Two new Paraparatrechina (Hymenoptera, Formicidae) species from the Seychelles, with notes on the hypogaeic weissi species-group

Abstract Recent survey work in the Seychelles has revealed two new species of Paraparatrechina that are here described: P. illusio sp. n. and P. luminella sp. n. A revised key to the workers of Paraparatrechina for the Afrotropical and Malagasy regions is provided. The taxonomy of the hypogaeic weissi species-group is also reviewed in light of recent field collections. The species P. sordida is revived from synonymy and given new status (as a full species) and a discussion of the morphologically peculiar species-group is provided. With the description of the two species and the removal of another species from weissi synonomy there are now 16 Paraparatrechina species known from the Afrotropical and Malagasy regions.


Introduction
The biology of the ant genus Paraparatrechina remains poorly known, but what is becoming clearer is that species diversity within the genus is certainly much higher than is currently recognized. LaPolla et al. (2010b) recently revised the Afrotropical and Malagasy species and found 8 new species (they found 13 total species within the two regions). While the Australasian species await taxonomic revision, preliminary data suggest that there are many undescribed species (S. Shattuck, pers. comm.). To emphasize this point of the genus having much higher species richness than is currently recognized, recent survey work in the Seychelles by BLF revealed two new species that were not included in LaPolla et al. (2010b). Here we describe those two species.
We also provide notes on a rather unusual group of Paraparatrechina, the Afrotropical weissi species-group (LaPolla 2004a;LaPolla et al. 2010a;LaPolla et al. 2010b). This group was last reviewed by LaPolla (2004a) and thought to contain two valid species (P. bufona and P. weissi). Recent collection work however in Uganda suggests that a third species, P. sordida, which is currently in synonymy with P. weissi, should be elevated to full species. Here we also discuss the taxonomic status of the weissi speciesgroup and provide images for all three species.

Materials and methods
Specimens examined for this study are deposited in the following institutions: CASC California Academy of Sciences, San Francisco, CA, USA MCZC Museum of Comparative Zoology, Cambridge, MA, USA USNM National Museum of Natural History, Washington, DC, USA All measurements were taken at 80× power with a Leica M125 microscope using an orthogonal pair of micrometers, recorded to the nearest 0.001 mm, and rounded to two decimal places for presentation. When more than one specimen was measured, minimum and maximum measurements and indices are presented. All measurements are given in millimeters. Digital color images were created using a Leica DFC425 digital camera. Leica Application Suite software (ver. 3.8) was used for images. Each imaged specimen is uniquely identified with a specimen-level unique identifier (e.g. CASENT0003099).
Morphological terminology for measurements and indices employed throughout are defined (following LaPolla et al. 2011a, b)    gaster, with pubescence appearing in neat rows with a silky appearance. Head subquadrate with nearly straight posterior margin; scapes surpass posterior margin by first 2-3 funicular segments; three ocelli apparent. Mesosoma compact with steeply rising pronotum in lateral view; metanotal area indistinct, only slightly impressed; propodeum with short, flat dorsal face and much longer, steep declivitous face.
Queen. Measurements (n=3)  Head brown, with bulging large eyes that occupy most of the lateral region of the head; head slightly broader than long. Palps distinctly lighter than head in color. A dense layer of pubescence covers head, with scattered erect setae along mid-region, posterior margin and clypeus. Scapes surpass posterior margin by about length of the first 2 funicular segments; antennae 13-segmented. Mandible with apical tooth and an indistinct basal angle. Mesosoma same color as head; pronotum short and collarlike; mesonotum large, rounded anteriorly, overarching pronotum; mesosoma dorsum flat, with erect setae. Gaster slightly ligher brown than head and mesosoma, covered with pubescence and erect setae. Parameres relatively broad then with a steep angle towards last third of paramere length; last third of paramere thinner and elongated with rounded apex; paramere with scattered erect setae.
Notes. This species falls into the small, yellow Paraparatrechina worker phenotype range (typically workers of these species vary by only slight difference in setation and color tones of yellow and brownish-yellow) and therefore identification of the species can be difficult. It is morphological similar to three Aftrotropical species: P. gnoma, P. oreias, and P. subtilis. In practice, a relatively straight forward, nonmorphological way, to separate P. illusio is it is only known from the Seychelles, while the remaining three species are from West and Central Africa. Morphologically, it differs from P. gnoma in being slightly larger overall and in coloration (P. gnoma is brownish-yellow with lighter yellow patches). From P. oreias the main difference also lies in color. The gaster of P. oreias is brownish-yellow, contrasting slightly with the remainder of the body. The metanotal area of P. oreias is also more distinctly defined than is seen in P. illusio. From P. subtilis, the difference is in the pubescence. Whereas the pubescence on the head of P. subtilis is decumbent throughout, on P. illusio it is only decumbent lateroanteriorly around eyes. Additionally the gastral pubescence is different between the two: in P. subtilis it is longer and slightly decumbent, contrasting with the shorter, tightly appressed pubescence observed in P. illusio.  Overall brown with patches of yellow to white; lighter area medially between eyes and above torulae; distinct patches of lighter areas (ranging from yellow to white) on pronotum (that typically extends onto mesonotum and occasionally onto dorsal face of propodeum) and gastral tergites (typically from posterior of T1 through anterior portion of T4; scapes proximally more brown becoming yellow to whitish midlength, lightening to white through apex of funiculus; procoxae golden yellow, meso/metacoxae and trochanters white; femur golden yellow then remainder of leg light yellow to white; body covered in dense, appressed pubescence; macrosetae placement as is typical in Paraparatrechina. Head ovate with nearly straight posterior margin; scapes surpass posterior margin by first 2-3 funicular segments; three ocelli apparent. Mesosoma compact with steeply rising pronotum in lateral view; mesonotum and metanotal area short; metanotal area indistinct, only slightly impressed; propodeum with short, flat dorsal face and much longer, steep declivitous face.

Paraparatrechina luminella
Queen. Measurements ( As in worker, with modifications expected for queen caste and the following differences: 1. Coloration overall more brown than in worker, with no distinct yellow to white patches on mesosoma or gaster; lighter antennae and legs.
3. Legs generally as in worker, except coxae brown.
Notes. The coloration pattern seen in the worker of Paraparatrechina luminella is very distinct and is unlike any pattern seen in other species from either the Afrotropical or Malagasy regions. The most similar coloration pattern among Paraparatrechina is seen in P. albipes, in which workers typically have a light patch of whitish coloration on the posterior pronotum and mesonotum that contrasts with the generally overall dark brown color of the body. Whether this is indicative of a close relationship between these two species or is simply convergence is unclear.
(2010a) found based on molecular evidence from 5 genes that these morphologically peculiar species in fact belonged in Paraparatrechina. In retrospect, there was some, albeit at the time seemingly rather weak morphological support for the placement within Paraparatrechina such as: the short, angular dorsal propodeal face (Figs 16,19,22), and although obscured in P. bufona by the presence of several erect macrosetae on the mesosoma, the typical Paraparatrechina mesosomal macrosetae pattern of 2:1:1 (pronotum, mesonotum and propodeum) is present on all species. Nonetheless, superficially the weissi species-group does resemble Pseudolasius. What this certainly reflects is that the weissi species-group species have become hypogaeic and have convergently taken on the suite of morphological characters common among subterranean formicines (for example see LaPolla 2004b). Of further interest is that like Pseudolasius at least two species of the weissi species-group have evolved majors (P. bufona and P. weissi). All Pseudolasius (of which most, if not all species are hypogaeic) presumably have majors (LaPolla et al. 2010a). There appears to be selection occurring in the Prenolepis genus-group among those with a hypogaeic lifestyle for the evolution of majors. For instance, the ground-dwelling and presumably largely hypogaeic (based on the morphology of the workers) Nylanderia amblyops known from Madagascar appears to be the only species in that genus to have evolved majors. In Euprenolepis at least one species, E. procera, has majors as well, although it is not hypogaeic (rather it appears to be nocturnal) (LaPolla 2009). LaPolla (2004a) reviewed what would later be called the weissi species-group, and recognized only two valid species (P. bufona and P. weissi), the others of which were considered synonyms (all of P. weissi). The study was hindered in two ways: the lack of recent material from nest series and the confounding variable of worker polymorphism. Recent fieldwork (J. Longino, pers. comm.) however resulted in a small collection of specimens of two species of the weissi species-group in sympatry with each other: P. weissi and another species that was clearly not P. bufona. Upon comparison with type material, we determined that the specimens clearly belonged to the species named P. weissi sordida (here treated as a full species), previously synonymized under P. weissi. Paraparatrechina sordida differs from P. weissi and P. bufona in several ways. Paraparatrechina bufona is distinct because of the presence of many erect macrosetae across the scapes and mesosoma. The other two species P. sordida and P. weissi are more similar to each other but differ in that P. sordida possesses a strongly impressed metanotal groove, a head without paired macrosetae medially from the posterior margin towards the cly peus and has no macrosetae on the posterior margin of the head. We therefore propose a revived and new status of P. sordida as a full species. The synopsis of the weissi species-group provides a reinterpretation of the valid species and where the synonyms should properly be placed. These findings nicely demonstrate the continued need for collection in the very poorly sampled Afrotropical region.