DIF-1 [Differentiation-Inducing Factor 1; 1-(3,5-dichloro-2,6-dihydroxy-4-methoxyphenyl)hexan-1-one] is a novel chlorinated signal molecule that induces stalk-cell differentiation during development of Dictyostelium discoideum. Here we introduce the use of the radioisotope 36Cl to label DIF-1 and other low-Mr chlorinated compounds produced during development. H.p.l.c. and t.l.c. were used to resolve the labelled compounds. We find the following. (1) At least 14 dialysable 36Cl-labelled compounds are released into the medium by cells labelled continuously through development with Na36Cl. (2) The compounds can be classified into two major groups according to their times of accumulation in development. The early group of compounds starts accumulating at the end of aggregation, co-ordinately with DIF-1; the late group is only made at the end of development, by mature fruiting bodies. There may also be an intermediate group made during culmination. (3) The early group of compounds has been identified as comprising DIF-1 and seven of its metabolites by co-chromatography with the authentic compounds. These metabolites had previously only been recognized in suspensions of living cells incubated with exogenous DIF-1. Their detection here, from cells undergoing normal development, suggests that endogenous DIF-1 is metabolized in normal development in much the same way as is DIF-1 added to cells in suspension. (4) The intermediate and late groups of compounds are not obvious DIF-1 metabolites. They may have some role unconnected with DIF signalling. (5) A group of 36Cl-labelled late compounds remain cell-associated after washing of the fruiting bodies, and these are greatly enriched in stalk, compared with spore, cells. (6) Other slime-mould species were labelled with 36Cl. All three tested, namely D. mucoroides, D. vinaceo-fuscum and P. violaceum, also produced chloro compounds. D. mucoroides produced DIF-1 by the criterion of h.p.l.c. co-elution with authentic DIF-1. A developmentally regulated metabolism of chlorinated compounds may therefore be widespread amongst slime moulds. To our knowledge, labelling with 36Cl in vivo has not been reported before and provides a powerful general method for investigating chlorinated compounds in diverse organisms.