Shallow-Water Northern Hemisphere Jaera (Crustacea, Isopoda, Janiridae) Found on Whale Bones in the Southern Ocean Deep Sea: Ecology and Description of Jaera tyleri sp. nov

The skeleton of a natural whale fall discovered in the Southern Ocean at 1,445 m was densely covered by one small, janirid isopod. Jaera tyleri sp. nov. is the first of its genus found in the southern hemisphere and in the deep sea and is described herein. Morphological and molecular investigations revealed the systematic position of this species new to science. Phylogenetic analysis of the 18S locus confirms that this species falls in a strongly supported monophyletic clade of Jaera species. The whale bone habitat of J. tyleri sp. nov. is quite different from those of other species of the genus Jaera. The analysis of bathymetric and distribution patterns of the Janiridae in general and Jaera specifically confirm the unusualness of the habitat for this isopod species. The abundance of J. tyleri sp. nov. on the whale bones and its absence from other nearby habitats suggests it to be a whale-fall specialist. The analysis of the size-frequency distributions of J. tyleri sp. nov. suggests multimodal population structure with continuous breeding activity throughout the year. The fecundity of the species is low but in line with reduced fecundity observed in polar and small-sized isopods.


Introduction
Until the discovery of whale falls as habitats of rich chemoautotroph communities in 1987 [1], studies of marine fauna discovered on dredged up whale bones have been more anecdotal and included taxonomic descriptions of the species [2,3] and the aspect of whale bones as organic food sources [4]. Since then, both natural and experimentally-implanted whale carcasses have been subject of research on the degradation of the carcass, the food enrichment of the deep-sea environment and its effects on the surrounding benthic biodiversity [1,[5][6][7][8] Smith and Baco [5] defined three succession stages in the decay of whale falls, which attract different types of scavengers and food specialists. During the first stage, the mobile-scavenger stage, soft tissues are removed by mobile vertebrate and invertebrate scavengers like sleeper sharks, hagfish, cirolanid isopods, lysianassid amphipods or lithodid crabs [5,9]. The second, enrichment-opportunistic stage is characterised by dense assemblages of heterotrophic, often invertebrate fauna, feeding on the remaining soft tissues and bones [5,10]. The third stage, the sulphophilic stage is distinguished by diverse, trophically complex assemblages living on the skeleton, including whale-bone feeders, bacterial grazers, species utilising chemoautotrophic endosymbionts, deposit feeders, facultative suspension feeders and predators [5]. The whale skeletons emit sulphides from the anaerobic breakdown of bone lipids, which form the basis of a chemoautotrophic environment that is host to rich and abundant assemblage of specialists, like the bone eating polychaetes (Osedax spp.) and gastropods (Pyropelta spp.), the bivalve Idas washingtonia (Bernard 1978) or the isopod Ilyarachna profunda Schultz 1966 [5,11].
The presence of isopod species at whale falls has been rarely reported, as research often focussed on the fish, decapod, gastropod, bivalve and polychaete species [5,6,9,12,13]. The giant cirolanid isopod Bathynomus giganteus Milne Edwards 1879 has been observed and filmed at relatively new whale falls, scavenging on the soft tissues [14]. The munnopsid Ilyarachna profunda has been recorded at whale skeletons in population sizes of 500-1800 [5] and appears to be the most numerous isopod species on whale falls. Smith and Baco [5] also mention the presence of family Janiridae at whale falls from southern California but do not specify the generic or species affiliation.
The marine isopod family Janiridae is globally distributed with records spanning from the Arctic to the Antarctic and from the Atlantic, Indian and Pacific oceans [15]. At present 174 species of the 23 genera are assigned to the Janiridae, although previous studies have shown that this family is not monophyletic and requires taxonomic revision [16,17] These species occur over a wide range of habitats, including the intertidal, estuarine areas, salt springs, anchialine caves and on whale carcasses [5,15] and from the intertidal to the hadal deep sea (e.g. [18,19,20]). Several of the intertidal and shallow water janirid species are herbivores and grazers, eating seaweed, dead wood and bacterial films; feeding strategies of the deep-water species are unknown [21][22][23][24][25][26][27]. Species of the Janiridae are known to be highly adjustable to their environments, including wide tolerances to salinity, temperature and oxygen stresses [28,29]. The genus Jaera with its main distribution in the northern hemisphere has been subject to several ecological and reproductive studies [30][31][32][33][34][35].
Here we describe the first deep-water, bathyal species of the genus Jaera, found on a natural Antarctic whale fall, present information on its ecology and review the depth and biogeographic distribution patterns of Jaera and the Janiridae in general.

Ethics statement
All necessary permits were obtained for the described field studies. Studies in the East Scotia Sea were undertaken under the permit S3-3/2009 issued by the Foreign and Commonwealth Office, London to section 3 of the Antarctic Act 1994.

Study site
During the expedition JC 42 of the RRS James Cook a baleen whale skeleton was discovered during dive 148 in a caldera next to the Kemp Seamount (59u41.6 S, 28u21.1 W (DDM)) in 1,444 to 1,447 m water depth (Fig. 1) in the vicinity of the South Sandwich Islands [8,36]. The whale fall was examined using the imaging systems of the remote operating vehicle (ROV) Isis, of the National Oceanography Centre Southampton, a Scorpio digital still camera with flash unit was used as well as two high definition video cameras (1080i) ( Fig. 2A, B).

Morphological analysis
Specimens were collected by picking up whale bones with the seven-function Kraft Predator arms of ROV Isis, which were stored in a biobox. At arrival on deck the whale bones were transferred to the 4uC lab and photographed (Fig. 2C, D) before epifaunal sampling began. Isopods were either picked by forceps or decanted from bottom of the sorting trays, fixed in pre-cooled 96% ethanol or 4% buffered formaldehyde and the latter after 3 months transferred to 70% ethanol. The material was sorted in the laboratory, measured and identified using a Leica MZ 12 stereomicroscope equipped with a camera lucida, and drawn using an Olympus SZX7 compound microscope, also equipped with a camera lucida. For scanning electron microscopy (SEM), whole animals were vibrated in single vials for 30 sec in soapy water, rinsed first with distilled water, than 96% ethanol and air dried. The specimens on the whale bone piece were air dried only. The specimens were mounted on aluminium stubs and viewed uncoated in a Hitachi TM3000 (Fig. 2E, F).
The terminology and measurements mostly follow Wilson [37]. Total body length was measured medially from the tip of the rostrum to the posterior tip of the pleotelson. The dorsal view was used for measuring the width, while the length of body segments was measured in lateral view. For the description of the body and pereopods the holotype was used, and for mouthparts and pleopods a male paratype was dissected and for some details that the paratype lacked or were different from the male, a female paratype was used.

DNA sequencing was performed at LGC Berlin
Germany. All sequences were edited and aligned in Codon-Code Aligner Version 3.5.6 (CodonCode Corporation 2006). Sequence quality was evaluated using ''Phred'' quality scores, excluding sequences with values ,300 [40,41]. Electropherograms were also manually examined for sequencing errors and, where possible, variable positions were confirmed by reference to the corresponding reverse sequences. Close relatives of the sequences were determined using the 'Blast' searching tool in GenBank and revealed closest matches with asellote isopod species reported in Raupach et al. [16]. The 18S fragments were aligned with additional Jaera species, Jaera albifrons (AF279609) and J. nordmannii (AF279610), as well as the 18S alignment obtained from Raupach et al. [16]. The whole fragment of 18S was analysed (2,302 bp) as well as a dataset truncated to the length of the J. tyleri fragment (680 bp). Outgroups from the Stenetriidae family were used, following Raupach et al. [16]. The 18S alignment was constructed using RNA Salsa v0.8.1 [42] after generating a 'guide' secondary structure using the 'mfold' web server (http://mfold.rna.albany. edu/?q = mfold). This analysis uses a thermodynamic folding algorithm to align DNA with respect to inferred stem and loop regions. Matching, mismatching and gap opening penalties were applied in this analysis with a default stringency of 0.6.

Molecular Phylogenetic Analysis
The 18S alignment was subjected to partitioned maximum likelihood (ML) analysis with RaXML v7.4.2 [43], using a mixed model to account for different evolutionary processes occurring in the stem and loop regions. A standard general time reversible DNA substitution model with a gamma correction for rate heterogeneity was applied to the loop regions. Stems were analysed using multi-state secondary structure models which allow for 6-16 movement probabilities between different paired states (RaXML models 6A-D, 7A-D and 16A-D, [44]). ML analyses were conducted ten times for each model choice. Models were then compared using Akaike Information Criterion (AIC) scores, calculated from the log-likelihood values and numbers of free parameters (P) for each model (AIC = 22LnL+2P). The model with the smallest AIC score was then chosen and a maximum likelihood bootstrap was conducted with 1000 replicates.

Biogeographic and depth analysis
Data on bathymetric and georeferenced distributions of the Janiridae were obtained from original literature references. The isopod world list [15] http://www.nmnh.si.edu/iz/isopod/, www. gbif.org) was used but with Jaera taxonomy edited following Harvey and Naylor [45], Tobias et al. [46] and Borza [47]. It is clear that because data are pooled from a variety of sources, not all taxa will necessarily occur together at one depth, however, they have at least been reported at a single location at that depth interval. For the distribution analysis of Jaera only species confirmed records were used.
For the analysis of bathymetric distribution patterns, water depth was divided into 100 m wide depth intervals. For the account of species numbers per depth the number of species found in the depth interval was taken. For the assessment of the depth range at genus level, the presence at each depth interval of species of the selected genus was taken. When breaks occurred in the depth distributions of two or more species within a family, a break is recorded in the results. The term bathyal was used for stations ranging from , 1,000 m down to 3,000 m, abyssal from , 3,500 -6,000 m where the hadal starts [48].

Ecological analysis
A representative subset of paratypes was measured, growth stages and sex identified using Leica MZ 12 and Zeiss Stemi SV6 stereomicroscopes equipped with eyepiece micrometers. Total body length was measured to the nearest 0.1 mm. Growths stages were separated following morphological characteristics into Manca I (6 pereomeres, 6 legs, tiny 7th pereonite), Manca II (small 7 th pereonite, short 7 th leg), immature male (short, not fully developed 1 st pleopod), male (fully developed 1 st pleopod), immature female (1 st pleopod absent, 2 nd pleopod formed the operculum, no oostegites present), female (oostegites present) and ovigerous female (eggs in brood pouch). If present, numbers of eggs in brood pouch were counted.

Nomenclatural Acts
The electronic edition of this article conforms to the requirements of the amended International Code of Zoological Nomenclature [49], and hence the new name contained herein is available under that Code from the electronic edition of this article. This published work and the nomenclatural act it contains have been registered in ZooBank, the online registration system for the ICZN. The ZooBank LSIDs (Life Science Identifiers) can be resolved and the associated information viewed through any standard web browser by appending the LSID to the prefix ''http://zoobank.org/''. The LSID for this publication is: urn:lsid:zoobank.org:pub: 112A7E66-FB6E-4821-A2AA-52E28156CFC0 and for Jaera tyleri sp. nov. is: urn:lsid:zoobank.org:act urn:lsid:zoobank.org:act: 2B77D993-22F1-4EB0-906A-7D210E3E0437 The electronic edition of this work was published in a journal with an ISSN, and has been archived and is available from NERC's Open Research Archive NORA.

Distribution patterns in Janiridae
The depth distributions of the 23 genera in the isopod family Janiridae and their 174 species have been analysed as well as the global and depth distributions of the genus Jaera itself. The majority of the species within the Janiridae occur in the intertidal and shallow shelf waters within 100 m depth (Fig. 3A), but single species can be found down to abyssal and hadal depths, like Janthura abyssicola Wolff 1962 and J. bougainvillea (Birstein 1963) (Fig. 3B). Of the 23 genera, 14 have so far solely been recorded within the first 100 m of water depth and only in 5 genera species exceed into the deep sea beyond 1,000 m ( Fig 3B). The presently known bathymetric records of Jaera species show a 1,300 m depth gap between the 20 previously known species and the new discovered species described herein (Figs. 3B).
Uropods inserted adjacent to each other, at least protopods not protruded behind outline of pleotelson, protopod parallel-sided, squat, much shorter than pleotelson, rami reduced, shorter than protopod.
Biogeographic remarks. In total 678 species-unique georeferenced locations for all species of Jaera have been analysed. All species of Jaera, apart from the new Antarctic one, are reported from shallow waters and the northern hemisphere (Fig. 4).    Maxilla 1 (Fig. 7) lateral endite 1.1 as wide as mesial endite, distal margin truncated with short spine-like distomedial seta and 11 stout curved comb-like setae, articulated more medially; mesial endite distal margin rounded with tuft of slender setae and 5 stout setulated setae.
Maxilla 2 (Fig. 7) mesial endite broadest, middle endite narrowest, lateral endite slightly wider than middle, all endites of approximately the same length; mesial endite distal margin truncated, with two rows of stout setae ending with small hook; lateral and middle endites each with 4 comb-like distal setae (2 long and 2 slightly shorter).
Additional description of female (Figs 11-15). Proportions of habitus and body parts in females are rather equal to those of male; differences will be described in the following. Body (Fig. 11) length 1.6 pereonite 3 width, pereonites 3-7 of same width, not broadened posteriorly like in male, caudal notch of pleotelson shorter than in male, 0.1 of pleotelson length. Antennae (Fig. 12) more slender, antenna 2 less setated than in male. Antennular article 1 length 1.2 width, with 4 lateral and 5 distomedial fine setae, article 2 0.8 as long and 0.4 as wide as article 1, with 7 lateral setae, 1 stout and 3 long slender  distomedial setae; articles 3-5 evenly tapering distally, with many fine setae, article 3 0.5 as long as article 2, articles 4 and 5 0.5 as long as article 3, article 5 with 4 long distal broom setae. Antenna 2 0.3 of body length, article 2 broadest (in male article 3 broadest), distolateral projection with 2 stout setae, articles 3-6 more slender than in male, articles 3 and 4 subequal in size, about 0.8 as long and as wide as article 2, article 3 with 1 distomedial stout seta, article 4 with 3 distomedial stout short setae, article 5 twice as long as article 4 medially, with 4 distal stout setae and many simple short setae, article 6 1.5 as long as article 5 with 1 stout distomedial seta and several simple slender setae. Flagellar article 1 longest, 0.4 as long as article 6, with only 2 (proximally) and 1 (distally) simple setae.
Mandibles (Fig. 13) with 3 long and 2 short teeth on left incisor (8 on right mandible), lacinia mobilis of left mandible slightly bent distally, with 5 teeth, 0.7 length of incisor process; spine row with 6 spines on left and 8 on right mandibles respectively; molar process 0.8 length and 0.6 width of incisor process on proximal part, triturative surface concave, with long dorsal and many lateral denticles and 3 ventral setae; condyle 0.1 times body length, 0.7 length of molar; palp 0.9 body length; article 1 with 1 slender distal seta; article 2 1.1 of length and 1.3 of width of article 1, with 2 bilaterally comb-like setae in distomedial half; article 3 slightly curved, 0.7 as long and as wide as article 2, with row of stout ventral setae and many fine setae distodorsally. Hypopharynx (Fig. 13) outer lobes 1.4 of inner lobes width, with dense distal fine setae, inner lobes without setae. Maxilliped (Fig. 12) as in male. Pereopods (Fig. 14) of similar shape and proportions, getting longer posteriorly. Pereopods more slender in comparison with male's pereopods (for example, length/width ratios of carpus of pereopods 2, 4 and 6 are 2.6, 2.5 and 3 respectively, for male it is 1.8, 2 and 1.8). Spine-like stout setae on ischium-carpus in female more setulated, but articles have less slender setae. Female pereopod 4 not shorter than pereopod 3 as in male.
Pleopod 2 (Figs 11, 15) about as long as wide, rounded, with shallow medial keel, many simple setae mediolaterally and distally, fine short setae scattered on surface. Pleopod 3 (Fig. 15) different Remarks: Jaera tyleri sp. nov. is the only blind species of Jaera that has ever been described and can therefore easily be distinguished from other species of the genus by this character as well as by the having the short antenna 2 (about a third of body  Table 1. Jaera tyleri sp. nov. Paratype information and collection numbers. length) and other characters mentioned in the presented diagnosis. By having small male pleopods 1 and 2 and enlarged opercular pleopods 3 J. tyleri is close to a group of Mediterranean species, especially to J. nordmanni, which possesses reduced eyes. Total views and details of J. tyleri are also provided in SEM images in Fig. 16.
Ecological remarks. Specimens of Jaera tyleri sp. nov. are found on all sites of the whale bones with exception of the parts that were submerged in the sediment ( Fig. 2A, B). The isopods were tightly holding on to the whale bone (Fig. 2C, D), having their pereopods actively stuck into the bone tissue (Fig. 2E, F). Population densities based on the number of specimens sampled per individual collected whale bone are estimated between 470 -6,000 specimens per m 2 .
In total 499 specimens were sex identified and measured, ranging from 0.8-3.8 mm in length (Fig. 17). Of these 58 specimens were in juvenile manca stages, 203 specimens were males and 238 specimens were females. The observed sex ratio showed a slight bias towards females (1:1.17 males: females). Sizefrequency distributions were analysed for stages and sexes of Jaera tyleri sp. nov. (Fig. 17), indicating a single cohort in the manca II stage (Fig. 8B) but multiple cohorts in immature females, males and females (Fig. 17D-G).
Twenty nine of the 104 mature females were ovigerous, carrying either eggs or juveniles (two specimens of 2.7 and 3.3. mm length) in their brood pouch (Fig. 17H). Their size-frequency distribution indicates that there are multiple reproducing cohorts (Fig. 17H). The numbers of eggs in the brood pouch varied from five to 11 and increases with size of the female (Fig. 18).

Molecular analysis
The sequencing was successful for 641 bp of 18S fragments for the specimens JC42F-0574-4 (GenBank KJ174500) and JC42-F-0574-6 (Genbank KJ174501). Both sequences were genetically identical. Sequence alignment with other Isopoda yielded full and truncated datasets of 2,302 and 680 bp respectively, containing 110 taxa. After RNA Salsa analysis, 18.7% and 13.8% of the two respective alignments contained gaps and undetermined characters. Estimated alpha = 0.22 within the loop regions for 678 bp, indicating a high degree of rate heterogeneity across this partition. For both alignments, the best model according to AIC scores was 6A.
The maximum likelihood phylogeny for secondary structure model 6A places J. tyleri basal to the two previously published Jaera species in a monophyletic group, with 100% bootstrap support from the 2,302 bp alignment ( Figure 19) and 86% from the 680 bp alignment ( Figure S1). The whole clade is placed basal within the Asellota, although bootstrap support for this position within the phylogeny is very weak (72% for the 2,302 bp alignment). In the 680 bp alignment, Janira maculosa is placed basal to the divergent Jaera clade but with weak (42%) support. In both phylogenies the Janiridae fall into four clades polyphyletic within the Asellota (Neojaera, Iais, Iathrippa and Jaera), but in each case the placement of these clades in the phylogeny is not strongly supported. Earlier analyses of these same genera using more sites (18S and 28S) showed the same pattern [16]. Both phylogenies provide strong support for the Jaera species as a highly distinct and evolutionarily divergent clade.

Taxonomic implications
The main difficulty in identifying the new species was the unusual habitat and the absence of eyes of the species. The new species has the usual shape of Jaera, but is blind. All known species of Jaera occur in shallow waters of the Northern Hemisphere, mainly the Atlantic and the Mediterranean and bear eyes. Besides a few characters, mainly the reduction of eyes and the short antenna 2, which distinguish the new species from other Jaera, the new species possesses a long list of characters (Table 2) which attributes it to Jaera. We therefore placing it into this genus what is also supported by the molecular analysis. Species of Jaera show different states of eye reduction, which is possibly connected with different life styles: from well developed eyes with 36 ommatidia in common littoral species like J. albifrons to reduced eyes with 5-6 ommatidia in J. hopeana, an ectocommensal living between the legs of Sphaeroma serratum. Species of the morphologically similar genus Iais Bovallius 1886 (Table 2), which are ectocommensals of isopods of the genera Sphaeroma, Exosphaeroma and other sphaeromatids have small eyes reduced to 2 ommatidia. By the reduction of eyes and by the short antenna 2 Jaera tyleri sp. nov. has similarities with some species of the janirid genus Caecijaera Menzies 1951, e.g. C.
The discovery of Jaera tyleri sp. nov. and the generic assignment of the janirid species found at southern Californian whale falls to cf. Janiura and an unidentifiable but not Jaera provides evidence for an underestimated species richness the family Janiridae especially in the deep sea. This study also indicates that the speciose family Janiridae urgently needs to be revised taxonomically, as both morphological and molecular characters, have shown that Janiridae are polyphyletic [16].
The isopod Jaera tyleri sp.nov. has been characterised with a highly divergent sequence of partial 18S rDNA. This species is challenging to align with other janirids but through careful alignment of this species with other janirids with reference to rDNA secondary structure, maximum likelihood cluster this species with significant (.95%) bootstrap support in a monophyletic group with other members of the Jaera genus. Raupach [personal comm.] previously reported extremely divergent 18S rDNA sequences of Jaera albifons and J. nordmanii, which were subject to long-branch attraction and therefore excluded from his Raupach et al. (2009) dataset [16]. Here we used the 18S rDNA Raupach et al. [16] dataset and included the sequences of Jaera albifons and J. nordmanii to analyse the phylogenetic position of Jaera tyleri sp.nov. Morphological characters support placement of Jaera within the Janiridae. The Jaera genus is however placed basal to other asellote isopods, with 72% support for this phylogenetic position. It is therefore still unclear where Jaera is placed phylogenetically in relation to other isopods. Further taxonomic and molecular survey of this unusual group is warranted to shed more light on this issue. Previous morphological and molecular phylogenetic studies on Isopoda including Janiridae showed evidence that this family is not monophyletic [16,17] and requires taxonomic revision.

Biogeography
The isopod family Janiridae is found in all oceans and seas from the intertidal to the hadal, although their distribution is predominantly in shallow waters and on the shelf [15]. Until the discovery of Jaera tyleri sp. nov., only four (Neojaera, Janiralata, Janira and Janthura) of the 23 janirid genera have been reported from the deep sea, including a record of Neojaera from 1,500 m in the Southern Ocean (SO) [52]. Of the 174 globally known janirid species, eleven species of six genera have been recorded from the SO [53]. The genera Austrofilus and Neojaera are most species rich with three of their species occurring in the SO; the fourth species of Austrofilus is known from Spain while the eight species of Neojaera show a Gondwanan distribution. Iathrippa is present with two species, while species of Caecianiropsis, Ianiropsis, Jaera are singletons. Here their overall bathymetric distribution ranges from the intertidal to 1,524 m, with more species living on the shelf than on the bathyal slopes and abyssal plains [52,53]. In their analysis of isopod abundances on SO slope and deep-sea Kaiser et al. [54] assessed the Janiridae to be patchily distributed.
The janirid genus Jaera has a northern hemisphere centric distribution and the presented new species J. tyleri sp. nov. is the first confirmed record of this genus for the southern hemisphere. The latitudinal as well as the bathymetric range of the genus has been extended.

Isopoda on whale falls
Detailed studies and identification of isopods associated with whale skeletons have been sparse in the past. The munnopsid Ilyarachna profunda and its abundance have been recorded from sulpophilic stage of whale falls [6]. The latter authors also reported unidentified Janiridae on whale skeletons in southern California. Their isopod collections from vertebrae picks and boxwashes of sampling since 1995 comprise seven morphospecies, now available to us for further identification and comparison with the Antarctic Jaera tyleri sp. nov. [Smith unpublished data]. No information is given on abundance of these morphospecies. The taxonomic identification revealed two species of Gnathiidae, two species of Munneurycope, one species similar to Acanthocope (both genera are Munnopsidae) and two species of Janiridae of which one is similar to the deep-sea genus Janthura and the other janirid was unidentifiable to genus. Amon et al. [8], describe the position, age and associated fauna of the Kemp Caldera whale fall, and remark on the presence of a Jaera sp. nov. and its occurrences within the whale skeleton, which relates to the species described herein. Their review on the Isopoda associated with whale falls names the Munnopsidae (4 species) and Janiridae (3 species) as the most frequently found and diverse families. While the Munnopsidae are very species-rich and commonly found in the deep sea, especially in the SO (reported with 219 species from the ANDEEP expeditions [55]), the Janiridae are less diverse and predominately shallow water taxa [53]. Gnathiidae are known to be parasitic, blood-sucking parasites on fish [56] and their occurrence near whale falls might be more related to the presence of scavenging fish.

Ecology
The isopod Jaera tyleri sp. nov. occurred in high abundance on whale bones and was absent from non-whale bone samples collected in the Kemp Caldera. The abundances estimated from the specimens sampled from the collected whale bones ranged from 470 -6,000 specimens per m 2 , which is within the ranges calculated by Amon et al. [8] for peracarids (lysianassid amphipods and Jaera tyleri sp. nov.) on the whale skeleton based on image analysis. Amon et al. [8] reported the peracarid abundances to be higher than the ones of lepetodrillid, osteopeltid and pyropeltid gastropods and siboglinid Osedax polychaetes. The abundances of the munnopsid Ilyarachna profunda from the southern Californian whale falls are lower and these had also been found in even fewer numbers on the deep-sea sediment and near seeps [6].
The feeding ecology of Jaera tyleri sp. nov. is still unknown. The shallow-water congeneric species are herbivores [26], detritivores  [23,25,27], even cannibals to some extent on weak or dead conspecifics [17,18] and probably bacterivores [24,26]. Amon et al. [8] report the abundance of peracarids on the whale bones to be positively correlated with the presence and percentage of bacterial mat cover on the bone and frequently seen on trunks and pulps of Osedax sp. nov. This indicated the potential of grazing on the bacteria but does not exclude the digestion of bone itself. Stable isotope or stomach content analyses might be able to clarify this. Within the Janiridae, species of genus Caecijaera are small commensals of the wood-borer Limnoria; they live in Limnoria's burrows and may feed on the fungi and bacteria living in the burrow walls [57]. Species of another closely related genus, Iais are ectocommensal of large sphaeromatids which may not be a specialised wood-borer, but at least they live in burrows of submerged decaying wood [58]. Jaera hopeana shares this habit [26]. Wolff [20] reported the presence of janirid isopods in sunken hollow sea grass rhizomes from the bathyal Caribbean trenches.
The observed female-biased sex ratio in J. tyleri sp. nov. (1:1.17) is in agreement with analyses on northern hemisphere Jaera species [19,31], although Piertney and Carvalho [59] reported up to twelve times more J. albifrons females found at a site in South Wales, United Kingdom. In her 20 months study on J. nordica and species of the J. albifrons group, Sjöberg [28] discovered a monthly fluctuation in sex ratios from 0.8:1 (males: females) in spring (March) to 2:1 in late autumn (October) after large specimens and ex-ovigerous females had died. Female-biased sex ratios have also been reported from other invertebrate taxa observed at whale falls e.g. polynoid polychaetes [60].
The analyses of the size-frequency distributions of J. tyleri sp. nov. suggest multimodal population structure (Fig. 8) with continuous breeding activity throughout the year. The studied northern hemisphere shallow water species of Jaera had shown synchronised reproduction with one annual cohort [31,32]. For Antarctic shelf isopods, Wä gele [61,62] proposed extended reproductive cycles of up to 32 months and seasonally synchronised reproduction.
The fecundity observed in Jaera tyleri sp. nov. increases with female size, in line with observations on northern hemisphere Jaera species [19,31,32]. In the northern species, ovigerous females were found at smaller sizes, carrying more eggs in comparison to the newly discovered species (5-11 eggs observed). In J. hopeana 7 -8 eggs and embryos in females of 1.8 mm length were observed [19], while in J. albifrons higher fecundity was observed, with 8 -84 eggs per females [19,28]. Similar fecundity has been reported for J. ischiosetosa, with 5 -30 eggs per female (2.1-4.8 mm length) [32] and Jaera nordica, with 5-28 eggs per female (1.5-4 mm length) [31]. In Isopoda in general the clutch size of females depends on the size of the female. For example, in Jaera albifrons a 3 mm large female produces 25 eggs, and a 5 mm large female 64 eggs. An Idotea species from the North Sea of 7.5 mm length produces 25 eggs, an 18 mm long female of the same species 324 eggs. In parasitic isopods the numbers of eggs are very high, for example Cymothoa oestrum produces around 2500 eggs. However, not all eggs of this ectoporasite develop well and produce offspring, usually about half of the eggs are lost during the brooding process. The breeding time strongly correlates with temperature [63]. A decrease in clutch size with decreasing temperatures, e.g. in cold waters of Antarctica and the deep sea, has been reported [61], but increased fecundity has been observed near hydrothermal vents for invertebrates [64]. In comparison with other Antarctic and deep-sea isopod species, the reported fecundity of J. tyleri sp. nov. is similar. In the Antarctic serolid Ceratoserolis trilobitoides (Eights 1833) broods of 50 -170 eggs were observed in females of 42 -80 mm length, but no correlation between clutch and female size was found [61,65].
With a single sampling event during the Antarctic summer, nothing can be said about seasonality in the breeding cycle. The presence of juvenile manca stages together with adults and the presence of several cohorts in the size-frequency distributions of immature and mature males and females suggest that the species lives for several breeding cycles. Some northern hemisphere species of Jaera showed strong seasonality of their breeding cycles with juveniles being released at certain months while these times vary between species [22,31,32]. Others, e.g. species of the Jaera albifrons group, breed throughout the year [66]. Sjöberg [28] postulated that J. albifrons usually die within a year and undertook one breeding cycle only. Harrison [67] found no evidence for seasonal activity in deep-sea asellote isopods from the north-east Atlantic Rockall Trough, but a strong signal in breeding intensity. More females were breeding through the winter months and more juveniles released during the summer months when the deposition of organic detritus began [67]. Figure S1 The tree from the RaXML secondary structure model 16A analysis of partial 18S rDNA (680 bp). The tree is rooted on the Stenetriidae. The numbers represent bootstrap support with 100 replicates.