Porcine gastric inhibitory peptide (GIP) has been used extensively as a standard and as [125I]-GIP in radio-immunoassays (RIAs) to determine immunoreactive GIP (IR GIP) in human plasma even though porcine and human GIP have slightly different amino acid sequences. A rabbit anti-porcine GIP serum (R65) which has been widely used in the RIA for IR GIP reacts with a part of the GIP molecule which is identical in human and porcine GIP. In order to validate the use of R65 the IR GIP contents of synthetic human and porcine GIP were measured, using natural porcine GIP as standard, and found to be 40% and 35% on a weight basis respectively. These materials were considered unsuitable for use as RIA standards. The IR GIP in samples of human plasma was then assayed with 125I-labelled synthetic human or porcine GIP using natural porcine GIP as standard. The values measured for IR GIP did not depend on the labelled peptide used (p = 0.84, n = 81) and were linearly correlated (r = 0.98, p less than 0.001, n = 81). It is concluded that the value of IR GIP measured in human plasma using R65 and standards of porcine GIP is linearly proportional to the level of human GIP in plasma.