REPRODUCTION OF VIRUSES: A COMPARATIVE SURVEY 559 



Here, lysogeiiic bacteria lose by mutation the ability to produce normal 

 mature phage, without losing some of the prophage-controlled properties, 

 such as immmiity to lysis by superinfection or production of phage-con- 

 troUed antigens. The mutations to defectiveness occur in the prophage itself, 

 and the nondefective prophage form may be restored by back mutation. 

 AVith some phages the defect leads to incomplete maturation. Lysis will then 

 result either in production of no recognizable phage elements, or of fragments 

 of phage coats, or of some particles that carry the genetic defect ( Appleyard, 

 1956). 



The notable fact is that the defective prophages, being genetically com- 

 petent in other respects, but incompetent to determine production of infec- 

 tious virus, have lost their "viral" aspect. They have become operationally 

 indistinguishable from any other fragments of genetic material of the cell. 

 Yet, we know the exogenous origin and the potential transmissibility of these 

 genetic determinants, revealed in some cases by their back mutations to non- 

 defectiveness. Since we have independent evidence, from transduction, that 

 most or all elements of the bacterial genome are transferable from cell to cell 

 if a suitable viral vehicle is available, we are led to ask how many of these 

 genetic elements either possess or can acquire by mutation the potentiality 

 to determine their own specific incorporation into a viral vehicle formed 

 under their own control. That is, we ask whether all portions of a cell genome 

 might become viruses and whether in so doing they would manifest an ever 

 present potentiahty, or acquire a novel cytomorphogenetic function, or 

 recover a function that had been lost by mutation. 



There are in bacterial genetics a number of situations that can be inter- 

 preted in terms of special genetic elements or "episomes" (Jacob and WoU- 

 m^an, 1958), with the abihty to behave at times as chromasomal elements, 

 at other times as units midtiplying vegetatively in the bacterial cell. Pro- 

 phages may be considered as a category of such episomes capable of assuming 

 an effectively transferable form. Other episomes might conceivably acquire 

 this capability by mutation. 



As we interpret phage infection as genetic parasitism, we identify phages 

 more and more closely with wandering portions of the cell genome. More 

 generally, we must ask what role infective heredity has played and may still 

 be playing in the evolution of genetic systems (J. Lederberg, 1952; Luria, 

 1953). 



III. Multiplication of Tobacco Mosaic Virus 

 A. RNA as tJie Initiator of Infection 



Little is known about the multiplication of tobacco mosaic virus (TMV) 

 at the cellular level, but several lines of evidence are relevant to our dis- 

 cussion. 



VOL. 1—37 



