The repeated use of particular species for experimental oncology in fish in
creases their future value by accumulating background information for these
models and justifies the establishment of genetic stock centers. However,
the wide diversity that exists within the class Osteichthyes and Chondricht
hyes suggests that the ideal surrogate models for studying some types of ne
oplasms might be found among lesser known species. To help assess cultured
fish as surrogates for some other types of human neoplasia, we examined cas
es in the archives of the Registry of Tumors in Lower Animals and reviewed
reports in the literature. Spontaneous and induced neoplasms originating fr
om a spectrum of cell types were seen in more than 215 fish species commonl
y raised in aquaria or cultured for study among 69 families. prominent fami
lies include the Poeciliidae (live-bearers), Cyprinidae (carps and minnows)
, Cichlidae (cichlids), Cyprinodontidae (killifish), Characidae (tetras), A
drianichthyidae (medakas), Aplocheilidae (rivulins), and Salmonidae (salmon
and trout). The following are examples of potential fish tumor models that
have received less consideration than some others: papilloma and carcinoma
of the urinary bladder in oscar (Astronotus ocellatus); osteogenic neoplas
ms, peripheral nerve sheath tumors, and ependymoblastoma in coho salmon fin
gerlings (Oncorhynchus kisutch); and nephroblastoma resembling Wilms' tumor
in Japanese eels (Anguilla japonica).