130 years from discovery to description: micro‐CT scanning applied to construct the integrative taxonomy of a forgotten moth from Southern Africa (Lepidoptera: Geometridae)

X‐ray micro‐computed tomography (micro‐CT) of dried and pinned museum specimens combined with advanced image processing can provide a useful, novel and non‐destructive tool for integrative insect taxonomy. This paper demonstrates how micro‐CT can be applied to provide unambiguous illustrations of diagnostic morphological characters for new taxa description and to understand how micro‐CT imaging may complement other imaging techniques. Following micro‐CT scanning, a semi‐automatic segmentation and volume rendering protocol was used to portray the wing venation and diagnostic structures and ornamentation of male genitalia from multiple angles. Using micro‐CT images, we provide the description of a conspicuous geometrid moth from southern Africa (Lepidoptera: Geometridae), which has been present in collections since 1894, but left without an available name. Using a multigenetic dataset comprising 273 terminal taxa from the superfamily Geometroidea, we constructed a molecular phylogeny to place our study species to an isolated lineage in Geometridae: Larentiinae, tribe Xanthorhoini sensu lato. We describe it as Chloecolora vergetaria new genus, new species Englund & Staude, and provide diverse ecological information on its distribution, habitat, host plant, adult and immature stages, and parasites. We found micro‐CT imaging particularly useful in two‐ and three‐dimensional imaging of wings, providing detailed information for instance on non‐tubular folds that may be difficult to distinguish using other techniques.


INTRODUCTION
Despite micro-computed tomography (micro-CT) imaging of dried insect specimens is non-destructive, authentic and minimally invasive, it has been sparsely used in descriptive insect systematics.In this paper, we focus on applying micro-CT imaging to yield diagnostic images in descriptive insect taxonomy in a non-destructive way using rare dried and pinned museum specimens.We scan air-dried museum specimens and render the micro-CT scans into two-and threedimensional images using a novel protocol comprising several software available for standard personal computers.
The necessity and urgency to describe biodiversity have become more acute as the biodiversity on earth is declining at an accelerating pace mainly due to anthropogenic influence (Hortal et al., 2015;Nori et al., 2023).While the decline of large and charismatic vertebrates has been well documented and widely recognised, even the megadiverse insect fauna may be in a state of general decline (e.g., Wagner, 2020).Besides being able to identify and name taxa, we need to understand the role of organisms in their ecosystems to facilitate the meaningful discussion and conduct of proper conservation measures to decelerate, halt or preferably reverse this decline.
Integrative taxonomy is a paradigm developed to delimit the units of diversity of life from multiple and complementary perspectives, such as biogeography, molecular phylogeny, comparative morphology, population genetics, ecology, development and behaviour (Dayrat, 2005).Morphology has remained indispensable for analysing insect diagnostic structures (e.g., Simonsen & Kitching, 2014, and references therein) even as new technologies have become available, especially but not only micro-CT followed by advanced image rendering protocols (Wipfler et al., 2016).
A micro-CT scan is created by shooting an x-ray beam through the target object to a detector from multiple angles recording the loss of radiation intensity typically proportional to the density variation among different structures inside the target (Simonsen & Kitching, 2014).The outcome of the micro-CT scanning is a slice image stack data file that can be rendered to two-(2D) or threedimensional (3D) images from any desired part, transect or angle of the target object using appropriate software.Micro-CT scanning followed by image rendering protocols enables the scrutiny of selected surface or inner structures of small target objects, such as insect specimens, without breaking or removing skin or other organs that obstruct the penetration of light to the target structures.
Since the first introduction to studies of insect diagnostic morphology (Hörnschemeyer et al., 2002), micro-CT imaging has been applied in various studies to illustrate diagnostic surface structures, such as head (Luo et al., 2021), mandibles (Beutel et al., 2021), thorax, legs, and the entire torso (Hita-Garcia et al., 2019) as well as internal structures such as brain (Rother et al., 2021) and pheromone glands (Ostwald et al., 2022), musculature (Richter et al., 2021), reproductive organs (Simonsen & Kitching, 2014), larval gut (Winfelder et al., 2023) and wing venation (Robinson et al., 2018).Mainly due to the long exposure times, micro-CT imaging in insect morphology is confined to dead specimens or sessile phases, such as pupae or eggs.
Micro-CT imaging has also extensively been used in studies of insect anatomy and physiology (Lowe et al., 2013;Martín-Vega et al., 2021), in forensic entomology (Richards et al., 2012) to time pupal metamorphosis, and to study the functional reproductive morphology (Jandausch et al., 2023;Zlatkov et al., 2023).For imaging soft tissues with synchrotron (e.g., Vommaro et al., 2023) or standard micro-CT, the use of fresh specimens and special preparation methods, such as critical point drying, metal coating and resin embedding (Mensa et al., 2022) or iodine staining (Gignac et al., 2016) are preferable.These sample preparation methods are, however, to a varying degree invasive and not routinely applicable to dried and pinned or mummified specimens.Even rather conservative dried sample manipulation such as rehydration, relaxing and remounting or repetitive freezing and thawing may degrade the relevant genomic DNA beyond accessibility using standard sequencing protocols, while micro-CT scanning is not known to have such effects (Hall et al., 2015).Living insect pupae are at least moderately tolerant to x-ray radiation (Lowe et al., 2013), but the use of MRI (Laussmann et al., 2022) instead of micro-CT imaging should be considered if the risk of adverse effects of ionising radiation to living specimens needs to be minimised, and a somewhat lower image resolution is acceptable.
The increasing hardship of acquiring necessary new sample collection permits together with the reluctance to sacrifice extant rare museum specimens may render micro-CT imaging the sole remaining option to study in detail the morphology of rare or undescribed species, which also are the most likely to be endangered with extinction (Liu et al., 2022).
The large and distinctive moth family Geometridae currently comprises approximately 24,000 described species worldwide, with the highest species diversity recorded at low latitudes (Rajaei et al., 2022).
While there are about 1000 described geometrid moth species in Europe, specimens of more than 1500 morphologically distinguishable species are present in extant collections originating from the Republic of South Africa, hereafter referred to as South Africa, and at least 240 of these remain undescribed with little data on their ecology or phylogeny (Krüger, 2020;Müller et al., 2019;Rajaei et al., 2022).For this study, we chose a large, conspicuous, putatively undescribed geometrid moth species from South Africa as the model taxon.We

Materials
Six pinned, dry and set specimens, three males and three females, a DNA sample (HSS-18391) and another reared male specimen with the pupal skin from the HSS research collection were initially examined in detail for this study.After sequencing the standard barcode region (658 bp fragment of the mitochondrial cytochrome c oxidase subunit I 'COI' gene (Hebert et al., 2003)) from this material, we discovered a virtually identical, unpublished sequence on BOLD database under curation by Axel Hausmann.This specimen was collected in 2006 from South Africa and identified as belonging to the species Xanthorhoë chloëphora Prout.As we failed to find Xanthorhoë chloëphora in the scientific literature, following our request, the curator of Lepidoptera of the State Museum of Natural History Stuttgart Hossein Rajaei conducted a search at NHMUK, where most of the Prout's type material is deposited.He found three old specimens belonging to the same putative species as our study specimens in a unit box labelled 'Xanthorhoë chloëphora Prout m.s.', with a (holo)type label attached to the pin of a female specimen (Figure 1).
Louis Beethoven Prout (1864Prout ( -1943) ) was a British lepidopterologist and musicologist who specialised in geometrid moths.He was active particularly in the first few decades of the 20th century and became known for his numerous descriptions of mainly geometrid species.However, he never published a description of Xanthorhoë chloëphora, and hence, this name remains a manuscript name and is not available under the International Code of Zoological Nomenclature (ICZN, 1999).The Prout's (holo)type-labelled specimen bears labels indicating that the specimen originated from the collection of another British entomologist, William Lucas Distant , whose insect collection comprising more than 50,000 specimens was purchased by NHMUK in 1920.According to another label on the pin, the specimen was collected from 'Transvaal Pretoria' in January of 1894, making it the oldest known specimen at 130 years of age.
The historical province of Transvaal no longer exists, but the eastern part of it is now the Mpumalanga province, where our study specimens were collected.'Pretoria' possibly refers to the place where Distant was resident, as suitable habitat and host plants do not occur anywhere near Pretoria.
After confirming the identity of Xanthorhoë chloëphora Prout m.s., label data from a total of 51 specimens belonging to this taxon were gathered from the following museum and private collections:

Morphology and life history
The six adult specimens examined were collected using custom-made battery-operated light traps in Mount Sheba Nature Reserve, Mpumalanga, South Africa.The larvae were collected at the same site by knocking them down from branches of Sclerochiton harveyanus Nees and reared to adult in the lab, feeding them with the leaves of the host plant in hard plastic containers.All the specimens were collected under relevant collection permits held by Hermann Staude.
The diagnostic images of wing colouration and external body structures were constructed from multifocal images shot with a Canon EOS 5D SRL camera mounted on a Cognisys StackShot automated macro rail using Canon EF 100 mm f/2.8 Macro lens.Established preparation protocols (e.g., Hardwick, 1950;Robinson, 1976;Sihvonen, 2001) were followed to dissect and slide mount the abdomen of one male and one female specimen.For comparison, several specimens of other species of the Larentiinae suspected to be the closest relatives of the target species were also dissected.Some structures were photographed during the dissection in situ, and eventually, all structures were mounted on slides in Euparal.To study the wing venation using traditional dissection, we detached the right-hand-side pair of wings from a male specimen and submerged the wings in a commercial chlorine bleach solution 'Klorin' containing 36 g of sodium hypochlorite per litre for 5 minutes.Next, we removed the bleach solution from the preparate by submerging the wings in 99% ethanol for 5 minutes and finally slide-mounted the transparent, nonstained wings in Euparal for photography.
The images from slide mounts were produced applying stacking technique to multifocal images shot through a Leica DM1000 LED stereo microscope with an attached Leica MC170 HD camera.Each Two dried and pinned specimens, a male (holotype, Figure 13f) and a female (paratype, Figure 13g), were used to make micro-CT scans with a Nikon XT H 225 Micro CT scanner at Zoological Museum of Helsinki (ZMH).The entire female specimen was scanned to obtain a scan from the wing venation, but for the male specimen, only the tip of the abdomen was scanned to obtain a high-resolution scan of the genitalia and to avoid artefacts and noise resulting from the metallic insect pin.A schematic flow chart of the micro-CT imaging process is shown in Figure 2.
Scans were performed using a multi-metal target with a molybdenum setting, with 74 kV beam energy, 94 μA beam current, 500 ms exposure time for 9998 projections for the male and 4476 projections for the female, with an average of eight frames per projection for both specimens.Detector binning was set to 1 Â 1 and gain to 24 dB, and the imaging was conducted using limited dynamic range.Voxel size was 5.44 μm for the male and 16.74 μm for the female.The scanning process took 11.5 h for the male and 5.5 h for the female specimen.
The CT scans were processed using Nikon CT Pro3D version XT 6.9.1 software, and the resulting 3D models were exported to VGSTUDIO 3.5.2(Volume Graphics GmbH, Heidelberg, Germany) in 16-bit.The 3D models were then aligned and visually examined to identify regions of interest.The 3D model of the abdomen of the male specimen was exported from VGSTUDIO as a 16-bit TIFF slice image stack to 3D Slicer 5.0.3 (Fedorov et al., 2012) for segmentation.The parts of the genitalia were manually demarcated on every 20th slice image, and the treated slice image stack was exported from 3D Slicer to Biomedisa (Lösel et al., 2020), an open-source online platform for segmenting.After processing the slice image stack file on Biomedisa, the resulting segmented file was uploaded back to 3D Slicer for masking to remove any remaining non-demarcated areas from the slice image stack.The masked stack was exported as multipage tiff file from 3D Slicer, and individual tiff slice images were extracted in FIJI 1.53t (Schindelin et al., 2012).The resulting 3D model was imported back to VGSTUDIO for visual inspection to evaluate the need for further refinement.Based on the visual inspection, the original segmentation at every 20th slice image was refined to every 10th image, and the most complex areas were segmented at every second image.A few small, remaining problematic areas were segmented manually.
The same procedure was used to further segment a version of the 3D model showing only the valvae and selected other structures of the male genitalia.The 2D images were enhanced with Adobe Photoshop (v 10.0).
We also screened six specimens for infection with the common endosymbiotic bacterium Wolbachia by PCR amplification of two Wolbachia genes, ftsZ and gatB (Baldo et al., 2006).

Systematic position of the study species and phylogenetic analyses
Initial systematic position of the study species on subfamily level within Geometridae was examined by comparing the morphological structures against diagnostic characters (summarised in Murillo-Ramos, Friedrich, et al., 2021).
For our phylogenetic analysis, the molecular data extracted from three of these specimens were included in the dataset of a total of 273 terminal taxa.The remaining 272 terminal taxa, with up to 11 genes per species, were obtained from published data by Murillo-Ramos (Murillo-Ramos, Chazot, et al., 2021).All subfamilies of the family Geometridae were represented in the dataset, and two species from the family Uraniidae, the closest sister clade to Geometridae (Mitter et al., 2017;Rajaei et al., 2015) belonging to the same superfamily (Geometroidea), were set as an outgroup.
To construct the phylogenetic trees, we applied the maximum likelihood approach as implemented in IQ-TREE web server (Trifinopoulos et al., 2016).Best-fitting substitution models were selected by ModelFinder (Kalyaanamoorthy et al., 2017) with '-m MFP + MERGE' option.The phylogenetic analyses were carried out with '-spp' option (edge proportional), which allowed each partition to have its own evolutionary rate.We evaluated the node supports with ultrafast bootstrap approximations (UFBoot2) and SH-like approximate likelihood ratio test (Guindon et al., 2010;Hoang et al., 2018) using '-B 1000 -alrt 1000' option.To minimise the risk of overestimating branch supports in ultrafast bootstrap approximation analysis, we used the '-bnni' option, which optimised each bootstrap tree using a hill-climbing nearest-neighbour-interchange search.The resulting trees were visualised and rooted in FigTree v1.4.2 (Rambaut, 2015) and modified using CorelDRAW (24.4.0).

Micro-CT images and morphology
We consider the micro-CT imaging particularly suitable for examining insect wing venation.Earlier, the method has been shown useful to study venation even from unset, dried specimens (Robinson et al., 2018).The wing venation was clearly visible in the image (Figure 3a) reconstructed from a micro-CT scan of 4476 projections from a set, pinned and dried female paratype (Figure 13g) specimen.
Folds and curvature of the wings can be observed from a 3D image (Figure S1), facilitating a clear distinction between actual wing veins and other structures, such as non-tubular folds appearing similar and difficult to separate in a 2D image.
The diagnostic, chitinised structures of the abdomen, especially those of the male genitalia can be scrutinised from any desired point of view in the 2D and 3D images (Figures 4, 5 and S2) reconstructed from a micro-CT scan of the 9998 projections from a set, pinned and dried male holotype (Figure 13f) specimen.For example, the uneverted vesica inside the aedeagus, the natural position of the uncus, and the juxta can be seen from the micro-CT image (Figure 4).organ or structure.For example, the tubular shape of the valva is easily observable from a 3D (Figure S3) image or the 2D images with artificial light and reflection from multiple angles (Figure 5).
The general habitus of the adult specimens (Figures 6 and 13) does not bear close resemblance to any described species within the subfamily Larentiinae that we are aware of.The persistent and conspicuous green coloration is still recognisable after more than a century of museum preservation (Figure 1a,c).
The male bipectinated antenna has extremely long, setose branches.Both male and female antenna has a strong terminal spine at the apex of each branch that is not present in most species of Larentiinae (Figure 7a,d).
The male coremata (Figure 8a,b) are minuscule, vestigial and probably defunct as they do not bear specialised scales as do the prominent coremata typical to several species of the tribe Xanthorhoini sensu lato (Viidalepp, 2011).
The male genitalia (Figures 4, 5 and 14) resemble to a certain degree those of Palearctic Scotopteryx Hübner species; both typically have similar labides of transtilla and rather rudimentary uncus of the male genitalia (Hausmann & Viidalepp, 2012).The female genitalia (Figure 14) are compact and confined to the few distal segments of the abdomen bearing few diagnostic characteristics.
The joint between the two dorsal terminal segments of the pupal skin is decorated with a crumpled ridge (Figure 9c) exclusive and diagnostic at least to the European species of the genus Scotopteryx (Patočka & Turčáni, 2005).
Based on the constructed molecular phylogeny (Figure 12) and the branch length therein, genetic distance to the nearest voucher specimens, likely host plant isolation, distinct forewing coloration, the structures of antennae in both sexes, male coremata and genitalia in both sexes, we conclude the study specimens to belong to yet undescribed species and undescribed genus within the tribus Xanthorhoini sensu lato in the subfamily Larentiinae (Lepidoptera: Geometridae).

Life history
The new species has been found in mid-to-high elevation (700-1 700 m) moist Afromontane forests along the Great Escarpment in northern and eastern parts of South Africa, and in moist southern coastal forests, which matches the distribution of the host plant (Figure 10a).
The adult specimens exhibit nocturnal behaviour and positive phototaxis, and are confined to the inner parts of the forest.There are 19 species, all with an Ethiopian region distribution (Vollesen, 1991).
The species of the subgenus Sclerochiton are endemic to montane and submontane forests (700-2100 m) of south-eastern Africa, with S. harveyanus having the widest distribution (Figure 10).

Molecular work and phylogeny
The species shows no evidence of Wolbachia infection since none of the six specimens (Figure 13) screened tested positive for either of the two Wolbachia-specific genes.

Morphology supports the classification of the study species into
Larentiinae, as it shows the following diagnostic characters of this subfamily: the forewing ornamentation pattern consists of several parallel undulating transverse lines (Figure 13), the hindwing veins Sc + R and Rs are fused for a considerable length, two areolas are present in the forewings, and the forewing radial veins Rs1-Rs4 have a joint stalk arising from the distal areole (Figure 3), the ansa (Figure 8c)  The closest interspecific DNA barcode distance to reference sequences in our dataset was 9.86%-10.29% to EO0202 Euphyia unangulata Hawort and 10.42%-10.81% to EO0519 Euphyia intermediata Guenée (Table S5).The K2P genetic distance in COI between our study species and Scotopteryx species ranged from 12.89% to 19.8%.Pupa: Fusiform, dark brown, smooth; crumpled dorsal ridge between two last segments, cremaster elongated, with 3 min pairs of setae, apical pair large and basally swollen (Figures 9 and 11g).

Systematics
Based on molecular data and morphology, C. vergetaria is classified in Geometridae: Larentiinae, Xanthorhoini sensu lato (Figure 12), as sister to Scotopterygini.The classification of Xanthorhoini sensu lato lineage needs further research.

Etymology
Vergeet (Afrikaans) to forget; refers to the so far incomplete and forgotten efforts to describe this species.

DISCUSSION
With this study, we experimented and demonstrated how micro-CT imaging can be used in integrative lepidopteran taxonomy to produce diagnostic morphological images from rare dried and pinned museum specimens without destroying or damaging them.Additionally, we established the taxonomic identity of a previously undescribed moth species from South Africa and described it.We used multiple imaging methods to study the morphology of adult specimens, compiled the known distribution, phenology and life history traits of the target species from label data of several museum specimens, as well as behavioural observations from adult and immature stages of the species in its natural habitat.
While molecular methods are of great value in most cases, morphology remains the most widely used, inexpensive and quick approach to identify, study, and delimit species and other taxa.We show that using micro-CT, it is indeed possible to obtain diagnostic images from internal structures of rare museum specimens without dissecting, destroying or damaging the target specimens.This should make virtual loans of, for example, unique type specimens more accessible for morphological studies at lower cost and risk levels and reduce the acquisition costs of new material.We were able to produce two-dimensional images (Figures 3-5) of the wing venation and male genitalia and three-dimensional footage (Figures S1-S3) of the target specimens without dissecting them and consider these images in many ways more authentic and superior to those we were or could have been able to obtain by using traditional methods, such as photographing glass slide mounts or drawing images.We decided not to produce high-resolution micro-CT images from female genitalia, since the weak general sclerotization of female genitalia makes the demarcation of dehydrated target structures ambiguous compromising the authenticity and quality of resulting images.
In our experience, the micro-CT imagery of air-dried specimens is best applicable to flat or well chitinised, originally hard structures not prone to deformation because of dehydration.For imaging soft inner structures, MRI imaging, synchrotron micro-CT (e.g., Vommaro et al., 2023) or standard micro-CT imaging following special sample preparation methods, such as critical point drying, metal coating, resin embedding (Mensa et al., 2022) or iodine staining (Gignac et al., 2016) are preferable.While these sample preparation methods enhance the quality micro-CT scans, they may be rather expensive and compromise the integrity of the samples.Whenever density differences between neighbouring tissues are small, image demarcation and rendering from micro-CT scans may be inaccurate and subject to artefacts (e.g., Souza et al., 2023).
The proliferation of micro-CT in descriptive morphological entomology using air-dried specimens has not been particularly rapid despite the necessary technology has been available for some two decades.The acquisition costs of suitable scanning devices and the long manual and computational image processing times, which require highly skilled human expertise probably repel even those potential users, who can afford to use more invasive imaging methods.We, for example, spent some 25 h labour and processing time to complete the segmentation of the male genitalia alone.To better employ the potential of this non-destructive and authentic imaging technique and the bioinformation embedded in the extant insect collections, we could suggest the following three measures for consideration.
First, the micro-CT scans should be made publicly accessible for reuse and development of rendering protocols.There already are recommendations for standards and best practice for 3D digital data publication, storage, management and accessibility (Davies et al., 2017;Miralles et al., 2020).The cost of long-term data storage is in a general downward trend, and there should be quite a few repositories, who could assume this task (Pampel et al., 2023)  Even though the absorption properties of the relevant tissues to staining agents and x-rays may be quite different, well-chosen colouration would make it easier to compare the new images to the ones in extant literature.
Third, the rendering software and protocols should further be developed particularly to shorten the arduous and still largely manual segmentation of target structures.Rather promising attempts have lately been made to cut the manual segmentation time from several hours to a few minutes using a deep learning-based fully automated segmentation of micro-CT images from ant (Toulkeridou et al., 2023) and bee (Lösel et al., 2023) brains.To take the automation a step further, an entire automatic rendering process from the scan to a standardised set of images should be available at least for certain diagnostic structures mentioned above.
Overall, we believe the current state technology is not the principal obstacle to the proliferation the micro-CT imaging (Hipsley & Sherratt, 2019).The systematics community may need some more time to adopt this technique into the routinely available toolbox.
There is a considerable potential to further develop micro-CT image processing and interpretation protocols for dried and often unique museum specimens, and we recommend this technique to be preferred over destructive methods, whenever the necessary technology is available at an acceptable cost.Furthermore, we anticipate the preservation of extant rare museum specimens for future generations to be increasingly often prioritised above the current needs to study biodiversity using invasive methods.
Our molecular phylogenetic tree is well in accordance with our comparative morphological analysis for the new species and its closest relatives and suggests that our study species represents a sister lineage to Scotopterygini in the subfamily Larentiinae.Based on the molecular phylogeny, the branch length therein, barcode distance to other described taxa and the morphological characters unlike any related species or genera, we consider the description of a new species and genus to be justified.Viidalepp revised the tribes of Larentiinae (Viidalepp, 2011) and considered Scotopterygini as a valid tribe, but later synonymised it with Xanthorhoini (Hausmann & Viidalepp, 2012).In a more recent taxonomic work on Larentiinae (Õunap et al., 2016) explored and evaluated a comprehensive array of molecular and imaging techniques, including micro-CT to establish the accurate taxonomic position of the taxon and to provide a balanced species description compliant with the International Code of Zoological Nomenclature (ICZN, 1999).The systematic part, where we describe a new genus and species, Chloecolora vergetaria gen.n., sp.n.Englund & Staude, is based on museum specimens, but the life history part of the study is based on our own field and rearing observations.

•
DMNHDitsong Museum of Natural History, Pretoria, South Africa (33) • HSSPrivate research collection of Hermann Staude, Magaliesburg, South Africa (14) • NHMUKNatural History Museum, London, UK (3) • ZSMZoologische Staatssammlung München, Munich, Germany (1).The life-history information, photographs of the habitat, host plant, larvae, and parasitoid specimens reared from the larvae of the study species were gathered on-site in the Mount Sheba Private Nature Reserve, near Pilgrim's Rest, South Africa.
Type material of 'Xanthorhoë chloëphora Prout m.s.' in NHMUK.(a) The type specimen; (b) the labels of the same specimen; (c) the unit box containing the three specimens in NHMUK (Photos: Hossein Rajaei).
final image was created by stacking 5-200 exposures of different focal planes using Zerene Stacker v.1.04software.The photographic image of the wing venation is a mosaic combining five partially overlapping stacked pictures.Scanning electron microscope images were taken using a Quanta FEG 250 device at the Institute of Biotechnology Electron Microscopy Unit in Viikki, Helsinki to show structural details of the antennae.
Furthermore, any obstructing structures can be virtually removed from a micro-CT image to reveal a clear view to any desired internal U R E 2 The micro-CT imaging process.
Wing venation, the right pair of wings.(a) Female, ventral view, micro-CT 2D image; (b) male, dorsal view, slide mount preparate, photographed through microscope, light from underneath, slide preparate ME 24.observations of adults from all months of the year, except for the winter months of July and August.Given the absence of both negative observations and observation activity during the winter months, the presence of larvae at the same time as the adults were in flight, and the lack of diapause in rearing, we suspect the species to be multivoltine with consecutive and overlapping generations.Several half-grown larvae measuring 12-20 mm in length (Figure 11c,d) were observed between 10 and 12 February 2021 at Mount Sheba Private Nature Reserve on the evergreen shrub S. harveyanus (Acanthaceae) growing in the forest understory.According to a revision by Vollesen, the genus Sclerochiton Harvey comprises The larvae reached a length of 25-28 mm (Figure11e,f) and pupated (Figure11g) around 3 weeks after being captured.Five adults emerged by 26 March 2021 after a pupal stage of some 2 weeks.Additionally, parasitoid specimens belonging to Tachinidae (Diptera) and Microgastrinae (Hymenoptera: Braconidae) emerged from the larvae and pupae of the new moth species (Figure11h,i).Our observations of the larval host plant (S.harveyanus) and the matching distributional data suggest that the study species is possibly monophagous on S. harveyanus.
Micro-CT images of male genitalia of Chloecolora vergetaria from different viewpoints.(a) The entire tip of male abdomen virtually cut along the medio vertical plane with selected structures pointed out.The genitalia apparatus seen from different angles: (b) ventral, (c) ventrolateral, (d) lateral and (e) dorsal.The phylogenetic hypothesis based on the seven-gene dataset also placed our study species in Larentiinae, providing strong support (UBF and SH-aLRT = 100) for the genetic isolation of the three samples of our study species.The distinct lineage of our study species belongs to the Xanthorhoini sensu lato, being the sister group of the Scotopterygini Warren (Figures 12 and S4), which together form a sister group of the remaining Xanthorhoini.
Micro-CT images of male left valva of Chloecolora vergetaria cut off from other structures along the medio vertical plane.(a) Selected structures pointed out.The valva seen from (b) ventral; (c) medio-ventral; (d) dorsal and (e) lateral angles.Etymology Chloe (Greek) young, green foliage or shoots of plants in spring, yellowish green colour in general; refers to the persistent, conspicuous coloration of the type species; color (Latin) tint, colour.Chloecolora vergetaria Englund & Staude sp.n. http://zoobank.org/urn:lsid:zoobank.org:act:76B43F5B-16CA-4236-A0AE-2A7BA6769526Type material Holotype, ♂ (Figure 13f), SOUTH AFRICA: Mpumalanga, Pilgrimsrest, Mount Sheba Private Nature Reserve, Afromontane forest, 1780 m, 10 February 2021, S24.4938 E30.7083, Hermann Staude leg.(DMNH).Paratypes (3♂♂ and 6♀♀), ♂ (Figure 13b) same data as holotype (ZMH); ♂ (Figure 13j) SOUTH AFRICA: Northern Transvaal, Mariepskop, wet eastern slope forest, 1600 m, 8 November 1996, 24 o 39 0 S 30 o 51 0 E, Hermann Staude leg.(HSS); ♂ (Figure 1b top) SOUTH AFRICA: The Haven, Transkei, 6/11/81, N. J. Duke (NHMUK); ♀ (Figure 13c) same data as holotype (HSS); ♀ (Figure 13g) same data as holotype (HSS); ♀ (Figure 13k) same data as holotype (HSS); ♀ (Figure 1a,b bottom) SOUTH AFRICA: Transvaal, Pretoria, I. 1894, Distant Coll.1911-383, NHMUK 014173789 (NHMUK); ♀ (Figure 1b middle) SOUTH AFRICA: Marieps Mnt., IV. 1932, G. van Son (NHMUK); ♀ SOUTH AFRICA: [Mpumalanga] Peilgrimsrus, Crystal Springs, 1800 m, 9 April 2006, À24.9217, 30.6853,Hermann Staude leg., BC ZSM Lep 42468 (ZSM).Description Adults.Head (Figure 6b,c): Frons and vertex covered with light green scales, protruding light green and brown scales towards thorax, chaetosemata covered with light brown hair-like scales, compound eye naked.Labial palpi length roughly twice eye diameter, basal and medial segment scales green, laterally scattered with dark brown scales, frontal tips distinctive light ochreous.Proboscis well developed.Male antennae (Figures 6b and 7a,b) bipectinate, proximal flagellomeres and rami dorsally covered with light green scales, distal flagellomeres dorsally covered with light brown scales, ramus length up to 10 times the width of flagellum, setous, setae attached to ventrolateral edges of ramus, single straight, stiff spine protruding from the tip of each ramus.Female antennae unipectinate (Figure 7d), ramus length equal to diameter of flagellomere, scale coloration and seta arrangement analogous to that of male antenna, the spine in each ramus tip even more conspicuous than in male antenna.Thorax (Figure 7a): Dorsally covered with green scales, ventral parts with light brown hair-like and flat scales.Spur pattern on legs 0-2-4.Abdomen (Figures 7a and 13): Segments dorsally green, with two medial dark angular spots of varying clarity separated by narrow medial whitish dot forming light dorsomedial truncated line, frontal and caudal borders of each segment light brown.Ventral side of segments light brown with sharp lateral border to dorsal colouration.Male seventh segment reduced (Figure 8a,b), with vestigial bilateral, weakly sclerotised coremata arising from membrane between seventh F I G U R E 6 Body part details.Stacked macro photographic images of the body.(a) The entire torso, lateral view; (b) frons dorsal view and (c) head lateral view.and eighth segments.Tympanal organs on fused first and second segments, rather shallow, ansa short with hammer-shaped head.Wings (Figures 3 and 13): Forewings triangular, termen rather evenly concave.Wingspan 30-35 mm, females slightly larger than males.Male forewing ground colour yellowish green with multiple serrated blackish antemedial and postmedial lines.Female forewing ground colour more bluish green than that of the male.Postmedial line pronounced, with sharp darkened outward angle between M3 and CuA1, neighboured with suffuse hazel brown blotch around subterminal line, more pronounced on females.Wave line blackish, serrated.Terminal line dark, truncated, composed of narrow, opposed triangular spots.Fringes green, blackish at vein tips.Discal spot small, elongated, dark.Especially female medial and cubital veins covered with light scales on terminal area.Hardly notable individual variation.Hindwings brownish grey with 3-4 rounded dark lines, ground colour lighter between lines.Rather large tornal area echoing forewing green colouration and augmented lines.Termen weakly undulating, terminal line and fringes like those of the forewings.Discal spot small, obscure, somewhat more pronounced on underside and on females.The hind-and forewings attached together by a frenulo-retinacular wing coupling system.Underside yellowish grey-brownish, postmedial line distinct.Terminal area frequently darker than rest of the wing, suffused with brown or grey scales, most pronounced on forewings.Genitalia Male (Figures 4, 5 and 14): Uncus compact, triangular with short, pointed apex.Tegumen massive, margins sclerotised.Dorsal margin F I G U R E 7 Scanning electron microscope images of detailed structures of the antennae.(a) Medial flagellomeres of the male antenna, dorsal view, magnification 300Â; (b) male rami scales and setae details, inverted lateral view, magnification 778Â; (c) the stem of a pair of male rami branching from a flagellomere ventral view, magnification 1473Â; (d) the female antenna, arrangement of setae and the single strong spine protruding from the tip of each ramus, dorso-lateral view, magnification 724Â. of valva with large, dorsally protruding lobe, cucullus evenly arching, compact, slightly convex, covered with dark, medially orientated setae.Sacculus very large, roughly twice the size of cucullus.Juxta elongated, bowl-shaped, dorsally expanding.At each end of transtilla a sclerotised labid with long setae extending towards costa from distal tip.Aedeagus narrowing towards caecum, rostrum with two arching sclerotised caudal border lobes.Everted vesica showing two consecutive asymmetrical lobes without cornuti, second larger lobe with cone shaped terminal extension.Base of vesica (Figure 14c,e) covered with tiny sclerotised grains (microcornuti).

F
I G U R E 1 0 Distribution and life history.(a) Observations of Chloecolora vergetaria based on museum specimen labels and Sclerochiton harveyanus after Vollesen (Vollesen, 1991) in the south-eastern part of Africa; (b) habitat of C. vergetaria in Mount Sheba, Mpumalanga, South Africa; (c) Sclerochiton harveyanus at the same location.Map retrieved from https://www.simplemappr.net.green or greyish brown; ornamentation suffuse and variable; last segment with arching dorsal ridge (Figure 11c-f).
Immature stages and parasitoids.(a, b, c) The spherical eggs with a smooth surface pattern; (c-f) the variable ground coloration of the caterpillar exhibiting looping locomotion pattern in the absence of abdominal prolegs, typical to geometrid moths; (g) the pupa shown from a ventral view; (h, i) parasitoids emerged from the larvae and pupae, belonging to Tachinidae (Diptera) (h) and Braconidae (Hymenoptera) (i).

F
I G U R E 1 2 The phylogenetic tree based on the seven-gene dataset, with a concatenated length of 4980 bp comprising a total of 273 terminal taxa from the superfamily Geometroidea.Of these, 195 taxa are from the subfamily Larentiinae, whereas 41 taxa form the tribal lineage, we call Xanthorhoini sensu lato.The branch and cluster including the three samples from our study species are enclosed within the red rectangle.The full tree of Larentiinae is provided in FigureS4.publishing platforms could include 3D volume data in their data availability policies.Second, a set of standards for imaging important diagnostic structures, such as genitalia and wing venation including meta data, scanner settings, resolution, view angles and artificial colouration should be established.Artificial colouration based on x-ray absorption differences mimicking the outcome of staining the same dissected structures with commonly used staining agents should be experimented.
, the authors present the phylogeny of Larentiinae, where Xanthorhoini appears polyphyletic and forms together with Cataclysmini and Euphyini a sister lineage to Scotopterygini.C. vergetaria belongs to the Xanthorhoini sensu lato lineage, which merits yet another revision and we therefore leave the exact tribal assignation of our study species open at this point.

F
I G U R E 1 3 Type specimens of Chloecolora vergetaria.(b, f, j) Males; (c, g, k) females; (a, e, i) under side of the left pair of male wings; (d, h, j) under side of the right pair of female wings; (f) the holotype, all other specimens paratypes.F I G U R E 1 4 Genitalia.Male; (a) ventral view without aedeagus; (b) the aedeagus; (c) everted vesica with the tip of aedeagus caecum removed; (d) right labid of transtilla; (e) apex of aedeagus with arching sclerotised caudal border lobes, paratype (Figure 13b), slide preparate PS 2854.Female; (f) ventral view, the ductus seminalis joining ductus bursae in the circle, paratype (Figure 13k), slide preparate PS2855.