158 R. W. SCHLESINGER 



under which work on viruses is carried out. The extracellular, mature virus 

 particle is looked upon as an exogenous, infectious agent to which a complex 

 host reacts much in the same way as to other parasites. On the other hand, 

 once within the individual host cell, a viral particle behaves in the manner of an 

 intracellular organelle, imposing on the cell new biological, metabolic, and 

 genetic functions, including altered response to superinfecting virus. Thus, 

 all conceivable components of the systemic immune response — specific or 

 nonspecific — as well as strictly cell-bound, virus-induced modifications enter 

 into the evolution of the immune state. 



In this framework, viral interference occupies a special place. Its importance 

 is underlined by the fact that mutual exclusion among phages and immunity 

 of lysogenic systems are the only forms of acquired immunity found in 

 infected bacteria, and that both depend entirely upon the modifying effect 

 on the individual cell of a primary infection. Systemic immune mechanisms, 

 so important in animals, do not enter into the picture. Yet, even within the 

 narrow confines of the cosmos made up of Escherichia coli B and the T phages, 

 the occurrence of mutual exclusion is subject to control by various factors. 

 Among these are the relative serological and genetic compatibility of the two 

 phages involved and variations in timing and dosage. Information obtained 

 from studies on cell populations has been made more meaningful by analyses 

 of single cell yields. Technically, it is possible to integrate bacterial immunity 

 to infection as a logical phase into a unified concept of the virus-host relation- 

 ship. 



In the study of animal viruses, a unified concept of the infectious process is 

 lacking. Hence, interference, even when observed imder conditions not 

 obscured by systemic immune mechanisms (i.e., in cell cultures or 

 embryonic tissues), remains so far largely a descriptive term which cannot yet 

 be defined in terms of its relation to the viral life cycle. The need for separate 

 treatment of this subject is born of our ignorance. 



In the following pages, we try to establish criteria toward a unified defini- 

 tion, collate many and diverse experimental observations, and summarize the 

 meager evidence which may clarify the dynamics of the phenomenon but by 

 no means the underlying mechanisms. 



II. Definition of the Problem 



A. Qualitative Criteria for a Definition of Interference 



1. The Relation to Virulence 



At the qualitative level, interference expresses itself as protection against 

 injury, disease, or death, or, in general, against manifestations of "virulence." 

 Unless the superinfecting virus is pathogenic for the host system involved, 

 interference in this sense cannot be observed. 



