74 INHERITANCE, FERTILITY, AND SEX IN PIGEONS. 



rank. It is thus shown that if orientalis be mated with forms progressively more dis- 

 tantly related to it, the predominance of males progressively increases. It is further 

 shown that in inbred orientalis' there is an actual predominance of females. A 

 summary of these data is given in table 46. The data are taken from the breeding 

 records fully given in this and previous chapters. 



RECIPROCAL CROSSES OF ST. RISORIA AND T. ORIENTALIS. 



The amount of data for crosses of blond rings with Japanese turtles is not large, 

 but it coincides with the data for the very similar cross which has just been con- 

 sidered, particularly in the demonstration of a high degree of fertility, of great 

 length of life of off sirring, '" and of the linkage of sex and color in inheritance. The 

 details are presented in tables 47 and 48. 



PRIMARY HYBRIDS CROSSED INTER SE AND WITH PARENT SPECIES. 



Most of the possible crosses of Fi hybrids from white rings or blond rings and 

 Japanese turtles have been made. When back-crossed with the parent species 

 these hybrids usually prove somewhat more fertile than when crossed inter se.'^ 

 The results from 11 pairs are fully shown in tables 49 to 52. In every case — in 

 back crosses and in inter se crosses — fertility is low and the life-term is short. It 

 turns out, furthermore, that of 8 offspring A^•hose gonads were examined after death, 

 4 were found to be inadequately or abnormally sexed. An additional 2 or 3 of 

 such offspring (which have functioned, apparently normallj', as male or female) 

 are still alive. 



The details of these matings need not be further discussed here.^^ ^ reference 

 may be made, however, to table 51, in which two Japanese turtle males are shown 

 to have been fertile to about the same very limited degree with a risoria-orientalis 

 hybrid. It is of some interest to note that in these two matings the 4 eggs of 

 this female which showed some development were all the first egg of the clutch. 

 Again, in table 57, where a multi-hybrid is mated with a pure St. risoria female, 

 there are 5 pairs of eggs which gave rise to the two sexes. In all of these 5 cases 

 the first of the clutch produced a male and the second a female. It now seems 

 certain that fertility and sex problems in pigeons are closely bound up with the 

 matter of order of the individual eggs of the clutch (particularly, however, in non- 

 hybrids), and with the order, extent, and crowding of the clutches. When fertility 

 is very low, as in the series first mentioned above (table 51), the longevity is much 

 decreased; but if perchance a bird can live moderately long, as did Bl of pair 6, 

 it will often betray striking abnormalities; and these malformations are often 

 associated with sex. 



FURTHER BREEDING OF COMPLEX ORIENTALIS-ALBA-RISORIA HYBRIDS. 



In tables 53 to 62 are recorded the results of a long series of matings of hybrids 

 of the three species under examination. The data contribute in general to the 

 thesis that "fertility is a thing of all degrees." In particular they show: (1) the 



" Several birds nearly 11 years old from this series are still alive. 



" For four pairs testing absolutely infertile see list near close of Chapter II. The data left by the author do not 

 fully decide this point, but breeding of these hybrids mter se by the editor has made it clear that the degree of fertility 

 here shown for pairs 1, .3, 4, 9, and 10 is most unusual in inter se crosses. 



'^ The limitation of color by sex shown in these matings has already been described. 



