,•..

Proceedings, Third International Coral Reef SymposiumRosenstlel School of Marine and Atmospheric ScienceUniversity of MiamiMiami, Florida 33149, U.S.A.May 1977

REEF FISHES OVER SPONGE BO'ITOMS OFF THE MOUTH OF THE AMAZON RIVER

Bruce B. ColletteSystematics Laboratory

National Marine Fisheries ServiceNational Museum of Natural History

Washington, D.C •. 20560

and

Klaus RUtzlerDepartment of Invertebrate ZoologyNational Museum of Natural History

Washington, D.C. 20560

ABSTRACT

A typical West Indian reef-fish fauna (45 species) was found associated with a diverse West Indiansponge fauna (35 species) at 14 stations 48-73 m deep off French Guiana and northern Brazil between 5°and 0° N. These records greatly narrow the supposed discontinuity in reef faunas between the Caribbeanand Brazil and indicate that endemism in the tropical marine reef fauna of northeastern Brazil is pro­bably limited to species occurring shallower than 50 m.

KEY WORDS: Reef Fishes, Sponges, Zoogeography.

305

Proceedings, Third International Coral Reef Symposium Rosenstiel School of Marine and Atmospheric Science University of Miami Miami, Florida 33149, U.S.A. May 1977

REEF FISHES OVER SPONGE BOTTOMS OFF THE MOUTH OF THE AMAZON RIVER

Bruce B. Collette Systematics Laboratory

National Marine Fisheries Service National Museum of Natural History

Washington, D.C. 20560

and

Klaus RUtzler Department of Invertebrate Zoology National Museum of Natural History

Washington, D.C. 20560

ABSTRACT

A typical West Indian reef-fish fauna (45 species) was found associated with a diverse West Indian sponge fauna (35 species) at 14 stations 48-73 m deep off French Guiana and northern Brazil between 5° and 0° N. These records greatly narrow the supposed discontinuity in reef faunas between the Caribbean and Brazil and indicate that endemism in the tropical marine reef fauna of northeastern Brazil is pro- bably limited to species occurring shallower than 50 m.

KEY WORDS: Reef Fishes, Sponges, Zoogeography.

305

­,

REEF FISHES OVER SPONGE BOTTOMS OFF THE MOUTH OF THE AMAZON RIVER

Bruce B. Collette and Klaus RUtzler

Introduction

The distribution of reef fish communitiesin the tropical western Atlantic has been con­sidered to be disjunct (1-3), like that of reef­bui Iding corals (4), with a 2700 km gap betweenthe mouths of the Orinoco River in Venezuela andthe Amazon in Brazil. The large volume of silt­laden fresh water coming from the Orinoco, Para,and Amazon rivershas been considered as forminga barrier to the continuous distribution ofcorals and coral-reef fishes, isolating popula­tions north and south of the barrier, therebyleading to speciation (2, 3). Therefore, it isof great zoogeographic interest to report thatpopulations of typical West Indian reef fishesare present over sponges on hard bottoms withoceanic salinity from Cayenne, French Guiana,south to the equator.

Methods and Materials

Collections were made with a 40-foot flatshrimp trawl fished on the bottom at 57 of aplanned 68 stations within the 40-80 m depthrange from approximately 5°N to O· on cruise 58of the National Oceanographic and AtmosphericAdministration R/V Oregon I I in May 1975. Atypical reef fish fauna was found at 14 of these57 stations (Fig. 1). This fauna was probablyalso present at 11 additional stations at thesouthern end of the survey area but these sta­tions could not be sampled due to the very roughbottom. Temperature profiles were made withbathythermograph probes and surface and bottomwater samples were collected for salinity analy­sis at the beginning and end of each trawlSurface and bottom sal inities and temperaturesfor the 14 reef fish stations are summarized inTable 1. Detai led station data is avai lablefrom the National Marine Fisheries Service,Pascagoula, Mississippi.

Fishes were sorted and identified in thefield by the first author. Representative spec­imens were sent to specialists on the list ofthe Smithsonian Oceanographic Sorting Center toverify identifications. A diverse sponge faunawas present at 12 of these 14 stations, with anestimated 800 kg taken at sta. 17702 and 200 kgat sta. 17721. Two large samples of sponges andone smal I sample were frozen in the field andsubsequently identified by the second author.Voucher specimens of fishes, sponges, and someother invertebrates have been deposited in theNational Museum of Natural History.

306

Results

The most common species of reef-fishes takenat the 14 reef fish stations are listed in Table2. Reef fishes that were taken at only a fewstations include the fol lowing species (number ofstations in parentheses): GYmnothorax vicinus(1); AuZostomus macuZatus (2); Apogon pseudomaau­Zatus (2); CephaZopholis fuZva (2); DipZectrumbivittatum (1); EpinepheZus morio (3); Paranthiasfuraifer (l); Serranus phoebe (2); Pristipomoidesaqui linaris (3); Oayurus chPysurus (1); HaemuZonboschmae (2); H. P Zumieri (1); H. s teindachneri(1); MUZZoidichthys martinicus (1); EquetusZanceoZatus (1); HoZacanthus ciZiaris (2);Chromis muZtiZineata (1); Eupomacentrus sp. (1);Bothus Zunatus (1); B. oce ZZatus (3); AZuteramonoceros (2); Cantherhines puZ Zus (1); Stephan­oZepis hispidus (1); ChiZomycterus antennatus(2); C. antiZZarum (1); C. atinga (1). A numberof additional species of fishes, benthic and mid­water species not particularly associated withreefs, was also collected but are omitted as notrelevant to this paper.

Thirty-five species of sponges from 20 fam­i I ies were identified from the three stations(Table 3). A 3-cm pebble attached to the baseof one sponge (sta. 17718) proved to be wei Icemented quartz-rich (22%) calcarenite, contain­ing many mollusk shel I fragments and encrustedby a coral I ine alga, by a bryozoan and by severalspecimens of Homotrema sp. (Foraminiferal. Otherrepresentatives of the benthic population (sta.17702 and 17698) preserved were one species eachof Hydroidea, Zoantharia, Octocorall ia (PeZestosp.), two species of Bryozoa, one species ofOphiuroidea (basket star) and four species ofAscidiacea.

Surface sal inities ranged from 25.1 % 0 atsta. 17720 to 35.7%0 at sta. 17716 except forthe northernmost two of the 14 reef fish stations(17638, 17640) which had water of much lower sa­linity, 17.0 and 22.3%0. Bottom sal initiesshowed much less variation, 34.5-36.40/00, andwere not influenced by freshwater drainage fromthe Amazon. Thus, no inhibitory effect of fresh­water was noted at the bottom where the spongesand reef fishes were found.

-

REEF FISHES OVER SPONGE BOTTOMS OFF THE MOUTH OF THE AMAZON RIVER

Bruce B. Collette and Klaus RUtzler

Introduction

The distribution of reef fish communities in the tropical western Atlantic has been con- sidered to be disjunct (1-3), like that of reef- building corals (4), with a 2700 km gap between the mouths of the Orinoco River in Venezuela and the Amazon in Brazil. The large volume of silt- laden fresh water coming from the Orinoco, Para, and Amazon rivers has been considered as forming a barrier to the continuous distribution of corals and coral-reef fishes, isolating popula- tions north and south of the barrier, thereby leading to speciation (2, 3). Therefore, it is of great zoogeographic interest to report that populations of typical West Indian reef fishes are present over sponges on hard bottoms with oceanic salinity from Cayenne, French Guiana, south to the equator.

Methods and Materials

Collections were made with a 40-foot flat shrimp trawl fished on the bottom at 57 of a planned 68 stations within the 40-80 m depth range from approximately 5°N to 0° on cruise 58 of the National Oceanographic and Atmospheric Administration R/V Oregon II in May 1975. A typical reef fish fauna was found at 14 of these 57 stations (Fig. 1). This fauna was probably also present at 11 additional stations at the southern end of the survey area but these sta- tions could not be sampled due to the very rough bottom. Temperature profiles were made with bathythermograph probes and surface and bottom water samples were collected for salinity analy- sis at the beginning and end of each trawl Surface and bottom salinities and temperatures for the 14 reef fish stations are summarized in Table 1. Detailed station data is available from the National Marine Fisheries Service, Pascagoula, Mississippi.

Fishes were sorted and identified in the field by the first author. Representative spec- imens were sent to specialists on the list of the Smithsonian Oceanographic Sorting Center to verify identifications. A diverse sponge fauna was present at 12 of these 14 stations, with an estimated 800 kg taken at sta. 17702 and 200 kg at sta. 17721. Two large samples of sponges and one small sample were frozen in the field and subsequently identified by the second author. Voucher specimens of fishes, sponges, and some other invertebrates have been deposited in the National Museum of Natural History.

Results

The most common species of reef-fishes taken at the 14 reef fish stations are listed in Table 2. Reef fishes that were taken at only a few stations include the following species (number of stations in parentheses): Gyrrmothorax vicinus (1); Autostomus maaulatus (2); Apogon pseudomacu- latus (2); Cephalopholis fulva (2); Diplectrum bivittatum (1); Epinephelus movto (3); Paranthias furaifer (1); Serranus phoebe (2); Pristipomoides aquilinaris (3); Ocyuvus chrysurus (1); Haemulon bosahmae (2); H. plumieri (1); H. steindaohneri (1); MulZoidiahthys martiniaus (1); Equetus lanceolatus (1); Holaeanthus ailiavis (2); Chromis multilineata (1); Eupomaeentms sp. (1); Bothus lunatus (1); B. oeellatus (3); Alutera monoaeros (2); Ccmtherhines pullus (1); Stephan- olepis hisp-idus (1); Chilomyaterus antennatus (2); C. antillarum (1); C. atinga (1). A number of additional species of fishes, benthic and mid- water species not particularly associated with reefs, was also collected but are omitted as not relevant to this paper.

Thirty-five species of sponges from 20 fam- ilies were identified from the three stations (Table 3). A 3-cm pebble attached to the base of one sponge (sta. 17718) proved to be well cemented quartz-rich (22%) calcarenite, contain- ing many moilusk shell fragments and encrusted by a coralline alga, by a bryozoan and by several specimens of Homotvema sp. (Forami ni fera). Other representatives of the benthic population (sta. 17702 and 17698) preserved were one species each of Hydro Idea, Zoantharia, Octocoral I ia (Telesto sp.), two species of Bryozoa, one species of Ophiuroidea (basket star) and four species of Ascid iacea.

Surface salinities ranged from 25.1°/oo at sta. 17720 to 35.7°/oo at sta. 17716 except for the northernmost two of the 14 reef fish stations (17638, 17640) which had water of much lower sa- linity, 17.0 and 22.3°/oo. Bottom salinities showed much less variation, 34.5-36.4°/oo, and were not influenced by freshwater drainage from the Amazon. Thus, no inhibitory effect of fresh- water was noted at the bottom where the sponges and reef fishes were found.

306

Fig.]. Distribution of coral reef fishes (shaded area) in the tropical western Atlantic (basedon I, 2), Large dots indicate the 14 stations with sponges and reef fishes collected on R/V Oregon I ICruise 58; open circles indicate the stations without a reef fish fauna.

Discussion

Hermatypic representatives of mil leporid andsc Ieracti'n ian cora Is generate massive limestonebui Id-ups in many tropical coastal regi'ons.Their flourishing depends on a narrow spectrumof suitable environmental conditions which areequally advantageous to a great 'variety of ben­thic algae, invertebrates and fishes. Togetherthey form a complex community that can only sur­vive because the coral skeletons provide a highlystructured habitat with ample solid substratesand hiding places. Simi lar configurations withcomparable, although less diverse, associatedflora and fauna are a Iso known as "reefs," Theycan be created by sessi Ie organisms (coral linealgae, gorgonlans, polychaete worms, vermetldmollusks, bryozoans), by eroded rock, by boulderfields and even by man's activity (breakwaters,arti'Hdal reefs composed of autorrobiles, tires,

307

or concrete blocks), Reef fishes are known toassociate with many kinds of substrates thatprovide hiding places (5), as long as their otherecological requirements are met.

Sponges, together with certain algae andgorgonians, are among the dominant organisms ofAtlantic coral reef communities. They gain con­siderable quantitative importance below the depthof 30 m where the abundance of reef corals beginsto decline due to reduced light conditions. Manycora I reef sponges are ramose or are shaped liketubes or vases and give shelter to many speciesof reef fishes (6),

Little evidence has been avai lable previous­ly that "stepping stones" in the form of habitatssuitable for reef fishes exist between the WestIndies and Brazil. Freshwater outflow, fine sed­iments and a strong northwesterly current ih the

m„iMm,,i,,intgi,7hTm,,i,,hitgi,i..m,,i1,m,1i,,inM,.i,,^

Fs

l"mi'l<'wa miuiiii.nutm mm<um i"imi"i''wi'wu lampi_.uR25'

Fig. ]. Distribution of coral reef fishes (shaded area) in the tropical western Atlantic (based on 1, 2). Large dots indicate the 14 stations with sponges and reef fishes collected on R/V Oregon II Cruise 58; open circles indicate the stations without a reef fish fauna.

Discussion

Hermatypic representatives of mi Ileporid and scleractinian corals generate massive limestone build-ups in many tropical coastal regions. Their flourishing depends on a narrow spectrum of suitable environmental conditions which are equally advantageous to a great variety of ben- thic algae, invertebrates and fishes. Together they form a complex community that can only sur- vive because the coral skeletons provide a highly structured habitat with ample solid substrates and hiding places. Similar configurations with comparable, although less diverse, associated flora and fauna are also known as "reefs." They can be created by sessile organisms (coralline algae, gorgon tans, polychaete worms, vermetld mollusks, bryozoans), by eroded rock, by boulder fields and even by man's activity (breakwaters, artificial reefs composed of automobiles, tires,

or concrete blocks). Reef fishes are known to associate with many kinds of substrates that provide hiding places (5), as long as their other ecological requirements are met.

Sponges, together with certain algae and gorgonians, are among the dominant organisms of Atlantic coral reef communities. They gain con- siderable quantitative importance below the depth of 30 m where the abundance of reef corals begins to decline due to reduced light conditions. Many coral reef sponges are ramose or are shaped like tubes or vases and give shelter to many species of reef fishes (6).

Little evidence has been available previous- ly that "stepping stones" in the form of habitats suitable for reef fishes exist between the West Indies and Brazil. Freshwater outflow, fine sed- iments and a strong northwesterly current in the

307

Table 1. Depth (m), temperature (OC), and sal inity (0/00), for 14 reef fish stations of theOregon II off the coast of French Guiana and northern Brazi I .

co 0 co co N '<t l!"\ l!"\ \() r-- co (J\ 0

""' '<t r-- (J\ 0 0 0 N N\() \() \() \() r-- r-- r-- r-- r-- r-- r-- r-- r-- r--r-- r-- r-- r-- r-- r-- r-- r-- r-- r-- r-- r-- r-- r--

depth 71 73 73 62 64 66 59 48 53 68 69 66 64 55surface salinity 17.0 22.3 30.3 29.7 32.8 30.7 33.2 32.2 35.7 34.3 33.8 25.6 25.1 29.4bottom sa I Inity 36.2 36.3 36.4 35.6 36.1 35.8 34.9 34.5 35.7 35.9 36.1 36.0 36.1 35.9surface temp. 27.8 28.3 27.8 27.8 28.3 27.8 27.8 27.8 28.3 27.8 28.9 29.4 28.3 28.3bottom temp. 26.1 26.1 24.5 27.8 27.8 27.2 27.8 27.8 28.3 27.2 27.8 27.8 26.1 27.8reef fish spp. 14 6 10 16 13 6 8 9 14 20 9 10 8 6total fish spp. 31 9 24 27 17 9 8 10 14 25 9 11 10 6sponges present? fi re ? yes yes 800 yes yes yes yes yes yes yes yes 200

sponge kg kg(no. spp. ) 14 22 6

coastal zone of the Gulanas make the Orlnoco­Amazon area an effective barrier for the dis­persal of West Indian reef corals. Ten of 18hermatypic species (56%) known from Brazi I areendemics (7). In comparison, only 5 of 17 spe­cies of gorgonians (30%) are exclusively Brazi 1­ian (8). Simi lar estimates for sponges are moredifficult because of dubious identifications inthe older literature and the scarcity of rei iablyidentified collections from the West Indian andother regions. Hechtel (9), in a detai led zoo­geographical analysis, estimated that 64 of 156Brazi I ian sponge species (42%) are endemics. Ourpresent sponge material is incompletely identi­fied becaus~ some species are either new or re­quire more study and comparison with types.Names and diagnoses of 16 new species I isted byHechtel (9) are not yet avai lable. Of our iden­tified species, al I but one are of West Indianorigin (Homoeodictya grandis has hitherto onlybeen known from South Africa). On the generic

level, the other species al I have strong WestIndian affinities.

Bayer (8) recorded 7 species of gorgoniansfrom waters off the Guianas, al I with West Indiandistribution. Ottmann (10) mentioned the occur­rence of several benthic organisms (bryozoans,sponges, echinoderms and octocorals) from morethan 100 km off the mouth of the Amazon. Noneof these, however, were identified.

From presently avai lable data, it appearsthat taxonomic problems, lack of ecological in­formation and insufficient sampl ing are the maincauses for confl icting zoogeographical concepts.Light, water movement and avai labi I ity of solidsubstrates are the major factors governing thedepth distribution of sessi Ie organisms, likesponges, corals and gorgonians. Most sponges,I ike gorgonians, are less light dependent thanhermatypic corals. In shallow water, large

Table 2. Common reef fishes taken at 14 Oregon I I stations off the coast of French Guiana andnorthern Brazi I. Numbers indicate number of specimens taken at station.

cor-­\()r--

Nor-­r--

cor-­r--

oNr-­r--

Nr-­r--

Gymnothorax oceZZatus 2 2HoZocentrus ascensionis 1 2 67 3 2 2 6 11 2MYripristis jacobus 1 20 3 3 16 10Serranus atropPanchius 2 2 1Priacanthus arenatus 3 10 2 3 3 2 8Lutjanus purpureus 4 12 3 17 1 5 8Lutjanus synagris 1 2 4 11RhombopZites aurorubens 1 2 3 1 8 9HaemuZon auroZineatum 5 1 7 187 10Pseudupeneus macuZatus 15 2 7 1Chaetodon oceZZatus 3 12 2Chaetodon sedentarius 2 20 5 10Pomacanthus paru 31 3 7 5 3 2Pomacanthus arcuatus 1 1Acanthurus chirurgus 3 2 3 1 4sparisoma chrysopterum 28 3 2 3 3 4BaUstes vetuZa 120 31 8 19 8 2 2 2 2Acanthostracion poZygonius 20 2 5 1 1 5 4Acanthostracion quadricornis 6 25 3 4 1

308

CO

r-

o VO

CO r-

CO CM O r-

o in o

in

r» UD

r- K

CO o CM r~

71 73 73 62 64 66 59 48 53 68 69 66 64 55 17.0 22.3 30.3 29.7 32.8 30.7 33.2 32.2 35.7 34.3 33.8 25.6 25.1 29.4 36.2 36.3 36.4 35.6 36.1 35.8 34.9 34.5 35.7 35.9 36.1 36.0 36.1 35.9 27.8 28.3 27.8 27.8 28.3 27.8 27.8 27.8 28.3 27.8 28.9 29.4 28.3 28.3 26.1 26.1 24.5 27.8 27.8 27.2 27.8 27.8 28.3 27.2 27.8 27.8 26.1 27.8

14 6 10 16 13 6 8 9 14 20 9 10 8 6 31 9 24 27 17 9 8 10 14 25 9 11 10 6

f i re ? yes yes 800 yes yes yes yes yes yes yes yes 200 sponge

_ 14 kg

22 6 kg

level, the other species a I Indian affinities.

have strong West

Table 1. Depth (m), temperature (°C), and salinity (°/oo), for 14 reef fish stations of the Oregon II off the coast of French Guiana and northern Brazil.

depth surface salinity bottom salinity surface temp, bottom temp, reef fish spp. total fish spp. sponges present?

(no. spp. )

coastal zone of the Guianas make the Orinoco- Amazon area an effective barrier for the dis- persal of West Indian reef corals. Ten of 18 hermatypic species (56%) known from Brazil are endemics (7). In comparison, only 5 of 17 spe- cies of gorgonians (30%) are exclusively Brazil- ian (8). Similar estimates for sponges are more difficult because of dubious identifications In the older literature and the scarcity of reliably identified collections from the West Indian and other regions. Hechtel (9), in a detailed zoo- geographical analysis, estimated that 64 of 156 Brazilian sponge species (42%) are endemics. Our present sponge material is Incompletely identi- fied because some species are either new or re- quire more study and comparison with types. Names and diagnoses of 16 new species listed by Hechtel (9) are not yet available. Of our iden- tified species, all but one are of West Indian origin {Homoeodiatya grandis has hitherto only been known from South Africa). On the generic

Bayer (8) recorded 7 species of gorgonians from waters off the Guianas, all with West Indian distribution. Ottmann (10) mentioned the occur- rence of several benthic organisms (bryozoans, sponges, echinoderms and octocora I s) from more than 100 km off the mouth of the Amazon. None of these, however, were identified.

From presently available data, it appears that taxonomic problems, lack of ecological in- formation and insufficient sampling are the main causes for conflicting zoogeographical concepts. Light, water movement and availability of solid substrates are the major factors governing the depth distribution of sessile organisms, like sponges, corals and gorgonians. Most sponges, like gorgonians, are less light dependent than hermatypic corals. In shallow water, large

Table 2. Common reef fishes taken at 14 Oregon II stations off the coast of French Guiana and northern Brazil. Numbers indicate number of specimens taken at station.

Gyrrtnothorax ocellatus Holocentrus ascensionis Myripristis jacobus Serranus atrobranchius Pviacanthus avenatus Lutjanus purpureus Lutjanus synagvis Rhomboplites auvorubens Haemulon aurolineatum Pseudupeneus maoulatus Chaetodon ocellatus Chaetodon sedentarius Pomacanthus paru Pomacanthus arcuatus Acanthuvus chimrgus Sparisoma chrysopterum Batistes vetula Aaanthostracion polygonius Aaanthostracion quadricornis

CO

to r-

o

1~~

CO CO o\

l*«

CM O r»

o o in

f- I'-

CO

I-- r-

ON o CM

2 1 2 1 - - 2 67 3 2 2 6 ll 2 - - - 1 2 3

- - 20 2 2

3 3 - - 16 10 - - - -

_ 10 3 _ 1 3 1 2 - 8 - - 4 12 3 17 1 5 - - - 8 - - - - 1 - - 2 - - - - - 4 - 11 - - 1 2 3 1 - - - - - 8 - 9 - - - - - 5 1 - - - 7 187 1 10 - - - - - - 15 2 1 - - 7 - 1 - - 3 - - - 12 - - - - - - 2 1 - 2 - 20 - - 5 - - - 10 - - - - - - - - 31 - 3 7

1 2

5 1

3 - 2 1 -

1 3 _ 3 _ 1 4 - - - - 28 1 3 - 2 - 3 - 3 4 - - 120 - - - 31 8 19 8 2 2 2 2 - - - - 20 - - 2 5 1 1 - 5 4 6 25 3 4 - - - - 1 - - 1 - -

308

.,•

, : -Table 3. Sponges (Demospongea) from three Oregon II stations off the coast of northern Brazi I

(number of stations shown; asterisk indicates new record for Brazi I).

Order DictyoceratidaSpongi idae

Ircinia strobilina (Lamarck)-2Verongi idae

Fasciospongia sp., cf. turgida (Lamarck)-lVerongia caulifoY'l77is (Carter)-2Verongia fistularis (Pal las) f. fistularis-1Verongia fistularis (Pal las) f. fulva-1Pseudoceratina crassa (Hyatt)-l

Order HaploscleridaHa Ii c Ion idae

Haliclona sp.-1Xestospongia muta (Schmidt)-lXestospongia sp.-1

Ca II yspong i idaeCallyspongia pergamentacea (Ridley)-lCallyspongia villosa (Pallas)-l

Order Poeci loscleridaAdoci idae

Petrosia sp.-1Agelasidae

Agelas clathrodes (Schmldt)-lEsperiopsidae

Homo?odictya grandis Ri d Iey & Dendy-2*Coelosphaeridae

Coeiosphaera biclavata (Priest)-l*Inflatella bartschi Laubenfels-1*Inflatella sp.-1*

Myca 11 daeEchinostylinos (?) unguiferus Laubenfels-2Mycale sp.-1

Myxi II idae

sponges are more affected by wave action becausethey are readily dislodged. AI I three groupsneed solid substrates to attach themselves. Manysponges may have an advantage because they cananchor on coarse sand substrates as long as thereare only steady currents (boundary layer effect)and no osci I lating water movement (below thereach of waves). Some can stabilize their sub­strate by rooting, others by attaching to, andthus connecting many smal I pieces of shel I orrubb Ie (11),

Ottmann (10) indicated that a belt of shellysand arches around the fluviatile quartz sandsand clay-rich sediments beyond ]00 km from MarajoIsland in the mouth of the Amazon, at a depth of30-100 m. The majority of our stations extendfrom about this location to the edge of the con­tinental shelf and range from 48 to 73 m indepth. Due to the nature of the sampl ing gearused we do not have much information on substrateconditions but, from the behavior of the trawland from fathometer readings, we conclude thathard bottoms are abundant in this region.

Conclusions

The composition, diversity and biomass ofbenthic fauna obtained on Oregon cruise 58 showthat excel lent conditions for the survival of a

309

Iotrochota sp.-1Clathri idae

cf. Clathria sp.-1Order Hal ichondrida

Ha I ichondri idae~enus and species unidentified-1

Order Axi ne II IdaAxi nell idae

Perissinella sp., cf. madeirensis Topsent-2Pseudaxinella lunaecharta (Ridley &Dendy)-icf .. Teichaxinella sp.-2Thrinacophora funiformis Ridley &Dendy-2

Raspa i I I I daeEchinodictyum sp.-1Raspai.lia sp.-1

Euryponidaecf. Eurypon sp.-1

Order Hadromeri daChond ros i idae

Chondrilla nucula Schmidt-1Order CnoristidCl

GeodildaeGeodia neptuni (Sollas)-l

Stellettldaecf. Stelletta sp.-2Penares sp., cf~ S. mastoidea Schmidt-1

Order SpirophoridaTeti II idae

Cinachyra kuekenthali UI iczka-2Order Homosclerophorida

PlakinidaePlakortis halichondrioides (Wi Ison)-l

West Indian reef fauna exist on the continentalshelf off the Guianas and the Amazon. A verysimi lar reef fish fauna has been reported oversponge and dead coral bottom at 55-73 m offGuyana (12). Additional sampling is I ikely toproduce more evidence for zoogeographical "step­ping stones" between the West Indies and Brazi I.Collecting with scuba and ichthyocides wi I Idoubtless increase the number of sma I I speciesof reef f Ishes known f rom these stepp Ing stones.More and better identified collections shouldshow that the fresh si It-laden water of the Ama­zon and Orinoco acts as a fi Iter barrier to inter­rupt gene flow between populations of speciesdirectly associated with corals or occurring inreef habitats shal lower than 50 m. Endemism inthe tropical marine fauna of Brazi I Is probablyconfined to species with a bathymetric distribu­tion limit of approximately 50 m.

Acknowledgments

We thank the crew and scientific staff ofthe Oregon, the chief scientist of cruise 58,Alexander Dragovich, and Elmer Guthertz, fortheir cooperation and assistance durin~ thecruise. Voucher fish specimens were identifiedby Wi II iam D. Anderson, Jr., Eugenia B. andJames E. BOhlke, Margaret G. Bradbury, LeonardJ. V. Compagno, C. E. Dawson, Carter R. Gi Ibert,

Table 3. Sponges (Demospongea) from three Oregon II stations off the coast of northern Brazil (number of stations shown; asterisk indicates new record for Brazil).

Order Dictyoceratida SpongiIdae

Irainia strobilina (Lamarck)-2 Verongi idae

Fasoiospongia sp., cf. turgida (Lamarck)-1 Verongia cauliformis (Carter)-2 Verongia fistulavis (Pallas) f. fistularis-} Verongia fistulavis (Pallas) f. fulva-} Pseudoaeratina orassa (Hyatt)-1

Order Hap losclerida Haliclonidae

Halialona sp.-i Xestospongia muta (Schmidt)-1 Xestospongia sp.-l

CaI Iyspongi i dae Callyspongia pergamentaaea (Ridley)-] Callyspongia villosa (Pallas)-I

Order Poecilosclerida Adoci idae

Petrosia sp.-l Age Iasidae

Agelas elathrodes (Schmidt)-] Esperiopsidae

Homoeodiatya grandis Ridley & Dendy-2* Coelos'phaeridae

Coelosphaera bialavata (Priest)-]* Inflatella bavtschi Laubenfel s-1* Inflatella sp.-l*

Mycalidae Eahinostylinos (?) unguiferus Laubenfels-2 Myoale sp.-l

MyxiI Iidae

sponges are more they are read 11y need solid subst sponges may have anchor on coarse are only steady and no osci Ilati reach of waves), strate by root in thus connecting rubble (11).

affected by wave action because dislodged. All three groups rates to attach themselves. Many an advantage because they can sand substrates as long as there

currents (boundary layer effect) ng water movement (below the

Some can stabilize their sub- g, others by attaching to, and many small pieces of shell or

Ottmann (10) indicated that a belt of shelly sand arches around the fluviatile quartz sands and clay-rich sediments beyond 100 km from MaraJo Island in the mouth of the Amazon, at a depth of 30-100 m. The majority of our stations extend from about this location to the edge of the con- tinental shelf and range from 48 to 73 m in depth. Due to the nature of the sampling gear used we do not have much information on substrate conditions but, from the behavior of the trawl and from fathometer readings, we conclude that hard bottoms are abundant in this region.

Conclusions

The composition, diversity and biomass of benthic fauna obtained on Oregon cruise 58 show that excellent conditions for the survival of a

Iotrochota sp.-] ClathriIdae

cf. Clathria sp.-l Order Halichondrida

Ha Iichondri idae genus and species unidentified-!

Order AxinelI Ida AxinelIidae

Perissinella sp., cf. madeirensis Topsent-2 Pseudaxinella lunaeoharta (Ridley & Dendy)-/ cf. Teiehaxinella sp.-2 Thrinaoophora funiformts Ridley & Dendy-2

RaspaiIi idae Echinodiatyum sp.-l Baspailia sp.-]

Euryponidae cf. Eurypon sp.-]

Order Hadromeri da Chondrosi idae

Chondrilla nuaula Schmidt-1 Order Choristida

Geodi idae Geodia neptuni (Sol I as)-1

Stellettidae cf. Stelletta sp.-2 Penares sp., cf. S. mastoidea Schmidt-1

Order Spirophorida TetiI Iidae

Cinaahyva kuekenthali Uliczka-2 Order Homosclerophorida

PIakinldae Plakortis halichondrioides (Wllson)-I

West Indian reef fauna exist on the continental shelf off the Guianas and the Amazon. A very similar reef fish fauna has been reported over sponge and dead coral bottom at 55-73 m off Guyana (12). Additional sampling is likely to produce more evidence for zoogeographicaI "step- ping stones" between the West Indies and Brazil. Collecting with scuba and ichthyocides will doubtless increase the number of small species of reef fishes known from these stepping stones. More and better identified collections should show that the fresh silt-laden water of the Ama- zon and Ori noco acts as a filter barrier to inter- rupt gene flow between populations of species directly associated with corals or occurring in reef habitats shallower than 50 m. Endemism in the tropical marine fauna of Brazil is probably confined to species with a bathymetric distribu- tion limit of approximately 50 m.

Acknowledgments

We thank the crew and scientific staff of the Oregon, the chief scientist of cruise 58, Alexander Dragovich, and Elmer Guthertz, for their cooperation and assistance during the cruise. Voucher fish specimens were identified by William D. Anderson, Jr., Eugenia B. and James E. Boh Ike, Margaret G. Bradbury, Leonard J. V. Compagno, C. E. Dawson, Carter R. Gilbert,

309

and C. Richard Robins. Joseph Russo assisted inpreparing Table 2 and Fi~. 1 and Keiko HiratsukaMoore completed the map. Useful comments ondrafts of the ms were contributed by John C.Briggs, Daniel M. Cohen, Alexander Dragovich,Carter R. Gi Ibert, George C. Mi Iler, and C.Richard Robins.

References

1. BOhlke, J. E. and C. C. G. Chaplin. 1968.Fi shes of the Bahamas and ad jacent trop ica Iwaters. Livingston Publ. Co., Wynnewood,Pa., map on inside cover.

2. Gi Ibert, C. R. 1973. Characteristics ofthe western Atlantic reef-fish fauna.Quart. J. Fla. Acad. Sci. 12.: 130-144..

3. Briggs, J. C. 1974. Marine zoogeography.McGraw-Hi I I Book Co., New York, p. 68.

4. Wells, J. W. 1957. Coral Reefs. In:Treatise of marine ecology and paleoecology,J. W. Hedgpeth (ed. l, Geol. Soc. Amer. Mem.67: 609-631 .

5. Randa I I, J. E. 1963. An ana Iys is of thefish populations of artificial and naturalreefs in the Virgin Islands. Carib. Jour.Sci. 3: 31-47.

6. Tyler, J. C. and J. E. BOhlke. 1972.Records of sponge-dwel I ing fishes,primarily of the Caribbean. Bul I. Mar.Sci. 22: 601-642.

7. Laborel,J. 1969. LespeuplementsdeMadreporaires des cotes tropicale duBresi I. Ann. Univ. Abidjan (El~: 1-260.

8. Bayer, F. M. ]96] . The sha I low-waterOctocora I I ia of the West Irid ian reg Ion.Martinus NiJhoff, The Hague, pp. 321-326.

9. Hechtel, G. J. 1976. Zoogeography ofBrazi I ian marine Demospongiae. In:Aspects of~ biology (F. W. Harrisonand R. R. Cowden, eds.l. Academic Press,New York, pp. 237-260.

10. Ottmann, F. 1959. Estudo das amostras do~ fundo recolhidas pelo N. E. "Almirante

Saldanha," na regiao da embocadura do RioAmazonas. Trab. Inst. Bioi. Mar.Oceanogr. Univ. Recife 1: 77-]06.

I]. RUtzler, K. ]965. Substratstabi I it~t 1mmarlnen Benthos als okologischer Faktor,dargestel It am Beispiel adriatischerPori fera. Int. Revue Ges. Hydrobiol. ~:281-292.

310

12. Lowe (McConnell l, R. H. 1962. The fishesof the British Guiana continental shelf,Atlantic coast of South America, with noteson their natural history. J. Linnean Soc.London, zool., 44: 669-700.

and C. Richard Robins. Joseph Russo assisted in preparing Table 2 and Fig. 1 and Keiko Hiratsuka Moore completed the map. Useful comments on drafts of the ms were contributed by John C. Briggs, Daniel M. Cohen, Alexander Dragovich, Carter R. Gilbert, George C. Miller, and C. Richard Robins.

12. Lowe (McConnell), R. H. 1962. The fishes of the British Guiana continental shelf, Atlantic coast of South America, with notes on their natural history. J. Linnean Soc. London, zoo I., 44: 669-700.

References

Bohlke, J. Fishes of

E. the

and C. C. G, Bahamas and

waters. Livingston Pub I. Pa., map on inside cover.

Chaplin. 1968. adjacent tropicaI Co., Wynnewood,

2. Gilbert, C. R. 1973. Characteristics of the western Atlantic reef-fish fauna. Quart. J. Fla. Acad. Sci. 35: 130-144. '

3. Briggs, J. C. 1974. Marine zoogeography. McGraw-Hill Book Co., New York, p. 68.

4. Wells, J. W. 1957. Coral Reefs. In: Treatise of marine ecology and pa IeoecoIogy, J. W. Hedgpeth (ed.), Geol. Soc. Amer. Mem. 67: 609-631.

5. Randall, J. E. 1963. An analysis of the fish populations of artificial and natural reefs in the Virgin Islands. Carib. Jour. Sci. 3: 31-47.

6. Tyler, J. C. and J. E. Bohlke. 1972. Records of sponge-dwelling fishes, primarily of the Caribbean. Bull. Mar. Sci. 22: 601-642.

7. Lahore I, J. 1969. Les peupl ements de Madreporaires des cotes tropicale du Bnlsil. Ann. Univ. Abidjan (E) 2_: 1-260.

8. Bayer, F. M. 1961. The shaI low-water OctocoralIia of the West Indian region. Martinus Nijhoff, The Hague, pp. 321-326.

9. Hechtel, G. J. 1976. Zoogeography of Brazilian marine Demospongiae. In: Aspects of sponge biology (F. W. Harrison and R. R. Cowden, eds.). Academic Press, New York, pp. 237-260.

10. Ottmann, F. 1959. Estudo das amostras do " fundo recolhi das pelo N. E. "Almirante

Saldanha," na regiao da embocadura do Rio Amazonas. Trab. Inst. Bio I. Mar. Oceanogr. Univ. Recife U 77-106.

11. RUtzler, K. 1965. SubstratstabiI itSt im marinen Benthos a Is okologf scher Faktor, dargesteI If am Bel spiel adriati scher Porifera. Int. Revue Ges. Hydrobiol. 50: 281-292.

310