526 



The Journal of Heredity 



More definite conclusions can be this by assuming that dun and yellow 



• reached in rej^'ard to dun. The fol- are rcspsctively the heterozyj^'otes of 



lowing table, which is a slight modifica- black and red with the rather uncommon 



Black Dun Red Yellow Br indie Number 



Black X black 71 5 18 1 5 97 



Xdun 29 45 11 13 1 82 



X red 43 4 43 4 7 239 



X vcllow 27 24 16 29 3 74 



X brindlc 45 2 29 2 22 181 



Dun Xdun (6) (6) (88) (16) 



Xred 18 13 23 46 54 



X vcllow 2 53 11 30 5 64 



Xbrindle 18 32 9 18 22 87 



Red Xred 3 1 86 9 2 131 



X vellow 5 ? 46 35 12 121 



X brindle 4 67 6 23 188 



Yellow X vellow 5 2 46 35 12 73 



X brindle 3 4 22 49 23 111 



Brindle X brindle 13 35 2 50 80 



tion of Wilson's table for herd book 2 cream-colored light dun which he puts 

 of the Highland cattle, shows some very in the same scries as black and red. As 

 interesting points. The progeny of an alternative hypothesis we may sup- 

 each color are tabulated by per cent of pose that dun and yellow differ from 

 the total, the total numbers being given iDlack and red respectively by an inde- 

 in the last column for reference. pendent dominant dilution factor of 



A consideration of this table certainly class lb. 



justifies Wilson's contention that duns ^0 Triple allelomorphs 



contain the factor by which blacks 5S , "-"j^ ^r, , 



,.,-. . - _,, ■' ... liK black KK red 

 difler from reds. The crosses mvolvmg bl dun RL yellow LL light dun 

 only red, yellow and bnndle produce (2) Two factor hypothesis 

 ver>' few blacks and duns, probably no ud D(J dd 

 more than can be accounted for by errors iji? cream dun dun black 

 in the records. The cross of black with Ee. .. .cream dun dun black 

 black produces 71% blacks, not much ee ....cream yellow red 

 below the 75% expected if blacks are (light dun) 

 heterozygous (Ee) which should gener- Probably the best criterion between 

 ally be the case in Highland cattle, the two hypotheses which is present in 

 Black by red or brindle produces but the data is in the results of the cross 

 little less than the expected 50%. But dun by red. On Wilson's view (1) this 

 in the crosses of black by dun or yellow should produce only blacks and yellows 

 only 297o and 27% black calves are (BL X RR = BR + RL). On hypothe- 

 produced which shows clearly that sis (2), duns, blacks, yellows and reds 

 factor E must lie present in some other should all be produced (EeDdXeedd = 

 color than black. This color can only EeDd+Eedd+eeDd+eedd). The re- 

 be dun (except perhaps for an occa- suits, 7 duns, 10 blacks, 25 yellows and 

 sional black brindle). Combining blacks 12 reds point strongly toward the 

 and duns as mostly of formula Ee and hypothesis of independence, although 

 reds, yellow, and brindles as mostly ee, the excess of yellows and reds indicates 

 the agreement between figures and ex- that some of the jmrents called duns 

 pectation become^' fairly good. were probably really yellows lacking 

 It is further evident from the table factor E. The color called light dun 

 that dun and yellow are closely related appears generally to be extreme dilu- 

 in some way. Black by yellow pro- tion of red. It produces very few blacks 

 duces many duns (24'/( ) and dun liy red and duns unless crossed with these 

 produces many yellows (469o) w'hile varieties. The evidence for a unit 

 neither color is produced in reliable Mendel ian difference between red and 

 numbers unless at least one parent is light dun with yellow as the hetero- 

 either dun or yellow. Wilson explains 7,ygf)te is not as clear cut as could be 



