Prior to a study of the pathophysiological significance of chronic atrophic gastritis and hypergastrinemia, we evaluated a new radioimmunoassay kit for serum total gastrin (Diagnostic Products Corp.). The mean intra-assay CV ranged from 2 to 5% (2386 patients' samples in 47 assay runs done during four months). Total CVs for two controls ranged from 4 to 10%. Within-assay bias was 5%. Oleate decreased the values, indicating that intravenous heparin, which releases endothelial lipase, causing in vitro lipolysis, should be avoided if indwelling catheters are used for sampling, e.g., during provocation tests for gastrin release. Of three other commercial kits examined, two were affected by oleate. Other anions such as heparin and EDTA also affected the assay. Values for gastrin in heparinized plasma from surgical patients representing a variety of disorders agreed well with results obtained by a reference laboratory. We confirm the usefulness of this assay for discriminating clinical situations and conclude that ligand assays, besides those for thyroid assessment, should be assessed for interference from nonesterified fatty acids. Preliminary data also suggest a marked age dependence of serum gastrin concentration.