Eight tomato (Lycopersicon esculentum Mill.) genotypes were evaluated based on shoot dry weight for resistance to four isolates of Verticillium dahliae Kleb. race 2 in two greenhouse seedling experiments. The race 2 isolates, obtained from North Carolina, Brazil, and Spain, demonstrated no differences in pathogenicity on the eight lines tested, thus precluding the identification of a third V. dahliae race in this collection. However, highly significant differences in virulence were observed among the isolates. The Brazilian isolate was the most virulent. No tomato genotype showed resistance comparable to that conferred by the single dominant Ve gene to V. dahliae race 1. While all tomato lines were susceptible to all race 2 isolates tested, there were significant differences in susceptibility equal to differences in levels of resistance. IRAT L3, Morden Lac, Okitsu Sozai, and `UC82' significantly outperformed the lowest ranking line XXIV-a. `Earlypak 7', Morden Mel, and Philippine 2 performance was statistically indistinguishable from that of either the highest- or lowest-ranked lines. Genetic diversity in the host and pathogen and environmental conditions favoring the pathogen likely contributed to the genotype × isolate interactions observed in Expt. 1. These results suggest using diverse isolates when screening for improved race 2 resistance.