Abstract: On the hypothesis that abstract ideas can be better grasped through more abstract symbols than numerals, an experiment was done to see whether algebra could be taught to children eariler than is. customary.Two 5th grade classes comprising 87 children aged 10—11, Who Would normally not reach algebra for another two years, were taught simple equations in algebra and tested as regards the tendency to a higher level of abstraction through using algebra to solve arithmetical problems, the understanding and use of formulae and the grasp of the rules governing the relation between numbers.Within one month, with no extra hours of teaching, the children did at least one fifth of all the problems set in the arithmetic textbooks of the 5th and 6th year primary school grades and the junior school 1st grade.In the regular term-end municipal examination in arithmetic, though the two classes had missed some of the arithmetic teaching by reason of the experiment, they obtained 92% and 96% respectively, much higher than the general average.The experimental classes also stood out in. a comparative test on grasp of the rules of arithmetic and the rules governing the relation between numbers. Two other classes of the same grade in the same school also took part in this test.Tests in the grasp and use of formulae showed an appreciable level of understanding not based on Mechanical memory.After a four months gap, these four classes were again tested and the two experimental classes remained well ahead.The experiment shows that children of this grade reach a high level of abstraction and suggests that higher abstract symbols ale a better vehicle for abstract thought. It. also shows that the use of algebraic symbols and formulae gives better retention as regards methods of solving arithmetical problems. The experiment suggests the practicability of introducing, algebra into the teaching of arithmetical problems. Work has yet to be done to determine at what stage this is most effective, how widely the conclusions are applicable, etc.

Key words: NULL