ARTICLES 595 



recorded as frequently or as fully as families showing the other 

 type of abnormal proportion. This may be because such 

 families rarely attained the eminence which would lead to 

 records being left, or because they were only left as " pedigrees " 

 which make no pretence of showing more than the chief members 

 of the line. 



In point of fact, out of the large number of genealogies 

 examined, only one showed a preponderance of females, while 

 six showed a preponderance of males in excess of the normal. 



(3) A family, to be of any use, had to show the descendants 

 of a number of the female members, so that comparisons could 

 be made with the ratios of the whole family. This sounds a 

 trivial point, but actually it is of great importance, because, 

 while genealogies show all the traceable members of a family, 

 pedigrees, as such, only show the direct line of descent of a 

 person, or persons, and only incidentally is it possible to get 

 further records of any large proportion of people figuring in a 

 pedigree. The great majority of family tables available are in 

 this latter form, including, of course, all the ** Visitations," 

 which at present constitute almost our only source of family 

 history. It should be understood, however, that only gene- 

 alogies have been used for the purposes of this investigation, 

 and the figures are, therefore, as full and as accurate as it is 

 possible to get them. 



(4) With a view to avoiding inaccuracy, as far as possible, 

 parts of genealogies extending back into the Middle Ages have 

 not been included, and, as a merely arbitrar}'' date, no records of 

 before a.d. 1600 have been used. 



A word must be said with regard to the method of dealing 

 with the figures obtained, and of drawing conclusions from the 

 totals. It has not been possible to pursue the ordinary methods 

 in use ; that is to say, it has not been possible to compare the 

 actual number of male-bearing sons found descending from male- 

 bearing fathers with what there would have been if there was no 

 inheritance of the characteristic. This is the normal method 

 of the determining whether a characteristic is inheritable or not, 

 but it is not applicable in this case for the following reasons : 



(i) It was hardly possible to decide definitely from the 

 offspring of a single marriage whether or not an abnormal 

 tendency was present, because the fewness of the offspring 

 exaggerated the element of chance to an impossible extent. 

 Thus, if a man had four sons and three daughters, it would be 

 very rash to call him male-bearing ; yet if the whole family 

 contained 400 males and 300 females, it would be a perfectly 

 just assumption. 



(2) It was impossible to calculate what proportion of people 



