Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
COMMENSALISM AND REPRODUCTIVE BIOLOGY OF THE BRITTLE STAR OPHIOCREAS OEDIPUS ASSOCIATED WITH THE OCTOCORAL METALLOGORGIA MELANOTRICHOS ON THE NEW ENGLAND AND CORNER RISE SEAMOUNTS By Celeste V. Mosher B.S. University of the Virgin Islands, 2002 A THESIS Submitted in Partial Fulfillment of the Requirements for the Degree of Master of Science (in Oceanography) The Graduate School University of Maine August, 2008 Advisory Committee: Les Watling, Professor of Oceanography, Advisor Pete Jumars, Professor of Marine Sciences and Oceanography Kevin J. Eckelbarger, Professor of Marine Sciences ©2008 Celeste V.Mosher All Rights Reserved COMMENSALISM AND REPRODUCTIVE BIOLOGY OF THE BRITTLE STAR OPHIOCREAS OEDIPUS ASSOCIATED WITH THE OCTOCORAL METALLOGORGIA MELANOTRICHOS ON THE NEW ENGLAND AND CORNER RISE SEAMOUNTS By Celeste V. Mosher Thesis Advisor: Dr. Les Watling An Abstract of the Thesis Presented in Partial Fulfillment of the Requirements for the Degree of Master of Science (in Oceanography) August, 2008 While temperate-water coral ecosystems of seamounts have been gaining more attention in the last decade, the organisms that live within and rely upon these corals for survival are studied to a lesser degree. Throughout the New England and Corner Rise seamounts of the western North Atlantic, several ophiuroid species are conspicuously epizoic on octocorals. One objective of this study was to investigate the association between Ophiocreas oedipus and its host octocoral Metallogorgia melanotrichos on these seamounts. Coral colonies with their brittlestar epibionts were collected from 11 seamounts in 2003, 2004, and 2005 at depths between 1300 and 2200 m via submersible. O. oedipus is obligately associated with M. melanotrichos, leading a solitary existence on all octocorals observed. O. oedipus gains feeding and protective benefits while M. melanotrichos appears to neither benefit nor be disadvantaged by this commensalism. M. melanotrichos exhibits a distinct developmental pattern that can be categorized into three growth stages. The positive correlation between size of O. oedipus and growth stage of its host is highly significant suggesting the brittlestar may grow up with M. melanotrichos. A further objective of this study was to investigate the reproductive biology of O. oedipus. Paraffin histology reveals that O. oedipus is gonochoristic and likely a broadcast spawner with a lecithotrophic larval stage. The gonadal tubules of O. oedipus occur in pairs of up to four within the proximal arm segments of each arm. The tubules exhibit synchronous gametogenesis within each pair but asynchronous development between pairs within the same individual. Findings suggest O. oedipus may use continuous reproduction throughout the year to maintain its population while limited by its dependence on M. melanotrichos as a host. ACKNOWLEDGEMENTS I would like to thank Les Watling for the opportunity to immerse myself in this fascinating research, for his advisory contributions, and for his contagious excitement with all things related to this subject. I cannot thank enough the members of my advisory committee, Pete Jumars and Kevin Eckelbarger, for their helpful comments and intriguing discussion. I am grateful to Scott France of the University of Louisiana, Lafayette, for confirming through genetic analysis the hypothesis that the three growth stages of M. melanotrichos are indeed the same species, and to Walter Cho of Woods Hole Oceanographic Institution who provided further molecular insight. I am indebted to Tim O'Hara of the Museum Victoria, Melbourne, and Isaac Smirnov of the Zoological Institute of the Russian Academy of Sciences, St. Petersburg, for their species identifications of the octocoral-associated brittle star Ophiocreas oedipus. Much appreciation goes to the graduate students, faculty, and staff of the Darling Marine Center, without whom this research would have been unsuccessful, and especially to my fellow lab members, Anne Simpson and Mateja Nenadovic, who were of great help and continuously supportive. I would like to thank Sheryl Mosher, Nye Mosher, Michelle Mosher, and Torren Nehrboss for supporting me in my endeavors, past and present. Finally, I am forever grateful to Jason Nehrboss for his technical assistance, unending patience, and loving inspiration. This research was funded by a NOAA Ocean Exploration grant to Les Watling and research and teaching fellowships from the University of Maine's School of Marine Sciences. iii TABLE OF CONTENTS ACKNOWLEDGEMENTS iii LIST OF TABLES vi LIST OF FIGURES vii Chapter 1. INTRODUCTION 1 2. MATERIALS AND METHODS 5 Study and Collection Site 5 Sample Treatment 7 Commensalism 8 Reproductive Biology 9 3. RESULTS 11 Growth stages of Metallogorgia melanotrichos 11 Association between Metallogorgia melanotrichos and 4. Ophiocreas oedipus 13 Predation on Ophiocreas oedipus 15 Ophiocreas oedipus reproductive biology 15 DISCUSSION 21 Commensalism of Ophiocreas oedipus on Metallogorgia melanotrichos 21 Reproduction in Ophiocreas oedipus 25 iv REFERENCES 29 BIOGRAPHY OF THE AUTHOR 33 V LIST OF TABLES Table 1. Sampling and morphological data for Metallogorgia melanotrichos hosts and Ophiocreas oedipus commensals Table 2. 6 Ophiocreas oedipus estimated fecundity and egg sizes per gonadal tubule 19 VI LIST OF FIGURES Figure 1. Bathymetric map of the New England and Corner Rise seamounts, NW Atlantic 5 Figure 2. Growth stages of Metallogorgia melanotrichos Figure 3. Ophiocreas oedipus disc diameters of individuals residing in the three 12 growth stages of M. melanotrichos hosts 14 Figure 4. Ophiocreas oedipus gonad morphology 17 Figure 5. Change in gonadal tubule morphology of Ophiocreas oedipus as disc diameter increases Figure 6. 18 Transverse histological sections of gonadal tubules of Ophiocreas oedipus 20 vn Chapter 1 INTRODUCTION In the face of habitat destruction by ever more intensive and deeper fishing methods as well as chemical and thermal changes caused by rising atmospheric C0 2 levels, deep-sea coral communities are in increasing danger of being damaged or destroyed (Auster et al. 1996; Auster et al. 1998; Koslow et al. 2001; Fossa et al. 2002; Hall-Spencer et al. 2007; Clark and Koslow 2007; Probert et al 2007; Waller et al. 2007). Deep-sea organisms have been proposed as important bio-indicators of climate change (Guinotte et al. 2003; Danovaro et al. 2004; Barnett et al. 2005) and as sources of natural products for pharmaceuticals (Faulkner 2001). Seamount communities in particular have been characterized as sites of higher endemism (de Forges et al. 2000; Stone et al. 2004), possible hot spots of biodiversity (Worm et al 2003; Samadi et al 2006), and areas of higher productivity (Genin 2004; Samadi et al 2006). Such characterizations imply that these communities may be at an elevated risk of ecological damage and perhaps, extinction events. While cold-water coral communities of deep seamounts have been gaining more attention in the last decade, the organisms that live within and rely upon these corals for survival are studied to a lesser degree. Many of the "associates" hosted by the octocorals are thought to live symbiotically upon their hosts either as commensals or as parasites (Buhl-Mortensen and Mortensen 2004). For example, of the known symbiotic associations on deep-sea cnidarian taxa, Buhl-Mortensen and Mortensen estimate that 35% are obligately associated with their hosts. Obligate coral symbionts are even more 1 at risk than other associates as they depend on their hosts for survival (Buhl-Mortensen and Mortensen 2004), but the degree to which associates rely on their hosts is relatively unstudied for most seamount symbioses. Although conservation efforts are underway, more progress is needed towards the protection of seamount ecosystems (reviewed in Probert et al 2007). It is especially crucial to assess the interspecific relationships of these octocoral associations. Echinoderms dominate the non-sessile megafauna in the deep-sea (Gage and Tyler 1991). Throughout the New England and Corner Rise seamounts of the North Atlantic, several ophiuroid species are conspicuously epizoic on the octocoral assemblage. The prominence of organisms perched in arborescent corals has been related to food availability (Gage and Tyler 1991). Particulate matter is resuspended from the bottom within the benthic boundary layer (Jumars and Nowell 1984) and corals tend to align perpendicular to the prevailing current, providing food for the suspension-feeding corals themselves and also their symbionts. Climbing a small distance above the bottom provides access to higher fluxes because any decrease in suspended particle concentration is generally more than offset by the steep velocity gradient in the lowermost bottom boundary layer. There are likely further benefits to the organisms that reside in coral hosts, such as protection from predators and/or increased gamete dispersal for freespawning species. Asteroschematid brittle stars (Ophiuroidea: Euryalida) are one of the least studied families of ophiuroids, likely because most known species occur in the deep sea. Many are found associated with colonial cnidarians. There have been some studies of the feeding characteristics and basic biology of asteroschematid species (Stewart and 2 Miadenov 1997, Grange 1991, Fujita and Ohta 1988, Emson and Woodley 1987, Hendler and Miller 1984) yet the ecological dynamics of brittle star-host associations within the family are poorly known. Brittle star host fidelity ranges from specific to a single host species in a particular geographic location (Grange 1991, Emson and Woodley 1987) to occurrence on many different coral species (pers. obs.). The only asteroschematid association studied in detail is that of Astrobrachion constrictum, a euryalid living in association with a gorgonian in the fjords of New Zealand (Grange 1991, Stewart and Miadenov 1997, Stewart 1998). A. constrictum is an asteroschematid species found at SCUBA-accessible depths. The New Zealand fjords, however, are subject to diel cycles of light penetration and nutrient-rich run-off, neither of which is found in the more stable environment of the deep-sea. A. constrictum lives upon an antipatharian octocoral, Antipathes fiordensis, in what has been described as an obligate symbiosis (Grange 1991). They appear to remain on their hosts year after year and, when transplanted from their perches in A. fiordensis. to the seabed, individuals will rapidly find the nearest A. fiordensis and climb up the colony (Stewart and Miadenov 1997). Other arborescent features and protective niches are apparently not deemed suitable. In addition, there appear to be elements of mutualism (Grange 1991) as the host coral is cleaned by the brittle star that feeds on the detritus settled onto its branches. The asteroschematid ophiuroid Ophiocreas oedipus Lyman, 1879, has been found on the octocoral Metallogorgia melanotrichos Wright and Studer, 1889, on the New England and Corner Rise seamounts. The major objective of this study is to begin a basic investigation of this association and the biology of the two species. It is hypothesized 3 that the brittle star settles onto the young M. melanotrichos and the two species then grow up and live together their entire lives. As part of this study, the reproductive biology of O. oedipus was also examined. While studies of reproductive periodicity and general anatomical structure have been conducted on the shallow water A. constrictum (Stewart and Mladenov 1994, 1995), similar research has not been conducted on deep-sea asteroschematids. We know characteristics of the reproductive biology of some deep-sea ophiuroids collected from muddy bottom, bathyal and abyssal regions (Tyler and Gage 1980, Sumida et al 2000, Gage et al 2004), however these species represent much more distant taxonomic groups. 4 Chapter 2 MATERIALS AND METHODS Study and Collection Site The New England Seamounts form a chain from New England, USA, extending from Bear Seamount (located at approximately 39°50 N, 67°20 W) southeasterly to Nashville Seamount (34°30 N, 56°50 W). The Corner Rise complex clusters around 34° N,49° W (Figure 1). The summits of these undersea mountains range from ~1.3 to 3.5 km water depth. Temperatures at these depths fall between 3°C and 4°C. New England and Corner Rise Seamounts Figure 1 Bathymetric map of the New England and Corner Rise seamounts, NWAtlantic. As part of the NOAA Ocean Exploration program, research cruises to the Corner Rise and New England seamounts were conducted in July 2003, May 2004, August 2005, and October 2005. The submersible Alvin and support ship Atlantis was used in 2003 and during the October cruise of 2005. The ROV Hercules was deployed from the NOAA 5 ship Ronald H. Brown in 2004 and August 2005. Sixty-six coral colonies were collected at depths between 1300 and 2200 m from 13 seamounts in the New England and Corner Rise chains (Table 1). In addition, over 500 h of video and -30,000 high-definition frame-grabs taken during the submersible and ROV dives were used to supplement the other data for this study. Table 1 Sampling and morphological data for Metallogorgia melanotrichos hosts and Ophiocreas oedipus commensals. (NES = New England seamounts, CRS = Corner Rise seamounts, DD = disc diameter) Specimen Collection ID BAL101 BAL107-1 BAL107-2 BAL107-3 BAL110-2 BAL208-2 BAL210-1 BEA401-2 BEA506-1 MIL103-2 MIL105-1 GOO103-1 GOO106-1 GOO107-1 GOO109-1 KUK209-1 KEL103-1 KEL107-1 KEL201-2 KEL203-3 KEL301-2 KEL301-3 KEL301-4 KEL607-1 KEL610-1 MAN 706-2 MAN 710-1 date 5/21/04 5/22/04 5/22/04 5/22/04 5/22/04 9/2/05 9/2/05 5/11/04 5/12/04 8/17/05 8/17/05 8/20/05 8/21/05 8/21/05 8/21/05 8/22/05 7/15/03 7/15/03 7/16/03 7/16/03 5/18/04 5/18/04 5/18/04 8/31/05 9/1/05 5/15/04 5/15/04 Seamount Lattitude N Balanus, NES 39°21.30 Balanus, NES 39°21.51 Balanus, NES 39°21.51 Balanus, NES 39°21.51 39°22.30 Balanus, NES 39°24.88 Balanus, NES Balanus, NES 39°24.88 Bear, NES 39°57.10 Bear, NES 39°53.02 Caloosahatchee, CRS 34°48.92 Caloosahatchee, CRS 34°48.91 Corner, CRS 35°23.61 Corner, CRS 35°23.59 Corner, CRS 35°23.58 Corner, CRS 35°23.57 Corner, CRS 35°33.40 Kelvin, NES 38°47.33 Kelvin, NES 38°47.31 Kelvin, NES 38°51.60 Kelvin, NES 38°51.51 Kelvin, NES 38°49.20 Kelvin, NES 38°49.20 Kelvin, NES 38°49.20 Kelvin, NES 38°45.88 Kelvin, NES 38°45.92 Manning, NES 38°08.93 Manning, NES 38°08.78 6 Longitude> Depth i M. melanotrichos 0. oedipus W (m) growth stage DD (mm) 65°21.52 1912 3 65°21.82 1767 3 65°21.82 1767 3 65°21.81 1767 3 65°22.31 1562 1 65°24.66 1802 2 65°24.56 1717 3 67°24.60 1559 1 67°28.32 1491 2 50°30.35 1601 3 50°30.36 1592 3 51°15.93 2135 2 51°15.99 2082 3 51°16.00 2076 2 51°16.05 2051 3 51°48.89 1829 3 64°07.91 2029 3 64°07.83 1949 1 63°54.85 2173 3 63°54.85 2047 3 63°57.54 1773 1 63°57.54 1773 3 63°57.54 1773 3 64°05.41 2171 3 64°05.43 2144 3 61°06.13 1847 2 61°06.92 1668 3 12.26 12.64 11.57 11.90 7.25 10.13 12.15 6.29 10.21 10.19 11.81 8.87 11.85 7.95 12.17 12.66 12.05 6.61 10.19 13.8 6.50 12.41 12.36 13.74 9.97 8.02 10.69 MAN703-1 MAN706-1 MAN708-1 MAN708-2 MAN801-1 MAN803-2 NAS104-1 NAS105-1 NAS110-3 NAS203-1 NAS203-2 5/15/04 5/15/04 5/15/04 5/15/04 5/16/04 5/16/04 8/25/05 8/25/05 8/25/05 8/26/05 8/26/05 10/28/05 PIC101-3 REH 203-1 8/29/05 REH105-1 8/29/05 REH 112-3 8/29/05 REH212-1 8/30/05 REH215-1 8/30/05 LYM210-1 8/14/05 Manning, NES Manning, NES Manning, NES Manning, NES Manning, NES Manning, NES Nashville, NES Nashville, NES Nashville, NES Nashville, NES Nashville, NES Picket, NES Rehoboth, NES Rehoboth, NES Rehoboth, NES Rehoboth, NES Rehoboth, NES Yukatat, CRS 38°08.93 38°08.93 38°08.09 38°08.09 38°08.84 38°08.83 34°34.92 34°34.88 34°34.77 34°28.81 34°28.81 39°39.12 37°33.67 37°27.63 37°27.60 37°33.48 37°33.37 35°11.63 61°06.14 61°06.13 61°06.97 61°06.97 61°05.87 61°05.86 56°50.59 56°50.59 56°50.51 56°44.15 56°44.15 65°56.50 59M8.43 59°57.08 59°56.99 59M8.18 59M8.13 47°40.60 1867 1847 1718 1718 1692 1674 2233 2221 2136 2103 2103 2084 1678 1907 1820 1438 1372 2143 3 3 3 3 3 3 3 3 3 3 3 3 2 3 3 3 3 3 13.41 12.24 12.06 14.86 12.90 12.63 12.85 12.31 11.46 11.37 12.24 13.61 9.16 11.34 11.95 11.71 11.08 12.03 Sample Treatment After removal from the submersible or ROV bioboxes, the M. melanotrichos and associated brittle stars were briefly held in cold seawater until processed on-board the research vessel. Processing each octocoral began by photographing each one with its associates and making preliminary identifications if possible. Small portions of the corals were removed for future use in molecular studies as was one arm of each brittle star. The specimens were then fixed in a 5% formalin/seawater bath for approximately 812 h. Most brittle stars were extricated from the coral colonies before fixation, although some especially small individuals were left on the host throughout the preservation process. Following the formalin bath, all specimens were transferred to 70% ethanol for long-term storage. 7 Commensalism Of the 66 M. melanotrichos colonies collected, 45 of the associated ophiuroids were brought to the Darling Marine Center for this study (Table 1). The rest were placed in collections at the Smithsonian National Museum of Natural History, Yale-Peabody Museum, and Woods Hole Oceanographic Institution. At the Darling Marine Center, the ophiuroid specimens were digitally photographed using a dissecting microscope taking care to capture defining characters for species identification. Disc diameter was recorded as the distance across the aboral surface from the point between the radial shields above one arm to the opposing interradius. Number of arms with macroscopically visible signs of regeneration per individual was recorded. The brittle star arms were tightly coiled around the branches of M. melanotrichos and data from arms that were broken and/or missing after extraction from the host, as well as those that had been taken for molecular studies, could not be included in this study. M. melanotrichos colonies of a wide range of sizes were collected. They were divided into three growth stages. Stage 1 is the youngest juvenile stage, Stage 2 is intermediate, and Stage 3 is the full-grown adult stage, as more fully characterized in the Results. Observational data recorded from the video and frame grabs of submersible dives included the number of brittle stars on each of the 66 M. melanotrichos colonies, presence/absence of non-ophiuroid fauna associated with the octocorals, presence/absence of O. oedipus brittle stars on other corals and/or on the seabed, location of the brittle stars on the octocoral, and behavior of the brittle stars. Also, the number and species of brittle stars associated with uncollected M. melanotrichos were recorded. 8 The observations were made before and during collection while great care was taken to discern abnormal behavior caused by the submersible or ROV. The video was also used to ascertain that the brittle stars were on their respective hosts when collected and identify those few that traveled from their hosts after being stored in the bioboxes. Reproductive Biology In many asteroschematids the gonads develop as tubules in the proximal part of the arms. In O. oedipus, this portion of the arms becomes swollen as the brittle star matures to accommodate the expanding gonad tubules. The number of modified arm segments was recorded from one arm of each brittle star. This sampling was deemed sufficient because the number of modified arm segments was seen to be approximately equal among the arms of each individual. Gonadal portions from multiple arms of each individual were dissected microscopically to examine gross morphology. The swollen proximal portion of one arm was removed from each individual for examination by paraffin histology. These gonadal portions were hydrated and allowed to decalcify in Hollande's fixative for approximately two weeks. The portions were then dehydrated in ethanol, cleared with toluene, and embedded in paraffin. Serial transverse sections (1214 jxm thick) were hydrated and stained with Gomori's trichrome (aniline blue) before finally dehydrating and mounting for microscopy. Histological sections of O. oedipus gonads were examined under a compound microscope to determine the sex of each specimen. The number of eggs within one of each pair of gonadal tubules of the sectioned arm was counted by counting the distinct nucleoli. This number was multiplied by 10 to estimate the standing fecundity of an individual. Fecundity of four individuals 9 with different disc sizes was estimated. The minimum and maximum egg sizes were taken from the tubules examined for fecundity. Many eggs were nonspherical due to packing within the gonads. Egg size was estimated as an average of two diameters measured perpendicular to one another. Digital images were captured of the male, female, and immature gonads and examined further for details of reproductive anatomy and evidence of reproductive mode and periodicity. in Chapter 3 RESULTS Growth Stages of Metallogorgia melanotrichos Juvenile M. melanotrichos do not look like adults. The juvenile form (Stage 1) consists of a single stalk with axial polyps arising from the central axis and a single terminal polyp at the apex of the stalk. This juvenile develops branches along the central axis that give rise to more polyps (Figure 2 a,b). As the colony matures, the polyps disappear from the central axis, and a cluster of terminal branches replaces the terminal polyp (Stage 2). This cluster forms a burgeoning "parasol" of polyps while the axial branches are retained along the stalk (Figure 2 c). Finally, the axial branches disappear and the adult form (Stage 3) is fully manifested as a solitary stalk with a full crown of polyp-covered terminal branches (Figure 2 d,e). As the colony ages further, branches in the crown continue to subdivide until the crown is densely branched. The M. melanotrichos hosts of the 45 brittle stars used in this study were categorized into growth stages as follows: four Stage 1 juveniles, six Stage 2 intermediates, and 35 Stage 3 adults (Table 1). 11 Figure 2 Growth stages of Metallogorgia melanotrichos. Note the Ophiocreas oedipus (O) associated with each colony. (White scale bars = 5 cm) a Stage 1 M. melanotrichos 12 characterized by a single terminal polyp (tp) and axial polyps (ap) arising from a central axial stalk. Axial branches (ab) are present in the more mature Stage 1 colonies, b Stage 1 colony illustrating the terminal polyp that defines this stage, c Stage 2 M. melanotrichos. Terminal branches (tb) have replaced the terminal polyp. The axial branches still arise from the stalk but the axial polyps are no longer present, d In situ image of Stage 3 M. melanotrichos with a full crown of terminal branches. A single O. oedipus is curled around the branches in the center of the crown, e Stage 3 M. melanotrichos. Axial polyps and branches are no longer present in the adult colony. Association Between Metallogorgia melanotrichos and Ophiocreas oedipus All collected M. melanotrichos hosted one individual O. oedipus. In addition, there were 94 M. melanotrichos definitively seen to host a single brittlestar resembling O. oedipus in the frame grabs. The dive tracks did not coincide on repetitive visits to seamounts, therefore multiple observations of the same individual did not occur. O. oedipus was found on no other coral in our explorations nor was it observed on the seabed via submersible/ROV video camera and frame grabs. No other ophiuroid species was found on living M. melanotrichos across the New England and Corner Rise seamounts. Other brittle star species were observed around the base of M. melanotrichos colonies and sometimes hanging from the detritus-covered base of the stalk. In addition, no other macroscopic species were observed on this coral, with the exception of one octocoral on Balanus Seamount. This M. melanotrichos was swathed with numerous non-stalked crinoids and other brittle star species were hanging off the stalk that appeared 13 to be covered with dead tissue. One O. oedipus was present even though the octocoral appeared to be in poor health. There is a strong correlation between the disc diameters of O. oedipus and the growth stage of the respective octocoral hosts (Figure 3, Kruskal-Wallis H = 22.24, p < 0.001, n = 45). Disc diameters ranged from 6.29 to 14.86 mm (Table 1) with a mean of 11.23 mm. O. oedipus does not exhibit external sexual dimorphism (Mann-Whitney U = 270, p = 0.95, n = 43). Females had a mean disc diameter of 12.86 mm and males, 11.04 mm. The disc diameter of brittle stars perched on Stage 1 M. melanotrichos is 6.66 mm. The mean diameter of those residing on Stage 2 hosts is 9.06 mm and the mean size of those on Stage 3 octocorals is 12.13 mm. 1 2 3 Growth stage of host octocoral Figure 3 Ophiocreas oedipus disc diameters of individuals residing in the three growth stages of M. melanotrichos hosts. (Circles = outliers, t-bars = range, solid bars = medians, boxes = 25th to 75lh percentiles) 14 O. oedipus was found with its arms coiled around the branches in the center of an M. melanotrichos host (Figure 2 d, 4 d) and often the outer lengths of up to all five arms were extended into the surrounding water. O. oedipus did not appear to take care to avoid contact with the stinging polyps of M. melanotrichos as it waved its arms about. Brittle stars stopped any activity or feeding postures when disturbed physically by the submersible/ROV arm. O. oedipus resided always in the full crown of branches of an adult host or coiled around the axial stalk of hosts without crowns. (Figure 2 a-e). Of the six brittle stars on Stage 2 hosts, four were found in the young parasol of branches at the apex of the stalk and the other two were coiled around the stalk below the parasol. Predation on Ophiocreas oedipus Although there were 45 O. oedipus available for the analysis of regeneration, only 82.7% of their arms were available and intact enough to make a confident assessment of regeneration. Of them, 15.1% showed macroscopic signs of regeneration (Figure 4 d). All but one of the regenerating individuals were on Stage 3 host colonies. Only one specimen had 3 arms regenerating and the rest had 2, 1, or none. Ophiocreas oedipus Reproductive Biology O. oedipus is gonochoristic. Of the 45 specimens examined histologically, 22 were female, 21 were male, and two immature individuals could not be sexed with paraffin histology. Thus, the sex ratio appears to approximate unity. Male and female gonads consist of paired tubules that extend into the arms of O. oedipus from the bursae of the central disc. The gonadal tubules lie aboral to the 15 vertebral ossicles (Figure 4 a,b) and occur in pairs of up to four per arm (Figure 4 a-c). Two of the 45 brittle stars contained unpaired gonadal tubules, both having one tubule on one side of each arm. The other side of the arms housed three and four tubules within the two individuals. The growth of the gonadal tubules causes a modification of the surrounding calcified arm structures so that the area that houses the gonads is visible externally (Figure 4 d). The number of arm segments modified by the growth of the gonads within one arm of each specimen ranged from zero to 14. Of the five individuals with no modified arm segments, three were sexually immature with no gonads at all and two had one pair of gonadal tubules just beginning to grow. There is a sexual dimorphism in the size of gonads as estimated by the number of arm segments modified. Male gonadal tubules modify more arm segments than female tubules (Mann-Whitney U = 290.5,0.01 < p < 0.02, n = 40). The number of arm segments modified in females averaged 5.14 and in males, 7.74. These means do not include those immature but sexed individuals with no modified segments. The number of gonadal tubule pairs per arm weakly depends on the disc diameter of the brittle star while the number of segments modified is more strongly dependent on disc diameter (Figure 5). The gonadal tubules may continue to grow farther out the arms as the brittle star ages, but the number of tubule pairs does not exceed four. 16 Figure 4 Ophiocreas oedipus gonad morphology. (Scale bars = 1 mm) a Image of one pair of aboral gonadal tubules extending into the arm of a female O. Oedipus. The oocytes are visible within the gonad coelom. b Three pairs of male gonadal tubules. Male tubules are longer and more slender than female tubules, c Female gonadal tubules removed from one arm to show all four pairs. The maximum number of pairs found in any individual arm, male or female, is four, d The heavily calcified arm structure is modified to accommodate the gonadal tubules. The enlarged proximal arm portion is visible externally (white ^ ) . Note evidence of regeneration on missing arm (black ^ ) . Also, note the posture of the brittle star curled around the branches of the octocoral seemingly heedless of the stinging polyps. (Brittle star disc diameter =13.8 mm) 17 10.0 12.0 Disc diameter ( m m ) 10.00 12.00 Disc diameter ( m m ) Figure 5 Change in gonadal tubule morphology of Ophlocreas oedipus as disc diameter increases, a Number of arm segments modified by tubule growth generally increases with disc diameter, b Number of gonadal tubule pairs within one arm is less dependent on disc diameter. Of the four female specimens examined, the smallest individual with three tubule pairs per arm had an overall standing fecundity of 1750 while the largest individual had an estimated 15,230 eggs (Table 2). Another large individual had one tubule pair per arm 18 and an estimated 9,840 overall fecundity. The number of eggs within a tubule pair differs between pairs in the same individual. Egg size also differs between pairs (Table 2). The diameters of eggs range from 39.0 to 604.5 ^m. Tubules with the most eggs have the smallest eggs because females have many small immature eggs within a single tubule pair (hereafter referred to as the primary pair). If there are more pairs within each arm, larger, more mature eggs are seen (Figure 6 b) over the entire length of these tubules. Gametogenesis appears synchronous between both tubules of pairs, with the exception of the primary pair. Within the primary tubule pair, eggs closer to the disc are smaller and less mature than eggs at the distal end. Each pair of tubules develops asynchronously to other pairs in the same arm. This is the case for both males and females (Figure 6 a,b) and was exhibited by O. oedipus collected at all four time periods: mid May, mid July, end of August/beginning of September, and the one female individual collected at the end of October. There is no evidence suggesting reproductive periodicity in the present study. Table 2 Ophiocreas oedipus estimated fecundity and egg sizes per gonadal tubule. Disc diameter Total estimated Gonad tubules No. eggs Min. egg size (mm) fecundity (pairs/arm) per tubule (|jm) 50.7 9.16 1750 3 161 10 230.1 4 421.2 11.34 4040 4 283 27.3 315.9 45 40 397.8 36 124.8 984 39.0 12.66 9 840 1 13.41 15 230 4 1285 39.0 401.7 87 83 292.5 218.4 68 19 Max. egg size (pm) 198.9 300.3 444.6 136.5 604.5 479.7 331.5 109.2 222.3 436.8 409.5 292.5 Figure 6 Transverse histological sections of gonadal tubules of Ophiocreas oedipus. The vertebral ossicles (O) lie along the oral surface of the arms. (Scale bars = 0.5 mm) a Male gonadal tubules at varying stages of gametogenesis within a single individual. One pair of tubules (Tl) has spawned more recently than the others and is spent. The germinal epithelium has many invaginations ( ^ ) , and there is a lack of spermatozoa. Another tubule pair (T2) contains spermatozoa. The germinal epithelium is thickened but still invaginated. The tubule T3 has lost the invaginations of the epithelium and is packed with spermatozoa, b Female gonadal tubules at varying stages of gametogenesis within a single individual. Tl is the tubule at the youngest stage of development in this female. These small, early-vitellogenic oocytes stain darker than the more mature gametes. A second tubule pair (T2) has larger, mid-vitellogenic oocytes. A third tubule pair (T3) are at a stage close to the remaining tubule pair (T4). In these gonadal tubules, the cytoplasm of the late-vitellogenic oocytes stains lighter with a grainy appearance. 20 Chapter 4 DISCUSSION Commensalism of Ophiocreas oedipus on Metallogorgia melanotrichos This study of the relationship between O. oedipus and M. melanotrichos offers a unique opportunity to peer into the lives of some of the most elusive deep-sea brittle stars and cold-water corals. O. oedipus appears to be different from other deep-seamount coral associates, many of which are generalists using any coral as a means to gain height in the benthic-boundary layer (Buhl-Mortensen and Mortensen 2004). The commensalism of O. oedipus on M. melanotrichos is species-specific and appears obligate on the part of O. oedipus. A mutualistic relationship is not suggested since no benefit to the coral has been seen and, is difficult to infer. Until molecular work was completed after the 2004 cruise, the Stage 1 and Stage 2 host colonies could not be positively identified as M. melanotrichos. Their difference in external morphology from the adult Stage 3 caused some initial ambiguity. However, the development of the polyps along the stalk and the branching patterns illustrate the progression from a juvenile colony to an adult. The strong correlation between brittle star size and growth stage of the host suggests that O. oedipus matures on - and with - its host M. melanotrichos. Instances of obligate associations between brittle stars and octocorals have been seen before (Emson and Woodley 1987, Grange 1991, Pearse et al. 1998, Metaxas and Davis 2005), but the phenomenon of growing up together has not been previously observed. 21 The high fidelity between O. oedipus and M. melanotrichos has many implications. O. oedipus appears to need M. melanotrichos for survival on the New England and Corner-Rise seamounts. If anthropogenic influences in the deep sea continue to put seamount coral communities at risk, the organisms that live on the corals are equally at risk if they are not able to thrive in other habitats such as on the seabed. The presence of species-specific associations may be an indicator of higher biodiversity on seamounts (McClain 2007), however, biodiversity may be only slightly elevated as obligate associations are rare (Buhl-Mortensen and Mortensen 2004). Moore et al. (2003) explored the biodiversity of a single seamount in the New England chain in which associations were not recorded. However, such studies should be cautiously interpreted since artifacts of sampling on seamounts are easily overlooked (Stocks and Hart 2007). There is a dire need for further sampling of seamounts to assist management strategies since it is well-accepted that the ecological communities on seamounts, although often distinct from seamount to seamount, are different from the surrounding deep-sea and so are usually not characterized by the larger biogeographic region (Probert et al. 2007). Such a tight association would seem limiting and is likely to have persisted for thousands of years in the stable environment at these depths. It is worth noting that commensal associations between euryalids and other invertebrates can be found in the fossil record as early as the late Carboniferous. Ancestors of the Asteroschematidae, the Onychasteridae, are known as fossils coiled around the anal cone of stalked crinoids, 300-325 million years ago (Clark 1908, Wachsmuth and Springer 1897). 22 The exact mechanism that allows O. oedipus to recognize its host is unknown. Chemical cues likely signal the brittle star larvae to settle either directly onto a young coral or onto the seabed. Recently settled O. oedipus would not have been visible on the seabed at the resolution of our video. However, the cost of settling on the seabed and then searching for a particular host species would be very high because the corals are rare. If the number of settling larvae is high enough to sustain the population under such conditions, a settle-then-search mechanism could be maintained. O. oedipus larvae apparently travel long distances between seamounts (Cho pers. com.) and presumably larval density is not high over one specific seamount. It is more likely that the larvae settle onto the juvenile coral from the water column after encountering a chemical cue in the surrounding water. No other ophiuroids are found on M. melanotrichos despite the presence of many other ophiuroid species on the surrounding benthos and hosted by nearby corals. Also, only one O. oedipus resides on a single octocoral. When the coral is young, it is possible that its defensive polyps and chemicals are not as toxic to the settling brittle star and the brittle star develops a resistance to the coral as it grows. Other possible associates, brittle stars or any other commensals, may not be able to withstand the defense mechanisms of M. melanotrichos, thus explaining the lack of other associations. This hypothesis could be tested through in situ manipulations. Pelagic larvae likely recognize the presence/absence of O. oedipus on a young coral. Alternatively, territorial aggression could maintain O. oedipus' solitary lifestyle but this is unsupported by any video observations. The lack of regenerating arms on young brittle stars may be evidence that physical aggression is not taking place; rather a chemical signal may indicate that a host 23 is already occupied. Another possibility is that the resident brittle star may remove potential competitors when they are in a younger, post-larval stage. Finally, the phenomenon of growing up together may simply be a result of host octocorals being rarer than newly settling O. oedipus. On the Corner Rise and New England seamounts, it may be that young M. melanotrichos without a current resident represent the only options available to settling O. oedipus. If the brittle stars live as long as the hosts and fatalities from predation are low, few adult M. melanotrichos would ever become available. The brittle stars on intermediate Stage 2 hosts were either in the burgeoning parasol of polyps or curled around the stalk. At some point in the development of the two species, O. oedipus ascends into the crown to live. This likely happens when the octocoral's crown becomes large enough to protect the brittle star and/or when the brittle star achieves a state of maturity that allows it to withstand the density of stinging polyps found in the complex of branches of the M. melanotrichos crown. The evolutionary advantage of maintaining a uniform, solitary population distribution over hosts is not obvious. Food availability may play a role, but nearby arborescent corals hosting many brittle stars would seem to dispel that conclusion. Since there appear to be many possible dwelling locations within the larger adult M. melanotrichos, one colony could, in theory, host several O. oedipus. The population uniformity of one brittle star per coral is likely to have persisted since both coral and brittle star were young. When the coral colony was no more than an axial stalk with a few polyps, there were fewer places for the young brittle stars to perch and take advantage of the available currents. That no other larval O. oedipus settles on the coral may be the result of an evolved survival mechanism. In addition, a single ophiuroid 24 perched and protected within the center of a M. melanotrichos crown of polyp-covered branches is less likely to be harmed by predators than multiple ophiuroids located nearer the outer limits of the coral's crown. If there is a predatory attack, damage is not likely to be fatal if arms are clipped off as opposed to the whole ophiuroid being removed from the coral. Of the 15% incidence of regenerating arms among those arms analyzed, in most cases only one arm was damaged. As the brittle stars wave their arms into the prevailing current for food, an arm could be bitten off and the brittle star would have a chance to curl into the interior of its host's protective crown. The in situ observations of O. oedipus suggest such a defense mechanism as the brittle stars were seen extending their arms out into the water and retracting them upon being disturbed. Reproduction in Ophiocreas oedipus Up to four gonadal tubule pairs were found per arm of O. oedipus. This differs from the single pair per arm that is more common among the asteroschematids (Hendler 1991; Byrne 1994; Stewart and Mladenov 1994). The two specimens that did not have matching pairs in each arm are likely developmental anomalies; unmatched tubules in pairs are uncharacteristic of the population as a whole. Pairs of tubules were at varying maturity levels from other pairs within each individual that had more than one tubule pair. That this was the case at all four collection times, and in both males and females, implies that spawning is likely to happen in one of two different ways: either O. oedipus spawns multiple times per year, or it is a dribble spawner. Gonad morphology suggests that O. oedipus combines these two mechanisms by spawning at multiple times per year only a portion of an individual's gametes, one 25 tubule pair in all five arms at a time. In other words, a spawning individual releases gametes from 10 gonadal tubules at one time while other tubule pairs are continuing to mature. Whether the population spawns synchronously or timing differs at the level of the individual is unknown, but considering their dioecious, solitary lifestyle, synchronous spawning across the population would be more likely to result in successful fertilization. That the number of gonadal tubule pairs per arm appears to not depend on disc diameter implies this number does not progressively increase to four as the individual grows and then four pairs are maintained through the remainder of an individual's lifespan. Tubule pairs may be resorbed after the contents are released and then a new pair begins to develop. With the exception of the primary pair in females, each gonadal tubule pair contains a cohort of gametes developing synchronously. The oocytes in the primary pair are all less mature than those found in other pairs, however, oocytes are distinctly smaller and less mature at the ends of the tubules proximal to the disc. Oocytes are likely formed initially near the central disc where the tubules attach and are pushed more distally as they mature. This phenomenon has been seen before in ophiuroids (Smith 1940; Patent 1976; Byrne 1994) and is evidence that the source of gamete-producing germ cells is the genital rachis (Hendler 1991). The distinct pattern of one tubule pair being fairly immature or recovering from a recent spawning event and other tubule pairs being more mature within each individual implies that these brittle stars do not spawn all their gonadal contents at the same time. Reproductive periodicity thought to be associated with the annual plankton blooms of the North Atlantic as seen in other deep-sea ophiuroids (Tyler and Gage 1980, 26 Tyler et al. 1982, Sumida et al. 2000, Gage et al. 2004) was not evident in this study. The maturity-stage index method (Patent 1969) commonly used to evaluate periodicity could not be used because each individual manifested up to four stages of gametogenesis. A. constrictum represents the closest relative of O. oedipus for which reproductive studies have been completed. This species exhibited an annual breeding pattern at depths between 18 and 21 m in a New Zealand fjord (Stewart and Mladenov 1995) and it follows the general reproductive anatomy of the Asteroschematidae described by Hendler (1991). But this species, although closely related, is located in an environment with light penetration and nutrient run-off, a very different environment than that found on the seamounts. Deep-sea species for which reproductive periodicity has been supported are more distantly related, bottom-dwelling brittle stars. Further studies are necessary, yet the reproductive timing of O. oedipus, with its long, persistent association in the deep sea, may be phylogenetically constrained as suggested by Eckelbarger and Watling (1995), and the gonadal morphology described previously supports continuous reproduction. Fecundity increases dramatically with size of the female and less so with number of gonadal tubules. This weaker dependence is partly because one gonadal tubule pair holds the majority of the oocytes regardless of the number of pairs within an individual (Table 2). The oocytes within this pair are the least mature and smallest of the oocyte cohorts within any individual housing more than 1 tubule pair per arm. Successive tubule pairs hold considerably fewer oocytes, although these oocytes are larger and more developed. An oosorption mechanism is likely responsible for this developmental pattern and is common in echinoderm species. It is believed that the oosorbed material may be used for the development of other healthy oocytes (Hendler 1991; Byrne 1994). Oocyte 27 abortion and oogenesis occur simultaneously in continuously reproducing species (Byrne 1994). In the case of O. oedipus, most of the oocytes within the least developed gonadal pair may be oosorbed to supply nutrients for the remainder of the oocytes in this developing cohort. If so, the total fecundity estimated for the individuals in this study is much greater than the number of oocytes that will reach maturity to be released. The number of oocytes spawned would depend upon the rate of gametogenesis as well as the rate of oosorption. The mature oocytes of O. oedipus attain sizes greater than 600 ptm, indicative of a lecithotrophic larval stage. Molecular studies are underway assessing the genetic connectivity within the population of O. oedipus across the New England and Corner Rise seamounts. Findings indicate high connectivity (Cho pers. com.) despite the thousands of miles of open sea between some of these octocoral communities. A longlived larval stage would be necessary to reach such distances. A pi an kto trophic larval stage would assist this species in maintaining its high connectivity. Fertilization success may be low because O. oedipus is restricted to its habitat in colonies of M. melanotrichos and host octocorals can be found in clusters in only some locales. The brittle stars may be hundreds of meters apart in other spots. It appears that M. melanotrichos hosts are a limiting resource for larval O. oedipus because we did not find any of these octocorals without its resident brittle star. Populations of O. oedipus are likely sustained by a longlived larval stage that can travel far to find an unoccupied juvenile host on which to settle. 28 REFERENCES Auster PJ, Maltesa RJ, Langton RW, Watling L, Valentine PC, Donaldson CLS, Langton EW, Shepard AN, Babb IG (1996) The impacts of mobile fishing gear on seafloor habitats in the Gulf of Maine (Northwest Atlantic): implications for conservation offish populations. Rev Fish Sci 4 (2): 185-202 Auster PJ, Michalopoulos C,Valentine PC, Malatesa RJ (1998) Delineating and monitoring habitat management units in a temperate deep-water marine protected area. In: Munro NWP, Willison JHM (eds) Linking Protected Areas with Working Landscapes, Conserving Biodiversity. Science and Management of Protected Areas Association, Nova Scotia Barnett TP, Pierce DW, AchutaRao KM, Gleckler PJ, Santer BD, Gregory JM, Washington WM (2005) Penetration of human-induced warming into the World's Oceans. Sci 309: 284-287 Buhl-Mortensen L, Mortensen PB (2004) Symbiosis in deep-water corals. Symbiosis 37: 33-61 Byrne, M. (1994) Ophiuroidea. In: FW Harrison (ed) Microscopic Anatomy of Invertebrates Vol. 14. Wiley-Lisse Inc, New York Clark AH (1908) Descriptions of new species of crinoids, chiefly from the collections made by the US Fisheries Steamer 'Albatross' at the Hawaiian Islands in 1902; with remarks on the classification of the Comatulida. Proc US Nat Mus 34 (1608): 209-239 Clark MR and Koslow JA (2007) Impacts of fisheries on seamounts. In: Pitcher TJ, Morato T, Hart PJB, Clark MR, Haggan N, Santos RS (eds) Seamounts: Ecology, Fisheries and Conservation, Blackwell Publishing, Oxford, UK Danovaro R, Dell'Anno A, Pusceddu A (2004) Biodiversity response to climate change in a warm deep sea. Ecol Lett 7: 821-828 de Forges BR, Koslow JA, Poore GCB (2000) Diversity and endemism of the benthic seamount fauna in the southwest pacific. Nat 405: 944-947 Eckelbarger KJ, Watling L (1995) Role of phylogenetic constraints in determining reproductive patterns in deep-sea invertebrates. Invertebr Bio 114(3): 256-269 Emson RH, Woodley JD (1987) Submersible and laboratory observations on Asteroschema tenue, a long-armed euryaline brittle star epizoic on gorgonians. Mar Bio 96: 31-45 Faulkner JD (2001) Marine natural products. J Nat Prod Rep 18: 1-49 29 Fossa JH, Mortensen PB, Furevik DM (2002) The deep-water coral Lophelia pertusa in Norwegian waters: distribution and fisheries impacts. Hydrobiologia 471: 112 FujitaT,Ohta S (1988) Photographic observations of the life style of a deep-sea ophiuroid Asteronyx loveni (Echinodermata). Deep-sea Res 35: 2029-2043 Gage JD, Anderson RM, Tyler PA, Chapman R, Dolan E (2004) Growth, reproduction and possible recruitment variability in the abyssal brittle star Ophiocten hastatum (Ophiuroidea: Echinodermata) in the NE Atlantic. Deep-sea Res I 51:849-864 Gage JD, Tyler PA (1991) Deep-Sea Biology: A Natural History of Organisms at the Deep-Sea Floor. Cambridge University Press, Cambridge Genin A (2004) Bio-physical coupling in the formation of zooplankton and fish aggregations over abrupt topographies. J Mar Syst 50: 3-20 Grange KR (1991) Mutualism between the antipatharian Antipathes fiordensis and the ophiuroid Astrobrachion constrictum in New Zealand fjords. Hydrobiologia 216/217: 297-303 Guinotte JM, Buddemeier RW, Kleypass JA (2003) Future coral reef habitat marginality: temporal and spatial effects of climate change in the Pacific Basin. Coral Reefs 22: 551-558 Hall-Spencer JM, Pike J, Munn CB (2007) Diseases affect cold-water corals too: Eunicella verrucosa (Cnidaria: Gorgonacea) necrosis in SW England. Dis Aquat Org 76: 87-97 Hendler G (1991) Echinodermata: Ophiuroidea. In: Giese AC, Pearse JS, Pearse VB (eds) Reproduction of Marine Invertebrates, Vol VI Echinoderms and Lophophorates. Boxwood Press, Pacific Grove, CA Hendler G, Miller JE (1984) Feeding behavior of Asteroporpa annulata, a gorgonocephalid brittlestar with unbranched arms. Bull Mar Sci 34: 449-460 Jumars PA, Nowell ARM (1984) Fluid and sediment dynamic effects on marine benthic community structure. Amer Zool 24: 885-897 Koslow JA, Gowlett-Holmes K, Lowry JK, O'Hara T, Poore GCB, Williams A (2001) Seamount benthic macrofauna off southern Tasmania: community structure and impacts of trawling. Mar Eco Prog Ser 213: 111-125 Lyman T (1879) Ophiuridae and Astrophytidae of the Challenger Expedition II. Bull Mus Comp Zoo 6(2): 17-83 30 McClain CR (2007) Seamounts: identity crisis or split personality? Jour Biogeogr 34: 2001-2008 Metaxas A, Davis J (2005) Megafauna associated with assemblages of deep-water gorgonian corals in Northeast Channel, off Nova Scotia, Canada. Jour Mar Biol Assoc UK 85: 1381-1390 Moore JA, Vecchione M, Collette BB, Gibbons R, Hartel KE, Galbraith JK, Turnipseed M, Southworth M, Watkins E (2003) Biodiversity of Bear Seamount, New England Seamount Chain: Results of Exploratory Trawling. J Northw Atl FishSci31: 363-372 Patent DH (1969) The reproductive cycle of Gorgonocephalus caryi (Echinodermata: Ophiuroidea). Biol Bull (Woods Hole) 136: 241-252 Patent DH (1976) Gonadal histology of the basket star, Gorgocephalus eucemis. Thalassia Jugosl 12: 269-276 Pearse VB, Pearse JS, Hendler G, Byrne M (1998) An accessible population of Ophiocanops, off the NE Sulawesi, Indonesia. In: Mooi R, Telford M (eds) Echinoderms: San Fransisco. Balkema, Rotterdam. Probert PK, Christiansen S, Gjerde KM, Gubbay S, Santos RS (2007) Management and conservation of seamounts. In: Pitcher TJ, Morato T, Hart PJB, Clark MR, Haggan N, Santos RS (eds) Seamounts: Ecology, Fisheries and Conservation, Blackwell Publishing, Oxford, UK Samadi S, Bottan L, Macpherson E, de Forges BR, Boisselier M (2006) Seamount endemism questioned by the geographic distribution and population genetic structure of marine invertebrates. Mar Bio 149: 1463-1475 Smith JE (1940) The reproductive system and associated organs of the brittle-star Ophiothrix fragilis. Q J Microsc Sci 82: 267-309 Stewart BG (1998) Can a snake star earn its keep? Feeding and cleaning behaviour in Astrobrachion constrictum (Farquhar)(Echinodermata: Ophiuroidea), a euryalid brittle-star living in association with the black coral Antipathes fwrdensis (Grange 1990). Jour Exp Mar Biol & Ecol 221:173-189 Stewart BG, Mladenov PV (1994) Aspects of reproduction and anatomy of the euryalinid snake star Astrobrachion constrictum. In: David, Guille, Feral, and Roux (eds) Echinoderms through Time. Balkema, Rotterdam Stewart BG, Mladenov PV (1995) Reproductive periodicity in the euryalinid snake star Astrobrachion constrictum in a New Zealand fjord. Mar Bio 123: 543-553 31 Stewart BG, Mladenov PV (1997) Population structure, growth and recruitment of the euryalinid brittle-star Astrobrachion constrictum (Echinodermata: Ophiuroidea) in Doubtful Sound, Fiordland, New Zealand. Mar Bio, 127: 687-697 Stocks KI, Hart PJB (2007) Biogeography and biodiversity of seamounts. In: Pitcher TJ, Morato T, Hart PJB, Clark MR, Haggan N, Santos RS (eds) Seamounts: Ecology, Fisheries and Conservation, Blackwell Publishing, Oxford, UK Stone GS, Madin LP, Stocks K, Hovermale G, Hoagland P, Schumacher M, Etnoyer P, Sotka C,Tausig H (2004) Seamount biodiversity, exploitation and conservation. In: Glover LK, Earle S (eds) Defying Ocean's End: An Agenda for Action. Island Press, Washington, DC Sumida PYG, Tyler PA, Lampitt RS, Gage JD (2000) Reproduction, dispersal and settlement of the bathyal ophiuroid Ophiocten gracilis in the NE Atlantic Ocean. Mar Bio 137:623-630 Tyler PA, Gage JD (1980) Reproduction and growth of the deep-sea brittle star Ophiura ljungmani (Lyman). Oceol Acta 3: 177-185 Tyler PA, Grant A, Pain SL, Gage JD (1982) Is annual reproduction in deep-sea echinoderms a response to variability in their environment? Nature 300: 747750 Wachsmuth C, Springer F (1897) The North American Crinoidea Camerata. Harv Col Mus Comp Zool Mem 21-22 Waller R, Watling L, Auster P, Shank T (2007) Anthropogenic impacts on the Corner Rise seamounts, north-west Atlantic Ocean. J Mar Biol Ass UK 87: 1075-1076 Worm B, Lotze HK, Myers RA (2003) Predator diversity hotspots in the blue ocean. Proc Nat Acad Sci 100: 9884-9888 Wright EP, Studer T (1889) Report on the Alcyonaria collected by HMS Challenger during the years 1873-1876. Rept Sci Res Challenger Zool 31: i-lxxvii + 1-314, 42 pis 32 BIOGRAPHY OF THE AUTHOR Celeste Virginia Mosher was born in Rumford, Maine, on November 4 , 1974. She was raised in Wilton, Maine, and graduated from Mt. Blue High School in 1993. Celeste was awarded a Bachelor of Science degree in Biology from the University of the Virgin Islands in 2002 where she graduated summa cum laude. After her undergraduate career, she gained marine laboratory and field experience while working with researchers from the Center for Marine and Environmental Studies, St. Thomas, VI, Harbor Branch Oceanographic Institution, Ft. Pierce, FL, and Woods Hole Oceanographic Institution, Woods Hole, MA. She has participated in numerous research cruises conducting physical and biological oceanographic investigations throughout the Caribbean and Northwest Atlantic. Celeste returned to Maine to begin her graduate career at the School of Marine Sciences at the University of Maine in the summer of 2005. Celeste is a candidate for the Master of Science degree in Oceanography from the University of Maine in August, 2008. 33