Geoffrey Smith and Mrs Haig Thomas 51 



the perfectly sterile hybrids, but nevertheless the preponderance of 

 males is striking enough. 



Now by referring to the column Dead in shell, it will be seen that 

 52 chicks died in the shell, and on the supposition that all these were 

 females it would bring the total number of females up to 191, which is 

 still considerably below the total number of males, viz. 238. But in a 

 few cases the sex of the birds dead in the shell was determined and in 

 four cases the sex was female and in five male. There is therefore no 

 reason for supposing that the 52 chicks dead in the shell were all females ; 

 on the contrary as far as the few figures go there is a slight preponder- 

 ance of males. Taking this result in conjunction with the result 

 obtained from the first series of figures, it is clear that a selective death- 

 rate during developement, at any rate after the rudiment of the chick 

 is visible, does not occur. The cause of the disproportion of the sexes 

 must be sought in the 1011 eggs out of 1426 which were either not 

 fertilised or else ceased to develope at so early a stage that no trace of 

 a blastoderui or chick could bo detected. If we accept the theory which 

 at the present time has the largest number of supporters, that sex is 

 determined at fertilisation by the nature of the gametes that make up 

 the zygote, we must suppose that in these cases of hybridism successful 

 fertilisation occurs most frequently between gametes which will form a 

 male zygote than between those which will form a female. This is 

 equivalent to selective fertilisation, and the facts recorded certainly 

 constitute very strong evidence in favour of this process occurring, a 

 process which has often been suggested as a possibility though never 

 directly proved. 



Summary. 



1. Evidence is brought forward to show that the sterility of the 

 male and female hybrid pheasants is due to abnormalities occurring at 

 the synapsis stage of the reduction division. 



2. The spermatozoa of the hybrids are either deformed, with 

 chromatic beads and thickenings on them, or else of abnormal size, 

 double-sized spermatozoa being of frequent occurrence as in hybrid 

 pigeons. The oocytes in the hybrid females fail to grow owing, probably, 

 to the early synapsis which should take place in the 19 day old embryo 

 having gone wrong. 



3. Partial assumption of cock's plumage by the female as the 

 result of ovarian degeneration is shown to be of fairly common occur- 

 rence. The oviduct in these cases may be either atrophied or greatly 



4—2 



