A revision of the spider genus Selenops Latreille, 1819 (Arachnida, Araneae, Selenopidae) in North America, Central America and the Caribbean

Abstract The spider genus Selenops Latreille, 1819 occurs in both the Old World and New World tropics and subtropics and contains nearly half of the species in the family Selenopidae Simon, 1897. In this paper the members of the genus Selenops found in North America, Central America, and on islands of the Caribbean are revised, excluding Cuban endemics. No taxonomic changes are currently made to the species from the southwestern United States. In total, 21 new species are described, including Selenops arikok sp. n., Selenops chamela sp. n., Selenops amona sp. n., Selenops baweka sp. n., Selenops bocacanadensis sp. n., Selenops enriquillo sp. n, Selenops ixchel sp. n., Selenops huetocatl sp. n., Selenops kalinago sp. n., Selenops oviedo sp. n., Selenops morro sp. n., Selenops denia sp. n., Selenops duan sp. n., Selenops malinalxochitl sp. n., Selenops oricuajo sp. n., Selenops petenajtoy sp. n., Selenops guerrero sp. n., Selenops makimaki sp. n., Selenops souliga sp. n., Selenops wilmotorum sp. n., and Selenops wilsoni sp. n. Six species names were synonymized: Selenops lunatus Muma, 1953 syn. n. = Selenops candidus Muma, 1953; Selenops tehuacanus Muma 1953 syn. n., Selenops galapagoensis Banks, 1902 syn. n. and Selenops vagabundus Kraus, 1955 syn. n. = Selenops mexicanus Keyserling, 1880; Selenops santibanezi Valdez-Mondragón, 2010 syn. n. = Selenops nigromaculatus Keyserling, 1880; and Selenops salvadoranus Chamberlin, 1925 syn. n. = Selenops bifurcatus Banks, 1909. Lectotypes are designated for the following three species: Selenops marginalis F. O. Pickard-Cambridge, 1900 (♂), Selenops morosus Banks, 1898 (♂), and Selenops mexicanus Keyserling, 1880 (♀). The female neotype is designated for Selenops aissus Walckenaer, 1837. The males of Selenops bani Alayón-García, 1992 and Selenops marcanoi Alayón-García, 1992 are described for the first time, and the females of Selenops phaselus Muma, 1953 and Selenops geraldinae Corronca, 1996 are described for the first time. Almost all species are redescribed, barring Cuban endemics and a few species recently described. New illustrations are provided, including those of the internal female copulatory organs, many of which are illustrated for the first time. A key to species is also provided as are new distributional records.


Introduction
Spiders of the family Selenopidae Simon, 1897, also known as wall crab spiders or flatties, have a cosmotropical distribution. They are exceptional in that both their running and striking speeds place them amongst the world's fastest animals , and they are extremely dorsoventrally flattened . Prior to this work, the family currently comprises around 196 species in 5 genera (Corronca 1998;Platnick 2010): Anyphops Benoit, 1968, Garcorops Corronca, 2003, Hovops Benoit, 1968, Siamspinops Dankittipakul & Corronca, 2009, and Selenops Latreille, 1819. The first three genera are distributed in Africa and Madagascar, and Siamspinops in Southeast Asia. Five new genera are currently being described for a total of nine genera (Crews and Harvey 2011). Selenops is distributed throughout the tropics and subtropics worldwide, and the majority of the species in the family are placed in this genus. Walckenaer (1837) first recognized three groups based on characteristics of the chelicerae, labium and leg lengths. These characters were not substantiated by Simon (1880), who attempted to divide the family into Old and New World groups based on eye size. Corronca (1998Corronca ( , 2002 provided diagnoses of the genera, however there are many variations. The monophyly of none of the genera has been formally tested.
In the Western Hemisphere, F. O. Pickard-Cambridge (1900) first distinguished species using eye size and position along with genitalic characters. Petrunkevitch (1925Petrunkevitch ( , 1930 divided these species into groups and described new species based on leg proportions. Muma (1953) established six subgeneric species groups for species occurring from North and Central America and the Caribbean based on leg lengths, eye size and position, and genitalic characters. Current authors (Alayón-García 1992, 2005Valdez-Mondragón 2007 still use these characters and groupings despite variation and species that do not fit into any group. While selenopids are quite common in certain areas and are relatively large, museum collections of this family are often depauperate (Corronca 1998). As with most spiders, many species of the family are known from either a single male or single female specimen, known only from the type, many of which were collected over 100 years ago. Sexes can be difficult to match and even if two sexes were collected at the same place at the same time, this does not guarantee they are the same species, as many species have overlapping distributions. I have found that the best ways to determine if certain males and females are the same species are collecting both sexes from multiple localities, and use of independent data, such as DNA or other genetic data.
Revisionary and descriptive work for Anyphops, Hovops and Garcorops has been done by Lawrence (1940), Benoit (1968), and more recently by Corronca (1996Corronca ( , 2000Corronca ( , 2003Corronca ( , 2005. The genus Selenops has been revised primarily by region, with the African species examined by Lawrence (1940Lawrence ( , 1942, Benoit (1968), and Corronca (2001Corronca ( , 2002, and the South American species revised by Corronca (1996Corronca ( , 1998. In the West Indies and Central America, the last major revisionary work was conducted by Muma (1953), while Alayón-García (1992, 2005 revised the species in Cuba, and has completed other regional descriptions. Only a few species have been described from México (Valdez-Mondragón 2007 since Muma's (1953) work. The Asian and Australian species are concurrently being revised and described (Crews and Harvey 2011), but are still in need of collection, as only a handful are in museums and have been described (Simon 1901;Gravely 1931;Reimoser 1934;Tikader 1969).
Here a taxonomic revision of Selenops from Greater North America, including México, Central America and the Caribbean islands, except for Cuba, is presented. Twenty one new species are described, and the first descriptions of 2 males and 2 females from previously described species are included. Also, 6 species names are synonymized and comprehensive locality records are provided allowing a more thorough examination of distributional data. Additionally, valuable natural history data, and new illustrations, including those of internal female copulatory organs, which have largely been omitted in past revisions and descriptions, are provided. Many of the types are very old and in poor condition (e.g. -coloration and markings have faded, leaving the specimens an artificial orangey color; many of the hairs have worn off; in some cases abdominal setal tufts 192) may actually be present on newer specimens). For many specimens, all the legs, or parts of the legs, have been disarticulated; in such cases, previous author's leg formulae have been used. Descriptions of these along with descriptions of newer specimens for comparison, as well as photographs of live specimens  have been provided. Information on the locality of types is presented. Finally, data independent of morphology (Crews and Gillespie 2010) have been used to place juvenile specimens for the purpose of (1) providing locality records, (2) examining of the validity of hypothesized species groups (Muma 1953) and associated morphological characters, and (3) examining monophyly and relationships of the genera that are based on morphological characters. A key to species is provided. The fossil species S. beynai Schawaller (1984) is not included in the key, but images from high resolution CT-scanning are included .
Despite the spiders being large and conspicuous, and subject to recent descriptive and revisionary work (Alayón-García 1992, 2005Corronca 1996Corronca , 1998Valdez-Mondragón 2007, several previously undescribed species have been found. There is no doubt that there are still more species to be discovered from the region. Even though there are some widespread species, there are also several species with very small ranges, and many are found from the same area. Of note are the Greater Antillean Islands of Cuba and Hispaniola, as well as mainland México, all of which are extremely species-rich.

material and methods
In addition to the examination of museum specimens, several new specimens were collected from over 200 localities primarily in the Caribbean, Southwestern United States, México and Central America. In total, over 1600 specimens were examined. Voucher numbers for new collections were placed in the vial and can be found in Crews and Gillespie (2010). These numbers begin with sel_, for Selenops.
Measurements were taken using a Leica dissecting microscope and ocular micrometer, or from photographs taken with the Microptics system. The photos were taken with a Minitool scale in place, and then the images were imported into Adobe Illustrator where they could then be measured. All measurements are in millimeters (mm).
The width and length of the carapace and sternum were measured at the widest and longest parts. Legs were measured along the dorsal side. Eye dimensions were measured across the lens, and eye group measurements and inter-eye distances were measured from the farthest distances.
Genitalic illustrations were made from photographs as well as actual specimens. The left palpus of the male was removed from each specimen and photographed using the Microptics imaging system. If only a right palpus was available, it was photographed and the photo was inverted horizontally for consistency using Adobe Photoshop. In most descriptive work and revisionary studies, the left palpus has been figured. However, Muma (1953) figured the right palpus. In this study, to adhere to convention and not frustrate workers that have already removed the left palpus, the left palpus is illustrated. The internal copulatory organs were not figured by Muma (1953), but I have found that the internal genitalic characters are important for identification and for assessing evolutionary relationships, thus they are figured here. The internal copulatory organs of many female Selenops species are highly sclerotized and full of strong tissues. Attempts to remove the tissues with forceps could potentially cause damage to the underlying structures. A digestive enzyme mixture of pancreatin and borax following Alvarez-Padilla and Hormiga (2008) was used to clean the copulatory organs. In many species, the epigynal plate folds inward and is directed anteriorly. This has been referred to as a 'uterus externus' by Dankittipakul and Corronca (2009), though this seems to be an incorrect usage (see Snodgrass 1952 andFoelix 1996), and in this work it is referred to as the posterodorsal fold of the epigynum. This invagination covers the spermathecae or other parts of the internal copulatory organs. Thus, it can be difficult to distinguish characters. Where possible, half of the posterodorsal fold was removed, though sometimes it was removed in its entirety to allow the viewing of the internal copulatory organs. Great care was taken to not disturb the internal genitalic structures during removal. The left palpus is figured in ventral and lateral view, and the female copulatory organs are figured in ventral and dorsal view. The figures of the internal copulatory organs have been drawn so that the left half depicts the internal copulatory organs with the posterodorsal fold removed and the right half depicts the internal copulatory organs with the posterodorsal fold present. Terms used in reference to genitalic structures follow those of Corronca (1998) unless otherwise noted. Muma (1953) referred to the apex of the abdomen of some specimens as having a festooned appearance, caused by the dark coloration of the posterior lateral margins of the abdomen. The same terminology is used here. Descriptions were written primarily using CSRIO DELTA v1.04 for windows (Dallwitz et al. 2000), and edited manually. In many cases the types are quite old and in poor condition. In the instances where newer specimens were available, these are described alongside the type specimens. Immature specimens were identified to species using molecular genetic data (Crews and Gillespie 2010).
Abbreviations used in the text are as follows:  Muma (1953) defined six species groups (banksi group, debilis group, insularis group, lindborgi group, mexicanus group and spixii group) based on characters that even he stated were subject to some variation. There were several species which did not fit into any of the Muma (1953) groups. However, given the lack of a more robust framework, modern workers (Alayón-García 19922005Valdez-Mondragón 2007 continue to place species in these groups without any evidence that these groups are phylogenetically valid. The first character, relative leg lengths, is problematic in that spiders tend to easily autotomize the legs to avoid being trapped or eaten, and this is especially common in selenopids, where they are missing legs on both sides of several specimens. If the spider is not yet an adult and will continue to molt, the leg will grow back, however, the leg will be shorter. Also, many of the specimens are in poor condition and legs are disarticulated from the specimen. Finally, leg lengths can differ between males and females of the same species, within the same sex of species, and sometimes on the left and right sides of a single specimen. Corronca (1996Corronca ( , 1998 defines the genus Selenops as having the second legs longer than the fourth, however, Muma (1953) defines species groups by having the second legs longer than the others or the fourth leg shorter than the others. In this study, the leg formulae are given; however, this character appears to be misleading. The comparative eye size also is subject to variation and can differ between the sexes within a single species, and also within species of the same sex. Genitalic characteristics and genetic data appear to be the most powerful tools for species determination, or determination of evolutionary relationships in these spiders, however, it does appear that there is some genitalic homoplasy, as species in different genera may sometimes have superficially similar copulatory organs.

monophyly of species groups and the genus Selenops
Previously, a comprehensive phylogenetic hypothesis was assembled for Selenops from North America, Central America and the Caribbean based on a large amount of molecular data (Crews and Gillespie 2010). Specific relationships and their association with Caribbean biogeography are discussed in that paper. Here I review those results in connection with previously defined species groups and the monophyly of the genus Selenops.
The monophyly of Muma's (1953) species groups cannot be tested thoroughly as there are some species from nearly all of the six species groups that I could not include in a previous study (Crews and Gillespie 2010). However, the data generated statistical support for only a few of Muma's (1953) species groups. The 'banksi' group, consisting of three species, S. banski, S. micropalpus and S. minutus is statistically well-supported, though S. minutus was not included here (see Crews and Gillespie: fig. 6, clade F). The 'debilis' group, consisting of S. bifurcatus, S. abyssus, S. lepidus, S. buscki, S. scitus, S. debilis, S. actophilus and S. nesophilus is recovered, but is lacking statistical support (see Crews and Gillespie 2010: fig. 6, clade I). The 'insularis' group consisting of S. insularis, S. trifidus, S. submaculosus, S. simius, S. alemani, S. vinalesi and S. candidus is not recovered. The 'lindborgi' group consisting of S. lindborgi, S. hebraicus and S. formosus is not statistically supported as S. lindborgi and S. hebraicus occur in distant parts of the tree. It is unknown which taxa were in Muma's 'spixii' group, as these were not explicitly stated. The 'mexicanus' group, originally consisting of only two species S. mexicanus and S. gracilis, and additional species, is statistically supported (see Crews and Gillespie 2010: fig. 6, clade H). Several 'unplaced' species were not available for these analyses, but the following were: S. phaselus, S. curazao, and S. marcanoi. Muma (1953) did not recognize S. phaselus as fitting into any of the defined groups and its placement in all of the analyses is ambiguous and unsupported. Alayón-García (2001) could find few affinities between S. curazao and South American or Cuban taxa, but left this species unplaced. In all three analyses it is closely allied with species from Trinidad and Tobago and the northern South American mainland (see Crews and Gillespie 2010: fig. 6, clade C). Alayón-García (1992) also did not place S. marcanoi, which in all three analyses is closely allied with S. insularis and several other Hispaniolan species (see Crews and Gillespie 2010: fig. 6, clade M). Placing species in groups based on the aforementioned commonly used morphological characters is unfounded. Based on these analyses, the Muma (1953) species groups are not reliable, and the practice of placing species within this genus in these species groups should be abandoned.
Based on the results of my molecular (Crews and Gillespie 2010) and morphological studies (Crews and Harvey 2011) it would appear that the monophyly of the genus Selenops is somewhat questionable. In the molecular phylogenies, Selenops is either para-or polyphyletic, though these relationships are not supported. The para-and polyphyly occurs between the Old and New World selenopids. That is, the New World selenopids are monophyletic, and the Old World selenopids are not -either with respect to one another or with the New World selenopids. Benoit (1968, p. 118) noted that American Selenops are very different from the African ones, and that they do not have anything at the generic level in common with types from the Old World. He suggested that they should be the object of a new classification. Unfortunately, he did not elaborate further. This conundrum is discussed further in Crews and Harvey (2011), though no better conclusion than Benoit's is reached.
Etymology. The specific epithet refers to the type locality, Arikok National Park, and should be treated as a noun in apposition.
Diagnosis. This species is most similar to S. curazao and S. isopodus, but can be distinguished by the more quadrangular and more hyaline median field than S. curazao, and unlike S. isopodus, the median field is not completely enclosed anteriorly. The shape of the epigynal plate and the internal copulatory organs also differ significantly (Figs 1-2). Males unknown.
Remarks. Even though only two adult female specimens have been collected, some variation is seen in the shape of border of the median field of the epigynum. In the holotype it is wider anteriorly, and more quadrangular in general, while in the other specimen, the median field is more rounded, and not wider anteriorly than it is posteriorly.

Selenops geraldinae
Note. The holotype of S. geraldinae Corronca is stated to be deposited in the CAS (Corronca 1996: 95), but the specimen has not arrived at CAS. Nevertheless, the description and illustrations are adequate to recognize as this species the specimens described and listed below.
Diagnosis. Males of this species can be distinguished from others by the small, finger-like MA directed distally, located distally on the palpal bulb . Females of S. geraldinae can be separated from others by the large central, lateral extending depression of the epigynum, lateral lobes that come together medially and diverge laterally (Figs 9-10).
Remarks. The male of S. geraldinae was described by Corronca (1996). In Trinidad and several small nearby islands, I collected several males of S. geraldinae and only with females of what Corronca (1998) had previously described as S. willinki. Along with this natural history data, molecular genetic data (Crews and Gillespie 2010) confirmed that these males and females were conspecific, exclusive of any other nearby specimens (e.g., from Tobago).
In Corronca (1998), the records of S. geraldinae are given as Mara and Bolívar, Venezuela. They are shown on the map as being in Zulia, near Lago Maracaibo, in the east of the country. However, there are also localities with the same names in the western part of Venezuela. As these specimens have not been examined, it is unclear where exactly they are from. I have shown the distribution following Corronca (1998).
Diagnosis. Males can be distinguished from other species by the sclerites which obscure the cymbium laterally, and the RTA is directed lateroventrally, and the very small, finger-like, distally directed MA . Females can be distinguished from all other species by the anteriorly located genital openings, and medially located epigynal pockets (Figs 13-14).
Remarks. The female of S. willinki is described here for the first time. Corronca (1998) associated the female of S. geraldinae with this species. Additional collecting and molecular genetic data (Crews and Gillespie 2010) confirmed that the male and female described here as S. willinki are conspecific.
Type material. Holotype male: Ayotzinapa, Guerrero, México, 11.I.1941, (AMNH, examined). Paratypes: Female, Taxco, Guerrero, México, IV.1946, L. Isaacs (AMNH, examined). Diagnosis. This species is similar to S. mexicanus, S. malinalxochitl, S. ecuadorensis and S. aztecus. Males can be easily distinguished by a very short embolus which arises from behind a larger sclerite, and also by the long, distally rounded RTA, . Females can be distinguished from other species by a hyaline median septum that goes behind a slightly m-shaped posterior area with lateral indentations, coming to a point medially, and internally, having laterally directed, non-branching internal ducts, with fertilization ducts located and directed laterally, and a rectangular posterodorsal fold (Figs 23-24).
Remarks. There is some variation in the external female copulatory organs. In all specimens viewed thus far, the posterormedial margin is always pointed, and internally, the ducts are all identical, unbranched, with an identically shaped rectangular posterodorsal fold.
Etymology. The specific epithet comes from the Nahua goddess Malinalxochitl, the goddess of desert-dwelling snakes and arthropods, and refers to the type locality and the indigenous Nahua people of this species' distribution. It is to be treated as a noun in apposition.
Diagnosis. This species is most similar to S. gracilis, but the posterior sclerotized area is u-shaped, with the lateral indentations directed anteriorly, and the posteromedian area of the caudal margin is rounded, rather than pointed (Fig. 25). Internally, the posterodorsal fold is medially depressed and the internal ducts are branched (Fig. 26). Males unknown.
Remarks. As in other closely related species, there is some epigynal variation. The internal ducts and posterodorsal fold are uniform across specimens examined.
Remarks. Several species are synonymized with S. mexicanus. The types of these species were compared with the available lectotype of S. mexicanus and/or DNA evidence from spiders from the same localities (Crews and Gillespie 2010) and these are being used as evidence for these synonymies. In the types of S. tehuacanus and S. galapagoensis, no differences were detected between these types and the types of S. mexicanus. The types of S. vagabundus were unable to be examined. However, from the illustrations of Kraus (1955), there are some differences in the male. I collected specimens that match the illustrations of Kraus (1955), but using DNA data, the specimens are deeply nested within the S. mexicanus clade. The differences include a shorter, blunter RTA and part of the MA is obscured by a tergite from which the embolus arises. From the illustration of the female copulatory organs (Kraus 1955) it is difficult to determine if there are any differences, however, new specimens were taken from the same locality and sequenced, and no differences were found.
There is some variation as might be expected in such a widespread species. Genitalic details reveal consistency in males in the RTA, MA and embolus, as well as in the internal copulatory organs in the female. Though the RTA may not be as long as in the holotype specimen (Muma 1953: figs 1-2), it is sinuous and tapers distally. MA with two branches in all specimens, with two bent, finger-like processes, though one branch may be obscured by the sclerite from which the embolus arises in some specimens. The lateral edge of the conductor is slightly sinuous in some specimens. The shape of the median septum and the degree of sclerotization of the median field varies. The median septum may be round to angular and the median field may only be lightly sclerotized or more strongly sclerotized. The area around the median field may be very defined or raised, or may be flat. The posterior indentations also differ in their shape and size. The posterodorsal fold is typically large and quadrangular, though may be rounded or smaller, but a portion of the spermathecae can always be seen above the anterior margin, and the fertilization ducts can usually be seen laterally.
It is clear that other closely related species may yet be discovered (e.g. the female of S. aztecus). The genitalic features mentioned above as well as DNA will be useful in distinguishing these new species from S. mexicanus. This species appears to be able to escape detection on produce or landscape plants and has been introduced into the US and St. Maarten. In St. Maarten, a male and young female were collected on a palm that had been shipped from Florida, via México, indicating the species is well-suited for travel. In most areas where it has been found in the US, such as Wisconsin and Washington, climate would likely not allow the spider to become established. In Florida or St. Maarten, the climate is similar to the native range and the species may be able to become established. This could be detrimental to local or endemic species, such as S. souliga sp. n. in St. Maarten.
Distribution. This species is one of the most widespread species of selenopids, naturally occurring from northern México, south to northern South America and the Galápagos Islands (Map 6). Diagnosis. This species can be distinguished from others by the inconspicuous lateral lobes, and the triangular epigynal plate, with a central v-shape (Fig. 31). Internally, the spermathecae are large and round, located laterally (Fig. 32). Males unknown.
Diagnosis. Males can be differentiated from other species by the embolus which has a round base, and is bladelike distally. Embolus located more medially than laterally. Also, MA with quadrangular base and strongly hooked distally (Fig. 33). Females can be separated from other species by the quadrangular median area of the epigynum (Fig. 35).
Remarks. In the illustration of the ventral view of the palp of the male by F. O. Pickard-Cambridge (1900: fig. 15), the MA looks somewhat different than on the lectotype. Everything else looks exactly the same. It is my conclusion that the MA was articulated differently, and that the equipment used to view the palp such a long time ago was perhaps not as sophisticated as equipment used today, and thus, this is definitely the same species. This species has been listed as occurring in Panamá (Nentwig et al. 1993), however, I believe this to be a misidentification.
Remarks. Muma (1953) placed this species with S. micropalpus and S. banksi, stating that the embolus and long palpal tibia from the illustrations of F. O. Pickard-Cambridge (1900) were of the same type. Muma (1953) was unable to examine the actual specimens. The illustrations of F. O. P-Cambridge (1900), however, do not show a long palpal tibia, or an embolus similar to that of S. banksi or S. micropalpus. F. O. Pickard-Cambridge (1900) also mentioned that none of the females were adults, yet he illustrated an adult. After close examination of the specimens, I do not believe this species to be aligned with S. micropalpus or S. banksi, but rather with other species found in México and Guatemala. Unfortunately, this species has not been collected for over 150 years and its precise range is unknown, as the label says only 'Guatemala.' This species is listed by Nentwig et al. (1993) as being found in Panamá. This is because Banks (1929) included this in his list of spiders from Panamá, rather than based on a collection made by the former authors. Banks (1929) included no figure of this spider in his publication, and the spider has not been found in any collection examined. I regard Panamá as an erroneous locality for this species.
Natural history. No data. Distribution. All that is known is that the species is found in Guatemala, which may have had different borders at the time the collection was made (Map 5). Type material. Lectotype male (designated here): Tepic, México, X-XI, Eisen and Vaslit (MCZ, examined). Female syntype has been lost or destroyed, and thus the female is unknown.
Diagnosis. Males can be distinguished by the RTA with two small pointed, angular projections, and the MA, which has a large triangular base and is directed somewhat ventrally (Figs 41-42).
Remarks. Banks (1898) illustrated a male and a female specimen and noted that 'several specimens' had been collected. Muma (1953) reported that the female type was lost or destroyed, but that females of S. abyssus may go with males of S. morosus, as the illustration of the epigynum by Banks (1898) is similar to that of S. abyssus. Muma also mentions two males and one young female from Tepic at the MCZ, but only a single male was located. An additional specimen from much further north (Sonora) was found in the AMNH. The epigynum illustrated by Banks (1898) looks nothing like that of S. abyssus, and I conclude that S. abyssus and S. morosus are distinct species. Banks (1898) did indeed have the female of S. morosus. The geographic range of S. abyssus encompasses that of several species known from only one sex, as does S. morosus. More collecting from this region and independent data, such as that of molecular or behavioral data, will be needed to determine which males and females belong to-gether. The male specimen is also in poor condition and the description may not reflect well the coloration and markings of this species.

Selenops nigromaculatus
Remarks. Until very recently (Valdez-Mondragón 2010) there was much confusion regarding this species, in particular the female. The male and female were described by Keyserling (1880). The illustrations are of poor quality and it is difficult to distinguish details from them. F. O. Pickard-Cambridge (1900) produced several beautiful illustrations of the male of S. nigromaculatus. For unknown reasons Simon didn't show him the female, and unfortunately it was not illustrated. In 1918, Mello-Leitão copied Keyserling's (1880) description and illustration, claiming he had collected a similar female, based on Keyserling's description, in Pará, Brazil. Finally, Muma (1953) was not able to examine the specimen himself, but he had M. Vachon describe and illustrate the male. Apparently, Vachon also illustrated the female, but this was disregarded by Muma because "There is no doubt that Keyserling's female type is the same as gracilis, new species," which added to the confusion. Although Keyserling's (1880) illustrations are lacking in certain details, his illustration of the female of S. nigromaculatus looks nothing like S. gracilis. Other illustrations of female Selenops copulatory organs (e.g. S. mexicanus) do resemble what the species actually looks like. I have come to two conclusions, both having the same outcome. First, Keyserling (1880) may have illustrated a juvenile. In the vials of S. nigromaculatus types at the BMNH, there is no adult female. There is a single adult male, which is the type, and, in addition, there is a juvenile spider in another vial labelled as S. nigromaculatus, from Panamá, Koch collection, with a blank label, numbered 15.3.5.6416. There is no female, which looks like S. gracilis or otherwise. The second involves the actual female of S. nigromaculatus being lost. Luckily, a male of S. nigromaculatus was collected with a female, described and beautifully illustrated by Valdez-Mondragón (2010). The specimens were described as S. santibanezi. The male type and illustration by Valdez-Mondragón (2010) clearly reveal these are the same species, and thus, S. santibanezi is a junior synonym of S. nigromaculatus.
There also appear to be some erroneous locality records. Mello-Leitão's (1918) collection of a female in Pará, Brazil is dubious, given that the actual specimen was not illustrated, and that it resembled Keyserling's description. In addition, the thorough work on South American selenopids by Corronca (1996Corronca ( , 1998 has turned up no specimens of S. nigromaculatus. Finally, it is mentioned by Simon (1880) and Mello-Leitão (1918) that this spider was found in Antigua by Grenadier. This particular specimen has not been located. This could refer to the vicinity of Antigua, Guatemala, rather than the Caribbean island nation of Antigua and Barbuda. The type locality is México and a different species has been found in Antigua and Barbuda which is not closely allied with S. nigromaculatus.

Selenops makimaki
Diagnosis. This species can be distinguished from all other species by the presence of a thin median septum, while the region surrounding the median septum is more heavily sclerotized, giving the appearance of a keyhole-shaped opening (Fig. 45). Males unknown.

Selenops debilis group (sensu Muma 1953)
The following three descriptions are of the spiders that belong to what Muma (1953) called the Selenops debilis group. These spiders were all described nearly 100 years ago, and the male and female types of S. debilis and S. actophilus are not from the same locality (though collecting from the same locality does not guarantee the same species as known in other species). Additionally, the three species have overlapping distributions in the desert southwest of North America and all three may be found in a single locality (Fig. 198, Map 7). Finally, there also appears to be genitalic variation away from the types, and several specimens have been found that share morphological features of more than one species. DNA data indicates at least three species in the region (Crews and Gillespie 2010). Morphologically determining which specimen belongs to which species is a daunting task, and in some instances is not possible at the present time.
Unfortunately recent collections did not contain adults. At this time, no taxonomic changes are made. Below, the types are re-described and locality data are included only when a specimen could, with reasonable certainty, be assigned to a particular species (i.e., when they did not differ from the type). Additional locality records are included below for members of the S. debilis group that cannot be sorted to species. Variations of the internal female copulatory organs are found in Figs 65-68. In order to sort out the variation and uncertainty in this group, to determine the correct number of species, and to learn more about biogeography and possible hybridization, this group is in need of revision. This will include the collection and analysis of molecular, morphometric, and behavioral data.  Diagnosis. Males can be distinguished from S. debilis and S. nesophilus by a combination of the embolus beginning at the 5 o'clock position and a MA that is broad, but tapers suddenly to a finger-like extension. Additionally, the conductor and RTA are shaped differently than those of S. debilis and S. nesophilus (Figs 59-62). Females can be distinguished by a more triangular epigynal plate, with a distally narrower septum. Internally, there is a much smaller invagination of the posterodorsal fold than in the other two species, and the spermathecae consist of two nearly parallel branches extending anteriorly (Figs 55-56).
Other Diagnosis. Females can be distinguished from others by the combination of a wider median septum with parallel lateral margins, and long, cylindrical spermathecae directed anterolaterally (Figs 57-58). Males can be distinguished from other species by the RTA, which is a very small stalk, distally bifurcated into a v-shape, with a small angular lateral projection (Figs 59-60).
Remarks. The male was matched to the female neotype by Muma (1953) based on the general appearance, including coloration, pattern and structure. They are from distant localities, and there are other species in this group that are very similar, and spe-cies boundaries are unclear. Thus, it is entirely possible that this male and this female are not the same species.
Natural History. Found under rocks and logs in both arid regions (Fig. 198) and oak woodlands.
Remarks. This male and female were not collected together and it is presumed Muma (1953) placed them together based on their size. However, a second small species (S. chamela sp. n., see below) known from a single female has been found in this region, and nearer to the type locality of the male of S. lepidus. Thus, it is possible that what is currently described as the female S. lepidus is actually a different species. In this region of México, there are 3 females known and 2 males known, and it is unclear if any of them go with the other, or if they are all separate species. At this time I make no changes.
Paratype female: Color: carapace orange-brown with white setae, markings indistinguishable; sternum light brown; chelicerae brown; maxillae light orange-brown lightening distally; labium brown lightening distally; abdomen dorsally grey-tan, broad creamcolored w-shaped mark posteriorly, laterocaudal festoon present; ventrally dusky grey with no markings; legs orange-brown with annulations. Carapace: 1.01 times longer than broad. Eyes: AER slightly recurved; PER recurved; PME larger than AME, PLE largest, ALE smallest; eye diameters, AME 0.12, ALE 0.10, PME 0.20, PLE 0.28; interdistances AME-PME 0.03, PME-ALE 0.10, ALE-PLE 0.30. PME-PME 0.85. ALE-ALE 1.40; ocular quadrangle AME-AME 0.40, PLE-PLE 1.57; clypeus 0.13 high. Mouthparts: chelicerae with a few stout setae medially and anteriorly; maxillae longer than broad, with tuft of conspicuous setae distally; labium distally rounded. Sternum: 1.25 times longer than broad, posteriorly indented. Legs: leg formula 4321 (Muma, 1953); tarsi I-IV with strong claw tufts; prolateral claw per foot slightly toothed. Abdomen: without terminal setal tufts. Epigyne: median septum located in center of epigynal plate, narrow and parallel sided, lateral lobes covering terminus of septum, one lobe atop the other, making the epigynum appear slightly asymmetrical, genital openings located posterolaterally to septum, epigynal pockets present; internally, openings connect to a large, somewhat triangular atrium that connects to lateral anteriorly directed ducts, one small, the other larger with spiralled appearance, fertilization ducts located and directed posterolaterally, posterodorsal fold present, but barely covers part of internal ducts (Figs 71-72 Etymology. This species is named after the type locality, Chamela, and is to be treated as a noun in apposition. Diagnosis. Females can be separated from all other species by the epigynum which has a median septum and epigynal pockets, and internally, the ducts are coiled at least 5 times (Figs 73-74). Males unknown.

Remarks. It is unclear whether this is a new species or is the female of S. lepidus.
See abover under 'Remarks' of the latter species.
Natural history. Collected under bark, rocks and concrete blocks in cloud forests (Fig. 200). It has been found with S. mexicanus.
Distribution. Southern México in the state of Chiapas (Map 5).  Banks (1909) and Muma (1953). The illustration of the female was clearly that of S. salvadoranus as illustrated by Muma (1953). It seems as though Kraus (1955) was aware of Muma's (1953) work, but perhaps was unable to look at the actual specimens. In many instances, the innermost branch of the MA is obscured by the conductor, in which case, the MA appears to be single branched, rather than double branched. It would appear that this was the case when Muma (1953) illustrated the type, which upon closer examination does have a two branched MA. Several species of selenopids have overlapping distributions. Some species have widespread distributions, while others have small distributions. Thus, even collecting a male and female from the same locality does not guarantee that they are same species. Unless a male and female are collected mating, or collected simultaneously from multiple localities, or their identities can be cross-checked with DNA data, it is recommended that they not be described as the same species, as this case illustrates. I have collected the same male and female from multiple localities across Central America, they have mated and produced offspring in the lab, and they have matching DNA sequences for multiple genes. In conclusion, the name S. salvadoranus has been synonymized with S. bifurcatus, and what was originally designated as the female of S. bifurcatus is here described as S. oricuajo sp. n. (see below).
Distribution. Found from southeastern Guatemala to northwestern Costa Rica (Map 8). Diagnosis. Females can be separated from all other species by the centrally located median lobe, the epigynal pockets, and the shape of internal ducts (Figs 83-84). Males unknown.
Natural history. All that is known of this species' natural history is that it has been collected with S. bifurcatus.
Distribution. Known only from the type locality (Map 8). Etymology. The specific name comes from the Taino word 'amona', the indigenous name for the type locality. It is to be treated as a noun in apposition.
Remarks. Despite the large distribution this species occupies, little to no variation was noticed in size or in genitalic morphology.

Diagnosis.
Females can be separated from other species by the posterior margin of the epigynal plate which extends below the epigastric furrow and forms two points (Figs 93-94). Males can be separated from other species by the very large RTA which extends laterally, and curves back toward the bulb. The longest apophysis also has a small process (Figs 95-96).
Remarks. Walckenaer (1837) described this species from Trinidad. The type was lost. This species is not found in Trinidad. Either Walckenaer (1837) described another species, or this type was from another locality where S. aissus actually occurs. The drawing of S. aissus from Muma (1953) is clearly the same as that provided by Keyserling (1884). Petrunkevitch (1925) doubted that Keyserling's (1884) S. aissus was the same as the type described by Walckenaer (1837), and described S. confusus. Petrunkevitch (1925) based this on leg lengths, which have problems associated with them, including intraspecific variation. Petrunkevitch (1925) also illustrated two species of 'S. aissus' from Panamá, but the illustrations are clearly those of S. mexicanus. Alayón-García (2005) synonymized the male of S. vexillarius with S. aissus. Although the type was not figured by Walckenaer (1837), Muma (1953) believed this specimen to be the same as the one he was describing as S. aissus. This leads me to two conclusions, both with the same outcome: (1) Walckenaer (1837) had incorrect locality data for the specimen he described as S. aissus from Trinidad, or (2) the specimen Walckenaer (1837) described was introduced on materials shipped from somewhere within the range of the species. In either case, the name is valid, but the locality data is not. Thus, no changes have been made to the name. Yet, in order to stabilize the species name, the female neotype collected from the Bahamas has been designated.
The specimen from Montgomery, Alabama is regarded as a likely importation, and it is unlikely this species is established there, especially in the winter.
There is some variation in size. In particular, the specimens from Stocking Island are larger than specimens from elsewhere. For example, one specimen from Stocking Island (sel_325) has a body length of 15.20, while one from Great Exuma (sel_324) has a length of 8.20.There is variation in the morphology of the epigynum. In some species it is wider and shorter vs. others where it is narrower and longer. There is also variation in the markings on the abdomen. In some specimens there is only a festoon present. Other specimens have a festoon and faint pairs of spots anteromedially, some with festoon, spots and chevrons, caudally. There were not enough male specimens to assess variation.
Distribution. Found in the Florida Keys, the Bahamian islands of Abaco, Andros, Eleuthera, Grand Bahama, Great Exuma, Long Island, Staniel Cay, Stocking Island and San Salvador, as well as on the Greater Antillean island of Cuba (Map 10).
Diagnosis. Males can be separated from all other species by the palpus, which is similar to that of other Jamaican species in having a two-pronged embolus and a tibial apophysis with 3 branches instead of two, however, the dorsal branch of the RTA is wider distally, and the base of the MA is more quadrangular (Figs 97-98). Females can be distinguished from other species by the quadrangular to round median field (Figs 99-100).
Remarks. The female of this species was described by Petrunkevitch (1925) as S. aissus based on a description only, and without viewing Walckenaer's (1837) specimen. Muma (1953) described the female of S. lunatus, designating a new type. Muma also described the male as S. candidus and noted that, based on their overall appearance, S. candidus and S. lunatus might be the same species (these specimens were apparently not collected together, and the precise collection localities are unknown). Alayón-García (2003) re-described Petrunkevitch's (1925) specimens of S. aissus from the Peabody Museum as two different species, one as S. lunatus and one as the female of S. candidus. Muma (1953) noted that the two specimens designated as S. aissus by Petrunkevitch (1925) may demonstrate genitalic variation. In my extensive collecting, several males identified as S. candidus were collected, and one female from a second locality that was identified as S. lunatus. However, molecular analyses (Crews and Gillespie 2010) indicate that the female specimen described as S. lunatus is the same species as those specimens described as S. candidus. Therefore, it would seem the female previously described as S. lunatus by Alayón-García (2003) is a variant of S. candidus. Therefore, this species' name has been synonymised. Despite that Petrunkevitch's specimens were viewed recently (Alayón-García 2003), and should be in the PM, they cannot be located now. Female specimens show genitalic variation in the shape of the median field. It can be very square or roundish, and wide or narrow.

Selenops petrunkevitchi
Type material. Diagnosis. While similar to S. wilmotorum sp. n., the shape of the anterior margin of the lateral lobes is more angular, and less sinuous, and the internal ducts are distinctly coiled (Figs 101-102). Males unknown.
Distribution. Endemic to Jamaica, and appears to be widespread there (Map 11). Etymology. This species is named in honor of the Wilmot family for their hospitality and preservation and expansion of Jamaican culture. It is to be treated as a noun in apposition.

Selenops wilmotorum
Diagnosis. Females can be differentiated from other species by the sinuous margin and epigynal pockets (Figs 103-104) and males can be separated from other species by the conductor and thornlike medial branch of the RTA (Figs 105-106).

Selenops wilsoni
Diagnosis. The epigynym of S. wilsoni sp. n. somewhat resembles that of S. duan sp. n., in that they both have the lateral lobes fused to form an anteriorly projecting lobe (Figs 107, 133). In S. wilsoni sp. n., the lobes are completely fused anteriorly, the epigynal plate is not ovoid, and the posterodorsal fold is very large, over half the length of the plate, completely covering the spermathecae (Figs 107-108). Also, S. wilsoni sp. n. is only known from the island of Jamaica while S. duan sp. n. is known only from the island of Hispaniola. Males can be separated from other Jamaican selenopids by their smaller size and copulatory organs, including the shape of the large, robust RTA and the overall stoutness of the median apophysis, as opposed to only a stout base as in S. candidus (Figs 109-110).
Remarks. The copulatory organs of this species differ substantially from other Jamaican selenopids. The male palpal organ has a huge, bifid dorsal RTA and large median apophysis, though the branched embolus is the same as other Jamaican endemics. In the female, the anteriorly extending lobe of epigynum differs from that of all other Jamaican selenopids. Although a considerable amount of time was spent collecting in the vicinity of the type locality, only one juvenile of this species has been recovered, and many more individuals of S. wilmotorum sp. n.

Type material. Holotype male from northern Dominican Republic, in amber (SM-NHS).
Diagnosis. Males can be distinguished from other species by the shape of the RTA and the location of the MA and embolus (Figs 112-115).
Remarks. This species was first described by Schawaller (1984). Despite HR-CT scanning (Penney et al. 2007), details of this spider, in particular the copulatory organs, are still lacking. One palp is poorly preserved and the other is damaged. Other juvenile specimens from the SMNHS are beautifully preserved, however, it is unknown whether these are the same species. Dominican amber has been dated as being 16 myo (Iturralde-Vinent 2001).
Description. Holotype male: Color: overall greyish, though somewhat difficult to see due to poor preservation and color of the amber, no markings visible on the carapace or abdomen; legs yellowish with annulations on femora and tibiae. Carapace: 1.08 times longer than broad. Eyes: AER nearly straight; PER slightly recurved; AME larger than PME, PLE largest, ALE smallest; eye diameters, AME 0.33, ALE 0.19, PME 0.28, PLE 0.36. Mouthparts: Chelicerae with stout setae medially and anteriorly; maxillae longer than broad, with tuft of conspicuous setae distally; labium distally rounded. Sternum: longer than wide, posteriorly indented. Legs: Leg I only slightly shorter than legs II, III and IV; leg formula 2431; spination: leg I, Fm pr 1-1-1, d 1-1-0, rl 1-1-1; leg II, Fm pr 1-1-0, d 0-1-1, rl 1-1-1; leg III, missing; leg IV, Fm pr 0-1-1, d 0-1-1, rl 1-1-0. Natural history. This species is only known from a single fossil specimen, and thus, the natural history is unknown. However, as it was found in amber and many species of Caribbean Selenops are found under bark, it is likely that this species was also found under the bark and on the trunks of trees.
Distribution. Known only from a single specimen in Dominican amber. Diagnosis. This species is most similar in habitus to S. bocacanadensis sp. n., but the females can be easily distinguished by the copulatory organs in that in S. insularis the genital openings occur along a small v-shaped opening (Figs 119-120). The male is most similar to S. trifidus, but can be distinguished by the copulatory organs in that the dorsal branch of the RTA of S. insularis widens distally, and is longer anteriorly, and the MA is slender throughout (Figs 121-122).

Selenops insularis
Remarks. It is likely this species isn't established in Jamaica, as both records were from Kingston, a major shipping port, and are quite old. Recent searching in Jamaica turned up no specimens of this species. Muma (1953) mentioned that the females of this species are variable, but he did not elaborate. There is some variation in size in both males and females, but none was noticed in the copulatory organs of the many specimens examined. Muma (1953) suggested that some variants may be females of S. trifidus. Given the isolation of that species on Navassa Island, I believe this to be unlikely.
Distribution. Found in the Greater Antillean islands of Cuba, Hispaniola, Mona and Puerto Rico. It has historically been collected in Jamaica (see Remarks). It is also found in southern Florida (Map 12). Diagnosis. Females can be separated from others by having only a lightly sclerotized median area, with a sinuous opening and large, round spermathecae (Figs 123-124). The male palpal organ is most similar to that of S. insularis, though the circular space formed by the conductor connections is more semi-circular, and the RTA is not as wide distally and is truncate (Figs 125-126).

Selenops bocacanadensis
Etymology. The specific epithet refers to a locality where the species is found, La Boca de la Cañada. It is to be treated as a noun in appostion.
Diagnosis. The females of S. bocacanadensis sp. n. are similar to S. insularis in size and coloration, but females can be distinguished by the epigynal plate, as it is not as straight along the posterior margin, the genital opening is more rounded, rather than v-shaped, the posterodorsal fold is smaller, and the internal ducts are shaped differently (Figs 127-128). Males unknown.
Diagnosis. Males can be easily distinguished from other species by the large base and distal hook of the MA, and the small process on the distal end of the dorsal branch of the RTA (Figs 131-132). Females unknown.
Remarks. The type was apparently not in great shape when Muma (1953) viewed it 60 years ago. Since Muma's (1953) descriptions, it appears to be even more discolored.
Etymology. The species name comes from the indigenous Taíno word for the region, Duan, and is to be treated as a noun in apposition.
Diagnosis. Females can be separated from all other species by the lateral lobes which come together medially and form an anterior rounded projection that the genital openings are located behind (Figs 133-134). Males unknown.
Etymology. This species is named in honor of Denia Veloz for all of her work on arthropod biodiversity in the Dominican Republic. It is to be treated as a noun in apposition.

Diagnosis.
Males can be separated from all other species by the conductor which arises anteromedially on a short stalk with a rounded projection. The conductor is directed opposite the RTA and is hammer shaped (Figs 135-136). Females unknown.
Natural history. This species has been collected under bark and on buildings and fence posts at night, primarily at higher elevations (> 550 m).
Etymology. This species is named in honor of Kelvin Guerrero for all of his work on the biodiversity of arthropods of the Dominican Republic, and is to be treated as a noun in apposition.
Diagnosis. This species is most similar to Selenops baweka sp. n., but can be separated having the genital opening nearly straight, differently shaped spermathecae, and a very small posterodorsal fold (Figs 137-138). Males unknown.

Selenops baweka
Etymology. The specific epithet comes from the name the indigenous people of the Turks and Caicos Islands used to refer to the Caicos Bank, Baweka, where this species is endemic. It is to be treated as a noun in apposition.
Diagnosis. This species is similar to S. guerrero sp. n., but females can be differentiated by having the genital openings located behind a slightly sinuous margin, and internally, the ducts are in contact anteriorly, and the posterodorsal fold is large particularly medially (Figs 139-140). Males can be separated by the very short embolus and the MA which arises at 4 o'clock (Figs 141-142).

Selenops kalinago
Distribution. Known from the islands of Antigua, Guadeloupe, including Les Saintes, and Montserrat (Map 14). Etymology. The specific epithet comes from the indigenous word Souliga, for the island of St. Maarten, the type locality. It is to be treated as a noun in apposition.

Selenops souliga
Diagnosis. Females can be separated from all others by the median field of the epigynum, which is heart-shaped, and the huge spermathecae (Figs 147-148). Males can be differentiated from other species by the extremely dorsoventrally flattened palpus and the very small RTA (Figs 149-150).
Distribution. Found in the northern Lesser Antilles on the islands of Anguilla, St. Maarten, and Saba (Map 9). Diagnosis. Females can be separated from other species by the squarish lateral lobes, the sinuous opening which extends the width of the epigynal plate, and the internal genitalic structures; the sperm ducts and spermathecae are very small and simple (Figs 151-152). Males can be distinguished from other species by the c-shaped distal end of the conductor and the ventrally projecting large MA (Figs 153-154).
Natural history. Collected from under rocks in dry forests. Distribution. Known only from the Dominican Republic on the island of Hispaniola (Map 13). Muma, 1953 http://species-id.net/wiki/Selenops_pensilis Figs 155-156, Map 13
Remarks. This species is closely related to S. enriquillo sp. n. It differs in genitalic details and molecular genetic makeup. As the specimens from Alayón-García (1992) were unable to be examined, it is unclear if he was referring to S. pensilis or S. enriquillo sp. n.
Remarks. Although the male and female were not collected at the same time or from the exact same place, the overall similarity in habitus and the amount of collecting historically accomplished from the type locality indicate these are likely the same species. There is variation in the female copulatory organs, including how far posteriorly the lateral lobes extend beyond the epigastric furrow, how far apart or close together these extensions are, and the shape of the posterodorsal fold. Although one species was taken from Kingston, Jamaica, I do not believe this species naturally occurs there.
Distribution. Endemic to Hispaniola (Map 13). Diagnosis. Males can be separated from others by the very small RTA, with a quadrangular lateral apophysis and a rounded ventral apophysis, and the embolus which does not curve around the edge of the cymbium (Figs 161-162). The external female copulatory organs are similar to those of S. insularis, but the epigynal pockets are as long as the v-shaped genital opening, the internal copulatory organs are shaped differently, and there is no posterodorsal fold (Figs 163-164).

Selenops phaselus
Remarks. The leg formula given by Muma (1953) was 2341. My measurements give 2314. All of the leg lengths are very close, except for leg II, which is quite long. This is further evidence that leg lengths are probably not the best character to use to determine relationships in this genus. There is some variation in the female copulatory organs. In the majority of specimens, the longitudinal line separating the two lateral lobes is visible between the v-shaped genital opening and the epigynal pockets. In one specimen (sel_160), this line is not present, thus the lateral lobes are inconspicuous. There are no other differences noted in this specimen in genitalic or somatic characters. There is some variation in coloration in both males and females. Preserved specimens can range from a yellowish coloration to a darker more brownish-red coloration in both males and females. In life, the spider is a grey-green to blue-green, almost metallic, perfectly matching the wet, cloud forest lichens covering the trees and rocks on which it lives.

Selenops morro
Etymology. The specific epithet comes from the name of the type locality. It is to be treated as a noun in apposition.
Diagnosis. This small species is most similar to several endemic Cuban species, however females can be differentiated from all other spcies by a sinuous opening extending the width of the epigynal plate, and there appears to be epigynal pockets. These are separated and facing outward rather than facing one another as in other species (Figs 165-166). Males can be separated from other species by the presence of a unique conductor and embolus, both sinuous, the conductor arising behind another sclerite, extending beyond the edge of the bulb, and the embolus is s-shaped (Figs  167-168).

Selenops simius
Remarks. Muma (1953) reported the female to have the leg formula of 2314, however my measurements indicate 3142, once again attesting to the problems of using leg lengths to examine species relationships. This species is similar to its sister taxon, S. submaculosus, and can be difficult to distinguish. Muma (1953) distinguished males in the key, but not females. He mentioned that they were very similar, but that it was very easy to distinguish them based on structural differences, but did not suggest how to do so. Alayón-García suggested the two species can be distinguished by  Muma 1953), including those than cannot be identified to species (black circles) either because they are juveniles or they have diagnostic characteristics of more than one species (see text), S. actophilus Chamberlin (white circles), S. debilis Banks (white diamonds), and S. nesophilus Chamberlin (white stars).  Copulatory organs of Selenops geraldinae Corronca male from Gaspar Grande Island, Trinidad and Tobago (EME sel_224) 7-8 female from Gaspar Grande Island, Trinidad and Tobago (EME sel_225) 9-10 Selenops willinki Corronca male from Little Tobago, Trinidad and Tobago (EME sel_251) 11-12 female from Little Tobago, Trinidad and Tobago (EME sel_236) 13-14, 7, 11 male pedipalp, ventral view 8, 12 male pedipalp, retrolateral view 9, 13 epigyne, ventral view 10, 14 spermathecae, dorsal view. Scale bar = 0.45 mm (7)(8), 0.35 mm (9)(10)(13)(14), 0.50 mm (11-12).