Abstract: Japanese racial elements are supposed to have been immigrated through almost all possible "roads"from surrounding seas. Japanese anthropologists have consistently paid attention to the surrounding regions, especially the continent of Asia, in seeking the "root"of Japanese people.
The present study considers the affinities of the inhabitants in the Japanese Islands within a broad spectrurn of the surrounding regions. The Japanese groups studied consist of the Minatogawa man, the Tsugumo, Yoshiko and South Kanto Jomon period groups, the Doigahana and Mitsu Yayoi period groups, and the rmodern Japanese groups of the Kanto, Kinai, Hokuriku, Ryukyu and Hokkaido Ainu regions. In accordance with this composition, late Paleolithic, Neolithic and modern Chinese inhabitants are used for comparison. Both Northern and Southern Chinese groups of the three periods are included. In addition, Neolithic and modern North Asian populations, as well as modern populations of Korea, Southeast Asia and Oceania, 38 groups in all, are included in this comparison.
All of the 38 groups have neans of the following 13 variables in common: cranial length, Cranial breadth, basi-bregmatic height. cranial base length, minimum frontal preadth, bizygomatic breadth, facial length, upper facial height, nasal height, nasal breadth, orbital height, orbital breadth (maxillofrontale-ektokochion) and total facial prognathism. These 13 means of the 38 groups form the data-set for statistical study.
Cluster analysis based on the Penrose's distances, factor analysis and discriininant analysis have been carried out.
The mean square distances produced a dendrogram which shows that the Upper Cave man and the modern Andaman group are distant from the remaining 36 groups, while these 36 groups compose 4 blocks: 1) Paleolithic Liujiang man and Minatogawa man; 2) Chinese Neolithic and Japanese Jomon groups; 3) modern Chinese, Japanese, Korean and Southeast Asian groups; 4) Japanese Yayoi groups of the North Kyushu and Yamaguchi regions and North Asian groups.
As to the factors analysis, the first factor explains 35% of the total variance. As in the comparison of Chinese groups only, which was reported in the author's previous paper, almost all of the linear variables, especially orbital breadth, cranial base length, facial length, cranial length, minimum frontal breadth, nasal breadth and bizygomatic breadth, have large factor loadings. This factor can be explained as size factor. In contrast, orbital height has a negative loading on the first factor. This implies that orbital height has a negative correlation with the other variables with regard to this factor. Also prehistoric groups have higher factor scores than modern groups of the same region. For the second factor, the variables which have the largest loadings are upper facial height, orbital height and nasal height. This factor reflects, mainly, geographical variation, and secondly, secular change from the Neolithic (Jomon period in Japan) to recent times. In this point, also, the analysis is comparable with the analysis of Chinese groups only.
A scattergram based on the first two factors documents a similar trend in terms of secular change from Neolithic (or Jomon) period to recent times in North China, South China and Japan. The general trend is a reduction in cranial size. This reduction concerns the 7 variables which have large loadings on the first factor. At the same time, orbital height increases slightly.
The purpose of the discriminant analysis was to detect the ability of each variable to distinguish between the groups of Neolithic (or Jomon) period on the one hand, and the groups of recent times on the other, regardless of whether Japanese or Chinese were represented. Its use was intended to test the significance of the differences between the two temporal periods in each variable. The results indicate that there are very significant differences between these two periods in facial length, cranial base length, nasal breadth, bizygomatic breadth, orbital height, cranial breadth and minimum frontal breadth.
Summing up the results produced by the above three methods, one can see that: I) the early inhabitants of the Japanese Islands were most similar to the South Chinese of the same period; 2) Yayoi period people in the North Kyushu and Yamaguchi regions were characterized by high facial height. Their racial elements were affected by a North Asian race; 3) modern Japanese are most similar to modern Chinese among the modern populations studied; 4) there is parallelism in terms of secular change over the period from the Neolithic (or Jomon) to recent times between North China, South China and Japan.
The statistical study described above leads to a further discussion about the racial relationships between Chinese and Japanese peoples.
1. Starting with Prof. Kotondo Hasebe, Japanese anthropologists have generally recognized that the earliest ancestors of modern Japanese were derived from South China. The results of the present study are in good agreement with this viewpoint. This immigration seems to have happened in a time earlier than late Pleistocene—the age of the Liujiang man and Mi- natogawa man. Although the Liujiang man and Minatogawa man were very similar to each other, they were still different in stature, cranial breadth, bizygomatic breadth, orbital breadth, facial prognathism etc. If these two samples could represent their respective populations, it is a logical inference that this couple of "cousins", as Prof. Hisashi Suzuki said, had diverged for a period of time. Furthermore, the existence of cultural remains and perhaps human fossils of a period earlier than that of the Minatogawa man gives support to this statement.
In the Quaternary glacial epoch, sea level was reduced by about 100 meters or more. The sea between the Continent and the Japanese Islands became a vast land or at least a land bridge linked them up. It is this land or land bridge that was the "road"of the immigrants from South China.
2. Although the Yayoi people in the North Kyushu and Yamaguchi regions were affected by a North Asian race in certain elements, modern Japanese people are still most similar to modern Chinese. With the fact in mind that they share a common ancestor, two alternative hypotheses can be put forward: that repeated immigration or parallel secular change accounts for the similarities between them.
3. With the close of the Quaternary Glaciation, the Japanese Islands separated from the Continent. The geographical basis for immigration on a large scale has not existed since the end of the glacial age. Also important is the way of life of Neolithic people. Agriculture, domestic animals and pottery occurred in this age. Residents lived settled lives increasingly. Immigration did not occur as frequently as before.
The Yayoi period in Japan was analogous to the Qin and Han dynasties in Chinese history in chronology, which developed foreign trade through ocean-going freighters. A nomadic nationality Xiongnu in the north border of the Qin and Han was also strong.Nevertheless, it is still hard to image that the barrier of the sea could be overcome by immigrants on such a scale so that the local inhabitants (but not the immigrants) were assimilated and absorbed gradually.
4. A certain parallelism in different regions in terms of secular change during the period from Neolithic to recent times implies that common motive forces were responsible for this process. In the list of candidates for these forces might be included cultural and technical progress, improvement of implerments, change in food composition and food preparation, as suggested by some authors. In other words, for the secular change of a population in Holocene time inter-population factors are primary or more important; external influence is secondary or less important.
As to the affinities of Japanese, a common rermote ancestor and a common trend of secular change resulted in the racial similarity of modern South Chinese and Japanese peoples.
The statistical analyses were carried out under the instruction of Prof. Kazuro Hanihara during the author's stay in the University of Tokyo as visiting scholar. Gratitude is expressed to him and his colleagues, especially Dr. Makiko Kouchi and Mt. Kiyotaka Koizumi for their help.

Key words: Chinese people; Japanese people; Crania