Palaemon, Weber, 1795

Carvalho, Fabrício L., Magalhães, Célio & Mantelatto, Fernando L., 2019, A molecular and morphological approach on the taxonomic status of the Brazilian species of Palaemon (Decapoda, Palaemonidae), Zoologica Scripta 49 (1), pp. 1-16 : 10-11

4.1.2 | Palaemon sensu stricto

The species of Palaemon s.s. which occur in Brazil ( P. argentinus , P. paivai , P. northropi and P. octaviae ) can be separated by the presence or absence of a mandibular palp and the projection on the fifth male thoracic sternite, proportion between carpus and chela and relative size of stout setae on the distal margin of the telson. Rostral characters, appendix masculina and the anterolateral projection of the first segment of antennular peduncle are also useful for distinguishing the Brazilian species of Palaemon s.s.

Although the external morphology of P. ritteri and P. paivai is very similar and overlaps in the diagnostic characters proposed by Fausto Filho (1967), new characters of the appendix masculina proposed herein allow the distinction between them. Palaemon paivai has a stronger and curved appendix, with a dense row of setae on almost the entire length of the anterior face of the appendix masculina. Palaemon ritteri has a slender and straight appendix masculina, with a sparse row of setae on the entire length of the anterior face of the appendix (Figure S4). The name P. ritteri has been attributed to some populations found in Pará and Ceará ( Ferreira et al., 2010; Gomes‐Correa, 1980). However, a population earlier on attributed to P. ritteri from São João de Pirabas (Pará), and included in our molecular analysis, was allocated outside the clade that includes the Pacific population of this species. Although histone H3 is a conserved gene, the sequence obtained from the topotype of P. paivai allowed the confirmation of a close relationship between individuals from Pará and Ceará, which were separated from the populations of the Pacific. In addition, analysing the appendix masculina from specimens used in previous works, we discovered that P. ritteri in Brazil represent a misidentification, being P. paivai in fact. The analysis of the samples from the Atlantic slope of Panama and Venezuela cited by Holthuis (1952) also revealed higher similarity of the appendix masculina with P. paivai . Therefore, we suggest that P. ritteri and P. paivai are sibling species separated after the complete closure of the isthmus of Panama, supporting the use of this lineage as a calibration point for recent phylogenetic divergence in Palaemonidae .

Palaemon northropi was not recovered as monophyletic. It was intercalated by P. floridanus specimens, which was described by Chace (1942) from western Florida specimens. Some of the characters proposed by Chace (1942) for distinguishing both species were inconsistent in the analysis conducted by Holthuis (1952). Additionally, Knowlton and Vargo (2004) emphasized that the larval morphology of the two species is almost identical. Despite this similarity, Holthuis (1952) maintained P. floridanus as a valid species and reported that P. northropi has three or four teeth on the ventral margin of the rostrum and fingers of the second pereopod up to 2/3 of the length of the palm, while P. floridanus has five to seven teeth and more than 2/3.

The topology indicates that the distinction between P. northropi and P. floridanus based on the number of dorsal teeth and the relationship between the size of the fingers and the size of the chela may not represent natural groups. Although the morphological distinction between the two species may be easy when P. floridanus individuals from the Gulf of Mexico are compared to P. northropi from other regions, this distinction is not very clear when specimens of P. floridanus from the Caribbean and P. northropi are compared. Baeza and Fuentes (2013) reported the genetic differentiation between P. floridanus from the Gulf of Mexico (west coast of Florida) and those from the east coast of Florida. Additionally, they described high similarity between individuals from eastern Florida and Panama. These results indicate that the populations of the Gulf of Mexico have a distinct phylogeographic history from the remaining populations, being the result of an event of recent divergence ( Baeza & Fuentes, 2013), which may be associated with the establishment of a morphological pattern in this area, which was recognized by Chace (1942) as compatible with the species level. A similar phylogeographic pattern for the Gulf of Mexico populations of the freshwater shrimp Atya scabra ( Leach, 1816 ) was discovered by Oliveira, Terossi, and Mantelatto (2019). However, the phylogenetic relationships recovered in the present work indicate that the diagnostic characters of P. floridanus may be autapomorphies that do not represent a lineage other than P. northropi . In this context, an expansion of sampling in the Gulf of Mexico, with the inclusion of individuals with more than five teeth in the ventral margin of the rostrum, is essential for a better understanding of the phylogenetic relationships between P. floridanus and the populations of P. northropi and to test the representativeness of this character on a possible distinct lineage.

A small genetic structuring was observed, with a tendency of separation, between specimens of the North and South Atlantic. As both the type localities of P. northropi and P. floridanus are in the North Atlantic, taxonomic adjustments may be necessary if the topology is confirmed after the amplification of samples from the Gulf of Mexico, resulting in P. floridanus being placed in the synonymy of P. northropi . The genetic structuring could justify the revalidation of another synonym, Palaemon brachylabis Rathbun, 1902 , whose topotypic specimen (Mamanguape River, Paraíba) was included in our analyses and, in general, showed a closer relationship with individuals from the South Atlantic. Nevertheless, low genetic divergence and the absence of morphological characters or other supporting evidence indicate that the genetic structuring found can be attributed to a reduction in gene flow, and this divergence is considered to be at the population level.

An unrecorded species for Brazil was identified after the molecular comparison of some populations of Rio de Janeiro and Pará in relation to the other Palaemon from Brazil. Although specimens of the type locality of P. octaviae ( Chace, 1972a ) were not included in the molecular analysis, the absence of distinctive characters between the later species and newly recorded one from Brazil allows us to assign the name P. octaviae to the populations found in Brazil. In addition, the first record of P. octaviae is made herein for Venezuela (data in Appendix S1), thus expanding the distribution of the species previously restricted to Mexico and Guadalupe to South America ( Venezuela and Brazil —states of Pará and Rio de Janeiro).

Palaemon octaviae has high morphological similarity with P. argentinus , as indicated by Chace (1972a). However, the molecular data indicate a clear separation between the two species and differences in the appendix masculina, male thoracic sternite, stout setae of the telson and shape of the rostrum allow the separation between them. Distinguishing P. octaviae from the species of Palaemon s.s. occurring in Brazil can be done by the absence of a mandibular palp in P. octaviae and comparing characters of the rostrum, distal margin of the telson, appendix masculina and the fifth thoracic sternite.

Both P. octaviae and P. northropi have a wide disjunct distribution, extended larval development (judging by the observed number and size of the eggs), and occur in the same environments. Considering the distribution of P. northropi in both North Atlantic and South Atlantic with low intraspecific genetic divergence, it is suggested that P. octaviae may also show the same geographical amplitude. The species has a markedly disjunct distribution in Brazil, without records from the north‐east region so far. A comparable pattern was also reported for a recently described alpheid species (Almeida, Terossi, & Mantelatto, 2014), which in Brazil occurs in Pará, Bahia and São Paulo, in similar environments to those in which P. octaviae was collected. Sampling efforts at such environments but including other regions of Brazil may reduce distribution gaps, since the absence of structuring on the analysed genes may suggest the existence of recent gene flow in both Alpheus petronioi ( Almeida et al., 2014 ) and P. octaviae , and the existence of intermediate populations along the distribution.

Alternatively, the current distribution of P. octaviae may represent a relict distribution as a result of interactions with other species of the group. The sympatric distribution of P. octaviae and P. northropi , the ecological similarity between the two species in terms of habitat and some reproductive features, and the larger size of P. northropi do not allow the rejection, a priori, of the hypothesis of competitive exclusion of P. octaviae by P. northropi . This hypothesis is reinforced by the fact that the two species are the result of two distinct colonization events on the American continent (see Carvalho et al., 2017). However, comparative ecological studies with these species are necessary so that the competitive exclusion hypothesis can be tested.