34 



The Journal of Heredity 



the inheritance could not be explained 

 short of two factors, as belt by belt 

 mic^ht produce solid black, and self by 

 self mi[;;ht produce belt. No doubt part 

 of the variation is developmental, and 

 not j.!;enetic, as has been suggested by 

 Simpson. ■• 



Another type of white pattern is found 

 in the irregular splashes formerly com- 

 mon on the sides of Poland-China, and 

 Bcrkshires. In a reduced form, this 

 tj'pe of white seems to be responsible 

 for the six white points on feet, nose and 

 tail of the present representatives of the 

 above breeds. A third type of white is 

 found in the solid white breeds, as the 

 Chesters, Yorkshires and German Edel- 

 schwein. The white in red roan Tam- 

 worth- Yorkshire crosses, and in blue 

 roan Sapphires, are additional kinds. 



THE WHITE OF YORKSHIRES 



The relations between the solid white 

 of Yorkshires and the black with white 

 ppints of Bcrkshires and Poland-Chinas 

 has been worked out by W. W. Smith. ^ 

 In his crosses, Fi was always solid white. 

 A Berkshire backcross gave thirty-two 

 white to thirty-three black, the latter 

 with more or less extensive irregular 

 splotching with white. A Poland-China 

 backcross gave nine white to twelve 

 black, the whites in this case showing 

 small spots of black in most cases while 

 the blacks were splotched with white as 

 in the Berkshire cross, F2 from the 

 Berkshire cross gave twenty pure white 

 to six splotched black. The evidence is 

 clear for a unit Mendelian difference 

 which determines a big difference in level 

 in a series from clear whi^.e through white 

 with black spots, to black with white 

 splotches. It is also clear that there are 

 independent subsidiary factors favoring 

 white in the Yorkshire and black in the 

 Bcrkshires and Poland-Chinas. One 

 point of great importance is the failure of 

 red spots to appear in F2 or the back- 

 crosses. There is no evidence that 

 Yorkshires tran.smit anything which 

 tends to change black to red and, indeed, 



the occasional small spots which may 

 appear on Yorkshires themselves are 

 black. 



When Yorkshire is crossed with a 

 solid red as Tamworth or Duroc, Fi is 

 again pure white. In F2 and in back- 

 crosses with the red (Spillman,^ Simp- 

 son^) there is clear-cut segregation 

 between red and white or nearly white. 

 The latter show more or less red roan. 

 If the writer understands Simpson cor- 

 rectly nothing approaching a solid black 

 ever ajjpears in any generation. Cur- 

 iously enough in the Duroc cross small 

 spots of black may appear but no red 

 sjjots. It has, been general to consider 

 the white of Yorkshire as a dominant 

 white, which is able to inhibit either red 

 or black, in short a factor of class lai. 

 The results cited above, however, show 

 great difficulties in the way of this view. 

 As we have just seen, white Yorkshires 

 neither transmit red in crosses with 

 black, nor black in crosses with red. 

 This fact at once distinguishes them 

 from the dominant whites of cattle, 

 horses and other animals. Further, 

 Yorkshire white does not act on color in 

 a way irrespective of its quality. In 

 crosses with black, the white aj^pears to 

 be the extreme in a spotting scries. In 

 the cross with red, it appears to be an 

 extreme roan and such spots as may ap- 

 pear are not red but black. 



ALTERNATIVE HYPOTHESES 



Spillman^ attempted to account for 

 such results as these on a system of three 

 allelomorphs like the polygamous factors 

 of Wilson in horses and cattle. White 

 is WW, black BB and red RR. This 

 will evidently explain the mam genetic 

 facts above, assuming that WB is white 

 with a tendency toward black spotting, 

 and WR white with a tendency toward 

 roan. It does not work so well in more 

 complex crosses, examples of which are 

 cited later. As an alternative hyjiothe- 

 sis, it may be supposed that white of 

 Yorkshire differs from red of Tamworth 

 for a reason wholly independent of that 



* Simpson, Q. T. I'M 4. 

 'Smith, W.W. V>1.\. 

 «Si)illman, W. J. 1906. 

 ^ Simpson, Q. I. and ]. P 

 «Spillm.'m, W. J. l')()6. 



Lor. n'l. 



Imcr. Breed. Ma^., 4:113-12.?. 



Set., 24:441. 

 1911. Amer. Breed, /lis. Rrl?., 7:266-275. 



Lnr. cit. 



