Experimental transmission of the Stetsonville, Wisconsin, U.S.A. source of transmissible mink encephalopathy (TME) to outbred Syrian golden hamsters resulted in two distinct syndromes, termed hyper (HY) and drowsy (DY), that diverge by the third hamster passage. The syndromes differed with respect to clinical signs, incubation period, brain titre, brain lesion profile and pathogenicity in mink. HY hamster TME had an incubation period of 65 +/- 1 days and was characterized by clinical signs of hyperaesthesia and cerebellar ataxia. Lethargy and the absence of hyperexcitability or cerebellar ataxia were representative of DY hamster TME which had an incubation period of 168 +/- 2 days. At endstage, HY and DY infected animals had brain titres of 10(9.5) LD50/g and 10(7.4) LD50/g of tissue, respectively, indicating that the replication kinetics of these two strains is different. Hamster TME passaged back into mink revealed that only DY retained mink pathogenicity. This suggests that the DY agent is the major mink pathogen in the Stetsonville TME source that is also pathogenic in hamsters after a long incubation period. The HY agent is likely to be a minor component of the original TME mink brain that replicates more rapidly than DY agent in hamsters, but alone is non-pathogenic in mink. The presence of the HY and DY strains of agent that retain their biological characteristics on repeated hamster passage in the Stetsonville TME source requires that the informational molecule encoding these transmissible agents has the capacity to account for this biological diversity.