First record of the Ligia baudiniana species complex in the American Gulf of Mexico Coastline, as confirmed by morphological and molecular approaches

Ligia isopods exhibit a constrained morphology that makes identification difficult. In the Greater Caribbean, a convoluted taxonomic history has left the distributional limits of Ligia baudiniana unclear. To date, no confirmed records of this species exist from the American Gulf of Mexico. Herein, we report the presence of L. baudiniana in Sarasota-Manatee Florida, as confirmed by morphological and molecular approaches. This is the first record of this species in the region and a ~300Km extension of its range. Specimens were collected in mangroves, underscoring the importance of protecting these habitats.


Introduction
The isopod genus Ligia includes ~40 nominal species, most of which inhabit a narrow range in the upper rocky intertidal habitats.
In the Greater Caribbean Region (i.e. the Caribbean and adjacent regions), a single endemic species is currently considered valid: Ligia baudiniana 1,2 . The species has been reported from Brazil 3 , the Caribbean islands 4 , the Pacific coastlines of Central and South America 5-7 , Bermuda 8 , Bahamas 4 , and in southern Florida 9,10 and the Everglades 4 ; however, doubt over historical records have left the distributional limits of L. baudiniana unclear.
L. baudiniana was described from specimens collected in the San Juan de Ulua Fort in Veracruz, Mexico. Milne-Edwards' original species description 11 focuses on characters that are of limited taxonomic importance 12 , lacks illustrations, and does not provide an account of male reproductive structures now known to be useful in Ligia taxonomy [12][13][14] . Indeed, the terse description and source origin of the type material (i.e., artificial substrate) have led to confusion on whether L. baudiniana is a synonym of L. exotica or a valid species 15,16 , and to records and specimens identified as L. baudiniana to be re-classified as L. exotica (3 and references). This is particularly true for specimens from the American Gulf of Mexico coastlines, as most records appear to have been reclassified as L. exotica. Furthermore, a wide-ranging survey of Ligia in the Gulf of Mexico from Texas to Florida has shown artificial habitats in the region to harbor only L. exotica (unpublished study; Hurtado LA, Mateos M, Wang C, Santamaria CA, Jung J, Khalaji-Pirbalouty V, and Kim W).
The taxonomic confusion between L. baudiniana and L. exotica is complicated by the presence of a Ligia species endemic to habitats throughout the Greater Caribbean, Gulf of Mexico excluded, that is easily recognized by a unique male gonopod morphology that is readily distinguishable from L. exotica (Figure 1), and that has been attributed to L. baudiniana by Andersson 5 , Rouse 9 , Schultz 4,8 , and Schultz and Johnson 10 . A recent molecular study demonstrated that Ligia exhibiting this trait form a well-supported monophyletic clade composed of several cryptic and highly divergent lineages endemic to the region 14 . The combination of these studies suggests that L. baudiniana as currently recognized: (a) is an endemic species to the Greater Caribbean Region; (b) can be identified using both molecular and morphological tools; and (c) appears to have a broad geographic range that includes the Caribbean islands, the Pacific coastlines of Central America to Ecuador, Bermuda, Bahamas, and southern Florida.
In southern Florida, L. baudiniana is reported from the Florida Keys 9,10 and the Everglades 4 , while no confirmed records from the American Gulf of Mexico exist to date. In this study, we use molecular and morphological approaches to identify specimens collected from Sarasota and Manatee counties in Florida as L. baudiniana. Our findings extend the confirmed range of this species ~300-km into the Gulf of Mexico coastline of Florida and represent the first confirmed record of L. baudiniana in the American Gulf of Mexico coastline.

Methods
Ligia specimens were collected by hand across the Sarasota-Manatee counties in Florida (Table 1, Figure 2) and field preserved in 70% EtOH. No permits were necessary for collections. Specimens were identified to species by inspecting the morphology of the apex of the endopod of the second pleopod of 15-25 male Ligia specimens per site, with individuals putatively identified as L. baudiniana if they exhibited a large process bifurcating close to the apex of the appendix masculina ( Figure 1A), as proposed by Schultz 4,8 and confirmed by Santamaria et al. 14  We combined nucleotide sequences produced in this study with publicly available ones for L. baudiniana and L. exotica (Table 1). We used default settings to align the resulting dataset using the MUSCLE Alignment 18 tool in Geneious R v8.1.7. No signs of misaligned regions or pseudo-genes were observed in the resulting alignment. The final alignment was imported into MEGA v7.0.18 19 , where we estimated a neighbor-joining tree under    Table 1.
The smaller panel presents the distribution of L. baudiniana lineages reported to date throughout the Caribbean and its adjacent region.
Kimura's 2-parameter model (hereafter K2P; 20) and uniform rates. Support for the relationships within the tree were estimated by conducting 1,000 bootstrap replicates. Lastly, we calculated K2P genetic distances between haplotypes produced by this study, L. exotica, and previously reported L. baudiniana clades 14 .

Results
Molecular identifications produced results congruent with morphological identifications. We obtained 12 unique COI haplotypes from a total of 25 individuals putatively identified as L. baudiniana.
Haplotypes produced in this study were highly similar to each other (COI K2P 0.00-2.81%, Table 2) and to those reported from localities in the Florida Keys, The Bahamas, northern Cuba, Cozumel, and Bermuda (COI K2P 0.50-6.08%, Table 2). Haplotypes were moderately to highly divergent from L. baudiniana from other localities in the Caribbean (COI K2P 14.44-24.90%, Table 2), and highly divergent from L. exotica (COI K2P 20.32-25.18%). The neighbor-joining analysis produced similar results (Figure 3),   nesting all haplotypes produced in this study in a well-supported clade (Bootstrap Support = 100) with the Clade C reported by Santamaria et al. 14 . All unique haplotypes have been deposited in GenBank (Table 1).

Discussion
Morphological and molecular evidence confirm that our sampled individuals represent L. baudiniana. These new records represent the first confirmed presence of this species in the Gulf of Mexico coastlines of the USA and extend the recognized range of the species ~300 km northward from a previous confirmed record from Florida Bay. Positive identifications in this study were made using both morphological and molecular characters. These findings are important as Florida's rich coastal biodiversity faces serious threats such as sea-level rise, introduction of alien species, urbanization, habitat loss, and species displacements 21 .
All L. baudiniana specimens collected in our surveys were found in coastal mangrove forests with no specimens found in >10 surveyed artificial habitats. This suggests that coastal development in the American Gulf of Mexico may have led to the replacement of a native species with an introduced one via the removal of mangrove habitats for the establishment of artificial substrates. Additional work is needed to establish whether L. baudiniana is present in other mangrove habitats along the Gulf of Mexico, thus clarifying the northern limits of this species' range.  ..I wonder if further work will show that the reports from the eastern Pacific constitute a sibling species but this information is not pertinent to acceptance of the current manuscript. It might be worthwhile to point out that mangroves either do not occur or do not support characteristic communities in the northern Gulf of Mexico.

If applicable, is the statistical analysis and its interpretation appropriate? Yes
Are all the source data underlying the results available to ensure full reproducibility? Yes

Are the conclusions drawn adequately supported by the results? Yes
No competing interests were disclosed.

Competing Interests:
Referee Expertise: Marine Biology I have read this submission. I believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.