Technical Abstract: To complement existing control strategies, grape growers in humid climates desire cultivars with resistance to downy mildew, caused by Plasmopara viticola. Numerous disease resistance screens of diverse Vitis germplasm have previously been conducted to identify downy mildew resistance, but ratings of named cultivars were inconsistent and identities of resistant individuals in wild species were not typically provided. Because inconsistencies between studies could be due to differences in pathogen genetics, we used controlled inoculations of a single pathogen isolate onto two leaf ages of 883 Vitis accessions and compared these results to natural infection in a five-fold replicated vineyard of 80 Vitis accessions in 2006 and 2007. Of the accessions rated in both assays, 16.2% were resistant to a single isolate but susceptible in the vineyard. Otherwise, there was a good correlation of ratings between the field assay and the rating of older leaves (r=0.62-0.71). Five accessions from V. cinerea, V. labrusca, and V. x champinii averaged zero severity in both vineyard years, yet some individuals of V. cinerea and V. labrusca were moderatley or highly susceptible in the field. Similarly, although significant differences in mean severity separated V. vinifera, V. hybrid, V. riparia, and V. labrusca for single isolate inoculations (from susceptible to resistant) notable within species variation was identified for all well-represented species. Resistant individuals were identified in most species with the prominent exceptions of V. vinifera and V. acerifolia. Single-isolate resistance ratings in 2006 corresponded well (94.6%) to 2007 ratings using a separate isolate collected from the same vineyard. Categorizing the ratings for this and previous studies, ratings infrequently corresponded among previous studies (31.9%) as well as between previous studies and single-isolate (34.9%) or vineyard (46.4%) ratings. These results highlight important factors for downy mildew resistance screens: leaf age, pathogen genotype(s), and host species and genotype. The results further underscore the importance to breeders of uniform testing in multiple environments.