The existence of possible differences among tobacco (Nicotiana tabacum L.) genotypes with regard to the activity of nitrate reductase and the changes in the activity of this enzyme at various plant growth stages were studied. The optimum conditions for the enzyme's activity and the relationship between the in vivo and the in vitro assays were evaluated. Ten genotypes and 19 F1 generations, obtained from crossings among the former, were used. The inheritance of the enzyme's activity in seedlings of four crossings was also studied. Plants were grown in nutrient solutions and in the soil under greenhouse conditions. Various rates of nitrogen fertilizer were applied in the growth media. Results showed that the optimum conditions for the nitrate reductase activity in tobacco were : temperature = 30-32.degree.C, pH of the incubation medium = 7.8, duration of the assay for the in vitro and in vivo = 20 and 60 min, respectively. The activity was more fully exhibited when ab undant nitrates were present in the growth medium. All genotypes showed the same response with respect to the enzyme's activity to nitrate concentration in the nutrient media. The activity was found 6-10 times greater when measured by the in vitro assay as compared to the in vivo one the correlation between the two methods was higher in the seedling stage and became lower as plants matured. Differences among genotypes with regard to the activity of nitrate reductase were identified. These differences were greater when the in vitro assay was used. In general, the activity was low at the early growth stages, then it rose reaching a maximum value during the development of the floral axis and fell abruptly during full flowering and the formation of the first seeds. The results suggested that young seedlings were most suitable to be used for evaluating the mean activity of the enzyme at the various stages of tobacco's biological cycle. It was suggested that the activity of nitrate reductase in tobacco was under genetic control. No heterotic phenomena were found with regard to the activity of this enzyme. No clear classes of the activity were observed in the parents, F1, F2 and in the backcrosses B1, and B2 to facilitate the determination of the number of genes participating in the expression of the activity. The percentages of variance which were due to the additive and dominant gene action were dependent on the genetic constitution of the parents. Broad-sense heritability was high (0.67 to 0.93). Narrow-sense heritability was lower (0.17 to 0.71) and depended on the genetic constitution of the parents. Both types of heritability were higher in the plants which were fertilized with nitrogen than in the plants which were not fertilized.