We have explored aspects of a suggested relationship between the expression of simian virus 40 tumor-specific transplantation antigen (TSTA) and tumor antigen (TA). A unique rat embryo cell line transformed by a temperature-sensitive A mutant that loses TA during exposure to the nonpermissive temperature (A28-RE) was found to lose TSTA. On the other hand, a typical control tsA-transformed cell line (A239-MB) expressed both TA and TSTA at the non-permissive temperature. TA in lysates obtained from A239-MB cells was found to be three to four times more thermolabile by covwt-mb) when incubated at either 33 or 40 degrees C. These data complement previous reports using TA from lytic infection and are consistent with the suggestion that TA is virus encoded. In contrast to TA, which even in wild-type-transformed cells was completely destroyed in less than 10 min at 50 degrees C, TSTA, assayed in vivo by tumor rejection, and tumor-specific surface antigen(s) TSSA) defined by an in vitro cytolytic assay, were thermostabile. Even after 24 h of incubation of extracts of 50 degrees C, high levels of TSTA and TSSA activity were present. Since these surface antigens when obtained from cells transformed by tsA mutants were also thermostabile, they could not be distinguished from the wild-type antigens. These results (i) indicate a coordinate expression of TA and TSTA in transformed cells; (ii) confirm that TA is virus encoded; and (iii) confirm that tha antigenic and immunogenic determinants that characterize TA and TSTA activities are distinct. However, the possibility that TSTA, like TA, is of viral rather than cellular origin is not excluded.