The specimen was examined and photographed by Mathias Vielsäcker at SMNK using a Canon EOS R5 with either a Canon EF 100MM F/2.8L MACRO IS USM for the dorsal, ventral and lateral images, or a Laowa 25 mm f/2,8 Ultra Macro 2,5–5X for the images of smaller details. Stacked imaging was performed using a WeMacro rail. Images were then stacked using Helicon Focus.

Measurements were made using ImageJ 1.53t software (Rueden et al. 2017). Taxonomy follows the OSF (Cigliano et al. 2024). Morphological terminology and measurements follow Tumbrinck (2014), with some additional measurements. The following measurements were taken: body length (BL), pronotum length (PL), pronotum width (shoulders) (PWS), pronotum width (maximal) (PWM), pronotum height (maximal) (PHM), vertex width (VW), compound eye width (CEW), antennal groove width (AGW), scutellum width (SW), fore femur length (FFL), fore femur width (FFW), mid femur length (MFL), mid femur width (MFW), hind femur length (HFL), hind femur width (HFW), tegmen length (TL), tegmen width (TW). Nomenclature is in accordance with the rules of the International Code of Zoological Nomenclature (ICZN 1999).

The specimen was scanned in six height steps using a filtered polychromatic beam at the high-throughput tomography station at the Imaging Cluster of KIT Light Source. We employed an air-bearing rotary stage (RT150S, LAB Motion Systems) for sample rotation and a fast indirect detector system consisting of a 200 µm LuAG scintillator, a double objective white beam microscope (Optique Peter; Douissard et al. 2012) and a 12-bit pco.dimax high-speed camera (Excelitas PCO GmbH) with 2016 × 2016 pixels of 11 µm physical size. We employed a magnification of 2×, resulting in an effective pixel size of 6.11 µm. For each scan, we took 200 dark field images, 100 flat field images, and 3000 equiangularly spaced radiographic projections in a range of 180° with an exposure time of 30 ms. The control system concert (Vogelgesang et al. 2016) served for automated data acquisition and online reconstruction of tomographic slices for data quality assurance. The 3D tomographic reconstruction was performed by tofu (Faragó et al. 2017) and additionally included ring removal and 8-bit conversion.

The individual tomographic volumes were registered and merged in Amira 2022.2, which was also employed for pre-segmentation. Pre-segmented labels served as input for semi-automatic segmentation with Biomedisa (Lösel et al. 2020). The Biomedisa result was imported back into Amira and minor errors were corrected. The needle was digitally removed and the holes filled. The final label field was converted into a polygon mesh and exported as OBJ file. CINEMA 4D R20 was employed for smoothing the surface and for polygon reduction. The polygon-reduced model was imported into MeshLab, converted into U3D format and integrated into an interactive 3D PDF document (Suppl. material 1).

The new species is at first glance quite similar to Phelene maroon. The latter was until recently known only from a single male specimen (Cadena-Castañeda et al. 2021), but a recent examination of collections from French Guiana revealed a number of new individuals, of which many were female (Itrac-Bruneau et al. 2023). This, coupled with our own P. turgida specimens, allowed us to thoroughly compare our only female specimen with P. maroon females and find some key differences, most notably in the shape of the subgenital plate. Furthermore, P. reinschmidti sp. nov. was collected in Ecuador, while P. maroon is known only from French Guiana. Interestingly, P. turgida is known from both Peru and French Guiana, and is sometimes syntopical with P. maroon. The specimens of P. turgida reported by Itrac-Bruneau et al. (2023) differ from the specimen from Marcapata, Peru by the shape of the pronotal crest and may represent a separate species. However, our specimens collected in central Peru (Fig. 1) exhibit the same amount of variability in this trait as between the French Guiana and Marcapata specimens, indicating either that this trait can vary somewhat or that the distribution of Phelene species is very complex. Unfortunately, photographs of the P. turgida types are not available. It is still possible to describe tetrigid species based on clear morphological characters, but DNA studies are becoming increasingly necessary, especially to address interesting biogeographical observations.

Until now, the question of the position of Lophotettiginae in the Tetrigidae phylogeny remained completely unaddressed. The brief synonymization of Phelene with Chiriquia (Cadena-Castañeda and Cardona 2015) was never elaborated despite being interesting and potentially premonitory. These two genera are clearly not synonymous, but their similarities cannot simply be explained as convergence (Cadena-Castañeda et al. 2021). Phelene and Chiriquia are alike in their facial morphologies, lengths of first tarsal segments, the placement of pronotal crests, and the sideways projected lateral lobes, which could point to some degree of relatedness. The genus Otumba Morse, 1900 seems to be related to Chiriquia and could provide further hints upon its urgently needed revision.

Further, we have to consider the genus Metrodora Bolívar, 1887. Some of its species, e.g., M. lutosa Bolívar, 1887, M. gibbinota (Bruner, 1910), and M. panamae (Hebard, 1924) also exhibit the earlier listed similarities with Phelene, while M. gibbinota and M. panamae also have white-tipped antennae. This genus and by extension its tribe, Metrodorini Bolívar, 1887, is composed of likely related but ill-defined taxa, which is hindering further research (Kasalo et al. 2023). The relationship between Metrodorinae and Lophotettiginae should be closely examined and it may even find the latter to be synonymous with the former.

Lastly, all these genera share the mentioned characters with many members of the predominantly African tribe Xerophyllini Günther, 1979, currently attributed to the subfamily Cladonotinae. Therefore, the possibility that Lophotettiginae, Metrodorinae, and Xerophyllini form a monophyletic group should be seriously considered and examined from both morphological and molecular angles. The morphology akin to Morphopus (Bolívar, 1887) (Xerophyllini) could be basal in this group, which could explain all the different derived morphologies within it, as well as its distribution across the southern half of Gondwana. It should be noted that we do not consider Asian Metrodorinae as true members of that subfamily and that all of the mentioned higher taxa require extensive revisions.

We here provide the first 3D data for any Tetrigidae, though more are in preparation. Such 3D data may become an important resource in future taxonomic works, especially in concert with artificial intelligence and machine learning approaches. One of the large advantages of including 3D data in taxonomic descriptions is that virtual specimens will be readily available around the world, without the danger of sending valuable and irreplaceable holotypes and risking them getting lost or damaged. In addition, such virtual specimens are at least a minor backup in case a type gets lost. The 3D models are scaled to size and allow easy and accurate measurement of distances and also have the benefit of allowing for 3D geometric morphometric measurements. Modern high-throughput X-ray imaging methods already offer the possibility to digitize large numbers of insects in a short time (van de Kamp et al. 2018; Rühr et al. 2021). In the future, artificial intelligence will help to further automate data acquisition as well as large-scale data analysis (Heethoff and van de Kamp 2023). Even statistical species delimitation may become a viable approach potentially speeding up the process of taxonomy, an important improvement in a time of heavy diversity decline.