Emerald ash borer, Agrilus planipennis Fairmaire (Coleoptera: Buprestidae), is an aggressive invader from Asia that has killed millions of trees in North America, causing substantial ecosystem effects and economic losses. All North American ash, Fraxinus spp., are thought to be susceptible, but recently emerald ash borer has been documented developing in a novel host, white fringetree, Chionanthus virginicus. We evaluated larval performance in two common ash species and white fringetree by infesting bolts with emerald ash borer eggs. In addition we evaluated cambial nitrogen, carbon, carbon:nitrogen, stem density, and response to artificial wounding, to determine which host plant characteristics most influence larval development. We also conducted choice and no choice assays using the classical biological control agent, Tetrastichus planipennisi Yang (Hymenoptera: Eulophidae), to assess its ability to locate larval emerald ash borer in the different host plants. We found significantly lower survival rates of emerald ash borer larvae in white fringetree compared to white ash, F. americana. Larval phloem consumption and larval growth were lower in fringetree than in either white or blue ash, F. quadrangulata. Carbon content and density were greater in fringetree than in either ash species. Response to wounding, measured by callus tissue formation, was greatest in white ash. In choice assays, T. planipennisi only parasitized emerald ash borer larvae in ash bolts, and in no-choice tests failed to parasitize larvae in fringetree. Our findings corroborate studies showing that fringetree is a suitable host for emerald ash borer larvae. Failure of T. planipennisi to parasitize larvae within fringetree has implications for the efficacy of this classical biological control agent in regulating emerald ash borer populations. Coupled with the use of white fringetree as a reservoir host, the enemy free space provided to emerald ash borer through use of this alternate host may have significant repercussions for emerald ash borer invasion dynamics.