T as a single group for now based on the distinctive morphological traits. Hosts: Crambidae, Riodinidae. The described AMG9810 price species are from ACG. Key to species of the keineraragoni group 1 Fore wing with vein r 1.4 ?as long as vein 2RS, vein 2M 1.5 ?as long as vein (RS+M)b; flagellomerus 2 2.7 ?as long as wide; interocellar distance 1.3 ?as long as posterior ocelli diameter; metatibia dark brown to black on posterior 0.8 (Figs 136 a, c) [Hosts: Crambidae] ………………………………………………… ………………………Apanteles keineraragoni Fern dez-Triana, sp. n.(N=3) Fore wing with vein r 1.7 ?as long as vein 2RS, vein 2M 0.7 ?as long as vein (RS+M)b; flagellomerus 2 3.2 ?as long as wide; interocellar distance 1.7 ?as long as posterior ocelli diameter; metatibia dark brown to black on posterior 0.4?.5 (Figs 137 a, c) [Hosts: Riodinidae] ………………………………………….. ………………….. Apanteles ronaldnavarroi Fern dez-Triana, sp. n. (N=1)?Jose L. Fernandez-Triana et al. / ZooKeys 383: 1?65 (2014)leucostigmus species-group This group, by far the largest in Mesoamerica, comprises 39 species in this paper. It is defined by a thick ovipositor (as thick or thicker than the width of the median flagellomeres, and with anterior width 3.0?.0 ?its posterior width beyond the constriction), ovipositor sheaths 0.5?.1 ?as long as metatibia, propodeum with strong sculpture limited to anterior half, the posterior half mostly smooth; mesoscutellum with lateral face bearing a polished area 0.7 ?or more the height of the face, pterostigma and most of fore wing white or transparent, and mediotergite 1 widening towards posterior 0.7, then narrowing toward posterior margin. The group is supported by the Bayesian molecular analysis (PP: 0.74, Fig. 1). Hosts: Hesperiidae. Widely distributed in the Neotropics; we have seen many more undescribed species in collections. This is the only group where we extensively used molecular (i.e., barcoding) and biological (i.e., host records) characters in the key. Likewise, the species descriptions were also simplified and only include some morphological traits (plus full details on barcoding and host data). This was mostly due to the paucity of morphological characters that serve to distinguish different species. Relying solely on DNA barcoding and/or host data to describe and key species has been done before in Braconidae (e.g., Butcher et al. 2012). However, we did some preliminary study of using morphometrics to separate species, and the results (unpublished) suggest that morphometrics may work for many, although not all, of the species in this group. We describe here the species that have been found in ACG for the sake of completing its inventory of Apanteles. Key to species of the leucostigmus group The species Apanteles albinervis, included in this group because of its morphology, is only known from the male holotype, and our key is only to females. There are no hosts or molecular data available for the holotype, collected in “Mexico” in 1904. It is therefore impossible to key this species by any of the character systems used here. 1 ?2(1) ?3(2) Metatibia entirely or mostly (>0.7) dark brown to black, with Sinensetin site yellow to white usually restricted to anterior 0.2 at most (rarely with pale area extending up to anterior 0.3 of metatibia) (as in Figs 166 a, d) ………………………………….2 Metatibia light yellow to orange-yellow from 0.4 to almost entire metatibia (as i.T as a single group for now based on the distinctive morphological traits. Hosts: Crambidae, Riodinidae. The described species are from ACG. Key to species of the keineraragoni group 1 Fore wing with vein r 1.4 ?as long as vein 2RS, vein 2M 1.5 ?as long as vein (RS+M)b; flagellomerus 2 2.7 ?as long as wide; interocellar distance 1.3 ?as long as posterior ocelli diameter; metatibia dark brown to black on posterior 0.8 (Figs 136 a, c) [Hosts: Crambidae] ………………………………………………… ………………………Apanteles keineraragoni Fern dez-Triana, sp. n.(N=3) Fore wing with vein r 1.7 ?as long as vein 2RS, vein 2M 0.7 ?as long as vein (RS+M)b; flagellomerus 2 3.2 ?as long as wide; interocellar distance 1.7 ?as long as posterior ocelli diameter; metatibia dark brown to black on posterior 0.4?.5 (Figs 137 a, c) [Hosts: Riodinidae] ………………………………………….. ………………….. Apanteles ronaldnavarroi Fern dez-Triana, sp. n. (N=1)?Jose L. Fernandez-Triana et al. / ZooKeys 383: 1?65 (2014)leucostigmus species-group This group, by far the largest in Mesoamerica, comprises 39 species in this paper. It is defined by a thick ovipositor (as thick or thicker than the width of the median flagellomeres, and with anterior width 3.0?.0 ?its posterior width beyond the constriction), ovipositor sheaths 0.5?.1 ?as long as metatibia, propodeum with strong sculpture limited to anterior half, the posterior half mostly smooth; mesoscutellum with lateral face bearing a polished area 0.7 ?or more the height of the face, pterostigma and most of fore wing white or transparent, and mediotergite 1 widening towards posterior 0.7, then narrowing toward posterior margin. The group is supported by the Bayesian molecular analysis (PP: 0.74, Fig. 1). Hosts: Hesperiidae. Widely distributed in the Neotropics; we have seen many more undescribed species in collections. This is the only group where we extensively used molecular (i.e., barcoding) and biological (i.e., host records) characters in the key. Likewise, the species descriptions were also simplified and only include some morphological traits (plus full details on barcoding and host data). This was mostly due to the paucity of morphological characters that serve to distinguish different species. Relying solely on DNA barcoding and/or host data to describe and key species has been done before in Braconidae (e.g., Butcher et al. 2012). However, we did some preliminary study of using morphometrics to separate species, and the results (unpublished) suggest that morphometrics may work for many, although not all, of the species in this group. We describe here the species that have been found in ACG for the sake of completing its inventory of Apanteles. Key to species of the leucostigmus group The species Apanteles albinervis, included in this group because of its morphology, is only known from the male holotype, and our key is only to females. There are no hosts or molecular data available for the holotype, collected in “Mexico” in 1904. It is therefore impossible to key this species by any of the character systems used here. 1 ?2(1) ?3(2) Metatibia entirely or mostly (>0.7) dark brown to black, with yellow to white usually restricted to anterior 0.2 at most (rarely with pale area extending up to anterior 0.3 of metatibia) (as in Figs 166 a, d) ………………………………….2 Metatibia light yellow to orange-yellow from 0.4 to almost entire metatibia (as i.