Description of a new fossil Thelyphonida (Arachnida, Uropygi) and further record of Cratosolpuga wunderlichi Selden, in Selden and Shear, 1996 (Arachnida, Solifugae) from Crato Formation (Aptian/Albian), Araripe Basin, Brazil

Geological setting

The Araripe Sedimentary Basin is located in the interior part of the northeast region of Brazil and includes areas of three different states: Ceará, Pernambuco, and Piauí. The Araripe Basin is about 12,000 km² and is the largest intratectonic basin in Brazil forming a sedimentary unit including nine geological formations with Devonian, Jurassic, and Cretaceous deposits (Saraiva et al., 2007; Fambrini et al., 2020). Under the Chapada do Araripe, we found a stratigraphic sequence of ∼1,000 m with pre-, syn-, and post-rift phases (Saraiva et al., 2007).

The Santana Group (syn-rift phase) is a Cretaceous sedimentary unit subdivided into four formations: Barbalha, Crato, Ipubi, and Romualdo (Assine, 1994). Barbalha, Crato, and Ipubi formations are commonly associated with freshwater environments, whereas the Romualdo Formation is associated with lacustrine and transitional marine sediments (Valença, Neumann & Mabesoone, 2003; Fara et al., 2005).

The Crato Formation is mainly composed of laminated micritic limestones and clay-carbonate rhythmites, with the color of these rocks being bluish in their unweathered state. However, upon weathering, the colors vary from beige to brown and from light gray to bluish gray, which alternate between shales and fine sandstones (Viana & Neumann, 2002; Saraiva et al., 2021). In the laminated limestone, there are pseudomorphs of salt (halite) and various types of well-preserved fossils (Viana & Neumann, 2002; Martill, 2007). Outcrops of the Crato Formation are more commonly found in the towns of Nova Olinda and Santana do Cariri, as well as in the quarries of Rio Batateiras, Santa Rita, and Caldas (Viana & Neumann, 2002).

The holotype of Mesoproctus rayoli n. sp. studied here was recovered along with several other fossil specimens in the “Operation Santana Raptor” of the Brazilian Federal Police. The materials recovered by this operation are largely from the Crato Formation, as the famous “Ubirajara jubatus”. As this material is from a federal police operation that retrieved smuggled fossils, the precise locality is unknown; however, through the characteristic of the matrix, it is possible to assume its Crato Formation origin. The paratype of Mesoproctus rayoli n. sp. was donated to the Museu de Paleontologia Plácido Cidade Nuvens (MPPCN) and its origin or any other information about sampling site was not given, although, as is the case of the holotype, we can assume its Crato Formation origin based on the matrix where the fossil is found.

The Solifugae Cratosolpuga wunderlichi Selden, in Selden & Shear, 1996, was donated through the “Projeto Força Tarefa” from the Inhumas neighborhood in Santana do Cariri, Ceará, Brazil. This initiative encourages local children to collect and contribute fossils to the MPPCN. Additionally, this project aligns with the museum’s campaign, “#lugardefossilenomuseu” which aims to solicit fossil donations from the regional community.

Systematic paleontology

Mesoproctus sp.–Dunlop & Martill, 2002: 332, figs. 3b, 4b.

Type material.—Holotype MPSC A4295, body length 65.9 mm; paratype MPSC A4205, carapace length 28.2 mm.

Additional specimen.—Partially complete prosoma MB.A. 1041, carapace length 32.5 mm.

Type locality.—Unknown (see Geological setting).

Stratigraphic unit.—Crato Formation, Santana Group, Araripe Sedimentary Basin, Brazil.

Type age.—Lower Cretaceous (Aptian—Albian).

Diagnosis.—Body length including pygidium around 66 mm; pedipalpal coxae apophysis with one terminal tooth, one accessory tooth on the inner margin, outer margin of the apophysis serrated proximally; opisthosoma distinctly oval shaped; leg IV much longer than opisthosoma.

Description.—Holotype (Figs. 1, 2A–2E, 3) almost complete, in ventral view; body length (including pygidium) 65.9 mm; carapace length 25.7 mm, width 14.3 mm (L/W 1.8 mm). Y-shaped anterior sternum between second pair of leg coxae (Fig. 2B). Posterior sternum not well preserved.

Pedipalps massive, folded, with coxae, femur, patella and tibia partially preserved (Figs. 2A, 3A); trochanter not well preserved. Pedipalpal coxae apophysis short, rounded, with one terminal tooth, and 1 accessory tooth on the inner margin, outer margin serrated proximally (Figs. 2A, 2C, 3B). Trochanter poorly preserved, with 1 spine visible on the right side (Fig. 2C) and 3 spines on the left side (Fig. 3A). Femur twice as long than wide; patella 1.4 longer than wide; patellar apophysis well developed. Tibia 1.6 longer than wide (excluding fixed finger and tibia apophysis) (Fig. 2D); fixed finger and basitarsus well developed.

Legs mostly preserved flattened (L4 > L2 = L3), prolateral face uppermost, partially complete, distal end not well preserved, telotarsus apparently not divided. Antenniform leg I not preserved. Legs II and III about same size, robust femora, patellae and tarsi ventral corner with stiff setae (clearly preserved on left side of the specimen), right leg II with pair of small claws. Leg IV longer than II and III, well exceeding pygidium.

Opisthosoma elongate; longitudinally oval, length 37.2 mm, width 16.5 mm (L/W ratio 2.2) (Figs. 1, 2E); first and last three segments (pygidium) poorly preserved; sternites II–IX partially preserved; telson flagelliform, length 72.5 mm, width 1.1 mm; articles approximately twice as long as wide.

Paratype (Fig. 4) partially preserved in dorsal view with only first three segments of opisthosoma preserved; carapace elongate, densely granulated; granules uniformly distributed; length 28.2 mm, width 16.6 mm (L/W ratio 1.7 mm).

Etymology.—The specific epithet is in honor of Rafael Ribeiro Rayol, Brazilian federal attorney, who helped, along with the Brazilian Federal Police, in the retrieve of the material from the “Operation Santana Raptor”.

Remarks.—When adding new characters to the description of Mesoproctus rowlandi Dunlop, 1998, Dunlop & Martill (2002) proposed that a larger specimen (MB.A. 1041) examined might represent a distinct species. This suggestion primarily stemmed from the significant size difference between Mesoproctus sp. and M. rowlandi. Alternatively, they also suggested that this size discrepancy might be attributed to the differing ontogenetic stages of the specimens, with the smaller one potentially being a juvenile. In the light of the new material studied, here we found characters to differentiate Mesoproctus rayoli n. sp. from Mesoproctus rowlandi Dunlop, 1998 and include Mesoproctus sp. of Dunlop & Martill (2002) in the synonym of the new species.

The size of Mesoproctus rayoli n. sp. (holotype 65.9 mm length) is impressive, without comparison in the fossil record. Among the whipscorpions, the only group with a similar size is the extant Mastigoproctus Pocock, 1894, with species that can reach as far as 73 mm in length including the pygidium (Barrales-Alcalá, Francke & Prendini, 2018). Within the known specimens of Mesoproctus rowlandi Dunlop, 1998, the type specimen has 25 mm (Dunlop, 1998), and additional specimens with 16.8 and 17.1 mm in length (without pygidium) were described by Dunlop & Martill (2002). Mesoproctus sp. from Dunlop & Martill (2002), which we consider here as a junior synonym of Mesoproctus rayoli n. sp., had a carapace length of 32.5 mm, being even larger than the holotype. It is true that in many species, mature specimens can be distinguished from their immature counterparts based on size; however, as observed in the recent species Mastigoproctus giganteus, ornamentation of the pedipalp remains somewhat constant during ontogeny (Weygoldt, 1971). Thus, Mesoproctus rayoli n. sp. differs from Mesoproctus rowlandi (Dunlop, 1998) by the pedipalpal coxae apophysis having one terminal tooth, one accessory tooth on the inner margin, and the outer margin of the apophysis serrated proximally (Figs. 2A, 2C, 3B) (vs. endites of coxa of pedipalps without ornamentation in M. rowlandi). Mesoproctus rayoli n. sp. also differs from M. rowlandi in the proportions of the opisthosoma, distinctly oval shaped in M. rayoli n. sp. (Figs. 1, 2E) and considerably cylindrical in M. rowlandi (Dunlop, 1998: 292, fig. 1; Dunlop & Martill, 2002: 327, fig. 2B; 332, fig. 4A). For instance, the L/W ratio of the opisthosoma sixth segment is 4.0 in Mesoproctus rayoli n. sp. (Fig. 2E), and the L/W ratio of the opisthosoma sixth segment in M. rowlandi is 2.1. Also, leg IV is much longer than the opisthosoma in M. rayoli n. sp. (Figs. 1, 2E), whereas the fourth pair of legs were just reaching (MB.A 975) or slightly overreaching the pygidium in M. rowlandi (holotype) (Dunlop, 1998: 292, fig. 1; Dunlop & Martill, 2002: 327, fig. 2B; 332, fig. 4A).

Remarks.–When describing Cratosolpuga wunderlichi, Selden, in Selden & Shear, 1996, attributed the new taxon to Ceromidae based on the morphology of the flagellum, the propeltidium shape with distinct lateral lobes, the eye tubercle with anteriorly directed spines, the spine pattern of legs, and the presence of long tarsal claws on legs III and IV (see also Punzo, 1998). The specimen we report here aligns with all of the characteristics described by Selden (Selden & Shear, 1996). However, the new material suggests the possibility of a segmented tarsomere on leg IV and leg I with one tarsal claw. Such features were neither observed in the holotype by Selden & Shear (1996) nor confirmed by Dunlop & Martill (2004). In our analysis, we identified what appears to be segmentation on the tarsomere of leg IV and at least one leg I tarsal claw. Although these characteristics aid in family definition (Muma, 1976), the potential number of tarsomeres observed aligns with the criteria set for Ceromidae as per Selden & Shear (1996).

Genus Cratosolpuga Selden, in Selden & Shear, 1996

Diagnosis (after Selden, in Selden & Shear, 1996).–Male cheliceral flagellum attached to dorsomedial side of chelicera near base of fixed finger, consisting of globose base and styliform whip extending directly backwards to base of chelicera, stiff membrane partly enclosing flagellum base and running length of flagellum forming a narrow gutter.

Cratosolpuga wunderlichi Selden, in Selden & Shear, 1996

Figs. 5, 6A–6F, 7A–7F, 8A–8D, 9

Cratosolpuga wunderlichi Selden, in Selden & Shear, 1996: 584, 588–595, 601–603, pl. 1, figs. 2–4, pl. 2, text-figs. 1, 3–6 a–g.

Cratosolpuga wunderlichi— Punzo, 1998: 213–214, figs. 7.7–7.10; Dunlop & Martill, 2002: 325; Harvey, 2002: 366; Harvey, 2003: 212; Dunlop & Martill, 2004: 145, figs. 1–8; Dunlop, Menon & Selden, 2007: 120, figs. 9.7, 9.8; Dunlop & Penney, 2012: 72, fig. 46.

Cratosolpuga [wunderlichi]—(Dunlop, 1996): 85, fig. 5.

Type material.–Holotype (n^o Sol 1), part only, deposited in the private collection of J. Wunderlich, Germany.

Additional specimens.–SMNK 1268 PAL 1, presumed juvenile, total length 5.8 mm; MB.A. 1087, total length 20.3 mm; MB.A.1088, total length 14.4 mm; SMNS 65417, total length 14.7 mm; SMNS 54418, total length 23.1 mm.

Stratigraphic unit.–Crato Formation, Santana Group, Araripe Sedimentary Basin, Brazil.

Type age.–Lower Cretaceous (Aptian–Albian).

Material examined.–MPSC A6696, total length 16.5 mm, probably an adult male; collected from Santana do Cariri, Ceará, Brazil.

Remarks.–Measurements of the studied specimen: Total body length (including chelicerae): 16.48 mm (Fig. 5); prosoma (with chelicerae): 4.97 mm; prosoma (without chelicerae): 4.06 mm; opisthosoma: 9.27 mm; chelicera: 0.91 mm. Total length of leg 1: 8.55 mm; femur: 1.03 mm, patella: 2.21 mm, tibia: 2.21 mm, basitarsus: 1.5 mm, telotarsus: 1.6 mm. Total length of leg 2: 5.18 mm; femur: 0.95 mm, patella: 1.22 mm, tibia: 1.46 mm, basitarsus: 0.97 mm, telotarsus: 0.58 mm. Total length of leg 3: 5.97 mm; femur: 1.01 mm, patella: 1.43 mm, tibia: 1.58 mm, basitarsus: 1.23 mm, telotarsus: 0.72 mm. Total length of leg 4 (exposed part): 5.38 mm. Protopetidium: length = 3.26 mm, width = 2.07 mm.

The specimen studied here presents what is probably the flagellum of the chelicera, a structure present only in adult males (Figs. 6A, 6B). Although this structure is not exceptionally preserved, it seems to be a styliform flagellum with a bulbous base. The prosoma has an exterior lobe clearly separated from the main part of the propeltidium (Fig. 7A), and both features closely mirror the observations made by Selden & Shear (1996) for ceromids.

The morphology of the pedipalp, including its shape and spinulation, aligns well with the descriptions given by Selden & Shear (1996). In particular, the tarsus is short and bulbous, with the tibia-tarsus joint being distinctive (Figs. 6C, 6D, 7B). Moreover, an inverted “C” structure located toward the anterior portion of the tarsus could represent the sucker organ (Figs. 6C, 6D).

Leg I exhibits a distinct tarsal claw (Figs. 6E, 6F, 7C). The tarsal segment formula (leg II, III, and IV) identified for Cratosolpuga in this study is 1, 1, 2(3), presenting a slight difference from the 1, 1, 1 formula detailed by Selden & Shear (1996) (Figs. 7F, 8A, 8B). It is noteworthy that the added tarsal segmentation observed in leg IV is consistent with patterns in Ceromidae (1, 2, 2), though it differs from the configuration seen in leg III (Figs. 7B, 7E).

Ceromids are typified by a unique pulvillus located on a small apotele of legs II–IV. However, this characteristic was not highlighted by Selden & Shear (1996). In the specimen we studied, a clear pulvillus is visible on the left leg IV (Fig. 8C). It is essential to emphasize that the presence of a pulvillus is not exclusively diagnostic for Ceromidae, as mentioned by Selden & Shear (1996). Additionally, the dorsal positioning of the anus on the anal segment is likely due to taphonomic processes (Fig. 8D), considering that it is ventrally located in Rhagodidae, and is terminally placed in other families (Harvey, 2003).