Hybrid vigor or heterosis for growth and survival was demonstrated in experimental crosses of inbred Pacific oysters by Hedgecock et al. [Hedgecock, D., McGoldrick, D.J., Bayne, B.L., 1995. Hybrid vigor in Pacific oysters: an experimental approach using crosses among inbred lines. Aquaculture 137, 285–298.]. Substantial evidence for the pervasiveness of heterosis accumulated since then suggests a role for crossbreeding in commercial improvement. Here, we summarize evidence for yield heterosis in juvenile (seed) and adult oysters resulting from four diallel mating experiments. In pair crosses of parental inbred lines, we quantify heterosis by potence, hp=Q/L>1.0, where Q is twice the deviation of a hybrid from the mid-parent value and L is the absolute difference between the mean trait-values of the two parental inbred lines, these contrasts being estimated from ANOVA. For larger incomplete diallel crosses, in which partially inbred parental lines were not reared, we present estimates of Griffing's [Griffing, B., 1956. Concept of general and specific combining ability in relation to diallel crossing systems. Aust. J. Biol. Sci. 9, 463–493.] general and specific combining abilities (GCA, SCA) and, where possible, reciprocal hybrid effects (R). GCA is significant in two of four crosses analyzed, while SCA is significant in all four crosses, particularly at the seed stage, and R is significant in all three crosses, in which reciprocal comparisons were possible. The reciprocal effect is partitioned into maternal (extra-nuclear) and non-maternal (extra-nuclear×nuclear interaction) effects; the latter are significant in 4 of 5 cases, while maternal effects are significant in only 2 of 5 cases. Improvement of commercial oyster seed can be achieved by a combination of selection among inbred lines and crossbreeding of elite inbred lines; pervasive differences between reciprocal hybrids may constrain the direction of interline crosses. Because correlation of seed and adult yield is positive but weak, we propose to retain top inbred parent lines based on seed yield and to re-test the most promising crosses on a much larger production scale.