290 J/r Batexons Revisions of Mendels Theonj of Heredity 



a division of the socond gcneration into thicc groups, of ponstitution and relative 

 frequcncy indicated by thc füriiuila 



GG + 2GG' + G'G' 



wliere the homozygotes GG are pink-eyed albinos, the homozygotes G'G' are pink- 

 eyed " with some colour in the coat," and the heterozygotes GG' have " dark eyes 

 " anil some colour in the coat." 



In thi.s fornmla either coat-colour and eye-colour togcther are trcated as 

 a single unit character, ur tliey are treated as a series of unit elements combined in 

 a "Compound allelomorph " which is transmitted to half the gerra-ceils of either 

 sex conipletely unchanged, and the Mendelian i)rini'i})le that Compound characters 

 are reduccd to their elements and redistiibuted in germ-cell formation is abandoued. 

 The formula gives 25 per cent. pink-e^'ed albinos, 50 per cent. dark-eyed mice 

 " with some colmir in the coat," and 25 per cent. pink-eyed mice with some colour 

 in the coat ; thc numbers given in Mr Darbishire's last Report, froni which 

 Mr Bateson deduced his formula, are 



18 i)ink-eyed albinos, 



3G clark-eyed with colour in the coat, 



17 pink-eyed with colour in the coat, 



and con.'^idering the small number of individuals, this record fits Mr Bateson's non- 

 Mendelian formula vcry well. Any Mendelian hypothosis, which should attempt 

 to account for the great variability in coat-colour by postulating a resolution of 

 the " Compound allelomorph " would lead to onc of two results, either of which 

 would be inconsistent with the experiments. 



For if we suppose the elements which represent coat-colour and eye-colour to 

 break np into oiily two coniponeuts during the formation of gametes by the hybrid, 

 tlien it both of these be supposed dominant over albinism, the number of pure 

 albinos, or of pure dominants in the second generation is reduced to one in sixteen, 

 a Proportion which is not in accord with Observation ; if ou the other band one of 

 the two colour elements be supposed recessive to albinism, we may get f^^ of the 

 second generation showing albinism ; such a proportion would fit the facts so far 

 obtained, but two out of every three such albinos should be dominant h^'brids, 

 producing some coloured young, and no such albinos have been observed by 

 Mr Darbishire or by any one eise. 



Mr Bateson's is thercfore the only pseudo-Mendelian formula which will 

 account for the observed distribution of eye-colour and albinism. But it is fatal 

 to the "great fact of gamctic purity," either in Mr Bate.^on's .sense or in Mendels. 

 Kor this "great fact" involves much more than is implied by the Statement that 

 the G'G' individuals are "pink-eyed, with some colour in the coat"; these in- 

 dividuals are said to be formed of gametes identical in Constitution with those of 

 their pure-bred waltzing grandparents, at least in respect of eye-colour and coat- 



