The metabolism of 2'-deoxyadenosine S'-phosphate (dAMP) by exponentially growing mouse fibroblasts (L-cells) in suspension culture has been studied. Cells incubated for four hours with 0.10 pmole [3H,32P]dAMP/ml in medium containing 9.6 pmole 31Pi/ml incorporate both activities at nearly linear rates into acid-soluble and acid-insoluble fractions. The 3H/32P value increased about 30-fold in each fraction during the incubation, indicating extensive dephosphorylation of dAMP. The DNA from treated cells was degraded enzymatically to 5'-mono-nucleotides, which were fractionated by ion-exchange chromatography. 3H was associated exclusively with dAMP (53%) and dGMP (47%). 32P, associated with all deoxynucleotides, was at a 20-fold higher specific activity in dAMP than in either dGMP or dCMP. The specific activity of [ 32P]dAMP incorporated into DNA in four hours was 24-fold greater than that of 32Pi in the cellular pool. In experiments in which cells were incubated with 32Pi plus or minus 0.10 @mole dAMP/ml, the specific activity of dAMP was slightly less than that of dGMP or dCMP, which were equal. These results suggest that the higher specific activity of [32P]dAMP in the DNA of cells after incubation with doubly-labeled dAMP was due to the intact penetration of some dAMP into the cells with its subsequent incorporation into DNA. Calculations, based on the amount of exogenously added [3H]thymidine incorporated into the cellular DNA in parallel cultures and the 32P of dAMP isolated from this DNA, suggest that 1% of the total dAMP residues incorporated during the 4-hour incubation were derived directly from intact, exogenously added dAMP.We have previously reported that upon incubation of L-cells with the toxic nucleotide 9-P-D-arabinosyladenine 5'-phosphate (araAMP), small quantities of the compound entered the cells without prior dephosphorylation (Plunkett et al., '74). After penetration, the original araAMP moiety was phosphorylated to the triphosphate, araATP, and incorporated into DNA.These results provided a biochemical basis for the suggestion that, in certain cases, nucleotides might be more toxic formulations for chemotherapeutic purposes than nucleosides. As nucleotides, the compounds might prevent the degradation of the nucleoside moiety by cellular enzymes or could possibly bypass cellular deficiencies in the enzymes required for activation of the toxic compound. In the former case, we have provided evidence that araAMP and its cyclic derivative, 3', 5'-cyclic araAMP, are more toxic to L-cells than is 9-p-D-arabinofuranosyladenine (araA) (Ortiz et al., '72; Plunkett and Cohen, '75b). In the latter case, we have shown