148 ALICK ISAACS 



same time as standard virus particles, there is a sharing of the available 

 nucleic acid among the progeny. If the incomplete virus is given sufficiently 

 long before the standard virus, there is time for viral interference to become 

 established and a diminished yield of total virus results. The suggestion that 

 in multiple infection of cells the nucleic acid may be shared among the viral 

 progeny is intended as an analogy to ideas of viral recombination. 



Another method of producing incomplete virus is to use as seed standard 

 virus which has had its infectivity greatly reduced by incubation for a few 

 days at 37 or 22°C. (Henle, 1953; Horsfall, 1954; Paucker and Henle, 1955a). 

 In an experiment described by Paucker and Henle (1955a), a preparation of 

 virus with an I /HA ratio of 10 6 - 1 was incubated at 37°C. for 5 days, when its 

 infectivity had dropped more than 100,000-fold, the I /HA ratio being 10 * 9 . 

 When this was used undiluted as seed the progeny had an I /HA ratio of 

 10 2 - 3 ; seed diluted 1/10 gave progeny with an I /HA ratio of 10 2,9 , while seed 

 diluted 1/100 gave progeny with an I/HA ratio of 10 6 * 4 . These findings show 

 that hemagglutinating virus of low infectivity is produced by seed virus which 

 has been inactivated at 37°C; there is a striking resemblance to the production 

 of incomplete virus by von Magnus's method in that only concentrated 

 inocula produce this effect, while on dilution of the seed 1/100 standard virus 

 is produced. One difference is that when incomplete virus is prepared by von 

 Magnus's method, passage of incomplete virus results in progeny with a 

 lower I /HA ratio, whereas inoculation of virus inactivated at 37°C. leads to 

 progeny with a slightly higher I /HA ratio than the seed. This finding argues 

 against the possibility that incomplete virus formation in von Magnus's 

 experiments is caused by accumulation of virus spontaneously inactivated at 

 37°C, as does the fact that the incomplete virus released in the first 2 hours 

 of the growth cycle has a low I /HA ratio, similar to that formed later in the 

 cycle (Finter et al., 1955). At the moment, there is no evidence on the nucleic 

 acid content of progeny from virus inactivated at 37°C. ; this would be an 

 interesting point to investigate. If the nucleic acid were low, the results 

 might be explained by postulating that during prolonged incubation at 

 37°C. the virus nucleic acid becomes sufficiently damaged to prevent the 

 normal synthesis of viral nucleic acid, but, provided multiple infection of 

 cells occurs, it is still able to take part in the synthesis of a lower quota of 

 nucleic acid with a resulting yield of incomplete virus. 



The finding by Fazekas de St. Groth and Graham (1955) that, in eggs 

 treated with metaperiodate, influenza virus growth results in the production 

 of hemagglutinating virus of low infectivity may be due to the action of 

 aldehydes on the virus rather than to an action on the cells, since a similar 

 effect could be demonstrated on virus in vitro (Liu et al., 1956; Schlesinger and 

 Karr, 1956). Fazekas de St. Groth and Graham (1954) also investigated 

 production of incomplete virus by different strains of influenza virus, using 



