The University of Chicago
Deep-Sea Species Richness: Regional and Local Diversity Estimates from Quantitative Bottom
Samples
Author(s): J. Frederick Grassle and Nancy J. Maciolek
Source: The American Naturalist, Vol. 139, No. 2 (Feb., 1992), pp. 313-341
Published by: The University of Chicago Press for The American Society of Naturalists
Stable URL: http://www.jstor.org/stable/2462414 .
Accessed: 12/07/2013 13:50
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].
.
The University of Chicago Press, The American Society of Naturalists, The University of Chicago are
collaborating with JSTOR to digitize, preserve and extend access to The American Naturalist.
http://www.jstor.org
This content downloaded from 143.107.244.76 on Fri, 12 Jul 2013 13:50:21 PM
All use subject to JSTOR Terms and Conditions
Vol. 139, No. 2
The AmericanNaturalist
Februar-y1992
DEEP-SEA
SPECIES RICHNESS:
REGIONAL AND LOCAL DIVERSITY
ESTIMATES
FROM QUANTITATIVE
BOTTOM SAMPLES
J. FREDERICK GRASSLE* AND NANCY J. MACIOLEKt
*Woods Hole Oceanographic Institution,Woods Hole, Massachusetts02543 and Instituteof
Marine and Coastal Sciences, RutgersUniversity,New Brunswick,New Jersey08903;
tBattelle Ocean Sciences, Duxbury,Massachusetts02332
SuibmittedOctober 24, 1990; Revised March 11, 1991; Accepted March 22, 1991
Abstract.-The deep-sea communitiesof the continentalslope and rise offthe easterncoast of
the United States have a remarkablyhighdiversity-measured regionallyor locally eitheras
species richness or as the evenness of relativeabundance among species. In a 1,500-2,500-m
depth range offNew Jerseyand Delaware, 233 30 x 30-cm samples contained798 species in
171 familiesand 14 phyla. Additionof stationsfromsites to the northand southapproximately
doubled the numberof samples and doubled thenumberof species to 1,597.Species-area curves
do not level offwithinstationsor when stationsare added together.Moreover,the proportion
of species representedby single individualsis highat all scales of sampling,which indicates
that much more samplingis needed to adequately representthe species richnesseitherlocally
or regionally.Diversitychanges verylittlethroughtimeat a singlesite or withdistancealong a
180-kmtransectat a depth of 2,100 m. Diversityis maintainedby a combinationof biogenic
microhabitatheterogeneityin a systemwithfew barriersto dispersal, disturbancecreated by
feedingactivities of larger animals, and food resources divided into patches of a few square
metersto square centimetersinitiatedby specifictemporallyseparatedevents.
Our results,fromthe firstextensivequantitativesamplingof deep-sea communities,indicatea muchgreaterdiversityof species in thedeep sea thanpreviously
livingon or in the sediments
thought.Thousands of species of smallinvertebrates
of the deep-sea floorcoexist in a shifting
mosaic of microhabitats.Several space
and time scales need to be considered in studyingdeep-sea communities;however, small-scaleprocesses appear to be particularlyimportantin thisecosystem.
Deep-sea microhabitatstend to persistlonger,be smallerin spatial extent,and
be separated by longerdistances between patches in comparisonwiththe most
closely studied shallow-watermarinecommunities.
Communityecologists have long been interestedin explainingpatternsof species diversityof marineand terrestrialecosystems,especially the large number
of species thatcoexist in tropicalregions(Wallace 1878;Hutchinson1959;Fischer
1960; MacArthur1965; Pianka 1966). Hessler and Sanders (1967) discovered an
unexpectedlyhighspecies richnessin thedeep sea, whichled to extensivespeculation on the reasons for the observed patternsof species richness in marine
habitats(Sanders 1968; Dayton and Hessler 1972; Grassle 1972, 1989; MacArthur
1972; Grassle and Sanders 1973; Rex 1973, 1976;Jumars1976;Menge and Sutherland 1976; Van Valen 1976; Osman and Whitlatch1978; Abele and Walters 1979;
Am. Nat. 1992. Vol. 139, pp. 313-341.
All rightsreserved.
C) 1992 by The Universityof Chicago. 0003-0147/92/3902-0005$02.00.
This content downloaded from 143.107.244.76 on Fri, 12 Jul 2013 13:50:21 PM
All use subject to JSTOR Terms and Conditions
314
THE AMERICAN NATURALIST
Thistle 1979; Smith 1986; Grassle and Morse-Porteous1987). The potentialfor
tremendousloss of species in thetropicshas led to increasedinterestand concern
about species diversityand associated ecosystemfunctions,in both marineand
terrestrialenvironments(E. 0. Wilson 1985, 1988; May 1988; National Science
Board 1989; di Castri and Younes 1990). Furtherdevelopmentof theoryrelevant
to the processes that controlspecies diversityrequires bettermeasurementsof
regional species richnessand investigationof the interactionof thisspecies pool
witha more local scale of directcontactamongspecies (e.g., predation,competition, symbiosis;Ricklefs1987; May 1988; Petraitiset al. 1989; Menge and Olson
1990).
Until the late 1960s, deep-sea species richnesswas thoughtto be lower than
species richness in shallower environments(Ekman 1953). The earliestrecognition of relativelyhighspecies richnessin the deep sea was based on fivequalitative epibenthicsled samples fromthe northwestAtlantic(Hessler and Sanders
1967). The largestnumberof species in one of these samples, 365, was obtained
froman epibenthicsled thatwas pulled over the bottomat 1,400-mdepthforan
indeterminatedistance of more than 1 km. Numerous taxonomic studies based
on these and othersimilartrawlsamples supportthe initialreportsof highdiversityfor portionsof the fauna such as polychaetes(Hartman 1965; Hartmanand
Fauchald 1971; Maciolek 1981, 1985, 1987), isopods (Hessler 1967, 1968, 1970a,
1970b; Wilson and Hessler 1974, 1980, 1981; Thistle and Hessler 1976, 1977;
G. D. Wilson 1976, 1980a, 1980b, 1981; Siebenallerand Hessler 1977, 1981; Thistle 1980; Kensley 1982), and bivalves (Allen and Sanders 1966, 1969, 1973; Sanders and Allen 1973, 1977; Allen and Turner 1974; Allen and Morgan 1981). Prior
to the present study,quantitativedata on deep-sea species diversityhave been
limitedto analyses of only a few (<10) box-core samples (0.25 m2) fromany one
geographicarea. The paucity of quantitativedata on deep-sea species diversity
led Nybakken (1982, p. 147) to argue that "the idea of a highlydiverse deep-sea
fauna mustremainspeculative."
Our results are based on analysis of macrofaunafromquantitativesamples
from10 stationsalong 176 km of the 2,100-mdepth contouroffNew Jerseyand
Delaware and fouradditionalstationsat 1,500-mand 2,500-mdepth(fig.1). These
samples were takento describe the deep-sea communitiesas partof a monitoring
programto evaluate the impactsof oil and gas explorationactivities.Exploratory
drillingtook place near one of the 2,100-mstations(Maciolek et al. 1987a), and
most of the stationswere at this depth on the lower continentalslope. There is
no ecological significanceto the choice of bottomdepths. Additional samples
fromthe slope and rise offNorthand South Carolina (Blake et al. 1987) and off
New England (Maciolek et al. 1987b) were used to show the numberof species
from a larger area of the western Atlantic. This large numberof quantitative
samples enabled us to address the issue of regionalversus local diversityand to
answer some basic questions concerningspecies diversityin the deep sea: Does
diversitydifferbetween stations in an apparentlyhomogeneous region? Does
diversityvary seasonally? Does the numberof new species decrease as the number of samples increases? What is the species richnessat any given site? Can the
This content downloaded from 143.107.244.76 on Fri, 12 Jul 2013 13:50:21 PM
All use subject to JSTOR Terms and Conditions
DEEP-SEA
SPECIES
315
RICHNESS
11 NAUTICALMILES
41eF
0
..
10 20 30 40 50
DEPTH IN METERS
400
NEW JERSEY
2
39.t
DELAWARE
38"
70
i;
k
*
2 12
3f
10
ATLANTICOCEAN
37.L
75"
74
73*
72
71
FIG. 1.-Location of deep-sea box-coringstationson thecontinentalslope offNew Jersey
and Delaware.
numberof westernAtlanticcontinentalrise species be estimated?The answers
to these questions are needed to understandthe processes controllingspecies
diversityin the deep sea.
Species diversitydepends on the numberof species, the relativeabundance of
those species, and the numberof individualssampled(Pielou 1966). Wherepossible, we have used graphicmethodsto emphasize the relationshipof species richness to the samplingpattern,and we illustrateour resultswithplotsofactual data,
of
avoidingmeasures of diversityderivedfromassumptionsabout thedistribution
individualsamong species. Relationshipsbetween numbersof species and samplingarea, between species and individualsin samples of differing
size, relative
abundances in different
areas, and expected numbersof species shared between
areas are used to provide an overview. Estimates of species presentin a single
area and how this varies spatiallyallow us to make a conservativeestimateof
how many macrofaunalspecies may be presentin the deep sea.
This content downloaded from 143.107.244.76 on Fri, 12 Jul 2013 13:50:21 PM
All use subject to JSTOR Terms and Conditions
316
THE AMERICAN NATURALIST
TABLE I
STATION LOCATIONS AND DEPTH IN METERS
Station Number*
M6
M5
M4
M11
M3
Ml
M2
M12
M7
M8
M9
M14
M13
M10
Reference
Depth
(m)
Depth
Range
(m)
No.
of Samples
Latitude
N
Longitude
W
2,090
2,065
2,100
1,515
2,055
2,195
2,020
2,505
2,100
2,150
2,105
1,500
1,613
2,095
2,045-2,091
2,055-2,090
2,091-2,124
1,502-1,540
2,045-2,064
2,165-2,209
2,005-2,024
2,495-2,509
2,085-2,110
2, 148-2,159
2;100-2,114
1,409-1,515
1,605-1,619
2,093-2,114
17
18
18
17
18
18
18
18
17
8
18
12
18
18
38?05.54'
38?50.49'
38?44.47'
38?40.17'
38?36.84'
38?35.98'
38?35.78'
38?29.30'
38?27.36'
38?27.31'
38?17.28'
3753.91'
37?53.33'
3751.80'
72?02.97'
72?33.01'
72?33.01'
72?56.37'
72?51.35'
72?52.97'
72?53.65'
72?42.15'
73?03.44'
73?04.87'
73?14.51'
73?44.62'
73?45.09'
73?19.84'
*Orderedfromnorthto south.
MATERIALS
AND METHODS
StludySites and Sampling Methods
The 176-kmtransectextendedfrom37.9? to 39.1?N and 72.10to 73.8?W. Three
replicatesamples were taken at each of 14 stations(table 1) threetimesper year
for 2 yr (March/April/May,August, and November/Decemberin 1984; May,
August,and November in 1985). Carefulattentionto navigationalpositionusing
Loran C ensuredthatsamples were generallywithin200 m of the stationlocation
regardlessof samplingdate.
The samples were taken usinga modifiedOcean InstrumentsMark III 0.25-m2
box corer (Hessler and Jumars 1974). The box corer was partitionedinto 25
subcores, each witha surfacearea of 0.01 M2 . A block of nine contiguouscentral subcores was designated for infaunalanalysis, with the remainingundisturbedsubcores reservedforsedimentchemistryand grain-sizeanalyses (data in
Maciolek et al. 1987a). Subcores for infaunalanalysis were sieved on board
using 0.3-mm-meshscreens and fixedin 10% bufferedFormalin.
Laboratory Methods
In the laboratory,each sample was resieved on a 0.3-mm-meshscreen and
fromFormalinto 80% ethanol. Samples were stainedwithRose Bentransferred
gal at least 4 h priorto sortingunder a dissectingmicroscope. The single most
and time-consuming
partof thisworkwas the taxonomicidenticritical,difficult,
ficationof each specimen. We were veryfortunateto have assembled a team of
experts, most of whom had many years of experience withdeep-sea taxa (see
Acknowledgments).Individualsof uncertainspecies identity(because theywere
very youngjuveniles or were missingmorphologicalfeaturesneeded foridentification) were not includedin the statisticalanalyses. Samples were gentlyhandled
This content downloaded from 143.107.244.76 on Fri, 12 Jul 2013 13:50:21 PM
All use subject to JSTOR Terms and Conditions
DEEP-SEA SPECIES RICHNESS
317
juveniles) of the
duringprocessing so that only 2.8% (includingunidentifiable
total number of individuals collected were not identifiedto species. Animals
attached to hard surfaces such as rocks and shells and parasitic or planktonic
species were also excluded fromthe analyses. Meiofaunal taxa such as nemaand mites were not intodes, harpacticoid copepods, ostracods, foraminifera,
cluded because theywere inadequatelysampled by the0.3-mm-meshscreensand
were thereforenot considered partof the macrofauna.
Results were based on 233 box-coresamples from14 stations;however,particular emphasis was placed on the 160 samplesfromthenine2,100-mdepthstations
(for comparison of combined samples withinstations,station8 was leftout because it was sampled only threeratherthan six times). Only the total numberof
species and individualsfroman additional323 box cores fromoffthe Carolinas
and New England were used. Data entrywas replicatedand cross-checked,and
the tabulateddata were returnedto the taxonomicspecialistswho eliminatedany
redundantentriesor similarerrors.In the analysis involvingrandomcombinations of all samples from2,100-mdepth, samples fromthe stationsat eitherend
of the along-contourtransect(6 and 10) were also droppedbecause of the limited
capacity of the VAX 11/780used in the calculations.
Hurlbert's (1971) modificationof the rarefactionmethod (Sanders 1968) was
used to generate species-individualscurves. The points on each curve are the
numberof species calculated to occur in computer-generated
subsamples of m
individualsfromsingle or combined samples (Smithand Grassle 1977). The rarefactioncurves were estimatedfromthe total fauna and also on separate faunal
groups: polychaetes, mollusks, and peracarid crustaceans. Shannon-Wienerdiversityindexes (H') were calculated forindividualsamples (Pielou 1966). Faunal
similaritieswere measured usingthe normalizedexpected species shared in samples of mn
individuals(NESS; Grassle and Smith 1976; Smithet al. 1979). In this
article,NESS is expressed as the percentageof species shared.
Plots of species versus area (numberof samples) were computed using the
methods of Gaufinet al. (1956) to calculate the average numberof new species
contributedby randomcombinationsof 1, 2, 3, . . ., n replicatesamples. Speciesaccumulation curves were also calculated by adding samples in sequences orwas estimated
dered accordingto timeor station.Fit to thelognormaldistribution
using the Gauss-fitprogramdescribed by Gauch and Chase (1974).
RESULTS
Diversityand Abundance of Higher Taxa
A total of 798 species representing171 familiesand 14 phyla were identified
from90,677 individualsin 233 box cores thathad a combinedsurfacearea of 21
m2 (table 2). An additional 64 macrofaunalspecies were presentin the samples
but were planktonicor lived on hard surfacesand thereforewere omittedfrom
the analyses (see Maciolek et al. 1987 for a complete list of species). Annelids
accounted for48%, peracarid arthropods(amphipods,isopods, tanaids, and cumaceans) 23%, and mollusks 13% of the macrofaunalindividuals.
This content downloaded from 143.107.244.76 on Fri, 12 Jul 2013 13:50:21 PM
All use subject to JSTOR Terms and Conditions
318
THE AMERICAN NATURALIST
TABLE 2
TAXONOMICCOMPOSITIONOF SOFT-SEDIMENT BENTHOS FROM
U.S. MID-ATLANTIC CONTINENTALSLOPE SAMPLES
Species and Family
Cnidaria
Hydrozoa
Anthozoa
Scyphozoa
Nemertea
Priapulida
Annelida
Polychaeta
Oligochaeta
Echiurida
Sipuncula
Pogonophora
Mollusca
Bivalvia
Gastropoda
Scaphopoda
Aplacophora
Arthropoda:
Cumacea
Tanaidacea
Isopoda
Amphipoda
Pycnogonida
Bryozoa
Brachiopoda
Echinodermata
Echinoidea
Ophiuroidea
Asteroidea
Holothuroidea
Hemichordata
Chordata
Total
No. of Species
No. of Families
19
6
12
1
22
2
385
367
18
4
15
13
106
45
28
9
24
185
25
45
59
55
1
1
2
39
9
16
3
11
4
1
10
3
6
1
1
1
49
47
2
2
3
5
43
18
18
4
3
40
4
8
11
16
1
1
1
13
2
6
3
2
1
1
798
171
Of all species, 58% (460) are new to science. Among the peracaridcrustacea
and polychaetes, 69% (127 species) and 64% (236 species), respectively,are undescribed. Withinseveral polychaete families,such as the Dorvilleidae, Cirratulidae, Spionidae, Flabelligeridae, and Terebellidae, 75%-93% are new. The
mollusksare somewhatbetterknown,however,and 37% (42 species) are undescribed.
CommunityHomogeneity
The relativeabundances of species at the nine 2,100-mstationssampled over
the entire 2-yrperiod were similarfromstationto station(fig.2). The relative
proportionsof these 10 species abundances withinthese stationsdid not change
radically when samples were combined according to samplingdate, station,or
entireregion (table 3). When the samples were combined,only 10 species had
an average abundance greaterthan 2%. The single most common species (the
This content downloaded from 143.107.244.76 on Fri, 12 Jul 2013 13:50:21 PM
All use subject to JSTOR Terms and Conditions
DEEP-SEA
6
5 4
3
1
2
7
9
i0
sta/Con
SPECIES
319
RICHNESS
no.
5.0
1.0
0A
.05
0
0oo
200
300
species sequences
FIG. 2.-Percentage abundance of species ranked fromthe most common to the least
common (species sequences) at each of nine stationsordered by positionon a NE-to-SW
transectalong the 2,100-mdepth contour.The intervalsof 10 on the baseline are numbers
of species (offsetby 10 species foreach station). Each stationhas a similardistributionof
individualsamong species.
TABLE 3
PERCENTAGE CONTRIBUTION
Species Ordered by Rank
Aurospio dibranchiata(P)
Pholoe anoculata (P)
Spathoderma clenchi (A)
Tharyxsp. 1 (P)
Prionospio sp. 2 (P)
Tubificoidesaculeatus (0)
Prochaetodermayongei (A)
Aricidea tetrabranchia(P)
Glycera capitata (P)
Nemertea sp. 5 (N)
OF THE IO MOST ABUNDANT SPECIES AT 2,IOO-M DEPTH
All
Samples
Combined
7.1
4.6
3.9
3.8
3.1
3.0
2.8
2.2
2.1
2.1
Replicates
Combined,
Replicates and
Averaged
Times Combined,
across
Averaged across Stations and
Stations(%)
Times (%)
7.2
5.6
4.2
3.6
3.4
3.1
2.8
2.6
2.4
2.2
(9.5)
(17.4)
(19.5)
(15.4)
(17.6)
(13.1)
(12.3)
(10.2)
(8.1)
(5.4)
7.7
5.8
4.6
3.9
3.4
3.2
2.9
2.6
2.4
2.3
(16.6)
(19.2)
(18.8)
(17.8)
(15.4)
(14.9)
(14.1)
(13.5)
(11.2)
(11.4)
Averaged across
Stations, Times,
and Replicates (%)
8.3
6.2
4.9
4.2
3.8
3.4
3.2
2.9
2.7
2.5
(26.9)
(21.9)
(18.1)
(17.0)
(15.9)
(14.9)
(14.9)
(14.3)
(13.0)
(13.5)
NOTE.-A,
Aplacophora; N, Nemertea; 0, Oligochaeta; and P, Polychaeta. Coefficientsof variations are shown in parentheses.
polychaete Aurospio dibranchiata) comprised 7%-8% of the total individuals
regardlessof scale of sampling.The abundance of this and the otherabundant
species changed littleregardlessof stationor samplingdate (fig.3).
About 20% of the species were found at all 10 2,100-m stations, and 34%
occurred at only one station. For the entire soft-sediment
fauna, 28% of the
species occurred only once and 11% only twice. Peracarid crustacea have the
narrowestdistributionwithonly 15% (21 species) occurringat all 10 stationsand
43% (60 species) occurringat a single station. The breadth of distributionof
This content downloaded from 143.107.244.76 on Fri, 12 Jul 2013 13:50:21 PM
All use subject to JSTOR Terms and Conditions
320
THE AMERICAN NATURALIST
50
52X8
4030-
2 3 4 5 6
1234
56
1 2 3 4 5 6
1234
56
1
I
S/oaion
St/tion2
50-
12
3 4 5 6
1 2 3 4 5 6
1 2 3 4 5 6
12
3 4 5 6
12
3 4 5 6
1 2 3 4 5 6
S/ot/on
3
2100m
S/otion
4
S/oaion
5
40
EQ.
20
Stat/ion
6
1
Station
7
Stto2
11
Station
e0150m-12500m
-
Station8
Station
9
~~~~~2100m7
10
Station
13
14
S/at/ion
S/at/ion
1500w -
FIG. 3.-Temporal variationin abundance (no./900cm2? 1 SD) ofAurospio dibranchiata
over a 2-yrperiod at each of 10 stationsalong the 2,100-rndepthtransect(stations 1-10), at
depths of 1,500 m (stations 11, 13, 14) and 2,500 m (station12). Each bar indicatestheX and
SD of three replicatesamples fromeach samplingdate.
polychaetes was near the mean forthe entirefauna; 22% (65 species) occurred
at all 10 stations and 36% (106 species) at one station. Bivalves, in additionto
being betterknown (see above), had broaderdistributions
with37% (14 species)
at all 10 stations and only 21% (8 species) at one station. Althoughvariable in
proportion,endemismwas highboth across and withinall taxa.
Based on the six northernmost
stations(stations1-6), the normalizedexpected
species shared (NESS) between samples of 50 individuals(m = 50 individuals)
fromadjacent pairs of stationswas 83% ? 5% (95% confidencelimits[CL]), and
the mean NESS similarityfor samples of the same size (m = 50) withineach
stationwas 92% ? 2%. The mean NESS similarity
(m = 200 individuals)between
samples fromthe most distantstations(6 and 10) was 79% ?-5% (within-station
similaritieswere 93% ? 5% and 91% ? 4%, respectively).
The communityat 2,100-indepth does not have sharp boundaries: a 500-in
change in depth in eitherdirectionfromstation 1 produced a NESS similarity
between stationsof 68% ? 3% in the shallowerdirectionand 64% ? 4% in the
deeper direction(N. J. Maciolek and J. F. Grassle, unpublishedmanuscript).
Clearly, althoughthereare manyendemic species and manyare rare, the variation in the identityof relativelyabundantspecies over a large scale (kilometers)
is relativelysmall.
This content downloaded from 143.107.244.76 on Fri, 12 Jul 2013 13:50:21 PM
All use subject to JSTOR Terms and Conditions
SPECIES
DEEP-SEA
32 1
RICHNESS
1
/2
130-
120 -
/7
3
2
/8
/
110 _
|
/73 0 3?
/ 93
235
7 50
7196
q)~~~~~~~~
61
/
~
80
2
02
6
86 16
53
3 6
1
7
2
49
49451
1
9
511
93
|
90 _
?
3
2
3 ~~~~~~~~~~~~1
2
4
4
0~~~~~7
3 6~~~~3
4 9 4 51
4
/
X
5
8
8 53
/7
/
%
tfi
4
OD
9 59
/
/3
100~~~~~~
2
20
64
61
265
5
09
9
60~~~~~~~~~
13
55
4. Nube
FIG.~~~8
4
l36 l
l
100 200
of speie
an4niiul
henum er inicatesthesttion
1-10.
i apotin
cnnctngtheX'
lin
all 168 individual
o
300~
l9 4 l
l
l
4050607080
dua/
/nd9
9
nec
elcte9
1,3 stto8
arfatin
te
samples.~~~~~~~~~~~~~~~~
;...
;2
tto
crv
clclaedfro
2sml
rmsain
,sain1)
asumaio
h
o
Species~Diest4fEchRpiaef
~~~
2,10-
SttosTetdSprtl
umer
, n hic te
s hon
n fgue
a th 210-msttios
Thedierit
o
plttd.Thenube
90-m2samlear
pecesineac
ad
of ndviual
pce
hsmaueo
o 3,wt
oerne
rm5
pe40-m
species~~~~~~~~~~~~~~~~~~~~~
b
arfatio
lneinicte
ensty Te
t funl
arin
acorin
ricnes
otane
b smmngth
inlesapl
ws
alulte
foma
theX' i fgue
nmbes
f pecesan
ofth
ne xcpton,al
smpes.Wih
1682,00-
This content downloaded from 143.107.244.76 on Fri, 12 Jul 2013 13:50:21 PM
All use subject to JSTOR Terms and Conditions
Uo
c00
H
0 1P
.
dO
0
Io,1+
It
CO
=
+
_l
.=
>
:
el+
_l
:
t
Nf _l
_l
I
-|:
+l
t
C
=
CZ
I:ts;E
~.- ~ CZ~fi ~L I
0 7
Q0I)0
fX
M
.Q
M .
N
C+1 "C
m"C+1 +1
m
+1 t+1 Itf
+1 0of
+1 CDf
+1 +1 C\f
+1 +1
]f
fof
Of_
)0
t-
oO
6
z,
v)
C)
c
E- :
CZ~~~~~~~~~~~~~~~~~C
CO
CO~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~T
_V)
z'
8
-_f
f^
f
Of ft
"C
r.
P
lWf
t|--
=1-f0w to
ON
'P
fofI C)f cr-I 0<f 0<f
t
0
0
f((
f
+1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1
f
N0
C)CN
-N C0
0 ON0N
r
Q
rg
0i
H E
En
0
0..
fo
COf
fZ
f
--
fE?g
0
-
-
-f
>
f
f
f
_
_
>
f
COL-
=
U
~ ~ N ~ ON~ ~ OsN000
? ~ af
Q
~fz'f-af
~ Q~ ~w ~ ~ N~ ~ - ~ ~N ~fff,;
~~~ ~~ ~~cOO~~~~~c
~~
CO
fON
00'f
-
:tf-]f3]f>is]f
,C oo >
t
>
>
ON= 00g
c
c
-
ONE
En
-1
En
Qo
-
o..>
.
En En
En
.t
>
C
N"C
c
z
OOOE
OC
0C
f
CO_
. COf-C
\OCO
i s ff [ ,
0:]f z ff > f
f~~~~f ~
t
x>
-
-
c
?
Mr.
, <
,
f _f _
t
fz
fL
0
_ f
if -
This content downloaded from 143.107.244.76 on Fri, 12 Jul 2013 13:50:21 PM
All use subject to JSTOR Terms and Conditions
SPECIES
DEEP-SEA
RICHNESS
323
60
58 -
Sta. 1 +
Sta. 2 X
Sta. 3 o
Sta. 4
Sta. 5 0
Sta. 6 v
Sta. 7 A
Sta. 8
Sta. 9 *
Sta.40
A
A
A
56 -
x
A
A
A
A
A
54
00
40
A
52 -
X
O
0
0~~~~0
-
.)
/v O4
U
+
0
5.2
*
~~
+
46
42
v*
5.
48-
44 -
A
xQ0
+
V
5.4
5.6
5.8
6.0
- Wiener
index
ShaGnnon
dliversify
6.2
6.4
FIG. 5.-Relationship between the rarefactionestimateof species per 100 individualsand
the Shannon-Wienerdiversityindex foreach sample fromthe 2,100-mdepth transect.The
regressionline forall pointswas r2= 0.83.
individuals per 900 cm2 are to the rightof the line, which indicates that the
individualsof each species are not distributedrandomlyamong the actual samples. This observation is not surprisingin view of the obvious patchiness that
was observed in bottomphotographs(Grassle et al. 1975; Hecker in Maciolek
et al. 1987a, 1987b). Numbers of species per 900 cm2,Shannon-Wienerdiversity
indexes, and rarefactionestimates of species per 100 individualsare shown in
table 4. At this scale of sampling,the Shannon-Wienerand species per 100 individuals estimateswere correlated(fig.5).
Species Diversityat Each 2,100-mStation, When
Replicates and Sampling TimesAre Pooled
Species-accumulationcurves for stations 1-7, 9, and 10 were calculated from
the species means of every combination of each 1, 2, 3, . . . , n samples (fig. 6).
This content downloaded from 143.107.244.76 on Fri, 12 Jul 2013 13:50:21 PM
All use subject to JSTOR Terms and Conditions
THE AMERICAN
324
NATURALIST
1
350
200 |
B250
SH
150
100
0
6
3
9
12
15
18
NUMBER OF SAMPLES
0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
AREA (im2)
FIG. 6.-Species-area curves based on random addition of samples fromeach of the
2,100-m depth stations sampled six times over a 2-yrperiod. The area of 18 samples is
1.62 M2.
Using this method, we found that diversitywas greatestat the southernmost
station, 10. The next most diverse stations-2, 3, and 5-did not representany
particulargeographicalpattern.
The rarefactioncurves forthese stations(fig.7) indicateddifferencesin diversitysimilarto those representedby the species-accumulationcurves,withstation
10 being the most diverse and station4 the least diverse. The northernmost
station, 6, was similarin species compositionto the majorityof stations,whereas
the southernmoststation, 10, showed the greatestdissimilarity
to other2,100-m
stations. Forty-twospecies occurred only at station 10; in contrast,17 species
occurred only at station6. The relativelyhighspecies richnessat station10 may
reflecta contributionof species dispersingfromdenser populationsto the south.
Rarefaction curves for the three major taxa-Polychaeta, Peracarida, and
Mollusca-are shown in figure7 as shaded areas delimitedby the curves forthe
least and most diverse stationsand include the actual end pointsof each of the
nine curves foreach taxa. Station 10 had the mostdiversefauna in each of these
taxa as well.
Fluctuations in diversitythroughtime at each station are illustratedby the
rarefactionnumberof species per 500 individualson each samplingdate (fig.8).
This content downloaded from 143.107.244.76 on Fri, 12 Jul 2013 13:50:21 PM
All use subject to JSTOR Terms and Conditions
~ ~~ ~~~~~~
?40~
Individuals
FIG. 7.-Rarefactionestimatesof diversity
of totalfauna,Polychaeta,Peracarida,and
Molluscafromeach ofthe2,100-rn
stationssampledsixtimes.
0100_
Ill~
180 _
~ ~ ~
\
D
_
._~~~~~~~~~~~~120_
/98
My
Aug
Nov
/9
Moy
Aug
Av
FIG. 8.-Temporal variationof species per 500 individuals
over a 2-yrperiodat nine
2,M100-stations,identified
bydistancealongthedepthcontour.
This content downloaded from 143.107.244.76 on Fri, 12 Jul 2013 13:50:21 PM
All use subject to JSTOR Terms and Conditions
THE AMERICAN NATURALIST
326
.90
replicates
separate
.88
.86 -
replicates
combined
.84 -
.82 q)
.80
times
combined
.78.76 .74 _
.72 -
.70
5.8
stations
combined
1, 4-6,8,9
5.9
6.0
6.1
6.2
6.3
6.4
6.5
6.6
Shannon- Wienerdiversityindex
FIG. 9.-Mean
and standard deviation of evenness (H'/Hmax) and the Shannon-Wiener
diversityindex (H') of each 900-cm2,2,100-mdepth sample separately,combined within
samplingperiods,combinedover timewithinstations,and combinedamongstations(stations
1-5, 7, 9 and stations 1, 4-6, 8, 9).
Diversityfluctuatedvery littlefromstationto stationor fromtimeto time. The
abundance of the most common species, the polychaeteAurospio dibranchiata,
littlewithtimeand site along the depth contour(fig.
also fluctuatedsurprisingly
3). Similarhistogramscould be shownformostof the abundantspecies along the
transect(Maciolek et al. 1987a).
Rarefactionestimatesof species diversitydid notchange drasticallywhen samples were combined (table 4). This was not the case with the Shannon-Wiener
index (H') or evenness (H'/Hmax) (fig. 9). For estimates of regional diversity,
neitherH' nor evenness are appropriatemeasures.
When the distributionsof individualsamong species at each stationwere fitted
to a lognormaldistribution(Gauch and Chase 1974; also see pros and cons of
this approach in Nelson 1987), only the extremeright-handtail of a lognormal
distributionwas observed (curves not shown). In six of the 10 fittedcurves, the
This content downloaded from 143.107.244.76 on Fri, 12 Jul 2013 13:50:21 PM
All use subject to JSTOR Terms and Conditions
DEEP-SEA SPECIES RICHNESS
327
mode of the curve was more than 1 SD unitto the leftof the portionof the curve
representedby our data. When the 2,100-mstationswere combined to include
more of the total pool of species in the region,the fitto the lognormaldecreased
(explaining 91% of the variance instead of 95%). The calculated mode of the
curve, based on all 2,100-msamples, was 1.6 SD unitsto the leftof the actual
data. Assumingthisdistribution,
we have collected only5.5% of the species from
the entiredistribution,
and thereare 11,800species in the community.Since only
one extremetail of the curve is present,the lognormaldistribution
may or may
not be a good representationof the community.Because more completelognormal plots have been fittedto other communities(Preston 1948; Patrick 1973;
Sugihara 1980) with relativelyfew rare species, it seems unlikelyto be a good
approximationfor deep-sea data. Hughes's communitydynamicsmodel of species abundance (Hughes 1984, 1986) providesa more completeapproximationof
the data than the lognormalor log series distributions.The Hughes model, however, does not provide a predictionof the total species pool.
Species Richness-Stations along the2,100-mContolur
Combined as a Single Assemblage
Faunal similaritieswere very high between stationsalong the 2,100-mdepth
contoursampled. Stations6 and 10, the northernmost
and southernmost
stations,
respectively,and station8, which was sampled for only 1 yr, were eliminated,
so thatthe lengthof the 2,100-mcontourspannedby theremainingseven stations
was 87 km. Faunal similaritieswere high between all pairs of these stations
(NESS > 85%), and the combineddata providea representationof regionalspecies diversity(fig. 10). Both a rarefactioncurve (+'s, based on combiningthe
same samples intoa singlecollectionand calculatingthe numberof species represented in successively smallerrandomdraws of individualsfromthe combined
collection) and a species-accumulationcurve (based on the mean numberof species in all combinationsof 1, 2, 3, . . . , n samples) are illustrated(fig.10). Because
are notrandom,themean numberof species in combinations
species distributions
of samples accumulated species at a fasterrate than chance combinationsof
individualsdrawn froma singlecombined sample.
The individual points in figure10 are the data fromHessler and Sanders's
(1967) five epibenthicsled trawl samples thatfirstshowed the highdiversityin
the deep sea. The epibenthicsled inefficiently
sampled a 0.81-m swath approximately 1-2 km long or 103 Mi2. Althoughthe sled samples were fromthe same
general region as our study area, the diversitywas much lower than our data
show, especially when the much smallersurfacearea thatwe sampled is considered. Three reasons account forthe difference.First,the box-core stationswere
spaced over a geographicalarea largerthanthatcovered by a singletrawlsample,
so trawlsmiss manyof the burrowingspecies (Gage 1975). Second, a finersieve
was used (0.3 mm instead of 0.42 mm). Finally, and most important,major advances have been made in deep-sea systematicsin the intervening2 decades,
which has resultedin the recognitionof manymore species thanwere identified
by Hessler and Sanders.
We used the data fromall nine stationssampled six times over a 2-yrperiod
This content downloaded from 143.107.244.76 on Fri, 12 Jul 2013 13:50:21 PM
All use subject to JSTOR Terms and Conditions
THE AMERICAN
328
NATURALIST
2100-Meter Stations
600
+
500
+
/
400 F
+
+
++
400
300
+
+
200
l
0
,
I
0
0
1400 m
29000m
+
100
+
* 2500m
* 2900 m
4700m
10,000
30,000
20,000
Individuals
.
.
.
.
.
...
.
50
Numberof Samples
2
4
Area
6
I
50,000
40,000
8
100
10
(m2)
FIG. 10.-Upper curve is a computer-generated
species-area plot of the mean numberof
species foreach combinationof samples fromthe 2,100-mstations1-5, 7, and 9, regardless
of samplingdate. Plulssymbols(+) markthe species-individualsrelationshipcalculated by
rarefactionfroma singlesummationof the 125separatesamples. Trawl samplesfromHessler
and Sanders (1967) are indicatedby asterisks.
to develop the species-accumulationcurves by adding the samples togetherin
order of time and order of distance along the transect(fig. 11). The rarefaction
curve is also presentedfor reference.When the samples fromseparate stations
were combined in an orderproceedingfromnorthto southwithineach sampling
date and then according to time of samplingfromthe firstto the last sampling
date, the species-accumulationcurve (closed circles) was very close to the rarefactionline. When samples withineach stationwere orderedfirstby sampling
date and then according to the order from northto south along the 2,100-m
contour, the curve (open circles) fell below the rarefactionline and had a pronounced stepwisecharacter.The horizontalpartof the step consistedof individuals of species already representedin the cumulativecollection.The verticalpart
This content downloaded from 143.107.244.76 on Fri, 12 Jul 2013 13:50:21 PM
All use subject to JSTOR Terms and Conditions
DEEP-SEA
SPECIES
RICHNESS
329
600 -
l113Akm
0-
500-
+
400-
9
87k-m
Z 68km
(f)
+
300 ()
0
52km
200 -+
100 -
0
I
0
10,000
20,000
I
I
30,000
I
I
40,000
I
I
I
50,000
I
I
60,000
IND/VIDUALS
FIG. 11.-Individual samples combinedin an orderedsequence by two different
methods.
The open circles are samples combined over time withinstationsand then combined according to station at the designated distances along a NE-to-SW transect.The staircase
patternis formedby horizontallines representingindividualsof species alreadyrepresented
in the samples accumulated at thatpoint.The closed diaimond-slaiped
poinitsare rarefaction
curves using individualsof the species added at thatstation.The closed circles are samples
accumulated accordingto stationwithineach timeinterval.This curve closely approximates
the rarefactioncurve representedby plus symbols(+). The rarefactionpointsare calculated
fromthe combined collectionof all samples.
of each step added veryfew individualsbecause species not previouslycollected
were rare. This is as expected if the 2,100-mcontouris all one assemblage. The
steps became less steep with distance except for a markedadditionof species
fromstation 10 at the end of the transect.Station 10 may be at the startof a
transitionto a different2,100-m community(see communityhomogeneityresults). Because widely separated small patches are expected in the deep sea, we
would predicta similarbut somewhatlower curve ifall the samples were froma
single site. The data fromthe hundredsof samples needed to fullycharacterize
a species-area curve at a singlesite are unlikelyto be obtainedin the near future
because of the expense of processinga singledeep-sea sample.
Rarefactioncurves of the threemajor taxa, which used samples fromstations
1-10 combined forall samplingperiods, showed thatpolychaeteswere the most
diverse group, followed by the peracaridcrustaceansand thenthe mollusks(fig.
12). Smaller groups such as tanaids,isopods, and bivalves (or the even less abundant gastropods and amphipods) gave an even less complete representationof
the diversity(fig. 13).
This content downloaded from 143.107.244.76 on Fri, 12 Jul 2013 13:50:21 PM
All use subject to JSTOR Terms and Conditions
THE AMERICAN
330
POLYCHAETES,
NATURALIST
CRUSTACEA, MOLLUSKS
300
7
250
/
_
_
//
~//
~~//
200
150~~~~~~~~0
_
_I*
i~
/
/
-/
150
/X
100-
/
~M
_ t
50
0 0
I'
I
10,000
I
II
20,000
I
I
30,000
FIG. 12.-Rarefaction curves for polychaetes (P), peracarid crustaceans (C), and mollusks (M).
DiversityIncludingStations offNew England and the Carolinas
A combinationof samples from31 additionalstationssouth of New England
(Maciolek et al. 1987b) and offNorth and South Carolina (Blake et al. 1987) at
depthsfrom255 to 3,494 m was used to extendtheplot of species and individuals
northand south beyond the regionoffNew Jerseyand Delaware (fig.symbols).
The total number of species added was 799, which broughtthe total to 1,597
for 556 900-cm2samples (272,009 individuals).A plot of the numberof species
representedby singleindividualsagainstthe totalnumberof individualscollected
showed thatthe rate of additionof rare species was undiminishedwiththe addi-
This content downloaded from 143.107.244.76 on Fri, 12 Jul 2013 13:50:21 PM
All use subject to JSTOR Terms and Conditions
DEEP-SEA
SPECIES
RICHNESS
331
ISOPODS,
TANAIDS,
VES
RIVAL
50 45 -
4035 -
15-
10 -
5-
0
2000
4000
6000
/ND/V/DUALS
FIG. 13.-Rarefaction curves forisopods (I), tanaids (T), and bivalves (B)
tion of samples fromotherslope environments(fig.14, othersymbols).This is a
criticalpointto consider in estimatingwhere the species-abundancecurve might
eventuallylevel off.
DISCUSSION
Faunal Homogeneity
The relative-abundancecurves (fig.2) show thatthe communitystructurewas
similarat each of the 2, 100-mstations.The species listedin table 3 were consistentlythe most abundant at each of the 2,100-mstations,and the single most
abundantspecies, thepolychaeteAurospiodibranchiata,had a remarkablyhomogeneous abundance over the entirearea sampled (fig.3). Despite manyrare species (90% of the species were representedby less than 1% of the individuals),
about one-fifth
of the species foundat any of the 2,100-mstationsoccurredat all
10 stations.Mean similarity
of samples withinstationswas slightlygreater(NESS
This content downloaded from 143.107.244.76 on Fri, 12 Jul 2013 13:50:21 PM
All use subject to JSTOR Terms and Conditions
THE AMERICAN NATURALIST
332
0
replicatesseparate
timesseparate
m replicates
combined,
A
stationsseparate
vtimes
combined,
176 kmlong,2100 mcontour
0 stationscombined,
0 stations
combined,
acrossregionanddepth
v
103
25
10.
A
0
U)~~~~~
v
102
A
101.5 2
10
13
10
4
10
5
10
Individuals
FIG. 14.-Relationship between species and individualsin successively largercombinations of samples. The lower set of pointsshows the relationshipbetween numberof species
representedby a single individualand the total numberof individuals.The closed symbols
are withinsingle 2,100-m stations. The open circles representdata from45 stationstaken
from the continentalslope off the east coast of the United States fromdepths of 2553,494 m.
at 50 individuals= 92% ? 2%) thanbetweenadjacent2,100-mstations(NESS
at 50 individuals = 83% ? 5%) or betweenthe two mostdistant2,100-mstations
(NESS at 50 individuals = 79% ? 5%). A change of 500 m in depth resultedin
a drop in NESS similarityto 68% ? 3% in the shallowerdirectionand to 64% ?
4% in the deeper direction.
Our results support evidence from other studies that species composition
changes more rapidly across rather than along depth contours (Sanders and
Hessler 1969; Grassle et al. 1979; Carney et al. 1983).
However, broad distributionsof manyof the abundantspecies both along and
across depthcontours(N. J. Maciolek and J. F. Grassle, unpublishedmanuscript)
are evidence thatboundariesbetween deep-sea communitiesare farless distinct
than between communitiesin shallow water. This is shown by the small spatial
This content downloaded from 143.107.244.76 on Fri, 12 Jul 2013 13:50:21 PM
All use subject to JSTOR Terms and Conditions
DEEP-SEA SPECIES RICHNESS
333
and temporalvariationof faunaldiversityin these samplescomparedwithsimilar
surveysfor shallow-watercommunities(see figs.4, 5, 8).
The threemost abundanttaxa differedsimilarlyin breadthof distribution
both
across and along depthcontours.Species of peracaridcrustacea had the narrowest distribution,polychaete species had distributionssimilarto the mean forthe
whole fauna, and bivalves had the broadestdepthdistributions.A similarobservation was made by Sanders and Grassle (1971) fromepibenthicsled trawldata.
How Many Species Occuirin the Deep Sea?
The discoveryof 798 species froma totalsurfacearea of 21 m2clearlyindicates
a high species richness in a single area at a uniformdepth. Even thoughother
quantitativestudies were based on veryfew samples, resultsfromsmallersamplingeffortscan be comparedwithours. The numberof species (278-351 species
at the 2,100-mstationsand 324-363 species at the 1,500-mstations)presentin
the 1.62-M2 surfacearea representedby the 18 samples fromeach single station
was similarto the 315 species obtainedin a slightlysmallersamplingeffort(1.25
mi2) at 1,230-mdepth in the San Diego Trough offsouthernCalifornia(Jumars
1976). Gage (1979) obtained up to 146 species per 0.25-M2 core at 1,800-2,900-m
depths in the northeastAtlanticRockall Trough.In clustersof fouror fivereplicate 0.25-M2 samples fromthe centralPacific(3,934-5,229-mdepth),Hecker and
Paul (1979) obtained approximately60 species per 80 individualseven though
densities were only about 120 individualsper square meter.Rowe et al. (1982)
obtained about 130 species from eight 0.04-m2 samples from approximately
2,800-mdepth in the northwestAtlantic.Grassle and Morse-Porteous(1987) obtained 250 species in fourbox cores (two 0.25 m2and two 0.09 m2) from3,600-rn
depth in the same region.
Our data show that the numberof species continuedto rise steadilyas more
samples and more individualswere collected. At a single station,species were
added at a rate of about 25 per 0.5 m2(fig.6). Osman and Whitlatch(1978) have
asked whythereare not more species when one considersthe surfacearea of the
deep sea. One answer is thatthe numberof deep-sea species has previouslybeen
greatlyunderestimated.Afterthe initialrapid increase in species as samples are
added along the 176-kmtransect,the data in figure11 suggesta rate of increase
in number of species with distance on the order of 100 species per 100 km.
The rate of addition of species with distance across depth contours or in less
homogeneousregionsis even greater(thisstudy;Blake et al. 1987). The deep sea
at depths greaterthan 1,000 m occupies on the orderof 3 x 108km2.If a linear
rate of additionof one species per kilometeris conservativelygeneralizedto one
species per square kilometer,a deep-sea reservoirof undescribedspecies on the
order of 108 is indicated. Since the deepest and most oligotrophicparts of the
ocean have densities of life more than an order of magnitudelower (1 15/M2;
Hessler and Jumars1974) than the depths consideredhere (4,597/m2;Maciolek
et al. 1987a), an estimateof 107 maybe a betterestimate.This estimateis probably
conservativegiventhataccumulationof species across depthcontoursappears to
be muchfasterthanwithin,and within-depth
contouraccumulationat a relatively
homogeneous site is perhaps the slowest in the deep sea.
This content downloaded from 143.107.244.76 on Fri, 12 Jul 2013 13:50:21 PM
All use subject to JSTOR Terms and Conditions
334
THE AMERICAN NATURALIST
The projectionwe have made froma verysmalldata set has severe limitations.
The veryhighproportionof species representedby singleindividualsis a measure
of the extremeinadequacy of samplingas well as some of the best evidence for
a universe with orders of magnitudemore species. Predictionsconcerningthe
numberof species added as more sites are studieddepend on theassumptionthat
rare species that turnup at more distantsites will not be the same as the rare
species already collected. In the more heterogeneousgroup of stationsadded
rare species were found,and
fromoffNew England and the Carolinas, different
the proportionof species representedby singleindividualsincreased(fig.14). As
expected, the rate of addition of species also increases with the broadeningof
depths and habitatssampled.
In contrastto the deep sea, shallow-watermarinecommunitiesoutsideof tropical areas have relativelyfew species. Shallow tropical marine areas with the
highestpotentialspecies richnessare not well enough studied for an adequate
communitiesare in intercomparisonto be made. The best-studiedsoft-bottom
tidal areas or coastal embaymentswhere the numberof species per numberof
individualscollected usually reaches an asymptoteat fewerthan 100 species for
collections largerthan 1,000 individuals(see, e.g., Hessler and Sanders 1967).
The continentalshelfis more diverse; however,data from70 0.04-M2grab samples, taken over a 3-yrperiod froma stationat 80-mdepthon Georges Bank and
analyzed usingthe same methods,indicatedthatthe numberof species per number of individualsleveled offatjust over 200 species forcollectionsofgreaterthan
100,000individuals(N. J. Maciolek and J. F. Grassle, unpublishedmanuscript).
In general,deep-sea communitiessubjected to obvious, frequentphysicaldisturbancehave reduced species diversity(Grassle 1989). Since mostdeep-sea studies are based on veryfew samples, thecomparisoncan onlybe made on the basis
of relative abundance of species. Our resultsand otherdeep-sea studies show
thatthe most abundantspecies is less than 10% of the fauna,and the majorityof
species in deep-sea communitiesare representedby less than 1% of the fauna.
In shallow-waterenvironments,the most common species generallymakes up
30% or more of the fauna.
A fewareas in thedeep sea are knownto be subjectedto disturbance,and these
are more like shallow-watercommunitiesin havingspecies thatare numerically
in theenvironment
dominant.For example, large-scalefluctuations
such as those
produced by widespreaddeep-sea currentsor shallow-waterstormsare notfavorable to the developmentof species-richcommunities.In a deep-sea area where
currentsmay be 20-25 cm/sforperiodsof severaldays, Thistleet al. (1985) found
thatthe polychaetePaedampharete acutiseries(see Russell 1987forsynonomyof
two species reportedby Thistleet al.) made up 50%-64% of the fauna.
Features of the Deep-Sea EnvironmentFavorable to
MaintainingHigh Species Richness
The patchy inputof food to the deep sea can be consideredboth as a disturbance and, in the sense ofAtkinsonand Shorrocks(1981), a patchyand ephemeral
resource. Withina patch, single species may be abundant,and the diversityis
This content downloaded from 143.107.244.76 on Fri, 12 Jul 2013 13:50:21 PM
All use subject to JSTOR Terms and Conditions
DEEP-SEA SPECIES RICHNESS
335
reduced on a microhabitatscale. Smithet al. (1986) reportedabundances of 67%
+ 1% of the polychaeteLevinsenia oculata in backgroundsamples froma deepsea area characterizedby dense concentrationsof megafaunalmounds. Artificial
mounds also produced similarproportions(68% + 2%) of the same species after
50 d. Grassle and Morse-Porteous(1987) foundthatthepolychaeteOph,yotrocha
sp. A comprised 38% of the individualsin cores dominated by decomposing
Sargassum weed from3,600-mdepthoffNew England; anothermemberof this
genus, 0. akessoni, comprised more than 90% of the individualsin deep-sea
sedimentsaffectedby hydrothermal
venting(Grassle et al. 1985).
Althoughdiversitymay initiallybe reduced by disturbanceon a microhabitat
scale, highoverall diversityappears to be maintainedby theinputof smallpatches
of ephemeral resources and the disturbancethat resultsfromthe activitiesof
individualanimals. Bottommounds(Jumars1976; Smithet al. 1986),vacant burrows (Aller and Aller 1986), activitiesof scavengers(Smith 1986), inputof wood
(Turner 1973, 1977), animals such as glass sponges (Jumars1976) and xenophyophoreans (Levin et al. 1986) projectingabove the sedimentsurface,and sunken
patches of seaweed or salp blooms thathave accumulatedin topographicdepressions (Grassle and Morse-Porteous1987) are all importantsources of small-scale
spatial heterogeneity.Spatial patchinessis produced in partthroughdisturbance
of existingpopulationsbut, more importantly,
throughrelativelyhighconcentrain the absence of
tions of food resources thatresultin long-lastingheterogeneity
widespread disturbanceof the sedimentor otherlarge-scalecatastrophicchange.
In most cases biogenicstructurescontributeto patchinessof organicinput.Mechanisms by which patchy organic resources maintainhigh species diversityare
reviewedby Hanski (1990) and Shorrocks(1990). As is trueon coral reefs(Grassle
1973), patch structurerelevantto individualspecies does not show up in quadrat
sampling.
Pulses of food tend to arrive sporadicallyand to collect in depressions and
burrowsproduced by the activitiesof animalson the deep-sea floor.Many of the
rare species live only in association with these rare and ephemeral resources.
Larvae or dispersingjuveniles of deep-sea animals colonize these patches even
thoughthey are so rare as to be missed even by extensive samplingprograms.
The contributionof species thatinhabita varietyof microhabitats(square centimetersto less than a few square metersin area) separated by hundredsto thousands of metersformsthe species pool available to settleat any specificsite. In
this environment,local diversitydepends less on species interactionsand more
to each area. The lack
on the totalspecies pool and therateof species recruitment
of barriersto dispersal allows distantmigrantsto contributeto local diversity.If
competitionamong species is weak, as is suggestedby low overall densitiesof
macrofauna,then locally high species diversitycan be explained solely on the
basis of the regionalspecies pool.
Croppingactivitiesof megafaunalanimalsalso play a role in maintainingdeepsea diversity(Dayton and Hessler 1972). Species withthe highestpotentialrates
of increase are preventedby predationfrommonopolizingresources(Grassle and
Morse-Porteous 1987).
This content downloaded from 143.107.244.76 on Fri, 12 Jul 2013 13:50:21 PM
All use subject to JSTOR Terms and Conditions
336
THE AMERICAN NATURALIST
Temporal Variationin Deep-Sea Communities
Our results showed no detectable seasonal or annual variationin abundance
or richness (figs. 3, 8). Most existingdescriptionsof deep-sea communitiesare
inadequate to detect even trulydrastic changes in communitystructure.Only
counts of individualspecies will provide a sensitivemeasure of change. Without
these data, climatic shiftsor gradual increases in pollutantinputsover decades
could resultin undetectedwidespread extinctions.Casual observationsof deepon the state of
sea communitiesare not possible. Even the simplestinformation
these communitiescan only be obtained at considerableexpense and involves a
number of specialists workingtogether.The continued quantitativestudy of
deep-sea communitieswith appropriatetaxonomic descriptionof the faunas is
particularlyneeded to determinewhetherthe communitiesthatcharacterizesuch
a large area of the globe are changing.
SUMMARY
Extensive quantitativesamplingof deep-sea macrofaunalcommunitiesfrom
the continentalslope and rise of the eastern United States shows thatthereare
manymore species in thedeep sea thanwere indicatedfromqualitativesampling.
In partbecause the surfacearea of the deep sea is so largeand we have sampled
so littleof it, Thorson's (1971, p. 39) estimateof 160,000marinespecies is certainlytoo low. Our results show a continualincrease in the numberof species
along a depthcontourand even greaterrates of species additionor species accumulationacross depth contours. Even veryconservativeextrapolationsof these
data to the enormous surfacearea of the deep ocean suggestthatthe numberof
species inhabitingthe deep-sea floorhas been greatlyunderestimated.As more
of the deep sea is sampled, the numberof species will certainlybe greaterthan
I millionand may exceed 10 million.
ACKNOWLEDGMENTS
fromthe U.S. DepartThis work was supportedby Contract14-12-0001-30064
mentof the Interior,Minerals ManagementService, to Battelle Ocean Sciences.
We particularlythankJ. A. Blake for data fromoffNorth and South Carolina
and R. F. Petrecca, who served as chief scientiston each of the six cruises off
New Jerseyand Delaware. We appreciate the assistance of E. A. Baptiste and
V. Starczak with data analysis and the very able technicalassistance of many
individualsat Battelle Ocean Sciences and the Woods Hole OceanographicInstitution (WHOI). We especially thank the taxonomistsat Battelle and WHOI:
T. K. Biksey, J. A. Blake, B. Brown, B. Hilbig, H. Jones, M. J. Kravitz, and
R. E. Ruff,polychaetes;L. S. Brown-Leger,amphipodsand isopods; L. S. MorsePorteous, bivalves and scaphopods; I. P. Williams,tanaids; P. W. Nimeskern,
gastropods,aplacophorans,nemerteans,bryozoans,and echinoderms;S. Y. Freitas, echinoderms; M. D. Curran, R. W. Williams,and R. D. Winchell,oligochaetes; R. D. Winchell, decapods and miscellaneous crustaceans; M. Collins,
This content downloaded from 143.107.244.76 on Fri, 12 Jul 2013 13:50:21 PM
All use subject to JSTOR Terms and Conditions
DEEP-SEA SPECIES RICHNESS
337
pogonophorans; and P. S. Winchell,-sipunculans. We are also gratefulto the
followingspecialists who confirmedidentifications:L. E. Watling(University
of Maine), amphipods and cumaceans; G. D. F. Wilson (Scripps Institutionof
Oceanography),isopods; M. A. Rex (UniversityofMassachusetts-Boston), gastropods; J. A. Allen (Dove Marine Laboratory, Scotland), thyasiridbivalves;
E. B. Cutler (Union College), sipunculans;A. Scheltema (WHOI), aplacophorans; and K. Sebens (NortheasternUniversity),anthozoans. We thankR. Etter,
V. Gibson, J. P. Grassle, P. Petraitis,R. F. Petrecca, P. V. Snelgrove,V. Starczak, and J. Weinbergforcommentson the manuscript.
LITERATURE CITED
Abele, L. G., and K. Walters. 1979. Marine benthicdiversity:a critiqueand alternativeexplanation.
Journalof Biogeography6:115-126.
Allen, J. A., and R. E. Morgan. 1981. The functionalmorphologyof Atlanticdeep water species of
the familiesCuspidariidae and Poromyidae(Bivalvia): an analysis of the evolution of the
septibranchcondition.PhilosophicalTransactionsoftheRoyal SocietyofLondon B, Biological Sciences 294:413-546.
Allen, J. A., and H. L. Sanders. 1966. Adaptationsto abyssal life as shown by the bivalve, Abra
profundorum(Smith). Deep-Sea Research 13:1175-1184.
1969. Nucinella serreiLamy (Bivalvia: Protobranchia),a monomyariansolemyidand possible
livingactinodont.Malacologia 7:381-396.
1973. Studies on deep-sea Protobranchia.The familiesSiliculidae and Lametilidae. 11. Museum of ComparativeZoology, Harvard University145:263-310.
Allen, J. A., and J. F. Turner. 1974. On the functionalmorphologyof the familyVerticordiidae
(Bivalvia) withdescriptionsof new species fromtheabyssal Atlantic.PhilosophicalTransactions of the Royal Society of London B, Biological Sciences 268:401-536.
Aller, J. Y., and R. C. Aller. 1986. Evidence forlocalized enhancementof biologicalactivityassociated withtube and burrowstructuresin deep-sea sedimentsat the HEBBLE site, western
NorthAtlantic.Deep-Sea Research 33:755-790.
Atkinson, W. D., and B. Shorrocks. 1981. Competitionon a divided and ephemeral resource: a
simulationmodel. Journalof Animal Ecology 50:461-471.
Blake, J. A., B. Hecker, J. F. Grassle, B. Brown, M. Wade, P. D. Boehm, E. Baptiste, B. Hilbig,
N. Maciolek, R. Petrecca, R. E. Ruff,V. Starczak, and L. Watling.1987. Studyof biological
processes on the U.S. South Atlanticslope and rise. Phase 2. Final reportpreparedforU.S.
Departmentof the Interior,Minerals ManagementService, Washington,D.C.
Carney, R. S., R. L. Haedrich, and G. T. Rowe. 1983. Zonation of fauna in the deep sea. Pages
371-398 in G. T. Rowe, ed. Deep-sea biology. Wiley, New York.
Dayton, P. K., and R. R. Hessler. 1972. Role of biologicaldisturbancein maintaining
diversityin the
deep sea. Deep-Sea Research 19:199-208.
di Castri, F., and T. Younes, eds. 1990. Ecosystem functionof biologicaldiversity.Biology International (special issue 22).
Ekman, S. 1953. Zoogeographyof the sea. Sidgwick& Jackson,London.
Fischer, A. G. 1960. Latitudinalvariationin organicdiversity.Evolution 14:64-81.
Gage, J. D. 1975. A comparisonof the deep-sea epibenthicsledge and anchor-boxdredge samplers
withthe Van Veen grab and hand-coringby diver. Deep-Sea Research 22:693-702.
1979. Macrobenthic communitystructurein the Rockall Trough. Ambio Special Report
6:43-46.
Gauch, H. G., and G. B. Chase. 1974. Fittingthe Gaussian curve to ecological data. Ecology
55:1377-1381.
Gaufin,A. R., E. K. Harris,and H. J. Walter.1956.A statisticalevaluationof streambottomsampling
data obtained fromthreestandardsamplers.Ecology 37:643-648.
This content downloaded from 143.107.244.76 on Fri, 12 Jul 2013 13:50:21 PM
All use subject to JSTOR Terms and Conditions
338
THE AMERICAN NATURALIST
uncertainty.Pages 19-26
Grassle, J. F. 1972. Species diversity,geneticvariabilityand environmental
in B. Battaglia,ed. FifthEuropean Marine Biological Symposium.Piccin, Padua.
1973. Varietyin coral reefcommunities.Pages 247-270 in R. Endean and 0. A. Jones,eds.
The geology and biologyof coral reefs.Academic Press, New York.
1989. Species diversityin deep-sea communities.Trends in Ecology & Evolution4:12-15.
Grassle, J. F., and L. Morse-Porteous.1987. Macrofaunalcolonizationofdisturbeddeep-sea environmentsand the structureof deep-sea benthiccommunities.Deep-Sea Research 34:1911-1950.
Grassle, J. F., and H. L. Sanders. 1973. Life historiesand therole of disturbance.Deep-Sea Research
20:643-659.
Grassle, J. F., and W. Smith. 1976. A similaritymeasure sensitiveto the contributionof rare species
and its use in investigationof variationin marinebenthiccommunities.Oecologia (Berlin)
25:13-22.
Grassle, J. F., H. L. Sanders, R. R. Hessler, G. T. Rowe, and T. McLellan. 1975. Patternand
zonation: a studyof the bathyalmegafaunausingthe research submersibleALVIN. DeepSea Research 22:457-481.
Grassle, J. F., H. L. Sanders, and W. K. Smith. 1979. Faunal changes withdepth in the deep-sea
benthos. Ambio Special Report 6:47-50.
Grassle, J. F., L. S. Brown-Leger,L. Morse-Porteous,R. Petrecca, and I. Williams. 1985. Deep-sea
vents. Bulletinof the Biological Society
fauna of sedimentsin the vicinityof hydrothermal
of Washington6:443-452.
Hanski, 1. 1990. Dung and carrioninsects. Pages 127-145 in B. Shorrocksand I. R. Swingland,eds.
Living in a patchyenvironment.Oxford UniversityPress, Oxford.
Hartman, 0. 1965. Deep-water benthic polychaetous annelids off New England to Bermuda and
otherNorthAtlanticareas. Pt. I. Allan Hancock FoundationPublicationsOccasional Paper
28:1-378.
Hartman,O., and K. Fauchald. 1971. Deep-waterbenthicpolychaetousannelidsoffNew Englandto
Bermuda and otherNorthAtlanticareas. Pt. II. Allan Hancock Monographof Marine Biology 6:1-327.
Hecker, B., and A. Z. Paul. 1979. Abyssal communitystructureof the benthicinfaunaof the eastern
equatorial Pacific: DOMES sites A, B, and C. Pages 287-308 in J. F. Bischoffand D. Z.
Piper, eds. Marine geology and oceanographyof the Pacific Manganese Nodule Province.
Marine Science 9. Plenum, New York.
Hessler, R. R. 1967. A record of Serolidae (Isopoda) fromthe NorthAtlanticOcean. Crustaceana
12:159-162.
1968. The systematicpositionof DactylostylisRichardson(Isopoda, Asellota). Crustaceana
14:143-146.
1970a. The Desmomatidae (Isopoda, Asellota) of the Gay Head-Bermuda transect.Bulletin
of Scripps Institutionof Oceanography 15:1-185.
1970b. A new species of Serolidae (Isopoda) frombathyaldepths of the equatorial Atlantic
Ocean. Crustaceana 18:227-232.
Hessler, R. R., and P. A. Jumars.1974. Abyssal communityanalysis fromreplicatebox cores in the
central NorthPacific. Deep-Sea Research 21:185-209.
Hessler, R. R., and H. L. Sanders. 1967. Faunal diversityin the deep sea. Deep-Sea Research
14:65-78.
communiHughes, R. G. 1984. A model of the structureand dynamicsof benthicmarineinvertebrate
ties. Marine Ecology ProgressSeries 15:1-11.
1986. Theories and models of species abundance. AmericanNaturalist128:879-899.
Hurlbert,S. H. 1971. The nonconcept of species diversity:a critiqueand alternativeparameters.
Ecology 52:577-586.
Hutchinson, G. E. 1959. Homage to Santa Rosalia, or why are there so many kinds of animals?
American Naturalist93:145-159.
Jumars,P. A. 1976. Deep-sea species diversity:does it have a characteristicscale? Journalof Marine
Research 34:217-246.
Kensley, B. F. 1982. Deep water AtlanticAnthuridea(Crustacea: Isopoda). SmithsonianContributions to Biology 346:1-60.
This content downloaded from 143.107.244.76 on Fri, 12 Jul 2013 13:50:21 PM
All use subject to JSTOR Terms and Conditions
DEEP-SEA SPECIES RICHNESS
339
Levin, L. A., D. J. DeMaster, L. D. McCann, and C. L. Thomas. 1986. Effectsof giantprotozoans
(Class Xenophyophorea) on deep-seamount benthos. Marine Ecology Progress Series
29:99-104.
MacArthur,R. H. 1965. Patternsof species diversity.Biological Reviews 40:510-533.
1972. Geographical ecology: patternsin the distributionof species. Harper & Row, New
York.
Maciolek, N. J. 1981. A new genus and species of Spionidae (Annelida: Polychaeta) fromthe north
and south Atlantic.Proceedingsof the Biological Society of Washington94(1):228-239.
. 1985. A revisionof the genus PrioniospioMalmgren,withspecial emphasis on species from
the AtlanticOcean, and new recordsof species belongingto thegeneraApoprionlospioFoster
and Paraprionospio Caullery (Polychaeta, Annelida, Spionidae). Zoological Journalof the
Linnean Society 84:325-383.
. 1987. New species and records of Scolelepis (Polychaeta: Spionidae) fromthe east coast of
NorthAmerica, witha review of the subgenera.Bulletinof the Biological Society of Washington7: 16-40.
Maciolek, N. J., J. F. Grassle, B. Hecker, P. D. Boehm, B. Brown, B. Dade, W. G. Steinhauer,
E. Baptiste, R. E. Ruff,and R. Petrecca. 1987a. Study of biological processes on the U.S.
mid-Atlanticslope and rise. Final report prepared for U.S. Departmentof the Interior,
Minerals ManagementService, Washington,D.C.
Maciolek, N. J., J. F. Grassle, B. Hecker, B. Brown, J. A. Blake, P. D. Boehm, R. Petrecca,
S. Duffy,E. Baptiste, and R. E. Ruff. 1987b. Study of biological processes on the U.S.
North Atlanticslope and rise. Final reportpreparedfor U.S. Departmentof the Interior,
Minerals ManagementService, Washington,D.C.
May, R. M. 1988. How many species are there on earth? Science (Washington,D.C.) 241:14411449.
Menge, B. A., and A. M. Olson. 1990. Role of scale and environmentalfactors in regulationof
communitystructure.Trends in Ecology and Evolution5(2):52-56.
Menge, B. A., and J. P. Sutherland. 1976. Species diversitygradients:synthesisof the roles of
predation,competition,and temporalheterogeneity.AmericanNaturalist110:351-369.
National Science Board. 1989. Loss of biological diversity:a global crisis requiringinternational
solutions. NSB-89-171. National Science Foundation,Washington,D.C.
Nelson, W. G. 1987. An evaluation of deviationfromthe lognormaldistribution
among species as a
pollutionindicatorin marinebenthiccommunities.Journalof ExperimentalMarine Biology
and Ecology 113:181-206.
Nybakken,J. W. 1982. Marine biology. An ecological approach. Harper & Row, New York.
Osman, R. W., and R. B. Whitlatch.1978. Patternsof species diversity:factor artifact?Paleobiology
4:41-54.
Patrick, R. 1973. Use of algae, especially diatoms, in the assessment of water quality. American
Society forTestingand Materials, Special Technical Publication528:76-95.
Petraitis,P. S., R. E. Latham, and R. A. Niesenbaum. 1989. The maintenanceof species diversity
by disturbance.QuarterlyReview of Biology 64:393-418.
Pianka, E. R. 1966. Latitudinalgradientsin species diversity:a reviewofconcepts. AmericanNaturalist 100:33-46.
Pielou, E. C. 1966. Shannon's formulaas a measureof specificdiversity:its use and misuse. American
Naturalist100:463-465.
Preston,F. W. 1948. The commonness,and rarity,of species. Ecology 29:254-283.
Rex, M. A. 1973. Deep-sea species diversity:decreased gastropoddiversityat abyssal depths.Science
(Washington,D.C.) 181:1051-1053.
1976. Biological accommodationin thedeep-sea benthos:comparativeevidence on the importance of predationand productivity.Deep-Sea Research 23:975-987.
Ricklefs,R. E. 1987. Communitydiversity:relativeroles of local and regionalprocesses. Science
(Washington,D.C.) 235:167-171.
Rowe, G. T., P. T. Polloni, and R. L. Haedrich. 1982. The deep-sea macrobenthoson the continental
marginof the northwestAtlanticOcean. Deep-Sea Research 29:257-278.
Russell, D. E. 1987. Paedamzpharetaactutiseries,a new genus and species of Ampharetidae(Poly-
This content downloaded from 143.107.244.76 on Fri, 12 Jul 2013 13:50:21 PM
All use subject to JSTOR Terms and Conditions
340
THE AMERICAN NATURALIST
chaeta) fromthe NorthAtlanticHEBBLE area, exhibitingprogenesisand broad intraspecific
variation.Bulletinof the Biological Society of Washington7:140-151.
Sanders, H. L. 1968. Marine benthic diversity:a comparative study. American Naturalist 102:
243-282.
Sanders, H. L., and J. A. Allen. 1973. Studies on deep-sea Protobranchia(Bivalvia), prologueand
the Pristoglomidae.Bulletinof the Museum of ComparativeZoology, Harvard University
145:237-262.
1977. Studies on the deep-sea Protobranchia(Bivalvia), the familyTindariidaeand thegenus
Pseludotindaria. Bulletin of the Museum of Comparative Zoology, Harvard University
148:23-59.
Sanders, H. L., and J. F. Grassle. 1971. The interactionsof diversity,distributionand mode of
reproductionamong majorgroupingsof the deep-sea benthos.Pages 260-262 in M. Uda, ed.
The ocean world: proceedings of JointOceanographic Assembly. Japan Society for the
Promotionof Science, Tokyo.
Sanders, H. L., and R. R. Hessler. 1969. The ecology of the deep sea benthos.Science (Washington,
D.C.) 163:1419-1424.
Shorrocks, B. 1990. Coexistence in a patchyenvironment.Pages 91-106 in B. Shorrocksand I. R.
Swingland,eds. Living in a patchyenvironment.OxfordUniversityPress, Oxford.
Siebenaller, J. F., and R. R. Hessler. 1977. The Nannoniscidae (Isopoda, Asellota): Hebefiustisn.
gen. and Nannoniscoides Hansen. Transactionsof the San Diego Society of NaturalHistory
19:17-44.
1981. The genera of the Nannoniscidae (Isopoda, Asellota). Transactionsof the San Diego
Society of Natural History 19:227-250.
Smith,C. R. 1986. Nekton falls, low-intensity
disturbanceand communitystructureof infaunalbenthos in the deep sea. Journalof Marine Research 44:567-600.
Smith,C. R., P. A. Jumars,and D. J. DeMasters. 1986. In situstudiesof megafaunalmoundsindicate
rapid sedimentturnoverand communityresponse at the deep-sea floor.Nature (London)
323:251-253.
Smith,W., and J. F. Grassle. 1977. Samplingpropertiesof a familyof diversitymeasures. Biometrics
33:283-292.
Smith, W., D. Kravitz, and J. F. Grassle. 1979. Confidenceintervalsfor similaritymeasures using
the two sample jackknife. Pages 177-191 in L. Orlocci, C. R. Rao, and W. M. Stiteler,
eds. Multivariatemethodsin ecological work. InternationalCo-operativePublishingHouse,
Fairland,Md.
Sugihara, G. 1980. Minimal communitystructure:an explanationof species abundance patterns.
AmericanNaturalist116:770-787.
Thistle, D. 1979. Deep-sea harpacticoidcopepod diversitymaintenance:the role of polychaetes.
Marine Biology 52:371-376.
1980. A revision of Ilyarachna (Crustacea, Isopoda) in the Atlanticwithfournew species.
Journalof Natural History 14:111-143.
Thistle, D., and R. R. Hessler. 1976. Origin of a deep-sea family,the Ilyarachnidae(Crustacea:
Isopoda). SystematicZoology 25:110-116.
1977. A revisionofBetainorpha(Isopoda: Asellota) in theworldocean withthreenew species.
Zoological Journalof the Linnean Society 60:273-295.
Thistle, D., J. Y. Yingst,and K. Fauchald. 1985. A deep-sea benthiccommunityexposed to strong
near-bottomcurrentson the Scotian Rise (westernAtlantic). Marine Geology 66:91-112.
Thorson, G. 1971. Life in the sea. McGraw-Hill,New York.
Turner,R. D. 1973. Wood-boringbivalves, opportunisticspecies in the deep sea. Science (Washington, D.C.) 180:1377-1379.
1977. Wood, mollusks,and deep-sea food chains. Pages 13-19 in Bulletinof the American
Malacological Union for 1977.
Van Valen, L. 1976. Energyand evolution. EvolutionaryTheory 1:179-229.
Wallace, A. R. 1878. Tropical natureand otheressays. Macmillan,London.
Wilson, E. 0. 1985. The biodiversitycrisis: a challengeto science. Issues in Science and Technology
2:20-29.
This content downloaded from 143.107.244.76 on Fri, 12 Jul 2013 13:50:21 PM
All use subject to JSTOR Terms and Conditions
DEEP-SEA SPECIES RICHNESS
341
1988. The currentstate of biological diversity.Pages 3-18 in E. 0. Wilson and F. M. Peter,
eds. Biodiversity.National Academy, Washington,D.C.
Wilson, G. D. 1976. The systematicsand evolutionof Haplonlnnnaand its relatives(Isopoda, Haplomunnidae,new family).Journalof Natural History 10:569-580.
1980a. New insightsinto the colonizationof the deep sea: systematicsand zoogeographyof
the Munnidae and Pleurogoniidaecomb. nov. (Isopoda: Janiroidea).Journalof Natural History 14:215-236.
1980b. Systematicsof a species complex in the deep-sea genus Eltycope, with a revision
of six previouslydescribed species (Crustacea, Isopoda, Eurycopidae). Bulletinof Scripps
Institutionof Oceanography25:1-64.
1981. Taxonomy and postmarsupialdevelopmentof a dominantdeep-sea eurycopidisopod
(Crustacea). Proceedingsof the Biological Society of Washington94:176-294.
Wilson, G. D., and R. R. Hessler. 1974. Some unusualParselloidea (Isopoda, Asellota) fromthedeep
benthos of the Atlantic.Crustaceana 27:47-67.
1980. Taxonomic charactersin the morphologyof the genus Eut-ycope(Crustacea, Isopoda)
witha redescriptionof E. corniutaSars 1864. Cahiers de Biologie Marine 21:241-263.
1981. A revision of the genus Eurycope (Isopoda, Asellota) withdescriptionsof threenew
genera. Journalof Crustacean Biology 21:401-423.
Associate Editor: JohniPastor
This content downloaded from 143.107.244.76 on Fri, 12 Jul 2013 13:50:21 PM
All use subject to JSTOR Terms and Conditions