In dogs with gastric and pancreatic Thomas fistulas the effect of different cholecystokinin-like peptides upon pancreatic polypeptide (PP) release was studied in three ways: (a) Plasma PP concentrations were determined by radioimmunoassay in response to 135 pmol/kg/h of the synthetic C-terminal octapeptide of cholecystokinin (CCK-OP) (A), of three of its analogues (B, C, D) where methionine has been replaced by methoxinine, and in response to 45 pmol/kg/h of caerulein. The greatest rise in plasma PP concentration expressed as delta PP was achieved with caerulein (327 +/- 37 pM), when taking into account the threefold smaller dose used, followed by CCK-OP (536 +/- 67 pM) and analogues B (343 +/- 51 pM), C (87 +/- 46 pM), and D (32 +/- 15 pM). The order of potency with respect to stimulation of exocrine pancreatic secretion was the same: E and A precede B, C, and D. delta PP correlated linearly with the pancreatic protein output (r = 0.98, p less than 0.01). (b) CCK-OP was infused in four doses of 35, 70, 135, and 270 pmol/kg/h, and plasma PP concentrations and exocrine pancreatic secretion were monitored. The correlation between pancreatic protein output and delta PP was very close (r = 0.98, p less than 0.01). (c) Atropine sulfate (0.1 mg/kg, i.v.) reduced the PP response to the 135 pmol/kg/h dose of CCK-OP by 61%. We conclude from this that CCK-OP and related peptides do release PP and that their effect on exocrine pancreatic secretion is closely correlated with their PP-releasing capacity. The PP release may therefore be used as an indicator of the CCK-like activity of CCK fragments and analogues. CCK-OP may well represent one of the humoral stimulatory factors contributing to the release of PP, but this action appears to depend on a cholinergic background.