Several variables were studied as they relate to vividness of imagery in recall. Thirty-two subjects, pre-selected for high and low imagery, worked with two types of block designs. With one type they were repeatedly asked either to summon up visual images and copy them, or else to recall the designs with the aid of imagery and other resources of memory. The results showed no differences in accuracy between the two groups of subjects, but the better recalled designs were generally described as being accompanied by more vivid imagery. Reported vividness (but not accuracy) increased after an introspective inquiry. There was a small experimenter effect. The second type of design was presented under incidental learning conditions. The images reported in the subsequent (unexpected) recall were much more vivid than those in the remainder of the experiment. These findings point to the effects of procedural variables on vividness of imagery, and suggest that imagery may play a special role in the recall of incidentally presented material.