A CONTRIBUTION TO THE STUDY OF THE SCOTTISH SKULL. 377 



the two German groups than in the Scottish group, the reverse relationship holds in 

 regard to the orbital index. 



The palatal indices in the Scottish and Whitechapel series are practically equal, 

 and are higher in value than those found in the two German groups. 



As regards the alveolar, basilar, and nasial angles of the triangle ANB, there 

 is no great difference in their respective values for the English, Scottish, and 

 Naquada groups. 



The profile angle was computed in the series of 100 female skulls by the aid of 

 dioptrographic median sagittal tracings. The angle was found in the female to be 

 on an average about 1° greater than in the male, a relationship which was also 

 present in the English group, whereas in the German series the mean angle in the 

 female was slightly smaller than in the male. 



The upper facial index is approximately equal in the Scottish and English groups 

 and distinctly smaller in the German groups, which points to the relatively shorter 

 and broader face in the latter race, a condition found also in the male sex. 



(B) The Variability of the Scottish Skull. 



One of Turner's principal aims in regard to the formation of his collection of 

 Scottish skulls was that it should be thoroughly representative, and should contain 

 specimens not only from many different but from definite places in Scotland. This 

 he has been successful in accomplishing, but diversity is apparent in the high degree 

 of variability exhibited by the skulls when grouped together. He has obtained a 

 representative series, but at the expense of homogeneity. The collection being 

 derived from such varied districts and sources lacks uniformity, although it loses no 

 value on that account, but rather receives an enhanced value, as it gives us a 

 knowledge of the different types in the different parts of Scotland. 



There has been, till recently, much difficulty in defining what is exactly implied 

 in the term " homogeneous series," but Pearson (32) has made a definite statement 

 on the subject which has removed the difficulty. The statement is to the effect that 

 " the heterogeneity of any series the variability of which for skull length exceeds 6*5 

 (standard deviation), or for skull breadth exceeds about the same amount, should be 

 suspected and the series subjected to close examination. If the variability of skull 

 length be less than 5 '5, or of skull breadth less than 3*3, then we must suspect that 

 the series is a rather stringently selected sample. This rule will generally enable us 

 to distinguish between heterogeneity due to a mixture of crania from diverse races, 

 and the homogeneity of a single race which may, indeed, be the product of a number 

 of generations of cross-breeding such as we may assert of modern English, French, 

 and German, but hardly with the same certainty of Ainos or Bengal castes." 



Turner in his summarised results in his memoir gives the minimum, the 

 maximum, and mean values for any measurement taken by him. Macdonnell has 



