NONCHROMOSOMAL GENES 267 



gration with their host by attaching in a mapable position to the bac- 

 terial chromosome. They can also exist in an infectious and unintegrated 

 form not attached to the bacterial chromosome. When the phage is un- 

 integrated, it replicates rapidly free of cellular regulation and eventually 

 kills the cell. The replication rate of F is not known, but indirect evi- 

 dence suggests that it can multiply out of synchrony with the host cell. 

 The term episome was coined initially to generalize from the behavior of 

 these two kinds of infectious particles. 



Sex Ratio in Drosophila 



An example of a trait with considerable evolutionary implications, 

 determined by an infectious particle, is "sex ratio" in Drosophila, in 

 which female flies give rise largely or exclusively to female progeny. 

 The trait is transmitted only by the female. Any exceptional male 

 progeny of sex-ratio females do not transmit the trait to their offspring. 



Examination of fertilized eggs laid by sex-ratio females reveals three 

 classes: normal developing fertilized eggs, and abnormal dying fertilized 

 eggs in about a 1 : 1 ratio, and in addition, considerable numbers of un- 

 fertilized eggs. The abnormal dying fertilized eggs are the best source 

 of the sex-ratio agent, which can be transmitted by injecting cytoplasm 

 from these eggs into normal females. Injected females begin to give 

 broods consisting entirely or largely of females after a latent period of 

 10-14 days. The female sex-ratio progeny from injected females do not 

 differ from other sex-ratio females, transmitting the factor to their off- 

 spring with the same high efficiency. 



The abnormal fertilized eggs probably represent the male zygotes, that 

 is the XY zygotes, and their susceptibility to the sex-ratio agent can 

 explain the absence of males among the progeny of sex-ratio females. 

 This hypothesis is strongly supported by the fact that cytoplasm from 

 dying eggs, which converts normal females to sex-ratio, is capable of 

 killing even adult males. 



There are strain differences in susceptibility to the sex-ratio agent: 

 some strains are more favorable than others for establishment of the 

 agent after infection, and some strains are so susceptible that XX tissues 

 are damaged in the female germ line, so that unfertilized eggs are killed 

 and females are consequently sterile. These differences in strain suscep- 

 tibility probably reflect gene-determined influences on rate of replication 

 of the infectious agent. 



The sex-ratio agent is particularly interesting because of the evidence 

 that different tissues of the fly respond differently to it. It can only 

 be found in the female germ line in infected flies, perhaps because it 



