178 R. W. SCHLESINGER 



UV-virus. Subsequently, it was shown by Burnet and Lind (1954a), and by 

 Baron and Jensen (1955), and Gotlieb and Hirst (1956), through use of 

 inactivated and active viruses with distinct genetic markers, that reactivation 

 did indeed occur and that the inactivated parent contributed markers to 

 combinant progeny particles. Of special interest in connection with the 

 problem of interference is the observation by Gotlieb and Hirst (1956) that 

 such interaction occurred even when the interval between inoculations of 

 inactivated (UV-M-) and active (W+) virus was extended to 16 hours. 

 Moreover, the yield of recombinants under these conditions was surprisingly 

 high. 



Burnet (personal communication) finds that under similar conditions the 

 effect depends on whether heated M + and active WS~ (recombination, no 

 interference) or heated WS _ and active M+ (interference, no recombination) 

 are used. This suggests to him utilization of different processes for the two 

 phenomena. It is clear that critical distinction between partial interference 

 and recombination in homologous systems can be made only through genetic 

 analysis of the viral yield. 



4. Autointerference and the Genesis of "Incomplete" Influenza Virus 



The subject of incomplete virus production is fully discussed in Chapter 

 VI. The phenomenon requires comment here because it has been interpreted 

 as a manifestation of autointerference in the sense that simultaneous 

 infection of a cell with infectious and noninfectious particles leads to libera- 

 tion of immature particles (von Magnus, 1951a,b, 1952). This concept 

 presupposes (a) that the liberated noninfectious HA ("incomplete" virus) is 

 the equivalent of a normal development stage in the genesis of infectious 

 virus; (b) that its premature release depends on multiplicity of infection of 

 single cells; (c) that interference here takes the form of arrested viral develop- 

 ment, not of mutual exclusion. 



Several cogent reasons have been cited for looking upon the genesis of 

 incomplete virus in different light. Cairns and Edney (1952) suggested that 

 multiplicity of infection is not required to obtain a high proportion of non- 

 infectious virus in first-cycle yields. Burnet et al. (1954) were led to conclude 

 that there were various degrees of incompleteness and that some forms were 

 capable of limited reproduction. Henle and associates (Finter et al., 1955; 

 Paucker and Henle, 1955a,b; Liu et al., 1956), working with partially 

 inactivated virus as seed inoculum, also offered evidence supporting the view 

 that cells infected only with inactivated viral particles yielded newly 

 produced noninfectious hemagglutinin. One may add to these points the 

 following considerations: (1) strain-specific variations in the ease with which 

 incomplete virus can be obtained in eggs (von Magnus, 1954; Fazekas de St. 

 Groth and Graham, 1954); (2) the restriction of multiplication of some viruses 



