Between 1956 and 1991, 8349 samples from dogs and cats were received for investigation of suspected dermatophytosis, and 1368 (16 per cent) yielded positive cultures. Cats had a significantly higher proportion of positive cultures (26 per cent) than dogs (10 per cent), and of these Microsporum canis accounted for 92 per cent in cats and 65 per cent in dogs. The other isolates were diverse but mainly sylvatic dermatophytes, and M gypseum was isolated on only four occasions. Different breeds of dog and cat had significantly different prevalences of infection, with pedigree and long-haired cats, and Jack Russell and Yorkshire terrier dogs having a particularly high proportion of positive cultures. Animals less than one year old appeared to be predisposed to infection, but there was no apparent sex predisposition and no conclusive evidence of any seasonal variation in the incidence of the disease. In comparison with the results of dermatophyte culture, examination under Wood's lamp had a positive predictive value of 90 per cent and a negative predictive value of 94 per cent in determining M canis infection, and direct microscopy had positive and negative predictive values of 93 per cent in determining the presence of dermatophytosis. However, cultural examination alone was insufficient for the diagnosis of dermatophytosis owing to the occurrence of false positive and false negative results.