Certain popular models of the regulation of food intake predict a positive correlation between the size of a meal and the preceding and/or following intermeal interval. However, the reported strength of these prandial correlations has varied widely in the literature. To determine what factors may influence the strength of these relationships, we measured the timing of and amount consumed in meals of laboratory rats as a function of 1) whether they were housed in isolation or with partial access to peers or a running wheel, 2) whether they were disturbed daily or weekly for maintenance procedures, and 3) whether food was free or contingent on operant responding. We also compared two definitions of "meal" and "intermeal interval." Strong prandial correlations were found only occasionally. Caging, food cost, and maintenance frequency did not affect the size of the correlations, although these factors did influence the rats' meal patterns. We conclude the regulation of food intake cannot be explained by models relying on a regular relationship between meal size and intermeal interval. (C) 1999 Elsevier Science, Inc.