Over the last quarter century, more than 532 genes have been put forward as regulatory targets of retinoic acid. In some cases this control is direct, driven by a liganded heterodimer of retinoid receptors bound to a DNA response element; in others, it is indirect, reflecting the actions of intermediate transcription factors, non-classical associations of receptors with other proteins, or even more distant mechanisms. Given the broad range of scientific questions continually under investigation, researchers do not always have occasion to classify target genes along these lines. However, our understanding of the genetic role of retinoids will be enhanced if such a distinction can be made for each regulated gene. We have therefore evaluated published data from 1,191 papers covering 532 genes and have classified these genes into four categories according to the degree to which an hypothesis of direct versus indirect control is supported overall. We found 27 genes that are unquestionably direct targets of the classical pathway in permissive cellular contexts (Category 3 genes), plus 105 genes that appear to be candidates, pending the results of specific additional experiments (Category 2). Data on another 267 targets are not evocative of direct or indirect regulation either way, although control by retinoic acid through some mechanism is clear (Category 1). Most of the remaining 133 targets seem to be regulated indirectly, usually through a transcriptional intermediary, in the contexts studied so far (Category 0).