A world checklist of Onychophora (velvet worms), with notes on nomenclature and status of names

Abstract Currently, the number of valid species of Onychophora is uncertain. To facilitate taxonomic work on this understudied animal group, we present an updated checklist for the two extant onychophoran subgroups, Peripatidae and Peripatopsidae, along with an assessment of the status of each species. According to our study, 82 species of Peripatidae and 115 species of Peripatopsidae have been described thus far. However, among these 197 species, 20 are nomina dubia due to major taxonomic inconsistencies. Apart from nomina dubia, many of the valid species also require revision, in particular representatives of Paraperipatus within the Peripatopsidae, and nearly all species of Peripatidae. In addition to extant representatives, the record of unambiguous fossils includes three species with uncertain relationship to the extant taxa. For all species, we provide a list of synonyms, information on types and type localities, as well as remarks on taxonomic and nomenclatural problems and misspellings. According to recent evidence of high endemism and cryptic speciation among the Peripatidae and Peripatopsidae, previous synonyms are revised. Putative mutations, subspecies and variations are either raised to the species status or synonymised with corresponding taxa. In our revised checklist, we follow the rules and recommendations of the International Code of Zoological Nomenclature to clarify previous inconsistencies.


introduction
Although onychophorans play an important role in studies of animal evolution, biogeography and conservation (Sedgwick 1908;Brues 1923;Brinck 1957;Ruhberg 1985;Monge-Nájera 1995;New 1995;, the taxonomy and species diversity of this taxon are understudied. According to a recent estimate, a total of 180 extant onychophoran species have been described since the first description in 1826, including 73 species of Peripatidae and 107 species of Peripatopsidae ( Fig. 1; Mayer and Oliveira 2011). However, the validity of many of these species is uncertain. Although a revision at the species level seems timely, it represents a challenge because several original descriptions and revisions are difficult to access or they have been published in different languages, including Latin (Guilding 1826), Russian (Sänger 1871), Dutch (Weber 1898), French (Bouvier 1905), Italian (Clark 1913a), Spanish (Scorza 1953), German (Ruhberg 1985), Portuguese (Oliveira and Wieloch 2005) and English (e.g., Oliveira et al. 2011).
In this study, we compiled a checklist of all described species of Onychophora and re-evaluated their species status following the International Code of Zoological Nomenclature (ICZN). Our checklist summarizes information on valid species names, Figure 1. Diagram illustrating the number of species descriptions of Onychophora per year since the very first description of an onychophoran species by Guilding (1826). Note the numerous gaps in the taxonomical work. synonyms, description language, type locality, type designation and location of holotypes. Additional remarks highlight aspects relevant for future work, which will help improve our knowledge of the taxonomy and species diversity of Onychophora.

Methods
The checklist of all described species of Onychophora was compiled by gathering, translating and interpreting information from the literature and from museums' databases (Natural History Museum of London -NHM, United Kingdom, and Muséum National d'Histoire Naturelle of Paris, France). Information on cryptic speciation and point endemism, common among species of Peripatidae and Peripatopsidae (Trewick 1998;Daniels et al. 2009;Oliveira et al. 2011), as well as precise collecting data were used to assess the validity status of species. Previous synonyms were critically evaluated and ambiguous synonyms reconsidered. Putative mutations and variations, based on characters known to be intra-specifically variable, were synonymised, if they were reported from the species type locality. Putative subspecies and variations with precise collecting data, which occur far from the type locality (at least 30 km), were raised to the species status. This approach is justifiable, given high cryptic diversity and endemism among onychophorans, including representatives of both Peripatidae and Peripatopsidae. The data available suggest that most clades, which are regarded as species or cryptic species, are found at localities lying at least 10-30 km apart Lacorte et al. 2011;Oliveira et al. 2011). We used this information as a guideline for assessing the validity of the species, i.e., if two putative clades occur at a distance of over 30 km from each other, they are likely to be separate species.
In the list, valid species names are sorted in alphabetical order and numbered consecutively within Peripatidae and Peripatopsidae. Fossil species are numbered separately at the end of the list. Nomina dubia are left unnumbered and listed after the valid names of each genus. No abbreviations are used for genus names in order to avoid confusion among species with a similar epithet. Each taxon/species name is accompanied by the corresponding author and year. Synonyms are arranged in chronological order and only the first reference mentioning a particular synonym is cited. Misspellings were not regarded as synonyms but are discussed in the Remarks section for each species. Only the data on the holotype are considered since it represents a single specimen, while syntypes may originate from different localities and, thus, they might represent different species. Only the type locality data are provided for each species rather than its putative range of distribution. Old locality names were updated and country names are provided in capitals according to their current political borders. The International System of Units (SI) is used throughout the list. The original information found in the literature is provided along with the converted units (1ft = 0.3048 m; 1 mile = 1.6 km). All available data on the type localities of valid species (Fig. 2) as well as on the localities of species regarded as nomina dubia herein (Fig. 3) are included into a world map based on the information obtained using the freeware Google Earth®. Checklist ONYCHOPHORA Grube, 1853 I. PERIPATIDAE Evans, 1901a Type genus: Peripatus Guilding, 1826 Remark: A thorough revision of the group, particularly neotropical genera, is required (Peck 1975: 343).

Eoperipatus Evans, 1901a
Type species: Eoperipatus horsti (Evans, 1901a), designated herein (see Remarks). Remarks: So far, no type species has been designated for the genus. Taking into account the recommendations of the ICZN (Art. 67) and the amount of information available in the literature, we designate Eoperipatus horsti as the type species of the genus since it is the only originally included nominal species for which data from both sexes are available.

Eoperipatus horsti
3. Eoperipatus weldoni Evans, 1901a Synonyms: None. Holotype: Not designated. Type locality: MALAySIA, Bukit Besar, "on the boundary line between the States of Nawng Chick and Jalor, a full day's journey from the town of Patani", 686 m (2,250 ft) (see Remarks). Language of species description: English. Remarks: The name Bukit Besar is currently attributed to different localities in Malaysia but evidence indicates that the putative type locality might be within the limits of the Bukit Perangin Forest Reserve, northern Malaysia, next to the border with the southern part of Thailand. An area referred to as Bukit Besar, located in the western part of the Forest Reserve, fits with the information on the altitude (690 m) and the distance from Patani (104 km from Pattani, South Thailand) provided by Evans (1901a). Moreover, the Bukit Perangin Forest Reserve is situated next to the border of the province yala (also known as Jala or Jolor), which might have been spelt as Jalor by the author. The current name and position of the state of Nawng Chick could not be found. Kloss (1926) stated that in contrast to Eoperipatus horsti, Eoperipatus weldoni is found in mountainous habitats. Van der Lande and Holthuis (1986: 18) discuss the possibility that the species is a variation of Eoperipatus sumatranus (van der Lande and Holthuis 1986: 18), and later, all Malaysian species were treated under the name Eoperipatus sumatranus (van der Lande 1988: 13), which we regard as a nomen dubium (see Remarks for Eoperipatus sumatranus). Thus, Eoperipatus weldoni is most likely a separate species, although it requires revision.

Nomen dubium
Eoperipatus sumatranus (Sedgwick, 1888) Synonyms: Peripatus sumatranus, by original designation (Sedgwick 1888: 485); Eoperipatus sumatranus (Evans 1901a: 484). Holotype: Not designated (see Remarks). Type locality: Unknown. Van der Lande and Holthuis (1986: 19) assumed it might be Mount Arjuno in East Java rather than a locality in Sumatra. However, apart from the doubtful record of Eoperipatus sumatranus, no onychophorans have been reported from Java or Sumatra thus far (see Remarks). Language of species description: English. Remarks: Holotype not clearly designated in the original description. The type locality of this species is unlikely to be Sumatra since the collector might have never been on this island (van der Lande and Holthuis 1986). The suggestion of Mount Arjuno in East Java as a putative type locality (van der Lande and Holthuis 1986) is in our view a mere guess. Therefore, we regard Eoperipatus sumatranus as a nomen dubium since neither the holotype nor the type locality is known, which makes a revision difficult.
9. Epiperipatus biolleyi (Bouvier, 1902a) Synonyms: Peripatus biolleyi, by original designation (Bouvier 1902a: 258); Epiperipatus biolleyi (Peck 1975: 345 (Peck 1975: 345 Carter", but it is uncertain whether "Santarem" represents a species name or the locality name on the label. Due to this uncertainty and due to the long usage of the name "brasiliensis", we favour the latter name (ICZN Art. 11.6.1). For the sake of stability, we therefore consider Peripatus santarem as a nomen oblitum and Peripatus brasiliensis as a nomen protectum, following the ICZN (Art. 23). The holotype has not been clearly designated in the original description. According to Bouvier (1905: 270), the first specimens of this species collected by W.H.J. Carter and placed in the Natural History Museum of London are not type specimens as the author used other specimens from the same collection for species description. The species name is commonly misspelt as braziliensis (e.g., Arnett 1947;Eakin and Brandenburger 1966). We consider Epiperipatus brasiliensis as a separate and valid species due to the great distance between the type localities of the two putative subspecies (brasiliensis and vagans). Requires revision.
According to Bouvier (1907a: 519), a type has been deposited in the Museum National d'Histoire Naturelle de Paris, France. The description contains imprecise type locality data. A redescription of this species (Bouvier 1905) (Oliveira & Wieloch, 2005) Synonyms: Macroperipatus machadoi, by original designation (Oliveira and Wieloch 2005: 61); Epiperipatus machadoi (Oliveira et al. 2010: 25 (Lacorte et al. 2010) is invalid according to the ICZN since it was suggested without providing a formal description (Art. 13) and without depositing type specimens (Art. 16). Moreover, according to the ICZN (Art. 9.9), the proposition of this name in a proceedings abstract cannot be regarded as a valid publication for purposes of zoological nomenclature. Epiperipatus paurognostus is very similar morphologically to Epiperipatus adenocryptus. The species diagnosis contains morphological and molecular characters ). (Bouvier, 1899b) Synonyms: Peripatus simoni, by original designation (Bouvier 1899b: 271); Peripatus (Epiperipatus) simoni (Clark 1913b: 18); Epiperipatus simoni (Peck 1975: 346). Holotype: Not designated (see Remarks). Type locality: VENEZUELA, Caracas (see Remarks). Language of species description: French. Remarks: The holotype has not been clearly designated in the original description.

Nomina dubia
Epiperipatus nicaraguensis (Bouvier, 1900a) Synonyms: Peripatus nicaraguensis, by original designation (Bouvier 1900a: 395); Peripatus (Epiperipatus) nicaraguensis (Clark 1913b: 18); Epiperipatus nicaraguensis (Peck 1975: 346). Holotype: Not designated (see Remarks). Type locality: NICARAGUA, Matagalpa (see Remarks). Language of species description: French. Remarks: The holotype has not been clearly designated in the original description. Röhlig et al. (2010: 227) refer to a holotype placed in the Museum für Naturkunde Berlin, which should be regarded as a syntype instead. The description contains imprecise type locality data (Matagalpa in Nicaragua occupies an area of 8,523 km 2 ). Bouvier (1905: 327) stated that a specimen collected by Belt (1874: 140) might belong to this species. However, this record is doubtful since the author might have misinterpreted the information provided by Belt (1874: 140), who referred to San Benito Mine located in Santo Domingo rather than to San Benito located in San Antonio Valley (see Bouvier 1905: 328). Revision of this species will be difficult since no precise locality data are available in the literature.
Epiperipatus tucupi (Froehlich, 1968) Synonyms: Peripatus (Epiperipatus) tucupi, by original designation (Froehlich 1968: 168); Epiperipatus tucupi (Peck 1975: 346). Holotype: Deposited in the Museu de Zoologia da Universidade de São Paulo, São Paulo, Brazil. Type locality: BRAZIL, Pará (see Remarks). Language of species description: English. Remarks: The description contains imprecise type locality data (the Pará State of Brazil occupies an area of 1,247,689.515 km 2 within Amazonia). Since no further work with more precise locality data is available, a revision of this species based on topotypes will be difficult.
Type locality: PANAMA, Azuero Peninsula, province of Veragua, north base of the ridge supporting Piedra del Tigre, near western border of Veragua, "two days south of Las Minas" (according to Dunn 1943), 792 m (2,600 ft). Language of species description: English. Remark: Note that the name clarki has also been used by Arnett (1961) for Macroperipatus insularis clarki (see Synonyms for Macroperipatus clarki below). The species might have been included in Heteroperipatus based on an ambiguous character and, therefore, requires revision. Zilch, 1954a Synonyms: None. Holotype: Likely to have been deposited in the Senckenberg Research Institute and Natural History Museum, Frankfurt, Germany, as the author was employed at this institution at that time and used the acronym SMF in his description (Zilch 1954a: 150 (Clark 1913b: 17). Remarks: As for Epiperipatus. Most species might have been assigned to Macroperipatus based on a fixation artefact (Oliveira et al. 2010: 31) and Macroperipatus torquatus might be the only species belonging to the (monotypic) genus. The entire genus requires revision . Arnett, 1961 Synonyms: Macroperipatus insularis clarki, by original designation (Arnett 1961: 215). Holotype: Deposited in the Science Museum of the Institute of Jamaica, Kingston, Jamaica. Type locality: JAMAICA, Portland, John Crow Mountains, ca. 8 km (5 miles) southwest of the Priestman's River, 457 m (ca. 1,500 ft). Language of species description: English. Remarks: Note that Dunn (1943) used the name clarki for Peripatus clarki (see Synonyms of Heteroperipatus clarki above). We regard Macroperipatus clarki and Macroperipatus insularis as separate species rather than subspecies due to the great distance between their type localities (430 km) on different islands. Revision is required as it might reveal morphological and molecular differences between Macroperipatus clarki and Macroperipatus insularis. (Evans, 1903) Synonyms: Peripatus guianensis, by original designation (Evans 1903: 145); Peripatus ohausi var. guianensis (Bouvier 1904a: 53); Peripatus (Macroperipatus) guianensis (Clark 1913b: 17); Macroperipatus guianensis (Peck 1975: 346). Holotype: Not designated (see Remarks). Type locality: BRITISH GUIANA, Demerara-Haimaca, eastern bank of the river Demerara. Language of species description: English. Remarks: The holotype has not been designated explicitly, but the author used the terms "male type specimen" and "female type specimen" in his figure legends (Evans 1903: 159-160). Requires revision. Clark, 1937 Synonyms: Macroperipatus insularis insularis (Peck 1975: 347). Holotype: Deposited in the Smithsonian National Museum of Natural Science, Washington D.C., USA. Type locality: HAITI, between Jacmel and Tronin (see Remarks). Language of species description: English. Remarks: The current name and geographical position of Tronin could not be found. We regard Macroperipatus insularis and Macroperipatus clarki as separate species rather than subspecies due to the great distance between their type localities (430 km) situated on different islands. Revision is required as it might reveal morphological and molecular differences between Macroperipatus clarki and Macroperipatus insularis. (Bouvier, 1900b) Synonyms: Peripatus ohausi, by original designation (Bouvier 1900b: 67); Peripatus (Macroperipatus) ohausi (Clark 1913b: 17); Macroperipatus ohausi (Peck 1975: 347). Holotype: Not designated. Type locality: BRAZIL, Rio de Janeiro, Petrópolis. Language of species description: French. Remarks: Although the holotype has not been clearly designated in the original description, Weidner (1959: 93) refers to an holotype found in the Zoologisches Staatsinstitut und Zoologisches Museum Hamburg, Hamburg, Germany. Requires revision.
According to Bouvier (1905: 201;1907a: 518), a type has been deposited in the Museum National d'Histoire Naturelle de Paris, France. The species name was misspelt as geagy by Jerez-Jaimes and Bernal-Pérez (2009: 567) and as geagi by Morera-Brenes and Léon (1986: 278). The description contains imprecise type locality data (the region of Calçoene in the Amapá State occupies 14,269 km 2 ). The information found in the type specimen label might refer to the river with the same name, which crosses the region. Since no further work with more precise locality data is available, a revision of this species based on topotypes will be difficult.

Mesoperipatus tholloni
According to Bouvier (1905: 349;1907a: 519), a type has been deposited in the Museum National d'Histoire Naturelle de Paris, France. The original description and the putative type contain imprecise type locality data (Gabon occupies an area of 267,667 km 2 ). However, Bouvier (1905: 348) refers to additional specimens found by M. Haug in Ngolé (spelt Ngômô) along the river Ogowe (spelt Ogôoué) and discusses that this might be the more precise locality for the species since the label of the type says only "Gabon" (a common practice at that time according to Bouvier 1905). Bouvier (1907a: 519) refers also to an additional specimen of this species collected in Ndjolé, along the same river. Therefore, it cannot be ruled out that M. tholloni is a species complex, which requires revision.
Language of species description: French. Remark: Requires revision. (Fuhrmann, 1913) Synonyms: Peripatus bimbergi, by original designation (Fuhrmann 1913: 242); Oroperipatus bimbergi (Clark 1915: 14). Holotype: Not designated. Type locality: COLOMBIA, Amagatal in the central mountain range (900-1,800 m) and Guaduas (800 m) towards Bogota, in the eastern mountain range (see Remarks). Language of species description: German. Remark: The current location of Amagatal could not be found. Requires revision. Fuhrmann, 1915 Synonyms: none (see Remarks). Holotype: Not designated. Type locality: PERU, Loreto, Shapajilla, Samiria River, 120 m. Language of species description: German. Remarks: In addition to Oroperipatus bluntschlii, the author used the name Peripatus bluntschlii in the original description (Fuhrmann 1915: 35 32.5.1), we suggest the above modification and consider, from now onwards, corradi as a misspelling of the species name. The description contains imprecise lo-cality data and the redescription of the species (Bouvier 1905: 120) was based on specimens from different localities, suggesting that it might be a species complex, which, thus, requires revision.
According to Bouvier (1907a: 518), a type has been deposited in the Museum National d'Histoire Naturelle de Paris, France. The description contains imprecise type locality data (Mexico occupies an area of 1,972,550 km 2 ). Since no further work with more precise locality data is available, a revision of this species based on topotypes will be difficult.
Remarks: Information on species description was obtained from the second edition of Schmarda's text book (Schmarda 1878: 74-77) and from Bell (1887). Although the species name suggests that the species locality is the environs of Quito, neither data on precise type locality nor information on type specimens is found in the literature. Since no further work with more precise locality data is available, a revision of this species based on topotypes will be difficult.

Peripatus Guilding, 1826
Type species: Peripatus juliformis Guilding, 1826, by subsequent monotypy. Remark: This oldest genus is referred to by some authors as Peripatus sensu stricto (e.g., Clark 1913b;Froehlich 1962). Brues, 1935 Synonyms: Peripatus dominicae var. basilensis, by original designation (Brues 1935: 62); Peripatus dominicae basilensis (Peck 1975: 348). Holotype: Not designated. Type locality: HAITI, Morne Basile (Mount Basil), northwestern part of the island (Brues 1939: 36 (Brues 1935: 61-62). Peripatus basilensis has also been recorded from additional localities (Brues 1939: 36), indicating that it might be a species complex. Revision of these five species is required, which might reveal additional morphological and molecular differences between them.  (Clark, 1913b: 17). Holotype: Not designated. Type locality: VIRGIN ISLANDS, Saint Thomas Island. Language of species description: French. Remarks. Peck (1975: 348) used the old species name suggested by Bouvier rather than that suggested by Clark (1913b: 17). The description contains imprecise type locality data. However, the area is relatively small (Saint Thomas Island occupies an area of 81 km 2 ) and clearly separate from other islands and from the mainland by seawater. Hence, we regard Peripatus danicus as a valid species, although we cannot rule out that additional species might be found on the same island. A revision of this species, including specimens from different localities on the island, is required. Brues, 1935 Synonyms: Peripatus dominicae var. darlingtoni, by original designation (Brues 1935: 62); Peripatus dominicae darlingtoni (Peck 1975: 348 (Peck 1975: 348). Holotype: Not designated. Type locality: DOMINICA ISLAND, Laudat (see Remarks). Language of species description: English. Remarks: As for Peripatus basilensis, except that this species was recorded from a single locality. Locality data obtained from specimens deposited in the Natural History Museum of London, United Kingdom. Requires revision.
Holotype: Not designated (see Remarks). Type locality: ANTIGUA ISLAND, Barlar, near Warburton (see Remarks). Language of species description: French. Remarks: The holotype has not been designated explicitly in the original description. According to Bouvier (1907a: 519), a type has been deposited in the Museum National d'Histoire Naturelle de Paris, France. The current name and position of the type locality could not be found. Since no further work with more precise locality data is available, a revision of this species based on topotypes will be difficult.
Holotype: Not designated. Type locality: AUSTRALIA, Tasmania, northwest (see Remarks). Language of species description: English. Remarks: Imprecise type locality data have been provided for this species, covering an area of over 2,000 km 2 in north-western Tasmania, although the "main population" occurs in the Christmas Hills, Arthur River, Rapid River area (Jackson and Taylor 1994: 167). Although the cover date for the species description is 1994, the work was not published until early 1995 (R. Mesibov in litt.). According to the provisions of the ICZN for publications after 1985 and before 2000, this does not constitute a published work for purposes of zoological nomenclature since it does not contain a statement that "the new name is intended for permanent scientific record" (ICZN Art. 8.5.2). Therefore, the name Ooperipatellus cryptus is a nomen dubium, which requires revision.

Ooperipatus Dendy, 1900a
Type species: Ooperipatus oviparus (Dendy, 1895), by subsequent designation (Dendy 1902: 367 (Dendy, 1895) Synonyms: Peripatus oviparus, by original designation (Dendy 1895: 195); Ooperipatus oviparus (Dendy 1900a: 510); Symperipatus oviparus (Cockerell 1913b; as a footnote in Clark 1913b: 19; see Remarks). Holotype: Not designated. A lectotype has been deposited in the Muséum National d'Histoire Naturelle, Paris, France (Reid 1996: 830). Type locality: AUSTRALIA, Victoria, Mount Macedon, 37°23'S, 144°35'E, 1,001 m. Language of species description: English. Remarks: A new genus Symperipatus was suggested for the species by Cockerell (1913b; as a footnote in Clark 1913b: 19). However, since Ooperipatus oviparus is the type species of Ooperipatus, the name Symperipatus has to be regarded as an objective synonym because it is based on the same name-bearing type (ICZN Art. 61.3.3). The species has been revised and a lectotype has been designated by . Reid, 2000a Synonyms: None. Holotype: Deposited in the Museum Victoria, Melbourne, Australia.  (Purcell 1899: 349) 55. Opisthopatus amatolensis Choonoo, 1947 Synonyms: Opisthopatus cinctipes var. amatolensis (Choonoo 1947: 71). Holotype: Not designated. Type locality: SOUTH AFRICA, Eastern Cape. The precise locality might be southeast of Houghton's farm, along the road from Alice towards Hogsback, 1,158 m (3,800 ft) (Choonoo 1947: 71). Language of species description: English. Remarks: The variation was considered as invalid by Ruhberg (1985: 85) and Hamer et al. (1997: 292) due to the lack of consistent differences to other subspecies. Nevertheless, we regard Opisthopatus amatolensis as a valid species because of the great distance between the type localities of this species and Opisthopatus cinctipes (161 km) and the apparent point endemism and cryptic speciation among the South African Peripatopsidae species (Daniels et al. 2009;Daniels and Ruhberg 2010; Ruhberg & Hamer, 2005 Synonyms: None. Holotype: Deposited in the Natal Museum, Pietermaritzburg, South Africa. Type locality: SOUTH AFRICA, KwaZulu-Natal, Mount Currie Nature Reserve, near Kokstad, alongside road between main entrance and pass, in forest patch near ravine, 30°17'13"S, 29°13'40"E (30.28 713°S/29.22 781°E). Language of species description: English. Lawrence, 1947 Synonyms: Opisthopatus cinctipes var. laevis by original designation (Lawrence 1947: 168). Holotype: Not designated. Type locality: SOUTH AFRICA, KwaZulu-Natal, East Griqualand, Bulwer. Language of species description: English. Remarks: The variation was considered as invalid by Ruhberg (1985: 85) due to the lack of consistent differences to other subspecies. Nevertheless, we raise Opisthopatus laevis to a species level and regard it as a valid species because of the great distance between the type localities of this species and Opisthopatus cinctipes (570 km) and the apparent point endemism and cryptic speciation among the South African Peripatopsidae species (Daniels et al. 2009;Daniels and Ruhberg 2010;. Requires revision. Bouvier, 1900d Synonyms: Opisthopatus cinctipes var. natalensis, by original designation (Bouvier 1900d: 368). Holotype: Not designated. Type locality: SOUTH AFRICA, Kwa-Zulu-Natal, Durban. Language of species description: French. Remarks: The variation was considered as invalid by Ruhberg (1985: 85) due to the lack of consistent differences to other subspecies. Nevertheless, we raise Opisthopatus natalensis to a species level and regard it as a valid species because of the great distance between the type localities of this species and Opisthopatus cinctipes (656 km) and the apparent point endemism and cryptic speciation among the South African Peripatopsidae species (Daniels et al. 2009;Daniels and Ruhberg 2010; (Willey, 1898b), by subsequent monotypy. Remarks: Initially, Paraperipatus was described as a subgenus of Peripatus and the name has been used as a generic name without an explicit statement since Bouvier (1900d: 369). Most species of the genus are understudied and require thorough revisions. (Muir & Kershaw, 1909) Synonyms: Peripatus ceramensis, by original designation (Muir and Kershaw 1909: 737); Paraperipatus ceramensis (Horst 1910: 218 (Sedgwick, 1910) Synonyms: Peripatus papuensis, by original designation (Sedgwick 1910: 369); Paraperipatus papuensis (Bouvier 1914b: 222 (Brues 1921: 51). Language of species description: English. Remarks: A holotype has not been designated for this species, although Brues (1921: 51) refers to a female specimen as a type. The species was synonymised with Paraperipatus papuensis (Ruhberg 1985: 151). However, Brues (1921: 51) described morphological differences between these two species, with Paraperipatus lorentzi showing more similarities to Paraperipatus stresemanni than to Paraperipatus papuensis (Brues 1921: 52). Moreover, according to Brues (1921) Ruhberg (1985: 146), the syntypes of this species have been lost. The type locality is imprecise (Ambon Island covers an area of 775 km 2 ) and a revision of the species based on topotypes will be difficult.  Brongersma (1932: 411) suppressed the name leopoldi and synonymised it with the older name Paraperipatus papuensis. Ruhberg (1985: 151) stated that, although the name Paraperipatus leopoldi is invalid, it is still available because it was accompanied by a proper species description. Accordingly, we regard Paraperipatus leopoldi as a nomen dubium. However, we disagree with Ruhberg's (1985: 151) suggestion that Paraperipatus leopoldi is a junior synonym of Paraperipatus papuensis since the type localities of the two species lie over 145 km apart from each other.

Paraperipatus leopoldi
Paraperipatus schultzei Heymons, 1912 Synonyms: Paraperipatus schultzei var. ferrugineus (Heymons 1912: 216;see Remarks). Holotype: Not designated. Type locality: PAPUA NEW GUINEA, north of New Guinea, inland region, on a mountain at 1,570 m (German New Guinea; see Remarks). Language of species description: German. Remarks: The holotype was not clearly designated in the original description. According to Röhlig et al. (2010: 228), the putative holotype (in fact a syntype) might have been placed in the Museum für Naturkunde Berlin but it has been lost. The more precise locality for the species might be the Sepik river system ("Sepikstrom 1,570 m"), according to Röhlig et al. (2010: 228). The putative variation Paraperipatus schultzei var. ferrugineus was found apparently in the same locality as Paraperipatus schultzei; this variation was suggested based on differences in colour pattern and in the number of leg pairs (Heymons 1912: 216). These characters are known to be variable intra-specifically in the Paraperipatus species and we therefore agree with Ruhberg (1985: 153) and regard Paraperipatus schultzei var. ferrugineus as synonym of Paraperipatus schultzei. The description of Paraperipatus schultzei contains imprecise locality data. Since no further work with more precise locality data is available, a revision of this species based on topotypes will be difficult.  Leloup (1931). The description contains imprecise locality data (Seram occupies an area of ~17,100 km 2 ). Since no further work with more precise locality data is available, a revision of this species based on topotypes will be difficult.

Peripatoides Pocock, 1894
Type species: Peripatoides novaezealandiae (Hutton, 1876), by original designation (Pocock 1894: 519 Remarks: The original description is based exclusively on molecular data and contains no type designation, which was not mandatory until 1999 (ICZN Art. 16). However, in contrast to other species described by Trewick (1998), based on specimens from different localities, Peripatoides kawekaensis shows a restricted distribution. We therefore regard it as a valid species, which requires revision.

Nomina dubia
Peripatoides aurorbis Trewick, 1998 Synonyms: None. Holotype: Not designated (see Remarks). Type locality: NEW ZEALAND, central and mid-northern North Island. Language of species description: English. Remarks: The original description is based exclusively on molecular data from specimens collected at different localities (Trewick 1998). Types were not designated and the description contains imprecise type locality data. Although molecular methods were used to define the species, no voucher specimens are available. Revision will thus be difficult.
Peripatoides morgani Trewick, 1998 Synonyms: None. Holotype: Not designated. Type locality: NEW ZEALAND, eastern North Island in a narrow coastal strip including southern and central Hawke's Bay and north to Lake Tikitapu. Language of species description: English. Remarks: The original description is based exclusively on molecular data from specimens collected at different localities (Trewick 1998). Types were not designated and the description contains imprecise type locality data. Although molecular methods were used to define the species, no voucher specimens are available. Revision will thus be difficult.
Holotype: Not designated. Type locality: NEW ZEALAND, southern North Island (Trewick 1998: 321). The precise locality might be near Wellington (Hutton 1876: 361; see Remarks). Language of species description: English. Remarks: The species name was commonly used for every species with 15 leg pairs found in New Zealand (Trewick 1998: 309). The original species description was based on specimens from different localities on both North Island and South Island. According to Trewick (1998), the species is restricted to the southern part of the North Island and the only specimens used in the original description from this area were those collected in the environs of Wellington (Hutton 1876: 361), suggesting that this might be the putative type locality of the species. However, the data available are still imprecise and different localities attributed to the species (Hutton 1876;Trewick 1998) Table Mountain. Language of species description: English. Remarks: The author designated two types in the original description but did not specify, which of the two specimens is the holotype (Lawrence 1931: 104). Furthermore, Lawrence (1931) did not provide any information on deposition of specimens, but they might have been deposited in the South African Museum (Cape Town, South Africa) as the author worked at this institution while describing the species.
72. Peripatopsis balfouri (Sedgwick, 1885) Synonyms: Peripatus balfouri, by original designation (Sedgwick 1885: 450); Peripatopsis balfouri (Purcell 1899: 341). Holotype: Not designated. Type locality: SOUTH AFRICA, Cape Town, Table Mountain Nature Reserve  (see Remarks). Language of species description: English. Remarks: The species comprises a species complex according to Daniels et al. (2009) and we suggest that the name should be used only for specimens obtained from the type locality. Requires revision.  ). Language of species description: German. Remarks: A type specimen has not been designated in the original description. Bouvier (1907b: 146) stated that the syntypes were lost and therefore a neotype has been designated by Ruhberg (1985: 94). The species was regarded as a species complex (Daniels et al. 2009) Daniels et al. (2009).
Thus, the name should be applied only to specimens obtained from the type locality. Requires revision. Hutchinson, 1928 Synonyms: None. Holotype: Deposited in the South African Museum, Cape Town, South Africa (see Remarks). Type locality: SOUTH AFRICA, 11.26 km (7 miles) east of Montagu. Language of species description: English. Remarks: A holotype has been designated in the original description under the name type. Ruhberg (1985: 91) has synonymised the species with Peripatopsis balfouri due to the bad condition of the type (known from the original description) and on putatively unreliable characters, on which the species was based. However, we believe it is premature to conclude that these species names are synonyms as long as no additional specimens of Peripatopsis intermedia have been collected and the species re-described. Moreover, the type localities of Peripatopsis balfouri and Peripatopsis intermedia lie 168 km apart. We therefore consider Peripatopsis intermedia as a valid species, which requires revision. Remarks: A type specimen has not been designated explicitly in the original description, although Weidner (1959: 93) and Ruhberg (1985: 98) refer to paratypoids and ex-types, the taxonomic status of which is unclear. No further record of this species has been made since 1912 (NHM-1936.4.28.4), suggesting that the species is either extinct or critically endangered (Brinck 1957: 13;Ruhberg 1985: 98;Daniels et al. 2009). Revision is required but will be difficult to accomplish, given the rarity of the specimens and the critically endangered status of the species.  Ruhberg (1985: 102) due to its putative long usage (see also the ICZN Art. 11.6.1). For the sake of stability, we consider Peripatopsis dewaali as a nomen oblitum and Peripatopsis sedgwicki as a nomen protectum, following the ICZN (Art. 23). However, since the epithet sedgwicki has also been assigned to Peripatus sedgwicki, a representative of Peripatidae, the abbreviation P. sedgwicki may cause confusion between these two species (see Röhlig 2010: 230). Recently, Expressed Sequence Tags (ESTs) have become available from a putative specimen of Peripatopsis sedgwicki (Meusemann et al. 2010). Thus, reverting the species name back to Peripatopsis dewaali at this point would create additional instability. We therefore suggest that the name Peripatopsis sedgwicki should be used with caution, in particular when additional data become available for both Peripatus sedgwicki and Peripatopsis sedgwicki. Furthermore, the name Peripatopsis sedgwicki should be applied only to specimens from the type locality as recent evidence suggests that it represents a species complex (Daniels et al. 2009 Remarks: Species description is based on an incomplete specimen preserved in amber of ambiguous age. Grimaldi et al. (2002) assigned the species to the Peripatidae, but this relationship is uncertain due to a limited preservation of the specimen. Note that the preserved body portion might be the posterior rather than the anterior end and the structure labelled mouth by Grimaldi et al. (2002: fig. 17 Poinar (2000) assigned †Tertiapatus dominicanus to a new taxon, Tertiapatidae, the validity of which is doubtful (Grimaldi et al. 2002).

Discussion
Our revised checklist revealed 177 valid onychophoran species worldwide (Table 1). However, this number most likely does not reflect the actual diversity of Onychophora as recent evidence suggests a high number of undescribed and cryptic species in both Peripatidae and Peripatopsidae Daniels et al. 2009;Oliveira et al. 2011;. Despite our attempt to retain as many valid species as possible, about 10% of described species represent nomina dubia (Table 1). The designation of nomina dubia in our list means that the species names are still available but difficult to revise due to the lack of designated types and precise locality data. Besides the nomina dubia, our study shows that many of the valid species also require revision, in particular representatives of Paraperipatus within Peripatopsidae, and nearly all species of Peripatidae. It is evident from previous studies that numerous species have been described based on characters that are ambiguous for some onychophoran subgroups, such as colour pattern of the skin, number of leg pairs, number of jaw denticles, and number and arrangement of spinous pads (for critical discussions, see Read 1988a;Oliveira et al. 2011). Typically, these species cannot be identified reliably without their locality data. Thus, a thorough revision of these species using scanning electron microscopy and molecular markers is desirable because these methods have proven useful for studies of the onychophoran taxonomy (e.g., Read 1988aRead , 1988bDaniels et al. 2009;Oliveira et al. 2011). The use of these methods will inevitably alter the traditional classification since several onychophoran genera, in particular within Peripatidae, are not based on unique features or synapomorphies but might be subjective, non-monophyletic assemblages .
Notably, 70% of the Peripatopsidae species (81 species) but only 34% of the Peripatidae (28 species) have designated holotypes (Table 1; Appendix 1: Checklist_numbers). This finding reflects the understudied nature of Peripatidae, with only one thorough revision at the species level (Bouvier 1905), while there have been at least four comprehensive revisions of the Peripatopsidae species from different geographic regions (Bouvier 1907b;Ruhberg 1985;Hamer et al. 1997). This highlights the urgent need of more taxonomic work on the Peripatidae species.