SummaryLaurel wilt, caused by the fungus Raffaelea lauricola and transmitted by the exotic ambrosia beetle Xyleborus glabratus, has killed members of the Lauraceae plant family throughout the southeast United States. A series of experiments were conducted to examine the effects of inoculum concentration on the development of laurel wilt in swamp bay, Persea palustris, and avocado, Persea americana. In each experiment, host plants were inoculated with aqueous suspensions of 102, 103, 104 or 105 conidia of R. lauricola, and plants were rated periodically for external symptom development (wilting and foliar dieback). At the end of experiments, plants were rated for internal symptoms of the disease (discoloration of sapwood) and assayed for R. lauricola on a semi‐selective medium. Symptom severity in swamp bay was significantly lower for the 102 treatment than at higher (103–105) concentrations, whereas 102 and 103 conidia caused less disease than 104 and 105 conidia in avocado. At the lowest inoculum concentration, 67% of the swamp bay plants and 20% of the avocados died by the time the respective experiments were terminated. The pathogen was recovered from a high proportion of the symptomatic sapwood of swamp bay (100%) and avocado (94%), and sapwood discoloration and recovery of R. lauricola from inoculated stems of swamp bay were highly correlated with recovery of the pathogen and symptom development in roots. Clearly, swamp bay and avocado are very sensitive to R. lauricola. The ability of only 100 conidia of this pathogen to kill these hosts suggests that few individuals of X. glabratus or other ambrosia beetles that carry low levels of the pathogen would be sufficient to transmit conidia that infect and lead to disease development. The results are also relevant to the development of disease‐tolerant host selections, as they indicate levels of the pathogen appropriate for use in screening plants for disease resistance.