Summary.The purpose and organisation of remedial education and possible methods of selection for it are discussed. At the outset of a new scheme of adjustment groups, the opportunity occurred to compare the effectiveness of selection by teachers' judgment with the conventional method of selection by test results. In the experiment, no significant difference was found in the effectiveness of the two methods as judged by gains in reading or arithmetic achievement. The achievement gains were also classified according to I.Q., A.Q. (R.A./M.A.) and amount of backwardness (R.A./C.A.) and the results of this analysis are reported. In addition, an attempt was made to assess personality changes during the remedial treatment, but the results were inconclusive.