SUMMARY1A distinction is drawn between the types of infection‐dilution series given by unpurified samples of the tobacco mosaic group of viruses and viruses less resistant to loss of virulence.2Distortion of the series for the tobacco mosaic group, when inocula were diluted with distilled water or buffered near the neutral point, was shown to exist more in the concentrated than in the dilute end of the series.3An acid hydrogen‐ion concentration was found to change the form of the dilution series.4The addition of plant juices to the inocula tended to produce two opposing effects on the production of local lesions, a stimulation and a depression.5Unpurified samples of viruses less resistant to loss of virulence than the tobacco mosaic group gave low values for lesions at high dilutions.6The addition of concentrations of the order of 0.01 and 0.1Mof neutral potassium phosphate‐phthalate buffer to unpurified suspensions ofXvirus depressed the numbers of lesions produced.7Added in various concentrations to purified samples ofXvirus, the buffer produced high values for lesions at a concentration of about 0.00005M, low values between 0.0001 and 0.001M, a maximum between 0.001 and 0.01M, and a decline at higher concentrations.8The presence of the buffer in inocula was shown to affect the dilution series ofXvirus.9A tentative hypothesis was advanced to explain the effects of salts on the production of lesions byX