H4 



Scientific Proceedings (112). 



In human families with their relatively small numbers, it would not 

 be at all surprising to find that all the sons in such families would 

 be hemophilic and that by the selective elimination of their 

 normal brothers an excess of hemophilic above the 1 : 1 ratio 

 would be produced. The same would hold true for color blindness. 



When a tabulation of the data available in Bulloch and Fildes 

 monograph on hemophilia and at the Eugenics Record Office of 

 the Carnegie Institution of Washington was made such an excess 

 was found to exist even after due allowance is made mathe- 

 matically for the one hemophilic male occurring in each family. 

 The excess of hemophilics over the expected is so great that the odds 

 are greater than one to a thousand million that it is due solely to 

 chance. The actual numbers are 551 observed to 457 expected. 

 The data available on parallel matings of color blindness are much 

 more meager but offer supporting evidence. The odds are greater 

 than 1 to twenty-six that the result is due solely to chance. The 

 observed number of color blind is 106 and the expected number 90. 



In both cases, therefore, in spite of the fact that both hemo- 

 philic and color-blind individuals are certainly no better fitted 

 for survival than their normal sibs under equal opportunities, there 

 is an excess of the abnormal types. This clearly suggests the inter- 

 vention of a sex-linked lethal factor which eliminates the otherwise 

 "normal" males in certain families and leaves an excess of abnormals. 



Families. 



Sex I 

 Males. 



latio. 

 Females. 



Ratio of Males 

 to 100 Females. 



Difference. 



Part males hemophilic 



413 

 1,070 



337 

 678 



122.55 ± 2.73 

 I57-8I ± 2.02 



35-26 ±3-39 

 10.4 X P.E. 



Part males color blind 



114 

 184 



100 

 119 



114.00 ± 4-4 

 154-02 ± 4.83 



30.62 ± 6.52 

 4.6 X P.E. 



Another characteristic of sex-linked lethal factors is, that in 

 such forms as man a decreased proportion of female offspring 

 should be produced in the matings in which no such excess of 

 affected males exists. If the families in which all males are 

 hemophilic or color blind, are contrasted as regards sex ratio, with 

 those in which part of the males are abnormal and part normal a 

 basis for comparison is afforded. Errors of classification, should 

 they occur, should militate against finding a significant difference- 



