This paper describes laboratory comparisons of precision and accuracy of some widely used methods for the determination of lead in blood and for lead and cadmium on filters. With the exception of the tantalum boat atomic absorption technique, the methods studied were found to be rugged for a variety of minor procedural changes. Recovery of cadmium from spiked filters was quantitative over a wide range of dilutions and for a variety of instrumental readout and calculation combinations. In addition to the tantalum boat atomic absorption methods, results were obtained with the dithizone, extraction/aspiration, graphite furnace, and flame atomic absorption method. The use of the tantalum boat method for lead in blood is not recommended. For the determination of cadmium filters, the most accurate and precise results are obtained by the use of external calibration curves rather than the method of standard additions.