458 GENETICS OF SOMATIC CELLS 



exploration of all possibilities, since the right experimental conditions cannot be 

 predicted and lack of success may depend on purely technical factors. As an alterna- 

 tive to the exposure of large populations of neoplastic or tissue culture cells to DNA 

 derived from variants with different markers, the injection of DNA directly into 

 eggs in cleavage may also be attempted. A possible advantage of this procedure would 

 be the use of multiple dominant markers, controlling visible phenotypic characteristics. 

 A mosaic individual would be the result of a successful experiment. 



The feasibility of a transduction type of approach depends on the way in which a 

 virus can be integrated with the genome of a somatic cell, its random or obligatory 

 localization if truly integrated, the amount of genetic information it can incorporate 

 from the host cell's genetic material before undergoing maturation, its ability to become 

 integrated with the recipient cell without preventing it from multiplying, and the 

 possibility of genetic exchange between the information carried by virus and the corre- 

 sponding sites of the recipient cell. While this may appear to be a chain of events 

 highly unlikely, it is not more unlikely than bacterial transduction which was hardly 

 predictable but has now become one of the routine methods of bacterial genetics. 



Since the only viruses capable of infecting higher cells without killing them are 

 certain tumor-viruses, 1080, 1350 they are the most likely candidates for such experiments. 

 Their possible integration with the genome of the host cell is still a matter of conjec- 

 ture, 1081, 135 ° although some form of integration is very probable. Nothing is known 

 about the localization of the integrated form of the virus. In this connection it may 

 be of interest to point out that two types of transduction are known in bacteria: genera- 

 lized or nonspecific and limited or specific. In the first type, discovered by Zinder 

 and Lederberg, 1470 certain phages are able to transmit any of the genetic characters 

 of the donor bacteria of the recipient strain. The probability of transmission of any 

 given character per phage is small. Phages capable of performing this type of trans- 

 duction do not appear to have an obligatory chromosomal localization. 1380 In the 

 second type of transduction, discovered by Morse, Lederberg, and Lederberg, 899 

 the prophage has an obligatory chromosomal localization and can transduce markers 

 located in its immediate vicinity at a low frequency, or, under special circumstances, 

 at a high frequency. The generalized type of transduction resembles transformation 

 of DNA with transmission of any genetic character in suitable systems and to competent 

 cells, while the restricted type is more similar to lysogenic conversion. The main 

 difference between the two mechanisms apparently depends on whether or not the 

 corresponding prophage occupies a specific position on the bacterial chromosome. 1380 

 It is impossible to guess the type of integration, if any, that tumor viruses may exhibit. 

 To remain on the safe side, it will be wisest to include as many marker differences as 

 possible in a system designed to detect the possible occurrence of transduction among 

 somatic cells. DNA viruses such as polyoma may be more hopeful than RNA viruses 

 from the point of view of integration, although in the absence of concrete knowledge 

 about the role of RNA as possible carrier of genetic information in higher cells it cannot 

 be excluded that RNA viruses may also become useful. Since most tumor viruses are 



