The purple acid phosphatases (PAPs) are bimetallophosphatases that are reported to lack phosphodiesterase activity based on the absence of reactivity with bis(p-nitrophenyl) phosphate. Yet, model systems designed to mimic the PAP active site preferentially hydrolyze phosphodiesters. We report that the PAPs from pig and from red kidney bean efficiently hydrolyze the diesters methyl nitrophenyl phosphate and ethyl nitrophenyl phosphate. The details of the diester reaction give insights into the sequential steps involved in binding and hydrolysis by the metal center. Methyl phosphate monoester, the initial product of the reaction with methyl nitrophenyl phosphate, is subsequently hydrolyzed without release into solution. The turnover number for this reaction is faster than that for the reaction of free methyl phosphate. This is consistent with a mechanism for monoester hydrolysis consisting of initial coordination to the divalent metal, followed by slower partially rate-limiting displacement of hydroxide from Fe(III) to give the catalytically active bidentately bound substrate. The reaction with the diester substrate is proposed to proceed by coordination to the divalent metal ion followed by nucleophilic attack by the Fe(III)-coordinated hydroxide to give a bidentately bound monoester poised for attack by the bridging mu-hydroxide. This mechanism eliminates the need for the second, slow binding event in monoester hydrolysis.