An eight-concentration intermittent-flow proportional diluter was modified to provide continuous flow to four separate avoidance chambers and used to test the avoidance response to phenol of rainbow trout (Salmo gairdneri) in three separate 5-day tests. When single fish had been acclimated for 24 h, a flow of phenol was initiated on one side of each avoidance chamber and well water was started on the other. After 48 h the phenol and well water sides were switched, and the experiment was continued for another 48 h; the positions of the fish in the chambers were monitored photographically throughout the test. Results of tests with 12 rainbow trout indicated that the threshold avoidance level was between 6.5 and 3.2 mg/L phenol, which was between 58 and 73% of the phenol 96-h LC50 for rainbow trout. Since long-term effects for rainbow trout exposed to phenol have been documented at concentrations as low as 0.2 mg/L, the avoidance tests were not nearly as sensitive as toxicity tests with embryos and larvae.