Indole-3-acetic acid (IAA) and its putative precursors, L- and D-tryptophan, indole-3-pyruvate, and tryptamine were isolated from tomato (Lycopersicon esculentum (L.) Mill.) shoots, identified by mass spectrometry, and measured using capillary gas chromatography with an electron capture detector and radioactive internal standards. Average amounts present were 7.9ng · (g FW)(--1) IAA, 5.7ng · (g FW)(--1) indole-3-pyruvate, 132 ng · (g FW)(--1) tryptamine, 103 ng · (g FW)(--1) D-tryptophan, and 2250 ng · (g FW)(--1) L-tryptophan. Indole-3-acetaldoxime was not found; detection limits were less than 1ng · (g FW)(--1). When tomato shoots were incubated for 6, 10 and 21 h in 30% (-2)H2O, up to four positions in IAA, L- and D-tryptophan, tryptamine and indole-3-pyruvate became labelled with (-2)H. Compounds became labelled rapidly with 10% of IAA molecules containing (-2)H after 6 h. The percentage of labelled molecules of IAA and L-tryptophan increased up to 10 h but then decreased again, correlating with an increase in the total shoot tryptophan and presumably a result of protein hydrolysis in the excised, slowly senescing tissue. The amount of (-2)H in D-tryptophan also showed an increase followed by a decrease, but the proportion of labelled molecules was much less than in L-tryptophan and IAA. Tryptamine became labelled initially at a similar rate to IAA but continued to accumulate (-2)H up to 21 h. We conclude that tryptamine is synthesized from a different pool of tryptophan from that used in IAA synthesis, and is not a major endogenous precursor of IAA in tomato shoots. Indole-3-pyruvate was the most heavily labelled compound after 6 and 10 h incubation (21-h data not available). Furthermore, the proportion of (-2)H-labelled indole-3-pyruvate molecules was quantitatively consistent with the amount of label in IAA. On the other hand, a quantitative comparison of the IAA turnover rate and the rate of (-2)H incorporation into both L- and D-tryptophan indicates that IAA is not made from the total shoot pool of either L- or D-tryptophan. Instead IAA appears to be synthesized from a restricted pool which is turning over rapidly and which has access to both newly synthesized tryptophan and that from protein hydrolysis.