Yield and fruit size were determined for 49 strawberry (Fragari ×ananassa Duch.) genotypes during a 7 year period, in soils prepared with and without preplant soil fumigation using 2 methyl bromide: 1 chloropicrin (wt/wt). Strawberries were grown in alternate years, with the nonfumigated treatment representing the first, second, third, and fourth strawberry crop cycles initiated without soil fumigation. Highly significant (P < 0.01) effects of soil fumigation treatment were present for yield in a combined analysis over all years; fumigation increased yield by 41% over nonfumigated soils in the first nonfumigated cultivation cycle and by 68% to 74% for subsequent nonfumigated cycles. Fruit size was less affected by soil treatment but increases due to fumigation (2% to 18%) were significant (P < 0.05) in the third or fourth nonfumigated crop cycle. Genotypic variances were highly significant in the combined analysis, whereas geneti × fumigation interaction variances were significant only for fruit size and contributed <8% of the total phenotypic variance for either trait. Genetic correlations were rg = 0.77 and 0.92, respectively, for yield and fruit size treated as independent traits across soil fumigation environments. There was no evidence for genes that confer specific adaptation to nonfumigated soils, or that these genes emerge as important contributors to the phenotypic variation as the soil environment deteriorates with repeated cultivation of strawberry in nonfumigated soil. Chemical names used: trichloronitromethane (chloropicrin).