Chap. VII.] THE RACES OF MAN. 233 



cause the distribution of the variously-colored races, most 

 of whom must have long inhabited their present homes, 

 does not coincide with corresponding differences of cli- 

 mate. Weight must also be given to such cases as that 

 of the Dutch families, who, as we hear on excellent au- 

 thority," haA^e not undergone the least change of color, 

 after residing for three centuries in South Africa. The 

 uniform appearance in various parts of the world of gyp- 

 sies and Jews, though the uniformity of the latter has 

 been somwhat exaggerated,^^ is likewise an argument on 

 the same side. A very damp or a very dry atmosphere 

 has been supposed to be more influential in modifying the 

 color of the skin than mere heat ; but as D'Orbigny in 

 South America, and Livingstone in Africa, arrived at di- 

 ametrically opjDOsite conclusions with respect to dampness 

 and dryness, any conclusion on this head must be consid- 

 ered as very doubtful." 



Various facts, which I have elsewhere given, prove 

 that the color of the skin and hair is sometimes corre- 

 lated in a surprising manner with a complete immunity 

 from the action of certain vegetable poisons and from the 

 attacks of certain parasites. Hence it occurred to me, 

 that negroes and other dark races might have acquired 

 their dark tints by thedarker individuals escaping during 

 a long series of generations from the deadly influence of 

 the miasmas of their native countries. 



I afterward found that the same idea had long ago oc- 

 curred to Dr. Wells." That negroes, and even mulattoes, 



excellent summary of the evidence is given by Godron, 'De TEspece,' 

 1859, vol. ii. p. 246, etc. 



*" Sir Andrew Smith, as quoted by Knox, ' Races of Man,' 1850, p. 473. 



*' See De Quatrefages on this head, ' Revue des Cours Scientifiques,' 

 Oct. 11, 1868, p. 731. 



^^ Livingstone's ' Travels and Researches in S. Africa,' 1857, p. 338, 

 329. D'Orbigny, as quoted by Godron, ' De I'Espfece,' vol. ii. p. 266. 



*^ See a paper read before the Royal Soc. in 1813, and published io 

 11 



