Forty-eight spring barley genotypes were evaluated for deoxynivalenol (DON) concentration under natural infection across 5 years at Harrington, Prince Edward Island. These genotypes were also evaluated for Fusari- um head blight (FHB) severity and DON concentration under field nurseries with artificial inoculation of Fusarium graminearum by the grain spawn method across 2 years at Ottawa, Ontario, and one year at Hangzhou, China. Additionally, these genotypes were also evaluated for FHB severity under greenhouse conditions with artificial inoculation of F. graminearum by conidial suspension spray method across 3 years at Ottawa, Ontario. The objective of the study was to investigate if re- actions of barley genotypes to artificial FHB inoculation correlate with reactions to natural FHB infection. DON concentration under natural infection was positively correlated with DON concentration (r = 0.47, P < 0.01) and FHB incidence (r = 0.56, P < 0.01) in the artificially inoculated nursery with grain spawn method. There- fore, the grain spawn method can be used to effectively screen for low DON. FHB severity, generated from greenhouse spray, however, was not correlated with DON concentration (r = 0.12, P > 0.05) under natural infection and it was not correlated with DON concentration (r = -0.23, P > 0.05) and FHB incidence (r = 0.19, P > 0.05) in the artificially inoculated nursery with grain spawn method. FHB severity, DON concentration, and yield were affected by year, genotype, and the genotype × year interaction. The effectiveness of green- house spray inoculation for indirect selection for low DON concentration requires further studies. Nine of the 48 genotypes were found to contain low DON under natural infection. Island barley had low DON and also had high yield.