270 Y. DODO and H. ISHIDA NAGAI, 1982). The latest cranial nonmetric studies have revealed that (1) the Jomon and Ainu are closely related with each o

Transcription

1 J. Anthrop. Soc. Nippon 98(3): (1990) Population History of Japan as Viewed from Cranial Nonmetric Variation Yukio DODO and Hajime ISHIDA Department of Anatomy, Sapporo Medical College Abstract Incidence data of cranial nonmetric traits were analysed in eight population samples from Japan, ranging in age from 4,000 B.P. to the present time. The trait frequencies were extremely homogeneous in the Japanese samples during the last 600 years from early medieval to modern times when no significant gene flow from overseas was evident in Japan. It was inferred that the incidence pattern of cranial nonmetric traits faithfully reflected the Japanese genetic constitution of historic times. SMITH's Mean Measures of Divergence (MMDs) based on the incidence data of 20 nonmetric traits between the protohistoric and historic Japanese samples were statistically insignificant, whereas those between the Neolithic Jomon and the foregoing Japanese samples far exceeded the significance level of The MMDs as well as the cluster analysis and principal coordinate analysis of MMDs suggested that (1) there existed population discontinuity between the Jomon and the protohistoric Kofun people; (2) the Aeneolithic Yayoi period was a phase when a number of different populations - natives and immigrants from the Continent - were struggling for their dispersals; (3) genetic constitution of the immigrants had predominated over that of the natives by the end of the Yayoi period and as a result the direct ancestral population of the modern Japanese was established during the protohistoric Kofun period. As for the Jomon and Ainu, their close affinity was demonstrated. Keywords Nonmetric trait, Skull, Incidence pattern, Population history, Japan Introduction On the origin of the modern Japanese two major hypotheses have been presented: the transformation theory and the immigration theory. The former claims that the Neolithic Jomon evolved without admixture into the modern Japanese (SUZUKI, 1956a, 1963, 1969, 1981, 1983), while the latter postulates that immigrants from the Asian continent during the Aeneolithic Yayoi and the protohistoric Kofun periods made a considerable contribution to the formation of modern Japanese (KANASEKI, 1956, 1959, 1966). Many of the recent craniometric and nonmetric studies as well as dental ones on the origin of the modern Japanese have yielded results in favor of the immigration theory (HOwELLS, 1966; YAMAGUCHI, 1982, 1985, 1987; HANIHARA, 1983, 1985; MIZOGUCHI, 1988; OSSENBERG,1986; MOURI, 1986; D0DO, 1986a, 1987; DODO and ISHIDA, 1988; KOZINT- SEV, 1990; TURNER, 1976, 1989; BRACE and Article No Received April 10, 1990

2 270 Y. DODO and H. ISHIDA NAGAI, 1982). The latest cranial nonmetric studies have revealed that (1) the Jomon and Ainu are closely related with each other (DODO, 1983a, 1986a, 1987; OSSENBERG, 1986; MOURI, 1986, 1988; KOZINTSEV, 1990) and (2) the protohistoric Kofun and/or the Aeneolithic Yayoi are much closer to the modern Japanese of Mongoloid stock than to the Neolithic Jomon (YAMA- GUCHI, 1985; MouRI,1986; DOD0,1987; DODO and ISHIDA, 1988; KOzINTSEV, 1990). In this study, an attempt is made to reconstruct more systematically the population history of Japan in terms of cranial nonmetric traits, using eight cranial samples from Japan whose dates range from 4,000 B.P. to the present time. Materials Eight cranial series from Japan, ranging in age from around 4,000 B.P. to modern times, were investigated for the presence or absence of cranial minor nonmetric variants. Sample sizes and provenances of these materials are listed in Table 1. The cranial series of the Yayoi, Edo, modern Japanese and Ainu are the same as those reported in our previous papers (DODO, 1974, 1975; DODO and ISHIDA, 1987, 1988). The Yayoi series examined in this study is generally considered to represent immigrants from the Continent and their offspring. It is derived from western Japan, while the remaining seven are from eastern Japan. These cranial materials are from the collections of Sapporo Medical College; Tohoku University School of Medicine; School of Medicine, Chiba University; the University Museum, the University of Tokyo; the National Science Museum, Tokyo; and the Faculty of Table 1. Materials examined 1),2),3),4): The same materials as reported in DODO and ISHIDA (1988)1), DODO (1975)2), and DODO (1974)3),x) or DODO and ISHIDA (1987)3),4)

3 Population History of Japan 271 Medicine, Kyushu University. The investigation consisted of only adult crania. The sex determination of the crania from dissecting room subjects was based on their documents and that of the archaeologically excavated crania was made by our own morphological observations. Methods The following 22 cranial nonmetric traits were examined for presence or absence. 1. Metopism* (For the criterion of trait judgment, see DODO, 1974) 2. Supraorbital nerve groove (Ditto) 3. Supraorbital foramen (Ditto and DODO, 1987) 4. Ossicle at the lambda* (DODO, 1974) 5. Biasterionic suture vestige (Ditto) 6. Asterionic ossicle (Ditto) 7. Occipitomastoid ossicle (Ditto) 8. Parietal notch bone (Ditto) 9. Condylar canal patent (Ditto) 10. Precondylar tubercle (DODO, 1975) 11. Paracondylar process (DODO, 1974) 12. Hypoglossal canal bridging (Ditto and DODO, 1980) 13. Tympanic dehiscence (DODO, 1974) 14. Foramen ovale incomplete (Ditto) 15. Foramen of VESALIUS (Ditto) 16. Pterygospinous foramen (Ditto) 17. Medial palatine canal (Ditto) 18. Transverse zygomatic suture vestige (Ditto) 19. Clinoid bridging (Ditto) 20. Mylohyoid bridging (Ditto) 21. Jugular foramen bridging (DODO, 1986b) 22. Sagittal sinus groove left* (YAMAGUCHI, 1977) * Median trait All the data of presence or absence of cranial nonmetric traits were gathered by the senior author (Y.D.). For both bilateral and median traits, incidence was detemined based on skull and pooled-sex sample. In the Japanese series of the Edo period, no observations have yet been made for traits #21 and #22. Frequency differences of each trait between samples were tested by chi-square statistics. Distances between different cranial series were assessed by C.A.B. SMITH's Mean Measure of Divergence (MMD) and its standard deviation (SJ*VOLD,1973; YAMAGUCHI, 1985). In order to graphically represent mutual relationships of cranial samples, cluster analysis and principal coordinate analysis of MMDs were attempted. The group average method was used in the cluster analysis. Incidence Data Skull-incidences of the 22 traits in the eight cranial series from Japan are given in Table 2. The data of traits #21 and #22 are missing in the early modern Edo series. Bilateral and unilateral trait occurrences in the cranial series of the Jomon, Kofun, Kamakura and Muromachi are shown for each sex in Appendix Tables 1-4. For those in the Yayoi, Edo, modern Japanese and Ainu series, see DODO and ISHIDA (1988), DODO (1975), and DODO (1974) or DODO and ISHIDA (1987). Comparison Incidences of each nonmetric trait were first compared between the cranial series of early medieval Kamakura, late medieval Muromachi, early modern Edo and modern Japanese. The significance of the frequency differences of each trait was tested by chi-square. As shown in Fig. 1, the differences are significant at the 0.05 level in two traits - the supraorbital nerve groove and the parietal notch bone. The frequencies of the two traits show a secular trend to increase. In the remaining 20 traits, however, frequency differences between the four series of historic

4 272 Y. DODO and H. ISHIDA

5 Population History of Japan 273 Fig. 1. Comparisons of trait incidences between the four Japanese cranial samples of historic times.

6 274 Y. DODO and H. ISHIDA Japanese never reach the significance level of The incidences of the supraorbital foramen and the hypoglossal canal bridging are surprisingly homogeneous in these four Japanese samples. Table 3 shows the matrices of SMITH's MMDs and their standard deviations for the eight cranial series from Japan computed from the incidence data of 20 traits, with traits #21 and #22 excluded. The Jomon and Ainu are rather close to each other, but are consistently distant from the six other population samples. The Yayoi are much closer to the protohistoric and historic Japanese than to the Jomon and Ainu. The closest relationships are observed in pairs among the protohistoric and historic Japanese samples. The MMDs between Kofun-Kamakura, Kofun- Muromachi, and Kofun-Edo are insignificant at the 0.05 level. Those between Kamakura- Muromachi, Kamakura-Edo, Muromachi-Edo, Muromachi-Modern, and Edo-Modern are also insignificant at the 0.05 level. Cluster analysis (group average method) was attempted for the MMD matrix of Table 3, with the negative MMD values obtained between Kofun-Kamakura and Kamakura-Muromachi being replaced by zeroes. The result is illustrated in Fig. 2 as a dendrograph. In this dendrograph, two major clusters are discernible: Jomon-Ainu vs. Yayoi-protohistoric and historic Japanese. In the latter cluster, the Kofun, Kamakura, Muromachi, Edo and Modern are joined tightly with each other, while the Yayoi form a slightly loose connection with these protohistoric and historic Japanese saniples. On the basis of the MMD matrix given in Table 3, principal coordinate analysis was carried out. The result is depicted three-dimensionally in Fig. 3. Proportions of the first, second and third principal components are 71.2%, 13.0% and 11.1 %, respectively. Also in this figure, the population samples from Japan are divided into two major groups: Jomon-Ainu vs. the remaining six samples. In the latter group, the Yayoi occupy a slightly deviated position from the protohistoric and historic Japanese samples which are lumped within a limited space. Fig. 4 shows the result of the cluster analysis of MMDs extended for 12 cranial series consisting of eight from Japan and four from overseas. The incidence data of cranial nonmetric traits in those overseas samples as the Mongolian, Aleut, Alaskan Eskimo and Canadian Eskimo are given in DODO and ISHIDA (1987). Even in this figure, Table 3. SMITH's MMDs (left lower half) and their standard deviations (right upper half) between the eight cranial series from Japan computed from frequency data of 20 traits MMDs within the solid line frame are insignificant at the 0.05 level

7 Population History of Japan 275 Fig. 2. Dendrograph of the eight population samples from Japan based on the cluster analysis of MMDs. JA: modern Japanese, ED: early modern Edo, MU: late medieval Muromachi, KA: early medieval Kamakura, KO: protohistoric Kofun, YA: Aeneolithic Yayoi, HA: early modern Hokkaido Ainu, JO: Neolithic Jomon. Fig. 3. Three-dimensional display of the result of the principal coordinate analysis of MMDs. JO: Jomon, HA: Hokkaido Ainu, YA: Yayoi, KO: Kofun, KA: Kamakura, MU: Muromachi, ED: Edo, JA: modern Japanese.

8 276 Y. DODO and H. ISHIDA Fig. 4. Dendrograph of 12 population samples from Japan and overseas based on the cluster analysis of MMDs computed from the incidence data of 20 cranial nonmetric traits. CE: Canadian Eskimo, AE: Alaskan Eskimo, AL: Aleut, MO: Mongolian, JA: modern Japanese, ED: Edo, MU: Muromachi, KA: Kamakura, KO: Kofun, YA: Yayoi, HA: Hokkaido Ainu, JO: Jomon. two major clusters are discernible: the one comprising the Jomon and Ainu, and the other socalled classic Mongoloids from East Asia and North America. The Mongoloid cluster is further divided into two subclusters: North American Mongoloid group and East Asian one. Inside the East Asian Mongoloid subcluster, the Mongolian connect at some distance with the protohistoric and historic Japanese samples, and the Yayoi join them at a still lower level of affinity. Discussion Within Population Homogeneity and Between Population Variability In Table 4, nine cranial measurements are compared between four male Japanese series of historic times. These craniometric data are quoted from SUZUKI et al. (1962). The cranial series of the Kamakura, Muromachi and Edo periods are almost the same as those examined in this study. Differences in measurements between samples were tested by the analysis of variance. The results are that in seven of the nine cranial measurements the differences between the four Japanese samples are significant at the 0.05 level. However, these results are not surprising, because it has been clarified that secular changes in cranial measurements are conspicuous even during the last 600 years from early medieval to the present time when no significant gene flow from overseas was evident in Japan (SUZUKI, 1956b, 1963, 1969, 1981; SUZUKI et al., 1962). In contrast to these cranial dimensions, most of the cranial nonmetric traits show a remarkable homogeneity in frequency within the same time span of historic times in Japan, as seen in Fig. 1. Perhaps this homogeneousness of trait incidence is a reflection of the unchanged genetic constitution of the Japanese population during the last 600 years.

9 Population History of Japan Table 4. Secular changes of main cranial measurements in male Japanese series of historic timesl) 1) Craniometric data are given in SUZUKI et al. (1962) * Significant at the 0.05 level Table 5. Chi-square test for homogeneity of trait incidence among four Mongoloid cranial series from East Asia and North Americal) 1) Skull-incidences in the cranial series of modern Japanese, Mongolian, Aleut and Alaskan Eskimo were tested * Significant at the 0.05 level

10 278 Y. DODO and H. ISHIDA Table 5 shows the results of the chi-square test for homogeneity of trait incidence among four Mongoloid populations. The cranial series of the modern Japanese, Mongolian, Aleut and Alaskan Eskimo were used for this analysis. The null hypothesis of trait-frequency homogeneity among the four Mongoloid populations is rejected at the 0.05 significance level in as many as 15 of the 22 traits. Therefore, population variability in occurrences of cranial nonmetric traits is obvious even in rather closely related Mongoloid populations from East Asia and North America. Such properties of cranial nonmetric trait occurrence as "within population homogeneity and between population variability" have been demonstrated also in different cranial samples (OSSENBERG, 1976), and thus they can be considered to provide good evidence for the effectiveness of cranial nonmetric traits in discriminating between closely related skeletal populations. Discriminating Power of Each Trait The results of cluster analysis and principal coordinate analysis indicated a dichotomy between the Jomon and Ainu on the one hand and the protohistoric and historic Japanese on the other (Figs. 2 and 3). Accordingly, the cranial Table 6. Contribution of each trait to the total measure of divergence between the Jomon-Ainu series and the pooled Japanese series 1) Skull-incidences are given 2) The cranial series of Kofun, Kamakura, Muromachi, Edo and Modern Japanese are pooled 3) Measure of divergence defined as 4) Total measure of divergence (EMD) =

11 Population History of Japan 279 series of the Jomon and Ainu, and those of the Kofun, Kamakura, Muromachi, Edo and modern Japanese were combined respectively to make two major cranial series: Jomon-Ainu and protohistoric-historic Japanese. Then, between the two cranial series, the measure of divergence by each trait and its contribution to the total measure of divergence were calculated, as shown in Table 6. In discriminating between the Jomon-Ainu series and the pooled Japanese series, the following five traits are more effective: Supraorbital foramen 22.1% Transverse zygomatic suture vestige o Mylohyoid bridging 10.6% Hypoglossal canal bridging 9.1% Medial palatine canal 8.4% Although the senior author previously considered supraorbital foramen and hypoglossal canal bridging most effective in discriminating between various racial groupings of man (DODO, 1987), the present study shows that the contribution of hypoglossal canal bridging is not as great as those of transverse zygomatic suture vestige and mylohyoid bridging. In this connection, it is noteworthy that supraorbital foramen and transverse zygomatic suture vestige, rather than hypoglossal canal bridging, have been demonstrated to be especially powerful in differentiating between various Mongoloid populations (KOZINTSEV, 1990). Population History of Japan It is obvious from Table 3, Fig. 2 and Fig. 3 that the protohistoric and historic population samples from Japan are very closely related with each other. The MMDs between protohistoric Kofun - early medieval Kamakura, Kamakura - late medieval Muromachi, Muromachi - early modern Edo, and Edo - modern Japanese are insignificant at the 0.05 level. Thus these protohistoric and historic Japanese samples are considered to belong to a single population. The close relationship between the Kofun and the modern Japanese has already been demonstrated based on cranial nonmetric variations (YAMA- GUCHI,1985; MOURI,1986; D0DO,1987). Consequently, from the evidence of cranial nonmetric variations it is asserted that the first emergence of the modern Japanese physical constitution dates back to the protohistoric Kofun age. On the other hand, since all the MMDs between the Neolithic Jomon and the foregoing Japanese samples far exceed the significance level of 0.01, it is reasonable to regard them as derived from two different populations: the Jomon and Japanese. The results of cluster analysis and principal coordinate analysis support this view (Figs. 2, 3 and 4), and similar results have been reported in other cranial nonmetric studies (YAMAGUCHI, 1985; OSSENBERG, 1986; MouRI, 1986; DODO, 1986a, 1987; DODO and ISHIDA, 1988; KOZINTSEV, 1990). In view of such properties of cranial nonmetric variation as "within population homogeneity and between population variability", as described before, it can be postulated that the dichotomy observed between the Jomon on the one hand and the Kofun and their descendents on the other is due to population discontinuity caused probably by immigration from the Continent during the Aeneolithic Yayoi and/or the protohistoric Kofun periods. This view is roughly consistent with the immigration theory proposed by KANASEKI (1956, 1959, 1966) and is considerably different from the transformation theory of SUZUKI (1956a, 1963, 1969, 1981, 1983) who claims that the Jomon evolved without admixture into the Kofun and finally the modern Japanese of the greater part of Japan. The cranial series of the Yayoi examined in this study is generally accepted as being comprised of immigrants from the Continent and their offspring. In the cluster analysis, it joins the cluster of the protohistoric and historic Japanese

12 280 Y. DODO and H. ISHIDA samples at a somewhat low level of affinity (Figs. 2 and 4). The MMDs between the Yayoi and the protohistoric or historic Japanese samples are all significant at the 0.01 level. This means that the schema for the post-jomon population history of Japan, Yayoi * Kofun *modern Japanese, advocated in our previous papers (DODO, 1987; DODO and ISHIDA,1988) may be oversimplified. Perhaps the population history from the Yayoi to the Kofun period was more complicated. In Japan, excluding Hokkaido and Okinawa, two types of Yayoi skeletal remains are recognized. The first type which resembles the Jomon in osteological characteristics and thus is considered to represent native inhabitants of Japan has been discovered in such districts as northwestern Kyushu, southern Kyushu, Shikoku, Chugoku, Kinki, and Kanto (NAIT0, 1971, 1981, 1984; KANASEKI, 1959, 1966; Dol and TANAKA, 1987; NAKAHASHI et al., 1989; YAMAGUCHI, 1979; MATSUSHITA, 1987; IKEDA, 1988; SUZUKI, 1963, 1969; KOIZUMI and IMAMURA, 1983). The second type, more or less similar to the physical constitution of Asiatic Mongoloid and regarded as comprising immigrants from the Continent and their offspring, has been found in the northern Kyushu, Chugoku, Kinki, and Tokai districts (USHIJIMA, 1954; KANASEKI, 1956, 1959, 1966; NAGAI, 1977; NAKAHASHI et al., 1985; KANASEKI et al., 1960: NAGAI et al., 1985; BABA et al., 1990; IKEDA, 1988; EHARA and OGATA,1989). Moreover, the Yayoi skeletal remains exhibiting mosaic features of the above two types have been unearthed from several sites in western Japan (MATSUSHITA, 1987; IKEDA, 1988). From these findings, it is inferred that the skeletal remains of the Yayoi period from various sites in Japan show considerable diversity in morphological characteristics, and this diversity must have been caused by such factors as isolation and microevolution on the one hand and immigration and hybridization on the other. In the population history of Japan, the Yayoi period may be a phase when a number of different populations were competing with each other for their gene dispersion. Since our data of cranial nonmetric variations indicate a close relationship of the second type of Yayoi people to the protohistoric and historic Japanese (Figs. 2, 3 and 4), it can be supposed that the genetic constitution of the immigrants from the Asian continent had predominated over that of the natives in various parts of Japan by around the end of the Yayoi period and as a result the direct ancestral population of the modern Japanese was established during the Kofun period. As for the Jomon and Ainu, their close affinity with each other has been noted by craniometric studies (HOWELLS, 1966; YAMAGUCHI, 1967, 1973, 1974, 1982; DODO, 1982; HANIHARA, 1983, 1985), by dental morphology (TURNER, 1976; BRACE and NAGAI, 1982; MATSUMURA, 1989), and by cranial nonmetric studies (DODO, 1983a, 1986a, 1987; DODO and ISHIDA, 1988; OSSENBERG, 1986; MOURI, 1986, 1988; KOZINTSEV, 1990). Although the same result is obtained in this study (Figs. 2, 3 and 4), the MMD between the Jomon and Ainu is significant at the 0.01 level. Therefore, in spite of their close affinity, they must be regarded as constituting two different populations. However, the recently discovered skeletal remains of the Epi-Jomon period in Hokkaido, contemporary with those of the Yayoi and Kofun periods in the mainland of Japan, have been disclosed to be transitional in morphological characteristics from the Jomon to the Ainu (OBA et al.,1978; DOD0,1983b; DODO et al., 1990). Accordingly, the Jomon people, especially those living in the northeastern part of Japan, must have progressed via the Epi-Jomon phase to the modern Ainu through either microevolutionary process or admixture with

13 Population History of Japan 281 their neighbors. Acknowledgments We are deeply indebted to Dr. Takeru AKAZAWA, Department of Anthropology and Prehistory, the University Museum, the University of Tokyo and Dr. Bin YAMAGUCHI, Department of Anthropology, National Science Museum, Tokyo for permission to investigate the cranial materials under their care. This study was supported in part by Grantin-Aid for Scientific Research, Ministry of Education, Science and Culture (No ) and Grant-in-Aid for Scientific Research on Priority Areas, Ministry of Education, Science and Culture (No ). References BABA, H., N. SHIGEHARA and K. HANIHARA, 1990: Human skeletal remains of Yayoi period from the Karako-Kagi site in Nara (abstract). J. Anthrop. Soc. Nippon, 98: 181. BRACE, C.L. and M. NAGAI,1982: Japanese tooth size: Past and present. Am. J. Phys. Anthropol., 59: DODO, Y., 1974: Non-metrical cranial traits in the Hokkaido Ainu and the northern Japanese of recent times. J. Anthrop. Soc. Nippon, 82: DODO, Y., 1975: Non-metric traits in the Japanese crania of the Edo period. Bull. Natn. Sci. Mus., Ser. D, 1: DODO, Y., 1980: Appearance of bony bridging of the hypoglossal canal during the fetal period. J. Anthrop. Soc. Nippon, 88: DODO, Y., 1982: A metrical analysis of Jomon crania from the Tohoku district. J. Anthrop. Soc. Nippon, 90 (Suppl.): (In Japanese with English summary) DODO, Y., 1983a: Jomon people in eastern Japan viewed from the incidence of non-metric minor cranial

14 variants (abstract). Acta Anat. Nippon., 58: 423. (In Japanese) DODO, Y., 1983b: A human skull of the Epi-Jomon period from the Minami-Usu-Six site, Date, Hokkaido. J. Anthrop. Soc. Nippon, 91: DODO, Y., 1986a: Metrical and non-metrical analyses of Jomon crania from eastern Japan. In: AKAZAWA, T. and C.M. AIKENS, eds., Prehistoric Hunter- Gatherers in Japan. Univ. Mus., Univ, of Tokyo, Bull., 27, Univ, of Tokyo Press, Tokyo, pp , pls DODO, Y., 1986b: Observations on the bony bridging of the jugular foramen in man. J. Anat., 144: DODO, Y., 1987: Supraorbital foramen and hypoglossal canal bridging: The two most suggestive nonmetric cranial traits in discriminating major racial groupings of man. J. Anthrop. Soc. Nippon, 95: DODO, Y. and H. ISHIDA, 1987: Incidences of nonmetric cranial variants in several population samples from East Asia and North America. J. Anthrop. Soc. Nippon, 95: DODO, Y. and H. ISHIDA, 1988: Nonmetric analyses of the Doigahama crania of the Aeneolithic Yayoi period in western Japan. In: Nippon Minzoku-Bunka no Seisei 1. Rokko Shuppan, Tokyo, pp (In Japanese with English summary) DODO, Y., N. OHSHIMA, H. 1SHIDA, M. KIDA and H. MATSUMURA, 1990: Human skeletal remains from Usu-Moshiri site, Date, Hokkaido (abstract). J. Anthrop. Soc. Nippon, 98: 180. (In Japanese) DOI, N, and Y. TANAKA, 1987; A geographical cline in metrical characteristics of Kofun skulls from western Japan. J. Anthrop. Soc. Nippon, 95: EHARA, A. and T. OGATA, 1989: Human skeletal remains from Shinmido (Yayoi-Era), Nishio-City, Aichi (abstract). J. Anthrop. Soc. Nippon, 97: 262. HANIHARA, K., 1983: Establishment of the modern Japanese. Saiensu (Scientific American, Japanese version), 13: (In Japanese) HANIHARA, K., 1985: Origins and affinities of Japanese as viewed from cranial measurements. In: KIRK, R. and E. SZATHMARY, eds., Out of Asia: Peopling the Americas and the Pacific. The Journal of Pacific History, Canberra, pp HOWELLS, W.W., 1966: The Jomon population of Japan: A study by discriminant analysis of Japanese and Ainu crania. Papers, Peabody Mus. Archaeol. Ethnol., Harvard Univ., 57: IKEDA, J., 1988: Yayoi skeletal remains from the western Tokai, Kinki and Setouchi regions. In: Nippon Minzoku-Bunka pp (In Japanese) no Seisei 1. Rokko Shuppan, Tokyo, KANASEKI, T., 1956: The question of the Yayoi people. In: Nippon Kokogaku Koza, 4. Kawade Shobo, Tokyo, pp (In Japanese) KANASEKI, T., 1959: The physical characteristics of Japanese in the Yayoi-period. Proc. 15th General Assembly, Japan Medical Congress 1, pp (In Japanese) KANASEKI, T., 1966: People of the Yayoi period. In: Nippon no Kokogaku, III. Kawade Shobo, Tokyo, pp (In Japanese) KANASEKI, T., M. NAGAI and H. SANG, 1960: Craniological studies of the Yayoi-period ancients, excavated at the Doigahama site, Yamaguchi Prefecture. Jinruigaku Kenkyu, 7 (Suppl.): (In Japanese with English summary) KOIZUMI, K. and K. IMAMURA, 1983: Human skeletal remains of the Yayoi period from the Iwatsubo cave site in Gunma Prefecture (abstract). J. Anthrop. Soc. Nippon, 91: 238. (In Japanese) KOZINTSEV, A., 1990: Ainu, Japanese, their ancestors and neighbours: Cranioscopic data. J. Anthrop. Soc. Nippon, 98: MATSUMURA, H., 1989: Geographical variation of dental measurements in the Jomon population. J. Anthrop. Soc. Nippon, 97: MATSUSHITA, T., 1987: Regional differences of the Yayoi people. In: Symposium Seinan Nippon-jin. Kikan Jinruigaku, 18: (In Japanese) MIZOGUCHI, Y., 1988: Affinities of the protohistoric Kofun people of Japan with pre- and proto-historic Asian populations. J. Anthrop. Soc. Nippon, 96: MOURI, T., 1986: Geographical and temporal variation

15 Population History of Japan 283 of Japanese populations as viewed from nonmetric traits of the skull. Doctoral dissertation. Kyoto University. (In Japanese) MOURI, T., 1988: Incidences of cranial nonmetric characters in five Jomon populations from West Japan. J. Anthrop. Soc. Nippon, 96: NAGAI, M., 1977: Human skeletal remains. In: Tateiwaiseki. Kawade Shobo Shinsha, Tokyo, pp (In Japanese) NAGAI, M., T. NAKAHASHI, N. Do!, Y. TANAKA and K. FUNAKOSHI, 1985: On the human remains from Koura site, Shimane Pref. (abstract). Acta Anat. Nippon., 60: 202. (in Japanese) SUZUKI, H., 1956b: Changes NAITO, Y., 1971: On the human skeletons of Yayoi period excavated at sites in north-western Kyushu. J. Anthrop. Soc. Nippon, 79: (In Japanese with English summary) NAITO, Y., 1981: Human skeletal remains of the Yayoi period. In: Jinruigaku-koza 5. Yuzankaku Shuppan, Tokyo, pp (In Japanese) NAITO, Y., 1984: The transition from the Jomon to the Yayoi skeletons in Kyushu. In: Jinruigaku. Nikkei- Saiensu Sha, Tokyo, pp (In Japanese) NAKAHASHI, T., N. DOI and M. NAGAI, 1985: Human skeletal remains of the Yayoi period from the Kanenokuma site, In: The Kanenokuma Site. Fukuoka City Board of Education, Fukuoka, pp (In Japanese) NAKAHASHI, T., N. DOI, Y. TANAKA and M. NAGAI, 1989: Human skeletal remains of Yayoi period from Hirota site, Kagoshima Prefecture (abstract). J. Anthrop. Soc. Nippon, 97: 265. OBA, T., S. MIZOGUCHI and B. YAMAGUCHI, 1978: Human skeletal remains from Etomo site, Muroran, Hokkaido. Bull. Natn. Sci. Mus., Ser. D, 4: (In Japanese with English summary) OSSENBERG, N.S., 1976: Within and between race distances in population studies based on discrete traits of the human skull. Am. J. Phys. Anthropol., 45: OSSENBERG, N.S., 1986: Isolate conservatism and hybridization in the population history of Japan: The evidence of nonmetric cranial traits. In: AKAZAWA, T, and C.M. AIKENS, eds., Prehistoric Hunter- Gatherers in Japan. Univ. Mus., Univ. Tokyo, Bulletin, 27, pp SJQ VOLD, T., 1973: The occurrence of minor nonmetrical variants in the skeleton and their quantitative treatment for population comparisons. Homo, 24: SUZUKI, H., 1956a: Human skeletal remains of the Jomon age. In: Nippon Kokogaku Koza, 3. Kawade Shobo, Tokyo, pp (In Japaeese) in the skull features of the Japanese people from ancient to modern times. In: WALLACE, A.F.C., ed., Men and Cultures. Univ. Pennsylvania Press, Philadelphia, pp SUZUKI, H., 1963: Skeletons of the Japanese. lwanami Shoten, Tokyo. (In Japanese) SUZUKI, H., 1969: Microevolutional changes in the Japanese population from the prehistoric age to the present-day. J. Fac. Sci. Univ. Tokyo, Sec. V, 3: , pls. I-Vu. SUZUKI, H., 1981: Racial history of the Japanese. In: SCHWIDETZKY, I., ed., Rassengeschichte der Menschheit. 8 Lieferung, Asien I. R. Oldenbourg, Munchen, pp. 7-69, pls. 1-V11. SUZUKI, H., 1983: The Root of the Japanese as Viewed from Skeletal Remains. Iwanami Shoten, Tokyo. (In Japanese) SUZUKI, H., H. SAKURA, T. HAYASHI, G. TANABE and Y. 1MAI, 1962: Craniometry of the Japanese skulls of the final Edo era. J. Anthrop. Soc. Nippon, 70: (In Japanese with English summary) TURNER II, C.G., 1976: Dental evidence on the origin of the Ainu and Japanese. Science, 193: TURNER II, CO., 1989: Teeth and prehistory in Asia. Scientific American, 260(2): USHIJIMA, Y., 1954: The human skeletal remains from the Mitsu site, Saga prefecture, a site associated with the "Yayoishiki" period of prehistoric Japan. Jinruigaku Kenkyu, 1: (in Japanese with English summary) YAMAGUCHI, B., 1967: A comparative osteological

16 284 Y. DODO and H. ISHIDA study of the Ainu and the Australian Aborigines. Occasional Papers 10, Austral. Inst. Aboriginal Studies, Canberra, pp YAMAGUCHI, B., 1973: Multivariate analysis of Ainu craniometry (abstract). J. Anthrop. Soc. Nippon, 81: (In Japanese) YAMAGUCHI, B., 1974: Paleoanthropology in Hokkaido. Dai-4-Ki Kenkyu (The Quaternary Research), 12: (In Japanese with English summary) YAMAGUCHI, B., 1977: A comparative study of the skulls of the Ontario Iroquoians and of Asiatic populations. Bull. Natn. Sci. Mus., Ser. D, 3: YAMAGUCHI, B., 1979: A human skeletal remain of Yayoi period from Shakameyama site, Ehime prefecture, Shikoku. Bull. Natn. Sci. Mus., Ser. D, 5: YAMAGUCHI, B., 1982: A review of the osteological characteristics of the Jomon population in prehistoric Japan. J. Anthrop. Soc. Nippon, 90 (Suppl.): YAMAGUCHI, B., 1985: The incidence of minor nonmetric cranial variants in the protohistoric human remains from eastern Japan. Bull. Natn. Sci. Mus., Tokyo, Ser. D, 11: YAMAGUCHI, B., 1987: Metric study of the crania from protohistoric sites in eastern Japan. Bull. Natn. Sci. Mus., Tokyo, Ser. D, 13: 1-9. Appendix Table 1. Bilateral presence (RL), unilateral presence (RO or OL), and bilateral absence (00) of cranial nonmetric traits in the prehistoric Jomon series * Median trait

17 Population History of Japan 285 Appendix Table 2, Bilateral presence (RL), unilateral presence (RO or OL), and bilateral absence (OO) of cranial nonmetric traits in the protohistoric Kofun series * Median trait

18 286 Y. DODO and H. ISHIDA Appendix Table 3. Bilateral presence (RL), unilateral presence (RO or OL), and bilateral absence (OO) of cranial nonmetric traits in the early medieval Kamakura series * Median trait

19 Population History of Japan 287 Appendix Table 4. Bilateral presence (RL), unilateral presence (RO or OL), and bilateral absence (00) of cranial nonmetric traits in the late medieval Muromachi series * Median trait Yukio DODO Department of Anatomy, Sapporo Medical College South 1, West 17, Chuo-ku, Sapporo 060, Japan