THE BIOCHEMISTRY OF PLANT VIRUSES 43 



characterizes a virus and differentiates it from nonviral material is its ability 

 to cause disease. Unfortunately, the most rapid of infectivity tests takes at 

 least 24 hours and then only under exceptional circumstances. Local lesions 

 usually take 3 days to develop, while systemic symptoms take a week or 

 more. Consequently, other methods are used whenever possible. Of these, 

 probably the most useful one is the ultracentrifugal examination. Most 

 viruses can be followed by this means after the first stage of concentration 

 has been passed, and usually viruses have a characteristic sedimentation 

 rate and very sharp boundaries. Such a test takes less than an hour. 



Spectrophotometry is also a very useful test. Most plant viruses absorb 

 with a maximum at about 260 m^u, and a minimum at about 240 m/x. The 

 extinction at the peak for virus solutions of 0.1 mg./ml. is usually between 0.2 

 and 0.8. Impurities usually remove the trough in the 240 m/x region. Pure 

 virus usually has very little absorption at 320 mfx, but when the larger plant 

 viruses come to be investigated appreciable absorption can be expected in 

 this region because of their scattering potentialities. 



Serological testing is also of considerable use in following a purification, 

 but the results may be complicated by the presence of noninfectious antigens 

 characteristic of the virus infection. As an example, if one were to follow 

 the purification of the Rothamsted strain of the tobacco necrosis virus (q.v.) 

 by infectivity measurements and by serology, one might end with two 

 different substances. 



1. Infectivity measurements 



Most infectivity measurements are of a low precision, and consequently 

 it is of little use to make dilutions at finer intervals than 10-fold. Comparisons 

 are made on opposite halves of leaves, because the leaves of a batch of plants 

 may vary a 100-fold in susceptibility. It is also important that the inoculations 

 of control and experimental dilutions should be made more or less simultane- 

 ously because plants show a diurnal change in susceptibility, which is much 

 greater than one might imagine. 



Provided that the number of lesions per half leaf is not excessive, preferably 

 less than one hundred, there is a roughly linear relationship between the 

 logarithm of the local lesion comit against the logarithm of the weight of 

 virus per milliliter which may be used for interpolation. Usually this is not 

 warranted unless large numbers of plants are used. 



With plants which do not give local lesions the only possible solution 

 is to inoculate batches of 10 or 20 plants with each dilution. Because of the 

 variation between the plants this method is very inaccurate and not to be 

 compared with the apparently similar titration of viable organisms using 

 10-fold dilutions and inoculating several samples of each dilution. 



