The Seed Pathology Committee of the Brazilian Seed Technology Association performed, in 1984/85, a comparative seed health testing for detection of the most important pathogenic fungi, aiming to select a suitable seed health testing at national level, which showed uniform figures within statistical limits for all participant laboratories. The same cotton seed sample was sent to 16 Brazilian laboratories, enclosing ?preliminary instructions? concerning to all over uniform steps to be followed by everyone. Nine out of the 16 laboratories carried out 4 methods corresponding to 4 variable factors: seed with and without lint, and incubation at 22 and 28°C. However, only Fusarium spp. were present in the work sample. It was clearly demonstrated that the results for incubation at 22°C and for delinted seeds did not present interaction with laboratories. This points out the last factors as the most appropriated for the Brazilian laboratories. However, we must have in mind that this is valid only as for a first approach, and only for Fusarium spp.