Morphological studies on the Sulawesi macaques I: Phyletic analysis of body color

PRIMATES,29(1): 65-80, January 1988 65 Morphological Studies on the Sulawesi Macaques I: Phyletic Analysis of Body Color YUZURU HAMADA, Japan Monkey Centre TSUYOSHI WATANABE, Sugiyama Women's University OSAMU TAKENAKA, Kyoto University BAMBANG SURYOBROTO,Bogor Agricultural University and YOSHI KAWAMOTO, Nagoya University ABSTRACT. The body color of Sulawesi macaques was measured quantitatively and compared among the different monkeys. As a result, divergence models for extant Sulawesi macaques, with tonkeana as the starting point and fading as the sole direction of color change, were inferred as follows: (1) fading slightly on the upper half of the body--nigra, fading more on the proximal part of the body--nigrescens; (2) fading over the whole body--maura; (3) fading greatly on the legs--hecki; and (4) fading on the distal part of the body---ochreata, fading more over the whole body, including the proximal part of the body--brunnescens. The color changed progressively in the order of (1) through (4). The divergence model, excluding the position of hecki (3), supports the speciation model of FOODEN (1969). If the proto-Sulawesi macaques had a body color pattern similar to the living nemestrina, darkening would have been necessary for the evolution of the Sulawesi macaques after their immigration, and it may have been acquired as an adaptation to the ground (forest floor) living nature of the Sulawesi macaques, together with influences deriving from the insularity and/or from the absence of predators. Key Words: Macaca; Sulawesi; Body color; Speciation. INTRODUCTION Sulawesi island, Indonesia, is situated just at the middle of the two border lines of the Asian (Indo-Malay) and Papuan (Australian) faunal regions, that is, so-called "Wallace's" line and the Weber line. It contains many representatives of both faunas and many endemic species (GRovEs, 1976; CRANBROOK, 1981). The island forms a very curious K-shape at the present. Its tectonic history is thought to have been very dynamic. For example, AUDLEY-CHARLES (1981) suggested that the island was a product of collision between its western (Laurasian origin) and eastern halves (Gondwana origin) in the middle of the Miocene (ca. 15 million years ago). The precise sequence of tectonic changes and the history of the migration, isolation, and speciation of the fauna and flora, which may have been strongly influenced by geological events, have not yet been fully elucidated. Sulawesi macaques are known to have a very different morphology from other macaques, and to differ among themselves. Because of the paucity of fossil materials, their history has not yet been determined. Such problems as the nature of the ancestral population, from where and via what land masses it came and how speciation has proceeded, remain unresolved. For example, the candidate site from which the ancestral Sulawesi macaques began their migration has been proposed as Kalimantan (FOODEN, 1969), Java [as a migration route for Asian faunas (GRovEs, 1976; AUDLEY-CHARLES, 1981)] and Philippine-Sanghir 66 Y. HAMADAet al. and Talaud [HILL, 1974; no precise description was given but he appeared to infer these islands as original or relaying places at least for northern Sulawesi macaques (Cynopithecus), granting the hypotheses of VAN BEMMELEN(1949) and HOOIJER (1950)]. Although many authorities have agreed that M. nemestrina represents a candidate macaque species having the most recent common ancestor with the Sulawesi macaques (FOODEN, 1969, 1980 ; DELSON, 1980), their opinions differed concerning the most primitive monkeys within the Sulawesi macaques. For example, from the viewpoint of morphology, the proto-Sulawesi macaques were estimated to have been most similar to M. tonkeana (FOODEN, 1969), M. maura (ALBRECHT,1978) or M. ochreata (WATANABEet al., 1985). A similar situation prevails in the case of studies of their biochemistry (TAKENAKAet al., 1987a, 1987b) and population genetics (KAWAMOTOet al., 1985). Interdisciplinary field work on the Sulawesi macaques was therefore carried out in 1981, 1984, and 1986, and some of the results have been published (e.g., TAKENAKAet al., 1987b). The present study focuses on the body (coat) color. Although body color had been used previously, because it is handy for species or subspecies discrimination, etc., the quantitative variation in body color within natural populations was not examined. There is indeed the hypothesis of metachromic pathways suggested by HERSHKOVITZ(1977), but the evolution of body color in primates has not yet been thoroughly investigated. We applied an apparatus to Sulawesi macaques, which measured body color digitally. Since Sulawesi macaques are thought to be phylogenetically very close to each other, in spite of their estimated multifold speciation, elucidation of a mechanism for the variation in body color and possibly for the evolution of the Sulawesi macaques should be possible. SAMPLES AND METHODS COLLECTIONOF SAMPLES In 1984 and 1986, we made round trips through the routes shown in Figure 1 to cover all places on Sulawesi that the seven species of macaques inhabit. We collected samples and information (blood, somatometrical data, and observations of external physical features) from monkeys which were kept as pets, and also data on where those monkeys came from, as reported by TAKENAKAet al. (1987b). As regards the border lines between the species, TAKENAKAet al. (1987b) drew up a map modifying the one made by FOODEN(1969), as will be discussed elsewhere. Measurements of body color were also carried out on wild monkeys, as follows: on M. maura in the Karaenta Nature Reserve, Maros, South Sulawesi; on M. nigra in the Tangkoko Batuangus Nature Reserve, Minahasa, North Sulawesi; and on M. nigrescens in the Dumoga-Bone National Park, Bolaang-Mongondow, North Sulawesi. The monkeys were temporarily provisioned and captured with traps constructed from timber and nets or from timber and rattans. Additional samples were taken from M. n. nemestrina individuals which were brought to the Cenkareng Primate Laboratory, Tangerang, West Java, from Sumatra. The captured and pet monkeys were anesthesized with ketamine hydrochloride at a dose of less than 10 mg/kg body weight. After sampling, the monkeys were allowed to recover and thert released without accident. Table 1 lists the numbers of samples in relation to species, age, and sex. The ages of the Body Color of Sulawesi Macaques 67 Manadotua~. Is. I rangkoko Batuangus N.R, Lembeh I s . H Malenge Xo~ o Is. i I Dumoga -Bone National Park C~Togianoc I O N.R. - )~ ! 1 ~bluna I s . Fig. 1. Routes of our round trip surveys and capturing sites on Sulawesi. Species boundaries are indicated by dotted lines and species names are abbreviated as follows: A: nigra; B: brunnescens; H: heck; M: rnaura; O: ochreata; S: nigrescens; T: tonkeana. monkeys were estimated from the state of their dentition, e.g., eruption and attrition. On the basis of FOODEN'S classification of Sulawesi macaques (1969), we assumed that there were seven species of Sulawesi macaques. At first, we basically applied the mode of discrimination of species which FOOOEN (1969) has suggested, but soon the discrimination could be made at a glance. METHODS The colors of the animal body have so far been expressed in ordinary terms, such as pale gray, pale brown, brown, blackish, glossy black, pale ochreceous-buff, brownish gray, or ochreceous-gray. Since delicate hues and their differences can hardly be expressed objectively even by such a rich terminology, quantitative analysis of body color was found to be impossible by the observational (qualitative) method. Using a Color Analyzer (CR-100, Minolta Co., Ltd.), we measured the body color of the back, upper arm, forearm, hand, thigh, leg, and foot. The precise measuring points were at the middle between the scapulae for the back, at the center of the lateral aspects for each of the upper arm, forearm, thigh and leg, and at the center of the metacarpal and metatarsal parts for the hand and foot, respectively. The Color Analyzer emits standard light (D65) over a circle of 8 mm in diameter with a Pulse Xenon Lamp by the d/o method under the Y. HAMADAet al. 68 Table 1. Numbers of samples in relation to species, age, and sex. Species Age 0 years 1 3 1-7 years 20 9 Adult ( > 7 years) 8 10 Total 29 22/51 M . maura m. f. M . tonkeana m. 9 2 f. 12 1 m. 12 12 f. m. f. m. 6 12 5 9 6/18 12 5/17 10 M. ochreata M . brunnescens M . hecki M. nigra .VI. nigrescens M. n. nemestrina f. m. f. m. f. m. f. 1 Overall total 5 30 17 9 12 18 8 1 5 6 4 2 11 13/24 5/15 35 23/58 10 12/22 22 10/32 237 definition of JIS-Z78722, and receives only reflected light in the vertical direction. Skin color may influence the result more or less, but this method is regarded as fitting best for the evaluation of color by the human eye. Color is physically composed of three factors: Value (lightness), Hue and Chroma. In the present study, we analyzed Value and Chroma statistically. The Color Analyzer output consists of three numerals, L* (0-100), a* ( r e d + 6 0 - g r e e n - - 6 0 ) and b* (yellow + 6 0 - b l u e --60). Based on these numbers, three factors of color are obtained from the following equations: Value = L*, Hue ( H ~ = t a n - l ( b * / a * ) , and Chroma = ~/a*~+b*2. Besides the quantitative measurements described above, we observed the thorax-belly skin color and color pattern of each hair. RESULTS SEXUAL DIFFERENCES OF BODY COLOR Sexual differences of body color were examined by the t-test between averages for the sexes (>= 1 year old) in m a u r a and nigra. Table 2 summarizes the results. Nearly all items showed no significant difference in Value or Chroma. Even differences which were significant were very small. In addition, here was no clear tendency for the averages of one sex to be larger than those of another. AGE DIFFERENCES OF BODY COLOR Although the results of the t-test for m a u r a did not reveal any significant differences (Table 3) between the averages of newborns (younger than 3 months old) and those of animals older than 0 years old in all items, except for the hand, the averages in the former did tend to be smaller than those in the latter. Plots of the body color of the back and foot against age are shown in Figure 2. In each item, no marked age change in body color was found, although the hairs of some monkeys Body Color of Sulawesi Macaques 69 Table 2. Sexual difference of body color (age > 1 year old). M. maura Male 1. Back 2. Upper arm 3. Forearm 4. Hand 5. Thigh 6. Leg 7. Foot L* Ch* L* Ch* L* Ch* L* Ch* L* Ch* L* Ch* L* Ch* M. niera Male Female Female :>* <* All n.s. Significance of t-test: * p <0.05. Table 3. Age differences of body color between 0-year-old monkeys and those older than 0 years in M. maura. 0 years 1. Back 2. Upper arm 3. Forearm 4. Hand 5. Thigh 6. Leg 7. Foot L* Ch* L* Ch* L* Ch* L* Ch* L* Ch* L* Ch* L* Ch* > 1 year < < < < < < < > < < < < < < All n.s. Significance of t-test: * p <0.05. older t h a n 10 years d i d t u r n gray. T h e hairs o f one a d u l t male h a d t u r n e d gray o v e r the w h o l e body, a n d the c o l o r o f the h a n d a n d f o o t o f eight a d u l t females h a d b e c o m e whitish. DIFFERENCES BETWEEN SPECIES Comparison by Averages F i g u r e 3 shows plots o f the Values a n d C h r o m a s for each species. Based o n a c o m p a r i s o n o f the Values, the Sulawesi m a c a q u e s can be g r o u p e d into two t y p e s : one in which the c o l o r v a r i a t i o n within the b o d y is slight, a n d the o t h e r in w h i c h the v a r i a t i o n is wide. T h e f o r m e r type includes maura, tonkeana, nigra, a n d nigreseens, a n d the latter includes oehreata, brunneseens, a n d heeki. In the f o r m e r type, maura has a lighter c o l o r in all items t h a n tonkeana; nigra is slightly lighter t h a n tonkeana in items o f the u p p e r h a l f o f the b o d y ; a n d nigreseens is lighter t h a n nigra in items o f the p r o x i m a l p a r t o f the b o d y . In the latter type, heeki is very light in the leg 70 Y. HAMADA et al. L* 5O L 5O Male 40 Female 40 ~ 30' " !:, 20- back ee 9 9 9 Sn 0 i'0 L 9 . 9 Age i0 9 30 i0 ' 1S I I: 9 9 Age ! L 9 50 50 40 40 foot 30 ee ~ " r 9 ,~e 30 " 20 20 :.-.. ":. i:. Age I0 0 , , , 5 i0 15 Age I0 S I0 15 Fig. 2. Age changes of body color on the back and the foot. and slightly light in the forearm, thigh and foot, as compared to tonkeana. The proximal part of the body in ochreata is very dark, nearly as dark as tonkeana, but the distal part of the body is very light. Thus, ochreata has an obvious two-tone body color pattern. A two-tone pattern is also observed in brunnescens with a lighter proximal part of the body than in ochreata. Comparison of the Chromas reveals that all of the Sulawesi macaques are duller than nemestrina, and they are difficult to type. All items except the hand and foot o f maura are more vivid than the averages of the Sulawesi macaques. Contrast within the body is not striking in tonkeana and nigra, and the Chroma itself is very low. The forearm, hand and foot of nigrescens are very dull, but the rest is vivid. The chromas of the distal parts of ochreata, e.g., the forearm, hand, leg, and foot, are higher than those of the proximal parts. The Chromas in brunnescens are higher than the averages of the Sulawesi macaques with slight variations within the body, except for the leg. The back, upper arm, and hand are more vivid than the rest in hecki. Cluster Analysis Cluster analysis was applied to each of Value and Chroma using the Euclid distance which was calculated on the basis of the species averages and the nearest neighbour method. The matrix of the Euclid distance is shown in Table 4. Figure 4 illustrates the dendrogram. In the analysis of the Value, the first cluster is composed of tonkeana, nigra, and nigrescens, and then hecki and rnaura are added to it in that order. Another cluster is formed by ochreata O Ch* L* 4O tonkeana maura nigra nigrescens o_ 15 e- 30 9 0 9 @ 00 O O O 00 10 O 00 20 O 0 O NO. 10 I 1 * ' 23 | * 45 | 67 9 O O O O 0 0 I , 0 n 34 12 0 O @ 9 00 0 0 ! 00 0 9 00 9 I 56 ! ! 7 12 | i 0 | i , O 0 9 0 @ 0 0 7 12 34 56 7 Ch * L* hecki ochreata brunnescens Onemestrina 50 20 0 9 IIe eo 40 0 O 30 @ @ 20 9 0 O 9 9 $o O 9 9 oO 9 9 9 0 15 0 10 8 o @ O 9 0 @ 9 O O 00 No. 10 S a 56 54 9 O 0 0 ! 0 '*'' 1 23 45 67 * * I 12 34 56 I I s 7 12 i , i 5 4 , 56 , | , 7 1 r 2 | ! , 5 67 34 Fig. 3. Plots of the Values (L*, closed circles) and Chromas (Ch*, open circles). | 72 Y. HAMADA et al. Table 4a. Matrix of Euclid distance between species: Value. tonkeana ochreata brunnescens hecki nigra nigrescens nemestrina 15.94 23.05 30.78 26.46 39.71 16.58 14.04 13.25 21.80 30.42 16.41 6.13 31.89 40.55 15.85 15.18 7.09 32.85 40.30 15.24 5.00 maura tonkeana ochreata brunnescens hecki nigra nigrescens 41.80 56.01 39.77 24.86 49.13 57.30 56.56 Table 4b. Matrix of Euclid distance between species:Chroma. maura tonkeana ochreata brunnescens hecki nigra nigrescens tonkeana ochreata brunnescens hecki nigra nigrescens nemestrina 9.67 8.97 9.96 7.01 15.52 10.18 8.01 10.66 5.66 8.50 8.20 3.45 10.75 14.81 11.31 6.03 6.45 11.43 12.86 11.11 3.63 nisra nigrescens I I 20 r i J i0 0 ~ 30.73 38.45 31.11 24.51 31.20 38.19 36.42 tonkeana nigra tonkeana nigrescens hecki maura maura ochreata brunnescens ochreata l brunnescens hecki nemestrina nemestrina I (a) I 20 b I i0 I 0 (b) Fig. 4. Dendrograms calculated for Value (a) and for Chroma (b). and brunnescens but the distance between the two species is relatively large, while nemestrina is farthest from any of the Sulawesi macaques. As for the Chroma, tonkeana, nigra, and nigrescens are grouped in the first cluster. Then, maura and brunnescens join the cluster in that order, and hecki and ochreata form another cluster. Again, nemestrina is away from the groups of Sulawesi macaques. Discriminant Analysis Canonical discriminant analysis (using a program package; TANAKA et al., 1984) was applied for each of Value and Chroma. The canonical values of the Values obtained were as listed in Table 5. The back (--), upper arm ( + ) and leg ( + ) largely contribute to the first factor. The upper arm (--), forearm ( + ) , thigh (--) and leg ( § have a large influence on the second factor. The factor scores are alotted in Figure 5a. Clearly, tonkeana, nigra, and nigrescens are very close to each other, while maura, hecki, brunnescens, and ochreata are situated away from the cluster in that order, and nemestrina is located farthest from the cluster. Similar statistical analysis was undertaken for the Chroma. The items showing a strong B o d y C o l o r of Sulawesi Macaques 73 Table 5a. Canonical values: Value. I 1. 2. 3. 4. 5. 6. 7. Back Upper arm Forearm Hand Thigh Leg Foot I1 --0.1194 0.1371 0.0269 0.0512 0.0502 0.1594 0.0168 0.0036 --0.2055 0.1124 --0.0458 --0.1690 0.2598 --0.0576 Table 5b. Canonical values: Chroma. I 1. 2. 3. 4. 5. 6. 7. Back Upper arm Forearm Hand Thigh Leg Foot II --0.1155 0.1873 0.0114 0.5540 0.0516 --0.1069 0.0874 0.2186 0.2392 0.0231 --0.0183 0.2328 --0.5969 --0.0188 O0 rll OB OH sA~I)o T Op Ca) So AO M T OB H0 0 [b) F~g. 5. Plots of the first and second scores in the canonical discriminant analysis of Values (a) and C h r o m a s (b). Abbreviations of species names are as follows: A : nigra; B: brunnescens; H : hecki; M : maura; O: ochreata ; P: nemestrina ; S: nigrescenes ; T : tonkeana. 74 Y. HAMADAet al. Table 6a. Results No. of Species cases maura (51) tonkeana (24) ochreata (18) brunneseens (17) hecki (15) nigra (58) nigrescens (22) nemestrina (32) of discrimination using stepwise discriminant analysis: Value. No. of cases classified maura tonkeana ochreata brunneseens heeki n i g r a nigrescens nemestrina 40 6 1 4 18 3 2 1 15 1 2 1 16 1 14 5 40 13 1 3 18 32 Table 6b. Results No. of Species cases maura (51) tonkeana (24) oehreata (18) brunneseens (17) hecki (15) nigra (58) nigrescens (22) nemestrina (32) of discrimination using stepwise discriminant analysis: Chroma. No. of cases classified maura tonkeana ochreata brunneseens hecki n i g r a nigrescens nemestrina 33 3 5 4 6 1 20 1 2 1 15 1 1 1 16 1 1 14 2 4 1 39 12 2 1 4 16 32 contribution to the first factor were the hand ( + ) , back (--), upper arm ( + ) and leg (--). The leg (--) contributed strongly and the back ( § upper arm ( + ) and thigh ( + ) did so considerably to the second factor. Figure 5b plots the scores of the first and second factors. Clearly, maura, tonkeana, nigra, and nigrescens are together, and of the remaining three species, brunneseens is the closest to them. Variations among the Sulawesi macaques were observed mainly in the second score, but nemestrina was remote from the maura-nigreseens cluster in the first score. Table 6 lists the probability of error discrimination by stepwise linear discriminant analysis. A better result for correct discrimination was given by using the Values. Error discriminations were made between maura, tonkeana, nigra, and nigreseens. Some samples of oehreata tended to be discriminated to hecki or to brunneseens. One sample of brunneseens was discriminated to oehreata, and one of heeki was discriminated to tonkeana. All samples of nemestrina were correctly discriminated. Observation o f Skin Color from the Thorax to Belly A wide area (about 80 ~ ) of the thorax-belly skin was pale blue in ochreata and brunneseens, which had a tint of the same kind as observed in such macaques asfuseata and siniea. Their pale blue skin seems to be due to pigmentation in the deeper stratum of the skin tissue. The five other Sulawesi macaques revealed a more or less black skin which had a drab tint, and nigra and nigreseens had the darkest skin. The darkening appeared to proceed proximally with age from the medial aspects of the distal part of the extremities, but the reference to the darkness in the thorax-belly skin was not ascertained. The black skin is of the same kind as observed in greater apes and the silenus macaque, and it seemed to be produced by pigmentation of the superficial stratum of the skin tissue. As for nemestrina, almost all areas of its skin were unpigmented. Body Color of Sulawesi Macaques 75 DISCUSSION GEOGRAPHICAL DISTRIBUTION AND BODY COLOR VARIATIONS OF THE SULAWESI MACAQUES Sulawesi island is shaped like a curious K, which is made up of four peninsulas (see Fig. 1). Seven species of macaques inhabit the island at present. They are distributed linearly within the peninsulas, and hecki, nigrescens, and tonkeana border on more than one other species. However, they live in a fundamentally aUopatric manner with respect to each other. Among the species, brunnescens is found only in the Butong and Muna islands which are separated from the southeast peninsula by the Tiworo strait. Analysis of the similarity of body color between the species revealed first that nigra and nigrescens are very close to each other, while tonkeana has a remarkable similarity to them and rnaura is close to the three. Thus, the macaques from the top of the northern peninsula to the end of the southern peninsula via the central part of the island are similar to each other from the standpoint of body color. They have a common feature in that the Value of the color of the body does not vary widely. A delicate difference between nigra and nigrescens can be seen in the proximal part which is lighter and more vivid in the latter, while tonkeana is darker than nigra in the upper half of the body, and maura is lighter than tonkeana over the whole of the body. It is difficult to apply one pattern from the four species mentioned above to hecki, which inhabits intermediate places among the four species. The hecki macaque has a rather characteristic body color pattern in that the leg is very light, and each of the forearm, thigh, and foot is lighter than in tonkeana. It is difficult therefore to regard hecki as a variant with slight alterations. By the same token, it is difficult to select one species from among the four forms which is the most similar to ochreata or brunnescens. The ochreata macaque shows a clear two-tone body color pattern; that is, while the proximal part of the body is as dark as the respective parts in tonkeana, the distal parts are much lighter than those of tonkeana. The body color pattern of brunnescens looks biased towards a lighter direction from that of ochreata, including the proximal part of the body. There is thus no simple way to relate body color variation to geographical distribution among the Sulawesi macaques. Assuming, however, that fading (becoming lighter or diluted) is the sole direction of color change, we can present four courses for Sulawesi macaques, starting from the color pattern of tonkeana, as follows: (1) fading slightly on the upper half of the body--nigra, fading more on the proximal part of the body--nigrescens; (2) fading over the whole body--maura; (3) fading strongly on the legs--hecki; and (4) fading on the distal part of the body--ochreata, fading more over the whole body, including the proximal part of the body--brunnescens. The degree of fading increases in the order from (1) to (4). The reason why tonkeana is considered as the starting point is that tonkeana has the weakest color pattern, and fading was taken as the sole direction of color change. The position of tonkeana supports the hypothesis of divergent speciation proposed by FOODEN(1969). The reverse order of (4) through (1) may provide a model of branching (speciation), but a consideration of the geographical distribution of the macaques poses some difficulties with this model; namely, the position of hecki and the reversed order of nigra and nigrescens. FOODEN (1969) and ALBRECHT(1978) demonstrated a morphological cline which is parallel to the geographic one. Although we cannot make any convincing proposals regarding this 76 Y. HAMADAet al. problem at the moment, elucidation of the detailed tectonic history of Sulawesi island is expected to afford some overall picture of the biogeography of the macaques. The geological history of Sulawesi island has clearly not been tranquil, as reflected in the many active faults, especially in the northern arm of the island, and the supposed rotation by more than 90 ~ of this arm (OTOFtm et al., 1980). Although the color of the thorax-belly skin is not a morphological character fully independent of the body color discussed above, its variation among Sulawesi macaques may contribute to the analysis of the model of body color divergence. The ochreata and brunnestens macaques have blue skin, while the others, especially nigra and nigreseens, have black skin. MACHIDA and PERI~INS (1967) reported that the blue skin is produced by pigmentation in the dermis and the black skin by pigmentation in the epidermis (sometimes, and in the dermis). They inferred from a systematic comparison of primates that those animals with black skin are more advanced than those with blue skin. If this is the case, ochreata and brunnescens would be more primitive than the others and nigra and nigrescens would be more advanced, and the model of divergence of body color would be partially supported. However, firm conclusions cannot be drawn in this connection because the physiological mechanism of control of the melanocytes which function in the pigmentation of the skin and hair has yet to be fully elucidated (SEARLE,1968). FADING AND INTERSPECIFICVARIATION OF BODY COLOR Fading (graying) has been noted in many species of Sulawesi macaques with different frequencies (FooDEN, 1969 ; HILL, 1974). We observed fading in some of the wild adult mauras in the Karaenta Nature Reserve. One adult male had turned gray over the whole body, and the hands and feet of three adult females had turned gray. We estimated the ages from the teeth but could not establish any exact relation between fading and aging because all animals older than 10 years did not turn gray. We observed vitiligo-like blotches on the face and cheiridia of many, if not all, wild adult mauras. This contradicts the proposal of HILL (1974) that such blotches resulted from artificial rearing. We also observed gray color on hairs of nigra macaques. In this case, however, the graying was only on part of each hair (annulation), not all of the hair as in maura, and all hairs did not turn gray. The coat of such animals thus became grizzled. One adult male which appeared to be the leader of a group in the Tangkoko Batuangus Nature Reserve, exhibited typical grizzling in the upper half of its body, mainly from the shoulders to the arms. This type of graying is not always parallel to the advancement of age because some younger animals of the same group showed graying to a lesser extent on the forearm or from the shoulders to the upper arms. Grizzling was also seen in some nigra pets, including one adult monkey which was said to have been captured in Kasoarina village which is situated just southwest of the Reserve. It is interesting that the graying was not observed in two young adult nigras who were members of a male group which overlapped in its foraging area with that of the above-mentioned group. HILL (1974) gave a subspecific rank to the population on Lembeh island (P. lembeh, as Cynopithecus niger lembicus), which faces Tangkoko Batuangus Nature Reserve and Kasoarina. He indicated that its discriminating morphological characters from the mainland population were as follows: the body is smaller, the hairs of the back and body side are glossy-black superficially but drab beyond the surface, and the lateral aspect of the arm is Body Color of Sulawesi Macaques 77 gray up to wrist in some animals. It is, however, rather hard to regard the Lembeh population as a subspecies. This is because, as we have reported, at least the graying is not characteristic to the Lembeh nigras and the body color ofnigra itself (our samples were all from the mainland, excepting one from Manadotua island) is lighter than that of tonkeana. The body color of the Lembeh population is therefore considered to be within the range of variation found irt the mainland population. Researchers on the Sulawesi macaques have reported graying of other Sulawesi macaques (FoOOEN, 1969; HILL, 1974). Two of their observations deserve attention. One concerns a report on the Malenge population of tonkeana, specimens of which were acquired from Malenge island of the Togian group (Kep. Togian) in the Tomini gulf. These specimen exhibited graying over a wider area of the body. The color was much paler (almost white) and the base of the dark colored black hair was pale brownish gray as compared with the mainland specimens. The other report concerns the color of specimens of brunnescens, indicating that specimens from Muna island, from which all of our samples derived, were lighter than those from Butong island which has a wider geographical area than the former island. Our proposed model of body color divergence stands on the supposition of fading as the sole direction of color change, and may be supported by the findings for graying. The reports and observations described above suggest a tendency whereby fading (graying) of body color appears in Sulawesi macaques as an age change (though not a strict one) or as a populational variation. It should be pointed out that populational variations tend to appear in a population whose living area (or size of deme) is limited. This suggests to us that the fading may be produced by a genetic change or a change in the controlling mechanism of body color of smaller scale, and that the fading may be fixed in the greographically peripheral populations, like the fixation of genetic features by random drift. Although Sulawesi is an island, it is presumed to have been composed of fragmentary islands [as supposed by FOODEN(1969) and HILL (1974)]. RELATIONSHIPS WITH OTHER MACAQUES Extant macaques are classified into 12-19 species (NAPIER &; NAPIER, 1967; FOODEN, 1969, 1980). Each species shows a variety of color patterning on each hair of the back, from clear light and dark bandings (e.g.,faseicularis) to unclear ones (e.g., silenus). All the species of Sulawesi macaques have hairs which are extremely unclear at the back, and silenus and nemestrina have unclear ones also. In this respect, the Sulawesi macaques may be regarded as being in the same species group as these macaques (FOODEN, 1980). In addition, monkeys which have entirely black hairs at the back are confined to the silenus and Sulawesi macaques. This fact suggests that such a character may be rarely attained in the speciation of the genus Macaca. The candidate macaque species which has the most recent common ancestor with the Sulawesi macaques is considered to be nemestrina. This is indicated by the geographical distribution, morphology, especially the morphology of the reproductive organs (FOODEN, 1969, 1980), paleobiology (DELSON, 1980), and genetics (NoZAWA et al., 1977; KAWAMOTOet al., 1985; MELNICK& KIDD, 1985). The Sumatran nemestrina from which we obtained body color samples reveals a body color pattern in which only the back is dark (moreover, the dark part is restricted to a median band of approximately 5 cm in width), the rest of the body is light and the color does not 78 Y. HAMADAe t a | . show a wide variation within the body. When we consider the evolution of the body color of the Sulawesi macaques, regarding the nemestrina pattern as the ancestral one, darkening becomes necessary, notwithstanding the selection of a species from outside the extant Sulawesi macaques as being the closest. There are two available routes in the process of darkening. One is to minimize the amount of color change. According to this idea, brunnescens and oehream would be regarded to be the closest to nemestrina. The other is that all parts of the body become darkened to a similar degree. According to this idea, maura, tonkeana, nigra, and nigrescens would be regarded as primitive. It seems natural to consider the color variation of brunnescens and ochreata within the body as the production of a color pattern not as a mere variation of the color itself. Based on a morphological examination of Hylobates, GROVES(1972) supported the general contention that the color pattern is more valuable as a taxonomical feature than the color itself. If this is the case, the model of body color divergence in the Sulawesi macaques proposed earlier, in which tonkeana is regarded as the most primitive (the starting point for divergence), does not contradict such a contention. It is presumed therefore that from an ancestral stock with a brownish gray body color, the proto-Sulawesi macaques may have become glossy-black (like tonkeana), and then the body color faded in several different ways. CAUSALITY OF THE CHANGES IN BODY COLOR The supposed body color transition of the ancestor to the proto-Sulawesi macaques, i.e., darkening, may have been caused partly by some kind of fixation like genetic drift, if the size of the ancestral population was limited, by the same token as the fading discussed above. As one possible main causal factor for the darkening, the idea of SCHAIKand NOORDWUI< (1985) is worthy of consideration. By comparing the coat color offascicularis between populations of the Sumatran mainland and of Simeulue island, they hypothesized that coat color had originally functioned as a form of camouflage but that as the risk of predation (mainly by felids) decreased, so the function changed to thermoregulation and the coat color of the island population turned dark. They gave several examples of coat color differences to support their hypothesis: other island populations offascicularis, a comparison between the n. nemestrina of mainland Sumatra and the n. pagensis of Pagai island, and the interspecific variation of Hylobates. Their hypothesis may be applicable to the Sulawesi macaques because the ancestral population had lost any harmful predators except for humans and wild dogs, as at present too, because of the isolation of the island from other land masses for a long period. Another factor which may have influenced the darkening concerns the ecology of the Sulawesi macaques. The Sulawesi monkeys are basically terrestrial, in spite of the fact that they spend time in the trees to take fruit. We observed in the Tangkoko Batuangus Nature Reserve that nigra moved between food patches mainly on the ground, that the social behaviours were conducted on the ground, and that individuals escaped from us by running on the ground when they noticed us. The maura macaque also is basically terrestrial rather than arboreal, as supported by observations at the Karaenta Nature Reserve. SCHAIr;and NOORDWIJK (1985) indicated that the Simeuluefascicularis are more terrestrial than the mainland populations. A comparison of nemestrina and fascicularis, which are sympatric over a wide range of South East Asia, shows that the former is more terrestrial (CRocKETT & WILSON, Body Color of Sulawesi Macaques 79 1980) and darker than the latter. It is possible therefore that the terrestrial (or forest-floor living) macaques have evolved a darker coat color or that the darker the living place, the darker is the body color that has evolved. This interpretation may also be applicable to body color variations within the Sulawesi monkeys. The tonkeana macaque, which is the darkest and has the least color variation, lives in the central part of the island where the amount of precipitation is much higher than in other areas on the island. Even in the minimum month, it is not less than 100 mm at Rantepao, a city situated in the central part of the island. On the other hand, south east Sulawesi, which is inhabited by brunnescens and ochreata macaques, is much drier and more seasonal than any other area on Sulawesi. The difference in climate may influence the forest, i.e., the development and kinds of constituent trees, and the amount of sunshine falling on the forest floor may vary with the type of forest. Clearly, in studies of the body color in relation to the ecology of the animals, morphological and ecological investigations, especially from the viewpoint of interspecific and/or intraspecific competition, are needed [as attempted by FOODEN (1982)]. Concerning the internal causality (physiological factors) of body color variation, HERSHKOVITZ (1977) proposed a model of metachromic pathways, involving saturation and dilution with two kinds of melanin pigments, eumelanin and pheomelanin. The eumelanin pathway might be applicable to the ancestor-Sulawesi macaque body color transition (darkening, saturation) and the divergence model of body color within the Sulawesi macaques (dilution). However, since our model is founded on many suppositions concerning the functional relations of the color change, we cannot exclude the possibility that the support for HERSHKOVITZ'Smodel may just be superficial. Acknowledgements. This study was supported by 1984 and 1986 Grants-in-Aid for Scientific Research (Overseas Scientific Survey) from the Ministry of Education, Science, and Culture of the Japanese Government (59041038, 61041044). We are grateful to the Indonesian Institute of Science (LIPI) and the Directorate of Nature Conservation and Wildlife Management (PPA) for kindly granting us permission to undertake and making arrangements for the present study. We would like to express our sincere thanks to Prof. D. O. SOEMARWOTO,Univ. of Padjadjaran; Mr. H. NAPITUPULU, LIPI; Dr. HUTABARAT,Chief of PPA Sulawesi; Mr. I. ISKANDAR, Chief of PPA South Sulawesi Branch; Mr. D. W. S~NAGA,Chief of PPA North Sulawesi Branch; and Mr. C. L. DARSONO, Indonesian Society for Primatology, for their kind cooperation. We are much indebted to the field staff of PPA at Karaenta Nature Reserve (South Sulawesi), Dunaoga-Bone National Park and Tangkoko Batuangus Nature Reserve (North Sulawesi) for capturing the wild monkeys. We are also grateful to Drs. M. IWAMOTOand K. WATANABE,Primate Research Institute, Kyoto University, for their help and encouragement. REFERENCES ALBRECHT,G., 1978. The craniofacial morphology of the Sulawesi macaques: Multivariate approach to biological problems. Contr. Primatol., 13: 1-151. AUDLEY-CHARLES,M. G., 1981. Geological history of the region of Wallace's line. In: Wallace's Line and Plate Tectonics, T. C. WHITMORE(ed.), Clarendon Press, Oxford, pp. 24-35. BEMMELEN,R. W. VAN, 1949. The Geology of Indonesia. Government Printing Office, The Hague. CRA~qBROOK,TUE E. OF, 1981. The vertebrate faunas. In: Wallace's Line and Plate Tectonics, T. C. WHITMORE(ed.), Clarendon Press, Oxford, pp. 57-69. CROCHET, C. M. & W. L. WILSON, 1980. The ecological separation of Macaca nemestrina and M. .tascicularis in Sumatra. In: The Macaques, D. G. LINDBURG(ed.), Van Nostrand Reinhold Co., New York, pp. 148-181. 80 Y. HAMADAet al. DELSON, E., 1980. Fossil macaques, phyletic relationships and a scenario of deployment. In: The Macaques, D. G. LINt~BURG(ed.), Van Nostrand Reinhold Co., New York, pp. 10-30. FOODEN, J., 1969. Taxonomy and Evolution of the Monkeys of Celebes. Bibliotheca Primatologica, Vol. 10, S. Karger, Basel. - - , 1980. Classification and distribution of living macaques (Macaca LACEPEDE,1799). In: The Macaques, D. G. LINDBURG(ed.), Van Nostrand Reinhold Co., New York, pp. 1-9. - - , 1982. Ecogeographic segregation of macaques species. Primates, 23: 574-579. GROVES, C. P., 1972. Systematics and Phylogeny of gibbons. In: Gibbon and Siamang. 1. Evolution, Ecology, Behavior, and Captive Maintenance, D. M. RUMBAUGH(ed.), S. Karger, Basel, pp. 1-89. - - , 1976. The origin of the mammalian fauna of Sulawesi (Celebes). Zeitschrfftfiir Saugetierkunde, 41: 201-216. HERSHKOVITZ,P., 1977. Living New WorM Monkeys (Platyrrhini), VoL 1. Univ. Chicago Press, Chicago. HILL, W. C. O., 1974. Primates: Comparative Anatomy and Taxonomy, Vol. VII, Cynopithecinae. Edinburgh Univ. Press, Edinburgh. HOOIJER, D. A., 1950. Man and other mammals from Toalian sites in south western Celebes. Verhandelingen der koninklijke Nederlandse Akademie van Wetenschappen, 46: 1-165. KAWAMOTO,Y., O. TAKENAKA,B. SURYOBROTO,& E. BROTOISWORO,1985. Genetic differentiation of Sulawesi macaques. Kyoto Univ. Overseas Research Report of Studies on Asian Non-human Primates, IV: 41-61. MACHmA, H. & E. M. PERKINS,JR., 1967. The distribution of melanocytic melanocytes in the skin of subhuman primates. In: Advances in Biology of Skin, VoL 8, The Pigmentary System, W. MONTAGNA& Hu FUMAN(eds.), Pergamon Press, New York, pp. 41-58. MELNICK,D. J. & K. K. KIDD, 1985. Genetic and evolutionary relationships among Asian macaques. lnt. J. Primatol., 6: 123-160. NAPIER, J. R. & P. H. NAPIER, 1967. A Handbook of Living Primates. Academic Press, London & New York. NOZAWA,K., T. SHOTAKE,Y. OHKURA,• Y. TANABE, 1977. Genetic variations within and between species of Asian macaques. Jap. J. Genet., 52: 15-30. OTOFUJI, Y., S. SASAJIMA,S. NISHIMURA,A. DHARMA, & F. HEHUWAT, 1980. Paleomagnetic reconnaissance results from northern arm of Sulawesi, Indonesia. In: Physical Geology of Indonesian Island Arcs, S. NtSmMORA(ed.), Kyoto Univ., Kyoto, pp. 23-34. SCHAIK,C. P. VAN& M. A. VAN"NOORDWIJK,1985. Evolutionary effect of the absence of felids on the social organization of the macaques on the island of Simeulue (Macaca ]ascicularis fuscata, MILLER 1903). Folia PrimatoL, 44: 138-147. SEARLE, A. G., 1968. Comparative Genetics of Coat Colour in Mammals. Logos Press, Academic Press, New York. TAKENAKA,O., M. HOTTA,Y. KAWAMOTO,B. SURYOBROTO,• E. BROTOISWORO,1987a. Origin and evolution of the Sulawesi macaques. 2. Complete amino-acid sequences of seven/~ chains of three molecular species. Primates, 28: 99-109. - - , A. TAKENAKA,Y. KAWAMOTO,B. SURYOBROTO,& E. BROTOISWORO, 1987b. Origin and evolution of the Sulawesi macaques. 1. Electrophoretic analyses of hemoglobins. Primates, 28: 87-98. TANAKa, Y., T. TARUMI,& K. WAKIMOTO(eds.), 1984. Handbook of Statistical Analyses for Personal Computer. H. Multivariate Analysis. Kyoritsu, Tokyo. (in Japanese) WATANABE,T., Y. HAMADA,& B. SURYOBROTO, 1985. Morphological studies of the Sulawesi macaques: Somatometry. Kyoto Univ. Overseas Research Report of Studies on Asian Non-human Primates, IV: 79-85. Received July 29, 1987; Accepted October 16, 1987 Authors' Names and Addresses: YUZURUHAMADA,Japan Monkey Centre, Inuyama, /tichi, 484 Japan; TsuYOSHIWATANABE,Sugiyama Women's University, Chikusa-ku, Nagoya, Aichi, 464 Japan; OSAMUTAKENAKA, Primate Research Institute, Kyoto University, Inuyama, ,4ichi, 484 Japan; BAMaANGSURYOBROTO,Bogor .4gricultural University, Bogor, Indonesia; Yosm KAWAMOTO,Nagoya University, Chikusa-ku, Nagoya, Aichi, 464 Japan.