The delta-selective opioid peptide deltorphin C(H-Tyr-D-Ala-Phe-Asp-Val-Val-Gly-NH2) (DEL C) was modified by para-substitution of Phe3 with halogens (F, Cl, Br, I), amino, or nitro groups. The bioactive potencies in peripheral tissues and brain receptor selectivities of these analogues depended upon the particular substituent; peptides containing halogen substituents exhibited the least disruptive effect. In the mouse vas deferens (MVD) bioassay, [p-ClPhe3]DEL C displayed equivalent bioactivities to DEL C; in combination with the guinea pig ileum (GPI) bioassay, [p-ClPhe3]DEL C and [p-BrPhe3]DEL C exhibited marked preference for delta sites (IC50GPI/IC50MVD = 11,250 and 6,363, respectively), which are approximately 4- and 2-fold greater than DEL C. In a receptor binding assay, none of the halogenated analogues had delta affinities (Ki) exceeding that of DEL C; however, in terms of delta selectivity (Ki mu/Ki delta), [p-BrPhe3]DEL C was nearly twice as selective as DEL C, while [p-FPhe3]DEL C was equivalent, and [p-IPhe3]DEL C only 25% less selective. The only correlation evident with the halogenated derivatives occurred between IC50GPI and Ki mu (r = 0.814) rather than between delta receptor studies (MVD or Ki delta); interestingly, IC50GPI also correlated with K' (r = 0.982). The p-amino or p-nitro substituents of Phe3 in DEL C and DEL B (= [Glu4]DEL C) were deleterious for bioactivity (MVD) (losses ranged from 400- to approximately 8,000-fold) and in receptor binding assays, where delta affinities decreased 140- to 840-fold and delta selectivities by 34- to 380-fold. p-Nitro-Phe3 was the most detrimental substitution for all the parameters measured for both deltorphins: the loss in MVD activity, however, was less with DEL B than with DEL C, which was the opposite for delta receptor affinity.