62 R. MARKHAM 



V. The Tobacco Mosaic Virus 

 A. Introdiiction 



As has already been mentioned, the tobacco mosaic virus has been investi- 

 gated more intensively and for a longer time than has any other virus. This 

 is, of course, because it was discovered before any other virus (Iwanowski, 

 1892), is widespread in nature, is economically important, and above all is 

 extremely infectious and stable (Allard, 1942). In the last two qualities, 

 the tobacco mosaic virus, or, more exactly, group of viruses, is by no means 

 representative of viruses in general. Most plant viruses are far less suitable 

 for laboratory investigation, and it is, indeed, only by the happy circumstances 

 enumerated above in the case of the tobacco mosaic and a few other viruses 

 that our knowledge of these pathogens has advanced as much as it has. 



The virus of tobacco mosaic was first isolated in quantity by Stanley 

 (1935) who used a salt-precipitation technique for isolating the infective 

 agent from diseased plant sap. Before this work numerous attempts had 

 been made to purify this virus, a procedure which was largely confused with 

 the removal of plant material rather than the preparation of virus in con- 

 centrated suspensions, but there is little reason to doubt that several workers 

 had obtained reasonably pure preparations without realizing it (Brewer 

 et al., 1930; Barton-Wright and McBain, 1933). Apart from the fact that 

 assay of the infective material was not really feasible until the development 

 of the local lesion method (Holmes, 1929; Samuel and Bald, 1933) and that 

 methods for transmitting the infections were so primitive that accidental 

 infection must have been nearly as common as purposeful transmission, 

 the major factor that delayed the recognition of the virus as a substance 

 having definite chemical properties was that nobody expected that it could 

 be present in the quantities which can be obtained from infectious sap. As 

 we now know, some 10^ particles of virus are needed to cause a single infection, 

 so that the quantity of virus obtained from a kilogram of plant tissue may 

 be measured in grams instead of fractions of a milligram, as one might have 

 anticipated. The reason for this singular lack of infectivity of plant virus 

 particles is not completely understood even now. It may be, as some hold, 

 that only a very small proportion of each preparation is potentially infectious 

 — a view which, incidentally, casts doubt on the validity of much of the 

 recent exciting work which has been carried out on plant viruses. It may 

 equally well be due to a statistical effect reflecting the improbability of any 

 indi ;idual virus particle reaching a suitable site in a viable plant cell, and 

 then being able to initiate an infection. It may well be a combination of 

 both of these, but at the moment it is not possible to tell with certainty. 

 The "Host probable tiling is that the chance of infection is small, but that 

 not all particles are potentially infectious. However, it should be borne in 



