Multigene phylogeny and taxonomic revision of American shrimps of the genus Cryphiops Dana, 1852 (Decapoda, Palaemonidae) implies a proposal for reversal of precedence with Macrobrachium Spence Bate, 1868

Abstract The freshwater shrimp genus Cryphiops Dana, 1852 has a disjunct distribution in North (Mexico) and South (Brazil, Chile) America, and is composed of only six species. The current classification of genera in the Palaemonidae is controversial, based on variable morphological characters, and still far from a clear definition. Cryphiops differs from the speciose genus Macrobrachium Spence Bate, 1868 only by the absence of the hepatic spines on the carapace. Previous studies with a limited dataset suggested the necessity to link morphology and phylogeny to create an internal rearrangement in the genus to resolve the paraphyletic status. Through a molecular phylogenetic approach, the evolutionary relationships are inferred based on four (mitochondrial and nuclear) genes, among all recognized species of Cryphiops and, in combination with a taxonomic revision, a rearrangement in the systematics of the genus is suggested. The absence of hepatic spines on the carapace, the only character used to separate the genus Cryphiops, is subjective and should be considered as a homoplasy. This implies that Cryphiops and Macrobrachium are subjective synonyms and, because the latter genus is much more diverse and widely known, with several economically important species, to avoid confusion and disturbance in nomenclatural stability and keep universality, a proposal for the priority of the older synonym (Cryphiops) to be partially suppressed in favor of maintaining the prevailing use of the younger synonym (Macrobrachium) is presented. As the species of Cryphiops should be accommodated in the genus Macrobrachium, new names to replace three preoccupied specific names that, by this action, resulted to be secondary homonyms are offered.


Introduction
During the 2010s, caridean shrimp systematics has undergone considerable changes at different levels (see De Grave et al. 2015a for review of the context and literature). The speciose Palaemonidae Rafinesque, 1815 is an example of this new tendency. The family consists of a large group of decapod crustaceans comprising 151 genera and approximately 780 species (WoRMS 2021), which reached a great evolutionary success, occupying marine, estuarine, and freshwater environments. Members of this group have a long taxonomic history and it can be considered a challenge to build a more natural classification since their morphology is highly conservative (Holthuis 1950(Holthuis , 1952aPereira 1997;Murphy and Austin 2005;Pileggi and Mantelatto 2010;De Grave and Ashelby 2013). Recently, considerable efforts have been taken to solve taxonomic incongruences and accommodate taxa in a more consistent classification (De Grave et al. 2009;De Grave and Fransen 2011). After the construction of this major guideline, some important specific and complementary taxonomic initiatives were developed focusing on different taxa (for a review see De Grave et al. 2015a). Despite this significant advance, the current knowledge is not sufficient to cover the tremendous diversity of palaemonids and the many questions that remain unanswered. One of these unsolved problems is that of Cryphiops Dana, 1852, a genus composed by six recognized species distributed in North (Mexico) and South (Brazil and Chile) America (Villalobos Hiriart et al. 1989;Baldari et al. 2010). Of the six species, only Cryphiops (C.) caementarius (Molina, 1782) needs estuarine water to complete its reproductive cycle while the other five [Cryphiops (Bithynops) brasiliensis Gomes Corrêa, 1973, Cryphiops (Bithynops) luscus (Holthuis, 1973), Cryphiops (Bithynops) perspicax (Holthuis, 1977), Cryphiops (Bithynops) sbordonii Baldari, Mejía-Ortiz & López-Mejía, 2010, and Cryphiops (Bithynops) villalobosi Villalobos Hiriart, Nates Rodríguez & Cantú Díaz Barriga, 1989] are restricted to inland waters with no apparent dependency of estuarine environments.
The taxonomic reappraisal of Cryphiops showed a close relationship with Macrobrachium Spence Bate, 1868, from which Cryphiops only differs by the absence of the hepatic spines on the carapace (Holthuis 1950(Holthuis , 1952a. The absence of one or both spines was also encountered by Short (2004) in some Australian species of

Molecular data
The molecular analysis was based on partial fragments of the 16S rDNA, COI mtDNA, 18S nDNA, and H3 nDNA genes, which have been effective in solving different levels of relationships among decapod species (Schubart et al. 2000;Porter et al. 2005;Pileggi and Mantelatto 2010;Mantelatto et al. 2011;Vergamini et al. 2011;Carvalho et al. 2013Carvalho et al. , 2017Rossi and Mantelatto 2013;Álvarez et al. 2020;Robles et al. 2020). DNA extraction, amplification and sequencing protocols followed Pileggi and Mantelatto (2010). Total genomic DNA was extracted from muscle tissue of the walking legs, chelipeds, or the abdomen. The amplification by polymerase chain reaction (PCR) was conducted with the following primers: 16Sar and 16Sbr  for the 16S mitochondrial gene; COI-a and COI-f (Palumbi and Benzie 1991) for the COI mitochondrial gene; 18Sai and 18Sb3.0 (Whiting et al. 1997) for the 18S nuclear gene; H3ar and H3af (Colgan et al. 1998) for the histone (H3) nuclear gene. PCR products were sequenced with the ABI Big Dye Terminator Mix (Applied Biosystems, Carlsbad, CA) in an ABI Prism 3100 Genetic Analyzer (Applied Biosystems automated sequencer) following Applied Biosystems protocols. All sequences were confirmed by sequencing both strands. Genetic vouchers generated were deposited in the CCDB and CNCR under the catalogue numbers listed in Table 1.
In total, 88 specimens were used for the analyses, eleven belonging to Cryphiops, 71 to Macrobrachium and six to Palaemon, to obtain a robust representation of the ingroup and a consistent rooting of the phylogeny (Table 1). The selection of species composing both internal and external groups was based on the phylogenies proposed from morphological (Pereira 1997) and molecular characters (Murphy and Austin 2005;Liu et al. 2007;Pileggi and Mantelatto 2010). With this careful selection we covered all possible close and related species to Cryphiops reported previously, either by morphology and/or molecular affinities.

Taxonomic revision
The species identification was carried by us based on diagnostic morphological features in accordance with the literature (Holthuis 1952a(Holthuis , 1993Gomes Corrêa 1973;Villalobos Hiriart et al. 1989;Baldari et al. 2010). We did not list all the synonyms for Cryphiops and Macrobrachium since a complete, detailed record can be found in Holthuis (1950Holthuis ( , 1993 and De Grave and Fransen (2011). A non-exhaustive synonyms list containing post-1950 citations focused mainly on taxonomic and faunistic studies is provided for all species and it is partially based on the "Carideorum Catalogus L.B. Holthuis", an extensive reference catalogue of scientific names of shrimps gathered by the late L.B. Holthuis during his 68 years of studying Crustacea (Fransen et al. 2010), which was digitized and kindly and unpretentiously made available by C.H.J.M. Fransen to the community of carcinologists in digital format on 7 April 2020. For pre-1952 citations regarding C. caementarius, see Holthuis (1952a, b).
The morphological data considered in this review for the comparative analysis of species were as follows. Measurements: total length (tl), from the anterior portion of the rostrum to the posterior portion of telson; and carapace length (cl), from the posterior margin of the orbit to the posterior margin of the carapace. Rostrum: shape, length in relation to scaphocerite, number of teeth and their distribution on the upper and lower margins. Orbit: shape of the lower margin. Scaphocerite: size and shape. Epistome: shape and arrangement. Carapace: presence of spinules, size and arrangement of hepatic and antennal spines. Pereiopods: size and shape of the first pereiopods (P1); size, shape, and proportion of the articles of the second pereiopods (P2); size and proportion of the articles of the third, fourth and fifth pereiopods (P3 to P5). Thoracic sternum: presence and shape of the median process (T4). Abdomen: surface roughness, shape of the pleura of the fifth somite. Pleopods: ratio appendix masculina/appendix interna of the second pair (PL2). Pre-anal keel: presence and shape in the inter-uropodal sclerite. Uropods: presence of external spines. Telson: general shape, shape of the posterior margin, presence, and distribution of dorsal spines, positioning of the posterior spines in relation to the posterior margin. Other aspects such as the size of males and ovigerous females, life cycle, color, distribution, systematic position, type locality and general considerations were also considered.

Molecular approach
The concatenated phylogenetic analysis included 45 species of Palaemoninae: six belonging to Cryphiops, 36 to Macrobrachium, and three to Palaemon. A total of 35 new DNA sequences was generated in this study: seven 16S and seven COI mitochondrial sequences, ten 18S, and eleven H3 nuclear sequences. The final alignment of the four markers totalized 1,982 bp.
The estuarine C. caementarius, which has extended larval development (ELD), is nested among the species of Macrobrachium that have the same type of larval development ( Fig. 1). Similarly, the species of Cryphiops from inland waters (C. brasiliensis, C. luscus, C. perspicax, C. sbordonii, and C. villalobosi) with abbreviated larval development (ALD) are positioned in clades with species of Macrobrachium that have ALD (Fig. 1). Interestingly, the recovered phylogeny follows the previous subdivision proposed by Villalobos Hiriart et al. (1989) into two subgenera based on morphological and life-cycle characters, i.e., Cryphiops as ELD and Bithynops as ALD groups. The species of Palaemon showed a stable position in an external branch.

Family
Diagnosis (modified from Holthuis 1950Holthuis , 1952a. Body compressed, generally robust, sometimes slender. Rostrum well developed, compressed, toothed, size varying from shorter to longer than distal margin of scaphocerite. Carapace with anterolateral portion smooth or bearing numerous small spinules. Carapace armed with antennal spine; hepatic spines present in most species, branchiostegal groove present, distinct. Mandible with 3-segmented palp. All maxillipeds with well-developed exopods. Pleurobranchs on third maxilliped and all pereiopods. P1 slender. P2 more robust than other pereiopods, usually longer than entire body in adult males, left and right legs often equal in size and shape or markedly different in several species. P3-P5 with dactylus simple. P5 with propodus bearing numerous transverse rows of setae in distal part of posterior margin. PL1 with endopod much smaller than exopod, endopod of male without appendix interna. Pleon with pleurae smooth in most species or with small spinules. Uropods overreaching telson; exopod with distolateral spine, endopod unarmed. Telson elongate, subtriangular, narrowing posteriorly, with two pairs of dorsal spines, anterior pair placed in middle, posterior pair usually placed midway between anterior pair and posterior margin; posterior margin ending in sharp median point, flanked by two pairs of spines, outer pair usually shorter than inner one, inner pair overreaching apex of telson in most species. See detailed description in Holthuis (1950).
Remarks. See Discussion. Description. Rostrum. Short, straight, reaching slightly beyond first third of third article of antennular peduncle; upper margin with 6-9 regularly spaced teeth, first one behind posterior edge of orbit; lower margin with 1-3 teeth.

Carapace.
Smooth; antennal spine small, slightly overreaching lower portion of orbit; hepatic spine absent. Lower orbital angle obtuse, moderately pronounced.
Pereiopods. P1: slender, reaching with distal third of carpus beyond scaphocerite; carpus slightly longer than merus; chelae 0.68 length of carpus. P2: moderately robust, with small spines, equal in form and size, reaching with distal third of merus beyond scaphocerite; ischium 0.75 length of merus; merus as long as carpus; carpus as long as palm, with slight basal constriction; propodus 2.5 × as long as dactylus, and 1.6 × as long as carpus; palm compressed, nearly 5 × as long as high; fingers 0.62 length of palm, with numerous spinules, not gapping, tips crossing, cutting edge with 3-6 teeth on proximal third in both fingers. P3-P5 with all joints covered with rows of small spinules. P3 reaching with entire dactylus beyond scaphocerite, propodus 2.5 × as long as dactylus, propodus nearly 2 × as long as carpus, propodus slightly longer than merus. P4 reaching with entire dactylus beyond scaphocerite, propodus 3 × as long as dactylus, nearly 2 × as long as carpus, propodus slightly longer than merus. P5 reaching with half-length of dactylus end of scaphocerite, propodus 3 × as long as dactylus, propodus nearly 2 × as long as carpus, propodus slightly longer than merus.
Pleopods. PL2 with appendix masculina less than 2 × length of appendix interna. Uropods. Exopodite with mobile spines as long as spiniform projection of outer margin.
Telson. Broad, smooth, slightly longer than abdominal somite 6, bearing two pairs of dorsal spinules close to posterior margin. Posterior margin ending in moderately acute triangular point; two pairs of posterior spinules with several plumose setae, inner pair overreaching distal margin of telson.
Etymology. Villalobos Hiriart et al. (1989) dedicated this species to Dr. Alejandro Villalobos Figueroa, eminent Mexican carcinologist and founder of the CNCR. We maintain this homage by just adding the first part of his name to the specific epithet.
Size. See in material examined.
Life cycle. Exclusive of inland waters, therefore independent of brackish waters to complete its life cycle. The eggs are few and large: 1.3-2.4 mm (Villalobos Hiriart et al. 1989). Its larval development is not known but given the characteristics of the eggs, it should be abbreviated, following the same pattern of congeners inhabiting continental waters (Magalhães and Walker 1988;Pereira and García 1995).
Remarks. The name Macrobrachium villalobosi was used by Hobbs Jr. (1973) for a new species from Mexico. Villalobos et al. (1989) used the same name for a new species of Cryphiops (Bithynops) also from Mexico. Since the synonymization of both genera makes these specific names secondary homonyms, Macrobrachium alevillalobosi is proposed as a replacement name for Cryphiops (Bithynops) villalobosi Villalobos Hiriat, Nates Rodríguez & Cantú Díaz Barriga, 1989.
Macrobrachium alevillalobosi nom. nov., comb. nov. differs from M. candango nom. nov. and M. perspicax comb. nov. mainly in the form, size, and proportion of the articles of the second pereiopod ( Table 2). The chelipeds are long and similar in size and shape, overreaching the scaphocerite with distal third of the merus; the ischium is shorter than merus; the palm is long and cylindrical, almost five times as long as high, and the dactylus is 0.62 times the length of the palm. Finally, M. alevillalobosi nom. nov., comb. nov. is the only species of this group in which the appendix masculina is almost as long as the endopod of the second pleopod. ( Description. Rostrum. Straight, short, nearly reaching first article of antennular peduncle; upper margin with 6-8 teeth, regularly spaced, one and/or two behind posterior margin of orbit; lower margin with 0-4 teeth.
Pereiopods. P1 slender, reaching with most of carpus beyond scaphocerite; fingers slightly longer than palm; carpus slightly shorter than chelae; ischium and merus distinctly spinulated; carpus and chelae smooth. P2 strong, with many spines, strong heterochely; largest cheliped reaching with half-length of merus beyond scaphocerite; ischium larger than half-length of merus; merus longer than carpus; carpus short, slightly shorter than half length of palm, with strong basal constriction; propodus 2.1 × as long as dactylus, 3.3 × as long as carpus; palm slightly inflated, more than 2.3 × as long as high; fingers shorter than palm, with numerous small spinules, cutting edges with 4-7 denticles of equal size. P3-P5 smooth, except for sparse setae and spinules along lower margin of propodus; propodus nearly 2 × as long as carpus; propodus slightly shorter than merus; P3 reaching with half-length of dactylus beyond scaphocerite, propodus 2 × as long as dactylus; P4 reaching with tip of dactylus end of scaphocerite, propodus 1.5 × as long Smooth. Somite 5 with posteroventral angle of pleuron acute; somite 6 slightly longer than somite 5. Inter-uropodal sclerite with strong, keel-shaped pre-anal carinae.
Pleopods. PL2 with appendix masculina 2 × as long as appendix interna. Uropods. Exopodite with mobile spines slightly longer than spiniform projection of outer margin.
Telson. Broad, smooth; 1.5 × as long as abdominal somite 6, bearing 2 pairs of dorsal spinules, first pair located in middle of telson, second pair located ¾ of length of telson. Posterior margin rounded, ending in truncated tip, with several plumose setae and two pairs of posterior spinules, inner pair not reaching end of telson.
Size. See in material examined.
Color. Yellowish green with light brown spots dorsally. P2 with reddish joints and greenish blue color.
Life cycle. Exclusive of coastal waters, dependent of brackish waters to complete its life cycle. The eggs are numerous and small: 0.43-0.62 mm of major diameter (Norambuena 1977;Yávar and Dupré 2007;Bazán et al. 2009). The larval development is long, with many free-swimming larval stages , following the usual pattern of coastal palaemonid species.
Remarks. For the heterochelia, the robustness and strong shape, as well as the ornamentation of the second pereiopod, M. caementarius comb. nov. is comparable with M. hancocki Holthuis, 1950, andM. occidentale Holthuis, 1950 from the Pacific slope. The species is morphologically similar to M. heterochirus (Wiegmann, 1836) from the Atlantic slope, particularly concerning the shape of the rostrum, carapace, and telson.
Macrobrachium candango nom. nov., comb. nov. Description of the holotype. Rostrum. Moderately high, nearly straight, distal end slightly directed upwards, reaching end of antennular peduncle, and little before the distal margin of scaphocerite; upper margin convex over orbit, with seven teeth, first and sometimes the second, slightly behind posterior edge of orbit; lower margin with one tooth.
Pereiopods. P1 reaching with almost half length of carpus beyond scaphocerite; fingers as long as palm; carpus 1.5 × as long as chelae, 1.5 × as long as merus; articles with scattered setae, fingers with tufts of setae. P2 similar in shape, different in size; largest one reaching with distal portion of merus beyond scaphocerite; smallest one reaching with distal end of carpus beyond scaphocerite, with fingers as long as palm; all articles with sparse setae and spines. Larger cheliped with ischium nearly as long as merus, with spinulation as in palm; merus as long as carpus, swollen, with spinulation as in palm; carpus slightly shorter than palm, swollen, with strong basal constriction; spinulation as in palm; propodus 2.5 × as long as dactylus, 2 × as long as carpus; palm with upper surface slightly compressed, somewhat swollen, covered with spinules, nearly 3 × as long as high; fingers 2/3 as long as palm, with numerous spinules; cutting edge of dactylus with large tooth in proximal third, slightly lower tooth in between large tooth and proximal part; cutting edge of fixed finger with tooth opposing two teeth of dactylus, with row of three denticles between proximal part and this tooth. P3-P5 smooth, except for sparse setae and spinules along lower margin of propodus; propodus nearly 2 × as long as carpus; propodus slightly shorter than merus; P3 reaching with half-length of dactylus beyond scaphocerite, propodus 2 × as long as dactylus; P4 reaching with tip of dactylus end of scaphocerite, propodus 2.5 × as long as dactylus; P5 reaching with tip of dactylus half-length of scaphocerite, propodus 2.5 × as long as dactylus.
Pleopods. PL2 with appendix masculina 2 × as long as appendix interna. Uropods. Exopodite with mobile spines slightly shorter than spiniform projection of outer margin.
Telson. Broad, smooth, 1.5 × as long as abdominal somite 6, bearing two pairs of dorsal spinules, first pair slightly behind middle portion of telson, second pair located closer to first pair than to posterior margin. Posterior margin distinct, ending in acute point, with several plumose setae and two pairs of posterior spinules, inner pair reaching end of telson.
Etymology. The specific epithet brasiliensis was used by Gomes Corrêa (1973) to refer to the type locality of the species, Brasília, the capital of Brazil. To keep that intention, we rename the species using the word candango, a demonym referring to those who are native to Brasília.
Size. See in material examined.
Color. From colorless to light brown, with dark brown carapace, mimicking the color of the substrate where they inhabit.

Distribution. Endemic of inland waters from Central Brazil (Distrito Federal) (Gomes Corrêa 1973; present paper).
Life cycle. Exclusive of inland waters, therefore independent of brackish waters to complete its life cycle. The fecundity is low, 38-61 eggs, and the eggs are large, their volume ranged from 4.41 to 7.71 mm 3 (Nogueira et al. under revision). Its larval development is not known but given its fecundity and egg size, it should be abbreviated, following the same pattern of congeners inhabiting continental waters (Magalhães and Walker 1988;Pereira and García 1995).
Remarks. Gomes Corrêa (1973) named Cryphiops brasiliensis a species from the vicinities of Brasília, Brazil. This specific epithet, however, was already used by Heller (1868) for a species of Macrobrachium described from the state of Mato Grosso, Brazil. With the synonymization of both genera, these specific names become secondary homonyms. We, therefore, propose the name Macrobrachium candango nom. nov., comb. nov. as a replacement name for Cryphiops brasiliensis Gomes Corrêa, 1973.
We examined specimens from three lots used by Gomes Corrêa (1973) to describe C. brasiliensis and deposited at the MNRJ: the holotype (MNRJ 903: male, cl 18.2 mm) and two others labeled as allotype (MNRJ 6464: 1 ovigerous female, cl 15.6 mm) and paratypes (MNRJ 2668: 1 male, cl 17.9 mm, 2 females, cl 15.3 and cl 15.3 mm), although the author did not explicitly designate the latter two as type material. We had this material on loan, which was returned to MNRJ in July 2008. After the fire at the Museu Nacional do Rio de Janeiro in September 2018, the lot MNRJ 2668 is missing, but the other two, including the holotype, preserved in alcohol, are safe and in good condition (I.A. Cardoso, curator of Crustacea, pers. comm. to FLM, Nov 2020). When carrying out aquatic surveys in the region around the type locality, we (FLM, LGP) visited the Reserva Ecológica do IBGE (Brasília, DF) and found a well-preserved collection of specimens (> 260, not listed herein) made during previous aquatic faunistic surveys in the area (Takahashi et al. 2019). The main area of occurrence of this species is in a protected reserve, which may avoid possible impacts. This species was classified in the IUCN's Data Deficient (DD) category (Mantelatto et al. 2016). However, due to anthropic pressures in the region, future monitoring is necessary to evaluate its conservation conditions. (Holthuis, 1973) Description. Rostrum. Short, directed slightly downwards, tip directed slightly upwards, reaching or slightly overreaching joint between second and third article of antennular peduncle, and at level of distal third of scaphocerite; upper margin convex over orbit, with 5-8 teeth regularly spaced, first over or slightly behind posterior edge of orbit; lower margin with none or one tooth.
Pereiopods. P1 slender, reaching with nearly entire chelae beyond scaphocerite; fingers slightly longer than palm; chelae 2/3 length of carpus. P2 moderately robust, with several spines, equal in form and size, reaching with proximal third of carpus beyond scaphocerite; ischium evidently shorter than merus; merus as long as carpus; carpus as long as palm, with basal constriction; propodus 2 × as long as dactylus, 2 × as long as carpus; palm inflated, nearly 3 × as long as high; fingers as long as palm, with numerous small spinules; cutting edge with two denticles of same size in both teeth. P3-P5 with all joints covered with row of small spinules on the lower margin; P3 reaching with entire dactylus beyond scaphocerite, propodus 2 × as long as dactylus, propodus slightly longer than merus; P4 reaching with tip of dactylus end of scaphocerite, propodus 3 × as long as dactylus, propodus slightly longer than merus; P5 reaching with tip of dactylus half-length of scaphocerite, propodus 3 × as long as dactylus, propodus slightly longer than merus. Pleon.
Pleopods. PL2 with appendix masculina 2 × as long as appendix interna. Uropods. Exopodite with mobile spines as long as spiniform projection of outer margin. Telson. Broad, smooth, slightly longer than abdominal somite 6, bearing two pairs of dorsal spinules closer to posterior margin of telson. Posterior margin ending in moderately acute triangular point, with several plumose setae and two pairs of posterior spinules, inner pair overreaching end of telson.
Size. See in material examined.
Color. Whitish to transparent. Type locality. México, Chiapas, Municipality of La Trinitaria, Gruta del Arco, El Rancho de San Rafael Del Arco, Lagunas de Montebello, altitude 1,470 m. Recent visits to the type locality showed an increasing contamination in the lakes that supply water to the underground stream of the Gruta del Arco, and the collections of specimens were not successful, at least in the closest access to the water pools. Possibly, M. luscus comb. nov. is seriously threatened.
Distribution. Only known from the type locality (Holthuis 1973;present paper). Life cycle. This is a cave species exclusive of inland waters, therefore independent of brackish to complete its life cycle. The eggs are few and large: 1.8-2.4 mm (Villalobos Hiriart et al. 1989). The duration of the embryonic development is probably long and with few larval stages following the pattern of other inland species.
Pereiopods. P1 slender, reaching with entire chelae or small part of carpus beyond scaphocerite; fingers slightly longer than palm; chelae 2/3 length of carpus. P2 moderately robust, with spines, equal in form and size, reaching with proximal third of carpus beyond scaphocerite; ischium evidently shorter than merus; merus as long as carpus; carpus as long as palm, with basal constriction; propodus 2.2 × as long as dactylus, 2 × as long as carpus; palm inflated, nearly 3 × as long as high; fingers slightly shorter (0.8) than palm, with numerous small spinules, not gaping, tips crossing, cutting edges with two similar denticles closer to proximal portion. P3-P5 with all joints covered with row of small spinules on lower margin; P3 reaching with entire dactylus beyond scaphocerite, propodus 2 × as long as dactylus, propodus nearly 2 × as long as carpus, propodus slightly longer than merus; P4 reaching with tip of dactylus end of scaphocerite, propodus 3 × as long as dactylus, propodus nearly 2 × as long as carpus, propodus slightly longer than merus; P5 reaching with tip of dactylus half-length of scaphocerite, propodus 3 × as long as dactylus, propodus nearly 2 × as long as carpus, propodus slightly longer than merus.
Pleopods. PL2 with appendix masculina nearly 2 × as long as appendix interna. Uropods. Exopodite with mobile spines as long as spiniform projection of outer margin. Telson. Broad, smooth, slightly longer than abdominal somite 6, bearing two pairs of dorsal spinules close to posterior margin of telson. Posterior margin ending in moderately acute triangular point, with several plumose setae and two pairs of posterior spinules, inner pair overreaching end of telson.
Size. See in material examined.
Distribution. Only known from the type locality (Holthuis 1977;present paper). Life cycle. Exclusive of inland waters, therefore independent of brackish waters to complete its life cycle. The eggs are few and large: 1.9-2.5 mm (Villalobos Hiriart et al. 1989). Its larval development is not known but given the characteristics of the eggs, it should be abbreviated, following the same pattern of congeners inhabiting continental waters (Magalhães and Walker 1988;Pereira and García 1995).
Remarks. Among the epigean forms of this group of species with abbreviated development and without hepatic spine, M. perspicax comb. nov. can be distinguished from M. candango nom. nov., comb. nov. and M. alevillalobosi nom. nov., comb. nov. by the total length of the body, and by the similar form and size of the second pereiopod and the proportion of its articles ( Table 2). Specimens of M. perspicax comb. nov. are generally smaller (31.1-43.3 mm) than those of the other two species; the second pereiopods are shorter, do not present heterochely like M. candango nom. nov., comb. nov. and the chelae are slender, the palm is 3 × as long as high, and the dactylus is slightly shorter.

Cryphiops sbordoni
Cephalon. Eyes reduced, globular cornea with facets, pigmented area reduced to a black point. Scaphocerite 2.4 × as long as wide.
Telson. Nearly 1.5 × longer than abdominal somite 6, shorter than uropodal rami, bearing two pairs of dorsal spines, first pair on distal fifth, second pair on middle section, with a single spine in the middle on left side; posterior margin broadly triangular bearing two pairs of lateral spines, inner pair 5 × longer than outer one, with plumose setae between inner spines, center ending in acute tip.
Etymology. Baldari et al. (2010) named this species in honor of Prof. Valerio Sbordoni, a studious of the cave fauna of Chiapas, Mexico, and collector of the specimens. We maintained this homage by forming the specific epithet with parts of his first and last name.
Size. See in material examined.
Color. Live specimens are white, without pigment in/on the body. Type locality. Mexico Chiapas, Las Margaritas, Cueva Chamburro. Distribution. Only known from the type locality (Baldari et al. 2010). Life cycle. Stygobitic species exclusive of inland waters, therefore independent of brackish waters to complete its life cycle. Female allotype with eggs (not measured).
Remarks. Mejía-Ortíz et al. (2008) described Macrobrachium sbordonii from Mexico, naming it after Valerio Sbordoni. Shortly thereafter, Baldari et al. (2010) pay homage to the very same person again by describing a new species of Cryphiops also from Mexico. Since the synonymization of both genera makes the names secondary homonyms, Macrobrachium valdonii nom. nov., comb. nov. is proposed as a replacement name for Cryphiops sbordonii Baldari, Mejía-Ortiz & López-Mejía, 2010. Similar to M. luscus comb. nov. (see remarks of that species and Table 2).

Taxonomic issues
The phylogenetic analysis presented here, including freshwater prawns of the genus Cryphiops and species of Macrobrachium from four different geographic regions revealed that they form an unnatural group inside the Palaemonidae. All the species of Cryphiops, however, were considered valid taxonomic entities and all of them were recovered in the proper group of Macrobrachium species in terms of distribution and type of larval development. Macrobrachium caementarius comb. nov. was consistently recovered associated to species with an estuarine affinity, supporting the taxonomic similarity showed in the phylogenetic analysis. The endemic species from Mexico, Macrobrachium luscus comb. nov., M. perspicax comb. nov., M. valdonii nom. nov., comb. nov., and M. alevillalobosi nom. nov., comb. nov., appear to have a joint position, always close to the species of Macrobrachium from Mexican inland waters (Fig. 1), which confirms the phylogenetic relationships among the four species. Similarly, the endemic species from central Brazil, M. candango nom. nov., comb. nov. is related to species of Macrobrachium also endemic to Brazil, in particular M. iheringi (Fig. 1), with a high degree of morphological similarity between these species.
The results of the taxonomic analysis of the species of Cryphiops corroborated the findings reported by Holthuis (1950Holthuis ( , 1952a, who listed a series of morphological and biological reasons to explain why the taxonomy of the genera within the family Palaemonidae is considered of difficult resolution and deserved more refined studies. Therefore, it is not surprising that the current systematics of the group used so far exhibited several inconsistencies at both the generic and specific levels, such as those already reported for other species when molecular analysis were contrasted with morphologically based classifications Austin 2002, 2003).
The morphological character used to define Cryphiops is clear and easily discernible: "This genus differs from Macrobrachium, with which it often is united, mainly by the absence of the hepatic spine on the carapace" (see Holthuis 1952a: 136). That is, the only synapomorphy separating the two genera is the absence of the hepatic spine in Cryphiops. In accordance with Short (2004), the presence or absence of a hepatic spine is a doubtful character in Palaemonidae because it sometimes can be absent from one or both sides in specimens of Macrobrachium. Therefore, this single character used to separate Cryphiops is subjective, and its usefulness should be reconsidered. Clearly, the absence of the hepatic spine refers to a case of homoplasy, in which the independently acquired apomorphies do not represent phylogenetic proximity. In this case, two hypotheses can be considered: 1) parallelism, losing the hepatic spine independently in the two lineages from a plesiomorphic with-hepatic-spine state, or 2) reversal, when the apomorphic state (absence of hepatic spine) becomes similar to the previous plesiomorphic state (absence of hepatic spine) present in the ancestor of the group. From a parsimony point of view, however, we believe that the first hypothesis seems more plausible, i.e., an independent loss of the hepatic spine that was propagated from generation to generation in different populations.

Nomenclatural issues
The obtained concatenated topology (Fig. 1) shows that there is high genetic similarity among the species of Macrobrachium and Cryphiops, coinciding with several previous studies that suggested that the latter should be part of Macrobrachium (Pereira 1997;Porter et al. 2005;Page et al. 2008;Pileggi and Mantelatto 2010). Following these studies, the robust results obtained here, considering all species of Cryphiops and almost all of the Neotropical species of Macrobrachium, corroborate the paraphyletic nature of these groups and indicate that the current classification should be amended accordingly. In this way, as De Grave and Ashelby (2013: 341) pointed out, such amendment will induce a nomenclatural issue regarding the priority of the names Cryphiops / Macrobrachium, a situation that demands extra caution and that will require an evaluation by the International Commission on Zoological Nomenclature (ICZN). The name Cryphiops Dana, 1852precedes Macrobrachium Spence Bate, 1868 and, if the Principle of Priority is strictly followed, the former should have priority over the latter (ICZN 1999, Art. 23). However, Macrobrachium is a much more speciose genus with many species of economic interest and importance, and extensively cited in the scientific literature. Therefore, a change in the generic name that at present is very well known would certainly cause taxonomic confusion and nomenclatural instability. The provisions of the Article 23.9.1 of the Code for a Reversal of Precedence cannot be applied in this case because the older synonym (Cryphiops) was used as a valid name after 1899 (see synonymic list under Macrobrachium). We, nevertheless, invoke the provision of Article 23.9.3 to propose herein that the younger synonym (Macrobrachium) keeps the priority over the older one. An application to the International Commission on Zoological Nomenclature to suppress the priority of Cryphiops and rule this proposal of Reversal of Precedence is being concurrently prepared. We also suggest that the prevailing use of the name Macrobrachium is maintained while the matter is under consideration by the Commission (ICZN 1999, Art. 82). Meanwhile, those who believe the taxa to be distinct could still use both names (L.B. Holthuis, pers. comm. to FLM on 27 Nov 2007). The arguments to support this proposal are detailed below.
In an essay on Chile's natural history, Molina (1782) introduced "Cancer caementarius" to name a freshwater shrimp abundant in the rivers of that country.
The genus remained monotypic for more than 120 years until Gomes Corrêa (1973) described Cryphiops brasiliensis, endemic to central Brazil. In that same year, Holthuis (1973) erected Bithynops to include a new cave species from Mexico, Bs. luscus. Soon after, Holthuis (1977) included another new species from Mexico in this genus: Bithynops perspicax. Subsequently, in a review of the genera Cryphiops and Bithynops, Villalobos Hiriart et al. (1989) proposed the synonymization of both taxa based on the fragility of the characters used to separate them (e.g., eyes with reduced cornea in Bithynops), but kept both taxa with subgeneric status. They retained C. caementarius under Cryphiops s. s., moved C. brasiliensis, Bs. luscus, and Bs. perspicax into Cryphiops (Bythynops), in addition to describing a new species, Cryphiops (Bithynops) villalobosi Villalobos Hiriart, Nates Rodriguez & Diaz Cantú, 1989. Later, Baldari et al. (2010 described a new cave species from Chiapas, Mexico, named Cryphiops sbordonii Baldari, Mejía-Ortiz & López-Mejía, 2010. It is noteworthy that Holthuis (1993), in his robust review of the caridean genera, did not follow this subgeneric arrangement, which is widely accepted (De Grave and Fransen 2011; WoRMS 2021).
The genus Macrobrachium was erected by Spence Bate (1868) to accommodate four species with males presenting "immensely developed" second pair of pereiopods without, however, designating a type species. This was subsequently done by Fowler (1912), who chose an American species, Macrobrachium americanum Spence Bate, 1868, as the type species.
Holthuis and Ng (2010) gave a historical overview of the nomenclatural situation of the name Macrobrachium, in particular regarding the confusing usage of the names Macrobrachium and Palaemon Weber, 1795, which led the matter to be ruled by the International Commission of Zoological Nomenclature in the Opinion 564 (ICZN 1959). Due to the very conservative nature of the morphological traits used to differentiate this group of palaemonid shrimps both to generic and specific ranks, the taxonomic status of Macrobrachium has undergone several changes, especially until the first half of the 20 th century. Spence Bate (1868) confessed his hesitation in creating the new genus, since he did not perceive any structural differentiation separating the new species of Macrobrachium from those of Palaemon but considered that the extremely long P2 would be a strong evidence that both taxa formed a natural classification. Shortly thereafter, the author did not follow his own system and treated Macrobrachium as a junior synonym of Bithynis (see Spence Bate 1888: 788). Ortmann (1891), based on characters of the P2 (shape of the palm and length ratio between carpus and merus), split up Palaemon into four subgenera: Eupalaemon Ortmann, 1891; Brachycarpus Spence Bate, 1888; Parapalaemon Ortmann, 1891; and Macrobrachium. His system was followed by Coutière (1901), but not by Stebbing (1908), who, in view of the inconsistency of such arrangement, argued that the retention of the name Macrobrachium was not justified and replaced it with Macroterocheir, a genus defined by one of the chelipeds of the second pair being exceedingly longer than the other. Henderson and Matthai (1910) found the subgeneric arrangement of doubtful utility, since those characters were age dependent, and kept all species under the genus Palaemon. Holthuis (1950Holthuis ( , 1952a presented a comprehensive discussion on the difficulties of studying the taxonomy of this group regarding the few useful differential characters and their large variability individually, ontogenetically or between the sexes. Holthuis (1950: 104) also considered the subgeneric division unfeasible, as it could lead to confusion, and treated Macrobrachium as a unity.
Since Holthuis' revision (1952a) of the American Palaemoninae and, particularly, after the Opinion 564 (ICZN, 1959), the taxonomic and nomenclatural status of the genus has remained stable. As a pantropical and subtropical genus occurring in a wide variety of habitats, the number of species from around the world added or described in it grew so rapidly that 41 years after his revision, Holthuis (1993) himself remarked that "it is now a quite respectable generic name". Today, the genus is one of the most speciose of the infraorder Caridea, with 243 valid species until 1 June 2011 (De Grave and Fransen 2011) and 259 until 5 Jan 2021 (WoRMS 2021), with this number varying either due to the frequent addition of new species (e.g., Mejía-Ortíz and López-Mejía 2011; Unnikrishnan 2012, 2013;Pillai et al. 2014Pillai et al. , 2015Fujita et al. 2015;Cai and Vidthayanon 2016;Lan et al. 2017;Saengphan et al. 2018Saengphan et al. , 2019Xuan 2019;Zheng et al. 2019;Rossi et al. 2020;Siriwut et al. 2020;Zhu et al. 2020;Myo et al. 2021) or due to synonymization or revalidation of species (e.g., Pileggi and Mantelatto 2012;Castelin et al. 2017).
The high diversity and worldwide tropical-subtropical distribution of Macrobrachium, combined with the scarcity of morphologic characters for accurate generic and specific delimitation, has long been intriguing taxonomists regarding its systematics, phylogenetic affinities, and biogeographic patterns. Several studies have been published on these topics using both morphological and molecular data, and, more recently, applying integrative approaches (Pereira 1997;Bowles et al. 2000;Austin 2003, 2005;Short 2004;Hernández et al. 2007;Liu et al. 2007;Valencia and Campos 2007;Wowor and Ng 2007;Parhi et al. 2008;Wowor et al. 2009;Pileggi and Mantelatto 2010;Acuña Gómez et al. 2013;Rossi and Mantelatto 2013;Pileggi et al. 2014;Jose et al. 2016;Jose and Harikrishnan 2019;Mokambu et al. 2019;Molina et al. 2020). Among other factors, the high number of species has been hampering a comprehensive study on the phylogeny of the genus, but several articles were published on this at a regional level, either based on American (e.g., Pileggi and Mantelatto 2010;Acuña Gómez et al. 2013;Rossi and Mantelatto 2013;Pileggi et al. 2014), African (e.g., Mokambu et al. 2019), or Indo-West Pacific species (e.g., Murphy and Austin 2005;Liu et al. 2007;Parhi et al. 2008;Chen et al. 2009;Wowor et al. 2009;Jose and Harikrishnan 2019;Siriwut et al. 2020). As one of the most conspicuous constituents of the aquatic fauna in estuarine and continental aquatic environments, a multitude of studies on the biology, ecology, reproduction, development, and physiology of many of its species have already been published. Jayachandran (2001) and Anger (2013) made a comprehensive review on the biology, ecology, and biogeography of Macrobrachium (see also the references therein).
The large size, high fecundity, and abundance of some species of the genus have made them an economically valuable fisheries and aquaculture resource and, consequently, numerous scientific and technical publications on different aspects related to their culture and fisheries have been made around the world (see New and Valenti 2000;Jayachandran 2001;New et al. 2008New et al. , 2010. Macrobrachium rosenbergii (De Man, 1879) and Macrobrachium nipponense (De Haan, 1849[in De Haan 1833-1850) are the most commercially important species, but other species of Macrobrachium have also been used for aquaculture or studied as potentially cultivable species (New and Valenti 2000;Jayachandran 2001;New et al. 2008New et al. , 2010Hongtuo et al. 2012;New and Mohanakumaran Nair 2012;FAO 2020). Holthuis and Ng (2010), considering the circumtropical, disjunct geographic distribution of this highly diverse group, raised doubts as to whether the genus would form a monophyletic clade. To this regard, we included eight Asian and two African species of Macrobrachium (Table 1); however, they were recovered either nested within American species or well within what is considered the genus Macrobrachium (Fig. 1). Although our study is limited to the available sequences and species that we were able to analyze and sequence, it contributes to the assumption that the genus is monophyletic and is supported by a multigene analysis. Other studies using molecular approaches, but also including a limited number of representatives either with preponderance of Indo-Pacific species (Murphy and Austin 2005;Liu et al. 2007;Parhi et al. 2008;Wowor et al. 2009;Jose and Harikrishnan 2019) or focused on American species (Pileggi and Mantelatto 2010;Acuña Gómez et al. 2013;Rossi and Mantelatto 2013;Pileggi et al. 2014;García-Velazco et al. 2017, have also pointed to a monophyletic status of the genus. Anger (2013) assumed that all Macrobrachium species originated from the same ancestor in proposing a robust scenario for explaining the origin, evolutionary history, and modern biogeography of the genus. Assuming that it is indeed monophyletic and considering that the type species of Macrobrachium is an American species, then our proposal of Reversal of Precedence of Macrobrachium over Cryphiops, if so ruled by the ICZN, should not affect the status and situation of the African and Indo-West Pacific species of Macrobrachium. On the other hand, if future, more comprehensive studies including a large number of worldwide representatives of the genus will eventually not corroborate its monophyly, then the taxonomic and nomenclatural situation of the non-American species might become somewhat complicated. Among the other generic names available, Eupalaemon Ortmann, 1891 cannot be used because its type species, designated by Holthuis (1955), is Macrobrachium acanthurus (Wiegmann, 1836), a well-established American species. If the African and Asian species constitute a separate clade, then Parapaleomon Ortmann, 1891 would be the name to be used, as Holthuis (1955) established its type species as being Macrobrachium dolichodactylus Hilgendorf, 1879, a species from the eastern coast of Africa (Mozambique). If, however, the results of such a study pose more atomized groups, the introduction of new generic names for those clades might be necessary, since the type species of Macroterocheir Stebbing, 1908, the only other name available for this group, is Macrobrachium lepidactylus Hilgendorf, 1879 (designated by Holthuis 1955), also from Mozambique.

Conclusions
Our phylogenetic analysis of all species of Cryphiops, including species of Macrobrachium from America, Africa, and the Indo-Pacific, using morphological and multigene approaches in combination with a taxonomic revision, revealed that the morphological character used to separate the genus Cryphiops is subjective and homoplasic, and that all Cryphiops species are nested within Macrobrachium. Such results corroborate the assumption about the monophyly of the genus Macrobrachium, which implies that Cryphiops Dana, 1852 andMacrobrachium Spence Bate, 1868 are subjective synonyms and, as a consequence, three specific secondary homonyms are established: M. brasiliense (Heller, 1862) × C. brasiliensis Gomes Corrêa, 1973M. villalobosi Hobbs Jr, 1973 × C. (Bithynops) villalobosi Villalobos Hiriart, Nates Rodríguez & Cantú Díaz Barriga, 1989;andM. sbordonii Mejía-Ortíz, Baldari &López-Mejía, 2008 × C. sbordonii Baldari, Mejía-Ortiz &López-Mejía, 2010. We therefore present a systematic rearrangement in which all species of Cryphiops are included in Macrobrachium and introduce replacement names for the three resulting specific secondary homonyms.
The available genetic data argues for the synonymy of Macrobrachium Spence Bate, 1868 under Cryphiops Dana, 1852. Considering the large number of species under both names and the fact that they have a pan-tropical distribution, it is likely this taxonomy may be challenged by new genetic techniques and finer morphological analyses. To change the generic names at this stage would be very disruptive, resulting in nomenclatural instability and causing confusion for other researchers, especially since there are several economically important species (notably Macrobrachium rosenbergii). Moreover, many species are also important in conservation efforts and used for a wide variety of biological studies in many parts of the world. Therefore, until a larger data set can be assembled, we recommend maintaining the status quo with regards to the generic names, i.e., use Macrobrachium sensu lato and restrict the use of Cryphiops for C. caementarius (Molina, 1782) and its immediately allied species. Under the current code (ICZN 1999: Arts. 23.9.3, 81.2.2), the senior synonym (Cryphiops) should be partially suppressed in favor of maintaining the prevailing use of the junior synonym (Macrobrachium) under the provision of the Article 82 of the Code (ICZN 1999). In this sense, an application is concurrently being prepared to the ICZN for using their Plenary Powers to partially suppress the priority of the name Cryphiops over the name Macrobrachium and rule a case of Reversal of Precedence regarding these names.

Acknowledgements
The present study is part of a long-term project started ten years ago to evaluate the taxonomy of freshwater decapods in America, and was completed as a multidisciplinary and collaborative research effort under financial support to FLM from Fundação de Amparo à Pesquisa do Estado de São Paulo -FAPESP (Temáticos Biota 2010/50188-8 andINTERCRUSTA 2018/13685-5;Coleções Científicas 2009/54931-0;PROTAX 2016/50376-5). Additional support was obtained by the Coordenação de Aperfeiçoamento de Pessoal de Nivel Superior -CAPES Código de Financiamento 001 (Auxílio N° 2823/2013, Processo N° 23038.009263/2013Ciências do Mar II, 2005-23038.004308/2014 and Conselho Nacional de Desenvolvimento Científico e Tecnológico - CNPq -Brazil (473050/2007-2, 471011/2011490314/2011, and2504322/2012-5) to FLM. LGP and JAFP were supported by post-doctoral fellowships (Proc. 02630/09-5 -PNPD/CAPES and Proc. 151105/2019-7 -CNPq/PDJ, respectively). CM and FLM thank the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq -Brazil) for research grants (308968/2018-2 and 302253/2019-0, respectively). FA gratefully acknowledges the financial support obtained through grant IV200319 "Área Experimental de Lagos Tropicales" from PAPIIT-DGAPA-UNAM to conduct fieldwork. We thank Barbara Schmidt, Dione Seripierri, K.V. Jayachandran, Mauro L.B. Ribeiro, Michel Hendrickx, Valerio Sbordoni, and Wagner C. Valenti for kindly helping us to obtain some difficult-to-find references. We are deeply indebted to Peter Ng for all the help and support regarding taxonomical and nomenclatural rules and procedures, as well as assistance with literature. We are also grateful to many colleagues and friends for their help in collections (Antonio Baeza, Ray Bauer, Juan Bolaños -in memoriam, Charles Fransen, Luis Pardo, Ingo Wehrtmann) for making available some essential fresh specimens, for lending material from collections used in our research, and for critical discussion during the preparation of this manuscript. Thanks are also due to Mayara Miyazaki and Pedro Peres for help in some molecular protocols, to Emerson Mossolin for his help in some field collection in Brazil and Chile, and to Charles Fransen and Ingo Wehrtmann for useful comments during the review process.