The failure problem in high schools has led to numerous trials in providing special attention for pupils who are dull. Favorable results have been secured from certain methods such as ability grouping, special classes, and the employment of additional teachers ; yet schools must be operated within their own limitations and many administrators find it impossible to provide extra teachers, special classes, or other equally de sirable factors. Hence, the field is open for further experiment. Norwin High School, Irwin, Pennsylvania, has tried a plan involving alternative dismissal. This experiment was conducted through April and May, 1929, during which time passing pupils were dismissed at 3:00 o'clock and failing pupils were detained for additional instruction until 3:40. This arrangement was made possible by dividing the afternoon session into four periods instead of three. The regular program was thus completed at 3:00 o'clock, and for the remainder of the school day the teachers gave special attention to their failing pupils. In other words, afternoon classes were made shorter in order to provide an extra period of instruc tion for failing pupils ; the benefits accruing to these pupils were derived from time that was taken from pupils who were doing satisfactory work. It is not our purpose to discuss the details and possibilities of this plan or to report upon the pupil reaction to it, though much could be said. Rather, our purpose is to report the results obtained. Data are provided in terms of the number of failures and of the average attain ment for different groups. The school in which this experiment was conducted is a joint high school with 22 teachers and an enrollment of 540 pupils. The pupils come from small towns, villages, country and mining communities, and from grade schools of every description without uniform supervision ex cept as provided by the County Superintendent who serves a pupil population of approximately 60,000. Preparation for high school en trance covers a wide range, and I Q's group themselves into a normal distribution ; throughout the population are scattered about 100 second generation pupils. The school is located 20 miles from Pittsburgh ; the community is classified as urban residential, and the East Pittsburgh