THE PHYSICAL PROPERTIES OF INFECTIVE PARTICLES 297 



that portion of a virus particle that is essential for infectivity, combined 

 with determinations of the size of the whole particle, can be expected to 

 elucidate some details of virus structure. 



Ill, Application of Physical Methods to Virus Studies 



A. Homogeneity and Purity as Related to Virus Identification 



When one prepares a suspension of particles from cells known to be 

 infected with a virus disease, a question of paramount interest is whether or 

 not there are virus particles in the suspension. It is in the nature of the word 

 "virus" that this question can be unequivocally answered only in the positive 

 sense, and then only after a biological assay has been made. A negative 

 answer does not prove the absence of virus particles; it may only show that 

 the host-cell system used for assay or the conditions of inoculation w^ere not 

 adequate to demonstrate infectivity. If the preparation proves to be infective, 

 a reasonable conclusion is that it contains virus particles which may be 

 experimentally "identified." Virus identification is a process that means 

 different things to different people, but for one interested in the physical 

 properties of viruses it means the establishment of a certain class of particles, 

 distinguished on physical grounds, as being associated with infectivity. 

 What is looked for, then, in establishing the identification of a class of 

 particles as viruses is a number of physical parameters that the suspected 

 particles are found uniquely to have. These may be size, shape, particle 

 weight, and electrophoretic mobility, for example. An attempt is then made 

 to correlate the existence of infectivity with those particles that have one 

 or more of these parameters. The degree to which a class of particles may be 

 said to include virus particles, and hence "identified," depends upon the 

 number of agreements between physical properties and infectivity, upon the 

 sensitivity of the biological assay, and upon the precision of the determina- 

 tions of physical characteristics. As we shall see later, ultracentrifugation, 

 diffusion, and electron microscopy are particularly apt methods for the 

 identification of virus particles. 



At some stage in the purification of a virus suspension it is usually possible 

 to pick out a certain class of particles, physically defined, and know with 

 some certainty that virus particles are within this class. Before making 

 determinations of the physical properties of the virus particles, however, it 

 is necessary to have some indication as to the homogeneity of the suspension. 

 Some would prefer to use the word "purity" in this context, and the advis- 

 ability of this choice will be mentioned below. The word "homogeneity" 

 itself, however, is not without ambiguity. It denotes a degree of uniformity, 

 but only in regard to the criteria used for detecting lack of miiformity. 



