For over 25 years there has been uncertainty over the pathway from CO(2) to acetyl-CoA in chloroplasts. On the one hand, free acetate is the most effective substrate for fatty acid synthesis by isolated chloroplasts, and free acetate concentrations reported in leaf tissue (0.1-1 mM) appear adequate to saturate fatty acid synthase. On the other hand, a clear mechanism to generate sufficient free acetate for fatty acid synthesis is not established and direct production of acetyl-CoA from pyruvate by a plastid pyruvate dehydrogenase seems a more simple and direct path. We have re-examined this question and attempted to distinguish between the alternatives. The kinetics of (13)CO(2) and (14)CO(2) movement into fatty acids and the absolute rate of fatty acid synthesis in leaves was determined in light and dark. Because administered (14)C appears in fatty acids within < 2-3 min our results are inconsistent with a large pool of free acetate as an intermediate in leaf fatty acid synthesis. In addition, these studies provide an estimate of the turnover rate of fatty acid in leaves. Studies similar to the above are more complex in seeds, and some questions about the regulation of plant lipid metabolism seem difficult to solve using conventional biochemical or molecular approaches. For example, we have little understanding of why or how some seeds produce >50% oil whereas other seeds store largely carbohydrate or protein. Major control over complex plant biochemical pathways may only become possible by understanding regulatory networks which provide 'global' control over these pathways. To begin to discover such networks and provide a broad analysis of gene expression in developing oilseeds, we have produced microarrays that display approx. 5000 seed-expressed Arabidopsis genes. Sensitivity of the arrays was 1-2 copies of mRNA/cell. The arrays have been hybridized with probes derived from seeds, leaves and roots, and analysis of expression ratios between the different tissues has allowed the tissue-specific expression patterns of many hundreds of genes to be described for the first time. Approx. 10% of the genes were expressed at ratios > or = 10-fold higher in seeds than in leaves or roots. Included in this list are a large number of proteins of unknown function, and potential regulatory factors such as protein kinases, phosphatases and transcription factors. The arrays were also found to be useful for analysis of Brassica seeds.