Lysmata rauli, Laubenheimer & Rhyne, 2010

Guéron, Rodrigo, Almeida, Alexandre Oliveira, Aguilar, Robert, Ogburn, Matthew B., Prakash, Sanjeevi & Baeza, J. Antonio, 2022, Delimiting species within the Lysmata vittata (Stimpson, 1860) (Decapoda: Lysmatidae) species complex in a world full of invaders, Zootaxa 5150 (2), pp. 189-216 : 202

Lysmata rauli is a valid species

Here we formally resurrect L. rauli to valid species status, concluding it is not a junior synonym of L. vittata as previously suggested by Soledade et al. (2013) based upon morphological similarity between newly collected Brazilian material and published accounts of L. vittata and genetic similarity between Brazilian L. rauli and L. vittata from Thailand (see below). While Soledade et al. (2013) provided a detailed morphological analysis of material collected in Brazil (the type locality for L. rauli ), there was no corresponding analysis of L. vittata from its putative native range. Importantly, their conclusions were also hindered by historical uncertainty regarding accessory branch structure of L. vittata . Counter to Bruce’s (1990) redescription of L. vittata based on a topotypic specimen, Soledade et al. (2013) stated L. vittata possesses a one-segmented accessory branch, similar to Brazilian material identified as L. rauli or L. vittata . We posit this supposition was based on misidentified material. The present study, as well as Aguilar et al. (2022) and Bruce (1990), clearly show the neotype /topotypic material of L. vittata possesses a uniramous accessory branch, in additional to L. vittata from New Zealand and the USA. Further, the genetic analyses presented in Soledade et al. (2013) were hindered by the lack of material from across the full distribution of L. vittata . Much of the strength of Soledade’s et al. (2013) argument was based on genetic similarity between Brazilian L. rauli and L. vittata from Thailand. However, we posit this Thailand material (also included in the present analyses) was not L. vittata (clade LV1), but rather L. rauli (clade LV4).

In our analyses, which included material from across the western Atlantic, the northern Indo-Pacific, and Oceania, we recovered L. vittata (clade LV1) and L. rauli (clade LV4) as strongly supported clades ( Fig. 4 ). Genetic differences (p -distances) between these two clades were considerable (0.168 –0.194) and greater than several interspecific cross-clade comparisons within Lysmata , e.g., L. seticaudata / L. debelius (0.139) and L argentopunctata / L. boggessi (0.157). Our genetic analyses are supported by Aguilar et al. (2022), which recovered L. vittata and “ L. vittata / rauli ” from Brazil in widely divergent and strongly supported clades and delineated both clades as putative species based on ABGD analysis. Given the morphologic uncertainty surrounding L. vittata and lack of available data at that time, Soledade’s et al. (2013) conclusions were not unreasonable and further highlights the urgency of solving the longstanding taxonomic ambiguity of L. vittata .

In addition to accessory branch structure, there are other important morphological differences which support the recognition of L. rauli as a valid species. The number of meral and carpal segments of the second pereopod (P2) vary widely across Lysmata species and can be useful characteristic in delimitating species (see Chace, 1997). On average, L. rauli had fewer carpal (16.75 vs. 20.79) and meral (7.36 vs. 10.32) second pereopod segments in comparison to L. vittata . In our PCA and DFA, both aforementioned characters had high loading factors (table 3) in the first principal component, which explained nearly a third of the total variance. Although rostrum length is variable and should be treated with caution, L. rauli appear to have a slightly shorter rostrum, reaching the midpoint of the second article of antennular peduncle, as opposed to the distal margin of the same article in L. vittata ( Soledade et al. 2013; Aguilar et al. 2022). Additionally, coloration patterns are an important and powerful character in distinguishing Lysmata species (Rhyne & Lin 2006; Rhyne et al. 2012; Baeza & Behringer 2017), particularly among species that are visually similar, such as between L. grabhami and L. amboinensis (see fig. 3 in Baeza 2010), and L. intermedia and L. jundalini ( Rhyne et al. 2012) or morphologically similar species (Rhyne & Lin 2006). For example, the cleaner shrimp L. amboinensis was originally described as a subspecies of L. vittata due to a superficially similar morphologies (from preserved material) but displays an unmistakable live coloration pattern compared to any entity in the wider L. vittata species complex. Lysmata rauli can be easily distinguished from L. vittata by the presence of dark transverse bands at the first and between the third and fourth pleonal segments, further supporting the resurrection of L. rauli .

Figures

FIGURE 4. Phylogenetic tree obtained from Maximum Likelihood (ML) analysis of the partial 16S rRNA gene for shrimps from the Lysmata vittata species complex, selected lysmatids, and Merguia. Numbers above or below the branches represent bootstrap values obtained from ML in the webserver W-IQ-TREE (Trifinopoulos et al. 2016; http://iqtree.cibiv.univie.ac.at/). Low node values (<50) were removed from the final topology. Asterisk indicates the neotype of L. vittata. Vertical bars represent dissimilar taxonomic units suggested by ASAP. Numbers below vertical bars indicate the number of species identified by ASAP (27) and ASAP-score (2.0) in the corresponding partition.