The experiments reported herein were designed to identify the best assay of diacetylmonoxime (DAM) in biological fluids, with particular emphasis on detection limits. Initially, four variations of the assay of urea with excess DAM were compared. The best method, published in 1977, was one that includes thiosemicarbazide, 4-aminoantipyrine and ceric ammonium sulphate in the acid reagent; it is fast, gives a reasonably stable chromophore and displays good linearity. However, the reaction was two or more times less sensitive when applied to the assay of DAM, with urea in excess, by interchanging the amounts of urea and DAM. Further, the calibration graphs did not pass through the origin, and one of the methods gave a mixture of two chromophores. None approached the sensitivity of a DAM-urea reaction specifically designed to assay biacetyl (formed from DAM by acid hydrolysis) and published in 1968. This method, using antipyrine and arsenicosulphuric acid, is also fast, with good linearity and a stable chromophore, but is sensitive to interference by plasma and urine. An alternative photometric assay that does not involve urea was equally sensitive. It had the advantage of less interference by plasma and urine but was more time-consuming. Both of these photometric methods had a limit of detection of ca. 0.2 microgram DAM, similar to that of a high-performance liquid chromatography (HPLC) assay. Sample clean-up is necessary before application of the HPLC assay.