Thirteen cotton (Gossypium barbadense L.) genotypes were evaluated for Fusarium-wilt incidence, under greenhouse conditions, in 2008 growing season. The genotypes were divided into 5 distinct groups, i.e. resistant, moderately resistant, moderately susceptible, susceptible, and highly susceptible. The genotypes showed considerable variation in healthy seedlings, which ranged from 0.00% on genotype 491/2002 to 90.08% on genotype 507/2002. A total of 13 fungi were isolated from the nonsterilized seeds of the 13 genotypes. The isolated fungi were Alternaria alternata, Aspergillus flavus, A. niger, Aspergillus sp., Chaetomium sp., Cladosporium sp., Fusarium moniliforme, F. oxysporum, Nigrospora sp., Penicillium sp., Rhizopus stolonifer, Stemphylium botryeosum, and Trichoderma sp. Genotypes 27/99 and 72/99 yielded the highest number of fungi (8 fungi), while 31/99 yielded the lowest number (3 fungi). The other genotypes yielded a number of fungi ranged from 4 to 6. Rhizopus stolonifer was the only fungus, which was isolated from all the tested genotypes. The mean percentage of fungal recovery from seeds of the 13 genotypes showed that A. flavus (24.77%), A. niger (60.46%), Penicillium (18.15%), and R. stolonifer (65.38%) were the most dominant fungi isolated from the seeds. Other fungi occurred at frequencies ranged from 0.15 to 9.69%. Data for healthy seedlings (dependent variable) and frequencies of the fungi isolated from the seeds (independent variables or predicators) were entered into a computerized stepwise multiple regression analysis. Using the predicators supplied by stepwise regression, a six-variable model was constructed to predict healthy seedlings. This model showed that the differences in healthy seedlings were due largely to the effects of R. stolonifer, Cladosporium, F. oxysporum, Nigrospora, F. moniliforme, and A. alternata, which collectively accounted for 96.10% of the total variation in healthy seedlings- that is, the total variation in wilt incidence. The model also showed that R. stolonifer followed by Cladosporium were the most important seedborne fungi contributing to the variation in healthy seedlings. As far as we know, the results of the present study demonstrated, for the first time, that seed mycoflora play an important role in modifying the reaction of cotton genotypes to FOV. Therefore, it is suggested that the role of seed mycoflora should be considered more than it has been in the past for understanding the variations, among cotton genotypes, in resistance or susceptibility to Fusarium wilt disease.