It has become customary to regard the various glycerophospholipids as quite similar, and the acyl groups are considered to have little influence on the behaviour of the lipids in membranes or metabolism. Nevertheless, a number of recent observations by the authors and others indicate a high degree of metabolic compartmentation and substrate specificity with regard to the acyl substituents (acyl specificity) of glycerophospholipid metabolising enzymes in intact cells. 1. [32P]Orthophosphate and [3H]glycerol are incorporated into phosphatidylcholine (PC) and phosphatidylethanolamine (PE) of platelets and Swiss 3T3 fibroblasts with a [32P]/[3H]-ratio several fold lower than in glycerol-3-phosphate, phosphatidic acid (PA) and phosphatidylinositol (PI), suggesting distinct metabolic separation (probably by cellular compartmentation) of the glycerol and choline (or ethanolamine) branches of de novo phospholipid biosynthesis. 2. In fibroblasts the [32P]/[3H]-ratio varied 50-fold among the molecular species of PC, PE, PI and PA, which indicates that the enzymes involved in these conversions have some degree of acyl specificity. 3. In vitro assays for lipid-converting enzymes employ detergents, which affect acyl specificity of the enzymes (lipid kinases) both by their chemical nature and concentrations. 4. Thrombin stimulation of platelets causes formation of a multitude of diacylglycerol (DAG) molecular species, but only one major molecular species of PA is formed indicating that the DAG kinase may have distinct acyl specificity in the intact cell. 5. However, this specificity could also result from the net reactions of DAG kinase(s) and PA phosphohydrolase(s), which would constitute an ATP-utilising, paired regulation of the molecular species of PA and the inositol lipids on one hand, and PC, PE phosphatidylserine and triacylglycerol on the other. These findings indicate a high complexity of glycerophospholipid metabolism and a distinct acyl specificity in intact cells that are not apparent from studies in vitro. A major challenge for future research in this area is to bridge the apparent discrepancy between in vivo and in vitro observations regarding glycerophospholipid metabolism, an endeavour that will require more knowledge about the physical chemistry of naturally occurring molecular species than is available today. The most prevailing appreciation of glycerophospholipids among biological scientists to-day is that they can be distinguished functionally, topographically and metabolically only by their head groups and that they form the bilayer in biological membranes. Most of us know that the fatty acid in the sn-2 position is unsaturated and have been indoctrinated that the higher the degree of unsaturation, the greater the fluidity of the membrane.(ABSTRACT TRUNCATED AT 400 WORDS)