on the Surface of the Earth, 105 



quence of the differences between the progenitors and descendants. 

 Amongst these slight modifications, are some which, wanting in the 

 parents themselves, may yet be transmitted by them, because their 

 predecessors possessed them. The parents may, therefore, possess 

 certain qualities in a latent and virtual manner, and give to their 

 descendants a title to an inheritance derived from their forefa- 

 thers. Thus two black dogs may have white descendants, if one of 

 their own predecessors was of the latter colour. This is a law which 

 has been long known under the name of cetavism, and which modi- 

 fies a little our first and simple notion of a species by adding to ap- 

 parent characters modifications which have lain dormant for some 

 generations. 



Among the accidental modifications exterior circumstances may 

 produce, there are some which, in consequence of a continued influ- 

 ence on many generations, transmit themselves in the way of repro- 

 duction. Thus horses with heavy limbs and a lymphatic tempera- 

 ment, when conveyed to the dry regions of Arabia, are subjected to 

 an influence which will gradually diminish the size of their limbs, 

 dry their cellular tissue, make their heads smaller, and, after a cer- 

 tain number of generations, their direct descendants will acquire the 

 characters of an Arabian horse. After a suflftcient time, these qua- 

 lities will be transmitted, even although the circumstances which 

 produced them have ceased, and the descendants of these same 

 horses, brought back to the humid climates of temperate Europe, 

 will yield a progeny for some generations which preserve the marks 

 of the prolonged influence of Arabia. Remarkable instances are 

 mentioned of accidents altogether artificial reproduced in this way 

 by generation, when they are repeated (always artificially) on many 

 progenitors.* The new families (ascending and descending) modi- 

 fied by these circumstances, will differ in certain respects from the 

 primitive type, and the groups into which the descendants, thus 

 more or less modified and deviating from the common origin, are 

 divided, are named races. We thus have a new and important mo- 

 dification of the first notion of a species, for the characters which 

 limit it can no longer be recognised by the direct comparison of as- 

 cendants and descendants, but only by the study of a long series of 

 generations, during which different types may have been formed. 



The history of races presents us with important facts. \st^ The 

 individuals composing them are, in general, capable of uniting with 

 those of races proceeding from the same common origin, giving birth 

 to descendants which may themselves become the source of new be- 

 ings. 2cf, We observe in races which have deviated from the pri- 

 mitive type, a tendency to revert to that type. When the circum- 



* One of the most remarkable and best known is that of hounds, in which 

 the constant habit of cutting the tail has produced races in which this organ 

 has naturally only the length which was artificially left in their progenitors. 



