Osteology of Amur Goral NEMORHAEDUS CAUDATUS RUDDEANUS
from Sikhote-Alin.
(Proc. 2nd World Conf. Mt. Ungulates (1998): pp. 197-205)

Voloshina I.V.1, Myslenkov A.I.1, Alexeeva E.A.2
1Sikhote-Alin State Biosphere Reserve, 46 Partizanskaya, Terney, Primorsky Krai
692150, Russia.
2Institute of Biology and Soil Science, 159 Stoletie Pr., Vladivostok, 690022, Russia.

ABSTRACT
Comparison of the biometric parameters of four skull samples from recent populations and one fossil ones of Nemorhaedus caudatus is discussed. The postcranial skeletons of recent gorals from Middle Sikhote-Alin and fossil ones from South Sikhote-Alin were analyzed. Dispersion and discriminate analyses of recent gorals from China and Russia by skulls showed the significant differences between these subspecies N. c. griseus and N. c. raddeanus.
Key words : osteology, skeleton, skull parameters, population, subspecies, goral, significant differences, fossil, recent.

INTRODUCTION.
Analysis of skull and postcranial skeleton of the genus Nemorhaedus has not been carried out. A detailed description of the skull is present in Sokolov (1959), while age classification and the change of skull was studied by Bromley (1963) and by Solomkina & Khokhrjakov (1993). Egorov (1955) compared metacarpus and metatarsus bones of Siberian Ibex (Capra sibirica Meyer) and goral (Nemorhaedus sp.) from China. The fossil remains of Nemorhaedus were found in Upper Pleistocene deposits in France and Italy (Schaub 1927).

Amur goral (N. caudatus ruddeanus) fossil remains were found in four locations in Primorsky and Khabarovsky Krais. Ovodov (1977) found 153 fossil bones in three caves : "Geographical society cave," "Tiger cave," and "Spyaschaya Krasavitsa". In the Holocene layer of the cave "Chertovy Vorota" in the central Sikhote-Alin 12 bones were found by E. Alexeeva at the depth of 20 cm at the human neolitic site.

PROBLEM STATEMENT
According to Simpson (1945), the tribe Rupicaprini (Caprinae, Bovidae) consists of four genera: Oreamnos, Rupicapra, Nemorhaedus, and Capricornis. Lydekker (1913) divided five species within Nemorhaedus.

Other authors, on the contrary, unite all goral forms into one species (Ellerman and Morrison-Scott, 1951). Hayman (1961) and Walker (1968) believed that there are two species of goral: N. goral (Russia, Korea, China, and Burma) and N. cranbrooki (India, Tibet). Following Groves & Grubb (1985) we argue that there are three goral species: N. goral, N. baileyi, and N. caudatus.

The later consists of four subspecies: N. c. evansi (Burma, Thailand), N. c. griseus (South and Central China), N. c. caudatus ( North China), and N. c. raddeanus (Northeast China, Korea and Russian Far East). Thus, the classification of genus Nemorhaedus
is not clear. It is necessary more investigation on goral morphology and genetics (Myslenkov & Voloshina 1989).
We suggest some hypotheses:
1. Significant differences between several skull parameters between separate populations indicate that these populations have been separated a long time ago.
2. Absence of significant differences among skull parameters of Khingan, Middle Sikhote-Alin, and South Sikhote-Alin goral populations points that all Russian gorals belong to one form of subspecies rank N. caudatus raddeanus.
3. Significant differences between Russian and Chinese goral populations show that we deal with different subspecies: N. c. raddeanus and N. g. griseus.
4. There are no differences in body size between fossil and recent gorals. To test this, we compared measurements of bone length.
5. Fossil gorals have more massive bones than recent ones.

MATERIALS AND METHODS

The general sample of goral skulls was of 107: 90 adults (>3-years-old), 12 subadults (1 to 2-years-old), and 5 juveniles (<1-year-old). For analysis we used only adults. The recent skulls included 45 from Abrek and Dalnegorsk in the Middle Sikhote-Alin Range, 8 from the Zapovednyj mountains in the South Sikhote-Alin Range, 4 from Khingan near Amur River, 1 from Sidimi, 5 from the North China (Shansi, Hubei), 19 from the South China (Sichuan, Yunnan), and 2 from Prague Zoo (hybrids). There were 6 fossil skulls from the South Sikhote-Alin Range ( fig. 1). Beside the osteological collection of the Sikhote-Alin Reserve we used collections of Zoological Institute of Russian Academy Science (Saint Petersburg), Zoomuseum of Moscow State University, and Alexeeva's collection (Vladivostok). The measurements of 15 Chinese gorals and 2 hybrids were taken from Volf (1976).

The material of postcranial skeleton of recent gorals is represented by 2475 bones of 63 individuals, and 2685 fossil bones of 37 ones. Many of them were damaged and were presented by fragments which cannot be measured. The goral skeleton consists of 156 bones, but we can measure only 15 ones, since many of them are too small or in fragment state. In total, 106 fossil bones and 186 recent ones were measured. The postcranial bones were measured by Duerst method (Duerst 1926). The main skull measurements were made according to Sokolov (1959) by a sliding vernier caliper.

Fig. 1. Goral range in Russia and the sample locations.
1 - Khingan,
Middle Sikhote Alin: 2 - Abrek, 3 - Dalnegorsk
South Sikhote-Alin: 4 - Lazo Reserve, 5 - Lozovyi Mountains (fossil)

We measured 31 parameters for recent gorals and 10 additional measurements for the cranium of both the recent and the fossil gorals, because of the facial bones of the fossil skulls were poorly preserved while craniums were well preserved.

Most fossil bones were found in the Bliznetz cave on the southern slope of the Lozovyj Mountain Ridge on the spur of the Sikhote-Alin mountains, 25 kms from Nakhodka, Prymorie Kry, Russia. The cave begins with a well, on the bottom of which we carried out excavations (Tiunov et al, 1992). Age of goral remains from a depth of 3 m was estimated by radiocarbon (C 14) dating at about 12965-years-old.

Statistical analysis was conducted with the program "Statistica for Windows". We tested for differences between samples using the T-test and lambda (Kcholmogorov-Smirnov -test). We considered samples different only if both methods demonstrated significant differences. The main database was divided into 8 groups based on samples from different existing populations, which were compared according to above hypotheses. Pair comparison was supplemented by research of intrapopulation variability. We also compared recent and fossil forms.

RESULTS

Intrapopulation Variability

Middle Sikhote-Alin. Thirty-six standard statistical measurements of skull parameters taken from 23 to 45 skulls (some parameters were not measurable on all skulls) from the Abrek-Dalnegorsk are discussed ( fig. 2). The variability coefficient of most parameters is 5% to 10%. The quantity of rings on the horns as an age feature and horn spread were the most changeable, 31,7% and 31% respectively. The variability of other horn parameters was 15% to 20 %.

Southern Sikhote-Alin.-- Twenty five skull parameters from Zapovednyj Mountain Ridge are discussed. The measurements are fewer than taken from Middle Sikhote-Alin specimens because during research of fossil skulls, 10 additional measurements were added and only 52 goral skulls from Ternei District were used. The quantity of rings and spread of horns were the most changeable ( 22%). In general, variability of a parameter was similar between populations. The mastoid width was the least changeable (2,6% ).

Khingan Mountain Ridge, Amur River.--The most changeable parameters were quantity of rings (37%) and spread of horns (26,2%), whereas the least changeable was mastoid width (5%, n=5).

Southern China , Yunnan and Sichuan provinces.-- The most changeable parameters were the spread of horns (37,8 %) and quantity of rings (32,4%, n=11). The length of the skull was the least changeable (2,1%) (fig. 3).

Lozovyj mountain ridge (fossil population).-- We had only six broken skulls of fossil gorals. The remains of horns were absent. Circumference of horn cores was the most changeable character (22,4%, n=6), whereas for large samples, for example, the Abrek sample, it was 12,3 % (n=52). The mastoid width was the least changeable (3% ).

Fig. 2. Skull of adult male of the Amur Goral from Abrek Mountain.

POSTCRANIAL SKELETON

Vertebral Column (Columna vertebralis).-- In the goral, the vertebral column consists of 42-44 vertebrae. Its formula is next: cervical - 7, thoracal - 13, lumbar - 6, sacrum -5, and caudal - 12-14. The cervical vertebra length was 208 - 219 - 241 mm (Min - Mean - Max), dorsal length 232 - 260 - 277 mm, lumbar length 231 - 258 - 279 mm, sacrum length 71 - 81 - 96 mm, and caudal length 210 - 225 - 235, (n=12).

Atlas (atlas, fig. 4 ).-- The length of the ventral arch was 19,8 - 21,4 - 24,4 mm and the maximum width of wings is 64,3 - 66,7 - 70,4 mm (n=12). Unfortunately, the fossil remains of vertebrae are damaged and haven't been measured.

Epistropheus (axis, fig. 5 ).-- In the goral, the tooth-shaped appendix of the epistropheus has forms a hollow half-cylinder and the epistropheus crest forms a right-angled plate similar to that in the domestic goat (Capra hircus). The bone length was 46.4 - 52.9 - 56.1 mm, the crest length was 39,4 - 42.2 - 46.7 mm, The width of tooth-shaped appendix was 20.8 - 22,8 - 24.3 mm (n=14).

Scapula.-- We measured 6 fossil specimens and 15 recent specimens from the Abrek population. The bone length for recent gorals was 107,3-160-175,6 (N=15). We could only compare two measurements: the width of neck and the width of facet .
Both criteria have shown that the neck width (22 mm) was identical, but the facet width differed (Xfossil = 29,7 mm, n=6, Xrecent = 24,1 mm, n=14).

Humerus.-- The humerus length of recent goral was 179,5 - 193,6 - 204,3 mm (n=14). This bone was compared by 5 measurements. The humerus of recent gorals was longer than that of fossil ones. It is necessary to note that among the fossil humeri there were two individual bones that were considerably longer than the others ( 322 and 326 mm). These two were located ot the depth of 2,1 to 2,4 m and near each other. The width and the diameter two distal characters show that fossil bones were thicker than recent bones.

Radius and Ulna.-- These bones are knitted together. The total length of the forearm was 212,6 - 235,9 - 246,4 mm (n=14). Among fossil remains, the ulna was preserved better than the radius. A comparison was carried out by two measurements: the olecranon length and the head diameter. Olecranon length was significantly different.

Metacarpus ( fig. 6).-- Metacarpi were well preserved and easily distinguishable from other ungulate metacarpi. It's length in recent gorals was 131,8 - 140,2 - 150,8 (n=15) and in fossil gorals was 134,0 - 138,6 - 143,8 (n=11). Four characters, proximal diameter, proximal width, distal diameter, and distal width were significantly different.

Phalanx.-- The phalanges of the foreleg are more large and massive than those of the hindleg. From the fossil sample, we collected 162 phalanges I, 98 phalanges II, and 72 phalanges III. Phalanx I was different on distal width only. The phalanx I length in recent gorals was 37,7 - 39,9 - 42,2 mm (n=15). Phalanx II differed in both length and distal width. Phalanx II length in recent gorals was 23,0 - 26,0 - 28,7 mm (n=12). Phalanx III did not differ in any characters.

Pelvis .-- Length in recent animals is 210,0 - 220,6 - 233,0 mm (n=13). The fossil pelvis bones were damaged and represented by fragments only.

Femur .-- The femur length in recent gorals was 210,3 - 213,4 - 222,4 mm (n=14). Fossil bone length was 212,0 - 217,0 - 224,0 mm (n=5). Differences between fossil and recent gorals were detected only by the T-test for three characters: proximal width, proximal diameter, and distal width.

Patella.-- Length was 31,6 -33,5 - 37,0 mm (n=10).

Tibia. --Length in recent gorals was 241,0 - 254,2 - 269,0 mm (n=16). A comparison between fossil and recent animals has shown significant differences by T-test for proximal width and distal diameter.

Calcaneus.-- This bone was well preserved among the fossil remains. Length in recent gorals was 62,7 - 66,0 - 69,3 mm (n=12). Three characters differed: bone length, bone width, and head length. The fossil bones were more massive than recent bones.

Astragalus.-- This bone was also well preserved and was the most numerous of the fossil bones measured. Length in recent animals was 30,7 - 32,9 - 35,2 mm (n=17) and 32,2 - 34,8 - 36,6 (n=13) in fossil bones. The fossil bones were more massive than recent .

Metatarsus (Fig. 6 ).-- This bone was one of the most easily identified and it can indicate the presence of genus Nemorhaedus among ungulate remains. Schaub (1927) defined Nemorhaedus philisi from Tuscany, Italy using this bone. We carried out a comparison on five characters. Five characteristics differed when tested with the T-test, but only three of these differed when tested with the K-S-test. The bone length in recent gorals was 140,5 - 148,9 - 158,1 mm (n=14) and 138,0 -143,6 - 151,0 mm (n=8) in fossil goral.

Phalanx.-- Phalanx I of recent animals was 39,4 - 41,3 - 42,3 mm (n=13) long, while that of fossil animals was 38,0 - 42,2 - 44,0 mm (n=7) long. There was significant differences by both tests for the distal width only. Phalanx II length in recent animals was 25,5 - 27,6 - 29,5 mm (n=11). Only bone length differed. Phalanx III length in recent animals was 31,1 - 32,0 - 32,7 mm (n=7).

Fig. 3. Differences among three parameters of skull length in two subspecies - N. c. griseus (China) and N. c. raddeanus (Primorsky Krai, Russia).
Box represents M-Sx, and M+Sx. Whisker - Min and Max.
* indicates significant differences (P< 0.05) for both the T-test and K-S-test.
1 - main length, 2 - condilobasal length,* 3 - total length*

Fig. 4. The atlas of the recent goral.

Fig. 5. The epistropheus of the recent goral.

Fig. 6. Metacarpal (left) and metatarsal (right) bones of the recent goral.

DISCUSSION

Postcranial bones analysis showed, that the significant differences between the total bone length of recent and fossil gorals are absent , on the contrary the bone massive ratio has significant differences : the fossil is more massive, then recent( Fig.7).

The analysis of skull differences between the Middle Sikhote-Alin and Amur regions demonstrates that for the T-test four parameters differed: total length, greatest width, width between horn lug centers, and nasal bone length, but the K-S-test detected a difference only for the latter. For all four measurements goral from the more northern Amur region were larger than those from the Middle Sikhote-Alin.

Comparisons between bones from the Middle Sikhote-Alin and Zapovednyj mountain ridge populations demonstrated that only horn length and horn curvature differed. Skull measurements did not differ in any case. However, the opposite was true for the comparison between the Middle Sikhote-Alin and Southern China populations. Thirteen and 9 measures differed for the T-test and K-S-test, respectively. Eight parameters differed for both methods: condilobasal length, total length, greatest width, inter eyed width, mastoid width, alveolar length, spread of horns, and distance between centers of horn lugs.

The differences indicate that goral from Sikhote-Alin had a bulkier skull, which may be explained by Bergman's rule; that northern forms are larger than southern forms. Among the three parameters on skull length the main length did not differ.

Condilobasal length in southern-china goral (19,4 mm ) was less than that in northern. Based on the large number of differences in bone and skull measurements, we believe that these two forms are different subspecies which diverged long ago.

We found no differences between 5 skull parameters from recent bones from Abrek and fossil bones from Lozovyj populations. Therefore, this comparison confirms the absence of differences between Southern and Middle Sikhote-Alin goral populations.

Fig. 7. Comparison the postcranial bones between recent and fossil gorals from Sikhote-Alin.
A - total bone length, B - massive index (width / length)
1 - humerus, 2 - metacarpus, 3 - front phalanx I, 4- front phalanx II, 5 - front phalanx III, 6 - femur, 7 - tibia, 8 - calcaneus, 9 - astragalus, 10 - metatarsus, 11 - hind phalanx I, 12 - hind phalanx II.

Thus, all samples can be divided into two groupings by skull parameters, a northern group including all Russian goral and a southern group including all southern China goral. from Southern China. Analysis by both tests gives 11 significant parameters: condilobasal length, total length, face axis, greatest width, intereyed width, mastoid width, alveolar length, spread of horns, spread of horn lugs, circle of horn, distance between centers of horn lugs.

Discriminate analysis of 16 skull characteristics from three populations provided similar results. Southern and Middle Sikhote-Alin goral populations were not significantly different, but gorals from China and Russia were ( Fig. 8).

Based on the morphological differences between these two groups, we argue that the two groups may be divided at least subspecificly: N. c. griseus and N. c. raddeanus, or perhaps the two groups represent two different species. Data by Volf (1976) indicates that hybrids between the two groups are sterile, which supports the latter conclusion.

Fig. 8. Results of discriminate analysis of 16 skull parameters among three groupings.
1 - South China, 2 - South Sikhote-Alin, 3 - Middle Sikhote-Alin.

ACKNOWLEDGMENTS
The authors are grateful to the INTERNATIONAL SCIENCE FOUNDATION and the OPEN SOCIETY INSTITUTE for partial financial support by grants 52536-By-3, 05A601, and 05A62D. The age of remains was identified thanks to the staff of the laboratory of geochronology of the Institute of Geology and Geophysics SB AS USSR in 1977 (Novosibirsk). We wish to express our appreciation to Mrs. S. Bondarchuk for picture drawing, and to Mr. S. Chikhunov for picture scanning.
 

REFERENCES

Bromley, G., F. (1963) - Biology of the Amur Goral (Nemorhaedus goral caudatus
Milne-Edwards 1867). Trudy Sikhote-Alinskogo zapovednika, 3:191-267. (in Russ)
Duerst, I., U. (1926) - Vergleicheende Untersuchungs - methoden am Skelett bei
Saugen. Handbuch,der biologischen Arbeitsmethoden. Abt.7., Heft. 2 - Berlin.-
529 .
Egorov, O., V. (1955) - Ecology of Siberian Ibex (Capra sibirica Meyer). Trudy
Zologicheskogo Instituta, 17: 7-134 . (in Russ)
Ellerman, J., & T., Morrison-Scott (1951) - Checklist of palearctic and Indian
Mammals. British Museum. London.
Groves, C., P. & P., Grubb (1985) - Reclassification of the serows and gorals
(Nemorhaedus: Bovidae). Pp. 45-50. In: The Biology and Management of
Mountain Ungulates. Croom-Helm Ltd., London.
Hayman, R. (1961) - The Red Goral of the North-east Frontier Region. Proc. Zool.
Soc., London, 136: 317-323.
Lydekker, R. , (1913) - Catalogue of the Ungulate mammals in the British Museum.
1., British Museum, London.
Myslenkov , A. & I., Volosina (1989) - Ecology and behaviour of the Amur Goral.
Nauka, Moscow. (in Russ)
Ovodov, N., D. (1977) - Late anthropogenic fauna of mammals (Mammalia) in South
Ussurian Krai. Trudy Biolog. Instituta SO AN SSSR, 31: 157-177. (in Russ)
Schaub, S., (1927) - Die Antilopen des Toskanishen Oberpliocans. Mit 4 Textfingen.
Eclogaegeol. Helv., 21: 260-266.
Simpson, G., C. (1945) - The principles of classification and a classification of
mammals. Bull. Amer. Mus. Natur. Hist., 85: 1-350.
Solomkina, N., V. & S., A. Khokhrjakov (1993)- Sex and age dependent variations in
the skull of long-tailed goral, Nemorhaedus caudatus(Artiodactila, Bovidae).
Zool. zurnal., 72: 153-159. (In Russ)
Sokolov, I., I. (1959) - Fauna SSSR, I., 3. Moskva - Leningrad. (in Russ)
Tiunov, M.,P., A.,V. Kosmach & E.,V. Alexeeva (1992) - On the formation of the bat
fauna in the south of the Soviet Far East. In: Prague Studies in Mammology.
Praha, :207-211.
Volf, J. (1976) - Some Remarks on the Taxonomy of the Genus Nemorhaedus H. Smith
1827 ( Bovidae:Rupicaprini). Vest. Cesk. Spol. Zool. 40, 1 : 75-80.
Walker, E., P. (1968) - Mammals of the world. Baltimore: Johns Hopkins Press.