Redescription and new records of Synopia ultramarina Dana (Amphipoda: Synopiidae) from off the northeastern Brazil, with comments on its morphological variations

July 2020 DOI 10.1590/1678-4766e2020017 ABSTRACT. Synopia ultramarina Dana, 1853 is considered a circumtropical species and was originally described based on material collected from two localities near halfway between the southwestern Africa and northeastern Brazil. The species is redescribed here with samples from Fernando de Noronha Archipelago, Pernambuco state, northeastern Brazil. The material examined was collected by SCUBA diving, from rhodolith beds, in Ressureta Channel, between Meio and Rata Islands. Additional material was collected from off Ceará state coast, during environmental consultancies, by Petrobras S. A., and off Pernambuco state coast, during the Project REVIZEE NE IV. Discussion on morphological variations of the species is presented, based on the studied material and previous

Synopiidae Dana, 1853 is a cosmopolitan family commonly found in deep sea and, less often, in shallow waters (Barnard, 1972). According to Barnard (1969), "the characters of this family are so subtle that they are practically indefinable, yet taxonomists have little trouble in recognizing a synopiid". However, Lörz & Coleman (2013) diagnose the family emended after Ruffo (1993), being characterized mainly by presenting head with strong recurved rostrum, galeate or with plough-shaped protuberance, pereonites sometimes with sternal hooks in males, urosomites free, antennae 1 and 2 well developed, accessory flagellum present, coxae 1-3 long, narrow or distally expanded, gnathopods simple or subchelate, linear, carpus elongated, uropods1-3 biramous, lanceolate, and telson very long (except in Synopia Dana, 1852), entire or incised.
According to Boltovskoy (1999), the species in the genus Synopia are epipelagic, "practically hypo-neustonic", bathymetrically distributed from 0 to 50 m deep. However, there are records of S. ultramarina reaching up to 3,000 m deep (Schellenberg, 1926). Although it has been recorded from several parts of the world, with many morphological variations, the type material is considered probably lost (Serejo & Siqueira, 2018), which makes taxonomic resolutions difficult. Additionally, some studies that have presented descriptions for this species around the world, have shown morphological variations, mainly related to the mandibles and the telson (Dana, 1853;Bovallius, 1886;Schellenberg, 1926;Shoemaker, 1945;Ledoyer, 1986;Hirayama, 1988;Barnard & Thomas, 1989;Hughes, 2009).
In this paper, we redescribe S. ultramarina, with new records from Northeastern Brazil and discuss the morphological variations between the analyzed specimens and the descriptions of other specimens found in the literature. Studied samples come from three localities: off the coast of Ceará and Pernambuco states and Fernando de Noronha Archipelago.

MATERIAL AND METHODS
The material examined was collected by scuba diving from rodolith beds in Ressureta Channel (Fig. 1
outer plate, medial margin with 1 plumose seta, apically with 2 slender and 1 stout plumose seta; outer plate about 0.8× the length of palp articles 1 and 2 combined, medial margin with 3 plumose setae, apical and lateral margins with 1 long plumose seta each; palp 4-articulate, article 2 with a row of 9 long plumose setae on the medial margin; article 3 with two pairs of subapical plumose setae; article 4 short, bearing a long apical slender seta.

DISCUSSION
Synopia ultramarina is the only species of the genus recorded for Brazilian waters. It is currently considered a circumtropical species (Boltovskoy, 1999) and, as stated by us, presents some variations in the past literature among many individuals regarding especially to mouthparts and the lateral margin of telson. With exception of two records from the Pacific Ocean (Hirayama, 1988;Hughes, 2009) and one record from the Indian Ocean (Ledoyer, 1986), all the other records presented here are from the Atlantic Ocean, with bathymetrical range from 0 to 219 m deep, plus one extralimital record to 3,000 m deep ( Fig. 30; Tab. I). It is worth mentioning that two works discussed herein, Bovallius (1886) and Boltovskoy (1999),were omitted from the distributional map, as the authors mentioned the species distribution as Tropical Atlantic, without geographic coordinates, nor any specific location. Bovallius (1886) cites that he found six specimens of S. ultramarina when analyzing the Captain George von Schéeleʼs pelagic Crustaceans collection, carried out around the world on board the Swedish vessel Monarch. Location and depth information is not provided. In the Habitat section, the author only mentions "Tropical parts of the Atlantic" and in brackets includes the acronyms of the Collections of the Royal Swedish State Museum of Stockholm and the Zoological Museum of the University at Upsala. On the other hand, Boltovskoy (1999) presents an identification key to species of Synopia and a diagnosis for S. ultramarina, but the illustrations of this species were modified after Bovallius (1886).
Regarding the mandibular palp, there are different patterns of number and distribution of elongate setae. In the original Dana's description, palp bears two simple setae in the inner margin of the article 2 and two other apical simple setae in the article 3. The specimen described by Bovallius (1886) presents one plumose seta in the inner margin of the second article and two apical plumose setae in the third. In Ledoyer (1986) and Dana (1853) the pattern of setae is the same. Hirayama (1988) illustrated the mandibular palp with two plumose setae in the inner margin, but these placed  (Schellenberg, 1926); C, Bermuda (Shoemaker, 1945); D, Tulear Reef, South Madagascar (Ledoyer, 1986); E, Tomioka Bay, Japan (Hirayama, 1988); F, Florida Keys and Grand Bahama Island (Barnard & Tomas, 1989); G, Espírito Santo state coast, Brazil (Wakabara et al., 1991); H, Coroa Vermelha, Abrolhos Bank, and California Reef, Parcel of Abrolhos, Bahia state, Brazil (Young & Serejo, 2005); I, Lizard Island, Australia (Hughes, 2009); J, Fernando de Noronha Archipelago (current study); K, off Ceará State coast (current study); L, off Pernambuco State coast (current study).
distally, and two apical plumose setae. Additionally, the article 2 of the palp bears some short slender setae proximally. The specimens described by Barnard & Thomas (1989) present three inner marginal and three apical plumose setae in this article. The material studied by Hughes (2009) has two inner marginal and two apical plumose setae. Finally, the specimens studied here present two variations: one inner marginal and two apical plumose setae (UERJ 542), or two inner marginal and three apical plumose setae (MOUFPE 19642), in both cases, inner margin of second article with a row of small slender setae.
While each material examined in this study and in other past works have a unique and different pattern of setae on the palp of mandible from the others, the shape variations of the lateral margins of telson can be grouped into only three morphotypes. Type 1 is the most common, with serrations at the distal quarter of the lateral margin, and it is present in populations from the Tropical Atlantic (off Ceará state coast, MOUFPE 19642;Bovallius, 1886), Madagascar (Ledoyer, 1986), Florida Keys and Bahama (Barnard & Thomas, 1989), and Australia (Hughes, 2009). Type 2, with serrations at the distal half of the lateral margin, is observed in specimens from Bermuda (Shoemaker, 1945), Japan (Hirayama, 1988), and off Pernambuco state coast (MOUFPE 14566). Type 3 is the least common morphotype, with a smooth lateral margin, which a juvenile was recorded from Madagascar (Ledoyer, 1986) and herein, from Fernando de Noronha Archipelago (UERJ 542). First of all, we must bear in mind that we do not know what the holotype telson of this species looks like, since Dana has not described or illustrated it, and as has been said earlier, the type material is lost. It is noteworthy that just as the specimens described by Ledoyer have morphotypes 1 and 3, the latest being a juvenile, the material examined in this study from off northeastern Brazilian coast presents all three morphotypes. Additionally, the specimens from Ceará (MOUFPE 19642) and Pernambuco (MOUFPE 14566) show another variation in the telson, which is less cleft (about 50%) than those of Tab. I. Occurrences of Synopia ultramarina Dana, 1853 worldwide: Letters A-L, records indicated on map (Fig. 30); x 1-2 , records omitted on map; *estimated coordinates; ** without longitude data; + J. F. Souza-Filho (pers. comm.).  (1886) x² ---0-50 m Circumtropical Epipelagic Boltovskoy (1999) Fernando de Noronha (UERJ 542) and others previously described in the literature (about 70%). It is possible, but not conclusive, that the morphological divergences between the different populations, especially in relation to the shape and ornamentation of telson, are the result of the growth stage of the specimens. It is also interesting to note that specimens recorded from regions other than the Tropical Atlantic have unique character states, differentiating them from the different populations sampled between Africa and Brazil. By Ledoyer (1986), from Madagascar, coxa 3, posterior lobe is broadly rounded (versus subquadrate). By Hirayama (1988), from Japan, mandibular palp article 2 is subrectangular, about twice longer than wide (versus inflated, i.e. 1.6× longer than wide -MOUFPE 19642), and maxilla 2, the inner margin of inner plate is naked (versus densely setose). By Barnard & Thomas (1989), from Florida Keys, the inner margin of inner plate is strongly concave (versus almost straight). And Hughes (2009), from Australia, male accessory eyes have 3 ommatidia (versus 2 ommatidia) and in the pereopod 4, the carpus is slender, about 2.5× longer than wide (versus inflated, about twice longer than wide).

Occurrences
The number and position of setae in each structure is always considered a subjective character as setae may be lost during fixation and handling of the material. Also, in some cases, we may simply be dealing with artifacts, from poor preparation, or mistaken observation of the material. On the other hand, a serrated/spinose or smooth margin is a state of character very commonly used by taxonomists to separate species. Even though it is a benthic species, most of the records of S. ultramarina mentioned here were made from plankton samplings with trawls, since it presents planktonic behavior at night, expanding its possibilities of dispersion. Although there is no data to indicate how far the specimens could disperse at night, this pelagic behavior could, in principle, favor gene flow and, thus, reduce morphological variability between populations. To date, with the data and specimens available, we cannot safely determine that all different populations, or even some of these, correspond to different species in a complex. Perhaps a future study using molecular data might shed light on whether we are really dealing with a cosmopolitan species or a complex of morphologically closely related species.