ON EXPERIMENTS IN INHERITANCE. 299 



' Purity,' lie says, ' is dominance over latency.' Cuenot, on the other 

 hand, supported by Wilson and favoured by Lock, suggests that pure 

 yellow-bearing germ-cells of both kinds are indeed formed by heterozygous 

 yellow mice, but that there is a selective fertilisation ; so that a yellow- 

 bearing ovum and a yellow-bearing sperm are either mutually repellent 

 or mutually sterile, though capable of fertile union with germ-cells bearing 

 other colours. Or it may be that yellow is due to the association of 

 several factors, as appears to be the case in certain colours of sweet peas 

 investigated by Bateson, Saunders, and Punnett, and reported on by 

 them to the Royal Society. Castle, discussing various alternatives, 

 commits himself to none. 



Further experiments are evidently most desirable. 



Objects of the Research. — This brings me to the more precise aims of 

 the present investigation, which are : — 



(a) To repeat Cu^not's experiment with another strain of yellow 

 mice. 



(b) To test analytically, by as many different kinds of matings as 

 possible, the gametic constitution of a series of yellow mice. 



(o) To endeavour to produce a race of homozygous yellow mice. 



Progress of the Research \at present in initial stage]. — With the 

 above objects I obtained, during December 1907, certain mice from 

 sundry fanciers. These comprised (a) yellow mice, (b) mice of other 

 colours to cross with the yellows. 



(a) The yellow mice were three clear yellows (Castle's terminology), 

 of which one was a male and the other two females, these mice are of an 

 orange-yellow colour, their hairs containing yellow pigment only ; there 

 are also two yellows, which contained, in addition, a good deal of 

 chocolate pigment associated with the yellow (Castle's sooty yellow) ; these 

 I call chocolate yellow. I use the term sootg yellow to include all mice 

 which, commencing life -yellow, develop in their hairs later any other 

 colour or colours in addition to the yellow. It will be convenient to 

 follow the line of descent from a pair of the clear yellows. Clear yellow 

 (J No. 15 mated with clear yellow ?18 gave an Fj generation of one 

 clear yellow (^4.5) and three blacks. Of this F, generation the clear 

 yellow member mated with a chocolate gave two clear yellows and three 

 blacks ; this seems to imply that it contains black recessive. The same 

 mouse ( $ 4.5) mated with a blue gave three yellow youngsters (of which 

 one grew to a clear yellow adult, while the other two developed a con- 

 siderable quantity of dark pigment, producing a blue colour associated 

 with the yellow) and two blues ; this seems to imply that it contains blue 

 recessive. That is to say, (?45 seems to contain both black and blue in 

 the recessive condition. The black was probably derived through the 

 father, (?15, since a black youngster resulted from a mating of the latter 

 with a sooty yellow ( ? 44) which had been shown to contain recessive 

 chocolate. The source of the blue is not so clear ; it is of course possible 

 that it depends upon the presence or absence of a ' diluting ' factor, blue 

 being, as pointed out by Bateson, a diluted form of black. 



Returning to d'lS and ? 18, the clear yellow commencers of this 

 pedigree, it may be noted that $ 1 8 has not littered further. Additional 

 matings of ^ 15 are : 



(i.) With clear yellow ? 1 9, giving two yellows in which much dark 

 pigment developed in the adult, producing in one of them a considerable 



