Microgloma Sanders & Allen, 1973 (Nuculanidae) and Pristigloma Dall, 1900 (Pristiglomidae) (Pelecypoda) in the Campos Basin off Brazil

Abstract As a secondary result of oil prospecting in Brazil, samples from the Campos Basin continental slope became available. In these samples, specimens of the genera Microgloma Sanders & Allen, 1973 and Pristigloma Dall, 1900 were found. This contribution provides the southernmost record of the genus Microgloma, the first record of Microgloma mirmidina (Dautzenberg & Fischer 1897) from the western Atlantic, the descriptions of Microgloma macaron sp. n. and Microgloma nhanduti sp. n.as new species, and the shallowest record of Pristigloma alba Sanders & Allen 1973.


Introduction
Knowledge of the deep-sea mollusks from the Campos Basin has increased in the last ten years, and several new species have been described (Absalão et al. 2001Absalão andPimenta 2003, 2005;Caetano et al. 2006;Zelaya et al. 2006;Pimenta et al. Basin Environmental Heterogeneity" in 2008 and. Both programs were sponsored by the Brazilian oil company Petrobras S.A. Of the material obtained we observed 260 samples taken between the isobaths of 400 and 2500 m. The list of localities with Microgloma and/or Pristigloma specimens is given in Tables 1 and 2. Most of the shells were found in a good state of preservation, with valves attached, ligaments intact, and often with the mass of the animal body inside the shell. Unfortunately, there were no preserved organs in these cases. Each specimen was examined under stereoscopic microscope (Nikon SMZ 800), and selected specimens were photographed with a scanning electron microscope (ZEISS EVO 40), at the Gerência de Bioestratigrafia e Paleoecologia Aplicada (BPA), of the Petrobras Research Center (Centro de Pesquisas da Petrobras, CENPES).
Taxonomic identifications were made through comparison with the figures of type specimens [Microgloma pusilla (Jeffreys, 1879)] and descriptions available in the literature (Sanders and Allen 1973, Salas 1996, Ockelmann and Warén 1998, La Perna 2008, Oliver et al. 2009). The species were characterized considering traditional criteria used in pelecypod orientation and terminology (Figs 1-2) (Fischer 1886, Sanders and Allen 1973, Mikkelsen and Bieler 2008, Baylei 2009). In view of the importance of the features of the hinge plate for the discrimination of other protobranch species , and also some subjective concepts in taxonomy (e.g., 'thin' or 'thick'), we described the species using certain quantitative criteria such as the ratios of the hinge teeth (wht) and hinge plate (whp) measurements (Figs 1-2), which are described as follows: 'thin' for width of hinge plate/total height ratio < 0.1; 'thick' for width of hinge plate/total height ratio ≥ 0.1. The width of the hinge teeth was measured just above (dorsal) and below (ventral) the limit of the big-
Characterization. Shell H/L ratio about 1.04. W/H ratio approximately 0.86. Muscle scars rarely visible; anterior adductor scar 2/3 of posterior scar. Anterior and posterior parts of the hinge plate usually of same length. Width of posterior row of teeth occupies about 65% of width of hinge plate, which is thin for its size (whp/H approximately 0.07). Posterior part of the hinge plate longer than anterior one. Prodissoconch smooth, with length approximately 120 μm.
Remarks. This species was recently well described and for this reason we add only new information on the proportions of the shell (H/L and width) and hinge plate characteristics. Sanders and Allen (1973) noted no evident muscle scars or pallial line. In the Campos Basin specimens, no pallial line is discernible on the valves, but faint muscle scars are apparent in some specimens. It is a common species in the Atlantic, and its occurrence in the Campos Basin was expected, since it was previously recorded from the northern Brazilian coast (Pernambuco) and from Argentina. The present study provides new points to the geographical distribution of P. alba, and is also the shallowest (1200 m) record for the species, which was previously known only from abyssal depths (2100-4898 m).
Description. Shell minute, ovate, H/L ratio about 0.81 (n 38, min 0.75, max 0.92), W/H ratio about 0.84 (n 14, min 0.77, max 0.87), glossy, translucent, robust for its size, equilateral; umbones prominent, large, posterior to midline, orthogyrous. Antero-dorsal margin straight, oblique; anterior margin rounded, extended. Antero-ventral margin, postero-ventral margin shorter and rising up to the short posterior end; posterior margin slightly truncated, forming a small shoulder. Entire shell margin thickened and covered by an extension of the periostracum, resembling a macaroon in ventral view (Figs 21-22). Surface with commarginal growth lines. Hinge plate with 5-7 anterior and 4-5 posterior teeth, interrupted by a large, rectangular and shallow resilifer. Width of both anterior and posterior row of teeth occupies about 70% of width of hinge plate, which is thick for its size (whp/H approximately Campos Basin, #54, 12/12/2002 0.14). Posterior part of the hinge plate shorter than anterior one. Prodissoconch surface nacreous, with one subtle radial striae (Fig. 20), and length approximately 244 μm. Maximum adult shell length 1.20 mm.
Remarks. In some other species of Microgloma, the shell is expanded ventrally, around the valve, which enlarges the internal volume of the animal and counterbal-Remarks. The radial sculpture on the prodissoconch surface is absent in M. mirmidina, but admittedly it is not always clearly developed in the other species of Microgloma (La Perna 2008). Microgloma mirmidina differs from other species of the genus in the elongated outline, more inflated shells and small hinge plate. In relation to the type material of M. mirmidina figured by La Perna (2008) the Campos Basin material presents a smaller hinge plate. The figured specimens presents H/L ratio of 0.72, 0.76 and 0.83 (La Perna 2008 fig 6 B, E, I, respectively). Despite geographical distance between the two records of the present species, these conchological differences are not pronounced enough to affirm they belong to distinct species.
Type locality. Campos Basin, #54, 12/12/2002, 750m, 21°57'17,5"S, 39°56' 01,1"W. ances the effects of miniaturization (Ockelmann andWarén 1998, La Perna 2008). However, in M. macaron this expansion is thicker, and the periostracum surrounds the entire margin (Fig 22). This characteristic of the periostracum deserves special attention since it not only covers the shell to the margins in the usual way, but is more conspicuous in this area, giving the impression that valves do not articulate. We are not sure how this system works, and how the animal, in spite of having these fibers surrounding the valve apertures, can have water flux in the mantle cavity. Compared to M. pusilla and M. mirmidina, M. macaron is distinguished by the ovate outline, umbones at midline and much more projecting, and a thicker hinge plate. Microgloma macaron is similar to M. yongei in outline, but compared with the paratypes figured by Ockelmann and Warén (1998, page 14, fig 6-D), the former has more prominent umbones, a thicker hinge plate, as well as a larger resilifer. Microgloma macaron can be distinguished from M. tumidula by the shape of the teeth, which are not as inclined as in this latter species. The anterior and posterior areas of the hinge plate form a less obtuse angle compared to those of M. tumidula.
Except for two, probably worn, valves found at two stations at approx. 3000 m depth, and one at 1970 m, this species is concentrated at depths between 400-750 m. Untill now this species is reccorded solely in Campos Basin.
Type locality. Campos Basin, #54, 12/12/2002, 750m, 21°57'17,5"S, 39°56'01,1"W. Etymology. Nhanduti is a Tupi-Guarani term (the language spoken by the largest groups of native people living in Brazil prior to the European colonization) for a spider web-like structure, similar to those present on the prodissoconch of this species. The species epithet is proposed as a noun in apposition.
Remarks. Microgloma nhanduti sp. n. is similar to M. pusilla and M. tumidula in the oval outline. but has more prominent umbones (Figs 26-28). Compared to M. macaron sp. n. and M. yongei, M. nhanduti sp. n. has a more elongated outline, a longer anterior area, and a more evident prodissoconch sculpture, with a web-like pattern (Figs 29-30). In M. nhanduti sp. n. the umbo is not as prominent and the shell margin not as thick as in M. macaron sp. n. Untill now this species is reccorded solely in Campos Basin.

Discussion
Ockelmann and Warén (1998) carefully evaluated the systematic position of the Nuculoidea and Nuculanoidea, and placed Microgloma within the Nuculanidae (a position with which we agree). However, these authors did not use the subfamily rank introduced by Allen and Hannah (1986). Considering differences such as prominence of the rostral area, presence of carena and keels, foot grooves, and the characteristics of the ligament among some groups within the Nuculanidae such as in Ledella Verrill and Bush, 1897, Propeleda Iredale, 1924, and Nuculana Link, 1807, we believe that the proposed subfamilies should be used for taxonomic purposes. Whether they represent a natural division or not is a matter to be discussed later in a phylogenetic study.
The genus Microgloma is still in need of review since, as stated by Ockelmann and Warén (1998), "similarity in hinge structure to juvenile specimens of Yoldiella and other nuculanids directly suggests progenesis. We assume that the species of Microgloma simply are species derived from Yoldiella or Ledella, which mature at a much smaller size than normal in these taxa. (…) Possibly the genus Microgloma is polyphyletic, since progeneses may have taken place more than once. This will be difficult to prove or disprove." The similarities with some species of the genera cited above may confuse many researches in the identification of Microgloma species and the validity or status of the genus must be revisited. We believe this is an issue to be resolved with molecular analysis. At present we can only assume the genus to be valid.
The sculpture pattern on the prodissoconch surface is not a character commonly used in taxonomy of the protobranchs and, considering the confused taxonomy of the protobranchs (as seen by the genus Microgloma), we believe it might be useful to better determine the genera. This character has been recorded recently in the literature, and some species, from different families, show particular patterns. The reticulated sculpture on the prodissoconch surface seems to be a common character for the Nuculanidae, and has been recorded for the genera Nuculana, Sacella Woodring 1925(Allen 1993, Ockelmann and Warén 1998, La Perna 2007, Propeleda (seen by Natalia Benaim; unpublished data), and the Bathyspinulidae, in Tindariopsis agathida Dall, 1889 (seen by the present authors; unpublished data). Some members of Nuculidae have ridges or knobs on the prodissoconch surface (Gofas andSalas 1996, Zardus 2002). The genus Yoldiella presents a smooth prodissoconch surface, but the species Yoldiella philippiana (Nyst, 1845) presents a radial prodissoconch sculpture with a web-like pattern (Ockelmann andWarén 1998, Salas 1996) as seen here in M. nhanduti sp. n. Salas (1996) also illustrated radial ridges on the prodissoconch of M. pusilla and M. tumidula from the Iberian Peninsula. The radial ridges present in M. macaron sp. n. and M. nhanduti sp. n. are distinctive characters that should be considered in future descriptions of species of Microgloma to aid in resolving the status of the genus.

Conclusion
The apparent absence of species of the genus Microgloma along the Brazilian coast wais an artifact, reflecting the logistical difficulties in obtaining material from these depths. Once this material became available, additions to the fauna was brought to light. The description of two new species of Microgloma and the new information on the conchology, and bathymetric and geographical distributions of M. mirmidina and P. alba contribute to knowledge of the biodiversity of deep-sea mollusks of the Campos Basin and Brazil.