Comparisons have been made of the rates of desaturation of stearic to oleic acid using a series of differently labeled forms of the substrate and two different methods of assay. The reaction has been measured either by analyzing the amount of labeled oleic acid produced, or by analyzing the amount of tritium released into water from [9,10-3H2]-labeled substrates. The rate of desaturation of [1-14C] stearic acid has been compared with that of [threo-9,10-3H2] and [erythro-9,10-3H2] stearic acids in which about 99% of the label was at the 9 and 10 positions and in which most of the labeled molecular species contained two tritium atoms. A further series of [erythro-9,10-3H2] stearates was used in which there was a greater proportion of tritium at positions other than 9 and 10 or a greater proportion of species containing only one tritium atom, or both. In the tritium release assay the enzyme discriminated against the tritium substrates, the discrimination being greater witherythro-ditritio-compounds than with monotritio orthreo-ditritio-compounds. The isotope effect was also observed when [3H] stearoyl-CoA thiol esters were the substrates, and when the source of the enzyme was the green algaChlorella vulgaris. Despite the isotope effect, the release of tritiated water can be used as an assay of desaturase activity. If the absolute value of enzymic activity is required, however, the location and stereochemistry of the tritium atoms should be known or the method standardized against [1-14C] stearic acid.