Swimming suspensions of Chlamydomonas eugametos were pelleted and homogenized, and the metabolism of inositol polyphosphates by cellular homogenates or supernatants was investigated. Ins(1,4,5)P3 was dephosphorylated under physiological conditions to yield a single InsP2, Ins(1,4]2. In the presence of ATP it was phosphorylated to give Ins(1,3,4,5)P3 as the only InsP4. The Ins(1,4,5)P3 3-kinase activity was predominantly soluble, was not detectably affected by calmodulin or Ca2+, and had a Km for Ins(1,4,5)P3 of 50 microM (two orders of magnitude higher than its mammalian counterpart). Ins(1,3,4,5)P4 was dephosphorylated by the cellular supernatants to Ins(1,3,4)P3 and Ins(1,4,5)P3, and could be phosphorylated to Ins(1,3,4,5,6)P4. No Ins(1,3,4)P3 6-kinase activity could be detected, and experiments with [3H]Ins(1,4,[32P]5)P3 revealed that Ins(1,3,4,5,6)P5 is formed from Ins(1,4,5)P3 with little loss of the 5-phosphate, i.e. the predominant route of synthesis is probably by a direct 6-phosphorylation of Ins(1,3,4,5)P4. Similar experiments with an (NH4)2SO4 fraction of turkey erythrocyte cytosol gave essentially the same result, i.e. direct phosphorylation of Ins(1,3,4,5)P4 in the 6 position is the predominant route of synthesis of InsP5 from that InsP4 in vitro. No InsP6 formation was detected in any of these experiments, but labelling of intact C. eugametos with [3H]inositol revealed that the cells do synthesize InsP6. The lipids of C. eugametos cells contain PtdIns, PtdIns(4)P and PtdIns(4,5)P2 [Irvine, Letcher, Lander, Drøbak, Dawson & Musgrave (1989) Plant Physiol. 64, 888-892]. Further examination of 32P-labelled lipids revealed that about 20% of the PtdInsP was the PtdIns(3)P isomer, and about 1% or less of the PtdInsP2 was the PtdIns(3,4)P2 isomer. The overall inositide metabolism of C. eugametos resembles that of a mammalian cell more closely than it does that of a plant cell or slime mould, and this suggests firstly that the known metabolism of inositol polyphosphates arose at an early time in eukaryotic evolution, and secondly that Chlamydomonas might prove a useful organism for genetic and comparative studies of inositide enzymology.