This paper presents a synopsis of detailed investigations carried out by the Chemists Sub-Committee, subsequently restyled the Analytical Committee, of the Fertiliser Manufacturers Association during the period 1949-1959, into the accuracy and precision of methods for the determination of potassium in fertilisers. Because it has had access to a large number of laboratories in different companies through the medium of circulation of samples under a routine check analysis scheme, it has been possible to test exhaustively the effect of modifications to methods, prompted by both practical and theoretical considerations. This review covers the older classical chloroplatinate and perchlorate methods, the development of the chloroplatinate method due to C.H. Perrin and the value of flame photometry. Each stage of the work has been proven by practical results on suitable samples in a large number of laboratories and the evidence is presented here in summarised statistical form.
It is suggested that the superiority in both precision and accuracy of the chloroplatinate method due to C.H. Perrin over the older, classical methods is clearly demonstrated, while, up to a certain level of potassium concentration, the flame photometer method, as developed in the course of the investigations, has similar qualifications. Both methods have the additional advantages that they are more rapid and require less operator skill.