Avicularia Lamarck, 1818

Fukushima, Caroline Sayuri & Bertani, Rogerio, 2017, Taxonomic revision and cladistic analysis of Avicularia Lamarck, 1818 (Araneae, Theraphosidae, Aviculariinae) with description of three new aviculariine genera 01, ZooKeys 659, pp. 1-185 : 22-27

Genus Avicularia Lamarck, 1818 Figs 1, 9-10, 13, 15-17, 19, 20, 21-27, 28-35, 36-43, 44-49, 50, 51, 52-53, 54-57, 58-65, 66-73, 74-75, 76, 77, 78-79, 80-87, 88-89, 90, 91-92, 93-100, 101-104, 105, 106-108, 109-116, 117-122, 123-126, 127, 128-130, 131-138, 139-141, 142-147, 148-153, 154-157, 158-165, 166-171, 172, 173-180, 181-183, 184-187, 188-195, 196-198, 199-200, 201-208, 209, 210-217, 218, 299-300, 302-304, 307, 309, 311, 312-315, 316-317

Aranea Linnaeus, 1758: 622 (in part: Avicularia avicularia ); Fabricius 1775: 438 (in part: Avicularia avicularia ); 1793: 424 (in part: Avicularia avicularia ); De Geer 1778: 313, pl. 38, fig. 8 (in part: Avicularia avicularia ); 1783: 122, pl. 38, figs 8-10 (in part: Avicularia avicularia ); Latreille 1802: 49 (in part: Avicularia avicularia ). Name abandoned as generic name to be used for the Order name ( Thorell 1870: 151).

Mygale Latreille, 1802: 49, 1804: 152, pl. LXII, fig.1 (in part: Mygale avicularia ); 1806: 85 (in part: Mygale avicularia ); Walckenaer 1805: 4 (in part: Mygale avicularia ); Hahn 1820: 5 (in part: Mygale avicularia ); Simon 1864: 70, figs 26-28 (in part: Mygale avicularia ). Name preoccupied in Mammalia by Mygale Cuvier, 1800 ( Thorell 1870: 162).

Avicularia Lamarck, 1818: 107 (type species Aranea avicularia Linnaeus, 1758 by subsequent designation in direction 67 of ICZN (1957: 116)); Thorell 1870: 168; Ausserer 1871: 128, 201; 1875: 138; Simon 1889: 213; 1892: 171; 1903: 958, 960; F. O. Pickard-Cambridge 1896: 740, 746; 1899: 42; Pocock 1901: 548; Mello-Leitão 1923: 320, 376; ICZN 1928: 395; Petrunkevitch 1928: 81; Roewer 1942: 253; Bonnet 1955: 826; Valerio 1979: 307; Raven 1985: 119; Gallon 2008: 243; World Spider Catalog 2016.

Eurypelma C. L. Koch, 1850: 73, 74 (in part); Simon 1864: 67 (in part); Roewer 1942: 238 (in part); Bonnet 1955: 1828; Raven 1985: 146, 153. Considered objective synonym of Avicularia by Raven (1985: 146).

Avicuscodra Strand, 1908: 771 (type species by original designation Avicuscodra arabica Strand, 1908, female, Egypt, El-Tor [28°14'N, 33°37'E], (Tor, Arabien [sic]) Rüppell, SMF 2660, examined); Petrunkevitch 1928: 81; Roewer 1942: 256; Gallon 2008: 243. First synonymized by Gallon (2008: 243).

Ancylochiros Mello-Leitão, 1920: 41 (type species by original designation Ancylochiros taunayi Mello-Leitão, 1920, immature male, Brazil, Minas Gerais, Mariana [20°22'S, 43°25'W], J. P. Fonseca leg., MZUSP 327, examined,); 1923: 318; 376; Roewer 1942: 225; Raven 1985: 149. First synonymized by Raven (1985).

Anchylochyrus Petrunkevitch, 1928: 83 (unjustified emendation per Bonnet 1955); Strand 1929: 12.

Type species.

Aranea avicularia Linnaeus, 1758, by subsequent designation ( ICZN 1957).

Species included.

Avicularia avicularia , Avicularia caei sp. n., Avicularia glauca , Avicularia hirschii , Avicularia juruensis , Avicularia lynnae sp. n., Avicularia merianae sp. n., Avicularia minatrix , Avicularia purpurea , Avicularia rufa , Avicularia taunayi and Avicularia variegata stat. n.

Diagnosis.

Avicularia resembles Caribena gen. n., Ybyrapora gen. n., Iridopelma and Typhochlaena by the procurve anterior row of eyes (Fig. 13). It can be distinguished from Typhochlaena by the digitiform apical article of PLS (Fig. 10). It differs from Iridopelma by male lacking tibial apopysis on tibiae II and female by spermathecae long, with accentuated outwards curvature medially (Fig. 21). It differs from Caribena gen. n. by stout urticating setae on abdomen dorsum of male and female (Figs 15-17) and by rounded process on cymbium (Fig. 307). From Ybyrapora gen. n. it can be distinguished by sclerotized spermathecae in females (Fig. 21) and presence of developed tibial apophysis in males (Fig. 311), except Avicularia minatrix , Avicularia lynnae sp. n., Avicularia caei sp. n., and A hirschii that lack tibial apophysis. Males of these Avicularia species can be distinguished from Ybyrapora diversipes comb. n. by the embolus less curved in frontal view (Fig. 30). From Ybyrapora sooretama comb. n. and Ybyrapora gamba comb. n. they differ in terms of the presence of a well-developed process on cymbium (Fig. 307), except Avicularia minatrix . Avicularia minatrix can be distinguished from Ybyrapora sooretama comb. n. and Ybyrapora gamba comb. n. by the abdomen dorsum black having orange spots on its side (Fig. 89).

Description.

Carapace slightly longer than wide, cephalic region slightly raised. Cephalic and thoracic striae inconspicuous due to setae density. Fovea deep or shallow, slightly recurve (most species) or straight. Chelicera without rastelum. Eye tubercle distinct, raised or slightly raised, wider than long. Anterior row of eyes procurve (Fig. 13), posterior recurve or slightly recurve (most species). Clypeus narrow (most species) or absent. Labium subquadrate, longer than wide, with 50-133 cuspules spaced by one or two diameters from each other on the anterior third center. Maxillary lyra absent. Maxilla subretangular, anterior lobe distinctly produced into conical process, inner angle bearing 85-215 cuspules. Sternum longer than wide, posterior angle acute, not separating coxae IV. Three pairs of sigilla, some pairs sometimes not evident. Anterior ellipsoidal or rounded, middle rounded, posterior ellipsoidal (most species) or rounded. All positioned one diameter or less from margin. Leg formula: I=IV II III (most species) or IV I II III ( Avicularia avicularia , Avicularia rufa and Avicularia hirschii female). Clavate trichobothria on distal 2/3 or 1/2 of tarsi I–IV. Tarsi I–IV fully scopulate, IV divided by a band of setae or divided by a bald stripe. Metatarsi I–II fully scopulate in most species, III 1/3 to 1/2 distal scopulate and IV 1/5 to 1/2 distal scopulate. Metatarsi IV divided by a row of setae. Scopulae of tarsi and metatarsi I–II very extended laterally giving them a spatulate appearance (Figs 299-300). Femora IV without retrolateral scopulae. Stridulatory setae absent. Legs aspinose. ITC absent; STC without denticles. Posterior lateral spinnerets digitiform (Fig. 10). Typical stout type II urticating setae on dorsal abdomen in females and males, 0.36 to 0.72 mm long, 0.11-0.019 mm wide in females and 0.78 to 1.11 mm long, 0.013-0.025 mm wide in males (Figs 15-17). Male tibiae I with apophysis (most species) (Fig. 311) or with discrete elevation covered by a cluster of setae on apical portion (Fig. 309), on prolateral side. Tibial apophysis with a single branch having well-developed base and grouped spiniform setae distally (Figs 33-35). Tibiae II lacking apophysis. Male metatarsus I touches retrolaterally tibial apophysis’ setae when folded. Globous bulb with small subtegulum; prominence on prolateral tegulum developed (most species) (Fig. 314), well-developed ( Avicularia variegata stat. n. and Avicularia juruensis ) (Fig. 315), weakly-developed ( Avicularia caei sp. n.) (Fig. 313) or absent ( Avicularia minatrix ) (Fig. 312). Embolus not flattened, without keels, about 3.0 to 3.5 times tegulum’s length (most species) or more than 4 times tegulum’s length ( Avicularia lynnae sp. n. and Avicularia caei sp. n.) in retrolateral view. Embolus medial portion and tegulum’s margin form an acute angle in retrolateral view (Fig. 323). Embolus with basal part very curved in frontal view (Figs 312-315), thin distal width and tip narrowing abruptly (Figs 316-317) or tapering (Figs 318-319). Cymbium subtriangular with subequal lobes, having a well-developed rounded process on retrolateral lobe (most species) (Fig. 307) or lacking any process (Fig. 162). Cymbium process bearing thick setae (most species) (Fig. 307) or thin setae (Fig. 205). Spermathecae with an accentuated outwards curvature medially, completely separated, and long (Fig. 21). Spermathecae not-twisted (except Avicularia hirschii ) with walls lacking projections or lobes (except Avicularia taunayi ) (Figs 21-27). Spermathecae midwidth as wide as its base width (most species) (Fig. 21) or midwidth expanded, about 1.5 times the apex width ( Avicularia variegata stat. n., Avicularia juruensis and Avicularia taunayi ) (Fig. 54). Spermathecae with weakly-sclerotized area shorter than half the length of well-sclerotized area (most species) (Fig. 21) or at least same size of well-sclerotized area ( Avicularia purpurea and Avicularia merianae sp. n.) (Fig. 154). Abdomen dorsum of females with homogeneously distributed setae (most species) (Fig. 304) or with long guard-setae grouped on lateral and dorsal anterior areas (Figs 302-303). Legs and palps with long guard-setae having homogeneous coloration along its length ( Avicularia avicularia , most morphotypes; Avicularia juruensis morphotype 3, Avicularia purpurea , Avicularia taunayi and Avicularia merianae sp. n.) (Fig. 299) or long guard-setae with darker base and contrasting whitish apex ( Avicularia avicularia morphotypes 6 and 7; Avicularia variegata stat. n.; Avicularia rufa ; and Avicularia juruensis morphotype 2) (Fig. 300). Leg rings on distal femora, tibiae and metatarsi whitish (most species) (Fig. 299) or yellowish ( Avicularia avicularia , morphotype 5; Avicularia variegata stat. n., morphotype 2; Avicularia juruensis morphotype 2; and Avicularia rufa ) (Fig. 300). All species except Avicularia minatrix showing drastic ontogenetic changes on abdominal color pattern. Brownish juveniles lacking metallic green or blue sheen, with black tarsi contrasting with other lighter articles (except Avicularia purpurea ) (Fig. 44). Juveniles having dorsal reddish abdomen with a single central longitudinal stripe and some transversal black stripes on each side (except Avicularia hirschii ). Central longitudinal stripe disconnected from transversal stripes ( Avicularia avicularia , Avicularia variegata stat. n., Avicularia juruensis and Avicularia rufa ) (Fig. 44), connected with all transversal stripes in each side of abdomen ( Avicularia purpurea and Avicularia minatrix ) (Fig. 167), connected only with the anterior pair of transversal stripes ( Avicularia merianae sp. n.), or connected with first and second anterior pair of transversal stripes ( Avicularia taunyai ) (Fig. 101).

Distribution and habitat.

Costa Rica, Panama, Trinidad and Tobago, Venezuela, Guyana, Suriname, French Guiana, Colombia, Ecuador, Peru, Bolivia and Brazil. In Brazil, it occurs in the states of Roraima, Amapá, Amazonas, Pará, Maranhão, Tocantins, Acre, western Bahia, Rondônia, Mato Grosso, Goiás, Minas Gerais, São Paulo, and the Distrito Federal (Fig. 19). Valerio (1979) reports presence of Avicularia species in Costa Rica (see note on Avicularia avicularia redescription and on Avicularia glauca taxonomy). Beside this, there are reports in Lago Catemaco and other localities in the state of Veracruz, as well as in the state of Chiapas, Mexico (R. West and J. Mendoza, pers. comm.), but no Mexican specimens could be examined. Thus, despite the reliable records, it is not possible to know which species are in Mexico.

Remarks.

When erecting the genus Avicularia , Lamarck (1818) included in it three species described originally in Aranea , in this order: Avicularia canceridea (Lamarck, 1818), Avicularia blondii (Latreille, 1804), and Avicularia fasciata (Latreille, 1804). The first species, Avicularia canceridea , could be found in the Antilles in "Meridional America" (South America)( Lamarck 1818). As synonyms of Avicularia canceridea , Lamarck (1818) cited Aranea avicularia Linnaeus, 1758 (sensu Linnaeus 1758 and Fabricius 1793) and Mygale avicularia Latreille, 1804 (sensu Latreille 1804 and Walckenaer 1805). It is unclear why Lamarck (1818) did not use the pre-existent specific name avicularia instead of canceridea .

Thorell (1870) considered Aranea avicularia Linnaeus "as the type for Avicularia ", and Avicularia canceridea Lamarck and Aranea vestiaria De Geer as its synonyms. Apparently, this was the reason for the Lamarckian combination Avicularia canceridea being buried.

Ausserer (1871) considered Avicularia canceridea Lamarck, Aranea avicularia Linnaeus, Aranea vestiaria De Geer and Mygale avicularia Latreille as synonyms; however, he considered valid the binomy Avicularia vestiaria and stated it as the type species. F. O. Pickard-Cambridge (1899) also considered Avicularia canceridea Lamarck as a synonym of Avicularia avicularia , but stated this last one as the type species. Despite the specific epithet avicularia being the most used for the type species of Avicularia over the years, it was only officially considered as such in 1928 through Opinion 104 ( ICZN 1928) and introduced in the official list of specific names in Zoology through Direction 67 of ICZN ( ICZN 1957).

Herein, we tried to establish the real identity of Avicularia avicularia as the original description is uninformative and useless for species identification. The type locality, America, is also vague and the existence of types was controversial. Taking into consideration that Aranea avicularia is a very old species described by Linnaeus in the 10th Edition of Systema Naturae ( Linnaeus 1758), it would be a difficult task to track a type.

Apparently, Linnaeus (1758) did not designate holotypes. According to Papavero and Llorente-Bousquets (1995), the oldest citation regarding the type concept as we use, nowadays, is dated 1817 by the German dipterologist C. R. W. Wiedmann. In his publication, Wiedmann (1817 apud Papavero and Llorente-Bousquets 1995) approved the idea about designation of types in description of new species and creation of a Normalmuseum where all types should be deposited. Therefore, Linnaeus could not determine types since this concept was elaborated many years after publication of Systema Naturae ( Papavero and Llorente-Bousquets 1995).

Linnean holotypes are also a problem for other specialists. Many botanists, for example, argue it is very difficult to know when Linnaeus used a well-defined specimen to describe a plant species. Thus, The Linnean Plant Name Typification Project was created to choose types from the specimens and illustrations that Linnaeus used in arriving at his concept of a species in question (The Linnean Society of London 2016). The creation of this project strengthens the ideia that Linnaeus did not designate holo types. Specifically in Avicularia 's case, this idea is supported by the expressed citation by F. O. Pickard-Cambridge (1896) that the type is a figure of two specimens drawn by Merian (1726), and not a specimen.

Even with a doubtful existence of the Avicularia avicularia holotype, Gabriel et al. (2007) tried to locate it in The Linnean Society of London Collection. The authors found two specimens of Aranea avicularia considered by the curator as Linnean material ( Gabriel et al. 2007) and observed they are two different species: a female of Stromatopelma calceatum Fabricius, 1793; and a male of Harpactira atra Latreille, 1832, both African species ( Gabriel et al. 2007). They concluded that none of the African specimens in the Linnean collection could be considered types of Avicularia avicularia ( Gabriel et al. 2007).

The ICZN states on the article 72.1.1 of its Code that "in the absence of holotype designation, or the designation of syntypes, or the subsequent designation of a lectotype, all [specimens] are syntypes and collectively they constitute the name-bearing type" ( ICZN 1999). Additionally, the article 72.4.1. ( ICZN 1999) states that "the type series of a nominal species-group taxon consists of all the specimens included by the author in the new nominal taxon (whether directly or by bibliographic reference)". Thus, as there is no designation of a specific holotype in the original description, all the specimens used in the publications mentioned by Linnaeus in the description of Aranea avicularia are part of a syntypic series.

In that description, Linnaeus (1758) cited his own previous work ( Linnaeus 1749), as well as Clusius (1611), Laet (1633), Piso and Marcgrave (1648), Worm (1655), Olearius (1666) and Merian (1705).

The spider drawn by Clusius (1611) seems, in fact, to be an aviculariine, since it has a hairy body and very developed subungueal tufts (Fig. 2). The spider was illustrated with low cephalic region and could be found in Baya de todos los Sanctos [sic] ( Clusius 1611). The animal depicted resembles specimens of Pachistopelma spp., which can be found in Baía de Todos-os-Santos, a bay in the city of Salvador, state of Bahia, northeastern Brazil. Laet (1633) and Piso and Marcgrave (1648) also reproduced this illustration and indicated that this species is very common in Saint Aleixo Island (state of Pernambuco, Brazil), area where Pachistopelma rufoniger specimens are frequent and where also there is no record of Avicularia spp. Other authors ( Olearius 1666, Worm 1655) used the same Clusius’ spider illustration to refer to this Brazilian species which we found to be a Pachistopelma species.

Merian (1705), on the other hand, illustrated two spiders with very developed and conspicuously colored subungueal tufts, unmistakably Avicularia sp. specimens (Fig. 1). One of them is leaving a typical retreat and feeding on ants and the other specimen is feeding on a small bird. Smith (2000b) stated that, probably because of this illustration, Linnaeus used the specific name " avicularia ", meaning "small bird" in Latin. Merian (1705) had drawn these spiders based on specimens observed during her expedition to Suriname. In the illustration’s comments, there is no indication of locality, nor any information that could lead to an inequivocal identification of the illustrated species.

Linnaeus (1758), in his Aranea avicularia description, also cited two illustrations of Seba (1734) of a spider from Ceylon, now Sri Lanka, which is probably a Poecilotheria specimen, and an unidentified specimen from America, which clearly is not an aviculariine.

Thus, among the syntypic Avicularia avicularia series are specimens of Pachistopelma spp. ( Clusius 1611, Laet 1633, Piso and Margrave 1648, Worm 1655, Olearius 1666), Avicularia ( Merian 1705) and unidentified specimens (Seba 1704). Apparently, the specimens used and/or illustrated by these authors are lost or have never been deposited in a zoological collection, but only observed in their natural habitat. Smith (2000a) believe that Linnaeus (1758) used the engraving done by Merian (1705) for describing Aranea avicularia , which was considered perfectly acceptable at that time. However, Smith (2000a) did not discard the possibility that Linnaeus had examined specimens collected by Merian, which could be deposited in particular collections, although the author states that much of Merian’s material is lost.

The ICZN Code article 72.4.1.1 states that "for a nominal species or subspecies established before 2000, any evidence, published or unpublished, may be taken into account to determine what specimens constitute the type series" ( ICZN 1999). The Code also considered part of the syntypic series specimens of which there is evidence that they were known to the author and recognized by him when the nominal species was established (see example in article 72.4.1.1 of ICZN 1999). After searching some collections, we found three specimens of Aranea avicularia belonging to the Linnean collection at the Museum of Evolution of Uppsala University. Even though there is no irrefutable proof that these specimens once belonged to Linnaeus and were examined by him when describing the species (Dr. Mats Eriksson, pers. comm.), we think that two of three specimens found in the collection were used by Linnaeus in his work. The spiders’ donor was Adolf Fredrik, King of Sweden in the time Linnaeus was writing his Systema Naturae ( Kullander 2001). The King had an extensive biological collection composed of plants and animals from all over the world, a common habit of wealthy and significant people of that time ( Kullander 2001). During nine weeks between 1751 and 1754, Linnaeus, during that time a professor at Uppsala University, was housed in the Royal Castle to catalogue the royal collection; the result of this work was published in a book ( Kullander 2001). It is probable that the King donated some material to Linnaeus, which were brought with him to Uppsala instead of going with most of the King’s material to the Swedish Museum of Natural History in Stockholm. Thus, those specimens of Aranea avicularia could have been really used by Linnaeus (1758) to describe the species and, therefore, they are part of the syntypic series.

King Adolf’s specimens are in a dry, pinned collection, thus they cannot be properly examined and handle (Fig. 20). However, even by photography, it is clear that both are Avicularia avicularia specimens, probable from the morphotype found in the state of Pará, Brazil, since morphological features such as size and coloration match with those seen in specimens found in that area. Therefore, herein, we establish the specimen UUZM 61 from the Uppsala collection as lectotype and the specimen UUZM 62 as paralectotype, solving the oldest and one of the most vexing problems in mygalomorph taxonomy.

Key to Avicularia species

Male of Avicularia glauca and female of Avicularia caei sp.n. and Avicularia lynnae sp. n. are unknown.

Female