April 7, 1905.] 



SCIENCE. 



523 



pigs. Further, I have been able to find no 

 reference to such a thing in the literature 

 of the group, though I have several times 

 since found this same mutation in other 

 herds of guinea-pigs. The mother of my 

 four-toed pig never produced another sim- 

 ilar individual, though she was the mother 

 in all of thirty young. The father, how- 

 ever, who sired in all 139 young, had five 

 other young with extra toes, but these were 

 all by females descended from himself, so 

 that it seems certain that the mutation had 

 its origin in this particular male. By 

 breeding together the four-toed young and 

 selecting only the best of their offspring I 

 was able within three generations to estab- 

 lish a race with a well developed fourth toe 

 on either hind foot. This race was not 

 created by selection, though it was im- 

 proved by that means. Like the poet, in 

 the proverb, it was torn, not made. Any 

 amount of selection practised on other 

 families of my guinea-pigs would probably 

 never produce a four-toed race, for though 

 carefully watched through as many as 

 seven generations no four-toed pig has ap- 

 peared among them. 



In a second family of my guinea-pigs, 

 which, like the other, was for the purpose 

 of a particular experiment inbred, a dif- 

 ferent mutation made its appearance. A 

 few individuals were found to have hair 

 about twice as long as that of their parents 

 and grandparents. Intermediate condi- 

 tions did not occur. Long-haired indi- 

 viduals mated together were found to pro- 

 duce only long-haired young, so that a new 

 breed was already fully established without 

 the exercise of any selection. It was found, 

 in short, that the long-haired character is a 

 Mendelian recessive in relation to the nor- 

 mal short coat, so that matings between long- 

 haired and short-haired animals produce 

 only short-haired young. But these young 

 bred inter se produce a definite proportion 

 (about one fourth) of long-haired young, 



and if no selection is practised among their 

 offspring, but all are allowed to breed free- 

 ly, the race will continue to contain this 

 proportion of long-haired individuals. 



If such a mutation as this occurred in a 

 state of nature, and such a possibility we 

 can scarcely question, a dimoi*phic species 

 would be the immediate result, containing 

 two varieties alike in every respect except 

 length of hair, in which they would be 

 sharply separated. The two varieties 

 would coexist in the same habitat and 

 might continue to interbreed freely with- 

 out the destruction or necessary modifica- 

 tion of either. Natural selection would 

 now come into operation to choose between 

 the two that one which was more advan- 

 tageous and the other condition would be 

 gradiially eliminated from the race. Or if 

 the two conditions were each the better in 

 a different habitat, then by the gradual 

 destruction of the other in that habitat the 

 two varieties would become geographically 

 separated, though they might continue to 

 coexist in an intermediate zone. 



' The second method which I mentioned 

 for the artificial production of new breeds 

 is to combine in one race characters al- 

 ready found in different races. This is ac- 

 complished through cross-breeding and is 

 made possible by the facts (1) that muta- 

 tions are alternative in heredity to the 

 normal condition, and (2) that one muta- 

 tion is entirely independent of another in 

 heredity. If, for example, we cross long- 

 haired with short-haired guinea-pigs, we 

 get, among the second-generation offspring, 

 a mixture of long-haired and of short- 

 haired animals, but, as a rule, no interme- 

 diates. Further, if in the original cross 

 one parent was four-toed and the other 

 three-toed, then in the second generation 

 offspring we get all possible combinations 

 of the characters involved in the cross, viz., 

 long-haired four-toed animals, long-haired 

 three-toed, short-haired four-toed and 



