Cali's approach to accuracy is reviewed, and it is concluded that the main factor limiting its extension to analytes other than elements is the lack of suitable definitive methods of analysis. The work published on calcium, andwork known to be under way on other elements, is based on the use of stable isotope dilution-mass spectrometry (I.D.M.S.) as the definitive technique. This paper outliones the theorectical considerations involved in extending I.D.M.S. to the determinatioons of molecules; in particular, exact equations for the calibration graph obtained by I.D.M.S. are developed, and special cases of practical significance are defined. It is concluded that in theory there is no reason why accurate resluts should not be realised. Ther determination of isotope rations is central to the tecnique of I.D.M.S. The means available for such measurments are discussed, and it is concluded that adequate precision cnan be achieved with standard equipment/ To illustrate the application for I.D.M.S. to molecules the steps involved in an assay for inorganic phosphate are defined, and the protocol this evolved is applied to specimens from the Wellcome Group Quality Control Programme. Although good correlation was found between I.D.M.S. results and the participants' mean results, there was a substantial bias. It is argued that this is a result of the difficulty of defining "inorganic phosphate" as a chemical entity. Some requirements for...