Electrospray mass spectrometry coupled with reverse-phase HPLC was used to identify peptides in the molecular mass range 3000-6000 Da in extracts of the pancreata of the clawed frog Xenopus laevis (Anura: Pipidae) and the red-bellied newt Cynops pyrrhogaster (Caudata: Salamandridae). Amino acid sequences of insulins, peptides derived from the post-translational processing of proglucagons and pancreatic polypeptide were determined by automated Edman degradation. Three molecular forms of insulin were isolated from the tetraploid organism X. laevis that represent insulin-1 and insulin-2, as deduced from the nucleotide sequences of previously characterized cDNAs, and a third form which differed from insulin-2 by the single amino acid substitution Asp(21)-->Glu in the B-chain. The amino acid sequence of Xenopus preproglucagons (genes 1 and 2 ) may be deduced from the nucleotide sequences of cDNAs but the pathways of post-translation processing of the precursors are not known. Two molecular forms of glucagon with 36 amino acids, derived from genes 1 and 2 and representing glucagon-29 extended from its C terminus by different heptapeptides, and five molecular forms of glucagon-like peptide 1 (GLP-1) were isolated. The GLPs represent proglucagon-(77-113), -(122-158) and -(160-191) from gene 1, and proglucagon-(77-113) and -(160-191) from gene 2. A single molecular form of insulin, glucagon-36, a C-terminally alpha-amidated GLP-1 with 30 amino acid residues, a 33 amino acid residue GLP-2 and pancreatic polypeptide were isolated from the pancreatic extract of the diploid organism C. pyrrhogaster. This study has illustrated the power of electrospray mass spectrometry for the rapid and reliable identification of peptides in chromatographic fractions without the need to use radioimmunoassay, radioreceptor assay or bioassay.