Living tubers were inoculated by a standard method with small quantities of the soil to be tested and held under conditions suitable for infection. The number of dry‐rot lesions that developed was regarded as giving a measure of the soil infectivity; this number, expressed as a percentage of the total number of inoculations, was termed the ‘Infectivity Index’ of the soil.In these experiments, 91% of the lesions were caused either byFusarium caeruleumor by a few otherFusariumspecies of which much the most common wasF. avenaceum.In most instances it proved possible to distinguish the lesions caused by these two groups of fungi in tubers of the variety Doon Star by the colour of the rotted tissues as seen on cutting; isolations showed that 85% of the separations made in this way were correct.Data are given indicating the sensitivity and consistency which may be expected from the m