SUMMARY: The effect of dilution on numbers of local lesions produced by plant viruses was tested graphically and statistically for compatibility with contrasting hypotheses. Experimental results are incompatible with the hypothesis that lesions are produced because of chance encounters between single virus particles and susceptible regions of a uniform type, and that variations in response to similar inoculation between different leaves or plants occur solely because of variations in numbers of such regions. The results are compatible with the hypothesis that individual susceptible regions vary in susceptibility so that different doses of virus are needed to produce a lesion (the variations being such that the logarithms of minimal effective doses are normally distributed). The second hypothesis is, therefore, more probably correct, but the first can be fitted to experimental results by introducing various auxiliary assumptions, such as the existence of qualitatively different virus particles and susceptible regions. Conclusive evidence is unlikely to come from mathematical treatment of existing results, but only from a new experimental approach.
BaudJ. G.1937a; The use of numbers of lesions for comparing the concentrations of plant virus, suspensions. I. Dilution experiments with purified suspensions. Ann. appl. Biol 24:88
BaldJ. G.1937b; The use of numbers of lesions for comparing the concentrations of plant virus suspensions. 11. Distortion of the dilution series. Ann. appl. Biol 24:56
BaldJ. G.1937c; The use of numbers of lesions for comparing the concentrations of plant virus suspensions. IV. Modification of the simple dilution equation. Austr. J. exp. Biol. med. Sci 15:211