RESULTS OF HYBRIDIZING RING-DOVES. 313 



In the case of the canaries, Durham and Marryat found that 

 their formula (i. e., that given on pp. 310-1 of this paper) did 

 not explain the occurrence of two FI "black-eyed" females when 

 the male bird was pink-eyed. The same difficulty appears with 

 the ring-doves as maybe seen on comparing Table II. where it 

 will be seen that three blond FI females appeared. The difficulty 

 is no greater in the scheme just presented in this paper, and it 

 would seem to be due possibly to the occasional failure of the 

 tw sperm to produce the w effect. 



It may be noted also that neither scheme explains the great 

 excess of FI males in the ring-dove and canary experiments. 

 The occurrence of a preponderance of males is a common and 

 well known result of crosses between bird species, for which no 

 satisfactory explanation has been given. It is of course con- 

 ceivable that female-determining sperms may in such cases 

 encounter difficulties in the egg of another species which may be 

 great enough to prevent the development of femaleness, more or 

 less frequently. (See Guyer, '090.) 



The gametic analysis is carried still further as follows, and it 

 may be noted that corresponding matings bear the same num- 

 bers. 



3. Blond FI ring-dove male X White ring-dove female. 



composition B cf (W 9 ) W 9 W (cf) 



B cf 



B, cf 



producing gametes W cf all W 9 



W Iw cf 



result Blond males. Blond females. 



B cf (W 9) B 9 (W cf ) 



White males. White females . 



W cf W (9) W 9 W 



See Table XII. of this paper where the birds crossed are 

 probably potentially the same as the forms which are implied by 

 this mating.) 



4. Blond FI ring-dove male Blond ring-dove female. 



composition B cf (W 9 ) B 9 B (cf) 



