IDENTICAL TWINS IN CATTLE. 5 



their body markings, tongue color and switch color are approxi- 

 mately the same for the four groups of cattle. The greatest dif- 

 ference is between the twins which are the males and the twins 

 which are the male and female. Here the difference favors the like 

 sex twins by about 5 per cent. It will be noted that the like sex 

 twins resemble each other between 3.43 and 4.87 per hundred more 

 than do the twins which are male and female. If the like sex 

 twins are compared for resemblances with the sister, sister com- 

 bination of column five, the difference becomes less. These facts 

 all point to not more than a low percentage of identical twins in 

 cattle if they exist at all. 



If the differences are compared and their significance deter- 

 mined, it will be seen that none of the combinations, female twins, 

 male twins, male and female twins or sister and sister, are signifi- 

 cantly different. In ' fact, the differences are at no time greater 

 than two times the probable error of such difference. If the whole 

 distributions are compared by the \ z method, it is seen that they 

 resemble each other closely. Such being the case, it can not be 

 argued that twins of identical markings, tongue color or switch 

 color exist in cattle. 



In the Armadillo, 1 Dasypus novcinc'mctits, when the 4 offspring 

 are known to be of monozygotic origin it has been shown that the 

 degree of resemblance between the members of the quadruplet is 

 very high. This resemblance is nearly twice as great as that which 

 exists between brothers of known dizygotic origin. It is conse- 

 quently fair to assume that the production of identical twins, in 

 the sense that they arise from a single egg, will increase the pro- 

 portion of identical offspring in like sex twins as compared with 

 twins of unlike sex, or sister and sister combination. The above 

 data have shown that the proportion of like sexed twins do not 

 differ appreciably in the proportions which resemble each other 

 from those which are of unlike sex or of sister, sister combination. 

 Such being the case, it follows that identical twins (monozygotic) 

 are rarely or never produced in cattle. 



The importance of this conclusion as related to the freemartin 

 problem has previously been discussed by F. R. Lillie 1 and need 

 only be mentioned here. Hart, Cole and others have assumed that 



i Loc. cit. 



