Bornean forests contain some of the highest levels of vascular plant diversity in the world, and although many plant stems in the plots were identified to species, the inclusion of stems identified only to genus meant that all analyses were done at this higher taxonomic level. Previous work on a variety of taxa has shown that lower taxonomic resolution is appropriate when identification to the species level is not possible or feasible.We used rarefied species accumulation curves to assess the dietary richness of gibbon and leaf monkey diets, both overall and for the frugivorous portion of the diet. We report dietary richness as the observed cumulative number of plant genera consumed given the number of months in which feeding observations were recorded for each primate species. We constructed diet richness curves using the “specaccum” function from the “vegan” package in R 3.0.0 statistical software. We described diet breadth of gibbons and leaf monkeys both in terms of the number plant taxa eaten and the relative proportion of feeding observations on each of five food classes .
We exclude the synconia of Ficus from the category of fruits and treat them separately because their reproductive parts are not true fruits, they are important foods for many vertebrates , square pot and they have unusual phenological behavior rendering them qualitatively distinct from other fruiting plants considered herein.Our study focused on the fruit portion of the diets, because fruits are expected to be the most limiting class of resources on which these species feed. Forest productivity was defined for gibbons as the proportion of stems bearing fruit that were mature or ripe; for leaf monkeys stems bearing immature and mature fruits were also included, reflecting the fact that gibbons avoid unripe fruits whereas leaf monkeys consume them frequently and tend to avoid ripe fruits because foods rich in starch or sugar can disrupt the fore stomach pH and cause acidosis. Despite this fact, we included ripe fruits in the estimate of food availability for leaf monkeys as our phenology categories were based on the most advanced stage, meaning that trees with one ripe fruit were scored as ripe, even if most of the fruits on the tree were still mature, so plants scored as ripe often still contained some food for leaf monkeys. We followed the Design I Protocol for calculating selectivity ratios, the appropriate approach when animals are not individually identified, the availability of a given resource is known, and resource use is sampled randomly. We calculated Manly Selectivity Ratios using the “widesI” function in the “adehabitat” package in R.
Genus use is simply the total number of independent feeding observations recorded on a genus, i. We followed the general convention for MSRs and calculated availability as the total number of stems of genus i that were observed to fruit during the study period. We calculated selection ratios for each genus that was observed to have been fed upon at least once by either gibbons or leaf monkeys. We report MSRs that are standardized so that they add to 1; these values can be interpreted as the probability that for a selection event the primate would choose the genus of interest over other available genera. Values close to zero indicate “avoidance”, meaning the genus was eaten less than would be expected based on its availability. Thus, in the MSR context, the term “avoidance” of an item does not necessarily indicate the item is never consumed . Large values indicate “preference” wherein genera were selected more than predicted based on availability. We conducted a chi-squared test of the null hypothesis that the animals were randomly feeding . The chi-squared test was significant for both animals, so we computed 95% confidence intervals for proportions of used and available resources. If there are fewer than five resource units per category , the corresponding confidence intervals should be interpreted with caution. Relatively few of the genera in our study had five or more feeding observations , so we caution against strong interpretations based on the calculated confidence intervals of the selectivity values.We calculated the proportion of the diet comprising each of the five food classes in three month blocks. We combined data into three-month periods in order to increase sample size per period and thereby improve the reliability of estimates. We then compared the number of feeding observations for each food class in each three-month block to the corresponding average fruit availability values during the same period.
Availability values were calculated as the proportion of the overall stems in the forest that were bearing fruit. Availability differed for gibbons and leaf monkeys because immature fruits were included in calculations for the latter but not the former . We fit linear models using ordinary least squares regression that predicted use based on the following predictors, calculated for a three-month block: fruit availability , seed availability , flower availability , and fig availability . We also included the following environmental predictors: minimumtemperature , maximum temperature , and rainfall . We then compared each of these models and a null intercept model using biased-corrected Akaike’s Information Criterion using the “AICctab” function in the R package “bbmle”. We obtained estimates of the 95% confidence bands for the regressions by simulation using the R package “rethinking”. Due to random sampling variation, some three-month blocks had observations of leaf monkeys but none for gibbons, so we fit the models using 15 three-month blocks for gibbons and 20 three-month blocks for leaf monkeys.This research complied with all applicable laws of the Republic of Indonesia and the United States of America. Per regulations of the Institutional Animal Care and Use Committee at the University of California-Davis, as our research entailed solely non-invasive observation of wild animals, no formal IACUC review was required. Permission to conduct research at Gunung Palung National Park was kindly granted by the Indonesian Institute of Sciences , the State Ministry of Research, Technology, and Higher Education , the Directorate General for Nature Conservation and the Gunung Palung National Park Bureau .We collected 145 feeding observations from gibbons and 219 feeding observations from leaf monkeys . Mean monthly survey effort across the five forest types sampled was 61.45 km/month . Approximately 41% of feeding observations were recorded on census routes , and the remainder were made during focal follows. Most feeding observations were made in the morning. There was no systematic bias towards males or females in either species, nor any difference in the proportion between them. Our observations were mostly of adults , but the proportion of adults and sub-adults did not differ between species. The sampling period included a mast fruiting event from December 2009 though February 2010, during which plant reproductive output spiked. The proportion of plant stems fruiting in our plots during these three months ranged from 0.08 to 0.15 and averaged over 0.10, blueberries in containers more than three times higher than lowest levels of productivity. This resulted in substantial variation in availability of food for both gibbons and leaf monkeys over the course of our study. Lumping of months to ensure adequate sample sizes for statistical analyses meant that variation in the proportion of fruiting stems in our 3-mo sampling blocks was more muted. For gibbons, the proportion of stems with mature and ripe fruit ranged from 0.026–0.077; for leaf monkeys the proportion of stems with immature, mature and ripe fruit ranged from 0.04–0.11 . The species accumulation curves illustrate that leaf monkey diets had higher overall taxonomic richness than gibbon diets, when considering all food classes .
Neither curve reached an asymptote, reflecting that additional richness would be predicted with additional sampling.Nevertheless, the gibbon curve clearly separates from the leaf monkey curve and begins to flatten earlier, indicating that the differences in dietary diversity between the taxa are not the result of different sample sizes. In contrast to the overall diet, gibbons and leaf monkeys exhibited high similarity in the richness of the frugivorous portion of the diet . Therefore, the higher overall dietary richness of leaf monkey diets is due to food classes other than fruit. Leaf monkey diets were more diverse than gibbon diets both in terms of the total number of taxa fed upon and the distribution of feeding observations across food classes . Leaf monkey diets were composed of 31.1% leaves , 25.6% fruits , 26% seeds , 8.2% figs , 6.4% flowers , and 2.7% unidentified items, whereas gibbon diets were composed of 50% fruits , 20% figs , 9.7% flowers , 9.7% leaves , and 6.9% unidentified items. Leaf monkeys consumed food sources from 69 genera in 37 families while gibbons consumed them from 46 genera in 40 families. Gibbons and leaf monkeys used different food classes from different genera. Leaf monkeys ate 28 genera of leaves, 13 genera of flowers, 32 genera of seeds, and 29 genera of fruits. Gibbons had a narrower dietary breadth, consuming 9 genera of leaves, 9 genera of flowers, 5 genera of seeds, and 31 genera of fruits.Our measure of dietary overlap included the number of genera consumed in common and the relative importance of each item in the diet, producing an index that permits asymmetry in dietary overlap measures. From the gibbon’s perspective, there was higher overall dietary overlap than from the leaf monkey’s perspective . This asymmetry is exemplified in the dietary overlap measures for the food class figs. From the leaf monkey perspective, the fig dietary overlap was 1; from the gibbon’s perspective it was 0.62. This asymmetry reflects differences in the importance of figs for the two primate species. In the fruit portion of the diet, leaf monkeys exhibited higher dietary overlap with gibbons then vice versa , a pattern that was reversed in the leaf portion of the diet, with leaf monkeys exhibiting lower dietary overlap with gibbons than gibbons with leaf monkeys . Leaf monkeys also had lower dietary overlap with gibbons in terms of seeds consumed than gibbons had with leaf monkeys .Fruit availability was in the top model as a predictor of resource use for all classes of foods for both primates, with the exception of leaves and flowers . The responses of gibbons and leaf monkeys to variation in fruit availability were qualitatively similar within food classes. We found that as overall fruit availability increased, the consumption of leaves and figs declined, while the consumption of fruit and seeds increased. For gibbons, the intercept only null model was the top model for leaf consumption , and this was also the case for flowers . For leaf monkeys, the strongest predictor of flower consumption was rainfall . For the other food classes, fruit availability was the only predictor included in the top models. The slope estimate for the model of leaf consumption was negative , meaning as the percentage of trees in the forest with fruit increased there was a decrease in the proportion of leaves in the diet. The proportion of figs in the diet also decreased in the diets of leaf monkeys and gibbons as fruit availability increased. The pattern was reversed in the models of seed consumption, with fruit availability having a positive slope in models for both leaf monkeys and gibbons . We also found a positive effect of fruit availability on fruit consumption for leaf monkeys . But for gibbons, the best predictor of fruit consumption was availability of flowers . The proportion of fruit in the diet for gibbons was generally greater than 50% even during periods of low availability ,while this proportion rarely exceeded 50% in leaf monkey diets and only then during periods of high availability .We report the results of a long-term comparative study of the feeding ecology of two sympatric primates. In the masting Dipterocarp forests of Gunung Palung National Park in Indonesia, leaf monkeys had higher dietary richness and diversity than gibbons, which is likely due to leaf monkeys’ physiological ability to process and digest a broader range of foods. We analyzed resource selection from two perspectives: 1) by genus in the selectivity analyses, and 2) and by food class to investigate how these primates alter their diets in response to variation in fruit availability.