140 



The Journal of Heredity 



The same factor reduces orange to 

 cream, and the ordinary black-and- 

 orange tortoiseshell to a maltese-and- 

 cream tortoiseshell. The effects are 

 very similar to those produced by the 

 dilution factors of mice and rabbits 

 and the factor thus fits in well with 

 class las. 



A very different kind of dilution is 

 present in the Siamese cats. These 

 are born white but become a sort of 

 fawn color with dark chocolate on the 

 ears and feet. The eyes are deficient 

 in pigment and appear blue. Bateson^ 

 noted the similarity to the Himalayan 

 rabbit. The mode of inheritance is 

 not known, but the appearance strongly 

 suggests that the variation is homolo- 

 gous with imperfect albinism as found 

 in rats, rabbits, guinea-pigs, and dogs 

 and belongs in class lb. If this is true, 

 blue-eyed Siamese and the ordinary 

 blue-eyed white variety are variations 

 of very different kinds. The first is 

 near the extreme in a dilution or albino 

 series; the second, the extreme in a 

 piebald series. 



TORTOISESHELL AND ORANGE 



The first attempt to explain the curi- 

 ous correlation between tortoiseshell 

 color and female sex in cats was made 

 by Doncaster,® who suggested that 



Black 9 X Orange & 

 Tortoise 9 X Black & 

 Tortoise 9 X Orange cf 

 Orange 9 X Black cf 

 Orange 9 X Orange cf 



Black 



13 

 14 

 6 



Females 

 Tort. 



51 

 22 

 45 

 16 

 3 



there is a difference in dominance, in 

 the sexes, in the opposed pair of char- 

 acters, orange and black. Thus he 

 supposed that orange is completely 

 dominant over black in males, but im- 

 perfectly dominant in females, which, 

 therefore show a patchwork of orange 

 and black. Little^ made crosses which 

 led him to advance the hypothesis 

 that the orange factor is linked with 

 the sex-determining factor in heredity, 

 a mode of inheritance which had been 

 discovered, in the mean time, by Don- 

 caster in the moth Abraxas. He as- 

 sumed that females are homozygous 

 for the sex-determining factor, males 

 heterozygous. Thus females may be 

 of any of the three types, EE, Ee, or 

 ee, black, tortoiseshell, and orange 

 respectively, while males can only be 

 of the types, E- or e-, black and orange. 

 This hypothesis was also adopted by 

 Doncaster^ in his next paper as explain- 

 ing the main facts, but he noted the 

 existence of rather frequent discrep- 

 ancies between expectation and obser- 

 vation. The following table gives the 

 results which he obtained from breeders. 

 Tabby and maltcsc are included with 

 black, the various kinds of tortoiseshells 

 are combined, and some cases called 

 doubtful are omitted. The discrep- 

 ancies are in italics. 



Orange 



53 



Black 



51 

 32 



45 



Males 

 Tort. 



1 

 1 

 1 



40 



22 discrepancies among 263 females. 

 3 discrepancies among 294 males. 



Orange 



37 

 58 

 20 

 48 • 



Doncaster advanced several hy- 

 potheses to explain the discrepancies, 

 e. g., imperfect sex-linkage, difficulty 

 in distinguishing dilute tortoise from 

 maltese, and, later, non-disjunction. 

 Whiting^ found it possible to produce 

 all grades of tortoise from nearly self 

 orange to nearly self black, and sug- 

 gested that occasional overstepping of 

 the limits between the colors, due to 

 cooperation of independent minor fac- 

 tors for extension or restriction of 

 black, would account for the aberrant 

 kittens. Ibsen'" advanced an hypo- 



theses invoh'ing a high degree of link- 

 age between a factor for tortoise as 

 opposed to yellow and a factor for 

 black as opposed to yellow, but was 

 obliged to fall back on WHiiting's 

 explanation in some of the cases. 



DATA FOR FEMALES LESS RELIABLE 



The writer is inclined to accept 

 Whiting's explanation in the main. 

 There is, however, a difllculty with all 

 attempts at explaining the discrep- 

 ancies which should be pointed out. 

 In Doncaster's data there arc man}* 



