Spathius galinae Belokobylskij and Strazenac (Hymenoptera: Braconidae) is a recently discovered gregarious idiobiont larval ectoparasitoid currently being evaluated for biological control against the invasive emerald ash borer, Agrilus planipennis Fairmaire (Coleoptera: Buprestidae) in the United States. To aid in the development of laboratory rearing protocols, we assessed the influence of various emerald ash borer stages on critical fitness parameters of S. galinae. We exposed gravid S. galinae females to emerald ash borer host larvae of various ages (3.5, 5, 7, and 10 wk post egg oviposition) that were reared naturally in tropical (evergreen) ash (Fraxinus uhdei (Wenzig) Lingelsh) logs, or to field-collected, late-stage emerald ash borers (nonfeeding J-shaped larvae termed “J-larvae,” prepupae, and pupae) that were artificially inserted into green ash logs. When exposed to larvae in tropical ash logs, S. galinae attacked 5 and 7 wk hosts more frequently (68–76%) than 3.5 wk (23%) and 10 wk (12%) hosts. Subsample dissections of the these logs revealed that 3.5, 5, 7 and 10 wk host logs contained mostly second, third, fourth, and J-larvae, respectively, that had already bored into the sapwood for diapause.NoJ-larvae were attacked by S. galinaewhennaturally reared in tropical ash logs. When parasitized by S. galinae, 7 and 10 wk hosts produced the largest broods (≈6.7 offspring per parasitized host), and the progenies that emerged from these logs had larger anatomical measurements and more female-biased sex ratios. When exposed to emerald ash borer J-larvae, prepupae, or pupae artificially inserted into green ash logs, S. galinae attacked 53% of J-larvae, but did not attack any prepupae or pupae. We conclude that large (fourth instar) emerald ash borer larvae should be used to rear S. galinae.