594 SCIENCE PROGRESS 



proportions of the sexes, the second question arises : Is the 

 tendency vested in the male or in the female, or is it lodged 

 indiscriminately in both ? In agricultural circles there is a 

 strong belief that this tendency is vested in the female. This 

 would mean, however, that the female is the heterozygous 

 parent, because, so far as can be seen, the homozygous parent 

 can have no determining effect on the sex of the offspring, 

 owing to all the gametes being constitutionally the same. But 

 farm animals, along with man and Drosophila, are believed to 

 be heterozygous for sex in the male ; so that once again there 

 is a distinct breach between practice and science. It was partly 

 to secure information on the agricultural aspect of the subject, 

 and partly to ascertain whether or not hereditary numerical 

 inequality between the sexes fitted in with the view that in 

 man the female is homozygous, that the following figures relating 

 to human genealogies were collected. 



At the outset it was proposed to work with animal pedigrees, 

 but it soon became self-evident that the very close inter-breeding 

 which is the vogue with our pedigree animals interweaves the 

 strains beyond hope of distinguishing the potentialities of one 

 family from those of another. When, for instance, a bull was 

 mated to his own mother, who in turn sprang from a cross 

 between sister and brother, it is more than difficult to say 

 whether any preponderance of one sex in the offspring of the 

 bull came through the male or the female line. In the human 

 genealogies dealt with below very few cases of marrying back 

 into some other branch of the same family occurred, so that 

 this factor of complexity may be considered as being negligible. 

 Also, it is fair to assume that, on the whole, the various members 

 of the families married normal people ; people, that is to say, 

 whose tendencies were towards equal numbers of the sexes. 



A word may now be said with regard to the selection of the 

 genealogies. To be of any use, these had to present several 

 well-defined features. The chief of these are enumerated below : 



(i) The family had to be large enough to eliminate the 

 element of chance as far as possible. Purely arbitrary limits 

 had to be fixed upon as regards this, and, just as a matter of 

 convenience, no family containing less than 150 members was 

 considered. Naturally, the larger the family, the better. 



(2) From the genealogies which could be found answering 

 to the above requirement, all those which showed a distinct 

 preponderance of male over female were selected. It appeared 

 easier to work with this characteristic than the reverse one. 

 There must be famihes with a preponderance of female over 

 male — no doubt every reader is familiar with one or more 

 examples of this happening through generations — but for a 

 variety of reasons such families do not seem to have been 



