Revision of the southern Andean genus Sadocus Sørensen, 1886 (Opiliones, Gonyleptidae, Pachylinae)

Abstract Species of the genus Sadocus Sørensen, 1886 are conspicuous gonyleptids that occur in Chile and Argentina. Here, the genus is revised for the first time and the cladistic analysis based on morphological characters does not corroborate its monophyly unless a phylogenetically unrelated species is excluded (explained further on). A new classification is proposed for the seven species left in the genus and considered valid, of the 13 nominal species previously recognized. Two out of the seven valid species are considered as species inquirendae: Sadocus allermayeri (Mello-Leitão, 1945) [= Carampangue allermayeri Mello-Leitão, 1945] and Sadocus nigronotatus (Mello-Leitão, 1943) [= Carampangue nigronotatum Mello-Leitão, 1943]. The following synonymies are proposed: Sadocus bicornis (Gervais, 1849) [original combination = Gonyleptes bicornis Gervais, 1849] is a junior synonym of Sadocus asperatus (Gervais, 1847) [= Gonyleptes asperatus Gervais, 1847]; Sadocus conspicillatus Roewer, 1913, Sadocus exceptionalis (Mello-Leitão, 1946) [= Araucanoleptes exceptionalis Mello-Leitão, 1946] and Sadocus guttatus Sørensen, 1902 are junior synonyms of the valid name Sadocus polyacanthus (Gervais, 1847) [= Gonyleptes polyacanthus Gervais, 1847]; and Sadocus calcar (Roewer, 1913) [= Lycomedes calcar Roewer, 1913] is a junior synonym of the valid name Gonyleptes horridus Kirby, 1819. Sadocus brasiliensis Soares & Soares, 1949 is not congeneric with Argentinean/Chilean species of the genus according to the cladistic analysis and is here synonymized with Discocyrtus catharinensis (Mello-Leitão, 1923 [= Sadocus catharinensis Mello-Leitão, 1923]).


Introduction
, in his complete catalogue of New World Laniatores, proposed the synonymy of Lycomedicus, Carampangue, and Araucanoleptes with Sadocus. Hence, Sadocus comprised 14 species (actually, there are entries for 15 species, but that of S. subsimilis is clearly a mistake, which should be listed as a junior synonym under S. asperatus). Finally, Pessoa-Silva et al. (2020), transferred S. planiceps to Eubalta. Sadocus hitherto was composed of 13 species (Kury et al. 2020b). In the present publication, we accept only seven species of Sadocus as valid.

Materials and methods
Material examined belongs to the following institutions (curators in parentheses) listed below: The following abbreviations are used throughout the text, including synonymic listings: cat catalogue; cit citation; coll collected; desc description; eco ecology; rdesc redescription; syst systematic discussion.
In the examined material: fe female; ma male; juv juvenile;

MS A-E penis ventral plate pairs of macrosetae A-E.
[

X(y)]
where X is the character number and y, the character state.
The topological nomenclature follows Acosta et al. (2007), nomenclature of integumentary ornamentation of dorsal scutum and legs, dorsal scutum outline and ventral plate penial macrosetae follows DaSilva and Gnaspini (2010), Kury and Medrano (2016) and Kury and Villarreal (2015), respectively. Nomenclature of ovipositor morphology generally follows Townsend et al. (2015). We adopted the orientation of the captured images to reference the ovipositor lobes, because we had no topological reference after detaching it. It is unlike the penis, which has a sclerotized ventral feature, thus being easily referenced topologically. In Sadocus, we realized that leg IV is twisted retro-laterad from the trochanter (gradually untwisting along the femur), rendering the otherwise prolateral structures as dorsal (Fig. 3A,E). To standardize the topological nomenclature, we opted to consider (and call) those as prolateral, despite being functionally dorsal (in situ). We illustrated the external morphology using a stereomicroscope with a camera lucida and the material immersed in 70% ethanol. We prepared male and female genitalia according to Pinto-da-Rocha (1997) to take pictures using a scanning electron microscope (SEM) or to illustrate using a compound microscope with a camera lucida. The generic characteristics are not repeated in the specific (re)descriptions. Only characters differing from those of the males are listed in the female (re)descriptions. The variation on the number of tubercles on the dorsal scutum and other parts of the body or legs were included in the intraspecific variation. The color descriptions are based on specimens preserved in 70% ethanol and living photograph examples presented in the section "variation in males (or females)" under each species. Many species of Sadocus present a white patch on the body, commonly known as a dry-mark (Kury in DaSilva and Gnaspini 2009). It is an external serose layer of the cuticle that often forms white patches/shapes. Distribution maps for Sadocus species were prepared using QGIS 3.10 (QGIS.org 2019). The identification key is only for males. Synonymic listings follow Kury's catalogue (2003), to which we add the category of its content between parentheses (see abbreviations section above). All measurements are in millimeters. We followed the view of Kury et al. (2020a) regarding the use of the correct inflection of specific epithets that are adjectives throughout the article to avoid inviting further confusion for the reader. Therefore, despite Kury himself (2003) proposing the combination Sadocus funestis (Butler, 1874), we use Sadocus funestus (Butler, 1874) as Kury et al. (2020b) in all sections (except for synonymic listing), including the historical aspects of Sadocus.

Cladistic analysis
To test the monophyly of Sadocus, we used a matrix of morphological characters composed of 18 taxa (13 outgroups and five ingroups) and 64 characters ( Table 1). The 64 characters are distributed as follows: 18 from dorsal scutum, four from free tergites, one from the chelicera, 23 from male legs, 17 from male genitalia, and one from coloration. We only included the valid Sadocus species with known males, as most of the characters are based on armature of male leg IV and penis. The cladistic analysis resulted in a single most parsimonious tree (182 steps, L = 182; C.I. = 45; R.I. = 53, Fig. 1).
According to the retrieved tree, Sadocus is not monophyletic, as it excludes Sadocus brasiliensis (Soares & Soares, 1949). Acosta (2020) first mentioned that S. brasiliensis may not belong to this genus based on the overall distribution of the other Sadocus species. That suspicion is corroborated here, and we propose its synonymy with the Brazilian Discocyrtus catharinensis (Mello-Leitão, 1923) (see taxonomic notes in this species' entry). In turn, the close relationship of D. catharinensis with Roeweria bittencourti  (1)]. Discocyrtus catharinensis is probably a Roeweriinae especially based on the shape of the penial ventral plate as well as the overall penial morphology. However, we refrained further taxonomic actions as Roeweriinae diversity grows further fueled by the dismemberment of Discocyrtus, which is currently under revision (Kury and Carvalho 2016;Kury 2018, 2021). Once we settled the issue related to S. brasiliensis, we propose a new concept of Sadocus. Under the new definition, Sadocus is monophyletic and supported by seven synapomorphies, four of which are exclusive: leg IV twisted from the trochanter to patella [28(0)]; trochanter IV with a finger shaped, robust prodorsal apical apophysis, its basal half of ca. uniform diameter [32(3)], trochanter IV twice longer than wide [37(1)] (modified from Hara 2016); and penis ventral plate with slightly concave distal margin [47(1)]. Sadocus is also the best supported clade of the analysis, with a high Goodman-Bremer support (4).
So far, Sadocus (represented especially by S. polyacanthus, its type species) has often been used in cladistic analysis as outgroups Hara 2016) or as an ingroup taxon of a more comprehensive analysis testing monophyly of Gonyleptidae or Gonyleptoidea (Pinto-da-Rocha et al. 2014;Benavides et al. 2021). According to the analyses based on morphological characters Hara 2016), Sadocus is nes-  Table 1, and the data matrix is in Table 2.  tled in a clade mainly composed of Brazilian species. However, we have to stress that the clade with Sadocus in Hara's analyses is not well supported (Bremer support: 1), its sole homoplastic synapomorphy being the proventral apical armature of tibia IV as a tubercle. In the present analysis, we have a roughly similar outcome, as the clade including Sadocus (Sadocus + Neogonyleptes) is sister group to a clade composed of solely Brazilian species. This outcome differs considerably from Pinto-da- Rocha et al. (2014) or Benavides et al. (2021): in those analyses based on molecular data, Sadocus is often retrieved closely related to Chilean Pachylinae genera. Regarding this, Pinto-da- Rocha et al. (2014) indicate that Sadocus is in a clade with other Chilean species (Neogonyleptes karschii and Tumbesia aculeata), which in turn is sister group to Pachylinae sensu stricto. Benavides et al. (2021) also corroborates a close relationship of Sadocus with Chilean genera.
The sister taxon closest to Sadocus is also an unsettled issue, mainly because different taxa are employed in those analyses. In the present analysis, Sadocus sister group is the Chilean genus Neogonyleptes, supported by seven synapomophies, two of them exclusive: ventral row of granules increasing in size apically on tibia IV [46(1)]; and apex of glans ventral process flattened circle shaped [57 (2)]. This sister group relationship is similar to Pinto-da-Rocha et al. (2014) sampling wise. On the other hand, Benavides et al. (2021) did not include Neogonyleptes in their analysis, and the Chilean Eubalta planiceps is the sister taxon to Sadocus. It is interesting to note that Sadocus sister taxon is strongly affected by the Chilean Pachylinae sampling not belonging to Pachylinae sensu stricto.
The main goals of this study were to revise and to test the monophyly of Sadocus, a hundred-year-old genus, with a convoluted taxonomic history. We believe that we succeeded in those, and the present study is an important step towards the understanding the evolution of the genus. Considering all the evidence (including the taxonomic history), Sadocus seems to be related to the Chilean-Argentinean Pachylinae. We understand that Sadocus relationship within Gonyleptidae is still an unsettled issue that deserves further investigation. As a mean to tackle that, we can suggest the inclusion of more Chilean Pachylinae genera (especially those already used in previous analyses and not belonging to Pachylinae sensu stricto) and Brazilian species as well, such as DRMN (Carvalho and Kury 2018) and K92 (Kury 1992 Sørensen (1902).
Diagnosis. Sadocus are large Pachylinae (dorsal scutum maximum length 5.5-13.8 mm) with paired spines on ocularium and prominent frontal hump on dorsal scutum anterior margin. Dorsal scutum shape types gamma triangular and gamma pyriform, its posterior margin concave. Dorsal scutum mid-bulge placed close to scutal groove IV (scutal groove III in S. funestus) and transversal (S. funestus, S. ingens, S. polyacanthus) or oblique (S. asperatus, S. dilatatus); free tergites II and III each with a pair of spines. Legs IV are twisted retro-laterad from the trochanter and gradually distorted along the femur and patella (except in S. funestus). Coxa IV bearing a long, large prodorsal apical apophysis and a short, retro-apical one (except in S. funestus, in which is lacking). Trochanter IV with a short, blunt prolateral sub-basal apophysis and a long, robust prodorsal apical one. Penis glans turgid and dorsally projected (with antero-lateral projections), with ventral process (half stylus length) and without dorsal process. General color (in living specimens) of the body and most parts of legs and ventral area dark brown, with lighter tones at the tips of podomeres. Yellowish to reddish tone in scutal area, scutal posterior margin, free tergites, part of legs and apophysis. Green on the arthrodial membranes between the free tergites.
Redescription. Male: Dorsum. Anterior margin of carapace with a prominent median frontal hump (bell shaped in dorsal view). Ocularium with one pair of spines posterior to the eyes. Dorsal scutum type varying from gamma to gamma triangular and gamma pyriform, its posterior margin concave, mid-bulge slightly asymmetrical and displaced posteriorly, widest at the scutal groove IV (scutal groove III in S. funestus). The curvature of mid-bulge can be transversal (S. funestus, S. ingens, and S. polyacanthus) or oblique (S. asperatus and S. dilatatus). Four scutal areas (three in S. polyacanthus); scutal area I divided into right and left halves by a longitudinal median groove. Scutal area III with one pair of paramedian spiniform tubercles or spines. Two pairs of ozopores close to coxa II. Lateral margin of dorsal scutum with an external and internal rows of tubercles (the external row of slightly larger tubercles) (except S. asperatus, with granules covering most of the lateral margin of dorsal scutum and S. ingens, smooth or with only few granules). Posterior margin of dorsal scutum and free tergite I each with one paramedian pair of tubercles (except S. funestus and S. polyacanthus, unarmed). Free tergites II and III each with one paramedian pair of spines. Venter. Coxa I-IV granulate; coxa I with a median longitudinal row of granules increasing in size apically, becoming tubercles. Chelicerae. Isomorphic in males and females. Segment I with well-marked bulla. Segment II fixed finger and segment III toothed. Pedipalps. Trochanter dorsal face inflated; ventral face with one or two setiferous tubercles. Femur bearing sub-apical mesal seta; dorsal face with few granules; ventral face with one basal setiferous tubercle. Tibiae and tarsi dorsal and lateral faces with few minute granules and variable setation. Legs. Coxae I-III each with one prodorsal and one retro-dorsal spiniform tubercles, ventral faces granulate (except S. polyacanthus, coxa I with tubercles and others with setae). Coxa IV dorso-lateral face with sparsely distributed granules, ventral face entirely granulated, with one long, oblique, bifid prodorsal apical apophysis (transversal in S. dilatatus, uniramous in S. funestus), dorsal branch longest and curved ventrad and ventral branch short and blunt; and one ventro-apical retro-lateral spine. Trochanters I-III granulate. Leg IV twisted retro-laterad from the trochanter, gradually untwisting along the femur (except S. funestus, straight). Trochanter IV longer than wide; prolateral face with one short, conical, blunt sub-basal apophysis, and one robust, blunt dorso-apical apophysis. Femora I-IV with granules roughly organized in six longitudinal rows (prodorsal, retro-dorsal, pro-and retro-lateral, proventral and retro-ventral rows); femora I and II unarmed. Femur IV curved, with marked inner curvature on the distal half (S. asperatus and S. ingens) or almost straight (S. dilatatus, S. funestus, and S. polyacanthus). Patellae I-III granulate, unarmed; patella IV dorsal face granulate, ventral face tuberculate. Tibiae I-III granulate, unarmed (except S. dilatatus and S. funestus, tibia III dorsal face granulate with retro-ventral row of tubercles increasing in size apically). Tibia IV dorsal face granulate, ventral face with tubercles sparsely distributed. Metatarsi I-IV minute granulate, unarmed. Tarsus III and IV each with ventral process, tarsal claws smooth. Penis. Ventral plate distal margin with slight (but conspicuous) to moderate concavity, two or three pairs of MS A, one pair of MS B or entirely absent, four or five pairs of MS C, one or two pairs of MS D, and one or two pairs of MS E. Glans sac tall, turgid, dorsally projected with antero-lateral projections, forming a sheath for the stylus. Glans without dorsal process; stylus inserted ventrally and smooth. Glans ventral process is short (half of stylus length), parallel to the stylus, apex curved ventrad, with a short semi-circular antero-lateral projection.
Geographic distribution (Fig. 2). Central Chile, Región XIV Los Ríos; Metropolitan Region of Santiago; Región V Valparaíso; Región VIII Bio-Bío; Región IX Araucanía; and Region X Los Lagos. There are other localities mentioned in the literature besides the material studied here for S. polyacanthus in Neuquén (Argentina) and Magallanes, in the extreme south of Chile, however, we did not examine any material from there. The record of Sadocus funestus for Ecuador (Chimborazo, Riobamba) by Roewer (1913) is certainly a mislabeling because it does not agree with the known generic distribution (Cekalovic 1985;Kury 2003). Two species are widely distributed (S. asperatus and S. polyacanthus) and three others occur mainly in coastal mountains of Central Chile (S. funestus, S. dilatatus, and S. ingens).
Key to males of Sadocus species 1 Coxa IV with a bifid prodorsal apical apophysis (Fig. 3A, C) and one retroventral apical apophysis (Fig. 3A)  Trochanter IV dorso-apical face only with one prolateral apophysis of similar length as the podomere, strongly curved (in lateral view), pointing frontwards (Fig. 6C) (-37.766667, -73.333333), 9.II.1965 Diagnosis. Sadocus asperatus resembles S. ingens, S. polyacanthus, and S. dilatatus by the bifid prodorsal apical apophysis on coxa IV. S. asperatus can be distinguished from the latter species by the combination of the following characters: lateral margin of dorsal scutum covered by granules; trochanter IV with a blunt retro-dorsal apical apophysis being half of the podomere length, and a rhombus retro-ventral apical tubercle; femur IV curved (in dorsal view), with a retro-lateral row of spiniform apophysis (the middle one longest). Redescription. Male (CAS 9055035). Measurements. Dorsal scutum maximum length 6.3; dorsal scutum maximum width 7.2; prosoma maximum length 2.5; prosoma maximum width 3.2; leg femora I 3.2; II 6.5; III 5.5; IV 7.2. Dorsum (Fig. 3A). Dorsal scutum type gamma triangular. Carapace with granules sparsely distributed. Scutal areas I-IV with eight, four, two and four granules, respectively; scutal area III with one pair of paramedian spiniform tubercles; scutal area IV incompletely divided. Lateral margin of dorsal scutum mostly covered in granules (from posterior half of carapace to posterior margin of dorsal scutum). Posterior margin of dorsal scutum and free tergite I each with a pair of paramedian tubercles. Chelicerae. Segment I with basal tubercle, bulla with small setae, each finger with five teeth. Pedipalps. Coxa dorsal face smooth, ventral face with two apical tubercles. Trochanter dorsal and ventral faces smooth. Femur ventral face granulate. Patella with sparsely distributed setae. Tibial setation: prolateral IiiiIi/IiiiIi; retro-lateral iIiIi/IiIi. Tarsal setation: prolateral and retro-lateral IiIi/IiIi. Legs (Fig. 3B-E). Coxa IV with one long, oblique, bifid prodorsal apical apophysis and one retro-ventral apical spine. Trochanters I and II each with one pair of prodorsal spiniform tubercles. Trochanter III with one medio-ventral tubercle and three retro-lateral ones. Trochanter IV prodorsal and proventral faces with few granules, the prodorsal apical apophysis long (ca. half the podomere length); retrolateral face with one basal, one central, and one apical tubercles; one retro-dorsal apical spiniform apophysis (ca. ¼ podomere length); ventral face with setiferous tubercles sparsely distributed. Femur III with one retro-basal tubercle. Femur IV sigmoid, with dorsal row of apophysis on the basal half abruptly decreasing in size apically, becoming granules; prolateral row with central-subapical tubercles; retro-lateral row of granules with spiniform apophysis (basal most and central one longer than the others, apical most oblique, curved ventrad); ventral face with two short retro-lateral sub-apical spiniform apophyses and one proapical spiniform apophysis. Patella IV ventral face mostly smooth, with one probasal, one proapical and one retro-apical large, spiniform tubercles. Tibia IV roughly with two ventral rows of granules increasing in size from central to apical becoming spines. Tarsal counts: 6, 9, 7, 8. Penis (Fig. 11A, B). Ventral plate of penis with moderate cleft on anterior margin, three pairs of MS A, one pair of MS B, four or five pairs of MS C and one pair of MS D, without MS E.
Redescription. Dorsum (Fig. 8A, B). Scutal areas I, II, and IV with three, five, and two granules, respectively. Legs. Coxa IV with one prodorsal apical apophysis and one retro-ventral apical spine shorter than on the male, trochanter IV with retro-lateral row of tubercles, the apical one longest, femur IV with pro-and retro-ventral rows of tubercles, patella-metatarsi IV unarmed. Tarsal counts: 6, 9, 7, 8. Ovipositor (Fig. 12A-C). Two main groups of lobes delimited by a constriction, ovipositors peripheral setae inserted into sockets that are a mixture of dorsal and ventral sockets, the dorsal lobe with five setae and the ventral one with six; each main group of lobes divided by a fissure.
Geographical distribution (Fig. 2). Chile, Región de Los Ríos, Valdivia, Corral; Región Metropolitana de Santiago, Santiago; Región de Valparaíso; Región de  Taxonomic notes. After examining the original description and the drawing of Gonyleptes bicornis, we concluded that it is of a male of S. asperatus. In the original description, the spines on the free tergite, the two apical apophyses on the trochanter IV and uneven spines in the inner part of the "leg" (referring to the femur IV) are mentioned. Those characters lead us to conclude that it is S. asperatus. Diagnosis. Sadocus dilatatus resembles S. polyacanthus by the lesser-armed femur IV (compared to other species) and by the posterior large tubercle on the lateral margin of dorsal scutum. Sadocus dilatatus can be distinguished from the other species of the genus by the single retro-ventral central apophysis on femur IV and the very long prodorsal apical apophysis on coxa IV (ca. ⅔ of the scutum width).

Sadocus dilatatus
Redescription. Male (SMF 886). Measurements. Dorsal scutum maximum length 7.5; dorsal scutum maximum width 10.4; prosoma maximum length 3.2; prosoma maximum width 4.2; leg femora: I 6.0; II 13.0; III 10.3; IV 11.0. Dorsum (Fig. 4A). Dorsal scutum type gamma triangular. Anterior margin of dorsal scutum with median frontal hump bearing six tubercles and three granules on each side. Carapace with granules sparsely distributed. Scutal areas I-IV with 13, 16, eight and six granules, respectively; scutal area III with one pair of paramedian spines; scutal area IV completely divided (from area III). Lateral margin of dorsal scutum mostly covered by granules (from the posterior half of carapace to posterior margin of dorsal scutum), with one large tubercle near scutal area IV. Posterior margin of dorsal scutum and free tergite I each with few granules on the corners. Chelicerae. Segment I with one seta on mesal side of the bulla, each finger with five or six teeth. Pedipalps. Coxa mostly smooth, with one ventro-central tubercle. Trochanter dorsal face smooth, with one retro-ventral apical tubercle and one retro-apical spine. Femur dorsal face smooth, ventral row of few granules and one retro-ventral spine. Patella smooth. Tibial setation: prolateral Iiii/IiiIi; retro-lateral IiIi/IiIi. Tarsal setation: prolateral IiIii/IiIi; retro-lateral iiIiIii/iIiIii. Legs (Fig. 4B-E). Coxa IV with one long, oblique, bifid prodorsal apical apophysis and one retro-ventral apical spine. Trochanters I-III granulate; trochanter III with one retro-apical spine. Trochanter IV with sparse granules, the prodorsal apical apophysis long, (ca. ⅓ of podomere length), pointing prolaterad. Femur III with proventral and retro-ventral rows of granules increasing in size apically, becoming tubercles. Femur IV with three blunt dorsal spines on basal ⅓; proventral row of pointed granules increasing in size apically becoming spines; retro-lateral row with two spines and one central apophysis and one proventral basal spine. Patella IV with ventral row of tubercles. Tibia IV with three retro-ventral apical spines (apical one the largest). Tarsal counts: 6, 11, 7, 10. Penis (Fig. 11G, H). Ventral plate of penis with attenuated cleft on anterior margin; three pairs of MS A, one pair of MS B, four pairs of MS C, one pair of MS D, and two pairs of ventral MS E.
Note. The allotype MZSP 7874 was not used for the variation or in the distribution maps because it is a female, which lacks the diagnostic characters of the species (solely based on male characters).
Redescription. Male (CAS 9026265). Measurements. Dorsal scutum maximum length 9.3; dorsal scutum maximum width 10.0; prosoma maximum length 3.9; prosoma maximum width 5.0; leg femora: I 4.5; II 8.3; III 7.0; IV 8.0. Dorsum (Fig. 5A). Dorsal scutum type gamma pyriform. Dorsal scutum anterior margin with nine granules, lateral margins with a row of granules and cluster of granules near scutal groove I. Ocularium with one pair of granules on anterior face. Scutal areas I-IV with six (three on each side), three, one, and six granules, respectively; scutal areas I-III each with one pair of paramedian tubercles. Scutal area IV completely divided. Lateral margin of dorsal scutum with row of granules between ozopore area and anterior part of scutal area IV. Posterior margin of dorsal scutum and free tergites I-III each with a row of granules. Chelicerae. Segment I with setae on bulla; fixed finger with four teeth, movable finger with three teeth. Pedipalps. Coxa with one ventro-apical spine. Trochanter dorsal face smooth, with one pair of geminated ventro-apical setiferous tubercles. Femur with a row of ventro-basal granules. Patella smooth. Tibial setation: prolateral IiIi/ IiiI; retro-lateral iIiIi/iIiIi. Tarsal setation: prolateral IiIii/IiIi; retro-lateral iiIIIiii/iiIiiIii. Legs (Fig. 5B-E). Coxa IV with one long, oblique, uniramous prodorsal apical apophysis, without retro-ventral apical spine. Trochanters I and III granulate; trochanter III with one retro-apical spine. Trochanter IV proapical spiniform apophysis with large base (ca. ⅓ of podomere length); additionally with two retro-dorsal apical apophyses with the ridges touching each other; retro-lateral face with two central granules and one apical spine. Femur III dorsal face granulate; ventral face with two rows of granules increasing in size apically. Femur IV with prodorsal row of large tubercles decreasing in size apically; prolateral row of tubercles; one retro-ventral row of granules; proventral row of granules increasing in size apically; one proventral apical spiniform apophysis, one retro-sub-apical spiniform apophysis, and two or three retro-ventral apical apophyses. Patella IV with five retro-ventral spines. Tibia III with ventral row of tubercles. Femur IV with retro-lateral row of spines and four ventro-apical spines. Tarsal counts: 6, 12, 7, 8. Penis (Fig. 11I, J). Ventral plate of penis with deeper (than moderate) cleft on anterior margin, three pairs of MS A, four pairs of MS C, one pair of MS D, without MS B or E.
Geographical distribution (Fig. 2). Chile, Región de Los Ríos, Provincia de Valdivia, Curiñanco & Chiguayco. The record for Ecuador is doubtful (Kury 2003). Figures 2, 6A-E, 8G, H, 10A Diagnosis. Sadocus ingens can be distinguished from the other species of the genus by being the largest among them (and quite large among gonyleptid harvestmen); by the prodorsal apical apophysis on trochanter IV of the same length as the podomere (in other Sadocus species, that apophysis length is up to ½ the podomere length); lateral margin of dorsal scutum smooth posterior to scutal area II.

Sadocus ingens (Mello-Leitão, 1937)
Redescription. Male (MZSP 36965). Measurements. Dorsal scutum maximum length 12.0; dorsal scutum maximum width 15.5; prosoma maximum length 4.5; prosoma maximum width 7.2; leg femora: I 8.7; II 17.4; III 11.7; IV 19.4. Dorsum (Fig. 6A). Dorsal scutum type gamma pyriform. Anterior margin of dorsal scutum with 13 granules, lateral margin of dorsal scutum with three to nine granules on carapace, seven granules behind ocularium. Scutal areas I-IV with four or five, nine, six, and four granules, respectively; scutal area III with one paramedian pair of spines. Scutal area IV completely divided by fading scutal groove IV. Lateral margin of dorsal scutum and free tergites I and II with three or four, eight and two granules, respectively. Free tergite III smooth. Chelicerae. Segment I with one probasal spine on bulla, one retro-and one prolateral pair of filiform spines; segment II, fixed finger with four teeth, movable finger with three teeth. Pedipalps. Coxa smooth and barely visible. Trochanter ventral face with one retro-apical tubercle and one prolateral spine. Femur dorsal face granulate, one ventro-basal tubercle, retro-ventral row of granules and a ventral row of tubercles. Patella smooth. Tibial setation: prolateral IiIi/IiiIi; retro-lateral iIiIi. Tarsal setation: prolateral iIiIi; retro-lateral iIiIii/iiIiIii. Legs (Fig. 6B-E). Coxa IV covered by setae, with one robust, long, bifid prodorsal apical apophysis and one short retro-ventral spiniform apophysis. Trochanters I-III granulate; trochanters I and III dorsal face smooth. Trochanter IV with few granules on ventral central and apical areas; one retrolateral tubercle; the prodorsal apical apophysis long, curved, as long as the podomere length. Femur IV with retro-lateral row of spines (five prominent) decreasing in size apically, becoming blunt tubercles; retro-ventral row of granules with one basal pointed tubercle, few tubercles on the middle ⅓ and two spines on apical area; ventro-apical face with one retro-lateral spine and one pointed, prolateral tubercle. Patella IV with three or four retro-ventral spines. Tibia IV dorsal face granulate and with retro-ventral row of granules increasing in size apically, becoming spines on distal half. Tarsal counts: 8,16,8,8. Penis (Fig. 11C, D). Ventral plate of penis with moderate cleft on anterior margin, two or three pairs of MS A, four or five pairs of MS C and one pair of MS D.
Coloration. Immersed in ethanol: carapace, trochanters, femora, patella IV, and tibia IV dark brown. Scutal areas II and III, free tergites, patellae and tibiae I-III orange. Live specimens (Fig. 10A, B): carapace, scutal areas and legs I-IV black; lateral margin of dorsal scutum dark brown with green pleurites; posterior margin of dorsal scutum and free tergites orange, arthrodial membranes green.
Variations (n = 6) -Free tergites II and III with one paramedian pair of spines which length varies from similar to slightly longer than the tergite length, its apex varying from blunt to pointed; femur IV with granules in between the retro-lateral spines. Measurements. Dorsal scutum maximum length 12.0-13.8; dorsal scutum maximum width 12. . Measurements. Dorsal scutum maximum length 12.0; dorsal scutum maximum width 11.5; prosoma maximum length 5.0; prosoma maximum width 7.0; leg femora: I 8.4; II 16.2; III 12.0; IV 17.0. Dorsum (Fig. 8G, H). Scutal areas I-IV with six, six, four, and four granules, respectively. Legs. Coxa IV with discreet apophysis, not seen in ventral view. Tarsal counts: 8; 15; 8; 9. Ovipositor (Fig.12 G-I). Two main groups of lobes delimited by constriction, ovipositors peripheral setae inserted into sockets that are a mixture of dorsal and ventral sockets, left lobe with six setae and right lobe with six. Each main group of lobes divided by a fissure.
Coloration. Immersed in ethanol: carapace, trochanters I-IV and femur IV dark brown. Legs with a gradient from brown to caramel. Live specimens (Fig. 10C-F): carapace, patellae, and tibiae I-IV black, except the areas of dorsal scutum that can vary (yellow, orange or red). Coxa and trochanters black. Femora I-IV varying between black and orange. Posterior margin of dorsal scutum to free tergite III arthrodial membranes green.
Taxonomic notes. After examining the holotypes of S. conspicillatus, S. guttatus, and S. exceptionalis, we concluded that they were males within S. polyacanthus size variation. The apophyses size and shape on trochanter IV and the armature of femur IV (especially the retro-dorsal and retro-lateral row of spines size pattern) of all species are the same.
Taxonomic notes. The type material, belonging to MNRJ, was lost in the fire that destroyed most of the arachnid collection (Kury, pers. comm.;. The original description is poor for modern standards and it has no illustrations. However, the description allows to be diagnosed by: the presence of scutal area IV on dorsal scutum, coxa IV with one prodorsal apical bifid apophysis; trochanter IV with one retro-basal apophysis and three proapical apophyses. Sadocus asperatus, S dilatatus, and S. ingens have the scutal area IV in dorsal scutum and coxa IV with a bifid prodorsal apical apophysis; but none of them has three apical apophyses on trochanter IV. The only species with four scutal areas on the dorsal scutum and trochanter IV with three apical apophyses is S. funestus, but the prodorsal apical apophysis on coxa IV is uniramous. Assuming that the description is correct, it implies that this is a valid species that we have not yet located among the material gathered for this revision of Sadocus. Figure 8K Carampangue nigronotatum Mello-Leitão, 1943: 8, fig 7 [desc]; Soares and Soares 1954: 242 [cat];Cekalovic 1968Cekalovic : 7 [cat]: 1985 (type material MNRJ, 1 fe holotype lost, not examined). Sadocus nigronotatus: Kury, 2003, 191 [cat]; Kury et al. 2020b [cat].
Taxonomic notes. The type material, belonging to the MNRJ, is lost (for the same reason as S. allermayeri). The original description is poor for modern standards, and both description and figure are based on a female. The females of the different Sadocus species are very similar and difficult to identify unequivocally. According to the original description, S. nigronotatus resembles S. polyacanthus in the presence of a spiniform apophysis on the lateral margin of the dorsal scutum, but can be distinguished by the presence of scutal area IV.

Species removed from Sadocus
Discocyrtus catharinensis (Mello-Leitão, 1923) Figure 11K, L Sadocus catharinensis Mello-Leitão, 1923 Taxonomic notes. We examined the type material of Sadocus brasiliensis and its external and penial morphology did not match that of other Chilean species of the genus. Based on its type locality, we examined other Brazilian Pachylinae genera and found striking similarities between S. brasiliensis and D. catharinensis. We examined detailed pictures of the type material kindly shared by Rafael N. Carvalho and additional material from the MZSP collection. Those species are the same, and we propose that S. brasiliensis is the junior synonym of D. catharinensis. Many Discocyrtus spp. have been transferred to revalidated genera or newly created subfamilies, such as Roeweriinae (Carvalho and Kury 2018) or Neopachylinae (Carvalho and Kury 2021). Indeed, the penial features of D. catharinensis (Fig. 15K, L) resembles those of that subfamily, which is corroborated in the present analysis: D. catharinensis is the sister species of R. bittencourti. Based on male genitalia and leg IV (see Carvalho and Kury 2018), D. catharinensis seems to belong to Discocyrtanus Roewer, 1929. However, because there is an ongoing study revising Discocyrtus conducted by Rafael N. Carvalho (MNRJ) and taking into account that it will have serious taxonomic consequences, we opted to propose only the synonymy of S. brasiliensis with D. catharinensis.
Taxonomic note. The holotype of Sadocus calcar is in a very bad state of preservation; only part of the carapace, with the ozopores, and leg IV remain. The rest of the prosoma and all of the other legs are absent. Even in this condition, we noted that S. calcar lacks the large tubercles and spines on the free tergites, which are diagnostic of Sadocus. Furthermore, the armature of trochanter IV and the long, bifid, C-shaped dorso-basal apophysis on femur IV are strikingly similar to those of Gonyleptes horridus, a common species in the state of Rio de Janeiro. Therefore, we propose S. calcar as a junior synonym of G. horridus. This synonymy made us conclude that the provenance of S. calcar is mistaken, because G. horridus is endemic to the Brazilian Atlantic rainforest. It is widely known that Roewer, unfortunately, indicated wrong provenance of a few species, and this seems to be the case for this species.

Eubalta planiceps (Gervais, 1842)
Remarks. Sadocus planiceps (originally Gonyleptes planiceps Gervais, 1842) has a convoluted taxonomic history, with many previously unknown synonyms detected, which will be published elsewhere (briefly commented in Pessoa-Silva et al. 2020). We excluded it from Sadocus because it lacks the synapomorphies of the genus. It also lacks the diagnostic characters of the genus, such as the typical ocularium shape and type of armature, dorsal scutum shape, just to name a few. Comparing with other species of Chilean Pachylinae, we detected striking similarities with Eubalta meridionalis. This synonymy did not go unnoticed by Kury et al. (2020a) in his catalogue, who also detected this in parallel with this revision. Finally, Kury et al. kindly invited us to publish this synonymy that resulted in a publication of that synonymy (Pessoa-Silva et al. 2020).
do Estado de São Paulo (Fapesp 2018/07193-2). We also thank the financial support from CNPq (304933/2014-7). Rafael Carvalho kindly shared pictures of Discocyrtus catharinensis that resulted its synonymy with Sadocus brasiliensis. We are grateful to the editor Gonzalo Giribet and reviewers of an early version, Cristina Rheims, Cibele Bragagnolo, and James Cokendolpher.