COMMUNICATIONSdepression in rats; this did not occur with nalorphine, bremazocine and cyclazocine. Furthermore, in contrast to morphine and EK, these agents had poor efficacy in the rat tail flick test and in inhibiting gastrointestinal transit (unpublished results; Green 1959).Against this background, we have recently learned that EK does not substitute for morphine in rats receiving an i.p. infusion of morphine (Teiger 1974) (Dr M. E. Feigenson, Sterling-Winthrop, personal communication).Taken together, these data show that the in vivo pharmacological profile of EK, in rats, resembles that of morphine far more than the profiles of other proposed K compounds such as nalorphine, bremazocine and cyclazocine. The major difference between morphine and EK is the lack of substitution by EK for morphine inTeiger's rat model. In summary, we call attention to the separation between cross-tolerance and cross-dependence in the rat and pose the following question: what criteria should be used to classify opioids in this species?Generous samples of ethylketocyclazocine and nalox-one were obtained from Sterling-Winthrop and Endo, respectively.Muscle damage including fibrosis and contracture has been reported as a complication of intramuscular drug administration (Todd