196 A. LWOFF 



responsible for fundamental differences. It happens that a number of small 

 filterable, infectious agents of plant and animal diseases share with bacterio- 

 phage those essential features which allow one to separate bacteriophage 

 from organisms and organelles. 



These infectious agents can be united into a special category for which 

 the following definition is proposed: strictly mtracellular and potentially 

 pathogenic entities with an infectious phase: (a) possessing only one type of 

 nucleic acid; (b) reproduced from their genetic material and multiplied in 

 the form of their genetic material; (c) unable to grow and to undergo binary 

 fission; and (d) devoid of a Lipmann system. It shoidd be noted that any one 

 of the characters (a), (b), (c), (d) is sufiicient to identify an uifectious entity as 

 a virus and that these characters are probably correlated and subordinated. 



The definition of viruses which has been proposed applies to any of the 

 phases of the life cycle — proviral, vegetative, infectious. If one wants to put 

 emphasis on the infectious particle, the following definition can be proposed: 

 viruses are infectious and potentially pathogenic entities, reproduced from 

 their genetic material and midtiplied as genetic material, unable to grow and 

 to undergo binary fission and devoid of a Lipmann system. 



The problem of the origin of viruses is not solved. Whatever their ancestor 

 might have been, viruses noiv differ from organisms as well as from cellular 

 organelles. 



An object, a category, or a concept has an individuality or a reality only 

 because of the existence of different objects, categories, or concepts. The art 

 of definition is founded on difi"erences as weU as on resemblances. A definition 

 has to exclude as well as to reassemble. Our definition of viruses is valid 

 only because, at the same time, it includes a homogeneous class of entities, 

 viruses, and excludes other homogeneous classes of entities, such as micro- 

 organisms and cellular organelles. Therefore, we do have the right to unite 

 viruses into a special class of objects. The concept of virus has thus been 

 estabfished on a firm ground and the term "virus" has, at last, a definite 

 meaning. 



Our position might appear somewhat too dogmatic. One can, of course, 

 point out that between "small" viruses and large micro-organisms, one finds 

 intermediary classes of large viruses and of small microorganisms. Taking 

 advantage of this situation, one can decide, as some people do, that viruses 

 are small microorganisms and that microorganisms are large viruses. This is 

 all right, but why then should some infectious agents be called viruses? 

 Because obviously they differ from bacteria, algae, fungi, and protozoa, and 

 because they differ from other microorganisms. If one tries to find out the 

 nature of the difference, one necessarily lands where we have landed. Either 

 viruses do not exist and we should stop utilizing the term "virus," or viruses 

 do exist and we have to state why something is a virus. 



