A study of the effectiveness of music and rhythm used in classroom activities as a technique for developing short-term memory for phonological learning had as subjects 25 adult Cambodian, Lao, Hmong, and Vietnamese immigrants, students in a course in English as a second language. The subjects were given a pretest of their ability to distinguish similar and dissimilar initial and final phonemes in 20 word pairs. The phonemes focused on were sh/s/z/l/r/c/ch/p/f. Those who followed the directions in their responses were assigned to two groups, a treatment and a control group. Both groups were given 20 minutes of instruction twice a week, for two weeks. The control group was given minimal pairs highlighting the above phonemes, and the experimental group was given various songs and rhythmic chants presenting the same sounds contextually. A posttest similar to the pretest was administered after two weeks. The experimental group had slightly higher posttest scores than the control group. This finding suggests an advantage in the experimental treatment, especially in view of the previous experience of the individuals composing the two groups. In addition, the experimental group's scores were all bunched around a central score, suggesting that the experimental group advanced more as a group than did the control group, whose scores were distributed over a full range of scores. (MSE)