Circadian rhythms in diet and habitat use in red ruffed lemurs (Varecia rubra) and white-fronted brown lemurs (Eulemur fulvus albifrons).код для вставкиСкачать
AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 124:353–363 (2004) Circadian Rhythms in Diet and Habitat Use in Red Ruffed Lemurs (Varecia rubra) and White-Fronted Brown Lemurs (Eulemur fulvus albifrons) Natalie Vasey* Department of Anthropology, Portland State University, Portland, Oregon 97207-0751 KEY WORDS daily rhythms; nutrition; thermoregulation; sex differences; reproduction ABSTRACT Daily variation in niche use among vertebrates is attributed to a variety of factors, including thermoregulatory, reproductive, and nutritional requirements. Lemuriform primates exhibit many behavioral and physiological adaptations related to thermoregulation and sharp, seasonal reproduction, yet they have rarely been subjects of a quantitative analysis of circadian (or daily) rhythms in niche use. In this study, I document daily rhythms in diet and microhabitat use over an annual cycle in two sympatric, frugivorous lemurs, Varecia rubra and Eulemur fulvus albifrons. Data on diet, forest site, and forest height were recorded at 5-min time points on focal animals and divided into three time blocks for analysis (06:00 –10:00 hr, 10:00 –14:00 hr, and 14:00 –18:00 hr). I employed multivariate tests of independence to examine daily rhythms in diet and microhabitat use according to sex, season, and reproductive stage. Throughout the day, V. rubra is frugivorous and dwells in the upper canopy, with notable departures (especially for females) during the hot seasons, gestation, and lactation. E. f. albifrons has heterogeneous daily rhythms of food choice and microhabitat use, particularly across seasons, and both sexes are equally variable. These daily rhythms in diet and microhabitat use appear related to thermoregulatory and nutritional requirements, seasonal food availability and circadian rhythms of plant (and possibly insect) palatability, predator avoidance tactics, and in the case of Varecia, to reproduction. Daily rhythms of food choice in V. rubra support two previously suggested hypotheses explaining why primates consume more nonfruit items late in the day, whereas those of E. f. albifrons are too variable to lend support to these hypotheses. Am J Phys Anthropol 124:353–363, 2004. © 2004 Wiley-Liss, Inc. Many organisms vary their diets, microhabitat use, and activity patterns on a regular schedule throughout the course of the 24-hr light-dark cycle of the earth. These outward manifestations of an endogenous, time-measuring function (i.e., “biological clock”) are referred to as daily, or circadian, rhythms. In mammals, the suprachiasmatic nucleus of the hypothalamus is the pacemaker responsible for daily rhythms in physiology and behavior. However, daily rhythms are not strictly ﬁxed. They respond to environmental stimuli, especially light (e.g., light-entrainment can lengthen or shorten cycle length from 24 hr), they can be phase-shifted (e.g., peaks in periodicity can be altered), they have a mutable genetic basis, and they even respond to social context and domestication (Plyusnina et al., 1991; Akiyama et al., 1998; Zordan et al., 2000; Toma et al., 2000; Suarez et al., 2001). Some daily rhythms are found across a broad array of vertebrate taxa. For example, feeding, coprophagy, locomotor activity, and thermoregulatory mechanisms often show regular daily peaks (Salmo salar (Atlantic salmon), Kadri et al., 1991; Lepus brachyurus (Japanese hares), Hirakawa, 1994). In some cases, these daily peaks differ among age-sex classes (Misgurnus anguillicaudatus (loach ﬁshes), Naruse and Oishi, 1996) and show seasonal periodicity, referred to as circannual rhythms (Agrionemys horsﬁeldi (turtles), Wuntke and Siegmund, 1992; Cricetus cricetus (European hamsters), Wollnik and Schmidt, 1995; Cynops pyrrhogaster (Japanese newts), Nagai and Oishi, 1998; Tiliqua rugosa (Australian sleepy lizards), Firth and Belan, 1998). Reproductive activity can also be subject to circadian rhythms (e.g., Arvicanthis niloticus (Nile grass rats), McElhinny et al., 1997; Cynops pyrrhogaster (Japanese newts), Nagai and Oishi, 1998). Daily rhythms have a more complex basis in some taxa. For exam- © 2004 WILEY-LISS, INC. Grant sponsor: Wenner-Gren Foundation; Grant sponsor: Leakey Foundation; Grant sponsor: National Science Foundation; Grant sponsor: Primate Conservation, Inc.; Grant sponsor: Boise Fund; Grant sponsor: Sigma Xi. *Correspondence to: Natalie Vasey, Department of Anthropology, Portland State University, Portland, OR 97207-0751. E-mail: firstname.lastname@example.org Received 26 September 2002; accepted 6 June 2003. DOI 10.1002/ajpa.10357 Published online 19 November 2003 in Wiley InterScience (www. interscience.wiley.com). 354 N. VASEY ple, foraging peaks in Amblyrhynchus cristatus (Galapagos marine iguanas) appear to be governed by a combination of circadian as well as circatidal rhythms (Wikelski and Hau, 1995), while coprophagy in Octodon degus (the degu), an herbivorous Chilean rodent, is governed by both a circadian rhythm of nocturnal coprophagy and a digestive physiology requiring continuous intake into the gastrointestinal (GI) tract and a constant rate of output from the colon (Kenagy et al., 1999). There is every reason to suspect that wild primates express daily rhythms in some ecological variables, but few such rhythms have been documented. One study of Colobus badius, an anthropoid primate, showed a link between feeding and thermoregulation tied in with seasonal shifts. These African colobine monkeys remained high in the canopy all day in the cold season, but regularly descended lower in the forest canopy to feed in the morning and evening during the hot dry season, thereby avoiding high temperatures and circumventing thermal stress (Clutton-Brock, 1973). A study of two Malagasy lemurs, Lemur catta and Eulemur fulvus rufus, showed that daily activity rhythms were related to behavioral thermoregulation. In cooler regions of Madagascar, both species sunned themselves early in the morning after experiencing cool nighttime temperatures, whereas in warmer regions of Madagascar, both species rarely sunned themselves, and feeding bouts occurred in the earlier, cooler hours of the day in L. catta (Sussman, 1974, 1975). Where the two species were sympatric, resting and feeding peaks differed between them, thereby contributing to niche separation (Sussman, 1974, 1977). Lastly, the two species differed in their daily rhythms of forest strata use, whether in allopatry or sympatry (Sussman, 1974). Lemurs exhibit many behavioral and physiological adaptations related to thermoregulation (e.g., sunning, torpor, and seasonal fattening) and photoperiodically cued seasonal reproduction (Van Horn, 1980; Rasmussen, 1985). Thus it seems highly likely that predictable daily rhythms should form part of their behavioral and ecological proﬁle. Yet daily rhythms have rarely been studied in lemurs (Sussman, 1974, 1975; Freed, 1986). In recent studies examining interspeciﬁc and intraspeciﬁc patterns of niche separation in Varecia rubra and Eulemur fulvus albifrons (two closely related, frugivorous, sympatric lemurs), I found that multiple factors inﬂuenced niche use, including seasonal differences in climate and food availability, thermoregulatory needs, reproductive patterns, nutritional needs, and predator avoidance tactics (Vasey, 2000, 2002; for revised taxonomy, see Groves, 2001; Vasey and Tattersall, 2002). Given this intra-annual variation, daily rhythms in diet and microhabitat use in these two lemurs are unlikely to be homogeneous throughout the day and from season to season, but rather should reﬂect species-speciﬁc thermoregulatory, reproductive, and nutritional requirements in addi- tion to predator-avoidance tactics (niche use hypothesis 1). A more speciﬁc set of predictions can be proposed concerning daily rhythms in diet. It has been a challenge to explain daily rhythms in this niche parameter fully, because its underlying adaptive basis has been difﬁcult to identify. Many wild primates consume more fruit in the early part of the day and diversify their diets in the later part of the day (reviewed in Clutton-Brock, 1977). Leaves, in particular, are fed upon late in the day by many primates (reviewed in Chapman and Chapman, 1991). A variety of hypotheses have been offered to explain this phenomenon, e.g., diet hypothesis 1: feeding on fruit early in the day may provide quick energy after nighttime rest when food energy is depleted (Clutton-Brock, 1977; Raemakers, 1978); and diet hypothesis 2: there may be an advantage to consuming leaves late in the day because, being naturally high in ﬁber and secondary compounds, nutrients in leaves can be slowly extracted during nighttime inactivity (Milton, 1979; Glander, 1982). Hypotheses 1 and 2 concerning diet do not conﬂict with each other, yet Ganzhorn and Wright (1994) observed that neither is sufﬁcient in the case of Aotus, the night monkey, whose leaf-eating peaks late in the day at the start of its nocturnal activity cycle. Their phytochemical study of Malagasy lemur food plants indicated that soluble carbohydrates in leaves increase throughout the day, whereas protein content does not vary much. They therefore suggested that by eating leaves late in the day, primates may be optimizing their energy intake by acquiring proteinaceous leaves while they are also highest in carbohydrate value (diet hypothesis 3). Hence, when to eat leaves may be, in some measure, dictated by circadian rhythms of plant palatability. This hypothesis, unlike earlier ones, takes into account the need for primates to acquire dietary protein. To date, our understanding of daily rhythms in primate diets is based on ﬁeld studies of anthropoid primates (see above references). On what basis might we expect dietary diversiﬁcation late in the day in lemurs? Phytochemical work on lemur food plants indirectly suggests such a pattern (Ganzhorn and Wright, 1994). Direct observations that formed part of a ﬁeld study of V. rubra and E. f. albifrons also suggest that lemurs might share this pattern with other primates (Vasey, 2000). Females of both species simultaneously diversify their diets with more seasonally available low-ﬁber, high-protein plant foods (young leaves, ﬂowers) during the hot seasons and during costly reproductive stages (gestation and lactation), whereas males do not. In addition, V. rubra shows pronounced sex differences in diet during costly reproductive stages, with females acquiring more low-ﬁber, high-protein plant foods than males. E. f. albifrons shows far fewer sex differences in diet (Vasey, 2002). These interspeciﬁc and intraspeciﬁc patterns have been attributed to CIRCADIAN RHYTHMS IN V. RUBRA AND E. F. ALBIFRONS differing energetic investments in reproduction between the two species and between the sexes. Sex differences in diet cannot be attributed to body-size dimorphism or male dominance, because most lemurs are monomorphic in body size (Kappeler, 1991), and none show male dominance (reviewed in Wright, 1999). The relatively pronounced seasonal shifts in diet for V. rubra females and their sharp dietary sex differences function in tandem with their higher energetic investment in reproduction (Vasey, 2000, 2002). Relative to E. f. albifrons, Varecia has a shorter gestation, larger, heavier litters, more concentrated milk, and nonclinging, rapidly growing altricial young that are kept in nests or stashed (Foerg, 1982; Rasmussen, 1985; Pereira et al., 1987; Young et al., 1990; Tilden and Oftedal, 1997; Vasey, unpublished ﬁndings). In their quest for high-protein foods, could the behavior of these lemurs, and of females in particular, be explained by hypotheses 1–3, described above? This would be cost-efﬁcient, especially for a primate with relatively high reproductive investment such as V. rubra. If daily peaks in consumption of high-protein, nonfruit items occur, we would predict such peaks to occur late in the day, to be more pronounced in females (especially Varecia), and to occur during costly reproductive periods (i.e., during gestation and lactation) or when such foods are abundant (diet: prediction 1). This study, in which sex, season, and reproductive state are partitioned, allows a controlled examination of the inﬂuences on daily rhythms of food choice and microhabitat use. METHODS Study site The study site is located in the Masoala National Park in northeastern Madagascar in a region of primary lowland coastal rain forest known locally as Andranobe. Average annual rainfall over the course of the study was 5,110.26 mm. Average monthly temperature maxima ranged from 22.5–31.6°C, and average monthly temperature minima ranged from 19 –23.5°C. There are four distinct seasons: 1) hot rainy (January–March), 2) transitional cold (April– May), 3) cold rainy (June–August), and 4) hot dry (October–December). Assessments of plant phenology from northeastern Madagascar indicate that fruit, ﬂowers, and young leaves are more abundant in the hot seasons, with additional increases in ﬂower and young leaf availability at the end of the cold rainy season (Andrianisa, 1989; Rigamonti, 1993). Increases in ﬂower and young leaf availability therefore occur during the earlier parts of gestation and lactation (see Table 1 in Vasey, 2002). Reproductive stages of V. rubra were highly synchronized among individuals in the study community: individuals mated in July and gave birth at the very end of October (Vasey, 1997a). Reproductive stages of E. f. albifrons females within the study group were more protracted but still synchronous: 355 females gave birth in mid-October and early December (Vasey, 1997a). At 4 months of age, infants of both species were largely weaned (Vasey, unpublished ﬁndings). “Nonreproductive” throughout the text refers to the period of the year when adult females were neither pregnant nor lactating. For both species, births occur when seasonal food availability and diversity are increasing. More extensive descriptions of the study site, region, climate, food availability, and reproductive schedules can be found in Vasey (1997a,b, 2000, 2002). Study population and data collection At Andranobe, V. rubra lives in large multimalemultifemale communities with a ﬁssion-fusion type of social organization, whereas E. f. albifrons lives in small, cohesive multimale-multifemale groups (Vasey, 1997a). Data were collected on adult animals on 5– 8 consecutive days per species per month over 12 consecutive months (January–December 1994), with the help of two ﬁeld assistants who assisted with animal tracking. Focal animal observation periods lasted from 8 –13 hours, depending upon seasonal differences in day length and time needed to locate animals at dawn. V. rubra was observed for 672 hr (5 乆, 463 hr; 3 么, 209 hr) during 78 focal animal observation periods, and E. f. albifrons was observed for 619 hr (4 乆, 410 hr; 2 么, 209 hr) during 64 focal animal observation periods. Observations were made on one community of V. rubra and on one group of E. f. albifrons. To facilitate location of animals at the beginning of each observation period, three V. rubra (2 乆 and 1 么, each from a separate core group within the community) and one E. f. albifrons (么) were ﬁtted with radio collars manufactured by Telonics (Mesa, AZ). An additional E. f. albifrons individual (乆) was ﬁtted with a nylon collar without a transmitter. At 5-min ﬁxed-interval time point samples (Crook and Aldrich-Blake, 1968; Altmann, 1974), I recorded diet and habitat use for the following variables: food item consumed (fruit, ﬂowers, mature leaves, young leaves, and miscellaneous); estimated height of focal animal from the ground in 5-m increments; and forest site (ground, trunk, major branch, crown, liana, or liana within tree crown). Tree height and crown diameter estimates were made visually and were periodically veriﬁed for accuracy, using a clinometer. Data analysis I pooled time-point samples for all focal animals for each species, following recommendations by Leger and Didrichsons (1994). For analysis of daily rhythms in diet, forest sites, and forest heights, frequencies were calculated based on the number of scores for each variable category (e.g., forest site: ground, trunk, crown, or liana) divided by the total number of records for that variable subdivided by sex, season, and time block. For dietary data, I also 356 N. VASEY subdivided time-point data by reproductive stage. For females in the latter analyses, I included data only for those who were reproductively active during the study period (i.e., those who were either gestating or lactating; V. rubra, n ⫽ 4; E. f. albifrons, n ⫽ 2). As E. f. albifrons females showed less reproductive synchrony, data collected during different, but overlapping, sets of months were included in each reproductive stage for each respective female. All time points that fell within the following time blocks were used for analysis regardless of the total length of observation period: 06:00 –10:00 hr (morning), 10: 00 –14:00 hr (midday), and 14:00 –18:00 hr (late afternoon/early evening, i.e., late in the day). Bivariate and multivariate analyses of frequencies were used to examine the extent of association between each variable and their respective number of categories (or states). In effect, these analyses test for signiﬁcant departures from an independent assortment of variables, an expectation of independence being the null hypothesis. I report the Cochran-Mantel-Haenzel (CMH) statistic for analyses of multiway contingency tables, and the Mantel-Haenzel (MH) chi-square statistic for two-way tables (Bishop et al., 1975), using standard notation for signiﬁcance levels (*P ⬍ 0.05; **P ⬍ 0.01; ***P ⬍ 0.001; Sokal and Rohlf, 1981). Statistical tests were run using the SAS System for Windows, release 6.12 (1989 –1996, SAS Institute, Cary, NC). For further details on the study population and on data collection and analysis, see Vasey (2000). RESULTS Daily feeding rhythms V. rubra. V. rubra feeds on fruit more than on any other item in every time block in every season. Furthermore, time spent feeding on each food item is similar from time block to time block on an annual basis, across seasons, within seasons, and between sexes (Table 1, Fig. 1a). The only departure from this overall pattern is that males feed signiﬁcantly more often on ﬂowers late in the day during the hot dry season (31% of time points, n ⫽ 21; Table 1 contains sample sizes, statistics, and signiﬁcance levels for analyses done on combined sex and on divided sex data). As with seasonal data, time spent feeding on each food item is similar from time block to time block across reproductive stages. Importantly however, there are differences within stages and according to sex (Table 2). Late in the day, V. rubra supplements its diet with more ﬂowers and foliage (especially young leaves) during gestation and lactation (Fig. 2a). When these data are divided by sex, males contribute signiﬁcantly to increased ﬂower intake late in the day during gestation (25%, n ⫽ 20). However, only females consume young leaves late in the day during gestation (3%, n ⫽ 8) and lactation (17%, n ⫽ 30; Table 2 contains sample sizes, statistics, and signiﬁcance levels for analyses done on combined sex and on divided sex data). In fact, males were rarely ever recorded eating foliage. E. f. albifrons. For E. f. albifrons, time spent feeding on each food item is similar from time block to time block on an annual basis and within each season. There are departures from independent assortment across seasons by sex, but these departures do not show an increase in nonfruit consumption exclusively late in the day (Table 1, Fig. 1b). Females are responsible for a morning peak in ﬂower feeding in the cold rainy season (19%, n ⫽ 22), and they also consume more ﬂowers late in the day in the hot dry season (44%, n ⫽ 34), whereas males eat more miscellaneous items (mainly faunal material) in the morning during the hot rainy season (30%, n ⫽ 10). Thus, seasonal data show sex-speciﬁc peaks in consumption of nonfruit items both in the morning and late in day, and these peaks occur during seasonal food abundance as well as seasonal food scarcity. As with seasonal data, time spent feeding on each food item in each time block differs across reproductive stages by sex, but there is no uniform pattern of nonfruit consumption late in the day (Table 2; Fig. 2b). During lactation, females eat more young leaves midday (23%, n ⫽ 11) and more ﬂowers late in the day (42%, n ⫽ 14), whereas during periods of female gestation, males eat more ﬂowers late in the day (24%, n ⫽ 18). Daily rhythms of forest site use V. rubra. V. rubra is primarily a crown dweller all day long in every season, particularly during the cold seasons. Signiﬁcant departures from this heavy use of tree crowns occur within the hot seasons according to sex (Table 1, Fig. 1c). At midday in the hot rainy season, females in particular spend more time on major branches (乆, 7.5%, n ⫽ 42 vs. 么, 3.75%, n ⫽ 7), and use more crown lianas (乆, 7%, n ⫽ 41 vs. 么, 1%, n ⫽ 2). In the hot dry season, females spend more time than males on crown lianas in the morning (乆, 13.5%, n ⫽ 57 vs. 么, 2.75%, n ⫽ 6) and on lianas outside of tree crowns late in the day (乆, 3%, n ⫽ 15 vs. 么, 1%, n ⫽ 2). E. f. albifrons. E. f. albifrons is characterized by signiﬁcant heterogeneity in use of forest sites from time block to time block on an annual basis and across seasons according to sex. Unlike Varecia, they also use the ground. Departures from independent assortment are signiﬁcant for one or both sexes within every season (Table 1, Fig. 1d). The annual pattern is due overall to extensive use of crown lianas in the morning and both crown lianas and lianas at midday. Additionally, there is substantial seasonal and intraspeciﬁc heterogeneity, such as the marked use of tree crowns, primarily by females (乆, 92%, n ⫽ 417 vs. 么, 86%, n ⫽ 118), at the expense of crown lianas in the morning during the hot dry season, and the decreased use of crown lianas (乆, 4.5%, n ⫽ 16; 么, 2.75%, n ⫽ 6) and a concomitant — (68) — (78) — (97) 19.68*** (199) 0.53 (398) — (200) — (386) 1.42 (479) 1.26 (466) — (278) — (483) 0.31 (678) 1.87 (2,132) MH 0.55 (1,905) 0.07 (1,905) CMH — (401) — (409) — (516) 0.00 (794) 3.00 (1,235) — (647) 0.61 (1,332) 42.69*** (1,442) 4.80* (1,636) 3.70 (1,056) 0.09 (1,848) 36.91*** (2,236) 6.05* (7,448) MH 7.64* (6,776) 8.10* (6,776) CMH Forest site Varecia rubra 16.47*** (402) 24.41*** (405) 5.30* (516) 4.21* (790) 8.02** (1,240) 0.00 (636) 20.44*** (1,321) 0.99 (1,425) 0.54 (1,642) 17.31*** (1,041) 21.4*** (1,837) 3.22 (2,215) 15.32*** (7,420) MH 4.65 (6,735) 5.13 (6735) CMH Height 5.43* (93) 0.01 (69) 3.63 (149) 0.06 (136) 0.18 (104) 0.13 (125) 10.26*** (272) 4.99* (337) 1.70 (197) 0.15 (194) 1.80 (421) 1.54 (473) 0.88 (1,422) MH CMH 15.23*** (1,285) 15.98*** (1,285) Diet 0.12 (619) 7.45** (458) 19.33*** (666) 0.03 (452) 4.54* (625) 2.39 (657) 15.74*** (1,133) 12.84*** (1,285) 2.50 (1,244) 0.26 (1,115) 31.08*** (1,799) 9.82** (1,737) 19.95*** (6,607) MH 49.11*** (5895) 48.07*** (5895) CMH Forest site Eulemur fulvus albifrons 0.00 (648) 12.58*** (471) 3.73 (670) 7.55** (453) 22.21*** (654) 1.19 (677) 6.10* (1,143) 6.13* (1,279) 89.99*** (5,995) 87.64*** (5,995) CMH Height 14.29*** (1,302) 9.19** (1,148) 9.15** (1,813) 11.75*** (1,732) 6.96** (6,716) MH 1 d.f. ⫽ 2 for season ⫻ block ⫻ niche variable, d.f. ⫽ 2 for season ⫻ sex ⫻ block ⫻ niche variable, and d.f. ⫽ 1 for all other tests. MH, Mantel-Haenzel chi-square statistic; CMH, Cochran-Mantel-Haenzel row mean scores statistic. Dashes indicate that tables were not produced because niche variables were so homogeneous that they resulted in multiple column sum zeros. Sample sizes (n) are indicated in parentheses. * P ⬍ 0.05. ** P ⬍ 0.01. *** P ⬍ 0.001. Hot dry Cold rainy Transitional cold Males Hot rainy Hot dry Cold rainy Transitional cold Females Hot rainy Season ⫻ sex ⫻ block ⫻ niche variable Hot dry Cold rainy Transitional cold Hot rainy Season ⫻ block ⫻ niche variable Block ⫻ niche variable Niche variable Diet TABLE 1. Intraspecies tests of independence between niche variables and time-blocks by season1 358 N. VASEY Fig. 1. Time spent during each time block feeding on various food items (a, b), in various forest sites (c, d), and at different forest heights (e, f), according to season based on combined sex data, for V. rubra (a, c, e) and E. f. albifrons (b, d, f). Time blocks include morning (06:00 –10:00 hr), midday (10:00 –14:00 hr), and late in day (14:00 –18:00 hr). Table 1 contains sample sizes and statistics for analyses based on combined sex data (shown here) and for analyses based on divided sex data. Salient percentages from divided sex analyses are provided in text. CIRCADIAN RHYTHMS IN V. RUBRA AND E. F. ALBIFRONS TABLE 2. Intraspecies tests of independence between diet, time block, and reproductive stage1 Varecia rubra Eulemur fulvus albifrons MH MH Stage ⫻ block ⫻ diet Gestation Lactation Nonreproductive Stage ⫻ sex ⫻ block ⫻ diet Females Gestation Lactation Nonreproductive Males Gestation Lactation Nonreproductive CMH 3.24 (2,046) 4.82* (788) 10.53*** (677) 1.52 (581) CMH 22.64*** (1,014) 0.72 (477) 6.05* (270) 0.51 (267) 22.64*** (2,046) (1,014) 3.77 (286) 3.98* (130) 2.51 (140) 13.41*** (208) 4.34* (178) — (140) 3.84* (191) 2.52 (140) 0.26 (127) 15–20 m). However, there is enormous heterogeneity in how time is divided between strata from time block to time block across seasons, within seasons, and when divided by sex (Table 1, Fig. 1f). This is particularly true in the morning: E. f. albifrons spends the largest portion of its time between the ground and 5 m in the hot rainy season, between 5–10 m in the cold seasons, and between 10 –15 m in the hot dry season. DISCUSSION 3.20 0.25 (580) 17.21*** (499) 1.42 (441) 359 d.f. ⫽ 2 for stage ⫻ block ⫻ diet, d.f. ⫽ 2 for stage ⫻ sex ⫻ block ⫻ diet, and d.f. ⫽ 1 for all other tests. MH, Mantel-Haenzel chi-square statistic; CMH, Cochran-Mantel-Haenzel row mean scores statistic. Dash indicates that table was not produced because niche variables were so homogeneous that they resulted in multiple column sum zeros. Sample sizes (n) are indicated in parentheses. * P ⬍ 0.05. ** P ⬍ 0.01. *** P ⬍ 0.001. 1 increase in the use of tree trunks (乆, 8.75%, n ⫽ 30; 么, 6%, n ⫽ 13) late in the day during the cold seasons (both sexes). Daily vertical ranging rhythms V. rubra. V. rubra spends signiﬁcantly different amounts of time in each forest level from time block to time block, yet the pattern is largely similar across seasons and between the sexes (Table 1, Fig. 1e). In nearly every time block within every season, V. rubra spends most of its time between 15–20 m, the second largest bulk of time between 20 –25 m, and the third largest bulk of time between 10 –15 m. The transitional cold season illustrates the most extreme version of this pattern and the most uneven use of forest levels (e.g., over 80% of time is spent over 15 m in the morning time block). In the opposite extreme, during the hot rainy season, V. rubra uses forest levels more equally in each time block. The amount of time spent below 15 m in the second and third time blocks is markedly greater during the hot rainy season (Fig. 1e). E. f. albifrons. In each time block, E. f. albifrons distributes its time relatively evenly in four 5-m height categories (0 –5 m, 5–10 m, 10 –15 m, and Sex differences and daily rhythms in microhabitat use As predicted, daily rhythms in diet and microhabitat use in V. rubra and E. f. albifrons are not homogeneous throughout the day and from season to season (niche use hypothesis 1). In terms of microhabitat use, daily rhythms appear linked to species-speciﬁc thermoregulatory requirements, predator-avoidance tactics, and in the case of Varecia, to reproduction. V. rubra carries out most of its activities in the crowns of high-canopy trees regardless of time of day or season (Vasey, 2000, 2002), and this use of high crowns is particularly evident during the morning in the transitional cold season. During the cold seasons the sun appears brieﬂy throughout the day amid torrential rain and fast-moving clouds, and V. rubra seizes these brief opportunities to sun itself in open areas in the canopy. Sun bathing occurs exclusively in the exposed upper canopy. The few departures from this upper canopy usepattern shown in Results are linked to other thermoregulatory needs and to reproductive and predator-avoidance strategies. From midday on in the hot rainy season, V. rubra descends to lower canopy levels. These descents illustrate attempts to remain cool during a season and time of day when ambient temperature is highest (mean seasonal maximum, 30.25°C). V. rubra rests, drinks from tree holes, and travels at these lower levels, and as a consequence, uses major branches more often. Another departure occurs during the hot dry season, when females use lianas in the crowns of trees more often than males in the morning and late in the day. This use of crown lianas corresponds to the long periods of time females spend in nests after infants are born (November) and in stashing depots thereafter (November– January), keeping their young warm and nourished. After approximately 2 weeks of age, infants are moved to concealed, protected, and supportive spots in the canopy created by liana tangles or other foliage. Infants are difﬁcult to see and access in such spots, and are effectively protected against predators and the elements (e.g., heat loss) when left unattended (Vasey, 1997a, 2000, unpublished ﬁndings). E. f. albifrons carries out its maintenance activities in a wide variety of forest sites and vertical strata, showing heterogeneity throughout the day and from season to season. Males and females are 360 N. VASEY Fig. 2. Time spent during each time block feeding on various food items according to reproductive stage based on combined sex data. a: V. rubra. b: E. f. albifrons. Time blocks include morning (06:00 –10:00 hr), midday (10:00 –14:00 hr), and late in day (14:00 –18:00 hr). Table 2 contains sample sizes and statistics for analyses based on combined sex data (shown in this ﬁgure) and for analyses based on divided sex data. Salient percentages from divided sex analyses are provided in text. equally variable. This pattern of microhabitat use demonstrates how E. f. albifrons uses its relatively small home range very thoroughly for a wide variety of food resources, resting locations, and travel substrates. As with V. rubra, their microhabitat niche seems well-designed to avoid predators and to meet thermoregulatory (but not reproductive) requirements. On a regular basis, animals enter crown lianas and lianas for their midday rest period where they are well-camouﬂaged, tucked out of sight from both aerial and ground predators. These sites often have very restricted arboreal access, requiring the lemurs to enter and exit the spot single-ﬁle using the single arboreal pathway available. Resting in dense lianas also protects them from high ambient temperatures (ranging as high as 40.5°C), the cold, the rain, and the wind. In addition, E. f. albifrons is a “social thermoregulator;” animals huddle together and wrap their tails around each other, a behavior never observed among V. rubra adults (see also Morland, 1993, and Discussion below of nesting in Varecia). During the one relatively dry season, E. f. albifrons spends more time in open tree crowns in the morning. Sex differences and daily rhythms in diet Daily rhythms in diet for E. f. albifrons do not show clear support for the prediction that daily peaks in consumption of high-protein, nonfruit items would occur late in the day, be more pronounced in females, and occur during costly reproductive periods or when such foods are abundant (diet: prediction 1). Nor do they show any clear pattern. In E. f. albifrons, both males and females show peaks in consumption of high-protein, nonfruit items, and these peaks do not uniformly occur during costly reproductive phases or when such foods are abundant. Furthermore, such peaks occur in the morning, at midday, and late in the day. Thus, E. f. albifrons has a relatively diverse diet throughout the day, and various daily rhythms in food choice are evident. Because E. f. albifrons males alone show a morning peak in consumption of miscellaneous items (mainly insect material) in the hot rainy season, this could indicate that males have a propensity for seeking out mobile sources of protein which are more costly to obtain and process (see also Vasey, 2000; Rose, 1994), and/or that these insects are more easily digested early in the day. Millipedes, in particular, require extensive preparation time once captured. Even once they are rolled in the hands and salivated on, presumably to reduce toxicity, they often still escape uneaten. On the other hand, sexspeciﬁc peaks in daily rhythms of nonfruit consumption (including the apparent penchant of E. f. albifrons males for insects) may simply be due to sampling a species with a highly diverse diet. Daily rhythms of food choice in E. f. albifrons may in fact reﬂect multiple inﬂuences in combination: seasonal patterns of food availability, dietary heterogeneity, and sex differences (see also Vasey, 2002), in addition to circadian rhythms of plant or insect palatability. In contrast to E. f. albifrons, daily rhythms in diet for V. rubra do support prediction 1 concerning diet, showing a clearly interpretable pattern. High-protein, nonfruit items are fed on late in the day during costly reproductive phases and when such foods are abundant. Quantitative analyses presented here indicate that both sexes show this pattern, with males increasing intake of ﬂowers late in the day during gestation and the hot dry season (Tables 1 and 2), and females feeding more often on young leaves late in the day during lactation (Table 2). These increases correspond to peaks in ﬂower and young leaf availability during the earlier parts of gestation and CIRCADIAN RHYTHMS IN V. RUBRA AND E. F. ALBIFRONS lactation (see Methods). Although the number of males under observation (3) was smaller than the number of females (5), the results for males appear representative, as the frequency distributions for each male are similar and a large number of time point records were employed in analyses (Tables 1 and 2). Young leaves are the most easily digested source of plant protein in the forest due to their high protein-to-ﬁber ratio, while ﬂowers contain water, protein (in pollen), and simple sugars (Richard, 1985). High-protein foods are critical for milk production and to replace lost energy reserves. Protein requirements of pregnant and lactating primates are estimated to increase 20 – 46% above baseline levels, and perhaps more in those (like Varecia) that produce rich milk (Oftedal, 1991). However, while nutritional requirements of gestation and lactation adequately explain sex differences in the diet of V. rubra (see also Vasey, 2002), they do not, by themselves, explain the daily rhythm of leaf-eating in females. In this regard, my case study from Andranobe (see below) suggests prominent roles for digestive physiology (diet hypothesis 2), circadian rhythms of plant palatability (diet hypothesis 3), and yet a third factor: the thermoregulatory needs of Varecia infants. All factors indicate that an energy conservation strategy is at work. High-protein foods generally require more time for nutrient extraction; digestion, absorption, transport, and protein synthesis are parts of a lengthy process of creating milk from leaves (and other plant foods). V. rubra females may increase intake of high-protein foods late in the day, which take longer to digest (diet hypothesis 2), to tide them through the night while they are ﬁrst starting to lactate and cannot leave their nestlings unattended, susceptible to cold and vulnerable to nocturnal predators. Only mothers were observed nesting infants, so other group members were not a source of heat to newborns (see also Morland, 1993, p. 200). These females may also be optimizing their energy intake by ingesting leaves at the time of day when they are highest in carbohydrate value (diet hypothesis 3). At Andranobe, lactating females began feeding extensively on young leaves immediately after giving birth, while newborns were still in the nest. Mothers left their nestlings for rapid foraging bouts several times a day during their ﬁrst week of life (Vasey, unpublished ﬁndings). The last feeding bout of the day occurred shortly before dusk and ended with an extensive feed on young leaves growing close to their nests. Often these were immature liana leaves growing though the canopy of the very same trees in which they built their nests. V. rubra give birth during the hottest, driest time of the year (October–November), which should ensure that altricial V. rubra infants do not undergo thermal stress while mothers are away feeding during the day. However, nighttime temperatures fall regularly (Vasey, 2000), and seasonal weather patterns 361 are not always predictable. In 1994, infants were born and nested during an unseasonably cold and rainy week (for climatic data from 1992–1996, see Vasey, 1997a). One mother showed gross physical signs of reproductive stress: she became emaciated and lost much of her lustrous coat while nesting her newborn twins. She may have used her fur as nesting material, as observed in captivity (Petter-Rousseaux, 1964), though it remains unclear whether this behavior was an artifact of poor captive management. Regardless of how much climatic conditions ﬂuctuate from birth season to birth season, this case study illustrates that daily rhythms of food choice coalesced with digestive, thermoregulatory, and nutritional requirements, as well as with seasonal peaks in availability of protein-rich plant foods, to ensure infant survivorship at Andranobe in 1994. All three hypotheses concerning diet in large measure assume that daily rhythms in diet are dictated by choice. However, daily variation in diet may also be governed by shifting food availability during the course of an organism’s active period and food perception capabilities. For example, if fruit feeding peaks in the early part of the day, choosy diurnal frugivores (i.e., many primates) may ﬁnd less palatable fruit available within their home ranges toward the end of the day, spurring a daily dietary rhythm of consuming nonfruit items. In effect, decreasing food availability throughout the day could spur competition for fruit late in the day. This potential factor is probably of negligible importance for V. rubra. Because V. rubra appears to be a predominant frugivore in its biotic community with an exclusive home range relative to conspeciﬁcs (Rigamonti, 1993; Vasey, 1997a), and because females are dominant to males in all contexts in this genus (e.g., social, feeding; Morland, 1991; Kaufman, 1991; Raps and White, 1995), it is likely that any increases in consumption of nonfruit items by females toward the end of the day are due primarily to choice and not to inter- or intragroup competition. In this regard, it is noteworthy that some Varecia females possess trichromatic vision (those that are heterozygous at the X-linked M/L opsin gene locus), a trait known only in one other prosimian, Propithecus verreauxi coquereli (Tan and Li, 1999; Jacobs and Deegan, 2003). These females may have an advantage in the long-range detection of young leaves that ﬂush red against a background of mature green foliage (e.g., Dominy and Lucas, 2001) a factor clearly linked here to their reproductive success. In summary, daily rhythms in diet and microhabitat use appear related to thermoregulatory and nutritional requirements, seasonal food availability and circadian rhythms of plant (and possibly insect) palatability, predator-avoidance tactics, and in the case of Varecia, to reproduction. V. rubra descends into the lower canopy during the hot rainy season when ambient temperature is highest, and females feed on high-protein leaves late in the day during 362 N. VASEY lactation and seek out discrete microhabitats for nesting and stashing infants early and late in the day in the hot seasons. E. f. albifrons has highly heterogeneous daily rhythms of food choice and microhabitat use, although midday resting sites are well-designed for insulation from harsh elements and predators. Daily rhythms of food choice in V. rubra support two general hypotheses explaining why primates consume more nonfruit items late in the day, whereas those of E. f. albifrons are too variable to lend support to any such hypotheses. It was recently demonstrated that vertical stratiﬁcation, food-patch sizes, and forest-site use separate the niches of V. rubra and E. f. albifrons more than do gross dietary categories (Vasey, 2000, 2002). The differing daily rhythms in diet and microhabitat use of these two frugivorous lemurs function as yet another set of niche-partitioning parameters. ACKNOWLEDGMENTS I thank my Ph.D. committee for guidance with various aspects of this work, and particularly Bob Sussman and Alan Templeton for encouraging me to undertake an analysis of lemur circadian rhythms. I also thank Urs Thalmann and Bob Sussman for comments on an earlier version of this paper. I acknowledge the Tripartite Committee of the Malagasy Government for permission to complete this research under an accord between the Department of Paleontology and Biological Anthropology, University of Antananarivo and the Department of Anthropology, Washington University. LITERATURE CITED Akiyama M, Moriya T, Shibata S. 1998. Physiological, pharmacological and molecular aspects of mammalian biological clocks. Folia Pharmacol Jpn 112:243–250. Altmann J. 1974. Observational study of sampling methods. Behavior 49:227–267. Andrianisa JA. 1989. Observation phénologiques dans la Réserve Spéciale de Nosy Mangabe, Maroantsetra. Mémoires de ﬁn d’études, Ecole Supérieure des Sciences Agronomiques, Université d’Antananarivo, Madagascar. Bishop YMM, Fienberg SE, Holland PW. 1975. Discrete multivariate analysis. Cambridge, MA: MIT Press. Chapman CA, Chapman LJ. 1991. The foraging itinerary of spider monkeys: when to eat leaves? Folia Primatol (Basel) 56: 162–166. Clutton-Brock TH. 1973. Feeding levels and feeding sights of red colobus monkeys in the Gombe National Park. Folia Primatol (Basel) 19:368 –379. Clutton-Brock TH. 1977. Some aspects of intraspeciﬁc variation in feeding and ranging behaviour in primates. In: CluttonBrock TH, editor. Primate ecology. New York: Academic Press. p 539 –556. Crook JH, Aldrich-Blake P. 1968. Ecological and behavioral contrasts between sympatric ground dwelling primates in Ethiopia. Folia Primatol (Basel) 8:192–227. Dominy NJ, Lucas PW. 2001. Ecological importance of trichromatic vision to primates. Nature 410:363–366. Firth BT, Belan I. 1998. Daily and seasonal rhythms in selected body temperatures in the Australian lizard Tiliqua rugosa (Scindidae): ﬁeld and laboratory observations. Physiol Zool 71: 303–311. Foerg R. 1982. Reproductive behavior in Varecia variegata. Folia Primatol (Basel) 38:108 –121. Freed BZ. 1996. Co-occurrence among crowned lemurs (Lemur coronatus) and Sanford’s lemurs (Lemur fulvus sanfordi) of Madagascar. Ph.D. dissertation, Washington University, St. Louis. Ganzhorn JU, Wright PC. 1994. Temporal patterns in primate leaf eating: the possible role of leaf chemistry. Folia Primatol (Basel) 63:203–208. Glander KE. 1982. The impact of plant secondary compounds on primate feeding behavior. Yrbk Phys Anthropol 25:1–18. Groves C. 2001. Primate taxonomy. Washington, DC: Smithsonian Institution Press. Hirakawa H. 1994. Coprophagy in the Japanese hare (Lepus brachyurus): reingestion of all the hard and soft faeces during the daytime stay in the form. J Zool Lond 232:447– 456. Jacobs GH, Deegan JF II. 2003. Photopigment polymorphism in prosimians and the origins of primate trichromacy. In: Mollon JD, Pokorny J, Knoblauch K, editors. Normal and defective colour vision. Oxford: Oxford University Press. p 14 –20. Kadri S, Metcalfe NB, Huntingford FA, Thorpe JE. 1991. Daily feeding rhythms in Atlantic salmon in sea cages. Aquaculture 92:219 –224. Kappeler PM. 1991. Patterns of sexual dimorphism in body weight among prosimian primates. Folia Primatol (Basel) 57: 132–146. Kaufman R. 1991. Female dominance in semi-free-ranging blackand-white ruffed lemurs, Varecia variegata variegata. Folia Primatol (Basel) 57:39 – 41. Kenagy GJ, Veloso C, Bozinovic F. 1999. Daily rhythms of food intake and feces reingestion in the degu, an herbivorous Chilean rodent: optimizing digestion through coprophagy. Physiol Biochem Zool 72:78 – 86. Leger DW, Didrichsons IA. 1994. An assessment of data pooling and some alternatives. Anim Behav 48:823– 832. McElhinny TL, Smale L, Holecamp KE. 1997. Patterns of body temperature, activity, and reproductive behavior in a tropical murid rodent, Arvicanthis niloticus. Physiol Behav 62:91–96. Milton K. 1979. Factors inﬂuencing leaf choice by howler monkeys: a test of some hypotheses of food selection by generalist herbivores. Am Nat 114:362–378. Morland HS. 1991. Social organization and ecology of black and white ruffed lemurs (Varecia variegata variegata) in lowland rain forest, Nosy Mangabe, Eastern Madagascar. Ph.D. dissertation, Yale University. Morland HS. 1993. Seasonal behavioral variation and its relationship to thermoregulation in ruffed lemurs (Varecia variegata variegata). In: Kappeler PM, Ganzhorn JU, editors. Lemur social systems and their ecological basis. New York: Plenum. p 193–203. Nagai K, Oishi T. 1998. Behavioral rhythms of the Japanese newts, Cynops pyrrhogaster, under a semi-natural condition. Int J Biometeorol 41:105–112. Naruse M, Oishi T. 1996. Annual and daily activity rhythms of loaches in an irrigation creek and ditches around paddy ﬁelds. Environ Biol Fishes 47:93–99. Oftedal OT. 1991. The nutritional consequences of foraging in primates: the relationship of nutrient intakes to nutrient requirements. Philos Trans R Soc Lond [Biol] 334:161–170. Pereira ME, Klepper A, Simons EL. 1987. Tactics of care for young infants by forest-living ruffed lemurs (Varecia variegata variegata): ground nests, parking and biparental guarding. Am J Primatol 13:129 –144. Petter-Rousseaux A. 1964. Reproductive physiology and behavior of the lemuroidea. In: Buettner-Janusch J, editor. Evolutionary and genetic biology of the primates. New York: Academic Press. p 91–132. Plyusnina IZ, Trut LN, Selina NM. 1991. The daily rhythm of locomotor activity in silver foxes (Vulpes fulva Desm.) and its changes during domestication. Sibirsk Biol Zh 4:11–15. Raemakers J. 1978. Changes through the day in the food choice of wild gibbons. Folia Primatol (Basel) 30:194 –205. Raps S, White FJ. 1995. Female social dominance in semi-free ranging ruffed lemurs (Varecia variegata). Folia Primatol (Basel) 65:163–168. CIRCADIAN RHYTHMS IN V. RUBRA AND E. F. ALBIFRONS Rasmussen DT. 1985. A comparative study of breeding seasonality and litter size in eleven taxa of captive lemurs (Lemur and Varecia). Int J Primatol 6:501–517. Richard A. 1985. Primates in nature. New York: W.H. Freeman and Co. Rigamonti MM. 1993. Home range and diet in red ruffed lemurs (Varecia variegata rubra) on the Masoala Peninsula, Madagascar. In: Kappeler PM, Ganzhorn JU, editors. Lemur social systems and their ecological basis. New York: Plenum. p 25–39. Rose LM. 1994. Sex differences in diet and foraging behavior in white-faced capuchins (Cebus capucinus). Int J Primtol 15:95– 114. Sokal RR, Rohlf FJ. 1981. Biometry. San Francisco: W.H. Freeman. Suarez LP, Wheatley K, Robson F, Onouchi H, Valverde F, Coupland G. 2001. CONSTANS mediates between the circadian clock and the control of ﬂowering in Arabidopsis. Nature 410: 1116 –1120. Sussman RW. 1974. Ecological distinctions in sympatric species of Lemur. In: Martin RD, Doyle GA, Walker AC, editors. Prosimian biology. London: Duckworth. p 75–108. Sussman RW. 1975. A preliminary study of the behavior and ecology of Lemur fulvus rufus AUDEBERT 1800. In: Tattersall I, Sussman RW, editors. Lemur biology. New York: Plenum Press. p 237–258. Sussman RW. 1977. Feeding behavior of Lemur catta and Lemur fulvus. In: Clutton-Brock TH, editor. Primate ecology. New York: Academic Press. p 1–36. Tan Y, Li W-H. 1999. Trichromatic vision in prosimians. Nature 402:36. Tilden CT, Oftedal OT. 1997. Milk composition reﬂects pattern of maternal care in prosimian primates. Am J Primatol 41:195– 211. Toma D, Bloch G, Moore D, Robinson GE. 2000. Changes in period mRNA levels in the brain and division of labor in honey bee colonies. Proc Natl Acad Sci USA 97:6914 – 6919. 363 Van Horn RN. 1980. Seasonal reproductive patterns in primates. In: Reiter RJ, Follet BK, editors. Seasonal reproduction in higher vertebrates. Progress in reproductive biology, volume 5. Basel and New York: Karger. p 181–221. Vasey N. 1997a. Community ecology and behavior of Varecia variegata rubra and Lemur fulvus albifrons on the Masoala Peninsula, Madagascar. Ph.D. dissertation, Washington University, St. Louis. Vasey N. 1997b. How many red ruffed lemurs are left? Int J Primatol 18:207–216. Vasey N. 2000. Niche separation in Varecia variegata rubra and Eulemur fulvus albifrons: I. Interspeciﬁc patterns. Am J Phys Anthropol 112:411– 431. Vasey N. 2002. Niche separation in Varecia variegata rubra and Eulemur fulvus albifrons: II. Intraspeciﬁc patterns. Am J Phys Anthropol 118:169 –183. Vasey N, Tattersall I. 2002. Do ruffed lemurs form a hybrid zone? Distribution and discovery of Varecia, with systematic and conservation implications. Am Mus Novitates 3376:1–26. Wikelski M, Hau M. 1995. Is there an endogenous tidal foraging rhythm in marine iguanas? J Biol Rhythms 10:335–350. Wollnik F, Schmidt B. 1995. Seasonal and daily rhythms of body temperature in the European hamster (Cricetus cricetus) under semi-natural conditions. J Comp Physiol [B] 165:171–182. Wright PC. 1999. Lemur traits and Madagascar ecology: coping with an island environment. Yrbk Phys Anthropol 42:31–72. Wuntke B, Siegmund R. 1992. Preferred temperature and locomotor activity in turtles (Agrionemys horsﬁeldi GRAY), diurnal and seasonal rhythmicity. J Int Cycle Res 23:56 – 64. Young AL, Richard AF, Aiello LC. 1990. Female dominance and maternal investment in strepsirrhine primates. Am Nat 135: 473– 488. Zordan M, Costa R, Macino G, Fukuhara C, Tosini G. 2000. Circadian clocks: what makes them tick? Chronobiol Int 17: 433– 451.