This paper describes a system for learning rules using axis-parallel or rotated hyperboxes as individuals of a genetic algorithm (GA). Our method attempts to find out hyperboxes: aligned with the coordinate axes, such C4.5; or at any orientation by combining deterministic hill-climbing with GA. The system uses a decision queue (DQ) as method of representing the rule set. It means that the obtained rules must be applied in specific order, that is, an example will be classified by the i-rule only if it doesn't satisfy the condition part of the i-1 previous rules. With this policy, the number of rules may be reduced because the rales could be one inside of another one. We have tested our system on real data from UCI repository. We have also designed some two-dimensional artificial databases to graphically represent the experiments. The results are summarized in the last section.