356 



BUREAU OF AMERICAN ETHNOLOGY 



[Bull. 182 



The cranial index in the two populations shows distinct dilfferences 

 (table 4). Except for two rather aberrant individuals (USNM 

 380892 in wliich the headf orm seems somewhat abnormal ; and USNM 

 380906 which is quite large and round), the Tollifero crania are rela- 

 tively narrow. If these two were excluded from the series, the mean 

 cranial index would be about 74. On the other hand, the Clarksville 

 crania are considerably wider and somewhat higher, with a mean 

 cranial index of about 80. Since neither series is large enough to al- 

 low calculating standard deviations, "mean sigmas" such as Stewart's 

 (1943 a) will have to serve as indicators of the variability. The mean 

 cranial indices are well over one mean sigma apart, and therefore are 

 probably significantly different. This difference is not likely to be 

 due to undetected deformity in the later population, for the relatively 

 greater width is also found in the face, orbits, nose, and palate. The 

 very slight deformity seen occasionally is not likely to have caused 

 these differences. 



Table 3. — Comparison of craniofacial modules and long "bone lengths in Tollifero 



and Clarksville males 



In table 3, the major difference between the two male series seems to 

 lie in the somewhat greater size of the Clarksville males. Yet an 

 examination of the individual measurements in tables 4 and 5 indicates 

 that this difference is not simply one of enlargement but also of change 

 in proportion. The major changes in the Clarksville face seem 

 to lie in the breadths — the bizygomatic, nasal, orbital, and palatal. 

 Of these, only the change in the bizygomatic breadth is likely to be 

 statistically significant. Height increases in the face and its com- 

 ponents seem to have been relatively slight, and there has been a de- 

 crease in the height of the lower jaw. Some correlation between the 

 facial heights and breadths and the components they cover is to be 

 expected (Pearson and Davin, 1924), but the correspondence in this 

 case is not very close. Orbital breadth has increased more than 

 height, resulting in relatively lower orbits, but total size, relative 

 to the skull and face, is about the same. Nasal height has increased 



