American Journal of Primatology 50:227–246 (2000) RESEARCH ARTICLES Foraging Ecology of the Mountain Monkey (Cercopithecus l’hoesti): Implications for its Evolutionary History and Use of Disturbed Forest B.A. KAPLIN1* AND T.C. MOERMOND2 1 Department of Environmental Studies, Antioch New England Graduate School, Keene, New Hampshire 2 Department of Zoology, University of Wisconsin, Madison, Wisconsin We present the first systematic field study on the feeding ecology of the mountain monkey (Circopithecus l’hoesti), a semi-terrestrial guenon. We compare our results with findings from a concurrent study of blue monkeys (C. mitis doggetti, which have an overlapping home range) conducted over ten months in the Nyungwe Forest Reserve, Rwanda. The mountain monkeys spent 35% of observation time feeding on terrestrial herbaceous vegetation. Thirty-five fruit and seed species comprised 42% of their diet, and invertebrates composed 9%. They spent 38% of observation time on the ground and 27% of observation time in canopy trees. Individuals were observed for equivalent proportions of time foraging on the ground for herbs and in canopy trees for fruits. Although mountain monkeys primarily foraged in undisturbed open areas, they also used disturbed open habitats for this purpose. Synthesizing our results with Pleistocene forest history and C. l’hoesti evolutionary history, we suggest that, although these monkeys readily use disturbed forest for herb foraging, they are a forest-adapted species which has adopted a terrestrial lifestyle to exploit the abundant herb layer common to montane forests. Am. J. Primatol. 50:227–246, 2000. © 2000 Wiley-Liss, Inc. Key words: Cercopithecus l’hoesti; foraging ecology; mountain monkey; Rwanda; terrestrial herbaceous vegetation INTRODUCTION In this paper we explore two foci of primate ecological research in a study of a group of mountain monkeys (Cercopithecus l’hoesti), previously known only from anecdotal accounts. First we examine how primates alter foraging patterns in response to temporal changes in resource abundance [Gautier-Hion, 1980]. Second, we examine how primates use disturbed forest [Johns, 1986; Skorupa, 1986; Contract grant sponsor: USAID Program for Science and Technology Cooperation; Contract grant number: DRH-5542-G-SS-9033-00; Contract grant sponsor: Wildlife Conservation Society; Contract grant sponsor: Davis Fund Fellowships, University of Wisconsin at Madison. *Correspondence to: Beth A. Kaplin, Department of Environmental Studies, Antioch New England Graduate School, 40 Avon St., Keene, NH 03431-3516 Received 19 April 1999; revision accepted 17 December 1999 © 2000 Wiley-Liss, Inc. 228 / Kaplin and Moermond Struhsaker, 1997]. We relate our findings to the relationship between morphology and ecology. We also relate our findings to the evolutionary history and current distribution of the mountain monkey. To strengthen our analyses we use a comparative approach, incorporating results from a concurrent study of one group of sympatric blue monkeys (C. mitis doggetti). The typical forest-dwelling Cercopithecus monkey, or guenon, is considered primarily an arboreal frugivore [Gautier-Hion, 1988a]. The ecology of the more arboreal guenons has been well studied, while that of the mountain monkey is relatively unknown. The mountain monkey spends a considerable amount of time on the ground, leading some to call it semiterrestrial [Gautier-Hion, 1988b] or fully terrestrial [Gebo & Sargis, 1994]. Its limb characteristics have been associated with terrestrial adaptations [Gebo & Sargis, 1994]. An analysis of its dental morphology suggests that it is much more folivorous than other forest guenons [Kay & Hylander, 1978]. It is considered highly susceptible to habitat destruction, and is listed as vulnerable by the IUCN [Lee et al., 1988]. The species has a limited and disjunct distribution, restricted to montane and mature lowland forest [Lernould, 1988]. Surveys conducted in Kibale Forest, Uganda, indicate that it is a mature forest species highly susceptible to disturbance [Skorupa, 1986; Lee et al., 1988; Struhsaker, 1997]. A survey conducted in the Impenetrable Forest, Uganda, found that mountain monkeys also frequent forest edges and secondary growth [Butynski, 1985]. We first describe the foraging ecology of one group of mountain monkeys and a sympatric blue monkey group in relation to use of forest substrates, habitat types, and fruit abundance in the Nyungwe Forest Reserve in Rwanda, central Africa. We used scan sampling methods to sample diet, activity budgets, and habitat use of one habituated group of each species for 10 months. We then synthesized our results with published findings on the dental and morphological adaptations of the mountain monkey, and with Pleistocene forest history, to interpret its current distribution. METHODS Study Site We conducted the research in the Nyungwe Forest Reserve (2°17′–2°50′S and 29°07′–29°26′E), a 950 km2 tropical montane forest in Rwanda, ranging in elevation from 1,600 to 2,900 m. Daily maximum and minimum temperatures and rainfall statistics have been collected at the study site since 1988. Daily temperatures varied little during the study period, with average maximum and minimum temperatures of 19.6°C and 10.9°C, respectively. Average annual rainfall was 1.744 m. The main dry season is July and August, with a short dry period in December and January. The study site was between 2,100–2,700 m in elevation and included dense secondary forest, closed canopy forest, openings of terrestrial herbaceous vegetation, revegetating landslides, and wet valleys, many of which have been mined for gold in the recent past. Behavioral Observations We began habituating one group of mountain monkeys in April, 1990. The group included a single mature male, and females, subadults, juveniles, and infants composed the remainder of the group, which varied between 26 and 29 individuals during the course of the study. We concurrently began habituation of a group of blue monkeys, composed of 29 individuals [Kaplin et al., 1998] whose home range overlapped that of the mountain monkey study group. We present C. l’hoesti Foraging Ecology / 229 data from both groups for February through September, 1991 and July through August, 1992. Methods of sampling diet composition and activity budgets were identical for both species. A trail system was cut following contour lines and drawn onto an enlarged (1:6389) topographic map. Both groups were sampled four to eight days per month from 0600–1800 hours, usually providing twelve hr per day of observations. Only full-day (at least eight hr) observations were used in the analyses. Consecutive partial days were combined to provide a full day of observations when the hours sampled during each partial day covered complementary periods between 0600–1800 hours. We used methods based on scan sampling [Altmann, 1974]. If the interval between samples is short relative to the average duration of the behavior, this method yields an estimate of the time devoted to eating particular food items [Martin & Bateson, 1986]. The activity budgets of each group were observed using scan samples with a duration of five minutes on the hour, quarter-hour, and half-hour; there were thus four activity scans every hour. The first activity (feeding, handling, moving, resting, scanning, or social behavior, including play, aggression, and grooming) of each individual observed was scored. Handling, which refers to foraging, was scored when an individual was handling a food item but was not observed to insert it into the mouth. These scores were combined with feeding scores unless otherwise indicated. We recorded substrate and estimated height of the animal off the ground. Because we were interested in the habitats mountain monkeys used when they were on the ground, we recorded the habitat type used by each scanned individual. The habitats were categorized as: 1) relatively undisturbed areas with terrestrial herbaceous vegetation and no obvious signs of human disturbance; 2) road and trail edges characterized by no trees, an herb layer, and colonizing woody shrub species; and 3) revegetating landslides characterized by no trees, an herb layer, and exposed soil. These classifications did not apply to the blue monkeys, as they rarely descended to the ground. We summarized the data by time of day to detect diurnal patterns of activity budgets and items eaten, using six periods: 0600–0800, 0801–1000, 1001–1200, 1201–1400, 1401–1600, 1601–1830. Monthly variations in activity budgets were determined by pooling scores from each month and determining the proportion of scores devoted to each activity each month. Immediately following each 5-min scan interval for activity budget sampling, we conducted scans to collect detailed information on diet composition (e.g., 0800– 0805, activity scans; 0805–0815, feeding scans; 0815–0820, activity scans). Two different scan sampling methods for diet composition data were used. We began using the “frequency method” [Struhsaker, 1975; Butynski, 1990] with a 10-min time interval every quarter-hour to scan the group and record food items eaten. Individuals were scored more than once if they switched items (plant part or species) within the same scan sample [Kaplin et al., 1998]. This method lacks a systematic procedure for observing individuals and their switches within one scan sample, and assumes that one can observe all switches simultaneously. It also may underestimate common items and overestimate uncommon ones. For the second half of the study we used “instantaneous” 3-min scan samples, in which an individual was scored only once during each scan. Observations of an individual were for only a few seconds, and the first item consumed by each visible individual was scored. Although this method may overestimate common items and underestimate uncommon ones, it is systematic for observing and scoring individuals, and provided a better estimate of feeding duration. Three 3-min scans were conducted every 15 min. 230 / Kaplin and Moermond Items were defined as food only if inserted into the mouth. “Herbs” refers to all terrestrial herbaceous vegetation consumed by the monkeys. We noted the habitat of individual mountain monkeys when they were scored for consumption of terrestrial herbs. These habitat types were identical to the categories described above for the activity scans, with the addition of a fourth type: wet valley bottoms characterized by high herb density and few trees of low stature. Diet composition data were collected from both monkey species groups from February, 1991 through June, 1991 using the 10-min “frequency method.” The 3min “instantaneous” scan sampling method was used July, 1991 to September, 1991, and July, 1992 to August, 1992. We collected data from the blue monkey group using both methods in July and August, 1991, and used an ANOVA test with feeding scores as a dependent variable to determine if the two methods differed significantly [Conover, 1980]. Based on the outcome of this test (Table I), we combined the 10-min and 3-min scan data for each species to analyze diet composition for the study period. Otherwise, diet composition data and activity budget data were analyzed by pooling scores for each sample month (sampling method constant within months) and computing proportions of items from the total, because sample days and scans within days are not independent. For analyses comparing diet or activity budgets between rainy and dry seasons, we divided the data into three categories unless otherwise indicated: rainy months (February, March, April, May, and June, 1991); 1991 dry months (July, August, and September); and 1992 dry months (July and August). Monthly dietary overlap was measured for the mountain and blue monkey by summing shared percentages of specific dietary items [Struhsaker, 1975]. A mean of 5.1 ± 3.3 mountain monkey individuals was observed during each scan. Following each 15-min scan interval (which included one 5-min activity budget scan and one 10-min or three 3-min scans for diet composition), an “estimated center of mass” for the group was located on a topographic map [Cords, 1987]. Nine months of diet composition data and ten months of activity data, including two dry seasons, are presented for the mountain monkey, comprising 574 contact hours with the study group. These data are compared with ten months of data collected concurrently (655 contact hours) on the blue monkey study group, with a mean of 4.2 ± 2.99 blue monkey individuals observed per scan sample. Phenological Data Collection We collected data on phenological patterns of 362 individuals of 25 fruitproducing species within the groups’ home range on a monthly basis beginning in January 1991 [Sun et al., 1996]. Species included those most common in the monkeys’ home ranges; they represented 71% of the fruit and seed species in the mountain monkey diet. Mean phenology sample size was 11 ± 3.87 (range 4 to 22, n = 25) individuals per species. Individuals were reproductively mature and TABLE I. Results of ANOVA Test Source df SS F P value Method Item Month Error 1 4 1 13 0.45 566.00 11.25 86.80 0.07 21.19 1.68 0.7992 0.0001 0.2168 df = degrees of freedom; SS = sum of squares; F = F-statistic. C. l’hoesti Foraging Ecology / 231 were assumed to represent the phenology patterns within the home ranges. Species identifications followed Troupin [1982]. Sampling took place over 2–3 days at the same time each month. Percentages of new leaves, flowering buds, flowers, and fruit in a given tree canopy were estimated and assigned a score from 1 to 4: 0% of the canopy = 0, 1–25% = 1, 26–50% = 2, 51– 75% = 3, and 76–100% = 4. Fruit phenology patterns are presented as the monthly overall density of trees, with a fruiting score of one or more as a representation of fruit abundance within the home range. Density does not take into account differences between species in tree size distribution and intensity of phenological activities, but it does take into account the relative abundance among tree species and improves the accuracy of depicting abundance of plant resources [Chapman et al., 1994]. Plant Abundance and Distribution Plant species distribution and abundance sampling took place during the last two months of the study. We classified and mapped 12 forest types in the study site by walking trails transecting the site at 50–100 m intervals, to record changes in the following variables: dominant canopy species, dominant understory species, dominant herbaceous vegetation, canopy height, elevation, aspect, and topography. The resulting map was digitized, and the area of each forest type was measured using the digitizing software CANVAS™ (Deneba Software, Miami, Florida). We used a random stratified sampling method, placing five to seven 0.1 ha plots in each forest type we identified. A total of 52 plots were sampled, covering an area of 5.2 ha, and 38 tree species (dbh ≥ 10 cm) were sampled in these plots. For all trees greater than 10 cm dbh we recorded species, dbh, and an estimate of height in meters. The density of each species found in the study site was calculated from the sum of the tree density in each forest type, multiplied by the proportion of each forest type in the study site. RESULTS Overall Diet Data from the activity scan samples indicate that the mountain monkeys spent an average of 45% (± 8.86, n = 10,990 scores) of their awake time engaged in feeding (Table II; blue monkey n = 12,972 scores). A total of 83 different plant species from at least 47 families was used by the mountain monkeys for food, and 104 different items were consumed. Many epiphytic orchids and ferns which the monkeys were observed to eat were not identified as to species due to the difficulty of collection. This created an underestimate of the total number of different food items eaten during the study period. Table III provides a list of the plant foods which mountain monkeys were observed to consume during the study. Based on the feeding scan samples, terrestrial herbaceous vegetation was TABLE II. Activity Budgets of Mountain and Blue Monkeys* Mountain monkey Blue monkey *Mean proportion ± SD. Feed Handle Move Scan Rest Social behaviors 45.1 (± 8.9) 44.6 (± 9.3) 2.5 (+ 1.6) 4.6 (± 2.4) 22.7 (+ 6.9) 20.4 (± 10.8) 5.8 (± 3.2) 3.5 (± 0.9) 13.6 (± 2.4) 15.5 (± 5.8) 10.4 (± 2.9) 11.4 (± 8.7) 232 / Kaplin and Moermond TABLE III. Plant Food Items in the Diet of the Mountain Monkey Family Species Life form Part eaten Acanthaceae Mimulopsis arborescens Mimulopsis sp. Rauvolfia sp. Ilex mitis Schefflera goetzenii Dryopteris sp. Asplenium elliotii Mikania cordata Mikania sp. Vernonia lasiopus Vernonia or Volkensia sp. Impatiens gesneroidea Impatiens mildbraedii Impatiens purpureo-violacea Basela alba Begonia meyeri-johannis Harungana montana Commelina benghalensis Ipomea involucrata Coccinea mildbraedii Cuscuta kilimanjari Cyathea manniana Cyperus atroviridis Kyllinga sp. Philippia sp. Blaeria kivuensis Alchornea hirtella Macaranga neomildbraediana Neoboutonia macrocalyx Dasylepsis racemosa Salacia erecta Isodon ramosissimus Plectranthus sylvestris Beilschmiedia troupinii Ocotea michelsonii Hibiscus sp. Acacia sp. Xymalos monospora Ficus oreodryadum Myrianthus holstii Embelia libeniana Embelia schimperi Maesa lanceolata Rapanea melanophloeios Syzyguim parvifolium Ochna afzelii Strombosia scheffleri Olea welwitschii Olinia rochetiana Cynorkis symoensii Disa stairsii orchid spp. Shrub Shrub Shrub Tree Vine Terr. fern Terr. fern Vine Vine Shrub Shrub Herb Herb Herb Herb Herb Tree Herb Herb Herb Parastic herb Tree fern Herb Herb Shrub Shrub Shrub Tree Tree Shrub Vine Herb Herb Tree Tree Shrub Tree Treelet Strangler Treelet Vine Vine Treelet Treelet Tree Treelet Tree Tree Tree Herb Herb Epiphyte Flower buds Tendrils, leaves Flowers, fruits Fruit, leaves Leaves, pith Leaflets Leaflets Stalks Tendrils, leaves Pith Pith Flowers Flowers Flowers Leaves Flowers, shoots Fruit Leaves Flowers Petioles, leaves Stems Fiddleheads Leaves Leaves Leaves Leaves Leaves, seeds, pith Seeds, pith Pith Young leaves Fruit Stalks, leaves Stalks, leaves Fruit, seeds Fruit Flowers Seeds Pith Fruit Fruit Fruit, tendrils Fruit Fruit Fruit Fruit Fruit Seeds Fruit Fruit Flowers, stalk Flowers Leaves Apocynaceae Aquafoliaceae Araliaceae Aspidiaceae Aspleniaceae Asteraceae Balsaminaceae Basellaceae Begoniaceae Clusiaceae Commelinaceae Convulvulaceae Curcibataceae Cuscutaceae Cyatheaceae Cyperaceae Ericaceae Euphorbiaceae Flacourtiaceae Hippocrateaceae Lamiaceae Lauraceae Malvaceae Mimosaceae Monimiaceae Moraceae Myrsinaceae Myrtaceae Ochnaceae Olacaceae Oleaceae Oliniaceae Orchidaceae (continued) C. l’hoesti Foraging Ecology / 233 TABLE III. (continued) Family Species Life form Part eaten Poaceae Isachne mauritiana Panicum brevifolium Panicum calvum Polygonum setulosum Polygonum nepalense Clematis hirsuta Rubus sp. Canthium sp. a Canthium sp. b Chassalia subochreata Galiniera coffeoides Oxyanthus troupinii Pavetta pierlotti Pentas sp. Pseudosabicea arborea Psychotria mahonii Rytigynia sp. Sericanthe leonardii unknown sp. Chrysophyllum rwandense Lindernia subracemosa Balthasarea schliebenii Ficalhoa laurifolia Stapfiela sp. Urera cameroonensis Clerodendrum sp. Cyphostemma sp. Herb Herb Herb Herb Herb Vine Shrub Vine Treelet Shrub Shrub Shrub Shrub Shrub Shrub/vine Treelet Shrub Treelet Treelet Tree Herb Tree Tree Shrub Vine Vine Herb Flowers Leaf blades Leaf blades Leaves Leaves Tendrils Fruit, leaves, pith Fruit Fruit Fruit Fruit, pith Fruit Leaves Leaves Buds Fruit Leaves, pith Fruit Fruit, pith, flowers Fruit Leaves Fruit, flowers Fruit Flowers Petiole, pith Pith Tendrils Polygonaceae Ranunculaceae Rosaceae Rubiaceae Sapotaceae Scrophulariaceae Theaceae Turneraceae Urticaceae Verbenaceae Vitaceae – = 35.17% the most frequently consumed food in the diet of the mountain monkey x ± 10.19, range 23.34–51.78, n = 9 mo), followed by the fruit category and then the seed category (Table IV). The bulk (3.2%) of the “other” category (Table IV) was comprised of tree leaves, with the young leaves of the canopy tree Ilex mitis and of the understory shrub Rubus sp. predominating. The second-highest component (1.8%) of the “other” category was fungi. This category also included pith from 15 different species of canopy trees, understory shrubs, and vines; tree leaf galls; lichens; dead wood; bird eggs; and sap. The mountain monkeys ate a total of 31 fruit species and 4 seed species; the majority of fruits was taken from canopy tree species (Table V). Although the mountain monkeys consumed a variety of different food items during the course of the study, a few items made up a large proportion of the diet. The top three food items in the diet, terrestrial herbs (of many different species), seeds of Macaranga neomildbraediana, and seeds of Alchornea hirtella (both seeds from the family Euphorbiaceae), together comprised 50% of the diet (Table VI). The blue monkey diet consisted of greater proportions of fruit, invertebrates, and tree leaves (Table IV). The “other” category was composed of leaves from lianas and epiphytic orchids, as well as fungi and pith. Temporal Variation in Diet There were diurnal fluctuations in the diet of both the mountain and the blue monkey (Table VII). Terrestrial herb consumption increased from the early 234 / Kaplin and Moermond TABLE IV. Diet Composition of the Mountain Monkey and the Blue Monkey Item Fruit Seeds Invertebrates Flowers Terrestrial herbs Tree leaves Other Mountain monkey mean % (+ SD) Blue monkey mean % (+ SD) 24.5 (9.8) 17.7 (11.7) 8.8 (8.5) 4.0 (3.3) 35.2 (10.2) — 9.8 (8.0) 47.4 (15.8) 9.3 (12.4) 24.9 (9.3) 6.2 (8.9) — 6.2 (6.5) 6.2 (6.1) morning and remained high until the mountain monkeys climbed into their sleeping trees at approximately 1800 hours. Changes in the proportion of herbs consumed during the day were significant (Kruskal-Wallis, H = 17.72, P = 0.003, df = 5). We found no difference in the proportion of fruit consumed in the mountain monkey diet during the time periods studied (Kruskal-Wallis, H = 7.39, P > 0.05, df = 5). There were also fluctuations in diurnal activity budgets (Fig. 1A and B). The mountain monkeys showed an increase in feeding activity as the day progressed, while moving decreased. There was no peak in resting activity, which remained constant throughout the day (Fig. 1A). The blue monkeys showed a bimodal pattern in feeding, with peaks in the morning and late afternoon, while resting and associated behaviors such as grooming peaked in the middle of the day (Fig. 1B). The proportion of time that mountain monkeys spent in different activities varied among months (Fig. 2). Feeding varied significantly between the rainy (February through June) and dry seasons; the proportion of time spent feeding was highest during the dry seasons (Kruskal-Wallis, H = 6.295, P = 0.043, df = 2). Moving also varied significantly between the rainy and dry seasons (Kruskal-Wallis, H = 7.182, P = 0.028, df = 2), and was reduced during the 1992 dry season. TABLE V. Parameters of the Top Ten Fruit Species and the Seed Species in the Diet of the Mountain Monkey Life form Proportion in diet Density Fruit species (25 species total) Beilschmiedia troupinii Ficus oreodryadum Chrysophyllum rwandense Balthasarea schliebenii Chassalia subochreata Psychotria mahonii Maesa lanceolata Vine, unknown species Syzyguim parvifolium Embellia sp. Canopy tree Strangler Canopy tree Canopy tree Understory shrub Colonizing tree Colonizing tree Vine Canopy tree Vine 10.0 7.6 7.6 6.9 6.1 5.3 4.7 1.7 1.3 1.1 8.73 8.39 4.76 4.81 — 19.57 5.71 — 30.26 — Seed species (4 species total) Macaranga neomildbraediana Alchornea hirtella Beilschmiedia troupinii Strombosia scheffleri Colonizing tree Shrub Canopy tree Canopy tree 23.1 13.6 5.2 0.2 42.93 — 8.73 7.04 C. l’hoesti Foraging Ecology / 235 TABLE VI. Cumulative Frequencies of the Top Ten Food Items in the Diet of the Mountain Monkeys Item Terrestrial herbs Macaranga neomildbraediana seeds Alchornea hirtella seeds Beilschmiedia troupinii fruit Invertebrates from B. troupinii leaves Balthasarea schliebenii flowers Ficus oreodryadum fruit Chrysophyllum rwandense fruit Balthasarea schliebenii fruit Chassalia subochreata fruit Frequency in diet Cumulative frequency 33.4 10.4 6.1 4.5 4.2 3.6 3.4 3.4 3.1 2.8 33.4 43.8 49.9 54.4 58.7 62.3 65.7 69.1 72.3 75.0 The density of fruiting trees in the monkeys’ home range varied during the course of the study (Fig. 3). The dry seasons sampled during the study had different levels of fruit availability: fruiting activity was relatively high during the 1991 dry season (July, August, and September) and lower during the 1992 dry season sample. Fleshy fruit consumption by the mountain monkeys was greatest during the 1991 dry season, with a slightly significant difference between seasons (Kruskal-Wallis, H = 5.50, P = 0.064, df = 2) (Fig. 4). Although fruit consumption was lower during periods of reduced fruit abundance (Fig. 4), fruit consumption and phenology, expressed as the density of trees in fruit each month, were not correlated (rs = 0.350, P > 0.10, n = 9). In the 1992 dry season sample the density of fruiting trees was relatively lower than in the 1991 dry season, and seeds comprised a significantly greater proportion of the mountain monkeys’ diet than in other months of the study (Kruskal-Wallis, H = 5.44, P = 0.066, df = 2) (Fig. 4). In July 1992 the proportion of seeds (35.8%) to herbs (38.8%) in the diet was fairly equivalent. We found no significant difference in the proportion of time spent feeding on terrestrial herbs between wet and dry seasons (range 23.3–51.8%; Mann-Whitney U = 18, P = 0.50) (Fig. 4). Herb consumption was not correlated with rainfall (rs = 0.55, p > 0.1, n = 9), and we found no significant correlation between herb consumption and fruit density (rs = 0.133, P > 0.10, n = 9). Herb consumption was negatively correlated during the course of the study, with both fruit consumption (rs = –0.733, P < 0.05, n = 9) and invertebrate consumption (rs = –0.717, P < 0.05, n = 9). Furthermore, fruit and invertebrate consumption were positively correlated (rs = 0.733, P < 0.05, n = 9). During the dry seasons, invertebrates comprised a larger proportion of the mountain monkeys’ diet than during other months of the study (Mann-Whitney U = 0, P = 0.014, df = 1) (Fig. 4). For the entire study period, the mountain monkeys foraged primarily on live leaves of trees for – = 54.38 ± 22.87, n = 10,990 scores). During the majority of the invertebrates (x the dry season months of 1991 and 1992, the mountain monkeys collected the – = 41.49 ± 23.28, range 19.34–62.15%) of the invertebrates that they majority (x ate from leaves of the canopy tree Beilschmiedia troupinii. We assessed monthly the relative contribution of items to the diet, using specific food items and groups of food items when we were unable to identify species. Terrestrial herbs, invertebrates, and fungi were not identified as to species, and thus each was considered a separate food category composed of several species each. Remaining items were considered separately by species and by the part eaten. The top five items, or categories of items, were then ranked each Fruit and seeds mean SD Inverts mean SD Flowers mean SD Leaves mean SD Herbs mean SD Other mean SD 17.1 28.9 33.9 42.8 44.8 41.0 13.3 13.1 13.2 10.0 12.3 16.7 8.5 5.4 6.1 3.6 5.3 4.6 8.7 4.9 5.2 3.5 5.4 4.9 1.3 2.7 3.4 1.5 1.1 2.5 1.6 2.5 3.9 2.5 1.6 2.3 Mountain monkey 0600–0800 55.0 0801–1000 45.9 1001–1200 45.8 1201–1400 38.1 1401–1600 33.7 1601–1830 36.6 16.1 10.9 17.9 9.9 7.6 11.7 9.1 10.1 9.7 7.7 8.2 7.1 13.0 9.5 8.1 5.2 8.4 8.3 5.7 7.3 2.6 2.3 2.1 3.3 6.6 8.5 2.6 2.3 2.3 3.3 4.6 2.4 1.9 5.5 5.9 7.3 4.6 2.3 2.5 7.7 5.9 6.4 Blue monkey 0600–0800 0801–1000 1001–1200 1201–1400 1401–1600 1601–1830 12.0 13.2 15.6 17.9 18.9 15.6 22.7 29.9 27.6 26.6 27.3 25.0 7.9 8.4 12.5 8.8 11.4 8.8 6.3 5.4 3.2 3.9 2.5 3.9 11.4 10.8 5.3 7.4 3.2 4.9 6.7 10.3 10.8 13.3 14.3 13.7 7.4 11.4 9.8 13.9 13.9 18.7 63.0 51.6 55.0 54.7 54.8 55.9 236 / Kaplin and Moermond TABLE VII. Diurnal Fluctuations in the Diets of the Mountain Monkeys and Blue Monkeys as Percentages C. l’hoesti Foraging Ecology / 237 Fig. 1. A. Diurnal pattern of activity budget for the mountain monkey group. Clear bars = feed, solid bars = handle, horizontal lined bars = move, slashed bars = rest, cross-hatched bars = scan, and hatched bars = social behaviors. B. Diurnal pattern of activity budget for the blue monkey group. Clear bars = feed, solid bars = handle, horizontal lined bars = move, slashed bars = rest, cross-hatched bars = scan, and hatched bars = social behaviors. 238 / Kaplin and Moermond Fig. 2. Monthly proportion of mountain monkey group scans devoted to each activity. Clear bars = feed, solid bars = handle, horizontal lined bars = move, slashed bars = rest, and cross-hatched bars = scan. Fig. 3. Monthly rainfall (line) and fruit abundance of 25 species eaten by the mountain and blue monkeys (bars). C. l’hoesti Foraging Ecology / 239 Fig. 4. Monthly diet composition of the mountain monkey group. Clear bars = fruit, solid bars = terrestrial herbs, hatched bars = invertebrates, slashed bars = seeds, horizontal lined bars = flowers, and cross-hatched bars = other. month according to their contribution to the diet (Table VIII). Herbs ranked first in the diet every month except during the dry season of 1992, when seeds of Macaranga neomildbraediana, a pioneer species, were ranked higher. Table VIII shows that no single fruit comprised more than 6.6% of the mountain monkeys’ diet until the period of April, 1991 through September, 1991. In that period the density of fruiting trees was relatively high, and individual fleshy-fruit species (e.g., Ficus oreodryadum and Beilschmiedia troupinii) composed a greater proportion of the diet than at any other time. Percentages of dietary overlap in the plant component of the mountain and blue – = 24.72 ± monkey diets, based on species-specific items, ranged from 16.6 to 37.69 (x 7.64). Mean dietary overlap for each of the three seasons sampled were as follows: wet 1991 overlap, – x = 20.04 ± 5.19; dry 1991 overlap, – x = 24.54 ± 8.65; dry 1992 – overlap, x = 34.35 ± 4.73. Monthly dietary overlap was not correlated with fruit consumption (rs = –0.267, P > 0.1, n = 9 mo) nor with season (rs = 0.579, P > 0.1). Spatial Pattern of Foraging We divided the forest vertically into four layers to evaluate forest strata use by the mountain monkeys: 0 m (ground), >0–2 m, >2–12 m (corresponding to the height range of understory trees based on Troupin [1982] and observation by the present author, B.A. Kaplin), and >12 m. The mountain monkeys spent more time on the ground than at any other height categories in the forest strata (Kruskal-Wallis, H = 25.60, P < 0.001, df = 3). A mean of 38% (± 6.68, n = 10,990 scores across 10 mo) of the activity scan sample scores was from animals on the – = 26.55% ground. The second most frequently used substrate was canopy trees (x – = 13.06% ± 7.87). When not on ± 7.50), followed by species of colonizing trees (x the ground, the mountain monkeys were most frequently observed at heights of Species Feb 91 Mar 91 Apr 91 May 91 Jul 91 Aug 91 Sep 91 Jul 92 Aug 92 Terrestrial herbs Alchornea hirtella, seed Balthasarea schliebenii, flower Ilex mitis, fruit I. mitis, new leaves Ficus oreodryadum, fruit B. schliebenii, fruit Beilschmiedia troupinii, fruit Various fungi Liana species unknown, fruit B. troupinii, seed Invertebrates Maesa lanceolata, fruit Psychotria mahonii, fruit Chrysophyllum rwandense, fruit Macaranga neomildbraediana, seed Chassalia subochreata, fruit 1(31.5) 2(13.0) 3(8.3) 4(6.3) 5(6.3) 1(51.6) 2(7.6) 1(48.8) 2(12.4) 1(38.5) 4(6.3) 5(5.4) 1(27.9) 1(21.3) 5(6.9) 2(23.8) 5(5.4) 1(38.8) 3(3.5) 2(23.3) 5(4.3) *Numbers in parentheses are percent frequencies. a Most from B. troupinii leaves. 4(5.3) 3(6.6) 5(3.6) (20.5) 4(4.4) 3(5.4) 5(4.0) 5(2.4) 4(8.5) 3(7.0) 4(9.0) 4(12.9) 2(15.0)a 2(14.8)a 1(24.4)a 3(9.6) 5(8.6) 3(9.8) 3(16.4) 3(13.6 2(30.3) 4(2.5) 1(33.9) 4(11.2) 240 / Kaplin and Moermond TABLE VIII. Top Five Ranked Foods Consumed Each Month by the Mountain Monkeys* C. l’hoesti Foraging Ecology / 241 – = 28.17% ± 7.35, n = 10) or at heights greater than 12 m between 2 and 12 m (x – (x = 20.23% ± 7.54, n = 10). These differences were slightly significant (Wilcoxon Signed Rank Test, z = –1.89, P = 0.059, n = 10 mo). We found no difference in the amount of time that mountain monkeys spent on the ground between seasons (Wilcoxon Signed Rank Test, z = –1.753, P = 0.08). Feeding by the mountain monkeys took place most frequently on the ground, followed by feeding in the upper levels of the forest canopy (Table IX). The majority of observations of moving animals was of animals on the ground. When activity categories were analyzed by type of substrate used rather than strata, we found that feeding by the mountain monkeys took place in equal proportions on the ground and in species of canopy trees (Fig. 5). Conversely, the blue mon– = 1.8% keys were rarely observed on the ground during activity scan samples (x ± 1.6, n = 10 mo) (Table IX). Blue monkeys were observed to feed most frequently – = 31.2% ± 3.4), and to move most frequently at at heights greater than 12 m (x – = 63.2% ± 16.0). these heights (x The mountain monkeys spent significantly more time foraging than engaging in any other activity when on the ground in open undisturbed habitats (Wilcoxon Signed Rank Test, z = –2.191, P = 0.028, df = 9) (Fig. 6). Foraging was also the primary activity on the ground in revegetating landslides (Wilcoxon Signed Rank Test, z = –2.201, P = 0.028, df = 9). We found no difference in the amounts of time animals spent moving or foraging when they were on trails or road edges (Wilcoxon Signed Rank Test, z = 1.274, P = 0.203, df = 9). Of the terrestrial herbs consumed by the mountain monkeys during the course of the study (4,244 scores), 59% were from undisturbed forest habitats and 41% were from disturbed forest habitats (Wilcoxon Signed Rank Test, z = 2.31, P = 0.02, n = 9 mo). Twenty-six percent of the observations of mountain monkeys on the ground in disturbed habitat were from individuals consuming terrestrial herbs in open areas, such as trail and road edges and landslides devoid of trees, and 15% of these observations were in wet valley bottoms with few or no fruit-producing trees. We used a χ2 test to evaluate the hypothesis that mountain monkeys were randomly foraging for terrestrial herbs in open habitat types in relation to their area within the home range. We found that undisturbed open habitat and landslides, trails, and road edges were used more than expected for feeding on terrestrial herbs, while wet valleys were used less than expected (Table X). Herbs appear to be a determinant in the mountain monkeys’ use of the ground TABLE IX. Proportion of Observations in Each Activity by Forest Strata for Mountain and Blue Monkeys 0m >0–2m >2–12m >12m Mountain monkey Feed Move Rest Scan Social behaviors 16 12.8 3.7 1.7 3.8 5.5 2.3 3 0.7 2.5 12.9 4.9 5.5 2 2.9 13.7 2.5 1.9 1.5 0.7 Blue monkey Feed Move Rest Scan Social behaviors 0.4 0.6 0.2 0.1 1.1 0.8 0.4 0.9 0.2 0.6 14.6 6.3 7.8 1.4 4.8 33.6 13.4 7 2 5.5 242 / Kaplin and Moermond Fig. 5. The proportion of substrates used for each activity category for the mountain monkey group. Clear bars = ground, solid bars = canopy trees, horizontal lined bars = pioneer (secondary) trees, slashed bars = shrubs, cross-hatched bars = vines, and hatched bars = other. in undisturbed forest habitats. Not only did they spend the majority of their time in these habitats foraging, but herbs composed 77.7% of the diet when the animals were in these habitat types. Herbs were also related to the use of disturbed open habitats, where the vast majority of feeding observations were of herb consumption: 89.1% of the diet in road and trail-edge habitats, and 92.2% of the diet when animals were feeding in revegetating landslides. DISCUSSION Terrestrial Herbaceous Vegetation and the Mountain Monkey Diet Terrestrial herbs were the most important category in the mountain monkey diet in this study. Although important in the diets of certain African apes [Rogers et al., 1992; Wrangham et al., 1992], they have not been found to be a major component in the diet of any other guenon studied thus far [Gautier-Hion, 1988a]. The Cercopithecus mitis (blue monkey) group tends to consume more tree foliage than other guenons [Cords, 1986]. Lawes [1991] suggested that increasing foliage in the diet allows blue monkeys to coexist with other guenons. Mountain monkeys are thus not singular in their use of foliage, but they are unique among guenons in their use of terrestrial herbs. Fruit is a source of readily available energy and carbohydrates for frugivores who must obtain protein elsewhere. Protein levels in African forest foliage are generally higher than in fruits, but many plants possess chemical protection from herbivory [Waterman, 1984]. Terrestrial herbs, which in African forests are considered to be widely dispersed and abundant [Rogers et al., 1992], have lower levels of condensed tannins and fiber than mature tree leaves [Waterman, 1984], and more protein than fruits and seeds,. The inverse relationship between inver- C. l’hoesti Foraging Ecology / 243 Fig. 6. Observed activities of the mountain monkey group when on the ground. Clear bars = undisturbed open habitats, solid bars = trail and road-edge areas, and slashed bars = landslides. tebrate consumption and herb consumption, and the lower overall proportion of invertebrates in the mountain monkey diet compared with the blue monkey diet [Kaplin et al., 1998], suggest that terrestrial herbs are a source of protein for the mountain monkey. Further studies are needed to determine the nutritional content and availability of these herbs. Mountain Monkey Diet and Foraging Behavior Although the mountain monkeys consumed more foliage, both the blue and mountain monkeys spent an equivalent proportion of time foraging. Fruit consumption was greatest in the morning and foliage consumption increased as the day progressed, consistent with other primates studied thus far [Gautier-Hion, 1980; Cords, 1987]. As Cercopithecus monkeys tend to live in sympatry [Cords, 1986], diet overlap has been found to be exceptionally high [Gautier-Hion, 1988a]. This increases as the proportion of fruit in the diet increases, so that diets are more similar when fruit availability is highest [Gautier-Hion, 1980]. We found that during a period of fruit scarcity both monkey species turned to seeds [Kaplin TABLE X. Forest Habitat Use by the Mountain Monkey for Terrestrial Herb Consumption Compared to Expected Values and c2 Test Result Forest habitat type Undisturbed Landslides Wet valleys Observed us Expected use χ2 2504 1103 637 4244 523.23 523.23 3197.53 7498.46 642.41 2050.43 10191.3 P < 0.001 244 / Kaplin and Moermond et al., 1998]; diet overlap was negatively correlated with fruit consumption. In times of relative fruit abundance the diets of the two species differed more than those of other sympatric guenons [Gautier-Hion, 1988a]. In periods of fruit scarcity, diet may be more narrowly based as individuals concentrate on a small number of critical food resources [Garber, 1993]. The mountain monkeys spent more than half their time on the ground each month, where foraging and moving were the principal activities. Although Fleagle and Mittermeier [1980] concluded that it is difficult to predict diet from locomotor behavior or vice-versa, we found that previously described terrestrial and anatomical adaptations [dental: Kay & Hylander, 1978; postcranial skeleton: Kingdon, 1988; Gebo & Sargis, 1994] in the mountain monkey correspond with our behavioral and dietary observations. This suggests that this species serves as a model of accordance between morphology and diet. Implications for Mountain Monkey Distribution and Conservation The mountain monkey has a disjunct distribution in two montane forest zones: one in Burundi, Rwanda, Uganda, and the Democratic Republic of Congo (DRC), where it also inhabits lowland forest; and one in Cameroon, Nigeria, and Bioko Island, where C. preussi, a close relative and possible subspecies of C. l’hoesti, is found [Dutrillaux et al., 1988; Lernould, 1988]. These two zones mirror core Pleistocene refugia [Hamilton, 1988; Colyn et al., 1991], which may have been linked in some wet phases [Oates, 1988]. A terrestrial lifestyle allows C. l’hoesti to exploit open patches characteristic of montane forest, where increased light penetration encourages a dense herb layer. Competition from woodland or savanna primates may have limited their distribution beyond the forest. Kingdon [1988] has shown that C. l’hoesti hand and foot morphology represents adaptations for both terrestriality and arboreality. The Cercopithecus species closest to the ancestor compares anatomically with all other arboreal and semiterrestrial guenons [Ruvolo, 1988; Ponsa et al., 1994], but lacks the modifications for terrestriality seen in the C. l’hoesti group [Gebo & Sargis, 1994]. C. lhoesti is not considered primitive [Ruvolo, 1988], and may thus represent a lineage that has moved away from arboreality and readopted a terrestrial lifestyle [Gebo & Sargis, 1994]. Montane forests today are highly threatened [Chapman & Chapman, 1996], and C. l’hoesti is considered vulnerable to extinction [Lee et al., 1988]. Censuses in Kibale Forest, Uganda, found C. l’hoesti to be intolerant of humaninduced forest disturbance [Skorupa, 1986], yet we found that they regularly used disturbed forest. Certain forest-edge or savanna primates, such as Macaca mulatta in Pakistan, are terrestrial herb feeders believed to be adapted to disturbed zones [Goldstein & Richard, 1989; Richard et al., 1989]. Mountain monkeys are distinctly different: as a forest-adapted species [Leakey, 1988], they forage for native terrestrial herbs in disturbed patches that mimic natural forest openings imbedded in a forest matrix. Caution must be used in evaluating primate responses to forest disturbance [Struhsaker, 1997]; long-term studies are needed to determine the effects of this disturbance on survival and reproductive rates. CONCLUSIONS 1. Terrestrial herbs were an important component of the mountain monkey diet and probably provided protein to the diet. C. l’hoesti Foraging Ecology / 245 2. Mountain monkeys spent about 40% of their time on the ground, where foraging for terrestrial herbs and moving were the principal activities. 3. Feeding activity by the group took place most frequently both on the ground and in canopy trees. 4. The mountain monkeys most frequently used undisturbed forest for terrestrial herb consumption, but they selected disturbed patches (landslides and road edges) within the forest to forage for terrestrial herbs. ACKNOWLEDGMENTS We thank the government of Rwanda for allowing us to conduct research in the Nyungwe Forest Reserve. Funding was provided by USAID Science and Technology grant number DHR 5542-G-SS-9033-00, the Wildlife Conservation Society, the Davis Fund Fellowships from University of Wisconsin at Madison (BAK), and a P.E.O. Scholar Award (BAK). The Nyungwe Forest Conservation Project provided logistical support and hospitality. We are indebted to: our field assistants, Gakima Jean-Baptiste, Ngayabahiga Ferdinand, Semahoro Innocent, Ngabonziza Eduard, Gratien Kamarampaka, Gratien Ndiramye, and Manirafasha François; and to C. Sun, K. Kristensen, and J.W. Jordan. P. Arcese, C. Snowdon, K.B. Strier, and M.G. Turner and two anonymous reviewers provided thoughtful comments on an earlier version of this manuscript. This paper is dedicated to our lost colleague, Munyaligoga Viateur, whom we know is out there somewhere. REFERENCES Altmann J. 1974. Observational study of behavior: sampling methods. Behavior 49: 227–267. Butynski TM. 1985. Primates and their conservation in the Impenetrable (Bwindi) Forest, Uganda. Primate Conservation 6:68–72. Butynski TM. 1990. Comparative ecology of blue monkeys (Cercopithecus mitis) in highand low-density subpopulations. Ecol Monogr 60:1–26. Chapman CA, Chapman LJ. 1996. Mid-elevation forests: a history of disturbance and regeneration in East African ecosystems and their conservation. McClanahan TR, Young TP, editors. Oxford: Oxford University Press. p 385–400. Chapman CA, Wrangham R, Chapman LJ. 1994. Indices of habitat-wide fruit abundance in tropical forests. Biotropica 26: 160–171. Colyn MA, Gautier-Hion A, Verheyen W. 1991. A re-appraisal of paleo-environmental history in Central Africa: evidence for a major fluvial refuge in the Zaire Basin. J Biogeography 18:403–407. Conover WJ. 1980. Practical nonparametric statistics. New York: Wiley & Sons. Cords M. 1986. Interspecific and intraspecific variation in diet of two forest guenons, Cercopithecus ascanius and C. mitis. J Anim Ecol 55:811–827. Cords M. 1987. Mixed-species association of Cercopithecus monkeys in the Kakamega Forest, Kenya. Vol. 117. Berkeley: University of California Press. CANVAS, version 3.0. 1992. Miami, Florida: Deneba Software. Dutrillaux B, Muleris M, Couturier J. 1988. Chromosomal evolution of Cercopithecinae. In: Gautier-Hion A, Bourlière F, Gautier JP, Kingdon J, editors. A primate radiation: evolutionary biology of the African guenons. Cambridge: Cambridge University Press. p 150–159. Fleagle JG, Mittermeier RA. 1980. Locomotor behavior, body size, and comparative ecology of seven Suriname monkeys. Am J Phys Anthropol 52:301–314. Garber PA. 1993. Seasonal patterns of diet and range in two species of tamarin monkeys: stability versus variability. Int J Primatol 14:145–166. Gautier-Hion A. 1980. Seasonal variations of diets related to species and sex in a community of Cercopithecus monkeys. J Anim Ecol 49:237–269. Gautier-Hion A. 1988a. The diet and dietary habits of forest guenons. In: Gautier-Hion A, Bourlière F, Gautier J-P, Kingdon J, editors. A primate radiation: evolutionary biology of the African guenons. Cambridge: Cambridge University Press. p 257–283. Gautier-Hion A. 1988b. Polyspecific associations among forest guenons: ecological, behavioral and evolutionary aspects. In: Gautier-Hion A, Bourlière F, Gautier J-P, 246 / Kaplin and Moermond Kingdon J, editors. A primate radiation: evolutionary biology of the African guenons. Cambridge: Cambridge University Press. p 452–476. Gebo DL, Sargis EJ. 1994. Terrestrial adaptations in the postcranial skeletons of guenons. Am J Phys Anthropol 93:341–371. Goldstein SJ, Richard AF. 1989. Ecology of rhesus macaques (Macaca mulatta) in Northwest Pakistan. Int J Primatol 10: 531–567. Hamilton AC. 1988. Guenon evolution and forest history. In: Gautier-Hion A, Bourlière F, Gautier J-P, Kingdon J, editors. A primate radiation: evolutionary biology of the African guenons. Cambridge: Cambridge University Press. p 13–34. Johns AD. 1986. Effects of selective logging on the behavioral ecology of west Malaysian primates. Ecology 67:684–694. Kaplin BA, Munyaligoga V, Moermond TC. 1998. The influence of temporal changes in fruit availability on diet composition and seed handling in blue monkeys (Cercopithecus mitis doggetti). Biotropica 30: 56–71. Kay RF, Hylander WL. 1978. The dental structure of mammalian folivores with special reference to primates and phalangeroidea (Marsupialia). In: Montgomery GG, editor. The ecology of arboreal folivores. Washington: Smithsonian Institute Press. p 46–51. Kingdon J. 1988. Comparative morphology of hands and feet in the genus Cercopithecus. In: Gautier-Hion A, Bourlière F, Gautier J-P, Kingdon J, editors. A primate radiation: evolutionary biology of the African guenons. Cambridge: Cambridge University Press. p 184–193. Lawes MJ. 1991. Diet of samango monkeys (Cercopithecus mitis erythrarchus) in the Cape Vidal dune forest, South Africa. J Zool 224:149–173. Leakey M. 1988. Fossil evidence for the evolution of the guenons. In: Gautier-Hion A, Bourlière F, Gautier J-P, Kingdon J, editors. A primate radiation: evolutionary biology of the African guenons. Cambridge: Cambridge University Press. p 7–12. Lee PC, Thornback J, Bennett EL. 1988. Threatened primates of Africa: the IUCN Red Data Book. Gland, Switzerland. Lernould J-M. 1988. Classification and geographical distribution of guenons: a review. In: Gautier-Hion A, Bourlière F, Gautier JP, Kingdon J, editors. A primate radiation. Cambridge: Cambridge University Press. p 54–77. Martin P, Bateson P. 1986. Measuring Behavior. New York: Cambridge University Press. Oates JF. 1988. The distribution of Cercop- ithecus monkeys in west African forests. In: Gautier-Hion A, Bourlière F, Gautier J-P, Kingdon J, editors. A primate radiation: evolutionary biology of the African guenons. Cambridge: Cambridge University Press. p 79–103. Ponsa M, Egozcue J, Garcia M. 1994. The phylogeny of guenons using PAUP analysis of cytogenetic characters. In: Thierry B, Anderson JR, Roeder JJ, Herrenschmidt N, editors. Current primatology. I. ecology and evolution. Strasbourg: Université Louis Pasteur. Richard AF, Goldstein SJ, Dewar RE. 1989. Weed macaques: the evolutionary implications of macaque feeding ecology. Int J Primatol 10:569–594. Rogers ME, Maisels F, Williamson EA, Tutin CE, Fernandez M. 1992. Nutritional aspects of gorilla food choice in the Lopé Reserve, Gabon. In: Itoigawa I, Sugiyama Y, Sackett GP, Thompson RKR, editors. Topics in primatology. II. Behavior, ecology and conservation. Tokyo: University of Tokyo Press. p 255–266. Ruvolo M. 1988. Genetic evolution in the African guenons. In: Gautier-Hion A, Bourlière F, Gautier J-P, Kingdon J, editors. A primate radiation: evolutionary biology of the African guenons. Cambridge: Cambridge University Press. p 127–139. Skorupa JP. 1986. Responses of rainforest primates to selective logging in Kibale Forest, Uganda: a summary report. In: K. Benirschke, editor. Primates: the road to self-sustaining populations. New York: Springer-Verlag. p 57–70. Struhsaker TT. 1995. The red colobus monkey. Chicago: University of Chicago Press. Struhsaker TT. 1997. Ecology of an African rain forest. Gainesville: University Press of Florida. Sun C, Kaplin BA, Kristensen KA, Munyaligoga V, Mvukiyjmwami J, Kajonda K, Moermond TC. 1996. Tree phenology in a tropical montane forest of Rwanda. Biotropica 28:668–681. Troupin G. 1982. Flores des plantes ligneuses du Rwanda. Vol. Publication no. 21. Butare, Rwanda, Institut National de Recherche Scientifique. Waterman PG. 1984. Food acquisition and processing as a function of plant chemistry. In: Chivers DJ, Wood BA, editors. Food acquisition and processing in primates. New York: Plenum Press. p 177–211. Wrangham RW, Conklin NL, Chapman CA, Hunt KD. 1992. The significance of fibrous foods for Kibale Forest chimpanzees. In: Whiten A, Widdowson EM, editors. Foraging strategies and natural diets of monkeys, apes, and humans. Oxford: Clarendon Press. p 11–18.