The influence of the route and the frequency of IL 2 administration on the ability of IL 2 to induce the growth of activated T cells in vivo was evaluated. Initial pharmacokinetic studies confirmed that i.v. injection of IL 2 results in a relatively high peak serum concentration, but a short serum half-life. By contrast, i.p. or subcutaneous (s.c.) injection of IL 2 results in a lower peak concentration but a prolonged serum half-life. The bioavailability of IL 2 administered by these routes was assessed by measuring the in vivo growth of adoptively transferred T cells that had been previously cultured long-term with IL 2, because the growth of such cells in vivo has been shown to be proportional to the dose of IL 2 administered. The results demonstrated that i.p., s.c., or i.v. administration of IL 2 each resulted in marked donor T cell growth in vivo. Thus, IL 2 can function in vivo at sites distant to the sites of injection. In addition, the magnitude of T cell growth in vivo varied dependent on the route of IL 2 administration and correlated with the length of time IL 2 was detectable in serum, rather than the peak level achieved (i.e., IL 2 inoculated i.v. had the highest peak concentration but was least effective). As suggested by these findings, dividing the total dose of IL 2 into frequent low-dose injections was more effective in inducing T cell growth in vivo than was dividing the total dose of IL 2 into less frequent higher-dose injections. These studies confirm the great potential for IL 2 to induce the growth of activated of T cells in vivo and demonstrate that the rate of T cell growth reflects not only the dose but also the route and timing of IL 2 administration.