: HM367063, Rh1-B: HM367062, Rh2-1: NM131253, Rh2-2:
NM182891, Rh2-3: NM182892, Rh2-4:NM131254, SWS1: NM131319, SWS2:NM131192,
LWS1:NM131175, LWS2: NM001002443.
Figure 5.11. Amino acids sequences of the two rod opsins identified in Symbolophorus
evermanni (Rod A and Rod B) aligned with Stenobrachium leucopsarus rod opsin sequence
(GenBank accession number: EU407251). Identical residues are indicated by a dot, missing data
by a dash. The black arrows indicate the tuning sites identified in S. evermanni.
Rh1A and Rh1B nucleotide sequences from Symbolophorus evermanni were

aligned with all the myctophid Rh1 opsin sequences available in GenBank, plus the Rh1
opsin gene nucleotide sequences of the short fin pearleye, Scopelarchus analis, the
zebrafish, Danio rerio, the yellow river scaleless carp, Gymnocypris eckloni and the
elephant shark, Callorhinchus milii to serve as out-groups. A neighbour-joining tree of
the aligned nucleotide sequences (Figure 5.12) grouped both Symbolophorus evermanni
193
sequences with those from Electrona antarctica and the two Benthosema species.
Significantly, however, the two Symbolophorus evermanni sequences form a separate
lineage, indicating that the duplication that gave rise to the two Rh1 genes occurred
subsequent to evolution of the genus Symbolophorus.
Figure 5.12. Phylogenetic tree of opsin gene nucleotide sequences showing the origin of the rod
opsin duplication in Symbolophorus evermanni within the Myctophidae. The tree was
constructed by the neighbour-joining (Saitou and Nei, 1987). The bootstrap confidence values
are shown for each branch. The scale bar is calibrated at 0.02 substitutions per site. GenBank
accession numbers: Diaphus metopoclampus JN544536, Diaphus rafinesquii JN412587,
Diaphus watasei JN231003, Stenobrachius leucopsarus EU407251, Lampanyctus alatus
JN412575, Ceratoscopelus warmingii JN412573, Bolinichthys indicus JN412574, Benthosema
suborbitale JN412576, Benthosema pterotum JN231002, Electrona antartica AY141258,
Scopelarchus analis EF517404, Danio rerio HM367063, Gymnocypris eckloni EU606010,
Callorhinchus milii EF565167.
5.4.7. Modelling of the association of the yellow pigment and the visual pigments
From our observations and measurements in yellow and non-yellow pigmented
areas of Myctophum obtusirostre, we assumed that the yellow pigmentation was
associated with the long wave-shifted (LW-S) rod visual pigment only. Modelling of the
association of the yellow pigment with the LW-S rod visual pigment in both species
shows a decrease in absolute sensitivity at the peak, a slight shift of the peak toward
longer wavelengths and a narrowing of the spectral sensitivity function resulting from
the selective absorption of shorter wavelengths (Figure 5.13, 5.14). In M. obtusirostre,
194
the FWHM bandwidth of the LW-S visual pigment narrows from 248 nm to 142 nm,
without and with the yellow pigment, respectively (Figure 5.13). Moreover, the peak of
maximum absorbance of the rods would shift 6 nm toward longer wavelengths in the
presence of the yellow pigment.
Figure 5.13. Modelling of the quantal spectral sensitivity of the two visual pigments (473 nm:
black, 527 nm: grey) measured in Myctophum obtusirostre, (A) without the presence of the
yellow pigmentation and (B) with the yellow pigment associated with the long wave shifted
visual pigment (527 nm).
In contrast to M. obtusirostre, two LW-S rod visual pigments were found in

Symbolophorus evermanni (Figure 5.14A). However, it is unknown at this stage if the
yellow pigment is associated with both LW-S visual pigments or only one of the two.
As a consequence, three different scenarios were modelled and presented here (Figure
5.14B-D). While for the 503 nm visual pigment, the FWHM of the rods would decrease
from 125 nm to 93 nm with the addition of the yellow pigment, it decreases from 149
195
nm to 108 nm for the 512 nm visual pigment. In terms of the wavelength of maximum
absorbance of the rod spectral sensitivity function, a slight shift toward longer
wavelengths would be predicted with the addition of the yellow pigment (10 nm for
rods containing the 503 nm pigment and 8nm for the 512 nm pigment). Depending on
the scenario, S. evermanni could possess rods with peaks of quantal spectral sensitivity
of 476 nm, 513 nm and 520 nm if both of the LW-S visual pigments are associated with
the yellow pigment (Figure 5.14B), of 476 nm, 513 nm and 512 nm (Figure 5.14C) or of
476 nm, 503 nm and 520 nm (Figure 5.14D) if only one of the LW-S visual pigment is
associated with the yellow pigment.
Figure 5.14. Modelling of the effect the yellow pigmentation on the quantal spectral sensitivity
of Symbolophorus evermanni. (A) Quantal spectral sensitivity of the three visual pigments (476
nm: black, 503 nm: dark grey, 512 nm: light grey) without the presence of the yellow
pigmentation. (B-D) Different scenarios where the yellow pigment is associated with the two
the two long wave-shifted visual pigments (B), the 503 nm visual pigment only (C) and the
512 nm visual pigment only (D).
196
5.5. Discussion
5.5.1. The unique yellow pigment in the retina of myctophids
In this study, we describe a new specialisation, a photostable yellow
pigmentation, present in the retina of several species of lanternfishes, all belonging to
the subfamily Myctophinae. Retinal photostable yellow pigmentations have previously
been observed in a number of animals including the lamprey Geotria australis (Collin et
al., 2003), the lungfish Neoceratodus forsteri (Bailes et al., 2006), the ornate lizard
Ctenophorus ornatus (Barbour et al., 2002), and Amazonian cichlid fishes (Muntz,
1973). However, in contrast to these other animals, the yellow pigmentation in
lanternfishes is restricted to a well-defined area of the retina or patch, which appears to
be species-specific in terms of the number, size, location and pigment density of the
patch. Moreover, the location of the yellow pigmentation within the retina of
myctophids differs from what has already been observed previously. In fact, in the
lamprey, G. australis (Collin et al., 2003), Australian lungfish, Neoceratodus forsteri
(Bailes et al. 2006) and lizard, Ctenophorous ornatus (Barbour et al. 2002) the pigment
is found within the distal region of the inner segment and/or diffusely distributed within
the myoid region, while in the myctophids, the photostable yellow pigment is diffusely
distributed throughout the outer nuclear layer.
Another unique feature of the yellow pigmentation in lanternfishes is its sexual

dimorphism, observed so far in two species of the genus Myctophum. Although sexual
dimorphism in the visual system has already been observed in insects (houseflies,
Franceschini et al., 1981; Zeil, 1983; march flies, Hornstein et al., 2000; moth, Meyer-
Rochow and Law, 2008; and butterflies, Arikawa et al., 2005), crustaceans (copepods,
Land 1984, 1988) and primates (Neitz and Jacobs, 1986; Bowmaker et al., 1987), to our
knowledge, this is the first observation of sexual dimorphism in the visual system of any
non-primate vertebrate. Within these examples, sexual dimorphism in the visual system
has been observed at several levels, eye morphology (flies, Franceschini et al., 1981;
Zeil, 1983; moth, Meyer-Rochow and Law, 2008; and copepods, Land, 1984, 1988),
retinal organisation (flies, Zeil, 1983; Hornstein et al., 2000; moth, Meyer-Rochow and
Law, 2008) and spectral sensitivity (butterflies, Arikawa et al., 2005; primates, Neitz
and Jacobs, 1986; Bowmaker et al., 1987). Lanternfishes show a new type of sexual
dimorphism in the visual system, functionally similar to the visual system of the small
white butterfly (Arikawa et al. 2005), which contains a filter in their ommatidia to tune
197
visual sensitivity. However, in contrast to the small white butterfly, both male and
female lanternfishes possess the filter but in different areas of the retina.
The myctophid yellow pigmentation in discrete retinal regions acts as a cut-off

filter, maximally absorbing short wavelengths between 356 nm to 443 nm. Similar
intra-ocular yellow filters have been found in the lenses of some mesopelagic fishes and
squids (380 nm < λYPmax < 480 nm, Somiya, 1976; Muntz, 1976; Douglas and Thorpe,
1992), in the cornea of cichlids, pufferfishes and toadfishes (425 nm < λYPmax < 500
nm, (Muntz, 1973; Appleby and Muntz, 1979; Siebeck et al., 2003 ) and in the retina of
the cichlid Crenicichla lenticulata (λYPmax = 455 nm , Muntz, 1973). All these
intraocular yellow filters usually possess different spectra with different absorbance
maxima indicating the presence of different pigments (i.e. compounds), depending on
the structure in which they are located and/or the species.
Unfortunately, the composition of the yellow pigmentation in myctophids could

not be identified in this study. However, absorbance spectra of the yellow pigmentation
in the six species analysed indicate the existence of at least two distinct components
present in different concentrations depending of the species, and possibly the presence
of a third component (i.e. in Myctophum obtusirostre, Figure 5.7). Coloration in animals
is primarily due to the presence of melanins and carotenoids, although other types of
pigments have also been shown to be responsible for animal coloration (i.e. porphyrins,
pterins, flavins, psittacofulvins, McGraw, 2006; amino acids Thorpe et al., 1993).
Specifically, yellow pigmentation in ocular tissues has been identified as carotenoids (in
pufferfish cornea, Appleby and Muntz, 1979; in avian retinal oil droplets, Goldsmith et
al., 1984; in the human macula, Bone et al., 1985), tryptophan derivatives (in the lens of
terrestrial vertebrates, van Heyningen, 1971a, b; in the lenses of deep-sea fishes, Thorpe
et al., 1992) and mycosporine-like amino acids (in the lenses of deep-sea fishes, Thorpe
et al., 1993). Results from our study appears to exclude all of these previously described
pigment types and since the myctophid yellow pigmentation could not be extracted by
water and non-polar solvents, lipophilic or hydrophilic pigments (i.e. porphyrins and
psittacofulvins) can also be exluded.
As previously mentioned, melanins are often implicated in animal coloration.

Two distinct forms of melanin exist, the brown to black eumelanin and the yellow to red
pheomelanin. Melanin is also known to be difficult to extract from tissues due to its
198
heterogeneous nature, and usually requires separation from its associated protein (Taft,
1949; Liu et al., 2003). Therefore, the yellow colour and difficulty in extracting it from
its parent tissue make pheomelanin a good candidate for the yellow pigmentation in the
retina of lanternfishes. Although pheomelanin has yet to be directly observed in fishes
(Ito and Wakamatsu, 2003), the observation of agouti signalling proteins (ASIP), which
control the production of pheomelanin, in several fishes (goldfish, Cerda-Reverter et al.,
2005; zebrafish and several species of pufferfishes, Klovins and Schiöth, 2005) suggest
that it may be present.
5.5.2. Visual pigments and spectral tuning in myctophids

Microspectrophotometry of the rod photoreceptors in two lanternfish species
with retinal yellow pigmentation indicate the presence of several visual pigments, two
in Myctophum obtusirostre (473 nm and 527 nm) and three in Symbolophorus
evermanni (476 nm, 503 nm and 512 nm). While most lanternfish species possess a
single visual pigment falling between 480 nm and 492 nm (Turner et al., 2009), two
visual pigments have been observed in four other species, Ceratoscopelus warmingii
(468 nm, 488 nm, Douglas et al., 2003), Hygophum proximum (475 nm, 502 nm, Turner
et al., 2009), Myctophum aurolaternatum (485 nm, 495 nm, Turner et al., 2009) and
Myctophum nitidulum (468 nm, 522 nm, Hasegawa et al., 2008), with the last three
species all possessing a retinal yellow pigmentation (this study). The presence of two
spectrally distinct visual pigments in the retina can result either from the presence of
one visual opsin gene associated with different chromophores (A1 and/or A2), or from
the presence of two different opsins. Based on known relationships between the λmax
values of A1 and A2 visual pigment pairs (Whitmore and Bowmaker, 1989; Parry and
Bowmaker, 2000), the presence of two different opsins can be predicted for Hygophum
proximum, Myctophum nitidulum, (Hasegawa et al., 2008; Turner et al., 2009), M.
obtusirostre and Symbolophorus evermanni (this study). This was confirmed at the
molecular level for S. evermanni (this study).
A pure rod retina with a single Rh1 pigment is typical among deep-sea fishes
(Hunt et al., 2001). However, analysis of the rod ospin coding sequences expressed in
Symbolophorus evermanni revealed the presence of two Rh1 opsin genes (Rh1-A and
Rh1-B). This is the second case of a rod opsin duplication recorded for a deep-sea fish,
the first one being the pearleye Scopalarchus analis (Pointer et al., 2007). However, in
contrast to S. analis for which the duplication of Rh1 results in two spectrally similar
199
visual pigments (486 nm and 479 nm, Pointer et al., 2007), the duplication in S.
evermanni gives rise to two spectrally different visual pigments, with peaks at 476 nm
and 503 nm (this study). Moreover, while the two Rh1 genes are phylogenetically
different in S. analis, indicating an early duplication in evolutionary history (Pointer et
al., 2007), Rh1-A and Rh1-B in S. evermanni are phylogenetically similar, forming a
separate lineage subsequent to speciation within the genus Symbolophorus. The fact that
this duplication appeared at a specific stage in the evolutionary history of the family,
only present in the Myctophinae subfamily, could explain why it has not been reported
previously. The characterisation of the two myctophid subfamilies (Lampanyctinae and
Myctophinae) in term of primitive and advanced states of evolution is very problematic
and to date unresolved (Paxton, 1972, Poulsen et al., 2013). Although far from
exhaustive, with only 11 species analysed out of some 250 species, preliminary results
from the phylogenetic analysis of Rh1 genes in Myctophids suggests the Lampanyctinae
are closer to the ancestral state than the Myctophinae.
Rh1 duplication in teleost fishes is quite rare and, in addition to the two deep-sea
species mentioned above, has only been observed in a few cyprinid species including
the zebrafish Danio rerio (Morrow et al., 2011) and a few eel species (Beatty, 1975;
Archer et al., 1995; Zhang et al., 2000). The expression of the second Rh1 gene in
teleosts seems to be associated with a change in the spectral conditions of each species’
habitat (i.e. in the pearleye and the eel). The European eel Anguilla anguilla, for
example, lives in very different environments depending on developmental stage. When
reaching sexual maturity, individuals will move from brackish/freshwater to the deep-
sea, two very different spectral environments and as a result, the level of expression of
their two Rh1 genes changes accordingly (Archer et al., 1995). Myctophids also show
marked differences in habitat depending on their developmental stage with larvae
inhabiting the well-lit surface layers of the ocean (Sabates et al., 2003) and juvenile and
adults occupying the deeper mesopelagic zone. Moreover, larvae and juvenile/adults
also possess different types of photoreceptors, the larvae having both rods and cones
(Bozzano et al., 2007) and the juvenile/adults only rods. As a result, a change in visual
pigment expression is expected between larval and juvenile/adult myctophids. MSP
results show that both Rh1 genes are expressed in the retina of the adult lanternfish,
although, it is unknown at this stage if both genes are also expressed in the larvae. As
mentioned earlier, most deep-sea fishes possess a single visual pigment, specifically
tuned to the spectral properties of the deep-sea environment, i.e. blue-green light
200
(downwelling light and most bioluminescent emissions). However, even though most
bioluminescent emissions are in the blue-green range, several organisms produce longer
wavelengths signals (Widder, 2010), the most extreme case being the red-
bioluminescence produced by some stomiid dragonfishes (Widder et al., 1984). Some of
these far-red illuminating fishes are known to prey on myctophids (Sutton and Hopkins,
1996) and long wave-shifted visual pigment in some of these lanternfishes (i.e. M.
nitidulum and M. obtusirostre) could potentially have evolved to help detect these
predators (Hasegawa et al., 2008; Turner et al., 2009).
5.5.3. Putative functions of the yellow pigment in myctophids

Intraocular filters are thought to have many functions in vertebrates. In well lit
environments, they may improve visual acuity by reducing chromatic aberration and
scatter (yellow lens and cornea, Walls, 1931; Walls and Judd, 1933; Muntz, 1973,
Barbour et al., 2002), enhance the discrimination of colours by narrowing the spectral
sensitivity of the visual pigment (i.e. oil droplets, Hart 2001) or filter out the potentially
damaging short-wavelengths present in bright environments (Collin et al., 2003). In the
deep-sea, only two types of intraocular filters have been observed previously in teleosts:
yellow lenses and the retinal yellow pigmentation found in myctophids. In the
mesopelagic zone, yellow lenses may increase hue discrimination in order to better
visualise bioluminescent flashes or to break down the counterilluminating camouflage
of other species (Muntz, 1976; Somiya, 1976; Somiya, 1982; Douglas and Thorpe,
1992). In fact, during the day in the mesopelagic zone, the simultaneous presence of
downwelling sunlight and bioluminescent flashes might render the detection of both
signals difficult. More particularly, the presence of downwelling sunlight might reduce
the contrast, and therefore the visibility, of bioluminescent emissions which could be
detrimental in terms of survival. However, since most bioluminescent emissions have a
broader spectrum than downwelling light (Herring, 1983), yellow lenses might prevent
this problem by cutting off most of the background illumination thereby accentuating
the signal. Lanternfish species with a yellow retinal pigmentation are found at very
different depths during the day (de Busserolles et al., 2013), from 280 m for Myctophum
brachygnathum (Bourret, 1985) to 1000 m for Myctophum aurolaternatum (Flynn,
2012). However, at night they all migrate towards the surface to depths <4 m (de
Busserolles et al., 2013), where downwelling light produced by the moon and stars
(Johnsen et al., 2004) may reduce the visibility of bioluminescent emissions. Therefore,
in a similar way to yellow lenses, the presence of retinal yellow pigmentation could
201
potentially help myctophids to enhance the contrast of bioluminescent emissions against

the night downwelling light.
Douglas and Thorpe (1992) also suggested that intraocular yellow filters might
occur in species possessing two visual pigments. They based their hypothesis on the fact
that single pigment species, which have pigments maximally absorbing around the
maximum penetration limits of sunlight (468-495 nm, Douglas and Partridge, 1997;
Douglas et al., 1998), will lose some information by not being able to distinguish
between wavelength and intensity. In contrast, species with two visual pigments might
be able to accomplish this, where the yellow pigmentation would increase the
discrimination of different wavelengths of bioluminescent emissions. Our results agree
with this hypothesis since at least two visual pigments were found in species with a
retinal yellow pigmentation (Hasegawa et al., 2008; Turner et al., 2009; this study).
Moreover, from our observations, the yellow pigmentation appears to be associated with
the long wave-shifted rod, clearly narrowing and shifting the spectral sensitivity of the
long wave-shifted pigment, thereby allowing a better discrimination of a range of
wavelengths.
Different species of lanternfishes possess different areas of yellow retinal tissue

that will enhance sensitivity to bioluminescence in different parts of the visual field,
implying interspecific differences in behaviour. Myctophids show a variety of
interspecific differences in behaviour with respect to their depth distribution (Karnella,
1987), migration pattern (Watanabe et al., 1999) and diet (Young and Blaber, 1986;
Tyler and Pearcy, 1975; Podrazhanskaya, 1993; Kozlov, 1995; Ishihara and Kubota,
1997; Pusch et al., 2004; Shreeve et al., 2009). Since all the species which possess a
retinal yellow pigmentation vertically migrate to the surface at night, interspecific
differences in retinal yellow areas could potentially reflect differences in diet. Other
interspecific differences in the visual system of myctophids (i.e. photoreceptor
distribution) have also been linked to differences in food consumption (Chapter 4).
Unfortunately, data on diet are quite limited for lanternfishes and are not yet available
for the species analysed in this study. Moreover, no studies have been conducted on
possible diet differences between sexes in myctophids, which would potentially explain
the observed sexual dimorphism in the retinal yellow pigmentation. Sexual dimorphism
in food choices has been reported in a number of organisms, as a consequence of
different nutritional requirements, food competition, habitat and body size between
202
males and females (Shine, 1989). Sexual dimorphism in size has been observed in
several lanternfish species (Gartner, 1993; Braga et al., 2008; Flynn and Paxton, 2012)
and some differences in distribution between both sexes have been suggested (Flynn
and Paxton, 2012), reinforcing the possible role of the retinal yellow pigmentation in
food choice. The sexual dimorphism in the retinal yellow pigment could also indicate a
role in intraspecific/sexual communication. Several lanternfish species possess sexually
dimorphic luminous tissues (head and caudal luminous organs, luminous patches,
(Herring, 2007; de Busserolles et al., 2013), which are thought to play a role in
intraspecific communication (Edwards and Herring, 1977; Herring, 2000; Herring,
2007). Sexually dimorphic species in yellow pigment also possess sexual dimorphism in
the location and size of their luminous organs, with males presenting a supracaudal
organ and female a smaller infracaudal organ. If the retinal yellow pigmentation is
involved in sexual communication, interspecific differences in shape, size and location
of the retinal yellow pigment could potentially be attributed to differences in luminous
organ size and shape between species, taking in consideration the position of the fish in
the water column (Barham, 1966). The other species possessing a retinal yellow
pigmentation also possess sexually dimorphic caudal luminous organs. However, due to
a lack of samples, it is unknown at this stage if these species also possess a sexual
dimorphism in the yellow pigment (i.e. Myctophum obtusirostre, Table 1) or the
presence of only immature individuals (i.e. Symbolophorus sp, Table 1).
5.5.4. Conclusion
A new retinal specialisation is identified in several species of lanternfishes, all
belonging to the subfamily Myctophinae. This retinal yellow pigmentation is species-
specific, varying in location, shape, size and occurs in species possessing at least two
visual pigments. Although more analyses need to be conducted, we suggest that this
retinal yellow pigmentation is associated with the long wave-shifted rod and acts as a
filter to enhance contrast and improve the detection of a larger range of bioluminescent
organisms (signals). Since the yellow pigmentation is only present in a specific part of
the retina, spectral tuning will only occur in a specific part of the visual field. We
suggest this unique pigmentation plays a role in food choice or discrimination and/or
sexual communication. Further analyses of the retinal yellow pigmentation in both
sexes and at several different life stages will shed more light on the function of this new
visual specialisation.
203
5.6. Acknowledgment
Ferguson for sea time and sampling opportunities. We gratefully acknowledge John
Denton and the American Museum of Natural History, New York, USA for providing
the additional Myctophum brachygnathum specimen and John Paxton (Australian
Museum) for providing most of the fish identification. We thank Prof. Julian Partridge
(University of Bristol) for helpful discussions regarding the yellow pigment absorption
spectra and Eduardo Garza-Gisholt for his help in using his R script for the creation of
the yellow pigment maps. This work was funded by the Australian Research Council
(Discovery grant to SPC, DMH, NSH and the Deep-Australia Linkage Grant to NJM
and SPC) and the West Australian State Government (SPC). FdB was supported by a
Scholarship for International Research Fees (SIRF) and a University International
Stipend (UIS) at the University of Western Australia.
5.7. References
Appleby SJ, Muntz WRA. 1979. Occlusable yellow corneas in Tetraodontidae. The
Journal of Experimental Biology 83(1):249-259.
Archer S, Hope A, Partridge JC. 1995. The molecular basis for the green-blue
sensitivity shift in the rod visual pigments of the European Eel. Proceedings of
the Royal Society of London Series B, Biological Sciences 262(1365):289-295.
Arikawa K, Wakakuwa M, Qiu X, Kurasawa M, Stavenga DG. 2005. Sexual

dimorphism of short-wavelength photoreceptors in the small white butterfly,
Pieris rapae crucivora. The Journal of Neuroscience 25(25):5935-5942.
Asenjo AB, Rim J, Oprian DD. 1994. Molecular determinants of human red/green color
discrimination. Neuron 12(5):1131-1138.
Bailes HJ, Robinson SR, Trezise AEO, Collin SP. 2006. Morphology, characterization,
and distribution of retinal photoreceptors in the Australian lungfish
Neoceratodus forsteri (Krefft, 1870). The Journal of Comparative Neurology
494:381-397.
Bailes HJ, Davies WL, Trezise AEO, Collin SP. 2007. Visual pigments in a living
fossil, the Australian lungfish Neoceratodus forsteri. BMC Evolutionary
Biology 7(200).
204
Barbour HR, Archer MA, Hart NS, Thomas N, Dunlop SA, Beazley LD, Shand J. 2002.
Retinal characteristics of the ornate dragon lizard, Ctenophorus ornatus. The
Journal of Comparative Neurology 450(4):334-344.
Barham EG. 1966. Deep scattering layer migration and composition: observations from
a diving saucer. Science 151:1399-1403.
Beatty DD. 1975. Visual pigments of the american eel Anguilla rostrata. Vision
Research 15(7):771-776.
Bone RA, Landrum JT, Tarsis SL. 1985. Preliminary identification of the human
macular pigment. Vision Research 25(11):1531-1535.
Bourret P. 1985. Poissoins teleosteens: Gonostomatidae, Sternoptychidae, et

Myctophidae (MUSORSTOM II). Memoires du Museum National d'Histoire
Naturelle 133:55-82.
Bowmaker JK, Jacobs GH, Mollon JD. 1987. Polymorphism of Photopigments in the
Squirrel Monkey: A Sixth Phenotype. Proceedings of the Royal Society of
London Series B, Biological Sciences 231(1264):383-390.
Bowmaker JK, Hunt DM. 2006. Evolution of vertebrate visual pigments. Current
Biology 16(13):484-489.
Bowmaker JK. 2008. Evolution of vertebrate visual pigments. Vision Research

48(20):2022-2041.
Braga AC, Costa PAS, Nunan GW. 2008. First record of the firebrow lanternfish
Diaphus adenomus (Myctophiformes : Myctophidae) from the South Atlantic.
Journal of Fish Biology 73(1):296-301.
181.
Cerda-Reverter JM, Haitina T, Schioth HB, Peter RE. 2005. Gene structure of the
goldfish agouti-signaling protein: a putative role in the dorsal-ventral pigment
pattern of fish. Endocrinology 146(3):1597-1610.
Collin SP, Hart NS, Shand J, Potter IC. 2003. Morphology and spectral absorption
characteristics of retinal photoreceptors in the southern hemisphere lamprey
(Geotria australis). Visual Neuroscience 20(02):119-130.
205

deep sea. Journal of Experimental Zoology 256(S5):66-75.
Davies, WL, Cowing, JA, Bowmaker, JK, Carvalho, LS, Gower, DJ, Hunt, DM, 2009.
Shedding light on serpent sight: the visual pigments of henophidian snakes. The
Journal of Neuroscience 29 (23), 7519-7525.
Davies WL, Collin SP, Hunt DM. 2012. Molecular ecology and adaptation of visual
photopigments in craniates. Molecular Ecology 21(13):3121-3158.
Douglas RH, Thorpe A. 1992. Short-wave absorbing pigments in the ocular lenses of
deep-sea teleosts. Journal of the Marine Biological Association of the United
Kingdom 72:93-112.
17(4):597-636.
Douglas RH, Hunt DM, Bowmaker JK. 2003. Spectral sensitivity tuning in the deep-
sea. In: Collin SP, Marshall J, editors. Sensory Processing in Aquatic
Environments: Springer. p 323-342.
41(1):59-70.
Fasick JI, Robinson PR. 1998. Mechanism of spectral tuning in the dolphin visual
pigments. Biochemistry 37(2):433-438.
Flynn AJ. 2012. Ecology and zoogeography of lanternfishes (family Myctophidae). PhD
thesis.: The University of Queensland.
Flynn AJ, Paxton JR. 2012. Spawning aggregation of the lanternfish Diaphus danae
(family Myctophidae) in the north-western Coral Sea and associations with tuna
aggregations. Marine and Freshwater Research 63(12):1255-1271.
Franceschini N, Hardie R, Ribi W, Kirschfeld K. 1981. Sexual dimorphism in a

photoreceptor. Nature 291(5812):241-244.
Gartner JV. 1993. Patterns of reproduction in the dominant lanternfish species (Pisces:
Myctophidae) of the Eastern Gulf of Mexico, with a review of reproduction
among tropical-subtropical Myctophidae. Bulletin of Marine Science 52(2):721-
750.
206
Garza Gisholt E, Hemmi JM, Hart NS, Collin SP. (In press). A comparison of spatial
ONE:In press.
Goldsmith TH, Collins JS, Licht S. 1984. The cone oil droplets of avian retinas. Vision
Research 24(11):1661-1671.
Govardovskii VI, Fyhrquist N, Reuter T, Kuzmin DG, Donner K. 2000. In search of the
visual pigment template. Visual Neuroscience 17(04):509-528.
Hart NS, Partridge J, Cuthill I. 1998. Visual pigments, oil droplets and cone
photoreceptor distribution in the european starling (Sturnus vulgaris). Journal of
Experimental Biology 201(9):1433-1446.
Hart NS. 2002. Vision in the peafowl (Aves: Pavo cristatus). Journal of Experimental
Biology 205(24):3925-3935.
Hart NS, Lisney TJ, Marshall NJ, Collin SP. 2004. Multiple cone visual pigments and
the potential for trichromatic colour vision in two species of elasmobranch.
Journal of Experimental Biology 207:4587-4594.
Hart NS, Theiss SM, Harahush BK, Collin SP. 2011. Microspectrophotometric evidence
for cone monochromacy in sharks. Naturwissenschaften 98(3):193-201.
Hart NS, Coimbra JP, Collin SP, Westhoff G. 2012. Photoreceptor types, visual
pigments, and topographic specializations in the retinas of hydrophiid sea
snakes. The Journal of Comparative Neurology 520(6):1246-1261.

945.
Herring PJ. 1983. The spectral characteristics of luminous marine organisms.

220(1219):183-217.
Herring PJ. 2000. Species abundance, sexual encounter and bioluminescent signalling in
the deep sea. Proceedings of the Royal Society of London Series B, Biological
Sciences 355(1401):1273-1276.
207
Hornstein EP, O'Carroll DC, Anderson JC, Laughlin SB. 2000. Sexual dimorphism
matches photoreceptor performance to behavioural requirements. Proceedings of
the Royal Society of London Series B, Biological Sciences 267(1457):2111-
2117.
Hunt DM, Fitzgibbon J, Slobodyanyuk SJ, Bowmakers JK. 1996. Spectral tuning and
molecular evolution of rod visual pigments in the species flock of cottoid fish in
Lake Baikal. Vision Research 36(9):1217-1224.
Hunt DM, Dulai KS, Partridge JC, Cottrill P, Bowmaker JK. 2001. The molecular basis
for spectral tuning of rod visual pigments in deep-sea fish. Journal of
Ito S, Wakamatsu K. 2003. Quantitative analysis of eumelanin and pheomelanin in

humans, mice, and other animals: a comparative review. Pigment Cell Research
16(5):523-531.

Klovins J, Schiöth HB. 2005. Agouti-related proteins (AGRPs) and agouti-signaling

peptide (ASIP) in fish and chicken. Annals of the New York Academy of
Sciences 1040(1):363-367.
Land MF. 1984. Crustacea. In: Ali MA, editor. Photoreception and Vision in
Invertebrates. p 401-438.
Land MF. 1988. The functions of eye and body movements in Labidocera and other
copepods. Journal of Experimental Biology 140(1):381-391.
247(5437):155-157.
Levine JS, MacNichol Jr EF. 1985. Microspectrophotometry of primate photoreceptors:

art, artefact and analysis. In: Fein A, Levine Js, editors. The visual system. New
York: Liss. p 73-87.
208
Liu Y, Kempf VR, Brian Nofsinger J, Weinert EE, Rudnicki M, Wakamatsu K, Ito S,
Simon JD. 2003. Comparison of the structural and physical properties of human
hair eumelanin following enzymatic or acid/base extraction. Pigment Cell
Research 16(4):355-365.
Lythgoe J, Partridge J. 1989. Visual pigments and the acquisition of visual information.
Journal of Experimental Biology 146(1):1-20.
MacNichol Jr EF. 1986. A unifying presentation of photopigment spectra. Vision

Research 26(9):1543-1556.
McGraw KJ. 2006. Mechanics of uncommon colors: pterins, porphyrins and

psittacofulvins. Bird Coloration 1:354-398.
Meyer-Rochow VB, Lau TF. 2008. Sexual dimorphism in the compound eye of the
moth Operophtera brumata (Lepidoptera, Geometridae). Invertebrate Biology
127(2):201-216.
Morrow JM, Lazic S, Chang BSW. 2011. A novel rhodopsin-like gene expressed in
zebrafish retina. Visual Neuroscience 28(04):325-335.
Muntz WRA. 1973. Yellow filters and the absorption of light by the visual pigments of
some amazonian fishes. Vision Research 13(12):2235-2254.
Muntz W. 1976. On yellow lenses in mesopelagic animals. Journal of the Marine

Biological Association of the United Kingdom 56:963-976.

Research. p 13-265.

Neitz J, Jacobs GH. 1986. Polymorphism of the long-wavelength cone in normal human
colour vision. Nature 323(6089):623-625.

Ohtsuka T. 1985. Relation of spectral types to oil droplets in cones of turtle retina.
Science 229(4716):874-877.
209
Parry JWL, Bowmaker JK. 2000. Visual pigment reconstitution in intact goldfish retina
using synthetic retinaldehyde isomers. Vision Research 40(17):2241-2247.
Partridge JC, Archer SN, Lythgoe JN. 1988. Visual pigments in the individual rods of
deep-sea fishes. Journal of Comparative Physiology A: Sensory Neural and
Behavioral Physiology 162(4):543-550.
Kingdom 72(1):113-130.

13:1-81.
Podrazhanskaya SG. 1993. Feeding habits of Mesopelagic species of fish and estimation
19:79-85.
Pointer MA, Carvalho LS, Cowing JA, Bowmaker JK, Hunt DM. 2007. The visual
pigments of a deep-sea teleost, the pearl eye Scopelarchus analis. Journal of


Sabates A, Bozzano A, Vallvey I. 2003. Feeding pattern and the visual light
environment in myctophid fish larvae. Journal of Fish Biology 63(6):1476-1490.
Saitou, N., Nei, M., 1987. The neighbor-joining method: a new method for
reconstructing phylogenetic trees. Molecular Biology and Evolution 4 (4), 406-
425.
Sedmak, J.J., Weerasinghe, D., Jolly, S., 1990. Extraction and quantitation of
astaxanthin from Phaffia rhodozyma. Biotechnology Techniques 4 (2), 107-112.
Shine R. 1989. Ecological causes for the evolution of sexual dimorphism: a review of
the evidence. Quarterly Review of Biology 64(4):419-461.
210
Siebeck UE, Collin SP, Ghoddusi M, Marshall NJ. 2003. Occlusable corneas in
toadfishes: light transmission, movement and ultrastruture of pigment during
light- and dark-adaptation. Journal of Experimental Biology 206(13):2177-2190.
Sievers F, Wilm A, Dineen D, Gibson TJ, Karplus K, Li W, Lopez R, McWilliam H,

Remmert M, Soding J, Thompson JD, Higgins DG. 2011. Fast, scalable
generation of high-quality protein multiple sequence alignments using Clustal
Omega. Molecular Systems Biology 7:539.
Somiya H. 1976. Functional significance of the yellow lens in the eyes of

Argyropelecus affinis. Marine Biology 34(2):93-99.
215(1201):481-489.
Sun H, Macke JP, Nathans J. 1997. Mechanisms of spectral tuning in the mouse green
cone pigment. Proceedings of the National Academy of Sciences 94(16):8860-
8865.
Sutton TT, Hopkins TL. 1996. Trophic ecology of the stomiid (Pisces: Stomiidae) fish
assemblage of the eastern Gulf of Mexico: Strategies, selectivity and impact of a
top mesopelagic predator group. Marine Biology 127(2):179-192.
Taft EB. 1949. Melanin solubility in tissue sections. Nature 164:1133-1134.
Tamura, K., Dudley, J., Nei, M., Kumar, S., 2007. MEGA4: Molecular Evolutionary
Genetics Analysis (MEGA) software version 4.0. Molecular Biology and
Evolution 24, 1596-1599.
Thorpe A, Truscott RJW, Douglas RH. 1992. Kynurenine identified as the short-wave
absorbing lens pigment in the deep-sea fish Stylephorus chordatus.
Experimental Eye Research 55(1):53-57.
Thorpe A, Douglas RH, Truscott RJW. 1993. Spectral transmission and short-wave
absorbing pigments in the fish lens - I. Phylogenetic distribution and identity.
Vision Research 33(3):289-300.
Toomey, M.B., McGraw, K.J., 2007. Modified saponification and HPLC methods for
analyzing carotenoids from the retina of quail: implications for its use as a
nonprimate model species. Investigative Ophthalmology and Visual Science 48
(9), 3976-3982.
Toomey, M.B., McGraw, K.J., 2009. Seasonal, sexual, and quality related variation in
retinal carotenoid accumulation in the house finch (Carpodacus mexicanus).
Functional Ecology 23 (2), 321-329.
211
56(6):1003-1017.
(family myctophidae) off oregon, USA. Marine Biology 32(1):7-11.
wholemount tTechnique: a window to understanding the brain and behaviour.
van Heyningen, R., 1971a. Fluorescent glucoside in the human lens. Nature 230:393-
394.
van Heyningen, R., 1971b. Fluorescent derivatives of 3-hydroxy-L-kynurenine in the

lens of man, the baboon, and the grey squirrel. Biochemical Journal 123:30-31.

Walls GL. 1931. The occurrence of colored lenses in the eyes of snakes and squirrels,
and their probable significance. Copeia 1931(3):125-127.
Walls GL, Judd H. 1933. The intra-ocular colour-filters of vertebrates. The British
journal of ophthalmology 17(11):641.

Whitmore AV, Bowmaker JK. 1989. Seasonal variation in cone sensitivity and short-
wave absorbing visual pigments in the rudd Scardinius erythrophthalmus.
Journal of Comparative Physiology A: Sensory Neural and Behavioral
Physiology 166(1):103-115.
Widder EA, Latz MI, Herring PJ, Case JF. 1984. Far red bioluminescence from two
deep-sea fishes. Science 225(4661):512-514.
Widder EA. 2002. Bioluminescence and the pelagic visual environment. Marine and
Freshwater Behaviour and Physiology 35(1-2):1-26.
Widder EA. 2010. Bioluminescence in the ocean: origins of biological, chemical, and
ecological diversity. Science 328(5979):704-708.
Yokoyama R, Knox BE, Yokoyama S. 1995. Rhodopsin from the fish, Astyanax: role of
tyrosine 261 in the red shift. Investigative Ophthalmology and Visual Science
36(5):939-945.
212
Yokoyama S, Tada T, Zhang H, Britt L. 2008. Elucidation of phenotypic adaptations:

Molecular analyses of dim-light vision proteins in vertebrates. Proceedings of
the National Academy of Sciences 105(36):13480-13485.
Biology 93(1):147-156.
Zoology 607:1-69.
Zeil J. 1983. Sexual dimorphism in the visual system of flies: the compound eyes and
neural superposition in bibionidae (Diptera). Journal of comparative physiology
A: Sensory Neural and Behavioral Physiology 150(3):379-393.
Zhang H, Futami K, Horie N, Okamura A, Utoh T, Mikawa N, Yamada Y, Tanaka S,

Okamoto N. 2000. Molecular cloning of fresh water and deep-sea rod opsin
genes from Japanese eel Anguilla japonica and expressional analyses during
sexual maturation. FEBS Letters 469(1):39-43.
213
Chapter 6 Ganglion cells and spatial resolving power
CHAPTER 6
Retinal ganglion cell distribution and spatial resolving power

in lanternfishes (Myctophidae)
Fanny de Busserolles1, N. Justin Marshall2, Shaun P. Collin1.
Australia, Crawley, WA, 6012, Australia.
2. Sensory Neurobiology Group, Queensland Brain Institute, University of Queensland, St Lucia, QLD,
4072, Australia.
215
6.1. Abstract
Topographic analyses of ganglion cell density is very useful in providing
information about the visual ecology of a species by identifying areas of acute vision
within the visual field (i.e. areas of high cell density). In this study, we investigate the
neural cell distribution in the ganglion cell layer of a range of lanternfish species
belonging to ten genera. Analyses were performed on wholemounted retinas using
stereology. Topographic maps were constructed of the distribution of all neurons and
both the ganglion and amacrine cell populations in five different species from Nissl-
stained retinas using cytological criteria. Amacrine cell distribution was also examined
immunohistochemically in two of the five species using anti-parvalbumin. The
distribution of both the total neuron and amacrine cell populations were aligned in all of
the species examined, showing a general increase in cell density toward the retinal
periphery. However, when the ganglion cell population was topographically isolated
from the amacrine cell population, which comprised up to 80% of the total neurons
within the ganglion cell layer, a different distribution was revealed. Topographic maps
of the true ganglion cell distribution in 18 species of lanternfishes, revealed well-
defined specialisations in different regions of the retina. Different species possessed
distinct areas of high ganglion cell density, with respect to both peak density and the
shape of the specialised acute zone (i.e streak-like elongated areae, areae temporales and
large areae centrales). Spatial resolving power was calculated to be relatively low,
varying from 1.6 to 4.4 cycles per degree, indicating that myctophids may constitute
one of the less visually acute groups of deep-sea teleosts. The diversity in retinal
specialisations and spatial resolving power within the family is discussed in terms of
possible ecological functions and evolutionary history.
6.2. Introduction
Retinal topographic analyses of the distribution of neurons within the ganglion
cell layer are very useful in providing information on how animals perceive and process
visual information. Using the wholemount technique (Stone, 1981; Coimbra et al.,
2006; Ullmann et al., 2011), which allows for the examination of the entire retina, areas
of high cell density or retinal specialisation can be identified. These areas of high cell
density provide a greater acuity in a specific part of the visual field and usually closely
reflect each species’ habitat and behavioural ecology in both terrestrial and aquatic
vertebrates (Hughes, 1977; Collin and Pettigrew, 1988b; Collin and Pettigrew, 1988c;
216
Collin and Pettigrew, 1989; Bozzano and Collin, 2000; Collin and Shand, 2003). The
spatial distribution of ganglion cells can either form specialisations such as an “area” or
a “streak”. An area is a region of acute vision formed by a concentric increase of cell
density in a specific part of the retina usually found in species living in a three
dimensional, “enclosed” environment (Hughes, 1977; Collin and Pettigrew, 1988b) and
exhibiting predatory behaviour (Fritsches et al., 2003). Conversely, streaks, which are
formed by an elongated increase in cell density across the retinal meridian, allow an
animal to scan a broad horizon (panoramic field) with enhanced acuity without using
distinctive eye movements, thereby enhancing the detection of moving objects across
the panoramic field (Hughes, 1977; Collin and Pettigrew, 1988c). This type of
specialisation is usually found in species living in an open environment, where they
perceive their surrounding with an uninterrupted view of the horizon, defined in the
ocean by the air-water or sand-water interfaces (Collin and Shand, 2003).
The deep-sea, and more particularly the mesopelagic zone, is a relatively simple
visual environment, reliant on essentially two different light sources: the residual
downwelling sunlight and bioluminescence. Downwelling sunlight between 200 and a
1000 m, diminishes exponentially with depth creating a semi-extended visual scene,
where shadows (prey, predator or conspecific) can be detected against the lighter
background when viewed from below. Bioluminescent emissions, on the other hand,
which are produced by the animal themselves, occur at all depths and are thought to
play important roles in the deep-sea, subserving behavioural interactions with prey,
predators and congeners (Herring, 2002; Haddock et al., 2010).
Although the deep-sea environment is not as diverse as the well-lit shallow

environment, deep-sea species also possess a wide range of retinal specialisations to
allow acute vision in specific parts of the visual field (Collin and Partridge, 1996; Collin
et al., 1997; Wagner et al., 1998). Areae have been found in several tubular and non-
tubular eyed deep-sea species of teleosts, providing acute vision in different parts of the
visual field depending on the location of the specialisation (Collin and Partridge, 1996;
Wagner et al., 1998). A fovea, a pit-like invagination of the retina associated with an
area has also been observed in several species (Collin and Partridge, 1996; Wagner et
al., 1998; Collin et al., 2000), providing high acuity and possibly image magnification,
accurate fixation and depth perception (Walls, 1942). Finally, relatively unspecialised
retinas with a nearly uniform ganglion cell distribution, probably mediating high
217
sensitivity rather than acuity, have also been reported in a single group of deep-sea
fishes, the Myctophidae (Wagner et al., 1998).
Myctophids or lanternfishes are one of the most abundant groups of mesopelagic

fishes occupying all of the world’s oceans. Like most mesopelagic organisms, they are
bioluminescent and possess two kinds of bioluminescent structures thought to play
several different roles (camouflage, distraction, illumination, and intra- and interspecific
communication, Case et al., 1977; Edwards and Herring, 1977). Myctophids also show
marked intra- and interspecific variation in the depths they occupy (Karnella, 1987),
their migration patterns (Watanabe et al., 1999) and the location of their luminous
organs (Nafpaktitis and Nafpaktitis, 1969; Paxton, 1972; Nafpaktitis et al., 1977;
Nafpaktitis, 1978; Zahuranec, 2000; Herring, 2007; de Busserolles et al., 2013).
Recently, a broad diversity in the visual system of this large group of fishes has been
revealed (de Busserolles et al., 2013; Chapter 3-5), including marked differences in
photoreceptor distribution across the retina (Chapter 4).
The first study examining the distribution of neurons within the ganglion cell
layer in lanternfishes revealed a concentric increase of cells toward the periphery,
potentially enhancing acute vision throughout the peripheral field (Collin and Partridge,
1996). Although this type of cell arrangement might appear as a good strategy in an
environment where small bioluminescent flashes can appear from any direction and at
any time, it was later revealed that this specialisation did not represent the true
distribution of ganglion cells due to the inclusion of an unusually large population of
displaced amacrine cells. Once the amacrine cell population was removed, analysis of
the distribution of true ganglion cells revealed a rather unspecialised retina (Collin and
Hoskins, 1997; Wagner et al., 1998). However, these analyses were only realised for
three different species of lanternfishes all belonging to the same genus, Lampanyctus,
which might only represent one type of retinal specialisation within this large family. In
this study, we investigate the diversity of the visual system of lanternfishes by
examining the distribution and density of the true ganglion cell population in a large
range of lanternfish species from different genera. We also examine the contribution of
the “displaced” amacrine cell population to the total number of neurons present in the
ganglion cell layer. The diversity in retinal specialisations and spatial resolving power
within the family is discussed in terms of possible ecological functions and evolutionary
history.
218

6.3.1. Collection of species and preservation of ocular tissue
Australian waters, the Tasman Sea and the Bay of Biscay were acquired through
collaborators.
For each individual, the standard length, rostro-caudal eye diameter (measured
in situ) and lens diameter, were measured with digital calipers to 0.1 mm. For most of
the lanternfishes, measurements were performed on fresh specimens on board ship prior
to dissection and fixation. However, when samples were acquired from collaborators,
the measurements and dissection were made after fixation. Depending on the source of
the samples, tissue was fixed in 4% paraformaldehyde (PFA) in 0.1 M phosphate buffer
(PB, pH 7.4), 5% buffered formalin or Karnovsky’s fixative (2.5% paraformaldehyde
and 2% glutaraldehyde in 0.1 M cacodylate buffer, pH 7.4). Since formaldehyde has
previously been found to affect body length by only 0.8% (Kristoffersen and Salvanes,
1998), no measurement corrections were used between fresh and fixed specimens.
6.3.2. Preparation of retinal wholemounts and Nissl staining

Retinal wholemounts were prepared according to standard protocols (Stone,
1981; Coimbra et al., 2006; Ullmann et al., 2011). Radial cuts were performed in order
to flatten the eye and subsequently the retina in toto onto a glass slide, where the
orientation was confirmed by making a small additional cut in the nasal or dorso-nasal
part of the eye. The sclera, retinal pigment epithelium and tapetum (when present), were
gently removed with the help of No. 3 watchmaker’s forceps and a kolinsky hair paint
brush. The retina was wholemounted on a gelatinised slide, ganglion cell layer facing
up. In order to improve cell differentiation, wholemounts were left to dry at room
temperature overnight in formalin vapours and incubated for an additional hour in
formalin vapour at 60°C prior to staining (Stone, 1981; Coimbra et al., 2006; Coimbra
219
et al., 2012). Following the staining protocol of Coimbra et al. (2006), wholemounts
were rehydrated, stained in 0.1% Cresyl violet for 5 mins, dehydrated in an ethanol
series, cleared in xylene and mounted with Entellan New (Merck). Shrinkage associated
with the staining steps was negligible and (where present) confined to the retinal
margins, since retinal wholemounts were attached to the slide during all steps (Coimbra
et al., 2006).
6.3.3. Amacrine cell labelling using immunohistochemistry

Parvalbumin antibody was used to selectively labelled amacrine cells present in
the ganglion cell layer of two species of lanternfishes, Ceratoscopelus warmingii and
Myctophum asperum. Retinal wholemounts, fixed in 4% PFA, were dissected as
described above and rinsed three times in 0.1 M PB (5 mins each). In order to optimise
the immunolabeling, two pretreatments were performed; antigen retrieval and
endogenous peroxidise inactivation. Briefly, each wholemount was first incubated in
boric buffer at 60°C for 30 mins, cooled to room temperature for 5 mins and incubated
15 mins in a light tight vial containing a solution of 10% methanol, 3% H2O2 in 0.1 M
PB (Coimbra et al., 2012). Retinas were then permeabilised in 5% Triton solution (2 x 5
mins followed by 1 x 20 mins, Coimbra et al., 2012) before being rinsed three times (5
mins each) in 0.1M PB. Retinal wholemounts were incubated under gentle rocking at
room temperature for 24 h in a solution containing anti-parvalbumin (1:1000, product
no. P3088, mouse monoclonal, Sigma-Aldrich, USA), 0.3% Triton and 5% rabbit serum
(Vector Laboratories, USA), in 0.1M PB. Thereafter, the retinas were washed three
times in 0.1M PB (5 mins each) and further incubated for 1 h in a mixture of
biotinylated rabbit anti-mouse secondary antibody (1:200, Jackson ImmunoResearch,
USA) with 0.3% Triton in 0.1M PB. Finally, retinal wholemounts were incubated in an
avidin/biotin complex (Vectastain ABC kit, Vector Laboratories, USA) for 1 h and
rinsed three times in 0.1M PB (5 mins each) before being reacted with nickel enhanced
3,3 diaminobenzidine solution (DAB kit, Vector Laboratories, USA). Each retina was
then wholemounted on a gelatinised slide, dried overnight, dehydrated in an ethanol
series, cleared in xylene and mounted with Entellan New (Merck).
Anti-parvalbumin immunolabeling was also performed on retinal sections of

C. warminigii in order to verify the specificity of the labelling. Briefly, a piece of retina
was cryoprotected by sequential incubations (5 mins each) in a graded series of sucrose
solutions (10%, 20% and 30% in 0.1 M PB). After a 24 h incubation at 4°C in 30%
220
sucrose, the sample was embedded in OCT Tissue-Tek medium (Sakura Finetek, CA,
USA) and 20 μm sections were cut using a Leica CM1900 cryostat-microtome and
mounted on a microscope slide. Immunolabeling was performed directly on the slides as
described above for the wholemount with the exception of the pretreatments, which are
unnecessary on sections. Following the reaction, the sections were mounted in 40%
glycerol in 0.1 M PB and observed using light microscopy.
Photographs of the Nissl-stained and immunolabeled retinal wholemounts and

sections of Ceratoscopelus warmingii were taken using a digital video camera
(MicroFIRE, OPTRONICS) mounted on a compound microscope (Olympus BX50).
For display purposes only, the contrast and brightness of the entire photographs were
adjusted using Adobe Photoshop CS4 (Adobe System Inc., USA).
6.3.4. Stereological analyses and the construction of topographic maps

The total number and topographic distribution of the total neuron, amacrine cell
and ganglion cell populations present within the ganglion cell layer of several species of
lanternfishes, were assessed using the optical fractionator technique (West et al., 1991)
modified by Coimbra et al. (2009, 2012) for its use in wholemounted retinas. Briefly,
the retinal wholemount was considered as a single section and consequently, the
thickness sampling factor (tsf) was fixed at 1. The outline of the retinal wholemount
was then digitised using a x4 objective (numerical aperture 0.13) mounted on a
compound microscope (Olympus BX50) equipped with a motorized stage (MAC5000,
Ludl Electronics Products, USA), a digital video camera (MicroFIRE, OPTRONICS),
and a computer running Stereo Investigator software (Microbrightfield, USA). Using a
x60 oil immersion objective (numerical aperture 1.35), cells were randomly and
systematically counted using the parameters listed in Tables 6.1 and 6.2. The total
number of cells was estimated by multiplying the sum of total neurons counted by the
area of sampling fraction (i.e. ratio between the counting frame and sampling grid, West
et al., 1991; Coimbra et al. 2009).
221
Table 6.1. Summary of the parameters used for the analysis of the neural cell distribution (total
neural cells (TC), amacrine cells (AC), ganglion cells (GC)) along with the coefficient of error
(Schaeffer CE). * indicates a retina analysed with parvalbumin immunohistochemistry to label
amacrine cells. SL = standard length, ø = diameter, # = number.
Species Indiv SL Eye ø Counting Grid Site CE CE CE

(mm) (mm) frame (μm) (μm) # GC AC TC
Ceratoscopelus warmingii 1 44.1 4.1 50 x 50 320 x 320 190 0.040 0.030 0.031
2 38.6 3.5 50 x 50 280 x 280 185 0.029 0.026 0.025
3 36.3 3.5 50 x 50 270 x 270 186 0.032 0.031 0.029
3* 202 0.036
Myctophum asperum 1 78.4 8.5 50 x 50 750 x 750 165 0.037 0.033 0.030
1* 183 0.041
Bolinichthys nikolayi 1 37.7 4.1 50 x 50 340 x 340 201 0.033 0.031 0.029
Diogenichthys laternatus 1 19.3 1.9 50 x 50 140 x 140 180 0.034 0.033 0.032
Notoscopelus kroeyerii 1 95.2 6.1 75 x 75 550 x 550 202 0.024 0.026 0.023
Table 6.2. Summary of the parameters used for the analysis of the ganglion cell distribution
along with the coefficient of error (Schaeffer CE). SL = standard length, ø = diameter, n.a. = not
available.
Species Indiv SL Eye ø Counting Grid Site Schaeffer

(mm) (mm) frame (μm) (μm) numbers CE
Ceratoscopelus warmingii 1 44.1 4.1 100 x 100 320 x 320 198 0.033
2 38.6 3.5 100 x 100 280 x 280 199 0.030
3 36.3 3.5 100 x 100 270 x 270 196 0.029
Electrona risso 1 n.a. n.a. 100 x 100 680 x 680 205 0.039
Bolinichthys longipes 1 42.1 4.0 100 x 100 350 x 350 207 0.026
B. nikolayi 1 37.7 4.1 50 x 50 340 x 340 201 0.033
Diaphus mollis 1 66.2 6.5 100 x 100 580 x 580 200 0.029
D. parri 1 35.2 4.1 100 x 100 330 x 330 204 0.033
Diogenichthys laternatus 1 19.3 1.9 50 x 50 140 x 140 180 0.034
Lampadena luminosa 1 104.3 8.4 150 x 150 800 x 800 181 0.023
Lampanyctus alatus 1 49.8 3.3 75 x 75 295 x 295 197 0.045
L. parvicauda 1 54.2 3.3 75 x 75 260 x 260 211 0.040
Myctophum asperum 1 78.4 8.5 100 x 100 750 x 750 172 0.034
M. brachygnathum 1 67.7 7.7 100 x 100 600 x 600 206 0.045
M. lychnobium 1 51 5.0 100 x 100 400 x 400 165 0.043
M. nitidulum 1 77.9 6.9 100 x 100 650 x 650 168 0.042
M. obtusirostre 1 92.3 10.6 100 x 100 850 x 850 197 0.046
M. spinosum 1 60.7 5.1 100 x 100 500 x 500 159 0.041
Notoscopelus kroeyerii 1 95.2 6.1 100 x 100 550 x 550 206 0.024
Symbolophorus rufinus 1 73.7 6.3 100 x 100 500 x 500 198 0.035
Three types of cells can be identified and distinguished within the ganglion cell
layer by cytological criteria from Nissl stained retinas; ganglion cells, amacrine cells
and glial cells (Hughes, 1975; Collin and Pettigrew, 1988a; Collin, 1988). Ganglion
cells are characterised by their larger size, irregular shape and large nucleus with a
granular aspect. Amacrine cells are characterised by their small size, round or tear drop
shape and darkly-stained appearance, where glial cells are differentiated by their
distinctive elongated shape. While glial cells are easily discriminated from the other
cells, amacrine cells can sometimes be hard to distinguish confidently from ganglion
222
cells on cytological criteria alone, and as a result are often included in topographic
analyses. Fortunately, several studies have shown that the distribution of amacrine cells
in the ganglion cell layer matches the distribution of the true ganglion cells and that
consequently their inclusion in the neural cell counts does not change the overall
topography (Wong and Hughes, 1987; Collin and Pettigrew, 1988a; Bailes et al., 2006b;
Coimbra et al., 2006; Collin, 2008). However, lanternfishes seem to be an exception to
this rule. In fact, lanternfishes seem to possess a very large proportion of amacrine cells
in the ganglion cell layer (up to 80%, Collin and Hoskins, 1997; Wagner et al., 1998;
Bozzano et al., 2007), which masks the distribution of the true ganglion cell population
(Collin and Hoskins, 1997; Wagner et al., 1998).
In this study, we confirm the distribution of the amacrine cell population in

another six species of lanternfishes and estimate their proportion. To do so, we use the
StereoInvestigator System to simultaneously count amacrine cells and ganglion cells at
each sampling site, differentiated using the cytological criteria described previously
(Figure 6.1) and using two different markers. Using this approach, we were able to
generate data for amacrine cells alone, ganglion cells alone and both amacrine and
ganglion cells together (total neural cells). For two of the six species, Ceratoscopelus
warmingii and Myctophum asperum, immunolabeled amacrine cells present in the
ganglion cell layer were counted from the complementary retina (right eye), using the
same stereological criteria (Table 6.1), to validate the distribution obtained using Nissl-
staining (left eye). Due to the consistent discrepancy found between amacrine and
ganglion cell distributions, we chose to only count ganglion cells in the remaining 12
species analysed. Subsampling was carried out in the area of highest cell density to
estimate the peak density of ganglion cells. Intraspecific variability in neural cell
distribution was investigated for one species, Ceratoscopelus warmingii for which three
different individuals were analysed.
The counting frame and grid size were chosen carefully in order to maintain the
highest level of sampling and achieve an acceptable Schaeffer coefficient of error (CE).
The CE is a measure of the accuracy of the density estimates and is considered
acceptable below 0.1 (Glaser and Wilson, 1998; Slomianka and West, 2005). As a
result, grid size was modified between each specimen to allow the sampling of around
200 sites per retina and to enable us to compare different-sized individuals in the case of
Ceratoscopelus warmingii. The counting frame size was also adjusted between species
223
to allow a minimum average count of 40 cells per sampling site. However, due to the
large proportion of amacrine cells, the counting frame used to count amacrine cells and
ganglion cells simultaneously did not always allow this. To avoid this problem and
assess ganglion cell distribution confidently in these cases, ganglion cells were counted
a second time, separately, using a larger counting frame.
Figure 6.1. Light micrograph of the Nissl stained ganglion cell layer of Ceratoscopelus
warmingii (A) in a low density area (nasal part) and (B) in the peak density area (temporal).
gc = ganglion cell, ac = amacrine cell, g = glial cell. Scale bars = 50 μm.
Topographic maps were constructed using the statistical program R v.2.15.0 (R

Foundation for Statistical Computing 2012) with the results exported from the Stereo
Investigator Software according to Garza Gisholt et al. (in press). Garza Gisholt et al.
(in press) proposed several smoothing models to construct the iso-density maps and, for
this study, we chose to use the Gaussian Kernel Smoother from the Spatstat package
(Baddeley and Turner, 2005). For each map, the sigma value was adjusted to the
distance between points (i.e. grid size).
6.3.5. Phylogenetic comparison of visual characteristics

In order to assess the influence of phylogeny on the interspecific variability in
retinal specialisations, within the Myctophidae, we mapped the type of specialisation (or
area of high ganglion cell density) onto an existing phylogeny (Paxton et al., 1984)
including all of the species examined here. Paxton et al. (1984) have published a
phylogeny classifying genera using derived character states of adult osteology and
photophore patterns (Paxton et al., 1984), and of larvae (as described by Moser and
224
Ahlstrom, 1970, 1972, 1974) and is currently the most up to date morphological
phylogeny.
6.3.6. Calculation of spatial resolving power

The upper limit of spatial resolving power (SPR) was estimated for each species
using the peak density of ganglion cells, as described by Pettigrew and Collin (1989).
Briefly, Matthiessen’s ratio states that the focal length f, in teleosts fishes, is about 2.55
the radius of the lens (Matthiessen, 1882; Matthiessen, 1886). Using f, angle α, which is
the angle subtending 1 mm on the retina, can be calculated:
α = arc tan (1/f)
Knowing the peak density of ganglion cells (PGC in cells/mm2), angle α and the fact
that at least two ganglion cells are needed to distinguish the boundary of a black and
white cycle of a grating, we can calculate SPR in cycles per degree:
SPR = (PGC/α) /2
6.4. Results
6.4.1. Immunolabeling of amacrine cells in the lanternfish retina
Immunolabelling using anti-parvalbumin revealed the presence of two
populations of amacrine cells, a low density population in the inner nuclear layer and a
high density population in the ganglion cell layer (Figure 6.2). The morphological
aspect of the cells labelled with parvalbumin antibody was very similar to the amacrine
cells identified in the ganglion cell layer of the Nissl-stained retinas using cytological
criteria (small round or tear drop-shaped cells, Figures 6.1-2). Within the ganglion cell
layer, the staining intensity of the immunolabeled amacrine cells varied from lightly-
staining to darkly-staining possibly indicating the presence of different types of
amacrine cells.
Analyses of the amacrine cell distribution within the ganglion cell layer using
both immunolabelling and Nissl staining, revealed very similar results in terms of
topography with both species showing a concentric increase in amacrine cells toward
the periphery (Figure 6.3). However, small discrepancies in the total number of
amacrine cells were observed between both methods (~ 7% difference, Table 6.3).
225
Figure 6.2. Light micrographs from (A) a transverse section and (B-C) a wholemount of the
amacrine cells, immunolabeled with anti-parvalbumin, in the retina of Ceratoscopelus
warmingii. The black arrows in (A) indicate the location of the amacrine cells in the inner
nuclear layer (INL) and the ganglion cell layer (GCL). The panels B and C show the amacrine
cells present in the two levels within the retina, (B) at the level of the inner nuclear layer and (C)
at the level of the ganglion cell layer. PRL = photoreceptor layer, ON = outer nuclear layer.
Scale bars = 20 μm (A), 50 μm (B and C).
6.4.2. Neural cell distribution in the ganglion cell layer of lanternfishes

Topographic maps of total neural cells, amacrine cells and ganglion cells,
present in the ganglion cell layer of five different species of lanternfishes, were revealed
from Nissl stained retinal wholemounts (Figures 6.4 to 6.8). Results were similar for the
five species in terms of total neural cell and amacrine cell distributions. Both
topographies were aligned where the concentric increase of cells toward the periphery
were in register (Figures 6.4-8 B-C). However, when only the ganglion cell distribution
was assessed, well-defined specialisations were revealed in the temporal and ventro-
temporal parts of the retina in each species. Hence, Myctophum asperum (Figure 6.5C)
and Diogenichthys laternatus (Figure 6.7C) possess a streak-like elongated area ventro-
temporalis, and Ceratoscopelus warmingii (Figure 6.4C), Bolinichthys nikolayi (Figure
6.6C) and Notoscopelus kroeyerii (Figure 6.8C) possess either an area temporalis (B.
nikolayi and N. kroeyerii) or an area ventro-temporalis (C. warmingii).
226
Figure 6.3. Topographic map showing the amacrine cell distribution within the retina of two
lanternfish species (A-B) Ceratoscopelus warmingii, (C-D) Myctophum asperum. Two different
methods were used to identify the amacrine cells, (A-C) cytological criteria from Nissl stained
retinas, (B-D) immunolabeling using parvalbumin anti-body. The black lines represent iso-
density contours and values are expressed in densities x 103 cells/mm2. The black arrows
indicate the orientation of the retina. T = temporal, V = ventral. Scale bar = 1 mm.
Table 6.3. Summary of the neural cell (ganglion, amacrine and total neural cell populations)
quantitative data obtained using the optical fractionator method on the wholemount retina of
five species of lanternfishes. * indicate a retina analysed with parvalbumin
immunohistochemistry to label amacrine cells. GC = ganglion cells, AC = amacrine cells.
Species Indiv. Total GC Total AC Total cell % AC

number number number
Ceratoscopelus warmingii 1 83,927 296,755 380,682 78
2 90,755 279,762 370,518 76
3 79,198 221,266 300,464 74
3* 254,829
Myctophum asperum 1 248,625 809,325 1,057,950 77
1* 706,500
Bolinichthys nikolayi 1 123,738 276,515 400,253 69
Diogenichthys laternatus 1 28,937 58,925 87,863 67
Notoscopelus kroeyerii 1 152,352 544,446 696,799 78
227
Figure 6.4. Topographic distribution of the total neural cells (A), amacrine cells (B), and
ganglion cells (C) of Ceratoscopelus warmingii. The maps were produced using the same Nissl-
stained wholemount, counting frame and grid. The black lines represent iso-density contours
and values are expressed in densities x 103 cells/mm2. The black arrows indicate the orientation
of the retina. T = temporal, V = ventral. Scale bar = 1 mm.
228
ganglion cells (C) of Myctophum asperum. The maps were produced using the same Nissl-
229
ganglion cells (C) of Bolinichthys nikolayi. The maps were produced using the same Nissl-
230
ganglion cells (C) of Diogenichthys laternatus. The maps were produced using the same Nissl-
231
ganglion cells (C) of Notoscopelus kroeyerii. The maps were produced using the same Nissl-
232
From the topographic map, it appears that M. asperum possesses a second specialisation
in the dorsal part of the retina (Figure 6.5C). However, the ganglion cell density
gradient is shallow for this species (1,200 cells/mm2 to 4,500 cells/mm2) and cells are
usually not regularly distributed within the retina. As mentioned in the Material and
methods section, the counting frame chosen to simultaneously count amacrine cells and
ganglion cells together only allowed us to count a very small proportion of ganglion
cells per sampling site (less than 20 cells on average per site). As a result, even though
the Schaeffer coefficient is acceptable (< 0.05), cell counts in the regions of low cell
density revealed a specialisation in the dorsal part of the retina (probably caused by the
irregular cell arrangement) that does not exist if a larger counting frame is used,
allowing a larger number of ganglion cells to be counted in each sampling site (an
average of at least 40 cells per site, Figure 6.9).
Figure 6.9. Topographic distribution of Nissl stained retinal ganglion cells in Myctophum
asperum, using two different counting frame sizes, (A) 50 x50 μm and (B) 100 x 100 μm. The
black lines represent iso-density contours and values are expressed in densities x 103 cells/mm2.
The black arrows indicate the orientation of the retina. T = temporal, V = ventral. Scale bar = 1
mm.
In terms of cell number, the total number of neural cells (ganglion cells + amacrine
cells) varies greatly between species and ranges from 87,863 cells (Diogenichthys
laternatus) to 1,057,950 cells (Myctophum asperum) and is positively correlated with
eye size (Table 6.3). Similar levels of variation and a correlation with eye size is
observed for the total number of ganglion cells and amacrine cells when assessed
separately. However, the amacrine cell population is very large compared to the
ganglion cell population, representing between 67 and 78% of the total neural cells
present within the ganglion cell layer (Table 6.3).
233
6.4.3. Interspecific differences in ganglion cell distribution

Topographic analysis of the ganglion cell distribution of 18 different species of
lanternfishes reveals three main specialisations; elongated areae (streak-like), areae
temporales and large areae centrales (Figure 6.10). A streak-like elongated area was
found in the ventro-temporal part of the retina in seven species (Electrona risso,
Diogenichthys laternatus, Myctophum asperum, M. nitidulum, M. spinosum,
M. lychnobium, M. obtusirostre) and in the ventral part of the retina in a further two
species (M. brachygnathum and Symbolophorus rufinus, Figure 6.10A-I). An area
temporalis was observed in seven species, i.e. Bolinichthys longipes, B. nikolayi,
Diaphus mollis, D. parri, Lampadena luminosa, Notoscopelus kroeyerii and
Ceratoscopelus warmingii (Figure 6.10J-P). Moreover, in B. longipes and C. warmingii,
the area temporalis could also be qualified as an area dorso-temporalis and an area
ventro-temporalis, respectively. Finally, a large area centralis with a gradual decrease of
cells toward the periphery was observed in the remaining two species, Lampanyctus
alatus and L. parvicauda, with a peak density located centrally and extending towards
the dorso-temporal part of the retina (Figure 6.10Q-R).
We consider it important to assess the influence of phylogeny on the type of

retinal specialisations, especially given the paucity of information regarding the
behaviour and life history traits of each species. Therefore, we have mapped the three
main specialisation types and location onto an existing phylogenetic tree (according to
Paxton et al. 1984), that includes all of the species examined (Figure 6.11). The type of
retinal specialisation appears to follow closely the phylogeny, with species from the
Myctophinae subfamily possessing only streak-like elongated areae, and representatives
of the subfamily Lampanyctinae having both areae centrales and areae temporales. Even
within the subfamily Lampanyctinae the type of specialisation is well partitioned along
the phylogeny with the presence of an intermediate species showing the transition from
an area centralis to an area temporalis, i.e. Lampadena luminosa which possesses a
temporal area and a decrease of cells toward the periphery. Slight differences in the
location of the streak-like elongated areae (ventral or ventro-temporal) and areae
temporales (temporal, ventro-temporal and dorso-temporal) are also observed between
species, independent of the phylogeny (Figure 6.11). Apart from the areae centrales,
most specialisations are located in the temporal part of the retina or close by, a region
that will sample information in the field of view situated directly in front of the fish, in
front and above (ventro-temporal specialisation), or in front and below (dorso-temporal
234
specialisation). However, these differences have to be considered cautiously since only

one individual per species was analysed.
Figure 6.10. Topographic distribution of retinal ganglion cells in 18 species of lanternfishes.

(A) Myctophum asperum, (B) M. nitidulum, (C) M. spinosum, (D) M. lychnobium, (E) M.
obtusirostre, (F) M. brachygnathum. The black lines represent iso-density contours and values
are expressed in densities x 103 cells/mm2. The black arrows indicate the orientation of the
retina. T = temporal, V = ventral. Scale bar = 1 mm.
235
Figure 6.10. (Continued). (G) Electrona risso, (H) Symbolophorus rufinus, (I) Diogenicghthys
laternatus, (J) Notoscopelus kroeyerii, (K) Bolinichthys longipes, (L) B. nikolayi.
236
Figure 6.10. (continued). (M) Diaphus mollis, (N) D. parri, (O) Lampadena luminosa, (P)
Ceratoscopelus warmingii, (Q) Lampanyctus alatus, (R) L. parvicauda.
237
Figure 6.11. Type of retinal specialisation in each of the 18 species of lanternfishes analysed
and plotted against the phylogeny of Paxton et al. (1984). EA = elongated area, AC = area
centralis, AT = area temporalis, VT = ventro-temporal, DT = dorso-temporal, V = ventral,
C = central, T = temporal.
The total number of ganglion cells varies greatly between species and ranges
from 28,937 cells (Diogenichthys laternatus) to 284,448 cells (Myctophum obtusirostre,
Table 6.4) and seems to vary directly with eye size. Mean and peak cell density also
varies greatly between species (from 1,958 cells/mm2 to 8,918 cells/mm2 and from
5,200 cells/mm2 to 23,600 cells/mm2, respectively). However, contrary to the total cell
number, they do not appear to vary accordingly to eye size.
6.4.4. Spatial resolving power

Spatial resolving power (SPR) was estimated using the peak ganglion cell
density for each species (Table 6.4). SPR was relatively low in myctophids, varying
from 1.63 cycles per degree (Diogenichthys laternatus) to 4.38 cycles per degree
238
(Myctophum obtusirostre). Interestingly, the species possessing the highest ganglion

cell density (Lampanyctus parvicauda) did not have the highest SPR. SPR was
positively correlated with eye size and lens diameter. Intraspecific variability in the
estimated SPR was very low, with a standard deviation of 0.04 (Ceratoscopelus
warmingii).
Table 6.4. Summary of the ganglion cell quantitative data obtained using the optical
fractionator method on the wholemounted retinas of 18 species of lanternfishes. SPR = spatial
resolving power.
Species Indiv Peak cell density Mean cell density Total cell lens ø SRP
(cells/mm2) (cells/mm2) number (mm)
Ceratoscopelus warmingii 1 11,000 3,662 74,255 1.6 2.01
2 14,300 5,215 81,355 1.4 2.04
3 13,200 5,247 74,970 1.4 1.96
Electrona risso 1 8,000 2,402 227,685 3.4 3.40
Bolinichthys longipes 1 14,900 5,244 132,973 1.8 2.62
B. nikolayi 1 20,100 5,156 123,738 1.8 2.95
Diaphus mollis 1 7,200 1,945 130,859 3.2 3.08
D. parri 1 17,400 4,697 104,347 1.8 2.75
Diogenichthys laternatus 1 23,200 8,204 28,937 0.73 1.63
Lampadena luminosa 1 4,577 1,990 230,599 3.7 2.83
Lampanyctus alatus 1 19,600 8,227 141,065 1.3 2.25
L. parvicauda 1 23,600 8,918 127,051 1.4 2.63
Myctophum asperum 1 5,200 1,958 189,394 3.7 2.98
M. brachygnathum 1 10,600 3,256 241,452 3.3 3.85
M. lychnobium 1 8,200 3,916 103,392 2.3 2.39
M. nitidulum 1 8,500 2,023 143,608 3.1 2.85
M. obtusirostre 1 6,900 1,998 284,448 4.7 4.38
M. spinosum 1 9,100 3,969 157,750 2.5 2.74
Notoscopelus kroeyerii 1 6,000 2,304 141,721 2.7 2.39
Symbolophorus rufinus 1 11,200 4,269 211,325 2.8 3.38
6.5. Discussion
6.5.1. Ecological significance of the retinal specialisations in lanternfishes
Analysis of retinal ganglion cell distribution, including the determination of
regions of high cell densities (i.e. specialisations), is a powerful means of investigating
how a species visually samples its environment. (Hughes 1977; Collin and Pettigrew,
1988b; Collin and Pettigrew, 1988c; Collin, 2008). The distribution of the true ganglion
cell population (not including amacrine cells) has previously been examined in three
species of lanternfishes from the genus Lampanyctus (Collin and Hoskins, 1997;
Wagner et al., 1998). The results revealed a poorly specialised retina, showing a nearly
uniform distribution of cells within the ganglion cell layer, most likely mediating
increased sensitivity. The results from this study, however, show a range of specialised
retinas with different species having distinct regions of high ganglion cell densities,
239
with respect to both peak density and the shape of the specialised acute zone (i.e. streak-
like elongated area, area temporalis and area centralis). As previously found in other
studies (Fritsches et al., 2003; Litherland and Collin, 2008), ganglion cell distribution
was found to be aligned with the photoreceptor cell distribution, where both cell
populations were investigated (i.e. Myctophum brachygnathum and Lampanyctus
parvicauda, Chapter 4 and this study) most probably to optimise spatial resolving power
within a specific retinal region.
The retinal morphology of teleost fishes is highly diverse and usually correlated
with habitat complexity and behavioural ecology (Collin and Pettigrew, 1988b; Collin
and Pettigrew, 1988c, Collin, 1997). In the mesopelagic zone, habitat complexity is
relatively low and mainly influenced by the intensity gradient of the residual
downwelling sunlight along the vertical axis. Differences in retinal specialisations are
then more likely due to interspecific differences in behaviour. Unfortunately, very
limited information is available on the behavioural ecology of this family. However,
taking into consideration the retinal specialisations identified in this study and other
visual adaptations enhancing photon capture identified in other studies (Chapters 2 - 4),
different ecological behaviours can be predicted.
While a visual streak will allow an animal to scan a broad horizon with
enhanced acuity (panoramic field) without using distinctive eye movements, and may
enhance the detection of moving objects across the panoramic field (Hughes, 1977;
Collin and Pettigrew, 1988b; Collin and Pettigrew, 1988c), areae provide acute vision in
a specific part of the visual field. Several lanternfish species possess a streak-like
elongated area, an intermediate specialisation between a streak and an area. Assuming a
horizontal position of the fish in the water column, a ventro-temporal streak-like
elongated area will subtend the frontal and dorsal view of the visual field, allowing the
detection of silhouettes situated above them against the lighter background of the upper
mesopelagic zone. Lanternfishes possessing streak-like elongated areae also have
several other specialisations. For example, in addition to possessing high cell densities
in the temporo-ventral part of the retina providing higher sampling of light signals
emanating from above, Myctophum brachygnathum also has a ventral aphakic gap and a
tapetum lucidum covering the entire retina (Chapter 3 and 4). While the tapetum
lucidum will enhance the chance of photon capture from light signals coming from any
direction, the ventral aphakic gap in this species would facilitate the detection of
240
bioluminescent signals (prey, predator or mate) situated below the fish, within the
increased visual field produced by this ventral extension of the pupillary aperture. M.
brachygnathum also possesses a relatively large eye (de Busserolles et al. 2013) and one
of the highest densities of photoreceptors within the family (around 2,000,000 rod
photoreceptors per mm2, Chapter 4). The peak summation ratio of photoreceptors to
ganglion cells in the temporo-ventral part of the retina (peak cell density) is close to
200:1 and in the non-specialised region is close to 500:1, much higher than what has
previously been reported for other marine species (1.2:1 to 100:1, Collin et al., 1998;
Collin et al., 2000; Fritsches et al., 2003; Litherland and Collin, 2008) and an indication
of exceptionally high sensitivity. All these results indicate that M. brachygnathum
possesses a highly specialised retina optimised for the detection of both bioluminescent
emissions and silhouettes in different parts of the visual field, suggesting that this
species relies heavily on vision for survival.
Conversely to streak-like specialisations, areae temporales improve acuity in the

temporal part of the visual field and may facilitate binocular vision in species living in a
structurally complex habitat or exhibiting predatory behaviours (Hughes, 1977; Collin
and Pettigrew, 1988b; Collin and Pettigrew, 1988c; Fritsches et al., 2003). Myctophids
are predatory species, feeding on a wide range of organisms, mainly zooplankton but
also fishes (Podrazhanskaya, 1993). However, it is unknown how myctophids catch
their prey. It has been suggested that species possessing head luminous organs (i.e.
Diaphus sp.) may use their organs as head torches in order to search for prey (Herring,
1985). Diaphus mollis, for example, possesses large ventro-nasal (Vn) luminous organs
most likely emitting light directly in front of the fish (Figure 6.12). The area temporalis
found in D. mollis (this study) associated with the presence of a nasal aphakic gap
(Chapter 3), provide the potential for binocular vision (Figure 6.12) and may enhance
acuity directly in the illuminated region in front of the fish, a powerful predatory
strategy.
A large area centralis was found in two species of lanternfishes, Lampanyctus

alatus and L. parvicauda indicating higher acuity in the centro-lateral visual field
thereby allowing the fish to perceive signals in a large part of its monocular field of
view, suggesting that these species are visual generalists. In addition to the generalist
specialisation, L. parvicauda does not appear to possess a very sensitive visual system
due to its relatively small eye (de Busserolles et al. 2013, Chapter 1) and lower
241
summation ratios (around 20:1 in the peak area and 50:1 in the periphery, Chapter 4,
this study). Even though L. parvicauda possesses adaptations to enhance photon capture
over the whole retina (i.e. circumlental aphakic gap and tapetum lucidum, chapter 3),
these findings indicate that this species possesses a less specialised visual system than
the two other species mentioned above. Consequently, L. parvicauda might not rely as
much on vision and might instead depend more heavily on other sensory systems to
survive in the mesopelagic zone.
Figure 6.12. Frontal view of the head of Diaphus mollis showing the presence of large ventro-
nasal (Vn) luminous organs (the two large white tissue patches in front of the eyes). The
position of the eyes, slightly tilted forward, most likely allows binocular vision in the frontal
visual field.
6.5.2. Acuity in lanternfishes

Spatial resolving power (SPR) in lanternfishes is relatively low (1.3 to 4.4 cycles
per degree, Wagner et al. 1998, this study) compared to other deep-sea teleosts i.e.
tubular and non-tubular eyes, which range from 4.3 to 22.9 cycles per degree (Collin
and Partridge, 1996; Collin et al., 1997; Wagner et al., 1998). However, the estimation
of SPR for these other deep-sea teleosts did include the population of amacrine cells,
which could have resulted in an overestimation. Nevertheless, even if the peak total
neural cell population (ganglion cells + amacrine cells) is taken into consideration in
estimating SPR in lanternfishes (i.e. 20,000 cells/mm2 instead of 5,000 cells/mm2 in
Myctophum asperum), the myctophid SPR would still be low compared to other deep-
sea species. Moreover, the proportion of amacrine cells estimated in two tubular-eyed
242
species was reported to be relatively low (less than 20%, Locket, 1977; Collin et al.,
1998), reinforcing the fact that lanternfishes might constitute one of the less acute
groups of deep-sea fishes. Similar to shallow water and deep-sea tubular-eyed species
(Collin and Pettigrew, 1989; Collin et al., 1997), SPR in myctophids was also found to
vary in proportion to eye/lens size.
Collin and Pettigrew (1989) found a relationship between SPR and feeding
strategies in 12 different species of reef teleosts, with species not requiring high acuity
(i.e. grazers, slow moving organisms) having lower SPR values (3-8 cycles per degree,
Collin and Pettigrew, 1989). Very low SPR could also reflect a low reliance on vision in
prey/predator interactions and indicate instead the use of other sensory systems (Bailes
et al., 2006a). In a relatively dark environment, where interspecific interactions are
mostly achieved through bright, bioluminescent flashes, as found in the mesopelagic
zone, sensitivity became more important than acuity. As most lanternfishes possess very
sensitive eyes (Chapter 4), prey detections at longer distance are most likely achieved
by the detection of a light signal or through the use of other sensory systems, with
acuity playing only a role at close range.
6.5.3. Putative function of the amacrine cell population

Amacrine cells are usually present in the inner nuclear layer. However, amacrine
cells have also been identified in the ganglion cell layer of several vertebrates (cat,
Wong and Hughes, 1987; rabbit, Vaney et al., 1981; newt, Ball and Dickson, 1983)
including teleosts (dragonet, van Haesendonck and Missotten, 1987; tench and rainbow
trout, Weruaga et al., 2000; hake Bozzano and Catalán, 2002; cichlid, Mack et al., 2004)
and for this reason have been referred to as displaced amacrine cells.
A large population of displaced amacrine cells was also identified in the

ganglion cell layer of myctophids. Contrary to what has usually been found in other
vertebrate retinas (Collin and Pettigrew, 1988a; Collin and Pettigrew, 1988a; Wong and
Hughes, 1987; Bozzano and Catalán, 2002; Bailes et al., 2006b), the inclusion of the
amacrine cell population in total neural counts of cells within the ganglion cell layer in
myctophids, which comprised up to 80% of the total neural cells in the ganglion cell
layer, masked the distribution of ganglion cells. As ganglion cells are the only neurons
possessing axons sending information to the visual centres of the brain, it is therefore
essential to exclude the amacrine cell population when assessing neuron distribution in
243
order to unravel the true distribution and peak density of the ganglion cells (at least in
myctophids).
Similar to the retina of the cat (Wong and Hughes, 1987), rabbit (Vaney et al.,
1981), and cichlid (Mack et al., 2004), the extremely large proportion of amacrine cells
in the ganglion cell layer of the lanternfish retina cannot be regarded as displaced and
must represent a true cell population within the ganglion cell layer to serve a particular
function. Mack et al. (2004) suggested that amacrine cells present in the ganglion cell
layer probably have a strong influence on the output activity of the ganglion cells due to
their proximity. Amacrine cells, in general, are in contact with every class of neuron
with the exception of the photoreceptors, and many different types of amacrine cells
exist (around 40 in teleosts, Wagner and Wagner, 1988). However the functional role of
many of these is still unknown.
Anti-parvalbumin, the antibody used in this study to label amacrine cells, is well
known for labelling a particular type of amacrine cell, i.e. AII amacrine cells, as
described in several vertebrate retinas (bat, Jeon et al., 2007; rabbit, Casini et al., 1995 ;
rat, Wässle et al., 1993; cat, Gabriel and Straznicky, 1992; and cichlid, Mack et al.,
2004). AII amacrine cells are usually involved in the rod signal pathway and may
improve the perception of small luminous objects against a dark background by
quickening the usually slow rod response (Nelson, 1982). Based on the labelling
intensity, several types of amacrine might occur in the lanternfish retina. Wassle et al.
(1993) also found two different populations of amacrine cells labelled with parvalbumin
in the rat retina; AII amacrine cells and widefield amacrine cells. In lanternfishes, AII
cells, if their presence is confirmed, might facilitate the detection of small
bioluminescent flashes against the mesopelagic background, especially in the peripheral
visual field, where they are most abundant.
6.5.4. Interspecific differences in retinal specialisations in lanternfishes: ecology or

phylogeny?
The study of retinal topography in a large number of vertebrates, and more
particularly mammals, has given rise to two main theories to explain the evolution of
retinal specialisations: the terrain theory by Hughes (1977) and the ancestry theory by
Stone (1983). While the terrain theory of Hughes (1977) suggests that retinal
topography results from an adaptation of each species to the symmetry and openness of
244
their perceived environment (habitat), Stone’s theory (Stone,1983) places more

emphasis on the influence of evolutionary history, where a part of the retinal
specialisation is inherited from a common ancestor.
The strong phylogenetic signal highlighted in this study, which reveals that a
similar type of retinal specialisation is found in closely related species, supports the
influence of evolutionary history on retinal specialisations. However, slight differences
in the location of the specialisations, independent of the phylogeny, may indicate the
influence of ecological and behavioural factors. Moreover, the presence of a streak-like
elongated area ventro-temporalis or ventralis, observed in several species, also indicates
a clear adaptation to the mesopelagic environment by allowing the detection of
silhouettes against the lighter background of the downwelling sunlit upper mesopelagic
zone. Furthermore, most species possessing a streak-like elongated area are also known
to be surface migrants and have been recorded in the first few meters of water at night
(i.e. Myctophum sp., Symbolophorus sp., Diogenichthys laternatus, de Busserolles et al.,
2013), where a weak visual horizon would be formed by the downwelling light emitted
by the moon and stars or by the air-water interface, explaining the need of a streak-like
elongated area in these species.
As recently demonstrated in a terrestrial vertebrate, the giraffe (Coimbra et al.,

2013) and an amphibious vertebrate, the penguin (Coimbra et al., 2012), our findings
support both theories and retinal specialisations in myctophids appear to be driven by
both ecology and phylogeny. However, more information on the ecology and behaviour
of each species will be necessary to better understand the evolution of the visual system
in lanternfishes and answer the following question. Has the visual system of each
lanternfish species adapted to a particular life style or did the each species develop
different behaviours based on an inherited retinal arrangement? If the type of retinal
specialisation is really inherited from a common ancestor, then some predictions could
be made on the possible visual capabilities of other myctophid species.
245
(Murdoch University), Dr. Brigitte Guillaumont (Ifremer) and Dr. Adrian Flynn
(University of Queensland), for providing additional samples. We gratefully
identification and, Adrian Flynn, Caroline Kerr and Alan Goldizen for their help during
field trips. We are indebted to Joao Paolo Coimbra for his assistance with retinal
wholemount preparation, immunohistochemistry and stereology analyses and to and to
Eduardo Garza-Gisholt for his help in using his R scripts for the creation of the
topographic maps. This work was funded by the Australian Research Council
(Discovery grant to SPC, the Deep-Australia Linkage Grant to NJM and SPC) and the
West Australian State Government (SPC). FdB was supported by a Scholarship for
International Research Fees (SIRF) and a University International Stipend (UIS) at the
University of Western Australia.
6.7. References
Bailes HJ, Robinson SR, Trezise AEO, Collin SP. 2006a. Morphology, characterization,
and distribution of retinal photoreceptors in the Australian lungfish
Neoceratodus forsteri (Krefft, 1870). The Journal of Comparative Neurology
494:381-397.
Bailes HJ, Trezise AEO, Collin SP. 2006b. The number, morphology, and distribution
of retinal ganglion cells and optic axons in the Australian lungfish Neoceratodus
forsteri (Krefft 1870). Visual Neuroscience 23(2):257-273.
Ball AK, Dickson DH. 1983. Displaced amacrine and ganglion cells in the newt retina.
Experimental Eye Research 36(2):199-213.
Bozzano A, Collin SP. 2000. Retinal ganglion cell topography in elasmobranchs. Brain,
Bozzano A, Catalán I. 2002. Ontogenetic changes in the retinal topography of the

European hake, Merluccius merluccius: implications for feeding and depth
distribution. Marine Biology 141(3):549-559.
181.
246
Casini G, Rickman DW, Brecha NC. 1995. AII amacrine cell population in the rabbit
retina: Identification by parvalbumin immunoreactivity. The Journal of
Coimbra JP, Trévia N, Marceliano MLV, Andrade-Da-Costa BLS, Picanço-Diniz CW,

Yamada ES. 2009. Number and distribution of neurons in the retinal ganglion
cell layer in relation to foraging behaviors of tyrant flycatchers. The Journal of
Coimbra JP, Nolan PM, Collin SP, Hart NS. 2012. Retinal ganglion cell topography and
spatial resolving power in penguins. Brain, Behavior and Evolution 80(4):254-
268.
Coimbra JP, Hart NS, Collin SP, Manger PR. 2013. Scene from above: Retinal ganglion
cell topography and spatial resolving power in the giraffe (giraffa
camelopardalis). The Journal of Comparative Neurology 521(9):2042-2057.
Collin SP. 1988. The retina of the shovel-nosed ray, Rhinobatos batillum
(Rhinobatidae): morphology and quantitative analysis of the ganglion, amacrine
and bipolar cell populations. Journal of Experimental Biology 47(4):195-207.
Collin SP, Pettigrew JD. 1988a. Retinal ganglion cell topography in teleosts: a
comparison between nissl-stained material and retrograde labelling from the
optic nerve. The Journal of Comparative Neurology 276(3):412-422.
Collin SP, Pettigrew JD. 1988b. Retinal topography in reef teleosts. I. Some species
Collin SP, Pettigrew JD. 1988c. Retinal topography in reef teleosts. II. Some species
with prominent horizontal streaks and high-density areae. Brain, Behavior and
Collin SP, Pettigrew JD. 1989. Quantitative comparison of the limits on visual spatial
resolution set by the ganglion cell layer in twelve species of reef teleosts. Brain,
Collin SP. 1997. Specialisations of the teleost visual system: adaptive diversity from
shallow-water to deep-sea. Acta physiologica Scandinavica Supplementum
638:5.
247
Collin SP, Hoskins RV. 1997. Topographic mismatching of the displaced amacrine and
ganglion cell populations in the retina of deep-sea lanternfish. Society for
Neuroscience 23:729.
Collin SP, Hoskins RV, Partridge JC. 1997. Tubular eyes of deep-sea fishes: a
comparative study of retinal topography. Brain, Behavior and Evolution
50(6):335-357.
Collin SP, Lloyd DJ, Wagner H-J. 2000. Foveate vision in deep-sea teleosts: a
comparison of primary visual and olfactory inputs. Proceedings of the Royal
Society of London Series B, Biological Sciences 355:1315-1320.
Collin SP, Shand J. 2003. Retinal sampling and the visual field in fishes. In: Collin SP,
Marshall NJ, editors. Sensory Processing in Aquatic Environments: Springer
New York. p 139-169.
Collins MA, Stowasser G, Fielding S, Shreeve R, Xavier JC, Venables HJ, Enderlein P,
Cherel Y, Van de Putte A. 2012. Latitudinal and bathymetric patterns in the
distribution and abundance of mesopelagic fish in the Scotia Sea. Deep-Sea
Research Part II: Topical Studies in Oceanography 59:189-198.
41(1):59-70.
Fritsches KA, Marshall NJ, Warrant EJ. 2003. Retinal specializations in the blue marlin:
eyes designed for sensitivity to low light levels. Marine and Freshwater
Research 54(4):333-341.
Gabriel R, Straznicky C. 1992. Immunocytochemical localization of parvalbumin- and

neurofilament triplet protein immunoreactivity in the cat retina: colocalization in
a subpopulation of AII amacrine cells. Brain Research 595(1):133-136.
Garza Gisholt E, Hemmi JM, Hart NS, Collin SP. In press. A comparison of spatial
ONE:In press.
Glaser EM, Wilson PD. 1998. The coefficient of error of optical fractionator population
size estimates: a computer simulation comparing three estimators. Journal of
Microscopy 192(2):163-171.
248
dimorphism in the sea. Journal of the Marine Biological Association of the UK
87(04):829-842.
Hughes A. 1975. A quantitative analysis of the cat retinal ganglion cell topography. The
Hughes A. 1977. The topography of vision in mammals of contrasting life style:

comparative optics and retinal organisation. In: Crescitelli F, editor. Handbook
of sensory physiology, vol VII/5. Berlin: Springer. p 613-756.
Jeon YK, Kim TJ, Lee JY, Choi JS, Jeon CJ. 2007. AII amacrine cells in the inner
nuclear layer of bat retina: identification by parvalbumin immunoreactivity.
Neuroreport 18(11):1095-1099.
Kristoffersen JB, Salvanes AGV. 1998. Effects of formaldehyde and ethanol

preservation on body and otoliths of Maurolicus muelleri and Benthosema
glaciale. Sarsia 83:95-102.
Litherland L, Collin SP. 2008. Comparative visual function in elasmobranchs: Spatial

arrangement and ecological correlates of photoreceptor and ganglion cell
distributions. Visual Neuroscience 25(04):549-561.
Handbook of sensory physiology, vol II/5. Berlin: Springer-Verlag. p 67-192.
Mack AF, Sussmann C, Hirt B, Wagner HJ. 2004. Displaced amacrine cells disappear
from the ganglion cell layer in the central retina of adult fish during growth.
Investigative Ophthalmology and Visual Science 45(10):3749-3755.
27:510-523.
Tiere 38:521-528.
249



Research. p 13-265.

Nelson R. 1982. AII amacrine cells quicken time course of rod signals in the cat retina.
Journal of Neurophysiology 47(5):928-947.

13:1-81.
Podrazhanskaya SG. 1993. Feeding habits of mesopelagic species of fish and estimation
19:79-85.
Slomianka L, West MJ. 2005. Estimators of the precision of stereological estimates: An

example based on the CA1 pyramidal cell layer of rats. Neuroscience
136(3):757-767.
Stone J. 1983. Parallel processing in the visual system. New York: Plenum Press. 438 p.
250
van Haesendonck E, Missotten L. 1987. Displaced small amacrine cells in the retina of
the marine teleost Callionymus lyra L. Vision Research 27(9):1431-1443.
Vaney DI, Peichl L, Boycott BB. 1981. Matching populations of amacrine cells in the
inner nuclear and ganglion cell layers of the rabbit retina. The Journal of
Wagner H, Wagner E. 1988. Amacrine cells in the retina of a teleost fish, the roach
(Rutilus rutilus): a Golgi study on differentiation and layering. Philosophical
Transactions of the Royal Society of London Series B, Biological Sciences
321:263-324.
17(4):637-685.
Wässle H, Grünert U, Röhrenbeck J. 1993. Immunocytochemical staining of AII-

amacrine cells in the rat retina with antibodies against parvalbumin. The Journal
of Comparative Neurology 332(4):407-420.

Weruaga E, Velasco A, Brinon JG, Arevalo R, Aijon J, Alonso JR. 2000. Distribution
of the calcium-binding proteins parvalbumin, calbindin D-28k and calretinin in
the retina of two teleosts. Journal of Chemical Neuroanatomy 19(1):1-15.
West MJ, Slomianka L, Gundersen HJG. 1991. Unbiased stereological estimation of the
total number of neurons in the subdivisions of the rat hippocampus using the
optical fractionator. The Anatomical Record 231(4):482-497.
Wong ROL, Hughes A. 1987. The morphology, number, and distribution of a large
population of confirmed displaced amacrine cells in the adult cat retina. The
Zoology 607:1-69.
251
Chapter 7 General discussion
CHAPTER 7
General discussion
253
7.1. Myctophid visual system

The main aim of this study was to better understand the visual adaptations of
deep-sea fishes in relation to their environment. The work presented in this thesis used a
multidisciplinary approach to investigate the visual adaptations, mainly present at the
ocular and retinal levels, of 61 species of lanternfishes, representing more than 50% of
the recognised genera within the family. Findings from this thesis confirmed that
myctophids possess a visual system particularly adapted to the visual constraints of the
mesopelagic environment by possessing very sensitive eyes, optimised for photon
capture. It was revealed that most lanternfishes possess large eyes, an aphakic gap, a
tapetum lucidum, a pure rod retina composed of a single bank of rod photoreceptors, a
high summation ratio of photoreceptors to ganglion cells and a single visual pigment
(Vilter, 1951; Lawry, 1974; O'Day and Fernandez, 1976; Pankhurst, 1987; Turner et al.,
2009, Chapter 2, 3 and 4).
Four new visual specialisations, a fundal pigmentation, microtubular-like

structures (MLS), an intra-ocular filter (retinal yellow pigment), and a streak-like
elongated area were discovered and described (Chapter 2, 5, 6). Although the function
of the MLS is still unclear, the three other specialisations are considered to be specific
adaptations to the visual environment of the mesopelagic zone. While the fundal
pigmentation may in fact be a reminiscent of a visual specialisation important at larval
stages, which changes as the visual environment transitions between the epipelagic zone
(larvae) and the mesopelagic zone (juveniles/adults), the yellow pigmentation enhances
the contrast and improves the detection of bioluminescent flashes, the main source of
light in the mesopelagic zone. Finally, the streak-like elongated areae, formed by a high
density of ganglion cells subtending the frontal and dorsal regions of the visual field,
may allow the detection of silhouettes situated above them against the lighter sunlit
background of the upper mesopelagic zone.
Myctophids also possess very poor spatial resolving power compared to shallow
water representatives and even other deep-sea teleosts, indicating that the family may
constitute one of the less acute groups of mesopelagic fishes (Wagner et al., 1998;
Chapter 6). In the relatively dark environment of the mesopelagic zone, where
interactions with other individuals (prey, predators, congeners) are mostly achieved
through bright, bioluminescent flashes, sensitivity is more important than acuity. The
visual system of lanternfishes has adapted well to these constraints by possessing
254
several specialisations to enhance sensitivity.
7.2. Interspecific variability

The Myctophidae show a marked diversity in their ecology and behaviour by
frequenting different depths, exhibiting different vertical migration patterns and
luminous tissue patterns. There are marked differences in the types of visual adaptation
and interspecific variability in the visual system of myctophids was surprisingly high at
all levels (i.e. ocular, retinal and molecular). At the two extremes are, for example,
representatives of the genera Myctophum and Lampanyctus. Species from the genus
Myctophum possessed highly specialised eyes with a large eye size, the highest rod
density, the highest spatial resolving power and the highest density of microtubular-like
structures (MLS). They also possessed a tapetum lucidum covering the entire retina, a
localised aphakic gap and no fundal pigmentation (an indication of high sensitivity).
Finally, Myctophum sp were one of a group of species possessing a sexually dimorphic
retinal yellow pigment, constituting the first record of sexual dimorphism in the visual
system of any vertebrate, in addition to possessing two rod opsins. On the other hand,
Lampanyctus species possessed a less specialised visual system with a small eye size,
low rod densities, a poorly specialised retina, low acuity, fundal pigmentation, variable
coverage of a tapetum lucidum and an aphakic gap.
The data reveal that both ecological and phylogenetic constraints drive the
evolution of the visual system in lanternfishes. While some visual specialisations are
strongly influenced by phylogeny, i.e. eye size (Chapter 2), fundal pigmentation
(Chapter 3), MLS (Chapter 3), yellow pigment (Chapter 5), type of retinal
specialisations (Chapter 6) and visual pigments (Turner et al., 2009), others appear to be
driven by ecological constraints, i.e. the aphakic gap and tapetum lucidum (Chapter 3),
rod diameter and length (Chapter 4) and some types of retinal specialisations (Chapter
6). The main ecological factor influencing photoreceptor design, and consequently
sensitivity, was the depth range at night, indicating that vision at night is of great
importance in lanternfishes (Chapter 4). This is not surprising when most myctophids
are known to perform their main survival tasks, i.e. feeding and reproducing in the
surface layers at night.
255
Myctophids possess visual pigments spectrally tuned to maximise their

sensitivity to the bioluminescent emissions present in their environment rather than to
the downwelling sunlight (Douglas et al., 1998; Turner et al., 2009; Chapter 5).
Bioluminescence in lanternfishes is thought to subtend several different interactions, i.e.
with prey, predators and congeners (Edwards and Herring, 1977). Due to the common
occurrence of sexually dimorphic luminous organs in myctophids, a role in sexual
selection has also been hypothesised (Herring, 2007). Tsarin (2001) noted that
lanternfish species possessing sexually dimorphic caudal organs had larger eyes than
non-sexually dimorphic species. However, findings from this study did not confirm this
observation and no correlation between relative eye size and sexual dimorphism in
luminous tissues could be revealed (de Busserolles et al., 2013; Chapter 2).
Nevertheless, some other relationships between visual characteristics and sexual
dimorphism in luminous tissues did exist. Species with a sexual dimorphism in
luminous tissues possessed higher rod densities, an adaptation for higher sensitivity
(Chapter 4). Moreover, species with the highest rod densities (Myctophum sp) also
possessed a species-specific sexually dimorphic intra-ocular filter (yellow pigment,
Chapter 5) enhancing contrast and improving the detection of bioluminescent signals.
The fact that species with sexually dimorphic luminous tissues also possessed more
sensitive eyes and a sexually dimorphic visual system strengthens a long standing
hypothesis proposed by a range of authors i.e. Barnes and Case, (1974), Edwards and
Herring, (1977) and Herring, (2007) that bioluminescence is used for intraspecific
communication in lanternfishes.
7.3. Conclusions and future directions

In conclusion, both ecological and phylogenetic constraints have influenced the
evolution of the myctophid visual system, resulting in highly specialised and sensitive
eyes that are particularly well adapted to the visual environment of the mesopelagic
zone. Moreover, the presence of new visual specialisations in some species and the
extreme interspecific variability observed, most likely indicates the presence of different
strategies for survival in the mesopelagic zone to match the ecological and behavioural
variability seen within the family. However, a lot remains to be learned about the
biology, ecology and physiology of this large and diverse family. More particularly, the
function and relative importance of their other sensory systems is still poorly
understood and constitutes an important area of research if one wants to better
256
comprehend their behaviour.
Findings from this study suggest that different myctophid species might use
various sensory strategies for survival in the mesopelagic zone. While some lanternfish
species may rely heavily on vision by possessing a well developed and sensitive visual
system, other species might rely on other sensory systems. Since the structure and size
of the different brain areas usually clearly reflects differences in sensory specialisations
in teleosts fishes (Lisney and Collin, 2006; Wagner, 2001a,b), analysis of the brain of
different species possessing different visual capabilities might shed some light on their
survival strategies. Wagner (2001a) investigated the size of different sensory brain areas
in six species of lanternfishes. While most of the species analysed had two sensory brain
regions above average in size, only two species out of six, from the genus Diaphus (D.
holti and D. mollis), were visually oriented. The other species were non-visual
specialists possessing somatosensory (lateral line + auditory), olfactory and gustatory
sensory brain regions with an above average size. However, Wagner (2001a) only
analysed representatives of the Lampanyctinae subfamily. Based on the results from this
study, we predict that most representative of the Myctophinae will be visual specialists,
with enlarged optic tecta.
Differences in the morphology of the pineal organ (a non-image-forming

photosensory organ located in the dorsal region of the brain) have also been observed
between species, and interspecific differences appear to follow phylogeny, with the
exception of the Diaphus genus (McNulty and Nafpaktitis, 1977). However, only seven
species were analysed and one can ask if the interspecific differences observed in the
morphology of the pineal organ could also be due to ecological constraints, especially
considering that myctophids appear to rely on this organ to detect the intensity of the
downwelling light (Young et al., 1979).
Regarding the visual system of myctophids, many more questions remain to be

answered. At the retinal level, the function(s) of the microtubular-like structures (MLS),
the large population of “displaced” amacrine cells within the ganglion cell layer and the
interspecific differences in inner segment length remain to be investigated. Analyses of
the type of retinal ganglion cells and their specific distribution would also bring
additional information about the visual capabilities of this family (i.e. distribution of the
giant ganglion cells implicated in movement detection). At the molecular level, more
257
work needs to be done to identify when, in the evolutionary history of the family, the
rod duplication event occurred and confirm the association of the long-wavelength
shifted rod with the retinal yellow pigmentation. In order to answer some of these
questions, studies of the ontogenetic development of the myctophid visual system
would need to be carried out, especially regarding the function of some of the new
specialisations described i.e. MLS, fundal pigmentation. Retrograde labelling of the
ganglion cells and immunohistochemical characterisation of the ganglion and amacrine
cells would allow a more informed prediction of their function, while an increase in the
number of individuals analysed per species would permit an estimation of the
intraspecific variability in the different visual specialisations.
Due to their high species diversity and range of ecological niches, myctophids
represent a very interesting group for visual adaptation studies. While 61 species were
analysed in this study, 189 species remain to be examined, some of which possess
atypical characteristics that would be of high interest to better understand the evolution
of the myctophid visual system. For example, it would be particularly interesting to
analyse the visual system of Taaningichthys paurolychnus, the only myctophid without
photophores, or species from the genus Protomyctophum, which uniquely possess
dorso-lateral rather than lateral eyes (Hulley, 1990). Moreover, species exhibiting a
particular behaviour, such as non-migrating species (i.e. Taaningichthys bathyphilus,
Paxton, 1967; Hulley, 1984; Stenobrahius nannochir, Watanabe et al., 1999; Moku et
al., 2000; Kojima et al., 2009) or inhabiting unusual environments (i.e. Benthosema
pterotum in the Red Sea) would also be of high interest. In fact, the Red Sea, for
example, constitutes an unusual habitat for deep-sea organisms as a result of the
extremely warm temperatures. While surface temperatures average 26°C, the
temperature in the mesopelagic zone is a constant 20°C (Dypvik and Kaartvedt, 2013).
One species of myctophid, B. pterotum is dominant in Red Sea and exhibits normal
depth distribution and diel vertical migration behaviours (Dypvik and Kaartvedt, 2013).
At low light levels, visual signals can significantly be contaminated by dark noise, an
occasional activation of the process of transduction without the presence of any light.
While the level of dark noise is unknown for deep-sea animals, low temperatures are
thought to keep the rate very low (Warrant, 2004). However, this would be a completely
different story in the Red Sea and one can ask how has the visual system of B. pterotum
adapted to cope with potentially high levels of dark noise associated with these warmer
temperatures? Finally, a comparative analysis of the visual system of the
258
Neoscopelidae, the sister family of the Myctophidae and a much more generalised
group with only six recognised species belonging to three genera, might bring some
additional information about the evolutionary history of vision in myctophids.
Since myctophids play a major role in the ocean ecosystem, good estimations of
their biomass is crucial in terms of conservation. Some lanternfish species are known to
actively avoid sampling nets resulting in an underestimation of their biomass (Kaartvedt
et al., 2012; Flynn and Kloser, 2012). If these species rely heavily on vision, a better
understanding of their visual capabilities may assist in the design of new and more
efficient sampling techniques. Since some of the specialisations appear to be highly
influenced by the evolutionary history of the family, predictions of the visual
capabilities of a range of species may be possible. However, in order to
comprehensively assess and understand the visual capabilities of myctophids, better
tools will be needed, a better resolved phylogeny, improved sampling techniques and
strategies for accurately assessing their depth distribution, and most importantly
information on their behaviour and visual tasks through observations of live animals in
situ and behavioural experiments in the laboratory.
7.4. References
17(4):597-636.
Dypvik E, Kaartvedt S. 2013. Vertical migration and diel feeding periodicity of the
skinnycheek lanternfish (Benthosema pterotum) in the Red Sea. Deep-Sea
Research Part I: Oceanographic Research Papers 72:9-16.
41(1):59-70.
259
Flynn A, Kloser R. 2012. Cross-basin heterogeneity in lanternfish (family Myctophidae)

assemblages and isotopic niches (δ13C and δ15N) in the southern Tasman Sea
abyssal basin. Deep Sea Research Part I: Oceanographic Research Papers
69:113-127.
178.
Kaartvedt S, Staby A, Aksnes DL. 2012. Efficient trawl avoidance by mesopelagic

fishes causes large underestimation of their biomass. Marine Ecology Progress
Series 456:1-6.
Kojima S, Moku M, Kawaguchi K. 2009. Genetic diversity and population structure of

three dominant myctophid fishes (Diaphus theta, Stenobrachius leucopsarus,
and S-nannochir) in the North Pacific Ocean. Journal of Oceanography
65(2):187-193.
247(5437):155-157.
Lisney TJ, Collin SP. 2006. Brain morphology in large pelagic fishes: a comparison
between sharks and teleosts. Journal of Fish Biology 68(2):532-554.
McNulty JA, Nafpaktitis BG. 1977. Morphology of pineal complex in 7 species of

lanternfishes (pisces Myctophidae). American Journal of Anatomy 150(4):509-
529.

myctophid fishes, Diaphus theta, Stenobrachius leucopsarus and S. nannochir,

Paxton JR. 1967. A distributional analysis for the lantern fishes (family Myctophidae)
of the San Pedro Basin, California. Copeia 1967:422-440.
260
Tsarin SA. 2001. Morphology of luminous organs of mesopelagic fishes in one group of
Myctophidae. Journal of Morphology (Abstract, Sixth International Congress of
Vertebrate Morphology) 248(3):292.
56(6):1003-1017.

17(4):637-685.
Wagner H-J. 2001a. Sensory brain areas in mesopelagic fishes. Brain, Behavior and
Evolution 57(3):117-133.
Wagner H-J. 2001b. Brain areas in abyssal demersal fishes. Brain, Behavior and
Evolution 57(6):301-316.
Warrant E. 2004. Vision in the dimmest habitats on Earth. Journal of Comparative

Physiology A: Sensory Neural and Behavioral Physiology 190(10):765-789.

261
Appendix 1
APPENDIX 1
263
Eye-Size Variability in Deep-Sea Lanternfishes
(Myctophidae): An Ecological and Phylogenetic Study
Fanny de Busserolles1*, John L. Fitzpatrick2,3, John R. Paxton4, N. Justin Marshall5, Shaun P. Collin1
1 Neuroecology Group, School of Animal Biology and the Oceans Institute, The University of Western Australia, Crawley, Western Australia, Australia, 2 Centre for
Evolutionary Biology, School of Animal Biology, The University of Western Australia, Crawley, Western Australia, Australia, 3 Computational and Evolutionary Biology,
Faculty of Life Sciences, University of Manchester, Manchester, United Kingdom, 4 Ichthyology, Australian Museum, Sydney, New South Wales, Australia, 5 Sensory
Neurobiology Group, Queensland Brain Institute, University of Queensland, Brisbane, Queensland, Australia
Abstract
One of the most common visual adaptations seen in the mesopelagic zone (200–1000 m), where the amount of light
diminishes exponentially with depth and where bioluminescent organisms predominate, is the enlargement of the eye and
pupil area. However, it remains unclear how eye size is influenced by depth, other environmental conditions and phylogeny.
In this study, we determine the factors influencing variability in eye size and assess whether this variability is explained by
ecological differences in habitat and lifestyle within a family of mesopelagic fishes characterized by broad intra- and
interspecific variance in depth range and luminous patterns. We focus our study on the lanternfish family (Myctophidae)
and hypothesise that lanternfishes with a deeper distribution and/or a reduction of bioluminescent emissions have smaller
eyes and that ecological factors rather than phylogenetic relationships will drive the evolution of the visual system. Eye
diameter and standard length were measured in 237 individuals from 61 species of lanternfishes representing all the
recognised tribes within the family in addition to compiling an ecological dataset including depth distribution during night
and day and the location and sexual dimorphism of luminous organs. Hypotheses were tested by investigating the
relationship between the relative size of the eye (corrected for body size) and variations in depth and/or patterns of
luminous-organs using phylogenetic comparative analyses. Results show a great variability in relative eye size within the
Myctophidae at all taxonomic levels (from subfamily to genus), suggesting that this character may have evolved several
times. However, variability in eye size within the family could not be explained by any of our ecological variables
(bioluminescence and depth patterns), and appears to be driven solely by phylogenetic relationships.
Citation: de Busserolles F, Fitzpatrick JL, Paxton JR, Marshall NJ, Collin SP (2013) Eye-Size Variability in Deep-Sea Lanternfishes (Myctophidae): An Ecological and
Phylogenetic Study. PLoS ONE 8(3): e58519. doi:10.1371/journal.pone.0058519
Editor: Eric James Warrant, Lund University, Sweden
Received November 16, 2012; Accepted February 5, 2013; Published March 5, 2013
Copyright: ß 2013 de Busserolles et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was funded by the Australian Research Council (Discovery grant to SPC and a Linkage Grant to NJM and SPC) and the West Australian State
Government. JLF is supported by an Australian Research Council Postdoctoral Fellowship. The funders had no role in study design, data collection and analysis,
decision to publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interest exist.
* E-mail: [email protected]
Introduction of downwelling sunlight. Although the low amount of sunlight in

this zone negates the process of photosynthesis, enough light is
The detection of light signals is one of the most important means present in the water column to create an extended visual scene
by which organisms perceive and react to their surroundings. In both vertically and horizontally [4], thereby allowing animals to
bony fishes (infraclass Teleostei), like all other vertebrates, light detect the silhouettes of potential prey items against a lighter
detection is achieved by a highly conserved structure, the eye, in background when viewed from below [5]. The mesopelagic zone
which a single lens focuses an image on to the neural retina. Since also contains the greatest biomass and diversity of animals in the
the cornea is not refractive underwater [1], the lens of most teleosts mid-waters of the ocean [6], many of which are bioluminescent,
is spherical and its radius and the distance from the lens centre to the producing light signals using either symbiotic bioluminescent
retina (focal length) are related by a ‘‘constant’’ defined, over 100 bacteria (i.e. anglerfish [7]) or their own enzymatic complex
years ago, as Matthiessen’s ratio [2], [3], regardless of the size of the (luciferin-luciferase, i.e. viperfish [8]). Due to the low levels of
eye. Thus, the acuity (quality of the image) of the teleost eye is sunlight and the predominance of small bioluminescent flashes,
directly related to eye size and lens diameter but is also influenced by vision is considered very important in the twilight zone and seems
other factors such as the quality of the optics, the amount of light to be the dominant sense used by its inhabitants [9]. To be able to
received (governed by the aperture or pupil size) and the degree of see in dim conditions and for viewing bioluminescence, the visual
overlap between the dendritic fields of neighbouring retinal system of mesopelagic fishes requires higher sensitivity than acuity
receptors. As visual acuity is limited by the amount of light [10]. To enhance sensitivity, species have adapted to optimize light
available, the evolution of the visual system in teleosts is also collection and extend the visual field [11]. One of the most
expected to be influenced by the depth at which an individual lives. extreme ocular adaptations at these depths (200–1000 m) is the
The mesopelagic zone (200–1000 m), also referred to as the tubular shape of the eyes of a number of species (i.e. Argyropelecus
twilight zone, is characterised by exponentially diminishing levels sp., Winteria sp.), which are directed upwards (and rostrally in some
PLOS ONE | www.plosone.org 1 March 2013 | Volume 8 | Issue 3 | e58519

Lanternfish Eye Size
species) to optimise light capture of the downwelling sunlight [12], be explained by ecological differences in habitat and lifestyle
[6], [13], [14]. However, the most important morphological within a single family of fishes characterized by broad intra- and
adaptation of the visual system of mesopelagic fishes is arguably interspecific variance in depth range and luminous patterns.
the enlargement of their eyes compared to body size [15]; a larger Specifically, we focus our investigation on lanternfishes (family
eye will increase the chance of photon capture (greater pupillary Myctophidae), one of the most abundant group of mesopelagic
aperture), thereby allowing improved detection of body silhouettes fishes in the world’s oceans [18], with some 250 species in 33
and bioluminescent flashes against an increasingly dim back- genera currently recognised [19]. Most species exhibit extensive
ground [5], [1], [16]. diel vertical migration toward the surface at night in order to feed
Deeper in the ocean, in the bathypelagic zone (1000–4000 m), and a great interspecific variability in their migration patterns has
where no downwelling sunlight penetrates, eye size tends to been observed [20]. Individuals can migrate from the epipelagic
decrease with some species of teleosts having such small eyes they zone at night to the upper part of the bathypelagic zone during the
were considered ‘‘degenerate’’ [17]. The reduction in eye size in day. Myctophids produce bioluminescence using a luciferin-
inhabitants of the bathypelagic zone could be explained by the luciferase reaction [21], [22]. They possess two kinds of luminous
predominant use of other sensory systems and/or the brighter structures, with highly variable patterns, that light up indepen-
appearance of the bioluminescent signals in the absence of residual dently. Those are the ventral and ventrolateral photophores which
downwelling sunlight. The bathypelagic zone represents a visual are thought to play a role in counter-illumination [23] and species
scene composed only of bright, point sources of light viewed recognition [24], and the luminous organs and tissue patches,
against a completely dark background. To detect these intermit- which are frequently sexually dimorphic, are located on the caudal
tent sources of light, the eyes do not need to be very large as the peduncle and/or head and/or body and may play several different
flashes will appear extremely bright compared to the background roles including intra- and interspecific communication, prey
[16]. In addition to the absence of residual daylight, the illumination and distraction [25]. Consequently, we hypothesised
bathypelagic zone is also characterised by a drastic diminution that lanternfishes with a deeper distribution and/or a reduction of
in animal biomass and biodiversity [6]. An individual living in this bioluminescent emissions will have smaller eyes and that ecological
zone will encounter fewer animals and therefore less biolumines- factors rather than phylogenetic relationships will drive the
cent signals (the only type of light present in the bathypelagic evolution of their visual system. We test these predictions by
zone). In this zone, vision may be used secondarily after a potential investigating the relationship between the relative size of the eye
prey or mate has been detected using other sensory modalities. (corrected for body size) and variations in depth and/or luminous-
More than half a century ago, Marshall [15] compared the eye organ patterns in the Myctophidae using phylogenetic compara-
size of three species of Gonostoma living in different oceanic zones, tive analyses.
the upper mesopelagic (G. denudatum), lower mesopelagic (G.
elongatum) and the bathypelagic (G. bathyphilum) zones and observed Materials and Methods
that the deeper the species the smaller the size of the eye. For
species that are restricted to a particular zone, this inverse Ethics statement
relationship between depth and eye size appears intuitive. For cruises 1–4 (Table 1), sampling was carried out under the
However, some species may frequent all three zones (epipelagic, following collection permits: Coral Sea waters (CSCZ-SR-
mesopelagic, bathypelagic) during their lifetime or even within the 20091001-01), Commonwealth waters (AU-COM2009051),
same day (diel vertical migrations). How does eye size vary in these GBRMPA (G09/32237.1) and Queensland Fisheries (133805),
species? Despite the importance of considering depth, in terms of (Marshall, AEC # SNG/080/09/ARC). For cruises 5–6,
light availability, when evaluating the evolution of the visual sampling permits were obtained by the Chief Scientist of the
system in deep-sea fishes, it remains unclear how eye size is shaped respective cruises (AIMS, University of Tübingen) for their target
by depth and other environmental conditions. Consequently, in species. We obtained our samples as by-catch and therefore no
this study we aim to assess to what extent variability in eye size can collection permits were required. Most individuals caught were
Table 1. Summary of the research cruises from which the samples were collected, together with their geographic location, fishing
equipment, time of sampling and fixative used.
Cruise Location Date Equipment Time of sampling Fixation
1 Coral Sea 11/2009 RMT*/plankton net/neuston net Night 4% PFA, Karnovsky

3 Coral Sea 12/2010 RMT* Night 4% PFA, Karnovsky
6 Peru-Chile Trench 09/2010 RMT*/neuston net Day, Night 5% formalin, 2% glutaraldehyde
7 Bay of Biscay, North East Atlantic 10/2009 Bottom trawl GV1001 Day Karnovsky
8 Balearic Islands, Western 07/2010 Double-warp modified commercial Day, Night Karnovsky
Mediterranean Sea mid-water trawl/IKMT2
9 Western Australia 07/2010 EZ net* Night 5% formalin
1
Sourced from Olivar et al. [74],
2
sourced from Mahé & Poulard [75]. EZ = multiple plankton net system; IKMT = Isaacs-Kidd Midwater Trawl; RMT = Rectangular Midwater Trawl.
*Indicates the use of an opening-closing device.
doi:10.1371/journal.pone.0058519.t001

already deceased; however, moribund animals were humanely presenting an enlarged Dn and/or Vn were given a special
euthanized following the guidelines of the NH&MRC Australian category in this study. Depending on the species, sexual
Code of Practice, under our University of Western Australia dimorphism in luminous tissues can be seen in the Dn/Vn,
Animal Ethic protocol (RA/3/100/917). Tissue samples obtained caudal luminous organs and/or luminous patches (Table 2).
from collaborators (cruises 7–9) did not require any UWA Differentiation of the type of sexual dimorphism in our analyses
collection or animal ethics permits. did not show significant differences. As a consequence, only results
for the presence/absence of sexually dimorphic features are
Data collection and morphometric measurements presented in this study (Table 2). For each species, the juvenile/
A total of 237 lanternfishes from 61 species and 19 genera were adult depth distributions during night and day were recorded,
analysed in this study. Samples were obtained from different taking into consideration the individual size and area of sampling
research cruises in different parts of the world and with different when possible (Table S2, S3). Published studies using opening-
methods of sampling (Table 1). For each individual, the standard closing sampling devices [34], [27], [35], [36], [37], [38], [39] and
length, rostro-caudal eye diameter (measured in situ) and lens differentiating life stages [40], [27], [35], [38] were given priority
diameter were measured with digital calipers to 0.1 mm. For most when making the dataset. Categorised depth ranges were created
of the lanternfish, measurements were performed on fresh for the statistical analyses based on the published data and our
specimens on board ship prior to fixation. However, when own capture-depth data. The same number of depth classes was
samples were acquired from collaborators (Cruises 7, 8, 9, see used for both diurnal and nocturnal depth distribution in order to
Table 1), the measurements were made after fixation. The fixatives statistically compare the results. The group’s cut off, during day
used in those cases were 5% buffered formalin and Karnovsky’s and night, was chosen at the most meaningful depth in terms of
fixative (2.5% paraformaldehyde and 2% glutaraldehyde in 0.1 M vision and accuracy of the depth data. Three group categories
cacodylate buffer). Since we were interested in the eye size: body were created in terms of the amount of downwelling light present:
size relationship and because formaldehyde has previously been moderate light level (0–5 m at night, 200–500 m during the day),
found to affect body length by only 0.8% [26], no measurement low light level (5–100 m at night, 500–900 m during the day), no
correction was used between fresh and fixed specimens and all the light (,100 m at night, ,900 m during the day). For the night-
morphometric data for each species were pooled. The life stage of depth range, species (excluding larvae) were classified according to
each specimen was estimated by length measurements published their shallowest depth recorded. In the surface category at night,
in the literature. With the exception of sexually dimorphic only species dip-netted or sampled with a neuston net (surface)
features, juvenile and adult lanternfishes are identical in appear- were included. For the day-depth range, species (excluding larvae)
ance (i.e. their body proportions, pigment and photophore were classified according to their deepest depth recorded.
patterns, [27]) and since the regression slopes of eye diameter
versus standard length were not significantly different between the Phylogenetic analyses
two stages (Table S1), these two groups were analysed together. Standard statistical analyses assume independence of the
One specimen of Scopelengys tristis of the family Neoscopelidae, samples. This assumption is unfortunately not met when
sister family of the Myctophidae, was also measured for comparing different species as more closely related species are
comparison. expected to be more similar to one another due to the share of a
common ancestor. Therefore, all data analyses were performed
Taxonomic remarks using phylogenetic comparative analyses to account for the shared
Most of the samples from the Coral Sea (Cruises 1–5) and Chile- history among species [41]. Unfortunately, no fully resolved
Peru Trench (Cruise 6) are registered as voucher specimens at the phylogeny is available for the family Myctophidae to date.
Australian Museum in Sydney, Australia. However, further Consequently, two different phylogenies, A and B (Figure 1), were
taxonomic analyses need to be carried out for six of our study built in the Mesquite program v. 2.75 [42] based on two published
species to confirm identification. A comment for each of these six phylogenies [43], [44]. Paxton et al.’s phylogeny classified genera
species is given below. Lampanyctus vadulus: requires confirmation of using derived character states of adult osteology and photophore
some northeastern Australian variants of L. nobilis currently under patterns, as described by Paxton et al. [43], and of larvae as
study. Myctophum spinosum/M. lychnobium: the characters distin- described by Moser and Ahlstrom [45], [46], [47]. The phylogeny
guishing these two species appear to form a continuum in the divided the family into two subfamilies (Myctophinae and
western South Pacific; the specimens studied here are the two Lampanyctinae) and seven tribes (Electronini, Myctophini,
extremes of the continuum that might be all M. spinosum. Gonichthyini, Diaphini, Gymnoscopelini, Lampanyctini and
Symbolophorus cf. boops: Symbolophorus from the eastern South Pacific Notolychnini, Figure 1A). The only difference between Paxton’s
require more study to allow species identification. Nannobrachium cf. originally described phylogeny and the one used in the present
nigrum: the specimens used match the description in Zahuranec analysis (phylogeny A) is the inclusion of the genus Nannobrachium,
[28], but not the figure and brief description of the holotype in which was added to Paxton’s phylogeny after Zahuranec [28]. The
Nafpaktitis et al. [29]. Triphoturus oculeus: detailed examination of phylogeny of Poulsen et al. [44] was the first molecular phylogeny
the specimens from the eastern South Pacific to distinguish any for the family and used mitogenomic results from DNA sequences
possible T. mexicanus has not been completed to date. and unique gene orders from 38 lanternfish species. Poulsen et al.
[44] confirmed the presence of the two subfamilies (Myctophinae
Ecological data and Lampanyctinae) and identified 10 monophyletic lineages or
A dataset of the location and sexual dimorphism of the luminous clades (Figure 1B). The genus Hygophum was added to Poulsen et
tissue was created using information found in the literature al.’s originally described phylogeny in our study (phylogeny B),
(Table 2). Presence-absence of luminous organs (head, caudal), although no clade was assigned, and its position in the phylogeny
additional luminous patches and sexual dimorphism in luminous was kept identical to Paxton et al.’s phylogeny. The main
tissues were noted [30], [31], [29], [32], [28], [33]. All lanternfish differences seen in Poulsen et al.’s phylogeny compared to Paxton
genera possess a dorsal nasal organ (Dn) and/or a ventral nasal et al’s are the taxon Notolychnus, which became a sister taxon of all
organ (Vn) on their head associated with the eye. Only species the remaining myctophids, and the tribe Diaphini, which became

Table 2. Dataset used in the phylogenetic comparative analyses of the location of the luminous tissue (Dn/Vn, Caudal, Body
patches), whether there was any level of sexual dimorphism present (Sex. dim. Sex. D., Sex. C. Sex. P.) and the depth categories
during night and day.
Benthosema glaciale 0 1 0 1 0 1 0 1 1
B. suborbitale 0 1 0 1 0 1 0 1 1
Bolinichthys longipes 0 1 1 1 0 0 1 1 1
B. nikolayi 0 1 1 0 0 0 0 2 1
B. supralateralis 0 1 1 0 0 0 0 1 1
Centrobranchus andreae 0 1 0 1 0 1 0 2 1
Ceratoscopelus maderensis 0 1 1 0 0 0 0 1 2
C. warmingii 0 1 1 0 0 0 0 1 2
Diaphus brachycephalus 1 0 0 1 1 0 0 1 1
D. danae 1 0 1 0 0 0 0 1 1
D. fulgens 1 0 1 1 1 0 1 1 2
D. garmani 1 0 1 1 1 0 0 1 1
D. holti 1 0 1 1 1 0 0 1 1
D. luetkeni 1 0 1 1 1 0 0 1 1
D. meadi 1 0 1 1 1 0 1 1 2
D. mollis 1 0 1 1 1 0 0 1 1
D. parri 1 0 1 1 1 0 0 1 1
D. phillipsi 1 0 1 0 0 0 0 1 1
D. regani 1 0 1 0 0 0 0 1 2
D. splendidus 1 0 1 1 1 0 0 1 1
D. termophilus 1 0 1 1 1 0 0 1 1
D. whitleyi 1 0 1 1 1 0 0 2 1
Diogenichthys atlanticus 1 1 0 1 1 1 0 1 2
D. laternatus 1 1 0 1 1 1 0 0 0
Electrona risso 0 1 0 1 0 1 0 1 1
Gonichthys tenuiculus 0 1 0 1 0 1 0 0 n.a.
Hygophum benoiti 0 1 0 1 0 1 0 1 2
H. hygomii 0 1 0 1 0 1 0 1 2
H. proximum 0 1 0 1 0 1 0 0 1
Lampadena luminosa 0 1 0 0 0 0 0 1 2
L. urophaos 0 1 0 0 0 0 0 1 1
Lampanyctus alatus 0 1 1 0 0 0 0 1 1
L. crocodilus 0 1 1 0 0 0 0 1 2
L. iselinoides 0 1 1 0 0 0 0 1 1
L. nobilis 0 1 0 0 0 0 0 1 2
L. omostigma 0 1 0 0 0 0 0 1 0
L. parvicauda 0 1 0 0 0 0 0 1 2
L. pusillus 0 1 0 0 0 0 0 1 2
L. vadulus 0 1 0 0 0 0 0 1 2
Lobianchia dolfleini 0 1 0 1 0 1 0 1 1
L. gemellari 0 1 0 1 0 1 0 1 1
Loweina interrupta 0 1 0 1 0 1 0 1 n.a.
Myctophum asperum 0 1 0 1 0 1 0 0 1
M. aurolaternatum 0 1 0 1 0 1 0 0 2
M. brachygnathum 0 1 0 1 0 1 0 0 0
M. lychnobium 0 1 0 1 0 1 0 0 0
M. nitidulum 0 1 0 1 0 1 0 0 1
M. obtusirostre 0 1 0 1 0 1 0 0 1

Table 2. Cont.
M. spinosum 0 1 0 1 0 1 0 0 1
Nannobrachium cf. nigrum 0 1 0 0 0 0 0 2 1
N. idostigma 0 1 0 0 0 0 0 1 0
N. phyllisae 0 1 0 0 0 0 0 2 n.a.
Notolychnus valdiviae 0 1 0 1 0 1 0 1 2
Notoscopelus elongatus 1 1 1 1 0 1 0 1 2
N. kroeyerii 1 1 1 1 0 1 0 1 2
Symbolophorus cf. boops 0 1 0 1 0 1 0 0 1
S. evermanni 0 1 0 1 0 1 0 0 1
S. rufinus 0 1 0 1 0 1 0 0 1
S. veranyi 0 1 0 1 0 1 0 0 1
Triphoturus nigrescens 0 1 0 0 0 0 0 2 2
T. oculeus 0 1 0 0 0 0 0 1 1
Sex. dim. = sexual dimorphism in luminous tissue, Sex. D. = sexual dimorphism in Dn/Vn luminous organs, Sex. C. = sexual dimorphism in caudal luminous organs, Sex.
P. = sexual dimorphism in luminous tissue patches, 0 = character absent, 1 = character present. For the night depth range, 0 = 0–5 m, 1 = 5–100 m, 2 = ,100 m. For the
day depth range, 0 = 200–500 m, 1 = 500–900 m, 2 = .900 m, n.a. = missing values.
a sister tribe of the Lampanyctini. Due to the lack of resolution, and ecological traits were assessed using phylogenetic generalised
both phylogenies are only resolved to generic level, resulting in least squares regressions (PGLS, [51]) with the package APE in R
several polytomies (i.e. unresolved relationship among species). [52]. PGLS are classic generalised least squares regressions that
Unfortunately, the presence of polytomies prevents the application additionally take into account the shared history of the different
of many phylogenetic analyses that require a fully resolved species by incorporating phylogenetic information into the
phylogeny. Therefore, to bypass this problem, 100 alternative analyses. PGLS regressions estimate a phylogenetic scaling
phylogenies were generated with polytomies randomly resolved to parameter, l, using maximum likelihood methods to determine
infinitesimally small (1026) branch lengths using the Mesquite the degree of covariance in the residuals of the model, while
program v. 2.75 [42]. Ten of these phylogenies with randomly controlling for phylogenetic effects. This approach also examines if
resolved polytomies were selected at random to perform the the scaling parameter l significantly differs from 0 or 1 using
different analyses and the results between each of the 10 likelihood ratio tests, where l = 0 indicates no phylogenetic
phylogenies compared for consistency. Moreover, to fit the dependence in the data and l = 1 indicates strong phylogenetic
statistical requirements for the phylogenetic linear models association in the data [50], [51]. PGLS models were used to assess
described below, branch lengths were transformed using Grafen’s the relationship between morphological traits, to identify if eye size
method [48] with rho transformation set at 2.5 before all analyses. differs between the two subfamilies (Myctophinae, Lampanyctinae)
All statistical analyses were performed, using both phylogenies when correcting for the effects of body size, and to assess if eye
separately, on Log10-transformed species averages with the diameter was related with various ecological parameters. Standard
statistical program R v.2.15.0 (R Foundation for Statistical length was added as a covariate in all models.
Computing 2012).
Phylogenetic ANOVAs
To identify differences at the tribal (phylogeny A), cladal
Estimating phylogenetic signal
(phylogeny B) and generic (phylogenies A and B) levels,
The phylogenetic signal for continuous and discrete traits was
phylogenetically corrected residuals of the eye diameter were
estimated with Pagel’s lambda (l) using the package GEIGER in
calculated from eye diameter - standard length regression fit lines
R [49]. Pagel’s l is a measure of the degree of phylogenetic using PGLS. Statistical analyses on residuals are usually not
dependence in the data [50], meaning to which degree closely recommended as they often lead to biased results, especially if the
related species are more similar to each other than what is variables tested are colinear with the controlled variable [53].
expected by random evolutionary processes. Pagel’s l varies from However, when too many groups are present, phylogenetic
0 to 1, with l value of 1 indicating that traits gradually accumulate ANCOVA (i.e. analysis of covariance incorporating phylogenetic
changes over time in a Brownian motion process (i.e. random information) cannot sort out the differences using the PGLS
change in any direction) and l values of 0 indicating that no approach. Consequently, residuals were used in this particular case
phylogenetic signal is present and that traits have evolved in to estimate differences between groups (tribes, clades, genera)
response to selective processes. The observed l value for each trait using phylogenetic ANOVA (i.e. a classic ANOVA incorporating
was compared to l values of 0 and 1 using likelihood ratio tests phylogenetic information [54]) followed by a sequential Bonferroni
with df = 1. post-hoc test using the GEIGER [49] and PHYTOOLS [55]
packages in R. At the generic level comparison, both phylogenies
Phylogenetic linear models A and B were used separately and the results compared. Only
Relationships between morphological traits (eye diameter, lens groups with at least three observations were included in those
diameter, body size) and the relationship between morphological analyses. The genus Hygophum was excluded from the analyses at

Figure 1. Phylogenies of Myctophidae reconstructed from (A) Paxton et al. [43], (B) Poulsen et al. [44]. The red branches indicate the main
differences between the two phylogenies. Branch lengths are arbitrarily ultrametricized on the figure. In A, the numbers identify the different tribes of
Paxton et al. [43], 1. Electronini, 2.Myctophini, 3.Gonichthyini, 4.Notolychnini, 5.Lampanyctini, 6.Gymnoscopelini, 7.Diaphini. In B, the letters identify
the different clades of Poulsen et al. [44], A. Notolychnini, B. Diaphini, C. Notoscopelini, D-E-F. Lampanyctini, G. Electronini, H. Myctophini, I.
Myctophini (cycloid-species-group), J. Myctophini (ctenoid-species group)+Gonichthyini.
doi:10.1371/journal.pone.0058519.g001
the cladal level due to its hypothesized position in Poulsen et al.’s

phylogeny.
Results
Morphometric measurements
Eye size varied greatly within the lanternfish family (Figure 2).
The range of values for standard length, eye diameter and lens
diameter for each species is given in Table 3 in addition to the life
stages and the origin of the specimens. In our dataset, standard
length varies from 17.6 mm (Diogenichthys laternatus) to 126.4 mm
(Diaphus danae); eye and lens diameter range from 1.5 mm to
11.1 mm and from 0.6 mm to 5.0 mm in Nannobrachium idostigma
and Myctophum lychnobium, respectively.
Figure 2. Difference in eye size compared to body size in two
Estimating phylogenetic signal species of lanternfish. (A) Myctophum brachygnathum,(B) Nanno-
Estimation of the phylogenetic signal using Pagel’s lambda gives brachium phyllisae. Scale bar, 10 mm.
relatively similar results with both phylogenies (Table 4). Results doi:10.1371/journal.pone.0058519.g002

Table 3. Range of standard length, eye diameter and lens Table 3. Cont.
diameter for each of the 61 species of Myctophidae analysed
in this study.
L. gemellari 4 32.3–52.9 2.2–3.4 0.8–1.3 A+J 3

Loweina 1 25.9 2.0 0.9 J 9
Benthosema 1 42 4.6 1.8 A 8 interrupta
glaciale
Myctophum 1 76.4 8.5 3.8 A 1
B. suborbitale 2 28.2–29.0 3.0–3.1 1.3 A 9 asperum
Bolinichthys 5 36.8–45.4 3.3–4.4 1.5–1.8 A 3 M. 1 57.78 5.3 2.3 J 4
longipes aurolaternatum
B. nikolayi 4 23.3–37.7 2.5–4.1 1.0–1.8 A+J 3 M. 7 65.7–69.7 6.8–7.6 1.7–3.6 A+J 2, 5
brachygnathum
B. supralateralis 1 41.2 4.2 2.0 J 2
M. lychnobium 4 51.0–106.9 4.8–11.1 2.3–5.0 A+J 2, 4
Centrobranchus 1 31.5 2.2 0.8 ? 1
andreae M. nitidulum 11 25.5–85.4 2.3–7.1 0.9–3.2 A+J 6
Ceratoscopelus 1 52 5.1 2.1 A 8 M. obtusirostre 3 90.9–97.6 10.1–10.5 4.2–4.7 A 2, 4, 5
maderensis
M. spinosum 5 32.7–87.5 3.2–8.7 1.4–3.9 A+J 1, 2, 4
C. warmingii 11 36.3–71.9 3.5–6.6 1.4–2.8 A+J 1, 3
Nannobrachium 7 36.3–90.0 1.8–4 0.8–2.1 A 1, 3
Diaphus 2 33.9–35.2 4.1–4.3 1.6–1.8 A 3 cf. nigrum
brachycephalus2
N. idostigma 8 31.1–72.2 1.5–3.9 0.6–2.0 A+? 6
D. danae1 16 81.2–126.4 7.2–9.0 2.9–3.8 A 3
N. phyllisae 3 48.9–72.5 2.3–3.4 0.8–1.4 A+? 6
D. fulgens2 1 41.2 3.9 1.8 ? 2
Notolychnus 1 21.4 1.4 0.6 A 3
D. garmani1 3 32.8–42.1 2.4–3.5 1.0–1.5 A+J 4, 9 valdiviae
D. holti2 1 40 5.1 2.1 A 8 Notoscopelus 1 42 3.2 1.2 A 8
elongatus
D. luetkeni1 5 35.2–40.8 2.3–2.9 2.4–3.1 A+J 1, 3
N. kroeyerii 4 95.2–105.1 6.1–6.7 2.7–2.9 A 7
D. meadi2 1 28.3 3.3 1.2 J 8
2 Symbolophorus 1 74.0 6.0 2.3 J 6
D. mollis 3 36.6–66.2 3.9–6.5 1.6–3.2 A 3, 9
cf. boops
2
D. parri 5 24.9–51.0 2.9–5.7 1.7–2.4 A+J 1, 2, 3, 4
S. evermanni 13 33.6–65.8 2.5–6.2 1.0–2.8 J 2, 4
D. phillipsi1 4 26.2–29.5 2.1–2.4 1.0 J 3
S. rufinus 8 30.5–73.7 2.1–6.2 1.2–2.9 J 2, 4
D. regani1 2 40.8–42.3 2.5–2.8 1.2–1.2 J 4
S. veranyi 1 85 6.7 2.9 A 8
D. splendidus1 4 26.7–48.1 1.5–2.8 0.6–1.3 A+J 3, 4
Triphoturus 1 35.4 2.0 0.9 A 3
D. termophilus1 6 48.3–77.3 4.2–6.5 1.9–3.2 A+J 1, 3, 4 nigrescens
D. whitleyi1 2 59.5–90.3 4.0–5.5 1.7–2.6 A+? 3 T. oculeus 8 30–58.3 1.7–4.0 0.6–1.6 ? 6
Diogenichthys 7 18.0–21.4 1.9–2.7 0.8–1.0 A 9
atlanticus For each species, the sample size (n), the life stage (A = adult, J = juvenile) and
the sample origin (Cruise) is given. For cruise number refer to Table 1. The
D. laternatus 11 17.6–31.1 2.0–3.2 0.7–1.3 A 6 superscript number for each Diaphus species indicates the group number made
Electrona risso 1 46 7.0 3.0 A 8 from the absence (1) or presence (2) of the So.
Gonichthys 3 40.6–49.4 2.7–3.5 1.3–1.6 A 6
tenuiculus
show that Dn/Vn and caudal luminous organs have a strong
Hygophum 1 45 6.2 2.3 A 8
benoiti phylogenetic signal, suggesting that they gradually accumulate
H. hygomii 2 22.2–57.3 3.0–7.5 1.1–3.1 A+J 9
changes over time in a random evolutionary process. On the
contrary, no phylogenetic signal is observed for the standard
H. proximum 4 25.8–38.2 3.1–4.8 1.3–2.1 A+J 3, 4, 5, 6
length and the day-depth distribution (phylogeny A only) variables.
Lampadena 1 104.3 8.4 3.7 A 3 The other variables (eye diameter, lens diameter, residuals eye
luminosa
diameter, luminous patches, sexual dimorphism in luminous tissue
L. urophaos 1 40 2.8 1.3 A 3 and night depth distribution) show intermediate values of Pagel’s
Lampanyctus 9 30.3–49.8 1.7–3.3 ?–1.3 A+J 3, 4, 9 lambda, which, although significantly different from 0 or 1, are
alatus
generally closer to 1 depending on the phylogeny used. However,
L. crocodilus 1 31 1.7 0.6 J 8 independent of the phylogeny used, the eye size corrected for body
L. iselinoides 1 34.3 2.0 0.8 J 6 size (residuals eye diameter), the luminous patches and the
L. nobilis 2 36.4–50.4 1.7–2.7 1.0 J 3 luminous tissue sexual dimorphism variables show a strong
L. omostigma 1 27.8 1.9 0.7 ? 6
phylogenetic signal very close to 1, again indicating that these
traits changed randomly over time during lanternfish evolution
L. parvicauda 9 28.4–106.0 1.7–6.4 0.9–2.8 A+J 6
(Table 4).
L. pusillus 1 37 2.1 0.8 A 8
L. vadulus 4 37.4–81.8 2.2–5.2 0.9–2.12 A+J 1, 3, 4 Relationship among morphometric traits
Lobianchia 2 27–33.3 1.9–2.2 1.0 A 8 Phylogenetic linear regression shows that lens diameter is
dolfleini
strongly, (positively) correlated with eye diameter (PGLS, n = 61,
R2 = 0.98, t-value = 53.48, P#0.001; Figure 3). Due to this close

Table 4. Estimates of the phylogenetic signal for each

variable using Pagel’s Lambda.
Variables l (Phylogeny A) l (Phylogeny B)
Eye diameter 0.94,0.001, ,0.001

0.750.02, ,0.001
Lens diameter 0.95,0.001, ,0.001

0.790.02, ,0.001
Standard length ,0.0011.0, ,0.001

,0.0011.0, ,0.001
,0.001, ,0.001 ,0.001, ,0.001

Residuals (eye/SL) 0.96 0.94
1.0
Dn/Vn organs 1,0.001, 1,0.001,1.0
,0.001, 1.0
Caudal luminous organs 1 1,0.001,1.0
,0.001, ,0.001
Luminous patches 0.99 0.99,0.001, ,0.001
,0.001, ,0.001 ,0.001, ,0.001

Luminous tissue sexual 0.93 0.97
dimorphism
Day depth ,0.0011.0, 0.04
0.790.113, ,0.001
Night depth 0.75,0.001, ,0.001

0.67,0.001, ,0.001
The results are presented for one of the ten randomly selected polytomy
resolved phylogenies for the two different phylogenies. A l value of 1 indicates
that the trait gradually accumulates changes over time in a Brownian motion
process. A l values of 0 indicates that no phylogenetic signal is present and
Figure 4. Relationship between eye diameter and standard
that traits have evolved in response to selective processes. The superscript
values are likelihood ratio tests different from 0 and 1. Sample size is 61 for all
length in 61 species of Myctophidae. Each point represents the
variables except day depth (58). mean for the species; individual details are in Table 2. Shapes represent
doi:10.1371/journal.pone.0058519.t004 the subfamilies, circles = Myctophinae, triangles = Lampanyctinae. Col-
ors represent the tribes of Paxton et al. [43], brown = Electronini,
red = Myctophini, blue = Lampanyctini, green = Diaphini, yellow = Go-
relationship between eye and lens diameter, we focused all nichthyini, pink = Gymnoscopelini, purple = Notolychnini. The fitted line
subsequent analyses solely on the eye diameter – standard length is the linear regression corrected for phylogeny (PGLS) using the
relationship. Phylogenetic linear regression reveals that the eye phylogeny of Paxton et al. [43]. The black square represents one
individual, Scopelengys tristis, from the sister family Neoscopelidae.
diameter is positively correlated with standard length (PGLS,
n = 61, R2 = 0.74, t-value = 13.05, P#0.001, Figure 4). However,
variations in eye size are observed between the different
myctophid species, both at the level of subfamilies and tribes with (Myctophinae and Lampanyctinae) in terms of eye size (PGLS,
representatives of the same subfamily or tribe having smaller or n = 61, standard length effect: tA = 12.95, pA#0.001, tB = 13.85,
larger eyes (Figure 4). Probably due to this great variability, no pB#0.001; subfamily effect: tA = 0.41, pA = 0.68, tB = 0.62,
significant difference was found between the two subfamilies pB = 0.54). The representative of the lanternfish sister family
Neoscopelidae, Scopelengis tristis, showed a relatively small eye
compared to all the Myctophidae analysed (Figure 4).
Morphometric comparisons among tribes and clades

Phylogenetic ANOVAs reveal differences at both tribal
(phylogeny A, n = 4, F = 12.4, P = 0.001, Figure 5A) and cladal
(phylogeny B, n = 6, F = 14.33, P = 0.001; Figure 5B) levels. At the
tribal level, post-hoc analyses revealed that the Myctophini possess
significantly larger eyes than the other tribes analysed statistically
and probably larger eyes than the tribes Notolychnini and
Gymnoscopelini, which were excluded from the analysis due to
insufficient sampling size (Figure 5A). Although also excluded from
the statistical analysis for the same reasons, the tribe Electronini
seems to possess the largest eyes. The tribes Diaphini and
Lampanyctini are significantly different from each other, with
the latter showing smaller eyes, but not significantly different from
the Gonichthyini. At the cladal level (Figure 5B), most of the clades
statistically analysed present similar eye sizes except clade E
(Lampanyctini), which has significantly smaller eyes than all the
other clades analysed. The three clades from the tribe Lampa-
nyctini (D, E, F) do not overlap and show very different eye sizes,
with clade D showing the largest eyes and clade E the smallest
eyes. To the contrary, within the tribe Myctophini (H, I, J), all the
Figure 3. Relationship between lens diameter and eye
diameter after correcting for phylogeny (PGLS). The fitted line
clades possess similar eye sizes. As at the tribal level, Clade G
is the linear regression corrected for phylogeny (PGLS) using the (Electronini) did not possess sufficient observations to be included
phylogeny of Paxton et al. [43]. in the statistical analysis, but appears to be the clade showing the
doi:10.1371/journal.pone.0058519.g003 largest eye size.

Figure 5. Residuals eye size corrected for body size of Myctophidae after correcting for phylogeny: (A) by tribes (phylogeny A, [43])
and (B) by clades (phylogeny B, [44]). Colours represent the tribes of Paxton et al. [43] for comparison, as in Fig. 4. Groups sharing the same
superscript letter are not significantly different from one another based on the post-hoc analyses. Groups with no superscript letters were not
included in the analysis due to the low samples size (n,3). #, $ indicates genera that were significantly different for one of the ten randomly resolved
polytomy phylogenies (P# = 0.04, P$ = 0.03).
Eye-size variation is also observed between and within genera eye sizes (i.e. Myctophum, Bolinichthys, Diaphus 2). Conversely, some
independently of the phylogeny used (n = 8, phylogeny A, genera belonging to the same tribe or clade show significantly
F = 39.15, P = 0.001, Figure 6; phylogeny B, F = 41.54, different eye sizes (i.e. Hygophum and Myctophum from the tribe
P = 0.001). The results are similar between the two phylogenies Myctophini and Symbolophorus and Gonichthys from clade I). Genera
except for two genera. When analysed with phylogeny A, Diaphus 2 from the tribe Gonichthyini (Loweina, Gonichthys, Centrobranchus)
is found to possess larger eyes than Bolinichthys (Post-hoc test, appear to have smaller eyes than all the genera present in the
P = 0.028). However, no significant difference was found between tribes Electronini and Myctophini and similar eye sizes to the
the two groups when analysed with phylogeny B (Post-hoc test, other tribes. A great variation in eye size is also observed within
P = 1). The genera Lampanyctus and Nannobrachium are not the same genus, i.e. Diaphus. This genus possesses great variability
significantly different from each other, but possess significantly in Dn/Vn luminous organs (Figure 7) and can be divided further
smaller eyes than the rest of the genera analysed. Despite into two groups based on the presence or absence of one light
belonging to different tribes/clades, several genera share similar organ, the So just below the eye. The two groups were found to be

Figure 6. Residuals eye size corrected for body size by genera of Myctophidae, corrected for phylogeny (phylogeny A, [43]). Colors
represent the tribes of Paxton et al. [43], as in Figure 4. The genus Diaphus was divided into two groups based on the presence/absence of the So.
Groups with the same superscript letters are not significantly different from one another, independent of the phylogeny used based on the post-hoc
analyses. Groups with no superscript letters were not included in the analyses due to the low samples size (n,3). # indicates genera that were not
significantly different when analysed with phylogeny B (P = 1) and for four of the ten randomly resolved polytomy phylogenies of phylogeny A
(P = 0.06–0.08).
significantly different in term of eye size, with Diaphus species scene in the mesopelagic zone will lead to a great diversity in eye
possessing an So (Diaphus 2) having larger eyes than Diaphus species size [5], [16], [4]. At depths where both bioluminescence and
without an So (Diaphus 1). downwelling sunlight are present, as in the mesopelagic zone,
adaptations of the eye toward the detection of one or the other will
Relationship between morphometric and ecological depend on the ecological task required. Adaptations to assess the
intensity of downwelling light are essential to aid a species to hold
traits a depth station during the day, to camouflage themselves by
Phylogenetically controlled multiple linear regression models do counter illumination, to vertically migrate, to set their circadian
not reveal any significant relationships or trends between eye rhythm and/or to detect the presence of other animals above
diameter, corrected for standard length, and any of the ecological them. Adaptations for viewing bioluminescent signals will be an
variables examined in this study (Table 5). This lack of any advantage in detecting other individuals (prey, predator, mate) in
significant relationships between eye diameter and ecological traits deeper zones, where bioluminescent cues predominate.
persists regardless of the phylogeny used in the analysis (Table 5). In addition to the ventral photophores, lanternfishes possess a
number of luminous tissues that are thought to have several
Discussion functions. Caudal luminous organs play a role in either escape
responses by producing a blinding flash [56] or in sexual
The aim of this study was to assess the variability in eye size
communication, where those organs are sexually dimorphic.
within a range of species of lanternfishes from different depths to Several hypotheses have been proposed for the function of the
assess the influences of both ecology and phylogeny. To our Dn/Vn organs in myctophids. They may be used as a head torch
knowledge, this is the first study to examine eye-size variability to search for prey [57], [56], to compare the intensity of their own
within the same family of vertebrates using representatives of more photophore emissions with the downwelling sunlight [58] and/or
than 50% of the recognised genera. for intraspecific communication in sexually dimorphic species.
The function(s) of the other luminous patches remains unclear.
Eye-size variation and ecology However, the occurrence of sexual dimorphism in some of those
With respect to assessing whether the variation in eye size patches suggests a role in sexual communication. Herring [33]
observed within the lanternfishes could be explained by ecological concluded that the variety and complexity of sexual dimorphism in
differences using phylogenetic comparative analyses, no significant luminous organs were most likely due to their role in sexual
relationship was found between relative eye size and any of our signalling. The fact that some lanternfishes possess additional
predictor variables. Our hypothesis that lanternfish species with a luminous patches and/or dimorphic luminous organs indicates
deeper distribution and/or less luminous tissues will have smaller than some species may rely more on bioluminescent signals than
eyes could not be validated with our dataset. This hypothesis was others. However, this hypothesis was not supported by our dataset
posed under the assumption than the gradual change in the visual in terms of eye size.

upper bathypelagic depths, while some Lampanyctus species with

small eyes are restricted to the mesopelagic zone, with L. omostigma
recorded at no deeper than 400 m (Supplementary Table S2, S3).
These results suggest that some species may be clearly disadvan-
taged in term of vision by their small eye size in the mesopelagic
zone (i.e. L. omostigma). After sampling very different eye types/
sizes in the abyss, Murray & Hjort [62] wrote ‘‘Nothing has
appeared more hopeless in biological oceanography than the
attempt to explain the connection between the development of the
eyes and the intensity of light at different depths in the ocean’’.
The conclusion of the present study seems to agree with this
statement. However, the same analyses using light intensities
rather than depth distribution might be more relevant here and
will need to be considered in any future analyses, in addition to a
more detailed analysis of the lanternfish visual system.
Even though the relative enlargement of the eye is an important
adaptation for vision in dim-light conditions and for viewing
bioluminescence, it is not the only visual adaptation found in the
mesopelagic zone. Deep-sea fishes possess several other visual
specialisations to enhance the sensitivity of the eye, especially at
the level of the retina, that will need to be considered. Moreover,
two types of light can be seen in the mesopelagic zone,
bioluminescence and downwelling light, which do not require
the same adaptations in terms of sensitivity for their detection. In
fact, while sensitivity to bioluminescence (point-like light sources) is
directly influenced by the size of the eye ([16], [10]), this is not the
case for downwelling light (extended light source), where sensitivity
is independent of eye size and is directly proportional to the size of
the visual pixel (photoreceptor diameter, [63]). As a result, small-
eyed species could potentially have a greater sensitivity to
downwelling light than larger-eyed species depending of the
diameter of their photoreceptors. This stresses the point that the
visual capabilities of a species cannot be solely assessed by the size
of the eye and that a number of other physical factors in addition
to the type of visual stimulus might need to be taken into
consideration in order to assess relationships with environmental
variables. The question that then remains is: Do small-eyed
lanternfishes really have a limited visual system or do they
Figure 7. Variation in the size and location of the Dn, Vn and So compensate for their small eye size with other visual specialisa-
luminous organs within the genus Diaphus. A = D. luetkeni, B = D. tions? Very few studies have examined the visual system of
brachycephalus, C = D. danae, D = D. mollis, E = D. phillipsi, F = D. parri,
G = D. termophilus, H = D. holti. A, C, E, G = Diaphus group 1 (So absent); myctophids. However, the group appears to have evolved eyes
B, D, F, H = Diaphus group 2 (So present). The yellow arrows indicate the designed to enhance sensitivity, i.e. with an aphakic gap
position of the So photophore. (Tarletonbeania crenularis, [58]), a pure-rod retina (Lampanyctus
doi:10.1371/journal.pone.0058519.g007 crocodilus [64]; Lampanyctodes sp [65]; Stenobrachius leucopsarus [66]),
a tapetum lucidum (Stenobrachius leucopsarus [66]), a high photore-
The visual capabilities of an eye are influenced by its size [59]. A ceptor density (Lampanyctus crocodilus [64], Lampanyctodes sp [65];
larger eye will provide an advantage in the mesopelagic zone as it Stenobrachius leucopsarus [66]), visual pigments tuned to view
will increase the chance of photon capture, since a large eye bioluminescence (58 species, [67]), and a rather unspecialised
normally has a greater pupillary aperture. However, the larger the retina with poor acuity (Lampanyctus macdonaldi, Myctophum
eye the more energetically costly it will be [60]. Smaller eyes are punctatum, [68], [11]). However, these data have been compiled
less energetic and can act as a ‘‘distance filter’’ by reducing the from very few species and most of the studies have only examined
visibility of a bioluminescent signal against a completely dark one or a few of these characteristics, which negated their inclusion
background (viewed within the bathypelagic zone and deeper, in our analysis. It is also possible that small-eyed species have
[61]). However, a small eye will be a disadvantage for species adapted to the mesopelagic zone in other ways by relying less on
adapted for survival higher up in the water column, where high vision and more on other sensory systems. This hypothesis seems
levels of background illumination are present and increased plausible considering the extremely high number of myctophid
sensitivity is required. The results from this study indicate that species present in the mesopelagic zone (,250) and the
lanternfishes show a great variability in eye size, independent of quantitative differences in the size of the optic tecta [9].
their depth distribution. While some species show the expected
pattern by being strictly mesopelagic with relatively large eyes (i.e. Eye-size variation and phylogeny
Myctophum nitidulum), or venturing down into the upper parts of the Results from this study showed great differences in eye sizes
bathypelagic zone during the day and possessing relatively small within the Myctophidae at all phylogenetic levels. At the
eyes (i.e. Lampanyctus crocodilus), others present an inverse pattern. subfamilial level both small and large eyes are present in the two
For example, Hygophum benoiti possesses large eyes and frequents different subfamilies, indicating that eye-size variations may have

Table 5. Regression model of eye diameter with different predictor variables when controlling for phylogeny (PGLS).
Phylogeny n l Predictor variables b t-value P

*,*
Phylogeny A 58 0.939 Standard length 0.925 11.927 ,0.001
Dn/Vn luminous organs 20.012 20.211 0.834
Day depth 0.008 0.408 0.685
Night depth 20.013 20.434 0.666
Phylogeny B 58 0.935*,* Standard length 0.922 12.123 ,0.001
Dn/Vn luminous organs 20.043 20.717 0.477
Day depth 0.008 0.419 0.677
Night depth 20.039 21.451 0.153
The results are presented for one of the ten randomly selected polytomy resolved phylogenies for the two different phylogenies. Standard length was added as a
covariate in the models. n = sample size, l = phylogenetic scaling parameter, the superscript * after the parameter l indicates whether the parameter was significantly
different from 0 (first position) and from 1 (second position) in the likelihood tests, b = partial regression slope. The significance levels are shown in bold.
evolved several times within the family. Paxton [31] discussed the of this taxon based on the So photophore. Future molecular,
evolution of the Myctophidae and its two subfamilies and morphological phylogenetic analyses will hopefully shed more light
considered the Neoscopelidae the more generalised of the two on this large complex group, which comprises some 75 species.
families and the Myctophinae closer to the ancestral state than the The subdivision of the tribe Myctophini and inclusion of the tribe
Lampanyctinae, while admitting the larval characters indicated Gonichthyini within the Myctophini by Poulsen et al. [44] seems
the Myctophinae were more specialised [45]. Poulsen et al. [44] less obvious from our results. Relative eye sizes between
presented the first molecular phylogeny for the family, but failed to Myctophini and Gonichthyini using Paxton et al.’s phylogeny
resolve this question, with the exception of Notolychnus, which was are significantly different and do not overlap, indicating that
used as the ancestral species of the two subfamilies. Notolychnus members of the Gonichthyini have systematically smaller eyes
valvidiae is a small eyed species, suggesting that small eyes represent than members of the Myctophini. We realise that eye size is only
the ancestral condition and that larger eyes might have evolved one character in the evolution of the group and our data set does
several times within the lanternfish family. not include all genera of either tribe.
The estimation of the phylogenetic signal by Pagel’s lambda
revealed that relative eye size within the family has a strong
phylogenetic signal independent of the phylogeny used. By Limits of the study
definition, members of a taxon are more closely related to one Sampling methods. In this study, an attempt was made to
another than to any members of another taxon. Based on this categorize species according to their diurnal and nocturnal depth
definition, and if relative eye size is a relatively conserved variable ranges. This task was complicated because of the lack of accurate
over time, as Pagel’s lambda seems to indicate, then eye size within depth-distribution data in the literature. Unfortunately, very few
a taxon (tribe or clade) would be more similar than between taxa studies accurately estimate the depth at which a specimen is
and as a result, eye size of a species might be estimated simply sampled, with the minority of sampling using opening-closing
based on the phylogenetic position of this species. In this sense, the devices. Moreover, even fewer studies report depth distribution by
sub-division of Paxton’s tribe Lampanyctini into three different life stages. Karnella [27] is to date the most comprehensive
clades by Poulsen et al. [44] appears to be supported by our lanternfish depth-distribution study, where fishes were sampled
relative eye-size data, since clades D, E and F present very seasonally from the Ocean Acre in the Northern Sargasso Sea at
different eye-size ranges that do not overlap. However, great day and night using predominantly discrete-depth sampling gear
variation in eye size is also observed within the single genus Diaphus every 50 m. Results from this study present accurate depth-
in our study. Kawaguchi and Shimizu [69] presented a taxonomic distribution data by life stages for 20 of our 61 species. In addition
key for Diaphus, which divided the genus into four groups (SuO- to the lack of accurate depth-distribution data in the literature,
group, So group, Ant-group and Dn-Vn group) based on the several other issues make the task of depth categorisation of species
presence or absence of different luminous organs associated with challenging. In addition to the high level of interspecific variability
the eye. The presence of the So below the eye is one of the first in depth pattern, there is observed intraspecific variability in depth
identifying characters used by several taxonomic keys to identify distribution depending on the season [27], moon cycle (i.e.
Diaphus species [32], (Paxton and Williams, unpublished data). Hygophum hygomii [70]), changes with size/age where larger/older
Further division of this genus based on the absence or presence of specimens become non-migrant (i.e. Benthosema glaciale [27]) and
the So in our study (Diaphus 1, Diaphus 2) shows that species ocean physics (i.e. Lampanyctus crocodilus [71]). Studies will often
possessing an So photophore have significantly larger eyes than record the depth of a species at a specific location, season and
species without this photophore, indicating a possible subdivision time, which might not reveal the general pattern for the species.

Finally, lanternfishes are able to efficiently avoid the net during the Supporting Information
day [72], increasing the chances of biased data for those species.
As a consequence, our depth categorisation may not represent the Table S1 Analyses of covariance (ANCOVA) of the eye
true distribution of several species, especially in cases where diameter versus standard length between juveniles and
information in the literature was from a different ocean than the adults in Myctophidae. n = sample size, b = partial regression
species analysed in our study. slope. Only species with at least three observations for each stage
(juvenile, adult) were analysed. Significant differences are shown in
bold. The slopes are not significantly different between juveniles
Phylogeny
and adults.
Phylogenetic comparative analyses are dependent on the
(DOC)
phylogeny used. To allow the most accurate analyses, a fully
resolved phylogeny with branch lengths is usually required. Table S2 Summary of the juvenile-adults night and day
Unfortunately, the phylogeny of the lanternfish family remains depth ranges found in the literature for the 61 species of
poorly resolved, even though they represent one of the most Myctophidae studied. The sampling area and depth range of
important groups of mesopelagic fishes in the ocean in term of our study samples is also given (N = night, D = day). Abbreviations
biomass and energy transfer [73]. Nevertheless, the first molecular for the areas can be found in Table S3. * Study using an opening-
phylogeny for the family presented by Poulsen et al. [44] appears closing device.
to support the basic morphological phylogeny of Paxton et al. [43], (DOC)
even though some differences in tribal arrangement are recog- Table S3 List of abbreviations used for the areas
nised. A better resolved phylogeny will undoubtedly improve our sampled in Table S2.
understanding of the relationship between the different tribes/ (DOC)
clades and the variation in eye size, especially as more ecological
data is accumulated (i.e. diet: luminescent prey vs non-lumines-
cent) for this group.
Acknowledgments
We wish to thank Dr. Mike Hall (AIMS) and the Masters and crews of the
Conclusion RV Cape Ferguson and, Prof. Hans-Joachim Wagner (University of
A great variability in relative eye size within the Myctophidae Tübingen) and the Masters and crews of the FS Sonne for sea time
opportunities. We thank Prof. Lynnath Beckley (Murdoch University), Dr.
was observed at all taxonomic levels. However, variability in eye Pilar Olivar (CSIC), Dr. Anna Bozzano (CSIC) and Dr. Brigitte
size within the family could not be explained by ecological Guillaumont (Ifremer), for providing additional samples and, Caroline
variables (bioluminescence and depth patterns) in this study and Kerr (UWA) and Alan Goldizen (UQ) for their precious help during field
seems instead to be driven by phylogenetic parameters. Further trips. Finally, we wish to thank Dr Kara Yopak (UWA) for her help and for
analyses including other environmental variables (i.e. diet, prey, initially suggesting the use of phylogenetic comparative analyses for this
light intensities) and a more complete phylogeny are needed to study.
understand the great variability in eye size within myctophids.
Moreover, examination of the visual system in more depth will be Author Contributions
an essential step in assessing the visual capabilities of each species Conceived and designed the experiments: FdB SPC NJM. Performed the
and to shed light on the visual adaptations of the lanternfish family experiments: FdB JRP. Analyzed the data: FdB JLF. Wrote the paper: FdB
in relation to their environment. JLF JRP SPC.
References
1. Land MF (1990) Optics of the eyes of marine animals. In: Herring PJ, Campbell 13. Collin S, Hoskins R, Partridge J (1997) Tubular eyes of deep-sea fishes: a
AK, Whitfield M, Maddock L, editors. Light and life in the sea: Cambridge comparative study of retinal topography (Part 2 of 2). Brain, Behavior and
University Press. pp. 149–166. Evolution 50: 347–357.
2. Matthiessen L (1882) Über die Beziehungen, welche zwischen dem Brechung- 14. Collin SP, Hoskins RV, Partridge JC (1998) Seven retinal specializations in the
sindex des Kernzentrums der Krystalllinse und den Dimensionen des Auges tubular eye of the deep-sea pearleye, Scopelarchus michaelsarsi: a case study in visual
bestehen. Pflügers Archiv für die gesamate Physiologie des Menschen und der optimization. Brain Behavior and Evolution 51: 291–314.
Tiere 27: 510–523. 15. Marshall NB (1954) Aspects of deep sea biology. London: Hutchinson 380 p.
3. Matthiessen L (1886) Über den physikalisch-optischen Bau des Auges der 16. Warrant E (2000) The eyes of deep-sea fishes and the changing nature of visual
Cetacean und der Fische. Pflügers Archiv für die gesamate Physiologie des scenes with depth. Philosophical Transactions of the Royal Society of London B
Menschen und der Tiere 38: 521–528. 355: 1155–1159.
4. Warrant E (2004) Vision in the dimmest habitats on Earth. Journal of 17. Munk O (1965) Ocular degeneration in deep-sea fishes. Galathea-Rep 8: 21–31.
Comparative Physiology A 190: 765–789. 18. Hulley PA (1981) Results of the research cruises of the FRV ‘‘Walther Herwig’’
5. Denton EJ (1990) Light and vision at depths greater than 200 metres. In: to south America, LVII: family Myctophidae (Osteichthyes, Myctophiformes).
Herring PJ, Campbell AK, Whitfield M, Maddock L, editors. Light and life in Archiv Fur Fischereiwissenschaft 31: 1–300.
the sea: Cambridge University Press. pp. 127–148. 19. Hulley PA, Paxton JR (In press) Myctophidae, lanternfishes. In: Carpenter KE,
6. Locket NA (1977) Adaptations to the deep-sea environment. In: Crescitelli F, editor. The living marine resources of the eastern central Atlantic. Rome: FAO.
editor. Handbook of sensory physiology. Berlin: Springer-Verlag. pp. 67–192. pp. In press.
7. Hansen K, Herring PJ (1977) Dual bioluminescent systems in the anglerfish 20. Watanabe H, Moku M, Kawaguchi K, Ishimaru K, Ohno A (1999) Diel vertical
genus Linophryne (Pisces: Ceratioidea). Journal of Zoology 182: 103–124. migration of myctophid fishes (Family Myctophidae) in the transitional waters of
8. Mallefet J, Shimomura O (1995) Presence of coelenterazine in mesopelagic fishes the western North Pacific. Fisheries Oceanography 8: 115–127.
from the Strait of Messina. Marine Biology 124: 381–385. 21. Tsuji FI, Haneda Y (1971) Luminescent system in a myctophid fish, Diaphus
9. Wagner HJ (2001) Sensory brain areas in mesopelagic fishes. Brain, Behavior elucens Brauer. Nature 233: 623–624.
and Evolution 57: 117–133. 22. Haygood MG, Edwards DB, Mowlds G, Rosenblatt RH (1994) Bioluminescence
10. Warrant EJ, Locket NA (2004) Vision in the deep sea. Biological Reviews 79: of myctophid and stomiiform fishes is not due to bacterial luciferase. Journal of
671–712. Experimental Zoology 270: 225–231.
11. Wagner HJ, Fröhlich E, Negishi K, Collin SP (1998) The eyes of deep-sea fish II. 23. Case JF, Warner J, Barnes AT, Lowenstine M (1977) Bioluminescence of lantern
Functional morphology of the retina. Progress in Retinal and Eye Research 17: fish (Myctophidae) in response to changes in light intensity. Nature 265: 179–
637–685. 181.
12. Munk O (1966) Ocular anatomy of some deep-sea teleosts. Dana Report 70: 24. Beebe W, Vander Pyl M (1944) Eastern Pacific expeditions of the New York
5–62. Zoological Society. XXXIII. Pacific Myctophidae (fishes). Zoologica 29: 59–95.

25. Edwards AS, Herring PJ (1977) Observations on the comparative morphology 48. Grafen A (1989) The phylogenetic regression. Philosophical Transactions of the
and operation of the photogenic tissues of myctophid fishes. Marine Biology 41: Royal Society of London B 326: 119–157.
59–70. 49. Harmon LJ, Weir JT, Brock CD, Glor RE, Challenger W (2008) GEIGER:
26. Kristoffersen JB, Salvanes AGV (1998) Effects of formaldehyde and ethanol investigating evolutionary radiations. Bioinformatics 24: 129–131.
preservation on body and otoliths of Maurolicus muelleri and Benthosema glaciale. 50. Pagel M (1999) Inferring the historical patterns of biological evolution. Nature
Sarsia 83: 95–102. 401: 877–884.
27. Karnella C (1987) Family Myctophidae, lanternfishes. In: Gibbs RH, Krueger 51. Freckleton RP, Harvey PH, Pagel M (2002) Phylogenetic Analysis and
WH, editors. Biology of midwater fishes of the bermuda Ocean Acre. Comparative Data: A Test and Review of Evidence. The American Naturalist
Washington, D.C.: Smithsonian Institution Press. pp. 51–168. 160: 712–726.
28. Zahuranec BJ (2000) Zoogeography and systematics of the lanternfishes of the 52. Paradis E, Claude J, Strimmer K (2004) APE: Analyses of phylogenetics and
genus Nannobrachium (Myctophidae: Lampanyctini). Smithsonian Contributions evolution in R language. Bioinformatics 20: 289–290.
to Zoology 607: 1–69. 53. Freckleton RP (2009) The seven deadly sins of comparative analysis. Journal of
29. Nafpaktitis BG, Backus RH, Craddock JE, Haedrich RL, Robison BH, et al. Evolutionary Biology 22: 1367–1375.
(1977) Family Myctophidae. In: Gibbs Jr RH, Berry FH, Böhlke JE, Cohen DM, 54. Garland T, Dickerman AW, Janis CM, Jones JA (1993) Phylogenetic analysis of
Collette BB et al., editors. Fishes of the Western North Atlantic: Memoir, Sears covariance by computer simulation. Systematic Biology 42: 265–292.
Foundation for Marine Research. pp. 13–265. 55. Revell LJ (2012) phytools: an R package for phylogenetic comparative biology
30. Nafpaktitis BG, Nafpaktitis M (1969) Lanternfishes (family Myctophidae) (and other things). Methods in Ecology and Evolution 3: 217–223.
collectedduring cruises 3 and 6 of the R/V ANTON BRUUN in the Indian 56. Haddock SHD, Moline MA, Case JF (2010) Bioluminescence in the Sea. Annual
Ocean: Science Bulletin, Los Angeles County Natural History Museum 5: 1–79 Review of Marine Science 2: 443–493.
p. 57. Herring PJ (1985) How to survive in the dark: bioluminescence in the deep sea.
31. Paxton JR (1972) Oeteology and relationships of the lanternfishes (family Symp Soc Exp Biol 39: 323–350.
Myctophidae). Science Bulletin, Los Angeles County Natural History Museum 58. Lawry JV (1974) Lantern fish compare downwelling light and bioluminescence.
13: 1–81. Nature 247: 155–157.
32. Nafpaktitis BG (1978) Systematics and distribution of lanternfishes of the genera 59. Walls GL (1942) The vertebrate eye and its adaptive radiation. Bloomfield Hills:
Lobianchia and Diaphus (Myctophidae) in the Indian Ocean. Science Bulletin, Los The Cranbrook Institute of Science.
Angeles County Natural History Museum 30: 1–92. 60. Laughlin SB, van Steveninck RRR, Anderson JC (1998) The metabolic cost of
33. Herring PJ (2007) REVIEW. Sex with the lights on? A review of bioluminescent neural information. Nature neuroscience 1: 36–41.
sexual dimorphism in the sea. Journal of the Marine Biological Association of the 61. Warrant EJ, Collin SP, Locket NA (2003) Eye disign and vision in deep-sea
UK 87: 829–842. fishes. In: Collin SP, Marshall NJ, editors. Sensory processing in aquatic
34. Kinzer J, Schulz K (1985) Vertical distribution and feeding patterns of midwater environments. New York: Springer-Verlag. pp. 303–322.
fish in the central equatorial Atlantic. I. Myctophidae. Marine Biology 85: 313– 62. Murray J, Hjort J (1902) The Depths of the Ocean. London: Macmillan.
322. 63. Land MF (1981) Optics and vision in invertebrates. In: Autrum H, editor.
35. Gartner JV, Hopkins TL, Baird RC, Milliken DM (1987) The lanternfishes Handbook of Sensory Physiology, Vol VII/6B. Berlin, Heidelberg, New York:
(Pisces: Myctophidae) of the eastern Gulf of Mexico. Fishery Bulletin 85: 81–98. Springer.
36. Robison BH (1972) Distribution of the midwater fishes of the Gulf of California. 64. Vilter V (1951) Bases cyto-architectoniques de l’acuité visuelle chez un poisson
Copeia 1972: 448–461. abyssal, le Lampanyctus crocodilus. Comptes Rendus des Séances de la Sociète de
37. Williams A, Koslow JA (1997) Species composition, biomass and vertical Biologie et de ses Filiales 145: 52–54.
distribution of micronekton over the mid-slope region off southern Tasmania, 65. Pankhurst NW (1987) Intra- and interspecific changes in retinal morphology
Australia. Marine Biology 130: 259–276. among mesopelagic and demersal teleost from the slope waters of New-Zealand.
38. Ross SW, Quattrini AM, Roa-Varon AY, McClain JP (2010) Species Environmental Biology of Fishes 19: 269–280.
composition and distributions of mesopelagic fishes over the slope of the 66. O’Day WT, Fernandez HR (1976) Vision in lanternfish Stenobrachius leucopsarus
north-central Gulf of Mexico. Deep-Sea Research Part II 57: 1926–1956. (Myctophidae). Marine Biology 37: 187–195.
39. Flynn A, Kloser R (2012) Cross-basin heterogeneity in lanternfish (family 67. Turner JR, White EM, Collins MA, Partridge JC, Douglas RH (2009) Vision in
Myctophidae) assemblages and isotopic niches (d13C and d15N) in the southern lanternfish (Myctophidae): Adaptations for viewing bioluminescence in the deep-
Tasman Sea abyssal basin. Deep Sea Research Part I: Oceanographic Research sea. Deep-Sea Research Part I 56: 1003–1017.
Papers. 68. Collin SP, Partridge JC (1996) Retinal specializations in the eyes of deep-sea
40. Ahlstrom EH, Stevens E (1976) Report of neuston (surface) collections made on teleosts. Journal of Fish Biology 49: 157–174.
an extended CalCOFI cruise during May 1972. Calif Coop Oceanic Fish Invest 69. Kawaguchi K, Shimizu H (1978) Taxonomy and distribution of the lantern-
Rep 18: 167–180. fishes, genus Diaphus (Pisces, Myctophidae) in the western Pacific, eastern Indian
41. Harvey PH, Pagel MD (1991) The comparative method in evolutionary biology. Oceans and the southeast Asian Seas. Bulletin of the Ocean Research Institute,
Oxford: Oxford University Press. University of Tokyo 10: 1–145.
42. Madison WP, Madison DR (2011) Mesquite: a modular system for evolutionary 70. Linkowski TB (1996) Lunar rhythms of vertical migrations coded in otolith
analysis. 2.75 ed. microstructure of North Atlantic lanternfishes, genus Hygophum (Myctophidae).
43. Paxton JR, Ahlstrom EH, Moser HG (1984) Myctophidae: relationships. In: Marine Biology 124: 495–508.
Moser HG, Richards WJ, Cohen DM, Fahay MP, Kendall AW et al., editors. 71. Stefanescu C, Cartes J (1992) Benthopelagic habits of adult specimens of
Ontogeny and Systematic of Fishes: American Society of Ichthyologist and Lampanyctus crocodilus (Risso, 1810) (Osteichthyes, Myctophidae) in the western
Herpetologist Special Publication. pp. 239–244. Mediterranean deep slope. Scientia Marina 56: 69–74.
44. Poulsen JY, Byrkjedal I, Willassen E, Rees D, Takeshima H, et al. (In press) 72. Kaartvedt S, Staby A, Aksnes DL (2012) Efficient trawl avoidance by
Mitogenomic sequenses and evidence from unique gene rearrangements mesopelagic fishes causes large underestimation of their biomass. Marine
cooroborate evolutionary relationships of Myctophiformes (Neoteleostei). BMC Ecology Progress Series 456: 1–6.
Evolutionary Biology: In press. 73. Cherel Y, Fontaine C, Richard P, Labat JP (2010) Isotopic niches and trophic
45. Moser HG, Ahlstrom EH (1970) Development of lanternfishes (family levels of myctophid fishes and their predators in the Southern Ocean. Limnology
myctophidae) in the California Current. Part I. Species with narrow-eyed and Oceanography 55: 324–332.
larvae. Science Bulletin, Los Angeles County Natural History Museum 7: 1–145. 74. Olivar MP, Bernal A, Moli B, Pena M, Balbin R, et al. (2012) Vertical
46. Moser HG, Ahlstrom EH (1972) Development of lanternfish, Scopelopsis distribution, diversity and assemblages of mesopelagic fishes in the western
multipunctatus Brauer 1906, with a discussion of its phylogenetic position in the Mediterranean. Deep Sea Research Part I 62: 53–69.
family Myctophidae and its role in a proposed mechanism for the evolution of 75. Mahé JC, Poulard JC (2005) Manuel des protocoles de campagne halieutique.
photophore patterns in lanternfishes. Fishery Bulletin 70: 541–564. Campagnes EVHOE (EValuation des ressources Halieutiques de l’Ouest
47. Moser HG, Ahlstrom EH (1974) Role of larval stages in systematic investigations Europe). Available: http://archimer.ifremer.fr/doc/00036/14707/. Accessed
of marine teleosts: the Myctophidae, a case study. Fishery Bulletin 72: 391–413. 2013 Feb 8.

Table S1. Analyses of covariance (ANCOVA) of the eye diameter versus standard length
between juveniles and adults in Myctophidae. n = sample size, b = partial regression slope.
Only species with at least three observations for each stage (juvenile, adult) were analysed.
Significant differences are shown in bold. The slopes are not significantly different between
juveniles and adults.
Species n juveniles n adults Predictors β T-value P

Ceratoscopelus warmingii 3 8 SL 0.92 6.93 <0.001
stage -0.03 -1.30 0.23
Lampanyctus alatus 5 4 SL 1.47 4.34 0.005

stage 0.05 1.00 0.35
Lampanyctus parvicauda 3 6 SL 0.91 9.53 <0.001

stage -0.05 -1.21 0.27
Table S3. List of abbreviations used for the areas sampled in Table S2.
Abbreviation Translation
CorS Coral Sea
PCT Peru-Chile Trench
Med Mediterranean Sea
wMed Western Mediterranean Sea
Aus Australia
wAus Western Australia
eAus Eastern Australia
Tas Tasman Sea
ChiS China Sea
Pac Pacific
cPac Central pacific
ePac Eastern Pacific
cNPac Central North Pacific
eNPac Eastern North Pacific
eSPac Eastern South Pacific
eAtl Eastern Atlantic
wNAtl Eastern North Atlantic
eNAtl Western North Atlantic
cNAtl Central North Atlantic
Phi Philippines
GMex Gulf of Mexico
GCal Gulf of California
SAfr South Africa
Table S2. Summary of the juvenile-adults night and day depth ranges found in the literature for the 61 species of Myctophidae studied. The
sampling area and depth range of our study samples is also given (N = night, D= day). Abbreviations for the areas can be found in Table S3. * Study
using an opening-closing device.
Benthosema glaciale 100-200 N wMed 750-1250 / 12-800 750-1050 / 375-800 wNAtl / Med [1]*/ [2]
B. suborbitale 0-200 N wAus 0-100 600-700 wNAtl [1]*
Bolinichthys longipes 87-107 N CorS 50-150 525-725 cNPac [3]
B. nikolayi 202-429 N CorS <300 <500 cEPac [4]
B. supralateralis 500 N CorS 200-700 / 60-700 600-650 / 250-600 wNAtl / GMex [1]*/ [5]*
Centrobranchus andreae 100 N CorS 100-165 640-650 cNPac [6]
Ceratoscopelus maderensis 65 N wMed 33-1000 / 12/800 751-1000 / 100-1000 wNAtl / Med [1]*/ [2]
C. warmingii 83-300 N CorS 0-500 800-1550 wNAtl [1]*
Diaphus brachycephalus 202-329 N CorS 150-225 / 50-300 250-450 / 300-600 wNAtl / GMex [1] */ [5]*
D. danae 253-377 N CorS 0-300 300-650 Tas [7]*
D. fulgens 500 N CorS 85 1000 ChiS [8]
D. garmani 100-150 N CorS / wAus surface / 0-125 <225 / 325-750 ePac / SAfr [9] / [10]
D. holti - wMed 80-235 500-675 Med [2]
D. luetkeni 202-325 N CorS 400-800 / 60-300 500-800 / 300-600 wNAtl / GMex [1]* / [5]*
D. meadi 50-100 N wMed 0 1250 Tas [11]*
D. mollis 202-429 N CorS / wAus 0-700 / 33-350 100-800 / 300-500 wNAtl [1] */ [2]
D. parri 200-329 N CorS 0 900 Aus [12]
D. phillipsi 107 N CorS 50-200 600 SAfr [10]
D. regani 50-400 N CorS 0 / <50 1000 Aus [12]/ [13]
D. splendidus 107 N CorS 51-250 / 30-550 501-650 / 300-600 wNAtl / GMex [1] / [5]
D. termophilus 200-477 N CorS 40-225 / 75-150 325-850 eNAtl / GMex [2] / [5]*
D. whitleyi 329 N CorS 136-152 170-710 Phi [14]
Diogenichthys atlanticus 50-150 N wAus 20-1000 / 50-700 500-1000 / 350-700 wNAtl / GMex [1]* / [5]*
D. laternatus 36-146 N PCT Surface / >50 100-500 / 200-400 eNPac / eSPac [15]/ [16]
Electrona risso 330-1000 D wMed 400-500 / 0-200 700-750 / 600-700 Med / cAtl [17] / [18]*
Gonichthys tenuiculus Surface N CPT surface - cPac [15]
Hygophum benoiti 410 D wMed 18-1050 450-1100 wNAtl [1]*
H. hygomii 0-202 N wAus 10-300 / 0-1000 400-800 / 500-1000 eNAtl / wNAtl [2]/ [1]*
H. proximum Surface N CorS / PCT 25-150 500-700 cNPac [3]
Lampadena luminosa 325 N CorS 75-250 / 65-600 525-725 / 500-1000 cNPac / GMex [3] / [5]*
L. urophaos 200 N CorS 50-600 600-750 wNAtl [1]*
Lampanyctus alatus 202-286 N CorS / wAus 50-300 700-850 wNAtl [1]*
L. crocodilus 600 D wMed 50-950 / 1200 750-1000 / 1200 wNAtl / Med [1]*/ [19]
L. iselinoides 450-707 D PCT <70 - ePac [9]
L. nobilis 107 N CorS 40-600 / 100-200 800-1000 / 300-<900 GMex / wNAtl [5] / [1]
L. omostigma 36-146 N PCT <50 200-400 eSPac [16]
L. parvicauda 61-302 N / 610-1200 D PCT 0-150 - GCal [20]*
L. pusillus 100-200 N wMed 50-325 / 50-1000 500-1000 / 550-850 Med / wNAtl [2] / [1]*
L. vadulus 100 N CorS 0 1000 Aus [12]
Lobianchia dolfleini 65 N wMed 25-400 375-600 Med [2]
L. gemellari 107-329 N CorS 20-210 / 50-600 300-450 / 400-850 GMex / wNAtl [5]* / [1]*
Loweina interrupta 100-150 N wAus 60-800 - wNAtl [2]
Myctophum asperum Surface N CorS Surface-150 / 0-125 400 / 425-750 GMex / SAfr [5]* / [10]
M. aurolaternatum Surface N CorS 0 1000 eAus [12]
M. brachygnathum Surface N CorS - 280-340 Phi [14]
M. lychnobium Surface N CorS 0 100 eAus [12]
M. nitidulum Surface N PCT 0-15 / surface 600-800 cNPac / eNpac [3] / [15]
M. obtusirostre Surface N CorS Surface-150 / 0-15 500-600 / 500-700 GMex / cNPac [5]* / [3]
M. spinosum Surface N CorS 0-15 600 cNPac [3]
Nannobrachium cf. nigrum 200-300 N CorS 100-310 640-900 cNPac [3]
N. idostigma <61 N PCT >50 250 eSPac [16]
N. phyllisae <108 N PCT <300 - ePac [9]
Notolychnus valdiviae 211-348 N CorS 50-250 / 30-1050 350-1050 / 400-850 GMex / wNAtl [21]* / [1]*
Notoscopelus elongatus 248 N, 414 D wMed 45-150 375-1000 Med [2]
N. kroeyerii 366-500 D eNAtl 0-200 325->1000 eNAtl [2]
Symbolophorus cf. boops Surface N CorS 0 900 eAtl [22]
S. evermanni Surface N CorS 0-125 600-900 cNPac [3]
S. rufinus Surface N CorS 0-125 / 0-900 425-850 / 750-900 SAfr / wNAtl [10] / [1]*
S. veranyi Surface N wMed 0-150 100-700 Med [2]
Triphoturus nigrescens 321 N PCT 200-1000 400-1000 cNPac [23]
T. oculeus 5-290 N <450 D CorS >50 >100 eSPac [16]
References
1. Karnella C (1987) Family Myctophidae, lanternfishes. In: Gibbs RH, Krueger WH, editors. Biology of midwater fishes of the bermuda Ocean Acre.
Washington, D.C.: Smithsonian Institution Press. pp. 51-168.
2. Hulley PA (1984) Myctophidae. In: Whitehead PJP, Bauchot M-L, Hureau J-C, Nielsen J, Tortonese E, editors. Fishes of the North-eastern Atlantic
and the Mediterranean: Unesco. pp. 510.
3. Clarke TA (1973) Some aspects of the ecology of lanternfishes (Myctophidae) in the Pacific Ocean near Hawaii. Fishery Bulletin 71: 401-434.
4. Hulley PA, Duhamel G (2009) A review of the lanternfish genus Bolinichthys Paxton, 1972 (Myctophidae). Cybium 33: 259-304.
5. Gartner JV, Hopkins TL, Baird RC, Milliken DM (1987) The lanternfishes (Pisces: Myctophidae) of the eastern Gulf of Mexico. Fishery Bulletin 85:
81-98.
6. Backus RH, Craddock JE, Haedrich RL, Shores DL, Teal JM, et al. (1968) Ceratoscopelus maderensis: Peculiar sound-scattering layer identified
with this myctophid fish. Science 160: 991-993.
7. Williams A, Koslow JA (1997) Species composition, biomass and vertical distribution of micronekton over the mid-slope region off southern
Tasmania, Australia. Marine Biology 130: 259-276.
8. Yang J, Huang Z, Chen S, Lee C (1996) The deep-water pelagic fishes in the area from Nansha Islands to the northeast part of South China Sea.
Beijing: Science Press. pp. 58-141.
9. Wisner RL (1976) The taxonomy and distribution of lanternfishes (family Myctophidae) of the eastern Pacific Ocean: Navy Ocean Research and
Development Activity. NORDA Rep. 3. 229 p.
10. Hulley PA (1986) Order Myctophiformes. Myctophidae. In: Smith MM, Heemstra PC, editors. Smiths' Sea Fishes. Heidelberg: Springer-Verlag.
pp. 282-322.
11. Flynn A, Kloser R (2012) Cross-basin heterogeneity in lanternfish (family Myctophidae) assemblages and isotopic niches (δ13C and δ15N) in the
southern Tasman Sea abyssal basin. Deep Sea Research Part I 69: 113-127.
12. Flynn AJ (2012) Ecology and Zoogeography of Lanternfishes (family Myctophidae). PhD thesis.: The University of Queensland.
13. Hartmann AR, Clarke A (1975) The distribution of myctophid fishes across the central equatorial Pacific. Fishery Bulletin 73.
14. Bourret P (1985) Poissoins teleosteens: Gonostomatidae, Sternoptychidae, et Myctophidae (MUSORSTOM II). Memoires du Museum National
d'Histoire Naturelle 133: 55-82.
15. Ahlstrom EH, Stevens E (1976) Report of neuston (surface) collections made on an extended CalCOFI cruise during May 1972. Calif Coop
Oceanic Fish Invest Rep 18: 167-180.
16. Cornejo R, Koppelmann R (2006) Distribution patterns of mesopelagic fishes with special reference to Vinciguerria lucetia Garman 1899
(Phosichthyidae: Pisces) in the Humboldt Current Region off Peru. Marine Biology 149: 1519-1537.
17. Nafpaktitis BG, Backus RH, Craddock JE, Haedrich RL, Robison BH, et al. (1977) Family Myctophidae. In: Gibbs Jr RH, Berry FH, Böhlke JE,
Cohen DM, Collette BB et al., editors. Fishes of the Western North Atlantic: Memoir, Sears Foundation for Marine Research. pp. 13-265.
18. Kinzer J, Schulz K (1985) Vertical distribution and feeding patterns of midwater fish in the central equatorial Atlantic. I. Myctophidae. Marine
Biology 85: 313-322.
19. Stefanescu C, Cartes J (1992) Benthopelagic habits of adult specimens of Lampanyctus crocodilus (Risso, 1810) (Osteichthyes, Myctophidae) in
the western Mediterranean deep slope. Scientia Marina 56: 69-74.
20. Robison BH (1972) Distribution of the midwater fishes of the Gulf of California. Copeia 1972: 448-461.
21. Ross SW, Quattrini AM, Roa-Varon AY, McClain JP (2010) Species composition and distributions of mesopelagic fishes over the slope of the
north-central Gulf of Mexico. Deep-Sea Research Part II 57: 1926-1956.
22. Hulley PA (1992) Upper-slope distributions of oceanic lanternfishes (family: Myctophidae). Marine Biology 114: 365-383.
23. Hulley PA (1986) A taxonomic review of the lanternfish genus Triphoturus Fraser-Brunner, 1949 (Myctophidae, Osteichthyes). Annals of the
South African Museum 97: 71-9

DeBusserolles Fanny 2013

Uploaded by

Copyright:

Available Formats

DeBusserolles Fanny 2013

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

DeBusserolles Fanny 2013

Uploaded by

Copyright:

Available Formats

Myctophid vision: seeing

and being seen in the

Fanny de Busserolles, BSc, MSc

This thesis is presented for the degree of Doctor of Philosophy of the

School of Animal Biology

Light is necessary, even to see the shadow.

A new short-wavelength intra-ocular filter, a yellow pigmentation, present in the outer

These findings show a surprising interspecific variability in the visual system of

I am grateful to The University of Western Australia for financially supporting my

Thesis format and authorship

In accordance with the University of Western Australia’s regulations regarding

Published in PLoS ONE as:

The overall estimated contribution of the candidate is 85%.

Accepted in the Journal of Comparative neurology as:

de Busserolles F, Marshall NJ, Collin SP. The eyes of lanternfishes (Myctophidae,

The overall estimated contribution of the candidate is 90%.

To be submitted to PLoS ONE as:

The overall estimated contribution of the candidate is 90%.

The overall estimated contribution of the candidate is 80%.

The overall estimated contribution of the candidate is 95%.

Peer reviewed journal articles

5. de Busserolles F., Marshall N. J., Collin S. P. 2010. Deep-sea Myctophid vision:

7. de Busserolles F., Marshall N. J., Collin S. P. 2010. Lanternfish vision:

Fanny de Busserolles Shaun P. Collin

CHAPTER 1: General introduction .............................................................................. 1

CHAPTER 2: Eye-size variability in deep-sea lanternfishes (Myctophidae): an

CHAPTER 3: The eyes of lanternfishes (Myctophidae, Teleostei): novel ocular

CHAPTER 5: Spectral tuning in the eyes of lanternfishes (Myctophidae):

CHAPTER 6: Retinal ganglion cell distribution and spatial resolving power in

CHAPTER 7: General discussion.............................................................................. 253

APPENDIX I ............................................................................................................... 263

1.1. Sunlight spectrum profile at the earth's surface ......................................................... 2

2.1. Phylogenies of Myctophidae reconstructed ............................................................. 57

4.1. Wholemount view of the rod photoreceptors ......................................................... 130

2.1. Summary of the research cruises.............................................................................. 50

3.1. Summary of retinal measurements for 53 species of lanternfishes. ......................... 93

4.1. Summary of the stereological parameters .............................................................. 130

5.1. Summary of the individual analysed ...................................................................... 174

α Combined attenuation coefficient of bioluminescence (Chapter 4)

1.1. Context: the mesopelagic zone

relatively constant colour and direction but exponentially diminishing intensity.

The mesopelagic zone is also characterised by unique physical and chemical

Figure 1.3. The distribution of bioluminescence emission maxima varies by marine

Deep-sea organisms rely on different sensory systems (vision, olfaction, hearing,

1.2. Fish model: the Myctophidae

1.2.2. Vertical migration behaviour

Intraspecific variability in migration pattern also exists and may depend on

of several species of myctophids appear to be non-migratory (i.e. Benthosema

Myctophids are mainly carnivorous, feeding mainly on zooplankton (copepods,

Stomach contents and isotope studies of myctophids indicate a high level of

always present (except Taaningichthys paurolychnus) and arranged in distinct groups

Among deep-sea fishes, two types of bioluminescence can be observed:

Symbiotic bacterial luminescence is the less common type of bioluminescence in

Self luminous bioluminescence is the most common mechanism used in oceanic

1.3.2.1. Interaction with prey by attraction or illumination.

Illumination. Some fishes possess luminous head organs (dragonfishes, lanternfishes).

Figure 1.7. Schematic diagram showing the functions of bioluminescence. Marine

1.3.2.2. Interaction with predator for the purpose of defence.

Flashes or squirts. Intimidation, distraction or disorientation are common uses of flashes

1.3.2.3. Interaction with congeners: intraspecific communication