602 SCIENCE IN AFRICA 



its indication of the complexity of the problem which has to be 

 solved. Data on the cephalic index are probably the easiest to 

 collect, since measurements can be made on the living head; some 

 data for central Africa were obtained by Struck (1922) and illus- 

 trated by 'contour lines' on a map which, however, is somewhat 

 misleading, as the areas where no observations were made, form 

 the larger proportion of the total shown. While it may be possible 

 to determine some degrees of relationship from a study of the living, 

 direct evidence of this type bearing on racial genealogies can only 

 be collected from skeletons. Skulls are more easily procurable in 

 large numbers than other bones, and have, therefore, received 

 most attention from the anthropological point of view. For 

 example. Miss Elisabeth Kitson (1931) has compared the measure- 

 ments of one hundred and twenty skulls found by Dr. Leakey in 

 the Teita region of Kenya with the measurements of six hundred 

 skulls published by other workers. No sharp distinction can be 

 drawn between the negro races of Western, Eastern and Southern 

 Africa. The Teita skulls are more closely related to those of the 

 Nguni and Hottentots than to those of their neighbours in Tan- 

 ganyika, while the latter resemble the Galla and Somali, and the 

 early negroid skulls from Egypt. A clear differentiation between 

 Bushmen, Hottentots, and Kaffirs was observed. These observa- 

 tions serve once again to emphasize the complex ancestry of the 

 present African peoples. 



The discovery of differences between the blood of various peoples 

 and the recognition of blood-groups, made in 191 9 by L. and H. 

 Hirszfeld, provides a new method of approach in classifying races. 

 It depends on the principle that the transfusion of serum from one 

 person's blood to that of another may cause the red corpuscles to 

 agglutinate. Research has led to the classification of all humans 

 into four groups or seriological races, but opinion is divided as to 

 whether this division came about as a result of inheritance from 

 anthropoid ancestors or from the evolution of man himself. Edwin 

 Smith (1935) has collected published results on about 8,500 Afri- 

 cans in a table showing the percentage of each of the four groups 

 in the tribes studied. The principal conclusions are that Bushmen 

 belong predominantly to the blood-group which is regarded as the 

 most primitive, and that the Bantu exhibit marked variation, indi- 



