103 urn:lsid:arphahub.com:pub:77d0745d-c3a1-5248-81de-8cdc02bed84a urn:lsid:zoobank.org:pub:F56F6CF9-7502-4001-A751-35D5F2EF6CA0 Arthropod Systematics & Phylogeny ASP 1863-7221 1864-8312 Senckenberg Gesellschaft für Naturforschung 10.3897/asp.84.e174388 174388 Research Article Bethylidae Chrysidoidea Coleoptera Curculionidae Curculionoidea Hexapoda Hymenoptera Insecta Polyphaga Identification key Molecular systematics Morphology & Anatomy Taxonomy Theory & Methodology Zoo- or Phylogeography Unraveling a complex of cryptic species near Cephalonomia hyalinipennis (Hymenoptera: Bethylidae) and their association with the coffee berry borer and other hosts Honsberger David dnh8@hawaii.edu https://orcid.org/0000-0002-3656-7258 1 Conceptualization Writing - original draft Writing - review and editing Data curation Formal analysis Investigation Methodology Resources Visualization Pérez-Lachaud Gabriela https://orcid.org/0000-0002-8924-1606 2 Writing - review and editing Investigation Methodology Resources Magnacca Karl N. https://orcid.org/0000-0003-3803-7506 3 Conceptualization Writing - review and editing Formal analysis Investigation Methodology Visualization Lorenzo-Elarco Jacob Hauʻoli 4 5 Investigation Kelly Sandor L. https://orcid.org/0009-0002-0774-673X 6 Writing - review and editing Investigation Resources Wright Mark G. https://orcid.org/0000-0002-1643-7560 1 Writing - review and editing Funding acquisition Gariepy Tara D. https://orcid.org/0000-0003-2533-3023 7 Conceptualization Writing - review and editing Data curation Formal analysis Investigation Methodology Resources Visualization Department of Plant and Environmental Protection Sciences, University of Hawai'i at Mānoa, Honolulu, Hawai'i, USA University of Hawaiʻi at Mānoa Honolulu United States of America Departamento de Conservación de la Biodiversidad, El Colegio de la Frontera Sur, Chetumal, Mexico El Colegio de la Frontera Sur Chetumal Mexico Hawai‘i Department of Land and Natural Resources, Division of Forestry and Wildlife, 1151 Punchbowl St., Honolulu, Hawai‘i, USA Hawai‘i Department of Land and Natural Resources, Division of Forestry and Wildlife Honolulu United States of America Ka Haka 'Ula o Ke'elikōlani, College of Hawaiian Language, University of Hawai'i at Hilo, Hilo, Hawai'i; USA Honolulu Community College Honolulu United States of America Honolulu Community College, Honolulu, Hawai'i, USA College of Hawaiian Language, University of Hawaiʻi at Hilo Hilo United States of America University of Central Florida Collection of Arthropods (UCFC), Department of Biology, University of Central Florida, Orlando, USA University of Central Florida Orlando United States of America Agriculture and Agri-Food Canada, London Research and Development Centre, London, Canada Agriculture and Agri-Food Canada London Canada

Corresponding author: David Honsberger (dnh8@hawaii.edu)

 2026 14 07 2026 84 609 645 2E835153-773B-59FE-BD12-7FBE3AAFF78E 85C0079D-A47F-4016-9A54-F383D67BF1F2 20 10 2025 29 12 2025 This is an open access article distributed under the terms of the CC0 Public Domain Dedication. https://zoobank.org/85C0079D-A47F-4016-9A54-F383D67BF1F2

Abstract

Despite the recorded presence in Hawai'i of Cephalonomia hyalinipennis, a presumed parasitoid of the coffee berry borer (Hypothenemus hampei), it has never been recovered from infested coffee berries. Here we show, using morphometric and molecular data, that what has been recorded in Hawai'i as C. hyalinipennis is actually a complex of morphologically similar species, each distinct from specimens attacking the coffee berry borer in Chiapas, and from the C. hyalinipennis types from Florida. The taxon from Chiapas utilizing the coffee berry borer is also found to be distinct from C. hyalinipennis sensu stricto, and is here described as C. soconuscensissp. nov. The five species found in Hawai'i are described as Cephalonomia lynneasp. nov., a parasitoid of Hypothenemus eruditus; Cephalonomia luijisp. nov.; Cephalonomia liulasp. nov., a parasitoid of a Leptophloeus sp. known to predate on the coffee berry borer; Cephalonomia kalasp. nov.; and Cephalonomia zandersp. nov., a parasitoid of an Ozognathus sp. Laboratory testing showed that C. lynnea has the ability to develop on coffee berry borer larvae, but it does not appear to do so in the wild. In addition to morphological data, DNA barcoding was used to provide further support for distinguishing species in the C. hyalinipennis species complex. The large dataset of morphometric measurements showing reliability of different ratios, and degenerate PCR primers developed to permit assembly of COI barcodes even from old museum specimens, should enable more rapid discernment within such cryptic species complexes in the future.

 cryptic diversity DNA barcoding integrative taxonomy morphometrics parasitoids parasitoid-host associations Scleroderminae Scolytinae 1. Introduction

Cryptic diversity may comprise a substantial part of the diversity of insects, and is especially important in parasitoid Hymenoptera due to their small size, taxonomic challenges, rarity and inconspicuous nature in the environment, unknown ecology of nearly all species, and overall enormous diversity (Smith et al. 2008; Forbes et al. 2018). Individual species within cryptic species complexes often display some degree of ecological partitioning among the component species, and supposed generalists may turn out to be a complex of more specialist species. For example, a supposed generalist, Ormyrus labotus (Hymenoptera: Ormyridae), associated with over 65 different types of oak galls was found to be actually 16–18 cryptic species each with only a few recorded hosts (Sheikh et al. 2022). A Ganaspis sp. (Hymenoptera: Figitidae) apparently parasitic on a variety of Drosophila (Diptera: Drosophilidae) appeared to have too broad a host range to be useful as a biocontrol agent against D. suzukii (Matsumura, 1931), but turns out to be a species complex, one member of which has host-habitat preferences that make it specific enough to be considered for biocontrol (Seehausen et al. 2020). Orlovskytė et al. (2016) used a combination of molecular, morphological, and ecological characters to make sense of a group of Chrysididae which appears to have undergone sympatric speciation, and found traditional morphological characters alone unable to resolve the complex.

Cephalonomia hyalinipennis Ashmead, 1893 (Hymenoptera: Bethylidae) has been regarded as somewhat of a generalist parasitoid. In recent decades, C. hyalinipennis has received attention because of its attack on Hypothenemus hampei (Ferrari, 1867) (Coleoptera: Curculionidae: Scolytinae), the coffee berry borer, in coffee farms (Pérez-Lachaud et al. 2004; Batchelor et al. 2005; Vega et al. 2015), though due to its generalist tendencies, its low attack rate in coffee, and possible interspecific interactions with other parasitoids of the coffee berry borer, is not thought to be an effective biological control agent against H. hampei (Batchelor et al. 2006). Cephalonomia hyalinipennis has been reported from Hawai'i (Perkins 1913; Bridwell 1920; Nishida 2002), but despite extensive sampling in coffee farms, it has not been found in association with H. hampei infested coffee berries. Motivated by this apparent inconsistency, this study examines the taxa that have been recorded as C. hyalinipennis in Hawai'i, and compares them to the C. hyalinipennis types and specimens known to attack the coffee berry borer, to uncover any cryptic diversity and, where possible, associate this diversity with biological information.

Cephalonomia hyalinipennis has been recorded from the immature stages of Hypothenemus eruditus Westwood, 1836 in a variety of plants (Chittenden 1893; Perkins 1913); from H. hampei in coffee berries (Pérez-Lachaud 1998; Pérez-Lachaud and Hardy 1999); from a variety of other Scolytinae in the tribe Scolytini including Conopthorus, Scolytus, and Pityophthorus; from a molytine weevil (Krombein 1979); as a facultative hyperparasitoid of hymenopteran parasitoids of H. hampei in coffee berries (Pérez-Lachaud et al. 2004); from galls of Cynipidae (Ashmead 1893); and from a variety of locations with hosts unrecorded. It has been considered to be native to the Americas and possibly also Europe (Evans 1978). Cephalonomia hyalinipennis was described by Ashmead (1893) and redescribed and keyed by Evans (1978), but neither description is especially meaningful in the context of the cryptic diversity now known to be present. A diagnosis of the genus Cephalonomia is found in Azevedo et al. (2018). Lists of described species in Cephalonomia are provided by Gordh and Móczár (1990) and Azevedo et al. (2018), and American species found north of México are keyed in Evans (1978). However, aside from Evans (1978), there have been no comprehensive regional revisions, and there are undoubtedly far more species than the <50 currently known (Azevedo et al. 2018).

Species within some genera of Bethylidae do not differ substantially in characters of topological significance, and are distinguished with the help of morphometric ratios, typically of the head or metapectal-propodeal complex (e.g. Magnacca 2020). Such morphometric characters are often utilized in the description of these species, and species are often described from only a few, or sometimes one, individual. Intraspecific variability in such characters is largely unknown. For the species complex explored here, the large dataset of morphological characters obtained may help elucidate the intraspecific variabilty of morphometric ratios, and allows us to test which of these morphometric ratios may be more reliable, and which may be subject to more variability.

To support the morphometric analysis, DNA barcoding was also used; this approach has been useful to distinguish cryptic species with varying degrees of host-specificity that were previously characterized as generalist parasitoids (e.g. Smith et al. 2006). The combination of morphology, natural history, and DNA barcoding has been used to separate presumed polyphagous parasitoid species into several more host-specific cryptic species in braconid wasps (Smith et al. 2008) and tachinid flies (Smith et al. 2006, 2007). This can be of particular value when morphological comparison and non-destructive DNA analysis can be applied to historical specimens, including the vast resources present in museum collections (Mitchell 2015). In the present study, specimens of Cephalonomia varied in their age and preservation status, with some collected recently and preserved in ethanol, and others collected and pinned 10–25 years ago. Obtaining a DNA barcode can present a challenge in older specimens due to DNA degradation (Mitchell 2015), and previous studies have used internal, degenerate primers to obtain several overlapping mini-barcodes that span the entire length of the DNA barcode region (e.g. Strutzenberger et al. 2012; Hebert et al. 2013). This approach was applied in the present study, where several Cephalonomia primers were designed to permit amplification of overlapping fragments of the DNA barcode region in order to obtain sequences from historical and museum specimens that could be used as a basis of comparison for newly-collected specimens from Hawai'i.

 2. Methods 2.1. <italic><tp:taxon-name><tp:taxon-name-part taxon-name-part-type="genus" reg="Cephalonomia">Cephalonomia</tp:taxon-name-part></tp:taxon-name></italic> specimens

Cephalonomia specimens from the Hawaiian Islands used in the morphological and genetic aspects of this study were opportunistically collected as part of a larger survey for parasitoids of bark beetles, or retrieved from insect collections. For collection, plant parts containing bark beetles were found by searching trees and fallen branches, and by cutting living branches, suspending them from the trees, and coming back in 5–8 weeks to check on them. The branches were then dissected by carefully peeling off the bark to search for potentially parasitized individuals within, or the plant parts were placed into rearing containers. Immature parasitoid Hymenoptera were reared to adulthood, and adult wasps were observed, and point mounted or placed into 95% ethanol upon their death. Those placed in ethanol were then mounted, photographed and measured, again placed into ethanol, DNA extracted as described below, and then remounted. Those retrieved from insect collections were from the University of Hawai'i Insect Museum and Hawai'i Department of Agriculture Collection; the glue was dissolved off their pins, their DNA extracted, remounted, and photographed and measured.

Specimens from a lab-reared colony whose founders were collected in H. hampei-infested coffee in the Soconusco region of Chiapas, Mexico were also examined (see Pérez-Lachaud 1998). Some of these were reared on H. hampei, most on the factitious host Caulophilus oryzae (Gyllenhal, 1838) (Coleoptera: Curculionidae), and a sample from the colony was saved in ethanol in 1998. Wasps from this same colony were the subjects of a variety of studies of H. hampei parasitoids, mainly examining intraguild predation (e.g. Pérez-Lachaud and Hardy 1999, 2001; Pérez-Lachaud et al. 2002, 2004). These wasps were mounted, photographed and measured, placed again into ethanol, DNA extracted, and then remounted.

A variety of specimens collected near Jacksonville, Florida, the type locality of Cephalonomia hyalinipennis, were present in the University of Central Florida Collection of Arthropods (UCFC), collected between 1997 and 2013. Individuals that appeared especially close to C. hyalinipennis were photographed and measured, removed from their pins, their DNA extracted, and remounted.

The types of C. hyalinipennis, and other specimens identified as C. hyalinipennis present in the Smithsonian National Museum of Natural History (NMNH) were examined, photographed, and measured. Due to their age and the importance of keeping these specimens intact, DNA extraction was not attempted.

 2.2. Identification of parasitoid-host associations

During dissection of wood, parasitized larvae were occasionally encountered. Ectoparasitic larvae, together with their hosts to which they were attached, were carefully picked up and placed into a small channel cut into a piece of wood, and covered with a glass slide to prevent desiccation and provide a fully enclosed space modeling that of its natural environment to allow the prepupa to construct a pupal cocoon. Parasitoid pupae and prepupae were also found next to the desiccated body of their host. Rearing of the parasitoid and association of the host larva with its adult counterparts determined the association.

Identity of the host was able to be determined if (1) the host larva or pupa was clearly associated with adults in the tunnels based on their relative positions, position of other progeny, and lack of connection to other tunnel systems containing related species, and (2) inspection of the larva or its exuvia, especially the head capsule, was consistent with the taxon and matched the head capsules of other larvae associated with identified adults in the tunnel system.

The discovery of an adult wasp in a tunnel with a parasitized host was not considered strong enough evidence to confirm the parasitoid-host association, as they could be using the tunnel simply as a refuge, searching for a host which is not present, or using the tunnels as pathways to access different hosts that coexist under the surface of wood.

Those reported here are a few of a diverse set of parasitoids, other wood-associated insects, and parasitoid-host associations similarly encountered when dissecting wood as part of a larger study of parasitoids of wood associated-insects in Hawai‘i (see Honsberger 2024).

 2.3. Observations of biology and behavior

Where possible, observations of biology and behavior were made upon discovery of the parasitoid within the plant material. Observations made in this way represent rarely more than a snapshot of their life as removal of the bark presents a drastic change in environment and presumably alters their behavior. Thus the following methods were also explored.

Observation chambers described by Honsberger et al. (2024), consisting of plant material clamped between a sheet of aluminum and a sheet of plexiglass with an aluminum spacer, were used to observe behavior in a semi-natural environment. The plant material consisted either of host wood with channels cut into it with a knife, or wood naturally bored by its phloem-feeding host beetles and peeled through the phloem layer to extract bark with outer half of host tunnels. Host frass and inhabitants of the wood were added back into the tunnels before the plexiglass cover was attached. Photos and videos of observed behavior in these observation chambers were taken using a cellular phone attached to an Olympus CX-31 compound microscope, illuminated using a fiber optic gooseneck microscope light powered by a halogen bulb.

 2.4. Morphological data

Adult female wasps and associated males identified as Cephalonomia nr. hyalinipennis were point mounted and photographed using a Macropod Pro imaging system (Macroscopic Solutions, East Hartford CT), and images were stacked using Zerene Stacker (Richland WA) on a desktop computer. The following five aspects were photographed. 1. Full dorsal view of the head (perpendicular to the ventral plane of the head). 2. Straight on view of the ocellar triangle (perpendicular to the plane of the ocellar triangle) 3. Lateral habitus. 4. Dorsal habitus (perpendicular to plane of metapectal-propodeal disc). 5. Backlit view of the forewing.

Measurements of every individual were obtained as in Fig. 1, using the photographs taken with the Macropod Pro. Measurements that could not be obtained using these images, e.g. if the antennae were obstructing the view of part of the face, were not included in the analyses. Where there were two measurements to be taken because of bilateral symmetry, the mean from each side was used.

10.3897/asp.84.e174388.figure1 7C1FDEB6-4CBC-5E89-8140-C05BD4DEC5F8

Morphometric measurements. Acronyms explained in section 2.4. ACephalonomia liula ♀ side view (paratype); BCephalonomia soconuscensis ♀ dorsal view; CCephalonomia lynnea ♀ face view of head (paratype); DCephalonomia lynnea ocellar view ♀ (paratype); ECephalonomia soconuscensis ♀ wings.

https://binary.pensoft.net/fig/1715967

Acronyms in Fig. 1 and descriptions of the measurements are as follows: LH Length of head; longitudinal line even with posterior of vertex to most anterior point of the face (excluding mandibles and antennae); measured in full dorsal view. – VOL Vertex-ocular line; longitudinal line even with posterior of vertex to even with posterior margin of eye; measured in full dorsal view. – LE Length of eye; posterior margin to anterior margin of compound eye; measured in full dorsal view. – WH Width of head; maximum width of head including eyes; measured in full dorsal view. – WF Width of frons; minimum distance between compound eyes on frons; measured in full dorsal view. – TH Thickness of head; maximum thickness of head along a line perpendicular to ventral plane of head; measured in side view of head. – WHE Width of head across compound eyes; maximum distance between lateral margins of compound eyes; measured in full dorsal view. – WHPE Width of head at posterior of compound eyes; width of head along the line connecting the posterior margin of each compound eye; measured in full dorsal view. – WT Width between toruli; minimum distance between the inner margins of the toruli; measured in full dorsal view. – WM Width across mandibles; distance between the lateral corner of the base of each mandible; measured in full dorsal view with the mandibles closed. – WOT Width of ocellar triangle; distance between outer margins of posterior ocelli; measured in full dorsal view. – OOL Ocellar-ocular line; shortest distance from margin of posterior ocellus to margin of compound eye; measured in full dorsal view. – AOL Lateral-ocellar line; shortest distance between margin of posterior ocellus and anterior ocellus; measured in straight on view of ocellar triangle. – POL Posterior ocellar line; shortest distance between inner margins of posterior ocelli; measured in full dorsal view. – AOT Angle between lines joining middle of anterior ocellus and middle of each posterior ocellus; measured in straight on view of ocellar triangle. – DAO Diameter of anterior ocellus; maximum width of anterior ocellus along a transverse line; measured either in face view or in straight on view of ocellar triangle. – HAO Height of anterior ocellus; length of anterior ocellus along longitudinal line; measured in straight on view of ocellar triangle – LPD Length of metapectal-propodeal disc; in dorsal view, distance along median line from anterior margin of metapectal-propodeal complex to declivity, or posterior carina if present; measured in plane perpendicular to overall surface of disc. – WPD Width of metapectal-propodeal disc; greatest transverse distance between lateral marginal carinae or lateral margin of disc, posterior of propodeal spiracles; measured in plane perpendicular to overall surface of disc. – LP Full length of propodeum; distance along the median line from anterior margin of metapectal-propodeal complex to petiolar foramen; measured in plane perpendicular to overall surface of disc. – HPL Height of propodeum lateral panel; distance from ventral margin of vertically inclined lateral panel of metapectal-propodeal complex, even with anterior reach of hind coxa, to lateral marginal carina of metapectal-propodeal disc; measured in side view. – LTP Length from tegula to petiole; distance along an antero-posterior line parallel with plane of metapectal-propodeal disc, even with anterior reach of tegula to even with petiolar foramen; measured in side view or dorsal view. – LW Length of forewing; distance along a line parallel to straight trailing edge of forewing, from wing base to apex. – WW Width of forewing; maximum width of forewing along a line perpendicular to straight trailing edge of forewing.

In species descriptions, ratios are reported as observed range and number of specimens measured. If the number of individuals was sufficient, the two-sided 90% prediction interval assuming a normal distribution, calculated in JMP Pro 16, is also included. This is written as Ratio = Rangemin–Rangemax, n, [PI90%, lower,PI90%, upper].

 2.5. DNA barcoding of <italic><tp:taxon-name><tp:taxon-name-part taxon-name-part-type="genus" reg="Cephalonomia">Cephalonomia</tp:taxon-name-part></tp:taxon-name></italic>

A total of 120 parasitoids that were either freshly collected or obtained from previous collections were used for DNA barcoding (Table 1). To maintain the integrity of the wasps for point-mounting and morphometric analysis following molecular analysis, DNA was extracted from whole wasps non-destructively using a Chelex extraction protocol. Prior to extracting the bulk of specimens, a trial was conducted with 10 specimens that were first subject to morphometric analysis, followed by Chelex extraction, and then the morphometric analysis was repeated to assess consistency of measurements before and after extraction. This trial confirmed that the characters were not distorted during the process of DNA extraction (data not shown), and thus was applied to the remaining 110 specimens. Extraction was as follows: using flame-sterilized forceps, each wasp was placed in an individual 200 µl well of a 96-well microplate, along with 2 µl of proteinase K (20 mg/ml; (Invitrogen, Carlsbad, CA, USA) and 100 µl of 5% Chelex 100 Molecular Grade Resin (Bio-Rad Laboratories, Hercules, CA, USA). A negative extraction control containing the Chelex and Proteinase K solutions, but no insect tissue, was included in each microplate. Sealed microplates were incubated overnight at 55°C, followed by 10 min at 99°C. Samples were centrifuged at 5800 g for 5 min to pellet the Chelex solution, and 50 µl of supernatant (containing DNA) was transferred to wells in a new plate, taking care not to transfer the Chelex residue along with the sample. Microplates containing the extracted DNA were stored at –20°C until further analysis.

Specimens of Cephalonomia obtained for DNA barcoding, age and preservation status at time of DNA extraction.

Collection location Collection year(s) Preservation status at time of extraction # specimens
Hawai'i, USA 2019–2022 < 3 years old, preserved in 95% ethanol 54
Hawai'i, USA 1996 29 years old, point-mounted museum specimens (University of Hawai‘i Insect Museum) 7
Hawai'i, USA 2013 12 years old point-mounted museum specimens (Hawai‘i Department of Agriculture Collection) 6
Chiapas, Mexico 1998 ~25 years old, preserved in ethanol (unknown whether it was 70% or 95%) 37
Florida, USA 1997–2013 10–26 years old point-mounted museum specimens (University of Central Florida Collection) 16
Σ = 120

PCRs were performed in a 25 µl volume containing 0.125 µl of Taq Platinum, 2.5 µl of 10× PCR buffer, 1.25 µl of 50 mM MgCl2, 0.125 µl of 10 mM dNTPs (Invitrogen, Carlsbad, CA, USA), 0.25 µl of 10 µM forward and reverse primer respectively, 19.5 µl ddH20, and 2 µl of template DNA. In an attempt to get a full DNA barcode sequence regardless of specimen age, a 658 bp sequence of the mitochondrial gene cytochrome oxidase subunit I (COI) was amplified by PCR using primers LCO-1490 and HCO-2198 (Folmer et al. 1994). Thermocycling conditions included initial denaturation at 94°C for 1 min, followed by five cycles of 94°C for 30 s, annealing at 45°C for 40 s, extension at 72°C for 1 min, followed by another 35 cycles of 94°C for 30 s, 51°C for 40 s and 72°C for 1 min and a final extension period of 5 min at 72°C. PCR products were visualized with a QIAxcel Advanced automated capillary electrophoresis system (Qiagen, Hilden, Germany) using the DNA screening cartridge and method AL320. Results were scored with QIAXCEL SCREENGEL Software (version 1.2.0), and only those samples of the expected fragment size with a signal strength exceeding 0.1 relative fluorescent units were scored as positive. Samples scored as positive were purified using ExoSAP-IT (Affymetrix, Santa Clara, CA, USA) following the manufacturer’s instructions. Purified PCR products were bidirectionally sequenced on an ABI 3730 DNA Analyser at the Robarts Research Institute (London Regional Genomics Centre, ON, Canada). Forward and reverse sequences were assembled and edited using CODONCODE ALIGNER, version 9.0.1 (Codon-Code Corporation, Centreville, MA, USA). Sequence data and trace files were uploaded to the Barcode of Life Datasystems (BOLD; www.boldsystems.org) in the Project CEPHH (Parasitoids in the Genus Cephalonomia (Bethylidae) in Hawai'i), which is publicly available.

 2.6. Development of PCR primers for older and/or point-mounted specimens

The ~25 year old specimens from Chiapas in ethanol, as well as all of the point-mounted museum specimens, failed to amplify and sequence using conventional universal barcode primers (LCO-1490/HCO-2198) for the full length of the DNA barcode region. To increase the likelihood of obtaining barcode sequences (preferably barcode-compliant sequences, > 500 bp), degenerate PCR primers nested within the DNA barcode region were designed that amplify up to 4 overlapping amplicons. The overlapping sequences could then ideally be assembled to re-construct a full-length DNA barcode sequence. In order to develop nested primers that would amplify these parasitoids, the DNA barcodes obtained from recently-collected specimens of Cephalonomia from a variety of hosts and host plants in Hawai'i were used as reference sequences to design degenerate primers that should amplify a variety of species within the genus Cephalonomia. Degenerate PCR primers that spanned areas of ~150–400 bp along the length of the DNA barcode region were developed (Table 2). The primers were then used in 4 different combinations (i.e., a separate PCR for each primer pair combination) in order to obtain overlapping fragments that could be assembled to generate DNA sequences covering as much of the barcode region as possible. The primer combinations were as follows: (a) LCO-1490/CeR-300 (~300 bp); (b) CeF-271/HCO-2198 (~400 bp); (c) CeF-271/CeR-431 (~150 bp); (d) CeF-215/CeR-431 (~200 bp). To increase the likelihood of amplifying the degraded DNA from these specimens, an initial PCR was performed using each combination of primers (with the same PCR conditions as described above for DNA barcoding, except with 5 µL of DNA template), and then a reamplification PCR (re-PCR) with the same primers was performed using 2 µL of the PCR product as the template DNA for the PCR reaction. PCR products from this re-PCR were visualized with a QIAxcel Advanced automated capillary electrophoresis system (Qiagen, Hilden, Germany) using the DNA screening cartridge and method AL320. Results were scored with QIAXCEL SCREENGEL Software (version 1.2.0), and only those samples of the expected fragment size with a signal strength exceeding 0.1 relative fluorescent units were scored as positive, and PCR products were purified and bi-directionally sequenced as described above for DNA barcoding. When amplification and sequencing were successful, DNA sequences for up to 8 different PCR fragments (i.e., a forward and reverse sequence for each primer combination) were generated for each specimen, and the fragments were assembled and edited using CODONCODE ALIGNER to generate a single sequence per specimen. Sequences and tracefiles were uploaded to BOLD in the same project (CEPHH) as described above.

Degenerate PCR primers for Cephalonomia.

Forward CeF-271 CCCNTCATTTATNNTTTTAATTTTAAG
Reverse CeR-300 GTNCCNACNCCTTTATTAATAATTC
Forward CeF-215 GAGCNCCNGATATAGCNTTNC
Reverse CeR-431 GTTGAAATAAAATTTGTTGCNCC
2.7. Intra- and interspecific variation in DNA barcode sequences

As the sequence length generated for the specimens was highly variable based on which primers yielded useable stretches of sequence (as expected with degraded / pinned museum specimens), comparison of sequences within a given BIN (i.e., intraspecific variation) was only applied to BINs with 3 or more individuals with sequences with > 400 bp, to ensure coverage over a reasonably informative stretch of sequences for multiple individuals per BIN. Clustal Omega Multiple Sequence Alignment (Madeira et al. 2024) was used to generate a percent identity matrix for specimens within each BIN to evaluate maximum and minimum percent pairwise sequence similarity. Similarly, to provide an indication of sequence variation between BINs (= interspecific variation), consensus sequences for each BIN were aligned and compared using Clustal Omega Multiple Sequence Alignment (Madeira et al. 2024) to generate a percent identity matrix and estimate percent pairwise sequence similarity between BINs. To illustrate distance relationships, a neighbor-joining distance tree was constructed in PAUP* 4.0a169 (Swofford 2021) under the Kimura 2-parameter model, with two Allobethylus sequences from BOLD as outgroups.

 2.8. Analysis of morphometric data

Beside molecular data, morphometric ratios and morphological characters were used together to determine species clusters. A variety of ratios were plotted, and the plots were inspected to determine which of these ratios best separated the taxa determined to be distinct. Concerns of allometry, specifically correlation of ratios with body size, were considered not to be especially important in this study because the range and distribution of body lengths was similar among the species.

The dispersion of morphometric ratios for each clade and sex was also analyzed to determine which may be subject to the least variation, and thus suggest which may more reliably be used in this and related taxa in assigning an individual to a certain species.

Coefficient of variance was calculated as CV=σμ;

Gini coefficients may be a more reliable measure (Valbuena et al. 2012), and were calculated as

G = ∑ i = 1 n ∑ j = 1 n | x i − x j | 2 n 2 μ ;

Where µ = sample mean of measurement; n = number of individuals measured; xi = measurement of ith individual; σ = standard deviation of sample. To compare dispersion among the morphometric ratios, Gini coefficients for each species with enough indivuals successfully sequenced, sexes analyzed separately, were compared using a Welch’s ANOVA and post-hoc Games-Howell tests.

 2.9. Repositories

BPBM Bernice Pauahi Bishop Museum, Honolulu, Hawai'i, USA – CNIN-UNAM National Insect Collection of the Institute of Biology at the National Autonomous University of Mexico, Mexico City, Mexico – ECO-CH-AR Arthropod Collection of El Colegio de la Frontera Sur, Chetumal, Quintana Roo, Mexico – HDOA Hawai'i Department of Agriculture Collection, Honolulu, Hawai'i, USA – UCFC University of Central Florida Collection of Arthropods, Orlando, Florida, USA – UHIM University of Hawai'i Insect Museum, Honolulu, Hawai'i, USA – USNM Smithsonian National Museum of Natural History, Washington DC, USA.

 3. Results 3.1. <italic><tp:taxon-name><tp:taxon-name-part taxon-name-part-type="genus" reg="Cephalonomia">Cephalonomia</tp:taxon-name-part></tp:taxon-name></italic> species considered here to be near <italic><tp:taxon-name><tp:taxon-name-part taxon-name-part-type="genus" reg="Cephalonomia">C.</tp:taxon-name-part> <tp:taxon-name-part taxon-name-part-type="species" reg="hyalinipennis">hyalinipennis</tp:taxon-name-part></tp:taxon-name></italic>

Considered here as part of the C. hyalinipennis species complex are species with female LH:WH ≈ 1.3, macropterous with forewing hyaline and lacking vein M+Cu and closed cells, and otherwise similar morphology to C. hyalinipennis. This combination of characters excludes all the other Nearctic species treated by Evans (1978), as well as the only two described Neotropical species, C. skottsbergi Brues (head nearly twice as long as wide) and C. ulrichi Brues (yellow, female wingless). Cephalonomia gallicola (Ashmead, 1887) and Cephalonomia tarsalis (Ashmead, 1893), the other species known in Hawai'i, are not considered here. Other Cephalonomia spp. which include some found in Hawai'i collections that appear unrecorded and possibly undescribed, are also not considered: in particular some with LH:WH substantially greater, and one series present in the Hawai'i Department of Agriculture Insect Collection with LH:LW ≈ 1.3 but all individuals apterous with associated smaller compound eyes and lacking ocelli.

 3.2. DNA barcoding of <italic><tp:taxon-name><tp:taxon-name-part taxon-name-part-type="genus" reg="Cephalonomia">Cephalonomia</tp:taxon-name-part></tp:taxon-name></italic> specimens

Using the standard DNA barcode primers (LCO-1490 / HCO-2198), full DNA barcode sequences (658 bp in length) were obtained for 51 of the 54 specimens from 2019–2022 collections in Hawai'i, with the 3 remaining specimens producing sequences of a contaminant (phylum Rotifera). In contrast, the older specimens (10–25 year old dry pinned, or ~30 years in EtOH; n = 66) failed to produce DNA barcode sequences using the standard barcode primers (LCO-1490 / HCO-2198). However, when the 4 different combinations of degenerate primers were used in sequential reamplification PCRs (re-PCR) to generate multiple fragments spanning the length of the DNA barcode region, 51 of 66 specimens (~77%) produced sequences of 148–658 bp. Barcode Index Numbers (BINs) were assigned by BOLD to most sequences over 300 bp, with only 7 sequences remaining unassigned to a BIN (due to poor quality sequences). Eleven BINs were generated from the older samples using the newly-designed Cephalonomia degenerate primers, and five BINs were assigned to the full-length sequences from the newly-collected Hawai'i specimens, suggesting a total of 16 BINs (potential species) among the samples analysed here (Table 3). Most of the BINs from specimens collected in Florida contained only one or two specimens, which indicates the presence of more cryptic taxa but makes it difficult to characterize morphological and molecular traits. We nonetheless included the information on these BINs to provide a basis for comparison with the Hawai'i and Chiapas specimens, and so that future studies can build upon the barcode database for Cephalonomia species and possibly provide more data to resolve their placement within this species complex.

Barcode Index Numbers (BINs) and corresponding species names assigned to Cephalonomia specimens.

Collection type Location BIN # specimens
Fresh samples Hawai'i, USA ADX0893 = C. lynnea 23
ACZ1832 = C. liula 19
ADE7970 = C. luiji 7
ADS1574 1
AEJ8807 1
No Sequence / Poor Sequence with no BIN 3
~25 year old specimens, ethanol Chiapas, Mexico AFO8725 = C. soconuscensis 21
No Sequence / Poor Sequence with no BIN 16
Museum specimens Hawai'i, USA - Hawai'i Department of Agriculture Collection ACP6610 = C. zander 7
Hawai'i, USA - University of Hawai'i Insect Museum AFU9765 = C. kala 6
Florida, USA - University of Central Florida Collection AJC1418 1
AET6794 3
AFN3489 1
AFN3490 1
AFN3491 1
ADT1058 1
ADW6560 1
AFN4835 2
No Sequence / Poor Sequence with no BIN 5

Point-mounted museum specimens collected in Hawai'i fell into two BINs: ACP6610 (7 specimens) and AFU9765 (6 specimens). From the recently-collected field samples, most specimens fell into three BINs: ADX0893 (23 specimens), ACZ1832 (19 specimens), and ADE7970 (7 specimens). Notes or species descriptions of members of these 5 BINs are provided below. The remaining recently-collected specimens fell into 2 BINs (ADS1574 and AEJ8807), each only containing a single specimen, but these were morphologically outside the range of the C. hyalinipennis complex as defined here (one with a substantially greater LH:WH ratio, the other micropterous), and not considered further in this study.

 3.3. Intra- and interspecific variation

Once the filters were applied to the available sequences (sequences ≥ 400 bp, BINs with ≥ 3 individuals), a percent identity matrix was constructed to assess intraspecific variation for specimens from 7 of the 16 BINs (Table 4; Table SS1). Similarly, percent identity matrices for interspecific variation between BINs (Table SS2) were constructed for all BINs (with the exception of the two BINs associated with the Hawai‘i specimens that were not part of the complex). Sequence similarity within each BIN was 98.6–100% (Table SS1). In contrast, sequence similarity between BINs was 85.6–97%. All genetic differences within and between BINs were substitional. No indels were observed.

Intraspecific variation between individuals within a BIN. Only those BINs containing three or more individuals with sequences >400bp were included.

BIN / Species # individuals with sequences > 500 bp Minimum pairwise similarity (%)
ADX0893 = C. lynnea 23 100
ACZ1832 = C. liula 19 99.54
ADE7970 = C. luiji 7 100
ACP6610 = C. zander 7 99.79
AFU9765 = C. kala 6 100
AFO8725 = C. soconuscensis 20 98.60
AET6794 3 98.85
3.4. Overview of clades determined from molecular and morphological data

Of the five BINs containing Cephalonomia nr. hyalinipennis from Hawai'i, sequence variation between BINs was ~10% or more, supporting the existence of five separate species. The DNA barcode region of the parasitoid of H. hampei from Chiapas (described here as C. soconuscensis sp. nov.), differed from the Hawai'i specimens by 7.6–13%, and thus it also appears to be distinct. The successfully sequenced specimens that were collected near the C. hyalinipennis type locality (Florida) and are morphologically close to C. hyalinipennis differ from any of the Hawai'i specimens by at least 4.3%, and from the specimens from the H. hampei parasitoid colony by almost 5%. Figure 2 shows a tree-based representation of distances within and between taxa (not to be taken as representative of evolutionary relationships). Two individuals collected in Hawai'i that are morphologically very different (CEPHH009-23 and CEPHH031-23) are also included, and their relatively close relationship to the others suggests that the “C. nr. hyalinipennis” morphotype is plesiomorphic. The C. hyalinipennis types were not sequenced due to their age, value of the specimens, and mounting method, and thus no molecular comparison can be made.

10.3897/asp.84.e174388.figure2 17B1C967-5DFC-58D9-8EAB-29309D480E48

Neighbor-joining distance tree, with specimens labelled by BOLD process ID and new species labelled by name. Cephalonomia nr. hyalinipennis from Florida in blue, two dissimilar Hawai'i specimens in red; both these groups are not treated in this paper.

https://binary.pensoft.net/fig/1715968

Morphological differences also distinguish each of five Hawai'i clades, the H. hampei parasitoid C. soconuscensis sp. nov., and the C. hyalinipennis types. Thus it appears the five species known from Hawai'i, C. soconuscensis sp. nov. from Chiapas, and C. hyalinipennis, are all distinct; the five Hawai'i clades and C. soconuscensis sp. nov. are subsequently described as new species. See Figs 15, 16, and the species notes and descriptions for details on morphological differences. It is entirely possible that none of the sequenced specimens from Florida were actually C. hyalinipennis, as their morphometric ratios appear to differ from those of the types.

The types of C. hyalinipennis are also somewhat morphologically close to Cephalonomia utahensis. Evans (1978) considered only infuscation of the forewing to distinguish them, but did not take into account minute details of sculpture and morphometrics as the current study shows are relevant. In this study, only the plesiallotype male of C. utahensis was examined, and its morphometric ratios differ substantially from each of the four species found in Hawai'i from which males are known, C. soconuscensis sp. nov., and specimens from near the C. hyalinipennis type locality. Thus we consider C. utahensis distinct from the H. hampei parasitoid, and from each of the species found in Hawai'i given that the only species described below for which males are unknown has subhyaline forewings. We cannot comment on its distinction from C. hyalinipennis because only female types of C. hyalinipennis and a male of C. utahensis were examined.

Animalia Gymnophiona Bethylidae 642CA967-5DFB-5BD7-ADE8-9349A82AE2DD Cephalonomia hyalinipennis Ashmead, 1893 Figures 3 , 17 Cephalonomia hyalinipennis Ashmead, 1893: 49 non Cephalonomia hyalinipennis – Pérez-Lachaud 1998; Nishida 2002: 149 (misidentifications) Material examined.

Lectotype (here designated): ♀; USA, Florida, Jacksonville, [reared from gall of] Holcaspis omnivora [= Disholcaspis quercusomnivora Ashmead], April 1887, Type No. 14057. — Paralectotype: ♀; USA, Florida, Jacksonville, [reared from gall of] Amphibolips cinerea [= Amphibolips quercuscinerea (Ashmead)], April 1887, Type No. 14057, USNMENT 01569950.

10.3897/asp.84.e174388.figure3 92C3F08B-B4F8-5C24-9557-A74DFCDB36B4

Cephalonomia hyalinipennis lectotype ♀ (A–E) and paralectotype ♀ (F–K). A, F Face view of head; B, G dorsal view; C side view; D, H ocellar view; E anterior view of head showing mandible; I, J gaster; K: wings.

https://binary.pensoft.net/fig/1715969 Diagnosis.

LH:WH ≈ 1.3 in female. Forewing subhyaline with 2r-rs+Rs vein reduced to fold. Mesoscutellum with reticulate sculpture similar to that of anteromesoscutum and dorsal pronotal area. Metapectal-propodeal disc with posterior carina complete, cuticle reticulate without smooth polished areas, similar sculpture on either side of the posterior carina (in paralectotype slightly smoother anterior of carina). Anterior ocellus circular. Second gastral sternum (S3) entirely conspicuously reticulate, similar to laterotergites.

Distinguished from C. lynnea and C. luiji by having the mesoscutellum entirely reticulate, rather than medially polished. Distinguished from C. liula by the nebulous 2r-rs+Rs vein (sclerotized and tubular distally in C. liula). Distinguished from C. kala by the narrower clypeus (C. kala with clypeus broadly triangular, the toruli separated by more than their width). Distinguished from C. zander by the weaker sculpture on the frons. Distinguished from C. soconuscensis and further distinguished from all the above species by S3 entirely conspicuously reticulate (S3 distinctly more polished than laterotergites in the others) (Fig. 17). All the above are also distinguished by various morphometric ratios: see Figs 15, 16 and species notes and descriptions for relative ranges.

 Relevant characters.

(For morphometric measurements, L = Lectotype, P = Paralectotype) Female: Length: 1.8(L),1.8(P) mm. Head: Frons shining with reticulate sculpture and sparse setae. LH:WH = 1.28(L), 1.27(P); LH:VOL = 2.08(L), 2.27(P); LH:LE = 2.60(L), 2.43(P); LH:TH = 1.93(L), 2.00(P); WF:WH = 0.53(L), 0.54(P); WF:LE = 1.08(L), 1.04(P); WHE:WHPE = 1.01(L), 1.03(P); VOL:LE = 1.25(L), 1.07(P). Anterior ocellus circular in shape. WOT:OOL = 0.44(L), 0.45(P); POL:AOL = 1.04(L), 1.01(P); AOT = 64(L), 61(P). Median lobe of clypeus narrow, slightly elevated but elevation broadly rounded and not acute; toruli not positioned especially widely on face, WT:WM = 0.13(L), 0.19(P). Mandibles tridentate with indistinct fourth tooth, second and third teeth somewhat blunt. Mesosoma: Dorsal pronotal area, posterior of anteromesoscutum, and mesoscutellum all with similarly reticulate sculpture. Metapectal-propodeal disc with lateral marginal carina present, posterior carina present and complete. Metapectal-propodeal disc also with reticulate sculpture, intensity of reticulations relatively even over its surface except subtly heavier medially to appear as weak median carina that becomes decreasingly distinct by posterior margin of disc, and subtly weaker near center of disc aside from faint extension of median carina; declivity also with reticulate sculpture; difference in sculpture on either side of transition from disc to declivity not especially strong. LTP:LH = 1.09(L), 1.06(P); LPD:WPD = 0.99(L), 1.05(P); LPD:LP = 0.73(L), 0.72(P). Metasoma: Generally polished dorsally, S3 distinctly and entirely reticulate; setae present near posterior margin of each segment. Forewing: Macropterous. Sc+R, prestigmal abcissa of radial 1, and small pterostigma sclerotized. 2r-rs+Rs vein absent, reduced to fold. Rs+M, M+Cu, and A indistinctly pigmented. Membrane subhyaline. LW:WW = 2.88(L), 3.04(P). Coloration: Dark brown except: basal antennal segments somewhat lighter, tibiae dark brown medially but fade to yellow-brown basally and apically, tarsi yellow.

 Distribution.

Reported widely from North and South America, Europe, and Hawai'i (Evans 1978; Nishida 2002); however, in light of the revisions here, all these records should be critically reexamined, as it is likely the true C. hyalinipennis has a much more restricted distribution.

 Hosts.

Reared from oak galls (Ashmead 1893), probably attacking inquilinous beetles rather than the gall-inducing wasp.

 Notes.

This species was redescribed by Evans (1978), but not sufficiently to separate it from the other species described here. The additional characters provided should distinguish it not only from those treated here but other potential semi-cryptic species in its native range. The list of incorrect referrals is obviously far from exhaustive, and given the large number of similar taxa found in this paper, likely includes most of the references to C. hyalinipennis cited in the Introduction.

The type status of this species has been somewhat confused. Ashmead (1893) originally described it from two syntype specimens, purportedly one male and one female, both reared from oak galls in Jacksonville, Florida. Evans (1978) later only mentions the female, referring to it as the “type;” his male description is based on a “plesioallotype” (an essentially meaningless designation) collected nearly 80 years later, and from a different locality and habitat. The more intact specimen is definitely a female, and is identifiable as the one Ashmead determined as such by the rearing record; it bears a lectotype label by C. Azevedo from 2003, but this does not appear to have been formally published, so it is designated as such here. The paralectotype, labeled as male by Ashmead, also appears to be a female and is noted as such by Azevedo’s label; it is in pieces glued to a card and with the very apex of its abdomen missing, but has the head and antennal shape more characteristic of females.

 Animalia Gymnophiona Bethylidae 1C4AF36C-8E93-5B04-BD3D-E01AAAF80B14 Cephalonomia lynnea https://zoobank.org/CFFDEF71-C5B8-4C86-8D01-5A1893DFC3E1 Honsberger, Magnacca & Lorenzo-Elarco sp. nov. Figures 1C , 1D , 4 , 5 , 14A , 18A–C , 19A–C Material examined.

Holotype (Fig. 4A–D): ♀; Hawaiian Islands, O'ahu, Wahiawā; 21.5143°N, 158.0419°W, 299 m; ex Spathodea campanulata branches; 29.v.2020; D. Honsberger; AT1-09/Ceph-0008 (BOLD ID CEPHH008-23); (point mounted, UHIM). — Allotype (Fig. 4E–H): ♂; same data as holotype; AT1-08/Ceph-0007 (BOLD ID CEPHH007-23); (point mounted, UHIM). — Paratypes: 14♀, 9♂; all point mounted. Hawaiian Islands, O'ahu, Mānoa; 21.3009°N, 157.8196°W, 43 m; ex Delonix regia seed pod; 23.iv.2021; D. Honsberger; 1♀ DR3-01/Ceph-0010 (BOLD ID CEPHH010-23); (USNM) • Hawaiian Islands, O'ahu, Waimānalo; 21.334°N, 157.711°W, 26 m; ex macadamia husks; 25.ii.2021; D. Honsberger; 1♀; MN4-01/Ceph-0042 (BOLD ID CEPHH042-23) (BPBM) • Hawaiian Islands, O'ahu, Keālia; 21.5768°N, 158.1924°W, 3 m; ex Leucaena leucocephala branches; 15.ii.2021; D. Honsberger; 1♀ HK1-03/Ceph-0029 (BOLD ID CEPHH029-23) (BPBM) • Hawaiian Islands, O'ahu, Keālia; 21.5768°N, 158.1924°W, 3 m; ex Leucaena leucocephala branches; 5.ii.2021; D. Honsberger; 1♀ HK3-01/Ceph-0030 (BOLD ID CEPHH030-23) (BPBM) • Hawaiian Islands, O'ahu, Mānoa; 21.3060°N, 157.8092°W, 57 m; ex Artocarpus altilis branch; 25.xii.2020; D. Honsberger; 4♀ UU1-01/Ceph-0043 (BOLD ID CEPHH043-23) (BPBM), UU1-02/Ceph-0044 (BOLD ID CEPHH044-23) (USNM), UU1-03/Cehy-0009 (BOLD ID CEPHH106-23) (UHIM), UU1-05/Ceph-0046 (BOLD ID CEPHH046-23) (BPBM); 2♂ UU1-04/Ceph-0045 (BOLD ID CEPHH045-23) (BPBM), UU1-06/Ceph-0047 (BOLD ID CEPHH047-23) (UHIM) • Hawaiian Islands, O'ahu, Wahiawā; 21.5143°N, 158.0419°W, 299 m; ex Spathodea campanulata branches; 29.v.2020; D. Honsberger; 1♀ AT1-10 (UHIM); 7♂ AT1-01/Ceph-0001 (BOLD ID CEPHH001-23) (USNM), AT1-02/Ceph-0002 (BOLD ID CEPHH002-23) (BPBM), AT1-03/Ceph-0003 (BOLD ID CEPHH003-23) (UHIM), AT1-04/Ceph-0004 (BOLD ID CEPHH004-23) (USNM), AT1-05/Cehy-0010 (BOLD ID CEPHH107-23) [apex of abdomen removed, genitalia slide mounted] (UHIM), AT1-06/Ceph-0005 (BOLD ID CEPHH005-23) (BPBM), AT1-07/Ceph-0006 (BOLD ID CEPHH006-23) (USNM) • Hawaiian Islands, O'ahu, Mānoa; 21.3288°N, 157.7930°W, 146 m; ex Trema orientalis branches; 12.ii.2020; D. Honsberger; 2♀ GP2-02/Ceph-0025 (BOLD ID CEPHH025-23) (UHIM), GP2-03/Ceph-0026 (BOLD ID CEPHH026-23) (USNM) • Hawaiian Islands, O'ahu, Waimānalo; 21.334°N, 157.711°W, 26 m; ex macadamia husks; 18.xii.2019; D. Honsberger; 3♀ MN2-01/Ceph-0034 (BOLD ID CEPHH034-23) (USNM), MN2-05/Ceph-0037 (BOLD ID CEPHH037-23) (UHIM), MN2-07/Ceph-0039 (BOLD ID CEPHH039-23) (USNM).

10.3897/asp.84.e174388.figure4 168876F6-FA67-55A2-AA3B-8817261DD7F0

Cephalonomia lynnea sp. nov. holotype ♀ (A–D) and allotype ♂ (E–H). A, E Side view; B, F dorsal view; C, G face view of head; D, H ocellar view.

https://binary.pensoft.net/fig/1715970 10.3897/asp.84.e174388.figure5 A93B5784-7718-5040-A93A-8B5F9FFA6671

Cephalonomia lynnea sp. nov. paratype ♀ (A–E) and paratype ♂ (F–J). A, F Side view; B, G dorsal view; C, H face view of head; D, I wings; E, J view of head showing mandible.

https://binary.pensoft.net/fig/1715971 Diagnosis.

LH:WH ≈ 1.3 in female, ≈ 1.2 in male. Forewing subhyaline with 2r-rs+Rs reduced to fold. Mesoscutellum smooth polished except near lateral margin, in distinct contrast with reticulate anteromesoscutum. Metapectal-propodeal disc lacking posterior carina, cuticle lightly reticulate on either side of the transition from disc to declivity. Anterior ocellus subtly oblong oval in shape. BOLD: BIN ADX0893

Distinguished from C. luiji by the metapectal-propodeal complex with posterior of disc and declivity both having somewhat similar weakly reticulate sculpture, and by sculpture on posterior frons above level of compound eyes distinctly weaker than on lower frons. Additionally, females can be distinguished by a combination of WHE:WHPE, LTP:LH, and POL:AOL; males by LH:WH, LH:VOL, VOL:LE, WF:WH, LH:LE, WHE:WHPE, LPD:WPD, and LPD:LP. Distinguished from C. liula by lacking vein 2r-rs+Rs. Distinguished from C. kala, C. zander, C. soconuscensis, and C. hyalinipennis by having the mesoscutellum broadly smooth and polished, not reticulate. Distinguished from the male plesiallotype of C. utahensis by various morphometric ratios, including LH:WH, LH:LE, WF:LE, VOL:LE, and LH:VOL. A female of C. utahensis was not examined, though is reported to have infuscate forewings (subhyaline in C. lynnea).

 Description.

Female: Length range: 1.0–1.4 mm (n = 16). Head: Frons shining with weak reticulate sculpture and sparse setae, sculpture fading posteriorly to become distinctly more smooth on posterior frons. LH:WH = 1.21–1.35, 13, [1.22,1.37]; LH:VOL = 1.91–2.11, 14, [1.95,2.16]; LH:LE = 2.60–2.97, 14, [2.60,2.97]; LH:TH = 2.01–2.23, 6, [1.94,2.30]; WF:WH = 0.59–0.63, 13, [0.58,0.63]; WF:LE = 1.22– 1.41, 13, [1.21,1.43]; WHE:WHPE = 1.00–1.03, 14, [1.00,1.03]; VOL:LE = 1.29–1.43, 14, [1.28,1.43]. Anterior ocellus subtly oblong oval in shape. WOT:OOL = 0.38–0.45, 12, [0.37,0.46]; POL:AOL = 0.91–1.11, 14, [0.88,1.15]; AOT = 57–71, 14, [56,70]. Median clypeal lobe narrow subtriangular, slightly elevated but elevation broadly rounded and not acute; toruli not positioned especially widely on face, WT:WM = 0.11–0.25, 16, [0.12,0.26]. Mandibles tridentate with indistinct fourth tooth, second and third teeth somewhat blunt. Mesosoma: Dorsal pronotal area and anteromesoscutum with reticulate sculpture, mesoscutellum smooth and polished except along lateral margin. Metapectal-propodeal disc with lateral carina present; posterior carina absent, rounded transition from disc to declivity. Anterior of metapectal-propodeal disc with reticulate sculpture, increasingly coarse medially appearing raised as indistinct median carina that fades in strength and width posteriorly, dissipating by posterior margin; reticulations light in posterior half of disc, especially medially aside from faint extension of median carina; declivity also with light reticulations; difference in sculpture on either side of transition from disc to declivity not especially strong. LTP:LH = 1.00–1.16, 14, [0.97,1.16]; LPD:WPD = 0.97–1.15, 15, [0.95,1.15]; LPD:LP = 0.71–0.76, 16, [0.71,0.77]. Metasoma: Polished, setae present near posterior margin of each gastral segment. S3 medially with sculpture faint and generally polished. Forewing: Macropterous. Sc+R, prestigmal abcissa of radial 1, and small pterostigma sclerotized. 2r-rs+Rs vein absent and reduced to fold. Rs+M, M+Cu, and A indistinctly pigmented. Membrane subhyaline. LW:WW = 3.14–3.44, 16, [3.14,3.46]. Coloration: Dark brown except: basal antennal segments and mandible somewhat lighter, tibiae dark brown medially but fading to yellow-brown basally and apically, tarsi yellow. — Male: Length range: 1.0–1.4 mm (n = 10). Similar to female except for the following. Head wider with eyes bulging, and various morphometric differences as below. Antennae with flagellar segments longer, and with longer setae. LH:WH = 1.15–1.23, 10, [1.14,1.23]; LH:VOL = 2.36–2.64, 10, [2.33,2.63]; LH:LE = 2.23–2.47, 9, [2.20,2.53]; LH:TH = 1.87–2.13, 5, [1.67,2.30]; WF:WH = 0.59–0.60, 10, [0.58,0.60]; WF:LE = 1.12–1.28, 9, [1.08,1.28]; WHE:WHPE = 1.05–1.10, 10, [1.04,1.11]; VOL:LE = 0.86–1.05, 9, [0.85,1.06]; WOT:OOL = 0.61–0.73, 8, [0.55,0.77]; POL:AOL = 1.21–1.42, 9, [1.18,1.47]; AOT = 66–76, 10, [66,80]; LTP:LH = 1.09–1.19, 10, [1.05,1.17]; LPD:WPD = 0.90–1.03, 10, [0.87,1.04]; LPD:LP = 0.67–0.75, 10, [0.67,0.78]; LW:WW = 2.80–3.13, 10, [2.75,3.13]. Genitalia (Fig. 14A): Genital capsule slightly wider than long. Harpe approximately two thirds length of gonostipes, posteriorly exceeding gonostipes; penisvalvae slightly exceeds digitus and cuspis, reaching apical of middle of harpe. Penisvalvae divergent apically, inner margin outcurved near apex; maximum width about one third length, and one third maximum width of genital capsule.

 Etymology.

This species is named for Lynnea Slutzky. The species name is to be treated as a noun in apposition. Pa'alāpu'upu'u was selected for the Hawaiian common name. This name describes the pu'upu'u (rough and bumpy) sculpture on this species as pa'alā (water-worn rocks which once used to be rough volcanic rock), which on both the face and metapectal-propodeal disc has subtle and gradual variation as the water-worn pa'alā stones.

 Distribution.

This species is known from the island of O'ahu in Hawai'i. Given its lack of native relatives, use of a host with no native relatives, and records coming from introduced forests, this species is likely adventive in Hawai'i.

 Hosts.

Ectoparasitoid of Hypothenemus eruditus (Coleoptera: Curculionidae: Scolytinae) larvae under bark of Artocarpus altilis, Spathodea campanulata, Trema orientalis, Leucaena leucocephala, in pods of Delonix regia, and husks of Macadamia integrifolia.

 Animalia Gymnophiona Bethylidae 16A3BB07-7A79-54FA-AACF-D5DB643F5F85 Cephalonomia luiji https://zoobank.org/55B52FB6-813E-422D-A033-FF6CAF3E7058 Honsberger, Magnacca & Lorenzo-Elarco sp. nov. Figures 6 , 7 , 14B , 18D–F , 19D–F Material examined.

Holotype (Fig. 6A–D): ♀; Hawaiian Islands, O'ahu, Waimānalo; 21.334°N, 157.711°W, 26 m; ex macadamia husks; 18.xii.2019; D. Honsberger; MN2-10/Cehy-0002 (BOLD ID CEPHH099-23); (point mounted, UHIM). — Allotype (Fig. 6E–H): ♂; Hawaiian Islands, O'ahu, Mānoa; 21.3288°N, 157.7930°W, 146 m; ex Trema orientalis branches; 12.ii.2020; D. Honsberger; GP2-01/Ceph-0024 (BOLD ID CEPHH024-23); (point mounted, UHIM). — Paratypes: 4♀, 1♂; all with same data as holotype; all point mounted. MN2-09/Ceph-0040 (♀) (BOLD ID CEPHH040-23) (USNM), MN2-11/Ceph-0041 (♀) (BOLD ID CEPHH041-23) (BPBM), MN2-03/Ceph-0035 (♀) (BOLD ID CEPHH035-23) (USNM), MN2-04/Ceph-0036 (♀) (BOLD ID CEPHH036-23) (BPBM), MN2-08/Cehy-0001 (♂) (BOLD ID CEPHH098-23) [apex of abdomen removed, genitalia slide mounted] (UHIM).

10.3897/asp.84.e174388.figure6 63DC68E2-8EAC-5F2E-A9F0-8DF522317E3F

Cephalonomia luiji sp. nov. holotype ♀ (A–D) and allotype ♂ (E–H). A, E Side view; B, F dorsal view; C, G face view of head; D, H ocellar view. Note that the apex of the gaster in (A) appears differently in this and other DNA extracted individuals; usually appearing as in Figs 7 and 18 in fresh specimens.

https://binary.pensoft.net/fig/1715972 10.3897/asp.84.e174388.figure7 67A39CA1-932B-54AF-9F42-CC85851683F6

Cephalonomia luiji sp. nov. paratype ♀ (A–D) and paratype ♂ (E–H). A, E Side view; B, F dorsal view; C, G face view of head; D, H wings.

https://binary.pensoft.net/fig/1715973 Diagnosis.

LH:WH ≈ 1.3 in female, ≈ 1.1 in male. Forewing subhyaline with 2r-rs+Rs vein reduced to fold. Metapectal-propodeal disc with weak posterior carina, cuticle smooth anterior of carina, and reticulate on declivity. Mesoscutellum smooth away from its lateral edges, in contrast with reticulate sculpture of anteromesoscutum. BOLD: BIN ADE7970.

See entry on C. lynnea for characters to distinguish these. Distinguished from C. liula by the nebulous 2r-rs+Rs vein. Distinguished from C. kala, C. zander, C. soconuscensis, and C. hyalinipennis by having the mesoscutellum broadly smooth and polished, not reticulate. Distinguished from the male plesiallotype of C. utahensis by various morphometric ratios, including a combination of LH:LE, WF:LE, VOL:LE, LH:VOL, WF:WH. A female of C. utahensis was not examined, though is reported to have infuscate forewings (subhyaline in C. luiji).

 Description.

Female: Length range: 1.5–1.7 mm (n = 7). Head: Frons shining with weak reticulate sculpture and sparse setae, reticulations of even intensity over frons. LH:WH = 1.25–1.33, 7, [1.24,1.35]; LH:VOL = 1.97–2.12, 7, [1.93,2.15]; LH:LE = 2.62–2.91, 7, [2.56,3.00]; LH:TH = 1.99–2.15, 6, [1.89,2.27]; WF:WH = 0.56–0.61, 7, [0.54,0.62]; WF:LE = 1.18–1.31, 7, [1.13,1.36]; WHE:WHPE = 0.99–1.01, 7, [0.99,1.01]; VOL:LE = 1.23–1.46, 7, [1.19,1.53]. Ocelli circular. WOT:OOL = 0.39–0.42, 7, [0.38,0.43]; POL:AOL = 1.01–1.19, 7, [1.01,1.26]; AOT = 62–67, 7, [61,69]. Median clypeal lobe subtriangular, posteriorly somewhat elevated to form a low ridge; toruli not positioned especially widely on face, WT:WM = 0.10–0.17, 7, [0.09,0.20]. Mandibles tridentate with indistinct fourth tooth, second and third teeth pointed. Mesosoma: Dorsal pronotal area and anteromesoscutum with reticulate sculpture, mesoscutellum smooth and polished except along lateral margin. Metapectal-propodeal disc with lateral marginal carina present; weak posterior carina present but may be especially weak medially. Anterior half of metapectal-propodeal disc with somewhat coarsly reticulate sculpture, at least medially and near marginal lateral carina; both lateral and medial reticulations fade in strength posteriorly, medial reticulations dissipating near or well before posterior margin; posterior half otherwise smooth and polished except near lateral marginal carina; declivity with reticulate sculpture; difference in sculpture on either side of posterior carina abrupt and strong. LTP:LH = 0.93–1.04, 6, [0.91,1.09]; LPD:WPD = 0.90–1.14, 7, [0.81,1.18]; LPD:LP = 0.71–0.77, 7, [0.70,0.78]. Metasoma: Polished, setae present near posterior margin of each gastral segment. S3 medially with sculpture faint and generally polished. Forewing: Macropterous. Sc+R, prestigmal abcissa of radial 1, and pterostigma sclerotized. 2r-rs+Rs vein absent, reduced to fold. Rs+M, M+Cu, and A indistinctly pigmented. Membrane subhyaline. LW:WW = 3.07–3.39, 7, [2.96,3.41]. Coloration: Shiny black except: basal antennal segments and mandible somewhat lighter, tibiae dark brown medially but fading to yellow-brown basally and apically, tarsi yellow. — Male: Length range: 1.2–1.3 mm (n = 2). Similar to female except for the following. Head wider with eyes bulging, and various morphometric differences as below. Antennae with flagellar segments longer, and with longer setae. LH:WH = 1.11–1.12, 2; LH:VOL = 2.66–2.76, 2; LH:LE = 2.20–2.23, 2; WF:WH = 0.54–0.58, 2; WF:LE = 1.08–1.15, 2; WHE:WHPE = 1.05–1.05, 2; VOL:LE = 0.81–0.83, 2; WOT:OOL = 0.72–0.78, 2; POL:AOL = 1.39–1.41, 2; AOT = 70–72, 2; LTP:LH = 1.09–1.12, 2; LPD:WPD = 0.86–0.91, 2; LPD:LP = 0.76–0.79, 2; LW:WW = 2.80–2.83, 2. Genitalia (Fig. 14B): Genital capsule narrow. Length of harpe approximately half length of gonostipes, posteriorly exceeding gonostipes; digitus, cuspis, and penisvalvae of subequal posterior extent, reaching apical of middle of harpe. Penisvalvae slightly divergent apically, inner margin straight near apex; maximum width about one third length, and one third maximum width of genital capsule.

 Etymology.

This species is named for Luiji Tamas. The species name is to be treated as a noun in apposition. 'Ōlohekalakala was selected for the Hawaiian common name. This name describes the 'ōlohe (smooth, almost glassy) area among the kalakala (rough and bumpy) sculpture on the metapectal-propodeal disc.

 Distribution.

This species is known from the island of O'ahu in Hawai'i. Given its lack of native relatives, and records coming from introduced forests, this species is likely adventive in Hawai'i.

 Hosts.

None confirmed, but likely Hypothenemus eruditus (Coleoptera: Curculionidae: Scolytinae), with which it was found. Tree hosts include Macadamia integrifolia husks and Trema orientalis branches.

 Animalia Gymnophiona Bethylidae 14091B00-08DF-5247-AD36-8C7AD207198B Cephalonomia liula https://zoobank.org/9DF60DBD-F478-47F0-BEA8-7A474C0688F8 Honsberger, Magnacca & Lorenzo-Elarco sp. nov. Figures 1A , 8 , 9 , 14C , 18G–I , 19G–I Material examined.

Holotype (Fig. 8A–D): ♀; Hawaiian Islands, O'ahu, Mānoa; 21.2946°N, 157.8136°W, 20 m; ex Ficus microcarpa branches; 2.viii.2022; D. Honsberger; FM3-01/Ceph-0011 (BOLD ID CEPHH011-23); (point mounted, UHIM). — Allotype (Fig. 8E–H): ♂; same data as holotype; FM4-05/Ceph-0017 (BOLD ID CEPHH017-23); (point mounted, UHIM). — Paratypes: 14♀, 6♂; all point mounted. Hawaiian Islands, O'ahu, Mānoa; 21.2946°N, 157.8136°W, 20 m; ex Ficus microcarpa branches; 2.viii.2022; D. Honsberger; 7♀ FM4-09/Ceph-0021 (BOLD ID CEPHH021-23) (USNM), FM3-02/Ceph-0012 (BOLD ID CEPHH012-23) (USNM), FM3-03/Ceph-0013 (BOLD ID CEPHH013-23) (BPBM), FM3-04/Ceph-0014 (BOLD ID CEPHH014-23) (USNM), FM3-05/Ceph-0015 (BOLD ID CEPHH015-23) (UHIM), FM4-08/Ceph-0020 (BOLD ID CEPHH020-23) (UHIM), FM4-11/Ceph-0022 (BOLD ID CEPHH022-23) (BPBM) • same data as previous except: ex Ficus microcarpa branch, reared from larva ectoparasitic on Leptophloeus sp. larva under bark; 2♀ FM4-01/Cehy-0003 (BOLD ID CEPHH100-23) (BPBM), FM4-02/Cehy-0004 (BOLD ID CEPHH101-23) (UHIM) • same data as previous except: ex Ficus microcarpa branch, reared from one of two pupal cocoons next to desiccated Leptophloeus sp. larva under bark; 2♀, 2♂ FM4-03/Ceph-0016 (♀) (BOLD ID CEPHH016-23) (USNM), FM4-06/Ceph-0018 (♀) (BOLD ID CEPHH018-23) (BPBM), FM4-04/Cehy-0005 (♂) (BOLD ID CEPHH102-23) (USNM), FM4-07/Ceph-0019 (♂) (BOLD ID CEPHH019-23) (UHIM) • same data as previous except: ex Ficus microcarpa branch, reared from pupal cocoon next to desiccated Leptophloeus sp. larva under bark; 1♀, 2♂ FM4-10 (♀) (UHIM), FM4-12/Ceph-0023 [gaster removed, genitalia slide mounted] (♂) (BOLD ID CEPHH023-23) (UHIM), FM4-13 (♂) (BPBM) • Hawaiian Islands, O'ahu, Wa'ahila; 21.2975°N, 157.8130°W, 56 m; ex Leucaena leucocephala branch, reared from one of two pupal cocoons next to desiccated Leptophloeus sp. larva under bark; 5.ii.2022; D. Honsberger; 1♀, 1♂ HK5-02/Ceph-0033 (♀) (BOLD ID CEPHH033-23) (USNM), HK5-01/Ceph-0032 (♂) (BOLD ID CEPHH032-23) (BPBM) • Hawaiian Islands, O'ahu, Waimānalo; 21.334°N, 157.711°W, 26 m; ex macadamia husks; 18.xii.2019; D. Honsberger; 1♀ (BPBM), 1♂ MN2-06/Ceph-0038 (BOLD ID CEPHH038-23) (USNM).

10.3897/asp.84.e174388.figure8 461F2735-FAAF-51A2-B268-1BEBF0783D07

Cephalonomia liula sp. nov. holotype ♀ (A–D) and allotype ♂ (E–H). A, E Side view; B, F dorsal view; C, G face view of head; D, H ocellar view. Note that the apex of the gaster in (A) appears differently in this and other DNA extracted individuals; usually appears as in Figs 9 and 18 in fresh specimens.

https://binary.pensoft.net/fig/1715974 10.3897/asp.84.e174388.figure9 23D8DB2C-7E51-58BC-AD1F-80759E344F6A

Cephalonomia liula sp. nov. paratype ♀ (A–E) and paratype ♂ (F–I). A, F Side view; B, G dorsal view; C, H face view of head; D, I wings; E anterior view of head showing mandibles.

https://binary.pensoft.net/fig/1715975 Diagnosis.

Forewing with 2r-rs+Rs vein pigmented distally, unsclerotized and not visible basally near pterostigma, sclerotized portion distinctly visible when backlit. LH:WH ≈ 1.3 in females, ≈ 1.1 in males. Mesoscutellum smooth and polished except near lateral margin, in contrast with reticulate sculpture of anteromesoscutum. Metapectal-propodeal disc with posterior carina present and complete; cuticle smooth just anterior of carina, reticulate on declivity. BOLD: BIN ACZ1832.

Both sexes can be distinguished from all other known Cephalonomia spp. by the forewing with the distal part of 2r-rs+Rs vein sclerotized and pigmented, basally reduced to a fold and not otherwise visible. No other known Cephalonomia sp. has a sclerotized 2r-rs+Rs vein.

 Description.

Female: Length range: 1.3–1.6mm (n = 18). Head: Frons shining with weak reticulate sculpture and sparse setae, reticulations of even intensity over frons. LH:WH = 1.26–1.33, 15, [1.26,1.34]; LH:VOL = 2.42–2.65, 15, [2.37,2.63]; LH:LE = 2.33–2.64, 15, [2.34,2.69]; LH:TH = 2.19–2.34, 6, [2.14,2.38]; WF:WH = 0.59–0.62, 15, [0.58,0.63]; WF:LE = 1.04–1.26, 15, [1.03,1.32]; WHE:WHPE = 1.00–1.06, 15, [1.00,1.06]; VOL:LE = 0.88–1.09, 15, [0.89,1.13]. Anterior ocellus somewhat oblong oval. WOT:OOL = 0.48–0.61, 15, [0.46,0.60]; POL:AOL = 1.12–1.41, 13, [1.09,1.47]; AOT = 65–75, 15, [65,77]. Median clypeal lobe narrow, especially constricted between toruli, and somewhat elevated forming a low ridge; toruli positioned somewhat widely on face, WT:WM = 0.17–0.30, 14, [0.14,0.30]. Mandibles tridentate with indistinct fourth tooth, second and third teeth somewhat blunt. Mesosoma: Dorsal pronotal area and anteromesoscutum with reticulate sculpture, mesoscutellum smooth and polished except along lateral edge. Metapectal-propodeal disc with lateral marginal carina; posterior carina present and complete. Anterior half of metapectal-propodeal disc with somewhat coarse reticulate sculpture; median carina indistinctly visible among the sculpture; both lateral and medial reticulations fade in strength and width posteriorly, medial reticulations dissipate by posterior margin; posterior half otherwise smooth and polished, declivity with reticulate sculpture; difference in sculpture on either side of posterior carina abrupt and strong. LTP:LH = 1.02–1.07, 7, [0.99,1.09]; LPD:WPD = 1.03–1.12, 11, [1.01,1.12]; LPD:LP = 0.70–0.83, 11, [0.68,0.83]. Metasoma: Polished, setae present near posterior margin of each gastral segment. S3 medially with sculpture faint and generally polished. Forewing: Macropterous. Sc+R, prestigmal abcissa of radial 1, and pterostigma sclerotized. 2r-rs+Rs vein distinctly sclerotized and pigmented distally, not visible basally other than as light fold, sclerotized portion beginning at about half length of wing. Rs+M, M+Cu, and A veins indistinctly pigmented. Membrane subhyaline. LW:WW = 2.98–3.41, 15, [2.98,3.44]. Coloration: Shiny black except: tibiae brown but sometimes lighter basally and apically, tarsi yellow. Basal antennal segments only slightly if at all lighter than rest of face. — Male: Length range: 0.9–1.2 mm (n = 7). Similar to female except for the following. Head wider with eyes bulging, and various morphometric differences, as below. Antennae with flagellar segments longer, and with longer setae. Metapectal-propodeal disc with median carina and differences in sculpture less distinct or absent. LH:WH = 1.09–1.12, 7, [1.08,1.13]; LH:VOL = 3.11–3.31, 7, [3.04,3.40]; LH:LE = 2.11–2.27, 7, [2.06,2.30]; LH:TH = 1.83–2.05, 5, [1.77,2.16]; WF:WH = 0.59–0.63, 7, [0.59,0.64]; WF:LE = 1.15–1.31, 7, [1.09,1.34]; WHE:WHPE = 1.08–1.12, 7, [1.07,1.13]; VOL:LE = 0.64–0.73, 7, [0.61,0.75]; WOT:OOL = 0.85–0.96, 7, [0.83,0.97]; POL:AOL = 1.47–1.63, 7, [1.42,1.66]; AOT = 79–85, 7, [77,87]; LTP:LH = 1.01–1.13, 6, [0.98,1.16]; LPD:WPD = 0.84–0.94, 6, [0.82,0.97]; LPD:LP = 0.71–0.85, 6, [0.67,0.86]; LW:WW = 2.84–3.16, 7, [2.74,3.25]. Genitalia (see Fig. 14C): Genital capsule slightly longer than wide. Harpe approximately three fourths length of gonostipes; posteriorly exceeding gonostipes; digitus, cuspis, and penisvalvae of subequal extent, reaching apical of middle of harpe. Penisvalvae not divergent, inner margin straight near apex; maximum width about one third length, and one third maximum width of genital capsule.

 Etymology.

The species name is Hawaiian, literally, mirage. A character that distinguishes this species from all other known Cephalonomia is on the wing ('ēheu), where the radial vein has only its distal section sclerotized and pigmented. It appears as a mirage (li'ulā), as sometimes seen over long distances of ocean or between islands in Hawai'i – trees, shoreline, houses, and waves appearing to float on clear air over the ocean. Its Hawaiian common name is 'ēheu li'ulā.

 Distribution.

This species is known from the Hawaiian Islands of O'ahu and Hawai'i. Given its lack of native relatives, its use of a host with no native relatives, and records coming from introduced forests, this species is likely adventive in Hawai'i.

 Hosts.

Leptophloeus sp. (Coleoptera: Laemophloeidae) larvae in tunnels under Ficus microcarpa and Leucaena leucocephala bark (Fig. 20D–F), also found in Macadamia integrifolia husks where host was not confirmed but the Leptophloeus sp. was present. This Leptophloeus sp. is known to be predatory within plant material, and to predate on H. hampei in coffee berries, and is currently being explored as a biocontrol agent against H. hampei in Hawai'i (Follett et al. 2016). Two parasitoids were typically observed on a single host larva, though single parasitism was also observed. There appears to be a circular component to their food web, where a larva of the Leptophloeus sp. was once observed predating on a pupa of C. liula, the parasitoid having developed on another Leptophloeus sp. larva (Fig. 20F).

 Animalia Gymnophiona Bethylidae 65991BAE-03B1-5A0A-B53A-A88B4B533A15 Cephalonomia kala https://zoobank.org/C9FBCA2E-F480-4D55-8BD6-D35568037C0A Honsberger, Magnacca & Lorenzo-Elarco sp. nov. Figures 10 , 18J–L Material examined.

Holotype (Fig. 10): ♀; MacFarms 8/22/96 // MA1-03/Ceph-0107 (BOLD ID CEPHH117-25)// UHIM2015.46859, Univ. of HI Insect Museum (UHIM). — Paratypes: 6♀; all point mounted. Same data as holotype, except MA1-01/Ceph-0105 (BOLD ID CEPHH115-25)// UHIM2015.46896 Univ. of HI Insect Museum (UHIM)• same data as holotype, except MA1-02/Ceph-0106 (BOLD ID CEPHH116-25)// UHIM2015.46856 Univ. of HI Insect Museum (UHIM) • [without collection data label but apparently from same series as holotype] MA1-04/Ceph-0108 (BOLD ID CEPHH118-23) (UHIM) • [without collection data label but apparently from same series as holotype] MA1-05/Ceph-0109 (BOLD ID CEPHH119-25) (UHIM) • [without collection data label but apparently from same series as holotype] MA1-06/Ceph-0110 (BOLD ID CEPHH120-25) (UHIM) • USA: HI, Hawaii; Waimea- DOFAW; tree nursery; 23.iii.2010 // ex. At large on Naio; coll. L. Ishii // Cephalonomia hyalinipennisAshmead 1893; det. B. Hoffman; 19.XI.2020/#2010-074 (HDOA).

10.3897/asp.84.e174388.figure10 A533CEC5-8AE8-582E-9640-FB4CFD85B2BD

Cephalonomia kala sp. nov. holotype ♀. A: Side view; B: Dorsal view; C: Face view of head; D: Ocellar view; E: Mandibles.

https://binary.pensoft.net/fig/1715976 Diagnosis.

LH:WH ≈ 1.2 in female. Forewing subhyaline with 2r-rs+Rs reduced to fold. Median clypeal lobe broadly triangular, toruli somewhat widely set on face. Mesoscutellum with reticulate sculpture; reticulations on pronotal disc fainter and about half the density (each reticulation about twice the size) compared with middle of anteromesoscutum. Metapectal-propodeal disc with posterior carina present only laterally, cuticle with weakly reticulate sculpture in posterior half, stronger reticulations in anterior half and on declivity, abrupt change in sculpture on either side of the posterior carina. Anterior ocellus small and subcircular. BOLD: BIN AFU9765.

Distinguished from other known species in the complex by the broadly triangular median clypeal lobe. Further distinguished from C. lynnea and C. luiji by the mesoscutellum entirely with reticulate sculpture. Further distinguished from C. liula by lacking vein 2r-rs+Rs. Further distinguished from C. zander and C. hyalinipennis by the relative density of reticulations on pronotal disc compared to mesoscutum medially (about half as dense in C. kala, may be slightly less dense but not so distinctly in C. zander and C. hyalinipennis), and the ratios LH:WH and WF:WH; and further from C. hyalinipennis by S3 polished medially. Further distinguished from C. soconuscensis by LH:WH, WF:WH, POL:AOL, and AOT. Distinguished from C. utahensis by its subhyaline forewings.

 Description.

Female: Length range: 1.7–2.2 mm (n = 6). Head: Frons shining with reticulate sculpture and sparse setae; sculpture fading dorsally to become distinctly more smooth on posterior frons, and small smooth area medially just posterior of median clypeal lobe. LH:WH = 1.17–1.22, 6, [1.16,1.24]; LH:VOL = 2.15–2.30, 6, [2.12,2.39]; LH:LE = 2.44–2.54, 6, [2.39,2.55]; LH:TH = 1.90–2.16, 6, [1.79,2.22]; WF:WH = 0.57–0.58, 6, [0.56,0.59]; WF:LE = 1.14–1.20, 6, [1.13,1.22]; WHE:WHPE = 0.99–1.02, 6, [0.99,1.03]; VOL:LE = 1.06–1.18, 6, [0.99,1.20]. Anterior ocellus small and subcircular in shape. WOT:OOL = 0.41–0.43, 6, [0.40,0.43]; POL:AOL = 0.91–1.00, 6, [0.86,1.01]; AOT = 56–60, 6, [54,61]. Median clypeal lobe broadly triangular, not raised; toruli positioned somewhat widely on face, WT:WM = 0.21–0.24, 5, [0.19,0.25]. Mandibles tridentate with indistinct fourth tooth, second and third teeth somewhat pointed. Mesosoma: Dorsal pronotal area, posterior of anteromesoscutum, and mesoscutellum all with reticulate sculpture; reticulations on dorsum of pronotum slightly fainter and distinctly less dense than on anteromesoscutum and mesoscutellum. Metapectal-propodeal disc with lateral carina present, posterior margin carinate only laterally. Anterior of metapectal-propodeal disc with reticulate sculpture, increasingly coarse medially to appear raised as indistinct median carina that fades in strength posteriorly, dissipating by posterior margin; reticulations lighter in posterior half of disc, especially medially aside from faint extension of median carina; declivity with somewhat heavier reticulations; difference in sculpture on either side of transition from disc to declivity distinct but not especially strong. LTP:LH = 0.98–1.07, 6, [0.97,1.10]; LPD:WPD = 0.93–0.99, 6, [0.91,1.01]; LPD:LP = 0.74–0.79, 6, [0.72,0.80]. Metasoma: Polished, setae present near posterior margin of each gastral segment. S3 medially with sculpture faint and generally polished. Forewing: Macropterous. Sc+R, prestigmal abcissa of radial 1, and small pterostigma sclerotized. 2r-rs+Rs vein absent and reduced to fold. Rs+M, M+Cu, and A indistinctly pigmented. Membrane subhyaline. Coloration: Dark brown except: basal antennal segments and mandible somewhat lighter, tibiae yellow-brown, tarsi yellow. — Male: Unknown.

 Etymology.

This species is named for Merkumungu Kaladine Berocan. The species name is to be treated as a noun in apposition. Kalā'anapa, which translates to the gleaming sun, was selected for the Hawaiian common name. Ka lā means both “the sun” and “the dorsal fin”, and connects to the widely triangular, fin-like shape of the median clypeal lobe of this species and appears as the sun rising from the anterior horizon of the face. Kalā'anapa also sounds like kala'anapa, this species with kala (rough) sculpture which shines in the light ('anapa).

 Distribution.

All but one of the specimens from which this species were described were from a series present in UHIM presumably representing one collection event. Some were labeled with “MacFarms // 8/22/96”, the rest unlabeled. No other collection data was present. MacFarms presumably refers to macadamia nut orchards managed by MacFarms on Hawai'i island in Hawai'i. The one additional known specimen is also from Hawai'i island, from Waimea. This species is likely adventive in Hawai'i.

 Hosts.

Unknown, but given its collection location at MacFarms, presumably found in association with macadamia trees.

 Animalia Gymnophiona Bethylidae 7F9360F2-1A8C-5B4C-A5BA-83C29DB903A0 Cephalonomia zander https://zoobank.org/4DB10676-57E4-49B0-80E8-D21F139DF41B Honsberger, Magnacca & Lorenzo-Elarco sp. nov. Figures 11 , 18M–O , 19J–L Material examined.

Holotype (Fig. 11A–E): ♀; “USA: HI, Oahu; Kalihi; 3.IV.2013 // ex Ozognathus sp. poss. exiguus in dead Ficus benjamina branches; coll. W. Nagamine // Cephalonomia sp.; det. W. Nagamine, 20.xii.2013, #2013-333.” // FB1-01/Ceph-0098; (BOLD ID CEPHH108-25) (point mounted, HDOA). — Allotype (Fig. 11F–I): ♂; same data as holotype; FB1-03/Ceph-0100; (BOLD ID CEPHH110-25); (point mounted, HDOA). — Paratypes: 4♀, 1♂, all point mounted in HDOA.”USA: HI, Oahu; Kalihi; 3.IV.2013 // ex. Ozognathus sp. poss. exiguus in dead Ficus benjamina branches; coll. W. Nagamine // Cephalonomia hyalinipennis Ashmead, 1893; det. K. Magnacca, 19.XI.2020, #2013-333.” Individual specimens with bottom label: FB1-05/Ceph-0102 (♀) (BOLD ID CEPHH112-25); FB1-06/Ceph-0103 (♀) (BOLD ID CEPHH113-25); FB1-07/Ceph-0104 (♀) (BOLD ID CEPHH114-25)• same data as holotype and allotype; individual specimens with bottom label: FB1-02/Ceph-0099 (♀) (BOLD ID CEPHH109-25); FB1-04/Ceph-0101(♂) (BOLD ID CEPHH111-25).

10.3897/asp.84.e174388.figure11 7F73761B-F6B0-5A0D-89BB-4F7BC04297D3

Cephalonomia zander sp. nov. holotype ♀ (A–E) and allotype ♂ (F–I). A, F: Side view; B, G: Dorsal view; C, H: Face view of head; D, I: Ocellar view.

https://binary.pensoft.net/fig/1715977 Diagnosis.

LH:WH ≈ 1.3 in female, ≈ 1.1 in male. Forewing subhyaline with 2r-rs+Rs reduced to fold. Mesoscutellum with reticulate sculpture of similar rugosity to anteromesoscutum and dorsal pronotal area. Metapectal-propodeal disc with posterior carina complete, cuticle with lightly reticulate sculpture in posterior half, heavier reticulations in anterior half and on declivity, abrupt change in sculpture on either side of the posterior carina. Anterior ocellus small and subcircular. BOLD: BIN ACP6610

Distinguished from C. lynnea and C. luiji by the entirely reticulate mesoscutellum (C. lynnea and C. luiji with mesoscutellum polished away from lateral margins). Distinguished from C. liula by the unsclerotized 2r-rs+Rs vein. Distinguished from C. kala by the narrower clypeus and toruli separated by less than their width. Distinguished from C. hyalinipennis and C. soconuscensis (both sexes) by the stronger sculpture on the frons; females by the anterior ocellus small and circular; males by the morphometric ratios WF:LE, WF:WH, WH:DAO, and LTP:LP. Further distinguished from C. hyalinipennis by S3 polished medially. All the above are also distinguished by various morphometric ratios: see Figs 21, 22 and species notes and descriptions for relative ranges.

 Description.

Female: Length range: 1.5–2.0 mm (n = 5). Head: Frons shining with reticulate sculpture and sparse setae, reticulations of even intensity over frons except a small smooth area medially just posterior of median clypeal lobe. LH:WH = 1.28–1.30, 5, [1.28,1.31]; LH:VOL = 2.17–2.21, 5, [2.15,2.23]; LH:LE = 2.46–2.57, 5, [2.43,2.61]; LH:TH = 2.07–2.19, 5, [1.99,2.28]; WF:WH = 0.52–0.53, 5, [0.51,0.54]; WF:LE = 1.00–1.05, 5, [0.97,1.07]; WHE:WHPE = 0.99–1.00, 5, [0.99,1.01]; VOL:LE = 1.11–1.18, 5, [1.09,1.21]. Anterior ocellus small and subcircular in shape. WOT:OOL = 0.38–0.42, 5, [0.37,0.44]; POL:AOL = 1.10–1.22, 5, [1.06,1.27]; AOT = 66–71, 5, [62,73]. Median clypeal lobe narrow; nearly flat; toruli not positioned especially widely on face, WT:WM = 0.13–0.15, 5, [0.12,0.16]. Mandibles tridentate with indistinct fourth tooth, second and third teeth somewhat blunt. Mesosoma: Dorsal pronotal area, posterior of anteromesoscutum, and mesoscutellum all with similarly reticulate sculpture. Metapectal-propodeal disc with lateral carina present and posterior carina weak but complete. Anterior of metapectal-propodeal disc with reticulate sculpture, increasingly coarse medially to appear raised as indistinct median carina that fades in strength posteriorly, dissipating by posterior margin; reticulations weaker in posterior half of disc, especially medially aside from faint extension of median carina; declivity with somewhat stronger reticulations; difference in sculpture on either side of transition from disc to declivity somewhat strong. LTP:LH = 0.99–1.07, 5, [0.93,1.09]; LPD:WPD = 0.90–0.95, 5, [0.88,0.96]; LPD:LP = 0.74–0.79, 5, [0.71,0.83]. Metasoma: Polished, setae present near posterior margin of each gastral segment. S3 medially with sculpture faint and generally polished. Forewing: Macropterous. Sc+R, prestigmal abcissa of radial 1, and small pterostigma sclerotized. 2r-rs+Rs vein absent and reduced to fold. Rs+M, M+Cu, and A indistinctly pigmented. Membrane subhyaline. Coloration: Dark brown except: basal antennal segments and mandible somewhat lighter, tibiae dark brown medially but fading to yellow-brown basally and apically, tarsi yellow. — Male: Length range: 1.2–1.5 mm (n = 2). Similar to female except for the following. Head wider with eyes bulging, and various morphometric differences as below. Antennae with flagellar segments longer, and with longer setae. LH:WH = 1.08–1.09, 2; LH:VOL = 2.88–2.94, 2; LH:LE = 2.20–2.22, 2; LH:TH = 1.72–1.76; WF:WH = 0.58–0.60, 2; WF:LE = 1.17–2.22, 2; WHE:WHPE = 1.06–1.07, 2; VOL:LE = 0.76–0.76, 2; WT:WM = 0.14–0.17, 2; WOT:OOL = 0.73–0.75, 2; POL:AOL = 1.48–1.51, 2; AOT 81–83, 2; LTP:LH = 1.06–1.18, 2; LPD:WPD = 0.84–0.86, 2; LPD:LP = 0.77–0.78.

 Etymology.

This species is named for Zander Wieger. The species name is to be treated as a noun in apposition. Kīwa'a'e'ena was selected for the Hawaiian common name. This name refers to the reticulate sculpture which is strong on the head and metapectal-propodeal disc with the kīwa'a (grooves) 'e'ena (extraordinary). 'E'ena means both shy, and extraordinary and fearsome, as this creature must appear wild and fearsome to its host when it appears in its tunnels, but small and unassuming to us humans.

 Distribution.

Known from O'ahu island, Hawai'i. The series from Ukumehame noted below was found on Maui island. This species is likely adventive in Hawai'i.

 Hosts.

This species has been found in Ficus benjamina, recorded from an Ozognathus sp. poss. exiguus (Coleoptera: Ptinidae). The likely conspecific specimens from Maui island were reared from Acacia farnesiana pods.

 Notes.

Nine female individuals on three pins, collected from Acacia farnesiana pods in Ukumehame Valley on Maui island in 1929, were present in UHIM and also appear very close to the specimens noted above. Given their age and mounting method, DNA extraction was not attempted, and their relationship is not fully clear.

 Animalia Gymnophiona Bethylidae 9ABE23C8-63E4-50C2-8049-73D55A78E7A8 Cephalonomia soconuscensis https://zoobank.org/44604A8A-32F2-435B-9869-05F358125A3B Honsberger, Magnacca & Pérez-Lachaud sp. nov. Figures 1B , 1E , 12 , 13 , 14D , 17 Cephalonomia hyalinipennis – Pérez-Lachaud 1998; Pérez-Lachaud and Hardy 1999, 2001; Pérez-Lachaud et al. 2002, 2004 (misidentification) Material examined.

Holotype (Fig. 12A–D): ♀; Mexico, Chiapas, Soconusco region; Parasitoid of Hypothenemus hampei, from colony reared on Caulophilus oryzae; 1998; G. Perez-Lachaud; 16♀ CC1-01/Ceph-0063 (BOLD ID CEPHH063-23); (BOLD ID CEPHH108-25) (point mounted, NMNH). — Allotype (Fig. 12E–H): ♂; same data as holotype; CC1-11/Ceph-0072 (BOLD ID CEPHH072-23); (point mounted, NMNH). — Paratypes: 32♀, 3♂; all point mounted. Same data as holotype; 15♀ CC1-02/Ceph-0064 (BOLD ID CEPHH064-23) (CNIN-UNAM), CC1-03/Ceph-0065 (BOLD ID CEPHH065-23) (BPBM), CC1-04/Ceph-0066 (BOLD ID CEPHH066-23) (NMNH), CC1-05/Ceph-0067 (BOLD ID CEPHH067-23) (NMNH), CC1-06/Ceph-0068 (BOLD ID CEPHH068-23) (CNIN-UNAM), CC1-07/Ceph-0069 (BOLD ID CEPHH069-23) (UHIM), CC1-08/Ceph-0070 (BOLD ID CEPHH070-23) (ECO-CH-AR), CC1-09/Cehy-0006 (BOLD ID CEPHH103-23) (ECO-CH-AR), CC1-10/Ceph-0071 (BOLD ID CEPHH071-23) (CNIN-UNAM), CC2-01/Ceph-0073 (BOLD ID CEPHH073-23) (NMNH), CC2-04/Ceph-0076 (BOLD ID CEPHH076-23) (UHIM), CC2-05/Ceph-0077 (BOLD ID CEPHH077-23) (ECO-CH-AR), CC2-07/Ceph-0079 (BOLD ID CEPHH079-23) (BPBM), CC3-01/Ceph-0080 (BOLD ID CEPHH080-23) (ECO-CH-AR), CC3-02/Ceph-0081 (BOLD ID CEPHH081-23) (NMNH); 3 ♂ CC2-02/Ceph-0074 (BOLD ID CEPHH074-23) (CNIN-UNAM), CC2-03/Ceph-0075 (BOLD ID CEPHH075-23) (NMNH), CC2-06/Ceph-0078 (BOLD ID CEPHH078-23) (ECO-CH-AR) • Mexico, Chiapas, Soconusco region; Parasitoid of Hypothenemus hampei, from colony reared on Caulophilus oryzae; xi.1998; G. Perez-Lachaud; 12♀ CC0-01/Ceph-0052 (BOLD ID CEPHH052-23) (ECO-CH-AR), CC0-02/Ceph-0053 (BOLD ID CEPHH053-23) (BPBM), CC0-03/Ceph-0054 (BOLD ID CEPHH054-23) (NMNH), CC0-04/Ceph-0055 (BOLD ID CEPHH055-23) (CNIN-UNAM), CC0-05/Ceph-0056 (BOLD ID CEPHH056-23) (ECO-CH-AR), CC0-06/Ceph-0057 (BOLD ID CEPHH057-23) (CNIN-UNAM), CC0-07/Ceph-0058 (BOLD ID CEPHH058-23) (BPBM), CC0-08/Ceph-0059 (BOLD ID CEPHH059-23) (NMNH), CC0-09/Ceph-0060 (BOLD ID CEPHH060-23) (BPBM), CC0-10/Cehy-0007 (BOLD ID CEPHH104-23) (CNIN-UNAM), CC0-11/Ceph-0061 (BOLD ID CEPHH061-23) (NMNH), CC0-12/Ceph-0062 (BOLD ID CEPHH062-23) (ECO-CH-AR) • Mexico, Chiapas, Soconusco region; From colony reared on Hypothenemus hampei; xi.1998; G. Perez-Lachaud; 5♀ CB1-01/Ceph-0048 (BOLD ID CEPHH048-23) (NMNH), CB1-02/Ceph-0049 (BOLD ID CEPHH049-23) (CNIN-UNAM), CB1-03/Cehy-0008 (BOLD ID CEPHH105-23) (ECO-CH-AR), CB1-04/Ceph-0050 (BOLD ID CEPHH050-23) (ECO-CH-AR), CB1-05/Ceph-0051 (BOLD ID CEPHH051-23) (BPBM).

10.3897/asp.84.e174388.figure12 83BA64E3-06DE-5377-9C8A-701BDA680F61

Cephalonomia soconuscensis sp. nov. holotype ♀ (A–D) and allotype ♂ (E–H). A, E Side view; B, F dorsal view; C, G: face view of head; D, H ocellar view.

https://binary.pensoft.net/fig/1715978 10.3897/asp.84.e174388.figure13 8618913F-75F5-5EFB-906D-71198276FB82

Cephalonomia soconuscensis sp. nov. paratype ♀ (A–E) and paratype ♂ (F–I). A, F Side view; B, G dorsal view; C, H face view of head; D view of head showing mandible; E, I wings.

https://binary.pensoft.net/fig/1715979 10.3897/asp.84.e174388.figure14 6FAA5E9A-469C-5665-948B-410188D3DE79

Genitalia (♂). ACephalonomia lynnea sp. nov.; BCephalonomia luiji sp. nov.; CCephalonomia liula sp. nov.; DCephalonomia soconuscensis sp. nov.

https://binary.pensoft.net/fig/1715980 10.3897/asp.84.e174388.figure15 BD33B12B-B764-59AD-AE3F-6BC34989642D

Morphometric ratios among Cephalonomia spp. nr. hyalinipennis (females): the five species found in Hawai'i; C. soconuscensis; C. nr. hyalinipennis from near the C. hyalinipennis type locality in Florida; and the C. hyalinipennis lectotype and paralectotype. Acronyms for morphometric ratios explained in section 2.4. and shown in Fig. 1.

https://binary.pensoft.net/fig/1715981 10.3897/asp.84.e174388.figure16 EA1C447B-EAEE-53AE-B3BD-7EEDF14856E3

Morphometric ratios among Cephalonomia spp. nr. hyalinipennis (males): the four species found in Hawai'i for which males are known; C. soconuscensis; and C. nr. hyalinipennis from near the C. hyalinipennis type locality in Florida. Acronyms for morphometric ratios explained in section 2.4. and shown in Fig. 1.

https://binary.pensoft.net/fig/1715982 10.3897/asp.84.e174388.figure17 C01E2280-208F-594D-BFC6-6006E3630EB4

Ventral view of C. hyalinipennis lectotype (above) and C. soconuscensis holotype (below). Sculpture of the second gastral sternum (S3) can be used to differentiate the two species.

https://binary.pensoft.net/fig/1715983 Diagnosis.

LH:WH ≈ 1.3 in female, ≈ 1.10–1.15 in male. Forewing subhyaline with 2r-rs+Rs vein reduced to fold. Mesoscutellum with reticulate sculpture similar to anteromesoscutum and dorsal pronotal area. Metapectal-propodeal disc with posterior carina complete but medially weak; sculpture smoother just anterior of carina, stronger on declivity. Anterior ocellus somewhat oblong oval. Second gastral sternum (S3) generally polished medially. BOLD: BIN AFO8725.

Distinguished from C. lynnea and C. luiji (both sexes) by having reticulate sculpture on the mesoscutellum similar to that on the anteromesoscutum (C. lynnea and C. luiji with polished sculpture on mesoscutellum away from lateral margins). Distinguished from C. liula (both sexes) by having 2r-rs+Rs unsclerotized and reduced to a fold (C. liula with distal section of 2r-rs+Rs sclerotized). Distinguished from C. kala (females) by the narrower clypeus (broadly triangular in C. kala). Distinguished from C. zander (both sexes) by the weaker sculpture on the frons, becoming polished posteriorly on posterior frons, and larger more oblong anterior ocellus (C. zander with frons entirely with stronger reticulate sculpture, and anterior ocellus small and circular). Distinguished from C. hyalinipennis (females) by metapectal-propodeal disc polished posterolaterally, and S3 generally polished (C. hyalinipennis with metapectal-propodeal disc entirely reticulate, and S3 evenly and distinctly reticulate) (Fig. 17).

 Description.

Female: Length range: 1.5–1.9 mm (n = 33). Head: Frons shining with reticulate sculpture and sparse setae, sculpture fading dorsally to become distinctly more smooth on posterior frons, a small smooth area medially just posterior of median clypeal lobe. LH:WH = 1.26–1.36, 31, [1.25,1.35]; LH:VOL = 2.12–2.37, 31, [2.11,2.28]; LH:LE = 2.38–2.66, 30, [2.44,2.68]; LH:TH = 1.93–2.23, 20, [1.92,2.20]; WF:WH = 0.53–0.59, 31, [0.53,0.59]; WF:LE = 1.00–1.18, 30, [1.04,1.18]; WHE:WHPE = 1.00–1.05, 31, [1.00,1.04]; VOL:LE = 1.01–1.24, 30, [1.08,1.26]. Anterior ocellus somewhat oblong oval in shape. WOT:OOL = 0.42–0.53, 31, [0.42,0.51]; POL:AOL = 1.10–1.43, 31, [1.10,1.39]; AOT = 65–76, 31, [65,74]. Median clypeal lobe narrow subtriangular; relatively flat; toruli not positioned especially widely on face, WT:WM = 0.12–0.23, 31, [0.13,0.24]. Mandibles tridentate with indistinct fourth tooth. Mesosoma: Dorsal pronotal area, posterior of anteromesoscutum, and mesoscutellum all with similarly reticulate sculpture, but reticulations somewhat less dense on dorsal pronotal area. Metapectal-propodeal disc with lateral carina present and posterior carina complete but may be weak medially. Anterior of metapectal-propodeal disc with reticulate sculpture, increasingly coarse medially to appear raised as indistinct median carina that fades in strength posteriorly, dissipating by posterior margin; reticulations weaker in posterior half of disc, especially medially aside from faint extension of median carina; declivity with somewhat stronger reticulations; difference in sculpture on either side of transition from disc to declivity somewhat strong. LTP:LH = 1.07–1.18, 28, [1.06,1.18]; LPD:WPD = 0.96–1.10, 26, [0.98,1.09]; LPD:LP = 0.72–0.79, 26, [0.72,0.80]. Metasoma: Polished, setae present near posterior margin of each gastral segment. S3 medially with sculpture faint and generally polished. Forewing: Macropterous. Sc+R, prestigmal abcissa of radial 1, and small pterostigma sclerotized. 2r-rs+Rs vein absent and reduced to fold. Rs+M, M+Cu, and A indistinctly pigmented. Membrane subhyaline. Coloration: Dark brown except: basal antennal segments and mandible somewhat lighter, tibiae dark brown medially but fade to yellow-brown basally and apically, tarsi yellow. — Male: Length range: 1.2–1.4 mm (n = 4). Similar to female except for the following. Head wider with eyes bulging, and various morphometric differences as below. Antennae with flagellar segments longer, and with longer setae. LH:WH = 1.08–1.16, 6, [1.07,1.18]; LH:VOL = 2.62–2.87, 6, [2.53,2.95]; LH:LE = 2.18–2.21, 4, [2.15,2.24]; LH:TH = 1.75–1.82, 3, [1.66,1.91]; WF:WH = 0.55–0.57, 6, [0.54,0.60]; WF:LE = 1.06–1.10, 4, [1.03,1.13]; WHE:WHPE = 1.08–1.15, 6, [1.05,1.15]; VOL:LE = 0.77–0.83, 4, [0.74,0.88]; WT:WM = 0.21–0.24, 5, [0.19,0.25]; WOT:OOL = 0.78–0.86, 6, [0.75,0.89]; POL:AOL = 1.57–1.84, 6, [1.46,1.94]; AOT = 78–83, 6, [77,84]; LTP:LH = 1.07–1.14, 5, [1.05,1.18]; LPD:WPD = 0.83–0.91, 3, [0.75,1.00]. Genitalia (Fig. 14D): Genital capsule slightly longer than wide. Harpe approximately two thirds length of gonostipes, posteriorly exceeding gonostipes; penisvalvae slightly exceeds digitus and cuspis, extent subequal to that of harpe. Penisvalvae slightly divergent apically, inner margin outcurved near apex; maximum width about two thirds length, and three fourths maximum width of genital capsule.

 Etymology.

The name refers to the Soconusco region in Chiapas, Mexico, where this species was initially collected.

 Distribution.

Type specimens are from a lab-reared colony whose founders were collected in Huehuetán and Tuxtla Chico in the Soconusco region of Chiapas, Mexico.

 Hosts.

Founders were collected from Hypothenemus hampei infested coffee berries, and laboratory reared on either H. hampei or Caulophilus oryzae.

 Notes.

The individuals on which the description is based are from laboratory colonies initiated from individuals emerging from H. hampei infested coffee berries in Chiapas, Mexico, and identified at the time as C. hyalinipennis; they were the focus of a variety of studies on H. hampei biocontrol and parasitoid competitive interactions (e.g. Pérez-Lachaud 1998; Pérez-Lachaud and Hardy 1999, 2001; Pérez-Lachaud et al. 2002, 2004). Some were reared on H. hampei, some were reared on Caulophilus oryzae (see Pérez-Lachaud et al. 2001). This species is a semi-gregarious ectoparasitoid, i.e. clutch size ranges from single to multiple eggs, and individual wasp size varies according to the size of the host and clutch size (Pérez-Lachaud and Hardy 1999, 2001). When developing hyperparasitically, specimens are smaller (Pérez-Lachaud et al. 2004). Relative to the other species considered in this study, this species is especially close morphologically to C. hyalinipennis.

 3.12. Key to female <italic><tp:taxon-name><tp:taxon-name-part taxon-name-part-type="genus" reg="Cephalonomia">Cephalonomia</tp:taxon-name-part> <tp:taxon-name-part taxon-name-part-type="species" reg="hyalinipennis">hyalinipennis</tp:taxon-name-part></tp:taxon-name></italic> species complex in Hawai'i and the coffee berry borer parasitoid <italic><tp:taxon-name><tp:taxon-name-part taxon-name-part-type="genus" reg="Cephalonomia">C.</tp:taxon-name-part> <tp:taxon-name-part taxon-name-part-type="species" reg="soconuscensis">soconuscensis</tp:taxon-name-part></tp:taxon-name></italic>

Considered here as part of the C. hyalinipennis species complex are: Cephalonomia spp. with female LH:WH ≈ 1.3, and macropterous with forewing lacking vein M+Cu and having no closed cells.

1 Distal part of 2r-rs+Rs vein sclerotized and pigmented; metapectal-propodeal disc with posterior carina complete, sculpture smooth anterior of carina and reticulate on declivity C. liula
1’ 2r-rs+Rs vein reduced to fold; other characters not in above combination 2
2 Mesoscutellum smooth and almost polished except along lateral margin (at most very faint reticulations may be visible under high magnification), distinctly contrasting with reticulate sculpture of dorsal pronotal area and anteromesoscutum 3
2’ Mesoscutellum with conspicuous reticulate sculpture, similar to that of pronotal disc and anteromesoscutum 4
3 Metapectal-propodeal complex lacking posterior carina; posterior of disc and declivity both with somewhat similar weakly reticulate sculpture. Reticulate sculpture distinctly weaker on posterior frons above level of compound eyes than on lower frons. Median clypeal lobe broadly elevated, not narrowed to a distinct ridge. Anterior ocellus somewhat oblong oval C. lynnea
3’ Metapectal-propodeal complex with weak posterior carina; posterior of disc with smooth sculpture contrasting strongly with reticulate sculpture on declivity. Strength and density of reticulations similar over entire frons. Median clypeal lobe somewhat more acutely elevated forming a weak longitudinal ridge. Anterior ocellus subcircular C. luiji
4 Frons entirely conspicuously and rather strongly reticulate, somewhat dull at low magnification, lines distinctly engraved at high magnification; clypeus narrow, toruli separated by less than their width; metapectal-propodeal disc with polished area posterolaterally; S3 mostly smooth medially C. zander
4’ Frons smooth and shining with very fine, weak reticulation between eyes, becoming faint and almost polished posteriorly on vertex; other characters often variable 5
5 Head broader, LH:WH 1.16–1.24; clypeus broadly triangular, toruli separated by distinctly greater than their width; ocellar triangle strongly acute, POL:AOL <1.01, WOT:OOL ≤0.43; propodeum entirely reticulate, with median carina at least on anterior half C. kala
5’ Head narrower, LH:WH 1.25–1.35; clypeus narrowly triangular, toruli separated by less than their width; ocellar triangle weakly acute, POL:AOL >1.10, WOT:OOL ≥0.42; propodeum variable, if entirely reticulate then without median carina 6
6 Head nearly polished laterally between eye and ocelli; metapectal-propodeal disc polished posterolaterally; gaster with second sternum (S3) polished medially, at most with very faint reticulation (Fig. 17) C. soconuscensis (coffee berry borer parasitoid)
6’ Head and propodeum smoothly but distinctly reticulate all over, without polished areas; gaster with S3 entirely conspicuously reticulate, similar to laterotergites C. hyalinipennis s.str.
3.13. Observations of the biology and behavior of <italic><tp:taxon-name><tp:taxon-name-part taxon-name-part-type="genus" reg="Cephalonomia">C.</tp:taxon-name-part> <tp:taxon-name-part taxon-name-part-type="species" reg="lynnea">lynnea</tp:taxon-name-part></tp:taxon-name></italic>

Figures 20A–C, 21, 22A–H; File S1

10.3897/asp.84.e174388.figure18 537AB876-3177-5CAE-8145-F3DA386EF333

Cephalonomia spp. females near C. hyalinipennis found in Hawai'i. C. lynnea (A–C), C. luiji (D–F), C. liula (G–I), C. kala (J–L), C. zander (M–O).

https://binary.pensoft.net/fig/1715984 10.3897/asp.84.e174388.figure19 E51B612C-BE23-5A59-83E2-813C0FCABF3C

Cephalonomia spp. males near C. hyalinipennis found in Hawai'i. C. lynnea (A–C), C. luiji (D–F), and C. liula (G–I), C. zander (J–L) (males are unknown for C. kala).

https://binary.pensoft.net/fig/1715985 10.3897/asp.84.e174388.figure20 96D9CF53-7834-5D73-A9E3-AB47FB7D84D7

Host associations of Cephalonomia lynnea (A–C) and Cephalonomia liula (D–F). All photographs except C are of naturally occurring situations, taken while peeling bark from branches found in forests. A: Hypothenemus eruditus larva with a Cephalonomia lynnea larva developing ectoparasitically on it, the adult wasp in the tunnel system possibly the mother of the ectoparasitic larva, guarding it; B: Cephalonomia lynnea pupa having developed on the desiccated H. eruditus larva next to it, the parasitoid removed from its pupal cocoon; C: Cephalonomia lynnea larva developing on a H. hampei larva, its mother having stung and parasitized the beetle larva in a phloem sandwich observation chamber, after switching out H. eruditus larvae in host wood with H. hampei larvae; D: Two Leptophloeus sp. larvae, each with a Cephalonomia liula egg attached; E: Two Cephalonomia liula cocoons, the parasitoids having developed on the dessicated Leptophloeus sp. larva adjacent and stuck to them; F: Leptophloeus sp. larva consuming the inside of a Cephalonomia liula cocoon, the parasitoid having developed on another Leptophloeus sp. individual.

https://binary.pensoft.net/fig/1715986 10.3897/asp.84.e174388.figure21 BF62C6DE-E803-50D8-9B6F-39CDC277C6C4

Behavior of C. lynnea in phloem sandwich observation chambers. A, B: Stinging; C–F: Chewing for host feeding or malaxation, mandibles moving in opposite phase as a tile nipper at the beginning of a chewing episode (E), and in phase as grinding for a prolonged period subsequently (F). The larvae in A, C–F are H. eruditus; in B is H. hampei.

https://binary.pensoft.net/fig/1715987 10.3897/asp.84.e174388.figure22 A2CA3602-C7D0-5F1F-BA6C-D14D46715B1B

Development of C. lynnea on H. eruditus larva (A–H), and C. liula on Leptophloeus sp. larva (I–L). Time elapsed after presence of an egg was first observed on the host are noted in the photographs. The parasitized beetles were found in natural situations upon opening wood, and the host with the ectoparasitoid was moved into a hole in the wood substrate and covered with a piece of glass slide cover to create an enclosed environment.

https://binary.pensoft.net/fig/1715988

Observation chambers were filled with Delonix regia pods containing H. eruditus gallery systems, into which were placed H. eruditus larvae and pupae collected from the same plant material.

Host searching, host encounter, and initial stinging. Female C. lynnea, collected from macadamia nuts and placed into the chambers, rapidly and agilely moved through the scolytid tunnels in the plant material searching for hosts, and often quickly encountered them. Upon encountering a larva the wasp did not always appear to show overt interest, but when she did, it manifested in antennation of the larva, sometimes followed by rapidly turning around and probing her gaster in the general direction of the larva to sting it, or climbing on the larva to sting it. There did not appear to be any effort to avoid stinging in the vicinity of the head of the larva, despite the presence of the larva’s active mandibles. Stinging typically was composed of a single or multiple probes of the gaster, or bouts of repeated probes interspersed with movement on or around the host; the sequence lasting from a few seconds to a few minutes. After this time, the larva had often become substantially less active than it had been before, though it was unclear if this was directly related to venom injected by the wasp or the result of exertion. The larva would eventually cease motion, probably due to the venom.

Chewing. Stinging was typically followed by a chewing behavior, either just subsequent to the stinging, or after a variable period of time where the wasp remained inactive in the tunnel near the larva. Two modes of chewing were observed. Often, in initial exploration of the larva with the mandibles, the mandibles would move in phase with each other (as pliers or tile nippers), and then in more prolonged chewing focusing on one spot, the mandibles would move in opposite phase (the teeth grinding) (both modes visible in Video 1, 0:58–1:06). This chewing, especially with the mandibles in phase, appeared to penetrate the cuticle, sometimes after the first few bites but often after substantial effort, and the wasp appeared to host feed. It was unclear in many instances whether the purpose of chewing was for host feeding, malaxation to explore the quality of the host, or an attempt at host feeding in which the cuticle was not penetrated. Stinging, host feeding, or malaxation of pupae was not observed, but this should not imply that it does not occur because relatively few were available to the wasps.

Oviposition and development. The act of oviposition was not observed, but young ectoparasitic larvae were found attached to their host, either a late instar larva or a pupa. When a young parasitoid larva was observed on a host, a female wasp was occasionally found nearby in the tunnels, sometimes immediately adjacent and motionlessly facing it, as if host guarding. Similar behavior was also observed in naturally inhabited wood from forest environments: in host galleries with early developing larvae, an adult female was occasionally observed in the nearby vicinity. Over the next few days, the parasitoid larva would transfer the hemolymph of the host into its own body, detach from the larva, and spin a silk pupal cocoon in which it would pupate. If attempting to pupate in an unenclosed environment such as a flat surface or a shallow depression, the prepupa often appeared to have trouble spinning a pupal cocoon. It would appear to use substantial energy over a few days in an often unsuccessful attempt to cover itself, and sometimes eventually constructing somewhat of a cocoon, or end up with just a ball of silk at its posterior. Pupation usually still occurred. When in a more fully enclosed environment resembling that of a tunnel, such as the indention covered with a glass slide used for rearing, the parasitoid larva created a more dense, uniform cocoon in which it pupated.

 3.14. Parasitism of <italic><tp:taxon-name><tp:taxon-name-part taxon-name-part-type="genus" reg="Hypothenemus">H.</tp:taxon-name-part> <tp:taxon-name-part taxon-name-part-type="species" reg="hampei">hampei</tp:taxon-name-part></tp:taxon-name></italic> by <italic><tp:taxon-name><tp:taxon-name-part taxon-name-part-type="genus" reg="Cephalonomia">C.</tp:taxon-name-part> <tp:taxon-name-part taxon-name-part-type="species" reg="lynnea">lynnea</tp:taxon-name-part></tp:taxon-name></italic> in a laboratory environment

Figures 20C, 21B

Cephalonomia lynnea adults that had emerged from macadamia wood, presented in the observation chambers with H. hampei larvae taken from coffee berries and placed into active H. eruditus gallery systems, successfully stung and oviposited on one of the five H. hampei larvae it was presented with. The larva developed and completed its life cycle on the H. hampei. Despite extensive collections from coffee in Hawai'i, this species has never been found inhabiting that environment, or parasitizing H. hampei.

 3.15. Dispersion of morphometric measurements

Figure 23

10.3897/asp.84.e174388.figure23 EC0EF32B-2702-5D67-974F-D1417AFD2331

Dispersion of morphometric ratios: coefficient of variation and Gini index, for females and males. Gini coefficients may be a more meaningful measure for statistical comparison, but coefficient of variation is also shown because it changes linearly with respect to standard deviation, and thus may provide a more easily interpreted visual comparison.

https://binary.pensoft.net/fig/1715989

For females, enough individuals were measured to calculate the Gini coefficients for six species: the five Hawai'i species and C. soconuscensis. The Welch’s ANOVA was significant (F22,40.6 = 11.03, p < 0.0001), and post-hoc Games-Howell tests show WHE:WHPE had significantly lower dispersion than many of the other ratios, and LH:WH and WF:WH had lower dispersion than some. While not significantly different from many of the other ratios, LH:VOL also had low dispersion, while ratios invoking WT and HPL had relatively high dispersion for some species, as did some of the ratios involving the ocelli and metapectal-propodeal disc.

For males, enough individuals were measured to calculate Gini coefficents for only three species: C. lynnea, C. liula, and C. soconuscensis. The pattern of dispersion among the measurement ratios appeared to follow a similar pattern to that of the females; Welch’s ANOVA was significant (F22,15.8 = 2.62, p = 0.0269), but in part because of the paucity of repetitions, and low power resulting from the large number of comparisons, post-hoc Games-Howell tests showed little of interest.

Despite an overall paucity of significant differences, some tentative patterns can still be discerned. The ratio found in both sexes to have near the least dispersion was WHE:WHPE (bulging of the eyes), and also appeared somewhat useful in distinguishing species in the complex. Various commonly utilized ratios invoking characters of the head and compound eyes, including LH:WH, LH:VOL, and WF:WH, also had low dispersion and distinguished species, and this analysis supports their utility. Other supported ratios include LH:TH and LTP:LH. Ratios of the metapectal-propodeal disc have proven to be useful among taxa with more morphological variety, but were helpful in distinguishing only some of the species in this complex, and also showed relatively high dispersion. The ratio with highest dispersion for some species was WT:WM, and to a smaller degree those invoking HPL, but a substantially greater WT:WM was a useful character to distinguish C. kala from the otherwise similar C. zander. The ocellar ratios WOT:OOL and POL:AOL are often utilized in Bethylidae and Hymenoptera in general, but despite their somewhat concerning dispersion (both were also found to sometimes vary bilaterally in a single individual), they still appeared useful in delineating some of the species in the C. hyalinipennis complex examined here. Thus, low dispersion did not always appear to correlate with utility to distinguish species, at least in this complex, but may provide one indication. It is also worth noting that ratios useful in delineating species were different among the sexes. For example, the ocellar measurements WOT:OOL and AOT distinguished some species of males where they did not in the females; LH:WH was similarly useful for distinguishing the males, despite females all having a similar ratio of near 1.3, which defines the complex. Between morphologically close groups, especially C. lynnea and C. luiji, female intraspecific variation was such that a single morphometric ratio to reliably distinguish all individuals was not identified, but multiple ratios when taken together appear to distinguish them with some confidence.

 4. Discussion 4.1. <italic><tp:taxon-name><tp:taxon-name-part taxon-name-part-type="genus" reg="Cephalonomia">Cephalonomia</tp:taxon-name-part> <tp:taxon-name-part taxon-name-part-type="species" reg="hyalinipennis">hyalinipennis</tp:taxon-name-part></tp:taxon-name></italic> species complex

Five species morphologically close to C. hyalinipennis (females with LH:WH ≈ 1.3, macropterous, forewing lacking vein M+Cu and no closed cells) were found in Hawai'i, each distinct from C. hyalinipennis as determined by the types. Cephalonomia soconuscensis, associated with coffee and H. hampei in Chiapas, and the focus of a variety of studies by Pérez-Lachaud and Hardy between 1998 and 2004, was also found to be distinct morphologically and molecularly from each of the five species in Hawai'i, and clearly though more subtly morphologically distinct from the true C. hyalinipennis. There is substantial additional genetic variation among morphologically similar specimens collected in the area of the type locality of C. hyalinipennis in Florida, and thus this species complex is likely to be substantially more diverse than is reported here. There is also host partitioning, the component species in this complex appearing to show apparent specificity: C. lynnea on H. eruditus (Curculionidae: Scolytinae); C. liula on Leptophloeus sp. (Laemophloeidae); C. kala on Ozognathus poss. exiguus (Ptinidae); C. sonocusensis from Chiapas on H. hampei (Curculionidae: Scolytinae) and factitious hosts in a laboratory and as a facultative hyperparasitoid of H. hampei; and C. hyalinipennis s.str. on Cynipidae galls.

Thus what has been referred to as C. hyalinipennis is a species complex, and no species associated with H. hampei currently appears to be present in Hawai'i. This explains why despite the recorded presence of C. hyalinipennis in Hawai'i it has not been found directly associated with the coffee berry borer. Interestingly, C. liula utilizes the same Leptophloeus sp. that is being explored for biocontrol of H. hampei in Hawai'i (Follett et al. 2016). A single C. liula individual of which we are aware has been collected from coffee berries from a farm on the Kona (west) side of Hawai'i island, presumably parasitizing the Leptophloeus sp. Depending on the rate of parasitism, it may have the potential to hinder control of H. hampei by reducing populations of Leptophloeus.

All the species from the C. hyalinipennis complex found in Hawai'i are considered to be adventive given that they have all so far been found exclusively attacking introduced beetles, in wood of introduced trees. Furthermore, despite the unreliability of COI phylogenetically, they are clearly not each other’s closest relatives (Fig. 2), indicating that they do not form a local endemic radiation. The sampled individuals of each Hawai'i species also showed little or no intraspecific variation in CO1, suggesting genetic bottleneck associated with the introduction of a small number of individuals.

An integrative approach combining molecular and morphological data clearly defined species boundaries and allowed morphological characters to be contextualized and analyzed among them, whereas there would have been uncertainty if using morphological data alone.

 4.2. Morphometric ratios, and implications regarding power to differentiate species

Morphometric ratios are often utilized to distinguish species in Bethylidae, as many taxa otherwise appear topologically similar. The large dataset in this study presents an opportunity to identify characters that may be either relatively constant or subject to variability among individuals of a species, for at least the C. hyalinipennis complex explored here: characters showing little intraspecific variation may tend to be more reliably utilized in distinguishing species that are otherwise morphologically close.

Ratios that show a combination of low intraspecific and high interspecific variability may provide an indication of how to create meaningful diagnoses for new species in light of the vast number of undescribed and unknown species in most bethylid genera. An indication of relative power of ratios may also be helpful when comparing an individual with a description based on few specimens, or when comparing individuals. Given that this study examined only a small species complex in the very diverse world of Bethylidae and Hymenoptera in general, the indications here should be taken as suggestions in a more general setting given the morphological diversity of these taxa, where characters likely show different stability in different groups. For example, the relatively flat heads of the C. hyalinipennis complex provided a convenient plane for obtaining consistent measurements, compared to the more convex heads of most Bethylinae and Epyrinae which are more difficult to consistently measure for OOL and VOL. The species utilized in the analyses here have also likely been subject to genetic bottleneck: C. soconuscensis came from a laboratory colony, while the Hawai'i taxa are likely nonnative and the result of relatively recent introductions of few individuals. Thus they may not accurately represent the entire range of variability present within the species.

 4.3. Future directions

Further investigation of the C. hyalinipennis complex may benefit from morphometric and molecular investigation to assess additional cryptic diversity and host-use patterns. The morphometric and molecular methods described here will be useful in this context to further untangle this species complex, and the degenerate Cephalonomia primers should be applicable across a range of Cephalonomia species, and possibly other Scleroderminae as well, although testing would be needed to verify how universal they are within and across the bethylid subfamilies. This would extend their use in assessing cryptic diversity beyond the species complex considered here.

Certainly the present study merely scratches the surface with respect to the diversity of Cephalonomia and the Cephalonomia hyalinipennis species complex. There appears to be much more diversity just within the UCFC collection from northern and central Florida than is resolved here, and thus both the C. hyalinipennis complex and Cephalonomia as a whole are in need of a broader revision. Investigation of additional collections from the field and/or existing museum collections using morphometic and molecular analyis may help clarify species boundaries and fill in some of the knowledge gaps brought to light here, e.g. the origin of Cephalonomia likely adventive in Hawai'i, and the diversity of species from the type location and their relationship to type specimens of C. hyalinipennis. Similarly, additional studies focusing on host-use patterns may identify whether species previously considered generalists are actually a complex of more host-specific cryptic species. The primers developed here could be used in a molecular diagnostic context to further evaluate trophic interactions between Cephalonomia species and their hosts (e.g. Gariepy et al. 2019; Lomeli-Flores et al. 2019) and between parasitoids that share the same hosts (e.g. Gariepy & Messing 2012; Slusher et al. 2024).

Further study of the ecology of the Hawai'i species may also be illuminating, e.g. whether C. liula has any effect on populations of Leptophloeus and its utility in H. hampei biocontrol, whether C. zander finds and utilizes endemic Ptinidae, or whether C. lynnea or possibly C. luiji utilize and may be beneficial against problematic Hypothenemus spp. Some of the methods here may also be applicable to the large Hawaiian endemic radiation of Sierola. We hope that this and other such studies will help inspire appreciation for the small creatures of the world, of which there is clearly increasingly great diversity the closer one looks.

 5. Acknowledgements

We acknowledge that insects that became subjects of this study were collected in the mokupuni (island) of O'ahu in the ahupua'a (land division) of Waikīkī in the moku (district) of Kona, Waimānalo in the moku of Ko'olaupoko, Kawaihāpai in the moku of Waialua, and Kamananui in the moku of Waialua (now Wahiawā), the ancestral and traditional land of Native Hawaiian people. We are very grateful to the Hawai'i Department of Agriculture and Biosecurity for providing funding and use of their lab space and insect collection. We thank the staff of the University of Hawai'i Insect Museum and Smithsonian Museum of Natural History for their support and assistance, and for providing us with specimen loans and workspace. We are also grateful to Hatch project HAW09041-H, administered by CTAHR, and USDA-ARS and APHIS for their funding.

 6. References Ashmead WH (1893) Monograph of the North American Proctotrypidae. Bulletin of the United States National Museum No. 45. https://doi.org/10.5479/si.03629236.45.1 Azevedo CO, Alencar IDCC, Ramos MS, Barbosa DN, Colombo WD, Vargas JMR, Lim J (2018) Global guide of the flat wasps (Hymenoptera, Bethylidae). Zootaxa 4489(1): 1–294. https://doi.org/10.11646/ZOOTAXA.4489.1.1 Batchelor TP, Hardy ICW, Barrera JF (2006) Interactions among bethylid parasitoid species attacking the coffee berry borer, Hypothenemus hampei (Coleoptera: Scolytidae). Biological Control 36: 106–118. https://doi.org/10.1016/j.biocontrol.2005.07.015 Batchelor TP, Hardy ICW, Barrera JF, Pérez-Lachaud G (2005) Insect gladiators II: competitive interactions within and between bethylid parasitoid species of the coffee berry borer, Hypothenemus hampei (Coleoptera: Scolytidae). Biological Control 33: 194–202. https://doi.org/10.1016/j.biocontrol.2005.02.010 Bridwell JC (1920) Some notes on Hawaiian and other Bethylidae (Hymenoptera) with the description of a new genus and species. 2nd paper. Proceedings of the Hawaiian Entomological Society 4(2): 291–314. http://hdl.handle.net/10125/15665 Chittenden FH (1893) Observations on some hymenopterous parasites of Coleoptera. Insect Life 5(4): 247–251. https://www.biodiversitylibrary.org/item/133722#page/5/mode/1up Evans HE (1978) The Bethylidae of America North of Mexico. Memoirs of the American Entomological Institute 27. Follett PA, Kawabata A, Nelson R, Asmus G, Burt J, Goschke K, Ewing C, Gaertner J, Brill E, Geib S (2016) Predation by flat bark beetles (Coleoptera: Silvanidae and Laemophloeidae) on coffee berry borer (Coleoptera: Curculionidae) in Hawaii coffee. Biological Control 101: 152–158. https://doi.org/10.1016/j.biocontrol.2016.07.002 Forbes AA, Bagley RK, Beer MA, Hippee AC, Widmayer HA (2018) Quantifying the unquantifiable: why Hymenoptera, not Coleoptera, is the most speciose animal order. BMC Ecology 18: 21. https://doi.org/10.1186/s12898-018-0176-x Gariepy TD, Bruin A, Konopka J, Scott-Dupree C, Fraser H, Bon MC, Talamas E (2019) A modified DNA barcode approach to define trophic interactions between native and exotic pentatomids and their parasitoids. Molecular Ecology 28(2): 456–470. https://doi.org/10.1111/mec.14868 Gariepy TD, Messing RH (2012) Development and use of molecular diagnostic tools to determine trophic links and interspecific interactions in aphid-parasitoid communities in Hawaii. Biological Control 60(1): 26–38. https://doi.org/10.1016/j.biocontrol.2011.06.011 Gordh G, Móczár L (1990) A Catalogue of the World Bethylidae. Memoirs of the American Entomological Institute 46. Hebert PD, Dewaard JR, Zakharov EV, Prosser SW, Sones JE, McKeown JT, Mantle B, La Salle J (2013) A DNA ‘Barcode Blitz’: Rapid digitization and sequencing of a natural history collection. PloS one 8(7): p.e68535. https://doi.org/10.1371/journal.pone.0068535 Honsberger DN (2024) Parasitoids of invasive Scolytinae in Hawai'i, their biology, and potential for biolgical control. PhD Thesis, University of Hawai'i at Mānoa, Honolulu, Hawai'i. https://hdl.handle.net/10125/108333 Honsberger DN, Honsberger M, Lorenzo-Elarco JH, Wright MG (2024) The genus Acerocephala and observations of the life history of Acerocephala hanuuanamu sp. nov. (Hymenoptera, Cerocephalidae) and its bark beetle host on the island of O'ahu, Hawai'i. Journal of Hymenoptera Research 97: 545–589. https://doi.org/10.3897/jhr.97.127702 Krombein KV (1979) Superfamily Bethyloidea. In: Krombein KV, Hurd PD Jr., Smith DR, Burks BD (Eds) Catalog of Hymenoptera in America north of Mexico, Vol. 2. Washington DC, Smithsonian Institution Press, 2209 pp. Lomeli-Flores JR, Rodríguez-Rodríguez SE, Rodríguez-Levya E, González-Hernández H, Gariepy TD, Talamas EJ (2019) Field studies and molecular forensics identify a new association: Idris elba Talamas, sp. nov. parasitizes the eggs of Bagrada hilaris (Burmeister). Journal of Hymenoptera Research 73: 125–141. https://doi.org/10.3897/jhr.73.38025 Madeira F, Madhusoodanan N, Lee J, Eusebi A, Niewielska A, Tivey ARN, Lopez R, Butcher S (2024) The EMBL-EBI Job Dispatcher sequence analysis tools framework in 2024. Nucleic Acids Research 52(W1): 521–525. https://doi.org/10.1093/nar/gkae241 Magnacca KN (2020) Review of Sierola Cameron (Hymenoptera: Bethylidae) of the Hawaiian Islands. Part 1: O'ahu. Bishop Museum Bulletin in Entomology 14. https://hbs.bishopmuseum.org/pubs-online/pdf/be14.pdf Mitchell A (2015) Collecting in collections: a PCR strategy and primer set for DNA barcoding of decades-old dried museum specimens. Molecular Ecology Resources 15(5): 1102–1111. https://doi.org/10.1111/1755-0998.12380 Nishida GM (2002) Hawaiian terrestrial arthropod checklist, 4th edn. Bishop Museum Technical Report 22: 1–313. https://hbs.bishopmuseum.org/pdf/tr22.pdf Orlovskytė S, Budrys E, Budrienė A, Radzevičiūtė R, Soon V (2016) Sibling species in the Chrysis ignita complex: molecular, mophological and trophic differentiation of Baltic species, with a description of two new cryptic species (Hymenoptera: Chrysididae). Systematic Entomology 41: 771–793. https://doi.org/10.1111/syen.12190 Pérez-Lachaud G (1998) A new bethylid attacking the coffee berry borer in Chiapas (Mexico) and some notes on its biology. Southwestern Entomologist 23(3): 287–288. https://www.sswento.org/_files/ugd/0646ec_3e4b0994a4f9494f855ae6e41243718e.pdf Pérez-Lachaud G, Batchelor TP, Hardy ICW (2004) Wasp eat wasp: facultative hyperparasitism and intra-guild predation by bethylid wasps. Biological Control 30: 149–155. https://doi.org/10.1016/j.biocontrol.2004.03.003 Pérez-Lachaud G, Hardy ICW (1999) Reproductive biology of Cephalonomia hyalinipennis (Hymenoptera: Bethylidae), a native parasitoid of the coffee berry borer, Hypothenemus hampei (Coleoptera: Scolytidae), in Chiapas, Mexico. Biological Control 14(3): 152–158. https://doi.org/10.1006/bcon.1998.0685 Pérez-Lachaud G, Hardy ICW (2001) Alternative hosts for bethylid parasitoids of the coffee berry borer (Hypothenemus hampei, Coleoptera: Scolytidae). Biological Control 22: 265–277. https://doi.org/10.1006/bcon.2001.0976 Pérez-Lachaud G, Hardy ICW, Lachaud J-P (2002) Insect gladiators: competitive interactions between three species of bethylid wasps attacking the coffee berry borer, Hypothenemus hampei (Coleoptera: Scolytidae). Biological Control 25: 231–238. https://doi.org/10.1016/S1049-9644(02)00103-2 Perkins RCL (1913) Introduction, being a review of the land-fauna of Hawaii. In: Fauna Hawaiiensis, Vol. 1. Cambridge, Cambridge University Press, 264 pp. Seehausen ML, Ris N, Driss L, Racca A, Girod P, Warot S, Borowiec N, Toševski I, Kenis M (2020) Evidence for a cryptic parasitoid species reveals its suitability as a biological control agent. Scientific Reports 10: 19096. https://doi.org/10.1038/s41598-020-76180-5 Sheikh SI, Ward AKG, Zhang YM, Davis CK, Zhang Linyi, Egan SP, Forbes AA (2022) Ormyrus labotus (Hymenoptera: Ormyridae): another generalist that should not be a generalist is not a generalist. Insect Systematics and Diversity 6(1): 8. https://doi.org/10.1093/isd/ixac001 Slusher EK, Cottrell T, Gariepy T, Acebes-Doria A, Querejeta Coma M, Toledo PF, Schmidt JM (2024) A molecular approach to unravel trophic interactions between parasitoids and hyperparasitoids associated with pecan aphids. Journal of Insect Science 24(4): 5. https://doi.org/10.1093/jisesa/ieae071 Smith MA, Woodley NE, Janzen DH, Hallwachs W, Hebert PDN (2006) DNA barcodes reveal cryptic host-specificity within the presumed polyphagous members of a genus of parasitoid flies (Diptera: Tachinidae). PNAS 103(10): 3657–3662. https://doi.org/10.1073/pnas.051131810 Smith MA, Wood DM, Janzen DH, Hallwachs W, Hebert PDN (2007) DNA barcodes affirm that 16 species of apparently generalist tropical parasitoid flies (Diptera, Tachinidae) are not all generalists. PNAS 104(12): 4967–4972. https://doi.org/10.1073/pnas.0700050104 Smith MA, Rodriguez JJ, Whitfield JB, Deans AR, Janzen DH, Hallwachs W, Hebert PDN (2008) Extreme diversity of tropical parasitoid wasps exposed by iterative integration of natural history, DNA barcoding, morphology, and collections. PNAS 105(34): 12359–12364. https://doi.org/10.1073/pnas.0805319105 Strutzenberger P, Brehm G, Fiedler K (2012) DNA barcode sequencing from old type specimens as a tool in taxonomy: a case study in the diverse genus Eois (Lepidoptera: Geometridae). PLoS One 7(11): e49710. https://doi.org/10.1371/journal.pone.0049710 Swofford DL (2021) Phylogenetic Analysis Using Parsimony (*and Other Methods). Version 4, build 169. Available at https://paup.phylosolutions.com. Valbuena R, Packalén P, Martín-Fernández S, Maltamo M (2012) Diversity and equitability ordering profiles applied to study forest structure. Forest Ecology and Management 276: 185–195. https://doi.org/10.1016/j.foreco.2012.03.036 Vega FE, Infante F, Johnson AJ (2015) The genus Hypothenemus, with emphasis on H. hampei, the coffee berry borer. In: Vega FE, Hofstetter RW (Eds) Bark Beetles: Biology and Ecology of Native and Invasive Species. London, Academic Press, pp. 427–494. https://doi.org/10.1016/B978-0-12-417156-5.00011-3 Supplementary materials 10.3897/asp.84.e174388.suppl1 58981D3E-D05D-5426-AB95-11FF2A266704

Tables S1, S2

: .zip

Table SS1. Within-BIN variation for Cephalonomia spp. nr. hyalinipennis from Hawai'i, Chiapas, and Florida [.xlsx file]. — Table SS2. Percent identity matrices for sequenced Cephalonomia spp. nr. hyalinipennis from Hawai'i, Chiapas, and Florida [.xslx file].

 https://binary.pensoft.net/file/1715990 This dataset is made available under the Open Database License (http://opendatacommons.org/licenses/odbl/1.0). The Open Database License (ODbL) is a license agreement intended to allow users to freely share, modify, and use this dataset while maintaining this same freedom for others, provided that the original source and author(s) are credited. Honsberger D, Pérez-Lachaud G, Magnacca KN, Lorenzo-Elarco JH, Kelly SL, Wright MG, Gariepy TD (2026) 10.3897/asp.84.e174388.suppl2 D205716E-79F7-5A80-B746-57315430A7F7

File S1 (Video)

: .mp4

: Cephalonomia lynnea behavior in observation chambers. Vimeo link: https://vimeo.com/994679495

 https://binary.pensoft.net/file/1715991 Copyright © 2024 David Honsberger. Licensed under CC BY 4.0. Honsberger D, Pérez-Lachaud G, Magnacca KN, Lorenzo-Elarco JH, Kelly SL, Wright MG, Gariepy TD (2026)