VIRUS PARTICLES AND THEIR FUNCTIONAL ACTIVITY 333 



B. Assay by Local Lesion Count 



1. Plant Viruses 



The counting of local lesions as a means of plant virus assay was first done 

 by Holmes (1929), who observed the development of discrete necrotic spots 

 on leaves of Nicotianaglutinosa rubbed with tobacco mosaic virus. The method 

 has been extended to many other plant viruses and, wherever applicable, has 

 completely replaced the end-point dilution assay based upon systemic 

 infection of plants. The local lesion assay is simple in principle. Serial 

 dilutions of a virus suspension are rubbed with some suitable appHcator on 

 the surfaces of leaves of susceptible plants. Lesions appear in numbers related 

 to the virus concentration. The phenomenon has been compared to the colony 

 count method of assaying viable bacteria. The two procedures are not 

 strictly analogous, however, sitice a plot of the number of lesions as a function 

 of virus concentration is not linear for plant viruses over a wide dilution 

 range ia contrast to the bacterial colony count assay. The necrotic lesion 

 count falls off with increasing log virus concentration and this is explained in 

 part by the fact that the number of susceptible sites on any one leaf is limited. 

 It is necessary, therefore, to determine the relative potencies by comparing 

 counts obtained with two virus preparations appHed to the host plant leaves 

 in as reproducible a manner as possible. For maximum sensitivity the assays 

 should be carried out over a range in which the curve representing lesion 

 count dependency upon virus concentration has its steepest slope. 



Improvements in experimental design of local lesion assay of plant viruses 

 have greatly reduced the errors due to inhomogeneity of the test plants. In 

 particular, the method of inoculating one-half of a leaf with the unknown 

 sample of virus and the other half with the standard sample has been employed 

 for more meaningful comparisons of virus samples (Samuel and Bald, 1933), 

 Variation in comparative counts have been further minimized by the system- 

 atic distribution of inocula of several virus samples according to some 

 prearranged scheme such as the Latin square (Youden and Beale, 1934). By 

 this method, inocula of the various samples to be compared are distributed 

 equally often on each plant and at each leaf position. It allows the comparison 

 of as many virus samples as there are inoculable leaves on a test plant and 

 permits a more meaningful analysis of the significance of differences between 

 lesion counts of the various preparations tested. 



Whatever the design of the assay experiment, it is important that the 

 dilutions of virus preparations to be compared are such that they produce 

 approximately the same number of lesions per section of leaf inoculated. This 

 is made clear from an examination of the curve relating lesion counts to log 

 virus concentration. The flattening of this curve at higher virus concentrations 

 necessitates that comparisons be made in the lower concentration range where 

 the slope is steepest. 



