198 HEREDITY AND EUGENICS 



same year occurred a miscarriage of four boys, and 

 the following year a miscarriage of three girls. This 

 makes a total of fort3'-two births and miscarriages 

 from the one mother. In this remarkable case the 

 tendency to multiple births has been traced in four 

 generations. 



Hayden (1922) describes a case of a pure-bred 

 Holstein-Friesian cow in the herd of the Ohio 

 Agricultural Experiment Station which has had twin 

 calves five times out of seven by three different bulls. 

 A more remarkable case is described by Pearl (191 2), 

 who remarks that: ''It is well established that a 

 tendency to multiple gestation in normally uniparous 

 forms may be inherited." A Guernsey cow pro- 

 duced fourteen calves in her first eight pregnancies, 

 bearing triplets twice, twins twice, and single 

 young four times. The triplets were two females 

 and a male, the former probably free-martins. 

 The male closeh' resembled the mother, while the 

 females were of different colour and pattern like the 

 father. 



That there are two types of twins — identical or 

 uniovular, of the same sex, and showing exceptionally 

 close resemblance; and biovular or fraternal, show- 

 ing no greater resemblance than ordinary brothers 

 and sisters — has, of course, long been recognised. 

 Occasionally attempts have been made to show that 

 twins do not fall into such categories, but without 

 success. Fisher (191 9), from a mathematical treat- 

 ment of some earlv data of Thorndike, thinks it 

 necessary to conclude that differences in the degree 

 of resemblance between twins could be accounted 

 for by assuming that the egg divides into two 

 halves, each of which is fertilised by a different 

 spermatozoon. vSuch an hypothesis seems to have 

 no biological or experimental evidence in its favour, 

 while the evidence for the existence of both biovular 



