Brazilian Miocene crabs I. Taxonomic review of Cyclocancer tuberculatus Beurlen, 1958 and Hepatella amazonica Beurlen, 1958 (Pancrustacea, Decapoda, Brachyura)

Abstract In describing Cyclocancer to accommodate his new species C. tuberculatus, Beurlen highlighted that his new genus was a cancroid intermediate in form between Atelecyclus and Cancer, but with enough characters to be placed in a new genus. Examination of C. tuberculatus type material revealed that Cyclocancer is a junior synonym of Hepatus. Therefore, C. tuberculatus is herein transferred into Hepatus as H. beurleni nomen novum because H. tuberculatus (Beurlen, 1958) is a junior homonym of H. tuberculatusSaussure (1858), a junior synonym of Hepatus pudibundus (Herbst, 1785). Miohepatus gen. nov., comb. nov. is proposed here to accommodate Hepatella amazonica Beurlen from the Pirabas Formation, Neogene, Pará, Brazil, based on new morphological evidence from the type material.


INTRODUCTION
Karl Beurlen (1901Beurlen ( -1985) ) was a German geologist and paleontologist who worked in Brazil from 1950 to 1969, invited by the Departamento Nacional de Produção Mineral (DNPM), currently Agência Nacional de Mineração (ANM).During this period, he described several fossil crustacean species, with special reference to the decapods.One of his important works was in the Pirabas Formation, north region of Brazil, where he described several Miocene species (Beurlen, 1958) in the first work describing decapod crustaceans from that formation (Aguilera & Páes, 2012).Although with undeniable importance, this work needs revision, mostly in the descriptions and updated images, with new photos of the species studied.
The revaluation of the species described by Beurlen revealed that Hepatella amazonica Beurlen, 1958, does not belong to Hepatella Smith, in Verril (1869) and cannot be assigned to an existing genus.Therefore, the new genus Miohepatus is described here with Miohepatus amazonicus as a new combination.Also, morphological evidence based on the type material from the Pirabas Formation revealed that Cyclocancer Beurlen, 1958, is a junior synonym of Hepatus Latreille, in Bosc, 1801.Cyclocancer tuberculatus Beurlen, 1958 is transferred herein into Hepatus as H. beurleni nomen novum.

MATERIAL AND METHODS
The material studied is deposited in the invertebrate paleontology collections of the Museu de Ciências da Terra (MCTer/SGB-CPRM), and Museu Nacional da Universidade Federal do Rio de Janeiro (MNRJ).The Hepatus beurleni nomen nov. is a single fossil specimen, with part and counterpart (Fig. 2).The redescription of Miohepatus amazonicus gen.nov., comb.nov. is based on the single type specimen from the Pirabas Formation.Descriptions, drawings, and photographs were made using a stereomicroscope Nikon SMZ800N equipped with camera lucida and a Leica EZ4W, both with digital camera attached.
The Brachyura classification of Ng et al. (2008) was mostly used in suprafamilial rankings.Infrafamilial rankings mostly follow De Grave et al. (2009) and Schweitzer et al. (2010).We also used the following abbreviations: cl, carapace length; cw, carapace width (taken at the widest point); coll., collector or collected by; † indicates taxa with recent and fossil representants; † † indicates taxa exclusively known from fossils; L/W, length/width ratio.
This published work and the nomenclatural acts it contains have been registered in ZooBank, the online registration system for the ICZN.Remarks: Aethridae currently embraces fossil and extant members which share similarities proposed by Guinot (1966Guinot ( , 1967) ) and they are currently positioned in Aethroidea (Ng et al., 2008;De Grave et al., 2009).Fossil specimens generally are attributed to this family by the general carapace outline and ornamentation, which is octagonal to subquadrate, wider than long; antero-and posterolateral margins clearly demarcated; and anterolateral margins entire to cristate, may be expanded to form subclypeiform structure (Davie, 2002).
A. Milne-Edwards, 1862; however, the presence of distinct elevations of the carapace in Hepatus makes possible to differentiate it from Metacarcinus.Hepatus beurleni nomen nov., has the diagnostic characters of Hepatus (as understood by Feldmann et al., 2005: 433), such as carapace much broader than long, broadly obovate, vaulted; crenated anterolateral margins; posterolateral margin concave; and posterior margin narrow.
Among the fossil species, H.The following set of characters differentiate H. beurleni nomen nov., from all of its extant congeners: 1) posterior margin narrow, crenulated, almost straight; 2) posterolateral margin strongly crenulated with a large lobe between the posterolateral and posterior margins; and 3) anterolateral margin crenated, with short lobes of about the same size.
The holotype MNRJ-4617-I had been missing for decades and was considered lost, but was recently found together in the MCTer collection, probably due to an old forgotten loan, posteriorly returned to the institution of origin (Museu Nacional do Rio de Janeiro/MNRJ).This fact allowed the specimen to escape the fire that hit the Museu Nacional in 2018 and destroyed a large part of the paleontological collections, which gives it even greater importance.
Diagnosis: As for the type species.

Etymology:
The generic name is a combination of the generic name Hepatus and Miocene in allusion to the age of the type species.
Remarks: Miohepatus gen.nov. is erected here to accommodate Hepatella amazonica Beurlen, 1958 from the early-middle Miocene, Pirabas Formation, Pará, Brazil, originally assigned to the recent genus Hepatella Smith, in Verril, 1869 (type species Hepatella amica Smith, in Verrill, 1869) from the tropical eastern Pacific Beurlen (1958) wrongly justified the assignment of his species to Hepatella by: the 1) presence of a carapace with remarkable elevations in the gastric, cardiac and branchial regions; 2) prominent rostrum; and 3) concave anterolateral margins.However, he was correct in suggesting a close relationship between Hepatella amazonica and H. peruviana, a recent species from the east Pacific coast.Accordingly, Hepatella peruviana is transferred along with H. amazonica to Miohepatus gen.nov.
The new genus differs from Hepatus in that the carapace dorsal surface is uneven, vaulted in the central region, with eight protuberances in the cardiac, gastric and branchial regions (vs carapace dorsal surface almost smooth in adults, convex, carapace regions poorly differentiated in Hepatus).
The new genus differs from Aethra Latreille, in Cuvier, 1816 [type species A. scruposa (Linnaeus, 1764)] by having an octagonal carapace and concave posterolateral margin (vs carapace elliptical and posterolateral margin slightly convex in Aethra).It differs from Actaeomorpha Miers, 1877[type species A. erosa Miers, 1877] in having three tubercles on the gastric region (vs two large elevations on the gastric region, behind each orbit in Actaeomorpha); and differs from Sakaila Manning & Holthuis, 1981[type species S. africana Manning & Holthuis, 1981] by having a crenulated posterolateral margins and carapace dorsal surface with eight protuberances (vs posterolateral margins with eight distinct teeth and carapace dorsal surface with six major protuberances in Sakaila).
Redescription: Carapace octagonal, wider than long (cw 18.5 mm, cl 15 mm; W/L ratio 1.2); greatest width at the posterolateral corner.Carapace dorsal surface uneven, vaulted centrally, concave marginally, anterior surface punctate, regions well defined.Cervical groove well defined.Rostrum partially preserved, bilobed, extending well beyond orbits; lobes separated medially by short cleft.Left orbit well preserved, small, circular, well visible in dorsal view.Carapace with eight distinct protuberances on cardiac, branchial, and gastric regions.Cardiac, mesobranchial and metagastric protuberances slightly higher than metabranchial and protogastric.Each protuberance bears a single large central granule, surrounded by some smaller ones.Anterolateral margins crenated, with 12 small rounded lobes.Posterolateral margins distinctly concave, arcuate, crenulated; shorter than anterolateral margins.Posterior margin narrow, crenulated, slightly convex.Remarks: Miohepatus amazonicus gen.nov., comb.nov. is closely related to M. peruvianus gen.nov., comb.nov.However, they differ in the presence of a small protuberance on each metabranchial region in M. amazonicus gen.nov., comb.nov., whereas the metabranchial protuberance is absent in M. peruvianus.
Miohepatus amazonicus gen.nov., comb.nov.differs from Hepatus beurleni nomen nov., comb.nov.[characters for H. beurleni within brackets] by having the carapace octagonal (vs carapace ovoid) and the carapace W/L ratio of 1.2 (vs carapace W/L ratio of 1.4).In addition, H. beurleni has a large knob-like lobe separating the posterolateral margins from the posterior margin, which is absent in M. amazonicus.Vega et al. (2009) mentioned the occurrence of M. amazonicus gen.nov., comb.nov.(as H. amazonica) from the lower Miocene of the Chiapas, Mexico.However, from the figures provided by Vega et al. (2009) it seems that their material is not attributable to M. amazonicus.The main differences are in the curvature of the posterolateral margin of the carapace, much more accentuated in M. amazonicus gen.nov., comb.nov.(Fig. 2) (vs almost straight in the specimen figured by Vega et al., 2009: pl. 1, figs. 15-18); and in the surface of the carapace with six main protuberances and two small protuberances on the metabranchial region in M. amazonicus gen.nov., comb.nov.(Fig. 2) (vs eight equally inflated elevations on the carapace in the material studied by Vega et al., 2009: pl. 1, figs. 15-18).Actually, the general carapace shape and the short bilobed rostrum of specimens from Mexico are similar to that of Eriosachila rather than to M. amazonicus gen.nov., comb.nov., although with quite different posterolateral margin (Vega et al., 2009;pl. 1).A reexamination of the material from Vega et al. (2009) is needed to clarify its taxonomic assignment.
lineatinus and H. biformis from the late Pliocene of Panama and Miocene-Pliocene of Panama and Costa Rica, respectively, differ from H. beurleni nomen nov.(characters within brackets) by having the carapace anterolateral margins tridenticulate (vs anterolateral margins with short lobes of about the same size).Hepatus guraboensis Collins, in Collins et al., 2009, from the Miocene of the Dominican Republic stands apart by having the anterolateral margins with short triangular spines increasing in size posteriorly (vs anterolateral margins strongly crenated).Hepatus nodosus Collins & Morris, 1976, from the Miocene of Trinidad differs by having the anterolateral margins divided into four blunt teeth (vs anterolateral margins strongly crenated, extending to anterior posterolateral margins).Hepatus bottomsi Blow, 2003, from the Pliocene of Virginia, United States, can be distinguished by the anterolateral margins of the carapace consisting of 40 blunt denticles situated from 12-13 or more bidentate or tridentate teeth (vs anterolateral margins consisting of simple lobes).Hepatus pauli Collins, Garvie & Mellish, 2014, from the Pleistocene of Texas, United States, differs by having 12-13 tridentate spines on the anterolateral margins (vs anterolateral margins consisting of simple lobes).Hepatus spinimarginatus Feldmann, Schweitzer & Encinas, 2005, from the Miocene of Chile differs by having the posterolateral margins rimmed by very finely beaded elevation (vs posterolateral margin strongly crenulated).