428 journal, r.a.s. (ceylon). [Vol. IX. 



those of Ceylon, to have ample and historically sure material 

 at our disposal. In the present case, I consider myself 

 justified in not foregoing the use of the skulls sent to me, 

 because through the comparison with the skull measurements 

 from other collections I have sufficient material under 

 control afforded me. Setting aside the not measured but 

 simply represented skulls in Davy's work, and also those of 

 Mr. Welcker (five) mentioned without any detailed state- 

 ments, and Sandifort's skulls given with a wholly different 

 method of measurement, we can use for our comparison 

 twelve Sinhalese skulls, viz., one from the Vrolik Museum, 

 nine from Mr. Davis's collection, and two from mine. These 

 would make a good broad basis for the decision, if important 

 numbers were not lacking in Mr. Davis's : for instance, 

 measures of the orbits, of the nose, and of the palate. Hence 

 my work in several respects can only serve as preparatory. 



Our study of details thus far has proved, first, that the 

 Sinhalese skull is considerably larger on the average than the 

 Vedda skull. Leaving out the two doubtful ones from nine 

 skulls of Mr. Davis's, and two of mine, we obtain an average 

 of 1,406 cub. cm., which exceeds the measure of the Vedda 

 average by 145 cub. cm. To be sure, the fluctuations here 

 also are very great : the extremes of 1,110 and 1,694 cub. cm. 

 afford a difference of 584, almost as great as we meet 

 with among the Yeddas. A comparison of them, however, 

 immediately shows that the numbers much more frequently 

 range high among the Sinhalese : — 



cub. cm. Veddas. Sinhalese. 



901-1,000 ... 1 j ... — 



1,001-1,100 ... if ... — 



1,101-1,200 ... 31 ... 3 



1,201-1,300 ... 10 * ... 1 



1,301-1,400 ... 2 ) ... 2^ 



1,401-1,500 ... 2 C 5 ... 2V7 



Over 1,600 ... 3J ... 3J 



Corresponding to the greater capacity, the measurements 

 are also larger with the Sinhalese. 



