Besides ATP, the sodium pump can hydrolyse phosphatase substrates like p-nitrophenyl phosphate (pNPP) in the presence of K ions; ATP, ADP, and their analogues act as low-affinity competitive inhibitors of the K⁺-phosphatase activity. Covalent block of an N-domain pocket (1) by FITC abolishes high-affinity ATP binding, ATP phosphorylation and the Na,K-ATPase activity, but the phosphatase activity and the low-affinity nucleotide effects remain (2,3). FITC modification aside, this effect on pNPP hydrolysis reflects the prominent low affinity ATP activation and ADP inhibition of the Na,K-ATPase activity. In fact, with our purified Na,K-ATPase preparations we now realise that ADP can be a better inhibitor (K_i of 0.27-0.42 mM) than ATP is an activator (K_0.5 of 0.6 mM). We wished to find out, therefore, whether the low affinity ATP and ADP effects on the Na,K-ATPase activity resulted from direct binding at the phosphatase substrate site. We applied two kinetic tests. If the substrate concentration is held fixed, increasing inhibitor concentrations should yield linear Dixon plots (1/v vs. [i]) if substrate and inhibitor compete for a unique site. However, we find that at 0.6 mM pNPP, the Dixon plot is clearly hyperbolic with ATP concentration, with a K_0.5 of 8.6 mM. This is diagnostic of partially competitive inhibition (4): at the saturating concentration, ATP will have unhindered access to the nucleotide site whilst pNPP may still bind at a distinct phosphatase site from time to time. The high fitted K_0.5 value should signal the ATP binding affinity at the nucleotide site when pNPP occupies the phosphatase site (4). The second test involves the use of 2 inhibitors, to decide whether they bind at the same or different sites to cause the inhibition. We plotted the reciprocal of the K⁺-phosphatase activity at a fixed 6 mM pNPP, as a straight line against 4-methylumbellipherone phosphate concentration (0-3 mM); this is another phosphatase substrate and a full competitor. The same was done again, but now in the presence of 4 mM ADP. Both straight lines were clearly convergent on the left, which indicates that the ternary complex is possible, of the pump with 4-methylumbellipherone phosphate and ADP. Mutual exclusion of the two inhibitors would have shown as roughly parallel straight lines (5). Finally, we did the converse experiment, and measured Na,K-ATPase activity at a fixed 0.25 mM ATP. The reciprocal of the ATPase activity was plotted against 0-2 mM ADP, with and without 15 mM pNPP. The straight lines were again clearly convergent on the left. Taken together, these results strongly suggest that the K⁺-phosphatase substrates bind elsewhere and not at the locus where nucleotides modulate the pump activities. Long-range effects should be the main cause for the mutual decrease in binding affinities.