Describes the method and results of experiments that were made to determine which of the three kinds of impressions (visual, auditory, motor) are best retained, and to discover the relation existing between recall and recognition. The subjects were pupils of a typical school and college in all grades from the third primary up. The results of these experiments are of more special interest to those concerned with pedagogical problems. They reveal very clearly the absurdity of the common assumption that any subject that requires memorizing gives valuable memory-training, and suggest for further observation and experiment questions which when answered will enable teachers to intelligently direct the exercise of the memory of pupils in general and to correct special defects in individuals.