﻿Ghatippuspaschima, a new species and genus of plexippine jumping spider from the Western Ghats of India (Salticidae, Plexippini, Plexippina)

﻿Abstract We propose a new genus of plexippine jumping spiders from the Western Ghats of India based on the new species Ghatippuspaschimagen. et sp. nov. While it bears a superficial resemblance to Pancorius in body form and Hyllus in membrane-bearing embolus, our UCE phylogenomic data—the first to resolve broad relationships within the Plexippina—as well as morphological features justify its status as a new genus. In addition to the molecular data and morphological descriptions, we provide photographs of living specimens of Ghatippuspaschimagen. et sp. nov. and information on their natural history.


Introduction
The Western Ghats of India, one of the hottest hotspots of biodiversity, awaits more than chance-based reporting of salticid spider diversity.Systematic surveys may reveal previously undiscovered salticids critical to understanding the region's ecosystems and the broader context of salticid diversity and phylogeny.Our 2019 surveys in a private estate in Kodagu, Karnataka, for instance, uncovered one such salticid lineage, of the subtribe Plexippina.Here, we describe that new species and propose a new genus for it based on phylogenomic evidence and morphology.
The subtribe Plexippina (Salticinae, Plexippini), an Old World group except for two New World species of Evarcha Simon, 1902, is species-rich, containing over 500 described species currently placed in 37 genera worldwide (Maddison 2015;Metzner 2023;World Spider Catalog 2023).Their combination of high diversity, conservative body forms, and simple genitalia have hindered the discovery of synapomorphies that could delimit genera, making the group taxonomically challenging.Placing new species in genera without evidence explicitly stated and interpreted phylogenetically has led to decisions about generic divisions (e.g.Prószyński's 2018 splitting of Evarcha) that are weakly supported and sometimes not broadly accepted (Kropf et al. 2019;World Spider Catalog 2023).Despite the taxonomic mess within the subtribe, what species are included in the Plexippina has remained more or less stable based on a combination of morphological (Maddison 1996(Maddison , 2015) ) and molecular data (Maddison and Hedin 2003;Maddison et al. 2008;Bodner and Maddison 2012).
The first steps to our modern concept of Plexippina were taken by Maddison (1996), based on the form of the male endite's serrula and the palp.Molecular data subsequently showed that some of the genera he included (e.g.Sibianor Logunov, 2001) are instead harmochirines (Maddison and Hedin 2003;Maddison et al. 2008;Bodner and Maddison 2012), leading to Maddison's (2015) refined concept of the Plexippina.Using these studies as context, we here examine phylogenomically the relationships of the newly discovered plexippine lineage from the Western Ghats.Its placement would be unclear by morphology alone, as it is morphologically similar to Hyllus C.L. Koch, 1846 in male genitalia and Pancorius Simon, 1902 in its body form.
In the course of this work, we provide the first-ever plexippine phylogenomic tree, based on ultraconserved element data (Faircloth 2017;Zhang et al. 2023), contributing to further understanding of the relationship among plexippine genera and salticids in general (see Maddison et al. 2020a, b).

Materials examined
The Indian specimens examined in this study are deposited in the Biodiversity Lab Research Collections of the National Centre for Biological Sciences (NCBS), Bengaluru, India (http://biodiversitycollections.in/).Individual specimens are identified by three-digit voucher codes prefixed with "IBC-BP" and "IBC-BX"; in addition, some are also identified by code numbers starting "AS19.".Non-Indian specimens are deposited in the University of British Columbia Spencer Entomological Collection.Codes beginning with "WPM#19-" indicate a collecting event of location and date, and thus may apply to more than one specimen.

Morphology
A drawing tube attached to a Nikon ME600L compound microscope was used to prepare illustrations.Clove oil was used for clear viewing of epigyna after digesting the internal epigynal soft tissues with pancreatin.Preserved specimens were photographed using an Olympus OM-D E-M10 II mounted on an Olympus SZX12 stereoscope (for bodies) and a Nikon D7000 mounted on a Nikon ME600L compound microscope (for copulatory organs).Photographs were stacked using Helicon Focus 8.2.1 Pro.Living specimens were photographed with an Olympus OM-D E-M10 II camera with a 60 mm macro lens.
Descriptions are based on ethanol-preserved specimens.The descriptions were written with primary reference to the focal specimen indicated, which was used for measurements and carefully checked for details, but they apply as far as known to the other specimens examined.Carapace length was measured from the anterior base of the median eyes to the posterior margin of the carapace.The abdomen was measured from its anterior edge to the posterior end of the anal tubercle.All the measurements are in millimetres.Leg measurements are represented as follows: total length (femur, patella, tibia, metatarsus, and tarsus).Abbreviations used here are as follows: ALE, anterior lateral eye; AME, anterior median eye; PME, posterior median eye; PLE, posterior lateral eye; RTA, retrolateral tibial apophysis.

Taxon sampling for phylogenomics
The set of 18 species (15 ingroup and 3 outgroup species) used in the phylogenomic analysis, and with their taxonomic authority indicated, is listed in Table 1.
The selection of ingroup taxa was determined based on the limits of Plexippina, informed by previous phylogenetic studies (Maddison et al. 2008;Bodner and Maddison 2012) and synthesis work by Maddison (2015).The taxon sampling strategy aimed to maximize the representation of plexippine genera and their morphological diversity, including those most similar and relatively least similar to the focal species of this work.The two genera viewed as morphologically most similar to the new species, and thus candidate genera to contain it, are Hyllus and Pancorius.Thus, two distinct species of each of those were included to give them the best chance of linking to the new species.Otherwise, 11 other plexippine genera representing diverse body forms were included, for a total of 15 ingroup taxa representing 13 genera.These 13 ingroup genera represent ~86% of the plexippine genera known from India.The selection of outgroup taxa, two harmochirines and one salticine, was based on previous salticid phylogenetic studies (Maddison et al. 2008(Maddison et al. , 2014(Maddison et al. , 2017;;Bodner and Maddison 2012;Maddison 2015).

Ultraconserved element (UCE) data
Molecular data was gathered for UCE loci using target enrichment sequencing methods (Faircloth 2017).One to four legs were used for DNA extraction using the Qiagen DNeasy Blood and Tissue Kit following the manufacturer protocol.

Phylogenetic analysis
Maximum-likelihood phylogenetic and bootstrap analyses were performed with IQ-TREE v. 1.6.12 (Nguyen et al. 2015) using the Zephyr v. 3.1 package (Maddison and Maddison 2020) in Mesquite v. 3.61 (Maddison and Maddison 2019) on the concatenated, unpartitioned UCE dataset with 15 ingroup and three outgroup taxa.For the phylogenetic tree inference, the option -m TEST (standard model selection followed by tree inference, edge-linked partition model, no partition-specific rates) was used with 10 search replicates.For the bootstrap analysis, the same option as the tree inference was used with 1000 search replicates.

Phylogenetic results
Table 2 lists the sequence data recovered from the taxa.On average 2844 UCE loci per taxa (minimum 2255, maximum 3092) were initially recovered.Of these total loci, on average 2807 loci survived per taxa (min.2225, max.3054) after removing suspected paralogous loci based on branch lengths, and on average 2722 loci remained per taxa (min.2205, max.2956) after removing loci represented in fewer than 10 taxa.In total, 3060 UCE loci were represented in the resulting dataset, which were concatenated into the final matrix, in which each taxon had on average ~2.2 million base pairs of sequence data (min.965482, max.2414600).
The phylogenetic results are shown in Fig. 1.The subtribes Plexippina and Harmochirina are recovered as reciprocally monophyletic, consistent with the previous phylogenetic studies with much less sequence data (Maddison and Hedin 2003;Maddison et al. 2008;Bodner and Maddison 2012).Within the Plexippina, two major clades are recognized (marked in Fig. 1).Bootstrap val-Table 2. Specifics of molecular data used for this phylogenomic analysis.Molecular data was generated based on RTA_ v2 probeset."SRA" is Sequence Read Archive accession number available through NCBI; "Reads pass QC" is the number of reads after the removal of adapter-contamination and low-quality bases using Illumiprocessor; "Total UCE loci" is the total number of UCE loci recovered with RTA_v2 probeset; "After paralogy filter" is the number of UCE loci after deletion of suspected paralogous loci based on branch length ratios; "In at least 10 taxa" is the number of UCE loci in at least 10 or more taxa after branch length criteria; "Filtered UCE sequence length" is the concatenated sequence length of filtered UCE loci; "Total loci" is the number of UCE loci represented among all taxa.ues are generally high, showing that the relationships are well supported, as might be expected with this volume of sequence data.Ghatippus gen.nov. is recovered as sister to all the genera in clade 1 (see Fig. 1): (Ghatippus, (Plexippus, (Ptocasius, (Anarrhotus, Artabrus)))).This phylogenetic position of Ghatippus gen.nov.necessitates its recognition as a new genus.Any other taxonomic decision apart from creating a new genus, whether to include it in a phylogenetically closely related genus or in another morphologically similar plexippine genus, would render the genus in which it is placed either paraphyletic or polyphyletic.The only other phylogenetically meaningful option, besides creating a new genus, would be to lump all the genera in clade 1 into a single genus.This would generate a massive genus of highly diverse body forms that would go against all traditions of salticid generic limits.A far better choice is to recognize Ghatippus gen.nov.as a new genus.

Species
The choice to establish a new genus is further substantiated by morphology.Within clade 1, Ghatippus gen.nov. is unique with its membrane bearing medium-long embolus.In contrast, Anarrhotus Simon, 1902 and Plexippus C.L. Koch, 1846 have a short embolus, while Artabrus Simon, 1902 and Ptocasius Simon, 1885 have a medium to long, thin embolus.Importantly, all four of these lack a membrane-bearing embolus.Etymology.The generic name Ghatippus gen.nov.combines the word 'Ghat', representing the collecting locality-the Western Ghats Mountain range-with the distinctive suffix found in several plexippine genera.The generic name is assigned to the masculine gender.
From Hyllus, Ghatippus gen.nov.differs in carapace (higher, box-shaped, PLEs on tubercles in Ghatippus gen.nov.vs relatively lower, rounder, no tubercles in Hyllus), RTA (simple, short vs serrated, wide), cymbium (laterally narrow with a narrow apex vs robust, laterally wide with a broader apex), and copulatory ducts (short vs long).From Thyene, Ghatippus gen.nov.differs in embolus length (medium in Ghatippus gen.nov.vs long and coiled in Thyene), copulatory ducts (short vs long), and carapace (higher, box-shaped, PLEs on tubercles vs relatively lower, rounder, no tubercles).From Vailimia, Ghatippus gen.nov.differs in embolus length (medium in Ghatippus gen.nov.vs long in Vailimia), RTA (simple, short vs curvy, long), and spermathecae (simple vs globular).Ghatippus gen.nov.also has an oval abdomen and open posture typical for salticids, unlike Vailimia's pointed abdomen and unusual stance, holding the legs close to the body in a compact crouch.
Ghatippus gen.nov. is most likely to be confused with Pancorius because of the high, box-shaped carapace with PLEs on tubercles, but Pancorius lacks the membrane-bearing embolus and has distinct epigynal coupling pockets.Distinct black bulge behind each ALE.Black around PMEs and PLEs.Thorax with steep slope, yellowish brown, sparsely covered with black hairs.With origin near front, brown band encircles carapace close to transition between ocular area and thorax.Brown along edges.Clypeus narrow, brown, covered with white hairs but more sparsely than in male.Chelicerae yellowish brown.Vertical, narrower than extent of carapace, not bulging as in male, with simple unbifurcated fangs (Figs 7,13).Legs mostly yellowish and some brown near joints.Abdomen ovoid, dark brown but with paler basal band (extended posteriorly to encir-cle the abdomen), muscle attachment points, and posterior medial chevron.On either of the chevron the brown is especially dark, almost black, and contains distinct pale spot (Fig. 24).Epigyne (Figs 4,5,20,21): two crescent-shaped anterior copulatory openings share common atrium.No epigynal coupling pocket visible, though there is slight medial indentation of the epigastric furrow.Simple round spermathecae with flattened (lamellar) copulatory ducts ventrally.Fertilizations ducts broad, placed anteriorly on spermathecae.

Ghatippus paschima
Additional materials.All from India: Kerala: near Thalappuzha, Fringe Ford, and deposited in Biodiversity Lab Research Collections, NCBS.Natural history.Ghatippus paschima sp.nov.was found commonly in both Kodagu and Kerala.Most collecting days in both locations were rainy and overcast.The spiders seemed to be exclusively vegetation dwellers, often found on small to medium-sized trees.Although they were collected from diverse habitats, they were mostly collected in the understorey, edge, and disturbed habitats of the evergreen forests of Honey Valley Estate in Kodagu.In Fringe Ford, Kerala, they were collected from the secondary evergreen growth of an inoperative tea estate.
While male and female salticids typically differ in colour, sexual dimorphism in the fangs is noteworthy.Male fangs are bifid, but female fangs are not (Figs 6,7,12,13).The bifid fangs may possibly be used to hold females during mating, in male-to-male combat, or have a sex-limited ecological function.
While we are beginning to see a steady uptick in the number of new plexippines being described (World Spider Catalog 2023), the unique endemic lineages and their radiations in India are still largely unexplored.With the addition of Ghatippus paschima sp.nov., potentially an endemic lineage, the number of plexippines stands at 46 species and 17 genera for India.

Figure 1 .
Figure 1.Maximum-likelihood tree, best tree of 10 replicates inferred using IQ-TREE, from concatenated dataset of 3060 ultraconserved element loci.Numbers at the nodes are percentage of 1000 bootstrap replicates recovering the clade.Ghatippus paschima sp.nov. is recovered distantly (see Clade 1) from morphologically similar Hyllus and Pancorius (see Clade 2).

Table 1 .
Specimens used in phylogenomic analysis.