HEREDITY IN SOMATIC CELLS 379 



that thev reproduce more slowly and soriK'times do not lyse the host cell 

 upon their release. There are even semistable associations made in tissue 

 culture between polio virus and resistant cells in which the cells may be 

 killed as they vield new virus, or they may multiply, carrying the virus 

 forward into future generations. In some ways this condition resembles 

 that of lysogenic bacteria; many other provocative examples exist among 

 plants and animals. 



In mice there is the classic example of the mammary tumor virus. 

 When transmitted at coitus or via the milk at nursing, it may induce a 

 tumor, depending on the genetic susceptibility of the host and the 

 endocrinological stress experienced. It may propagate in males and 

 genetically resistant females without giving rise to cancer. Sometimes 

 this virus arises spontaneously in stocks thought to be free of it, raising 

 the question whether it may be of genetic origin in the same sense as is a 

 prophage. Another example is Shope's papilloma virus which can be ex- 

 tracted from warts on the skin of the wild American cottontail rabbit; 

 when applied to the scarified skin of the same kind of rabbit, it produces 

 new papillomas and more virus. But in the domestic rabbit of European 

 origin it gives rise to tumors which become cancerous with age. In these 

 tumors the virus causing them cannot be demonstrated, although par- 

 ticles with similar antigens can be found. If the virus is injected into 

 cottontails in which papillomata have been induced by tar, cancerous 

 growths result. In these again the virus is not found, but virus antigen 

 is present. X-ray treatment causes the virus to be produced. This virus 

 seems to be able to enter a provirus stage, which in many ways resembles 

 that assumed by prophage. Many other cases of masked viruses exist; 

 more and more tumors are being found to be associated with viruses and, 

 as tissues from the body are cultured, more and more are found to pro- 

 duce viruses with time. The main factor limiting our ability to discover 

 this kind of association is the requirement for a tester system in each 

 case. It is beginning to appear that most cells contain proviruses and 

 that they may be normal features of the genotype. 



In Drosophila there are cases of infection with viruslike entities. One 

 causes a pathological sensitivity to high concentrations of CO,. It is not 

 contagious but is infective by organ transplantation or by the injection of 

 body fluids from one fly to another. After injection there is an eclipse 

 period of about one day, whereupon the virus multiplies and spreads 

 throughout the body. Before it has been stabilized, females transmit it 

 to some of their offspring, but males transmit it to none; after stabiliza- 

 tion, females transmit it to all, and males to some of their offspring, as 

 though its concentration had increased. This pattern is one of maternal 

 inheritance. Indeed, the virus is transmitted to offspring independently 



