232 PHYSICAL INVESTIGATION. [PART I. 



The old attempts, with regard to the natural division of the 

 human race, were chiefly founded on the visible differences ex- 

 hibited by human beings, such as the colour of the skin and the 

 quality of the hair. 1 It was, however, soon found that these pre- 

 sented too many deviations to allow great importance to be at- 

 tached to them. Thus the colour of the skin is not sufficiently 

 constant in the same people ; and the changes it presents from 

 one family of peoples into another, are uncertain. Generally 

 speaking, dark colour of the skin is found in combination with 

 black hair, and a black or brown iris ; on the other hand, fair 

 complexion is combined with light or red hair, and a light 

 brown, blue, or greenish iris. The second combination is 

 generally, though not exclusively, peculiar to the northern or 

 white race. If to this be added, that the hair of the Negro 

 frizzles in consequence of the elliptical form of its diameter, 

 we obtain a division of mankind into tribes of light complexion 

 and straight hair, into tribes of dark complexion and straight hair, 

 and, finally, into families of dark complexion and frizzly hair. 



This basis of a division according to colour and hair, has 

 recently been abandoned. Only D'Omalius d'Halloy 2 still con- 

 siders colour and hair as decided distinctions of race. He 

 includes the Finns and Turks in the white race ; and in the 

 brown (Malay), the Hindoos, as a mixed people of white Arians 

 and black aborigines, the Abyssinians, and the Fellahs. 

 Though he may be wrong in attributing greater constancy to 

 the colour of the skin than to the peculiarities of the skeleton, 

 it must be admitted that with many individuals of mankind it 

 would remain doubtful as to which family they belong, if the 

 colour and the hair did not furnish us with certain indications. 

 A greater degree of constancy is now generally attributed to 

 the shape of the skull. It is agreed, that it presents a basis 

 for a natural division; and this mark of distinction becomes 

 the more important as it gives some indications with respect 

 to past generations. 



With regard to the principal cranial types, from which the 

 rest may be considered as deviations, we find that Blumenbach, 



1 See Blumenbach, " De gen. hum. var. nat.," p. 296, 3rd edit. 



2 " L'Institut.," ii, p. 86, 1844. 



