AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 59:399-411(1982) Japanese Tooth Size: Past and Present

Mesial-distal and buccal-lingual crown measurements were made on male and female samples of recent Japanese teeth from three locations, Fukuoka, Kyoto, and Tokyo, and for Hokkaido Ainu and Koreans. Similar data were collected for prehistoric Middle-to-Late Jomon Japanese and from Yayoi specimens representing the first agriculturalists to appear in Japan. From a tooth-bytooth comparison of cross-sectional areas, it was shown that the modern Japanese samples did not differ from one part of Japan to another. Korean tooth size also is not significantly different from Japanese, while Ainu have the smallest teeth recorded in Asia. The Yayoi who brought rice to Japan about 300 B.C. came in with teeth that were the same size as Chinese Neolithic teeth. They encountered a resident Jomon population whose teeth were 10% smaller. From tooth size measures alone, it is most economical to suggest that, if the rates of reduction observed elsewhere in the world applied in Japan, the recent Ainu would best be regarded as the direct descendants of the Jomon, while the modern Japanese are the results of in situ reduction from the incoming Yayoi. Other aspects of craniofacial morphology suggest that some Jomon was incorporated by the Yayoi. The modern Japanese, then, while predominantly derived from the Yayoi, would include a Jomon component. Lateand Post-Pleistocene reductions in muscularity and robustness can be documented in virtually every part of the world where human populations have remained in situ and maintained their continuity up into historical times. Because of the ease of quantification and the sheer amount of data available, the most easily demonstrated of these reductions is the one that can be seen in the human dentition. Since more attention has been focused on Europe than anywhere else, it has produced the most complete and unequivocal evidence for this trend of dental reduction (Brace, 1977, 1979; Frayer, 1978). Where the evidence has been examined, however, it is apparent that parallel reductions have taken place in other regions, for example preEuropean North America (Brace and Mahler, 1971), the Near East (Smith, 1977), China (Brace et al., in press), Southeast Asia (Brace and Vitzthum, in press), and even Australia (Brace, 1980). The magnitude of the change in tooth size is impressive to say the least, with a drop of some 50% since the early Mousterian in Europe (Brace, 1979), but, because of problems in establishing precise dates for some prehistoric samples, there is less certainty about the rates of change. Similarly, it is not yet clear whether the changes have followed the path of phyletic gradualism or punctuated equilibria (Brace, 19811, although we would suggest that the evidence for gradual and continuous change is more convincing than the converse. It is abundantly clear, however, that reduction did not stop nor was stasis achieved with the first appearance of “modern” Homo sapiens more than 30,000 years ago. Instead, tooth size continued to drop by more than 20% through the Mesolithic, the Neolithic and up to the present. In the latter part of this time span, it would appear that the rate of change accelerated to more than 1% per thousand years (Brace, 1979). The evolutionary mechanism which has produced the observed change has been the subject of continuing differences of opinion (Brace, 1963, 1980; Brace and Ryan, 1980; Brace et al., in Received April 19, 1982; accepted June 29, 1982 0002 948318215904-0399$04.00 C‘ 1982 ALAN R. LISS, INC. 400 C.L. BRACE AND M. NAGAI press; Armelagos and Van Gerven, 1980; Frayer, 1978; Smith, 1976, 1977; Williams, 19781, and we do not intend to dwell on the controversy here. It is relevant to our present concern, however, to note that the argument relating reduction to selective force relaxation leads to various predictions that can be tested against the data available in Japan. Specifically, it has been suggested that the use of pottery in the preparation of food brought about a major relaxation of the selective forces that had previously served to maintain a functional dentition throughout life (Brace and Mahler, 1971). The prediction follows, then, that dental reduction should be most apparent in the people who have continued to inhabit the areas where pottery has been in use for the longest period of time. Since Jomon pottery in Japan goes back further in time (i.e., 12,500 B.P. at Fukui Cave in Kyushu) than any other known tradition in the world (Ikawa-Smith, 1976, 1978, 1980; Pearson, 19771, it should be associated with people who show a maximum degree of dental reduction, a t least when compared with their counterparts in adjacent parts of Asia. In this paper, we present the data for tooth size in modern and prehistoric Japan. These are compared with the data available for other recent and prehistoric Asian samples, and we consider what light this can shed on questions concerning the origin and shaping of modern Japanese form. MATERIALS AND METHODS Mesial-distal and buccal-lingual crown measurements were taken to the nearest 0.1 mm according to the conventions usually followed in dental anthropology (Brace, 1979). Although previous studies have reported on both modern and prehistoric Japanese dental measurements, they have been done by separate authors, fewer examples were treated, and the modern specimens did not represent the geographic spectrum of Japan (Miyabara, 1916; Yamada, 1932; Gonda, 1959; Hanihara, 1976, 1977). The tooth measurements reported here were all made by the senior author in September and October of 1980. If nothing else, the present work should at least have the merit of consistency of technique. As in previous studies (Brace, 1978, 1979, 1980; Brace and Hinton, 1981; Brace, Mahler, and Rosen, 1973; Brace and Ryan, 1980; Brace et al., in press; Brace and Vitzthum, in press), teeth were not measured if it was apparent that wear had made an appreciable alteration in crown size. Buccal-lingual measurements, which are taken relatively low on the crown, are less likely to be affected by wear than are mesial-distal measurements. The latter are especially affected by interproximal attrition (Brace, 1967; Wolpoff, 1971). As a consequence, one might think that the most effective comparisons would be made by using buccal-lingual measurements alone. Experience has shown, however, that the clearest group distinctions are made by using the mesial-distal and buccal-lingual measurements in combination (Brace, 1967; Brace, Mahler, and Rosen, 1973). Their product can be taken as a representation of cross-sectional area. If selection really does operate to maintain usable tooth substance, then given the impractability of measuring unworn crown height, the closest approximation to the trait which selection actually controls is represented by cross-sectional area (Brace, 1967, 1980). The data on which our comparisons are based are recorded in Tables 1 to 7. We only used individuals for which sexual identification could be made. For each individual used, both right and left teeth were measured when possible, and an average of the two was taken to represent that individual. Of course if only one side was measurable, that is what was used. In either case, the resultant figure was considered as a single instance for the subsequent calculation of sample statistics. These dimensions were treated in a number of ways to simplify what would otherwise have been an unmanageable number of separate figures. First, we multiplied the mesial-distal (MD) and buccal-lingual (BL) figures together to apTABLE 1. Mesio-distal and buccal-lingual measurements standard deviations, and n for female and male human teeth from the Edo period in Tokyo