ELSEVIER 
J.lJUI=IL^IJrl=L-U 
Available online at www.sciencedirect.com 
*??' ScienceDirect 
Deep-Sea Research II ? (IHI) Ill-Ill 
DEEP-SEA RESEARCH 
PART II 
www.elsevier.com/locate/dsr2 
Small-scale distribution of deep-sea demersal nekton 
and other megafauna in the Charlie-Gibbs Fracture Zone 
of the Mid-Atlantic Ridge 
J.D. Felley^'*, M. Vecchione^, R.R. Wilson Jr.'^ 
"^Office of the Chief Information Officer, MRC S03, Smithsonian Institution, Washington, DC 20560, USA 
^NOAA/NMFS Systematics Laboratory, MRC 153, National Museum of Natural History, Washington, DC 20013-7012, USA 
'^Department of Biological Sciences, California State University, Long Beach, 1250 Bellflower Blvd., Long Beach, CA 90840, USA 
Accepted 14 September 2007 
Abstract 
Videotapes from manned submersibles diving in the area of the Charlie-Gibbs Fracture Zone of the Mid-Atlantic Ridge were used to 
investigate the distribution of fishes, large crustaceans, epifaunal and sessile organisms, and environmental features along a series of 
transects. Submersibles MIR 1 and MIR 2 conducted paired dives in an area of mixed sediment and rock (beginning depth ca. 3000 m) 
and on a large pocket of abyssal-like sediments (depth ca. 4000 m). In the shallower area, the submersibles passed over extremely 
heterogeneous terrain with a diversity of nekton, epifaunal forms and sessile forms. In the first pair of dives, MIR 1 rose along the Mid- 
Atlantic Ridge from 3000 to 1700 m, while MIR 2 remained near the 3000 m isobath. Nekton seen in these relatively shallow dives 
included large and small macrourids (genus Coryphaenoides), shrimp (infraorder Penaeidea), Halosauropsis macrochir, Aldrovandia sp., 
Antimora rostrata, and alepocephahds. The last two were more characteristic of the upper areas of the slope reached by MIR 1, as it rose 
along the Mid-Atlantic Ridge to depths less than 3000 m. Distributions of some forms seemed associated with depth and/or the presence 
of hard substrate. Sessile organisms such as sponges and large cnidaria were more likely to be found in rocky areas. The second pair of 
dives occurred in an abyssal area and the submersibles passed over sediment-covered plains, with little relief and many fewer countable 
organisms and features. The most evident of these were holes, mounds, small cerianthid anemones, small macrourids and the holothurian 
Benthodytes sp. A few large macrourids and shrimp also were seen in these deeper dives, as well as squat lobsters {Munidopsis sp.). 
Sponges and larger cnidaria were mostly associated with a few small areas of rocky substrate. Holes and mounds showed distributions 
suggesting large-scale patterning. Over all dives, most sessile and epifaunal forms showed clumped distributions. However, large 
holothurians and large nekton often had distributions not significantly different from random. 
? 2007 Elsevier Ltd. All rights reserved. 
Keywords: Deep-ocean habitats; Submersibles; Video; Dispersion; Substrates 
1. Introduction 
Submersibles have extended the range at which we can 
directly observe the natural history of deep-sea organisms, 
in particular their behavior (e.g., Cohen, 1977; Drazen 
et al., 2003; Uiblein et al., 2003) and small-scale distribu- 
tion (e.g., Grassle et al., 1975; Smith and Hamilton, 1983; 
Kaufmann et al., 1989). Submersibles are especially useful 
*Corresponding author. Tel.: + I 202 633 0624; fax: + 1 202 633 2953. 
E-mail addresses: felleyj@si.edu (J.D. Felley), vecchiom@si.edu 
(M. Vecchione), rwilsonl@csulb.edu (R.R. Wilson Jr.). 
when investigating deep-sea regions of rough topography 
such as ridges and volcanic island slopes (e.g.. Chave and 
Malahoff, 1998), hydrothermal vents (Gebruk et al., 2000) 
and in the open ocean (Vinogradov, 2005). Interactions of 
deep-sea organisms with their environment can be docu- 
mented from videotapes taken from submersibles as well as 
by direct observations (Felley and Vecchione, 1994; Levin 
et al., 1994), providing insight into the environmental 
variables that affect the distribution of these forms. 
The MAR-ECO pilot project of the Census of Marine 
Life provided an opportunity for a short series of 
submersible dives in the Charlie-Gibbs Fracture Zone to 
0967-0645/S - see front matter ? 2007 Elsevier Ltd. All rights reserved. 
doi:10.1016/j.dsr2.2007.09.021 
Please cite this article as: Felley, J.D., et al., Small-scale distribution of deep-sea demersal nekton and other megafauna in the Charlie-Gibbs Fracture 
Zone of the Mid-Atlantic... Deep-Sea Research II (2007), doi:10.10I6/j.dsr2.2007.09.021 
inTOia?5?!5?C 
J.D. Felley et al. / Deep-Sea Research III ('llllj lll-l 
explore and photo-document the demersal nekton and 
benthic macrofauna in this very deep area of the Mid- 
Atlantic Ridge. The Charlie-Gibbs Fracture Zone is 'a 
major left lateral offset in the Mid-Atlantic Ridge at about 
52?30'N' (Fleming et al., 1970). Our submersible dives were 
the first in that area and produced the videotapes and 
observations that we discuss here. We analyzed patterns of 
distributions of organisms, comparing the different dives 
and comparing habitats seen within each dive. Our primary 
goal was to examine the small-scale distribution of nekton, 
but for comparison we also conducted similar analyses of 
epibenthic megafauna and signs of infauna. 
2. Materials and methods 
2.1. Dives 
Dives using the manned submersibles MIR 1 and MIR 2 
of the Russian Academy of Sciences were conducted in the 
northwestern region of the Charlie-Gibbs Fracture Zone in 
June 2003 during a trans-Atlantic crossing of the R.V. 
Akademik Mstislav Keldysh. The dives were conducted in 
tandem (using both submersibles). 
The first tandem set was on 11 June 2003 at 52?58'N; 
35?01'W in an area where the bottom rose from a depth of 
about 3000-1000 m. Both MIR 1 and MIR 2 reached the 
bottom at ca. 3100-3000 m and spent about 8 h each on 
the bottom. Both submersibles passed over extremely 
variable topography, including areas of flat bottom 
covered with sediments, slopes with both rock and areas 
of sediment, rocky slopes and wafls. MIR 1 followed the 
slope up to a depth of ca. 1700 m. The transect followed by 
MIR 2 proceeded along an isobath near 3000 m at the base 
of the slope. 
The second set of dives occurred on 13 June 2003 at 
52?47'N; 34?46'W in a large pocket of sediment similar to 
an abyssal plain at 4200^500 m. The bottom was mostly 
flat sediment, with a few rocks and/or clumps of mud. The 
submersibles each spent ca. 7 h on the bottom. Videotapes 
were recorded during the period when each submersible 
was near the bottom. 
2.2. Measurements and analyses 
since some of the dives were over extremely fiat topogra- 
phy). The percentage of sediment was an estimate of the 
proportion of the bottom traversed during the 1-min 
segment that was covered with sediment. Other variables 
were counts of particular features. Holes were large and 
were distinguished from 'fairy rings' (Gage and Tyler, 
1991), which consisted of sets of smafler holes forming a 
ring. Mounds were large, and were distinguished from 
large holothurian casts. One type of cast that we could 
identify specifically was a spiral cast formed by enterop- 
neusts (Hofland et al., 2005). Several morphotypes of 
sponges were counted separately. Tube sponges were single 
tubes rising from a base, whfle finger sponges had several to 
many tubes arising from a common base. Round sponges 
were essentially a short, fat tube, and round sponges that 
were bright yellow were counted separately. We also 
counted the remains of dead sponges. We counted 
morphotypes of sessile colonial cnidarians. Single stalks 
were called 'sea whips', those with a few branches were 
cafled 'branched gorgonians', while those with densely 
packed branches were called 'sea fans'. Several different 
types of echinoderms were counted, including stalked 
crinoids, feather stars (unstalked crinoids), brisingid stars, 
asteroids and various identifiable forms of holothurians. In 
MIR I's deep dive, many smafl holothurians of the genus 
Peniagone were seen covering the bottom. These were too 
dense to count, so the occurrence of these was coded 1 
(present) or 0 (absent) for segments in this dive. Finally, we 
counted gastropods. 
Nektonic organisms were classified into morphological 
types, some of which could consistently be identified to 
genus or species. For example, some large macrourids 
could confidently be identified as Coryphaenoides {Nema- 
tonurus) armatus. Not all macrourids could be identified 
consistently to species; they have therefore been lumped 
into the morphorype 'large macrourids'. A similar 'taxon' 
included all smafl macrourids. In the analyses that follow, 
such morphotypes, or 'taxa' shall be the units of analysis 
and interest. Supplementary information provides images 
of nekton and environmental features for each dive. 
Data from transects covered by the submersibles were 
described and analyzed separately. Statistical analyses were 
carried out with SYSTAT 11 (2006). 
We measured environmental variables and counted 
megafaunal forms including nekton, epifauna, and infau- 
nal signs (those benthic features that have been referred to 
as lebensspuren, e.g., Kaufmann et al., 1989). Counts were 
made from 1-min segments of the tapes, whenever the 
submersible was near the bottom and traveling slowly 
(ca. 1 m/s). This is similar to the visual fast count method 
shown by Strong et al. (2006) to be an effective way to 
quantify epibenthic organisms. The environmental vari- 
ables recorded are hsted in Table 1. Depth (m) was 
recorded whenever the value was noted by the observers 
either during or prior to the 1-min segment (the frequency 
that depth was stated varied by observer, and also by dive. 
3. Results 
3.1. Nekton forms seen 
Table 2 lists the numbers and frequencies of different 
nekton forms seen in the dives. Small macrourids were 
the most abundant, and may represent more than a single 
species. They were most abundant and more frequent in the 
shaflower dives, occurring in almost 50% of the 1-min 
segments. They were much rarer in the deep dives, 
occurring in one-quarter to one-third of the 1-min 
segments. Large macrourids were the next most common 
forms, and represented more than a single species. Those 
Please cite this article as: Felley, J.D., et al., Small-scale distribution of deep-sea demersal nekton and other megafauna in the Charlie-Gibbs Fracture 
Zone of the Mid-Atlantic... Deep-Sea Research II (2007), doi:10.1016/j.dsr2.2007.09.021 
J.lJUI=IL^IJrl=L-U 
J.D. Felley et al. / Deep-Sea Research III ('llli; lll-lll 3 
Table 1 
Means and standard deviations (in parentheses) of environmental variables recorded and features and organisms counted from 1-min segments of 
videotape taken by the MIR 1 and MIR 2 submersibles in dives in two different areas of the Charlie-Gibbs Fracture Zone of the Mid-Atlantic Ridge 
Variable Shallow Deep 
MIR 1 (n = 183) MIR 2(n = 87) MIR 1 {n = 105) MIR 2{n = 90) 
Depth (m) 3150-1740 3015-3000 4493-4361 4500-4200 
% Sediment 70.4 77.8 100 100 
(95% CI) (61.7-90.3) (69.4-84.9) - (99.9-100) 
Holes 0.62(1.282) 0.98 (1.294) 1.36(1.257) 3.59 (2.652) 
Fairy rings 0.27 (0.784) 0.47 (1.274) 0.26 (0.621) 0.53 (0.939) 
Mounds 0.22 (0.617) 0.68 (1.418) 28.5 (13.678) 18.7 (20.52) 
Spirals 0.66 (1.300) 0.18 (0.518) 0.02(0.137) 0.03 (0.235) 
Sponges 
Dead 0.21 (0.995) 0.07 (0.397) 0.02(0.211) 
Tube 2.80 (5.920) 0.12 (0.387) 0.05 (0.322) 0.16(0.652) 
Encrusting 1.21 (2.443) 0.16 (0.525) 
Finger 0.44(1.141) 0.12 (0.387) 0.08 (0.430) 
Vase 1.14(2.886) 0.12 (0.355) 0.06 (0.275) 
Round 0.57 (2.361) 0.07 (0.367) 
Yellow 0.30 (1.237) 0.02 (0.151) 
Cnidaria 
Sea fans 0.10 (0.616) 0.10 (0.571) 
Gorgonians 0.16 (0.720) 0.44 (1.148) 
Sea whips 0.45 (1.147) 1.67 (2.769) 
Small anemones 0.05(0.211) 0.83(1.105) 1.66(2.224) 
Large anemones 0.01 (0.104) 0.02 (0.151) 0.01 (0.098) 0.07 (0.328) 
Echinoderms 
Stalked crinoids 0.27 (0.681) 0.37 (0.764) 
Feather stars 0.44(1.318) 
Brisingid stars 0.04 (0.192) 0.36 (1.034) 
Benthothuria sp. 0.07 (0.248) 0.84(1.493) 
Benthodytes sp. 0.02 (0.147) 0.07 (0.297) 0.08 (0.267) 0.08 (0.269) 
Peniagone sp. 1.20 (3.509) 0.20 (0.662) 0.10(0.475) 
Green holothurian 0.03 (0.163) 
Peniagone sp. (presence) 0.61 (0.490) 
Gastropod 0.08 (0.376) 0.03 (0.184) 
Nekton 
Small macrourids 0.80 (1.517) 1.25 (2.354) 0.22 (0.460) 0.63 (0.999) 
Large macrourids 0.17 (0.443) 0.23 (0.564) 0.05 (0.255) 0.14(0.412) 
Halosauropsis macrochir 0.08 (0.339) 0.10 (0.342) 
Aldrovandia sp. 0.05 (0.260) 
Antimora rostrata 0.04 (0.230) 
Bathysaurus sp. 0.01 (1.04) 0.03 (0.184) 
Alepocephalids 0.04 (0.219) 
Shrimp 0.11 (0.405) 0.14 (0.347) 0.11 (0.320) 0.03 (0.181) 
Munidopsis sp. 0.01 (0.107) 0.04(0.192) 0.03 (0.181) 
The depth range is provided for depth, '?' is the number of usable 1-min transect segments on each dive. Means and 95% confidence intervals of 
percentage of sediment were calculated from arcsine transformed values of percentages. 
seen clearly enough to identify included the abyssal 
grenadier Coryphaenoides (Nematonurus) armatus and 
probably the ghostly grenadier Coryphaenoides (Chalinura) 
leptolepis (the most abundantly caught macrourid of the 
subgenus Chalinura in the eastern Atlantic, Merrett et al., 
1991). Both small and large macrourids were usually seen 
hovering over the bottom, and sometimes in contact with 
the bottom. Large macrourids (especially C. armatus) often 
approached the submersibles closely, investigating them. 
Macrourids large and small were the only fishes seen in the 
deeper dives in the 1-min segments analyzed. A bath- 
ysaurid species (suspected to be Bathysaurus mollis because 
the sighting was deeper than 3000 m; e.g., Sulak et al., 
1985; Haedrich and Merrett, 1988) was seen by MIR 2 in 
its deep dive but was not included in the analysis (for 
images of this individual see Supplementary Information). 
Other fishes were seen during the two shallower dives, 
most commonly the halosaur Halosauropsis macrochir. 
This form was usually seen on the bottom with its belly on 
the substrate and its tail extended perpendicular to the 
Please cite this article as: Felley, J.D., et al., Small-scale distribution of deep-sea demersal nekton and other megafauna in the Charlie-Gibbs Fracture 
Zone of the Mid-Atlantic... Deep-Sea Research II (2007), doi:10.1016/j.dsr2.2007.09.021 
inTOia?5?!5?C 
J.D. Felley et al. / Deep-Sea Research III ('llli; lll-l 
Table 2 
Nekton counted from 1-min segments of videotape taken by the submersibles MIR 1 and MIR 2 during dives in two different areas of the Charlie-Gibbs 
Fracture Zone of the Mid-Atlantic Ridge 
Taxon Shallow Deep 
MIR 1 (? = 183) MIR 2 (? = 87) MIR 1 (? = 105) MIR 2{n = 90) 
Small macrourids 147 (70) 109 (40) 23 (21) 57 (34) 
Large macrourids 31 (26) 20 (15) 5(4) 13(11) 
Halosauropsis macrochir 14(11) 9(8) 
Aldrovandia sp. 1(1) 4(3) 
Synaphobranchid eel 1(1) 1(1) 
Antimora rostrata 8(7) 
Bathysaurus sp. 2(2) 3(3) 
Ophidioid 2(2) 1(1) 
Alepocephalids 7(6) 
Shrimp 20 (15) 12 (12) 12 (12) 3(3) 
Spider crab 1(1) 
Munidopsis sp. 1(1) 4(4) 3(3) 
Pycnogonid 1(1) 
Cirrate octopod 1(1) 1(1) 
Given are total numbers of individuals of a taxon seen in each dive and (in parentheses) number of 1 -min segments in which a taxon was seen. 
bottom. Another halosaur was a very skinny, pale form, 
identified as Aldrovandia sp. This form was usually seen 
with its lower jaw touching or close to the bottom, its body 
rigidly straight and extended up at an angle to the bottom. 
Bathysaurs, again suspected to be B. mollis (depths being 
deeper than 3000 m) were also seen by both MIR 1 and 
MIR 2, always motionless on the bottom. Several 
alepocephalids, blue hake Antimora rostrata and Cory- 
phaenoides rupestris (classified as a large macrourid) were 
seen by MIR 1 during the shallower dive, as the 
submersible rose with the bottom. Alepocephahds typically 
were seen swimming several meters above the substrate, 
while individuals of A. rostrata were seen actively swim- 
ming close to the bottom. MIR 1 found two ophidioid 
individuals and a single synaphobranchid eel, all swimming 
a meter or two over the substrate. MIR 2 encountered one 
individual each of these forms. 
Invertebrate nekton were seen in all dives. Shrimp 
(infraorder Penaeidea) were the most common. The squat 
lobster Munidopsis sp. was seen in the deeper dives, and a 
single individual was seen by MIR 2 at the 3000 m depth. 
During the shallower dives, a spider crab and a pycnogonid 
were seen by MIR 1, both walking on the bottom, and in 
their shallow dives, MIR 1 and MIR 2 each saw a cirrate 
octopod hovering in the water column. 
3.2. Description of dives 
On its shallow dive, MIR 1 rose more than 1500 m, from 
below 3000 m to almost 1700 m (Table 1). The submersible 
traversed a variety of topographic features including fiat 
sediment areas, areas of sediment with rocky outcroppings, 
rocky slopes, and steep rocky cliffs. The means for 
environmental variables recorded in the 1-min segments 
are presented in Table 1, and show the variabihty of the 
environment.  In areas with sediment were found such 
features as holes, mounds and casts such as the spiral casts 
of enteropneusts, as well as the holothurians Benthothuria 
sp., Benthodytes sp., and Pentagone sp. Areas with rock 
were characterized by attached sessile invertebrates such as 
sponges and cuidarla. MIR 1 traversed an area with 
numerous sea whips, branched gorgonians and sea fans 
(a cuidar?an garden, where up to 16 were seen in a 1-min 
segment) and also an area with many sponges (a sponge 
garden, with up to 80 sponges seen in a 1-min segment). 
3.3. Analyses of distributions 
Means and standard deviations of numbers (Table 1) 
and the coefficients of dispersion that can be calculated 
from them (Sokal and Rohlf, 1981; Ludwig and Reynolds, 
1988) suggested that most forms showed clumped (con- 
tagious) distributions, with more 1-min segments with high 
counts and more with very low (0) counts than expected for 
a random distribution. The coefficient of dispersion can be 
tested against a null hypothesis of random (Poisson) 
distribution of numbers using a /^ test. For 1-min segments 
from the MIR 1 shaUow dive, we calculated -/^ statistics for 
counts of 25 of the benthic features, sessile invertebrates, 
and nekton, using a Bonferroni correction for tables of 
tests (Rice, 1989). We excluded forms or features with low 
numbers (counts < 6), which in this dive excluded counts of 
large anemones, Benthodytes sp., the green holothurian and 
bathysaurs. The only forms that did not show a clumped 
distribution were large macrourids and the holothurian 
Benthothuria sp. 
It was apparent that for some forms, clumped distribu- 
tions were due to their occurrence in particular depth 
ranges, and also due to the different habitats traversed in 
each depth range. For the MIR 1 shallow dive (the only 
one that traversed a wide range of depths) we therefore 
divided up the 1-min segments by depth into four groups 
Please cite this article as: Felley, J.D., et al., Small-scale distribution of deep-sea demersal nekton and other megafauna in the Charlie-Gibbs Fracture 
Zone of the Mid-Atlantic... Deep-Sea Research II (2007), doi:10.1016/j.dsr2.2007.09.021 
J.lJUI=IL^IJrl=L-U 
J.D. Felley et al. / Deep-Sea Research III ('llli; lll-lll 5 
Table 3 
Means and standard deviations (in parentheses) of environmental variables recorded and features and organisms counted from 1-min segments of 
videotape from MIR I's shallow dive 
Variable Depth 
>3000m (11 = 36) 3000-2500 m (? = 68) 2501-2000m(n = 63) <2000m (11 = 21) 
% Sediment (mean) 72.9 95.7 
(95% CI) (53.4-100) (92.0-100) 
Holes 0.68 (1.077) 1.25 (1.740) 
Fairy rings 0.24 (0.682) 0.60(1.122) 
Mounds 0.64(1.018) 0.24 (0.601) 
Spirals 0.64 (0.985) 1.46(1.723) 
Sponges 
Dead 0.13 (0.341) 
Tube 0.29 (0.588) 
Encrusting 0.26 (0.445) 
Finger 0.71 (1.039) 
Vase 
Round 
Yellow 
Cnidaria 
Sea fans 0.52 (1.387) 
Gorgonians 0.94(1.548) 
Sea whips 1.61 (2.155) 0.06 (0.382) 
Large anemones 0.03 (0.180) 
Echinoderms 
Stalked crinoids 0.74(1.182) 0.03 (0.170) 
Brisingid stars 0.03 (0.170) 
Benthothuria sp. 0.10 (0.301) 0.13 (0.341) 
Benthodytes sp. 0.03 (0.180) 0.04 (0.207) 
Peniagone sp. 0.03 (0.180) 3.29 (5.152) 
Green holothurian 0.07 (0.263) 
Gastropod 0.15 (0.466) 
Nekton 
Small macrourids 0.52 (0.926) 1.16(1.767) 
Large macrourids 0.13 (0.341) 0.18 (0.487) 
Halosauropsis macrochir 0.06 (0.250) 0.13 (0.486) 
Antimora rostrata 0.04 (0.207) 
Alepocephalids 
Shrimp 0.19 (0.477) 0.16(0.536) 
31.4 
(39.0-48.0) 
0.08 (0.326) 
0.06 (0.304) 
0.02(0.126) 
0.49(1.625) 
6.11 (8.619) 
2.56 (3.222) 
0.87(1.661) 
2.73 (4.289) 
0.93 (3.583) 
0.64(1.937) 
0.05 (0.280) 
0.30 (0.586) 
0.02(0.126) 
0.40 (0.661) 
0.08 (0.272) 
0.79(1.598) 
0.22 (0.490) 
0.05 (0.215) 
0.05 (0.280) 
0.05 (0.280) 
0.05 (0.215) 
48.4 
(30.5-64.4) 
0.10 (0.436) 
0.14 (0.478) 
5.62 (2.958) 
2.48 (2.839) 
0.14(0.478) 
1.76 (1.998) 
2.14(2.516) 
0.71 (1.140) 
0.48 (0.873) 
0.10(0.436) 
0.05 (0.218) 
0.10 (0.402) 
0.19 (0.402) 
This dive rose along a ridge from 3150 to 1750 m. 'n' is the number of 1-min segments in a depth range. Means and standard deviations for these forms are 
given for different depth ranges. Means and 95% confidence intervals of percentage of sediment were calculated from arcsine transformed values of 
percentages. 
(Table 3). When MIR 1 was below 3000 m, it traversed 
a heterogeneous bottom with expanses of sediment and 
also large areas of rocky outcrops. Coefficients of 
dispersion were assessed for counts in 1-min segments 
below 3000 m, Bonferroni corrections were again apphed, 
and the forms excluded because of low counts were dead 
sponges, large anemones, the three holothurians seen at 
that depth, large macrourids, and H. macrochir. Below 
3000 m, clumped distributions were seen for the cnidaria 
and stalked crinoids (which were abundant. Table 3), 
forms strongly associated with the rocky bottom areas. 
Of the soft-bottom features, only the fairy rings showed 
a clumped distribution. None of the other counts or 
features showed significant deviations from random 
distributions. 
MIR 1 traversed soft bottom environments from 3000 to 
2500 m, with very few rocks. No sponges, branched 
gorgonians or sea fans were seen, and stalked crinoids, 
brisingid stars and sea whips were too few to analyze. Also 
excluded from further analysis were Benthodytes sp. and 
A. rostrata. In this depth range, large numbers of 
Peniagone sp. were seen, up to 23 in a 1-min segment. 
Counts of holes, fairy rings, spirals, Peniagone sp., small 
macrourids, H. macrochir and shrimp indicated clumped 
distributions. Other forms did not show significant devia- 
tions from random distributions. 
From 2500 to 2000 m, the substrate was rocky with 
patches of sediment. Soft-bottom features such as holes 
and mounds were few and were excluded from analysis. 
No fairy rings were seen. Sponges were quite abundant 
Please cite this article as: Felley, J.D., et al., Small-scale distribution of deep-sea demersal nekton and other megafauna in the Charlie-Gibbs Fracture 
Zone of the Mid-Atlantic... Deep-Sea Research II (2007), doi:10.1016/j.dsr2.2007.09.021 
inTOia?5?!5?C 
J.D. Felley et al. / Deep-Sea Research III ('llllj lll-l 
(Table 3) and all forms had significantly clumped distribu- 
tions. Stalked crinoids showed a significantly over-dis- 
persed distribution, the only such distribution seen in these 
dives. Other than sea whips (which were apparently 
randomly distributed), cuidar?an forms were few and were 
not analyzed, and none of the holothurian forms were seen. 
The only nekton seen in enough numbers for analysis were 
small macrourids and large macrourids. Small macrourids 
had a clumped distribution, while large macrourids showed 
no significant deviation from random distribution. 
Above 2000 m, the substrate was very rocky and rugged. 
Sponges were abundant. While there were areas of 
sediment, the only soft-bottom features counted were two 
holes that appeared in a single 1-min segment. Finger 
sponges and the remains of dead sponges were also rare in 
this depth range. No stalked crinoids were seen, and none 
of the large holothurians we counted were found in this 
range. Sea whips were the only type of sessile cuidar?an 
seen and were distributed apparently at random. None of 
the nekton forms appeared in numbers large enough to 
allow analyses of distribution. Encrusting, vase, and round 
sponges showed significantly clumped distributions, while 
tube sponges and yellow sponges had distributions that 
were not significantly different from random. 
MIR 2, travehng along the 3000 m isobath, also found 
the bottom to be mixed sediment and rock and the 
topography to be extremely variable. Like MIR 1, MIR 2 
encountered a range of topographic features including flat 
sediment areas, areas of sediment with rocky outcroppings, 
rocky slopes, and steep rocky cliffs. The means for 
environmental variables recorded in the 1-min segments 
are presented in Table 2, and illustrate the variability of the 
environment. In areas with sediment were found such 
features as holes, mounds and casts such as the spiral casts 
of enteropneusts, as well as the holothurians Benthothuria 
sp., Benthodytes sp., and Peniagone sp. Areas with rock 
were characterized by attached sessile invertebrates such as 
sponges and cuidar?a. This dive encountered an area of 
densely distributed cuidar?a (up to 19 sea whips, gorgo- 
nians and sea fans in a 1-min segment), but did not 
encounter the type of dense sponge garden that was seen by 
MIR 1. Though individuals o? Peniagone sp. were seen, no 
dense aggregations were seen by MIR 2. 
Most of the forms commonly encountered by MIR 2 
along the 3000 m isobath had clumped distributions. Dead 
sponges, round sponges and yellow sponges were not often 
seen, and were not included in these analyses. Anemones 
(both large and small), gastropods and the holothurian 
Benthodytes sp. were also excluded from analysis. Of the 
soft-bottom features, holes, mounds, fairy rings and spirals 
all had clumped distributions, reflecting their occurrence in 
the 1-min segments with much sediment. Also character- 
istics of these areas were the holothurians Benthothuria sp. 
and Peniagone sp. and both forms had clumped distribu- 
tions. Sea fans, branched gorgonians, sea whips, brisingid 
seastars and stalked crinoids were more likely to be found 
in   rocky   areas,   and   they   ah   demonstrated   clumped 
distributions. Feather stars co-occurred with cuidar?a, 
and showed a clumped distribution. Results were not as 
clear-cut for sponges. Encrusting sponges were strongly 
associated with the rocks they encrusted, and had a 
clumped distribution. The distributions of tube sponges, 
vase sponges and finger sponges did not differ significantly 
from random. Of the nekton forms, small macrourids, 
large macrourids, H. macrochir and shrimp were common 
enough to examine distributions. Only small macrourids 
showed a distribution that was significantly different from 
random, and they showed a clumped distribution. 
The second set of dives occurred in a much deeper area 
with habitats more characteristic of the abyssal plain. The 
environment was almost entirely soft bottom with many 
mounds and holes. There were very few hard-substrate 
areas. Both submersibles traversed generally similar areas, 
though MIR 2 did encounter an area with rocks and 
accompanying sessile organisms. In the areas traversed by 
both submersibles, fairy rings, mounds, and smah cer- 
ianthid anemones had significantly clumped distributions. 
Benthodytes sp. had a distribution that did not differ 
significantly from random. In the MIR 1 deep dive, holes 
were distributed apparently at random, while in the area 
traversed by MIR 2, they were significantly clumped. 
Conversely, small macrourids were distributed in clumped 
fashion as seen by MIR 2, and were distributed apparently 
at random in the area covered by MIR 1. Shrimp 
(distribution apparently random) were seen by MIR 1, 
but not by MIR 2. For most of its dive, MIR 2 passed over 
a carpet of very small holothurians {Peniagone sp.) too 
numerous to count. These were seen in 61% of the 1-min 
segments. Large Peniagone sp. individuals were seen by 
MIR 2, and had a clumped distribution. MIR 2 passed a 
rocky area with tube and finger sponges. These two forms 
showed clumped distributions, as would be expected. 
Finally, MIR 2 saw several large macrourids, distributed 
apparently at random. 
During ah dives, data collection from videotape was 
interrupted at various times when the submersible stopped 
to coflect samples or moved too high above the bottom to 
aflow successful counts. Thus, contiguous 1-min segments 
were usuafly limited to single minutes to up to 7 min at a 
time, with interruptions of 1 to several minutes. In the MIR 
1 deep dive, we were able to obtain 37 min of uninterrupted 
videotape. We used these data to calculate serial correla- 
tions between adjacent 1-min segments, to see if counts 
from one segment could be used to predict counts in the 
next segment. For this data subset, counts of holes, fairy 
rings, mounds, small anemones, Benthodytes sp., and large 
and small anemones were subjected to autoregressive 
integrated moving average analysis (ARIMA, Systat 11) 
(Ripley, 1981). Only holes and mounds showed significant 
serial correlations (0.72 and 0.97, respectively) of counts 
transformed as logarithm (count+1). Inspection of the 
data showed that during the 37-min period, the submer- 
sible moved from an area with relatively more holes and 
mounds to an area with fewer (averages of 1.6 holes and 
Please cite this article as: Felley, J.D., et al., Small-scale distribution of deep-sea demersal nekton and other megafauna in the Charlie-Gibbs Fracture 
Zone of the Mid-Atlantic... Deep-Sea Research II (2007), doi:10.1016/j.dsr2.2007.09.021 
J.lJUI=IL^IJrl=L-U 
J.D. Felley et al. / Deep-Sea Research lit ('llli; lll-lll 
41.7 mounds per 1-min segment in the first lOmin, 0.2 
holes and 26.5 mounds per 1-min segment in the last 
10 min). 
4. Discussion 
The area of the Charhe-Gibbs Fracture Zone shows 
similarities to other surveyed deep-ocean habitats, but 
differences as well. The cuidar?a and sponge morphotypes 
seen were those characteristics of deep-ocean habitats. 
However, the densities of sessile cuidar?a in the shallower 
dives were much less than have been documented in much 
shaUower areas of Atlantic continental slopes (Gass and 
Wilhson, 2005; Reed et al., 2006; Roberts et al., 2006). The 
?shes found in the shallower of the two sets of dives were 
characteristic of Haedrich and Merrett's (1988) 'rise' fauna, 
found along continental slopes at depths < 4500 m, while 
the few forms seen in the second set of dives were those 
they associated with the abyssal plain (depths > 4500 m). 
The organisms counted were sparse in both sets of dives. In 
the shallower dives, there was an average of 9 countable 
organisms (sponges, cuidar?a, epifauna, and nekton) per 
1-min segment, ranging from 0 to 81 organisms per 
segment. The deeper set of dives covered areas much 
poorer in biota: an average of 2 countable organisms per 
segment, ranging from 0 to 13. Note that this count does 
not include the carpet of tiny Peniagone sp. seen in 55 
of the 90 1-min segments from MIR 2, dive 2. The 
shallower set of dives found a more diverse fauna than 
the deeper set: 14 nektonic forms vs. 4, 6 morphotypes of 
sponges vs. 3, 4 types of sessile cuidar?a vs. 2. No cuidar?an 
gardens were found in the deeper dives, although a sponge 
garden was found on a patch of hard substrate. Interest- 
ingly, during the second (deeper) pair of dives, the 
submersibles both traversed similar habitats and depths, 
but the faunal observations differed substantially between 
the submersibles. 
The lack of hard substrate in the deeper areas probably 
accounts for the fewer numbers of sessile forms such as 
sponges and sessile cuidar?a. In the deeper dives, the few 
areas of rocky substrate were colonized by sponges and 
other sessile forms. However, even in shallower dives, 
presence of hard substrate was not a guarantor of finding 
sessile forms. In the shaUow dives, there were 51 1-min 
segments with little sediment (<15% coverage), of which 
21 had 5 or fewer sessile organisms counted in the segment. 
None of these low-sediment samples had any of the soft- 
bottom features such as holes, mounds, spirals or fairy 
rings. Apparently, fairly large patches of sediment are 
necessary for the animals making these lebensspuren. 
Conversely, large expanses of soft substrate did not 
guarantee high counts of soft-bottom features. In the 
shahow set of dives, 107 1-min segments had 85% coverage 
of sediment or more. In 10 segments, there were no soft- 
bottom features counted, and 77 had 5 or fewer soft- 
bottom features. Much hke the deeper dives, hard substrate 
within these segments was colonized: in the  107 high- 
sediment areas, 1 or more sessile organisms were found in 
17 1-min segments. 
We tested hypotheses concerning patterns of distribution 
of the various features and organisms. Clumped or random 
patterns of distribution characterized the counts of both 
features and forms, with only a single example of an over- 
dispersed distribution (stalked crinoids in the MIR 2 
shaUow transect). Some patterns were easily explained. The 
clumped distributions of sponges and cuidarla could be 
accounted for by their requirements for hard substrate. 
However, as mentioned above, this could not be the entire 
answer, as these forms did not heavily colonize many areas 
rich in hard substrate. Of the soft-bottom features, fairy 
rings and the fecal spirals formed by enteropneusts had 
consistently clumped distributions, perhaps reflecting 
variabihty in substrate quahty. Holes and mounds were 
often distributed in clumped fashion, but an investigation 
of contiguous 1-min segments during the MIR 1 deep dive 
suggested mat there were larger-scale processes influencing 
distribution of holes and mounds, as there was significant 
autocorrelation in these segments: numbers of holes and 
mounds in one segment were predictable from counts in 
adjacent segments. For the other counted features and 
taxa, 1-min segments seemed to represent independent 
samples. 
Clumped patterns of distribution were the rule for soft- 
bottomed features in sediment-filled areas and for sessile 
organisms in rocky areas. Distributions of nekton and also 
of large holothurians were less easy to characterize. The 
blue hake A. rostrata and alepocephalids were seen only in 
the shallower parts of the ridge covered by MIR 1, 
coinciding with their known depth ranges (Merrett et al., 
1991). Shrimp and the squat lobster Munidopsis sp. were 
more often seen below 3000 m. Large macrourids as a 
group tended to have distributions not significantly 
different from random. The halosaur H. macrochir had a 
clumped distribution in the area between 3000 and 2500 m 
covered by MIR 1, but was apparently distributed at 
random in the area covered by MIR 2. Small macrourids 
usually had clumped distributions. Some large holothur- 
ians tended to distribute apparently at random (Bentho- 
dytes sp. in the deep dives, Benthothuria sp. in the shallower 
dive between 3000 and 2500 m), while others showed 
clumped distributions (i.e., large Peniagone sp.). 
The four MIR dives into the Charlie-Gibbs Fracture 
Zone are the first for that area and include the dive in MIR 
1 reported by Vinogradov (2005). We were surprised by the 
prevalence of soft substrate on this mid-ocean ridge, as well 
as by the patchy high abundance of some fauna, such as the 
'gardens' of sponges. We found evidence of spatial patterns 
ranging from depth-related structure to relatively small- 
scale patterns likely associated with habitat requirements 
and habitat quality. We also found some evidence of 
patterns on intermediate scales, as in the distribution 
of holes and mounds across large expanses of substrate. 
Some of the larger and more mobile forms seemed to 
be distributed in random fashion. These investigations of 
Please cite this article as: Felley, J.D., et al., Small-scale distribution of deep-sea demersal nekton and other megafauna in the Charlie-Gibbs Fracture 
Zone of the Mid-Atlantic... Deep-Sea Research II (2007), doi:10.1016/j.dsr2.2007.09.021 
mraia?5?!5?c 
J.D. Felley et al. / Deep-Sea Research III ('llli; lll-l 
distributional patterns suggest that there are interesting 
associations between deep-sea organisms and their envir- 
onment, and also among the organisms themselves. Such 
observations as these can only begin to unravel the 
complexity that may exist in these deep-sea areas. 
Acknowledgments 
We thank Audrey Gebruk and Odd Aksel Bergstad of 
the MAR-ECO Project of the Census of Marine Life for 
their assistance in organizing this expedition. We also 
thank Anatoly Sagalevich and the personnel of the MIR 
operations group as well as the Captain and crew of 
the R.V. Akademik Mstislav Keldysh for their support 
of the field operations. Funding was provided by the Alfred 
E. Sloan Foundation and NOAA's Office of Ocean 
Exploration. 
Appendix A. Supplementary data 
Supplementary data associated with this article can be 
found in the online version at doi:10.1016/j.dsr2.2007.09.021. 
References 
Chave, E.H., Malahoff, A., 1998. In Deeper Waters: Photographic Studies 
of Hawaiian Deep-sea Habitats and Life-forms. University of Hawaii 
Press, Honolulu, HI. 
Cohen, D.M., 1977. Swimming performance of the gadoid fish Antimora 
rostrata at 2400 m. Deep-Sea Research 24, 275-277. 
Drazen, J.C, Goffredi, S.K., Schlining, B., Stakes, D.S., 2003. Aggrega- 
tions of deep-sea brooding fish and cephalopods on the Gorda 
escarpment: a reproductive hotspot. Biological Bulletin 205, 1-7. 
Felley, J.D., Vecchione, M., 1994. Assessing habitat use by nekton on the 
continental slope using archived videotapes from submersibles. Fishery 
Bulletin 93, 262-273. 
Fleming, H.S., Cherkis, N.Z., Heirtzler, J.R., 1970. The Gibbs Fracture 
Zone: a double fracture zone at 52?30'N in the Atlantic Ocean. Marine 
Geophysical Research 1, 37-45. 
Gage, J.D., Tyler, P.A., 1991. Deep-sea Biology: A Natural History of 
Organisms of the Deep-sea Floor. Cambridge University Press, 
Cambridge, UK. 
Gass, S.E., Wilhson, J.H.M., 2005. An assessment of the distribution 
of deep-sea corals in Atlantic Canada by using both scientific and 
local forms of knowledge. In: Freiwald, A., Roberts, J.M. (Eds.), 
Cold-water Corals and Ecosystems. Springer, Berlin, Heidelberg, 
pp. 223-245. 
Gebruk, A.V., Southward, E.C., Kennedy, H., Southward, A.J., 2000. 
Food  sources,  behaviour,  and  distribution  of hydrothermal  vent 
shrimps at the Mid-Atlantic Ridge. Journal of the Marine Biological 
Association of the United Kingdom 80, 485-499. 
Grassle, J.F., Sanders, H.L., Hessler, R.R., Rowe, G.T., McLellan, T., 
1975. Pattern and zonation: a study of the bathyal megafauna using 
the research submersible Alvin. Deep-Sea Research 22, 457-481. 
Haedrich,  R.L.,  Merrett,  N.R.,   1988.  Summary atlas  of deep-living 
demersal fishes in the North Atlantic Basin. Journal of Natural 
History 22, 1325-1362. 
Holland, N.D., Clague, D.A., Gordon, D.P., Gebruk, A., Pawson, D.L., 
Vecchione,   M.,   2005.   'Lophenteropneust'   hypothesis   refuted   by 
collection and photos of new deep-sea hemichordates. Nature 434, 
374-376. 
Kaufmann, R.S., Wakefield, W.W., Genin, A.,  1989. Distribution of 
epibenthic megafauna and lebensspuren on two central North Pacific 
seamounts. Deep-Sea Research 36, 1863-1896. 
Levin, L.A., Liethold, E.L., Gross, T.L., Huggett, C.L., DiBacco, C, 
1994. Contrasting effects of substrate mobility on infaunal assemblages 
inhabiting two high-energy settings on Fieberling Guyot. Journal of 
Marine Research 52, 489-522. 
Ludwig, J.A.,  Reynolds,  J.F.,   1988.  Statistical Ecology. Wiley-Inter- 
science. New York, NY. 
Merrett, N.R., Haedrich, R.L., Gordon, J.D.M., Stehmann, M., 1991. 
Deep demersal fish assemblage structure in the Porcupine Seabight 
(Eastern North Atlantic): results of single warp trawling at lower slope 
to abyssal soundings. Journal of the Marine Biological Association of 
the United Kingdom 71, 359-373. 
Reed, J.K., Weaver, D.C., Pomponi, S.A., 2006. Habitat and fauna of 
deep-water Lophelia pertusa coral reefs off the southeastern US: Blake 
Plateau, Straits of Florida, and Gulf of Mexico. Bulletin of Marine 
Science 78, 343-375. 
Rice, W.R.,  1989. Analyzing tables of statistical tests. Evolution 43, 
223-225. 
Ripley, B.D., 1981. Spatial Statistics. Wiley-Interscience, New York, NY. 
Roberts, J.M., Wheeler, A.J., Freiwald, A., 2006. Reefs of the deep: the 
biology and geology of cold-water coral ecosystems.  Science 312, 
543-547. 
Smith, C.R., Hamilton, S.C., 1983. Epibentbic megafauna of a bathyal 
basin off southern California: patterns of abundance, biomass, and 
dispersion. Deep-Sea Research 30, 907-928. 
Sokal, R.R., Rohlf, F.J., 1981. Biometry. Freeman, New York, NY. 
Strong, J.A., Service, M., Mitchell, A.J., 2006. Application of the visual 
fast count for the quantification of temperate epibenthic communities 
from video footage. Journal of the Marine Biological Association of 
the United Kingdom 86, 939-945. 
Sulak, K.J., Wenner, CA., Sedberry, G.R., van Guelpen, L., 1985. The 
life-history of deep-sea lizard fishes genus Bathysaurus (Synodontidae). 
Canadian Journal of Zoology 63, 623-642. 
SYSTAT 11, 2006. Systat Software Incorporated, Richmond, CA. 
Uiblein, F., Lorance, P., Latrouite, D., 2003. Behaviour and habitat 
utilization  of seven demersal  fish  species  on  the  Bay  of Biscay 
continental slope, NE Atlantic. Marine Ecology?Progress Series 257, 
223-232. 
Vinogradov, G.M., 2005. Vertical distribution of macroplankton at the 
Charlie-Gibbs Fracture Zone (North Atlantic), as observed from the 
manned submersible "Mir-1". Marine Biology 146, 325-331. 
Please cite this article as: Felley, J.D., et al.. Small-scale distribution of deep-sea demersal nekton and other megafauna in the Charlie-Gibbs Fracture 
Zone of the Mid-Atlantic... Deep-Sea Research II (2007), doi:10.1016/j.dsr2.2007.09.021