The two goals of this investigation were to (1) examine the comprehension monitoring strategies adults employ when trying to understand procedural instructions and (2) determine how comprehension may be affected by varying such instructions. Sixty-eight college students, using instructions that consisted of either text alone, illustrations alone, or a combination of the two, were videotaped individually as they attempted to assemble a toy loading cart. In addition to differences in the mode of presentation, the instructions were also varied so that half of the students received directions that contained explicit operational or "how to" information while the other half received directions that contained more general information. The videotaped performances were then coded according to a taxonomy of comprehension monitoring strategies. The results showed that several of these strategies were related to comprehension as measured by the speed and accuracy of performance. The findings also indicated that, regardless of the mode of presentation, students using instructions that contained explicit operational information made fewer uncorrected errors than those using more general instructions. Suggestions based on these results were offered for the design and use of procedural instructions. (Author/KC)