ROUGH FUR. 



105 



INHERITANCE OF ROUGH AS OPPOSED TO SMOOTH. 



Castle (1905) demonstrated that all roughs differ from smooths by a 

 Mendelian unit-factor and that rough is dominant over smooth. The 

 writer's experience fully confirms this conclusion. 



In nearly 3,000 young recorded, smooth by smooth has never given 

 rise to rough (of series I) in spite of much rough ancestry, with one 

 possible exception. This exception was of a kind which was expected 

 and was being tested for when found. One of the smooth parents was 

 undoubtedly like rough E genetically. The case will be discussed later. 



TABLE 45. 



On the other hand, rough by rough has often given smooth. In the 

 cross of rough by smooths from stocks in which roughs have never 

 occurred, all or half of the young are rough. All wild cavies, for ex- 

 ample, are smooth and have only smooth descendants when crossed 

 with smooth guinea-pigs. They have numerous rough young in F x when 

 crossed with rough guinea-pigs. Thus it is clear that rough is domi- 

 nant. Table 45 is a summary of the rough crosses made by the writer. 

 Grades of roughness are ignored. No special attempt has been made 

 to obtain homozygous roughs. Male 4003, rough E, is the only one 

 which has been adequately proved homozygous. Male R197, rough 

 A (cross 46), is another which is probably homozygous. In the cross 

 Rr X Rr above, only matings are included in which both animals are 

 known to be heterozygous, either because of a smooth parent, or, 

 having had 12 or more young, because of smooth young. Tabulated in 

 this way, the expectation is not appreciably different from 3 rough to 

 1 smooth. Other litters of rough by rough are tabulated above. 

 Probably most of these are of the type Rr X Rr, although in some cases 

 there were no smooth young. In crosses Rr X rr, the only cases tab- 



