Changes in dNTP pools have been studied by a number of investigators, in a wide range of cell types. The in vitro pertubations in dNTP pool levels induced, in particular, by deoxynucleosides which act as allosteric modulators, are not totally consistent with current 'in vitro models' of ribonucleotide reductase function. This problem has also been addressed by Henderson et al. (1980) who stress the profusion of such models. Possible explanations, apart from the technical problems of the range of different experimental conditions (e.g. concentration of modulator used, time of incubation, etc.) for the various cell lines include: Modulators presumably have unpredictable 'network' effects by inhibiting or stimulating many other enzymes involved in the de novo and salvage synthesis of purines and pyrimidines. It is possible there are two separate forms of ribonucleotide reductase, one specifically reducing CDP/UDP, the other ADP/GDP. This, in particular, would explain the lack of decrease in dCTP levels after elevation of the dATP pool. There may be variations in ribonucleotide function which in vivo are cell specific, e.g. in thymic-derived compared with non-T-cell types. Peculiarities of T-cells include: Their ability to elevate their dNTP pools on exposure to very low exogenous concentration of deoxynucleoside. This may reflect very low rates of dNTP catabolism. The biological response of T-cells to elevation of the dATP or dGTP pool is reflected by a G1 block compared to an S phase block in cell-cycle progression in non-T-cell lines. The possibility that, in thymic cells, ribonucleotide reduction is restricted to ADP/GDP while pyrimidine dNTPs are synthesized by salvage pathways. As well, possible variation in the pool localization of dNTPs depending on production by either de novo or salvage synthesis could produce dNTP pool changes not clearly in accord with in vitro models. Clearly, the solution to these problems (although not easy) requires systematic comparative study, using cells of various origin (particularly T vs non-T), of dNTP pool responses to deoxynucleoside modulators, with an attempt to explore the factors described above. However, in the detailed pursuit of such an analysis the concept, that these variations in the control of nucleotide metabolism in T and non-T-cell systems may reflect quite significant differences in growth control and cell-cycle progression, should not be lost.