The morphological features of iridociliary epithelial tumors in 100 dogs and 17 cats were reviewed. Twenty-seven cases were in either Golden Retrievers or Labrador Retrievers. Affected globes were stained for light microscopy with alcian blue, periodic acid Schiff (PAS) and hematoxylin and eosin stains. Selected tissues were examined by immunohistochemistry for vimentin, desmin, cytokeratin, S-100, neuron-specific enolase (NSE), and glial fibrillary acid protein (GFAP). The presence or absence of hyaluronic acid was recorded by staining with alcian blue before and after digestion of the tissue with hyaluronidase. Canine tumors were divided into papillary and solid tumors based on the pattern of growth. Twenty-eight of 57 papillary tumors exhibited invasive behavior including eight of the 57 which showed infiltration of the sclera. Twenty-nine of 43 solid tumors were invasive including 13 of 43 with scleral invasion. Tumors with scleral invasion were designated adenocarcinoma. Tumors of both types could be pigmented or nonpigmented and often contained smooth basement membranes reminiscent of the inner membrane of the nonpigmented ciliary body epithelial cell. All of the feline tumors were nonpigmented and 14 of 16 feline tumors were solid and two of the tumors were papillary. Eighteen of 20 canine tumors and three of four feline tumors stained positive for vimentin. Cytokeratin stain was positive only in a few of the highly aggressive tumors. The finding of pigmented epithelial cells, thick, smooth basement membrane structures, positive staining for vimentin, S-100, and NSE as well as hyaluronic acid deposition were considered to be features which define iridociliary epithelial tumors in dogs. The positive staining for vimentin and NSE are highly specific markers which help to characterize iridociliary epithelium and distinguish this tumor from metastatic epithelial tumors. The finding of solid nonpigmented tumors with small epithelial cells packeted by thin PAS-positive membranes staining positive for vimentin were considered significant features defining iridociliary epithelial tumors in cats. Follow-up information on survival and cause of death was obtained on 43 canine cases and only two feline cases. The average follow-up interval in dogs was 25 months and only two dogs died with lesions that could have been due to metastasis although neither was confirmed. We concluded that neither iridociliary adenomas nor adenocarcinomas is likely to metastasize.