W&M ScholarWorks
Dissertations, Theses, and Masters Projects Theses, Dissertations, & Master Projects
1984
Food habits and body composition of some dominant deep-sea fishes from temperate and tropical regions of the western North Atlantic
Roy E. Crabtree College of William and Mary - Virginia Institute of Marine Science
Follow this and additional works at: https://scholarworks.wm.edu/etd
Part of the Biology Commons
Recommended Citation Crabtree, Roy E., "Food habits and body composition of some dominant deep-sea fishes from temperate and tropical regions of the western North Atlantic" (1984). Dissertations, Theses, and Masters Projects. Paper 1539616621. https://dx.doi.org/doi:10.25773/v5-xmzd-zb60
This Dissertation is brought to you for free and open access by the Theses, Dissertations, & Master Projects at W&M ScholarWorks. It has been accepted for inclusion in Dissertations, Theses, and Masters Projects by an authorized administrator of W&M ScholarWorks. For more information, please contact [email protected]. INFORMATION TO USERS
This reproduction was made from a copy of a document sent to us for microfilming. While the most advanced technology has been used to photograph and reproduce this document, the quality of the reproduction is heavily dependent upon the quality of the material submitted.
The following explanation of techniques is provided to help clarify markings or notations which may appear on this reproduction.
1.The sign or “target” for pages apparently lacking from the document photographed is “Missing Page(s)”. If it was possible to obtain the missing page(s) or section, they are spliced into the film along with adjacent pages. This may have necessitated cutting through an image and duplicating adjacent pages to assure complete continuity.
2. When an image on the film is obliterated with a round black mark, it is an indication of either blurred copy because of movement during exposure, duplicate copy, or copyrighted materials that should not have been filmed. For blurred pages, a good image of the page can be found in the adjacent frame. If copyrighted materials were deleted, a target note will appear listing the pages in the adjacent frame.
3. When a map, drawing or chart, etc., is part of the material being photographed, a definite method of “sectioning” the material has been followed. It is customary to begin filming at the upper left hand comer o f a large sheet and to continue from left to right in equal sections with small overlaps. If necessary, sectioning is continued again—beginning below the first row and continuing on until complete.
4. For illustrations that cannot be satisfactorily reproduced by xerographic means, photographic prints can be purchased at additional cost and inserted into your xerographic copy. These prints are available upon request from the Dissertations Customer Services Department.
5. Some pages in any document may have indistinct print. In all cases the best available copy has been filmed.
University/ Microfilms International 300 N. Zeeb Road Ann Arbor, Ml 48106
8428467
Crabtree, Roy Eugene
FOOD HABITS AND BODY COMPOSITION OF SOME DOMINANT DEEP-SEA FISHES FROM TEMPERATE AND TROPICAL REGIONS OF THE WESTERN NORTH ATLANTIC
The College of William a n d M ary in Virginia Ph.D. 1984
University Microfilms International300 N. Zeeb Road, Ann Arbor, Ml 48106
PLEASE NOTE:
In all cases this material has been filmed in the best possible way from the available copy. Problems encountered with this document have been identified here with a check mark V .
1. Glossy photographs or pages ______
2. Colored illustrations, paper or print ______
3. Photographs with dark background ______
4. Illustrations are poor copy ______
5. Pages with black marks, not original copy ______
6. Print shows through as there is text on both sides of page ______
7. Indistinct, broken or small print on several pages
8. Print exceeds margin requirements ______
9. Tightly bound copy with print lost in spine ______
10. Computer printout pages with indistinct print ______
11. Page(s)______lacking when material received, and not available from school or author.
12. Page{s)______seem to be missing in numbering only as text follows.
13. Two pages n u m b ered ______. Text follows.
14. Curling and wrinkled pages ______
15. O ther ______
University Microfilms International
FOOD HABITS AND BODY COMPOSITION OF SOME
DOMINANT DEEP-SEA FISHES FROM TEMPERATE AND TROPICAL
REGIONS OF THE WESTERN NORTH ATLANTIC
A Dissertation
Presented to
The Faculty of the School of Marine Science The College of William and Mary in Virginia
in Partial Fulfillment
of the Requirements for the Degree of
Doctor of Philosophy
by
Roy E. Crabtree
1984 APPROVAL SHEET
This dissertation is submitted in partial fulfillment of the
requirements for the degree of
Doctor of Philosophy
Crabtree
Approved, September 1984
John V. MeyrYner \ National Marine Fisheries xervice
Kenneth J\ Stilak Natiodl'al Museum of Natural History
Richar
Evon P.Ruzeclci TABLE OF CONTENTS
Page
ACKNOWLEDGEMENTS...... v
LIST OF TABLES...... vii
LIST OF FIGUR ES...... x
ABSTRACT...... xiii
INTRODUCTION...... 2
MATERIALS AND METHODS...... 6
C o llecti ons ...... 6
Food ha b i t s ...... 13
Body composition ...... 17
RESULTS...... 20
Food h a b i t s ...... 20
Squ alidae...... 20
Opb ichthidae ...... 22
Syn aphobranchi dae ...... 24
Halosauridae ...... 28
No t scanthidae ...... 41
Ale pocephalidae...... 47
Chi orophthalmidae ...... 57
Ogc -ocephalidae ...... 68
Macrouridae ...... 70
Ste phanoberycidae ...... 80
Cot tid a e ...... 80
iii Page
Cluster analysis ...... 83
Body composition ...... 37
Water content ...... 88
Ash free dry weight ...... 92
Carbon ...... 98
N itrogen ...... 98
DISCUSSION...... 103
Predator groups...... 103
Ontogenetic dietary shifts ...... 108
Overlap in d i e t ...... 109
Faunal composition ...... 114
Body composition...... 118
LITERATURE CITED...... 125
APPENDIX 1 ...... 132
APPENDIX 2...... 139
VITA...... 140
iv ACKNOWLEDGMENTS
I would like to thank my committee chairman, Dr. John A.
Musick, for his support and advise, and the members of my committee, Dr. Kenneth J. Sulalc, Dr. John V. Merriner, Dr.
Richard L. Wetzel, and Dr. Evon P. Ruzecki for their comments.
Special thanks to Dr. C. Richard Robins for access to specimens in the University of Miami fish collection. This study was supported by grants NSF-0CE-7600729, NSF-OCE-
7920567, to the Virginia Institute of Marine Science, John A.
Musick pricipal investigator, NSF-OCE-7306639 to the
University of Miami, C. Richard Robins, principal investigator, and NSF predoctoral grant 0CE-8104574 to the author.
Deep-sea work requires the cooperative efforts of many people. Foremost among those involved in this study was Ken
Sulak without whose advise and support this work would not have been possible. Also great thanks to Jacque Carter and
Eric Anderson for many discussions of deep-sea biology and endless hours at sea, and to former VIMS graduate students
George R. Sedberry, Douglas Markle, Charles Wenner, and John
Gartner, whose work on deep-sea fishes contributed greatly to this study. In addition, I thank James Colvocoresses for his advise concerning the computer analysis of data. Others whose imput was particularly valued include Daniel M. Cohen and
Robert Carney. I also thank the many graduate students past and present at VIMS who in one way or another contributed including Tom Munroe, B ill Raschi, John Olney, John Gourly,
Joe Smith, Mike Armstrong, Brian Bowen, Pat Duncan, Steve
Smith, Tom Sminky, Ruth Klinger, and many others.
I am grateful to Jacque van Montfrans, Marcia Bowen, and
Gary Gaston, VIMS, Michael Vecchione, McNeese State
University, and John Dickinson, National Museum of Canada, who identified many of the invertebrates found in fish guts. I also thank Gordon W. Thayer and W. Judson Kenworthy of the
National Marine Fishery Service Laboratory in Beaufort, N.C. who performed the carbon-nitrogen analysis of deep-sea fishes.
Finally, I thank my wife Jane who buys me boats,
Seamaster reels, sends me to the Keys, and has supported me throughout my long graduate career. LIST OF TABLES
Table Page
1. Cruises conducted by the Virginia Institute of Marine Science (VIMS) and the University of Miami (RSMAS) in the Middle Atlantic Bight and Bahamas study a re a s ...... 7
2. Fishes examined in food habits study ...... 14
3. Fishes examined in body composition study ...... 18
4. Gut contents of Centroscyllium fabricii. F = percent frequency of occurrence, W = percent weight, N = percent numerical abundance ...... 21
5. Gut contents of Deania calceus. F = percent frequency of occurrence, W = percent weight, N = percent numerical abundance ...... 23
6. Gut contents of Ophichthus cruentifer. F =percent frequency of occurrence, W =percent weight, N = percent numerical abundance...... 25
7. Gut contents of Synaphobranchus brevidorsalis. F = percent frequency of occurrence, W = percent weight, N = percent numerical abundance ...... 26
8. Gut contents of Aldrovandia affinis. F =percent frequency of occurrence, W =percent weight, N = percent numerical abundance ...... 29
9. Gut contents of Aldrovandia gracilis. F =percent frequency of occurrence, W = percent weight, N = percent numerical abundance ...... 33
10. Gut contents of Aldrovandia phalacra. F = percent frequency of occurrence, W =percent weight, N = percent numerical abundance ...... 37
vii Table Page
11. Gut contents of Aldrovandia oleosa. F = percent frequency of occurrence, W = percent weight, N = percent numerical abundance ...... 39
12. Gut contents of Polyacanthonotus sp. A. F = percent frequency of occurrence, W = percent weight, N = percent numerical abundance ...... 42
13. Gut contents of Polyacanthonotus rissoanus. F = percent frequency of occurrence, W = percent weight, N = percent numerical abundance ...... 44
14. Gut contents of Conocara macropterurn 175 mm,SL or less. F = percent frequency of occurrence, W = percent weight, N = percent numerical abundance ...... 49
15. Gut contents of Conocara macropterurn larger than 175 mm,SL. F = percent frequency of occurrence, W = percent weight, N = percent numerical abundance ...... 51
16. Gut contents of Conocara niger. F = percent frequency of occurrence, W = percent weight, N = percent numerical abundance ...... 56
17. Gut contents of Marcetes stomias. F = percent frequency of occurrence, W = percent weight, N = percent numerical abundance ...... 58
18. Gut contents of Bathypterois longipes. F = percent frequency of occurrence, W = percent weight, N = percent numerical abundance ...... 59
19. Gut contents of Bathypterois grallator 240 mm,SL or less. F = percent frequency of occurrence, W = percent weight, N = percent numerical abundance ...... 62
20. Gut contents of Bathypterois grallator larger than 240 mm,SL. F = percent frequency of occurrence, W = percent weight, N = percent numerical abundance ...... 64
viii Table Page
21. Gut contents of Bathypterois phenax. F = percent frequency of occurrence, W = percent weight, N = percent numerical abundance...... 65
22. Gut contents of Ipnops murrayi. F = percent frequency of occurrence, W = percent weight, N = percent numerical abundance ...... 67
23. Gut contents of Dibranchus atlanticus. F = percent frequency of occurrence, W = percent weight, N = percent numerical abundance ...... 69
24. Gut contents of Nezumia bairdii 35 mm,HL or less. F = percent frequency of occurrence, W = percent weight, N = percent numericalabundance ...... 73
25. Gut contents of Nezumia bairdii larger than 35 mm,HL. F = percent frequency of occurrence, W = percent weight, N = percent numerical abundance ...... 74
26. Gut contents of Ventrifossa occendentalis. F = percent frequency of occurrence, W =percent weight, N = percent numerical abundance ...... 78
27. Gut contents of Stephanoberyx monae. F = percent frequency of occurrence, W = percent weight, N = percent numerical abundance ...... 81
28. Gut contents of Cottunculus thompsoni. F = percent frequency of occurrence, W = percent weight, N = percent numerical abundance ...... 82
29. Body composition of selected deep-sea fishes. Values in parentheses are standard error of the mean andnumber of specimens examined...... 89
30. Mean body composition parameters for selected deep-sea fish families. Values in parentheses are standard error of the mean and number of species examined ...... 91
31. Mean percent water, total ash, skeletal ash, carbon and nitrogen content of benthic and benthopelagic species with and without sx^imbladder s ...... 93 ix LIST OF FIGURES
Figure Page
1. Location of Middle Atlantic Bight and Bahamas study areas in the western North Atlantic. Areas detailed in Figures 2, 3, 4, and 5 are indicated ...... 8
2. Station locations from the Norfolk Canyon area, Middle Atlantic Bight...... 9
3. Station locations in the vicinity of Hudson Canyon, Middle Atlantic Bight ...... 10
4. Station locations in the Bahamas study area and adjacent abyssal plain ...... 11
5. Location of stations within the Bahamian basins, Tongue-of-the-Ocean, Exuma Sound, Columbus Basin, and Inagua Basin ...... 12
6. IRI values of dominant prey taxa plotted against predator size for Aldrovandia affinis ...... 30
7. IRI values of dominant prey taxa plotted against predator size for Aldrovandia affinis ...... 31
8. IRI values for dominant prey taxa plotted against predator size for Aldrovandia gracilis ...... 34
9. IRI values of dominant prey taxa plotted against predator size for Aldrovandia gracilis ...... 35
10. IRI values of dominant prey taxa plotted against predator size for Aidrovandia phalacra ...... 38
11. IRI values of dominant prey taxa plotted against predator size for Conocara macropterum ...... 48
12. IRI values of dominant prey taxa plotted against predator size for Conocara macropterum ...... 50
13. Frequency of ingestion of sediment plotted against fish size for Conocara macropterurn...... 54
14. IRI values of dominant prey taxa plotted against predator size for Bathypterois grallator ...... 61
x Figure Page
15. IRI values of dominant prey taxa plotted against predator size for Bathypterois grallator ...... 63
16. IRI values of dominant prey taxa plotted against predator size for Nezumia bairdii ...... 71
17. IRI values of dominant prey taxa plotted against predator size for Nezumia bairdii ...... 72
18. IRI values of dominant prey taxa plotted against predator size for Ventrifossa occidentalis...... 79
19. Species groupings based on cluster analysis of gut contents data for dominant species from the Bahamas study area ...... 84
20. Water content of benthopelagic species plotted against capture depth ( r a=0.187; p<0.01; n=156). Points represent values for each specimen examined for the species listed in Table29 and as grouped according to life style in thete x t...... 94
21. Ash as percent dry weight plotted against percent water for selected deep-sea fishes ( r 3=0.415; p<0.01; n=167). Points represent values for each specimen examined of the species indicated in Table 29 ...... 96
22. Ash as percent dry weight plotted against capture depth for selected deep-sea fishes ( r a=0.073; p<0.001; n=167). Points represent values for each specimen examined of the species indicated in Table 29...... 97
23. Carbon as percent ash free dry weight plotted against capture depth for selected deep-sea fishes ( r 3=0.099; p<0.05; n=43). Points represent values for each specimen examined of the species indicated in Table 29 ...... 99
24. Carbon as percent wet weight plotted against depth of capture for selected deep-sea fishes ( r 3=0.261; p<0.01; n=43). Points represent values for each specimen examined of the species indicated in Table 29 ...... 100
25. Nitrogen as percent wet weight plotted against capture depth for selected deep-sea fishes ( r 3=0.203; p<0.01; n=43). Points represent values for each specimen examined for the species indicated in Table 29 ...... 101
xi Figure Page
26. Nitrogen as percent ash free dry weight plotted against depth for selected deep-sea fishes ( r 3=0.102; p<0.05; n=43). Points represent values for each specimen examined for the species indicated in Table 29 ...... 102
xii ABSTRACT
Food habits of 23 species of demersal deep-sea fishes from the temperate Middle Atlantic Bight and the tropical Bahamas region are described. In addition, body composition parameters including percent water, ash, carbon, and nitrogen are discussed for 48 demersal species from these study areas.
Food habits data on Bahamian species are combined with those from other studies in an attempt to describe the trophic structure of this tropical deep-sea fish assemblage. Numerical classification techniques are used to group species based upon similarity of diets. Four groups are evident, including a group which feeds largely on polychaetes, a second which feeds mainly on copepods along with other small crustaceans, a third group which feeds on small crustaceans but most heavily on mysids and amphipods, and finally a fourth group which feeds heavily on natant decapods and teleosts. Within each group a variety of taxa and feeding modes are apparent, including benthic species as well as benthopelagic species with and without swimbladders. Thus a variety of feeding mechanisms which result in similar diets are displayed by these fish assemblages.
Body composition parameters are variable; however, some trends are evident. Benthopelagic species without swimbladders tend to have higher water contents and more poorly ossified skeletons than other species. In addition, percent water shows a positive correlation with depth of occurrence, but it is apparent that at all depths, species with a variety of body compositions have successfully adapted to life in the deep sea.
The diets and body compositions of tropical species are compared with those of temperate Middle Atlantic Bight species in an attempt to account for taxonomic differences between the two areas. Species with high water contents appear to be more abundant in the Bahamas study area than in the Middle Atlantic Bight. Accordingly, species with relatively inactive life styles seem to be more important in the Bahamas region. Differences in trophic structure may account for other differences between the Middle A tlantic Bight and Bahamian faunas.
xiii Food habits and body composition of some dominant deep-sea fishes from temperate and tropical
regions of the western North Atlantic Introduction
Recent increases in the number of deep-sea demersal fish collections have led to the publication of a number of regional faunal descriptions (Markle and Musick, 1974;
Haedrich, Rowe, and Polloni, 1975, 1980; Haedrich and K refft,
1978; Musick, 1979; Merrett and Marshall, 1981; Pearcy,
Stein, and Carney, 1982; Sulak, 1982; and Anderson et a l., in press). These studies point out striking taxonomic differences between regions, even though many of these regions are not separated by obvious physical boundaries. Merrett and Marshall (1981) discuss differences in the taxonomic composition of dominant species off northwest Africa with those of the temperate western North Atlantic. Pearcy et a l.
(1982) found major taxonomic differences between the northeastern Pacific fauna and that of the Atlantic. Indeed, even within the western North Atlantic, Sulak (1982) found major differences between the deep-sea fish fauna of the
Bahamian basins and that of the USA Middle Atlantic Bight
(MAB), and Anderson et al (in press) pointed out differences between the fauna within the Caribbean basin and that of the
Bahamas region.
2 3
In addition to taxonomic differences, regional ecological differences also exist. The so called "bigger-deeper" phenomenon described for the temperate western North Atlantic by Haedrich and Rowe (1977), Musick (1979), and Polloni et al.
(1979), has not been observed in the Bahamas (Sulak, 1982), off northwest Africa (Merrett and Marshall, 1981), or in the northeastern Pacific (Pearcy et al . , 1982). Furthermore, differences in average fish size have been found between regions. Fishes from the Bahamas (Sulak, 1982) and off northwest Africa (Merrett and Marshall, 1981) are conspicuously smaller than those of the temperate western
North Atlantic (Sulak, 1982). Presently, most considerations of regional faunas have been broad in scope and considered overall differences community trends without detailed comparisons of the ecology of individual species. Sulak
(1982) discussed regional differences in trophic structure, energetics, and reproductive patterns; however, his discussion is limited by the scarcity of data on the life histories of deep-sea fishes.
The present study, is based upon bottom trawl collections initiated in 1973 by the Virginia Institute of Marine Science to describe demersal fish assemblages on the continental slope, rise, and abyss of the MAB (Musick, 1979) and a sim ilar study conducted by the University of Miami in the Bahamian deeps (Tongue-of-the-Ocean, Providence Channel, and Exuma
Sound) at about the same time. Similar collecting methods were used in both studies (see Sulak, 1982), thus providing an 4
opportunity for comparison of the temperate MAB fauna with that of the tropical Bahamas region.
Resulting studies have analyzed species composition (MAB
- Musick, 1979; Bahamas - Sulak, 1982), and in the MAB, feeding habits (Markle, 1976; Sedberry and Musick, 1978;
Wenner, 1978) and life histories (Markle, 1975, 1976; Markle and Wenner, 1979; Middleton, 1979; Wenner, 1975, 1976, 1978,
1979; Wenner and Musick, 1979). Only a few studies have addressed the biology of individual species from the Bahamas
(Marshall and Staiger, 1975; Sulak, 1977a, 1977b; Carter,
1984; and Crabtree, Sulak, and Musick, in press). Several of these studies are largely taxonomic in scope.
Sulak (1982) found a number of important differences in the faunal composition of the Bahamas and MAB study areas in terms of numerical dominance of both families and species.
The families Synaphobranchidae, Moridae, and Macrouridae are numerically dominant in the MAB, while in the Bahamas the fauna is comprised predominaitely of the Synaphobranchidae,
Ophidiidae, Chlorophthalmidae, and Halosauridae. Even within families occurring in both areas, different species often predominate. Sulak suggests that these differences may reflect differing levels of food availability in the respective areas with species adapted to low energy levels prevailing in the Bahamas. Additional data on food habits, energy requirements, and reproductive biology of deep-sea species are needed to evaluate the adaptations of tropical and temperate deep-sea fishes to their respective environments. The present study presents new data on the food habits of deep-sea fishes from the Bahamas and the temperate western
North Atlantic. In addition, data on body composition parameters, including water content, ash, carbon, and nitrogen, are presented for dominant species from each study area. Body composition parameters allow inferences of energy requirements and swimming capabilities of fishes from the two study areas. Materials and Methods
Collections
This study is based upon collections (cruises listed in
Table 1) made by the Virginia In stitu te of Marine Science
(VIMS) and the University of Miami's Rosenstiel School of
Marine and Atmospheric Science (RSMAS). General areas sampled are shown in Fig. 1. Seven VIMS cruises were conducted in the
MAB, concentrated in the vicinity of Norfolk Submarine Canyon
(Fig. 2) with additional stations near Hudson Submarine Canyon
(Fig. 3). Sampling in the MAB covered a depth range of 69-
4879 m. In the Bahamas four major basins were sampled, including Tongue-of-the-Ocean, Exuma Sound, Columbus Basin, and Inagua Basin (Figs. 4 and 5). Additional collections were made on the Blake Plateau, in Providence Channel, and on the abyssal plain adjacent to the Bahamas study area (Fig. 4).
Sampling in the Bahamas covered a depth range of 918-5345 m.
In all, seven cruises by RSMAS and three by VIMS were made in the Bahamas. Additional specimens for comparative life history work were taken aboard cruises by R/V Albatross IV
(NMFS), R/V Virginian Sea (VIMS), and FFS Anton Dohrn
(In stitu t flir Seef isc h e re i, West Germany, cruise 213-3) along the U.S. east coast.
6 ■H Table 1. Cruises conducted by the Virginia Institute of Mar c o Science (VIMS) and the University of Miami (RSMAS) 3 3 in the Middle Atlantic Bight and Bahamas study are • C_J n
I
O' O' O' *a I I co U 23E 3 (2 CD M O(0 CO M O pq CD aj CO p o - n oo ' O M M i—c x: O o o O 00 pq o c rH H •H 43 4J O3 CO 2c(2 c 12 3 a a c Cu o 0 CO CO a a a DC)CD CL) CD £ E * * • En : \ M M CO CO CO er* rH PQ 42 rH _d 2C a 43 HT rH OC CO CO CO OC OCO CO CO CO 3 3 3 3 3 E £ 3 E U U 00) (0 3 ** CO H H M r>- 2 4 Hr iH rH rH 2 4 fH pq o> si C o H»H »H - 3 3 3 3 O O O a a a CD h 3 3 3 • * 1 CO qeq pq * cn H i- ON r*- S3 HrH rH x: o o 42 o 3 4 0 U 3 > D CO n i CT* ^ r r-I si o 43 •H < 3 U £ CO 3 * CO 42 m H H pq r-H o x: a u rH rH 43 H HH ♦H c 3 3 3 3 E E 3 3 3 3 d d 00 42 00 3 O O OCO CO 3 3 a a CD OCO CO 3 3 * * v£> CO ^ r pq : s Si i-H D rH 42 co 7 8 Figure 1. Location of Middle Atlantic Bight and Bahamas study areas in the western North Atlantic. Areas detailed in Figures 2, 3, A, and 5 are indicated. ,'MOVA SCOTIA' IGi 2 .WESTER NORTH ATLANT 9 Figure 2 Station locations from the Norfolk Canyon area, Middle Atlantic Bight. ■ 1 1... 1 i _ "“ i • ' 00 ° 73 - ■-9 ' ' ...... „- 30 •• > ° 73 \ ..... • \ • N\ ‘\ • • /~ iijooSS" -- / / \ c - p # ‘ / 1 ' \! 1 i /7^ • . • / .'i • 00 n ° ill ft ix • I - • II t ! \J \ 74 ^ g J '"'r J Z l/llv o < 0 1 1 ' * g . • I \ l| 0, \ z r * i 't 11* i ’ \ i • “ XX ’> n r ^ KJm\V s-a „ XX ' > '“ ■“'-I - XJ 0 • * | 11 • n 1 30 • wr . , ° * • • * * • 74 ^ c s s s r " i •*>£ - — — • --X S 5vsp \ \ " Hs \ v N s x rvC *" §£' ' r - v vn 00 ° " n \ \ ,—* 75 \ rv I j' \ xw ' ^ *-"*N •v 1 ...... r ■■ - " i - i ■—~r "o O ro n "S 10 Figure 3 Station locations in the vicinity of Hudson Canyon, Middle Atlantic Bight. 0o 9 06 .OO. ,OOolA .OOoOZ. 00o69 ,00oQ9 z ,00oZl ,00a ZL to 11 Figure 4 Station locations in the Bahamas study area and adjacent abyssal plain. (Modified from Sulak, 1982) < CO CO (a ir cr< s^ ui =» 2 <13 O< A < V '-- ■ uiui'oooy OC i'66'cfe □□ ";"A‘ 9 ,••• o tr u. 82°00‘ 80*00’ 78*00' 76*00' 74*00' 72*00' 70*00' 68*00' CM 12 Figure 5. Location of stations within the Bahamian basins, Tongue-of-the-Ocean, Exuma Sound, Columbus Basin, and Inagua Basin. (Modified from Sulak, 1982) 26*00 ‘ V M 25*001 -| ^ V \ N . , \ \ \ 24* 00' % TONGUE OF THE OCEAN 23*00f 1 • COLUMBUS \ \ _ BASIN 7 W / i i \ \ VL* V.— .. * 22* 00' INAGUA SSH BASIN 78*00' 77*00' 76*00 75*00* 74*00* 13 Sampling on VIMS cruises was accomplished with a 13.7 m (45 ft) trawl while RSMAS cruises utilized a slightly smaller 12.5 m (41 ft) net. Both nets are semi-balloon otter trawls (Marinovich Gulf of Mexico shrimp trawls), similar to those used by Haedrich et a l . (1980) and Merrett and Marshall (1981) in other North Atlantic deep-sea demersal fish programs. Trawls were fished at a speed of approximately two knots for 60 minutes on all RSMAS cruises and for 30 minutes to 180 minutes on VIMS cruises depending on depth. For a detailed description of sampling gear and techniques see Sulak (1982). Food Habits Stomachs and intestinal tracts were removed from museum specimens housed at VIMS or RSMAS and from fresh specimens at sea. Lengths, depth ranges, number of fish examined, percent containing food, and percent containing sediment are given in Table 2. Depth of maximum abundance was estimated to be the modal capture depth of all collections in which a given species occurred. Guts were maintained in 40% isopropyl alcohol following fixation in 10% formalin. The percentage of guts containing prey in which a prey item occurred (frequency of occurrence), number of individuals of each food type as a percentage of the total number of individual prey items (percent numerical abundance), and wet weight as a percentage of the total weight of all prey items (percent weight) were determined. Weights were measured by blotting prey items on f ilte r paper and weighing on an analytical balance. The index Table 2. Fishes examined in food habits study. 2 x .o x j c£ X Z 0 o J J u B B u 1 e o . a 4 c § s 5-3 I8 tc 01 B B V q *oC 3 : .s x b * tu •D o • h < s e 3 E «> B E E . o 0) c s o • O £ q NO m CO 0 0 •H CO CO CM r f • ON T n * n tr> r>» n n i T i o CO \D M•—• CM «ti r m o • oo * m r*- 0 0 CM o « m m n i p r NO CM n m Mm CM O OO-T - O tO MC m CM CM O n * n m o m H n n c l O , 15 of relative importance (IRI) of Pinkas, Oliphant, and Iverson (1971) was calculated: IRI « F(N + W) where F = frequency of occurrence, N = percent numerical abundance, and W = percent weight. Ontogenetic changes in diet were analyzed using the computer program SELECT (Vodopich and Hoover, 1981). The continuum option of this program systematically adjusts the size range of predators included in calculations and allows graphic presentation of data across a size continuum. This program in itia lly ranks predators of a given species by length. Food habits are then analyzed for groups of four fish at a time. Groups are formed beginning with the smallest four fish and then incrementing by the two next largest fish while deleting the two smallest of that group. This continues until the four largest individuals are grouped. Resulting IRI values for each group of four predators are then plotted for each prey taxon as a function of mean length of each group of f ou r. Cluster analysis of stomach content data was used to group species based on sim ilarity of diets as indicated by the IRI. This analysis was conducted only for the Bahamas study area. Cluster analysis was not attempted for MAB species due to the disparate data analysis methodologies utilized in previous food habits studies pertinent to the fauna of this region (Haedrich and Henderson, 1974; Haedrich and Polloni, 1976; Sedberry and Musick, 1978; Wenner, 1978; Farlow, 1980). The analysis was restricted to species characteristic of a depth range of 1000-2000 m, and for \*hich at least 20 specimens containing prey were available. Food habits data from the present study as well as those of Carter (1984) for ophidiid fishes were incorporated into this analysis. These combined data from fishes collected in the same sampling program include many species common in the Bahamas. Stomachs were clustered by treating predators as collections and following normal cluster analysis based on prey similarity as indicated by percent standardized IRI values. IRI values were percent standardized for each predator (because of the great variability in the sum of all IRI values for each predator) in a manner similar to that in which measures of abundance are standard!zed when unequal sample sizes are encountered (Clifford and Stephensen, 1975). Flexible sorting, with Beta=-0.25, was used along with the Bray-Curtis similarity measure (Clifford and Stephensen, 1975). The Bray-Curtis sim ilarity measure is: S ,= 1 - i ( where is the similarity between predator species "j" and "k", Xjk is the abundance of the "i"th prey species for entity " j ”; and X^- is the abundance of the "i"th attribute for 17 entity "K". Predator species are the entities and prey species are the attributes by which predators are classified. Body Composition Specimens for body composition analysis were frozen at sea immediately after capture and held frozen for up to 12 months before analysis. Fish were weighed at sea and immediately before analysis to correct for any water loss during storage. Specimens examined are listed in Table 3 (detailed collection data are given in Appendix 1). Dry weights of small fish were determined by cutting specimens into small pieces which were then weighed and dried to a constant weight at 55°-60°C. Large fish were ground whole in a meat grinder while frozen, and three subsamples of up to ten grams each taken of the resulting mixture. After drying, samples were pulverized with mortar and pestle and subsampled in triplicate for ashing at 500°C for five hours. Skeletal ash was estimated by subtracting the estimated solute ash (40% that of sea water) from total ash as done by Childress and Nygaard (1973). Carbon and nitrogen were determined on a Carlo Erba Elemental Analyzer model 1106, standardized with cyclohexanone 2-4 dintrophenylhydrazone ) at the National Marine Fisheries Service Laboratory in Beaufort, North Carolina. Table 3.3. FishesFishes examinedexamined in in body body composition composition study study. Length* Wet Wt Depth Range {m) Rax1mum family Species <=> (g) Minimum Maximum Abundance Myxinidae Myxine .s,lutinosa 250-400 17-71 lOB 613 285 Synaphobranehidae S.f.!!al!hobranchus ltaul!i 285-635 14-236 316 2200 918 S. affi~ 290-360 29-53 452 1550 699 S. brevidorsalis 335-640 25-301 918 2960 1496 Ilyophis brunneus 350-430 22-50 1109 3032 1568 Simencheli! parasiticus 275 24 SOl 1823 918 Haloaauridae Halosauropsia macrochir 240-270 183-311 1286 2933 1803 Aldrovandia affinia 135 29 1142 2188 1430 A. gracilis 82-174 6-50 1086 2354 1383 A. phalacra 8/.o-162 6-51 699 1645 1332 Alepocephal idae Narcetes stomias 495 1329 1478 2728 1848 Alepocepha1us agasaizii 235-342 130-417 658 2400 1470 Conocara macropterum 125-220 10-75 1239 2166 1404 Chlorophthalmidae Chlorophthalmus agassizii 93-100 8-9 <200 613 270 Bathypterois longipes 125-145 11-21 3710 5345 5043 .!.• phenax 134 15 1239 2748 1435 .!· b ige lowi 134 17 377+ 986+ 625+ Ipnops murrayi 80-117 1-6 1239 4539 1559 Bathytyphlops marionae 294 184 1075 1525 1311 Bathymicropa regis 109-110 3 4380 5345 5043 Lophiidae Lophius amerieanus 157-220 100-231 <200 818 (200 18 Ogcocephalidae Dibranehua atlanticus 45-145 1-63 <200 1142 486 I-' CXl Gadidae Urophycis ~ 238-341 113-368 bO rM 00 OH cn CO Ox o OH OI \D © tn OH sO O CM «o © m o HO o 0 0 p** CO r- OH r-. d vO fO Ox OH CM 00 CO Ox © -4 r^ i/\ d HO d 00 d d CM r > CN ** -4 •* is £ 4 x ec I' a• I I to X •O u « B 1 OO 3 ^ C x e r»» N m hO o CM hO O o i * -' J J * -t o 00 vO CO r l d tn d OH o d © o 1 x 3 Cb c o m Food Habits Squalidae Centroscyllium fabricii was taken in the MAB at depths from 712-1803 m but was not encountered in the Bahamas sampling area. The species reaches a length of at least 800 mm,TL (Bigelow and Schroeder, 1948) and submersible observations suggest that it is capable of active swimming (Sedberry and Musick, 1978). For this reason, (). fabricii is probably able to avoid trawls and may be more abundant than indicated by VIMS survey data. Stomachs of 28 C^. fab ricii were examined of which nine contained recognizable prey (Table 4). Cephalopods (IRI=4731), of the families Ommastrephidae and Lepidoteuthidae, were the most common food items, making up 81.5% of the diet by weight. Teleost remains (IRI=4463) were also common, as were decapods (IRI=760). These findings agree with previous studies which report cephalopods, decapods, and teleosts from C^. fabricii (Bigelow and Schroeder,1948; Clarke and Merrett, 1972; Sedberry and Musick, 1978; Mauchline and Gordon, 1983a). Ommastrephid and lepidoteuthid squids have previously been reported from the squaloid shark Centroscymnus coelolepis 20 21 Table 4. Gut contents of CentroscylHum fabricii. F = percent frequency of occurrence, W =* percent weight, N ** percent numerical abundance, IRI =* index of numerical abundance. Taxon F W NIRI M ollusca Cephalopoda Qnmastrephidae 11 0.7 6.3 77 Pholidoteuthis adami 11 30.4 6.3 407 Unidentified Cephalopoda 22 50.4 12.5 1398 Total Cephalopoda 44 81.5 2 5 .0 4731 C rustacea Mysidacea 11 0.1 6 .3 71 Decapoda 33 4.1 18.8 760 Unidentified Crustacea 11 0.7 6 .3 77 Total Crustacea 56 4 .9 31 .3 2009 T e le o s te i 78 13.6 4 3 .8 4463 Number examined 28 Percent with food 32 Percent with sediment 0 22 (Clarke and Merrett, 1972), which occurs at similar depths to those inhabited by C. fab ricii in the MAB. Another squaloid, Deania caIce us, was not taken on either VIMS or RSMAS cruises; however, a series of specimens taken aboard FFS Anton Dohrn on the Blake Plateau at depths of 8 3 0 - 1026 m was available for study. Of 13 J). calceus examined, only two contained recognizable prey (fish remains) including a myctophid (Lampanyctus sp.) and a melamphaeid (Table 5). One fish contained a large amount of sediment. Mauchline and Gordon ( 1983a) examined stomachs of 111 JD. calceus and found mostly fish along with smaller amounts of crustaceans and squid. Clark and Merrett (1972) reported on two eastern Atlantic specimens which contained unidentified teleost remains. Marshall and Merrett (1977) found teleosts, primarily myctophids, in eight eastern Atlantic specimens. Ophichthidae The ophichthid eel, Ophichthus cruentifer is abundant in the MAB, occurring at depths between 36-1350 m, with most specimens found shallower than 400 m (Wenner, 1976). Staiger (1970) reported this species from the continental side of the northern S traits of Florida but it is not known from the Bahamas study area where most collections have been at greater depths. Ophichthus cruentifer is morphologically adapted for a fossorial existence, and has been observed burrowing ta il f ir s t in aquaria (Wenner, 1976). This fossorial life style, 23 Table 5. Gut contents of Deania calceus. F 3 percent frequency of occurrence, W = percent weight, N 3 percent numerical abundance, IRI = index of relative importance. Taxon FWN IRI T e le o s te i Myctophidae 33 41.2 50.0 3010 Lampanyctus sp. Melamphaeidae 33 58.8 50.0 3590 Total Teleostei 66 100.0 100.0 13200 Number examined 13 Percent with food 15 Percent with sediment 8 24 suggests that C). cruentifer is probably more abundant than indicated by trawl catches. Of 45 guts examined only five contained recognizable prey consisting of polychaetes (IRI=7806) and crustaceans (IRI=1914) (Table 6). All identifiable crustaceans were gammarid amphipods. In keeping with its fossorial habits, 20% of (). cruentif er guts examined contained significant amounts of sediment. These findings are similar to those of Wenner (1976) who reported crustaceans, including crabs (Cancer sp.) and an amphipod, a sipunculid, and polychaete remains from seven specimens of (). cruentif er . Goode and Bean (1896) suggested that 0 . cruentifer was a parasitic boring species feeding on other fishes. There appears to be little evidence for this contention. Instead benthic infauna and epifauna appear to be the dominant prey. Synaphobranchidae Synaphobranchus brevidorsalis is one of the most abundant species in the Bahamas study area between depths of 1000-3000 m, but is rare in the MAB where only one specimen has been collected. Sulak (1982) reports the species from more than 75% of successful collections from 918-2354 m in the Bahamas study area. The species is one of the largest synaphobranchid eels reaching a length of at least one meter. Despite the abundance of S_. brevidor salis, its life history has not been previously studied. Only ten (16%) of the 64 S^. brevidorsalis examined contained recognizable prey (Table 7). Teleosts (IRI=2159) 25 Table 6. Gut contents of Ophichthus. cruentifer. F = percent frequency of occurrence, W = percent weight, N a percent numerical abundance, IRI a index of relative importance. Taxon F W N IRI P olychaeta 60 80.1 50.0 7806 C rustacea M phipoda Gammaridea 20 5 .4 16.7 441 Unidentified Crustacea 40 14.5 3 3 .3 1914 Total Crustacea 60 19.9 50 .0 4194 Number examined 45 Percent with food 11 Percent with sediment 20 26 Table 7. Gut contents of Synaphobranchus brevldorBaliB. F =* percent frequency of occurrence, W = percent weight, N = percent numerical abundance, IRI *» index of relative importance. Taxon F W NIRI M olluscs Cephalopoda 10 <0.1 11.1 111 Crustacea Dec a pod a 30 15.3 33.3 1459 Teleostei 30 3 8 .6 33.3 2159 Debree 10 4 1 .4 11.1 525 Humber examined 64 Percent with food 16 Percent with sediment 0 27 were most commonly encountered, followed by decapod (IRI=l459) and cephalopod remains (IRI=111). All prey items were highly digested precluding any specific identifications. Some material appeared to have been scavenged including plastic fo il from one specimen and the upper and lower jaws of a shelf dwelling tetraodontiform fish. Scavenging has been suggested as a feeding mode for synaphobranchid eels by Robins (1968). Moreover, Merrett and Marshall (1981) reported the remains of fishes too large to have been ingested whole in S^. kaupi guts and suggested scavenging. The diet of S_. brevidorsalis has not been previously studied; however, its congener S. kaupi, a common species in the MAB and Bahamas, has been examined by Sedberry and Musick (1978), Farlow (1980), Saldanha (1980), and Merrett and Marshall (1981). These authors found teleosts, cephalopods, and decapods to be the main food items. Based on submersible observations, Sedberry and Musick ( 1978) suggest that _S. kaupi is capable of capturing active prey. The more flaccid body composition of brevidorsalis (mean water content = 87. 1%) when compared with other synaphobranchi ds such as S^. kaupi, jJ. a f f in i s , or Ilyophis brunneus (mean water contents of 76.5%, 76.7%, and 76.2%, respectively) is obvious when handling these species, and is discussed more completely be lo\ This, in conjunction with the finding of apparent debris and scavenged material in S^. brevidorsalis guts, suggests that S^. 28 brevidorsalis is more dependent on scavenged prey than other synaphobranchid eels. Halosauri dae Halosaurs of the genus Aldrovandia are abundant members of both the MAB and Bahamas fish assemblages at depths between 1000-2000 m. The four species examined here, af fi n i s . .A. gracilis, A_. phalacra, and A^. oleosa, are morphologically similar and are often found together in the same trawl (McDowell, 1973; Sulak, 1977a). Sedberry and Musick (1978) have reported on the food habits of Aldrovandia from the MAB. All specimens examined in the present study were taken in the Bahamas study area. In the Bahamas, _A. af finis occurred over a depth range of 1239-2188 m. Between depths of 1436-1977 m it occurred in 90% of successful trawls (Sulak, 1982). One hundred specimens of _A. affinis were examined of which 94 contained prey. The species fed most heavily on crustaceans (IRI=16048) with copepods (IRI=4483) the most abundant taxa (Table 8). Polychaetes (IRI=1947) were also important occurring in 68% of the fish examined. Significant amounts of sediment were found in 13% of the fish examined. L ittle evidence of diet shifts with size was found (Figs. 6 and 7) except that decapods and teleosts, taxa rarely consumed by A_. af fin i s , were found only in large specimens. Sedberry and Musick ( 1978) examined 11 A_. affinis from the MAB and found the major prey items to be polychaetes, benthic mollusks, ostracods, mysids, and ophiuroids, as well as copepods, amphipods, and sediment. 29 Table 8. Gut contents of Aldrovandia affinis. F “ percent frequency of occurrence, W a percent weight, N “ percent numerical abundance, IRI =■ index o f r e la t iv e im portance. Taxon F WN IRI P olychaeta 68 15.8 12.8 1947 M olluscs Scaphopoda 2 <0.1 0 .4 1 Gastropoda 1 0.1 0 .4 1 Pelecypoda 1 0.1 0 .2 1 Total Mollusca 4 0 .2 1.0 5 Crustacea Ostracoda 14 0 .1 3 .2 46 Copepoda 74 2.5 5 7 .7 4482 Mysidacea 1 <0.1 0 .2 1 Cumaeea 4 <0.1 1 .0 4 Tanaidacea 28 0 .3 7 .2 208 Isopoda 3 0.1 0 .6 2 Amphipoda Gammaridea 17 0 .2 4 .6 82 Decapoda 1 2.1 0 .2 2 Unidentified Crustacea 57 78.6 10.8 5136 Total Crustacea 95 83.9 85 .6 16048 Echinodermata Ophiuroidea 2 0.1 0 .4 1 T e le o s te i 1 <0.1 0 .2 1 Number examined 100 Percent with food 94 Percent with sediment 13 30 Figure 6 IRI values of dominant prey taxa plotted against predator size for Aldrovandia affinis. l/l O Nl Nl 1-2 kg < Q LlJ □Qj< >• UJLJ LxJ O M < m M O 2 1 X CL UJ tn p NJu in t/5 \-2 31 Figure 7 IRI values of dominant prey taxa plotted against predator size for Aldrovandia affinis. ALDROVANDIA AFFINIS CD L y igc! I g 2 y ^ 03 0 >- Q S BEcn l J QL Q Z -o •Zf ^_ot.(Kapu)ia d o JvS I tSH i => ry o c S W W V) 32 McDowell (1973) reported copepods, amphipods, pelecypods, and polychaetes from 25 _A. af fin is. Aldrovandia gracilis was the most abundant member of the genus found in Bahamian waters and occurred over a depth range of 1086-2354 m. Between 1239-1515 m the species occurred in 99% of successful collections (Sulak, 1982). Of 103 A^ g racilis guts examined, 91% contained food. Copepods (IRI=1386) were the most abundant prey occurring in 44% of those specimens examined, but accounted for only 1.7% of the diet by weight (Table 9). Polychaetes ranked second in importance (IRI=9241) and were found in 23% of the fish examined. As was found for _A. af fi n i s , there were no apparent ontogenetic diet sh ifts in A_. g r a c ilis . However, decapods and teleosts were consumed only by fish larger than about 150 mm,GPL and cumaceans seemed to decrease in importance with increasing size (Figs. 8 and 9). Decapods and teleosts occurred only infrequently in _A. gracilis guts, yet comprised a significant portion of the biomass consumed (Table 9). Natant decapods were present in only three specimens, but accounted for 13.2% of the diet by weight. Sediment occurred in 9% of those fish examined. Sedberry and Musick (1978) reported an amphipod, a pelecypod, and bivalve fragments from two MAB _A. gracilis. McDowell (1973) reported on 45 specimens which contained crustacean and polychaete remains. Copepods, the dominant 33 Table 9. Gut contents of Aldrovandia gracilis. F = percent frequency of occurrence, W = percent weight, N = percent numerical abundance, IRI ° index of relative importance. Taxon F W N IRI Hexactinellida 1 <0.1 0.4 1 Polychaeta 23 0.8 9.4 241 Pycnogonida 1 0.2 0.4 1 Crustacea Ostracoda 14 0.2 6.0 85 Copepoda 44 1.7 30.0 1386 Mysidacea 5 0.6 2.6 17 Cumaeea 9 0.1 5.2 45 Tanaidacea 10 0.1 3.9 38 Amphipoda Gammaridea 5 0.1 2.6 14 Dec a pod a 3 13.2 1.3 46 Unidentified Crustacea 91 75.2 36.9 10257 Total Crustacea 98 91.2 88.4 17576 T eleostei 3 7.8 1.3 29 Number examined 103 Percent with food 91 Percent with sediment 9 34 Figure 8 IRI values for dominant prey taxa plotted against predator size for Aldrovandia gracilis. £ - j 180 200 m 160 C/JX as t/)$ X(< 1/1 ^ 8 c UO t/j 120 t n > 100 o I— "T < dw o Q1 ^jai.fwpwjia 0 < Q Z 1 C£ O CD ‘ 3 120 120 UO KO no 200 50 SJZE Or FlSH(mm) Or FlSH(mm) SJZE Or nSH(mm) SIZE o 35 Figure 9. IRI values of dominant prey taxa plotted against predator size for Aldrovandia gracilis. ALDROVANDIA GRACIUS i a J U size or nsn(mfn) 36 prey in Bahamian A^. g r a c ilis , were not found by either Sedberry and Musick or McDowell, Aidrovandia phalacra occurred at depths between 918-1434 m and occurred in 75% of successful collections at depths of 918-1093 m (Sulak, 1982). All of the 51 Aj_ phalacra guts examined contained food consisting mostly of crustaceans (IRI=19793), with copepods (IRI=1443) the most frequently encountered taxa (Table 10). Copepods occurred in 43% of the fish examined and made up 32.1% of the diet by number, but accounted for only 1.4% of the diet by weight. Natant decapods, which occurred in only two specimens, comprised the bulk of the diet by weight making up 53% of the total prey. Decapods were consumed only by fish of about 160 mm,GPL or more (Fig. 10). No sediment was found. The findings reported here for Bahamian _A. phalacra are similar to those of Sedberry and Musick (1978) for MAB fish. Aldrovandia oleosa is the least common of the four Aldrovandia species considered in this study. The species occurred at depths between 1282-1977 m and was present in 45% of successful trawls between 1436-1977 m (Sulak, 1982). Of the 21 specimens examined, 90% contained prey. The diet of Aldrovandia oleosa consisted almost entirely of crustaceans (IRI=17794) with copepods (IRI=686) most prominent, followed by mysids (IRI=118) (Table 11). Most guts contained relatively few prey, and copepods were the only prey encountered more than once. No guts were found to contain sediment. Sedberry and Musick (1978) reported polychaete 37 Table 10. Gut contents of Aldrovandia phalacra. F = percent frequency of occurrence, W ■ percent weight, N 3 percent numerical abundance, IRI ° index of relative importance. Taxon F W N IRI Polychaeta 2 0.1 0.9 2 Mollusca Gastropoda 2 0.1 0.9 2 Crustacea Ostracoda 10 0.1 7.5 75 Copepoda 43 1.4 32.1 1443 Mysidacea 6 0.7 2.8 21 Tanaidacea 18 0.5 12.3 226 Amphipoda Gammar idea 2 <0.1 0.9 2 Dec a pod a 4 53.4 2.8 221 Unidentified CruBtacea 82 43.7 39.6 6859 Total Crustacea 100 99.8 98.1 19793 Number examined 51 Percent with food 100 Percent with sediment 0 38 Figure 10. IRI values of dominant prey taxa plotted against predator size for A Id rovand ia phalacra. ALDROVANDIA PHALACRA 8 £ £ 2 I < g UJ < r s o _ X X r ■ a N l/J oi*(«pw)ia s r Size or FlSH(mm) 39 Table 11. Gut contents of AldrovandjLa oleosa. F 3 percent frequency at occurrence, W 3 percent weight, N 3 percent numerical abundance, IRI 3 index of relative importance. Taxon FW N IRI Crustacea Ostracoda 5 0.2 3.8 21 Co pe pod a 26 3.0 23.1 686 Mysidacea 5 18.7 3.8 118 Unidentified Crustacea 84 73.7 61.5 11388 Total Crustacea 95 95.5 92.3 17794 T eleostei 5 0.2 3.8 21 Unidentified Tubes 5 4.3 3.8 43 Number examined 21 Percent with food 90 Percent with sediment 0 40 remains from seven MAB oleosa, but no traces of polychaete were found in Bahamian specimens. In their discussion of MAB Aidrovandia, Sedberry and Musick (1978) suggested the possibility of resource partitioning, with A^. af finis feeding more heavily on polychaetes and mollusks than other Aldrovandia. Bahamas data suggest that _A. affinis does feed more heavily on polychaetes than other Aldrovandia species. Polychaetes made up 15.8% by weight and 12.8% by number of the diet of _A. af f in i s , but were relatively less important to h_. gracilis where polychaetes accounted for 9.4% of the diet by number but only 0.8% by weight, and were rarely or never consumed by phalacra and oleosa. Furthermore, sediment was found only in gracilis and A_. af finis guts suggesting that these species may feed somewhat more heavily on benthic prey than other Aldrovandia examined. Mollusks were not important components of the diet of any Aldrovandia species in the Bahamas. The diets of A_. gracilis, _A. phalacra, and _A. oleosa appear to overlap broadly with little evidence of resource partitioning. Bahamas data differ from those of McDowell (1973) and Sedberry and Musick (1978) in that natant decapods and teleosts were occasionally included in the diet of large Aldrovandia from the Bahamas. Sedberry and Musick concluded that the small mouths of Aldrovandia probably restict their diets to small prey items, however large individuals appear capable of at least occasionally feeding on larger prey such as decapod shrimp and teleosts. Indeed, the diets of large 41 Aldrovandia may be similar to that of Halosauropsis macrochir at similar sizes. The latter species has been reported to feed heavily on natant decapods by Sedberry and Musick (1978). Notacanthidae Polyacanthonotus sp . A (Sulak, Crabtree, and Hureau, in press) was collected in both the MAB and Bahamas study areas at depths between 1000-2000 m. The species occurs at tropical and subtropical latitudes throughout the Atlantic, but has previously been confused with _P. af ricanus, a synonym of P^. challengeri (Sulak et al., in press; Crabtree et al., in press). Crabtree et al. have estimated the numerical abundance of Polyacanthonotus sp. A in the Bahamas to be 9 fish km 2 at depths between 1200-1500 m and in the MAB to be 12 fish km2 between 1300-1600 m. At depths of 1239-1515 m in the Bahamas Polyacanthonotus sp. A occurred in 59% of successful collections. Of the 109 stomachs of Polyacanthonotus s p . A examined, 44 contained prey items (Table 12). In order of importance by the IRI, crustaceans were the dominant prey (IRI=5139) with gammaridean amphipods predominating (IRI=590). Polychaetes ranked second (IRI=3405), but were the dominant prey by weight making up 67.6% of the total prey biomass. Identification of most prey items to lower taxonomic categories was generally precluded due to the advanced stage of digestion of food. Polyacanthonotus rissoanus is widely distributed in the O Atlantic but absent in the western Atlantic between 30 N and 30°S (Crabtree et al., in press) and did not occur in the 42 Table 12. Gut contents of Polyacanthonotus sp. A. F =* percent frequency of occurrency N 3 percent numerical abundance, W 3 percent w ight, IRI 3 index of relative importance. Taxon FWN IRI Foraminifera 2 1.9 0.1 4 Porifera Hexactinellida 2 0.6 1.9 6 Polychaeta 36 67.6 27.8 3405 Crustacea Copepoda 7 0.8 5.6 45 Mysidacea 5 2.5 5.6 39 Anphipoda Gammaridea 24 4.4 20.4 590 Unidentified Crustacea 2 23.0 24.1 1366 Total Crustacea 60 30.8 55.6 5139 Unidentified Tubes 19 1.1 14.8 303 Number examined 109 Percent with food 40 Percent with sediment 3 43 Bahamas study area. Polyacanthonotus rissoanus has been taken at depths between 402-2293 m in the Atlantic and occurred between 1190-1823 m in the MAB. Crabtree et al. estimate the numerical abundance of rissoanus to be 12 fish km at^2. depths between 1400-1800 m. Of the 50 stomachs of p_. rissoanus examined, 30 contained prey (Table 13). Crustaceans (IRI=8842), primarily gammaridean amphipods (IRI=3858) and mysids (IRI=1154), and polychaetes (IRI=8031) were dominant prey. As was found for Polyacanthonotus sp. A, polychaetes were the dominant prey by weight, making up 74.8% of the diet. Polyacanthonotus sp. A and P. rissoanus can be categorized as predators of the benthic macrofauna, consuming small benthic crustaceans and polychaetes. The diets of the two species were similar except that Polyacanthonotus sp. A consumed relatively fewer mysids than P^. rissoanus. It is unknown if this difference reflects feeding selectivity or differences in prey abundances between the Bahamas and MAB. These findings correspond with those of McDowell (1973) with respect to dominant prey items, but do not support McDowell's contention that crustaceans dominate the diet of _P. rissoanus, while polychaetes dominate that of "P^ africanus". McDowell's ecological conclusions with respect to "P_. africanus" are suspect because of his failure to distinquish Polyacanthonotus sp. A from challengeri. Sessile invertebrates seem less important in the diets of the small-mouthed species of Polyacanthonotus than in those of 4 4 Table 13. Gut contents of Polyacanthonotus rissoanus. F = percent frequency of occurence, N = percent numerical abundance, W percent weight, IRI =• index of relative importance. Taxon F W N IRI Polychaeta Polynoidae 34 55.7 9.9 2264 Unidentified Polychaeta 52 19.1 8.4 1423 Total Polychaeta 86 74.8 18.3 8031 Crustacea Myaidaea 48 5.6 18.3 1154 Amphipoda Gammaridea Ampeliscidae 3 0.4 0.5 3 Aoridae (Unicola sp.) 3 <0.1 0.5 2 Cressa sp. 3 <0.1 0.5 2 Eusiridae 24 1.5 3.7 124 Oedicerotidae 3 <0.1 0.5 2 Harpina sp. 3 0.1 0.5 2 Pleustidae 3 0.2 1.6 6 Stegocephalidae 7 0.2 1.6 13 Stegocephalus auratus 3 0.5 1.0 5 Podoceridae 3 0.1 5.8 20 Photidae (Podoceropsis sp.) 3 0.1 0.5 2 Unidentified Gammaridea 59 5.1 28.3 1957 45 Table 13. Continued. Taxon F W N IRI Total Gammaridea 72 8.3 45.0 3858 Caprellidae 7 3.4 10.5 96 Total Amphipoda 72 11.6 55.5 4862 Unidentified Crustacea 31 8.0 7.8 490 Total Crustacea 83 25.2 81.7 8842 Number examined 50 Percent with food 60 Percent with sediment 2 46 other notacanthiform fishes. The Atlantic species of Notacanthus, _N. chemnitzi. and JJ. bonapar te i , have specialized premaxillary teeth which form a continuous serrate cutting edge probably used to crop sessile invertebrates (McDowell, 1973). In keeping with this adaptation sea anemones and corals were reported by McDowell in stomachs of N_. chemnitzi. Lozano Cabo (1952) found bryozoans and colonial hydrozoans, as well as copepods and amphipods in a large series of N_. bonapartei. In contrast to Notacanthus, species of Polyacanthonotus do not possess specialized cutting teeth (McDowell, 1973) and only one sessile invertebrate (hexactinellid sponge in Polyacanthonotus sp. A) was found in Polyacanthonotus stomachs. The larger mouths of halosaurs probably enable them to ingest a broader size spectrum of prey items than consumed by Polyacanthonotus. Fish, decapod crustaceans, and infaunal bivalves have been reported from the stomachs of halosaur species (see above). These items were not found in Polyacanthonotus. Additionally, there is evidence that at least in some areas halosaurs apparently tend to indiscriminately root in the sediment as indicated by Sedberry and Musick's (1978) finding of sediment in 48.2% of Halosauropsis macrochir stomachs and 20.5% of Aldrovandia stomachs. This contrasts with Polyacanthonotus where sediment occurred in only one stomach of P^. rissoanus. Furthermore, only 8.4% of bahamas specimens of Aldrovandia examined contained sediment, thus the tendency to swallow sediment may reflect sediment type and/or prey availab ility . In this 47 regard it may be significant that benthic mollusks comprise a more important component of the diet in MAB Aldrovandia (Sedberry and Musick, 1978) than in Bahamas Aldrovandia. Alepocenhalidae Conocara macropterum is an abundant member of the Bahamas deep-sea fish assemblage between 1239-2166 m and occurred in 97% of successful stations between 1239-1515 m (Sulak, 1982). The species is known from the Gulf of Mexico, the Caribbean, and from off Brazil in the western Atlantic as well as from the eastern Atlantic (Krefft, 1973). Conocara macropterum was not captured in the MAB. The diet of macropterum changed dramatically with increasing size. Polychaetes, copepods, and ostracods decreased in importance with increasing fish size (Fig. 11). Guts of fish less than 175 mm,SL contained mainly crustaceans (IRI=12422) with copepods (IRI=2537) and ostracods (IRI=1252) the most abundant taxa (Table 14). Polychaetes (IRI=1196) were also of importance. At sizes greater than 175 mm decapods (IRI=1294) and teleosts (IRI=1914) were important prey (Fig. 12) and together made up 53.5% of the diet by weight (Table 15). Salps were important components of the diet at all sizes (Fig. 12). The diet of C_. macropter um has received little attention by previous workers. Bright (1968) examined one Gulf of Mexico specimen and found polychaete and crustacean remains. Mauchline and Gordon (1983b) examined guts of three species of 48 Figure 11. IRI values of dominant prey taxa plotted against predator size for Conocara macropterum. 3 _l rfl x o 2 2 UJ o Q. t—X 8 r-fl 01-(MpuOKN i< 3 X 1/9 § g or 2 § rxiLdJ 'ft 49 Table 14. Gut contents of Conocara macropterum 175 mm, SL or less. F » percent frequency of occurrence, W ° percent weight, N “ percent numerical abundance, IRI ™ index of relative importance. Taxon F W N IRI Polychaeta 43 20.3 7.7 1196 Mollusca Gastropoda 3 0.2 0.5 2 Pelecypoda 3 3.2 0.5 11 Total Mollusca 6 3.4 1.0 25 Crustacea Ostracoda 51 3.9 20.4 1252 Cope poda 43 9.2 50.0 2537 Mysidacea 3 0.9 0.5 4 Cumacea 11 0.8 5.1 67 Unidentified Crustacea 57 26.8 10.2 2116 Total Crustacea 97 41.7 86.2 12422 Thaliacea 23 34.3 4.1 878 T eleostei 6 0.4 1.0 8 Number <175 mm examined 38 Percent with food 92 Percent with sediment 11 50 Figure 12. IRI values of dominant prey taxa plotted against predator size for Conocara macropterum. THAUACEA DECAPODA ^jn^MpuQm r>o o d> * LJ o m I. I 0 200 ISO 300 330 400 SIZE OF riSH(mm) 51 Table 15. Gut contents of Conocara jnacropterum larger than 175 um, SL. F « percent frequency of occurrence, W =* percent weight, N ° percent numerical abundance, IRI “ index of relative importance. Taxon F W N IRI Anthozoa 4 11.1 1.0 45 Polchaeta 19 0.2 5.0 97 Mollusca Scaphopoda 15 0.2 10.0 150 Gastropoda 22 4.3 12.0 362 Pelecypoda 9 1.0 2.5 32 Cephalopoda 6 0.2 1.5 9 Ctnmastrephidae 2 0.1 0.5 1 Cranchiidae 2 0.4 0.5 2 Unidentified Cephalopoda 6 0.2 1.5 9 Total Cephalopoda 9 0.7 2.5 29 Unidentified Mollusca 2 <0.1 0.5 1 Total Mollusca 43 6.1 27.5 1431 Crustacea Ostracoda 17 0.1 8.5 143 Cope poda 15 0.1 10.0 150 Cumacea 2 <0.1 1.0 2 Amphipoda Gammaridea 2 0.1 0.5 1 Decapoda Oplophoris (spinicaudata?) 9 8.5 2.5 99 Glyphocrangon longirostris 2 2.6 0.5 6 52 Table 15. Continued. Taxon F W N IRI Unidentified Caridean 2 0.5 0.5 2 Paguridae 2 2.0 0.5 5 Unidentified Decapoda 19 13.3 5.5 357 Total Decapoda 31 31.6 9.5 1294 Unidentified Crustacea 32 3.3 7.1 330 Total Crustacea 74 34.7 57.6 6874 Thaliacea 37 25.2 13.0 1415 T eleostei Myctophidae 4 14.7 1.0 58 Unidentified Teleostei 48 7.2 14.0 1020 Total Teleostei 52 21.9 15.0 1914 Unidentified Tubes 4 0.3 1.5 7 Unidentified Eggs 2 0.2 1.5 3 Number >175 mm examined 62 Percent with food 87 Percent with sediment 47 53 benthopelagic alepocephalids from the eastern North Atlantic and found prey similar to those reported here for macropterum. Three Alepocephalus rostratus guts contained decapods and unidentified crustaceans. Eleven Alepocephalus agassizi contained mostly mysids along with polychaetes, amphipods, and cephalopod and teleost remains. A large series of 886 Alepocephalus bairdii contained a wide variety of prey including polychaetes, medusae, copepods, amphipods, euphausiids, and fish, along with framented salps and/or ctenophores. Moreover, A^ bairdii displayed a shift in diet similar to that reported here for (]. macropterum with smaller fish feeding more on polycheates, copepods, and amphipods, and larger fish more heavily on medusae, mysids, squid, and teleosts. Golovan and Pakhorukov (1975) found coelenterates, anglerfishes, and salps (Pyrosoma sp.) in guts of _A. bairdii and noted an absence of benthic prey from its diet. Markle (1976) reported mud, ctenophores, calanoid copepods, crustaceans, urchin tests, and polychaetes from MAB _A. a g a s s iz ii. Golovan and Pakhorukov (1980) found coelenterates, salps, decapods, teleosts, cephalopds, pteropods, and polychaetes in guts of Alepocephalus rostratus from off the coast off west Africa. The consumption of benthic mollusks by C^. macropter um increased with predator size (Fig. 11). This could reflect the increased ingestion of sediment by larger fish (Fig. 13). Overall, sediment occurred in 35% of the guts examined. Mixed with this sediment were large numbers of epipelagic pteropods. These pteropods and probably some of the scaphopods, 54 Figure 13 Frequency of ingestion of sediment plotted against fish size for Conocara macropterum. 100 o A 0 N 3 n 03 dJ CN o r o r S * ° r o r~ O CN o o ° o SIZE OF FISH(mm) 55 gastropods, and bivalves listed in Tables 14 and 15 could well have been empty shells when ingested and of no nutritive value. The habit of consuming large quantities of sediment has been reported for Gulf of Mexico £. macropterum (Bright, 1968), and by Mauchline and Gordon (1983b) who found sediment in 15% of 886 guts of Alepocephalus bairdii examined. The purpose of ingesting sediment is unknown; however, sediment does not appear to have been incidentally ingested in the trawl since large quantities were often found well back in the intestine just above the anus. Conocara niger is known from the tropical western Atlantic as well as from the Indian Ocean and western Pacific (Krefft, 1973). Conocara niger was not taken in the MAB. In the Bahamas study area the species occurred at depths between 1363-2404 m and in 20% of successful collections between 1587- 2354 m (Sulak, 1982). In general, C. niger occurred at greater depths and was much less abundant than C^. macropterum. The diet of C. niger was similar to that of (3. macropterum with salps (IRI=10683) and crustaceans (IRI=2642) the major prey (Table 16). Unlike its congener, £. niger guts contained no traces of decapod or teleost remains. Furthermore, only one of the specimens examined contained sediment contrasting with 35% of the C_;_ macropter um examined. The diet of C^. niger has not been previously studied. Narcetes stomias is known from the Atlantic, eastern Pacific, and Indian oceans (Markle, 1976; Sazonov and Ivanov, 1980). In the Bahamas and MAB it was taken between 1478-2728 56 Table 16. Gut contents of Conocara -niger. F = percent frequency of occurrence, W = percent weight, N =* percent numerical abundance, IRI ° index of relative importance. Taxon F W N IRI Polychaeta 15 <0.1 2.7 42 Crustacea Ostracoda 31 0.1 9.5 293 Copepoda 23 0.1 17.6 407 Amphipoda Gammaridea 15 6.2 2.7 137 Unidentified Crustacea 38 0.1 6.8 264 Total Crustacea 62 6.4 36.5 2642 Thaliacea 69 93.5 60.8 10683 Number examined 25 Percent with food 52 Percent with sediment 4 57 m. The species occurred in 37% of successful stations in the Bahamas between depths of 1586-2354 m. In the MAB only six specimens were taken. Narcetes stomias appears adapted to a predaceous life style. Its large body size, mouth gape, and dentition suggest that it is capable of feeding on large nektonic organisms. Unfortunately, of the 30 specimens examined only five contained recognizable prey including natant decapods (IRI=9403) and the beaks of an onchoteuthid squid (Table 17). The biology of JN. stomias has not been studied previously. Chlorophthalmidae The family Chlorophthalmidae was an abundant component of the Bahamian deep-sea fauna at all depths sampled, but was rarely encountered in the MAB. All specimens examined for food habits were from the Bahamas. Chlorophthalmids are benthic fishes (Sulak, 1977b; Heezen and H ollister, 1971), resting on the bottom on their pectoral and caudal fin rays, which may be quite elongated in Bathypterois. Bathypterois longipes is known only from depths below 3500 m where it is one of the more abundant species encountered. Sulak (1982) reports the species from 17% of successful collections below 2745 m in the MAB and 46% of successful collections below 4246 m in the Bahamas. Of the 23 specimens examined in this study, 16 contained recognizable prey. Crustaceans, including copepods (IRI=377), gammarid amphipods (IRI=156) and one decapod (IRI=432), were included in the diet (Table 18). Polychaete setae were found in one 58 Table 17, Gut contents of Narcetes -stomias. F m percent frequency of occurrence, W - percent weight, N = percent numerical abundance, IRI = index of relative importance. Taxon F W N IRI Mollusca Cephalopoda Onychoteuthidae 20 3.2 20.0 465 Crustacea Decapoda 60 96.7 60.0 9403 Unidentified Crustacea 20 <0.1 20.0 401 Total Crustacea 80 96.8 80.0 14140 Number examined 30 Percent with food 17 Percent with sediment 0 59 Table 18. Gut contents of Bathypternia longipea. F a percent frequency of occurrence, W = percent weight, N ** percent by number, IRI “ index of relative importance. Taxon F W N IRI Polychaeta 6 0.2 4.8 31 Crustacea Copepoda 19 1.1 19.0 377 Amphipoda Gammaridea 13 3.0 9.5 156 Decapoda 6 63.0 4.8 423 Unidentified Crustacea 81 32.8 61.9 7692 Total Crustacea 94 99.8 95.2 18284 Number examined 23 Percent with food 70 Percent with sediment 9 60 specimen. Marshall and Merrett (1977) reported copepods from a single eastern Atlantic specimen of £5. longipes. Bathypterois grallator was captured at depths between 1239-3032 m in the Bahamas. Although never taken in large numbers, it occurred in 71% of successful stations between 2519-3094 m in the Bahamas (Sulak, 1982). Bathypterois gralla tor attains a length of at least 368 mm.SL, larger than other members of the genus (Sulak, 1977b). The species has greatly elongated pelvic and caudal fin rays which are capable of propping large individuals at least a meter off the bottom. A shift in diet with size was apparent in 15. grallator. Copepods and gammarid amphipods were included in the diet at all sizes (Fig. 14) and were the major prey items of fish less than 240 mm,SL (Table 19). Mysids were found only in fish larger than 200 mm. At sizes greater than about 240 mm decapods and teleosts were prominent prey items (Fig. 15) with teleosts making up 78.7% of the diet by weight (Table 20). The diet of J3. grallator has been previously studied by Marshall and Merrett (1977) who reported crustaceans from a single specimen. Bathypterois phenax was taken from 1239-2748 m in the Bahamas where it occurred in 93% of successful stations between 1436-1977 m (Sulak, 1982). The species was rare in the MAB, where only three specimens were captured. All of the 44 specimens examined contained prey. Crustaceans, primarily copepods (IRI=4058), ostracods (IRI=879), and gammarid amphipods (IRI=239), were the dominant prey (Table 21). 61 Figure 14. IRI values of dominant prey taxa plotted against predator size for Bathypterois grallator. BATHYPTEROIS GRALLATOR M 31 8 o l £ r J o i 2 3 o 5 230 300 330 Size or nSH(mm) 62 Table 19. Guts contents of Bathvptarois grallator less than 240 mm, SL. F =* percent frequency of occurrence, W = percent weight, N =■ percent numerical abundance, IRI = index of relative importance. Taxon F W N IRI Polychaeta 10 1.7 4.9 63 Crustacea Oatracoda 10 0.4 4.9 51 Co pe poda 33 5.1 26.8 1064 Mysidacea 10 2.8 4.9 73 Amphipoda Gammaridea 24 13.6 14.6 672 Unidentified Crustacea 76 71.1 39.0 8392 Total Crustacea 100 93.0 90.2 18324 T eleostei 5 1.7 2.4 20 Unidentified Tubes 5 3.6 2.4 29 Number <240 mm examined 23 Percent with food 91 Percent with sediment 0 63 Figure 15. IRI values of dominant prey taxa plotted against predator size for Bathypterois grallator. fr_0T *(xopui)|a) 250 300 350 150 200 250 300 350 SI2E OF FlSH(mm) FlSH(mm) OF SI2E FlSH(mm) OF SIZE ot •(xopu)ia 6 4 Table 20. Gut contents of Bathypterois grallator 240 mm, SL or larger. F =* percent frequency of occurrence, W =* percent weight, N = percent numerical abundance, IRI 53 index of relative importance. Taxon F WN IRI Polychaeta 5 3.5 3.3 34 Crustacea Copepoda 25 0.8 16.7 438 My sidacea 15 1.9 10.0 179 Amphipoda Gammaridea 25 2.6 20.0 566 Decapoda 15 10.8 10.0 313 Unidentified Crustacea 25 1.3 16.7 449 Total Crustacea 85 17.5 73.3 7721 T eleostei 30 78.7 20.0 2962 Unidentified Tubes 5 0.3 3.3 18 Number >240 ran examined 24 Percent with food 83 Percent with sediment 0 65 Table 21. Gut contents of Bathypterois phenax. F =* percent frequency of occurrence, W 3 percent weight, N * percent numerical abundance, IRI 3 index of relative importance. Taxon F W N IRI Polychaeta 27 3.7 5.3 245 Crustacea 03tracoda 43 2.4 18.0 879 Copepoda 75 7.6 46.5 4058 Mysidacea 2 0.7 0.4 3 Cumaeea 9 0.3 1.8 19 Tanaidacea 14 3.6 3.1 90 Iso poda 5 0.2 0.9 5 Amphipoda Gammaridea 30 1.9 6.1 239 Decapoda 2 0.2 0.4 1 Unidentified Crustacea 89 79.4 17.1 8550 Total Crustacea 100 96.2 94.3 19049 T eleostei 2 0.1 0.4 1 Number examined 44 Percent with food 100 Percent with sediment 0 66 Polychaete remains were found in 27% of the specimens examined with decapod and teleost remains found in single specimens. Most of the fish examined were similar in size (range 95-153 mm,SL), thus no size related trends were discovered. The diet of ji. phenax has not been studied previously. Unlike the related Bathypterois, the caudal and pelvic fin rays of Ipnops murrayi are not greatly elongated and the species probably rests close to the bottom. Ipnops murrayi is rare in the MAB where only one specimen was taken; however, it is abundant in the Bahamas where 962 individuals were taken at depths between 1239-4539 m. It was most abundant between 1586-2359 m in the Bahamas where it occurred at 96% of successful stations (Sulak, 1982). Polychaetes were the major prey found in guts of JE. murrayi and made up 68.9% of the diet by weight and 50.8% by number (Table 22). In addition, bivalves occurred in four specimens, which in conjunction with the presence of polychaetes suggests benthic feeding. Crustaceans were also ingested with copepods (IRI=660) the most frequently occurring taxa. These findings agree with those of Nielsen (1966), Marshall and Staiger (1975), and Bright (1968). The small size range (62-115 mm,SL) available for study precluded an investigation of size related dietary shifts. Bathymicrops regis is a small abyssal species which, like I_. murrayi, lacks the greatly elongated fin rays found in Bathypterois and apparently rests directly on the bottom. Though absent from MAB collections, IJ. regis occurred in 31% of successful collections below 4246 m in the Bahamas (Sulak, 67 Table 22. Gut contents of Ipnops murrayi. F = percent frequency of occurrence, W «* percent weight, N = percent numerical abundance, IRI = index of relative importance. Taxon F W N IRI Polychaeta 74 68.9 50.8 8910 Mollusca Pelecypoda 9 3.0 7.9 102 Crustacea Copepoda 26 8.3 17.5 660 Amphipoda Gammaridea 2 0.8 1.6 5 Unidentified Crustacea 33 18.9 22.2 1340 Total Crustacea 56 28.0 41.3 3868 Number examined 51 Percent with food 84 Percent with sediment 0 68 1982). Of 18 II. regis guts examined, only 18% contained recognizable prey, consisting entirely of crustacean remains. Nielsen (1966) reported crustacean remains including amphipod fragments and other crustacean remains from three specimens. Qgcocephalidae Dibranchus atla n tic u s, is a benthic species common in the MAB and on the Blake Plateau at depths shallower than about 1300 m. The species is also known from Northwest Providence and Santaren channels in the Bahamas and from the Florida S traits (Staiger, 1970). For comparative purposes, specimens from both the MAB and southern Blake Plateau were examined. As expected based on external morphology, atlanticus fed mainly on benthic species. Echinoderms (IRI=5198), primarily ophiuroids (IRI=1550), along with polychaetes (IRI=2608), mostly Hvalinoecia a r ti f le x , were the major prey (Table 23). Mixed crustaceans and benthic mollusks were also consumed but were of relatively little importance. These data contrast somewhat with those of Rayburn (1975) who found crustaceans and bivalves to be the major prey of Gulf of Mexico JO. atla n tic u s. Stomach contents of MAB and Bahamas specimens of JO. atlanticus were quite different. Of the 16 stomachs with food examined from MAB collections, 81.3% contained H_;_ a rtif lex or unidentified polychaete remains, and only 12.5% contained echinoderm remains. In contrast, 88.9% of the 18 specimens examined from the Blake Plateau contained ophiuroid or 69 Table 23. Gut contents of Dibranchus atlanticus. F >* percent frequency of occurrence, W =* percent weight, N = percent numerical abundance, IRI = index of relative importance. Taxon FW N IRI Polychaeta Hyalinoecia artiflex 11 31.6 5.5 401 Unidentified Polychaeta 30 11.8 15.4 809 Total Polychaeta 41 43.5 20.9 2608 Mollusca Gastropoda 3 0.9 2.2 8 Pelecypoda 3 <0.1 1.1 3 Total Mollusca 5 0.9 3.3 23 Crustacea Mysidacea 3 0.1 3.3 9 Tanaidacea 3 0.1 1.1 3 Amphipoda Gammaridea 3 <0.1 1.1 3 Euphausiidae 3 0.6 1.1 5 Unidentified Crustacea 5 <0.1 2.2 12 Total Crustacea 14 0.8 8.8 130 Echinodermata Ophiuroidea 32 7.1 40.7 1550 Unidentified Echinodermata 35 44.8 14.3 2074 Total Echinodermata 49 51.9 54.9 5198 Unidentified Tubes 27 2.9 12.1 405 Number examined 66 Percent with food 56 Percent with sediment 6 70 unrecognizable echinoderm remains, and none contained polychaetes. All Blake plateau specimens came from a single trawl station, thus these data can not be interpreted as indicative of regional differences. Instead, these data suggest that I), atlanticus can feed opportunistically on what ever acceptable benthic prey is most abundant. Macrouridae Nezumia bairdii is one of the most abundant species from the MAB at depths shallower than about 1500 m but is rare in the Bahamas study area. Though macrourids are one of the most studied groups of deep-sea fishes, the diet of N_. bairdii has received little previous study. The diet of N_. bair dii showed a pronounced ontogenetic shift. The importance of polychaetes declined with increasing predator size (Fig. 16), while decapods and teleosts increased in importance (Fig 17). Fish less than 35 mm,HL fed largely on crustaceans, primarily gammarid amphipods (IRI=5181) and copepods (IRI=740). Polychaetes (IRI=1420) were also an important dietary component (Table 24). At sizes above 35 mm, decapods (IRI=3644) were the major prey item and made up 71.0% of the diet by weight (Table 25; Fig. 17). Gammarid amphipods (IRI=2177), copepods (IRI=458), polychaetes (IRI=236), and mysids (IRI = 238) were s t i l l included in the diet of large N^. b a ird ii, but were relatively less important than to small individuals. The shift from benthic to more pelagic feeding at larger sizes by Jjl. baird ii suggested by these findings is similar to that described by Pearcy and Ambler (1974) who 71 Figure 16. IRI values of dominant prey taxa plotted against predator size for Nezumia bairdii. -t8 o 8 < ^_ot. (x*pui)mi CD M o 5 3 , a £LU 1 . o .o nSH(mm) OF o Q_ Size Size -e 4* r» O o6 o o t _ot*(»puOia 72 Figure 17. IRI values of dominant prey taxa plotted against predator size for Nezumia bairdii. NEZUMIA BAIRD1I g 1 1 Q 0 £ Q f S .o -8 ry u I L l J fi fS SIZE Or FlSH(mm) 73 Table 24. Gut contents of Nezumia bairdii leas than 35 mm HL. F “ percent frequency of occurrence, W *» percent weight, N =< percent numerical abundance, IRI a index of relative importance. Taxon F W N IRI Polychaeta Fragments 45 24.0 7.2 1420 Mollusca Pelecypoda 9 0.3 1.4 16 Crustacea Copepoda 41 2.3 15.8 740 Mysidacea 9 5.7 2.2 72 Tanaidacea 5 0.2 0.7 4 Isopoda 5 1.4 0.7 10 Am phi pod a Gammaridea 77 12.4 54.7 5181 Caprellidea 5 0.2 0.7 4 Total Amphipoda 77 12.6 55.4 5250 Decapoda 5 15.0 0.7 71 Unidentified Crustacea 50 37.8 7.9 2286 Total Crustacea 91 74.9 83.5 14397 Unidentified Tubes 14 0.5 5.0 76 Unidentified Eggs 5 0.2 2.9 14 Number <35 mm examined 58 Percent with food 83 Percent with sediment 0 7 4 Table 25. Gut contents of Nezumia bairdii 35 mm, HL or larger. F = percent frequency of occurrence, W = percent weight, N 53 percent numerical abundance, IRI a index of relative importance. Taxon FWN IRI Folchaeta Fragments 35 3.3 3.4 236 Mollusca Gastropoda 2 <0.1 0.2 1 Pelecypoda 19 0.4 3.2 68 Unidentified Remains 2 0.4 0.2 2 Total Mollusca Crustacea Copepoda 33 0.3 13.7 458 Mysidacea 30 1.0 6.9 238 Cumaeea 5 <0.1 0.5 2 Amphipoda Gammaridea 58 1.5 35.9 2177 Decapoda Sergestes arcticus 23 46.3 6.9 1237 Sergestes? 12 19.4 3.4 266 Unidentified Decapods 9 5.3 1.1 60 Total Decapoda 44 71.0 11.4 3644 Unidentified Crustacea 63 11.0 6.2 1080 Total Crustacea 100 85.0 74.6 15957 Priapulida 7 3.5 1.1 33 75 Table 25. Continued. Taxon F W N IRI Echinodermata Ophiuroidea 2 0.1 0.2 1 T eleostei Cyclothone microdon 2 0.5 0.2 2 Lampanyctus sp 2 4.4 0.2 11 Unidentified Teleostei 5 1.3 0.5 8 Total Teleostei 9 6.2 0.9 66 Unidentified Tubes 33 0.8 4.1 162 Unidentified Eggs 2 0.1 11.2 26 Number > 35 mm examined 46 Percent with food 94 Percent with sediment 2 76 found a shift from benthic to pelagic prey among macrourids off the coast of Oregon. The dominance of decapods in the diets of large JN. bairdii is misleading. Of the 20 specimens containing decapods, 13 came from a single trawl station (FFS Anton Dohrn, Cruise 213-3, St. 6416, 36°23'N, 74°43'W, 815 m). Thirteen of the 14 specimens containing food at this station contained natant decapods. This could reflect a concentration of sergestid shrimp near the bottom and available as prey to _N. b a ird ii. Murdoch, Avery, and Smyth ( 1975) and Love and Ebeling (1978) have shown that shallow water fishes will switch from feeding on less abundant prey to near exclusive feeding on a single greatly abundant prey taxon. Some evidence of this was presented above for Dibranchus atlanticus. In a similar manner, JN. bairdii may feed almost exclusively on a single prey resouce, if that resource is in great abundance. The diet of N_, bair dii is similar to that reported for other Nezumia species except that decapods may be important in the diet of N_. bair dii (with the reservations noted above). Farlow (1980) examined guts of samples including both _N. bairdii and JN. aequalis but failed to distinquish the two species. He found small crustaceans including gammarid and caprellid amphipods, copepods, mysids, and euphausiids to be dominant prey. In addition, pycnogonids and teleost remains were reported, Bigelow and Schroeder (1953) reported euphausiids, amphipods, and polychaetes from JN. bairdii off 77 Georges Bank. Du Buit ( 1978) examined eastern Atlantic N_. aequalis and found crustacean remains and ophiuroids. Macpherson (1979) reported polychaetes, crustaceans, including gammarid amphipods and mysids, and ophiuroids from Mediterranean N.. aequalis. Marshall and Merrett ( 1977) reported mixed crustaceans, primarily copepods and ostracods, and polychaetes from eastern Atlantic N_;_ aequalis. Bright (1968) found polychaetes, gammarid amphipods, copepods, and cumaceans in Gulf of Mexico specimens of N_. aequalis (referred to as N_. hildebrandi) . The macrourid, Ventrifossa occidentalis. was taken in the MAB at depths to 647 m. Staiger (1970) reported V_. occidentalis from the continental side of the Florida S traits from the Gulf of Mexico to Miami at depths from 190-672 m. The species was not taken in Bahamian collections, all of which were below its depth range. Stomachs of V_. occidentalis contained mostly gammarid amphipods and natant decapods (Table 26). Evidence of a shift in diet was apparent (Fig. 18) with decapods consumed primarily by specimens larger than 30 mm,HL. Teleosts, which were of minor importance in the diet of _V. occidentalis. were consumed only by the largest specimens examined. These findings are similar to those of Marshall and Merrett (1977), who reported natant decapods from the stomachs of four eastern Atlantic _V. occidentalis. Marshall and Iwamoto ( 1973) reported euphausiids and gammarid amphipods from a single specimen. 78 Table 26. Gut contents of Ventrifossa occidentalis. F “ percent frequency of occurrence, W 3 percent weight, N = percent numerical abundance, IRI = index of relative importance. Taxon F W N IRI Mollusca Cephalopoda 2 0.1 0 .2 1 Crustacea Ostracoda 2 <0.1 0 .2 1 Mysidacea 10 1.0 3.9 48 Cumaeea 5 0.1 0.6 3 Isopoda 2 0.1 0.6 2 Amphipoda Gammaridea 56 19.4 85.5 5889 Decapoda 49 63.7 5.7 3384 Unidentified Crustacea 37 9.6 2.9 457 Total Crustacea 100 93.9 99.4 19331 Teleostei 5 6.1 0 .4 31 Number examined 57 Percent with food 72 Percent with sediment 0 79 Figure 18. IRI values of dominant prey taxa plotted against predator size for Ventrifossa occidentalis. VENTRIFOSSA OCCIDENTALIS Q -S tsl i u o Vi u. a a k o s ^_oi- (K»puQia ss Size or FiSH(mm) SIZE OF FI5H(mm) 80 Stephanoberycidae Stephanoberyx monae was abundant at depths between 1000- 2000 m in the Bahamas where it occurred at more than 60% of successful stations between 918-1515 m (Sulak, 1982). The species was not taken in the MAB. Stephanoberyx monae is relatively small in size with the largest specimen examined only 115 mm,SL. The size range examined (77-115 mm,SL) was limited thus no analysis of ontogenetic diet shifts was possible. Of the 64 J3. monae examined, 81% contained prey. The diet consisted almost entirely of crustacea, with gammarid amphipods (IRI=2366), mysids (IRI=1557), copepods (IRI=1074), and tanaids (IRI=953), the most commonly encountered prey items (Table 27). Bright (1968) examined nine Gulf of Mexico specimens and found gammarid amphipods, copepods, taniads, and cumaceans to be the dominant prey. Cottidae The benthic Cottunculus thompsoni was occasionally taken in the MAB at depths from 749-1719 m. Of 18 stomachs examined 10 contained food, most of which was in an advanced state of digestion (Table 28). Teleost and gammarid amphipods were the most frequently occurring prey. Most teleosts were present only as eye lenses or occasionally skull bones, and the actual weight ingested may be grossly underestimated. 81 Table 27. Gut contents of Stephanoberyx monae. F “ percent frequency of occurrence, W =* percent weight, N =* percent numerical abundance, IRI 13 index of relative importance. Taxon F W N IRI Polychaeta 13 1.6 2.6 57 Crustacea Ostracoda 27 3.2 8.2 309 Cope pod a 46 4.2 19.1 1074 Mysidacea 37 30.6 12.0 1557 Tanaidacea 48 2.2 17.6 953 Isopoda 4 0.1 0.7 3 Amphipoda Gammaridea 56 19.6 22.8 2366 Unidentified Crustacea 87 38.4 16.9 4782 Total Crustacea 100 98.4 97.4 19574 Number examined 64 Percent with food 81 Percent with sediment 0 82 Table 28. Gut contents of Cottunculus thompsoni. F ■ percent frequency of occurrence, tf » percent weight, N = percent numerical abundance, IRI = index of relative importance. Taxon F WN IRI Polychaeta 20 5.3 2.1 148 Mollusca Cephalopoda 20 2.4 2.1 89 Crustacea Copepoda 10 <0.1 1.0 11 Amphipoda Gammaridea 70 59.0 52.1 7773 Caprellidae 10 0.2 1.0 13 Total Amphipoda 70 59.2 53.1 7863 Unidentified Crustacea 50 6.1 5.2 567 Total Crustacea 100 65.4 59.4 12475 T eleostei 70 26.9 35.4 4362 Unidentified Tubes 10 <0.1 1.0 11 Number examined 18 Percent with food 56 Percent with sediment 0 Cluster Analysis Cluster analysis produced four major predator groups with one composed of two subdivisions (Fig. 19). The diets of all predators considered overlap to some extent; however, each predator group has one or more characteristic prey taxa consumed by all members. Group 1 includes predators of small crustaceans, primarily copepods and ostracods, as well as polychaetes. Within this group two subgroups are apparent. Subgroup A, feeding heavily upon copepods and polychaetes, contains Aldrovandia affinis, Dicrolene kanazawi. and small (<175mm,SL) Conocara macropterum. Subgroup B contains Aldrovandia gracilis . A_. phalacra, A^ oleosa, Bathypterois phenax, and Porogadus s ilu s . The members of this subgroup tend to feed more intensively on copepods than members of subgroup A, with other taxa such as polychaetes, assuming relatively less importance. This tendency to feed extensively on copepods is most pronounced in P.. s ilu s , where copepods make up 75.5% of the diet by number and 64.0% by weight (Carter, 1984). Feeding group 2 includes small (<240mm,SL) Bathypterois g ra lla to r, Bathyonus p ecto ralis, Stephanober yx monae, and Barathrodemus manatinus, species which feed almost entirely on 83 84 Figure 19. Species groupings based on cluster analysis of gut contents data for dominant species from the Bahamas study area. kanazawi o ir> V CQ CO •H CO o V 00 PQ E 0 0 tn u B 3 « Q3 u D m o o IT) T to _o .1 85 crustaceans. As in group 1, copepods are a major prey, but in group 2 gammarid amphipods and mysids are also important. Barathrodemus manatinus is included in this group, but differs from other members by consuming large quantities of tanaids which make up 44.6% of the diet by number and 29.5% by weight (Carter, 1984). In addition, II. manatinus feeds more heavily on polychaetes than other members of group 2, a feature which suggests that .B. manatinus might be more appropriately placed in group 1. Cluster analysis includes the species in group 2 probably because of sim ilarities between the diets of _B. manatinus and Stephanoberyx monae, which also consumes significant numbers of tanaids. The diet of II. manatinus is somewhat specialized and the species does not fit well into any grouping. Group 3 includes species \vhich feed heavily on decapods and teleosts. Bassozetus normalis and Dicrolene intronigra feed more heavily on natant decapods (Carter, 1984) while large (>240 mm,SL) Bathypterois grallator and large (>175mm,SL) Conocara macropterurn are more piscivorous. Dicrolene intronigra is included in this group because it feeds heavily on decapods, which make up 68% of the diet by weight (Carter, 1984). However, 1). intronigra is anomalous because it also feeds heavily on isopods which make up 48.3% of its diet by number. Finally, the diet of (I. macropter urn is also divergent with salps making up 25.2% of the diet by weight. Group 4 contains Polyacanthonotus sp. A (Sulak et al., in press), Acanthonus armatus, Barathrites p a rri, and Ipnops 86 murrayi all of which feed most heavily on polychaetes. The most atypical member of the group is _B. parri where polychaetes occurred in 98% of the fish examined, with elasipod holothurians the only other significant prey taxa making up 63% of the diet by weight (Carter, 1984). Body Composition Body composition parameters can be analyzed in several manners which require grouping of species for comparative purposes. Species can be grouped taxonomically and body composition parameters compared among groups. Perhaps more instructive however, is to group species ecologically and compare composition parameters among groups. In this study, species are broken into several groups based on the position in the water column typically occupied by a species and the mechanism by which the species maintains its vertical position, in this case presence or absence of a swimbladder. Benthic species typically rest directly on the bottom and include chlorophthalmids, zoarcids, the goosefish Lophius americanus, Pibranchus atlanticus, and Glyptocephalus cynoglossus. In keeping with their benthic life styles, none of these species possess swimbladders. The hagfish, Myxine glutinosa, which often borrows into the sediment (Bigelow and Schroeder, 1953) is also a benthic species, however it is excluded from ecological groupings because of its distant phylogenetic position. Benthopelagic species typically occur some distance off the bottom. This group includes species with and without 87 88 swimbladders. Benthopelagic species with swimbladders include the synaphobranchid eels, halosaurs, merluciids, ophidiids with the exception of Acanthonus armatus, macrourids, Antimora rostrata, and He1icolenus dactylopterus. Somewhat equivocal are the gadids, Urophycis chuss, JJ. tenuis, and Phycis chesteri, which have been observed resting directly on the bottom (Sulak, 1982); however, because they often occur in the water column, possess swimbladders, and feed mostly on pelagic prey (Bigelow and Schroeder, 1953; Sedberry and Musick, 1978) they are here considered benthopelagic. Several benthopelagic species do not possess swimbladders yet have been observed hovering motionless in the water column (Sulak, 1982). These include the Alepocephalidae and the ophidiid, Acanthonus armatus. Water Content Mean percent water ranged from 73.2% for Helicolenus dactylopterus to 91.9% for Acanthonus armatus (Table 29). Overall the families Alepocephalidae and Ophidiidae were characterized by the highest water contents, contrasting with zoarcids and chlorophthalmids which had much lower water contents (Table 30). Halosaurs and macrourids were intermediate in water content. Somewhat atypical were the synaphobranchid eels with three species Synaphobranchus af finis ( mean = 76 . 7% ) , S^. kaupi (mean = 76 . 5%) , and I lyophis brunneus (mean=76.2%) having water contents among the lowest of any species examined, and in contrast £3. bre vid o rsalis. with a water content of 87.1%; among the highest of any 1.0 89OJ 3) 36; of AFDW* % 61(0. Cirllon number 49. 49.50 52.93(2.14;2) 48.45(0.99;4) 52.08(0.71;3) 47.32 55.19(2.14;3) 48.58(1.93;3) 52.99(4.12;3) and mean 79;2) the AFDW* 52(1. of Nitro-gen % 13.03 14.88(0.06;3) 12. 14.83(0.39;4) 14.65 14.37(0.41;3) 14.69{0.64;3) 14.03(0.38;3) 12.21(1.17;3) error Aah Wt 1.8 2.2 1.4 2.7 1.7 3.7 2.5 3.0 1.4 0.8 1.9 0.3 2.2 2.0 1.8 1.7 l.5 2.0 2.7 2.3 1.1 Wet standard % lrl(eletal are 2;2) Wt l. 1.9;8) Ash Dry parentheses 15.1 16.8(0.6;4) 20.5(0.3;3} 17.5 11.9 20.1( 17.2(1.2;2} 19.6 8.8(0.4;6) % 26.8(0.6;7) 12.8 19.0(0.8;5) 16.1(1.0;7) 18.7(0.4;4) 15.4 14.2(0.4;5) 10.9 12.8(0.8;3) 21.5( 16.3(0.8;8) 14.4(2.0;ll) in 7) ll) Values 7;8) Wt 1.0;4) 1.3; 7(0. Wet Water 80.7(0.1;2) 78.2(0.5;3) 76.1(0.6;2} 87.1( 80.0(0.2;5) 87.3(0.5;5) 77.0 78.1 % 79.2 76.7(0.4;2) 78.8(0.8;2) 89.3(0.5;5) 89.9(0.5;7) 87.7 !10.9(0.9; 82.1(0.1;4) 79.8 75.2 79.6(0.8;5) 76.2(0.9;3) 76. 87.1(0.9;9) 76.5{ 75.4(0.7;6) fishes. deep-sea kauEi agassizii marionae agassizii macrochir regis longipes affinis atlanticua farasiticua ~ selected stomias macropterum brunneus americanua murrayi of g1utinoaa tenuis phenax pha1acra. !?.!&elovi gracilis brevidorsatie affinia Urophycia lpnops .!!· Bathytyphlopa BathymicroEs Dibranchus l.ophiua Bathypteroia e~amined. Chlorophtalmus Alepocepha1us Conocara .!!.· Narcetes A. .!!.· ~· Halosauropsis Aldrovandia 11yophia S. S. Simenchelys Species SlnaEhobranchus Hyxine co~position idae idae Body specimens 29. Ogcocephal Gadidae Lophiidae Chlorophthalmidae Alepocephal Halosauridae Synaphobranchidae ~·lllllily Hyxinidae Table 0 90U) AFI:N* % Carbon 59.67(0.85;2) 49.02 47.82 57.56(4.03;3} 55.40(1.2;3) 48.80(0.68;3) 49.15(0.42;3) 48.24 44.41 56.74(0.90;2) 3) 72;2) 1.00; AFOW* 20( 70(0.12;3) NLtrogen % 14.35 11.45(0.01;2) 11.81 13.21 12. 16.69(1.03;3) 14.45(0.53;3) 13. 12.99 ll.54(0. Ash Wt 1.8 2.7 3.4 3.3 2.2 1.8 2.3 1.7 1.3 2.7 3.3 3.2 2.9 1.8 1.7 4.3 1.6 2.5 1.8 4.5 2.0 3.2 1.8 Wet % Skeletal 7) ;2) ;6) 7;4) 0; 0;9) 2;2) 7;4) Wt 1.1 1. 1.2;7) 1.2;3} I. 1. 1.1 5( 7(0. 5( Ash Dry .6( 16. 19.4 19.9(0.8;5) 16.1(0. 16.5(0.3;4) J; 13.8( 18.6 13.5( 17.90.2;4) 19.1(1.1;4) 18.2(0.7;8) 17 27.4(4.8;5) 19.6 28.6 17.6 37.2( 25.3(2.2;3) 27.7 15. 15. 21.2(0.2;3) 11.4( 7) 7;5) 0; Wt 1. 2(1.0;2) 2( 3(0.9;9) Wet Water 73. 77.1(0. 77.5(0.8;5) 78.2(06;7) % 84.8 84.9(1.0;5) 79.5{0.3;4) 80.9(0.4;4) 82. 84.3 80.0(0.4;4) 75.8(0.8;14) 78.1(0.6;6) 91.9(1.1;3) 89.5 80.3(0.4;5) 17. 19.7 84.2 85.3(2.3;3) 79.8(0.4;3) 81.3(0.4;3) 80.2(0.9;5) 74.9(1.0;7) weight. dry cynoglossu~ rupestris ca~inatus gahthea free atlanticum verrilli daetylopterus £• armatus bilinearis pectoralis silua normalis intronigra ash rostrata atlantieus bairdii chesteri for paxilla le~ia armatus earapinus aequalia albiduiJ Helicolenus L. Ly~oi~a L~nchelys Helanosti£!! Glyptocephalus c. Antimora C. Coryphaenoides C. Coelorichua Bathyonus Nezumia Abys_sobrotula Aeanthonus Porogadua Basozetus !!• B.~ Herlucciua Dicrolene Species Phycia !:!· abbreviation idae Continued. an t idae idae i is 2'J. ty rpaen uronec Ple *AFDW is an *AFDW abbreviation for ash free dry weight. *AFDW Sco Zoarcidae Moridae Macrouridae rami Ophidiidae Mer1ucc Table 91 *-x •/■s N“N 0^4 d Mt CM >n • m. to 4 4k 4a #a 'O o M3 CO C 2 O' d o Hi c- o o • • ■ • * XJ a* o CM CM In < %■«/ v>^ w a in O o ST d CO CJ K CN CM in r- M3 O' cm o O' vO O' CM in in *a in sy in 4—p o-s n <—x d St CM d e •• • m • 4 4* ■ a 43 2 *y d d r* in • oo a r- m O' \D o species examined. Benthopelagic species (mean =80,2%) with swimbladders and benthic species (mean=79.3%) had similar water contents; however, benthopelagic species without swimbladders had significantly (Mann-Whitney U test) higher water contents with a mean of 89.7% (Table 31). Water content tended to increase with depth when only benthopelagic species were considered (Fig 20). Benthic species examined had uniformly low water contents (range 76.1- 79.2%) regardless of capture depth with the exception of three shallow living species, Lophius americanus, Dibranchus atlanticus, and Glyptocephalus americanus, which had water contents in excess of 85%. Increased water content with depth probably reflects the increased abundance of benthopelagic species without swimbladders at greater depths such as all alepocephalids and the ophidiid, Acanthonus armatus, which had high water contents and were relatively more important at depths below 1000m. Ash Free Dry Weight (AFDW) Skeletal ash (% dry weight) ranged from 0.3% in Narcetes stomias to 4.5% in Bassozetus normalis (Table 29). Trends among taxa in ash content are d iffic u lt to discern; however, as is the case with percent water, trends are evident when species are grouped by life style. Benthopelagic species without swimbladders had mean skeletal ash contents of only 1.1%, significantly (Mann-Whitney U test; p<0.05) lox^er than benthic (mean=2.6%) or benthopelagic species with swimbladders (mean=2.3%) (Table 31). 93 x : co • O n c CN CO r - cu CS CN CO CO 0) cu CU 0 rH a < Q w ’W ' 0) cu CO CN o rH 0 > *0 4 CU X c h - < r X 4J CO CN CO c cu • e 4* Xl CO CO x u 0 CO TZJ 0) E QN CO d "O CN co TO 0 H CO X CO Figure 20. Water content of benthopelagic species plotted against capture depth (r*=0.187; p<0.01; n=156). Points represent values for each specimen examied for the species listed in Table 29 and as grouped according to life style in the text. ET (m) DEPTH O o IO o o o ro 8 o CJ .o .o •8 r . 1 . 1 ------1 ------1 lN30U3d • • • :il i*. - . • • • • • • ••• •• •• U3±m ------•• • ••• ••• •• • . . • • •!*•»*•• • • • • o O o CD o f - i 1 •• •• i i 001 r 95 Ash as percent dry weight was positively correlated with percent water ( r J=0.415; p<0.001; Fig 21), and loosely correlated with depth ( r J=0.073; p<0.001; F ig.22) although there was much scatter present. Skeletal ash showed no significant correlations with depth. Positive correlations between water and ash could reflect the replacement of tissue in the form of lipid or protein by water in more flaccid species and a consequent increase in ash as a percentage of dry weight. The greater variation (mean=27.4%, standard error=4.8) in ash content found among individuals of Bathyonus pectoralis than in other species is noteworthy. The five individuals examined had ash contents of 18.3, 21.2, 22.3, 30.2, and 44.9% dry weight. All five specimens examined came from two trawl stations, both at depths of about 2700 m, thus the observed differences do not reflect adaptations to different depths. Furthermore, all five fish were similar in size ranging from 122-157 mm. It is unlikely that this wide range of values is artifactual since there is no single obvious outlier. Instead the values are evenly spread over the range of 18-45%, It is interesting that J3. pectoralis occurs over an extremely wide depth range of at least 3900 m (1438-5345 m). Perhaps the apparent p lasticity of the species in terms of body composition may in part explain its ability to survive over such a large depth range. 96 Figure 21. Ash as percent dry weight plotted against percent water for selected deep-sea fishes ( r a=0.415; p<0.01; n=167). Points represent values for each specimen examined of the species indicated in Table 29. HSV XN30y3d 97 Figure 22. Ash as percent dry weight plotted against capture depth for selected deep-sea fishes (r’=0.073; p<0.001; n=167). Points represent values for each specimen examined of the species indicated in Table 29. • 10o Depth Depth fm) S. *. f M * * +« •* -t>+****£ ■**+* t« r* msv 98 Carbon Percent carbon ranged from 44.4% AFDW in Basso zetus normal is to 59.7% in Coryphaenoides rupestris (Table 29), and showed no apparent patterns. Correlations of percent carbon with other parameters show a great deal of scatter casting doubt on their significance. Nevertheless, carbon as percent AFDW showed a slight negative correlation with depth (Fig. 23). This correlation showed much less scatter when carbon as percent wet weight was correlated with depth (Fig. 24). Changes in carbon largely parallel changes in lipid content as shown by Childress and Nygaard (1973) for midwater fishes, thus these trends are suggestive of a slight decrease in lipid content with increasing depth. Nitrogen Nitrogen as percent AFDW ranged from a low of 11.45% in Coryphaenoides rupestris to 16.69% in Bathyonus pectoralis (Table 29). No trends are evident and overall variations in percent nitrogen were relatively small. As percent wet weight, nitrogen showed a loose but significant negative correlation with depth (Fig. 25); however, as percent AFDW nitrogen showed a weak but significant positive correlation with depth (Fig. 26). The weakness of these correlations and their conflicting nature suggest that they have little biological meaning. 99 Figure 23. Carbon as percent ash free dry weight plotted against capture depth for selected deep-sea fishes ( r a=0.099; p<0.05; n=43). Points represent values for each specimen examined of the species indicated in Table 29. < < DEPTH (M) DEPTH < < <33 <<^<1 < 2000 3000 4000 5000 <3 < < < O J- ° <<3 < I------1------1------1------1------m o m o io o to to LO in ^ (m q j v %) Noayvo 100 Figure 24. Carbon as percent wet weight plotted against depth of capture for selected deep-sea fishes ( r 3=0.261; p<0.01; n=43). Points represent values for each specimen examined of the species indicated in Table 29. 20 -i <<< < n m o in im J. m i ( T <1 < < < < 1000 2000 3000 4000 5000 DEPTH (M) 101 Figure 25. Nitrogen as percent wet weight plotted against capture depth for selected deep-sea fishes ( r J=0.203; p<0.01; n=43). Points represent values for each specimen examined for the species indicated in Table 29. o “ o o DEPTH (M) DEPTH <1 < _ o o m m m cn ^ o ( iM 13M %) N390H1IN 102 Figure 26. Nitrogen as percent ash free dry weight plotted against depth for selected deep-sea fishes ( r a=0.102; p<0.05; n=43). Points represent values for each specimen examined for the species indicated in Table 29. DEPTH (M) DEPTH 2000 3000 4000 5000 _ o «J « l 00 CO o co (M0dV%) N300U1IN Discussion Predator Groups The cluster analysis used in this study is limited by several shortcomings and should be considered preliminary. The resolution of the analysis is limited by a failure to break species into size groups as delineated by ontogenetic diet shifts. Limitations in the available food habits data allow only two of the species considered to be sp lit into size groups which are then treated as separate en tities. For most species data are inadequate to allow the detection of ontogenetic diet shifts either because of small sample sizes or inadequate representation of the entire species size range. For some species this limitation may be negligible since only a small range of sizes are present; i.e . Ipnops murrayi, Bathymicrops regis, and Stephanoberyx monae among others, which do not attain large sizes. Many other deep-sea species probably under go pronounced shifts in diet with increasing size as has been pointed out by Macpherson (1981). Large predaceous species such as Synaphobranchus kaupi, S^. bre vidorsalis , Narcetes stomias , Bathy saurus mollis , II. f erox , Bassozetus species, and others probably feed on small invertebrates when small, shifting to larger prey with increasing size. 103 104 An additional problem with the cluster analysis is that prey were identified only to major taxon. Thus species which cluster together do not necessarily overlap in diet, since they may have identical diets with regard to major taxa yet share no common prey species. The value of this analysis is in its grouping of predators based on definable criteria, and its indication of groups of species where dietary overlap is possible. Finally, the analysis may be criticized for its reliance on the index of relative importance (Pinkas et a l., 1971). However, any index of importance combining such different measures of prey importance as frequency of occurrence, percent numerical abundance, and percent weight, is likely to be biased in some way. Furthermore, clusters based soley on frequency of occurrence, percent number, or percent weight produced obviously unrealistic pairings and were unsatisfactory. The IRI is more heavily influenced by frequency of occurrence than by either percent numerical abundance or by percent weight. This may be desirable in studies of species such as deep-sea fishes where most prey items are highly digested, precluding accurate determinations of either weight or abundance. For a number of species only a small amount of published data were available. Because of the incomplete nature of these data, these species were not included in the cluster analysis; however, it is possible to discuss potential placement of these species in the feeding groups defined by 105 cluster analysis based on a consideration of the available information. Data presented earlier in the present study suggest that Conocara niger can be grouped with its congener £. macropterum. The 13 C. niger guts which contained prey showed no evidence of feeding on either decapods or teleosts. This could be an artifact of the small sample size, particularly since many of the specimens examined were small in size. It is probable that Cl. niger displays an ontogenetic shift in diet similar to that described for C. macropterum with decapods and teleosts consumed increasingly by large individuals. Insufficient data are available to test this hypothesis. A number of predaceous species which feed on teleosts, decapods, and squid are present in the Bahamas study area in addition to those included in the cluster analysis. Marcetes stomias, Bathysaurus mollis, and j3. ferox are large predators which consume teleosts, squid, and decapods as shown by the limited data reported here on N_. stomias and by Sedberry and Musick ( 1978) and Sulak et al. (in prep.) on 13. mollis and J3. ferox. Additionally, four common synaphobranchid eels, Synaphobranchus kaupi , 53. bre vidorsalis, S^. af finis , and Ilyophis brunneus also probably feed to some extent on teleosts, cephalopods, and decapods. Sedberry and Musick (1978) found teleosts and cephalopods to be the major prey of 5^. kaupi. Data presented in this study suggest a similar diet for S_. brevidorsalis perhaps with a greater dependence on scavenged prey. The diet of Ilyophis br unneus and 53. af fin is 106 have not been studied in detail but based on morphological similarities are probably similar to those of S^. kaupi and S^. brevidorsalis . The diets of these large predators probably include larger and more mobile prey than the diets of any of the species grouped in predator group 3. One of the most obvious features of these feeding groups is that they do not closely reflect taxonomic relationships. Unrelated species often appear to have similar diets, while taxonomically related forms cluster into different feeding groups. For instance Conocara macropterurn and Bathypterois grallator were each broken into large and small size groups and large individuals of the respective species showed more dietary sim ilarity to unrelated species than to small individuals of the same species. Cluster analysis produced a number of groupings including species with grossly different morphologies and life styles. In each group unrelated taxa appear to have achieved similar diets through quite different feeding mechanisms. Examples are found in each of the feeding groups defined. All of the groups constructed contained benthopelagic species with and without swimbladders as well as benthic forms. Corresponding with differences in life styles, differences in body composition are apparent within groups. For example, feeding group 1 contains the benthopelagic halosaurs of the genus Aldrovandia, which possess swimbladders and have water contents ranging from 79.8% to 82.1% (Table 29). In contrast, small individuals (<175 mm,SL) of the 107 benthopelagic Conocara macropterum, are also included in this group. This species does not have a swimbladder, and appears to accomplish neutral bouyancy through a high water content (mean=89.9%, Table 29) and reduced skeletal ossification (mean skeletal ash content=1.4%; Table 29). Finally, group 1 also contains a benthic species, Bathpterois phenax, which has a relatively low water content (mean=79.2%) and a well ossified skeleton (mean skeletal ash content=3.0%). Group 4 is characterized by species which feed heavily on polychaetes and also contains species exhibiting a diverse array of life styles and body compositions. This group includes a notacanth, Polyacanthonotus sp. A (Sulak et al., in press), the ophidiids, Acanthonus armatus and Barathrites p a rri, and the chlorophthalmid Ipnops murrayi. Polyacanthonotus sp. A and jl. parri are both benthopelagic species with swimbladders. Acanthonus armatus, a benthopelagic species without a swimbladder, achieves neutral bouyancy via a high water content (mean=91.9%), low skeletal ash content (mean=1.7%), and the accumulation of fluids of decreased ionic concentration in its cephalic cavities (Horn et a l., 1978). In contrast to these species, Ipnops murrayi is a small benthic species of low water content (mean=78.2%) and relatively high skeletal ash content (mean=2.7%). Despite the gross differences between 1^. murrayi and A^. armatus, their diets are apparently quite similar. This unlikely observation has also been made by Nielsen who reported polychaete setae of the genus Aphrodite from guts of both murrayi (Nielsen, 1966) and _A. armatus (Nielsen, 1965). 108 A diverse array of species is also included in group 3 which comprised species feeding heavily on decapods and teleosts. Included in this group are the benthic predators Bathysaurus mollis and IB, f erox, two of the largest predators inhabiting this depth range, and Bathypterois grallator. Also included in this group are the benthopelagic alepocephalids Conocara macropterurn (>175 mm,SL) and Narcetes stomias, both of which are characterized by high water contents, poorly ossified skeletons, and the absence of swimbladders. Despite obvious differences, both these benthic forms and the benthopelagic alepocephalids are probably "sit and wait" or "float and wait" predators feeding in similar manners. Indeed, the elongate fin rays of J3. grallator may place it a similar distance off the bottom to that occupied by _C. macropterum and N_. stomias. In this regard it is of interest that large IB. grallator and C. macropter um are paired by cluster analysis. Lastly this group includes two benthopelagic taxa with swimbladders, the ophidiids, Bassozetus normalis and Dicrolene intronigra, and the synaphobranchid eels. On togene tic Dietary Shif ts Ontogenetic shifts in diet have been shown to be of wide spread occurrence in fishes (Tyler, 1972; Pearcy and Ambler, 1974; Ross, 1978; Werner, 1979; Sedberry, 1983), and occur in many species considered in the present study. Bathypterois grallator, Conocara macropterum, Ventrifossa occidentalis, and 109 Nezumia bairdii all show marked shifts in diet with increasing size. Indications of less pronounced shifts are also evident for species of Aldrovandia. Increased predator size allows large prey taxa such as decapod crustaceans and teleosts, which are too large to be ingested by small predators, to become potential prey items. In some predators such as C^. macropterum a corresponding decline in the importance of small crustaceans accompanies the addition to the diet of larger prey taxa (Figs. 11 and 12, Tables 14 and 15); however, in the case of IB. g ra lla to r, this does not occur and copepods are ingested at all sizes (Figs. 14 and 15). Thus it appears that in some species such as _C. macropter um dietary shifts may indicate a switch from benthic to more pelagic feeding modes, while in other species such as IB. grallator, an expansion of the diet to include a more diverse array of prey sizes is indicated . Overlap in Diet Predator species show considerable overlap in diet at the major taxon level. Furthermore, most species feed on a variety of prey taxa, although as shown by Carter (1984) some species such as Barathrites parri, which feeds almost exclusively on polychaetes and holothurians, and Porogadus silus, which feeds largely on copepods, appear to have rather restricted diets. In general, the deep-sea species considered in this study appear to have diets comparable to their better known shallow water counterparts occurring on the continental 110 shelf, where considerable interspecific overlap in diet has been reported among MAB demersal fishes (Sedberry, 1983). There seems to be little evidence to suggest that in general the diets of deep-sea species are any more or less generalized than those of shallow water species found in similar open bottom habitats. Dayton and Hessler (1972) considered deep-sea fishes to be non-selective predators on the benthic macrofauna. In addition, Schoener (1971) suggested that predators should have generalized diets in areas of low food abundance such as the deep sea. Cluster analysis indicates that although diets of deep-sea fishes overlap, species can be grouped according to diet similarities. Most species tend to feed most heavily on a few prey taxa, and differences between species suggests that these species are not non-selective feeders. Based on an analysis of depth distributions of some common deep-sea fishes, Sulak (1982) concluded that competition may be an important factor structuring deep-sea fish assemblages. Sulak examined the depth distributions of various groups of related species and showed that although depth ranges overlap in most cases, depths of greatest abundance are often different. Thus by implication, a species replacement occurs with increasing depth with species presumably better adapted to greater depths replacing those best equiped for life at shallower depths. Sulak's discussion is limited by a lack of food habits data on many of the species considered, making it impossible to evaluate the Ill potential for competition for common food resources. A number of the groups he considered include species reported in the present study and warrant further discussion. Two congenors displaying spatial displacement are Conocara macropterum and C, niger (Sulak, 1982). Though these species overlap in their depth distributions, C^. macropterum has a center of abundance shallower than niger. Data presented earlier suggest that the two species do overlap in diet, with salps being a major prey item in common. Conocara niger did not feed on decapods or teleosts as did C. macropterum; however, this could be an a rtifa c t of the small number of C. niger examined. Another pair of potential competitors considered by Sulak is Nezumia bairdii and Coryphaenoides carapinus. Again these species overlap in their depth ranges but the center of abundance of C. carapinus is substantially deeper than that of N.* bairdii. C^. carapinus is known to feed largely on benthic taxa (Haedrich and Polloni, 1976). Sulak assumes benthic feeding in bairdii based on the limited available data on _N. bairdii and studies of other Nezumia species; however as shown earlier N_. bairdii can feed heavily on natant decapods. Though competition between N_. bairdii and C. carapinus is possible, two other species, Nezumia aequalis and Coelorhyncus c^. carminatus seem more likely as potential competitors of N.. bairdii yet occur at roughly equivalent depths with l i t t l e suggestion of spatial displacement except that £. carminatus occurrs slightly shallower than either JJ. bairdii or N_. aequalis (Middleton, 1979). These species are most 112 abundant at depths shallower than 1000 m and were not considered by Sulak whose discussion was restricted to depths below 1000 ra. Among ipnopine species, Sulak suggests that some form of resource partitioning is likely among Ipnops murrayi, Bathypterois phenax, and Bathypterois g ra lla to r. Sulak bases this on the widely overlapping depth ranges of these three species. Data presented here support Sulak’s contention with little dietary overlap found among these species. Furthermore, the great difference in size attained by these species (see above) suggests l i t t l e probability of dietary overlap. Sulak proposes that competition among the synaphobranchid eels Synaphobranchus kaupi and S.. brevidorsalis could explain the truncated bathymetric range of S^. kaupi in the Bahamas in comparison to its range in the MAB. In the Bahamas, S_. kaupi is abundant to depths of 2000 m while it is abundant only at depths less than 1500 m in the MAB. Synaphobranchus brevidorsalis is rare in the MAB, however in the Bahamas it is one of the most abundant species at depths between about 1200- 3000 m. The success of J3. brevidorsalis at greater depths and apparently more energy impoverished regions may reflect the more flaccid body composition of S_. brevidorsalis (mean water content = 87.1%) when compared to the more widely distributed Jl* kaupi (mean = 76.5%). The higher water content of S_. brevidorsalis could in turn reflect lower energy requirements, thus enabling the species to survive in more impoverished 113 areas and depths than j3. ka upi. Furthermore this flaccid body may restrict the mobility of j5. brevidorsalis placing it at a competitive disadvantage with S^. kaupi and other more mobile predators in high energy areas such as shallower depths or as Sulak suggests the MAB. The presence of apparently scavenged material in guts of J3. brevidorsalis as discussed earlier may be significant in this regard and could reflect the apparent low mobility of the species. Four morphologically similar species of Aldrovandia overlap bathymetrically in both the MAB and Bahamas study areas at depths between about 1000-2000 m. There is some indication of food resource partitioning among Aldrovandia species with A_;_ affinis feeding more heavily on polychaetes and mollusks than other Aldrovandia species. Sulak (1982) suggests resource partitioning as a mechanism allowing the co occurrence of the four species of Aldrovandia. However, with the exception of _A. af finis, Aldrovandia species appear to overlap broadly in diet lending little support to Sulak's conjecture. Moreover, Sulak suggests that competition between Aldrovandia and the related Halosauropsis machrochir may result in the displacement of jl. macrochir to deeper water in the Bahamas compared to its distribution in the MAB. As suggested earlier the diets of large individuals of Aldrovandia and small II. macrochir may be similar, thus competition could occur. However, decapods and isopods are major components of the diet of II. macrochir (Sedberry and Musick, 1978) and of much less importance in the diets of Aldrovandia. Competition among the co-occurring species of 114 Aldrovandia appears more likely than between Aldrovandia and II. macrochir. It seems inconsistent to attribute the coexistence of Aldrovandia species to resource partitioning and simultaneously the displacement of II. macrochir to competition even though available evidence indicates that interspecific competition is at least as likely among Aldrovandia species as between Aldrovandia and jl. macrochir. Faunal Composition Numerous important differences in the taxonomic composition of the Bahamas and MAB fish assemblages have been discussed by Sulak (1982). Families of importance in the MAB but rare or absent from the Bahamas below 1000 m include morids, macrourids, pleuronectids, zoarcids, gadids, and cottids. In the Bahamas ophidiids, chlorophthalmids, and stephanoberycids are dominant groups. With the exception of the ophidiid Dicrolene intronigra, these families are rare in the MAB. The families Synaphobranchidae, Halosauridae, Notacanthidae, and Alepocephalidae are abundant in both study areas. Yet within these families different species tend to dominate respective study areas. For example, the alepocephalid, Alepocephalus agassizii, and the notacanth, Polyacanthonotus rissoanus, are common in the MAB, yet absent from the Bahamas. Conversely the alepocephalid Conocara macropterum and the synaphobranchid eel Synaphobranchus brevidorsalis are among the most abundant species in the Bahamas, but are both rare in the MAB. 115 Sulak (1982) attributes these taxonomic differences to differences in food av ailab ility between the MAB and Bahamas study areas. Based on a review of the existing literature Sulak concludes that due to differences in surface productivity, more food should be available to fishes in the MAB than in the Bahamas study area. Consequently, he proposes that successful MAB species have higher energy requirements as evidenced by more active foraging modes and probable high fecundities, in contrast to successful Bahamian species with less active foraging modes and lower fecundities. The information presented herein with regard to food habits and body composition can be interpreted in terms of food availability in the respective study areas and in general seem to support Sulak's hypothesis. Species which appear to utilize "sit-and-wait" or "float-and-wait" feeding modes are more abundant in the Bahamas than in the temperate Atlantic where more active foraging modes prevail. The benthic chlorophthalmids as well as the flaccid bodied alepocephalids and ophidiids probably swim little and simply wait on prey to d rift or swim within striking distance. These groups are less abundant in terms of both numbers of individuals, biomass, and numbers of species present in the MAB than in the Bahamas where they are a major component of the fauna. It seems plausible that this difference reflects lowered food availability in the Bahamas and a corresponding reduction in the importance of energetically less feasible active foraging life styles such as those apparently utilized by such common MAB species as Coryphaenoides rupestris, £. armatus, Nezumia 116 b a ird ii, and Antimora ro s tra ta . Accordingly, it appears that sedentary feeding modes have been relatively less successful in more food rich temperate latitudes. While Sulak's hypothesis seems plausible, it must be noted that it has not been conclusively demonstrated that the amount of food available to deep-sea demersal fishes in the MAB is actually higher than in the Bahamas. Sulak's conclusions with regard to energy availability are based upon estimates of surface productivity, sediment trap work, organic content of sediments, and sparse available data on the numerical density and biomass of benthic macrofauna in the respective study areas. For the Bahamas study area, primary production estimates are unavailable and must be extrapolated from adjacent areas. Present knowledge of energy flux between surface and bottom waters and of the trophic pathways followed by organic matter upon arrival at the bottom is incomplete. Thus a better understanding of the energetics of deep-sea systems as well as more data on primary production in the respective study areas is necessary before Sulak's hypothesis can be evaluated. A consideration of the trophic structure of the MAB and Bahamas fish assemblages suggests that Sulak's hypothesis of differences in food av ailab ility need not be invoked to explain some differences between the respective faunas. It is apparent from the discussion of food habits that benthic prey such as infaunal bivalves, gastropods, and echinoderms are of l i t t l e importance in the diets of Bahamian deep-sea fishes. 117 This is in contrast with the temperate Atlantic where several species of zoarcids common on the continental slope feed heavily on bivalves and ophiuroids. Sedberry and Musick (1978) found ophiuroids and pelecypods to be the two most abundant prey of Lycodes a tla n tic u s , a species common between 1000-2000 m in the MAB. Bivalves and gastropods were reported as the major prey items found in guts of Lycenchelys v e r r i l l i i , an abundant zoarcid on the upper slope of the MAB above 1200 m, by Wenner (1978) and Farlow (1980). In addition, Wenner reported bivalves and gastropods from Lycenchelys paxilla, a species which occurs between 600-2100 m in the MAB. It is interesting that these species which are common in the MAB are absent from Bahamian collections. In addition, the macrourid, Coryphaenoides carapinus, has been shown to feed heavily on benthic prey such as ophiuroids by Haedrich and Polloni (1976). This species was much more common in the MAB than in the Bahamas. Furthermore, there seems to be no group potentially replacing these species feeding on benthic mollusks and ophiuroids in the Bahamas. Perhaps the distributions of these fishes are dictated by that of their prey and reflect a lower abundance of benthic bivalves and ophiuroids in the Bahamas in comparison with the temperate Atlantic. This hypothesis does not rely on differences in the absolute level of food abundance between the two study areas, but instead on differences in the proportional abundances of diiferent prey types. Unfortunately, data on the abundance of benthic infauna and epifauna are not adequate to evaluate this hypothesis; 118 however, differences in sediment type between the two regions suggest that differences in the benthic infauna and epifauna are not unexpected. Slope sediments in the MAB are predominatly hemipelagic silts and clays and graded turbidites of t e r r e s t r i a l o rig in (Emery and Uchupi, 1972; Horn, Delach, and Horn, 1974). These sediments contain low concentrations of calcium carbonate (Sanders, H essler, and Hampson, 1965; Milliman, Pilkey' and Ross, 1972; Balsam, 1982). In contrast, bottom sediments in the Bahamas study area are mainly calcareous oozes (Emery and Uchupi, 1972; Horn et a l ., 1974). Body Composition No comprehensive study has examined the whole body composition of demersal deep-sea fishes, although several have examined white skeletal muscle. Childress and Nygaard (1973) examined 37 species of midwater teleosts, and in general their ranges for water, ash, carbon, and nitrogen are comparable to those reported here for demersal species. In addition, the trends with depth reported by Childress and Nygaard are similar to those observed here for demersal species, although there is much more scatter in the demersal fish data. Percent water shows a tendency to increase with depth. A consequence of this are accompanying decreases in carbon and nitrogen as percent wet weight. These decreases reflect corresponding decreases in protein and lipid as percent wet weight and consequently a decrease in caloric content. Similarly, Childress and Nygaard reported reduced caloric 119 densities at depth in midwater fishes. The most obvious interpretation of these data is that selection has favored energetically "cheaper" bodies as a response to decreasing food availability with increasing depth. However, the great amount of scatter present in all correlations indicates that a variety of adaptations have been successful in deep-sea fishes at all depths. In addition to energetic concerns, bouyancy mechanisms are important in determining body composition. Benthic and benthopelagic species with swimbladders have lower percent water and higher skeletal ash contents than do benthopelagic species without swimbladders. These findings agree with Childress and Nygaard (1973) who found higher water contents and lower skeletal ash contents in midwater fishes without swimbladders than in those with swimbladders. Fishes which achieve neutral bouyancy through high water content appear to be restricted to "float and wait" feeding modes. These along with benthic species may be seldom active and thus among the most energy efficient of deep-sea species. It may be significant that these species are far more prevalent both in terms of numbers of species and numbers of individuals in the Bahamas study area than in the MAB. Recently a number of authors have attempted to relate enzymatic activity levels with food habits of deep-sea fishes. Sullivan and Somero (1980) contend that L -la c ta te dehydrogenase (LDH) and pyruvate kinase (PK) activities are valid biochemical indices of the capacity for rigorous high speed locomotion in fishes. White muscle is important in 120 short term high speed locomotion of fishes (Beamish, 1978; D riedzic and Hochachka, 1978) and provides the power required for rapid movements such as prey capture or predator avoidance. Short bursts of swimming power are obtained via glycolytic enzyme activity and the breakdown of glycogen stores to lactic acid. This breakdown involves the action of both LDH and PK (Bilinski, 1974), thus the activities of these enzymes should reflect the capacity of a species for high speed locomotion. Sullivan and Somero (1980) examined the activities of L- lactate dehydrogenase (LDH), pyruvate kinase (PK), L-malate dehydrogenase (MDH), and citrate synthase (CS) in a number of deep-sea fishes including several demersal species occurring in the western Atlantic. Among these are Synaphobranchus bathybius (called Histiobranchus bathybius) , Coryphaenoides armatus, (1. carapinus, (1. leptolepis, Antimora rostrata, H alosauropsis m acrochir, and Nezumia b a i r d i i . Species most abundant at depths below 200 m were found to have extremely low enzyme activity levels compared to the shallow living species examined, although strict comparison of their groups is deceptive since the shallow group includes several pelagic species while the deep-sea group includes only benthic and benthopelagic species. Furthermore, comparisons were not confined within taxonomic groups with members living in both shallow and deep zones. Two enzymes LDH and PK showed significant depth related changes with activities up to five times lower in deeper living species, further indication that selection has favored energetically less expensive life styles at greater depths. Of the western Atlantic deep demersal species, C. armatus and J5. bathybius have much higher enzymic activities than do A_. rostrata, H_. macrochir. C. leptolepis, N_. bairdii, or C. c a ra p in u s. S ullivan and Somero speculate that C. armatus may employ a more activ e "search and capture" feeding strategy in contrast to a "float and wait" strategy utilized by the other species which would minimize energy expenditures. It is significant that C. armatus, which had enzyme activity levels among the highest of any deep-sea species, is an abyssal species most abundant at depths below about 3000 m. This again points out that d espite generalizations regarding the success of energetically inexpensive life styles at great depths, a variety of life styles have been successful. Furthermore, it is of interest that (I. armatus is much less abundant in the Bahamas than in the MAB, again suggesting that low cost life styles have been favored in the Bahamas as compared to the MAB. Similar findings were reported by Siebenaller, Somero, and Haedrich (1982) who examined five macrourids, Nezumia bairdii, Coryphaenoides rupestris. C. carapinus, C^. armatus, and C. leptolepis, as well as the non-macrourids Halosauropsis macrochir, Bathysaurus ferox, Synaphobranchus (Histiobranchus) bathybius, and Dicrolene introniara. Of these species C. armatus displayed the highest enzyme activities indicating a mobile existence as suggested by Sullivan and Somero (1980). Furthermore, C!. armatus displayed siginificant scaling of enzyme a c tiv ity to body mass for CS, PK, and LDH with the 122 a c t i v i t i t e s of the g ly c o ly tic enzymes LDH and PK increasing with rising body mass and that of the citric acid cycle enzyme CS decreasing. Siebenaller et al. speculate that this scaling may be related to the tendency for large individuals of C_. armatus to feed pelagically in midwater. Finally, of the remaining species examined, C_. ru p e s tris (which has also been captured in midwater; Haedrich, 1974) showed the second highest enzyme activities with extremely low levels found in others species with the exception of . bathybius, which was found to be extremely variable. There may be a relationship between enzyme activity and burst swimming c a p a b ility ; however, the lin k between enzyme activity and diet is unclear. For benthic feeding species, burst swimming may never be involved in feeding but may be more closely related to predator avoidance. Many benthopelagic species, such as halosaurs, ophidiids, as well as some macrourids probably cruise slowly over the bottom seeking prey. It is difficult to visualize the role of rapid bursts of speed when feeding on benthic prey such as polychaetes, bivalves, or ophiuroids, or even in the case of such small crustaceans as copepods or amphipods. Predator avoidance would seem to be more im portant than foraging with regard to the adaptive significance of burst swimming capabilities in many species. A species which is unsuccessful in capturing a prey item has only missed a meal, however an individual which is unsuccessful in avoiding a predator is removed from the population. 123 The enzyme activities reported by these workers in many cases are inconsistant with food habits data. The LDH a c tiv ity of Nezumia b a ird ii is reported as 6.9 (u n its /g wet weight; Siebenaller et al., 1982) and the data presented in the present study shows that the species is capable of feeding on pelagic shrimps, which could involve bursts of rapid swimming. Dicrolene intronigra, a species with a diet similar to that of _N. bairdii had a much higher LDH activity level of 46.4, giving it the third highest LDH activity level of all species examined. There is no evidence to suggest that Jl. intronigra has any more active a life style than N_. b a i r d i i . No s t a t i s t i c a l comparisons were made among any of the species examined; however, the large standard deviations reported suggest that the LDH activity of J3. armatus (mean=53.1, standard deviation=28.9) is probably no different than that of I), intronigra (mean=46.4, n = l) or Bathysaurus ferox (mean=35.4, standard deviation=6.0; reported as Bathysaurus agassizi), yet C^. armatus is interpreted as having a " re la tiv e ly high capacity for swimming" c o n tra sted with a " re la tiv e ly low capacity for activ e swimming" for D^. in tro n ig ra and _B. f erox (S ieb en aller et al . , 1982). Furthermore, no evidence of mass related scaling were shown for either NL_ bairdii or Coryphaenoides leptolepis even though pronounced shifts in diet have been demonstrated for both species (N_^ bairdii - present study; C_^ leptolepis - Pearcy and Ambler, 1974). Finally the declines in enzyme activity with increasing depth reported by these workers must be interpreted with 124 caution. All of the shallow water species examined are higher acanthopterygians with the exception of Squalus acanthias. In contrast all of the deep water species are lower (Anguilliformes) or intermediate (Gadiformes) teleosts. Comparisons of such distantly related groups may be highly biased. I t would seem more appropriate to compare deep-sea gadiforms such as macrourids with common shallow water gadiforms such as Urophycis or Gadus. Literature Cited Anderson, M. E., R. E. Crabtree, H. J. Carter, K. J. Sulak, and M. D. Richardson. In press. D istrib u tio n of demersal fishes of the Caribbean Sea found below 2000 m eters. Bull. Mar. S c i. Balsam, W. L. 1982. Carbonate dissolution and sedimentation on the Mid-Atlantic continental margin. Science, N.Y. 217:929-931 . Beamish, F. W. H. 1978. Swimming capacity. In: Fish physiology, Vol. VII, pp 101-187. Ed by W. H. Hoar and D. J . Randall. New York: Academic Press. Bigelow, H. B. and W. C. Schroeder. 1948. Fishes of the Western North Atlantic, Part I, 576p. Sears Foundation for Marine Research, Yale Univ., New Haven, Conn. Bigelow, H. B. and W. C. Schroeder. 1953. Fishes of the Gulf of Maine. U.S. Fish. Wildl. Ser. Fish Bull. 53(74):1-577. Bilinski, E. 1974. Biochemical aspects of fish swimming. In: Biochemical and biophysical perspectives in marine biology. Vol. I, pp 247-288. Ed. by D. C. Malins and J. R. Sargent. New York: Academic P ress. B right, T. J. 1968. A survey of the deep-sea bottom fish es of the Gulf of Mexico below 350 meters. Ph. D. Dissertation. Texas A&M Univ. 218p. C arter, H. J. 1984. The functional morphology and feeding stategies of demersal deep-sea neobythitine fishes. Ph.D. dissertation. College of William and Mary in Virginia. Williamsburg. 236p. Childress, J. J. and M. H. Nygaard. 1973. The chemical composition of midwater fishes as a function of depth of occurrence off southern California. Deep-Sea Res. 20:1093-1109. Clarke, M. R. and N. M errett. 1972. The sig n ific a n c e of squid, whale, and other remains from the stomachs of bottom-living deep-sea fish. J. Mar. Biol. Assoc. U. K. 52:599-603. 12 5 126 Clifford, H. T. and W. Stephenson. 1975. An introduction to numerical classification. Academic Press, New York. 229p. Crabtree, R. E., K. J. Sulak, and J. A. Musick. In press. Feeding, reproduction, and distribution of Polyacanthonotus (Pisces: Notacanthiformes) in the western North A tla n tic . Bull. Mar. Sci. Dayton, P. K. and R. R. H essler. 1972. Role of b io lo g ical disturbance in maintaining diversity in the deep sea. Deep-Sea Res. 19:199-208. D riedzic, W. R. and P. W. Hochachka. 1978, Metabolism in fish during exercise. In: Fish physiology, Vol VII, pp 503-544. Ed. by W. S. Hoar and D. J. R andall. New York: Academic P ress. Du Buit, Marie-Henriette. 1978. The diet of some teleost fishes in the Wyville Thomson ridge area. Oceanol. Acta. 1:129-134. Emery, K. 0. and E. Uchupi. 1972. Western A tla n tic Ocean: Topography, Rocks, Structure, Water, Life and Sediments. Am. Ass. petro. G eol., Mem. 17, 532 pp. Farlow, J. 0. 1980. A demersal fish - megabenthic in v e rte b ra te community from the upper co n tin en ta l slope off southern New England. Ph. D. d is s e r ta tio n . Yale Univ. 430 p. Golovan, G. A. and N. P. Pakhorukov. 1975. Some data on the morphology and ecology of Alepocephalus bairdii (Alepocephalidae) of the central and eastern Atlantic. Vopr. Iktiol. 15(l):44-50. Golovan, G. A. and N. P. Pakhorukov. 1980. New data on the ecology and morphometry of Alepocephalus rostratus (Alepocephalidae). Vopr. Iktiol. 20(3):77-83. Goode, G. B. and T. H. Bean. 1896. Oceanic ichthyology, Deep-sea and pelagic fishes of the world. Spec. Bull. U.S. Natl. Mus. 2:1-553. Haedrich, R. L. 1974. Pelagic capture of the epibenthic rattail Coryphaenoides rupestris. Deep-Sea Res. 21:977-979. Haedrich, R. L. and N. R. Henderson. 1974. Pelagic feeding by Coryphaenoides armatus. a deep benthic rattail. Deep-Sea Res. 21:739-744. 127 Haedrich, R. L., G. T. Rowe, and P. T. P olloni. 1975. Zonation and faunal composition of epibenthic populations on the continental slope south of New England. J. Mar. Res. 33:191-212. Haedrich, R. L. and P. T. Polloni. 1976. A contribution to the life history of a small rattail fish Coryphaenoides carapinus. Bull. So. Calif. Acad. Sci. 75:203-211. Haedrich, R. L. and G. T. Rowe. 1977. Megafaunal biomass in the deep-sea. Nature, Lond. 269:141-142. Haedrich, R. L. and G. Krefft. 1978. Distribution of bottom fishes in the Denmark Strait and Irminger Sea. Deep-Sea Res. 25:705-720. Haedrich, R. L., G. T. Rowe, and P. T. Polloni. 1980. The megabenthic fauna in the deep-sea south of New England, U.S.A. Mar. B iol. 57:165-179. Heezen , B. C. and C. D. Hollister. 1971. The face of the deep. Oxford Univ. Press., New York. 659p. Horn, M. H., P. W. Grimes, C. F.Phleger, L. L. McClanahan. 1978. Buoyancy function of the enlarged fluid-filled cranium in the deep-sea ophidiid fish Acanthonus armatus. Mar. Biol. 46:335-339. Horn, D. R., M. N. Delach, B. M. Horn. 1974. Physical properties of sedimentary provinces, North Pacific and North Atlantic Oceans. pp 417-441, In: Deep-Sea Sediments (A. L. Inderbitzen, ed.), Plenum Press, NY. Krefft, G. 1973. Alepocephalidae, pp. 86-93. In: Check-list of the fishes of the north-eastern Atlantic and of the M editerranean. Ed. by J. C. Hureau and Th. Monod. Unesco, Paris. Love, M. S. and A. W. Ebeling. 1978. Food and habitat of three switch-feeding fishes in the kelp forests off Santa Barbara, California. Fish. Bull. 76:257-271. Lozano Cabo, F. 1952. Estudio preliminar sobre la biometria, la biologia y la anatomia general de Notacanthus bonapar tei. Bol. Inst. Espanol Oceanogr. 49:1-30. Macpherson, E. 1979. Ecological overlap between macrourids in the western Mediterranean Sea. Mar. Biol. 53:149-159. Macpherson, E. 1981. Resource partitioning in a M editerranean demersal fish community. Mar. Ecol. Prog. Ser. 4:183-193. 128 Markle, D. F. 1975. Young witch flounder, Glyptocephalus cynoglossus, on the slope off Virginia. J. Fish. Res. Bd. Can. 32:1447-1450. Markle, D. F. 1976. Preliminary studies on the systematics of deep-sea Alepocephaloidea (Pisces: Salmoniformes). Ph. D. d is s e r ta tio n . College of William and Mary in Virginia. Williamsburg. 225p. Markle, D. F., and J. A. Musick. 1974. Benthic-slope fishes found at 900 m depth along a transect in the western N. A tla n tic Ocean. Mar. B iol. 26:225-233. Markle, D. F. and C. A. Wenner. 1979. Evidence of demersal spawning in the mesopelagic zoarcid fish, Melanostigma a tla n tic u m , with comments on demersal spawning in the alepocephalid fish Xenodermichthys copei. Copeia. 1979:363-366. Marshall, N. B. and T. Iwamoto. 1973. Family Macrouridae, pp 496-662. In: Fishes of the westernNorth Atlantic. Ed. by D. M. Cohen. Mem. Sears Found. Mar. Res. Marshall, N. B. and J. C. Staiger. 1975. Aspects of the structure, relationships and biology of the deep-sea fish Ipnops murrayi (family Bathypteroidae). Bull. Mar. Sci. 25:101-111. Marshall, N. B. and N. R. Merrett. 1977. The existence of a benthopelagic fauna in the deep sea. pp 483-497. In: A voyage of discovery, George Deacon 70th anniversity volume. Ed. by M. V. Angel. Mauchline, J. and J. D. M. Gordon. 1983a. Diets of the sharks and chimaeroids of the Rockall Trough, northeastern Atlantic Ocean. Mar. Biol. 75:269-278. Mauchline, J. and J. D. M. Gordon. 1983b. Diets of clupeoid, stomiatoid and salmonoid fish of the Rockall Trough. Mar. B iol. 77:67-78. McDowell, S. B. 1973. Order Heteromi (Notacanthiformes). pp 1-228. In: Fishes of the western North Atlantic. Part 6. Ed. by D. M. Cohen. Sears Found, for Mar. Res. Merrett, N. R. and N. B. Marshall. 1981. Observations on the ecology of deep-sea bottom-living fishes collected off northwest Africa (08 -27 N). Prog. Oceano. 9:185-244. Middleton, R. W. 1979. The abundance, distribution and biomass of the family Macrouridae in the Norfolk Canyon area. MA thesis. College of William and Mary in Virginia. Williamsburg. 72p. 129 Milliman, J. D., 0. H. Pilkey, and D. A. Ross. 1972. Sediments of the continental margin off the eastern United States. Bull. geol. Soc. Am. 83:1315-1344. Murdoch, W. W., S. Avery, and M. E. B. Smyth. 1975. Switching in predatory fish. Ecology. 56:1094-1105. Musick, J. A. 1979. Community s tru c tu re of fish es on the continental slope and rise off the Middle Atlantic coast of the U.S. Va. Inst. Mar. Sci. Spec. Sci. Rept. No. 96. Nielsen, J. G. 1965. On the genera Acanthonus and Typhlonus (Pisces, Brotulidae). Gal. Rept. 8:33-47. Nielsen, J. G. 1966. Synopsis of the Ipnopidae (Pisces, Iniomi) with descriptions of two new abyssal species. Gal. Rept. 8:49-75. Pearcy, W. G. and J. W. Ambler. 1974. Food h ab its of deep- sea macrourid fishes off the Oregon coast. Deep-Sea Res. 21:745-759. Pearcy, W. G., D. L. Stein, and R. S. Carney. 1982. The deep-sea fish fauna of the no rth eastern P acific Ocean on Cascadia and Tufts Abyssal Plains and adjoining continental slopes. Biol. Oceanog. 1:375-428. Pinkas, L., M. S. O liphant, and I. L. K. Iverson. 1971. Food habits of albacore, bluefin tuna, and bonito in California waters. Calif. Fish. Game Fish. Bull. 152: 1-105. Polloni, P., R. Haedrich, G. Rowe and C. H. Clifford. 1979. the size-depth relationship in deep ocean animals. Int. Revue ges. Hydrobiol. Hydrogr. 64(l):39-46. Rayburn, R. 1975. Food of deep-sea demersal fishes of the northwestern Gulf of Mexico. MS thesis. Texas A & M Univ. 119p. Robins, C. H. 1968. The comparative osteology and ecology of the synaphobranchid eels of the Straits of Florida. Ph.D. th e s is . Univ. of Miami. 149p. Ross, S. T. 1978. Trophic ontogeny of the leopard searobin, Prionotus scitulus (Pisces: Triglidae). Fish. Bull. 76:225-234. Saldanha, L. 1980. Regime alimentaire de Synaphobranchus kaupi Johnston, 1862 (Pisces Synaphobranchidae) au large des cotes Europeennes. Cybium 3e serie. 8:91-98. 130 Sanders, H. L., R. R. H essler, G. R. Hampson. 1965. An introduction to the study of deep-sea benthic faunal assemblages along the Gay Head-Bermuda transect. Deep-Sea Res. 12:845-867. Sazonov, Yu. I. and A. N. Ivanov. 1980. Slickheads (Alepocephalidae and Leptochilichthyidae) from the thalassobathyal zone of the Indian Ocean. Trudy Instit. Okeanol. 110:7-104. Sedberry, G. R. 1983. Food habits and trophic re la tio n s h ip s of a community of fishes on the outer continental shelf. NOAA Tech. Rept. NMFS SSRF-773. 56p. Sedberry, G. R. and J. A. Musick. 1978. Feeding strategies of some demersal fishes of the continental slope and rise off the Mid-Atlantic coast of the USA. Mar. Biol. 44:357-375. Schoener, T. W. 1971. Theory of feeding strategies. Ann. Rev. Ecol. System. 2:369-404. Siebenaller, J. F., G. N. Somero, and R. L. Haedrich. 1982. Biochemical characteristics of macrourid fishes differing in their depths of distribution. Biol. Bull. 163:240- 249. Staiger, J. C. 1970. The distribution of the benthic fishes found below two hundred meters in the Straits of Florida. Ph. D. d is s e r ta tio n . Univ. of Miami. 219p. Sulak, K. J. 1977a. Aldrovandia oleosa, a new species of the Halosauridae with observations on several other species of the family. Copeia. 1977:11-19. Sulak, IC. J. 1977b. The systematics and biology of Bathypterois (Pisces, Chlorophthalmidae) with a revised classification of benthic myctophiform fishes. Gal. Rept. 14:49-108. Sulak, K. J. 1982. A comparative taxonomic and ecological analysis of temperate and tropical demersal deep-sea fish faunas in the western North Atlantic. Ph. D. d is s e r ta tio n . Univ. of Miami. Sulak, K. J., R. E. Crabtree, and J. -C. Bureau. In press. Provisional review of the genus Polyacanthonotus (Notacanthidae) with description of a new Atlantic sp ecies, Polyacanthonotus m e rre tti. Cybium, ser 3, 8(4). Sulak, K. J . , C. A, Wenner, G. R. Sedberry, and L. Van Guelpon. In prep. The life history and systematics of deep-sea, lizardfishes, genus Bathysaurus (Synodontidae). Submitted to Can J. Zool. 131 Sullivan, K. M. and G. N. Somero. 1980. Enzyme activities of fish skeletal muscle and brain as influenced by depth of occurrence and habits of feeding and locomotion. Mar. Biol. 60:91-99. Tyler, A. V. 1972. Food resource division among northern marine demersal fishes. J. Fish. Res. Bd. Can. 29:997-1003. Vodopich, D. S. and J. J. Hoover. 1981. A computer program for integrated feeding ecology analyses. Bull. Mar. Sci. 31:922-925. Wenner, C. A. 1975. The occurrence of elvers of Synaphobranchus affinis on the continental slope of North Carolina. Fish. Bull. 73:687-691. Wenner, C. A. 1976. Aspects of the biology and morphology of the snake eel, Pisodonophis cruentifer (Pisces, Ophichthidae). J. Fish. Res. Bd. Can. 33:656-665. Wenner, C. A. 1978. Making a liv in g on the co n tin en ta l slope and in the deep sea: life history off some dominant fishes of the Norfolk Canyon area. Ph. D. dissertation. College of William and Mary in Virginia. Williamsburg. 294p. Wenner, C. A. 1979. Notes on fishes of the genus Paraliparis (Cyclopteridae) on the Middle Atlantic continental slope. Copeia. 1979 :145-146. Wenner, C. A. and J. A. Musick. 1977. Biology of the morid fish Antimora rostrata in the western North Atlantic. J. Fish. Res. Bd. Can. 34:2362-2368. Werner, E. E. 1979. Niche partitioning by food size in fish communities. JJ^n: Predator-prey systems in fisheries management. R. H. Stroud and H. Clepper, eds., Sport Fishing Inst., Washington. pp 311-322. Appendix 1. Fishes examined in body composition study. Cruise Station Date La t . Long. Depth Length Myxine glutinosa AL-80-05 13 8-V-80 39 58 N 70 20 W 338 370,TL AL-80-05 13 8-V-80 39 58 N 70 20 W 338 370,TL AL-80-05 13 8-V-80 39 58 N 70 20 W 338 280,TL AL-80-05 13 8-V-80 39 58 N 70 20 W 338 250,TL AL-80-05 13 8-V-80 39 58 N 70 20 W 338 350,TL AL-80-05 13 8-V-80 39 58 N 70 20 W 338 400,TL Synaphobranchus kaup i AL-80-05 6 7-V-80 39 46 N 71 21 W 1130 333,TL AL-80-05 6 7-V-80 36 46 N 71 21 W 1130 635,TL AL-80-05 5 7-V-80 3949 N 70 59 W 963 478,TL AL-80-05 5 7-V-80 39 49 N 70 59 W 963 490,TL AL-80-04 21 9-V-80 40 02 N 69 02 W 737 420,TL AD-213-3 6383 2-XI-79 29 52 N 77 09 W 1026 442,TL AD-213-3 6356 28-X-79 36 22 N 74 42 W 1035 439,TL AD-213-3 6364 29-X-79 34 40 N 75 30 W 868 350,TL CI-80-07 C041 15-IX-80 23 53 N 77 07 w 1341 360,TL CI-80-07 C041 15-IX-80 23 53 N 77 07 w 1341 285,TL CI-80-07 C041 15-IX-80 23 53 N 77 07 w 1341 380,TL VS-79-03 12 - 37 25 N 74 38 w 1550 579,TL Synaphobr anchus af fin is CI-80-07 C063 21-IX-80 29 06 N 77 08 w 1093 355,TL CI-80-07 C063 21-IX-80 29 06 N 77 08 w 1093 330,TL CI-80-07 C063 21-IX-80 29 06 N 77 08 w 1093 350,TL CI-80-07 C063 21-IX-80 29 06 N 77 08 w 1093 290,TL CI-80-07 C063 21-IX-80 29 06 N 77 08 w 1093 360,TL CI-80-07 C062 21-IX-80 29 46 N 77 09 w 918 355,TL CI-80-07 C062 21-IX-80 29 46 K 77 09 w 918 340,TL CI-80-07 C062 21-IX-80 29 46 N 77 09 w 918 325,TL Synaphobranchus brevidorsalis CI-80-07 C001 29-IIX-80 25 23 N 77 01 w 2423 470,TL CI-80-07 C008 21-IX-80 24 14 N 76 06 w 1778 335,TL CI-80-07 C008 21-IX-80 24 14 N 76 06 w 1778 640,TL AD-213-3 6387 4-XI-79 29 08 N 78 57 w 830 473,TL CI-80-07 C038 14-IX-80 24 38 N 77 32 w 1769 565,TL CI-80-07 C009 2-IX-80 23 55 N 75 29 w 2168 360,TL CI-80-07 C008 2-IX-80 24 14 N 76 06 w 1787 357,TL CI-80-07 C008 2-IX-80 24 14 N 76 06 w 1787 350,TL CI-80-07 C008 2-IX-80 24 14 N 76 06 w 1787 415,TL Ilyophis br unneus 'CI-80-07 C039 14-IX-80 24 25 N 77 24 w 1578 430,TL CI-80-07 C041 15-IX-80 23 53 N 77 07 w 1341 425,TL 132 133 Appendix 1. Continued. Cruise Station Date Lat . Long . Depth Length Illyophis; brunneus cont. CI-80-07 C041 15-IX-80 23 53 N 77 07 W 1341 340,TL Simenchelys parasiticus AL-80-05 21 9-V-80 4002 N 69 02 W 737 275,TL Halosauropsis macrchir AL-80-05 6 7-V-80 39 46 N 71 21 W 1130 250,GPL VS-79-03 8 13-XII-79 37 25 N 74 38 W 1550 270,GPL VS-79-03 8 13-XII-79 37 25 N 74 38 W 1550 253 ,GPL VS-79-03 8 13-XII-79 37 25 N 74 38 W 1550 240,GPL VS-79-03 8 13-XII-79 37 25 N 74 38 W 1550 249,GPL Aldrovandia affinis CI-80-07 C012 3-IX-80 23 51 N 75 19 W 1856 135,GPL Aldrovandia gracilis CI-80-07 C008 2-IX-80 24 14 N 76 06 W 1787 122,GPL CI-80-07 C008 2-IX-80 24 14 N 76 06 W 1787 174,GPL CI-80-07 C007 l-IX-80 24 31 N 76 29 W 1683 8 2 ,GPL CI-80-07 C007 l-IX -80 24 31 N 76 29 W 1683 135,GPL Aldrovandia phalacra AD-213-3 6383 2-XI-79 29 52 N 77 09 W 1026 106,GPL AD-213-3 6383 2-XI-79 29 52 N 77 09 W 1026 105,GPL AD-213-3 6383 2-XI-79 29 52 N 77 09 W 1026 84,GPL AD-213-3 6383 2-XI-79 29 52 N 77 09 W 1026 133,GPL AD-213-3 6383 2-XI-79 29 52 N 77 09 W 1026 94,GPL AD-213-3 6373 31-X-79 32 36 N 76 37 W 1003 162,GPL AD-213-3 6373 31-X-79 32 36 N 76 37 W 1003 103,GPL Narcetes ;stomias CI-80-07 C038 14-IX-80 24 38 N 77 32 W 1769 495,SL Alepocephalus agassizii VS-79-03 8 7-IX-80 37 25 N 74 38 W 1550 289,SL AL-80-05 25 9-V-80 39 58 N 68 54 W 1244 342,SL AD-213-3 6350 27-X-79 39 11 N 72 13 W 1031 235,SL AD-213-3 6350 27-X-79 39 11 N 72 13 W 1031 300,SL 134 Appendix 1. Continued. Cruise Station Date Lat . Long . Depth Length Conocara macropterum CI-80-07 C042 15-IX-80 23 48 N 77 04 W 1373 143,SL CI-80-07 C042 15-IX-80 23 48 N 77 04 W 1373 135,SL CI-80-07 C042 15-IX-80 23 48 N 77 04 W 1373 156,SL CI-80-07 C007 l-IX-80 24 31 N 76 29 W 1683 220,SL CI-80-07 C011 3-IX-80 23 46 N 75 48 W 1790 175,SL CI-80-07 C011 3-IX-80 23 46 N 75 48 W 1790 150,SL CI-80-07 C011 3-IX-80 23 46 N 75 48 w 1790 125,SL Chlorophthalmus agassizii AL-80-05 13 8-V-80 39 58 N 70 20 w 338 93, SL AL-80-05 13 8-V-80 39 58 N 70 20 w 338 9 3 ,SL Bathypterois longipes CI-80-07 C045 17-IX-80 26 30 N 76 05 w 4777 145,SL CI-80-07 C034 ll-IX -80 26 09 N 75 25 w 4539 125,SL Bathypterois phenax CI-80-07 C018 7-IX-80 22 45 N 75 34 w 2337 134,SL Bathypterois bigelowi AD-213-3 6387 4-XI-79 29 08 N 78 57 w 830 134,SL Ipnops murrayi CI-80-07 C016 6-IX-80 22 55 N 75 14 w 2443 110.SL CI-80-07 C019 7-IX-80 22 33 N 75 35 w 2565 80, SL CI-80-07 C041 15-IX-80 23 53 N 77 07 w 1341 117,SL Bathytyphlops marionae AD-213-3 6383 2-XI-79 29 52 N 77 09 w 1026 294,SL Bathymicrops regis CI-80-07 C034 ll-IX -80 26 09 N 75 25 w 4539 109,SL CI-80-07 C034 ll-IX -80 26 09 N 75 25 w 4539 110,SL Lophius americanus AD-213-3 6358 28-X-79 36 24 N 74 43 w 841 240,TL AD-213-3 6418 10-XI-79 36 51 N 74 39 w 143 160,SL AD-213-3 6418 10-XI-79 36 51 N 74 39 w 143 157,SL AD-213-3 6418 10-XI-79 36 51 N 74 39 w 143 210,SL AD-213-3 6418 10-XI-79 36 51 N 74 39 w 143 190,SL 135 Appendix 1. Continued Cruise Station Date La t . Long . Dep th Length Dibranchus atlanticus AD-213-3 6370 30-X-79 33 25 N 76 21 W 627 104,SL VS-79-02 9 27-IX-79 36 40 N 74 41 W 500 149,TL VS-79-02 9 27-IX-79 36 40 N 74 41 w 500 154,TL VS-79-02 9 27-IX-79 36 40 N 74 41 w 500 48, TL AL-80-05 13 8-V-80 39 58 N 70 20 w 338 125,SL AL-80-05 13 8-V-80 39 58 N 70 20 w 338 60, SL AL-80 05 13 8-V-80 39 58 N 70 20 w 338 123,SL Urophycis chuss AD-213-3 _ __ 238,SL AD-213-3 -— --- 294,SL AD-213-3 --- - - 341,SL AD-213-3 6428 12-XI-79 39 23 N 74 41 w 100 292,SL AD-213-3 6428 12-XI-79 39 23 N 74 41 w 100 325,SL Urophycis tenuis AD-213-3 _ __ _ _ 375,SL AD-213-3 6354 27-X-79 39 15 N 72 18 w 450 410,SL Phycis chesteri AD-213-3 6352 27-X-79 39 11 N 72 16 w 845 229,SL AD-213-3 6352 27-X-79 39 11 N 72 16 w 845 235,SL AD-213-3 6352 27-X-79 39 11 N 72 16 w 845 323,SL AD-213-3 6358 28-X-79 36 24 N 74 43 w 841 265,SL AD-213-3 6358 28-X-79 36 24 N 74 43 w 841 257,SL AD-213-3 6358 28-X-79 36 24 N 74 43 w 841 269,SL AD-213-3 6358 28-X-79 36 24 N 74 43 w 841 286,SL Merluccius bilinea r is AD-213-3 6418 10-XI-79 36 51 N 74 39 w 143 271,SL AD-213-3 6418 10-XI-79 36 51 N 74 39 w 143 243,SL AD-213-3 6418 10-XI-79 36 51 N 74 39 w 143 255,SL Merluccius albidus AD-213-3 6389 4-XI-79 29 00 N 79 47 w 624 182,SL AD-213-3 6371 30-X-79 33 35 N 76 34 w 427 180,SL Dicrolene in tro n ig ra CI-80-07 C063 21-IX-80 29 06 N 77 08 w 1093 234,SL CI-80-07 C063 21-IX-80 29 06 N 77 08 w 1093 230,SL CI-80-07 C063 21-IX-80 29 06 N 77 08 w 1093 211,SL 136 Appendix 1. Continued Cruise Station Date Lat . Long . Depth Length Porogadus silu s CI-81-13 D009 18-XI-81 21 27 N 74 21 W 2714 130,SL CI-81-13 D009 18-XI-81 21 27 N 74 21 W 2714 110,SL CI-81-13 D004 16-XI-81 21 57 N 74 46 w 2728 120,SL Bassozetus normalis CI-81-13 D004 16-XI-81 21 57 N 74 46 w 2728 157,SL Bassozetus taenia CI-81-03 D009 18-XI-81 21 27 N 74 21 w 2714 185,SL Abyssobrotula ealathea CI-81-13 D016 23-XI-81 24 04 N 68 03 w 5690 125,SL Bathyonus n e c to ra lis CI-81-13 D007 17-XI-81 21 50 N 74 56 w 2758 157 ,SL CI-81-13 D007 17-XI-81 21 50 N 74 56 w 2758 143,SL CI-81-13 D007 17-XI-81 21 50 N 74 56 w 2758 122,SL CI-81-13 D009 18-XI-81 21 27 N 74 21 w 2714 140,SL CI-81-13 D009 18-XI-81 21 27 N 74 21 w 2714 140,SL Acanthonus armatus CI-80-07 C011 31-IX-80 23 46 N 75 48 w 1790 196,SL CI-80-07 C011 31-IX-80 23 46 N 75 48 w 1790 196,SL CI-81-13 D009 18-XI-81 21 27 N 74 21 w 2714 155,SL Nezumia b a ird ii AD-312-3 6344 26-X-79 39 50 N 60 55 w 1035 45 ,HL AD-213-3 6344 26-X-79 39 50 N 60 55 w 1035 53,HL AD-213-3 6347 26-X-79 39 53 N 50 54 w 670 43 ,HL AL-80-07 15 8-V-80 39 53 N 69 47 w 436 15, HL AL-80-07 15 8-V-80 39 53 N 69 47 w 436 15 ,HL AD-213-3 6347 26-X-79 39 53 N 70 54 w 670 40, HL AD-213-3 6347 26-X-79 39 53 N 70 54 w 670 46 ,HL AD-213-3 6347 26-X-79 39 53 N 70 54 w 670 52,HL AD-213-3 6383 2-XI-79 29 52 N 77 09 w 1026 55,HL Nezumia aequalis AD-213-3 6373 31-X-79 32 36 N 76 37 w 1003 46,HL AD-213-3 6373 31-X-79 32 36 N 76 37 w 1003 4 2 ,HL AL-80-05 17 9-V-80 39 54 N 69 42 w 377 12,HL AL-80-05 17 9-V-80 39 54 N 69 42 w 377 16,HL AD-213-3 6373 31-X-79 36 36 N 76 37 w 1003 42 ,HL 137 Appendix 1. Continued Cruise Station Date Lat. Long. Depth Length Nezumia ae qualis cont. AD-213-3 6373 31-X-79 36 36 N 76 37 W 1003 42,HL Coelorinchus c . carminatus VS-79-02 12 28-IX-79 36 38 N 74 39 W 640 60,HL VS-79-02 12 28-IX-79 36 38 N 74 39 W 640 52,HL VS-79-02 12 28-IX-79 36 38 N 74 39 W 640 46,HL AL-80-05 13 8-V-80 39 58 N 70 20 W 338 35, HL AL-80-05 13 8-V-80 39 58 N 70 20 W 338 44,HL AL-80-05 13 8-V-80 39 58 N 70 20 W 338 31 ,HL VS-79-02 12 28-IX-79 36 38 N 74 39 W 640 61 ,HL VS-79-02 10 27-IX-79 36 37 N 74 40 W 630 56,HL AD-213-3 6361 29-X-79 34 42 N 75 29 W 612 49 ,HL AD-213-3 6361 29-X-79 34 42 N 75 29 W 612 42, HL AD-213-3 6361 29-X-79 34 42 N 75 29 W 612 52 ,HL AD-213-3 6361 29-X-79 34 42 N 75 29 W 612 48, HL AD-213-3 6389 4-XI-79 29 00 N 79 47 W 624 50,HL AD-213-3 6359 28-X-79 36 23 N 74 43 W 750 52,HL Coryphaenoides ru p e s tris AD-213-3 6344 26-X-79 39 50 N 70 55 W 1035 65 ,HL AD-213-3 6344 26-X-79 39 50 N 70 55 W 1035 93,HL AD-213-3 6344 26-X-79 39 50 N 70 55 W 1035 94 ,HL AD-213-3 6344 26-X-79 39 50 N 70 55 W 1035 105,HL AD-213-3 6344 26-X-79 39 50 N 70 55 W 1035 133,HL AL-80-05 5 7-V-80 39 49 N 70 59 W 963 65, HL AD-213-3 6346 26-X-79 39 50 N 70 55 W 1035 80 ,HL Coryphaenoides armatus VS-79-03 3 12-XII-79 36 38 N 73 22 W 2900 78,HL VS-79-03 3 12-XII-79 36 38 N 73 22 W 2900 69, HL VS-79-03 3 12-XII-79 36 38 N 73 22 W 2900 6 8 ,HL VS-79-03 3 12-XII-79 36 38 N 73 22 w 2900 76,HL Coryphaenoides le p to le p is CI-80-07 C034 ll-IX -80 26 09 N 75 25 w 4539 105,HL Coryphaenoides carapinus VS-79-03 4 12-XI1-79 36 37 N 74 05 W 2550 50,HL VS-79-03 4 12-XII-79 36 37 N 74 05 w 2550 49, HL VS-79-03 4 12-XII-79 36 37 N 74 05 w 2550 39,HL VS-79-03 4 12-XII-79 36 37 N 74 05 w 2550 5 6 ,HL 138 Appendix 1. Continued. Cruise Station Date La t . Long . Depth Length Antimora rostrata VS-79-03 8 13-XII-79 37 25 N 74 38 W 1550 295,SL VS-79-03 8 13-XII-79 37 25 N 74 38 W 1550 295,SL VS-79-03 4 12-XII-79 36 37 N 74 05 W 2550 339,SL VS-79-03 4 12-XII-79 36 37 N 74 05 W 2550 370,SL VS-79-03 4 12-XII-79 36 37 N 74 05 W 2550 365,SL AD-213-3 6344 26-X-79 39 50 N 70 55 W 1035 370,SL VS-79-03 8 13-XII-79 37 25 N 74 38 W 1550 375,SL Lycodes atlanticus VS-79-03 9 14-XII-79 1375 29 7,TL VS-79-03 9 14-XII-79 1375 252,TL VS-79-03 9 14-XII-79 1375 256,TL VS-79-03 9 14-XII-79 1375 310,TL Lycenchely s verrilli VS-79-02 10 27-IX-79 36 37 N 74 40 W 630 143,TL VS-79-02 10 27-IX-79 36 37 N 74 40 W 630 178,TL VS-79-02 10 27-IX-79 36 37 N 74 40 W 630 192,TL VS-79-02 10 27-IX-79 36 37 N 74 40 W 630 165,TL VS-79-02 10 27-IX-79 36 37 N 74 40 W 630 181,TL Lycenchely s p ax illa VS-79-02 12 28-IX-79 36 38 N 74 39 W 549 230,TL AL-80-05 5 7-V-80 39 49 N 70 59 W 963 252,TL AL-80-05 9 8-V-80 39 51 N 70 37 w 726 186,TL AL-80-05 5 7-V-80 39 49 N 70 59 w 963 252,TL AL-80-05 5 7-V-80 39 49 N 70 59 w 963 223,TL Melanostigma atlanticum AL-80-05 11 8-V-80 39 53 N 70 22 w 586 130,TL Helicolenu s dactylopterus 142,SL — 145,SL Glyptocephalus cynoglossus AD-213-3 6344 14-XI-79 39 46 N 71 33 w 848 202,SL AD-213-3 6344 14-XI-79 39 46 N 71 33 w 848 238,SL AD-213-3 6344 14-XI-79 39 46 N 71 33 w 848 178,SL AD-213-3 6344 14-XI-79 39 46 N 71 33 w 848 224,SL VS-79-02 10 27-IX-79 36 37 N 74 40 w 630 240,SL Appendix 2. Abbreviations used in text. AFDW - Ash free dry weight GPL - Gnathoproctal length HL - Head length SL “ Standard length TL - Total length 139 Vita The author was born on October 8, 1954, in Chapel Hill, North Carolina. He attended the fine public school system in Travelers Rest, South Carolina. In 1976 the author graduated from Furman U niversity and entered the marine science program at the University of South Carolina, recieving a Master's degree in 1978. He then entered the Ph. D. program at the College of William and Mary. The author is now employed as a cross-stitch tycoon. 140