Traditional and confocal descriptions of a new genus and two new species of deep water Cerviniinae Sars, 1903 from the Southern Atlantic and the Norwegian Sea: with a discussion on the use of digital media in taxonomy (Copepoda, Harpacticoida, Aegisthidae)

Abstract Aegisthidae is one of the most abundant and diverse families of harpacticoid copepods living in deep-sea benthos, and the phylogenetic relationships within the family are in state of flux. Females of two new deep-water species of harpacticoid copepods belonging to the Hase gen. n. (Aegisthidae: Cerviniinae) are described. The first taxonomic description of marine copepod species based on the combined use of interference and confocal microscopy for the study of the habitus and dissected appendages is presented here. CLSM (Confocal Laser Scanning Microscopy) is a non-destructive method, comparable in quality to SEM (scanning electron microscopy) at the same magnifications. To observe and reconstruct in detail the habitus and dissected appendages, whole specimens and dissected parts were stained with Congo Red, mounted on slides with glycerine for CLSM and scanned under three visible-light lasers. Hase lagomorphicus gen. et sp. n. and Hase talpamorphicus gen. et sp. n. were collected from the sediments of the Southern Atlantic and the Norwegian Sea, from 2270 m and 5468 m depths, respectively. Hase gen. n. is included within Cerviniinae based on the caudal rami which are relatively divergent. Hase gen. n. is the sister taxon of Cerviniella based on the following synapomorphies: sturdy body, exopodites 1–3 of pereopods 1–3 heavily built, transformed into digging limbs, with strong outer and distal spines/setae, two-segmented endopod on the pereopods 2 and 3, and a reduced pereopod 5. Compared to Cerviniella, Hase gen. n. exhibits a more developed armature on the pereopod 1, which has outer and distal elements transformed into strong and long spines vs. stiff setae on Cerviniella.Hase gen. n. has one or two strong and long spines on the inner margin of the exopodite 3 of pereopod 4 and pereopod 5 is fused to the somite, ornamented with three distal setae. The telson of Hase gen. n. is subquadratic, and the furca is among the shortest yet described for Aegisthidae. The new species differ in a number of diagnostic characters, three of which are: a) the somite bearing pereopods 3 and 4 with latero-distal spiniform processes in H. talpamorphicus gen. et sp. n. but smooth in H. lagomorphicus gen. et sp. n., b) antenna is armed with three stout spines on the lateral inner margin of the exopod in H. talpamorphicus gen. et sp. n. and two proximal setae in H. lagomorphicus gen. et sp. n., and c) pereopod 4 exopodite 3 has two long and strong spines on the inner margin in H. lagomorphicus gen. et sp. n. and one spine in H. talpamorphicus gen. et sp. n. The high quality of CLSM images should foster discussion about the use of high quality digital images as type or as part of the type series in zoological studies, especially when studying rare and small macrofaunal and meiofaunal taxa.

The paper describes two new species of copepod crustaceans designated to a new genus of the subfamily Cerviniinae (Harpacticoida: Aegisthidae) found in the deep waters of Southern Atlantic and Norwegian Sea. This is the first formal description of a marine copepod species based on combined use of interference and confocal microscopy in study of dissected appendages and the genital field. The methods for the acquisition of 3D rendered images are described by Corgosinho et al. (2017) and Kamanli et al. (2017). One of the most important advantages of using Confocal Laser Scanning Microscopy (CLSM) over Scanning Electron Microscopy (SEM) is that CLSM is a nondestructive imaging technique for the study of whole microscopic animals or small parts of them, such as millimetre or micrometre-long hydrated structures (see extensive discussion by Kamanli et al. 2017). In addition, the use of CLSM improves the presentation of appendages and structures in their natural 3-dimentional state, a property not easily transferable by the 2-dimentional inked drawings. High quality CLSM photos could substitute the need of taxonomists to acquire type specimens from Museums for comparisons, therefore lessening the burden of understaffed museums but also decrease the likelihood of a lost or damaged type material through the transfer of the specimens back and forth to the Museum. High quality CLSM photos depict structures as they appear in reality and potentially remove any shortcomings that an inexperienced taxonomist may have and will increase the quality of publications. The new species were sampled in the frame of the Census of Marine Life subproject Ce-DAMar (Census of the Diversity of Marine Abyssal Life, [2000][2001][2002][2003][2004][2005][2006][2007][2008][2009][2010] and the IceAGE (Icelandic marine Animals: Genetics and Ecology, since 2011) project. The CeDAMar was a ten-year multinational project (from 2000 to 2010) devoted to map the world biodiversity in the abyssal plains between 4,000 to 5,000 meters deep. The aim of the IceAGE project was to combine classical taxonomic methods with modern biodiversity research, in particular phylogeography and ecological modelling in the climatically sensitive region around Iceland.

Material and methods
The copepods were extracted from sediment samples of three scientific cruises of the Research vessel (RV) "Meteor". Sediment samples (5127-5455 m depth) were collected by a multi corer (MUC) during the DIVA-1 expedition of the RV "Meteor" (Cruise No. M48/1) to southeast Atlantic Ocean in July-August 2000. During DIVA-2 Expedition of the RV "Meteor" (Cruise No. M63/2), samples were taken by a MUC in the equatorial east Atlantic at depths >5000 m. Additional samples were collected by a box corer (BC) during the Overflow, Circulation and Biodiversity Expedition of the RV "Meteor" (Cruise No. M85/3) 307-2749 m deep (Fig. 1, Table 1), in the northernmost North Atlantic and the Arctic Ocean. Temperature and salinity were obtained by a CTD probe coupled to the MUC and BC.
For taxonomic studies, specimens were stained with Congo Red and mounted on slides for confocal laser scanning microscopy (CLSM) following Michels and Büntzow's (2010) protocol. We used the following equipment and settings: Leica DCR 5000 SP5 (Leica, Wetzlar, Germany) equipped with a Leica DM 5000B microscope (Leica, Wetzlar, Germany) and three visible-light lasers (DPSS 10 mW 561 nm; HeNe 10 mW 633 nm; Ar 100 mW 458 nm, 476 nm, 488 nm and 514 nm), combined with the software LAS AF Lite, Leica Application Suite Advanced Fluorescence (Leica, Wetzlar, Germany). Series of stacked images were obtained, collecting overlapping optical sections throughout the whole preparation. Final images were obtained by maximum projection, and CLSM illustrations were composed and adjusted for contrast and brightness using the software Adobe Photoshop CS6 (Adobe Systems, San José, U.S.A.).
The habitus was drawn from whole specimens temporarily mounted in slides with glycerine, adhesive plastic discs were used to support the cover slip and prevent destruction of the specimen (Kihara and Falavigna da Rocha 2009). After CLSM microscopy, specimens were dissected under a Leica MZ12.5 (Leica, Wetzlar, Germany). Dissected parts were mounted on slides using glycerine as mounting medium, and preparations were sealed with transparent nail varnish. Drawings were made under a Leica DMR microscope equipped with Nomarsky interference contrast and a drawing tube at 400× and 1000× magnification (Leica, Wetzlar, Germany). Final illustrations were "digitally inked" using Adobe Illustrator CS6 (DIVA-1 and DIVA-2 species) or free hand inked (IceAGE species).
The terms 'furca' and 'telson' are used according to Schminke (1976). Terminology and homologisation of maxillary and maxillipedal structures follow Ferrari and Ivanenko (2008). Therefore, by the application of serial homology, the nomenclature of Huys and Boxshall (1991) for Mx2 (fig. 1.5.5,p. 26) is modified as follows: praecoxa of Mx2 is hereafter recognized as syncoxa (praecoxa and coxa), coxa is considered as the basis, and the basis is recognized as the first endopodal segment with claw. Other morphological terms are used according to Huys and Boxshall (1991).
The following abbreviations are used in the text: Hyphen (i.e., "-") between figure numbers, structures, number of spines and setae, etc. indicates all between and is inclusive (ex: P1-P5 = P1, P2, P3, P4, and P5; A-C = A, B and C; etc.) The type material is deposited at the Forschungsinstitut und Naturmuseum Senckenberg (SMF) in Frankfurt, Germany.

Order Harpacticoida Sars, 1903 Family Aegisthidae Giesbrecht, 1893 Subfamily Cerviniinae Sars M., 1903
Hase gen. n. http://zoobank.org/158094D3-E3A5-427B-BFD7-1223015ABF72 Diagnosis. Female body sturdy with clear distinction between prosome and narrower urosome. Prosome 5-segmented, with cephalosome and four free pedigerous somites. Cephalosome with minute spinules covering surface and anastomosing reticulation towards rostrum and along margins; posterior margin slightly serrate. Pedigerous somites with reticulation along ventroposterior margins; lateral margins of third and fourth pedigerous somites smooth or expanded posteriorly forming hook-like projections laterally; posterior margins serrate. Urosome 5-segmented, comprising P5 bearing somite, genital double-somite, two free abdominal somites, and telson. Genital double-somite and two free abdominal somites with hook-like projections ventrolaterally. Genital double-somite original segmentation indicated by transverse, serrate surface ridge with reticulation and sensilla dorsal and laterally, completely fused ventrally; genital field with copulatory pore located in median depression; gonopores covered by operculum derived from sixth legs and by anteriorly directed flap arising from somite wall; P6 fused genital opercular plate armed with two setae. Telson with well-developed anal operculum; large anal opening with folded and reticulated cuticle; surface ornamentation consisting of pair of sensilla dorsally, minute spinules and pair of pores ventrally; ventral posterior margin with minute setules. Furca symmetrical; approximately 3.4× as long as maximum width; distinctly convergent. Each ramus with seven setae: setae I-III not inserted close to each other; seta I proximal, laterally inserted, spiniform and bipinnate; seta II median, dorsally inserted, spiniform, and bipinnate; seta III subdistal, laterally inserted, spiniform and bipinnate; setae IV and V distally inserted, bipinnate and fused basally; seta VI distally inserted, minute and naked; seta VII dorsally inserted, close to seta III, tri-articulate at base and pinnate.
Body ( Fig Genital double-somite (Figs 2B, C, 3A-C, 4E) original segmentation indicated by transverse surface ridge with reticulation and sensilla dorsal and laterally, completely fused ventrally; genital field (Figs 3C, 4E) with copulatory pore slightly covered by a proximal flap, pointing posteriorly, located in a soft median depression; gonopores covered by operculum derived from sixth legs and anteriorly directed flap, medially depressed, arising from somite wall; P6 bearing two naked seta.
Mx1 (Figs 6C(c1-c4), 8A, B). Praecoxa with row of spinules; arthrite well developed, with one pinnate and one smooth seta on anterior surface, four smooth spines, and three pinnate spines along distal margin (two ornate with two large spinules at     basis), four pinnate setae on aboral margin, two fused at basis. Coxa endite cylindrical, bearing five setae (four naked and one pinnate) distally; epipodite absent. Basis and enp fused; basis with eleven setae; enp incorporated into basis, represented by two naked setae. Exp 1-segmented, with three bipinnate setae.
Male unknown.
Genital double-somite (Figs 12C, 13B, E, 19D) original segmentation indicated by transverse, serrate surface ridge with reticulation and sensilla dorsal and laterally, completely fused ventrally; genital field (Figs 12C, 13B, E, 19D) with copulatory pore completely visible, not covered by a proximal flap as observed for the previous species, located in a well-developed median depression; gonopores covered by operculum derived from sixth legs and by anteriorly directed and straight flap arising from somite wall; P6 bearing two naked setae.
Md (Figs 16A, 17A). Coxa with well-developed musculature, gnathobase curved inwards, with several multicuspidate teeth and one bipinnate seta on inner distal margin; rows of spinules near insertion area of bipinnate seta. Palp well developed, with basis, enp and exp. Basis with four bipinnate setae and surface ornamentation as indicated in Figure 14A. Enp 1-segmented with three lateral setae (two bipinnate and one unipinnate) and seven apical setae (four naked, two bipinnate and one unipinnate). Exp 4-segmented, exp-1 as long as next three segments combined; armature formula:
Male unknown.

Taxonomic discussion and phylogenetic position within the Aegisthidae
According to Seifried and Schminke (2003), the monophyly of Aegisthidae is supported by the following female autapomorphies: 1) anal somite elongated, tapering posteriorly; 2) caudal rami more than twice as long as wide; 3) antennule of female 8-segmented; fusion of Oligoarthra segments 3 and 4; 4) antenna with allobasis or incomplete basis; 5) enp-2 laterally with one spine (III) and two setae (2 + 4), spine I lacking; 6) endopod of mandible of one large segment and at least two times longer than wide; 7) proximal segment of exopod elongated, considerably longer than remaining segments and at least 3 times longer than wide; 8) epipodite of maxillule represented by two setae; 9) exopod of maxillule reduced in size with three setae; endopodal element 11 of allobasis of maxilla developed as large, strong spine inserted on posterior surface; 10) P5 without endopodal lobe.
Hase gen. n. can be included within Aegisthidae on account of the above mentioned apomorphies 5, 6, and 10. Species of Aegisthinae are derived Cerviniinae and Cerviniopseinae (Seifried and Schminke 2003). Cerviniinae and Cervinopseinae are paraphyletic and as such, are defined by plesiomorphies. Therefore, Hase gen. n. cannot be placed within any of the subfamilies on the account of synapomorphies and its taxonomic position must be typological and on account of the close proximity to one of the taxa composing a given subfamily.
Hase gen. n. has an antenna with four-segmented exp and could be included within both Cerviniinae and Cerviniopseinae. However, Cerviniinae and Cerviniopseinae are to date separated according to the degree of divergence of the caudal rami (see Boxshall
In Cerviniella the whole exopod or the exopodite-2 and 3 are fused on the P1-P3 (sy), keeping the inner and outer armature of the original segments, the endopodite of P1 is never 3-segmented (sy) and the P4 undergoes a further reduction both in segmentation and/or armature of the exopod and endopod (sy) (viz. Kihara and Martínez Arbizu 2012). The strongest armature occurs on P2 (sy) and P3 (sy), with the P2 somite showing a very large proximal region of weakly sclerotized cuticle. Within Hase gen. n. the armature is more developed on the P1 (sy). They are longer and stronger on P1 and P2; shorter yet stout on P3 and P4. The P1, although keeping the plesiomorphic 3-segmented exopodite and endopodite, have the outer and distal elements transformed into strong and long spines (sy), kept as stiff setae on Cerviniella (pl). Hase gen. n. has one or two strong and long spines on the inner margin of the exopodite-3 of P4 (sy). The P5, which is fused to the somite (sy), is stalked and with three distal setae (sy). Additionally, the anal somite of Hase gen.n. is subquadratic, slightly tapering posteriorly, wider than longer (sy) and the caudal rami is one of the shortest yet described for this family (sy), with spiniform setae I to III (sy).
Interestingly, the same morphology of the P5, telson and furca is depicted by Brotskaya (1963) in the deep-sea genus Paracerviniella. This author briefly described Paracerviniella based on a male specimen only, as follows: Body without outgrowths. The first thoracic somite not completely separated from the cephalothorax. Posterolateral corners of body somites, except for the first thoracic, drawn into pointed outgrowths. The posterior edge of all somites, except the anal, armed with a number of small teeth. Furcal rami 1.5 times shorter than the anal somite, width at the base one and a half times less than the length. The first antenna six-segmented, with two enlarged basal segments; the second, third and sixth segments of the male with sausagelike sensory cylinders, the fourth segment with a hooked spine, the fifth segment with two sensory cylinders of the usual structure. Both branches of P1-P4 triplesegmented. Endopodite of P1 and P2 with clawed spine at the distal part. P5 and P6 1-segmented with three apical bristles.
Most of these characters are not informative enough to allow the inclusion of Paracerviniella within any monophyletic clade within the Aegisthidae. With exception of the clawed spine present on the endopodites of P1 and P2, the P5 morphology and armature and maybe body ornamentation, the remaining characters are gender-linked or plesiomorphic within the family. In addition, the illustration of some characters that could be informative, such as the mouthparts, is insufficient. Considering this and on 1 seta and 1 spine, 3 setae and 1 spine, and 2 setae and 1 spine (Fig. 17D).

Genital Field
Copulatory pore Slightly covered by a proximal flap, pointing posteriorly, located in a soft median depression (Fig. 3C).
Completely visible, not covered by a proximal flap as observed for the previous species, located in a well-developed median depression (Fig. 13E).

CLSM vs. SEM technology
There are some important differences among the scanning microscopy systems that produce high quality imaging, especially regarding to the subsequent fate of the specimens and the resolution limits. Some image systems (e.g., SEM) inevitably destroy type specimens; CLSM is highly desirable in this aspect as the studied specimen remains intact. According to Kamanli et al. (2017), the images obtained by CLSM are comparable in quality to SEM at the same magnifications, and the technique offers a 3D data set. In addition, the sample preparation routine for CLSM is simpler than that for SEM, it is practically a non-destructive method, and allows the study of hydrated material. It is difficult to establish a good SEM protocol for the study of miniaturized body parts of small macrofauna and meiofaunal specimens. Not infrequently they can be lost during manipulation, damaged before any observations are made (Michels 2007), or rendered unusable and in vain even during later processes such as coating in which the structure can become over-coated. CLSM also allows the appendages to be manipulated within the mounting medium to offer views of the specimen from multiple angles, which can be problematical to achieve using SEM since some viewpoints may be inaccessible due to the way that the specimen is mounted and the tilt limitations in SEM (Kamanli et al. 2017). After scanning, the material can be recovered intact and kept as a voucher. An example where CLSM is advantageous in the present species description is the dor-sal (5B) and outer (5C) view of the same A1. In addition it offers a clear view of the natural 3-dimentional state of the antenna and the exact position of overlapping A1 setae, an arduous task during the traditional drawings of this structure. The continuous technological advancements in the field of microscopy are reducing the resolution gap among the different technologies. The resolution of SEM is approximately 10 nm whereas confocal microscopes have the potential to resolve microstructures in the 50 to 100 nm range (Schrader and Hell 1996). Practically, CLSM has reached a resolution comparable to SEM (Butler et al. 2010). In many situations, enhancing resolution beyond this range does not result in an increase in useful biological information about the specimen (St. Croix et al. 2005). Now, even for the smallest meiofaunal larvae, this level of resolution is more than sufficient to fully capture and catalogue the most minute external details such as pore morphology or individual setal ornamentation.

The importance of digital image acquisition in taxonomy
Recent papers have highlighted the importance of image acquisition in taxonomy (e.g., Michels 2007, Neusser et al. 2009, Neusser et al. 2011, Faulwetter et al. 2013, Akkari et al. 2015. Garraffoni and Freitas (2017) argued that the International Code of Zoological Nomenclature should be modified to allow, in some cases, as in the study of rare or soft-bodied meiofaunal organisms, the deposit of high quality photographs and videos as Type material. This proposal has met with strong opposition among some researchers (Dubois 2017, Rogers et al. 2017. The evolution of optical systems has led to the exponential increase in the use of high quality imaging systems in all fields of biology, including taxonomy. Our opinion in this debate is that the image quality obtained by scanning through either CLSM, SEM or Micro CT is so high, that we should consider how viable it is to designate a photomicrography as Type material. Although this may sound provocative, we must consider that a well-curated image lasts potentially forever, whereas the type specimen, especially when we take into account small macrofauna and meiofaunal groups, may deteriorate fast during study or even when mounted on "permanent" slides, those have a half-life of only few decades or centuries. Diminution of trained museum staff to maintain collections only exacerbates this problem and highlights the need to seek alternative solutions to record and study taxonomically the world's biodiversity (Decker et al. 2018). Hence, the use of CLSM and other high quality image acquiring systems should be considered not only as complimentary evidence to a taxonomical study. In some cases, the images generated should be also considered if not the type alone, at least part of the type series.

Conclusions
This contribution highlights the diversity of exquisite bauplans in deep-sea copepods and the broad distribution of a meiobenthic crustacean genus in the Atlantic Ocean basins. Additionally, it is a showcase on how confocal microscopy can assist in provid-ing a better and more accurate description of small macrofaunal and meiofaunal organisms. We favour the inclusion of digital media at least as a component of the type series and we encourage the discussion for setting standards for such data. Additional studies and sampling effort must be continued to find the male of the genus Hase gen. n. to improve comparisons with Paracerviniella and Cerviniella.