6 



THE AMERICAN NATURALIST [Vol. LII 



the fantail ; selection of these birds has not been made ex- 

 clusively in regard to number of feathers, but in regard 

 also to their size and shape, their regularity of distribu- 

 tion, their method of spreading, etc. It was a priori un- 

 likely that the race itself is homozygous for all of the fac- 

 tors that influence the number of feathers. How far the 

 results would depend on whether the maximum effects 

 are produced by a homozygous condition in several of 

 the factors, with heterozygous condition in others, would 

 be a point not easy to ascertain in a race that produces as 

 few offspring as does the pigeon. Nevertheless, the re- 

 sults give, I believe, pretty clear indications that the effects 

 are due to several factors, and they indicate, moreover, 

 that the failure to recover the extreme type of the fantail 

 in F 2 is probably only a question of insufficient numbers — 

 in fact, the fantail type has probably reappeared in F 2 , 

 though not in its most extreme form, even with the rela- 

 tively few F 2 pigeons that I have been able to get. 



The work has extended over several years, owing to 

 lack of suitable quarters in which to keep the birds and of 

 assistance to take care of them. They had to be removed 

 to and from Woods Hole each year, with the consequent 

 loss of young and disturbance of the regularity of habits 

 essential to a bird as conventional as the pigeon. 



The original stock was obtained from Dr. F. D. Solley, 

 of New York City, a well-known breeder of high-grade 

 fantails. Dr. Solley has also supplied me with informa- 

 tion as to the number of tail feathers in birds of his strain. 

 Unfortunately these numbers were not obtained until a 

 year after these particular birds had passed out of his 

 hands. He assures me they are typical, and the birds of 

 his stock that I saw when my parent birds were ob- 

 tained were closely similar in tail number, etc., to those 

 here recorded. 



The birds with which the original fantails were bred 

 to get Fj stock were ordinary birds. As they were not 

 pedigreed stock there is a small chance that they might 

 have contained factors of the fantail type, but this is 



