Over the last decade, research has shown that the antifolate effects of the drug, methotrexate (MTX), are mediated by its polyglutamate metabolites. The polyglutamates are known to be responsible, directly or indirectly, for inhibition of both thymidylate and purine synthesis. Ultimately we are interested in formulating a biochemical kinetic model useful for exploring various hypotheses of purine inhibition by MTX, but before this is possible, the kinetics of MTX cellular transport and of its polyglutamation must be described. Accordingly, the polyglutamation kinetics of methotrexate (MTX) in MCF-7 human breast cancer cells have been formulated mathematically. The model accounts for glutamation and hydrolysis kinetics up through the pentaglutamate level, increased synthesis of dihydrofolate reductase following exposure to drug, reversible tight-binding to reductase, and membrane transport of all the drug polyglutamates. The glutamation, hydrolysis, and efflux parameters have been determined from fits to experimental MTX polyglutamate uptake and efflux data. The preferred substrate for folypolyglutamyl synthase in the intact cell has been shown to be MTX diglutamate, on average being 2 to 3 times as reactive as either the parent drug or the triglutamate. Hydrolysis rate constants exhibit no clear trend with chain-length because of the large uncertainty of each parameter estimate. However, the efflux of MTX polyglutamates from MCF-7 cells does show a trend with chain-length, decreasing with increasing length as expected. The model also shows quantitative agreement with the fraction of MTX polyglutamates found still to be bound to reductase in MCF-7 cells following 24 hrs of efflux.