A taxonomic review of the mygalomorph spider genus Linothele Karsch, 1879 (Araneae, Dipluridae)

ABSTRACT The neotropical diplurid spider genus Linothele Karsch, 1879 is reviewed. Three species of Linothele are newly described: Linothele septentrionalis n. sp. based on the absence of maxillary cuspules and presence of a sternal pattern; Linothele spinosa n. sp. based on the presence of preening-combs and its genital morphology; Linothele uniformis n. sp. based on its scopula, clypeus and genital morphology. Eight species names are newly synonymized: Linothele soricina (Simon, 1889) n. syn. is recognized a junior synonym of Linothele curvitarsis Karsch, 1879; Linothele bitaeniata (Mello-Leitão, 1941) n. syn. and Linothele nigerrima (Mello-Leitão, 1941) n. syn. are removed from the synonymy with Linothele aequatorialis (Ausserer, 1871) and instead considered junior synonyms of Linothele sericata (Karsch, 1879) together with Linothele megatheloides Paz & Raven, 1990 n. syn. Linothele longicauda (Ausserer, 1871) is recognized a senior synonym of Linothele aequatorialis (Ausserer, 1871) n. syn. and Linothele cousini (Simon, 1889) n. syn.; Linothele paulistana (Mello-Leitão, 1924) is recognized a senior synonym of Linothele annulifila (Mello-Leitão, 1937) n. syn.; Ischnothele caudata Ausserer, 1875 is recognized a senior synonym of Linothele dubia (Caporiacco, 1947) n. syn.; Linothele borgmeyeri (Mello-Leitão, 1924) is removed from the synonymy with Linothele gymnognatha (Bertkau, 1880) and considered a nomen dubium. Linothele gymnognatha and Linothele keithi (Chamberlin, 1916) are transferred back to Diplura C. L. Koch, 1850 and Brachythele Ausserer, 1871, respectively, due to original designation and considered nomina dubia. New distribution data and information on several species are presented.


INTRODUCTION
The genus Linothele Karsch, 1879 was proposed based on a single specimen of L. curvitarsis Karsch, 1879. Simon (1889a proposed the monotypic genus Uruchus for Uruchus gaujoni Simon, 1889. F. O. Pickard-Cambridge (1896 proposed the monotypic genus Neodiplura for Neodiplura jelskii F. O. Pickard-Cambridge, 1896, but the genus was found to be a junior synonym of Uruchus by Simon (1903). As most subsequent authors followed Simon and placed newly described species in Diplura, Linothele was monotypic until Strand (1908) described L. macrothelifera Strand, 1908. Pedroso et al. (2016 noted "[…] that the first description of a lyra in the maxilla of Dipluridae was made by Blackwall (1867), this structure was not mentioned again until the reevaluation of its form and function by Pocock (1896)." All species of Diplura, except for Diplura monticolens Chamberlin, 1916, described between the mention of the lyra by Pocock and until Raven discovered its significance for taxonomic understanding, were proposed by Mello-Leitão (1924, 1937, 1941a, b, 1945. Mello-Leitão placed newly described species in Diplura, until he proposed Evagrella Mello-Leitão, 1923 andTrechoninae Mello-Leitão, 1923, which were later found to be junior synonyms of Diplura and Diplurinae, respectively, by Raven (1985: 73, 74). Main (1969) described Troglodiplura Main, 1969 based on a fragmented specimen of T. lowryi Main, 1969 and placed it in Diplurinae. Bücherl et al. (1971) reexamined material described by Mello-Leitão and transferred some of the species to Uruchus. Raven (1980) initially considered Linothele a junior synonym of Diplura C. L. Koch, 1850following Simon (1903, who claimed Diplura does not bear a lyra on the prolateral maxillae. Raven (1985) found the type species of Diplura, Diplura macrura C. L. Koch, 1841, was in fact bearing a lyra. He therefore removed Linothele from the synonymy of Diplura and transferred all species of Diplura, except its type species, Diplura macrura, to Linothele. Raven (1985) synonymized those genera with a lyra consisting of a single row of bristles with Diplura and those with a lyra consisting of several rows of bristles with Trechona C. L. Koch, 1850. He synonymized the alyrate genus Uruchus with Linothele and tentatively transferred Troglodiplura to Nemesiidae Simon, 1889, so that of formerly ten genera included in the subfamily, only three remained. After Raven (1985), the subfamily Diplurinae, as well as the genus Linothele were better defined with variation in the lyra, including its absence, more clearly associated with generic boundaries.
As Raven (1985) defined Linothele only by the absence of a lyra, Goloboff (1994) stated the genus might well be a paraphyletic group. One year later, Goloboff (1995) provisionally transferred Brachythele keithi Chamberlin, 1916to Linothele. Main (1993 transferred Troglodiplura back to Diplurinae. Pedroso et al. (2008) transferred Trechona sericata Karsch, 1879 to Linothele. Based on molecular data, Harvey et al. (2020) transferred Troglodiplura to Anamidae Simon, 1889, which was formerly considered a nemesiid subfamily by Raven (1985), but granted family rank by Opatova et al. (2020). As a result, only the three genera originally placed in Diplurinae by Raven (1985) and Harmonicon F. O. Pickard-Cambridge, 1896, which was revalidated from the synonymy of Diplura by Maréchal & Marty (1998), remain in the subfamily. The redefinition of Linothele by Raven (1985) was then followed and several species newly described. Opatova et al. (2020) granted family rank to the former diplurid subfamilies Euagridae Raven, 1979 andIschnothelidae F. O. Pickard-Cambridge, 1897. As a result, only Diplurinae and Masteriinae remain in Dipluridae, further supporting ZOOSYSTEMA • 2021 • 43 (10) Goloboff's (1993) hypothesis that the family probably has to be restricted to those genera currently placed in Diplurinae. Although there might not be a closer relationship between the genera formerly assigned to Dipluridae and Linothele, many of the diagnostic features of Linothele can also be found in species of these families.
The genus Linothele has never been reviewed, or revised before and examined material of Linothele proved difficult to identify at the species level. The problem arose from a lack of up to date taxonomic information in the literature and the unavailability of type material. Knowledge of many species of Linothele rests on the publications of Mello-Leitão, or earlier authors, who worked at a time when modern key characters to distinguish species of Linothele were not practically used.
Most species of Linothele build extensive sheetwebs ending in a tubular retreat (Figs 1;2). The spiders fastly retreat upon disturbance, making them hard to collect. Due to their webbuilding habits, Linothele usually have to be collected by hand and specimens of the genus may not be found very often in museum collections. Yet, we were able to locate most of the relevant types and even some additional material. Resulting from our research, we here present a first comprehensive list of the species of Linothele with descriptions of three new species, a diagnostic key to the species, alongside new information and updated distribution data for many species. Some species, especially those from Brazil, are known only from the literature and in need of a revisit.

MATERIAL AND METHODS
Examinations of material were made using a stereo zoom microscope Leica MZ12.5. Illustrations of relevant structures were obtained as vectors from transparent layers above digital images photographed with varying quality and camera models in Adobe Illustrator. Dotted lines in illustration indicate broken structures. Carapace length was measured from its anterior margin to its posterior margin in a perpendicular line through the fovea. Pedipalp measurements are given as: total length (femur, patella, tibia, tarsus). Leg measurements are given as: total length (femur, patella, tibia, metatarsus, tarsus). Spinneret measurements are given as: total length (basal, medial, apical). All measurements are given in mm. Comparative measurements of male genitalia characters follow Coyle (1995) (also see Fig. 3). Male palpal organs have been recorded in prolateral and retrolateral view. Male megaspines and metatarsal protuberances have been recorded in retrolateral view, slightly turned to show the maximal elevation of the metatarsal protuberances. The spermathecae have been removed and cleaned in lactic acid according to von Wirth (2006). Where possible, the abbreviations proposed by Goloboff & Platnick (1987) were used to indicate relative positions. The term "scopula" refers to pseudo-scopula as defined by Pérez-Miles et al. (2017) on ventral anterior leg tarsi and metatarsi. Raven (1985) distinguished between sparse and dense scopulae in Linothele. We found what Raven (1985) referred to as "sparse scopula" is in fact scopula interspersed (divided) by lines of spiniform setae, whereas "dense scopula" is not interspersed with such setae (undivided). For easier references the terms divided (Fig. 5A,B) and undivided (Fig. 5C,D) are herein used instead of sparse and dense, respectively. The definition of a wide clypeus follows Gertsch & Platnick (1979) and is as follows: The clypeus is defined as the distance from the anterior margin of the eye tubercle to the anterior margin of the carapace; it is considered wide if it is at least "equal in width to long diameter of anterior lateral eye". The definition of maculae follows Decae et al. (2007) and is as follows: "[…] dark pigmented blotches […] on the external leg segments and/ or on the external basal segment of the PLS […]". Definition of vesicles follows . The term preening-combs refers to a field at ventrodistal posterior metatarsi that is densely covered with short spines (Fig. 11A, B). Some material examined for this study derived from lab reared specimens and is labelled accordingly: In examined material, the parental generation is marked with an asterisk (*) while following generations are labelled indicating from which generation they were taken (F1-2). For variability, only sexed specimens from the parental and F1 generation were considered. Unfortunately, the loan request on holotypes to NHRS was rejected by QCAZ for all species described by  and, additionally, it was not possible to arrange a loan on any of the other types. N. Dupérré provided images and confirmed a "weak" scopula with "[…] two parallel lines" for all described species (pers. comm.).

AbbreviAtions
As mentioned by Ríos-Tamayo & Goloboff (2012), it seems common for mygalomorphs with many spermathecal lobes, or processes to often show minor differences in the number or shape of lobes and processes of both sides. Presumably, lobes and processes with thin, weak ducts may easily get broken during moults, producing minor differences in successive moults of the adult female. Our research presented here suggests the same applies to vesicles.
Linothele soricina is recognized a junior synonym of L. curvitarsis on account of flexible apical segments of the PLS, the distinct pattern of immatures and females on opisthosoma, and its type locality which is partly consistent with the type locality of L. curvitarsis. Dupérré & Tapia (2021) designated a male lectotype and a single female paralectotype. Following ICZN 74.1.3 all specimens of the syntype collection, except for the lectotype, are to be considered paralectotypes. A juvenile (F1) in NHRS was examined: CL = 7.9. CT = 9-10. MC = 18-20. As this specimen is clearly conspecific to other specimens in NHRS, we were able to observe the maximum range of variability for CT (± 5). The immature holotype of L. curvitarsus was smaller CL (5.5), but had more CT (11) and MC (30-32).
nAturAl History According to D. Reimann and B. Striffler (pers. comm.) the spiders settle in coastal forests of Venezuela, where they can be found under stones or between buttress roots. The species seems to be synanthropic. It takes about one year for males and one and a half years for females to reach maturity. Males mature from July to October. Usually they produce less extensive, but more three-dimensional webs than other species of the genus. The tubular retreat, where the spider stays during the day, ends in a funnel-web which is approximately 30-40 cm in diameter.

remArks
According to the first description (Mello-Leitão 1926), the type locality is "Alto Juruá". Bücherl et al. (1971) and Silva-Moreira et al. (2010) referred to the type locality as Juruá, Amazonas, ignoring the "alto" part, which might actually refer to the "upper" Jurua river at Peru and Acre, Brazil; thus, the type locality is somewhat ambiguous. The holotype could not be located by Silva-Moreira et al. (2010).
Material from Bolivia was found to match the descriptions by Mello-Leitão (1926) and Bücherl et al. (1971), as well as the illustration of the spermatheca provided by the latter.
nAturAl History Linothele fallax can be found in natural crevices near ground level, but also in burrows in the ground. The spiders seem not to burrow, but occupy existing crevices. They usually produce less extensive funnel-webs, which end in a short funnel at the entrance of their burrow. Females produce an egg-sac with up to 120 eggs as a fixed hammock, usually attached to the entrance funnel or shortly behind in the tubular retreat. Unfortunately, we lack information on the time of the year the spiders mature and produce offspring in the wild. Under artificial conditions females started to build their egg-sacs after the humidity has been raised, indicating that mating and oviposition take place at the start of the wet season.

remArk
The vial sent to us from MNHN contained two females. Hence, we cannot comment on the male.

Female
Unknown.
remArks F. O. Pickard-Cambridge (1896) stated a leg formula of 4123 in the genus description of Neodiplura, but shows a leg formula of 1423 for the adult male syntype, which we hereby confirm. The tip of the only embolus still attached to the adult male type is damaged (Fig. 9A, B). The immature syntype of N. jelskii bears undivided scopula and has a narrow clypeus as the adult male syntype. A loose opisthosoma ZOOSYSTEMA • 2021 • 43 (10) can be found in the vial (Fig. 18G), whereas another one has been pinned to the prosoma of the adult male, both showing a distinct dorsal pattern consisting of chevrons. Bücherl et al. (1971) mentioned that they examined two females from Peru (without specified locality) of 'Uruchus costatus' from a vial labelled by Mello-Leitão. They stated the species has not been mentioned in any of Mello-Leitão's works and consider the specimens conspecific with Uruchus jelskii. The specimens of U. costatus are supposed to have a wide clypeus (as wide as the "diameter of an anterior lateral eye"; Bücherl et al. [1971]), clearly distinguishing them from both examined specimens of N. jelskii. Hence, we doubt Bücherl et al. examined the types of N. jelskii and that the specimens they examined were truly conspecific with N. jelskii. We can verify the species has never been published, but also that no such material could be located in MNRJ collection by Silva-Moreira et al. (2010). From the short description alone, the specimens of U. costatus could be distinguished from Linothele gaujoni only by their type locality. We are confident the short notes Bücherl et al. (1971) provided on U. costatus were never intended to serve as a first description. We therefore do not consider the species to be formally described and therefore a nomen nudum.
nAturAl History Unknown.
The medial and apical segments of the PLS of the holotype of Diplura aequatorialis are missing. The epigastrium has been dissected by an earlier examiner, but no preparation can be found in the vial of the holotype. Bücherl et al. (1971: 117) Bücherl et al. (1971: 117) are rejected for the following reasons: apparently, the apical segments of the PLS in D. nigerrima and D. bitaeniata were very elongated. Unfortunately, Mello-Leitão (1941a) did not explicitly state on the structure of the apical PLS segment in any of the two species, but in no other species, except for L. sericata, the apical PLS segment is as elongated as mentioned for D. nigerrima; PLS: 23.4 (5.6, 5.6, 12.2). Sizes (22.6 for D. bitaeniata, 37.0 for D. nigerrima; both probably incl. chelicerae) and CT (9 for D. bitaeniata, 8 for D. nigerrima) might well fall within the range observed for Linothele sericata. Additionally, the dorsal pattern on the opisthosoma, which Mello-Leitão (1941a) described for D. bitaeniata and Paz & Raven (1990) described for L. megatheloides, resembles that of younger L. sericata, whereas the holotype of D. aequatorialis bears no distinct pattern. Furthermore, the type localities (both Bogotá) of Diplura bitaeniata and D. nigerrima are consistent with the one of L. sericata and notably disjunct from the rather unspecific type locality of D. aequatorialis. As a result, D. bitaeniata n. syn. and D. nigerrima n. syn. are removed from the synonymy of L. aequatorialis and instead considered junior synonyms of L. sericata. The holotype of D. aequatorialis matches variation and overall appearance observed in the types of D. longicauda and D. cousini. The holotype of D. longicauda is indistinguishable from the types of D. cousini and D. aequatorialis. As a result, L. aequatorialis n. syn. and L. cousini n. syn. are considered junior synonyms of L. longicauda.
nAturAl History Unknown. Strand, 1908 (Figs 11;17F-H)  diAgnosis. -Females of Linothele macrothelifera differ from those of most other species of Linothele by the presence of preening-combs (Fig. 11A, B). Females may further be distinguished from those of L. spinosa n. sp. by their shorter spermathecae stalks lacking an elongated vesicle (Fig. 11C, D) and their distribution.

Female
Colouration in alcohol: Prosoma, chelicerae, legs and pedipalps pale, yellow; opisthosoma with distinct patterns, mid-dorsally consisting of quadrate spots anteriorly, which become more rectangular posteriorly, or can be interconnected, forming longitudinal lines, laterally with several spots, ventrally with longitudinal lines and spots (Fig. 17F-H); maculae absent. Clypeus: narrow. Leg formula: 4123. Scopula divided. Preening-combs present, see Figure 11A, B. Leg tarsi pseudo-segmented. Spinnerets: apical segments of the PLS rigid. Spermathecae: consisting of two stalks with broad bases, bearing several vesicles at 1:2-1:3A and a short globular apex, see Figure 11C, D. Linothele melloleitaoi (Brignoli, 1983) Diplura maculata Mello-Leitão, 1941b: 236. -Brignoli 1983: 124 (preoccupied by Thorell, 1890.  mention Linothele melloleitaoi and refer to Diplura melloleitaoi (Brignoli, 1983) with no further mention. It remains unclear if the type was deposited in MNRJ collection. Unfortunately, we have to assume the holotype to be lost. According to the first description, maxillary cuspules were present. A single fragmented male of a Linothele species from Northern Colombia (MCZ) has maculae and a palpal organ of similar shape as the one in males of L. fallax, which certainly differs from the one in L. quori. Without having examined more material from Colombia, it can currently not be assured that the examined male (MCZ) is the conspecific male of L. melloleitaoi. However, this suggests there is another species of Linothele with maculae in northern Colombia that differs from L. quori.
Unfortunately, MCZ did not allow dissections on type material and we therefore cannot comment on the female genitalia. The sex of the holotype is referred to by original designation. Legs are partly disarticulated from prosoma, but on no leg an undivided scopula, or preening-combs could be observed.
desCription Male Unknown.
Female CT = 7(?)-14. MC = 0. Colouration in alcohol: prosoma, chelicerae, legs and pedipalps brown; opisthosoma dorsally and ventrally without distinct patterns; maculae absent. Clypeus: narrow. Sternum, labium and maxillae: see Figure 12B. Spinnerets: apical segments of the PLS flexible, see Figure 10A. Spermatheca: see Figure 12C-E. Before the type material of D. annulifila was lost in the fire of 2010, A. Brescovit (IBSP) provided images of the holotype (all lacking scales) of the holotype of D. annulifila. The preparation of the spermatheca was broken and the holotype clearly lacked maxillary cuspules. All legs were disarticulated with some leg tarsi appearing less flexible, with only few medial cracks. As the images were not highly resolved, many characters have to be reconsidered. Mello-Leitão (1937) reported a length of 9.0 for the cephalothorax of D. annulifila. D. annulifila n. syn. is recognized a junior synonym of D. paulistana on account of the absence of maxillary cuspules and patterns, but presence of flexible apical segments of the PLS, alongside their close type localities.   nAturAl History According to , specimens "[…] of L. pukachumpi were collected in a primary cloud forest at 2225m, they build webs of 50 × 40 cm on dirt talus along trails, the retreat is located in soil crevices." Linothele quori Dupérré   Female CL = 6.0. CT = 9. MC = ~25. Colouration alive: as for male, see Dupérré & Tapia (2015: figs 41, 47). Leg formula: 4123. Scopula divided. Leg tarsi weakly pseudo-segmented. Spermathecae: "joined at base, wide and short, with small vesicles medially, with or without stalks", see Dupérré & Tapia (2015: figs 42, 43).
remArks Unfortunately, we were unable to obtain images of the relevant structures, but the illustrations for male characters in the first description show a megaspine and MP at prolateral tibia and metatarsus I. The description mentions those to be situated retrolaterally. As a prolateral megaspine and MP have not been observed in any other Linothele, it is likely the terms "prolateral" and "retrolateral" have been swapped in the illustration legends.
nAturAl History According to , specimens of L. distribution. -Only known from the type locality.
diAgnosis. -The female of Linothele septentrionalis n. sp. differs from those of all other Linothele by the combined presence of maculae (Fig. 13B), presence of a pattern on the sternum and absence of maxillary cuspules (Fig. 13A).
desCription Male Unknown.

remArks
The holotype of Trechona sericata is a dried and pinned specimen, which seems to be in fragile condition. For this reason, photographic examination by the curators is clearly to be preferred and the ZMB team was so kind to send us images of the holotype, allowing for a proper diagnosis. As stated by Pedroso et al. (2008), we confirm the specimen described as Trechona sericata by Karsch (1879) is a misidentified Linothele. Karsch was not able to observe a prolateral lyra on the maxillae, since the specimen was already dried when he examined it (Karsch 1879: 544). The type specimen shows maxillary cuspules, two rows of teeth on paired tarsal claws, divided scopula, and flexible apical segment of the PLS; a combination of features that can only be found in Linothele.
The type material of Linothele megatheloides has not been studied and identification is based on the very detailed description by Paz & Raven (1990), alongside examination of the additional material (AMNH_IZC 00327625) mentioned in the first description. L. megatheloides n. syn. is recognized a junior synonym of L. sericata on account of the presence of maxillary cuspules, flexible apical segment of the PLS (at least twice the length of the medial segment), the homogeneous colouration described by Karsch (1879), as well as the type locality close to the one of L. sericata. For information on other synonymies established here refer to remarks at L. longicauda.
As observed in lab-reared specimens, younger living females, immature males of L. sericata and even early instars bear distinct patterns on dorsal (see Paz & Raven 1990: fig. 9) and ventral opisthosoma. In older females, patterns can become indistinct and the opisthosoma of the spider appears almost black. Early instar specimens of Linothele sericata (CL = 2.3-2.7) bear fewer cheliceral teeth (CT = 7-9), fewer maxillary cuspules  and only few scopuliform setae on anterior leg tarsi.
nAturAl History According to Paz (1988), L. sericata build large sheet webs ending in a tubular retreat at the base of tree trunks, in which several different symbionts and cleptoparasites may be found. The reproduction strategies of the species have been well documented by Paz (1993). (Schiapelli & Gerschman, 1945) (Fig. 14) Diplura sexfasciata Schiapelli & Gerschman, 1945: 177, pl distribution. -Only known from the type locality.

Linothele sexfasciata
diAgnosis. -The female of Linothele sexfasciata resembles L. monticolens, but can be distinguished from it by its pseudo-segmented leg tarsi, alongside its distribution.  Figure 14A.

remArks
The holotype is partly disarticulated with no leg, nor pedipalp still attached to the prosoma. Fortunately, one leg II has been dissected including coxa with other coxae still attached to the prosoma. We therefore could evaluate the scopula on tarsus II, which is divided. On no other leg an undivided scopula or preening-combs could be observed. Schiapelli & Gerschman (1945: Pl. VI) show a distinct dorsal pattern consisting of chevrons, which is still evident in the holotype.  (Fig. 15C, D), as well as their proportionally long [(PL*100)/BD = 385] embolus bearing no keels (Fig. 15A, B). Females differ from those of most other species of Linothele by the presence of preening-combs and can further be distinguished from those of L. macrothelifera by their longer spermathecae stalks bearing an elongated vesicle (Fig. 15G) and their distribution.

remArks
The male holotype has been preserved immediately after the moult to adulthood which might have caused its very light colouration.
nAturAl History Unknown.
nAturAl History According to  specimens of L. tsachilas " […] were collected in a foothill forest at 884m elevation. This species builds webs approximately 50 × 30 cm near the ground, and the retreat is underneath or in a hole of a dead tree trunk. The two specimens collected live in sympatry with L. zaia and L. quori, but seem to prefer distinct microhabitats." Linothele uniformis n. sp. (Fig. 16)  distribution. -Only known from the type locality.
diAgnosis. -The female of Linothele uniformis n. sp. can be distinguished from those of most other species of Linothele by its undivided scopula. It resembles females of L. fallax, but differs from them by the absence of patterns on the opisthosoma, a significantly lower number of maxillary cuspules (Fig. 16A) and its spermatheca stalks not bearing a single retrolateral lobe (Fig. 16D).
desCription Male Unknown.
nAturAl History Unknown.
nAturAl History According to  the "[…] holotype female was collected in a foothill forest at 884m elevation.  (10) The web of approximately 60 × 40 cm was along a vertical dirt talus or landslide 6m from the ground, and the retreat was concealed in a soil crevice. The only specimen collected lives in sympatry with L. zaia and L. quori, but was found in a somewhat distinct microhabitat."
Mello-Leitão described the species twice (Mello-Leitão 1924, once as Diplura borgmeyeri and later as Diplura borgmeieri. In both cases, Mello-Leitão considered the specific epithet a patronym of the collector, "Thomaz Borgmeyer", but apparently misspelled the family name. Thomas Borg-meier was a German priest and entomologist who lived at Petropolis (the type locality mentioned in both descriptions) and joined the National Museum at Rio de Janeiro in 1923, where Mello-Leitão worked at that time. The German family name suffix "-meyer" is very common, even as a standalone name and comes in various spellings: The "ey" may be spelled "ei", "ay", or "ai"; all pronounced the same. Despite the incorrect spelling of the collector's family name, the description by Mello-Leitão (1926) clearly indicates the epithet to be a patronym dedicated to "Borgmeyer". Considering this, there is clear evidence of an inadvertent error in the original publication (without recourse to any external source of information) and the epithet in the second description has to be corrected to "borgmeyeri" (following ICZN 32.5.1), as correctly noted by Bonnet (1956) and followed by Silva-Morreira et al. (2010), who considered Mello-Leitão's (1926) description to be redundant. Mello-Leitão (1924) states the size of the specimen was 25.0, whereas Mello-Leitão (1926) mentions a size of 18.0 (opisthosoma: 8.0. spinnerets: 7.0). Both descriptions state the number of cheliceral teeth to be 8; (Mello-Leitão 1924. In the first description of D. borgmeyeri, Mello-Leitão (1924) mentioned a dorsal chevron pattern and maculae, whereas in the second description (Mello-Leitão 1926) he mentioned the legs to be uniformly colored, but reports a dorsal pattern consisting of bright spots and the presence of a ventral pattern on the opisthosoma. Later, a key is provided in which Mello-Leitão (1926) stated the PLS to be short. None of the descriptions mentioned the presence, or absence of maxillary cuspules. Bücherl et al. (1971) apparently did not follow Bonnet (1956) and used the uncorrected spelling of "D. borgmeieri". They considered Diplura borgmeyeri a junior synonym of D. gymnognatha based on the spermatheca and their close type localities. The authors clearly stated the examined spermathecae to be rudimentary, coming from a juvenile specimen. Yet, they considered it to be similar to the illustration provided by Bertkau (1880). We are confident the illustration provided by Bertkau lacks detail making it hard to compare to the spermathecae examined by Bücherl et al. (1971), who, unfortunately, did not illustrate it. Mello-Leitão (1924 described only a single measured specimen in both descriptions. Contrary, Bücherl et al. (1971) mentioned a female holotype and two female syntypes, while Silva-Moreira et al. (2010) mentioned 3 female and 1 male syntypes, of which only the male syntype could be located at the museum collection at that time. Mello-Leitão is known to have added material to types after describing a species, or switch labels (Silva-Moreira et al. 2010). In the case of D. borgmeyeri only a single male from the supposed syntype series could be located, while neither descriptions by the first describer (Mello-Leitão 1924, nor the following work on this species (Bücherl et al. 1971) mention a male to be part of the type series. It is highly unlikely that consecutive authors failed to identify a male as a part of a syntype series. The only logical conclusion is that the male in MNRJ collection (MNRJ 1592) has not been added by Mello-Leitão ( †1948) himself. Consequently, this male was not part of the original type series and its name-bearing status cannot be confirmed.
As all types are probably lost and descriptions on the species are contradictory, Linothele borgmeyeri is herein considered a nomen dubium.  (2003), most of the arachnids collected during the Belgian mission to Brazil and described by Bertkau (1880) were deposited at IRSNB. Unfortunately, the holotype could not be located in IRSNB collection (L. Baert, pers. comm.) and might have been lost during WWII. Bücherl et al. (1971) synonymized D. borgmeyeri with D. gymnognatha and provided variation data for D. borgmeyeri . They further added measurements for the opisthosoma (10.5) and a PLS: 8.9 (2.5, 2.6, 3.8). It becomes clear that these measurements do not match the ones provided for D. gymnognatha by Bertkau (1880), who reported the spinnerets (6.0) to be a lot shorter than the opisthosoma (10.0). Bücherl et al. (1971) stated they examined the epigastrium of 3 juvenile females, of which two lacked a spermatheca (= imm. male) and the only spermatheca found was rudimentary. Yet, they considered it to equal the illustration of the spermatheca of D. gymnognatha provided by Bertkau (1880). Contradictory to Bertkau (1880), who mentioned the presence of 20 maxillary cuspules, Bücherl et al. (1971) noted the absence of maxillary cuspules in D. borgmeyeri. Considering the possible lapsus in the first description of D. gymnognatha, the scopula on anterior tarsi and metatarsi was dense and possibly undivided. The fact that Bücherl et al. (1971) did not transfer D. borgmeyeri to Uruchus, as they did with D. fallax, indicates the scopula on all tarsi of D. borgmeyeri was divided.
As the holotype of D. gymnognatha is presumably lost since WWII, Bücherl et al. certainly did not examine it and the specimens they examined were not conspecific. As a result, the synonymy established by Bücherl et al. (1971) is herein rejected.
Bertkau (1880) neither mentioned the presence, or absence of a maxillary lyra for D. gymnognatha. In the same work, Bertkau (1880) proposed Thalerothele for T. fasciata Bertkau, 1880 without mentioning a lyra. Thalerothele was found to bear a lyra by Simon (1903), clearly indicating Berkau did not check for the presence of such a structure. Considering the possible lapsus, the description might match either Linothele, or Trechona.
As the absence of a lyra cannot be confirmed by original designation and the type is lost, the species cannot be placed in any diplurine genus with certainty. As a result, Linothele gymnognatha is transferred back to Diplura due to original designation and Diplura gymnognatha comb. rev. is considered a nomen dubium. remArks Chamberlin (1916) proposed B. keithi based on a single specimen. He noted the labium to be much wider than long for B. keithi and Diplura monticolens. In the latter, we observed a more subquadrate labium. According to Chamberlin (1916) legs II and IV are missing, but tarsal scopulae are "dense, extending to base; none divided by a setose line or band." Goloboff (1995) mentioned the holotype to be in extremely poor condition ("reduced to a series of loose fragments") and tentatively transferred the species to Linothele, due to the "loose fragments of spinnerets accompanying the specimen, which appear to have corresponded to long spin- nerets." Chamberlin (1916) provided measurements for the cephalothorax and the spinnerets, of which he reported "the three joints subequal in length". As the "cephalothorax" usually includes the chelicerae and the opisthosoma may vary in size, we have to agree the spinnerets seem to have been elongated. Chamberlin (1916) explicitly mentioned "conspicuously curved" tarsi for Diplura monticolens, but not so for Brachythele keithi. Goloboff (1995) further mentioned only 10 maxillary cuspules and "the epigastrium dissected and at least some parts of it placed in a separate microvial, but with no detectable spermathecae". The largest intact part we were able to find in the vial of the holotype was a single maxilla with all cuspules broken off and no lyra visible. As none of the spinnerets, or leg tarsi was still intact, the transfer to Linothele by Goloboff (1995) cannot be confirmed with certainty. Despite its incorrect placement in Brachythele, the type is too fragmented to allow for a certain placement. As a result, Linothele keithi is transferred back to Brachythele due to original designation and Brachythele keithi comb. rev. is considered a nomen dubium.