Skeleton, with especial Reference to the Naqada Race. 399 



variation and the relationships which obtain between one variation 

 and another ; these may be expressed by diagrams or by means of a 

 few numerical constants. After the data have thus been expressed 

 in a convenient form, they are examined to see if any legitimate 

 theory or generalisation can be drawn from them. If not, the facts 

 nevertheless possess a value, for light is sure to be shed on them by 

 future investigations, since the subject of variation is still in its 

 infancy. 



To obtain an accurate knowledge of the variability displayed by 

 man and the mutual relationships of the different organs is surely a 

 subject of great interest and of both practical and theoretical import- 

 ance. An exceptionally favourable opportunity for studying the 

 variation exhibited by the human skeleton in a fairly homogeneous 

 race was presented by the kindness of Professor Petrie in placing at 

 my disposal the skeletons of the " New Race " which he has recently 

 brought over from Egypt. There were remains of about 400 

 skeletons. 



Although numerous bone-measurements have been made, yet 

 generally the number in each race has been too small to be of any 

 statistical use, and even when fairly numerous measurements have 

 been made on a single race, the results have been dealt with in a 

 very inadequate manner. So far as I am aware, the present investi- 

 gation is the first where the limb-bones have been measured for 

 statistical purposes, and it is also, I believe, the largest series of 

 measurements that have been made on the human skeleton. 



The paper deals primarily with the nature of the variation in the 

 length of the limb-bones and of the correlation between the bones. 

 The sacrum and pectoral girdle have also been studied as far as the 

 material would allow. Besides this I have made some observations 

 on the more important ethnological characters which the material 

 exhibited. 



The measurements of each bone are recorded in the tables, and 

 these constitute the foundation for all theory, and in the future will 

 be of use for discovering fresh relationships and for correcting any 

 erroneous ideas. 



In the following paragraphs some of the more important conclu- 

 sions are briefly recapitulated : — 



(1) The variability of the limb-bones is roughly proportional to 



their absolute length, and hence expressing the variability in terms 



of the mean we tend to get a constant ratio. 



mi . • standard deviation X 100 „ . . i . V nrr ' i 



lne ratio fluctuates about 5. We do 



mean 



not admit that this ratio (called the " coefficient of variation ") 

 always expresses the variability of an organ as it concerns the indi- 

 vidual. 



