SEE RETRACTION FOR THIS ARTICLE

Monarch butterflies (Danaus plexippus L.) use a

magnetic compass for navigation
Jason A. Etheredge*, Sandra M. Perez†, Orley R. Taylor*‡, and Rudolf Jander*

*Department of Entomology, University of Kansas, Lawrence, KS 66045; and †Department of Ecology and Evolutionary Biology, University of Arizona,
Tucson, AZ 85721

Edited by May R. Berenbaum, University of Illinois, Urbana, IL, and approved August 10, 1999 (received for review March 10, 1999)

Fall migratory monarch butterflies, tested for their directional 40 cm long). Upon reaching the top of the tube, they hesitated,
responses to magnetic cues under three conditions, amagnetic, often rotated, and took flight. The point of contact by the
normal, and reversed magnetic fields, showed three distinct pat- butterfly with the off-white paper-covered sidewall of the arena,
terns. In the absence of a magnetic field, monarchs lacked direc- estimated to the nearest 10° interval, was recorded as the
tionality as a group. In the normal magnetic field, monarchs compass heading. For the amagnetic environment, we used a
oriented to the southwest with a group pattern typical for mi- room encased in Mu-metal, a nickel-iron alloy (77% Ni, 15% Fe,
grants. When the horizontal component of the magnetic field was plus Cu and Mo), which provided an effective magnetic screen
reversed, the butterflies oriented to the northeast. In contrast, and an internal volume with an extremely low residual magnetic
nonmigratory monarchs lacked directionality in the normal mag- field. The reversed magnetic environment was created by using
netic field. The results are a direct demonstration of magnetic a set of Helmholtz coils to cancel out the normal magnetic field
compass orientation in migratory insects. and then recreate it in the opposite direction at the same strength
(11). All experiments were conducted in enclosed rooms. The
only source of light was a 120-watt bulb placed 40 cm above the
E ach fall, monarch butterflies migrate up to 4,000 km from
arena to give an omnidirectional lighting effect. The top of the

ECOLOGY
their breeding grounds in the northeastern United States and
Canada to overwintering sites in the transvolcanic mountain arena was covered with Plexiglass with a 10-cm cardboard disc
range of central Mexico. These insects, with a mass of 0.5 g, are
descendants, 3–5 generations removed, of monarchs that mi-
grated north from Mexico the previous March. There are many
unanswered questions concerning this migration; among them,
how do naive autumnal migrating monarchs navigate as they
cross the continent to the few mountainsides on which they
overwinter? Monarchs are not strong fliers and they use winds
and thermals to move in a southwestern direction (1). They are
Downloaded from https://www.pnas.org by 107.77.234.49 on May 5, 2023 from IP address 107.77.234.49.

easily blown off course, and, to reach overwintering sites,

monarchs would seem to need a general geographic sense (2, 3),
perhaps not present in nonmigratory generations. To adjust for
changes in location, the butterflies need the ability to set a
compass heading based on their present location and to reset or
remodel the compass direction should they be blown off course.
Although the overall pattern of migration within the United
States, as determined from vanishing bearings (3) and mark and
recapture records (2), suggests that monarchs adopt regional
headings consistent with magnetic compass orientation, mag-
netic orientation per se has not been established for this species.
In a previous orientation study, to determine whether mon-
archs use celestial information, Perez et al.(4) showed that clock
shifted monarchs used the position of the sun to orient them-
selves. Orientation by means of a sun compass, however, does not
address the question of how butterflies navigate on days when
the sun is not visible. Such ability could be explained by magnetic
compass orientation (3). Monarchs are thought to contain
magnetite (5, 6), a magnetically active, biosynthesized mineral
suspected to mediate orientation in organisms that respond to
magnetic fields (6). To determine whether migratory monarchs Fig. 1. Mean laboratory heading data giving resultant vector direction (␮)
respond to magnetic fields, Perez et al. (7) subjected the but- and length (r) for subjects in three environments (a) amagnetic (no mean
terflies to a magnetic pulse. As in birds, (8–10), orientation was heading for random distribution; n ⫽ 39); (b) normal magnetic field (␮) ⫽
altered after this treatment. These results demonstrate a sensi- 213.98°; r ⫽ 0.60; n ⫽ 40); (c) reversed magnetic field (␮ ⫽ 61.15°; r ⫽ 0.65;
tivity to magnetic fields, but they do not directly implicate use of n ⫽ 40); (d) reproductive, non-migratory, subjects ⬎ 7 days of age (no mean
a magnetic compass. heading for random distribution; n ⫽ 75). Each circle represents one subject.
Here we present direct experimental evidence that monarch
butterflies use an internal magnetic compass to maintain their
This paper was submitted directly (Track II) to the PNAS office.
migratory direction. Compass headings of field-collected fall ‡Towhom reprint requests should be addressed at Department of Entomology, 7005
migratory monarchs were recorded in amagnetic, normal, and Haworth Hall, University of Kansas, Lawrence, KS 66045. E-mail: [email protected].
reversed magnetic environments with the use of a circular (0.5 The publication costs of this article were defrayed in part by page charge payment. This
m radius and 0.43 m sidewall) arena. Butterflies entered the article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C.
arena from below by climbing up a central tube (10 cm i.d. and §1734 solely to indicate this fact.

PNAS 兩 November 23, 1999 兩 vol. 96 兩 no. 24 兩 13845–13846

 in the center to prevent direct light from entering the entrance least 7 days post eclosion were also tested in the normal magnetic
tube. field. Headings of this reproductive group were randomly dis-
Butterflies tested under amagnetic conditions showed no tributed (Rayleigh test of uniformity, P ⫽ 0.07, n ⫽ 75, Fig. 1d),
consistent directionality as a group (Fig. 1a) (Rayleigh test of indicating that they lacked a consistent response to the normal
uniformity, P ⫽ 0.23, n ⫽ 39), which is indicative of the absence magnetic field. Reproduction, however, does not appear to limit
of orientation stimuli. The mean direction under normal mag- response to the magnetic field because monarchs are reproduc-
netic conditions (Fig. 1b) was southwest (␮ ⫽ 213.98; r ⫽ 0.60; tive on the return spring migration. Thus, the response to the
n ⫽ 40) and did not differ significantly from the average magnetic field appears to be coupled with the migratory con-
direction (␮ ⫽ 200) of naturally migrating monarchs (F ⫽ 3.51; dition but not the reproductive state. The external cues and/or
P ⬎ 0.05) in eastern Kansas (4). Butterflies tested in a reversed physiological changes that trigger migration and cause monarchs
magnetic field (Fig. 1c) flew toward the northeast (␮ ⫽ 61.15; r ⫽ to respond to magnetic information have not been determined.
0.65; n ⫽ 40), that is, in the reverse direction of the normal Although monarchs use both magnetic and sun compass
migratory behavior. The results from this reversed group are information to orient and navigate, the system is probably more
significantly different from the normal field group (Watson’s complex. Once monarchs reach Mexico, the predominant course
F-test, F ⫽ 121.36; P ⬍ 0.01). These tests demonstrate that changes from southwest to southeast along the Sierra Madre
monarchs navigate by means of a magnetic compass even in the Oriental (13), suggesting that structural features or other cues
absence of celestial information. Whether monarchs use the are used for navigation.
magnetic dip angle or horizontal polarity, or both, is an open
question. The site of magnetoreception may be the thorax, which We thank the University of Kansas Department of Geology and Paul
contains 65% of the magnetite (5), rather than the head, where Montgomery for use of the amagnetic room and Bill McGregor for
in rainbow trout, it is associated with olfactory lamellae (12). assistance in the design and construction of the arena. Support for this
The butterflies used in these tests were fall migrants that study was provided by Monarch Watch, The George E. Gould Student
remain nonreproductive throughout the migration. A different Assistance Fund, The University of Kansas Division of Biology, and a
set of monarchs that were reproductive, nonmigratory, and at National Science Foundation Postdoctoral Fellowship to S.M.P.

1. Gibo, D. L. & Pallett, M. J. (1979) Can. J. Zool. 57, 1393–1401. 6. Jungreis, S. A. (1987) Florida Entomologist 70, 277–283.
2. Rogg, K. A., Taylor, O. R. & Gibo, D. L. (1999) North American Conference 7. Perez, S. M., Taylor, O. R. & Jander, R. (1999) Naturwissenschaften 86,
on the Monarch Butterfly, eds. Hoth, J., Merino, L., Oberhauser, K., Pisanty, I., 140 –143.
Price, S. & Wilkinson, T. (Commission for Environmental Cooperation, 8. Kirschvink, J. L & Gould, J. L. (1981) BioSystems 13, 181–201.
Montreal), in press. 9. Wiltschko, W., Munro, U., Beason, R. C., Ford, H. & Wiltschko, R. (1994)
3. Schmidt-Koenig, K. (1993) in Biology and Conservation of the Monarch Experientia 50, 697–700.
Butterfly, eds. Malcom, S. B. & Zaluki, M. P. (Natural History Museum of Los 10. Beason, B. C., Wiltschko, R. & Wiltschko, W. (1997) Auk 114, 405–415.
Angeles County, Los Angeles), pp. 275–283. 11. Wiltschko, R. & Wiltschko, W. (1995) Magnetic Orientation in Animals.
4. Perez, S. M., Taylor, O. R. & Jander, R. (1997) Nature (London) 387, 29. (Springer, Berlin).
5. MacFadden, B. J. & Jones, D. S. (1985) in Magnetite Biomineralization and 12. Walker, M. M., Diebel, C. E., Haugh, C. V., Pankhurst, P. M., Montgomery,
Magnetoreception in Organisms, eds. Kirschvink, J. L., Jones, D. S. & MacFad- J. C. & Green, C. R. (1998) Nature (London) 320, 371–376.
Downloaded from https://www.pnas.org by 107.77.234.49 on May 5, 2023 from IP address 107.77.234.49.

den, B. J. (Plenum, New York), pp. 407–415. 13. Rzedowski, J. (1957) Acta Zool. Mex. 2, 1–4.

13846 兩 www.pnas.org Etheredge et al.