Analgesic potencies of turkey, ostrich and des-acetyl salmon beta-endorphins have been measured in the tail-flick test and binding affinities determined by radio-receptor assay. The duration of analgesia and the slope of the dose-response curves generated by these peptides are similar to those elicited by mammalian beta-endorphins. This suggests that they act in vivo and in vitro on the same population of opiate receptors. The ratio of binding to analgesic potencies observed for these peptides varies nearly sixfold. Structure-activity analysis suggests that a basic side-chain at position 9 is required in order to produce a high opiate activity both in vivo and in vitro. A reexamination of the biological activities of camel beta-endorphin shows that the analgesic potency and binding affinity of this peptide are respectively 171 and 2.7 times higher than human beta-endorphin. His-27 and/or Gln-31 may contribute to this increased potency. The dissociation of radioreceptor binding affinity from analgesic potency in these naturally occurring beta-endorphin homologs suggests that either the conditions under which the binding assay is performed mask the true binding potency in the brain or that, once bound to the appropriate receptor, these homologs do not possess equal ability to produce biological effects.