The genus Homoeomma comprises thirteen species distributed in Argentina, Brazil, Colombia, Peru and Uruguay (World Spider Catalog c2018). Homoeomma uruguayense (Mello-Leitão, 1946) is known from Buenos Aires, Entre Ríos and Santa Fe provinces, central eastern Argentina (Gerschman and Schiapelli 1972) and Uruguay (Costa and Pérez-Miles 2002). In a recent survey to Córdoba province, central Argentina, we recorded individuals of Homoeomma uruguayense. The location of this population is far away from the known geographic distribution of the species, thus we used a species distribution model to predict the potential distribution under different scenarios.
The specimens were deposited in the Arachnological collection of the Laboratorio de Zoología de Invertebrados II, Universidad Nacional del Sur, Bahía Blanca, Buenos Aires, Argentina. Photographs were taken with a digital camera attached to a Leica S8APO stereoscopic microscope and processed using the Leica Application Suite version 4.6.0. The following abbreviations are used: DA = digitiform apophysis, PI = prolateral inferior keel, PS = prolateral superior keel. We carried out eight encounters between four females and two males under laboratory conditions. All interactions took place in cages measuring 40 x 30 and 25 cm high with soil as substrate. For species distribution models we used 59 presence localities. Bioclimatic variables (19) were obtained from the WorldClim 2.0 (www.worldclim.org), at a resolution of 30 arc sec (Fick and Hijmans 2017). From those, using Pearson's correlation as implemented in ENMTools ver. 1.4.4, we selected a subsample. The choice of a variable from a correlated (r > 0.75) pair was done in a preliminary run of the model with all the variables, retaining those with the best percent contribution and permutation importance in MaxEnt. The following subset of bioclimatic variables was selected: BIO1 = Annual Mean Temperature, BIO3 = Isothermality, BIO4 = Temperature Seasonality, BIO14 = Precipitation of Driest Month and BIO15 = Precipitation Seasonality. Past climatic conditions comprised 19 bioclimatic variables for the Last Inter-Glacial event on about 120 000 and 140 000 years ago obtained from WorldClim 1.4 (www.worldclim.org) (Otto-Bliesner et al. 2006). We predicted the potential distribution of Homoeomma uruguayense under different climate scenarios using MaxEnt 3.3.3k (Phillips et al. 2006). This algorithm searches for the maximum entropy density using Robust Bayes Estimation and requires only presence points as input data (Elith et al. 2011). MaxEnt estimates the relation between species presence and environmental variables in a geographic space and draws a model of environmental suitability for the occurrence of a given organism.
Homoeomma uruguayense (Mello-Leitão, 1946)
Diagnosis. Homoeomma uruguayense differs from all other species of the genus by the serrated base of prolaterial inferior keel (PI) (Figs. 2c-d) and metatarsus I curved at middle (Fig. 1g). Females of H. uruguayense are characterized by a spermathecae with membranous base and semicircular receptacles as high as wide (Fig. 3e). For further taxonomic information see Gerschman and Schiapelli (1972) and Pérez-Miles et al. (1996).
Distribution. This species is distributed in Argentina and Uruguay (Fig. 5). New record from Córdoba province (Fig. 5), at the locality of Villa Berna.
Natural history. Females of this species were found inhabiting short burrows under small stones (Figs. 4a-b).
Burrows usually presented more than one entrance (Figs. 4c-d). The median length of burrows was 11.52 cm ± 3.63 SD (n = 9). One female captured constructed an egg sac on October 30, 2015 and 40 juveniles emerged after 42 days.
Sexual behavior. We facilitate nine sexual encounters using three males and three females during August 2017 according to the mating period reported for this species (Costa and Pérez-Miles 2002). All males courted before contacting the females. The male courtship was characterized by body vibrations caused by leg III and palpal drumming as an alternating movement of palps against the substrate. We recorded three mating events. None female accepted a second copulation. All females responded to male courtships by making leg tapping with the first and second pairs of legs against the substrate before contacting the male. The mean duration of copulation was 38.5 s ± 4.94 SD with a range of one to three palpal insertions from males. The courtship and mating behavior observed was similar to that reported by Costa and Pérez-Miles (2002) with the exception that the female response was observed before individuals make contact.
Material examined. Argentina, Córdoba: five females, Valle de Calamuchita, Villa Berna, 31°54'35.60"S - 64°44'21.35"W, 1308 meters, 9 Sept 2015, N. Ferretti, S. Copperi, G. Pompozzi; three males, Valle de Calamuchita, Villa Berna, 31°54'35.60"S - 64°44'21.35"W, 1405 m, molted on laboratory, captured as juveniles on 16 Jun 2017, N. Ferretti, S. Copperi, G. Pompozzi; four females and thirteen juveniles, Valle de Calamuchita, Villa Berna, 31°54'35.60"S - 64°44'21.35"W, 1405 m, 16 Jun 2017, N. Ferretti, S. Copperi, G. Pompozzi.
Species distribution models. The model of current suitable climatic areas yielded an AUC value of 0.992 suggesting a good performance of the model. The bioclimatic variables that provided the highest contribution were: Bio4 (temperature seasonality) with 47.8 of percent contribution and Bio 14 (precipitation of driest month) with 29.8 of percent contribution. We observed that highest suitability areas for the species corresponded to the north-eastern areas and coasts of Buenos Aires province, southern Entre Ríos province in Argentina and central-eastern Uruguay (Fig. 5a). Moreover, it was possible to see that the suitable areas extended towards the new record provided in this work but not including it as a suitable area (Fig. 5a). From the model (AUC value of 0.993) under past climate conditions (~120 000 years ago), suitable habitats were identified at central Buenos Aires province, southern Entre Ríos province in Argentina and western Uruguay (Fig. 5b). The location of the population found would not be a possible refuge, it is probable that other factors even oldest than the last Inter-Glacial events could isolated this population. For example, the marine transgressions that took place during the middle Miocene (~15 Ma years ago) could have affected their distributional pattern, as was evidenced for other arachnids (Ferretti et al. 2012, Ojanguren-Affilastro et al. 2017). Moreover, during the marine transgressions, the emerged lands during this period corresponded to south of Buenos Aires province, the mountainous systems of central Argentina, in Córdoba province, and central-eastern Uruguay (Ortiz-Jaureguizar and Cladera 2006). The findings of the present work not only expand the known geographic distribution of the species but also remark the isolation of this population in relation to the suitable areas predicted with the species distribution model.