N.E. Miller (1959) and others have stressed the effectiveness and creativity of working backward in problem-solving; but Newell, Shaw, & Simon (1962) argued that working backward is not inherently superior to working forward. An experiment is reported in which undergraduate subjects were set a series of route-finding tasks in a maze-situation modified to allow observation of working forward and working backward. The effectiveness of the two directions of work was controlled. The method and results of the experiment are discussed in relation to the views of the above-named writers and others.