Data are assembled on the clutch–size strategies adopted by extant species of parasitoid wasp. These data are used to reconstruct the history of clutch–size evolution in the group using a series of plausible evolutionary assumptions. Extant families are either entirely solitary, both solitary and gregarious, or else clutch size is unknown. Parsimony analysis suggests that the ancestors of most families were solitary, a result which is robust to different phylogenetic relationships and likely data inadequacies. This implies that solitariness was ubiquitous throughout the initial radiation of the group, and that transitions to gregariousness have subsequently occurred a minimum of 43 times in several, but not all lineages. Current data suggest that species–rich and small–bodied lineages are more likely to have evolved gregariousness, and contain more species with small gregarious brood sizes. I discuss the implications of these data for clutch–size theory.