82 ME. W, BATESON ON COLOUR-HEREDITY [May 26, 



whites. The types are in fact definite, and cannot be built up by 

 cumulative selection. 



I am indebted to Miss Douglas for much information as to the 

 varieties of rats and for the loan of specimens. She tells me that 

 rats coloured otherwise than the varieties named are exceedingly 

 rare. Irregvdarly piebald or spotted rats occasionally occur, but 

 she has tried recently to obtain such rats from fanciers without 

 success. In her experience the type 6 breeds true or nearly so. 

 Of the blacks examined by me this type had less of the brown hair 

 than type 7. 



A striking feature appears from the rat-evidence, namely the 

 absence of yellow, blue, chocolate, and indeed most of the varieties 

 so familiar in fancy mice. On superficial examination, the 

 colour of a wild rat is not very greatly different from that of a 

 wild mouse. In rabbits also the yellow as well as the black forms 

 are common. Yellows or yellow-and- whites are also familial' in 

 guinea-pigs, fowls (buff", and " pile "), and pigeons. Miss Douglas 

 has heard of a pair of cream-coloured rats, but otherwise I can 

 find no records of any kind of yellow in the fancy. As blacks 

 are so easily produced by resolution in the rat, the absence of 

 the corresponding yellow and chocolate is remarkable. One is 

 tempted to inquire whether the existence of black gametes does 

 not suggest that yellow or at least chocolate gametes must also 

 exist. The problem of their disappearance raises many important 

 questions as to selective union between gametes, and others too 

 elaborate to develop now. As there are no yellows, so also there 

 are no chocolates. 



Another noteworthy fact is the complete absence of blue rats. 

 This particulai- stage in the diminution of the amount of dark 

 pigment is well known in mice, rabbits, cats, and several birds, 

 but it is unknown in rats. There is of course no question that 

 such forms would have been preserved if they had been seen by 

 fanciers. Either yellow or blue rats would be worth several 

 pounds. We may take it therefore that these particular resolu- 

 tions, or perhaps mutations, cannot be produced by any of the 

 means by which they have been produced in other forms. Con- 

 ceivably, if some distinct species were crossed with our fancy rats, 

 some of these forms might be created. ' Similarly there are no 

 " Himalayan " rats, i. e. pink-eyed with patches of colour (blackish 

 or yellow), forms well known in rabbits, guinea-pigs, and in the 

 " Japanese" waltzing mice. 



To sum up the evidence as to rats, we have clear proof of the 

 segregation of certain types of gamete — the albino, the black-and- 

 white, and the grey-and-white, though the ratios in which they 

 are produced by heterozygotes are not yet determined. Further, 

 there is proof that certain of the colour-types exhibit definite 

 valency {Werthigkeit of Tschermak) and dominate over each other 

 according to a regular system. Of the other colour-types one, viz., 

 type 2, is almost certainly a definite heterozygote form, and is 

 probably incapable of being made into a pure race. 



