368 DON C. WARREN. 



the females and these exceptional individual ratios caused the 

 relative number of females in the female strain to be higher than 

 normal*. These unusual ratios can be readily explained by as- 

 suming that the female which Moenkhaus selected as the mother 

 of his female strain carried a recessive sex-linked lethal factor 

 and it will be pointed out later that all of Moenkhaus's data 

 substantiate, very precisely, this assumption. The existence 

 of such factors has been conclusively demonstrated by the recent 

 work of Morgan and his students. A recessive lethal factor is 

 any factor that brings about the death of the individual in 

 which it occurs, provided its effect is not counteracted by the 

 action of its normal allelomorph. Then if a lethal is sex-linked, 

 all males which get it will die for they cannot carry its normal 

 allelomorph since they possess but one X-chromosome. Since 

 all males receiving the lethal factor die, this factor is never trans- 

 mitted by the male and as a consequence the female can never 

 be homozygous for it. Therefore the lethal factor has no effect 

 on the female but she, by transmitting it to half of her sons, 

 causes their death. Since half of the males die, a 2 : i ratio will 

 result. A female carrying a lethal will transmit it to half of her 

 daughters and they will always be heterozygous for it since they 

 cannot receive it from their father. 



Then the female which Moenkhaus used as the mother of his 

 female line was probably a lethal female. It should be said that 

 Moenkhaus is not to be criticised for not considering lethals in 

 his paper, for lethals were not known till two years after his work 

 was published. The original female gave a ratio which was an 

 approximate 2 : i ratio. As mothers of the succeeding genera- 

 tions, Moenkhaus probably selected lethal bearing females 

 (excluding female 5 of generation 4). These females should 

 transmit the lethal factor to half their daughters and this expec- 

 tation is realized, for from the 31 matings made in the female 

 strain (excluding the offspring of female 5 of generation 4), 

 fifteen 2:1 ratios resulted. Those considered 2:1 ratios are 

 matings 3, 9 and II of generation i ; 7, 8 and 10 of generation 

 2; 3, 4 and 7 of generation 3; i, 7, 8, and 10 of generation 4; and 

 7 and 8 of generation 5. So the number of 2: i ratios obtained 

 would justify the assumption of the presence of a lethal factor. 



