A field experiment with 12 sowing dates, ranging from February 2, 1989 to January 19, 1990, in monthly intervals, was carried out in a randomized block design, with four replications, with the objective to evaluate the development of Alternaria disease in three sunflower genotypes. The disease level was evaluated at 10-day intervals, up to physiological maturity. The interaction between sowing dates and genotypes was statistically significant; furthermore. the genotype Conti 621 showed less disease severity than VNIIMK 8931 and IAC-Anhandy. August sowing date showed, on the average, the lower area under disease progress curve (AUDPC) and disease index (DI), which should be a consequence of the low rainfall during this period.