|USFDC Home | USF Electronic Theses and Dissertations||| RSS|
This item is only available as the following downloads:
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001914270
007 cr mnu|||uuuuu
008 071022s2006 flu sbm 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001746
Burghart, Scott E.
Micronektonic community composition and trophic structure within the bathypelagic zone in the eastern Gulf of Mexico
h [electronic resource] /
by Scott E. Burghart.
[Tampa, Fla] :
b University of South Florida,
ABSTRACT: The Earth's most extensive living space is found in the bathypelagic zone of the oceans, yet research in these areas is scant. The micronekton of the bathypelagic zone in the eastern Gulf of Mexico (EGOM) was investigated with the goals of comparing its community structure and trophic interactions with those of the well-studied overlying mesopelagic micronekton. Significant changes in faunal structure were found, including shifts in dominant families as well as species. Compared to the mesopelagic zone, the bathypelagic community had increased abundance and biomass contributions from the Gonostomatidae, Oplophoridae,and Eucopiidae, with a simultaneous decrease in the importance of the Myctophidae and the Dendrobranchiata. The changed faunal structure within the crustacean assemblage includes a distinct difference in reproductive strategies. There is increased prevalence of taxa which feature egg brooding and abbreviated larval development. In addition, the bathypelagic zon e was characterized by relatively large biomass contributions from rare but large species, particularly those within the families Oplophoridae and Nemichthyidae. The faunal shifts, in combination with a high percentage of bathypelagic species absent from mesopelagic samples (~50% of crustacean and ~37% of fish species), suggest the bathypelagic zone is home to a distinct pelagic community, with a biology and ecology fundamentally different from that of the mesopelagic zone. The broad zoogeographic distributions of bathypelagic species suggest the EGOM assemblage is possibly similar to that of other geographic locations at similar latitudes. Diet analysis was performed on several prominent species and revealed 2 major feeding strategies based on diet composition and prey size. Species of Cyclothone and Eucopia preyedon small planktonic crustaceans, while the decapods examined were primarily piscivorous. The fraction of fish in the diets of decapods was greater than in their mesopelagic^ counterparts. It is suggested the primary trophic players in the system are oplophorid shrimps, followed by Cyclothone spp.
Dissertation (Ph.D.)--University of South Florida, 2006.
Includes bibliographical references.
Text (Electronic dissertation) in PDF format.
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
Title from PDF of title page.
Document formatted into pages; contains 95 pages.
Adviser: Thomas L. Hopkins, Ph.D.
x Marine Science
t USF Electronic Theses and Dissertations.
Micronektonic Community Composition and Trophic Structur e Within the Bathypelagic Zone in the Eastern Gulf of Mexico by Scott E. Burghart A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy College of Marine Science University of South Florida Major Professor: Thomas L. Hopkins, Ph.D. Thomas G. Bailey, Ph.D. Tamara M. Frank, Ph.D. Joseph J. Torres, Ph.D. Gabriel A. Vargo, Ph.D. Date of Approval: October 27, 2006 Keywords: abundance, biomass, diet, zoogeography, reprod uction Copyright 2006, Scott E. Burghart
A CKNOWLEDGEMENTS I am very grateful for the contributions the follo wing people and organizations made to this project. Richard Bohrer, Joe Donnelly, Steve Geiger, Ralph Kitzmiller, James Locascio, Ernst Peebles, Nadina Piehl, Andrew Remsen, Chris S imoniello, Tracey Sutton, Gregory Tolley, and Jos Torres all provided valuable aid, input an d moral support. I am thankful for the identifications provided by Karsten Hartel, Heather Judkins and Ted Pietsch, Tracey Sutton and Mike Vecchione. I am also thankful for the technic al assistance provided by Bill Flanery, Mike Hall, Charlie Jones, Chad Lembke, Jimmy Mulhollon, Jim Mullins, Jim Patten, Randy Russell, Scott Samson, Rich Schmid, Graham Tilbury, the Flor ida Institute of Oceanography (FIO), the Louisiana Universities Marine Consortium (LUMCON), and Harbor Branch Oceanographic Institution (HBOI). The crews of the R/V Suncoaste r, and the R/V Pelican, deserve praise for their hard work and dedication, without which no sa mples would have been collected. I would also like to thank the BBC for providing some of ou r ship time, as well as Jean Claude Sorbe, Brian Kensley, and Mike Vecchione of the Smithsonia n Institution for their helpful input. I am grateful to my committee members: Tom Hopkins, Jos Torres, Gabe Vargo, Tammy Frank, and Tom Bailey, as well as S. Gregory Tolley for agreei ng to chair my defense. Finally, I would like to thank my wife, Terri, and my family for their patie nce, understanding and unfailing support.
i Table of Contents List of Tables iii List of Figures v Abstract vii Chapter 1: The Bathypelagic Decapoda, Lophogastrida and Mysida of the Eastern Gulf of Mexico 1 Introduction 1 Methods 3 Results 8 Hydrography 8 Sample Contamination 11 Crustacean Faunal Composition 11 Vertical Distribution 14 Shrimp Biomass 18 Discussion 19 Biomass Calculations 19 Comparison with Mesopelagic Assemblage 20 Zoogeography 26 Conclusion 30 Chapter 2: The Bathypelagic Fishes of the Eastern G ulf of Mexico 31 Introduction 31 Methods 32 Results 33
ii Assemblage Composition 33 Vertical Distribution 41 Discussion 42 Comparison with Other Studies 42 Comparison with Mesopelagic Assemblage 44 Zoogeography 49 Conclusion 52 Chapter 3: Community and Trophic Structure of the B athypelagic Micronekton in the Eastern Gulf of Mexico 53 Introduction 53 Methods 54 Results 56 Overall Community Structure 56 Diet 60 Discussion 63 Mesoand Bathypelagic Community Comparison 63 Sources of Bias in Diet Data 65 Diet Composition 68 Vertical Distribution of Dietary Copepods 73 Primary Trophic Pathways 75 Conclusions 79 References 81 Appendicies 90 Appendix A 91 Appendix B 92 Appendix C 95 About the Author End Page
iii List of Tables Table 1: List of bathypelagic trawl samples with th eir corresponding depths and type of gear 5 Table 2: Sample occurrence, abundance and biomass o f bathypelagic shrimp species in the eastern Gulf of Mexico. 15 Table 3: Vertical distribution results for the 14 m ost abundant species of bathypelagic shrimp using Fisher's LSD procedure. 18 Table 4: Zoogeography of bathypelagic shrimp specie s found in the eastern Gulf of Mexico. 29 Table 5: Sample occurrence rates, abundances, and b iomass of bathypelagic fish species in the eastern Gulf of Mexico. 37 Table 6: Worldwide occurrence of fish species found in the bathypelagic zone of the eastern Gulf of Mexico. 50 Table 7: Familial composition, in terms of abundanc e and biomass, of the bathypelagic micronekton in the eastern Gulf of Mexico. 57 Table 8: Percent contribution the twenty largest co ntributors of abundance and biomass in the bathypelagic zone of the eastern Gulf of Me xico. 59 Table 9: Comparison of biomass distribution between prominant families in the mesopelagic and bathypelagic zones of the eastern Gulf of Mexico. 64 Table 10: Diet composition results for fourteen spe cies of bathypelagic micronekton from the eastern Gulf of Mexico. 69
iv Table 11: Percentage contribution of major prey gro ups to the diets of two decapod families in the mesopelagic and bathypelagic zones of the eastern Gulf of Mexico. 71 Table 12: Copepod taxa occurring in the diets of ba thypelagic micronekton expressed as percentage of total number of copepods. 73
v List of Figures Figure 1: Location of the study area in the eastern Gulf of Mexico. 3 Figure 2: Representative depth trace (from SC97B 03 ) depicting oscillation of the net within targeted depth horizon. The arrows indicat e times at which the net was opened and closed. 7 Figure 3: Potential temperature and salinity profil es from the bathypelagic zone of the eastern Gulf of Mexico. Data is from all tows dur ing cruise P98. 9 Figure 4: T-S diagram from the eastern Gulf of Mexi co during cruise P98. 10 Figure 5: Number of bathypelagic shrimp species ide ntified against samples collected. 12 Figure 6: Vertical distribution of abundance (A) an d Biomass (B) of bathypelagic shrimp in the eastern Gulf of Mexico. 17 Figure 7: Relative abundance of shrimp families bet ween the mesopelagic and bathypelagic zones in the eastern Gulf of Mexico. 24 Figure 8: Relative biomass of shrimp families betwe en the mesopelagic and bathypelagic zones in the eastern Gulf of Mexico. 25 Figure 9: Number of fish species identified versus number of samples collected in the bathypelagic zone of the eastern Gulf of Mexico. 34 Figure 10: Vertical distribution of fish numbers be tween depth strata in the bathypelagic zone of the eastern Gulf of Mexico. 42 Figure 11: Dominance comparison of fish species abu ndance between the mesopelagic and bathypelagic zones in the eastern Gulf of Mexico. 45 Figure 12: Relative abundance of fish families betw een the mesopelagic and bathypelagic zones in the eastern Gulf of Mexico. 48
vi Figure 13: Relative biomass of fish families betwee n the mesopelagic and bathypelagic zones in the eastern Gulf of Mexico. 48 Figure 14: Cluster results for diet composition bas ed on prey biomass. 61 Figure 15: Cluster results for diet composition bas ed on prey size. 63 Figure 16: Trophic impact of major families in the mesopelagic (A) and bathypelagic (B) zones in the eastern Gulf of Mexico. 78
vii Micronektonic Community Composition and Trophic Str ucture Within the Bathypelagic Zone in the Eastern Gulf of Mexico Scott E. Burghart Abstract The Earths most extensive living space is found i n the bathypelagic zone of the oceans, yet research in these areas is scant. The micronekton of the bathypelagic zone in the eastern Gulf of Mexico (EGOM) was investigated with the goals of comparing its community structure and trophic interactions with those of th e well-studied overlying mesopelagic micronekton. Significant changes in faunal structu re were found, including shifts in dominant families as well as species. Compared to the mesop elagic zone, the bathypelagic community had increased abundance and biomass contributions f rom the Gonostomatidae, Oplophoridae, and Eucopiidae, with a simultaneous decrease in the importance of the Myctophidae and the Dendrobranchiata. The changed faunal structure wit hin the crustacean assemblage includes a distinct difference in reproductive strategies. Th ere is increased prevalence of taxa which feature egg brooding and abbreviated larval develop ment. In addition, the bathypelagic zone was characterized by relatively large biomass contr ibutions from rare but large species, particularly those within the families Oplophoridae and Nemichthyidae. The faunal shifts, in combination with a high percentage of bathypelagic species absent from mesopelagic samples (~50% of crustacean and ~37% of fish species), sugg est the bathypelagic zone is home to a distinct pelagic community, with a biology and ecol ogy fundamentally different from that of the mesopelagic zone. The broad zoogeographic distribu tions of bathypelagic species suggest the EGOM assemblage is possibly similar to that of othe r geographic locations at similar latitudes. Diet analysis was performed on several prominent sp ecies and revealed 2 major feeding strategies based on diet composition and prey size. Species of Cyclothone and Eucopia preyed
viii on small planktonic crustaceans, while the decapods examined were primarily piscivorous. The fraction of fish in the diets of decapods was great er than in their mesopelagic counterparts. It is suggested the primary trophic players in the system are oplophorid shrimps, followed by Cyclothone spp.
1 Chapter 1: The Bathypelagic Decapoda, Lophogastrida and Mysida of the Eastern Gulf of Mexico Introduction Most of the earths habitable volume is found in th e open ocean, specifically, those areas with a bottom depth of greater than 200 m. The ave rage depth of the oceans is 3800 m, hence 88% of its volume lies below 1000 m depth (Herring 2002). Consequently, the planets largest ecosystem is the bathypelagic zone. Despite the vo lumetric dominance of this habitat, little effort has been dedicated to assessing even its most basic features. Primary reasons for this include the fact that most production, biomass, and commerc ial value is in the epipelagic zone, and that exploration of the bathypelagic zone is costly and logistically difficult. In contrast, most oceanic research has been conduc ted in the oceans upper 1000 m. As an example, the mesopelagic zone of the eastern Gul f of Mexico (EGOM), which is ecologically similar to other low latitude oceanic systems, has been the subject of regular sampling for over twenty years; consequently much information is avai lable on community structure and trophic relationships. To summarize, the system supports a total micronekton biomass of 375-450 kg DW km -2 (derived from data in Hopkins and Lancraft 1984) d ominated by decapods and fish (Hopkins and Lancraft 1984), and is characterized b y the high diversity characteristic of low latitude oceanic systems. The community contains o ver 250 species of fish (Hopkins and Lancraft 1984; Gartner et al. 1987; Sutton and Hopk ins 1996a; Hopkins et al. 1997), at least 30 species of decapods (Heffernan and Hopkins 1981; Ho pkins et al. 1989; Flock and Hopkins 1992; Hopkins et al. 1994), and at least 47 species of ce phalopods (Passarella and Hopkins 1991). Typically, micronektonic species vertically migrate on a diel basis, and diets within the community are most often planktivorous. Although the microne kton largely depend on zooplankton for food,
2 its predation does not appear to have a large impac t on that assemblage (Hopkins et al. 1997). Predation between micronekton groups, however, can be intense (Sutton and Hopkins 1996b; Hopkins et al. 1997). Significantly, despite high diversity, niches are well defined when sufficient environmental factors are considered (Hopkins and G artner 1992; Hopkins et al. 1994; Hopkins and Sutton 1998). While oceanic mesopelagic systems seem homogenous compared to epipelagic and coastal waters, they contain vertical gradients in light, temperature, and food concentration. Below 1000 m, this is not the case. Sunlight is ab sent, hence the system is disassociated from a source of primary production (hydrothermal vents ex cepted), and there is little variation in temperature. Bathypelagic organisms reside in a un ique environment characterized by high pressure and low temperature, low temporal variabil ity, and diffuse resources. Despite the unique and extensive nature of such systems, the de arth of data has left community composition and energy cycling very poorly understood. Between 1996 and 2000, a trawling program was condu cted to obtain fundamental information on the composition, distribution and tr ophic relationships of the micronekton in the bathypelagic zone of the EGOM. As mentioned above, decapods and fishes overwhelmingly dominate the biomass of trawl catches in the mesope lagic zone of the EGOM (e.g. Hopkins and Lancraft, 1984) and our catches revealed the same i s true for the bathypelagic zone. This chapter reports on the bathypelagic populations of decapods, lophogastrids, and mysids (hereafter referred to collectively as shrimp). A search of bathypelagic literature indicates that th e present study is unique in that it treats the speci es composition, abundance, vertical distribution and trophic relationships among the micronekton in one comprehensive study. Furthermore, the wealth of data available from the overlying mesopel agic zone provides a strong basis for comparison of assemblages within the same geographi c area.
3 Methods The study site was in the eastern Gulf of Mexico w ithin a 30-km radius of 27N 86W, a location that has been sampled by the University of South Florida for more than 22 years (Figure 1). Previous work in the area suggests it may be u sed as an analogue for low latitude oceanic gyre systems (Hopkins and Gartner 1992; Hopkins et al. 1994), although productivity and standing stock levels are closer to that of oceanic boundary currents (Hopkins 1982; Remsen et al. 2003). All samples were collected in the warm months between May and September. 2 0 0 m 1 0 0 0 m 2 0 0 m 1 0 0 0 m 2 0 0 m 1 0 0 0 m Figure 1: Location of the study area in the eastern Gulf of M exico.
4 Except for one cruise in 1999 (Table 1), samples we re collected using an opening/closing rectangular midwater trawl (Tucker Trawl) with a mo uth area of 9-m 2 and a mesh size of 4-mm tapering to a meter net with 1-mm mesh. The cod en d bucket itself was lined with 1000m mesh netting. The volume of every tow was measured usin g a General Oceanics flowmeter mounted in the mouth of the net, and some tows also included a nested 162m m mesh plankton net (Hopkins and Baird, 1975). Depth of the Tucker Trawl was monitored in real-tim e during tows using either a MOCNESS depth sensor (1996, 1999 and 2000) or a dep th sensor designed specifically for the trawl by the USF Center for Ocean Technology (COT) (Table 1). During the first four cruises, depth was also monitored with a Time Depth Recorder (TDR). On cruise P98, a cable failure prevented transmission of electronic data from the COT depth sensor and resulted in the use of a SeaBird CTD with data logging capability to record trawl depth, temperature, and salinity. Two mechanisms were employed to close the trawl, e nsuring discrete depth sampling. Prior to 1999, trawls were opened and closed using electronic clock timers. The first cruise in 1999 (P99) utilized an entire 4-m 2 MOCNESS system, while the later cruises in 1999 an d 2000 utilized the MOCNESS electronics to open and close the Tucker Trawl as well as monitor its depth (Table 1). When the clocks were in use, the net was first lowered to the bottom of the desired depth interval until it opened. At the end of the trawl, the net was held at depth for a minimum of 30 minutes after the scheduled closing t ime prior to being brought to the surface to ensure closure prior to retrieval.
5 Table 1: List of bathypelagic trawl samples with their corre sponding depths and type of gear. Cruise Number Trawl Number Depth Net Type SC96 5 1000 1500 Tucker SC96 6 1500-2000 Tucker SC96 10 1000-1500 Tucker SC96 11 1500-2000 Tucker SC96 B 4 1000-1500 Tucker SC96 B 5 2000-2500 Tucker SC96 B 9 1500-2000 Tucker SC96 B 16 1000-2000 Tucker SC96 B 18 1000-1500 Tucker SC96 B 22 1000-1500 Tucker SC96 B 23 1500-2000 Tucker SC97 1 1000-1500 Tucker SC97 2 1000-1500 Tucker SC97 3 2000-2500 Tucker SC97 4 1500-2000 Tucker SC97 5 1000-1500 Tucker SC97 6 1000-1500 Tucker SC97 8 2500-3000 Tucker SC97 9 1500-2500 Tucker SC97 10 1500-2000 Tucker SC97 B 2 1000-1500 Tucker SC97 B 3 1500-2000 Tucker SC97 B 4 1500-2000 Tucker SC97 B 5 1000-1500 Tucker SC97 B 6 1500-2000 Tucker SC97 B 7 1000-1500 Tucker SC97 B 9 1500-2000 Tucker SC97 B 10 1500-2000 Tucker P98 7 1500-2000 Tucker P98 11 1500-2000 Tucker P98 15 1500-2000 Tucker P98 20 2000-3000 Tucker P98 23 2000-3000 Tucker P98 30 1000-2000 Tucker SC98 1 1000-1500 Tucker SC98 4 2000-2500 Tucker SC98 9 1500-2000 Tucker SC98 11 1000-1500 Tucker SC98 15 1500-2000 Tucker SC98 18 1500-2000 Tucker SC98 21 1000-1500 Tucker P99 2 1000-1500 MOCNESS P99 12 1000-1500 MOCNESS P99 17 1000-1500 MOCNESS SC99 3 2000-2500 Tucker
6 The water column below 1000 m was divided into 500 -m discrete depth intervals (Table 1), and a majority of the trawls took place entirel y within one of these intervals. Utilizing the depth-monitoring device, the trawl was oscillated w ithin the target depth interval by varying ship speed. (an example of this can be seen in Figure 2) During the 1999 cruise, during which the 4m 2 MOCNESS was used, only the 1000-1500 m interval wa s sampled, in 100-m intervals, but these were pooled for the purpose of this paper. O verall, the majority of sampling took place in the upper bathypelagic zone, with 20 tows sampling the 1000-1500 m and 16 sampling the 15002000 m depth horizons (Table 1). Seven tows succes sfully sampled depth intervals below 2000 m, and for the purposes of analysis were grouped to gether and categorized as >2000 m trawls. In addition, three trawls successfully opened and c losed within the bathypelagic zone, but did not remain within one of the 500-m depth intervals, or below 2000 m (e.g., trawl SC97 09 fished between 1500 and 2500 m). All organisms were counted, identified to lowest po ssible taxonomic level (usually species), and measured to the nearest mm. Individu als were included in the analysis only if they exceeded 20 mm TL, which was taken to be the lower size limit of micronekton. Raw counts were standardized by volume filtered and then repre sented as individuals km -2 over a 500-m deep interval of the water column. Species that ha ve not previously been recorded below 700 m in the EGOM were considered contaminants, and exclu ded from analysis. There are two probable sources of sample contamination. First, t he closing mechanism may have failed to operate before the net rose above 1000 m. However, as stated above, retrieval of the net did not begin until at least 30 minutes after the scheduled closing time. Nets observed to be even partially open upon arrival at the surface were not counted as discrete depth tows, and therefore excluded. Error resulting from a tow in which the n et hung open temporarily (but closed completely before reaching the surface), however, w ould not have been detected. The second possible source of contamination involves drift of the net above 1000 m for a significant period of time. The ability to monitor the depth of the net in real time minimized such error, and any tows that strayed above 1000 m for a significant period of time were excluded.
7 0 500 1000150020002500 0100200300400500600700 Time (minutes)Depth (m) Figure 2: Representative depth trace (from SC97B 03) depictin g oscillation of the net within targeted depth horizon. The arrows indicate times at which the net was opened and closed. Species numbers have been converted to biomass usin g relationships between total or standard length and weight established from mesopel agic samples (Torres and Donnelly, unpublished; Hopkins, unpublished). Whenever possi ble, regressions were generated using data obtained from non-preserved material (Torres and Do nnelly, unpublished). Dry weights were calculated by adjusting for water content reported for the closest possible taxonomic group (Childress and Nygaard 1973; Childress and Nygaard 1974; Donnelly and Torres 1988; Stickney and Torres 1989; Childress et al. 1990). At times, it was necessary to apply some equations to multiple genera (Appendix A). In such cases, phylo geny and morphology were matched as
8 closely as possible. For statistical comparison of vertical distribution, all tows were divided into three depth categories: 1000-1500 m, 1500-2000 m, a nd >2000 m. The three tows with depths not within one of these categories were discarded. Abundance and biomass data were log transformed and the variance was checked using Levi nes test. They were then analyzed for difference between depth horizons using ANOVA combi ned with Fishers LSD procedure. Results Hydrography Representative temperature and salinity data, acqu ired during the 1998 R/V Pelican (cruise P98), are shown in Figure 3. Overall, pote ntial temperature varied little between 1000 m and 3000 m, remaining between 4 and 4.75 o C, with most of the variation occurring between 100 0 and 1500 m. Salinity also varied little, remaining between 34.9 and 35.0. The Gulf of Mexico is connected to the Atlantic basin in the east by the Florida Strait (depth 800-1000 m) and to the Caribbean basin in the south by the Yucatan Strait (depth ~1900 m). A T-S curve (Figure 4) shows water mass signatures typical of the eastern Gulf of Mexico and adjacent Atlantic. In the Gulf of Mexico, North Atlantic Central water overli es remnants of Antarctic Intermediate Water, represented by a salinity minimum at about 1000 m ( Figures 3 and 4). North Atlantic Deep Water was represented by higher salinities that coincided with lower temperatures within this water mass (Nowlin 1971; Pickard and Emery 1990). The sa linity maximum visible in Figure 4 is characteristic of subtropical underwater.
9 1000 14001800220026003000 34.534.7 34.9 35.135.335.5 4.04.2 4.4 22.214.171.124 (oC) SalinityDepth (m) Temperature Salinity 1000 14001800220026003000 34.534.7 34.9 35.135.335.5 4.04.2 4.4 126.96.36.199 (oC) SalinityDepth (m) Temperature Temperature Salinity Salinity Figure 3: Potential temperature and salinity profiles from th e bathypelagic zone of the eastern Gulf of Mexico. Data is from all tows during cruis e P98.
10 0 5 10 15 20 25 30 35 34.53535.53636.537 SalinityTemperature Figure 4: T-S diagram from the eastern Gulf of Mexico during cruise P98.
11 Sample Contamination The total abundance of all species deemed to be me sopelagic contaminants amounted to 1.4% of the total number of shrimp collected. The two species most clearly satisfying the criteria for contamination were Pasiphaea merriami (two individuals), and Parapandalis richardi (five individuals) (Hopkins et al. 1989). Gennadas scutatus and G. talismani were also classified as contaminants (see Heffernan and Hopkins 1981), and both were present in very low numbers (a combined six specimens in four samples). The large st source of contamination was due to individuals of Sergestes (25 individuals), which were conservatively consid ered contaminants despite some evidence of sporadic occurrence below 700 m in the EGOM (Heffernan and Hopkins 1981; Flock and Hopkins 1992). Crustacean Faunal Composition The three dominant crustacean orders in our catche s were Lophogastrida, Mysida, and Decapoda (following Martin and Davis 2001). Other crustacean groups, namely euphausiids and amphipods, were encountered in very low numbers (34 and 38 total individuals, respectively), and with very few exceptions, were too small to be cons idered micronekton (< 2 cm TL). The total abundance of euphausiids and amphipods combined was 1% that of all other crustacean groups. Therefore, the scope of this study was restricted t o the three dominant orders. For the sake of convenience, these orders will hereafter be referre d to collectively as shrimp, an umbrella term that does not imply phylogenetic affinity. In tota l, we collected 48 species from eight different families, and Figure 5 indicates the sampling effor t effectively characterized the bathypelagic shrimp assemblage. The oplophorids were the most speciose family with 24 different species identified; four times the number of species than the next closest f amily of crustaceans (Table 2). Collectively, oplophorids were the most commonly encountered fami ly, with every tow containing at least one individual from the family. In addition, 13 differ ent species had occurrence rates greater than
12 10%, three of which were encountered in at least ha lf of the samples: Hymenodora glacialis (91.3%), Acanthephyra stylorostratis (78.3%), and A. curtirostris (58.7%). Hymenodora glacialis was absent from only five tows and was the most com monly encountered shrimp overall. Species Count (Cumulative)Samples 0 10 20 30 40 50 13579111315171921232527293133353739414345 2468101214161820222426283032343638404244 Figure 5: Number of bathypelagic shrimp species identified ag ainst samples collected. Altogether, the oplophorids accounted for about one quarter of the individuals collected. Two oplophorids, Acanthephyra stylorostratis and Hymenodora glacialis were found in numbers greater than 10,000 ind. km -2 (Table 2). Six oplophorids were among the 20 most abundant shrimp taxa overall: H. glacialis (4 th ), A. stylorostratis (7 th ), A. curtirostris (10 th ), H. gracilis (12 th ), A. gracilipes (15 th ), and A. purpurea (17 th ). A majority of the oplophorid species, however, were rare, as 16 of the 24 species were represented by t en or fewer specimens. Five species of Sergia were positively identified, the most common being Sergia splendens occurring in 39.1% of the samples. All other spe cies of this genus had sample occurrence rates of less than 15%. The Sergestidae accounted for 6.9% of the total shrimp
13 numbers, and with the exception of S. splendens which had an abundance of 5600 ind. km -2 (9 th most abundant species overall), all species were fo und in numbers less than 5000 ind. km -2 Sergia that were either immature or too damaged to be ide ntified to species comprised the second most abundant taxon within the family. There were four benthesicymid species identified. Bentheogennema intermedia the most commonly encountered was present in 69.6% of the tows (Table 2). Gennadas valens also occurred in more than half the samples (54.3%) while the third most commonly encountered member of the family was G. capensis (39.1% of samples). Gennadas bouvieri occurred in some samples (~10%), but usually as solitary individuals resulting in a total catch of only seven specimens. The benthesicymids accounted for 18.9% of the shrim p identified. The most abundant member of the family was Bentheogennema intermedia, which at an average of 13,000 ind. km -2 was the 5 th most abundant shrimp overall. No other benthesicy mids were present in numbers greater than 10,000 ind. km -2 although Gennadas valens (9700 ind. km -2 ), the 8 th most abundant shrimp overall, was nearly so. The four species within the Eucopiidae were among the most commonly collected organisms, occurring in at least half of the sample s (Table 2). Only three trawls contained no specimens of this family. The two most frequently collected species were Eucopia australis (87.0%) and E. sculpticauda (84.8%). In addition, the 6 th most commonly encountered shrimp overall was E. grimaldii which occurred in 67.4% of the samples, while the 4 th E. unguiculata was found in 50.0% of the samples.
14 In addition to being commonly encountered, the euc opiids were the most abundant family. Collectively they accounted for 46% of the shrimp collected. All four species were among the ten most abundant shrimp. Eucopia sculpticauda E. australis and E. grimaldii were all present in numbers close to 20,000 ind. km -2 while E. unguiculata was present at 12,700 ind. km 2 These four species were the 1 st 2 nd 3 rd and 6 th most abundant shrimp species, respectively. The families Mysidae, Pasiphaeidae, and Bresiliida e made small contributions to the catch, combining to add four taxa to the list (Tabl e 2). While two taxa, Boreomysis spp. and Lucaya bigelowi were encountered in more than 20% of the trawls, all of the members of these families were present in numbers less than 1000 ind km -2 Vertical Distribution Abundance for the whole shrimp assemblage was high est in the 1000-1500 m depth zone (Figure 6a), as was biomass (Figure 6b). An i dentical procedure was applied to the abundance of each species that numbered greater tha n 1000 ind. km -2 Five of 14 species in this abundance category did not show significant trends in abundance with depth (Table 3). For one species, Gennadas capensis there was a significant trend at the P < 0.1 leve l (P value was 0.06) in which abundance was highest in the 1000-1500 m d epth zone. Only one of the ten most abundant species, Hymenodora glacialis displayed no significant trend in depth distribut ion. For six of the nine species that did display significan t trends, abundance was higher in the 1000-1500 m depth zone than either of the other two zones, wh ile no difference was displayed between the lower two zones. In the remaining three cases, no differences were discerned between the shallowest and deepest depth zones.
15 Table 2: Sample occurrence, abundance and biomass of bathype lagic shrimp species in the eastern Gulf of Mexico. % occurrence Individuals km -2 (total) % of numbers kg DW km -2 (total) % of biomass Lophogastridae Gnathophausia gigas 17.4 500 0.3 0.2 0.7 Gnathophausia gracilis 13.0 400 0.3 0.1 0.3 Gnathophausia ingens 2.2 <100 0.0 0.1 0.3 Gnathophausia zoea 10.9 300 0.2 0.1 0.3 Pseudochalaraspidum hanseni 8.7 200 0.1 <0.1 0.1 Eucopiidae Eucopia australis 87.0 20,300 13.2 1.6 5.2 Eucopia grimaldii 67.4 19,600 12.8 0.9 3.1 Eucopia sculpticauda 84.8 20,300 8.3 0.8 2.1 Eucopia unguiculata 50.0 12,700 13.2 0.6 2.8 Mysidae Boreomysis sp 23.9 800 0.5 <0.1 <0.1 Benthesicymidae Bentheogennema intermedia 69.6 13,000 8.5 3.7 12.5 Gennadas bouvieri 13.0 300 0.2 <0.1 0.1 Gennadas capensis 39.1 4,000 2.6 0.4 1.4 Gennadas valens 54.3 9,700 6.3 1.0 3.4 Gennadas spp 28.3 2,600 1.7 0.1 0.4 Sergestidae Sergia grandis 8.7 200 0.1 0.1 0.3 Sergia japonica 13.0 400 0.3 0.1 0.3 Sergia regalis 10.9 800 0.5 0.3 1.0 Sergia splendens 39.1 5,600 3.5 0.4 1.4 Sergia wolffi 10.9 800 0.5 0.3 1.0 Sergia spp 10.9 1,900 1.2 0.3 1.0 Sergestes spp 19.6 1,200 0.8 0.1 0.4 Pasiphaeidae Parapasiphaea macrodactyla 10.9 200 0.1 0.2 0.1 Parapasiphaea sulcatifrons 15.2 600 0.4 0.5 1.7
16 Table 2 continued % occurrence Individuals km -2 (total) % of numbers kg DW km -2 (total) % of biomass Oplophoridae Acanthephyra spp. 8.7 300 0.2 0.1 0.3 Acanthephyra acanthitelsonis 15.2 500 0.3 1.3 4.4 Acanthephyra acutifrons 13.0 400 0.3 4.7 15.9 Acanthephyra curtirostris 56.5 4,100 2.6 1.7 5.8 Acanthephyra exima 2.2 <100 0.0 <0.1 0.0 Acanthephyra gracilipes 37.0 1,900 1.2 0.4 1.4 Acanthephyra pelagica 4.3 100 0.1 <0.1 0.1 Acanthephyra purpurea 21.7 1,200 0.8 0.4 1.4 Acanthephyra quadrispinosa 6.5 200 0.1 0.2 0.7 Acanthephyra stylorostratis 73.9 12,300 7.7 1.7 5.8 Ephyrina benedicti 10.9 300 0.2 0.8 2.7 Ephyrina ombango 17.4 700 0.4 0.6 2.0 Hymenodora glacialis 89.1 13,000 8.2 1.2 4.1 Hymenodora gracilis 43.5 400 2.5 0.5 1.7 Janicella spinicauda 6.5 300 0.2 <0.1 0.0 Meningodora marptocheles 2.2 <100 0.0 <0.1 0.0 Meningodora miccyla 2.2 <100 0.0 <0.1 0.0 Meningodora mollis 17.4 500 0.3 0.6 2.0 Meningodora vesca 21.7 500 0.3 0.2 0.3 Notostomus gibbosus 15.2 700 0.0 3.3 10.8 Systellaspis braueri 6.5 100 0.4 0.1 0.3 Systellaspis cristata 2.2 <100 0.0 <0.1 0.0 Systellaspis debilis 8.7 400 0.3 0.1 0.3 Systellaspis pellucida 6.5 100 0.1 <0.1 0.0 Bresiliidae Lucaya bigelowi 21.7 800 0.5 <0.1 0.1
17 Figure 6: Vertical distribution of abundance (A) and biomass (B) of bathypelagic shrimp in the eastern Gulf of Mexico.
18 Table 3: Vertical distribution results for the 14 most abund ant species of bathypelagic shrimp using Fisher's LSD procedure Species 1000 1500 & 1500 2000 1000 1500 & >2000 1500-2000 & >2000 Eucopia sculpticauda X X NS Eucopia australis X X NS Eucopia grimaldii X NS NS Eucopia unguiculata X X NS Hymenodora glacialis NS NS NS Bentheogennema intermedia X X NS Acanthephyra stylorostratis X X NS Gennadas valens X NS NS Sergia splendens X X NS Acanthephyra curtirostris X NS NS Gennadas capensis NS NS NS Hymenodora gracilis NS NS NS Acanthephyra gracilipes NS NS NS Acanthephyra purpurea NS NS NS X indicates a significant difference (p<0.05). NS indicates no significant difference. Shrimp Biomass The entire shrimp assemblage totaled an estimated 29.8 kg DW km -2 Oplophorids accounted for 59.6% of the biomass, although they c omprised about a quarter of the numbers. Among species with prominent biomass, six of the to p ten, and 12 of the top 20 were oplophorids. Only eight specimens of Acanthephyra acutifrons were collected, yet they accounted for the highest biomass of all shrimps (Table 2): 15.6% of the total. Similarly, Ephyrina benedicti, E. ombango, A. acanthitelsonis and Notostomus gibbosus were collected in very low numbers (7, 14, 10 and 14 specimens respectively), but contribu ted a disproportionate amount of biomass due to their large size. The family Benthesicymidae, with a combined 18% of the total, had the second largest biomass and included three species among the top 20 contributors. This was principally due to the biomass of Bentheogennema intermedia which was the second highest of all the shrimps
19 (3.7 kg dry weight km -2 ) and accounted for 12.5% of the total (Table 2). The other two species among the top 20 included the second highest total in the family, Gennadas valens (3.5% of the total), and G. capensis The four species of Eucopia together totaled 13.1% of shrimp biomass with the largest fraction being that of E. australis (5.2%). This species had the 6 th highest biomass of all shrimp species, while the other three species of the famil y ( Eucopia grimaldii, E. sculpticauda and E. unguiculata ) had the tenth, eleventh and thirteenth highest to tals, reflecting their similar size and abundance. Sergestids were the only remaining family to compr ise over 5% of the total biomass (5.6%). The principal biomass contributor was Sergia splendens (0.40 kg DW km -2 ; 1.4% of total), while S. regalis, S. wolffi, and unidentified members of Sergia all accounted for 1%. The remaining two species, S. grandis and S. japonica contributed little to the total. Two families, Lophogastridae and Pasiphaeidae, had similar biomasses (1.8% and 1.9%, respectively), and between them, only the pasiphaei d Parapasiphaea sulcatifrons was among the top twenty species. The largest contributor from t he Lophogastridae was Gnathophausia gigas with an estimated biomass of 0.22 kg DW km -2 (0.7% of the total). The remaining two families, Mysidae and Bresiliidae, added little to the assemb lage biomass, each amounting to 0.1% or less. Discussion Biomass Calculations Despite contributing only 5% of the numerical abun dance, the biomass of the bathypelagic shrimp was 17% that of the mesopelagic shrimp assemblage (according to Hopkins et al. 1994), indicating individuals below 1000 m a re generally larger (cf. Mauchline 1972). One of the more striking results was the high biomass c ontribution of species with low abundances, implying that energy flow through a small number of relatively large micronektonic individuals may
20 be important in the bathypelagic zone. Numerically important species, such as those in the genera Eucopia and Hymenodora, while still significant contributors in terms of b iomass (and thus energy cycling), were less so than their abundance would indicate. The relationships between size and biomass were der ived using data collected from mesopelagic shrimp in the EGOM (Torres and Donnelly unpublished). In the EGOM, the trend is for water content of crustaceans to increase with i ncreasing minimum depth of occurrence (MDO) (Donnelly et al. 1993a). For example, the water co ntent of Acanthephyra purpurea (MDO 300 m) averages 73.6%, while that of A. acutifrons (MDO 800 m) averages 85.2%. According to the same work, the migratory behavior of the species in question also has an effect on water content, exhibited in lower water content of migratory compa red to that of non-migratory species (73.1 3.8% versus 82.0 7.9% respectively). Even within a given genus, variation in relationships between carapace length and dry weight can be large Equations derived for mesopelagic migrating shrimp species will invariably lead to an overestimate of biomass for bathypelagic, nonmigrators such as Bentheogennema intermedia, Hymenodora spp. Notostomus spp., and some species of Acanthephyra The problem may be especially acute in the Oplop horidae as the same equation is applied across several genera. However biomass for all species of shrimp was calculated in the same manner and thus, any downwar d adjustment would include virtually all species to some degree, including those that are sm aller and more numerous. Thus, while the absolute values of biomass may vary depending on th e method used, the conclusion that relatively small numbers of large organisms are imp ortant to energy cycling in the bathypelagic zone remains unaltered. Comparison with Mesopelagic Assemblage The division between mesopelagic and bathypelagic z ones in the ocean is based mainly on light attenuation (Herring, 2002), but little ev idence has been advanced regarding any ecological validity of this boundary. Recently, th ere has been some effort devoted to resolving this, although it has tended to focus on plankton, particularly copepods (Yamaguchi et al. 2002;
21 Yamaguchi et al. 2004; Yamaguchi et al. 2005). Col lectively, those studies have found significant changes with depth in community composition, chemic al composition, and ontogenetic vertical migration patterns. In an extensive review on the vertical distribution of plankton, Vinogradov (1997) suggested that, while not representing a fir m ecological border, the boundary has some significance at lower latitudes where vertical stra tification is more pronounced. Childress et al. (1980), examining micronekton, compared life histor y strategies among fishes living in different depth zones and found deep-living, non-migratory sp ecies delayed reproduction until they approached maximum age. Due to the large body of w ork published on the EGOM mesopelagic ecosystem, a direct comparison between the two zone s at this location is possible. While access to only summary data precludes rigorous statistical comparison, there is enough information to establish patterns. To begin, there is notable overlap in the fauna of both zones. Four species, Gennadas valens, Sergia splendens, Eucopia sculpticauda, and E. unguiculata, rank among the top ten in abundance in both zones. Furthermore, six species are among the 20 highest biomass contributors to both assemblages: Acanthephyra curtirostris, G. valens, E. sculpticauda, E. unguiculata, G. capensis, and A. purpurea Finally, of the 67 species found in the EGOM, al most half have distributions that span the 1000-m isobat h. As mentioned above, several of those species were prominent in both sets of samples, ind icating that the distributions of several important micronektonic shrimp species have vertica l distributions best characterized as deep mesoto bathypelagic Other authors have also found evidence of species d istributions extending into both depth zones. Foxton (1970) found that some large m ales and large ovigerous females of A. purpurea reside below 1000 m during the day, and move above this depth at night. Vereshchaka (1994) and Donaldson (1975) similarly found that so me species of Sergia migrate from the bathypelagic zone to the mesopelagic zone on a diel basis, while others have broad distributions spanning the two zones. Thus, the delineation betw een mesopelagic and bathypelagic is often not absolute; however, there are several crucial di fferences between the two assemblages.
22 In the EGOM, shrimps are numerically more abundant above than below 1,000 m. In the mesopelagic zone, shrimps totaled 3.1 X 10 6 ind. km -2 (Hopkins et al. 1994), while in the bathypelagic zone there were 1.5 X 10 5 ind. km -2 Of the 67 total species reported from the area, slightly less than half were found in greater abund ance below 1000 m than above. Seventeen of those were oplophorids. Even though statistical co mparison of abundances between the two zones was not possible, the numbers are highly sugg estive. Another key difference is that 16 species found in the present study were absent from the mesopelagic zone despite decades of sampling. Thos e included four of the nine lophogastrids (Lophogastridae and Eucopiidae), as well as Boreomysis sp Importantly, three of the 16 new species, Acanthephyra gracilipes and both species of Hymenodora, were present in numbers greater than 1000 ind. km -2 The bathypelagic samples also added ten species of oplophorids to the list of species reported in the EGOM. There were also obvious changes in the dominant org anisms within families, most prominently in the Benthesicymidae and Oplophoridae Above 1000 m, Gennadas valens was not only the dominant benthesicymid, but also the d ominant shrimp in the EGOM. This single species accounted for almost 50% of the shrimp biom ass, and nearly equaled that of the entire fish family Myctophidae (Hopkins et al. 1994). Whi le G. valens was still present in high numbers below 1000 m, its abundance decreased by more than two orders of magnitude. Conversely, Bentheogennema intermedia, which was present in very low numbers above 1000 m became the most abundant benthesicymid and, along with Hymenodora glacialis exhibited the 4 th highest abundance (Table 2). Within the Oplophoridae the s hift involved replacement of two of the three most abundant species. The three most abundant spe cies above 1000 m were, in order, Systellaspis debilis, Acanthephyra purpurea and A. curtirostris, while below 1000 m they were Hymenodora glacialis, A. curtirostris, and A. stylorostratis. The distinctness of the two assemblages was further highlighted by the contribution each species made to overall abundance and biomass. In the mesopelagic zone, five species contributed 5% or more to total numbers ( Sergestes pectinatus Sergia splendens, Gennadas
23 valens, G. capensis, and Eucopia unguiculata ), while below 1000 m, there were eight ( Bentheogennema intermedia G. valens, Acanthephyra stylorostratis, Hymenodora glacialis, E. australis, E. grimaldii, E. sculpticauda, and E. unguiculata ). The similarity in gear used to sample both zones allowed diversity comparisons, and showe d a somewhat higher measure of species eveness in the bathypelagic zone (J 0.6179 versus 0.7094). A matching trend was observed for biomass, with four species contributing more than 5 % above 1000 m and six species contributing that level below 1000 m. Perhaps the definitive difference between the two z ones is the sharp contrast in abundance and biomass considered at the family leve l. The composition of the mesopelagic shrimp assemblage in the EGOM was dominated by the Benthesicymidae (again, due mainly to Gennadas valens ) and Sergestidae (Figures 7 and 8). When combined the two families accounted for about 85% of the individuals and 78% of the biomass. In contrast, within the bathypelagic zone, the Eucopiidae and Oplophoridae were the most numerous families (75% combined Figure 7), while the Oplophoridae and Be nthesicymidae were the two highest contributors of biomass (79% combined Figure 8). Although benthesicymids were still prominent below 1000 m, their biomass fraction decr eased from 56% to 18%. At the same time, the biomass contribution of the Eucopiidae increase d from 4% to 13%. Thus, between the mesopelagic and bathypelagic zones, the dominant fa milies change from Benthesicymidae and Sergestidae, to the Oplophoridae and Eucopiidae. The shifts in faunal structure highlight the distin ctive nature of the two assemblages, and indicate fundamental biological differences between the two communities. Above 1000 m, species that undergo diel migration dominate the co mmunity. Of the 50 shrimp species reported in the EGOM, 32 were diel migrants, which together accounted for almost 90% of the numbers and biomass. Angel et al. (1982) suggested a break between migratory and non-migratory behavior exists at 1000 m, and our data support thi s. Below 1000 m, only 12 of the 48 species have been reported as migratory and combined they r epresent 15% of the numbers and 9% of the biomass.
24 Although swimming rates of deep-sea shrimp dont pr eclude vertical migration based on time constraints (Bailey et al. 2005), it is possib le the physiological cost of migrating from the bathypelagic zones into shallow water outweighs the benefits (Torres and Childress 1983; Torres 1984). Another possibility is the absence of visua l cues in the form of a changing level of downwelling irradience at dusk and dawn. The reduc ed migration within the bathypelagic shrimp assemblage serves to further isolate the community from primary production. Figure 7 : Relative abundance of shrimp families between the mesopelagic and bathypelagic zones in the eastern Gulf of Mexico. 0% 20% 40% 60% 80% 100% MesopelagicBathypelagic% of abundance Sergestidae Benthesicymidae Oplophoridae Eucopiidae Other
25 Figure 8: Relative biomass of shrimp families between the mes opelagic and bathypelagic zones in the eastern Gulf of Mexico. In addition to reduced migration, the bathypelagic assemblage exhibited a notable shift in reproductive strategy. Above 1000 m, species that broadcast their eggs (Sergestidae and Benthesicymidae) made up 85% of the numbers and 79% of the biomass. In contrast, broadcast spawners made up only 23% of the numbers and 22% of the biomass in the bathypelagic zone. Further, the majority of the numbers and biomass of bathypelagic broadcast spawners were accounted for by only three species: Sergia splendens, Gennadas valens and Bentheogennema intermedia (18.5% of the numbers, 17.6% of the biomass). Of the three species, Sergia splendens and Gennadas valens were also prominent members of the mesopelagic com munity and are strong diel vertical migrators, leaving Bentheogennema intermedia as the only archetypal bathypelagic inhabitant that broadcasts its eggs. Consequently, the increased prominence of carideans and lophogastrids means the primary repro ductive strategy among bathypelagic 0% 20% 40% 60% 80% 100% MesopelagicBathypelagic% Biomass Sergestidae Benthesicymidae Oplophoridae Eucopiidae Other
26 species involves the brooding of eggs (Omori 1974; Mauchline 1980), and in some cases, may include semelparity (Childress and Price 1978). Th ere appears to be a similar trend in the eastern Pacific as Krygier and Pearcy (1981) found none of the dendrobranchiate shrimp species had distribution centers below 1000 m, while severa l carideans did. Likewise, Walters (1976) found that, among the penaeids, only Petalidium suspiriosum had a significant fraction of the population below 1000 m. It thus appears that amon g bathypelagic shrimp, there is a shift away from income breeding and towards capital breeding ( Stearns 1992). The prevalence of brooders in the bathypelagic zone is likely related to distance from productive surface waters. Dendrobranchiate shrimp hatch as nauplii and develop in epipelagic waters (Omori 1974). Eggs and early developmental stages are thus constrained to develop in shallow waters where they are exposed to numerous p redators and a patchy food supply for an extended period of time. Species that brood fewer but larger eggs hypothetically reduce risk to younger stages (Mauchline 1972; Bauer 2004) since t heir young hatch at a later stage of development (Aizawa 1974; Bauer 2004). As Omori (1 974) states, this abbreviated larval development, ...imparts to the larvae greater inde pendence from possible food shortages, greater swimming and feeding abilities, and greater safety from predators. The lophogastrids and mysids provide an even more extreme example by eliminating larval stages entirely and hatching as juveniles (Mauchline 1980). Apparently from the time of hatching, the young of bathypelagic shrimp are closer (developmentally/tem porally and spatially) to joining the adult population than their mesopelagic brethren. Zoogeography The lack of information on the bathypelagic fauna makes comparisons with other areas difficult. However, there are a handful of other p apers that address species composition. Two studies (Fasham and Foxton 1979; Hargreaves 1985) u sed principal component analysis to identify faunal groups in the eastern North Atlanti c, and both investigations were able to identify distinct species groups with bathypelagic distribut ions. Hargreaves (1985) found a faunal break
27 at 1000 m that appeared in decapods, euphausiids an d mysids. Below 1000 m, the numerically dominant decapods were the oplophorids Hymenodora gracilis and H. glacialis while the most numerous peracaridians were Eucopia unguiculata E. grimaldii E. australis Gnathophausia spp., and Boreomysis spp. Species with bathypelagic centers of distrib ution included: Acanthephyra pelagica, Hymenodora gracilis, H. glacialis, Sergia japonicus, Ephyrina bifida, Bentheogennema intermedia, Systellaspis braueri, E. grimaldii, E. sculpticauda, E. australis, Boreomysis microps, B. incisa, B. acuminata, and G. gigas The division between mesoand bathypelagic found by Fasham and Foxton (1979) was less clear. Looking only at decapods, they identified a total of fourteen faunal groups, with three centered below 1000 m. The species comprising these three groups were A. prionota A. curtirostris A. acutifrons Physetocaris microphthalma Meningodora miccyla Ephyrina bifida H. gracilis S. braueri A. stylorostratis Petalidium obesum B. intermedia and Sergestes submaximus Donaldsons (1975) examination of the sergestids in the western Atlantic included some sampling within the upper bathypelagic zone. As in both the mesoand bathypelagic zones in the EGOM, Sergia splendens was found to be the most abundant species. In add ition, several species had vertical distributions that extended be low 1000 m, including some common to our samples: S. splendens, S. japonica, and S. grandis. Of the species reported, S. japonica had the deepest vertical distribution. In the eastern Pacific, Krygier and Pearcy (1981) f ound twenty-nine species of decapods (Dendrobranchiata and Caridea) down to depths of 24 00 m. The assemblage was dominated by Sergestes similis a species with a shallow mesopelagic distribution that made up 94% of the individuals collected, while below 1000 m the most prevalent species were Hymenodora frontalis and H. gracilis Only five of their twenty-nine species had bathy pelagic centers of distribution. These included Parapasiphaea cristata, H. gracilis, H. glacialis, H. acanthitelsonis, Systellaspis braueri, and Acanthephyra sp. (an unknown species). Among the lophogastrids in the same area, Eucopia is reported as being the most abundant mesoto b athypelagic mysid in all the oceans, (Krygier and Murano 1988), and both E. sculpticauda and E. australis were reported as
28 having deep mesoto bathypelagic distributions. It should be noted Krygier and Murano considered it possible that E. australis, E. unguiculata, E. grimaldii and E. hanseni were part of a single species complex. They referred to them coll ectively as the E. australis complex, a practice that has not been followed here. A majority of the species collected in this study have been reported in more than one ocean basin (Table 4: Krygier and Pearcy 1981; Chac e 1986; Iwasaki and Nemoto 1987; Krygier and Wasmer 1988; Mller 1993; Wasmer 1993; Vereshch aka 1994; Prez-Farfante and Kensley 1997; Richter 2003). Of the ten most abundant shri mp species found in this study, only 2, Gennadas valens and Sergia splendens are restricted to the Atlantic Ocean. In additio n, only seven of the 42 species for which zoogeographical d ata was available are restricted to the Atlantic Ocean, and many species have distributions best described as circumtropical/subtropical, while a handful clearly appea r to be distributed globally (Table 4). Chace (1986) detailed ranges of the oplophorids and found several species ranging from tropical to subtropical areas of the Indian, Pacific and Atlant ic Oceans, including several found to be numerous in the EGOM such as Acanthephyra curtirostris, Hymenodora glacialis, and H. gracilis as well as the biomass dominant A. acutifrons According to M ller (1993), the distributions of all four species of Eucopia from this study are nearly pan global, as are thos e of all four species of Gnathophausia Although direct comparisons between different re gions were hindered by the scarcity of bathypelagic studies, it seems clear th at most of the bathypelagic shrimp species found in the EGOM have broad geographic distributio ns limited if at all, only by latitude.
29 Table 4: Zoogeography of bathypelagic shrimp species found i n the eastern Gulf of Mexico. Atlantic Pacific Indian Lophogastridae Gnathophausia gigas X X X Gnathophausia gracilis X X X Gnathophausia ingens X X X Gnathophausia zoea X X X Pseudochalaraspidum hanseni X Eucopiidae Eucopia australis X X X Eucopia grimaldii X X X Eucopia sculpticauda X X X Eucopia unguiculata X X X Benthesicymidae Bentheogennema intermedia X X X Gennadas bouvieri X X X Gennadas capensis X X X Gennadas valens X Sergestidae Sergia grandis X X Sergia japonica X X X Sergia regalis X X X Sergia splendens X Sergia wolffi X Pasiphaeidae Parapasiphaea sulcatifrons X X X Oplophoridae Acanthephyra acanthitelsonis X Acanthephyra acutifrons X X X Acanthephyra curtirostris X X X Acanthephyra exima X X X Acanthephyra gracilipes X Acanthephyra pelagica X X X Acanthephyra purpurea X Acanthephyra quadrispinosa X X X Acanthephyra stylorostratis X X X? Ephyrina benedicti X X X Ephyrina ombango X X X Hymenodora glacialis X X X Hymenodora gracilis X X X Janicella spinicauda X X X Meningodora marptocheles X X Meningodora miccyla X Meningodora mollis X X X Meningodora vesca X X X Notostomus gibbosus X X X Systellaspis braueri X X X Systellaspis cristata X X X Systellaspis debilis X X X Systellaspis pellucida X X X
30 Conclusion Almost half of the shrimps identified here were fo und in both the mesopelagic and bathypelagic zones. This included some, such as Gennadas valens, Eucopia unguiculata, E. sculpticauda and Sergia splendens, that were prominent in both zones. Thus, no abrup t faunal transition exists between the two zones. However, there was ample evidence to conclude that viewing the two zones as separate faunal assemblage s is useful and valid. 1) There were obvious shifts in the relative abunda nces of species in each zone. Some numerous bathypelagic species, such as Bentheogennema intermedia, Acanthephyra stylorostratis, Hymenodora glacialis, H. gracilis, and E. grimaldii were either absent from the mesopelagic assemblage, or p resent only in low numbers. 2) Species that were dominant in samples from below 1,000 m generally were not diel vertical migrators. 3) There was a high percentage of bathypelagic ende mism (~50%). 4) Important differences in community composition o ccurred at the family level, implying significant biological differences between the comm unities. Most importantly, there was a reduced contribution by those species that di sperse their eggs (Dendrobranchiata) in favor of those that brood the m (Caridea, Lophogastrida and Mysida). Considering this point, the biology of B. intermedia is of special interest as, despite its life history strategy, it is an importa nt, even characteristic member of the bathypelagic shrimp assemblage in the EGOM. 5) Bathypelagic shrimp species generally have circu mglobal distributions, suggesting assemblages from different ocean basins, but simila r latitudes, are composed of similar constituent species.
31 Chapter 2: The Bathypelagic Fishes of the Eastern G ulf of Mexico. Introduction This chapter considers fish, the second of the two principal components of the bathypelagic micronekton occurring in the eastern G ulf of Mexico (EGOM). The micronektonic fishes reported here were collected in midwater tra wls made below 1000 m in the vicinity of the USF Standard Station located at 27N 86W (~3400 m depth). The epiand mesopelagic zones in this region have been intensively studied over a period of several decades and the fish fauna is well known in terms of species composition, abundan ces, vertical distribution and trophic relationships (Hopkins and Lancraft 1984; Hopkins a nd Baird 1985b; Gartner et al. 1987; Hopkins and Gartner 1992; Sutton et al. 1996; Hopkins et al 1997). As mentioned previously, the mesopelagic fauna at this location is similar to ot her mesopelagic, low latitude, oceanic ecosystems (Hopkins and Gartner 1992; Hopkins et al 1994). The bathypelagic fish fauna, on the other hand, is poorly known, not only in the EG OM, but elsewhere; a result of the high costs and logistical difficulties of bathypelagic samplin g. The present chapter addresses abundances and verti cal distributions of micronektonic fishes taken from below 1000 m in the upper bathype lagic zone and briefly discusses their global oceanic distributions and species diversity pattern s. The underlying question being addressed with this study is: How unique is the bathypelagic fauna of the EGOM? As mentioned in the previous chapter, there are little ecological data available to support the division of the mesopelagic and bathypelagic zones. There is a rel ated lack of data with which to compare the bathypelagic faunas between localities. Backus et al. (1977), considering mesopelagic species, indicated that while no endemic myctophid species w ere found in the Gulf of Mexico, the region was unique and faunistically separable from other r egions of the western Atlantic. The question
32 remains: is the faunal composition of the bathypela gic zone unique as well or is the ecosystem structure in the EGOM similar to that found in othe r tropical-subtropical basins worldwide? The present chapter and the earlier one on bathype lagic shrimps cannot resolve the question. No other papers that quantitatively addre ss the structure of the micronekton community in any other bathypelagic region could be located. These chapters then provide the basis for future comparisons, which become possible when the bathypelagic zones of other regions are investigated. Methods The sampling location, trawl summary, and collecti on methods have been described in Chapter 1. The same methods were applied to the fi shes, with the following variations. Species numbers have been converted to biomass using relati onships between total or standard length and weight established from mesopelagic samples (Su tton and Hopkins, 1996; Torres and Donnelly, unpublished; Hopkins, unpublished; Sutton and Suntsov, unpublished). Whenever possible, regressions were generated using data obt ained from non-preserved material (Torres and Donnelly, unpublished). For some species, it w as necessary to weigh and measure preserved material. In such cases, measurements we re adjusted upward by 12% for length and 20% for weight (Gartner et al. 1988). Dry weights were obtained by adjusting for water content reported for the closest possible taxonomic group ( Childress and Nygaard 1973; Childress and Nygaard 1974; Donnelly and Torres 1988; Stickney an d Torres 1989; Childress et al. 1990). Some equations were applied to multiple genera (App endix B). In such cases, an attempt was made to match phylogeny and morphology as closely a s possible.
33 Results Assemblage Composition The species-discovery curve (Figure 9) indicates t hat sampling of the bathypelagic fish assemblage was not comprehensive, but extensive eno ugh to collect all but the rarest species. In total, 93 species from 38 families were identifi ed (Table 5), 19 of which are new records for the Gulf of Mexico. Some of the 19 records are also ne w for the western Atlantic. Included among these are the bathylaconid Bathylaco nigricans, the alepocephalid Herwigia kreffti the cetomimid Cetomimus hempeli the chiasmodontid Dysalotus oligoscolus the melamphaid Melamphaes eulepis and the neoscopelid Scopelengys tristis. One new record, the oneirodid Chirophryne xenolophus, is new to the Atlantic Ocean. Of the 37 families collected, only five, the Gonost omatidae, Myctophidae, Sternoptychidae, Melamphaidae, and Stomiidae, accou nted for at least 1% of the individuals collected. Total biomass of the fish assemblage wa s 23.7 kg DW km -2 This biomass was more evenly distributed taxonomically than abundance, wi th 12 families supplying at least 1% of the total, although the five families mentioned above d id account for 64.4%. The additional families contributing at least 1% of the biomass were the Ne michthyidae (24%), Cetomimidae (1.7%), Platytroctidae (1.5%), Neoscopelidae (1.5%), Bregma cerotidae (1.5%), Melanonidae (1.3%), and Bathylagidae (1.1%).
34 Species Count (Cumulative)Samples 0 50 100 150 135791113151719212325272931333537394143 2468101214161820222426283032343638404244 Figure 9: Number of fish species identified versus number of samples collected in the bathypelagic zone of the eastern Gulf of Mexico. The Gonostomatidae was the predominant family in e very measure save number of species. Nine species were identified. Eight were in the genus Cyclothone and the ninth was Sigmops elongatum (Table 5). Together the gonostomatids accounted fo r 87.2% of the fish numbers, and 42.2% of the biomass. Members of the family were present in 43 of the 46 catches, and four of the species were collected in more than half of the samples: Cyclothone pallida (84.8%), C. obscura (82.6%), C. acclinidens (54.3%), and C. braueri (52.2%). In addition to having high sample occurrence rates, most gonostomatids were relatively abundant. The family included the three most abund ant fish species: Cyclothone pallida (105,100 ind. km -2 ), C. obscura (95,300 ind km -2 ), and C. acclinidens (9100 ind. km -2 ); additionally there were a large number of Cyclothone specimens too damaged to be identified to species, making Cyclothone spp. the fourth most abundant fish taxon. Three o ther gonostomatids had
35 abundances greater than 1,000 ind. km -2 (Table 5): C. alba, C. braueri and C. pseudopallida The numbers of C. pallida and C. obscura were relatively high and resulted in these two spe cies alone combining for over three-quarters of all fish colle cted. Together, the various species of Cyclothone accounted for 41.3% of the fish biomass with Cyclothone obscura and C. pallida contributing more than one-third (38.7%). In spit e of its lower numbers, C. obscura made up a larger fraction of biomass than C. pallida due its larger size (mean standard lengths of 39.7 mm versus 30.7 mm, respectively). Cyclothone acclinidens, which had the 3 rd highest biomass within the family, had the sixteen th highest total among all fish species. The family Myctophidae contained the most species ( 17), and was the second most abundant family. Myctophids were commonly encounte red, with at least one individual occurring in 36 of 46 tows, and seven species occurring in gr eater than 10% of the samples: Ceratoscopelus warmingii, Lampanyctus alatus, Lepid ophanes guentheri, Notoscopelus resplendens, Lampanyctus nobilis, Nannobrachium lin eatum, and Taaningichthys paurolychnus With a combined abundance of 10,400 ind. km -2 myctophids accounted for ~4% of the fishes collected, and four species were found in numbers g reater than 1000 ind. km -2 : C. warmingii, L. alatus, L. guentheri, and N. resplendens The first two species were among the ten most abundant fishes, while the latter two as well as Nannobrachium lineatum were among the 20 most abundant. Despite comprising only 4% of the n umbers, the Myctophidae represented 12.3% of the fish biomass, the 3 rd highest contribution among families. Within the f amily, N. lineatum contributed the most biomass, and ranked 4 th among all fish species (4.0%). Three other species ( N. lineatum, N. resplendens, and C. warmingii ) were among the ten largest contributors to fish biomass, while L. nobilis was among the top 20. Almost half of the tows (47.8%) contained members of the Sternoptychidae, with the most commonly encountered species being Sternoptyx. pseudobscura (30.4% of tows) followed by S. diaphana (28.3% of tows). As a family, Sternoptychidae had the 3 rd highest abundance (2.9% of total) and 4 th highest biomass (4.9%). Sternoptyx pseudobscura was the most
36 numerous species within the family (Table 5), and t he most abundant fish outside the Gonostomatidae. This species had the 5 th highest biomass, representing 4.0% of the total. Sternoptyx diaphana was also among the 20 largest contributors of biom ass (1.0%). There were five identified species from the family Melamphaidae, which combined occurred in 56.5% of the tows. Melamphaids had the 4 th highest abundance (2.9%) and 5 th highest biomass (2.7%). Scopeloberyx robustus was the most abundant and common melamphaid and, other than species of Cyclothone was the only fish present in more than 50% of the samples. This species was also the 8 th most abundant fish overall (3100 ind. km -2 ) and had the 9 th highest biomass (1.6%). No other melamphaids were present in numbers greater than 1000 km -2 The second most abundant fish in the family was Scopelogadus mizolepis mizolepis (400 ind. km -2 ). There were 12 species identified from the family S tomiidae, second only to the Myctophidae in species richness. As a family, stom iids represented 1% of the numbers and 2.2% of fish biomass. Half of the tows contained stomii ds, although as individual taxa they tended to be encountered rarely (seven of the species were re presented by single specimens). The three most commonly encountered occurred in similar propo rtions in the samples: Chauliodus sloani (21.7%), Photostomias guernei (19.6%) and Stomias affinis (15.2%), which together accounted for about 75% of the numbers within the family, alt hough none were present in numbers greater than 1000 ind. km -2 Chauliodus sloani had the highest biomass within the family, account ing for 2.0% of the total and ranking 7 th among fish species.
37 Table 5 : Sample occurrence rates, abundances, and biomass of bathypelagic fish species in the eastern Gulf of Mexico. % occurrence Individuals km -2 % of numbers kg dry weight km -2 % biomass Alepocephalidae 15.9 0.2 0.8 Herwigia kreffti 2.2 <100 + + + Photostylus pycnopterus 8.7 300 0.1 0.2 0.7 Talismania antillarum 2.2 <100 + + + Anoplogasteridae 2.2 + 0.8 Anoplogaster cornuta 2.2 <100 + 0.2 0.8 Bathylaconidae 2.2 + + Bathylaco nigricans 2.2 <100 + + + Bathylagidae 13.6 0.2 1.1 Dolicholagus longirostris 13 500 0.2 0.3 1.1 Bregmacerotidae 31.8 0.5 1.5 Bregmaceros atlanticus 19.6 600 0.2 0.1 0.3 Bregmaceros ASH sp. 5 15.2 700 0.2 0.3 1.2 Centrophrynidae 2.2 + + Centrophryne spinulosa 2.2 <100 + + + Cetomimidae 18.2 0.2 1.7 Cetomimus hempeli 4.3 100 + + 0.1 Cetomimus sp. 2.2 <100 + + + Cetostoma regani 2.2 <100 + 0.4 1.5 Ditropichthys storeri 10.9 200 + + + Chiasmodontidae 13.6 0.1 0.3 Chiasmodon cf. 2.2 <100 + + + Chiasmodon niger 8.7 200 0.1 + 0.1 Dysalotus oligoscolus 2.2 <100 + + 0.1 Chloropthalmidae 6.8 0.1 + Chlorophthalmus agassizi 6.5 200 0.1 + + Evermannellidae 2.2 + + Odontostomops normalops 2.2 <100 + + + Gempylidae 2.3 + + Diplospinus multistriatus 2.2 <100 + + + Gigantactinidae 27.3 0.2 + Gigantactis longicirra 8.7 200 0.1 + + Gigantactis Male 2.2 <100 + + + Gigantactis sp. 10.9 200 0.1 + + Rhynchactis macrothrix 2.2 <100 + + + Rhynchactis sp. 4.3 100 + + +
38 Table 5 continued % occurrence Individuals km -2 % of numbers kg dry weight km -2 % biomass Giganturidae 4.5 0.1 0.2 Gigantura chuni 2.2 <100 + + + Gigantura indica 2.2 100 + 0.1 0.2 Gonostomatidae 93.2 87.2 42.2 Cyclothone acclinidens 54.3 9100 3.4 0.2 1.0 Cyclothone alba 23.9 1800 0.7 + 0.1 Cyclothone braueri 52.2 1500 1.7 0.1 0.2 Cyclothone microdon 4.3 200 0.1 + + Cyclothone obscura 82.6 95300 36.3 5.9 24.8 Cyclothone pallida 84.8 105100 40.0 3.3 13.9 Cyclothone parapallida 6.5 300 0.1 + + Cyclothone pseudopallida 28.3 2900 1.1 0.1 0.5 Cyclothone sp. 47.8 8000 3.0 0.2 0.8 Sigmops elongatum 13 300 0.1 0.2 0.9 Himantolophidae 2.2 + + Himantolophus (brevirostris) 2.2 <100 + + + Howellidae 6.5 0.1 0.1 Howella brodiei 6.5 200 0.1 + 0.1 Linophrynidae 15.2 0.1 + Linophryne (Rhizophryne) 15.2 400 0.1 + + Megalomycteridae 8.7 0.1 + Ataxolepis apus 8.7 200 0.1 + + Melamphaidae 59.1 1.6 2.7 Melamphaes eulepis 2.2 <100 + 0.1 0.3 Poromitra crassiceps 2.2 100 + 0.1 0.3 Scopeloberyx opisthopterus 10.9 300 0.1 + 0.1 Scopeloberyx robustus 54.3 3100 1.2 0.4 1.6 Scopeloberyx sp. 4.3 100 + + + Scopelogadus m. mizolepis 6.5 400 0.2 0.1 0.4 Melanocetidae 18.2 0.2 0.1 Melanocetus murrayi 4.3 100 0.1 + 0.1 Melanocetus sp. 13 300 0.1 + + Melanonidae 6.5 0.1 1.3 Melanonus zugmayeri 6.5 300 0.1 0.3 1.3
39 Table 5 continued % occurrence Individuals km -2 % of numbers kg dry weight km -2 % biomass Mirapinnidae 2.2 0.1 + Eutaeniophorus festivus 2.2 <100 + + + Moridae 2.2 0.1 + Physiculus fulvus 2.2 200 0.1 + + Myctophidae 72.7 4.0 12.3 Bolinichthys photothorax 2.2 <100 + + + Bolinichthys supralateralis 4.3 100 + + + Ceratoscopelus warmingii 32.6 3300 1.3 0.4 1.7 Diaphus mollis 4.3 100 + + + Diaphus splendidus 6.5 100 0.1 + + Lampadena anomala 2.2 <100 + + 0.1 Lampadena luminosa 2.2 <100 + 0.1 0.4 Lampanyctus alatus 26.1 1800 0.7 0.1 0.6 Lampanyctus sp. A 2.2 <100 + 0.1 0.4 Lampanyctus nobilis 13 400 0.1 0.2 1.0 Lampanyctus sp. 2.2 <100 + 0.1 0.2 Lepidophanes guentheri 32.6 1500 0.6 0.2 0.7 Nannobrachium lineatum 10.9 500 0.2 1.0 4.0 Notoscopelus resplendens 19.6 1500 0.6 0.5 2.1 Taaningichthys bathyphilus 6.5 100 0.1 + 0.2 Taaningichthys paurolychnus 15.2 400 0.2 + + Nemichthyidae 10.9 0.1 24.0 Avocettina infans 10.9 200 0.1 5.7 24.0 Neoscopelidae 2.2 + 1.5 Scopelengys tristus 2.2 <100 + 0.4 1.5 Notosudidae 4.3 + + Scopelosaurus mauli 2.2 <100 + + + Scopelosaurus sp. 2.2 <100 + + + Omosudidae 10.9 0.1 + Omosudis lowei 10.9 300 0.1 + + Oneirodidae 13.6 0.1 0.3 Chirophryne xenolophus 2.2 <100 + + + Dolopichthys sp. 4.3 100 + + + Lophodolos indicus 4.3 100 + 0.1 0.3 Ophidiidae 9.1 0.1 0.1 Bassozetus compressus 8.7 200 0.1 + 0.1 Paralepididae 2.2 + + Stemonosudis gracilis 2.2 <100 + + +
40 Table 5 continued % occurrence Individuals km -2 % of numbers kg dry weight km -2 % biomass Platytroctidae 29.5 0.4 1.5 Barbantus curvifrons 8.7 200 0.1 0.3 1.2 Mentodus facilis 10.9 400 0.1 + 0.1 Mentodus longirostris 10.9 200 0.1 0.1 0.2 Platytroctes apus 2.2 <100 + + + Rondeletiidae 2.2 + + Rondeletia bicolor 2.2 100 + + + Scopelarchidae 2.2 + + Scopelarchus analis 2.2 <100 + + + Sternoptychidae 45.5 2.9 4.9 Sternoptyx diaphana 28.3 1600 0.6 0.2 1.0 Sternoptyx pseudobscura 30.4 5500 2.1 0.9 4.0 Sternoptyx sp. 2.2 <100 + + + Stomiidae 47.7 1.0 2.2 Aristostomias xenostoma 2.2 <100 + + + Astronesthes niger 2.2 100 + + + Astronesthes sp. 2.2 <100 + + + Chauliodus sloani 21.7 700 0.3 0.5 2.0 Eustomias acinosus 4.3 100 + + + Eustomias dendriticus 2.2 <100 + + + Eustomias fissibarbis 2.2 <100 + + + Eustomias schmidti 2.2 <100 + + + Eustomias sp. 4.3 100 + + + Idiacanthus fasciola 2.2 <100 + + + Leptostomias bilobatus 2.2 <100 + + + Photostomias guernei 19.6 600 0.2 + + Photostomias megistius 2.2 <100 + + + Stomias affinis 15.2 700 0.3 + + Thaumatichthyidae 4.3 + + Thaumatichthys binghami 4.3 100 + + + valu es less than 0.1 are indicated by +
41 Although represented by only five specimens of Avocettina infans the Nemichthyidae exhibited a high biomass (Table 5) relative to the total assemblage. Only the Gonostomatidae had a higher biomass at the family level, and the b iomass of Avocettina infans was almost identical to that of C. obscura (5.7 versus 5.9 kg DW km -2 respectively). Several other species from various families were present in lower numbers but contributed a disproportionate amount of biomass. For example, six different species sup plied at least 1% of the biomass while numbering less than 10 specimens: Cetostoma regani (1 specimen), Scopelengys tristis (1 specimen), Melanonus zugmayeri (6 specimens), Barbantus curvifrons (5 specimens) and Lampanyctus nobilis (8 specimens). The large biomass contribution of relatively rare species was due to the low biomass of the entire assemblage coupled with the large size of these species. All six of the species in question had av erage standard lengths of 50 mm or more. The highest average standard length belonged to A. infans at 401 mm, while the two species represented by single specimens, C. regani and S. tristus, were both over 100 mm SL (182 and 165 mm, respectively). Vertical Distribution Analyzed together, fish were more abundant in the 1 000-1500 m depth zone than either depth zone below it (Figure 10; p < 0.05). The thi rteen species with abundances greater than 1000 ind. km -2 were then tested individually, and only Cyclothone pallida showed a significant abundance trend with depth. With all species consi dered together, the shallowest of the three depth intervals displayed a significantly higher ab undance than the two deeper intervals for this species.
42 1000-15001500-2000 >2000 8.99.910.911.912.913.914.9Log of Individuals km-2Depth (m) 1000-15001500-2000 >2000 8.99.910.911.912.913.914.9Log of Individuals km-2Depth (m) Figure 10: Vertical distribution of fish numbers between depth strata in the bathypelagic zone of the eastern Gulf of Mexico. Discussion Comparison With Other Studies Though few studies have been directed towards bathy pelagic sampling, some published reports are available for comparison, especially re garding the Gonostomatidae. A report by Wisner (1962) included four oblique tows that sampl ed in the bathypelagic zone of the eastern tropical Pacific, and Cyclothone acclinidens was the most abundant species in each. Badcock and Merrett (1976), working in the eastern North At lantic, conducted sampling that included the bathypelagic zone and, from the Gonostomatidae, the y reported five species with distributions that included both the mesopelagic and bathypelagic zones: C. braueri, C. microdon, C. pallida,
43 C. pseudopallida, and Sigmops elongatum (their Gonostoma elongatum ). They also reported C. obscura and S. bathyphilum (their Gonostoma bathyphilum ), both collected in very low numbers, as having strictly bathypelagic distributions. Exc epting the absence of S. bathyphilum from our samples, there is a high degree of accord between t he two studies in both the species present and their vertical distributions. Badcock (1970), sampling down to approximately 1000 m, presented data implying the vertical distributions of S. elongatum, C. braueri, C. pallida, and C. acclinidens extend into the bathypelagic zone. The absence of C. obscura from their samples suggests a strict bathypelagic distribution similar to what has been found in the EGOM, a deduction that is supported by Badcock and Merrett (1976) well as Miya and Nishida (1996). Regarding the Myctophidae, Kinzer and Schulz (1985) sampled down to 1250 m in the equatorial Atlantic and found most species to be me sopelagic; however, they did capture five species below 1000 m. In order of abundance, they were: Hygophum machrochir, Lepidophanes guentheri, Notoscopelus resplendens, Lampanyctus no bilis, and Ceratoscopelus warmingii Additionally, Kinzer and Schulz (1988) examined the Sternoptychidae in the equatorial Atlantic down to depths of 1500 m and found eight species, n one of which occurred below 1200 m. Only Sternoptyx pseudobscura the most abundant sternoptychid in Kinzer and Sch ulz (1988), was collected below 1000 m, and then in very low number s. Other data show that S. pseudobscura displays the deepest vertical distribution within t he family in the EGOM (Baird 1971; Hopkins and Baird 1985b).
44 Comparison with the EGOM Mesopelagic Assemblage Comparison of the data presented here with those of the EGOM mesopelagic zone (cf. Hopkins et al. 1997) reveals that the bathypelagic fish assemblage has ~7% the numbers and ~8% the biomass of the mesopelagic assemblage. The larger decline in abundance than in biomass is similar, if less dramatic, to that seen in the shrimp assemblage (Chapter 1). Both the mesopelagic and bathypelagic zones in the EGOM are characterized by the high species richness typical of low latitude ecosystems. Howeve r, the richness of the mesopelagic fish assemblage was much higher than that of the bathype lagic assemblage. A total of 134 primarily mesopelagic species have been reported in the EGOM (Hopkins and Lancraft 1984; Gartner et al. 1987; Sutton and Hopkins 1996a), while 93 were found in the present study. It should be noted that Figure 9 suggests more species would lik ely be encountered with more sampling. Based on diversity measures, the abundance of fish species is more evenly distributed in the mesopelagic zone than in the bathypelagic zone (J = 0.555 versus 0.374 respectively). This point is further illustrated by comparing species d ominance curves for the two zones (Figure 11), which highlights the increased contribution of the most abundant species in the bathypelagic zone. The question was posed regarding the distinctness o f the bathypelagic and mesopelagic assemblages. In other words, does the composition of the fish assemblage indicate that the bathypelagic zone should be considered a separate c ommunity, or is it simply a continuation of the pelagic community that spans the entire water c olumn? There are, in fact, some notable similarities between the mesopelagic and bathypelag ic fish assemblages.
45 Mesopelagic BathypelagicCumulative Dominance%Species rank 0 20 40 60 80 100 1101001000 Figure 11: Dominance comparison of fish species abundance betw een the mesopelagic and bathypelagic zones in the eastern Gulf of Mexico. Of the 93 species present in the bathypelagic sampl es, 63 have distributions that include both the mesopelagic and bathypelagic zones, and to gether, they accounted for significant percentages of both numbers (62%) and biomass (68.3 %) below 1000 m. Eleven of those species were myctophids, including three of the mos t abundant mesopelagic species, Ceratoscopelus warmingii, Lepidophanes guentheri, and Lampanyctus alatus (Gartner et al. 1987). All 13 stomiids recorded from the bathypela gic zone were represented in the mesopelagial, including the three most abundant spe cies in both zones: Photostomias guernei, Chauliodus sloani, and Stomias affinis Also present in the mesopelagic zone were the nemichthyid Avocettina infans which was an important biomass contributor below 1000 m, and Cyclothone pallida the most numerous species in both zones. Additio nally, of the 21 species accounting for at least 1% of the fish biomass, onl y six are not known in the mesopelagic zone in
46 the EGOM, and of the eight species accounting for a t least 1% of the numbers, only two ( C. obscura and Scopeloberyx robustus ) were found exclusively in bathypelagic zone. At higher taxonomic levels, the most abundant famil ies in both zones were the Gonostomatidae and Myctophidae, and both zones feat ured a similar prominence of the Sternoptychidae and Stomiidae. Sternoptychids had the 3 rd highest abundance, and 4 th highest biomass in both zones. Stomiids were the 5 th most abundant and 6 th largest contributors of biomass in the bathypelagic zone, while in the meso pelagic zone the family ranked 4 th in abundance and 3 rd in biomass. However, as in the shrimp assemblage, there were key differences between the two assemblages, supporting the idea of distinct communities. The most dramatic difference between fish assemblag es in the two zones was the change in relative importance between the Gonostoma tidae and Myctophidae. While gonostomatids were the most numerous in both zones, their relative numerical contribution increased from 59.5% to 87.3% below 1000 m. This w as due entirely to the prevalence of various species of Cyclothone as they accounted for 86.4% of the total numbers of bathypelagic fishes (Figure 12). Conversely, myctophids decreas ed from 29.0% of total mesopelagic numbers to 4.0% in the bathypelagic zone. Above 1000 m gon ostomatid biomass was lower than that of the myctophids (27.4% versus 37.0%; see Figure 13), but below 1000 m the biomass fraction from myctophids dropped to 12.5%, while that of gon ostomatids rose to 42.7%. Other dominant families showed marked changes in r elative importance between the two zones. The Stomiidae, which were 2.8% of the numbe rs and 20.8% of the biomass above 1000 m, dropped to 1.0% of the numbers and 2.9% of bioma ss in the bathypelagic zone (Figures 12 and 13), while the Phosichthyidae, 2.2% of numbers and 1.2% of biomass in the mesopelagic zone, were absent in the bathypelagic zone. The Me lamphaids showed an opposite trend, increasing from 1% of numbers and biomass in the me sopelagic zone to 1.6% and 3.6% of bathypelagic numbers and biomass respectively. Alt hough low, those percentages were high enough to raise melamphaids to the 4 th highest in numerical abundance and biomass below 1 000 m. The most noticeable change in family compositio n involved the Nemichthyidae, represented
47 by A. infans which contributed a minor proportion of mesopelag ic biomass, but was 24.3% of the bathypelagic total, despite the collection of only five individuals. The increased biomass fraction attributed to rare species was similar to that foun d in the shrimp assemblage (previous chapter), suggesting a more complete picture of the bathypela gic community may be gained by utilizing additional types of gear more suited to capturing l arger species. Fock et al. (2004) investigated the bathypelagic fish assemblage in the north Atlan tic using a large commercial trawl and although the family Gonostomatidae was still the do minant bathypelagic family, the resulting faunal list appeared to be significantly different than that reported here. In addition to the shifts in relative importance of several families, the bathypelagic samples contained several species not collected in the mesopelagic zone. Of the 93 fish species identified in this study, 32 (~34%) were not found above 1000 m. In general, those deep-living species were present in low numbers, with only two present in abundances greater than 1000 ind. km -2 ; however, one ( Cyclothone obscura ) was the chief biomass constituent. The high percentage of species with bathypelagic centers of distribution, in combination with changes in the relative importance of families, mirrored resul ts seen in the shrimp assemblage (Chapter 1), and support the idea of two separate communities. There was a distinct difference in the reproductive ecology of the EGOM mesopelagic and bathypelagic shrimp assemblages (Chapter 1). A similar trend could not definitively be found between the fish assemblages, due at least in part to a lack of information regarding many of the groups. It is additionally possible the small gear used in this study sampled large adults poorly. Childress et al. (1980) suggested bathypelagic spec ies tend to be semelparous, while mesopelagic species tend to be iteroparous. A stud y of three species of Cyclothone with three different maximum depths of abundance (350, 450 and 600 m) revealed delayed reproduction and increased fecundity as depth increased (Miya an d Nemoto 1991). Fecundity values at maturity for the three species were 255, 590 and 99 0, being positively correlated with depth distribution.
48 0% 20% 40% 60% 80% 100% MesopelagicBathypelagic Gonostomatidae Myctophidae Sternoptychidae Stomiidae Melamphaidae Other Figure 12: Relative abundance of fish families between the mes opelagic and bathypelagic zones in the eastern Gulf of Mexico. 0% 20% 40% 60% 80% 100% MesopelagicBathypelagic Gonostomatidae Myctophidae Sternoptychidae Stomiidae Melamphaidae Other Figure 13: Relative biomass of fish families between the mesop elagic and bathypelagic zones in the eastern Gulf of Mexico.
49 Zoogeography The broad distributions recorded for most of the sp ecies found in this study suggests that classical zoogeographic boundaries do not apply in the bathypelagic zone. Of the bathypelagic species for which sufficient zoogeographic informat ion is available, only ten were recorded in a single ocean basin (Table 6: Badcock and Baird 1979 ; Miya and Nishida 1996; Sutton and Hopkins 1996a; Miya and Nishida 1997; McEachran and Fechhelm 1998; McEachran and Fechhelm 2006) and two of those had their ranges ex tended by this study. In contrast, ~75% of the species in our data set have been found in the Atlantic, Pacific and Indian Oceans, with the most typical distribution being circumtropical. Wh ile there is strong evidence that surface water characteristics affect bathypelagic community compo sition (e.g. Fock et al. 2004), it appears bathypelagic species have broad distributions that tend to be limited mainly by latitude (if at all). This is assuming that there is not a high degree of cryptic species present in the deep-sea, which Miya and Nishida (1997) raised as a possibility.
50 Table 6 : Worldwide occurrence of fish species found in the bathypelagic zone of the eastern Gulf of Mexico. Atlantic Pacific Indian Nemichthyidae Avocettina infans X X Bathylagidae Dolicholagus longirostris X X Alepocephalidae Herwigia kreffti X X X Photostylus pycnopterus X X X Talismania antillarum X X X Bathylaconidae Bathylaco nigricans X X X Platytroctidae Barbantus curvifrons X X Mentodus facilis X X X Mentodus longirostris X X Platytroctes apus X X X Gonostomatidae Cyclothone acclinidens X X X Cyclothone alba X X X Cyclothone braueri X X X Cyclothone microdon X X X Cyclothone obscura X X X Cyclothone pallida X X X Cyclothone parapallida X X Cyclothone pseudopallida X X X Sigmops elongatum X X X Sternoptychidae Argyropelecus aculeatus X X X Sternoptyx diaphana X X X Sternoptyx pseudobscura X X X Stomiidae Aristostomias xenostoma X X X Astronesthes niger X Chauliodus sloani X X X Eustomias acinosus X Eustomias dendriticus X Eustomias fissibarbis X X X Eustomias schmidti X X X Idiacanthus fasciola X X X Leptostomias bilobatus X Photostomias guernei X X X Stomias affinis X X X Giganturidae Gigantura chuni X X X Chloropthalmidae Chlorophthalmus agassizi X X X Notosudidae Scopelosaurus mauli? X X X Scopelarchidae Scopelarchus analis X X X Evermannellidae Odontostomops normalops X X X Omosudidae Omosudis lowei X X X Paralepididae Stemonosudis gracilis X X X
51 Table 6 continued Atlantic Pacific Indian Myctophidae Bolinichthys photothorax X X X Bolinichthys supralateralis X X X Ceratoscopelus warmingii X X X Diaphus mollis X X X Diaphus splendidus X X X Lampadena anomala X X X Lampadena luminosa X X X Lampanyctus alatus X X X Lampanyctus nobilis X X X Lepidophanes guentheri X Nannobrachium lineatum X X X Notoscopelus resplendens X X X Taaningichthys bathyphilus X X X Taaningichthys paurolychnus X X X Ophidiidae Bassozetus compressus X X Bregmacerotidae Bregmaceros atlanticus X X X Bregmaceros macclellandi X X X Melanonidae Melanonus zugmayeri X X X Moridae Physiculus fulvus X Centrophrynidae Centrophryne spinulosa X X X Gigantactinidae Gigantactis longicirra X X Rhynchactis macrothrix X X Himantolophidae Himantolophus (brevirostris) X Melanocetidae Melanocetus murrayi X X X Oneirodidae Chirophryne xenolophus X Lophodolos indicus X X X Thaumatichthyidae Thaumatichthys binghami X Melamphaidae Melamphaes eulepis X X X Poromitra crassiceps X X X Scopeloberyx opisthopterus X X X Scopeloberyx robustus X X X Scopelogadus m. mizolepis X X X Cetomimidae Cetomimus hempeli X Cetostoma regani X X X Ditropichthys storeri X X X Megalomycteridae Ataxolepis apus X Mirapinnidae Eutaeniophorus festivus X X X Rondeletiidae Rondeletia bicolor X Anoplogasteridae Anoplogaster cornuta X X X Howellidae Howella brodiei X X Chiasmodontidae Chiasmodon niger X X X Dysalotus oligoscolus X X X Gempylidae Diplospinus multistriatus X X X
52 Conclusion About two-thirds of the fish species identified in this study have been recorded from both above and below 1,000 m, including several prominen t species such as Cyclothone pallida, Ceratoscopelus warmingii, Chauliodus sloani and Sternoptyx pseudobscura However, it is concluded that viewing the two zones as separate fa unal communities is useful and valid based on the following evidence. 1) The species with the highest biomass, C. obscura, is absent from the mesopelagic assemblage. 2) There were significant faunal changes at the fam ily level, most notably, the increased prominence of the Gonostomatidae coupled with the d ecreased prominence of the Myctophidae and Stomiidae. 3) A large percentage of fish species identified (3 4%) were not present in mesopelagic samples. 4) The mesopelagic fish assemblage exhibited more s pecies with more even abundance distributions among them. 5) The bathypelagic zone is characterized by signif icant biomass contributions from large, rare species, suggesting substantial differe nces in how energy is transferred through the system. Like the bathypelagic shrimps, the fish species ge nerally have circumglobal distributions, suggesting assemblages from different ocean basins, but similar latitudes, are composed of similar constituent species. The broader implicati ons of this are that the community structure of the bathypelagic EGOM may closely resemble that of other low latitude systems, and that there is decoupling with processes occurring in the surface layer.
53 Chapter 3: Community and Trophic Structure of the B athypelagic Micronekton in the Eastern Gulf of Mexico Introduction In the two previous chapters, aspects of the ecolo gy of micronektonic shrimps and fishes in the bathypelagic zone of the eastern Gulf of Mex ico (EGOM) were considered. Included were data on shrimp and fish species composition, abunda nce, biomass, vertical distribution, zoogeography, and reproductive strategies, as well as comparisons of the bathypelagic and overlying mesopelagic zones. The present chapter c ombines the shimp and fish data for a more complete look at the community structure and then a nalyzes trophic relationships through diet analysis. In the mesopelagic and bathypelagic zone comparisons, the bathypelagic zone was distinguished by lower abundance and biomass, incre ased biomass fractions attributable to rare species, high levels of endemism (50% for shrimp, 3 6% for fish), different species dominance patterns, different family dominance patterns, a re duced number of diel migratory species, and a significant shift in predominant reproductive strat egies within the shrimp. Together, those characteristics suggest the mesoand bathypelagial are two distinct communities with different ecological structures. The reduced incidence of diel vertical migration am ong bathypelagic residents equates to reduced access to mesoand epipelagic waters where plankton biomass is higher (Angel and Baker 1982; Hopkins 1982; Vinogradov 1997; Yamaguch i et al. 2005), and thus, raises the question of how non-migrating bathypelagic species obtain their food. Bathypelagic nonmigrators have two possible sources of nutrition. They can rely on the passive sinking of dead material from surface waters or, as Vinogradov (196 2) proposed, subsist on species with access to more productive depths in a ladder of vertical migration. There were four requirements
54 proposed for the operation of the latter mechanism: (1) prevalence of predatory species in deepwater (2) significant quantity of filter feeders th at feed within the surface layers on a regular basi s and co-occur with the predators (3) low percentage of detritus feeders (4) increased average size of deep-sea animals. With this in mind, the diets of dominant bathypelagic species in the EGOM were examined with the goals of investigating the f easibility of Vinogradovs ladder of vertical migration, and defining the main trophic pathways within the bathypelagic zone. Methods Organisms used in diet analysis came from the seri es of discrete-depth trawls used for community analysis. Collection of these samples was described in Chapter 1. In this section, the data from shrimp and fish were combined for the pur poses of presenting an overview of the dominant components of the bathypelagic micronekton ic assemblage. Animals used in diet analysis were measured to the nearest millimeter before removal of their digestive tracts. The entire tract was remov ed from fishes and, the entire fore and midgut was excised from shrimps, as was the posterior port ion of the intestine through at least abdominal segment three. Contents of the digestive tracts we re spread out on a glass slide in a mixture of acid fuchsin and glycerol, then examined at 40-600X magnifications. When possible, individuals from a species were dissected until a curve of prey taxa versus individuals analyzed became asymptotic. Diet items were identified to the lowest possible taxonomic level and, whenever possible, measured. For a positive occurrence of fish to be recorded, the presence of some material other than scales, such as bone or eye lenses, was requir ed. Occurrence of entire prey items within the digestive tract of shrimp was rare, and general ly, some digestion-resistant portion of the prey item had to be measured. In such cases, relationsh ips between part size and the overall size of the prey species (unpublished data) were applied to obtain a prey size. In cases where taxon specific relationships were unavailable, an analogo us relationship from the closest possible taxon
55 of similar morphology was applied. The presence of detritus and nematocysts was often recorded but could not be measured in a form conver tible to biomass. The use of regressions to establish overall size w as complicated for some groups. For example, there is variation in lens size relative t o overall body length among different species of fish, even within the same genus (e.g. Cyclothone ). In such cases, the regression used for converting to prey size represented an average acro ss two or more species. For some organisms, such as some polychaetes, relationships could not be established. For example, no relationship between setae length and body length f or Pelagobia longicirrata was determined. In these instances, prey size was based on an average size of the organism in the environment (Hopkins unpublished). In cases for which positive prey identifications were possible without the presence of any measurable body part, the average s ize of the same prey from within the same species/size predator group was used. After assign ing a size to every possible prey item, they were placed into one of 18 size categories for clus ter analysis: in 2-mm increments up to 20-mm, 5-mm increments up to 30-mm, and one group >30-mm. Having established or estimated a size for all pos sible prey items, regressions were used to convert prey size to biomass (unpublished data). In many cases, these were the same equations used to establish organism dry weight fro m the trawl samples. Diet taxa were grouped into 13 broad categories: calanoid copepods, cephal opods, chaetognaths, decapods, euphausiids, fish, gastropods, hyperiids, lophogast rids, miscellaneous crustaceans, non-calanoid copepods, ostracods, and polychaetes. Results, in terms of prey number, size, and biomass, were run through cluster analysis in Primer E (version 5.2.9) using the Bray-Curtis dissimilarit y index. In every case, groups were differentiated u sing the 40% dissimilarity level.
56 Results Overall Community Structure The total biomass of the bathypelagic micronekton assemblage in the EGOM was estimated to be 53.4 kg DW km -2 While fishes accounted for 62.2% of the numbers of organisms collected, slightly more of the biomass (56%) was a ttributed to shrimp. Six families contributed at least 5% of the biomass, while a total of 12 famili es accounted for at least 1% of the total biomass (Table 7). The various species within the family Oplophoridae contributed the highest biomass fraction (33.2% of total) despite comprising only 9 .6% of the total numbers (Table 7). Five species within the family were among the ten highes t biomass contributors, while only two were among the ten most abundant (Table 8). Within the family, the two largest biomass values were from large species present in low numbers ( Acanthephyra acutifrons 8.7% of total biomass and Notostomus gibbosus 6.1% of total biomass). In terms of numbers, gon ostomatids were the most prevalent organisms collected. Together, they accounted for 54.4% of all individuals (Table 7), mainly due to the large numerical contributions of Cyclothone pallida (24.9% of total numbers) and C. obscura (22.6% of total numbers) (Table 8). Both species were about five times as abundant as the next most abundant species, Eucopia sculpticauda Despite being numerically dominant, as well as containing the principal bioma ss species in C. obscura (11% of total biomass), the gonostomatids ranked 2 nd in terms of familial biomass (Table 7), accounting for 18.7% of the total.
57 Table 7: Familial composition, in terms of abundance and bi omass, of the bathypelagic micronekton in the eastern Gulf of Mexico. Family % of Total Biomass % of Total Numbers Oplophoridae c 33.2 9.6 Gonostomatidae f 18.7 54.4 Nemichthyidae f 10.6 0.1 Benthesicymidae c 10.0 7.2 Eucopiidae c 7.3 17.4 Myctophidae f 5.5 2.4 Sergestidae c 3.1 2.6 Sternoptychidae f 2.2 1.7 Melamphaidae f 1.2 1.0 Pasiphaeidae c 1.0 0.2 Lophogastridae c 1.0 0.3 Stomiidae f 1.0 0.6 Cetomimidae f 0.8 0.1 Platytroctidae f 0.7 0.2 Neoscopelidae f 0.7 + Bregmacerotidae f 0.7 0.3 Melanonidae f 0.6 0.1 Bathylagidae f 0.5 0.1 Anoplogasteridae f 0.4 + Alepocephalidae f 0.4 0.1 Oneirodidae f 0.2 0.1 Chiasmodontidae f 0.1 0.1 Giganturidae f 0.1 + Ophidiidae f 0.1 0.1 Bresiliidae c + 0.2 Howellidae f + + Melanocetidae f + 0.1 Pandalidae c + 0.1 Rondeletiidae f + + Gigantactinidae f + 0.1 Scopelarchidae f + + Chloropthalmidae f + + Megalomycteridae f + 0.1 Moridae f + + Mirapinnidae f + + Evermannellidae f + + Thaumatichthyidae f + + Linophrynidae f + 0.1 Gempylidae f + + Omosudidae f + 0.1 Centrophrynidae f + + Himantolophidae f + + Paralepididae f + + Mysidae c + 0.2 Values < 0.1 are noted by + f denotes fish families c denotes crustacean families
58 The Benthesicymidae had the 4 th highest abundance (7.2% of total) and biomass (10. 0% of total) among the families (Table 7). The abunda nce contribution was almost equally shared between two species, Bentheogennema intermedia (3.1% of numbers) and Gennadas valens (2.3% of numbers), while the biomass contribution w as chiefly due to B. intermedia (7.0% of the total) (Table 8). The 2 nd most abundant family was the Eucopiidae (17.4% of total numbers, Table 7); however, due to their small size, the four species within the family collectively accounted for only 7.3% of the total biomass, the 5 th highest total among families. All four species we re among the ten most abundant, but only the largest, Eucopia australis was among the ten highest biomass contributors (Table 8). The Myctophidae, which contributed 5.5% of the over all biomass and 2.4% of total numbers, had the 6 th highest total in both categories (Table 7). Withi n the family, Nannobrachium lineatum contributed the largest biomass fraction (1.8%), a nd was the only myctophid among the twenty highest biomass species (Table 8). All spec ies of myctophids taken contributed less than 1% to the total numbers, and only Ceratoscopelus warmingii was among the 20 most abundant species overall. The sergestid shrimps were taken in similar numbers to the myctophids (2.6% of total) and were the 5 th most numerous family (Table 7). None of the Serge stidae were among the ten most abundant species, and only Sergia splendens was among the top 20 (Table 8). In terms of biomass, no sergestids reached 1% of the overall to tal, and as a family they ranked 7 th collectively contributing 3.1% of the total.
59 Table 8 : Percent contribution of the twenty largest contri butors of abundance and biomass in the bathypelagic zone of the eastern Gulf of Mexico. % of Numbers % of Biomass Cyclothone pallida f 24.9 Cyclothone obscura f 11.0 Cyclothone obscura f 22.6 Avocettina infans f 10.6 Eucopia sculpticauda c 4.8 Acanthephyra acutifrons c 8.7 Eucopia australis c 4.8 Bentheogennema intermedia c 7.0 Eucopia grimaldii c 4.6 Cyclothone pallida f 6.1 Bentheogennema intermedia c 3.1 Notostomus gibbosus c 6.1 Hymenodora glacialis c 3.1 Acanthephyra curtirostris c 3.2 Eucopia unguiculata c 3.0 Acanthephyra stylorostratis c 3.1 Acanthephyra stylorostratis c 2.9 Eucopia australis c 2.9 Gennadas valens c 2.3 Acanthephyra acanthitelsonis c 2.4 Cyclothone acclinidens f 2.1 Hymenodora glacialis c 2.3 Sergia splendens c 1.3 Gennadas valens c 1.9 Sternoptyx pseudobscura f 1.3 Nannobrachium lineatum c 1.8 Cyclothone braueri f 1.1 Sternoptyx pseudobscura f 1.8 Acanthephyra curtirostris c 1.0 Eucopia grimaldii c 1.8 Gennadas capensis c 0.9 Eucopia sculpticauda c 1.5 Hymenodora gracilis c 0.9 Ephyrina benedicti c 1.4 Ceratoscopelus warmingii f 0.8 Eucopia unguiculata c 1.1 Scopeloberyx robustus f 0.7 Ephyrina ombango c 1.1 Cyclothone pseudopallida f 0.7 Meningodora mollis c 1.1 f denotes fish species; c denotes crustacean species Only three other families accounted for more than 1 % each of the overall biomass (Table 7): Nemichthyidae (10.6%), Sternoptychidae (2.2%) a nd Melamphaidae (1.2%). The Nemichthyidae had the 3 rd highest biomass within the community despite being represented by only five specimens of Avocettina infans. Conversely, the abundance ranks of the Sternoptychidae and Melamphaidae were comparable to that of their biomass (Table 7). Other micronekton groups were numerically negligibl e. As mentioned in the first chapter, euphausiids and amphipods were rare and generally f ell into the plankton size range. Cephalopods, although very rare in the quantitative samples (Appendix C), likely contributed significant biomass; however, lack of reliable regr essions prohibited an estimation of their biomass (Vecchione, pers. comm.).
60 Diet Diet analysis was performed on 850 specimens from 14 species belonging to the five most important families: Oplophoridae, Benthesicymi dae, Eucopiidae, Sternoptychidae, and Gonostomatidae. The species were chosen for analys is based on their relative abundance and biomass in the community. Together these species r epresented 78.5% of the numbers and 58.5% of the community biomass, while their respect ive families accounted for 90.2% of the numbers and 71.4% of the biomass. Avocettina infans was not analyzed due to the poor condition of the specimens adversely affecting an a lready small sample size, but the family is thought to strictly consume micronektonic crustacea ns. The proportion of empty digestive tracts varied widely between species. Five were empty les s than 10% of the time: Sternoptyx pseudobscura (0%), Acanthephyra curtirostris (4%), A. stylorostratis (5%), Notostomus gibbosus (8%) and Bentheogennema intermedia (4%), while the highest proportions of empty guts were found in two species of Eucopia : E. grimaldii (73%) and E. unguiculata (61%). Four species were empty about half of the time: Cyclothone obscura (45%), C. pallida (52%), E. australis (41%), and E. sculpticauda (45%), while the three remaining species, Gennadas valens, Hymenodora glacialis and A. acutifrons were empty 21%, 13%, and 13% of the time, respecti vely. As mentioned in the methods, it was not possible t o estimate biomass of detrital or cnidarian material in the digestive tracts although its occurrence suggested it was an important dietary component in some species. Cnidarian mater ial, usually in the form of nematocysts, occurred in at least 40% of individuals of Bentheogennema intermedia Gennadas valens, and Hymenodora glacialis It was less common in Acanthephyra stylorostratis (25%), A. curtirostris (19%), and A. acutifrons (13%), while all remaining species had less than a 10% occurrence of nematocysts. Detrital material was prevalent in B. intermedia occurring in 65% of individuals examined, and common in G. valens (40% of individuals). Acanthephyra stylorostratis was the only other species in which detritus appeared in mo re than 10% of the individuals (12%), and eight of the species examined showed no evidence of such material.
61 Cluster analysis of diet biomass resulted in four feeding clusters (Figure 14). Cluster A included both species of Cyclothone examined, and was characterized by a diet of small planktonic crustaceans, specifically calanoid copep ods and ostracods, which made up 62% and 25% of the diet, respectively. Other diet taxa inc luded polychaetes (5.5%), miscellaneous crustaceans (4.5%), chaetognaths (2.1%), and non-ca lanoid copepods (<1%). Cluster B, which contained three of the four species of Eucopia was characterized by a copepod dietary fraction o f 91.5%. Other prey categories included chaetognaths (4.7%), ostracods (3.7%) and non-calanoid copepods (<1%). Cobsra Cpall Eaust Egrim Eungc Spsud Esclp Hglac Ngibb Aacut Acurt Astylo Binter Gvlns 806040200 Dissimilarity Cluster A Cluster B Cluster C Cluster D Cobsra Cpall Eaust Egrim Eungc Spsud Esclp Hglac Ngibb Aacut Acurt Astylo Binter Gvlns 806040200 Dissimilarity Cluster A Cluster B Cluster C Cluster D Figure 14: Cluster results for diet composition based on prey biomass. In contrast to the first two clusters, clusters C and D contained species for which most of the biomass was attributed to fish. Cluster C cont ained only Sternoptyx pseudobscura, which had a diet that was composed of 60% fish but, with the presence of ten different diet categories, was diverse overall. Hyperiid amphipods contribute d the second highest diet percentage (23.6%), followed by decapods (6.8%) and euphausiid s (5.4%). All other categories contributed
62 less than 5% to the diet biomass. Containing eight of the 14 species, cluster D was the largest, and except for the lophogastrid, Eucopia sculpticauda was composed entirely of decapods. Together, species within the cluster consumed 12 of the 13 prey categories, but fish made up 73.7% of the diet. Only two other categories accou nted for more than 5% of the diet: decapods (9.3%) and calanoid copepods (7.2%). The analysis for prey size resulted in seven clust ers, although four of them (A,B,C and E) consisted of a single species (Figure 15). Similar to the results for diet biomass, the first four clusters were all characterized by diets consisting of small prey items (<10 mm). The three species in Cluster D had 70.6% of their biomass con tained within the two size bins from 2-3.9 mm. The prey-size distribution of Cluster C, Eucopia sculpticauda was similar to that of Cluster D, with 57.4% of the biomass between 2 mm and 4.9 m m; however, it differed in that there was a second biomass peak in the 12-13.9 mm size range. This was due to one individual consuming a single fish, a datum that impacted both prey compos ition and prey size clusters, and which will be examined later in the discussion. The peak prey si zes of E. grimaldii which alone comprised Cluster A, were shifted upward in size so that 79.8 % of its diet biomass was between 4 mm and 5.9 mm. The last of the planktivorous species, Cyclothone obscura which alone composed Cluster B, also fed on small particles, but its pre y-size distribution was more broadly distributed than the other three planktivorous groups with 85% of its diet being spread between 2 mm and 7.9 mm. The remaining three clusters (E, F, and G) all con tained species whose prey biomass was primarily derived from items larger than 10 mm. Of the three, Cluster G ( Acanthephyra acutifrons and Sternoptyx pseudobscura ) took the largest prey items. Together, the mode of their prey size was in the largest possible category (>30 mm) and accounted for 37.1% of their prey biomass. The oplophorid shrimp, Notostomus gibbosus was in its own cluster (F), and 79.7% of its prey biomass fell within 18-29.9 mm in length. Finally, Cluster E, which was the largest group, contained those species whose prey length was prima rily between 14 mm and 19.9 mm.
63 Egrim Cobsra Esclp Cpall Eaust Eungc Ngibb Aacut Spsud Gvlns Hglac Acurt Astylo Binter 806040200 Dissimilarity Cluster DCluster A Cluster B Cluster C Cluster E Cluster F Cluster G Egrim Cobsra Esclp Cpall Eaust Eungc Ngibb Aacut Spsud Gvlns Hglac Acurt Astylo Binter 806040200 Dissimilarity Cluster DCluster A Cluster B Cluster C Cluster E Cluster F Cluster G Figure 15: Cluster results for diet composition based on prey size. Discussion Mesoand Bathypelagic Community Comparison Total estimated biomass of the mesopelagic microne kton assemblage is between 380 and 430 kg DW km -2 depending on the source. The lower estimate is o btained by multiplying the wet weight measurement by 0.15 (Hopkins and Lancraf t 1984). The higher figure comes from adding estimates and measurements of individual gro ups. Using these results, the bathypelagic micronekton biomass is 12-14% that in the mesopelag ic zone. The list of families making up 2% or more of the b iomass is similar in both zones (Table 9), with seven families common to both lists (Benth esicymidae, Eucopiidae, Gonostomatidae, Myctophidae, Oplophoridae, Sergestidae, and Sternop tychidae). The only difference is the replacement of the Stomiidae with the Nemichthyidae (a family represented by only five
64 individuals of Avocettina infans ; see previous chapter). What distinguishes the tw o lists is the relative importance of those families. Of the seven families common to both lists in Tabl e 9, only two contributed a higher proportion of biomass below 1000 m than above. The Eucopiidae, only 2% of the mesopelagic biomass, increased to 7% of the bathypelagic biomas s. More dramatically, the Oplophoridae increased from 7% to 33% of the micronekton biomass and became the dominant biomass family. Three of the seven families decreased in r elative importance: Benthesicymidae, Myctophidae, and Sergestidae (Table 9), while the t wo remaining families, Gonostomatidae and Sternoptychidae, remained about the same. It is no teworthy that the biomass fraction due to gonostomatids remained constant because, in terms o f abundance, the family increased in importance from 34% in the mesopelagic zone to 54% in the bathypelagic zone. The constant biomass fraction in the face of an increase in rela tive abundance was likely a reflection of the drastic decrease in the abundance of Sigmops elongatum a species that is much larger than Cyclothone spp. Table 9: Comparison of biomass distribution between prominan t families in the mesopelagic and bathypelagic zones of the eastern Gulf of Mexico. Mesopelagic % of biomass Bathypelagic % of biomass Benthesicymidae 25 Oplophoridae 33.2 Gonostomatidae 20 Gonostomatidae 18.7 Myctophidae 10 Nemichthyidae 10.6 Sergestidae 10 Benthesicymidae 10.0 Stomiidae 9 Eucopiidae 7.3 Oplophoridae 7 Myctophidae 5.5 Sternoptychidae 2 Sergestidae 3.1 Eucopiidae 2 Sternoptychidae 2.2 Other 15 Other 9.4
65 Sources of Bias in Diet Data Of major concern in diet analysis is sample contam ination due to net feeding. While studies using similar types of gear have shown that net feeding was a minor source of error, at least within fishes (Hopkins and Baird 1975; Lancra ft and Robison 1980), it is of special concern in this study due to the extended tow times employe d. Three lines of evidence argued against significant net feeding. The first involved the co ndition of the organisms upon retrieval. The live recovery of bathypelagic organisms was very rare, e ven on the few occasions when a closing cod end was attached to the net. This was not surprisi ng given the turbulence within the cod end in combination with the watery and delicate structure of many bathypelagic species. Individuals of Eucopia spp., for example, were usually recovered with at least some damage to their delicate gnathopods. A second line of evidence was the diet differences evident between species. The turbulent environment of the cod end is a mixture o f prey items equally available to all species of micronekton, and it is thus likely that predators f eeding on this mixture would have very similar diets. To invoke diet differences under conditions of net feeding one must assume different groups are feeding selectively and distinctively wi thin the cod end. While interspecific differences in the amount of net feeding have been demonstrated (Lancraft and Robison 1980), interspecific differences in selective feeding have not. For exa mple, hyperiid remains were recovered in only two species: Notostomus gibbosus (onlyoneoccasion), and Sternoptyx pseudobscura (found in 74% individuals examined). To obtain this result u nder conditions of significant net feeding would require that S. pseudobscura consume hyperiids within the cod end while all oth er species ignore them. The third line of evidence against net feeding cam e from a comparison between the contents of stomachs and intestines. The premise w as that organisms feeding in the net would have noticeably different diet compositions between stomachs and intestines. Based on the diet items recorded from each segment, this was not the case. In 85% of the species, the most
66 numerous diet categories found in both the stomachs and intestines were identical. Further, in 75% of the cases, each section of the digestive tra ct had two of the top three diet categories in common. In some cases, a significant number of the diet items recorded were found in the intestine, suggesting the animals had not fed in so me time. The most extreme examples of this were found in Cyclothone pallida and Eucopia unguiculata for which ~80% of the items recorded were found in the intestine. Similarly, the other three species of Eucopia as well as C obscura had ~ 50% or more of their diet items contained in the intestine. Another uncertainty inherent diet analysis is the estimation of prey biomass. A bias exists towards prey items that are resistant to dig estion. Furthermore, when examining the diets of species that masticate their food, such as shrim p, only pieces of animals are available for identification. Even prey from species that swallo w items whole can be fragmented if the prey is in an advanced state of digestion. Biomass estimat ion is best in groups where it is possible to obtain high taxonomic resolution, accurate measurem ents of body parts that can be reliably related to overall prey size, and accurate regressi ons relating size to biomass. In many cases, all three of these criteria cannot be met. The followi ng discussion highlights groups that proved problematic, specifically fishes, lophogastrids, ce phalopods, polychaetes, cnidaria and detritus. With fishes and lophogastrids, it is difficult to o btain high taxonomic resolution. Generally, the eye lenses provide the most reliable measurements for fish while lophogastrids ( Eucopia spp.) were identified by one of three body parts: m andibles, the telson, or terminal segments of the gnathopods. While the first two pa rts of lophogastrids provided reliable size-tolength data, both the gnathopods of Eucopia spp as well as the lenses of fish were less reliable due to variation in size relative to body length. The resulting average lengths of fish and lophogastrid prey were 18.4 mm SL and 19.1 mm TL, r espectively. Previous diet studies from the mesopelagic zone in the EGOM found the average size of ingested fish to be 35% smaller than that reported here (12 mm SL according to Hopkins e t al. 1994). This does not necessarily imply an overestimation of prey size in this study, howev er, as both the average size of available prey and potential predators was larger below 1000 m.
67 Two other problematic diet categories were polychae tes and cephalopods. In both cases, it was impossible to relate identifiable par ts to overall body size, even in polychaetes where high taxonomic resolution was possible. Esti mates of cephalopod contribution to diet (a diet category that appeared only in Notostomus gibbosus ; see Table 10) were hampered by both the inability to achieve high taxonomic resolution, and lack of reliable relationships between size and biomass (Vecchione, personal communication). T he biomass results for this diet category, in particular, were considered dubious; however, its o ccurrence was unique to N. gibbosus (Table 10). Finally, there was no way to estimate the biomass c ontribution due to detritus and cnidaria, and thus, these two diet categories could not be included in the cluster analysis. Detritus, or marine snow, is a catch-all term refer ring to a diverse array of material including, but not limited to, dinoflagellate remains, tintinnids, diatom tests, phytoplankton resting stages, mucous, and unidentifiable greenish-brown material. Although the occurrence of those categories was relatively minor in most species (Ta ble 10), it was prominent in enough to preclude resolving interspecies diet differences. Three species within the Oplophoridae contained cnidarian material more than 10% of the t ime: Acanthephyra acutifrons (12.5%), A. curtirostris (18.9%), and A. stylorostratis (25.4%). Additionally, A. stylorostratis also contained detritus in 11.9% of the individuals examined. The inability to quantify this type of material was most important in the cases of Bentheogennema intermedia and Gennadas valens ., the latter of which has previously been shown to graze heavily on marine snow (Heffernan and Hopkins 1981). Both cnidarian material and detritus appear ed to be significant for both of these shrimp species (Table 10), suggesting they habitually graz e on marine snow. This is further supported by the fact that non-calanoid copepods, such as Oncaea spp., Oithona spp. and harpacticoids, copepod groups often associated with marine snow pa rticles (Alldredge 1972; Steinberg et al. 1994; Green and Dagg 1997), had the highest occurre nce rates in the diets of these two species.
68 In particular, the high occurrence of detritus in t he diet of B. intermedia (65.5%) had the potential to enhance the distinction of this species diet, a s only calanoid copepods appeared more frequently (74.5% of the individuals). Diet Composition Cluster results for diet composition generally fel l along phylogenetic lines: Cluster A contained both species of gonostomatids, Cluster B three of the four species of eucopiids, Cluster C the sternoptychid, and Cluster D was primarily ma de up of decapods (Figure 14). The one exception to this trend was Eucopia sculpticauda which appeared in Cluster D among the decapods due to the consumption of one fish by a si ngle individual. The result was that the diet of E. sculpticauda most closely resembled that of Hymenodora glacialis the smallest of the decapods examined, and the one with the smallest po rtion of fish in its diet. Running the analysis with this individual removed caused E. sculpticauda to cluster with its congeners. Roe (1984) found the diet of E. unguiculata was numerically dominated by copepods, and did not report fish material. However, the appearance of fish in the d iets of E. australis and E. unguiculata is documented (Hopkins et al. 1994), and it thus appea rs that members of the genus feed primarily on calanoid copepods, occasionally supplementing wi th fish One prominent feature of both the prey biomass and prey size analyses was the clear split creating two super-groups in each dendogram ( Figures 14 and 15). Again, Eucopia sculpticauda was exceptional in that it switched from one super -group to the other. While the phylogenetic trend was not as clear in the analysis of prey size, no other species switched supergroups, indicating that the Gonostomatidae and Euco piidae preyed primarily on small crustaceans. The Oplophoridae, Sternoptychidae and Benthesicymidae had diverse diets, but consumed larger prey items, and were primarily pisc ivorous.
69 Table 10: Diet composition results for fourteen species of ba thypelagic micronekton from the eastern Gulf of Mex ico. % of diet biomass Individuals analyzed % Empty % occurrence of cnidaria % occurrence of detritus Calanoida Ostracoda Euphausiacea Lophogastrida Hyperiidea Decapoda Misc. Crustacean Gastropoda Cephalopoda Chaetognatha Fish Eucopiidae 73.8 1.8 4.5 19.9 Eucopia australis 68 41.2 1.5 7.4 89.3 5.6 5.0 Eucopia grimaldii 60 13.3 3.3 93.9 5.6 Eucopia sculpticauda 71 45.1 2.8 57.6 4.3 38.1 Eucopia unguiculata 57 61.4 97.9 2.1 Benthesicymidae 11.6 1.1 2.6 1.0 2.5 81.0 Bentheogennema intermedia 55 3.6 40.0 65.5 11.8 4.5 1.7 2.5 78.0 Gennadas valens 53 20.8 47.2 41.5 11.3 2.6 85.0 Oplophoridae 5.5 1.0 0.1 4.9 0.1 11.2 0.2 2.8 1.9 72.0 Acanthephyra acutifrons 8* 12.5 12.5 0.2 8.4 21.6 0.1 69.7 Acanthephyra curtirostris 53 3.8 18.9 5.7 4.4 0.3 2.9 12.0 0.3 2.5 77.5 Acanthephyra stylorostratis 59 5.1 25.4 13.6 4.2 0.6 1.9 7.8 0.4 1.1 82.6 Hymenodora glacialis 83 13.3 3.6 3.6 25.3 8.3 8.8 6.4 51.0 Notostomus gibbosus 13* 7.7 7.7 7.2 0.6 4.7 0.5 0.6 23.5 2.1 60.7 Gonostomatidae 62.1 25.3 4.5 2.1 Cyclothone obscura 127 45.7 71.0 16.4 5.9 2.7 Cyclothone pallida 97 51.5 32.6 54.8 0.3 Sternoptychidae 1.3 Sternoptyx pseudobscura 35 1.3 0.1 5.4 23.7 6.8 0.3 + + 60.5 indicates all available specimens were dissected + indicates value <0.1
70 Also prominent was the greater reliance on fish by decapods in the bathypelagic zone relative to those in the mesopelagic zone. Hopkins et al. (1994) found the average size of fish eaten in their samples was 12 mm and subsequently i nserted the dry weight equivalent of a 12mm Cyclothone whenever a fish was recorded in the gut contents. In the present study, fish biomass was estimated for each individual occurrenc e. While the methods used to estimate prey biomass were slightly different, the major portion of the increase is likely a result of the size difference of the prey given that the average lengt h of fish consumed in this study was 18 mm. A comparison with mesopelagic data (Hopkins et al. 1994) suggests much of the difference was due to the replacement of chaetognat h and euphausiid biomass with fish (Table 11). In both cases, the drop in biomass due to cha etognaths and euphausiids almost exactly equaled the percentage increase due to fish. This is perhaps not surprising given the results of Hopkins (1982) as well as Kinsey and Hopkins (1994) in which chaetognaths and Stylocheiron resided primarily in the upper mesopelagic to epipe lagic zones. Biomass contributions to diet from all other groups remained similar in Table 11, excepting the percentage of other in the diets of Benthesicymidae. The difference here was due to the contribution of radiolarians in the mesopelagic study (9.7%), a group not encountered i n great numbers in this study, but a likely byproduct of consumption of marine snow, a diet com ponent also noted within the group. The reduced presence of some groups in the bathypelagic zone was offset by increased consumption of fish by bathypelagic decapods relative to their mesopelagic counterparts.
71 Table 11: Percentage contribution of major prey gro ups to the diets of two decapod families in the mesopelagic and bathypelagic zones of the eastern Gulf of Mexico. Benthesicymidae Oplophoridae Mesopelagic Bathypelagic Mesopelagic Bathypelagic Calanoida 13.3 11.6 6.7 5.5 Cephalopoda 0.0 0.0 4.1 2.8 Chaetognatha 13.3 2.5 23.3 1.9 Decapoda 0.0 2.6 9.0 11.2 Euphausiacea 27.3 0.0 17.9 0.1 Fish 31.7 81.0 33.2 72.0 Ostracoda 0.0 1.1 0.0 1.0 Other 14.4 1.2 5.8 5.2 Data from past studies relating vertical distribut ion of shrimps to the occurrence of fish in diets are equivocal. Fish were consumed little to none at all in sergestids in the EGOM (Flock and Hopkins 1992), and there was no trend for the d eeper living of the two genera, Sergia to consume a greater proportion of fish. Similarly, W alters (1976), reported no fish in the diets of Pacific sergestids. However, in the Atlantic, a st udy of the same family did indicate that Sergia consumed more fish than Sergestes (Donaldson 1975). Furthermore, in their examinati on of the mesopelagic shrimp assemblage in the EGOM, Hopkins et al. (1994) found the highest percentage of fish in the diet of Gnathophausia ingens a species with a deep mesoto bathypelagic vertical distribution. The decapod sp ecies had a diet consisting of 12.7% to 57.3%, but the species also examined in this study average d 39% above 1000 m, while all other decapods averaged 30%. In addition, Roe (1984) loo ked at the diets of seven decapod species in the Atlantic, including two species of Acanthephyra ( A. purpurea and A. pelagica ) with differing vertical distributions. No effort was made to conv ert prey to biomass, but an examination of the data reveals that the occurrence of fish between th e species differed in that the deeper of the two species, A. pelagica (Foxton 1972), displayed the higher incidence of fi sh in its diet. The strongest evidence suggesting a more piscivorous di et in deeper living decapods is the data presented in this study.
72 The diets of some species of Cyclothone have been previously examined, and results here agree well with earlier findings. The diet of the deep mesoto bathypelagic species, C. acclinidens, was investigated (DeWitt and Cailliet 1972) and id entifiable diet categories included (in ascending order of frequency), amphipods, ostra cods, chaetognaths, and copepods. In their extensive study of mesopelagic fish diets, Hopkins et al. (1997) found copepods comprised the largest fraction of the diets in Cyclothone spp., resulting in 11 of the 12 species/size-class grouping together in a guild characterized by > 72% copepod diets. Also, like bathypelagic individuals, mesopelagic members of the genus had a high percentage of empty guts (Sutton, personal communication). The diet composition of Sternoptyx pseudobscura was unique. Other than the increase in the proportion of fish consumed, these results a re similar to previous data (Hopkins and Baird 1985b; Kinzer and Schulz 1988). While 61% of the c onsumed biomass came from fish, it is the more minor elements of its diet that were noteworth y. To begin, it was the only species with significant fractions of euphausiids (chiefly Stylocheiron ) and hyperiids (chiefly Platyscelidae) in its diet. Additionally, the diet was unique in the occurrence of alciopid polychaetes, crab megalopae, and a gastropod. Finally, of the dietar y copepods, the principal taxon consumed was Candacia pachydactyla ; also one individual contained ten pontellid copep ods. Both taxa are associated with epipelagic (or, in the case of pont ellids, neustonic) waters. Sternoptyx pseudobscura is known to be a non-migrating member of the deep mesoto bathypelagic community (Baird 1971; Hopkins and Baird 1985b; Kin zer and Schulz 1988; Hopkins et al. 1997), and the paradox involving its consumption of shallo w water prey was noted previously. The occurrence of such taxa in the diet of fish collect ed within the bathypelagic zone reinforces existing data, but adds nothing to the explanation. The presence, however, of pontellids in only one of 35 individuals examined in this study is per haps indicative of reduced access to shallow water fauna in the bathypelagic residents of the sp ecies.
73 Table 12: Copepod taxa occurring in the diets of bathypelagic micronekton expressed as percentage of total number of copepods. All E. australis E. grimaldii G. valens A. acutifrons A. curtirostris A. stylorostratis H. glacialis N. gibbosus C. obscura C. pallida S. pseudobscura Augaptilidae 2.6 6.3 4.4 2.2 Euaugaptilus sp. 2.0 3.1 25.0 3.3 4.2 Haloptilus sp. 0.6 2.2 7.7 Heterorhabdus sp. 2.0 2.2 2.9 3.3 3.8 Lucicutia sp. 1.2 2.2 7.7 Metridiidae 7.3 10.9 4.4 8.8 8.9 Metridia sp. 2.3 1.6 25.0 6.7 Pleuromamma sp. 6.1 7.8 8.9 2.9 6.7 7.7 3.8 Candacia sp. 10.8 2.4 4.7 13.3 26.5 3.3 37.5 Labidocera sp. 0.3 4.2 Eucalanidae 5.5 4.8 44.4 4.7 4.4 5.9 1.1 7.7 2.6 4.2 Rhincalanus cornutus 15.9 21.9 25.0 8.9 8.8 20.0 23.1 20.5 4.2 Rhincalanus nausutus 1.7 25.0 15.4 5.1 4.2 Rhincalanus sp. 2.0 3.1 3.3 3.8 Spinocalanidae 0.3 2.6 Aetideidae 12.5 23.8 10.9 6.7 8.8 8.9 7.7 19. 2 8.3 Chirundina sp. 0.3 2.6 Euchirella sp. 1.2 7.7 2.6 8.3 Euchaetidae 1.2 2.2 2.9 1.1 7.7 Unid. Calanoid 24.2 69.0 55.6 25.0 35.6 32.4 35.6 23.1 48.7 69.2 25.0 n 42 9 64 4 45 34 90 13 39 26 24 Vertical Distribution of Dietary Copepods Ideally, the identification of prey items to speci es would allow discernment of the depth at which predators were typically feeding. Unfortunat ely, as a result of the mastication of prey by shrimp species and the tendency of diet items in Cyclothone often to be well-digested having been recovered from the intestinal tract, species l evel taxonomic resolution was often impossible. Among the prey groups identified during diet analys is, taxonomic resolution was greatest within the copepods. These data are valuable as the verti cal distributions of copepods are well
74 documented in the EGOM and Atlantic, thus providing evidence concerning the depth zone at which species are feeding, and the list of diet tax a (Table 12) is generally populated by genera and families that often have deep mesoto bathypel agic distributions, Of the copepods identified, Rhincalanus cornutus was consumed in highest numbers (15.9% of all copepods) and was the most abundant t axon found in six of the 14 species. The prevalence of this single species is especially rem arkable given that many of the taxa in Table 12 represent consolidations of multiple species. Rhincalanus cornutus has a broad deep mesoto bathypelagic distribution in the Atlantic and EGOM (Grice and Hulsemann 1965; Roe 1972b; Deevey and Brooks 1977), with Deevey and Brooks rep orting that its relative contribution to copepod numbers peaked below 1000 m where it accoun ted for 0.8-1.2% of numbers in the Sargasso Sea. Anecdotally, the abundance of this s pecies appeared to be extremely high in several of the nested net bathypelagic plankton sam ples (personal observation). Among the nine families represented in Table 12, a survey of literature addressing their vertical distributions reveals a majority of them c ontain several species whose distributions extend, or are limited, to the bathypelagic zone. For example, the Aetideidae and Augaptilidae were found to increase in relative importance in th e bathypelagic zone of the Pacific (Arashyevich 1972). Furthermore, in the Atlantic, over 50% of t he species belonging to the Aetideidae, Augaptilidae, Lucicutiidae, and Spinocalanidae have distributions that can be characterized as either deep mesoto bathypelagic, or bathypelagic (Grice and Hulsemann 1965; Deevey and Brooks 1977). Within the Metridiidae, Metridia spp. tended to occur deeper than Pleuromamma spp. (Grice and Hulsemann 1965; Roe 1972a; Deevey a nd Brooks 1977), with Metridia reported as the most abundant calanoid genus below 1000 m (D eevey and Brooks 1977). Exceptions to the prevalence of deeply occurring ta xa in Table 12 were the genera Labidocera Candacia and Pleuromamma Labidocera which occurred in a single individual of S. pseudobscura has the shallowest vertical distribution of the t hree. As previously mentioned, this neustonic (Hopkins unpublished) copepod has of ten been reported in the diet of Sternoptyx pseudobscura (Hopkins and Baird 1985b). Although the vertical distribution of Candacia does
75 extend down to at least 1000 m, the genus primarily occurs in the epito upper mesopelagic zones (Grice and Hulsemann 1965; Arashyevich 1972; Roe 1972a; Deevey and Brooks 1977; Yamaguchi et al. 2002). In spite of this, it occur red in the diet of eight species, and was prominent in Bentheogennema intermedia, Acanthephyra curtirostri s, A. stylorostratis and S. pseudobscura Regarding Pleuromamma, research in the mesopelagic zone of the EGOM repeatedly found the genus to be an important compo nent in the diets of planktivorous species, disproportionately so relative to its occurrence in the water column (Hopkins and Baird 1985a; Hopkins and Gartner 1992; Hopkins et al. 1997). In fact, Hopkins et al. (1997) determined the genus accounted for 40% of the copepod biomass cons umed by the mesopelagic fish assemblage, and its reduced dietary prominence in t he bathypelagic zone may be a reflection of the upper-mesopelagic distribution of some members of the genus (Grice and Hulsemann 1965; Roe 1972a; Bennett and Hopkins 1989). Primary Trophic Pathways Although data on gut evacuation rates are necessary to calculate daily ration, it is possible to infer information about the rations of bathypelagic animals using previous estimates from mesopelagic species and respiration data from mesoand bathypelagic species. At the low end, non-migratory bathypelagic fish species were e stimated to have a daily ration of 0.68% by Childress et al. (1980). Hopkins et al. (1997) use d the total undigested prey biomass recovered from mesopelagic fish as a daily ration for piscivo res and a minimum ration for planktivores. Their method resulted in an estimated daily ration of 1.8-3.6% of standing stock for the entire mesopelagic fish assemblage (Hopkins et al. 1997). Furthermore, based on similar weightspecific metabolic rates between shrimp and migrato ry myctophids, Hopkins et al. (1994) suggested a daily ration of 6% for mesopelagic shri mp species. A lower estimate of daily ration for shrimp of 1-2% is acquired by combining energy content values provided by Donnelly et al. (1993b) with an equation relating oxygen consumptio n rate to minimum depth of occurrence (temperature considered) (Donnelly and Torres 1988) Bailey et al. (1995) measured respiration
76 of the mesopelagic nemichthyid Serrivomer beanii, and employing their numbers yields a daily ration estimate of approximately 1.5%. Smith and L aver (1981) provide an estimate of 2% for Cyclothone acclinidens based on in situ metabolic measurements at 1300 m; however, these latter three estimates are based on metabolism only and do not account for growth or excretion, which Brett and Groves (1979) estimated at approxim ately 56% of the energy budget for carnivorous fish. Despite the organisms considered here violating two of the conditions provided by Brett and Groves (1979), that the fish were fed a surplus ration and temperatures were not extreme, as well as the fact that we are considerin g shrimp as well as fish, doubling the rations based on metabolism provides a very rough estimate of 2-4%, which is similar to ration estimates provided by Hopkins et al. (1997). Therefore, base d on these various sources, it is seems likely that the daily ration of most bathypelagic species is less than 4%, and probably averages approximately 2%. The predatory impacts of the primary families are shown in Figure 16 A (mesopelagic) and B (bathypelagic). Data for the mesopelagic zon e was compiled from Hopkins et al. (1994; 1997). A comparison of the two figures suggests in the mesopelagic zone the shrimp had a much larger impact than the fish, while the two assembla ges were more balanced below 1000 m. Doubtless, this is due in part to the differing met hodologies employed. Within the mesopelagic zone, fish daily rations were estimated from gut co ntents while it was assumed the overall ration for the shrimp community was 6%. For bathypelagic calculations, all groups were assumed to have daily rations of 2%. It is therefore prudent to limit comparisons to the distribution of impact between families within their respective assemblage s. In the mesopelagic shrimp assemblage, the Benthesic ymidae had by far the greatest trophic impact, its total being three times greater than the next closest family, Sergestidae. The Oplophoridae had the next greatest impact, meaning in terms of energy flow, the three predominant families were all decapods, which colle ctively accounted for an estimated 95% of crustacean impact. In the bathypelagic zone, a sli ghtly different picture emerges. The largest
77 mesopelagic player, Benthesicymidae, is reduced to second and replaced in prominence by the Oplophoridae. Furthermore, the lophogastrid family Eucopiidae figured more prominently. Among the fish families, the trophic impact within the mesopelagic zone was more evenly distributed than among the shrimp families, but the principal families were Myctophidae, Sternoptychidae, and Stomiidae. Within the bathype lagic zone, the Gonostomatidae is estimated to have the greatest impact, followed by the Nemich thyidae. The impact of the Myctophidae was approximately equal to that of the large grouping o f various families that mainly contained rare species.
78 0 50 100 150 200 250 300 350 400 Lophogastridae Euc opi id a e Bent h e s icym ida e Se r ge s t i da e Oplophoridae Misc D e capod G onos tomatidae My c tophi da e S te rnoptychid a e Stomiidae Nemichthyi da e Misc Fish 0 1000 2000 3000 4000 5000 6000 7000 Lophogastridae Euc o piid a e Benthesicymidae S e rge s t i da e Opl ophori da e M i s c Decapod Gon os tom a tid a e My ctop hi da e Sternoptychidae Stom ii da e Nemichthyida e M is c Fish mgDWkm-2day-1mgDWkm-2day-1A B 0 50 100 150 200 250 300 350 400 Lophogastridae Euc opi id a e Bent h e s icym ida e Se r ge s t i da e Oplophoridae Misc D e capod G onos tomatidae My c tophi da e S te rnoptychid a e Stomiidae Nemichthyi da e Misc Fish 0 1000 2000 3000 4000 5000 6000 7000 Lophogastridae Euc o piid a e Benthesicymidae S e rge s t i da e Opl ophori da e M i s c Decapod Gon os tom a tid a e My ctop hi da e Sternoptychidae Stom ii da e Nemichthyida e M is c Fish mgDWkm-2day-1mgDWkm-2day-1A B Figure 16: Trophic impact of major families in the mesopelagic (A) and bathypelagic (B) zones in the eastern Gulf of Mexico.
79 Conclusions This goal of this study was to address the lack of basic information regarding the community structure of the most prevalent ecosystem on Earth, the bathypelagic zone. Utilization of similar gear and the presence of a thorough data set from overlying waters facilitated a valuable comparison with other mid-water micronekto nic assemblages. The results presented here provide a clear picture of the micronektonic c omponent of this ecosystem by providing a thorough representation of the species present, the ir relative importance in terms of numbers and biomass, and trophic relationships involving a majo rity of the dominant species. Among the important finding of this study are: 1) The wide zoogeographical distributions found amo ng most species suggest that the structure and function of this system is likely to be similar to other bathypelagic systems at similar latitudes. 2) Species that are rare but large in size play a m ore important role in the cycling of energy relative to the mesopelagic zone. 3) The list of dominant families is similar to that from the mesopelagic zone; however, the bathypelagic zone is distinguished by increased importance of some families (such as Gonostomatidae, Oplophoridae, Nemichthyida e, and Eucopiidae) and decreased importance of others (Myctophidae, Serges tidae, Benthesicymidae, and Stomiidae). 4) These taxonomic shifts are indicative of shifts in reproductive strategies, at least between the two crustacean assemblages. 5) Species analyzed for diet fell into two basic ca tegories: planktivores ( Cyclothone spp. and Eucopia spp.) and piscivores (Decapods). Among the decapo ds, there is a greater reliance on fish as a source of food compar ed to their mesopelagic counterparts.
80 6) Community composition and diet results were gene rally supportive of the requirements of Vingradovs ladder of vertical migr ation such as that the community was composed primarily of predatory species larger than their mesopelagic counterparts and with access to mesopelagic migrato rs.
81 References Aizawa Y (1974) Ecological studies of micronektonic shrimps (Crustacea, Decapoda) in the western North Pacific. Bull. Ocean. Res. Inst. Univ Tokyo 6: 1-84 Alldredge AL (1972) Abandoned larvacean houses: a u nique food source in the pelagic environment. Science 177: 885-887 Angel MV, Baker AdC (1982) Vertical distribution of the standing crop of plankton and micronekton at three stations in the northeast Atla ntic. Biol. Oceanogr. 2: 1-30 Angel MV, Hargreaves PM, Kirkpatrick P, Domanski P (1982) Low variability in planktonic and micronektonic populations at 1000m depth in the vic inity of 42N 17W; evidence against diel migratory behavior in the majority of species. Biol. Oceanogr. 1: 287-319 Arashyevich YG (1972) Vertical distribution of trop hic groups of copepods in the boreal and tropical regions of the Pacific ocean. Oceanology 1 2: 265-274 Backus RH, Craddock JE, Haedrich RL, Robison BH (19 77) Atlantic mesopelagic zoogeography. In: Robert H. Gibbs J (ed) Sears Foundation for Mar ine Research. Yale Univ., New Haven, pp 266-287 Badcock J (1970) The vertical distribution of mesop elagic fishes collected on the SOND cruise. J. Mar. Biol. Assoc. UK 50: 1001-1044 Badcock J, Baird R (1979) Remarks on systematics, d evelopment, and distribution of the hatchetfish genus Sternoptyx (Pisces, Stomiatoidei) Fish. Bull. 77: 803-820 Badcock J, Merrett NR (1976) Midwater fishes in the eastern North Atlantic. I. Vertical distribution and associated biology in 30N, 23W, with developmen tal notes on certain myctophids. Prog. Oceanogr. 7: 3-58 Bailey DM, Bagley PM, Jamieson AJ, Cromarty A, Coll ins MA, Tselepidis A, Priede IG (2005) Life in a warm deep sea: routine activity and burst swim ming performance of the shrimp Acanthephyra eximia in the abyssal Mediterranean. Mar. Biol. 146: 1199 -1206
82 Bailey TG, Youngbluth MJ, Owen GP (1995) Chemical c omposition and metabolic rates of gelatinous zooplankton from midwater and benthic bo undary layer environments off Cape Hatteras, North Carolina, USA. Mar. Ecol. Prog. Ser 122: 121-134 Baird RC (1971) The systematics, distribution, and zoogeography of the marine hatchetfishes (Family Sternoptychidae). Bull. Mus. Comp. Zool. Ha rvard Univ. 142: 1-128 Bauer RT (2004) Remarkable Shrimps: Adaptations and Natural History of the Carideans. University of Oklahoma Press, Norman Bennett JL, Hopkins TL (1989) Aspects of the ecolog y of the calanoid copepod genus Pleuromamma in the eastern Gulf of Mexico. Cont. Mar. Sci.: 119 -136 Brett JR, Groves TDD (1979) Physiological Energetic s. In: Hoar WS, Randall DJ, Brett JR (eds) Fish Physiology. Academic Press, New York, pp 280-3 52 Chace FA (1986) The Caridean shrimps (Crustacea: De capoda) of the Albatross Philippine Expedition, 1907-1910, Part 4: families Oplophorida e and Nematocarcinidae. Smith. Cont. Zool. 432: 1-82 Childress JJ, Nygaard M (1973) The chemical composi tion of midwater fishes as a function of depth of occurrence off southern California. Deep S ea Res. 20: 1093-1109 Childress JJ, Nygaard M (1974) Chemical composition and buoyancy of midwater crustaceans as function of depth of occurrence off Southern Califo rnia. Mar. Biol. 27: 225-238 Childress JJ, Price MH (1978) Growth rate of the ba thypelagic crustacean Gnathophausia ingens (Mysidacea: Lophogastridae). I. Dimensional growth and population structure. Mar. Biol. 50: 47-62 Childress JJ, Price MH, Favuzzi J, Cowles D (1990) Chemical composition of midwater fishes as a function of depth of occurrence off the Hawaiian Islands: food availability as a selective factore. Mar. Biol. 105: 235-246 Childress JJ, Taylor SM, Cailliet GM, Price MH (198 0) Patterns of growth, energy utilization and reproduction in some mesoand bathypelagic fishes off southern California. Mar. Biol. 61: 27-40
83 Deevey GB, Brooks AL (1977) Copepods of the Sargass o Sea off Bermuda: species composition, and vertical and seasonal distribution between the surface and 2000 m. Bull. Mar. Sci. 27: 256-291 DeWitt FAJ, Cailliet GM (1972) Feeding habits of tw o bristlemouth fishes, Cyclothone acclinidens and C. signata (Gonostomatidae). Copeia 1972: 868-871 Donaldson DM (1975) Vertical distribution and feedi ng of sergestid shrimps (Decapoda: Natantia) collected near Bermuda. Mar. Biol. 31: 37-50 Donnelly J, Stickney DG, Torres JJ (1993a) Proximat e and elemental composition and energy content of mesopelagic crustaceans from the Eastern Gulf of Mexico. Mar. Biol. 115: 469480 Donnelly J, Stickney DG, Torres JJ (1993b) Proximat e and elemental composition and energy content of mesopelagic crustaceans from the Eastern Gulf of Mexico. Mar. Biol. 115: 469480 Donnelly J, Torres JJ (1988) Oxygen consumption of midwater fishes and crustaceans from the eastern Gulf of Mexico. Mar. Biol. 97: 483-494 Fasham MJR, Foxton P (1979) Zonal distribution of p elagic decapoda (Crustacea) in the eastern north Atlantic and its relation to the physical oce anography. J. exp. mar. Biol. Ecol. 37: 225-253 Flock ME, Hopkins TL (1992) Species composition, ve rtical distribution, and food habits of the sergestid shrimp assembledge in the eastern Gulf of Mexico. J. Crust. Biol. 12: 210-223 Fock HO, Pusch C, Ehrich S (2004) Structure of deep -sea pelagic fish assemblages in relation to the Mid-Atlantic Ridge (45 o -50 o N). Deep Sea Res. 51: 953-978 Foxton P (1970) The vertical distribution of pelagi c decapods (Crustacea: Natantia) collected on the SOND cruise 1965. I. The Caridea. J. Mar. Biol Assoc. UK 50: 939-960 Foxton P (1972) Observations on the vertical distri bution of the genus Acanthephyra (Crustacea: Decapoda) in the eastern North Atlantic, with parti cular reference to species of the purpurea' group. Proc. R. S. E. (B) 73: 301-313
84 Gartner JV, Conley WJ, Hopkins TL (1988) Escapement by fishes from midwater trawls: a case study using lanternfishes (Pisces: Myctophidae). Fi sh. Bull. 87: 213-222 Gartner JV, Hopkins TL, Baird RC, Milliken DM (1987 ) The lanternfishes (Pisces: Myctophidae) of the eastern Gulf of Mexico. Fish. Bull. 85: 81-98 Green EP, Dagg MJ (1997) Mesozooplankton associatio ns with medium to large marine snow aggregates in the northern Gulf of Mexico. J. Plank Res. 19: 435-447 Grice GD, Hulsemann K (1965) Abundance, vertical di stribution and taxonomy of calanoid copepods at selected stations in the northeast Atla ntic. J. Zool. 146: 213-262 Hargreaves PM (1985) The vertical distribution of D ecapoda, Euphausiacea, and Mysidacea at 42 N, 17 W. Biol. Oceanogr. 3: 431-464 Heffernan JJ, Hopkins TL (1981) Vertical distributi on and feeding of the shrimp genera Gennadas and Bentheogennema (Decapoda: Penaeidea) in the Eas tern Gulf of Mexico. J. Crust. Biol. 1: 461-473 Herring P (2002) The Biology of the Deep Ocean. Oxf ord University Press, Oxford Hopkins TL (1982) The vertical distribution of zoop lankton in the eastern Gulf of Mexico. Deep Sea Res. 29: 1069-1083 Hopkins TL, Baird RC (1975) Net feeding in mesopela gic fishes. Fish. Bull. 73: 908-914 Hopkins TL, Baird RC (1985a) Aspects of the trophic ecology of the mesopelagic fish Lampanyctus alatus (family Myctophidae) in the East ern Gulf of Mexico. Biol. Oceanogr.: 285-313 Hopkins TL, Baird RC (1985b) Feeding ecology of fou r Hatchetfishes (Sternoptychidae) in the Eastern Gulf of Mexico. Bull. Mar. Sci. 36: 260-277 Hopkins TL, Flock ME, Gartner JV, Torres JJ (1994) Structure and trophic ecology of a low latitude midwater decapod and mysid assemblage. Mar Ecol. Prog. Ser. 109: 143-156 Hopkins TL, Gartner JV (1992) Resource-partitioning and predation impact of a low-latitude myctophid community. Mar. Biol. 114: 185-197
85 Hopkins TL, Gartner JV, Flock ME (1989) The caridea n shrimp (Decapoda: Natantia) assemblage in the mesopelagic zone of the eastern Gulf of Mexi co. Bull. Mar. Sci. 45: 1-14 Hopkins TL, Lancraft TM (1984) The composition and standing stock of mesopelagic micronekton at 27 o N 86 o W in the eastern Gulf of Mexico. Cont. Mar. Sci. 27 : 143-158 Hopkins TL, Sutton TT (1998) Midwater fishes and sh rimps as competitors and resource partitioning in low latitude oligotrophic ecosystem s. Mar. Ecol. Prog. Ser. 164: 37-45 Hopkins TL, Sutton TT, Lancraft TM (1997) The troph ic structure and predation impact of a low latitude midwater fish assemblage. Prog. Oceanogr. 38: 205-239 Iwasaki N, Nemoto T (1987) Pelagic shrimps (Crustac ea: Decapoda) from the Southern Ocean between 150 E and 115 E. Mem. of Nat. Inst. of Pola r Res. 38: 1-40 Kinsey ST, Hopkins TL (1994) Trophic strategies of euphausiids in a low-latitude ecosystem. Mar. Biol. 118: 651-661 Kinzer J, Schulz K (1985) Vertical distribution and feeding patterns of midwater fish in the central equatorial Atlantic. Mar. Biol. 85: 313-322 Kinzer J, Schulz K (1988) Vertical distribution and feeding patterns of midwater fish in the central equatorial Atlantic II. Sternoptychidae. Mar. Biol. 99: 261-269 Krygier EE, Murano M (1988) Vertical distribution a nd zoogeography of oceanic mysids from the northeastern Pacific Ocean. Bull. of Ocean Res. Ins t. Univ. Tokyo 26: 109-122 Krygier EE, Pearcy WG (1981) Vertical distribution and biology of pelagic decapod crustaceans off Oregon. J. Crust. Biol. 1: 70-95 Krygier EE, Wasmer RA (1988) Zoogeography of pelagi c shrimps (Natantia: Penaeidea and Caridea) in the North Pacific Ocean (with synopses and keys to the species of the subarctic and transitional zones). Bull. of the Oce an Res. Inst., Univ. Tokyo 26: 43-98 Lancraft TM, Robison BH (1980) Evidence of postcapt ure ingestion by midwater fishes in trawl nets. Fish. Bull. 77: 713-715 Martin JW, Davis GE (2001) An updated classificatio n of the recent Crustacea. Natural History Museum of Los Angelas County, Los Angelas
86 Mauchline J (1972) The biology of bathypelagic orga nisms, especially Crustacea. Deep Sea Res. 19: 753-789 Mauchline J (1980) The biology of mysids and euphau siids. Adv. Mar. Biol. 18: 1-681 McEachran J, Fechhelm JD (1998) Fishes of the Gulf of Mexico, Vol. 1: Myxiniformes to Gasterosteiformes. University of Texas Press, Austi n McEachran J, Fechhelm JD (2006) Fishes of the Gulf of Mexico, Volume 2: Scorpaeniformes to Tetraodontiformes. University of Texas Press, Austi n Miya M, Nishida M (1996) Molecular phylogenetic per spective on the evolution of the deep-sea genus Cyclothone (Stomiiformes: Gonostomatidae). Ic hthyol. Res. 43: 375-398 Miya M, Nishida M (1997) Speciation in the open oce an. Nature 389: 803-804 Mller H-G (1993) World Catalogue and Biblography o f the Recent Mysidacea., Wetzlar, pp 491 Nowlin WDJ (1971) Water masses and general circulat ion of the Gulf of Mexico. Oceanology International and Offshore Technology 6: 28-33 Omori M (1974) The biology of pelagic shrimps in th e ocean. Adv. Mar. Biol. 12: 233-324 Passarella KC, Hopkins TL (1991) Species compositio n and food habits of the micronektonic cephalopod assemblage in the eastern Gulf of Mexico Bull. Mar. Sci. 49: 638-659 Prez-Farfante I, Kensley B (1997) Penaeoid and Ser gestoid Shrimps and Prawns of the World: Keys and Diagnoses for the Families and Genera. di tions du Musum national d'Histoire naturelle, Paris, Paris Pickard GL, Emery WJ (1990) Descriptive Physical Oc eanography. An Introduction. Pergamon Press, New York Remsen AR, Hopkins TL, Samson S (2003) What you see is not what you catch: a comparison of concurrently collected net, Optical Plankton Counte r, and Shadowed Image Particle Profiling Evaluation Recorder data from the northea st Gulf of Mexico. Deep Sea Res. 51: 129-151
87 Richter S (2003) The mouthparts of two lophogastrid s, Chalaraspidum alatum and Pseudochalaraspidum hanseni (Lophogastirda, Peracarida, Malacostraca), includin g some remarks on the monophyly of the Lophogastrida. J. of Nat. Hist. 37: 2773-2786 Roe HSJ (1972a) The vertical distributions and diru nal migrations of calanoid copepods collected on the SOND cruise, 1965. J. Mar. Biol. Ass. U.K. 5 2: 277-314 Roe HSJ (1972b) The vertical distributions and diur nal migrations of calanoid copepods collected on the SOND cruise, 1965. II. Sytematic account: f amilies Calanidae up to and including the Aetideidae. J. Mar. Biol. Ass. U.K. 52: 315-343 Roe HSJ (1984) The diel migrations and distribution s within a mesopelagic community in the north east Atlantic. 2. Vertical migrations and fee ding of mysids and decapod crustacea. Prog. Oceangr. 13: 269-318 Smith KLJ, Laver MB (1981) Respiration of the bathy pelagic fish Cyclothone acclinidens. Mar. Biol. 61: 261-266 Stearns SC (1992) The Evolution of Life Histories. Oxford University Press, Oxford Steinberg DK, Silver MW, Pilskaln CH, Coale SL, Pad uan JB (1994) Midwater zooplankton communities on pelagic detritus (giant larvacean ho uses) in Monterey Bay, California. Limnol. Oceanogr. 39: 1606-1620 Stickney DG, Torres JJ (1989) Proximate composition and energy content of mesopelagic fishes from the eastern Gulf of Mexico. Mar. Biol. 103: 13 -24 Sutton TT, Hopkins TL (1996a) The species compositi on, abundance, and vertical distribution of the stomiid (Pisces: Stomiiformes) fish assemblage of the Gulf of Mexico. Bull. Mar. Sci. 59: 530-542 Sutton TT, Hopkins TL (1996b) Trophic ecology of th e stomiid (Pisces: Stomiidae) assemblage of the eastern Gulf of Mexico: strategies, selectivity and impact of a mesopelagic top predator group. Mar. Biol. 127: 179-192 Sutton TT, Hopkins TL, Lancraft TM (1996) Trophic d iverstiy of a mesopelagic fish community. Proc. Second Intl. Conf. Pelagic Biodiversity. UNES CO: 377-381
88 Torres JJ (1984) Relationship of oxygen consumption to swimming speed in Euphausia pacifica. II. Drag, efficiency and comparison with other swim ming organisms. Mar. Biol. 78: 231237 Torres JJ, Childress JJ (1983) Relationship of oxyg en consumption to swimming speed in Euphausia pacifica. 1. Effects of temperature and pressure. Mar. Biol. 74: 79-86 Vereshchaka AL (1994) North Atlantic and Caribbean species of Sergia (Crustacea, Decapoda, Sergestidae) and their horizontal and vertical dist ribution. Steenstrupia 20: 73-95 Vinogradov ME (1962) Feeding of the deep-sea zoopla nkton. Rapp. Proc.-Verb. Cons. Intern. Explor. Mer 153: 114-120 Vinogradov ME (1997) Some problems of vertical dist ribution of mesoand macroplankton in the ocean. Adv. Mar. Biol. 32: 2-92 Walters JF (1976) Ecology of Hawaiian sergestid shr imps (Penaeidea: Sergestidae). Fish. Bull. 74: 799-836 Wasmer RA (1986) Pelagic shrimps of the family Oplo phoridae (Crustacea: Decapoda) from the Pacific sector of the Southern Ocean. In: Kornicker LS (ed) Biology of the Antarctic Seas XVII. AGU, Washington, D. C., pp 29-68 Wasmer RA (1993) Pelagic shrimps (Crustacea: Decapo da) from six USNS Eltanin cruises in the southeastern Indian Ocean, Tasman Sea, and southwes tern Pacific Ocean to the Ross Sea. In: Cairns S (ed) Biology of the Antarctic Sea s XXII. AGU, Washington, D.C., pp 4991 Wisner RL (1962) Midwater trawl surveys. Limnol. Oc eanogr. 7: xxxix-xli Yamaguchi A, Ikeda T, Watanabe Y, Ishizaka J (2004) Vertical distribution patterns of pelagic copepods as viewed from the predation pressure hypo thesis. Zoological Studies 43: 475485 Yamaguchi A, Watanabe Y, Ishida H, Harimoto T, Furu sawa K, Suzuki S, Ishizaka J, Ikeda T, Takahashi M (2002) Community and trophic structures of pelagic copepods down to
89 greater depths in the western subarctic Pacific (WE ST-COSMIC). Deep Sea Res. 49: 1007-1025 Yamaguchi A, Watanabe Y, Ishida H, Harimoto T, Maed a M, Ishizaka J, Ikeda T, Takahashi MM (2005) Biomass and chemical composition of net-plan kton down to greater depths (05800m) in the western North Pacific Ocean. Deep Sea Res. 52: 341-353
91 Appendix A: Length-weight regressions applied to shrimp genera Family Genus Equation Source Benthesicymidae Bentheogennema y = 0.4445x 2.5171 Lab unpub Benthesicymidae Gennadas y = 0.4445x 2.5171 Lab unpub Bresiliidae Lucaya y = 0.2021x 2.9374 Lab unpub (Acanthephyra) Eucopiidae Eucopia y = 0.4459x 2.4053 Lab unpub Lophogastridae Gnathophausia y = 0.4459x 2.4053 Lab unpub Lophogastridae Pseudochalaraspidum y = 0.4459x 2.4053 Lab unpub (Eucopia) Mysidae Boreomysis y = 0.0001x 3.6785 Lab unpub (Eucopia) Oplophoridae Acanthephyra y = 0.2021x 2.9374 Torres & Donnely unpub. (Acanthephyra) Oplophoridae Ephyrina y = 0.2021x 2.9374 Torres & Donnely unpub. (Acanthephyra) Oplophoridae Hymenodora y = 0.2021x 2.9374 Torres & Donnely unpub. (Acanthephyra) Oplophoridae Janicella y = 0.2021x 2.9374 Torres & Donnely unpub. (Acanthephyra) Oplophoridae Meningodora y = 0.2021x 2.9374 Torres & Donnely unpub. (Acanthephyra) Oplophoridae Notostomus y = 0.2021x 2.9374 Torres & Donnely unpub. (Acanthephyra) Oplophoridae Systellaspis y = 0.2021x 2.9374 Torres & Donnely unpub. (Acanthephyra) Pandalidae Parapandalis y = 0.0509x 3.6415 Torres & Donnely unpub. (Acanthephyra) Pasiphaeidae Parapasiphaea y = 0.3062x 2.4191 Lab unpub Pasiphaeidae Pasiphaea y = 0.3062x 2.4191 Lab unpub Sergestidae Sergestes y = 0.3690x 2.3124 Torres & Donnely unpub. (Acanthephyra) Sergestidae Sergia y = 0.5373x 2.3119 Torres & Donnely unpub. (Acanthephyra)
92 Appendix B: Length-weight regressions applied to fish genera. Family Genus Equation Source Alepocephalidae alepocephalid y = 0.0048x 2.6517 Sutton unpub Alepocephalidae Bathylaco y = 0.0048x 2.6517 Sutton unpub Alepocephalidae Herwigia y = 0.0048x 2.6517 Sutton unpub Alepocephalidae Photostylus y = 0.0048x 2.6517 Sutton unpub Alepocephalidae Talismania y = 0.0048x 2.6517 Sutton unpub Anoplogasteridae Anoplogaster y = 0.00218x 3.0 Fish Base Bathylagidae bathylagid y = 0.00051x 3.2335 Fish Base Bathylagidae Dolicholagus y = 0.00051x 3.2335 Fish Base Bregmacerotidae Bregmaceros y = 0.001x 3.3213 Hopkins unpub (myctophid) Centrophrynidae Centrophryne y = 0.001x 3.04 Sutton (1996) (Group III stomiid) Cetomimidae cetomimid y = 0.001x 3.04 Sutton (1996) (Group III stomiid) Cetomimidae Cetomimus y = 0.001x 3.04 Sutton (1996) (Group III stomiid) Cetomimidae Cetostoma y = 0.001x 3.04 Sutton (1996) (Group III stomiid) Cetomimidae Ditropichthys y = 0.001x 3.04 Sutton (1996) (Group III stomiid) Chiasmodontidae Chiasmodon y = 0.001x 3.04 Sutton (1996) (Group III stomiid) Chiasmodontidae Dysalotus y = 0.001x 3.04 Sutton (1996) (Group III stomiid) Chloropthalmidae Chlorophthalmus y = 0.00013x 3.38 Fish Base Evermannellidae Coccorella y = 0.00013x 3.38 Fish Base Evermannellidae evermannellid y = 0.00013x 3.38 Fish Base Evermannellidae Odontostomops y = 0.00013x 3.38 Fish Base Gempylidae Diplospinus y = 0.0005x 2.74 Sutton (1996) (Group I stomiid) Gigantactinidae Gigantactis y = 0.00104x 3.04 Sutton (1996) (Group III stomiid) Gigantactinidae Rhynchactis y = 0.00104x 3.04 Sutton (1996) (Group III stomiid) Giganturidae Gigantura y = 0.000424x 3.03 Sutton (1996) (Group II stomiid) Gonostomatidae Cyclothone y = 0.0015x 2.8519 Lab unpub Gonostomatidae Sigmops y = 0.0008x 2.8788 Lab unpub (Cyclothone) Himantolophidae Himantolophus y = 0.00104x 3.04 Sutton (1996) (Group III stomiid) Howellidae Howella y = 0.001x 3.3213 Hopkins unpub (myctophid) Linophrynidae Linophryne y = 0.00104x 3.04 Sutton (1996) (Group III stomiid) Megalomycteridae Ataxolepis y = 0.0005x 2.74 Sutton (1996) (Group I stomiid) Melamphaidae Melamphaes y = 0.0035x 2.9811 Lab unpub
93 Appendix B: continued Family Genus Equation Source Melamphaidae Poromitra y = 0.0035x 2.9811 Lab unpub Melamphaidae Scopeloberyx y = 0.0035x 2.9811 Lab unpub Melamphaidae Scopelogadus y = 0.0035x 2.9811 Lab unpub Melamphaidae Scopelosaurus y = 0.0035x 2.9811 Lab unpub Melanocetidae Melanocetus y = 0.00104x 3.04 Sutton (1996) (Group III stomiid) Melanonidae Melanonus y = 0.000424x 3.03 Sutton (1996) (Group II stomiid) Mirapinnidae Eutaeniophorus y = 0.0005x 2.74 Sutton (1996) (Group I stomiid) Mirapinnidae mirapinnids y = 0.0005x 2.74 Sutton (1996) (Group I stomiid) Moridae Physiculus y = 0.000424x 3.03 Sutton (1996) (Group II stomiid) Myctophidae Benthosema y = 0.001x 3.3213 Lab unpub (myctophid) Myctophidae Bolinichthys y = 0.001x 3.3213 Lab unpub (myctophid) Myctophidae Ceratoscopelus y = 0.001x 3.3213 Lab unpub (myctophid) Myctophidae Diaphus y = 0.001x 3.3213 Lab unpub (myctophid) Myctophidae Diogenichthys y = 0.001x 3.3213 Lab unpub (myctophid) Myctophidae Gonichthys y = 0.001x 3.3213 Lab unpub (myctophid) Myctophidae Hygophum y = 0.001x 3.3213 Lab unpub (myctophid) Myctophidae Lampadena y = 0.001x 3.3213 Lab unpub (myctophid) Myctophidae Lampanyctus y = 0.001x 3.3213 Lab unpub (myctophid) Myctophidae Lepidophanes y = 0.001x 3.3213 Lab unpub (myctophid) Myctophidae Myctophid y = 0.001x 3.3213 Lab unpub (myctophid) Myctophidae Myctophum y = 0.001x 3.3213 Lab unpub (myctophid) Myctophidae Nannobrachium y = 0.001x 3.3213 Lab unpub (myctophid) Myctophidae Notolychnus y = 0.001x 3.3213 Lab unpub (myctophid) Myctophidae Notoscopelus y = 0.001x 3.3213 Lab unpub (myctophid) Myctophidae Taaningichthys y = 0.001x 3.3213 Lab unpub (myctophid) Nemichthyidae Avocettina y = 0.000006x 3.5778 Sutton unpub Neoscopelidae Scopelengys y = 0.0015x 3.0253 Sutton unpub Omosudidae Omosudis y = 0.000424x 3.03 Sutton (1996) (Group II stomiid) Oneirodidae Chirophryne y = 0.00104x 3.04 Sutton (1996) (Group III stomiid) Oneirodidae Chirophryne y = 0.00104x 3.04 Sutton (1996) (Group III stomiid) Oneirodidae Dolopichthys y = 0.00104x 3.04 Sutton (1996) (Group III stomiid) Oneirodidae Dolopichthys y = 0.00104x 3.04 Sutton (1996) (Group III stomiid) Oneirodidae Lophodolos y = 0.00104x 3.04 Sutton (1996) (Group III stomiid)
94 Appendix B: continued Family Genus Equation Source Oneirodidae Lophodolos y = 0.00104x 3.04 Sutton (1996) (Group III stomiid) Oneirodidae oneirodid y = 0.00104x 3.04 Sutton (1996) (Group III stomiid) Ophidiidae Bassozetus y = 0.00104x 3.04 Sutton (1996) (Group III stomiid) Paralepididae paralepidid y = 0.0005x 2 .74 Sutton (1996) (Group I stomiid) Paralepididae Stemonosudis y = 0.0005x 2.74 Sutton (1996) (Group I stomiid) Paralepididae Uncisudis y = 0.0141x 2.7106 Hopkins unpub Pasiphaeidae Parapasiphaea y = 0.3062x 2.4191 Lab unpub Pasiphaeidae Pasiphaea y = 0.3062x 2.4191 Lab unpub Phosichthyidae Ichthyococcus y = 0.00104x 3.04 Sutton (1996) (Group II stomiid) Phosichthyidae Pollichthys y = 0.00104x 3.04 Sutton (1996) (Group II stomiid) Phosichthyidae Vinciguerria y = 0.00104x 3.04 Sutton (1996) (Group II stomiid)
95 Appendix C: Bathypelagic cephalopod individuals from bathypelag ic trawl series in the eastern Gulf of Mexico. Sample Quantitative Species Mantle Length SC96-10 Yes Bolitaena pygmaea 45 SC96-10 Yes Grimalditeuthis bomplandii 60 SC96-10 Yes Mastigoteuthis flammea 56 SC96B-04 Yes Joubiniteuthis portieri 92 SC96B-23 Yes Japetella diaphana 56 SC96B-22 Yes Mastigoteuthis flammea 74 SC96B-22 Yes Mastigoteuthis sp. 157 SC96B-22 Yes Vampyroteuthis infernalis 52 SC96B-22 Yes Haliphron atlanticus 70 SC97A-05 Yes Mastigoteuthis flammea 100 SC97A-07 Joubiniteuthis portieri 123 SC97A-09 Yes Grimalditeuthis bomplandii 84 SC97A-09 Yes Japetella diaphana 25 SC98-18 Yes Magnapinna n. sp. 65 SC98-21 Yes Bolitaena pygmaea 36 SC99-32 Chiroteuthis sp (sp. B2?) 55 SC99-32 Chiroteuthis sp (sp. B2?) 55 SC99-32 Chiroteuthis sp? 84 SC99-32 Japetella diaphana 58 SC99-32 Japetella diaphana 38 SC99-32 Japetella diaphana 33 SC99-32 Japetella diaphana 29 SC99-32 Japetella diaphana 25 SC99-32 Japetella diaphana 27 SC99-33 Mastigoteuthis flammea 73 P99-05 Japetella diaphana 37 P99-28 Cycloteuthis serventyi 32 P99-28 Grimalditeuthis bomplandii 53 P99-28 Japetella diaphana 44 SC00-02 Japetella diaphana 69 SC00-13 Grimalditeuthis bomplandii 58
About the Author Scott E. Burghart was born in Santa Clara, Califor nia, but spent most of his childhood in Texas. After receiving his Bachelor of Science Deg ree from Baylor University in Biology and Environmental Studies, he moved to Florida and bega n his graduate career at the University of South Florida. He received his Master of Science d egree in Biological Oceanography working with Antarctic zooplankton and began a career as a zooplankton taxonomist. While working in this capacity, he began pursuing his Ph.D., the cul mination of which is contained in this work.