A fresh look at the biodiversity lexicon for fiddler crabs (Decapoda: Brachyura: Ocypodidae). Part 1: Taxonomy

Fiddler crabs (Ocypodidae) have gone through a gradual series of taxonomic revisions and refinements over the last 40 years, culminating most recently with an expansion from a single genus into eleven different genera. I examine the opportunities presented by these revisions with respect to establishing formal names for previously established clades at a variety of taxonomic levels that were otherwise previously impossible to name due to historical compression of these crabs into a single genus, including the establishment or reestablishment of three tribes (Ucini, Gelasimini, and Minucini) and ten subgenera: Uca (Uca), Uca (Acanthoplax), Gelasimus (Gelasimus), Gelasimus (Mesuca), Austruca (Austruca), Austruca (Cuneatuca), Austruca (Sinduca), Tubuca (Tubuca), Tubuca (Australuca), and Tubuca (Angustuca). A previously overlooked synonymy between Gelasimus excisa (Nobili, 1906) and G. neocultrimana (Bott, 1973) is discussed, and the former name is adopted as valid.


INTRODUCTION
Fiddler crabs (Ocypodidae Rafinesque, 1815) are well-known brachyuran crabs inhabiting shorelines worldwide across the tropics and well into the temperate zones. Despite their rather small size and somewhat shy nature around humans, their colorful markings, aggressive waving and fighting behaviors, and the seemingly absurd claw asymmetry found in males makes them a charismatic group, popular with both amateur naturalists as well as in scientific studies. As with many other groups of organisms, the taxonomy of fiddler crabs has gradually shifted over time as new studies, technologies, and scientific attitudes have led to new and different insights into the evolutionary history of these species. The last truly comprehensive revision and study of the taxonomy of fiddler crabs (Crane, 1975) has served as a benchmark for all studies since and has allowed subsequent researchers to use a more piecemeal approach to updating the taxonomy and systematics as necessary (Thurman, 1981(Thurman, , 1982George & Jones, 1982;Barnwell & Thurman, 1984;von Prahl & Toro, 1985;von Hagen, 1987;von Hagen & Jones, 1989;Levinton et al., 1996;Sturmbauer et al., 1996;Rosenberg, 2001Rosenberg, , 2013Beinlich & von Hagen, 2006;Shih et al., 2009Shih et al., , 2010Shih et al., , 2012Shih et al., , 2013aShih et al., , 2015Shih et al., , 2016bShih et al., , 2018Shih et al., , 2019Spivak & Cuesta, 2009;Landstorfer & Schubart, 2010;Naderloo et al., 2010Naderloo et al., , 2016Thurman et al., 2018). Today we recognize 105 extant species (Rosenberg, 2014), with five additional named fossil taxa, a major shift from the 62 extant species (92 taxa with subspecies included) and two fossils recognized by Crane (1975). The nine subgenera described by Crane (1975) were based on morphological comparison in the absence of quantitative phylogenetic methodology. Advances in molecular phylogenetics resulted in the rearrangement, renaming, and gradual expansion of the recognized subgenera to twelve, until a recent study  moved away from the notion of treating fiddler crabs as a single genus and instead has raised eleven of the former subgenera to full genera.
The goal of this work is to consider the current status of fiddler crab taxonomy in light of these recent changes, identify opportunities in our current classification for better describing known and likely clades, clarify a taxonomic ambiguity that has been left unresolved, and highlight places where more work is necessary.

ABOVE AND BEYOND THE GENUS: THE HIGHER-LEVEL TAXONOMY
Fiddler crab species have usually been considered members of a single genus: Gelasimus Latreille, 1817 for most of the 19th century and Uca Leach, 1814 starting after 1897 when the priority of this name was recognized (Rathbun, 1897). Proposed subdivisions within the genus largely started with the split of the Eastern Pacific species into two subgenera by Bott (1954): Uca (the narrowfront species) and Minuca Bott, 1954 (the broad-front species). More widespread divisions of the genus began with the competing works of Crane (1975) and Bott (1973); Crane's names were more widely recognized and thought to better represent true systematic clusters, but Bott's names had taxonomic priority (von Hagen, 1976;Rosenberg, 2001). Since the mid-1990s, as additional morphological characterization and formal molecular phylogenetic methods were applied to the genus, a number of studies proposed to contract or expand the various subgenera (Rosenberg, 2001;Beinlich & von Hagen, 2006;Spivak & Cuesta, 2009;Shih, 2015;Shih et al., 2015). Outside of a number of largely ignored, informal superspecies designations by Crane (1975), little additional effort was made to further subdivide the species within subgenera. Because Crane (1975) treated many of the now-recognized species as subspecies the subsequent literature often refers to the larger groups of formerly-single-species-under-Crane as semi-formally recognized taxa, e.g., the vocans species complex (Shih et al., 2010(Shih et al., , 2016aRosenberg, 2013) or the lactea species complex (Shih et al., 2009(Shih et al., , 20102013b;Naderloo et al., 2010Naderloo et al., , 2016. Because all fiddler crabs were otherwise considered a single genus, higher-level taxonomic names only served to express relationships between fiddler crabs and closely related clades such as ghost crabs (Ocypode Weber, 1795) and mangrove crabs (Ucides Rathbun, 1897). Shih et al. (2016b) upended this system with their conclusion that fiddler crabs were paraphyletic without the inclusion of ghost crabs. This paraphyly had been seen in earlier works with similarly built but smaller data sets, starting with Levinton et al. (1996) and Sturmbauer et al. (1996), but these earlier authors appear to have written off the result as a taxonomic reconstruction error likely not reflecting the true evolutionary history of these crabs. With their larger data set and high phylogenetic support, Shih et al. (2016b) embraced this result and used it to reorganize higher-order names across fiddler crabs. The lack of fiddler crab monophyly forced them to abandon the traditional concept of the genus Uca as a single taxonomic name capturing all fiddler crabs; instead all but one of the former subgenera were raised to generic status and no single taxonomic name can be used to refer to all fiddlers (one of the former subgenera was abandoned as it was phylogenetically contained within another; see below). Shih et al. (2016b) organized these genera into two subfamilies: Ocypodinae Rafinesque, 1815 (containing the fiddler crab genera Uca and Afruca Crane, 1975 as well as the ghost crabs, predominantly the genus Ocypode) and the Gelasiminae Miers, 1886 (containing the other nine fiddler crab genera) (Fig. 1). A third subfamily, Ucidinae Dana, 1851, contains the mangrove crab genus Ucides. The relationship between Uca, Afruca, and Ocypode was left unresolved by Shih et al. (2016b), such that it was not clear whether the two fiddler crab genera formed a clade separate from Ocypode. The World Register of Marine Species (WoRMS, 2019) currently assumes these two fiddler crab genera are a clade, represented by the subfamily Ucinae Dana, 1851, reserving Ocypodinae for just the ghost crabs (Fig. 1).
The hypothesis/result that fiddler crabs are paraphyletic with respect to ghost crabs is not without controversy. To many researchers (MSR, unpublished data), this result seems biologically implausible (although by no means impossible), as it appears to require the evolution of a suite of key characters present in all fiddler crabs and absent from close relatives (e.g., extreme sexual dimorphism, extreme male asymmetry, long and thin eyestalks) followed by a nearly complete reversal of these same characters in the ghost crabs (sexual monomorphism, mild asymmetry in both sexes, short and thick eyestalks). While Shih et al. (2016b) included a large number of species and samples, their result was based on only three gene fragments (1,936 total sites), two mitochondrial and one nuclear, leading to both a limited number of sites per species and the admixture of two types of genes with known differential inheritance patterns and effective population sizes. Phylogenetic discordance between mitochondrial markers and nuclear markers is a common phenomenon and it is not unusual for phylogenetic studies based on thousands of independent markers from NextGen sequencing technologies to produce different results than those found from more limited mitochondrial-focused data sets (e.g., Fisher-Reid & Wiens, 2011;Jacobsen & Omland, 2011;Jockusch et al., 2014;Morgan et al., 2014;Hofmann, 2015;Thielsch et al., 2017;Platt et al., 2018). With all of this in mind, it is not clear whether the paraphyly of fiddler crabs with respect to ghost crabs represents the true evolutionary history of this group or is a gene-sampling artifact that will stand not up to further scrutiny and study when NextGen data becomes available.
A clear benefit of the Shih et al. (2016b) result is that it highlights a flaw in our traditional thinking: we tend to focus so much on evolution within groups that we often fail to consider the implication of evolution among groups. For example, fiddler crab researchers have tended to focus on evolution of the large claw in male fiddler crabs with a generally unstated assumption that this large claw evolved from the smaller claw (females, after all, have two small claws). The small claw of males and the two claws of females are at least as divergent, if not more divergent (certainly in size), from the claws of their close relatives (Ocypode and Ucides, both of whom have moderately large, mildly asymmetric claws in both sexes) than is the large claw of the male. The evolution of the claws of fiddler crabs might have been driven as much, if not more, by reduction of the small claw for more efficient deposit feeding with sexually selected retention of a large claw in males, rather than the other way around.
While the placement of ghost crabs within fiddler crabs required splitting fiddler crabs into multiple genera, these splits could have been justified in the absence of paraphyly, and all of the new fiddler crab genera are more or less identical to the previously recognized subgenera (the subgenus Australuca Crane, 1975 had previously been suggested by the same authors to likely be a subset of another subgenus, Tubuca Bott, 1973). Furthermore, the remainder of the phylogeny is uncontroversial and can serve as a basis for considering higher-order taxonomic divisions for fiddler crabs. All of the proposed names below are congruent with, if not directly based on, the phylogeny of Shih et al. (2016b) and are generally independent of the monophyly/ paraphyly question. It has long been accepted that fiddler crabs consist of three primary groups corresponding to broad geographic distributions Rosenberg, 2001). The first group to diverge consists of the Eastern Atlantic species (Afruca) and the American narrow-front species (Uca) (equivalent to Ucinae above, combined into a clade with Ocypode by Shih et al. 2016b). The remaining fiddler crabs form a clade (Gelasiminae) with two major subclades: the American broad-front species and the Indian and Western Pacific oceans species (Indo-West Pacific, or IWP, region). The American broad-front and IWP clades have long been recognized Sturmbauer et al., 1996;Rosenberg, 2001) but remain formally unnamed, in part due to taxonomic limitations imposed by the tradition of treating all fiddler crabs as a single genus.
With the expansion of fiddler crab species to multiple genera and subfamilies and to aid in clarity of communication it seems clear that these latter two subclades should receive formal names, with tribe being the obvious rank. The American broad-front species should be referred to as the tribe Minucini tribus nov. (containing Minuca, Leptuca Bott, 1973 andPetruca Shih, Ng &Christy, 2015), and the IWP species should be referred to as the tribe Gelasimini Miers, 1886 status nov. (containing Tubuca, Xeruca Shih, 2015, Gelasimus, Cranuca Beinlich & von Hagen, 2006, Paraleptuca Bott, 1973, and Austruca Bott, 1973. One could alternatively reserve Gelasiminae for just the IWP species and use Minucinae for the American broad-front species, but that division would fail to recognize the clear relationship of those two clades relative to the Ucinae and Ocypodinae. For rank consistency across all of the groups, the tribe Ucini Dana, 1851 would contain Uca and Afruca (Fig. 2), following the WoRMS (2019) classification under the assumption that Uca and Afruca form a monophyletic group. The latter tribe was originally used by Pretzmann (1983) for all fiddler crabs when he accepted the multiple genera of Bott (1973) but would be restricted to only these two genera under the current system.
A clear advantage of these tribes is taxonomic stability as, unlike the subfamilies, the tribes would likely retain their identical meaning whether fiddler crabs are monophyletic or paraphyletic. If future studies find fiddler crabs to be monophyletic, Ucinae would have priority as the subfamily representing all fiddler crabs, but the three named tribes could still be used to represent the three main internal clades without modification (Fig. 2). The subfamily currently called Gelasiminae would instead become the supertribe Gelasimitae status nov., with the supertribe Ucitae status nov. serving the role currently occupied by Ucinae.

THE SUBGENERA ARE DEAD, LONG LIVE THE SUBGENERA
With the former subgenera raised to generic status, there is now potentially room within the new genera for further delineation, which would seem useful in a few cases. Previous researchers have questioned the need for subgenera (e.g., von Hagen, 1976) and the majority of scientific publications since they were broadly introduced in the 1970s have tended to ignore them. It is fair to question whether increasing the complexity of our taxonomy by designating new subgenera is necessary now, when we have concluded splitting a single genus into multiple genera. In a few cases, I believe it is readily justified because we already use informal names to refer to some of these groups, e.g., the vocans species complex. Another important point to note is that the use of subgenera can be viewed as condition-dependent. Researchers can use them when they provide value or clarity but choose to ignore them when they do not. For each genus discussed below, the initial designated subgenus is already a well-recognized group in the literature, with additional support from the phylogeny of Shih et al. (2016b). Additional subgenera are suggested as place-holders for the remaining species in the genus, generally based on strongly supported clades from that same phylogeny (exceptions will be noted).
Austruca currently consist of 12 species; eight of these make up what has generally been referred to as the lactea species complex (Shih et al., 2009(Shih et al., , 2010(Shih et al., , 2013bNaderloo et al., 2010Naderloo et al., , 2016, defined as the subspecies grouped into a single species by Crane (1975). This group can formally be designated the subgenus Austruca, consisting of A. albimana (Kossmann, 1877), A. annulipes (H. Milne Edwards, 1837), A. cryptica (Naderloo, Türkay & Chen, 2010;see Naderloo et al., 2010), A. iranica (Pretzmann, 1971), A. lactea (De Haan, 1835), A. mjoebergi (Rathbun, 1924), A. occidentalis (Naderloo, Schubart & Shih, 2016), and A. perplexa (H. Milne Edwards, 1852). The other four species in the genus can be divided among two additional new subgenera: Cuneatuca subgen. nov., consisting of A. triangularis (A. Milne-Edwards, 1873), A. bengali (Crane, 1975), and A. variegata (Heller, 1862); and Sinduca subgen. nov., consisting of A. sindensis (Alcock, 1900). The former is derived from the Latin for tapering (describing the sharp pointing anterior-lateral angles of these species) and represents the pair of species formerly considered subspecies of Uca triangularis by Crane (1975) (the "triangularis-species complex" according to Shih et al., 2019), as well as A. variegata, a recently recognized (Shih et al. 2019) long-lost relative. In keeping with the current informal naming convention, Shih et al. (2019) suggest renaming this group the variegata species complex, demonstrating that a formal taxonomic name would be useful. Austruca sindensis is placed in its own subgenus because it occupies a basal division and does not appear to be closely related to any of the other species within the genus ; Sinduca derives its name from the same source as the type species, the Indus River where it was first found.
The genus Uca has one clear, differentiated subgroup consisting of three species, which can be considered the subgenus Acanthoplax Milne Edwards, 1852: U. insignis (H. Milne Edwards, 1852), U. maracoani (Latreille, 1803), and U. ornata (Smith, 1870). These species are among the largest of all fiddler crabs and have uniquely shaped major claws, with extraordinarily wide and flat dactyl and pollex that more resemble pruning shears than the major claws of most other species. The remaining six species are tentatively placed in the subgenus Uca: U. heteropleura (Smith, 1870), U. intermedia von Prahl & Toro, 1985, U. major (Herbst, 1782), U. monilifera Rathbun, 1915, U. princeps (Smith, 1870), and U. stylifera (H. Milne Edwards, 1852). The relationships among the species in this genus are generally not yet well enough known for confident subdivision.
Of the remaining seven genera, four are monospecific (Afruca, Cranua, Xeruca, and Petruca), while the other three (Leptuca, Minuca, and Paraleptuca) currently lack the systematic clarity necessary for further subdivision.

SPECIES TAXONOMIC NOTES
As with many other groups, named species of fiddler crab have gone through waves of consolidation and expansion. Over the last few decades, the relationships among most of the historical names have largely stabilized with taxonomic advances mostly revolving around the recognition/discovery of cryptic species within formerly recognized single species (Novak & Salmon, 1974;Thurman, 1981;Naderloo et al., 2010Naderloo et al., , 2016Shih et al., 2009Shih et al., , 2010Shih et al., , 2012Shih et al., , 2013aShih et al., , 2018Shih et al., , 2019Thurman et al., 2018), although a few purely novel species have been described as well (George & Jones, 1982;von Prahl & Toro, 1985;von Hagen, 1987;Landstorfer & Schubart, 2010).
One currently recognized species requires some discussion with respect to historical names and nomenclature: Gelasimus excisa (Nobili, 1906) versus G. neocultrimana (Bott, 1973). Desmarest (1817) described a fossil crab under the Latin name Goneplax nitida (using "Goneplace luisant" as a common name in French). Desmarest (1822) subsequently renamed this fossil Gelasima nitida, a reassignment later accepted by Milne Edwards (1837). Dana (1851) subsequently described a new species from Fiji as Gelasimus nitidus, apparently without reference or knowledge of Desmarest's earlier name. Dana's species became part of a complex of names, in particular with Uca marionis (Desmarest, 1823) and U. cultrimana (Adams & White, 1848), which were used for the same and/or very similar species or forms, all of which were eventually synonymized with Uca vocans by the 1970s. Crane (1975) recognized six subspecies of Uca vocans, including a supposedly new subspecies, Uca vocans pacificensis Crane, 1975, found in Fiji. This name was later found to be a junior synonym of Mesuca (Latuca) neocultrimana Bott, 1973, also described from Fiji (Rosenberg, 2001;Shih et al., 2010). Shih et al. (2010)  Of note is that Crane (1975) recognized that her name Uca vocans pacificensis could be a junior synonym of Dana's species but felt that nitida should be avoided due to the confusion with Desmaret's name. Crane (1975) also rejected use of excisa by Nobili (1906) as a replacement for nitida, because she claimed Nobili's meaning was unclear and not tied to specific type specimens or locations. Her reasoning was unfortunately faulty as Nobili specifically suggested a replacement name for Dana's species, thus inheriting his name-bearing type and type locality. (Bott, 1973) should therefore be recognized as a junior synonym of the senior name, Gelasimus excisa (Nobili, 1906).

THE FUTURE OF FIDDLER-CRAB SYSTEMATICS
While this work has been predominantly focused on taxonomy, three areas of inquiry with respect to the systematics of fiddler crabs stand out as critical to the next generation of studies.
1) What is the phylogenetic relationship between fiddler crabs and ghost crabs? Is it actually paraphyletic as suggested by Shih et al. (2016b) or was that result a data artifact? 2) What are the relationships of species within the genera and subgenera? A lot of progress has been made on understanding the broader relationships among the genera, but on the whole the species-level relationships are still quite uncertain, particularly within Minuca (18 species) and Leptuca (30 species). 3) Where will the next cryptic species be found? As molecular systematics has been more broadly applied to fiddler crabs, a number of geographically widespread species have recently been split into sets of more regional similar/cryptic species (Shih et al., 2009(Shih et al., , 2010(Shih et al., , 2012(Shih et al., , 2019Naderloo et al., 2010Naderloo et al., , 2016Thurman et al., 2018), and it is likely more are waiting to be found. Species such as Uca (Uca) princeps (Crane, 1975;MSR, unpublished data), Tubuca (Tubuca) forcipata (MSR, unpublished data; H.T. Shih, personal communication), and Minuca ecuadoriensis (Barnwell, 1988) have all been observed to encompass enough variation to raise questions as to whether they represent multi-species complexes currently hidden under a single name.

OUTLINE OF THE TAXONOMIC HIERARCHY OF FIDDLER CRABS
The lists of synonymous names and prior usages of each taxon are not included below as this outline, with the sole exception of Gelasimus (Gelasimus) excisa discussed above, does not change or challenge other recent revisions that otherwise contain identical such lists (e.g., Shih et al., 2016b). Species marked with † are only known from fossils.
Remarks: This subgenus is readily distinguished from subgenus Acanthoplax by the shape of the major chela, with the claws of Uca more similar to those of other fiddler crab species rather than the broad and flat shear-like shape of Acanthoplax. Uca (Uca) monilifera has a chela shape midway between the rest of the Uca and Acanthoplax making its placement within this subgenus tentative.
Remarks: Among the largest fiddler crabs, the subgenus Acanthoplax is readily distinguished from the sister subgenus Uca by the unique shape of the major chela, with only Uca (Uca) monilifera having a shape approaching those of the Acanthoplax.

Type genus: Gelasimus Latreille, 1817
Diagnosis: Small-to-medium sized species (carapace breadth 10-40 mm); front narrow or broad; gastric mill without large brownish setae at base of posterior tooth plate; pleonal locking mechanism present or absent.

Remarks:
The tribe Gelasimini is geographically restricted to the Indian and central-to-western Pacific oceans and includes all fiddler crab genera within these regions. It can be distinguished from its sister tribe, the American Minucini, by the absence of two large brownish setae at the base of the posterior tooth plate on the gastric mill.
Remarks: The species in this subgenus are readily distinguished from the sister subgenus Mesuca by the shape of the major chela, the tuberculate ridge on the inside of the major palm, the strong groove running along the outside of the pollex, and the relatively long fingers and gape in the minor chela.
Remarks: The single species in the subgenus, Gelasimus (Mesuca) tetragonon, is easily distinguishable from the species in the sister subgenus Gelasimus by the shape of the major chela, the lack of a strong groove on the outer pollex, the tuberculate ridge inside the major palm, and by the length of the fingers on the minor chela.
Remarks: The subgenus Austruca represents the "lactea species complex." It is generally distinguishable from the other subgenera by the lack of a projecting tube on the distal end of the gonopod; it is also distinguishable from subgenus Cuneatuca by its carapace shape and the lack of a tuberculate row on the minor merus. It is sister to the subgenus Cuneatuca.
Remarks: This subgenus represents one of the two crown clades of the genus and is sister to subgenus Australuca.
Remarks: Australuca was previously considered a subgenus within the single fiddler crab genus Uca but was subsumed within the subgenus Tubuca when greater phylogenetic resolution was achieved.
Within the now-raised-to-genus Tubuca, the traditional species within Australuca still represent a unique clade.
Remarks: This subgenus represents the "acuta species complex." It diverges basally from the other subgenera within the genus.

Remarks:
The tribe Minucini is geographically restricted to the Americas and outlying islands, and contains all three broadfronted genera from this region. It can be distinguished from its sister tribe, the Indo-West Pacific Gelasimini, by the presence of two large brownish setae at the base of the posterior tooth plate on the gastric mill.