The literature on the measurement of working memory is discussed, with emphasis on the validity of such measurement with elementary school students. Approaches by J. Pascual-Leone, R. Case, A. D. Baddeley, and C. Hulme are considered. An experimental study is presented, in which 100 boys and 91 girls (aged from 6 years to 10 years 11 months) were given a battery of 17 tests in four sessions, including well-known tests of verbal and spatial ability and measures that have already been proposed for various aspects of working memory. Eight subjects were excluded from some of the analyses because they made too many errors on the letter classification test. Three tests were group-administered. Two group sessions were followed by two individual sessions. The statistical analyses included factor analysis, partial correlations, and analysis of variance. The results show that five tests can be used in the measurement of the "central components" of working memory; this is consistent with the theories of Pascual-Leone and Case. On the other hand, Hulme's hypothesis that the speed of articulation of school children may predict the capacity of a "peripheral" component (i.e., the articulatory loop) seems to be disproved. Five tables summarize test data, and a 68-item reference list is provided. (SLD)