When a danger stimulus (a passing shadow) is presented to the crab Chasmagnathus granulatus, an escape response is elicited that habituates after repeated stimulation. Results from previous work suggest that this habituation might be mediated by endogenous opiates and, thus, that after an habituation session an analgesic effect on the response to a painful stimulus should be found. The purpose here was to test this contention. Each crab was lodged in an actometer where two 2-trial sessions with electrical shock stimulation were given (SS1, SS2), the intersession interval being 60 or 105 min, and the response activity recorded. In some experiments, repeated shadow stimulation was inserted during the interval, and the response activity also recorded. In Expt. 1, shadow sessions of 15 and 30 trials were tested and a significant decrease of the reactivity to the shock was only shown with 30 trials. In Expt. 2, crabs were immediately injected after SS1 with distilled water (d.w.) or naloxone (NX) (3.2 micrograms/g), and a shadow session of 30 trials was given to half of the d.w. animals and to half of the NX ones. NX injected crabs showed an increased responsiveness to the danger stimulus during the shadow session but no inhibitory effect on pain reactivity to the subsequent painful stimulus. Issues concerning dependence of the analgesia on danger stimulus iteration and on stimulus controllability, as well as the opioid nature of the analgesia, are discussed.