RESEARCH ARTICLE

The Use of Asian Ficus Species for Restoring

Tropical Forest Ecosystems
Cherdsak Kuaraksa1,2 and Stephen Elliott3

Abstract stock from cuttings. The most efficient method of produc-

ing Ficus spp. was from seed. Propagation from cuttings
Fig (Ficus spp.) trees have been promoted as frame-
work species for tropical forest restoration throughout was much less successful. Seedlings produced from seed
Asia, because they are considered to be keystone species. had the highest rates of growth and survival both in the
This article presents optimal propagation and planting nursery and in field trials. In field trials, use of planting
techniques for six Asian dioecious Ficus species, which stock from seed was also more cost-effective than direct
will enable their inclusion in forest restoration plantings seeding and vegetative propagation. Establishment costs
across the Asia-Pacific region: Ficus auriculata, F. fulva, calculated on the basis of “per plant established” were
F. hispida, F. oligodon, F. semicordata, and F. varie- $1.14 for seed, $6.95 for cutting, and $25.88 for direct
gata. Nursery experiments compared the growth perfor- seeding.
mance of propagating planting stock from seed and from
leafy cuttings, whereas field experiments assessed the cost- Key words: dioecious Ficus trees, framework species,
effectiveness and the relative performance of (1) direct planting stock types, propagation, seedling performance,
seeding, (2) planting stock from seed, and (3) planting tropical forest restoration.

Introduction Like many tropical countries, tree planting has become

Destruction of tropical forests continues to be a major threat to popular all over Thailand, in particular, as a response to
global biodiversity and also contributes significantly to global mitigate global climate change. However, many tree planting
climate change. Forest restoration could be a significant part projects fail, due to the planting of inappropriate species, and
of the solution, if efficient techniques can be developed and inadequate planting techniques and post-planting maintenance
regimes particularly when using forest native tree species,
sociopolitical conditions allow them to be practiced. Many
about which little is known (Blakesley et al. 2002). Since
tropical countries have recently started incentive programs to
1994, the Forest Restoration Research Unit of Chiang Mai
restore tropical forest lands, particularly for carbon sequestra-
University’s Biology department (FORRU-CMU) has been
tion and biodiversity conservation. However, the effectiveness
developing methods to restore forest ecosystems to deforested
of such restoration projects depends on the development of
sites within protected areas, for biodiversity conservation,
simple but effective methods to grow, plant, and take care of
and environmental protection in northern, Thailand. The unit
the many tree species that comprise tropical forest ecosystems.
successfully adapted the so-called “the framework species
Restoration ecology is still a young and developing science,
method” (Goosem & Tucker 1995) to restore forest ecosystems
and tropical forest ecosystems are complex, with different
in Thailand. Biodiversity recovery was achieved and the
sites requiring different restoration techniques. Several meth-
tree species composition of restored plots approached that
ods have been developed to restore degraded tropical forest
of the original primary forest more rapidly than by natural
lands such as maximum diversity plantings (Miyawaki 1993),
regeneration. Planting 29 framework tree species resulted in
nurse crops, or foster ecosystems (Parrotta 1993), staggered
more than 70 non-planted tree species recolonizing the plots
planting of primary forest species (Knowles & Parrotta 1995),
within 8 years and bird species richness increased from 30 to
and scattered tree planting (Lamb & Gilmour 2003). Most
88, within 6 years (Elliott & Kuaraksa 2008).
approaches involve planting of tree seedlings.
In the tropics, Ficus species are excellent framework species
for restoring tropical forest ecosystems, due to their role in
1 Environmental Science Program, Faculty of Science, Chiang Mai University,
ecosystems as keystone species, which sustain populations
Chiang Mai 50200, Thailand
2 Address correspondence to C. Kuaraksa, email [email protected] of seed dispersing animals (FORRU 2006). This is because
3 Forest Restoration Research Unit, Department of Biology, Faculty of Science, of their all-year-round production of figs, often in large
Chiang Mai University, Chiang Mai 50200, Thailand quantities, which can potentially maintain wildlife populations
© 2012 Society for Ecological Restoration International
when other fruits are in short supply (Lambert & Marshall
doi: 10.1111/j.1526-100X.2011.00853.x 1991). Thus, they help to maintain viable populations of seed

Restoration Ecology 1
The Use of Fig Trees in Forest Restoration Plantings

Figure 1. Average monthly rainfall (mm), maximum and minimum temperature (◦ C) at the Northern Meteorological Center, about 3 km from the
National Park (from May 2009 to October 2010).

dispersers, which are vital for recovery of tree species richness (May to October) and the dry season (mean monthly rainfall
in regenerating forest. Fig trees thereby favor the dispersal below 100 mm, November to April). Average annual rainfall
of other animal-dispersed plant species, influencing plant and temperature recorded at the nearest weather station are
community composition, and rates of succession (Compton presented in Figure 1.
et al. 1996). Furthermore, they have great potential to restore All seeds and cuttings were collected from trees beside
forest on sites with heavily compacted soils (e.g. open-cast dirt tracks which run through natural or disturbed forest
mines), since their roots can penetrate and break open even ecosystems of the park. All selected Ficus species were
the hardest substrates, including rock, improving aeration, and dioecious. In dioecious Ficus species (which comprise about
drainage of substrates, which allows establishment of other half of the genus worldwide), male and female functions
tree species (Thornton et al. 2001; FORRU 2006). However, are separated, with female trees producing only seed and
most research on Ficus species is focused on their taxonomy, male trees producing wasps and pollen (Harrison 2005). More
pollination, and the role of fig wasps. Their propagation and details of the six selected dioecious Ficus species are presented
their use for forest restoration have not been extensively in Table 1. Although the study site was in Thailand, the species
studied. Ficus species may be propagated from seed, cuttings, studied have wide distributions across the Asia-Pacific region,
layering, grafting, and tissue culture (Rahman et al. 2004; so results can be applied to many countries.
Dolgun & Tekintas 2008; Mathew et al. 2011). With the
exception of seeds and cuttings, these techniques are generally
Nursery Trials
considered unsuitable for mass propagation, particularly in
developing countries, due to high costs and the need for Propagation from Seed. Mature, ripe figs of the six selected
technical expertise. The purpose of this study was to compare Ficus spp. were collected from 10 or more individual trees
the use of three planting stock types to establish Asian Ficus of each species. Fruits were removed directly from the plant,
species in disturbed habitats, in order to enable their inclusion rather than harvested from the ground, principally to reduce
in forest restoration projects across the region. The emphasis the risk of disease infection (FORRU 2006). Figs were opened
was on developing methods which are practical for use in and the seeds scraped out with a spoon. The pulp was sieved
broad-scale restoration activities, but simple, inexpensive, and through a mosquito net in water, so that viable seeds passed
easily acquired by local communities. through the mosquito net and sank. Seeds were spread out on
paper and left to dry for 1–2 days being sown into modular
plastic trays, by placing them on the surface of the germination
Methods medium (1:1 mixture of sand and charcoalized rice husk),
uncovered. Seeds were watered by hand using a fine spray
Study Sites and Species bottle. Germination and survival were monitored weekly, and
The study site was Doi Suthep-Pui National Park in northern the experiment ended 30 days after the final germination event
Thailand. The area has two main seasons: the wet season was recorded.

2 Restoration Ecology
 The Use of Fig Trees in Forest Restoration Plantings

Table 1. Species studied.

Abundance No. of Seeds (per

Ficus Speciesa Distribution Rangeb (Country) Habitatc Elevationd (m) Ranke Fig, n = 50 )

Ficus auriculata Lour. Bhutan, Cambodia, China, India, Laos, Streams in dof, 891–1,319 Rare 5,969 ± 4,287
Myanmar, Nepal, Pakistan, Thailand, egf, eg/pine
Vietnam
Ficus fulva Reinw. Brunei, China, India, Indonesia, Malaysia, da, sg in egf 923–1,100 Rare 858 ± 311
Ex Blume Myanmar, Thailand, Vietnam
Ficus hispida L.f. Australia, Bhutan, Cambodia, China, da in bb/df, sg 326–1,351 Medium 1,140 ± 486
India, Indonesia, Laos, Malaysia,
Myanmar, Nepal, Sri Lanka, Thailand,
Timor, Vietnam
Ficus oligodon Bhutan, China, India, Malaysia, da, open bb/df, 605–1,336 Medium 8,692 ± 4,244
Miquel. Myanmar, Nepal, Thailand, Vietnam mxf, egf, sg
Ficus semicordata Bhutan, China, India, Malaysia, sg, da in bb/df, 418–1,531 Medium 712 ± 413
Buch.-Ham. ex Sm. Myanmar, Nepal, Thailand, Vietnam egf, eg/pine
Ficus variegata Australia, Cambodia, China, India, egf 899–1,343 Rare 1,825 ± 978
Blume Indonesia, Japan, Laos, Malaysia,
Myanmar, Philippines, Thailand, Vietnam
a All selected Ficus species were dioecious.
b Source: Flora of China (Zhekun & Gilert 2003); Flora Malesiana (Berg & Corner 2005).
c Source: Maxwell and Elliott (2001); dof, deciduous dipterocarp-oak forest; egf, primary evergreen forest; eg/pine, evergreen forest with pine; da, disturbed areas; sg, secondary
growth; bb/df, degraded teak and bamboo + deciduous forest.
d Altitude range (m) of the parent/donor trees was found along the seed/cutting collection trails of Doi Suthep-Pui National Park.
e Species abundance in the park is ranked by Maxwell and Elliott (2001); F. fulva is a synonym of F. hirta Vahl var. roxburghii (Miq.) and F. oligodon is a synonym of
F. fistulosa Reinw. ex Bl. var. fistulosa.

Propagation from Cuttings. A low-cost technique was used fertilizer (NPK 14-14-14, Osmocote, Marysville, CA, U.S.A.)
for cutting experiments, developed at the ASEAN Forest Tree were applied every 3 months. Seedling growth (root collar
Seed Centre, Muak-Lek, Thailand, using closed plastic bags to diameter, height, canopy width, and health) and survival rate
retain high humidity, a so-called non-mist propagation system were monitored monthly. A randomized complete block design
(Kantarli 1993). Cuttings were collected from both sexes of was used for all experiments, with three replications for each
the six selected adult Ficus tree species. Lateral branches were species. Each replicate consisted of 100 seeds or cuttings,
cut and mature and hardened branches selected. Each cutting except for seedling growth trials in which 30 seedlings were
was four nodes long (about 10–20 cm in length, depending used per replicate for each propagation types in the nursery.
on species), basal cuts at least 0.5 cm below a node, with
only one leaf attached, and the leaf on each cutting was cut Field Trials
in half. The cuttings were put in a plastic bag and placed in
Three different methods were tested for establishing Ficus spp.
a refrigerator (5◦ C) overnight to seal the wound and prevent
trees: (1) direct seeding; (2) planting stock from cuttings; and
bacterial infection. The basal ends of the cuttings were dipped (3) planting stock from seed. Three replicated blocks (each
in rooting powder hormone (Seradix; IBA̸=3; 4-(Indol-3-yl) approximately 30 × 30 m) were established to compare the
butyric acid, Leverkusen, North Rhine-Westphalia, Germany). field performance of the three planting stock types of six Ficus
Cuttings were planted to half of their length (two nodes) into species in disturbed habitats. Plots were prepared by weeding
black plastic bags (5 × 13 cm) filled with sand and charcoal- with hand tools about 1 week before planting. In each block,
ized rice husk (1:1). The small plastic bags were then placed different planting stock types of the six Ficus species (30
in larger plastic bags (60 × 90 cm; with 1 L of water added), individuals per block for each of the two planting stock types,
sprayed with water until the medium was saturated. Plastic and 60 units of Ficus seeds for direct seeding; one unit = 1
bags were closed firmly to prevent moisture loss and kept bamboo tube with 100 seeds) were planted randomly (except
under 70% shade. After that, the plastic bags were opened for direct seeding where seeds were sown in rows in order
weekly, for 5–10 minutes, to allow some moisture out, dead to find them easily when monitoring), with a mean distance
or yellowing leaves were removed. Plants were watered only between plants of 1.5 m. Fifty grams of fertilizer (NPK 15-15-
when the soil surface dried or if there was no evidence of con- 15) was applied in a ring around each tree (but not for direct
densation on the inside of the plastic bag. Time to emergence seeds). After planting out, plots were weeded and additional
of first shoot, root and survival was observed weekly. dose of fertilizer applied at 6 week intervals, during the rainy
Once mature leaves had expanded, seedlings from both tech- season for the first 2 years after planting. The planted trees
niques were pricked out and potted into new containers (black were monitored for field performance (survival and growth
plastic bags: 6.5 × 22 cm) using a medium of soil, peanut rates) three times; (1) immediately post-planting, (2) at the end
husk, and coconut husk (2:1:1). Ten granules of a slow-release of the first rainy season, and (3) at the end of the second rainy

Restoration Ecology 3
The Use of Fig Trees in Forest Restoration Plantings

season. Measurements included height (root collar to highest each species ranged from 35 ± 2 to 79 ± 2.5%. Differences
meristem measured by measuring pole), root collar diameter in germination among species were statistically significant
(measured with Vernier calipers), and canopy width (at widest (Fig. 2a). At the time of pricking out (after expansion of the
point using a tape measure), and the health of the trees was second true leaf pairs, which took about 4 months after ger-
scored on a scale of 0 (dead) to 3 (perfect health). mination) seedlings were tiny; only 1 to 2 cm tall. Subsequent
For direct seeding, seeds were sown on the surface of the seedling growth rate of all species was very high but it did
soil but protected inside bamboo tubes 8 × 10 cm (half of not differ significantly among species (Fig. 3a–c). By plant-
the bamboo tubes were stuck in soil), in order to prevent seed ing time (after 1 year in nursery), seedlings had mostly grown
movement due to wind and rain. Each tube contained about taller than 15 cm and seedlings of all species remained in
100 seeds but was counted as one unit irrespective of how good health throughout their growth in the nursery (Table 2;
many seeds germinated. Germination was monitored at weekly Fig. 3d). However, the mean overall success (proportion ger-
intervals for 3 months. After expansion of the second true leaf minated × proportion survival) of all species was fairly low.
pairs, seedlings were monitored for field performance, using F. variegata had the highest mean overall success compared
the standard silvicultural and monitoring methods as for the with all other species (SD = 0.03, p < 0.05; Fig. 2d).
other planting stock types.
Propagation from Cuttings. All species developed new
shoots 2–4 weeks after being placed in the propagation bags,
Data Analysis and most cuttings produced new roots about a week after above
Data were subjected to analysis of variance (ANOVA), T -test, ground growth initiated. Averaging across all species, the mean
and least significant difference or Scheffe test (where needed) shooting/rooting percentage was 66.1% (SD = 17). The mean
at p = 0.05 significance level, using SPSS 17.0. The median values for individual species ranged from 19.3 ± 5.1% for
length of dormancy (MLD) was calculated from the germi- F. fulva to 89.7 ± 2.5% for F. oligodon. Shooting/rooting
nation times of all seeds which germinated. Overall success ability also varied significantly among species (ANOVA, p <
of germination/cutting trials was defined as the probability 0.05; Fig. 2a). Within 7–8 weeks, cuttings of all species
of germination/shooting multiplied by the probability of early were ready for removal from the enclosed plastic bags
seedling survival, converted to a percentage, that is, the num- (mature leaves produced). F. auriculata cuttings success was
ber of seedlings that could be potted from sowing 100 seeds significantly higher when compared to all other species in the
or make 100 cuttings. The relative growth rates (RGR) were trial (ANOVA, p < 0.05; Fig. 2b & 2c). However, after the
calculated using the formula: cuttings were removed from the enclosed plastic bags and
replanted in a new container, their growth and survival rates
ln(G2) − ln(G1) were very low (Figs. 2d & 3a–c). F. fulva had, significantly,
RGR = × 365 × 100
T2 − T1 the lowest mean overall success, compared with all other
species (p < 0.05; Fig. 2d). By the first planting time, the
where G1 and G2 are the growth parameters (root collar mean height for individual species ranged from 15.7 cm for
diameter, height, and canopy width) at the beginning (T1) and F. oligodon to 21.3 cm for F. auriculata (Table 2).
end (T2) of the sampling period.
Cost per plant of each planting stock type was calcu- Propagation Type Comparison. During the first 12-month
lated for each stage of the process throughout the study period in the nursery, averaged across all species, the RGR
period (1.5 years for direct seeded and 2.5 years for both of seedlings (656.7 ± 40.6%/year for root collar diameter,
nursery grown plants). Operational costs and labor require- 1,091.8 ± 164.2%/year for height, and 1,192.0 ± 111.4%/year
ments for activities from seedling production in the nursery to for canopy width) was higher than that of rooted cuttings
post-planting maintenance were recorded. These included all (88.6 ± 8.3%/year for root collar diameter, 75.6 ± 13.3%/year
materials and labor costs associated with seed/cutting collec- for height, and 149.1 ± 9.6%/year for canopy width). Stock
tion, planting stock production, transportation, site preparation, plants derived from seed also showed higher success rates
plantation establishment, and maintenance. We used a rate of than that of stock plants derived from cuttings (0.22 and
US$6.53 per day (8 hours) to calculate labor expenses. 0.07, respectively, T -test, p < 0.05). However, by planting
time, seedling size varied among propagation types. Seedlings
derived from seed were significantly taller and had higher
Results mean health scores compared with stock plants derived from
cuttings (23.0 ± 4.9 cm and 17.7 ± 2 cm for height, 2.9
Nursery Trials ± 0.1 and 2.6 ± 0.2 for health score, respectively, T -test,
Propagation from Seed. Seeds of all species began to p < 0.05). Conversely, seedlings propagated from cuttings
germinate within 3–4 weeks after sowing and completed showed significantly higher mean root collar diameter than
germination within 7–8 weeks. MLD ranged from 18 days that of seedlings propagated from seed (10.1 ± 2 mm and
for Ficus semicordata (SD = 0.3) to 27 days for F. hispida 5.9 ± 0.4 mm, respectively, T -test, p < 0.05). However, there
(SD = 4.4), but MLD was not significantly different among were no significant differences in mean canopy width among
species (ANOVA, p = 0.15). Germination percentages for the propagation types (29.3 ± 2.1 cm for stock plants derived

4 Restoration Ecology
 The Use of Fig Trees in Forest Restoration Plantings

Figure 2. Comparison of means for each species by each propagation type in the nursery trials. FIAU, Ficus auriculata; FIFU, F. fulva; FIHI,
F. hispida; FIOL, F. oligodon; FISE, F. semicordata; FIVA, F. variegata; (a) germinating/shooting rate; (b) survival rate by transplanting time;
(c) survival rate by planting time; and (d) overall success rate (proportion germinated/rooted × proportion survival by planting time). Different letters
above the bars indicate significant differences among species (p < 0.05).

from seed and 26.3 ± 4.8 cm for stock plants derived from that survived the period December 2009 to October 2010 are
cuttings, T -test, p = 0.19). presented in Figure 4a–c.

Field Trials Planting Stock-Raised in Nursery from Seed. The trees

Direct Seeding. Seed germination in the field ranged from were approximately 20 cm tall when planted (averaged across
47.2 ± 8.1% (F. hispida) to 73.9 ± 2.8% (F. auriculata), the all species) but they mostly grew taller than 1.5 m after
MLD ranged from 20 (F. oligodon) to 28 days (F. variegata), 1.5 years with F. semicordata and F. variegata reaching
but the differences among species were not statistically sig- 2 m (Table 2). From May 2009 to October 2010, F. fulva
nificant (p = 0.27 for germination and p = 0.06 for MLD). achieved the highest growth rate (relative root collar diam-
However, seedling survival after germination of all species eter growth rate [RRGR] = 141.2 ± 11.3%/year and rela-
was very low. The highest mortality for all species occurred tive height growth rate [HRGR] = 183.1 ± 12.3%/year), while
during the first rainy season (>90% mortality by 1 month F. semicordata had the highest relative canopy width growth
after germination) due to damping-off diseases. Overall mor- rate (CRGR) (115.6 ± 7.2%/year). However, differences in
tality rate of most species was 100% during the first dry relative growth rates among species were not statistically
season, except for F. hispida of which only 1.7% (SD = 0.3) significant (Fig. 4a–c). At the end of the second rainy sea-
of seeds planted remained alive at the end of second rainy son, F. variegata had the highest mean root collar diame-
season (Fig. 4d). RGRs (%RGR/year) of F. hispida saplings ter and mean height (37.8 mm and 264.6 cm, respectively),

Restoration Ecology 5
The Use of Fig Trees in Forest Restoration Plantings

Figure 3. Mean relative growth rates and mean health scores by species for each propagation types after 1 year in the nursery. (a) Relative root collar
diameter growth rate (mm); (b) relative height growth rate (cm); (c) relative canopy width growth rate (cm); and (d) health score.

while F. semicordata had the highest mean canopy width RRGR and HRGR (p < 0.05; Fig. 4a & 4b). At the end of
(141.4 cm); however, ANOVA showed no significant differ- the second rainy season, F. auriculata had the highest mean
ences among the species means (ANOVA, p = 0.59 for root height (77.6 cm), followed by F. variegata (69.6 cm) and
collar diameter, p = 0.08 for height, and p = 0.55 for canopy F. oligodon (51.6 cm), significantly taller compared with the
width). Survival after two rainy seasons averaged 68.1% other species (ANOVA, p < 0.05). While, F. variegata had,
(SD = 5.7) across all species. The mean values for individ- significantly, the highest mean root collar diameter (23.3 mm)
ual species ranged from 63.3% for F. semicordata to 78.3% and mean canopy width (98.1 cm) compared with all other
for F. variegata. Overall, there were no statistically significant species (ANOVA, p < 0.05; Table 2).
differences for survival among species (Fig. 4d).

Planting Stock-Raised in Nursery from Cuttings. After Planting Stock-Type Comparison. By the end of the sec-
planting out in disturbed areas, 67% of plants propagated from ond rainy season (1.5 years after planting), averaging across
cuttings died within 7 months (averaged across all species). At the species, nursery-grown saplings from seeds performed sig-
the end of the second rainy season, mean survival of saplings nificantly better than other planting stock types. They had,
across all species was 15.1% (SD = 3.5). The mean values for significantly, the highest survival rate (68.1%), highest rel-
individual species ranged from 0% for F. fulva to 36.7% for ative growth rate (126.8 ± 14.1% RRGR/year, 162.2 ±
F. hispida, differences among species in relation to survival 16.7% HRGR/year, and 105.3 ± 9.8% CRGR/year), and the
rate (ANOVA, p < 0.05; Fig. 4d). The relative growth rate of largest seedling size (31.6 ± 4.1 mm for root collar diameter,
seedlings propagated from cuttings was also fairly slow, with 199.3 ± 45.1 cm for height, and 122.9 ± 13.3 cm for canopy
differences among species statistically significant for mean width).

6 Restoration Ecology
 The Use of Fig Trees in Forest Restoration Plantings

Table 2. Seedling growth comparison between the planting stock types for each species, which attained at the first planting time (1 yr in the nursery)
and at the end of the second rainy season (1.5 yr after planting out).

Nursery Trials Field Trials

Planting stock Planting stock Planting stock Planting stock
from seed from cuttings Direct seeding from seed from cuttings

Ficus auriculata
RCD (mm) 6.2b ± 0.7 12.4a ± 2.3 — 31.7a ± 2.8 21.1b ± 1.0
Height (cm) 18.3a ± 5.0 21.3a ± 1.9 — 162.4a ± 17.6 77.6b ± 5.4
Canopy width (cm) 30.6a ± 4.4 30.8a ± 2.6 — 119.1a ± 4.0 93.7a ± 2.9
Ficus fulva
RCD (mm) 5.9a ± 0.9 8.7a ± 2.0 — 30.5 ± 6.0 —
Height (cm) 23.0a ± 11.2 17.0a ± 1.7 — 196.1 ± 42.2 —
Canopy width (cm) 31.4a ± 9.3 27.7a ± 5.7 — 121.2 ± 35.3 —
F. hispida
RCD (mm) 5.2b ± 0.3 10.5a ± 0.8 35.5a ± 13.4 26.3a ± 8.7 18.6a ± 2.0
Height (cm) 25.9a ± 3.7 18.5b ± 0.6 188.5a ± 54.4 154.8ab ± 61.3 50.1b ± 2.8
Canopy width (cm) 25.9a ± 1.9 31.4a ± 4.0 141.0a ± 84.6 108.2a ± 27.6 73.1a ± 2.2
Ficus oligodon
RCD (mm) 6.0b ± 0.5 9.0a ± 1.0 — 28.8a ± 5.9 16.9b ± 1.7
Height (cm) 15.9a ± 4.8 15.7a ± 0.9 — 174.6a ± 33.8 51.6b ± 12.3
Canopy width (cm) 30.4a ± 2.0 25.7b ± 1.8 — 111.4a ± 23.4 75.6a ± 15.5
Ficus semicordata
RCD (mm) 5.6b ± 0.6 7.4a ± 0.5 — 34.3a ± 12.0 12.0b ± 6.9
Height (cm) 28.3a ± 6.5 16.8b ± 1.1 — 243.3a ± 70.0 29.2b ± 16.7
Canopy width (cm) 29.6a ± 1.2 18.7b ± 1.5 — 141.4a ± 27.0 68.7b ± 39.3
Ficus variegata
RCD (mm) 6.3b ± 0.2 12.3a ± 0.6 — 37.8a ± 9.7 23.3a ± 0.4
Height (cm) 26.5a ± 3.5 17.0b ± 0.0 — 264.6a ± 49.8 69.6b ± 9.1
Canopy width (cm) 27.8a ± 1.3 23.8a ± 2.5 — 136.0a ± 23.3 98.0a ± 3.2
RCD, root collar diameter; mean ± SD. Values in a row with different letters are significantly different according to T -test or LSD-test (where needed) at p < 0.05.

Cost Comparison. On the basis of our calculations, the cost These fungi species were observed when seedlings were tested
per plant of each planting stock types are presented in Table 3. in a plant pathogen laboratory (C. Kuaraksa 2009, Chiang
The major cost of all planting stock types was for labor which Mai University, personal observation). However, if the limiting
account for 79.5, 70.5, and 77.1% of total costs (for direct factor of damping-off disease could be overcome, then direct
seeding, planting stock-raised in nursery from seed and cutting, seeding could become a very cost-effective technique, because
respectively). If the survival rates are 100%, direct seeding was Ficus plants which survive the first rainy season grow rapidly
the cheapest technique, but when we took mortality rate into and there are no nursery costs.
account, planting stock-raised in the nursery from seed was the All six Ficus species had germination percentages in
most cost-effective, compared with the other two techniques excess of 35%, which is acceptable for nursery production
(Table 3). of native trees for forest restoration purposes (Elliott et al.
2002). All fig seedlings performed well after transplanting
in the high-light intensity conditions of forest restoration
Discussion plots, because most dioecious Ficus spp. are pioneer species
Because fig seeds are so small, they do not contain sufficient (Thornton et al. 2001). The recommended height of sapling for
food reserves (endosperm) to support early seedling growth, planting out for the framework species method is 30–60 cm
which probably makes them unsuitable for direct seeding. (FORRU 2006). The Ficus saplings, used in this experiment,
Previous studies have shown that large- or medium-sized were planted out when only about 20 cm tall. However,
seeds are most suitable for this technique (Manga & Sen this did not appear to affect their subsequent growth and
1995; Doust et al. 2006). Very high mortality rates after survival. Therefore, Ficus seedlings can probably be planted
direct seeding of Ficus spp. trees have also been reported for out as soon as they have grown 20 cm or taller. Two
F. stupenda planted in Borneo, of which only 1.3% remained species established from seed achieved excellent survival
alive at the end of 1 year (Laman 1995). From this study, rates (>70% after the second rainy season; F. oligodon and
most seedlings died of fungal infection (damping-off diseases), F. variegata), while F. auriculata, F. fulva, F. hispida, and
probably derived from soil pathogens including Penicillium F. semicordata had acceptable survival in field performance
spp., Fusarium spp., Aspergillus spp., and Rhizopus spp. standards for framework species (>50%). All species also

Restoration Ecology 7
The Use of Fig Trees in Forest Restoration Plantings

Figure 4. Mean relative growth rate and mean survival by species for each planting stock types after 1.5 years of planting out in disturbed habitats. (a)
Relative root collar diameter growth rate (mm); (b) relative height growth rate (cm); (c) relative canopy width growth rate (cm); and (d) survival rate.
Different letters above the bars indicate significant differences among species (p < 0.05).

achieved acceptable or excellent height growth by the end from main stems. Thus, after rooting, cuttings from branches
of the second rainy season (>1.5 m, Elliott et al. 2003). grow into plants which may not grow normally, particularly
Results of the study also revealed that the expense of planting developing reproductive parts rather than roots. In addition, the
stock-raised in nursery from seed is inexpensive, if most labor roots of plants, propagated from cuttings, are quite fragile and
cost can be replaced with volunteers. Moreover, Ficus spp. sensitive, and can be broken when placed directly into ground
seeds are orthodox and easy to collect as they are available or during transportation (Dolgun & Tekintas 2008). However,
all-year-round hence intensive phenology studies may not be in the case of dioecious Ficus species, vegetative propagation
needed. from cuttings might be useful to ensure a fairly even sex
Producing Ficus spp. planting stock from cuttings was ratio among the planting stock. Vegetative propagation from
inefficient for all tested species. Low percentage survival cuttings may also be useful to select tree phenotypes with
(15.1%, averaged across all species) and high cost were the desirable features (i.e. female Ficus spp. trees, which fruit
most important reasons to reject this method for use in large- when other foods are scarce, and male Ficus spp. trees, which
scale forest restoration projects, as it necessitates expensive fruit during the critical bottleneck for wasp dispersal). Further
replanting. The poor growth and survival rates of planting research on rejuvenation techniques of mature trees may help
stock derived from cuttings might be explained by cutting to increase rooting potential and survival rate (Hartmann
from branches of mature trees which do not generally root et al. 1990; Longman & Wilson 1993; Kibbler et al. 2004),
well (Smits et al. 1990; Kantarli 1993). Longman and Wilson but it seems expensive and requires intensive trained staff.
(1993) suggested that branches are unsuitable for cuttings Further field trials on the use of “truncheons” (i.e. large
because they have a different structure and growth hormones cuttings, approximately 2 m tall) which are commonly used

8 Restoration Ecology
 The Use of Fig Trees in Forest Restoration Plantings

Table 3. Establishment and maintenance costs (per plant) through of LITERATURE CITED
study.
Berg, C. C., and E. J. H. Corner. 2005. Moraceae—Ficus. Flora Malesiana
Planting Stock Type Series I (seed plants). Vol. XVII/Part 2. National Herbarium of The
Netherlands, Leiden, The Netherlands.
Planting Planting Blakesley, D., K. Hardwick, and S. Elliott. 2002. Research needs for restoring
Direct stock-raised in stock-raised in tropical forests in Southeast Asia for wildlife conservation: framework
seeding nursery from nursery from species selection and seed propagation. New Forests 24:165–174.
(U.S. $ per seed (U.S. cutting (U.S.
Compton, S. G., J. T. Wiebes, and C. C. Berg. 1996. The biology of fig trees
Items plant) $ per plant) $ per plant)
and their associated animals. Journal of Biogeography 23:405–407.
Nursery materialsa 0 0.03 0.04 Dolgun, O., and F. E. Tekintas. 2008. Production of fig (Ficus carica L.) nurs-
Labor cost in nurseryb 0 0.14 0.40 ery plants by stem layering method. Agricultural Conspectus Scientificus
Field materialsc 0.09 0.20 0.20 73:157–160.
Doust, S. J., P. D. Erskine, and D. Lamb. 2006. Direct seeding to restore
Labor cost in fieldd 0.35 0.41 0.41
rainforest species: microsite effects on the early establishment and growth
Total cost (100% 0.44 0.78 1.05
of rainforest tree seedlings on degraded land in the wet tropics of
survival)
Australia. Forest Ecology and Management 234:333–343.
Establishment coste (per 25.88 1.14 6.95
Elliott, S., and C. Kuaraksa. 2008. Producing framework tree species for
plant established)
restoring forest ecosystems in northern Thailand. Small-Scale Forestry
a Nursery materials include containers, media, fertilizer, and rooting hormone. 7:403–415.
b Labor for seed/cutting collection, seed/cutting preparation, seed sowing/cutting, Elliott, S., C. Kuarak, P. Navakitbumrung, S. Zangkum, V. Anusarnsunthorn,
potting, watering, fertilizing, weeding, and grading. During the period of study, and D. Blakesley. 2002. Propagating framework trees to restore season-
manual labor cost average $6.53 per day (8 hours); 1 US$ = 30.65 Thai baht. ally dry tropical forest in northern Thailand. New Forests 23:63–70.
c Field materials include bamboo stakes, bamboo tubes, fertilizers, and gasoline.
d Labor for seeds/seedling transferring, planting, weeding, and fertilizing. Elliott, S., P. Navakitbumrung, C. Kuarak, S. Zangkum, V. Anusarnsunthorn,
e On the basis of survival rates at the end of the second rainy season were 1.7% for and D. Blakesley. 2003. Selecting framework tree species for restoring
direct seeding (averaging only for Ficus hispida), 68.1% for seed, and 15.1% for seasonally dry tropical forests in northern Thailand based on field
cutting (averaging across all species). performance. Forest Ecology and Management 184:177–191.
FORRU (Forest Restoration Research Unit). 2006. How to plant a forest: the
for growing F. glomerata, F. lutea, F. natalensis, F. pertusa, principles and practice of restoring tropical forests. Biology Department,
and F. thoningii, should be carried out since they can be Science Faculty, Chiang Mai University, Chiang Mai, Thailand.
established simply by placing them directly in the ground. Goosem, S. P., and N. I. J. Tucker. 1995. Repairing the rainforest—theory and
practice of rainforest re-establishment in North Queensland’s wet tropics.
Zahawi (2008) reported that after 1 year, survival of F. pertusa
Wet Tropics Management Authority, Cairns, Australia.
“truncheons” was 83%, with high establishment success and Harrison, R. D. 2005. Figs and the diversity of tropical rainforests. Bioscience
rapid growth. However, this technique seems to be suitable for 55:1053–1064.
small-scale restoration plantings. Hartmann, H. T., D. E. Kester, and F. T. Davies. 1990. Plant propagation:
principles and practices. Prentice-Hall International Editions, Englewood
Cliffs, New Jersey.
Implications for Practice Kantarli, M. 1993. Vegetative propagation of Hopea odorata by cuttings: a low-
• In the case of Ficus, the use of planting stock, raised cost technology. Technical Publications No. 16, ASEAN-Canada Forest
Tree Seed Centre Project, Muak-Lek, Saraburi, Thailand.
from seed is simple, inexpensive, and effective. It can
Kibbler, H., M. E. Johnston, and R. R. Williams. 2004. Adventitious root
be applied in broad-scale restoration activities. formation in cuttings of Backhousia citriodora F. Muell - 1. Plant
• There were no differences in the establishment and genotype, juvenility and characteristics of cuttings. Scientia Horticulturae
growth performance of nursery-grown seedlings from 102:133–143.
seed among species studied. So, it is likely that the Knowles, O. H., and J. A. Parrotta. 1995. Amazonian forest restoration: an
techniques described in this article can be used for other innovative system for native species selection based on phenological
data and field performance indices. Commonwealth Forestry Review
dioecious Ficus tree species.
74:230–243.
• Further research is needed to prevent pathogenic fungi
Laman, T. G. 1995. Ficus stupenda germination and seedling establishment in
that cause damping-off diseases, particularly at the early a Bornean rain forest canopy. Ecology 76:2617–2626.
stage of seedling establishment. If successful, direct Lamb, D., and D. Gilmour. 2003. Rehabilitation and restoration of degraded
seeding could become more cost-effective. forests. IUCN, Gland, Switzerland and Cambridge, UK and WWF, Gland,
Switzerland.
Lambert, F. R., and A. G. Marshall. 1991. Keystone characteristics of bird-
dispersed Ficus in Malaysian lowland rain forest. The Journal of Ecology
Acknowledgments
79:793–809.
We thank D. Blakesley for helpful comments which improved Longman, K. A., and R. H. F. Wilson. 1993. Rooting cuttings of tropical
out manuscript. We also acknowledge Chiang Mai University, trees. Volume 1 of “Tropical trees: propagation and planting manuals.”
the Center of Excellence on Environmental Health, Toxicology Commonwealth Science Council, Westminster, London, U.K.
Manga, V. K., and D. N. Sen. 1995. Influence of seed traits on germination
and Management of Chemicals, and Doi Suthep-Pui National
in Prosopis cineraria (L.) MacBride. Journal of Arid Environments
Park for supporting this study and for permission to conduct 31:371–375.
fieldwork. This work was supported by The Thailand Research Mathew G., B. P. Skaria, and A. Joseph. 2011. Standardization of conventional
Fund through The Royal Golden Jubilee PhD Program (Grant propagation techniques for four medicinal species of genus Ficus Linn.
No. PHD/0033/2550). Indian Journal of Natural Products and Resources 2:88–96.

Restoration Ecology 9
The Use of Fig Trees in Forest Restoration Plantings

Maxwell, J. F., and S. Elliott, 2001. Vegetation and vascular flora of Doi Indonesia. Pages 449–450 in G. L. Gibson, A. R. Griffin, and A. C.
Sutep-Pui National Park, Chiang Mai Province, Thailand. Thai Studies in Matheson, editors. Breeding tropical trees: population structure and
Biodiversity 5. Biodiversity Research and Training program, Bangkok. genetic improvement strategies in clonal and seedling forestry. Oxford
Miyawaki, A. 1993. Restoration of native forests from Japan to Malaysia. Pages Forestry Institute, Oxford, United Kingdom.
5–24 in H. Leith and M. Lohman, editors. Restoration of tropical forest Thornton, I. W. B., S. Cook, J. S. Edwards, R. D. Harrison, C. Schipper, M.
ecosystems. Kluwer Academic Publishers, Dordrecht, The Netherlands. Shanahan, R. Singadan, and R. Yamuna. 2001. Colonization of an
Parrotta, J. A. 1993. Secondary forest regeneration on degraded tropical lands: island volcano, Long Island, Papua New Guinea, and an emergent
the role of plantations as ‘foster ecosystems.’ Pages 63–73 in H. Leith island, Motmot, in its caldera lake. Journal of Biogeography 28:
and M. Lohman, editors. Restoration of tropical forest ecosystems. 1389–1408.
Kluwer Academic Publishers, Dordrecht, The Netherlands. Zahawi, R. A. 2008. Instant trees: using giant vegetative stakes in tropical
Rahman, M. M., M. N. Amin, and M. F. Hossain. 2004. In vitro propagation forest restoration. Forest Ecology and Management 255:3013–3016.
of Banyan tree (Ficus benghalensis L.)—a multipurpose and keystone Zhekun, Z., and M. G. Gilert. 2003. Moraceae. Pages 21-73 in Z. Wu, P. H.
species of Bangladesh. Plant Tissue Culture 14:135–142. Raven and D. Y. Hong, editors. Flora of China. Science Press, Beijing
Smits, W. T. M., I. Yasman, D. Leppe, and M. Noor. 1990. Summary of results and Missouri Botanical Garden Press, St. Louis, Missouri.
concerning vegetative propagation of Dipterocarpaceae in Kalimantan,

10 Restoration Ecology

View publication stats