Full text of "Proceedings of the Second International Corbicula Symposium"

See other formats

XXAUW - SjSWj

m ^ ^ m w m ^ w

co — co £ — co \ 2 co

IAN INSTITUTION NOIlfUliSNI NVIN0SH1IWS S3iavaai3 LIBRARIES SMITHSONIAN INSTITUTION NOlinill

7 to 2 _ CO 2 ..... CO 21 CO

< -v4S\, 2 ,< . ,• ^ .v#x E xf^Tx < 2

s\a\\V -i t- AW. , „

_ _ 2

*> ’•v > *40* 2 > ','Wr' 2 ^ >

* 2 CO 2 CO V Z Jo ‘ 2

iws ssiavaan libraries Smithsonian institution NounmsNi nvinoshiiws SBiavaan librari

5 2 5 Z .«, 2 _ 5

cc
<

O _ ^ _ ^

lAN^INSTITUTION^ NOlinillSNI “NVINOSHIIWS SBiaVaail^LIBRARI ES^ SMITHSONIAN-* INSTITUTION NOlinill

r* z r~ , 2 i— 2 r~ Z

u*qx o m ^ 2 xEEii'cA “ xruJox 2 sM*., m /^"px 2 ^

CO
PO
>

m xx^lX' ^ m •' 2 xiA^i^ m m xovost^

iws^ss i avaa n“u brar i es smithsonian”institution NoiimiiSNi “nvinoshiiws saiavaan librari

CO 2 CO 2 T CO 2 CO

2 ^ i ^ A E < xT^qX 2

I II, . Jl I '3il I

^ j, , «g x^\os ~i> ’

lAN^lNSTITUTION^NOlinillSNI NVIN0SH1IWS°°S3 I a V8 8 11 LIBRARI ES^SMITHSONIAN INSTITUTION NOlinill

_ CO ^ Z \ 00 ^ 2 CO ~ CO

5 ^ X^am^X o ' ” XIVA^ O XCVostjvx — o

iws^saiavaan^LiBRARiEs smithsonian institution NouniiiSNi^NViNOSHiiws saiavaan librar

2 r- 2 r* 2 2

O ~ o X^X z j£k* o x^x m /ks8&

^ 'n^jiiS^V 2 vr-- X ^ — — - • - \. "Z. ‘ — - f n

^ ^ ^ — — co — c/> _ {/) \ f: ^

IIAN “INSTITUTION NOlinillSNI^NVINOSHllWS^SB IBVaail ^LIBRARI ES ^SMITHSONIAN^ INSTITUTION^ NOIlflllJ

5 %% 1 %»■

/ 5 ^

n^s°°s 3 1 avaan^ librar i es^smithsonian^ institution^noiioiiisni nvinoshiiws^sb i ava a n_ li b rar

co ^ CO ^ , . ^ ^ 5 vx,- .2 ^ 2

MSS^XA 2 x

cc V;,%, .K^iV 2 W\ CC

. ^ ttW// -’ VcPWXV m 5 2 5 ^2 5 ^2^:

'IIAN"J INSTITUTION ^ N0linillSNnNVIN0SHlllAIS^S3 I BVB a n_LI B RAR I ES^SMITHSONlAN^INSTITUTION^NOlinill

O ~ X>a X ° X^riri^

co \ co /c^sTm^A x 2 /xcffTATN

pp /j,

> £

m '' ^ xivA^>ix m 'UjAiost^x

Lms^sa iavaan~*LIBRARI ESt/,SMITHSONIAN_INSTITUTIONt/,NOIinillSNl“NVINOSHimS S3 lavaail LIBRAR

CO 2 CO 2 T CO 2 ^ - £

BAN

INSTITUTION NOimillSNI NVIN0SH1IINS S3iavaail LIBRARIES SMITHSONIAN JNSTITUTION NOIlflllJ

-r rn

co X t o \ it ¥*

noshiiins S3 1 a va an libraries Smithsonian institution

■' " z ^

s — xgs* -

"-*sc !Qi^ w

Z to V Z 00

thsonian institution NouniiJLSNi nvinoshiiims S3iavaan_

^ __ . co 2

in A' C/3

.« , /cr^ww^y\ — /o^TirvTA w/,

H /»*7AVi ~ H a. xS X\\

/* y/y. ' BB r '

'//^d

* ' \X f

O X^V Q

2 _l 2 ^ _

noshiiiais saiavaan libraries Smithsonian institution^

• z ^ ^ r- Z r- z

m xovos^v^ ^ m

to _ co

ITHSONIAN INSTITUTION NOIlfllllSNI NVINOSHIIIMS SSIBVaail
z co z co

v < £ <

SKES* I

t I , mr. /■'" t

s >'' s '<^o^y >• s

co • Z CO z 60

noshiiws saiavaan libraries Smithsonian institution

CO = CO —

ITHSONIAN INSTITUTION NOIlfllllSNI NVINOSHIIIMS S3l3Vaan

m -<zo^ »> N m X co Nccoi;iy$Z rn

CO £ (7) \ Z co

inoshiiims S3 lava an libraries Smithsonian institution

z .... co z ^ co

, V- ? /^Tx < 4K;s S

-I .,-/ A,, z ATnA *h z vW%-*

4?- ° X 3C (£Myrtb& o »T< X

J7/ ,n co 31 co wm v w

_ £

> 'W"* 2 > ^pTiisg^ 2 >

ITHSONIAN INSTITUTION NOlinillSNI NVINOSHIIIMS^SS I ava 3 Cl

co CO —

^ovJx cu ,n UJ /»«, U>

o "Xofj ocjA q ^ '^'x^oi'iTsgy o

NosHims S3 1 ava a h^librar i es^smithsonian~institution

. z r- z <“ z

^ ^ xfoo JvrTx O xToo VoX CD

33 /ii

33 W\

m Xo<os"X/ ^ m

ITHSONIAN INSTITUTION^NOlinillSNl'NVINOSHUlMS^Sa 1 HVa 3 II
z » co z to

- z -H W£///s 2 /®W\ -H W%4&/, Z

to MW'/xS: <o

X
co
O

'W'labP’jsy £ ^

2 N^v > ' 5 Xi\os^/ > 2

iNOSHims S3 1 ava an li brar i es^smithsonian institution^

\ CO X CO — CO

I • I

LI B RAR I ES^SMITHSONIAN^ INSTITUTI0N°°N0l

NoiiruiiSNi^NviNOSHiiiMS^sa i ava a n“*u e

(— z r-

LIBRARIES SMITHSONIAN INSTITUTION NOI

LIB RAR I ES^SMITHSONIAN^INSTITUTION^NOI

NouniiiSNi "‘nvinoshiiims^s 3 1 ava a n ^l ie

, PROCEEDINGS OF THE
'SECOND INTERNATIONAL
CORBICULA SYMPOSIUM

SPECIAL EDITION NO. 2 OF THE AMERICAN MALACOLOGICAL BULLETIN

SPECIAL EDITION NO. 2

JUNE 1986

CONTENTS

Preface

Historical review of Asiatic clam ( Corbicula ) invasion and biofouiing of waters and industries

in the Americas. BILLY G. ISOM 1

The zoogeography and history of the invasion of the United States by Corbicula fluminea

(Bivalvia: Corbiculidae). CLEMENT L. COUNTS, III 7

Biofouling of power plant service systems by Corbicula.

T. L. PAGE, D. A. NEITZEL, M. A. SIMMONS and P. F. HAYES 41

Engineering factors influencing Corbicula fouling in nuclear service water systems.

K. I. JOHNSON, C. H. HENAGER, T. L. PAGE and P. F. HAYES 47

Corbicula control at the Potomac River Steam electric station Alexandria, Virginia.

JEANNE MILES POTTER and LAWRENCE H. LIDEN 53

A mechanical strainer design for Corbicula fouling prevention in the service water system

at Arkansas Nuclear One, Unit 2. DAVID MACPHEE 59

Development of a Corbicula control treatment at the Baldwin power station.

JAMES A. SMITHSON 63

Corbicula fouling and control measures at the Celco Plant, Virginia.

DONALD S. CHERRY, ROB L. ROY, RICHARD A. LECHLEITNER, PATRICIA A.

DUNHARDT, GREGORY T. PETERS and JOHN CAIRNS, Jr 69

Asiatic clam control by mechanical straining and organotin toxicants.

YUSUF G. MUSSALLI, I. A. DIAZ-TOUS and JAMES B. SICKEL 83

Corbicula population mortalities: factors influencing population control.

JAMES B. SICKEL 89

Controlling Corbicula (Asiatic clams) in complex power plant and industrial water systems.

BILLY G. ISOM, CHARLES F. BOWMAN, JOSEPH T. JOHNSON and

ELIZABETH B. RODGERS 95

Power station entrainment of Corbicula fluminea (Muller) in relation to population
dynamics, reproductive cycle and biotic and abiotic variables.

CAROL J. WILLIAMS and ROBERT F. MCMAHON 99

Corbicula in Asia — an updated synthesis.

BRIAN MORTON 113

— continued on back cover —

SECOND INTERNATIONAL CORBICULA SYMPOSIUM

EDITOR

JOSEPH C. BRITTON

MANAGING EDITOR

ROBERT S. PREZANT

AMERICAN MALACOLOGICAL BULLETIN

BOARD OF EDITORS

EDITOR

ROBERT S. PREZANT

Department of Biological Sciences
University of Southern Mississippi
Hattiesburg, Mississippi 39406-5018

ASSOCIATE EDITORS

MELBOURNE R. CARRIKER

College of Marine Studies
University of Delaware
Lewes, Delaware 19958

ROBERT ROBERTSON

Department of Malacology
The Academy of Natural Sciences
Philadelphia, Pennsylvania 19103

GEORGE M. DAVIS

Department of Malacology
The Academy of Natural Sciences
Philadelphia, Pennsylvania 19103

R. TUCKER ABBOTT
American Malacologists, Inc.
Melbourne, Florida, U.S.A.

JOHN A. ALLEN
Marine Biological Station
Millport, United Kingdom

JOSEPH C. BRITTON
Texas Christian University
Fort Worth, Texas, U.S.A.

JOHN B. BURCH
University of Michigan
Ann Arbor, Michigan, U.S.A.

EDWIN I N. CAKE, JR.

Gulf Coast Research Laboratory
Ocean Springs, Mississippi, U.S.A.

PETER CALOW
University of Sheffield
Sheffield, United Kingdom

JOSEPH G. CARTER
University of North Carolina
Chapel Hill, North Carolina, U.S.A.

JAMES W. NYBAKKEN

Ex Officio

Moss Landing Marine Laboratories
Moss Landing, California 95039-0223

BOARD OF REVIEWERS

ARTHUR H. CLARKE
Ecosearch, Inc.

Portland, Texas, U.S.A.

CLEMENT L. COUNTS, III
University of Delaware
Lewes, Delaware, U.S.A.

WILLIAM K. EMERSON
American Museum of Natural History
New York, New York, U.S.A.

DOROTHEA FRANZEN
Illinois Wesleyan University
Bloomington, Illinois, U.S.A.

VERA FRETTER
University of Reading
Berkshire, United Kingdom

JOSEPH HELLER

Hebrew University of Jerusalem

Jerusalem, Israel

ROBERT E. HILLMAN
Battelle, New England
Marine Research Laboratory
Duxbury, Massachusetts, U.S.A.

W. D. RUSSELL-HUNTER

Department of Biology
Syracuse University
Syracuse, New York 13210

K. ELAINE HOAGLAND
Academy of Natural Sciences
Philadelphia, Pennsylvania, U.S.A.

RICHARD S. HOUBRICK
U.S. National Museum
Washington, D.C., U.S.A.

VICTOR S. KENNEDY
University of Maryland
Cambridge, Maryland, U.S.A.

ALAN J. KOHN
University of Washington
Seattle, Washington, U.S.A.

LOUISE RUSSERT KRAEMER
University of Arkansas
Fayetteville, Arkansas, U.S.A.

JOHN N. KRAEUTER
Baltimore Gas and Electric
Baltimore, Maryland, U.S.A.

ALAN M. KUZIRIAN
Laboratory of Biophysics
NINCDS-NIH at the
Marine Biological Laboratory
Woods Hole, Massachusetts, U.S.A.

ISSN 0740-2783

RICHARD A. LUTZ
Rutgers University
Piscataway, New Jersey, U.S.A.

EMILE A. MALEK
Tulane University
New Orleans, Louisiana, U.S.A.

MICHAEL MAZURKIEWICZ
University of Southern Maine
Portland, Maine, U.S.A.

JAMES H. McLEAN

Los Angeles County Museum

Los Angeles, California, U.S.A.

ROBERT F. McMAHON
University of Texas
Arlington, Texas, U.S.A.

ROBERT W. MENZEL
Florida State University
Tallahassee, Florida, U.S.A.

ANDREW C. MILLER
Waterways Experiment Station
Vicksburg, Mississippi, U.S.A.

JAMES J. MURRAY, JR.

University of Virginia
Charlottesville, Virginia, U.S.A.

WINSTON F. PONDER
Australian Museum
Sydney, Australia

CLYDE F. E. ROPER
U.S. National Museum
Washington, D.C., U.S.A.

NORMAN W. RUNHAM
University College of North Wales
Bangor, United Kingdom

AM ELI E SCHELTEMA

Woods Hole Oceanographic Institution

Woods Hole, Massachusetts, U.S.A.

ALAN SOLEM

Field Museum of Natural History
Chicago, Illinois, U.S.A.

DAVID H. STANSBERY
Ohio State University
Columbus, Ohio, U.S.A.

FRED G. THOMPSON
University of Florida
Gainesville, Florida, U.S.A.

THOMAS E. THOMPSON
University of Bristol
Bristol, United Kingdom

NORMITSU WAT ABE
University of South Carolina
Columbia, South Carolina, U.S.A.

KARL M. WILBUR
Duke University

Durham, North Carolina, U.S.A.

THE AMERICAN MALACOLOGICAL BULLETIN (formerly the Bulletin of the American Malacological Union) is the official journal publication
of the American Malacological Union.

AMER. MALAC. BULL. SPECIAL EDITION NO. 2

June 1986

PREFACE

The Second International Corbicula Symposium was held in Little Rock, Arkansas,
June 21-24, 1983. The meeting was sponsored by and the majority of support for publica-
tion of these proceedings was provided by grants from the Electric Power Research In-
stitute, Palo Alto, California and the Nuclear Regulatory Commission, Washington D. C.
Arkansas Power and Light Company kindly acted as host of this meeting. The Symposium
Organizing Committee (Louise Russert Kraemer, Robert West, Robert McMahon, Jack Mat-
tice, Paul Hayes, and Joseph C. Britton) express our sincere gratitude to these organiza-
tions for their assistance.

The Little Rock meeting was for the specific purpose of stimulating dialogue be-
tween basic and applied scientists with respect to Corbicula biofouling and control. Addi-
tional aspects of the basic biology of Corbicula were also considered. About 40 papers
were presented by the participants, and several Corbicula and power plant “tutorials” were
provided in informal evening sessions. The last day of the meeting was highlighted by two
panel discussions, the first devoted to basic biology and the second devoted to biofouling
and control. (A summary of each panel discussion appeared in Volume 8(2) of The Cor-
bicula Newsletter.) The delivered papers, informal rap sessions and panel discussions fo-
cused on many topics, most of which are formalized in the papers which appear in these
proceedings. A few topics, however, have received only modest coverage herein, but are
likely to continue to be issues of Corbicula biology or control in the years to come. The
first, an issue of basic science, has to do with the number of species of Corbicula now
in North America. There is increasing biochemical evidence supporting the concept of two
species of Corbicula in North American fresh waters (e.g., see McLeod’s paper herein),
but ecological data presented at the meeting suggests that the situation may not be con-
clusively settled by biochemical evidence alone. Jerry Landye raised the question of a Cor-
bicula fishery at the Little Rock meeting. Since that time (and independent of his presenta-
tion) I have received numerous inquiries concerning Corbicula fisheries or aquaculture for
human or livestock consumption. This may be another direction future Corbicula work may
lead. Another issue, one of applied biology, addresses the most appropriate mechanism
for Corbicula control in industrial water supplies. Several papers herein discuss various
methods for Corbicula control. There seems to be an increasing feeling among the industrial
community that there is no single-most appropriate measure. Effective Corbicula control
is highly site-specific. What works at Site A may not be effective at Site B. Control efforts
must focus on the one or combination of measures most effective for the specific industrial
operation.

Most of you who attended the Little Rock Symposium have communicated to me
that you found the Symposium useful, intellectually stimulating, and a smoothly run meeting.
There are several reasons for this positive reaction, all of which are the results of attention
given by members of the Local Organizing Committee. Louise Russert Kraemer and Bob
West were perhaps the most visible, and each clocked hundreds of hours working on many

details that eventually culminated in an efficient and successful meeting. Nancy Rogers
of the University of Arkansas’ Division of Continuing Education must also be commended
on effective management of registration and pre-meeting preparations. Many staff members
of Arkansas Power and Light Company contributed time and energy before and at the
meeting to insure its success. To all of you go our sincere appreciation.

As with the first Corbicula Symposium Proceedings, manuscripts submitted for this
volume received peer review. All papers were read by at least two independent reviewers.
I would like to express my appreciation to each of the following persons who graciously
provided time and expertise to serve as reviewers:

John H. Balletto

Gerald L. Mackie

Harry F. Bernhard

Jack S. Mattice

Joseph C. Britton

Robert H. McMahon

Peter Calow

Brian Morton

Donald S. Cherry

Harold D. Murray

Billy G. Isom

Jeanne M. Potter

Larry Eng

Robert S. Prezant

L. Barry Goss

Elizabeth B. Rodgers

James J. Hall

W. D. Russell-Hunter

Phillip S. Hartman

W. D. Sheppard

Paul F. Hayes

James B. Sickel

William Heard

Michael H. Smith

Chester M. Himel

James A. Smithson

Joe T. Johnson

Richard Sparks

James E. Joy

David H. Stansbery

Victor S. Kennedy

Robert L. Thomas

Allen W. Knight

Robert West

Louise Russert Kraemer

David S. White

J. Jerry Landye

Lynn L. Wright

Roland A. Leathrum

I am especially indebted to Robert Prezant, who was willing to accept the formidible
task of Managing Editor for these Proceedings. Bob received the reviewed manuscripts,
sent them for additional reviews, negotiated with the printer, marked all manuscript copy
for printing, and, in essence, served as overseer in the transition from manuscript to printed
page. He has also served as the primary liason between the Symposium Organizing Com-
mittee and the American Malacological Union. Finally, I am appreciative to the Executive
Committee of the American Malacological Union, who agreed to publish these proceedings
under the AMU name, and thereby insured a distribution for the document significantly
greater than would have been possible by private publication.

Joseph C. Britton
February, 1985

HISTORICAL REVIEW OF ASIATIC CLAM ( CORBICULA ) INVASION AND
BIOFOULING OF WATERS AND INDUSTRIES IN THE AMERICAS

BILLY G. ISOM

TENNESSEE VALLEY AUTHORITY
MUSCLE SHOALS, ALABAMA 35660, U.S.A.

ABSTRACT

Severe biofouling problems with Asiatic clams ( Corbicula ) were first reported in the United States
in 1956. Clam infestations were reported in 1961 in irrigation canals, and numerous reports of clams
biofouling irrigation canals in the west have appeared in subsequent years.

The first verified excursion of Asiatic clams beyond the Rocky Mountain barrier occurred when
they were discovered in the Tennessee River in 1959. Since that time there have been numerous
reports of range extensions into other river systems, including headwater streams. Clams have caus-
ed severe biofouling problems in water intakes, pumps, and industrial and power plant cooling water
(heat rejection) systems.

The current range extension in the Americas includes 35 of the contiguous United States. The
range extension of Corbicula “leans" and “fluminae" to Argentina, in South America, was reported
in 1981.

This paper provides a historical review of Asiatic ciam
(Corbicula) invasion and nuisance problems in the Americas,
principally the United States. The First International Corbicula
Symposium (Britton etal., eds., 1979) provided considerable
information on the invasion and nuisance of Corbicula in North
America. Mattice (1979), Goss ef at. (1979), and especially
Mattice, Eng, and Collier (1979) identified resources on this
problem. Isom (1975-1982) provided an updated bibliography
on mollusks, including Corbicula. Ralph Sinclair initiated a
“Conb/'cu/aCommunication” (1971a), as did Fox, for exam-
ple, in 1970, and later the Corbicula Newsletter became an
additional source of information. McMahon (1982) provided
a comprehensive review of the introduction, invasion, and
spread of Corbicula in the United States.

Ingram (1956) was the first to develop a reporting
system to indicate mollusk nuisance encountered in the water
works field to assist the American Water Works Associations
task group on biological infestation of purified waters. Until
1956 the principal mollusk problem was with the “faucet
snail” Bythinia tentaculata (Linnaeus), a nuisance in some
water supplies (Ingram, 1956).

The macroinvertebate and mollusk nuisance problem
in water supplies is an old one (see Isom 1971). Kraepelin
(1885) reported macroinvertebrates inhabiting water pipes at
Hamburg. Whipple ef a/. (1927) cited incidents of biological
nuisance in Rotterdam, Hartford, Boston, and Brooklyn. The
National Electric Light Association (1926) reported restrictions
in flow due to vegetable and animal growths in conduits. Con-
trol used was periodic manual cleaning of condensers and
pipes and , in one instance, backwashing of condensers.

Clarke (1952), Feigina (1954), Ingram (1956-1959), Ray
(1962), Krishnamoorthi and Rajagopalan (1967), and Morton
(1975) recorded fauna associated with nuisances in drinking

water supplies. Feigina (1959), Gruver (1968), Goss and Cain
(1975), Goss ef a/. (1979), Isom (1971, 1 976), Kirpichenko ef
a/. (1962), Mattice (1979), Mikheev (1961), Sinclair (1964),
Sinclair and Isom (1961,1963), Van Benthem Jutting (1953),
and others have recognized nuisance mollusk problems in
steam electric station water systems, hydrotechnical installa-
tions, and various industries.

HISTORY OF CORBICULA INVASION
IN THE U.S.

Ingram (1959) first reported the potential Corbicula
nuisance problems with the prophetic statement that,
“...Asiatic clams are seen as potential pests and a threat of
continued nuisance...”. Ingram reviewed nuisance problems
to the date including Coachella Valley Water District’s water
supply transported 198 km by open canal from the Colorado
River at Imperial Dam. The water supply of the Metropolitan
Water District of Southern California reported clam infesta-
tions in 1958. Dundee and Dundee (1958) extended the
known range of Corbicula outside coastal areas of the west
and central California to an irrigation canal in Phoenix,
Maricopa County, Arizona, in June 1956. Prokopovich (1969)
reported that in 1952, only one year after completion of the
Delta-Mendota Canal, Corbicula caused serious operational
problems at the Tracy Pumping Station, Tracy, California.
Sinclair and Isom (1963) reviewed these early papers and
records. From 1 963-1970, Fox (1970) reported to the Western
Society of Malacoiogists the invasion of the clams in the west,
and in the east from published accounts.

As the clam nuisance spread continued in the western
area of the U.S., Corbicula was found downstream of Pickwick
Dam on the Tennessee River, km 327-332, October 21, 1959
(Sinclair and Ingram, 1961). Heard (1964) recorded the range

American Malacological Bulletin, Special Edition No. 2 (1 986): 1 -5

CORBICULA SYMPOSIUM

extension of Corbicula to Florida, and Sickel (1973) to areas
of Georgia. Other range records can be found in Sinclair
(1971). However, essentially none of these papers indicate
any problem with nuisance aspect of the Corbicula range ex-
tensions. Following the report of Fuller and Powell (1973) on
the finding of Corbicula in the Savannah, Pee Dee, and
Delaware River systems, there was a spate of complaints
about Corbicula nuisance in power plants and industries in
the Atlantic Coastal drainages, which were confirmed by
Foster and Box (1976). Coincidental with the spread of Cor-
bicula to the southeastern and Atlantic Coastal states was
its spread to the middle and northen Interior Basin. Thomer-
son and Myer (1970) and Eckblad (1975) reported range ex-
tension and power plant problems with Corbicula on the
Mississippi River at Granite City, Madison County, Illinois,
and Lansing, Iowa. Subsequintly, the author discussed and
visited with personnel of Commonwealth Edison Company
of Illinois about real and potential problems associated with
Corbicula in power plants throughout their system (personal
communications).

Corbicula spp. are now in the northern Mississippi River
drainage basin in the St. Croix River east of St. Paul and north
of Hastings, Minnesota (personal communication). Clarke
(1981) reported Corbicula in Lake Erie, and Scott-Wasilk
(1982) reported Corbicula in a sewage plant outfall area
“within the Bay Shore Station thermal plume,” Ohio. Scott-
Wasilk also reported that Corbicula occurred in the thermal
plume of the Monroe Power Plant, Michigan, on the western
shore of Lake Erie.

Sinclair (1971) and McMahon (1982) reviewed the
spread of Corbicula. Ituarte (1981) Reported the introduction
of Corbicula to the “Argentine” and South America which is
further indication of its adaptability.

ASIATIC CLAM NUISANCE PROBLEMS

If not the first, the most significant early problem with
Asiatic clams was in the Delta-Mendota Canal, a part of the
California Central Valley project, which is about 188 km long
(Anonymous, 1963; Prokopovich and Herbert, 1965). They
reported the number of live and dead clams in shallow
sediments (15-30 cm) was 25,000-65,000/0.092 m2 in some
places. The clams contributed significantly to the deposition
of 1 7,330 cubic meters of sediments in 48.2 km of the canal.
Prokopovich (1969) noted that not only was shell deposition
a problem, but also the fact that clams clasticized suspend-
ed materials in their mucus (forming sediment bars) which
was also a major contributor to the clam problem in the Delta-
Mendota Canal. Eng (1979) reported population dynamics of
Asiatic clams in this canal and also noted problems.

Sinclair and Isom (1 961 , 1 963) compiled and discussed
various aspects of Corbicula biology, spread, nuisance, con-
trol, benefits, and economic value in Phillipine and Asian
habitats. Their 1961 report relates the first nuisance incidents
with Corbicula in U.S. power plants and industries. They
reported that on the Tennessee River a large number of small
cooling-water pipes became sluggish with clams at several
electric power stations. Two wheelbarrow loads of clams were
removed from one condenser inlet water box and clams had

plugged the inlets on about half of the tubes. Most of the
clams were small, ranging from about 12.7-22.0 mm in size,
some clams occured throughout the raw water service and
the fire protection systems.

The only control measure instituted was flushing the
line at regular intervals, with future plans to chlorinate the
water as a control measure since it was the only practical
approach to the problem at that time.

Sinclair and Isom (1961, 1963) further reported the pro-
blems that sand and gravel companies had with Asiatic clams
in concrete aggretates. One sand and gravel company ex-
ecutive stated, “Seeing moving concrete can be unnerving”
(Sinclair and Isom, 1963). Problems with clams migrating to
the surface of poured concrete resulted in many gravel
“beds” being abandoned.

Since the publication by Sinclair and Isom (1963), I
have probably received 1,000 telephone calls or letters of in-
quiry concerning Corbicula nuisance problems. Only a few
of these inquiries have ultimately been reported in the
literature. Most inquires were from power companies and
engineering consulting firms trying to solve immediate pro-
blems. Other inquiries were from sand and gravel companies,
the Environmental Protection Agency (EPA), the Nuclear
Regulatory Commission (NRC), Department of the Interior
agencies, including the Bureau of Reclamation, chemical
companies and vendors, State agencies, individuals repor-
ting clam mortalities or nuisances, and individuals interested
in location, utilization, and suitability of clams for food or
aquaculture.

Corbicula problems have been widespread in in-
dustries and power plants over the years (Goss and Cain,
1 975; and Goss et ai, 1 979); however, with the shutdown of
Arkansas Nuclear One (ANO) on September 3, 1 980, due to
waterline clogging with Corbicula shells, the issue received
National attention. Following this shutdown, the Nuclear
Regulatory Commission (NRC) issued IE Bulletin 81-03;
“Flow Blockage of Cooling Water to Safety System Com-
ponents by Corbicula sp. (Asiatic Clam) and Mytilus sp.
(Mussel),” (Anonymous, 1981) which mandated licensees
determine if they had a real or potential problem with Asiatic
clams. This shutdown, and the resultant NRC bulletin, alerted
scientists and engineers about the problems and essentially
everyone in the power industry became cognizant of the
potential for Corbicula fouling. NRC noted that Corbicula or
Mytilus were significant to reactor safety”... because (1) the
fouling represented an actual common cause failure, i.e., in-
ability of safety systems redundant components to perform
their intended safety functions, and (2) the licensee was not
aware that safety components were fouled.” Mollusk
blockage problems were subsequently reported at Brunswick
and Sequoyah unit 1.

The information sought by IE Bulletin 81-03 surveys
should be very helpful to the industry when published by the
NRC. They will also provide information on the distribution
of Corbicula nationally in relation to nuclear power stations.

Initial testing for Corbicula as required by IE Bulletin
81-03 costs untold numbers of dollars. If one assumes an
outage cost of $50,000 per hour per 1,000 megawatt unit,

ISOM: HISTORICAL REVIEW OF CORBICULA IN AMERICA

times roughly 79 operative units in the U.S., plus 50-60 under
construction, these costs alone would be $3.5 million for the
first evaluation and perhaps $1 million in the plants under
construction. Need for continued re-evaluation of these plants
due to Corbicula nuisance points to the tremendous economic
burden caused by this clam. Other industries have similar
problems with Asiatic clams that power plants do, although
in most cases safety systems are not as critical. The com-
bined outages, reductions in efficiency, capital investment
in equipment, labor, and chemical control of Corbicula pro-
bably for exceed $1 billion annually in the U.S.

There are several unstated but common “threads”
among the papers reviewed that may be of interest: (1) In
most, if not all instances, problems have been the result of
clams growing in the plant/industry water system, since most
have water supplies that have screens with openings smaller
than clams found within the systems; (2) based on the few
studies that reported quantitative sampling of Asiatic clams,
the average “carrying capacity” of interior Basin streams over
a long period of time is 100-200 sexually mature clams/m2
(Bickel, 1966).

Based on 1 ,187 TVA samples from 1971-1976 in the
upper Tennessee River, the Corbicula population mean was
93.5/m2. Results from hundreds of samples from 1969-1979
revealed 131 Corbicula! m2 in the middle Tennessee River.
Sicke! and Chandler (1981) reported 200 Corbiculalm 2 from
the lower Cumberland River (km 106-113) in the spring of
1980. Gardner et al. (1976) found much higher mean den-
sities of Corbicula in the Altamaha River, Georgia, when sex-
ually immature clams were included in their samples.
However, average density of first year, second year, and third
year sexually mature Corbicula was about 70/m2 each in 1979
and 1975, or a total of 210/m2. When sexually immature clams
were included in the data, an average density of 9,257/m2
was reported in July 1974. Eng (1979) and Sickel (1979) also
reported high densities from California and Georgia, respec-
tively. Gottfried and Osborne (1982) reported 212 Cor-
biculalm2 for Wekiva River, Florida, 1966-1967.

Ingram et al. (1 964) reported 1 ,400-2,900 Corbiculalm2
from the “main canal” near Parker, Arizona, in May 1963,
only 16 months after the canal had been cleaned.

Data from Villadolid and Rosario (1930) on Corbicula
taken from tributaries to Laguna de Bay, Phillipines, indicated
densities from 45-235/m2 which appeared to decline with
altitude.

Rodgers et al. (1977) reported a mean population den-
sity of 18-29 Corbiculalm2 from the New River, Virginia, at
Glen Lyn which may have been low due to its recent inva-
sion or perhaps the relatively low alkalinity (39 mg/L as
CaC03) or lack of food, this should be a good reference river
for comparing with more eutrophic streams and canals. There
are other meager quantitative data, but they were unavailable
to the author as of this writing. However, I suggest that more
than 200 Corbiculalm2 may constitute a potential for severe
nuisance problems, 100-200/m2 moderate problems, and
fewer than 100/m2 indicates potential for minor problems.
These values should be derived from at least two years
preoperational and two years post operational data, or four

or more successive years of recorded data.

MEDIATION (METHODS) OF CORBICULA
SPREAD

Disagreements continue about how freshwater
mussels were/are dispersed. The abailable data on historical
records of LJnionidae should be applicable to analysis of
spread of Corbicula. Even though Unionidae have a fish
host(s), similar claims are being made as to their dispersal
as those for Corbicula. My evaluation of the role of natural
dispersion of Corbicula vs. distribution by man is as follows.

The role of birds should be discounted as a significant
dispersal factor in the spread of Unionidae or Corbicula.
Juvenile Unionidae have byssal attachment organs, as do
Corbicula, and would have been dispersed by birds or water-
fowl, as has been attributed to the spread of Corbicula
McMahon (1982). However, all historical evidence is to the
contrary. There are no records of endemic, Cumberlandian
Unionidae, for example, outside their known ranges. Not even
a relic shell record from contiguous watersheds has ever been
reported. If aquatic birds, or even aquatic mammals, were
a factor in mussel dispersal, it would seem logical to expect
at least a few relic shell records to document these events
from the thousands of years of opportunity.

McMahon (1982) raises the possibility of Corbicula be-
ing spread in “fish digestive tracts.” In my opinion, this
avenue of spread is without merit. Gut contents of several
fish species that had consumed Corbicula, included
freshwater catfish, drum, sunfish, and carp, have never
revealed even a shred of Corbicula tissue. This was the case
regardless of whether the shells were crushed or not. Thomp-
son and Sparks (1977) discounted the dispersal of Corbicula
by the intestinal tracts of migratory waterfowl.

Invasive Unionidae such as Anodonta were absent from
the long historical record of the Tennessee River and
tributaries as were a number of big river or “Ohioan” species
upstream of Muscle Shoals prior to impoundment. A number
of Anodonta and other genera now occur in impoundments
upstream and downstream of Muscle Shoals. Unionid species
have parasitic glochidia which are transported by fish. Cor-
bicula do not have this type distribution due to lack of a
parasitic life stage. Jenkinson (1979) concluded that Cor-
bicula, as a result of their “non-swimming, pelagic, veliger
larval stage,” are dispersed unidirectionally as a result of
water currents.

Corbicula did not get to North America from its native
Asiatic habitat by natural distributional means. Corbicula did
not cross the western mountain barrier and get into the Ten-
nessee River by natural means of dispersal. The author’s
observations since 1959 and the historical records for
Unionidae strongly suggest spread of Corbicula has been by
two, and only two methods, by human mediation and passive
dispersal by water currents. Early studies on Unionidae in-
dicated that locomotion is not a significant factor in bivalve
dispersal because of their highly random movements.

BENEFICIAL USES OF CORBICULA

In conclusion a short review of Asiatic clam benefits

CORBICULA SYMPOSIUM

are presented. These benefits range from the intangible such
as their use in aquaria and by sportsmen for fish bait, to their
local harvest and use as food for human consumption.

History of Corbicula for use as food for humans,
domestic animals, and wildlife is documented to the extent
that we can note their benefits, but not to the extent that we
can juxtapose their benefits against costs of their nuisance.

Villadolid and Rosario (1930) reported that Corbicula
manilensis from the Laguna de Bay and its tributaries in the
Phillipines were used extensively as a protein source for
domestic ducks, for native “laboring class” inhabitants, and
that in some cases shells were “burned” for the manufac-
ture of commercial air-slaked lime, this was also a practice
in China.

Miller and McClure (1931) reported on human con-
sumption of Corbicula manilensis known in the vernacular as
“Wong Sha Hin” (yellow sand clam). When found on/in mud
the clams were known as “Nai Hin” or “mud clams” which
were a dark color and had an inferior taste to yellow sand
clams.

Caution is noted about use of raw clams for human
food. Sinclair (1971) (from Cheng, 1964; Sandground and
Bohne, 1940; and Van Benthem Jutting, 1953) noted that in
its native range the clam is the intermediate host for
Echinostome trematodes. Clams should be well cooked
before eating, like pork and some fish, etc.

Gonzales and Bersamin (circa 1956) reported that Cor-
bicula were a better source of dietary calcium than either
oysters or clams (Paros) Soletellina elongata Lamarck.

Cahn (1951), Sinclair and Isom Sinclair (1963),
reported that in Japan Corbicula sandai was protected by law
which provided for a closed season and size limit. Cahn (1951)
also discussed the Corbicula japonica fishery (for a synopsis
see Sinclair and Isom, 1963, page 28).

Fox (1970) (see Mattice et al. , 1979, for other Fox cita-
tions) reported that from 1963-1968, 2,240,822 pounds of Cor-
bicula were sold in California for bait at a value of $234,448.
Other papers relating the value of Corbicula as bait include
Sickel et al. (1980), and Sickel and Chandler (1981). There
was a large Corbicula fish bait industry in the lower Tennessee
River prior to massive mortalities in the summer of 1977. Cor-
bicula were sold for about two cents each by collectors, and
four cents each by wholesalers.

There is certainly a lot of opportunity to exploit the Cor-
bicula resource commercially. Historically, exploitation of
“wild” populations has provided control to the extent of
overharvesting and in many cases, unfortunately, to the point
of extinction. Whether or not exploitation of Corbicula will con-
trol their nuisance will be one more interesting aspect of deal-
ing with this introduced bivalve.

ACKNOWLEDGEMENTS

Although this review is not all inclusive, the following persons
are recognized as contributing to the awareness of the Corbicula pro-
blems and providing insight into solutions of the nuisance of Cor-
bicula: William Marcus Ingram, Ralph M. Sinclair, Jack Mattice, Ralph
Olen Fox, and N. P. Prokopovich.

LITERATURE CITED

Anonymous. 1963. Little creatures clog big canals. The Reclama-
tion Era, November: 96-98.

Anonymous. 1981 . Flow blockage of cooling water to safety system
components by Corbicula sp. (Asiatic clam) and Mytilus sp.
(mussel). IE Bulletin 81-03. NRC. pp. 1-5.

Bickel, D. 1966. Ecology of Corbicula manilensis Philippi in the Ohio
River at Louisville, Kentucky. Sterkiana 23: 19-24.

Britton, J. C., J. S. Mattice, C. E. Murphy, and L. W. Newland, eds.,
1979. Proceedings, First International Corbicula Symposium.
Texas Christian University Research Foundation Publication,
Ft. Worth, J. C. Britton, et al., eds., 313 pp.

Cahn, A. H. 1951. Clam culture in Japan. Reprinted as U. S. Fish
and Wildlife Service Fishery Leaflet No. 383. 80 pp.

Cheng, T. C. 1964. The Biology of Animal Parasites. W. B. Sanders
Company, Philadelphia. 727 pp.

Clarke, A. H. 1981. Corbicula fluminea in Lake Erie. Nautilus
95:83-84.

Clarke, K. B. 1952. The infestation of waterworks by Dreissena
polymorpha, a freshwater mussel . Journal of Institution of Water
Engineers, 6:370-378.

Dundee, D. S. and H. A. Dundee, 1958. Extensions of known ranges
of four mollusks. Nautilus 72(2):51-53.

Eckblad, J. W. 1975. The Asian clam Corbicula in the upper Mississip-
pi River. Nautilus 89:4.

Eng, L. L. 1979. Population dynamics of the Asiatic clam, Corbicula
fluminea (Muller), in the concrete-lined Delta-Mendota canal
of Central California. IN: Proceedings First International Cor-
bicula Symposium. J. C. Britton, et al., eds. pp. 39-68. Texas
Christian University.

Feigina, Z. S. 1984. Control of biological fouling of waterworks.
Gorodgkee Khoziaistro Moskvy, Moscow 3:26-28.

Feigina, Z. S. 1959. Control of Dreissena fouling by heated water in
thermal power plants. Elektvicheskie Stantsii, No. 10.

Foster, R. B. and Box, S. W. 1976. Procedures for evaluating
chemical control of larval Asiatic clams. Unpublished report
presented at the 24th Annual Meeting of the North American
Benthologicai Society, 6 p. 2 figs.

Fox, R. 0. 1970. The Corbicula story: Chapter Two. Read at the Third
Annual Meeting of the Western Society of Malacologists, Stan-
ford University. 10 pp.

Fuller, S. L. H. and Powell, C. E., Jr. 1973. Range extension of Cor-
bicula manilensis (Philippi) in the Atlantic drainage of the United
States. Nautilus 87(1): 59.

Gardner, J. A., Jr.; Woodall, W. R.; Staats, A. A.; and Napoli, J. F.
1976. The invasion of the Asiatic clam (Corbicula manilensis
Philippi) in the Altamaha River, Georgia. Nautilus
90(3):1 17-125.

Gonzales, O. N. and Bersamin, S. V. Circa 1956. Tulya (Corbicula
manilensis Philippi), Paros (Soletellina elongata Lamarck), and
oysters as good sources of available calcium. Bureau of
Fisheries, Manila. 6 pp.

Goss, L. B., J. M. Jackson, H. B. Flora, B. G. Isom, C. Gooch, S.
A. Murray, C. G. Burton, and W. S. Bain, 1979. Control studies
on Corbicula for steam-electric generating plants. IN: Pro-
ceedings, First International Corbicula Symposium. J. C. Brit-
ton, et al., eds. pp. 139-151. Texas Christian University.

Goss, L. B. and C. Cain Jr., 1975. Power plant and condenser ser-
vice water system fouling by Corbicula, the Asiatic clam. IN: Bio-
fouling Control Procedures Technology and Ecological Effects,
L. D. Jensen, ed., Marcel Dekker Publ., pp. 11-17.

Gottfried, P. A. and Osborne, J. A. 1982. Distribution, abundance,

ISOM: HISTORICAL REVIEW OF CORBICULA IN AMERICA

and size of Corbicula manilensis (Phiiippi) in a spring-fed con-
trol Florida stream. Scientist. 45(3): 178-1 88.

Gruver, M. L. 1968. Asiatic clams - A new freshwater pest. Iron and
Steel Engineer, October 91-94.

Heard, W. H. 1964. Corbicula fluminea in Florida. Nautilus 77:105-107.

Ingram, W. M. 1956. Snail and clam infestations of drinking water
supplies. Journal of the American Water Works Association
48(3):258-268.

Ingram, W. M. 1959. Asiatic clams as potential pests in California
water supplies. Journal of the American Water Works Associa-
tion 51(3):363-370.

Ingram, W. M., L. Keup, and C. Henderson, 1964. Asiatic clams at
Parker, Arizona. Nautilus 77(4): 121 -124.

Isom, B. G. 1971. Evaluation and control of macroinvertebrate
nuisance organisms in freshwater industrial supply system.
Presented at the 19th Annual Meeting of the Midwest (N.A.)
Benthological Society 13 pp. (8 of text).

Isom, B. G. 1976. Biofouling - state-of-the-art in controlling Asiatic
clams (Corbicula manilensis Philippi) and other nuisance
organisms at power plants. Presented at the 24th Annual
Meeting of the North American Benthological Society. 1 3 pp.

Isom, B. G. 1975-1982. Mollusca. In: Current and Selected
Bibiliographies on Benthic Biology. Published annually for
members of the North American Benthological Soceity. D. W.
Webb, Illinois Natural History Survey, Chairman.

Ituarte, C. F. 1981. Primera noticia acerca de la Introduction de
pelecipodos Asiaticos en el area Rio Platense. Neotropica
27(77):79-82.

Jenkinson, J. J. 1979. The occurrence and spread of Corbicula
manilensis in East-Central Alabama. Nautilus 93(4): 149-1 53.

Kirpichenko, M. Y., V. P. Mikheev, and E. P. Shtern, 1962. Control
of Dreissena polymorpha fouling at hydroelectric power sta-
tions. Elektr Station 5:30-32.

Kraepelin, K. 1885. Fauna in the Hamburg water system. Abhandl.
Naturn. Ver. Hamburg. 9:1 (from Ingram).

Krishnamoorthi, K. P. and Rajagopalan, S. 1967. Survey of mollusc
nuisance in some water supplies of Calcutta. Advance
abstracts of Contributions on Fish and Aquatic Sciences in
India 1(3):67-68.

Mattice, J. S. 1979. Interactions of Corbicula sp. with power plants
IN: Proceedings, First International Corbicula Symposium,
Texas Christian University, J. C. Britton, ef a/., eds. pp.
119-138.

Mattice, J. S., L. L. Eng, and B. N. Collier, 1979. Corbicula 1979:
A Bibliography. IN: Proceedings, First International Corbicula
Symposium, Texas Christian University, J. C. Britton, ef a/., eds.
pp. 289-313.

McMahon, R. F. 1982. The occurrence and spread of the introduc-
ed Asiatic freshwater clam, Corbicula fluminea (Muller), in
North America: 1924-1982. Nautilus 96(4): 134-1 41.

Mikheev, V. P. 1 961 . Experiments on destroying Dreissena polymor-
pha by heating the water. Bjul. Inst. Biol. Vodochr. 11:10-12.

Miller, R. C. and McClure, F. A. 1931. The fresh-water clam industry
of the Pearl River. Lingnan Science Journal, Canton. All in
English. 10(2-3):307-332 and plates 41-50.

Morton, B. 1975. the colonization of Hong Kong’s raw water supply
system by Limnoperna fortune i (Dunker 1857) (Biralria:
Mytilacea) from China. Malacologies I Review 8:91-105.

National Electric Light Association. 1926. Restriction in flow due to
vegetable and animal growths in conduits. Proceedings of the
49th Convention. 83:801-803.

Prokopovich, N. P. 1969. Deposition of clastic sediments by clams.
Journal of Sedimentary Petrology 39(3):891-901.

Prokopovich, N. P. and Herbert, D. J. 1965. Sedimentation in the
Delta-Mendota Canal. Journal of the American Water Works
Association 57:375-382.

Ray, H. C. 1 962. Chokage of filtered-water pipe systems by freshwater
molluscs. Proceedings, First All-India Congress of Zoology. Part
2. Scientific Papers: pp. 20-23, 3 figs.

Rodgers, J. H., Jr., D. S. Cherry, J. R. Clark, K. L. Dickson, and J.
Cairns, Jr. 1977. The invasion of Asiatic clam, Corbicula
manilensis, in the New River, Virginia. Nautilus 91(2):43-46.

Scott-Wasilk, J. L. 1982. Evaluation of the Asiatic clam, Corbicula
fluminea, in the western basin of Lake Erie. Davis-Besse Unit
No. 1, Annual Environmental Operating Report, January
1 -December 31, 1981. Also, 1982 Annual Meeting of the North
American Benthological Society.

Sickel, J. B. 1973. A new record of Corbicula manilensis (Philippi)
in the Southern Atlantic Slope Region of Georgia. Nautilus
87(1 ): 1 1-12.

Sickel, J. B. 1979. Population dynamics of Corbicula in the Altamaha
River, Georgia. IN: Proceedings First International Corbicula
Symposium, Texas Christian University, J. C. Britton, ef a/., eds.
pp. 69-80.

Sickel, J. B., D. W. Johnson, G. T. Rice, M. Heyn, and P. Wellner,
1980. Asiatic clam and commerical fishery evaluation. Ken-
tucky Federal Aid Project No. 2-344-R-1 and NOAA, Depart-
ment of Biological Sciences. Murray State University, Mur-
ray, KY. 43 pp.

Sickel, J. B. and C. C. Chandler, 1981. Commercial mussel and
Asiatic clam fishery evaluation. Kentucky Project No. 2-367-R
and NOAA, Department of Biological Science, Murray State
University, Murray, KY. 77 pp.

Sinclair, R. M. 1964. Clam pests in Tennessee water supplies. Journal
of the American Water Works Association 56(5):592-599.

Sinclair, R. M. 1971a. Corbicula Communication. September 22, EPA,
Cincinnati, 3 pp.

Sinclair, R. M. 1971. Annotated bibliography on the exotic bivalve
Corbicula in North America, 1900-1971. Sterkiana 43:11-18.

Sinclair, R. M. and W. M. Ingram. 1961. A new record for the Asiatic
clam in the United States, the Tennessee River. Nautilus
74(3):1 14-118.

Sinclair, R. M. and B. G. Isom. 1961. A preliminary report on the
introduced Asiatic clam Corbicula in Tennessee, Tennessee
Stream Pollution Control Board. Tennessee Department of
Public Health. 31 pp.

Sinclair, R. M. and B. G. Isom. 1963. Further studies on the intro-
duced Asiatic clam Corbicula in Tennessee. Tennessee
Stream Pollution Control board. 76 pp.

Thomerson, J . E. and D. G. Myer. 1970. Corbicula manilensis: Range
extension upstream in the Mississippi River. Sterkiana 37:29.

Thompson, C. M. and Sparks, R. E. 1977. The Asiatic clam Corbicula
manilensis in the Illinois River. Nautilus 91 :34-36.

Van Benthem Jutting, W.S.S. 1953. Systematic studies on the non-
marine mollusca of the Indo-Australian Archipelago. IV. Critical
revision of the fresh-water bivalves of Java. Treubia 22: 1 9-73,
22 figs.

Villadolid, D. V. and Del Rosario, F. G. 1930. Some studies on the
biology of Tulla (Corbicula manilensis Philippi), a common food
clam of Laguna de Bay and its tributaries. The Phillippine
Agriculturist 19(6):355-382, with 1 plate, 7 charts, and 2 text
figs.

Whipple, G. C., G. M. Fair, and M. C. Whipple. 1927. The Microscopy
of Drinking Water. John Wiley and Sons, New York. 585 pp.,
19 Plates.

THE ZOOGEOGRAPHY AND HISTORY OF THE INVASION
OF THE UNITED STATES BY CORBICULA FLUMINEA
(BIVALVIA: CORBICULIDAE)’

CLEMENT L. COUNTS, III
COLLEGE OF MARINE STUDIES
UNIVERSITY OF DELAWARE
LEWES, DELAWARE 19958, U.S.A.

ABSTRACT

A survey of the collections of Corbicula fluminea housed in 26 museums was collated with distribu-
tional information from state agencies, private collections, and available literature to determine the
present zoogeographic distribution and chronology of the invasion of these exotic bivalves in the United
States waters. Results revealed C. fluminea is presently found in 33 states. An analysis of the historical
zoogeography of C. fluminea in the United States indicates that man is the principal agent of its dispersal
into new drainage systems and that no large-scale geographic features act as a significant barrier
to dispersal. Two long-distance dispersal events have occurred in the United States; from the western
states to the Ohio River in 1957, and an infestation in the Escambia River, Florida, in 1960. Accounts
of the zoogeography of C. fluminea in each affected state are presented, as is a detailed chronology
of the invasion of the United States.

Bivalves in the genus Corbicula Muhlfeld , 1844 were
introduced into North America sometime during or before the
1920’s (Counts, 1981a). Our earliest record for bivalves in
the genus is at Nanaimo, Vancouver Island, British Colum-
bia and was collected in 1924 (Counts, 1981a) The first col-
lection of C. fluminea (Muller), 1774) in United States was
made along the banks of the Columbia River near Knapp-
ton, Pacific County, Washington in 1938 (Burch, 1944). Since
that first discovery in United States waters, C. fluminea has
invaded nearly every major river system of the country and
now threatens the Great Lakes (Clarke, 1981).

Several investigators have published summaries con-
cerning the spread of Corbicula fluminea in the United States.
Fox (1969-1971) presented yearly updates on its spread in
which he gave new stream and state records for the year of
the report. Sinclair (1961) published an annotated
bibliography on C. fluminea in the United States which also
described, in a general way, the invasion of these bivalves.
Sinclair and Isom (1961 , 1 963) described the invasion of the
Tennessee River system, and Dundee (1974) reported the
zoogeography of C. fluminea in the United States using litera-
ture reports as well as some museum records. Britton and
Morton (1979) discussed the systematics of C. fluminea in the
United States and gave some zoogeographic data chiefly in
the form of records from the Texas Christian University
Museum. However, their emphasis was almost entirely

’University of Delaware College of Marine Studies Contribution No. 171

systematic rather than zoogeographic.

Much of the published literature on Corbicula fluminea
in the United States is comprised of new locality reports, i.e.
new infestations of streams, lakes, or industrial water facilities.
The only synthesis of these reports is that of McMahon (1982)
in which he attempted to chronicle the invasion and subse-
quent spread of C. fluminea in the United States and draw
conclusions about the manner by which it has successfully
infested North American waters. However, his description
lacked many literature reports and did not include the exten-
sive resources of the malacological collections held in
zoological museums of the United States.

The present paper describes the zoogeography of Cor-
bicula fluminea in the United States and reports on the
chronology of the invasion using both distributional literature
and museum collection records.

TAXONOMY

There is some debate as to how many species of
bivalves in the genus Corbicula are currently present in the
United States. Corbiculid clams in this country have been
reported using the taxa Corbicula fluminea, Corbicula leana
Prime, 1864, Corbicula manilensis (Philippi, 1844), and Cor-
bicula sinensis nomen dubium. While McLeod and Sailstad
(1980) and Hillis and Patton (1982) have presented elec-
trophoretic evidence for two species, their studies are as yet
inconclusive. Thus, I will use the taxon C. fluminea, sensu

American Malacological Bulletin, Special Edition No. 2(1986): 7-39

CORBICULA SYMPOSIUM

Britton and Morton (1979), for all populations of corbiculid
bivalves reported in this study.

MATERIALS AND METHODS

Zoogeographic records for Corbicula fluminea were ob-
tained from the malacological collections of the museums
listed below. Also included are records provided by several
state environmental agencies. An acronym for each institu-
tion or agency, as it appears in the results section of this
paper, are given in parentheses. The museums are: the
Academy of Natural Sciences, Philadelphia (ANSP); the
California Academy of Sciences, San Francisco (CAS); the
Cincinnati Museum of Natural History (CMNH); the Dallas
Museum of Natural History (MNHD); the Delaware Museum
of Natural History, Greenville (DMNH); the Field Museum of
Natural History, Chicago (FMNH); the Florida State Museum,
University of Florida, Gainesville (FSM); the Fort Worth
Museum of Science and History (FWM); the Houston Museum
of Natural Science (HMNS); the Illinois State Museum,
Springfield (ISM); the Marshall University Malacological Col-
lection, Huntington, West Virginia (MUMC); the Milwaukee
Public Museum (MPM); the Mississippi Museum of Natural
Science, Jackson (MMNS); the Museum of Comparative
Zoology, Harvard University, Cambridge, Massachusetts
(MCZ); the Museum of Northern Arizona, Flagstaff (MNA);
the National Museums of Natural Science, Ottawa, Canada
(NMNS); the North Carolina State Museum of Natural History,
Raleigh (NCSM); the Ohio State University Museum of
Zoology, Columbus (OSUM); the Peabody Museum of Natural
History, Yale University, New Haven, Connecticutt (PMNH);
the San Diego Museum of Natural History (SDMNH); the San-
ta Barbara Museum of Natural History (SBMNH); the Stan-
ford University Museum, Stanford, California (SU); the State
Biological Survey of Kansas, Lawrence (SBSK); the Thomas
Burke Memorial Washington State Museum, University of
Washington, Seattle (TBWSM), the United States National
Museum of Natural History (USNM); the University of Cin-
cinnati Geological Museum (UCGM); and the University of
Oklahoma Museum, Norman (UOM). Museum records from
the Texas Christian University Museum (TCU) were published
by Britton and Morton (1979).

Other distributional records were provided by C. Dale
Snow, Oregon Department of Fish and Wildlife (ODFW), Alan
C. Buchanan and Ron D. Oesch, Missouri Department of
Conservation (MDC), and Robert Singleton, Arkansas Depart-
ment of Pollution Control and Ecology (ADPC). Gary A.
Coovert (GAC), Dayton, Ohio and David Metty (DM), Cincin-
nati, Ohio generously provided records from their private col-
lections. Constance E. Boone (CEB) of HMNS provided ad-
ditional information on Texas populations. R. Tucker Abbott
(RTA) and Robert Bullock (RB), University of Rhode Island,
and Russell Jensen (RJ) provided Florida records. Richard
L. Reeder, University of Tulsa, provided Oklahoma records
and Steven L. Coon (SLC) and James J. Hall (JJH) provided
records for southern California and North and South Carolina,
respectively. Stavros Howe (SH) provided information on
Delaware populations.

All information concerning locations of populations of
Corbicula fluminea gathered from these sources was collated
with the available zoogeographic literature and placed in a
computerized data base at the University of Delaware.
Categories of encoded information were month and year of
collection or first report of the population, body of water in-
fested, state, county, nearest city to population site, locations
and catalog numbers of museum specimens, literature report
citations, and latitude and longitude. Records were then
sorted by state, year, and major drainage.

All records were assembled into individual state ac-
counts and, the zoogeography of the species was plotted on
base maps for each state. Year records were used to plot
chronologic maps of the invasion of the United States waters.

RESULTS

The review of museum records and literature reports
revealed that Corbicula fluminea now inhabits the waters of
33 states. An account of the zoogeography of these bivalves
in individual states is presented below. Following the name
of each body of water infested by C. fluminea is the acronym
for the record source or literature citation.

ALABAMA

(Fig. 1)

Corbicula fluminea was first reported in Alabama in the
Mobile River in 1962 (Hubricht, 1963). In the fall of 1964 it
was found in the Alabama River (Hubricht, 1965) and the
following year it appeared in the Cahaba and Tombigbee
rivers and in Sucanochee Creek (Hubricht, 1966).

Shealy (1966) reported predation of Corbicula fluminea
by the Alabama map turtle, Graptemys puichra Baur, in the
Conecuh River of the Escambia River system. This preda-
tion did not significantly reduce the C. fluminea population
in that stream.

Jenkinson (1979) reported the occurrence of Corbicula
fluminea in the main streams and tributaries of the Chat-
tahoochee and Tallapoosa rivers in east-central Alabama in
1973. He noted that the population in Saugahatchee Creek,
a tributary of the Tallapoosa River, is apparently thriving in
waters receiving effluents from a textile mill, a sewage treat-
ment plant, and runoff from the experimental ponds of Auburn
University. C. fluminea was usually found in clay and sand
substrata in these streams.

Corbicula fluminea has been found in the following waters
of Alabama: Alabama River (MCZ; USNM; Hubricht, 1966), Big Cedar
Creek (OSUM), Big Nance Creek (OSUM), Black Warrior River
(FMNH; MCZ; OSUM; TCU), Buck Creek (OSUM), Burnt Corn Creek
(OSUM), Cahaba River (OSUM; Hubricht, 1966), Cedar Creek
(OSUM), Chattahoochee River (Jenkinson, 1979), Choctawhahatchee
River (FSM), Conecuh River (OSUM; UF; Shealy, 1976), Coosa River
(DMNH; NMNS; OSUM; USNM), Cypress Creek (USNM), Dauphin
Island (USNM), Drivers Branch (FSM), Elk River (FSM), Escambia
River (Hubricht, 1963), Flint River (FSM), Gantt Lake (FSM), Indian
Creek (OSUM), Limestone Creek (OSUM), Little Cypress Creek
(OSUM), Little Uchee Creek (OSUM; Jenkinson, 1979), Locust Fork
(NMNS), Mobile River (Hubricht, 1966), Mud Creek (FSM), Murder

COUNTS: CORBICULA FLUMINEA IN THE UNITED STATES

Figs. 1 - 6. Zoogeographic distribution of Corbicula fluminea in Alabama (1), Arizona (2), Arkansas (3), California (4), Florida (5), and Georgia
(6). Scale bar = 50 km.

CORBICULA SYMPOSIUM

Creek (OSUM), Neely Henry Lake (Britton and Morton, 1979), North
River (OSUM), Okatuppa Creek (FSM), Paint Rock River (DMNH;
FSM; MCZ; OSUM; USNM), Pea River (FSM), Peckerwood Creek
(OSUM), Piney Creek (OSUM), Santa Bouge Creek (FSM), Sauga-
hatchee Creek (Jenkinson, 1979), Second Creek (OSUM), Sucarno-
chee Creek (Hubricht, 1966), Sepulga River (FSM), Tallapoosa River
(Jenkinson, 1979), Tennessee River (ISM; NMNS; OSUM), Terrapin
Creek (OSUM), Tombigbee River (MCZ; OSUM; USNM; Hubricht,
1963, 1966), Town Creek (OSUM), Tubbs Creek (FMNH), and Uchee
Creek (Jenkinson, 1979).

ARIZONA
(Fig. 2)

Dundee and Dundee (1958) made the first report of
Corbicula fluminea in Arizona from collections made at
Papago Park, Phoenix in 1956. Keup et al. (1963) later
reported C. fluminea from Main Canal near Parker.

Minickley et al. (1970) found that Corbicula fluminea
was a food of three species of buffalo fishes (Ictiobus bubaius
/Rafinesque], Ictiobus cyprinellus /Valenciennes] and Ictiobus
niger [ Rafinesque/j in Apache and Roosevelt lakes in cen-
tral Arizona. Veligers, juveniles, and small adults were con-
sumed by these fishes. Rinne (1974) noted that the highest
densities of C. fluminea in these lakes occurred on rock rub-
ble slopes and increased directly with the complexity
(numbers of components) of the substrata. He also noted that
numbers of these clams increased with depth and position
downlake from inflow areas with high turbidity.

Bequaert and Miller (1973) reported Corbicula fluminea
to be common in the Colorado, Gila, and Verde rivers by 1972.
Dundee (1974) also reported C. fluminea from Lake Meade
in the Colorado River drainage.

Corbicula fluminea reported from the following Arizona waters:
Agua Fria River (Bequaert and Miller, 1973), Colorado River (ANSP;
NMNS; Bequaert and Miller, 1 973), Gila River (Bequaert and Miller,
1973), Lake Martinez (SBMNH; SDMNH), Salt River (CAS; SDMNH;
Dundee and Dundee, 1958), Verde River (DMNH; MNA), and several
irrigation systems (MCZ).

ARKANSAS
(Fig. 3)

Fox (1970a) reported Corbicula fluminea from the St.
Francis River in the northeastern portion of the state where
densities of these clams reached 21 /yd2. He also reported
C. fluminea from the lower Ouachita River in the southwestern
portion of the state. These were the first published reports
of C. fluminea in Arkansas.

Kraemer (1976) found Corbicula fluminea in the Arkan-
sas River at river mile 43. Kraemer (1977) later reported these
bivalves at river mile 171 and said that they were ubiquitous
from below Ft. Smith downstream to Lock and Dam No. 3.
Kramer (1976) believed it is unlikely that C. fluminea invaded
the Arkansas River before the mid-1960’s.

Corbicula fluminea has also been reported from the
Buffalo River (Kraemer, 1978, 1979), Chamagnoll Creek and
the Caddo River (Britton and Morton, 1979) as well as from
the White River (Kraemer, 1980).

Corbicula fluminea is reported from the following Arkansas
waters: Arkansas River (Kraemer, 1977), Bayou Bartholomew

(ADPC), Black River (MCZ; OSUM), Boeuf River (ADPC; OSUM),
Buffalo River (ADPC; OSUM), Caddo River (ADPC; TCU),
Chamagnoll Creek (TCU), Coon Bayou (OSUM), LaGrue Bayou
(FSM), L’Anguille River (OSUM), Little River (MCZ), Madison -
Mariana Diversion Canal (OSUM), Maniece Bayou (OSUM), McKin-
ney Bayou (OSUM), Mississippi River (OSUM), Ouachita River
(ADPC; FSM; OSUM; TCU), Red River (OSUM), Saline River (FSM),
St. Francis River (OSUM, Spring River (OSUM), Strawberry River
(ADPC; OSUM), and White River (FSM; OSUM; USNM; Kraemer,
1980).

CALIFORNIA
(Fig. 4)

Corbicula fluminea was first discovered in California in the
Sacramento River north of Pittsburg in 1945 (Hanna, 1966).
It apparently spread from this region to foul many of the canals
surrounding the San Francisco Bay estuary and the Central
Valley. It has been reported from Mayberry Cut and the Con-
tra Costa Cana! (Ingram, 1959), and the South Bay Aqueduct
(Morgester, 1967).

Invasion of the Delta-Mendota Canal in the Central
Valley has been well documented, prokopovich and Hebert
(1964) and Prokopovich (1969, 1970) noted that an increas-
ed deposit of clastic sediments was attributable to these
bivalves binding susupended sediments with mucus and in
their feces as well as decreasing water flow as a result of
mounds of dead shells.

Eng (1975) reported on the biology of Corbicula
fluminea in the Delta-Mendota Canal in a detailed study for
the California Academy of Sciences. Eng (1976) also found
the oligochaete, Chaetogaster limnaei von Baer, 1827, living
and feeding on the lateral grooves of the gills of C. fluminea
from the Delta-Mendota Canal. Of those clams exaimed in
March 1974, 87% were infested. Less than 3% were infested
in other months. These oligochaetes were believed to be com-
mensal. In a later paper, Eng (1979) discussed the popula-
tion dynamics and growth of these bivalves in the Delta-
Mendota Canal. Behrens (1975) reported the survival and
growth of these bivalves in beverage containers and, in one
instance, a locked fishing tackle box, taken from the canal.

Sigfried etal. (1980) discussed the occurrence of Cor-
bicula fluminea in the San Francisco Bay estuary in 1976
when conditions were dry. C. fluminea dominated the ben-
thos with the bivalve Macoma balthica (L. 1758) the am-
phipods Corphium stimpsoni and C. spinicoine, nematodes,
and a spinonid polychaete, Boccardia ligerica. They deter-
mined that the factors controlling the size and composition
of the benthos were salinity and sediment composition. They
further reported that the population of C. fluminea peaked in
March and noted that previous studies (Fisk and Doyle, 1962;
Hazel and Kelley, 1966) showed population peaks in January.
Their estimates of population densities for the estuary ranged
from 2,000 rrr2 to 14,500 nr2 as compared with the 312
nr2 estimate of Fisk and Doyle (1962) for 1960-1961.

Carlton (1973) discovered Corbicula fluminea in Stow
Lake, Golden Gate Park, San Francisco. This represented
the first finding of these bivalves on the peninsula.

Corbicula fluminea was discovered in the Imperial
Valley of southern California in 1953 (Fitch, 1953). It has since

COUNTS: CORBICULA FLUMINEA IN THE UNITED STATES

fouled many of the major aqueducts and reservoirs in this
portion of the state including the Colorado Aqueduct (Fox,
1970a), the Los Angeles Aqueduct (Fox, 1972), the San Jacin-
to Reservoir (Fox, 1972), Lake Jennings (Richardson et al.,

1970) , and Lake Matthews (Fox, 1972). The Coachella Water
District (Fitch, 1953) and the Gene and Mayfield pumping
plants of the Colorado Aqueduct have also experienced
biofouling by C. fluminea. Fox (1970b) and Bequaert and Miller
(1973) reported that the bivalve had infested the Salton Sea
basin and had crossed into the Baja of Mexico.

Richardson et ai. (1970) noted that the blue catfish,
Ictalurus furcatus (LeSueur) commonly fed on Corbicula
fluminea in Lake Jennings. They did not believe, however,
that /. furcatus would appreciably limit the size of C. fluminea
populations where the two species are found together.

Corbicula fluminea is found in the following California waters:
Alamao Canal (SDMNH), All American Canal (CAS), Anaheim Bay
(SBMNH), Coachella Valley Water District (CAS; FMNH), Cahuma
Lake (SBMNH), Colorado Aqueduct (DMNH; USNM), Columbia River
(CAS), Delta-Mendota Canal (DMNH; TCU; Eng, 1979; Prokopovich,
1969), Dyer Canal (Morgester, 1967), El Capitan Reservoir (Fast,

1971) , Evans Lake (ANSP; DMNH), Lake Casitas (SLC), Lake Jen-
nings (Richardson et al., 1970), Lake Murray (TCU), Lake Piru (SLC),
Livermore Canal (Morgester, 1967), Mayberry Cut (CAS: Ingram,
1959), Merced River (SBSK), Mokelumne Aqueduct (CAS),
Mokelemne River (PMNH, SDMNH, SU), Owens River (Fox, 1972),
Potatoe Slough (CAS), Russian River (CAS), Sacramento River
(ANSP, CAS, GAC, TCU, USNM; Hanna, 1962, 1966), Salinas River
(TCU), Salton Sea (CAS, SBMNH), San Diego City Water Works
(SBMNH), San Francisco Bay (CAS), San Jacinto Reservoir (Fox,
1970a), San Joaquin River (ANSP, CAS, SBMNH, SU), San Luis
Reservoir (TCU), Santa Ana River (ANSP), Santa Barbara Harbor
(SBMNH), Shasta Lake (CAS), South Bay Aqueduct (Prokopovich,
1968), Stanislaus River (OSUM), Stow Lake (Carlton, 1973), Tolumne
River (CAS, Ingram, 1959).

DELAWARE

Although Corbicula fluminea has been reported on the
Delmarva peninsula from states surrounding Delaware (Stotts
et al., 1977; Counts, 1981b) no reports of the species within
the geopolitical borders of the state have yet appeared in the
literature. The present report represents the first published
record of C. fluminea in Delaware.

Nanticoke River (SH).

FLORIDA
(Fig. 5)

The earliest reports of Corbicula fluminea in Florida are
those of Heard (1964, 1966). Heard (1964) reported C.
fluminea in the Apalachicola, Chipola, and Withlacoochee
rivers. He later hypothesized that upon the completion of the
Cross-Florida Barge Canal, C. fluminea could move across the
state to inhabit the St. Johns River system (Heard, 1966). This
hypothesis has since been proved correct. Gifford (1974)
found C. fluminea in the Cross-Florida Barge Canal and
reported that these bivalves reached a biomass of 500 g live
tissue/m2.

Schneider (1967) found Corbicula fluminea in the

Escambia River during a survey by the Florida State Board
of Health in 1960. He believed these bivalves invaded the
Escambia River sometime in 1957. Corbicula fluminea was
found in the Apalachicola River, near the Chattahoochee
River in 1961. By 1967, C. fluminea ranged from the Escam-
bia River near Century to the Withlacoochee Inglis.

Clench (1970) found that Corbicula fluminea’s range
extended into the Caloosahatchee River - Lake Okeechobee
system. He hypothesized that their presence in the system
may lead to the spread of C. fluminea throughout all of
southern Florida.

Deaton (1981) found Corbicula fluminea in the
Ochlocknee River in waters of about 0.1 ppt S and in associa-
tion with the native unionid bivalve Lampsilis claibornensis
(Lea, 1938).

Corbicula fluminea has been found in the following Florida
waters: Apalachicola River (DMNH; FSM; OSUM; Heard, 1964;
Schneider, 1967), Aucilla River (OSUM), Calossahatchee River
(DMNH; OSUM; Clench, 1970), Chipola River (FMNH; FSM; Heard,
1964), Chochtawahatchee River (FSM), Cypress Creek Canal
(OSUM), Escambia River (OSUM; Schneider, 1967), Ft. Lauderdale
Canal (USNM), Grassy Lake (OSUM), Holmes Creek (FSM),
Ichetucknee River (FSM), Indian Prairie Canal (DMNH), Kissimmee
River (FSM), Lake Buena Vista (RJ), Lake Hippochee (MCZ; OSUM;
Clench, 1970), Lake Jackson (OSUM), Lake Lucy (FSM), Lake
Okeechobee (ANSP; DMNH; FSM; MNHD; OSUM), Lake Oklawaha
(FSM), Lake Palatlakaha (FSM), Lake Talquin (FSM), Lake Tsala
(ANSP), Main Canal (FSM), Mayakka River (OSUM), Middle River
Canal (OSUM), Mosquito Creek (USNM), North Mosquito Creek
(OSUM), Ochlocknee River (FSM; OSUM; Heard, 1966), Oklawaha
River (FSM; OSUM; USNM), Rocky Creek (FSM), St. Joe Bay (FSM),
St. Johns River (OSUM; RTA), Santa Fe River (FSM), Sky Lake
(OSUM), Spring Creek (FSM; OSUM; USNM), Steinhatchee River
(DMNH), Suwannee River (FSM), Waccassa River (FSM), Wekiva
River (RB), Withlacoochee River (MCZ; USNM), and Yellow River
(FSM).

GEORGIA
(Fig. 6)

The first specimens of Corbicula fluminea reported in
Georgia were collected from the Altamaha River in 1971 near
river mile 116 (Sickel, 1973). Later, in 1971, populations were
also found in the Ocmulgee and Flint rivers (Sickel, 1973).
The Fiint River and the Altamaha River populations were
found in association with Lampsilis anodontoides floridensis
(Lea, 1852), Lampsilis uniominatus (Simpson, 1900), and
Quincuncina infucata (Conrad, 1834) in coarse sand substrata
(Sickel, 1973).

Sickel (1976) reported that the Altamaha River popula-
tion increased to a density of more than 2500/m2 between
1971 and 1974. However, Gardner et al. (1976) reported den-
sities of 10000/m2 in 1974 in some localities in the Altamaha
River. These population increases were accompanied by a
decline in populations of unionid bivalves (Gardner et al.,
1976).

Fuller and Powell (1973) reported Corbicula fluminea
in the Savannah River in 1972. Fuller and Richardson (1977)
noted the success of C. fluminea in that river and reported
that these bivalves actively uproot unionid bivalves.

CORBICULA SYMPOSIUM

Figs. 7-12. Zoogeographic distribution of Corbicula fluminea in Illinois (7), Indiana (8), Kentucky (9), Louisiana (10), Maryland (11), and Mississippi
(12). Scale bar = 50 km.

COUNTS: CORBICULA FLUMINEA IN THE UNITED STATES

Walker (1982) reported mass mortalities of Corbicula
fluminea in the Towaliga River, a tributary of the Ocmulgee
River, in 1981. He believed a similar mortality occurred in
the same stream in 1980. However, no causitive factor was
identified for these mass mortalities.

Corbicula fluminea has been found in the following Georgia
waters: Altamaha River (DMNH; FSM; OSUM; Gardner et al.; Sickel,
1973, 1976, 1979), Chattahoochee River (USNM), Chickamauga
Creek (OSUM), Chickasawhatchee River (FSM), Coahulla Creek
(OSUM), Consauga River (OSUM), Flint River (OSUM; USNM; Sickel,
1973), Lake Allatoona (OSUM), Little Ocmulgee River (OSUM), Oc-
mulgee River (OSUM; USNM; Sickel, 1973, 1979), Ogeechee River
(FSM), Ohoopee River (FSM; OSUM; USNM), Oostanula River
(OSUM), Potatoe Creek (OSUM), Pound Creek (OSUM), Savannah
River (FSM; Fuller and Powell, 1973), Towaliga River, and
Withlacoochee River (FSM).

HAWAII

Corbicula fluminea has been sold as food in Kailua,
Oahu Island in the Open Market (Burch, 1978). Burch (1978)
reported that several shipments have been intercepted by
the Department of Agriculture Plant Quarantine Office. These
clams were imported from the Orient and from California. As
yet, however, no populations have been reported from the
streams of Hawaii.

IDAHO

The only report of Corbicula fluminea in Idaho is that
of Ingram (1959). He noted the presence of a population in
the Snake River at the Wasington-ldaho border.

ILLINOIS

(Fig. 7)

The oldest record of Corbicula fluminea in Illinois is that
of Fetchner (1962) for Massac County along the Ohio River.
Parmalee (1965) later reported the spread of C. fluminea in
the Ohio River of Illinois and found specimens at 18 localities
in the rivers and streams of the southern border counties in-
cluding the Mississippi, Ohio, and Wabash rivers. Parmalee
believed that C. fluminea became established in the Ohio
River in Illinois in 1 961 and that populations in the three rivers
of the state were descendents of populations from the Cin-
cinnati Reach of the Ohio River (Keup et al., 1963). Popula-
tions in the Wabash River were the youngest being approx-
imately 2 years old in 1964 (Parmalee, 1964). Substrata for
the Illinois populations reported by Parmalee varied from silt-
sand and mud slab-like cobbles. However, densest popula-
tions were found in sand substrata.

Thomerson and Myer (1970) reported large popula-
tions of Corbicula fluminea inhabiting the cooling system of
the Granite City Steel Company’s plant at Granite City in
1969. The intake pipe for the plant’s water system was at Lock
and Dam 27 of the Chain and Rocks Canal of the Mississip-
pi River. They believed C. fluminea became abundant at that
site in 1966. Specimens from this locality were later used to
determine the effects of potassium on larval and adult C.

fluminea by Anderson et al. (1976).

Thompson and Sparks (1977) found populations of
Corbicula fluminea in the Illinois River and noted that waters
are generally warmer in areas where the bivalves are found.
They believed the clams became established sometime bet-
ween 1970 and 1971 .

Klippel and Parmalee (1979) found that Corbicula
fluminea were the most frequently recovered bivalve in Lake
Springfield of the Sangamon River. Lake Springfield is an
impoundment constructed in 1935 and C. fluminea probably
became established there in the 1970’s.

Lewis and Brice (1980) commonly found Corbicula
fluminea in the Kankakee River. The location at which C.
fluminea was abundant did not receive thermal discharge, a
condition that has been credited with the success of the nor-
thern populations in Minnesota (Cummings and Jones, 1978).

Corbicula fluminea has been reported from the following Illinois
waters: Crab Orchard Lake (Thompson and Sparks, 1977), Illinois
River (OSUM; Thompson and Sparks, 1977), Kankakee River (Lewis
and Brice, 1980), Kaskasia River (Thompson and Sparks, 1977),
Mississippi River (OSUM; Thomerson and Myer, 1970), Ohio River
(FMNH; ISM; NMNS; OSUM; Fetchner, 1962), Saline River (OSUM),
Sangmon River (Klippel and Parmalee, 1979; Thompson and Sparks,
1977), and Wabash River (FMNH; ISM; OSUM).

INDIANA

(Fig. 8)

Fox (1969) reported Corbicula fluminea in the Ohio
River of southern Indiana. No specific locality information was
reported. Taylor (1982) collected C. fluminea in three of seven
localities in Big Indian Creek, a tributary of the Ohio River.
No other published reports of C. fluminea in Indiana are
known.

Metty (Personal communication, 1979) found Corbicula
fluminea in the East Fork of the White River near the Indiana-
Ohio state line.

Corbicula fluminea has been found in the following Indiana
waters: Big Indian Creek (MUMC), Blue River (OSUM), Ohio River
(OSUM), Salt Creek (OSUM), Stoney Creek (OSUM), Wabash River
(OSUM), and White River (DM).

IOWA

The only report of Corbicula fluminea in Iowa is that
of Eckcald (1975). It was found in the thermal effluent from
an electric power generating station at Lansing in 1974.
Eckbald believed the oldest clams at this site were 2 years
old. They reached a density of 200/m2 in some areas.

KENTUCKY

(Fig. 9)

The first report of Corbicula fluminea in Kentucky is
that of Sinclair and Isom (1961) who reported the collection
of specimens at the Shawnee Steam Plant at Paducah on the
Ohio River in 1957. This initial collection was later described
by other investigators (Bates, 1962a, b; Stein, 1962) and

CORBICULA SYMPOSIUM

Figs. 13-18. Zoogeographic distribution of Corbicula fluminea in Missouri (13), New Jersey (14), New Mexico (15), North Carolina (16), Ohio
(17), and Oklahoma (18). Scale bar = 50 km.

COUNTS: CORBICULA FLUMINEA IN THE UNITED STATES

also represents the first collection of C. fluminea east of
Arizona.

Bates (1 962a) noted the presence of Corbicula fluminea
in the Tennessee River at Kentucky Reservoir and in a later
paper (Bates, 1962b) reported the presence of populations
in the Green River. Stein (1962) reported the collection of
more than 200 specimens from the Ohio River at Dayton, Ken-
tucky, that were living in filamentous green algal mats.

Bickel (1966) described the ecology of Corbicula
fluminea in the Ohio River at Louisville and noted that spring
mortalities seemed to be an annual occurrance.

Branson and Batch (1969) reported the occurrence of
Corbicula fluminea in the Kentucky and Red River systems
and noted that the first infestation of the Red River probably
occurred sometime after 1966. In 1968 they reported the col-
lection of specimens in the Cumberland and Tennessee
Rivers (Branson and Batch, 1971)

Williams (1969) described populations of Corbicula
fluminea from the Green and Tennessee Rivers and reported
that these bivalves contituted 99% of all living mussels col-
lected in those rivers. He further reported on the collection
of C. fluminea from the Green River at Mammoth Cave Na-
tional Park in 1968. Similar collections by Bates (1962b) fail-
ed to reveal the presence of C. fluminea at that locality in 1 962.
Isom (1974) noted the continued presence of C. fluminea in
the Green River.

Blankenship and Crockett (1972) reported the first oc-
currence of Corbicula fluminea in the Rockcastle River for col-
lections made in 1968 at Livingston.

Sickel et a!. (1980) and Sickel and Chandler (1981)
reported on mass mortalities of Corbicula fluminea popula-
tions in the Cumberland River in 1978 and noted similar oc-
currences in Barkley Lake. They further discussed the possi-
ble commerical exploitation of C. fluminea populations as a
fish bait.

Taylor (1980) reported Corbicula fluminea in Tygarts
Creek in Carter and Greenup counties of eastern Kentucky.
Substrata at his collection localities varied from mud covered
by algal mats to shale rubble and sand.

Sickel and Lyles (1981) noted the first occurrence of
the presumed commensal Chaetogaster lymnaei with Cor-
bicula fluminea living in Barkley Lake. The oligochaete was
found on the gills and foot with an incidence of infestation
greater than 80%.

Corbicula fluminea has been found in the following waters of
Kentucky: Buck Creek (OSUM), Coal River (MUMC), Cumberland
River (FMNH; OSUM; Branson and Batch, 1971; Sickel and
Chandler, 1981; Sickel and Lyles, 1981), Dix River (Branson and
Batch, 1971), Eagle Creek (MUMC), Elkhorn Creek (MUMC), Floyds
Fork (DMNH; MUMC; OSUM), Gasper River (OSUM), Green River
(FWM; MCZ; MUMC; NMNS; OSUM; USNM; Bates, 1926b; Williams,
1969), Kentucky River (OSUM; Branson and Batch, 1969), Licking
River (CMNH), Little River (OSUM), Mississippi River (FMNH), Mud
River (OSUM), Nolin River (MUMC), Ohio River (OSUM; Bickel, 1966;
Sinclair and Isom, 1961; Williams, 1969), Red River (OSUM; Bran-
son and Batch, 1969), Rockcastle River (FWM; OSUM), Salt River
(OSUM), Silver Creek (Branson and Batch, 1969), Slate Creek
(MUMC; OSUM), Tennessee River (Sickel ef a/., 1981), Tradewater
River (OSUM), and Tygarts Creek (MUMC).

LOUISIANA
(Fig. 10)

Stein (1 962) made the first report of Corbicula fluminea
from Louisiana waters. These bivalves were collected in the
Mississippi River in 1961. However, no precise locality data
were published concerning these specimens and none ac-
companied the specimen’s collection label (OSUM 4416).
Dundee and Harman (1963) found C. fluminea in benthic
samples taken from the Calcasieu River in 1961 and from
Bayou Magasille and Bayou Sorrel in 1962.

Gunning and Suttkus (1966) found Corbicula fluminea
throughout the lower Pearl River system and speculated, bas-
ed on shell measurements, that it became established in the
Pearl River in 1959. Corbicula fluminea was abundant in sand,
gravel, silt, and mixtures of these substrata types in the Pearl
River (Gunning and Suttkus, 1966).

Kuckyr and Vidrine (1975) reported Corbicula fluminea
to be the most abundant of 21 species of bivalves collected
in the Tensas River of southern Louisiana. Vidrine and
DeRouen (1976) found that C. fluminea inhabited most of the
streams of the Bayou Teche system of southeast Louisiana.

Corbicula fluminea has been reported from the following Loui-
siana waters: Bayou Cocodrie (Vidrine and DeRouen, 1976), Bayou
Magasille (Dundee and Harman, 1963), Bayou Sorrel (HMNH;
Dundee and Harman, 1963), Calcasieu River (Dundee and Harman,
1963), Mississippi River (OSUM), Pearl River (FMNH; USNM; Gun-
ning and Suttkus, 1966), Red River (Corbicula Newsletter, 1982), Ten-
sas River (Kuckyr and Vidrine, 1 975), and an unnamed creek in Lin-
coln Parish (FSM).

MARYLAND
(Fig. 11)

Stotts et al. (1977) made the first report of Corbicula
fluminea from Maryland. Specimens were collected at Sus-
quehanna Flats of the northern reaches of the Chesapeake
Bay on silt-sand to hard substrata near Turkey point and
Havre-de-Grace. Stotts et al. (1977) noted the Corbicula
fluminea have been in the bay since 1975. Nichols and
Domermuth (1981) noted that populations of C. fluminea are
present in the Susquehanna River at Conowingo Dam. These
clams may have been the population of origin for those
populations at Susquehanna Flats.

Dresler and Cory (1981) reported the discovery of Cor-
bicula fluminea in the Potomac River at the center of
Washington, D. C. (RM 89) to the mouth of Piscataway Creek
(RM 84.5) in 1976. They further discussed the fouling of the
Potomac Electric Power steam generating station that was
so seriously infested that live clams, dead shells, and silt
build-up caused operational problems.

Counts (1981b) made the first report of Corbicula
fluminea in waters of the Eastern Shore. Specimens were
found in the Wicomico River at Salisbury. No fouling problems
have been associated with Eastern Shore populations.

Records for Corbicula fluminea in Maryland include:
Chesapeake Bay (Stotts et al., 1977), Nasawango Creek (USNM),
Potomac River (OSUM; USNM; Dresler and Cory, 1980), Susquehan-
na River (Nichols and Domermuth, 1981), and Wincomico River
(Counts, 1981b).

CORBICULA SYMPOSIUM

MICHIGAN

The only report of Corbicula fluminea in Michigan is
that of Clarke (1981). Specimens were found on sand
substratum in 1 .3 m of water approximately 305 m south of
the mouth of Sandy Creek. No other bivalves were found at
this locality. Scott-Wasilk, (Personal communication, 1983)
reported that these bivalves are living in a thermal plume from
a power station and have survived the winters since their
discovery. Other specimens from this locality are in the col-
lection of OSUM and USNM.

MINNESOTA

Corbicula fluminea has been reported from only one
locality in Minnesota. Specimens were found in a small lake
that received thermal effluent from Northern States Power
Company Blackdog electric generating station, and others
were found in the channel of the Minnesota River downstream
from the plant (Cummings and Jones, 1 978). Live specimens
that were collected at these sites were approximately 3 years
old. The shells, both live and dead, varied in height from 6
to 44 mm and were thought to represent individuals ranging
in age from 1 to 5 years old. Although the proportionately large
number of empty shells suggested a mass mortality, no
causative agent could be identified.

MISSISSIPPI
(Fig. 12)

Heard (1966) first discovered Corbicula fluminea in
Mississippi in the Yazoo River at Vicksburg in 1963. Heard
also reported on populations in the Coldwater and Leaf rivers.
Grantham (1967) reported C. fluminea in the Chickasawhay,
Pascagoula, and Pearl rivers. Grantham (1969) later reported
C. fluminea in the Big Black, Deer, Leaf, Tennessee, and Tom-
bigbee rivers.

Cooper and Johnson (1980) found Corbicula fluminea
in Grenada Reservoir of the Yalobusha River during studies
conducted there between 1973 and 1976. Leard etal. (1980)
collected C. fluminea from the Big Black, Chickasawhay,
Chucky, Coldwater, Leaf, and Pearl rivers and from Steel
Bayou and Black Creek.

King and Miller (1982) reported Corbicula fluminea to
be most abundant bivalve species in the Tombigbee River
near Columbus. Hartfield and Cooper (1982) reported C.
fluminea to be absent in the southern part of the state bet-
ween, but not including, the Pascagoula and Pearl rivers and
streams that drain into the lower Mississippi River; Bayou
Pierre, Buffalo Bayou, and the Homochitto River.

Corbicula fluminea is reported from the following Mississippi
waters: Allan Branch (FMNH), Amite River (MMNS), Bear Creek
(MMNS), Big Black Creek (MMNS), Big Black River (FMNH; MMNS;
Leard et al., 1980), Bouge Phalia River (FSM), Buckatunna Creek
(FMNH; FSM), Buttahatchie River (MMNS; OSUM), Chickasawhay
River (FMNH; FSM; OSUM; Grantham, 1967), Chunky River (FMNH;
FSM), Coldwater River (Heard, 1966; Leard etal., 1980), Leaf River
(FMNH; FSM; MMNS; OSUM; Heard, 1966; Leard et al., 1980),

Mississippi River (MMNS; OSUM), Moss Creek (FMNH), Okatibee
Creek (USNM), Okatoma Creek (MMNS), Pascagoula River (MMNS;
Leard ef al., 1980), Pearl River (MMNS; OSUM; USNM; Leard etal.,
1980), Shubuta Creek (FMNH), Souinlovey Creek (FMNH), Steel
Bayou (Leard et al., 1980), Sunflower River (FSM), Talahala Creek
(MMNS), Tibbee Creek (FMNH; MMNS), Tombigbee River (MMNS;
OSUM), Woodward Creek (FMNH), Yalobusha River (Cooper and
Johnson, 1980), Yazoo River (FMNH; USNM; Heard, 1966), and
Yockanookany River (MMNS).

MISSOURI

(Fig. 13)

Fox (1969) reported Corbicula fluminea from the
Mississippi River of Missouri. Oesch (Personal communica-
tion, 1979) reported the species from the St. Francis,
Gasconade, Osage, and Meramec rivers. Buchanan (Per-
sonal communication, 1979) found C. fluminea from the Big
and Bourbuese rivers as well as in the little Black River system.

Corbicula fluminea is reported from the following Missouri
waters: Big Creek (MDC), Big River (ACB; MDC; OSUM), Black River
(DMNH; FMNH; MDC; MNHD; OSUM), Bourbeuse River (ACB; MDC;
OSUM), Bryant Creek (MDC), Cane Creek (MCZ; MDC; OSUM),
Gasconade River (MDC), Little Black River (ACB), Little River Canal
(MCZ; OSUM), Logan Creek (ACB), Meramec River (MDC; OSUM),
Mississippi River (MDC; OSUM; Fox, 1969), Missouri River (RDO),
Moreau River (MDC), Osage River (MDC; SBSK), St. Francis River
(DMNH; MDC), Thomas Hill Reservoir (MDC), and Whitewater River
(MDC).

NEVADA

There is but a single report of Corbicula fluminea in
Nevada. Ingram (1959) took specimens from Lake Meade of
the Colorado River.

NEW JERSEY
(Fig. 14)

The first report of Corbicula fluminea in New Jersey
was made by Fuller and Powell (1973) who recorded its
presence in the Delaware River between Trenton and
Philadelphia. Bivalves seemed to prefer muddy and fine
gravel substrata.

Crumb (1977) reported on the colonization of the
Delaware River by Corbicula fluminea and noted the exten-
sion of its range between Trenton and Burlington. Crumb
hypothesized that C. fluminea invaded the Delaware River
sometime during, or before, 1971 and noted that it was usually
found in sand or coarse sediments, as reported by Fuller and
Powell (1973). He also found that the bivalve Sphaerium
transversum (Say) was common in the Delaware River until
the expansion of the population of C. fluminea. Corbicula
fluminea, with Limnodrilus spp., Procladius culiciformis, and
Peloscolex ferox dominated the benthic community at his col-
lection sites.

Trama (1982) reported a population of Corbicula
fluminea in the Raritan River. Populations were successful
in both the tidal and non-tidal portion of the river as well as
in the North and South Branches of the river.

COUNTS; CORBICULA FLUMINEA IN THE UNITED STATES

Figs. 19 - 25. Zoogeographic distribution of Corbicula fluminea in Oregon (19), South Carolina (20), Tennessee (21), Texas (22), Virginia (23),
Washington (24), West Virginia (25). Scale bar = 50 km.

CORBICULA SYMPOSIUM

Other specimens of Corbicula fluminea from the
Delaware River are in ANSP.

NEW MEXICO
(Fig. 15)

The only published account of Corbicula fluminea in
New Mexico is that of Metcalf (1966). Clams were found in
the West Drain, a drainage ditch, in the Mesilla Valley, a part
of the Rio Grande drainage. The affected portion of the ditch
flows between Texas and New Mixico in a north - south direc-
tion from El Paso, Texas to Radium Springs, New Mexico
and crosses the state line in several places.

Records of Corbicula fluminea in New Mixico waters are:
Caballe Reservoir (USNM), Elephant Butte Reservoir (USNM), Pecos
River (USNM), and Rio Grande (MNHD; USNM; Metcalf, 1966).

NORTH CAROLINA
(Fig. 16)

The earliest report of Corbicula fluminea in North
Carolina was made by Fox (1971). He reported specimens
taken at the Allen Steam Station at Lake Wylie of the Catawba
River system in 1970. Fuller and Imlay (1976) later reported
C. fluminea in Lake Waccamaw, and speculated that the
population became established only where the habitat was
severely disturbed by man.

Records of Corbicula fluminea in North Carolina waters are:
Cape Fear River (NCSM), Catawba River (JJH; NCSM; Fox, 1971)
Eden River (JJH), Little River (NCSM), Long Mountain Island Lake
(USNM), Richardson Creek (NCSM), Rocky River (OSUM), Uhwar-
rie River (MCZ), and Waccamaw River (OSUM; Fuller and Imlay,
1976).

OHIO
(Fig. 17)

Pojeta (1964) first reported Corbicula fluminea in the
Ohio River from collections made in 1962. Horning and Keup
(1964) later reported a decline of the C. fluminea population
in the Cincinnati Reach of the Ohio River. They attributed
this mass mortality to the severe winter of 1 962-1963 during
which the river was ice-covered for seven days. Keup et al.
(1963) also noted the spread of C. fluminea to points above
Cincinnati at RM 465.5. Three years later, C. fluminea had
spread to Marietta at RM 172 (ORSANCO, 1966). Taylor
(1980) found C. fluminea along the entire length of the Ohio
River from the Ohio-Pennsylvania state line to just below the
mouth of the Scioto River.

The most recently reported infestation of Ohio waters
occurred in the Maumee River at the Davis Basse Nuclear
Power Station, Toledo (Scott-Wasilk et al., 1983).

Corbicula fluminea has been reported from the following waters
of Ohio: Brush Creek (DM), Hocking River (OSUM), Licking River
(OSUM), Little Muskingum River (OSUM), Maumee River (Scott-
Wasilk et al., 1983), Meigs Creek (OSUM), Miami River (OSUM),
Muskingum River (DMNH; MCZ; OSUM), Ohio River (MUMC; OSUM;
USGM; ORSANCO, 1966), Olentangy River (DM; OSUM), Olive
Green Creek (OSUM), Scioto River (FMNH), and Stillwater River
(GAC).

OKLAHOMA
(Fig. 18)

Few reports have appeared in the literature concern-
ing the occurrence and distribution of Corbicula fluminea in
Oklahoma. Clench (1970) first reported C. fluminea in the state
from Lake Overholser collections made in 1969. Britton and
Morton (1979) reported C. fluminea from Lake Texoma and
Lake Thunderbird. White (1977) commented that the Lake
Texoma population experienced a reduction in numbers due
to the droughts of 1 975 - 1 976 and 1 976 - 1 977. He also noted
that gravel and rip-rap habitats seemed to provide greater
protection from desiccation than did sand substrata when
water levels fluctuate. White and White (1977) found that C.
fluminea from Lake Texoma cannot withstand more than a
few days of aerial exposure and suggested that controlled
water draw-down during the winter months may be used as
a control method in reservoirs and other impoundments.

Oklahoma records for Corbicula fluminea include the follow-
ing bodies of water: Arkansas River (RLR), Caddo Creek (UOM), Little
River (RLR), North Canadian River (OSUM; TCU; Clench 1972;
O’Kane et al., 1977), and Red River (TCU; UOM).

OREGON
(Fig. 19)

Although Ingram (1948) did not report Corbicula
fluminea from Oregon, he did report the species as being col-
lected along the north bank of the Columbia River, Pacific
County, Washington. Hence, it is not unreasonable to assume
that C. fluminea was also present in Oregon during the 1940’s.
In a later paper Ingram (1959) mentions populations in the
Williamette River and it’s confluence with the Columbia River.

Fox (1969) noted that the entire Columbia River Basin
of Oregon was infested and reported the presence of Cor-
bicula fluminea in the Umpqua River for the first time.

Snow (personal communication, 1979) noted that Cor-
bicula fluminea is an important item in the diet of the white
sturgeon, Ancipenser transmontanus Richardson, in Oregon.

Corbicula fluminea has been reported from the following waters
of Oregon: Columbia River (MCZ; NMNS; ODFW; USNM), John Day
River (ODFW), Suislaw River (ODFW), Smith River (ODFW), Ump-
qua River (Fox, 1969), and Williamette River (Ingram, 1959).

PENNSYLVANIA

Corbicula fluminea has been reported only from the
Ohio River at Pittsburgh (Taylor, 1 980). Although the species
is found in the Susquehanna River below Conowingo Dam,
Maryland, it has not yet moved upstream into Pennsylvania
(Nichols and Domermuth, 1981). No other records, other than
those for the Delaware River, which is a part of New Jersey,
have been reported for the state.

SOUTH CAROLINA
(Fig. 20)

The first published report of Corbicula fluminea in
South Carolina is that of Fuller and Powell (1 973). They found
“gapers” and living specimens in the Pee Dee River on hard

COUNTS: CORBICULA FLUMINEA IN THE UNITED STATES

Figs. 26 - 27. Chronologic zoogeographic distribution of Corbicula fluminea by United States counties. 1938 - 1945 (26). 1946 - 1950 (27).
Scale bar = 500 km.

CORBICULA SYMPOSIUM

clay and sand substrata in 1972. They further reported
populations found in 1973 in the Savannah River near
Augusta, Georgia. Fuller later reported C. fluminea to be abun-
dant in the Cooper River (Fuller, 1974) and in the Santee-
Cooper River systems (Fuller, 1976).

Fuller and Imlay (1976) found numerous living and dead
specimens of Corbicula fluminea below the confluence of the
Waccamaw River with the Intracoastal Waterway. Fuller
(1978a) later noted the establishment of a population in the
Intracoastal Waterway in Georgetown County.

The infestation of the Savannah River Power Plant by
Corbicula fluminea has received wide attention in the
literature. Tille ef al. (1978) noted severe yearly infestations
at the plant and reported on irradication experiments using
gamma irradiation. Boozer and Mirkes (1979) discussed the
association of C. fluminea with Musculum partumenium (Say,
1822) in the sedimentation basin at the plant. Harvey (1981)
reported on the recolonization of the plants reactor water cool-
ing system and noted that the pumps, reactor basins, pump
wells, and emergency cooling system must be cleaned every
10 months to keep them free of these bivalves.

Corbicula fluminea has been found in the following waters
of South Carolina: Cooper River (TCU; USNM), Edisto River (OSUM),
Hartwell Reservoir (JJH), Intracoastal Waterway (Fuller, 1978a; Fuller
and Powell, 1973), Lake Keowee (JJH), Little Pee Dee River (Kool
et al., 1981), Pee Dee River (Coney et al., 1983; Fuller and Powell,
1973), Salkahatchie River (OSUM), Santee River (OSUM; Fuller,
1976), Savannah River (TCU), and Waccamaw River (Fuller and
Powell, 1973).

TENNESSEE
(Fig. 21)

The first account of Corbicula fluminea in Tennessee
appears in Sinclair and Isom (1961) and described the infesta-
tion below Pickwick Dam of the Tennesse River in 1959. By
1961 , C. fluminea had also invaded the Cumberland River of
Tennessee (Sinclair and Isom, 1961). By 1962, C. fluminea
had infested the Johnsonville Steam Plant on the Kentucky
Reserevoir and public and industrial water supplies as well
as a sand and gravel quarry at Chattanooga (Sinclair and
Isom, 1963). Introduction of C. fluminea into the Tennessee
River occurred sometime after 1958. A survey of mussels of
the Kentucky Reservoir by Bates (1962a) did not reveal their
presence in 1985.

Sinclair (1964) discussed infestations of the sand and
gravel industries of the Tennessee and Cumberland rivers
and described the deleterious effects of Corbicula fluminea
in freshly poured concrete.

Isom and Yokley (1968) found Corbicula fluminea at
10 stations in the Duck River between river miles 71 and 242.5
in 1965. Clench and Stansbery (1969) later reported the in-
festation of the Nolichucky River southeast of Warrensburg
and noted that these populations were found living in beds
of angular limestone, loose limestone rocks, gravel, sand, and
sandy mud substrata.

Isom (1971) reported Corbicula fluminea from the Fort
Loudoun Reservoir of the Tennessee River in 1970, and later
(Isom, 1972) reported an infestation at the Nickajack Dam

site first noted in 1965 before completion of the dam.

Isom et al. (1973) found Corbicula fluminea in the Elk
River basin during studies made in 1965 to 1967. Van der
Schalie (1973) further noted the presence of C. fluminea in
the Duck and Buffalo rivers, tributaries of the Tennessee River
and noted that they are commonly eaten by mink and
muskrat, thus replacing the once-numerous unionid bivalves
as a food source for these mammals.

Goss and Cain (1977) discussed the history of biofoul-
ing by Corbicula fluminea at the Tennessee Valley Authori-
ty’s Johnsonville Steam plant as well as the fouling of Brown’s
Ferry Nuclear Plant in late 1974. They also discussed various
techniques used to control these bivalves at industrial
facilities.

Eagleson and Morgan (1977) reported the growth rates
of populations of Corbicula fluminea in the Clinch River and
two small tributaries, Grassy Creek and Bear Creek, near Oak
Ridge in 1975 - 1976.

Corbicula fluminea has been found in the following waters
of Tennessee: Barren Fork River (OSUM), Big Bigby Creek (OSUM),
Big Hickory Creek (OSUM), Big Rock Creek (OSUM), Big Swann
Creek (OSUM), Buffalo River (ANSP; FSM; OSUM), Clinch River
(ANSP; OSUM; USNM), Collins River (OSUM), Cumberland River
(OSUM; Sinclair and Isom, 1963), Duck River (ANSP; MCZ; OSUM),
East Rock Creek (OSUM), Elk River (FSM; MCZ; OSUM), Emory
River (OSUM), Fall Creek (OSUM), Flat Creek (OSUM), Fountain
Creek (OSUM), Garrison River (OSUM), Greenlick Creek (OSUM),
Holston River (ANSP; USNM), Harpeth River (OSUM), Hatchie River
(OSUM), Lick River (FSM; OSUM), Little Duck River (OSUM), Little
Tennessee River (SBSK), Mississippi River (SBSK), Nine Mile Creek
(OSUM), Nolichucky River (ANSP; DMNH; FSM; MCZ; OSUM), North
Fork Creek (OSUM), Notchy Creek (OSUM), Obey River (OSUM),
Paint Rock River (OSUM), Piney River (OSUM), Red River OSUM),
Rich Creek (OSUM), Richland Creek (OSUM), Rutherford Creek
(OSUM), Sequatchie River (OSUM), Shoal Creek (USNM), Sinking
Creek (OSUM), South Chickamauga Creek (OSUM), Stones River
(OSUM), Sugar Creek (OSUM), Tellico River (OSUM), Tennessee
River (ANSP; FMNH; MCZ; NMNS; OSUM; USNM; Sickel et al.,
1981), and Weekly Creek (OSUM).

TEXAS
(Fig. 22)

Metcalf (1966) first reported Corbicula fluminea in
Texas. This first population was located in the Rio Grande
at El Paso. These bivalves were believed to have invaded
the Rio Grande in 1964 or earlier (Metcalf, 1966). Since its
initial discovery, C. fluminea has extended its range
downstream to Monte Alto Reservoir and Falcoln Lake (Mur-
ray, 1971a).

Corbicula fluminea has also invaded the Colorado River
system, specimens reported from this river have nearly all
been reported from reservoirs. Murray (1971b) reported C.
fluminea from Lake Lyndon B. Johnson. O’Kane ef al. (1977)
reported infestations in Lake Inks and Lake Travis. Britton
and Murphy (1977) found C. fluminea at Marble Falls, Austin,
and Bastrop.

Three instances of power plant fouling by Corbicula
fluminea occurred in Texas (McMahon, 1977; O’Kane et al.,
1977; Baker, 1978). In all three cases infestations required

COUNTS: CORBICULA FLUMINEA IN THE UNITED STATES

Figs. 28 - 29. Chronologic zoogeographic distribution of Corbicula fluminea by United States counties. 1951 - 1955 (28). 1956 - 1960 (29).
Scale bar = 500 km.

CORBICULA SYMPOSIUM

the shut-down of the plant to remove the clams and restore
normal operation.

Britton and Murphy (1977) reported that Corbicula
fluminea is a food of the fishes Lepomis microlophus (Gun-
ther) and Minytrema melanops (Rafinesque) in Lake Benbrook
and Aplocinotus Grunniens (Rafinesque) in Eagle Mountain
Lake.

Hillis and Patton (1982) examined the morphology and
genetic variability of populations in the Brazos River system.
It was their opinion that two species of bivalves in the genus
Corbicula may have been introduced into North America.
Their evidence is, however, preliminary and they are unable
to refer the presumed second species to a taxon with cer-
tainty. Britton (1982) reviewed the biogeography and ecology
of C. fluminea in Texas and also discussed the possibility of
two species being in the state.

Corbicula fluminea is reported from the following waters of
Texas: Angelina River (TCU; UOM), Big Cypress River (Pool and
McCullough, 1979), Blanco River (OSUM; Horne and Macintosh,
1979), Brazos River (HMNS; OSUM; TCU; UOM; Britton and Mor-
ton, 1979), Colorado River (HMNS; TCU; Baker, 1978; Britton, 1982;
Britton and Murphy, 1977), Concho River (Baker, 1978), Guadelupe
River (UOM; Britton and Murphy, 1977), Johnson Creek (Britton,
1982), Little Brazos River (OSUM), Llano River (UOM), Nueces River
(Britton, 1982; Britton and Murphy, 1977; Murray, 1971a, 1978),
Pecos River (Britton, 1982), Perdernales River (CED), Red River (Brit-
ton and Murphy, 1977), Rio Grande (HMNS; OSUM; SBMNH; Brit-
ton, 1982; Metcalf, 1966; Murray, 1971a), Sabine River (Pool and
McCullough, 1979), San Antonio River (CEB), San Gabriel River (Hillis
and Patton, 1982), San Jacinto River (MCZ; TCU; CEB; Hillis and
Patton, 1982), Spring Creek (OSUM), Trinity River (TCU; Aldridge
and McMahon, 1978; Britton and Murphy, 1977; Evans etal., 1979;
McMahon, 1977), and White River (Britton, 1982; Fontanier, 1982)

VIRGINIA
(Fig. 23)

Diaz (1974) made the first report of Corbicula fluminea
in Virginia. Specimens were taken in the James River be-
tween RM 80, at Richmond, and RM 45. He also found
populations at the confluence of the Appomattox and James
rivers. Clams were usually found on clay-silt substrate. Diaz
(1974) suggested that the triclad Dugesia trigrina may be a
predator of C. fluminea in the James River estuary. Shell
measurements indicated the James River was probably in-
vaded by C. fluminea in 1968 or earlier (Diaz, 1974).

Rodgers et al. (1977) reported the presence of Cor-
bicula fluminea in the New River. In a later paper Rodgers
et al. (1978) calculated the upstream rate of invasion to be
14.4 km/yr. The population dynamics, ecology, and control
of C. fluminea in the New River was discussed in detail by
Cherry et al. (1980).

Corbicula fluminea has infested three power stations
in Virginia: the Twelfth Street Generating Station of the
Virginia Electric and Power Company, Richmond (Diaz, 1974);
the Glen Lyn Power Plant (Rodgers et al., 1977, 1978; Graney
et al., 1980; Cherry et al., 1980); and the Potomac Electric
Power Company’s generating station at Alexandria (Dresler
and Cory, 1980). C. fluminea has also been found in the col-
lecting ponds of cooling towers at the Allied Chemical Com-

pany plant at Hopewell (Diaz, 1974). The thermal effluent of
the Glen Lyn plant has been implicated in the maintenance
of a stable population in the New River (Rodgers et al., 1 978;
Graney et al., 1980).

Corbicula fluminea has been found in the following waters of
Virginia: Appomattox River (USNM), Chickahominy River (DMNH),
Clinch River (MCZ; OSUM; USNM), James River (MPM; OSUM;
SBMNH; USNM; Diaz, 1974), New River (Rodgers ef a/., 1977), and
Potomac River (Dresler and Cory, 1980).

WASHINGTON
(Fig. 24)

The first report of Corbicula fluminea in the United
States was made from collections on the banks of the Col-
umbia River, Pacific County, in 1938 (Burch, 1944). Ingram
(1949) also reported C. fluminea from the north bank of the
Columbia River near Knappton-- the same population original-
ly described by Burch (1944). Fox (1971), commenting on this
discovery, noted that Burch’s materials were composed of
dead, drift shells and that the water of the Columbia River
where they were collected is saline. Fox (1971) further
reported C. fluminea in the Columbia River from near Knapp-
ton upstream to Richland.

Records for Corbicula fluminea in Washington include:
Chehalis River (CAS; DMNH; NMNS; USNM) and Snake River
(USNM).

WEST VIRGINIA
(Fig. 25)

The earliest report of Corbicula fluminea in West
Virginea was that of Thomas and MacKenthum (1964). They
found C. fluminea in the Kanawha River at two localities and
believed the clams became established populations of C.
fluminea from the Kanawha River not far from the localities
of Thomas and MacKenthum (1964). C. fluminea was col-
lected at five stations and found in association with unionid
bivalves where substrata was usually pebbly, the water clarity
good, and stream flow variable (Taylor and Morris, 1978).

Taylor and Counts (1977) reported populations in the
Ohio River and noted they were preyed upon by the Northern
Raccoon, Procyon lotor (Linne).

Markham et al. (1980) found that Corbicula fluminea
was the most abundant bivalve in the New River from the
Virginia - West Virginia state line downstream to the mouth
of Meadow Creek, a distance of Approximately 55 km.

Taylor and Hughart (1981) reported Corbicula flumiea
in the Elk River from its confluence with the Kanawha River
upstream to Sutton Dam, Braxton County. The bivalve was
found at all localities in association with unionid mussels.

Zeto (1982) reported the presence of Corbicula
fluminea in the Monogahela and West Fork rivers. Joy and
McCoy (1975) studied the correlation between shell and
visceral characters in a population of C. fluminea from a rif-
fle in the Mud River.

Records of Corbicula fluminea in West Virginia waters include:
Beach Fork Creek (MUMC), Big Seven Mile Creek (MUMC; OSUM;
Taylor and Hughart, 1981), Guyandotte River (MUMC), Hughes River
(MUMC), Kanawha River (DMNH; MCZ; MUMC; OSUM; USNM; Mor-

COUNTS: CORBICULA FLUMINEA IN THE UNITED STATES

Figs. 30-31. Chronologic zoogeographic distribution of Corbicula fluminea by United States counties. 1961 - 1965 (30). 1966 - 1970 (31).
Scale bar = 500 km.

CORBICULA SYMPOSIUM

ris and Taylor, 1978; Thomas and MacKenthum, 1964), Monongahela
River (Zeto, 1982), Mud River (MUMC), New River (OSUM; Markham
etal., 1980), Ohio River (DMNH; MUMC; OSUM; Taylor and Counts,
1977), Pocatalico River (MUMC), Twelve Pole Creek (MUMC), and
West Fork River (Zeto, 1982).

WISCONSIN

Corbicula fluminea was discovered in the St. Crox
River, near Hudson, in the summer of 1977 (Cummings and
Jones, 1978; Fuller, 1978b; Mathiak, 1979). This is the only
locality, thus far, for these bivalves that has been reported
for the state. Since they were not living in a thermal plume,
it is not known whether they have survived winters since their
discovery.

The Chronology of invasion of the United States by
Corbicula fluminea is presented in Table 1 and graphically
in Figs. 26-33. Records presented in Table 1 represent only
the first account of C. fluminea in a particular body of water
in a given state. Figures 26-33 present the complete distribu-
tion of C. fluminea in the form of records for the counties from
which specimens have been collected and/or reported. Col-
lections and reports cited in Table 1 are not necessarily the
first appearance of C. fluminea in a body of water but rather
represent the earliest documented detection.

DISCUSSION

Several hypotheses have been offered to explain the
introduction of Corbicula fluminea into North America.
The first states that the introduction was made by Chinese
immigrants who arrived on the west coast during the Gold
Rush of the late 1840’s (Fox, 1970a). These immigrants
played a prominant role in the development of the American
west as laborers in construction projects (such as the
Transcontinental Railroad) and in agriculture. Their migra-
tions, coupled with a traditional use of corbiculid clams as
food (Miller and McClure, 1931) and their penchant for settl-
ing in “China Towns” (although this was more a result of
highly restrictive laws governing the Chinese in the western
states at the time) suggest that they were the first to introduce
C. fluminea into North America. It is also significant that the
majority of the Chinese who immigrated to the west came
from, or had family origins in, Kwangtung Province in the
Pearl River Basin (of which Canton is the principal city) (Fox,
1971; Morton, 1973) where Cantonese have a traditional in-
dustry centered around the harvest and consumption of cor-
biculid clams (Miller and McClure, 1931).

It is unclear, however, how Chinese immigrants were
able to transport live Corbicula fluminea to the west coast of
North America in significant numbers. Voyages from China
to the western American coast in the mid 1800's involved the
use of slow sailing vessels, and the trip could take as along
as 180 days. Further, Caucasian ship captains and crews
were distrustful of the Chinese and, in many cases, locked
these passengers below decks with food and water just ade-
quate for subsistance during the voyage. It is difficult to im-

agine significant amounts of water being spared for clams
transported as a seed stock. Unless the Chinese were frugal
with their water rations, or corbiculids were able to withstand
long periods of only damp conditions, it is doubtful that many
of them arrived in North America in a viable condition by this
route. There were also many Chinese immigrants to the
Hawaiian Islands and the eastern coast of the United States
during this same period. Populations of C. fluminea have on-
ly just now been found in the Islands, and none was
discovered east of the Rocky Mountains before 1957. Con-
sidering the number of active malacologists of the period, it
would seem likely the C. fluminea would have been
discovered in the United States before 1938.

A second possibility for the introduction of Corbicula
fluminea to the Pacific Coast is their importation from the
Orient with the Giant Pacific Oyster, Crassostrea gigas
(Thunberg, 1793). Morton (1977) reported that C. fluminea
enters the Hong Kong area in shipments of C. gigas that are
imported to seed oyster beds. Abbott (1974) reported that
large numbers of C. gigas are imported into the western
coasts of the United States, Canada, and Mexico, and it is
conceivable that C. fluminea entered North American waters
by this route. However, no information is available concern-
ing the first importation of C. gigs.

The third hypothesis also involves the Chinese. Dur-
ing the mid- and late 1 930’s, mainland China was beset with
national and international hostilities. During that period, many
Chinese immigrated to the United States. This period would
allow more favorable shipboard conditions for the successful
transport of Corbicula fluminea to North America. Treatment
of passengers was somewhat improved over that of the
1 840 's and the time necessary to traverse the Pacific Ocean
was shorter. This is the theory subscribed to by Britton and
Morton (1 979) and seems to be most attuned to what is known
about the early history of C. fluminea in North America.
Although Counts (1981b) noted that C. fluminea were col-
lected as dead shells in British Columbia in 1924, before the
immigrations of the 1 930’s, this hypothesis best explains the
method of early introductions. The details of introduction,
however, are still unclear.

Earliest introduction of Corbicula fluminea into United
Staes waters was at Knappton, Pacific County, Washington
(Burch, 1944) (Fig. 34). McMahon (1982) suggested that C.
fluminea’s present zoogeographic distribution is the result of
this single introduction and possibly two subsequent, long-
distance introductions to the Ohio River at Paducah, Ken-
tucky, in 1957 (Sinclair and Isom, 1961) and to Lake
Overholser, Oklahoma, in 1969 (Clench, 1972).

Introduction of Corbicula fluminea into the Ohio River
in 1957 is indeed a dramatic leap across the continent. Until
its discovery at Paducah, C. fluminea appeared to spread
eastward across southern Washington into the Snake River
and across southern California into the Colorado River Basin
of Arizona and Nevada (Figs. 34 - 36). The transcontinental
leap to the Ohio River (Fig. 37), therefore, seems to be the
result of human activity, although the exact manner by which
it was transported is unknown.

Introduction into Lake Overholser is not as dramatic

COUNTS: CORBICULA FLUMINEA IN THE UNITED STATES

Figs. 32 - 33. Chronologic zoogeographic distribution of Corbicula fluminea by United States counties. 1971 - 1975 (32). 1976 - 1983 (33).
Scale bar = 500 km.

CORBICULA SYMPOSIUM

as depicted on the map of McMahon (1982), In reality, Cor-
bicula fluminea was collected in the Black and White rivers
in Arkansas in 1964 (MCZ 260919; OSUM 13992), in the St.
Francis River in 1966 (OSUM 20142), and in the Ouachita
River at Camden (OSUM 20409) and at Arkadelphia (FSM-
U) in 1969 (Figs. 38 - 39); the year of Clench’s (1972) Lake
Overholser discovery. Kraemer (1971; 1976) reported signifi-
cant populations of C. fluminea in the Arkansas River, of which
the North Canadian River (and hence Lake Overholser) is a
tributary, before 1964. Records indicate that C. fluminea was
well established in the Arkansas River drainage, and in the
rivers of Arkansas generally for at least five years before its
appearance in Oklahoma and was very likely in the Fort
Smith, Arkansas vicinity, on the Arkansas - Oklahoma state
line, before collections were made at Lake Overholser. It is
possible that a long distance introduction was made into Lake
Overholser from the Ohio - Mississippi basin and that occur-
rence of populations in the Arkansas River reported by
Kraemer (1971) was the result of normal downstream migra-
tion. An upstream dispersal in the Arkansas - Canadian river
system terminating at Lake Overholser is not temporally dif-
ferent from similar upstream movements by C. fluminea in
the Ohio River system.

Examination of the upstream range extension records
for Corbicula fluminea in the Ohio River reveal that after the
initial discovery at Paducah, Kentucky in 1957 (Sinclair and
Isom, 1961), these bivalves were collected at Ghent, Ken-
tucky, in 1961 (OSUM 6585), at Cincinnati, Ohio in 1963
(OSUM 9275), at Marietta, Ohio in 1966 (ORSANCO, 1966),
and at Pittsburgh, Pennsylvania in 1979-1980 (Taylor, 1980).
C. fluminea was also collected in the Kanawha River, West
Virginia in 1963 (Thomas and MacKenthum, 1964). In view
of these records, McMahon (1982) was incorrect in stating
that C. fluminea reached its upstream limit in the Ohio River
system with the establishment of the Kanawha River popula-
tion in 1963. Collections of C. fluminea have also been made
in the Monongahela River at Morgantown, West Virginia,
(Zeto, 1982) and in the Olentangy River, Delaware County,
Ohio, in 1972 (OSUM 33900), demonstrating that these
bivalves are still expanding their upstream range in the Ohio
River system. Populations reported in the New River at Glen
Lyn, Virginia (Rodgers et al., 1979), may also reflect the
upstream range extension of C. fluminea.

More dramatic than the Lake Overholser population
expansion was the extablishment of a population in the
Escambia River, near Century, Florida, in 1960 (Schneider,
1967). It seems unlikely that this population was established
by naturally mediated dispersal.

Corbicula fluminea has a short-term planktotrophic
veliger stage (Sinclair and Isom, 1963; Goss and Cain, 1977).
While a velum is present, Sinclair and Isom found no evidence
that this organ is used to keep larvae suspended in the water
column. They also noted the presence of an apical swim plate
but could not demonstrate that this is used for swimming. Eng
(1979) found that C. fluminea larvae in the Delta - Mendota
Canal are essentially benthic, and that a pediveliger stage is
released by the parent that broods larvae in a marsupium
located in the inner gills. Both Sinclair and Isom (1963) and

Eng (1979) observed that the larvae, though benthic, can be
transported in turbulent water.

Kraemer (1979) described the development of a
byssus in C. fluminea after marsupial release and observed
that Arkansas River larval populations frequently attach
themselves to sand grains with this structure. McMahon
(1982) reported populations of C. fluminea in Texas similarly
attached to sand grains frequently entangled in algal mats.
Stein (1962) found adult C. fluminea , in algal mats in Cincin-
nati, Ohio, in 1962.

McMahon (1982) argued that the rate of invasion by
Corbicula fluminea is higher when moving downstream than
moving upstream. However, by 1959, C. fluminea had expand-
ed its range from Paducah, Kentucky, into the Tennessee
River upstream to Pickwick Dam and had been collected in
1960 at Metropolis, Illinois, in the Ohio River (Table 1; Figs.
37 - 38). Establishment of the population in the Escambia
River would have required an extremely high rate of
downstream transport by natural means. In view of a short
larval life span, it is doubtful larvae could remain suspended
in the water column for periods of time sufficient to make
possible migration from the Ohio River to the lower Mississippi
River. Larvae may have attached byssally to logs or filamen-
tous algal mats and then be transported downstream but this
mode of travel seems doubtful since logs or mats would tend
to become entangled in vegetation near the river’s banks.

T ransport of Corbicula fiumiea by barges seems most
likely. Larvae could attach themselves to barges that in turn
could be towed downstream within the time necessary to link
the Ohio River infestation with that in the Escambia River.
However, once established in the lower reaches of the
Mississippi River, C. fluminea would have to reach the Escam-
bia River. While transport in the gastrointestinal tract of
migratory waterfoul is biologically impossible for any length
of time (Thompson and Sparks, 1977b), short term transport
by birds is likely.

Mackie (personal communication, 1979) has found that
sphaeriid bivalves may survive ingestion by migratory water-
fowl only if they are regurgitated after a few hours. McMahon
(1982) noted that larval Corbicula fluminea , byssally attach-
ed to sand grains enmeshed in filamentous green algae,
could become attached through entanglement on the feet of
wading birds or migratory ducks and subsequently be
transported to a new locality. However, McMahon (1982)
logically points out the long-distance transport does not seem
likely since these flights would result in death of the bivalves
by desiccation. Thus, transport of C. fluminea from the lower
Mississippi Valley to the Escambia River could have been
accomplished only by flights of short duration. Such flights
should have deposited C. fluminea in streams between the
Mississippi and the Escambia rivers. The chronological record
of C. fluminea's invasion (Table 1) and collection records do
not indicate that this happened. Hartfield and Cooper (1982),
for example, noted that in spite of C. fluminea’s presence in
this region of the United States for over 20 years it is still ab-
sent from the rivers and streams that empty into the lower
Mississippi River.

Chronological records for invasion of the United States

COUNTS: CORBICULA FLUMINEA IN THE UNITED STATES

by Corbicula fluminea (Table 1) indicate that in all probability
the Escambia River population was established by a long-
distance dispersal event, as was the Paducah, Kentucky,
population, and not by natural dispersal mechanisms.

McMahon (1982) believed that the Appalachian Moun-
tains are a significant barrier to the eastward migration of Cor-
bicula fluminea into streams below the fall line of the east
coast. However, he also suggests that invasion of the New
River at Glen Lyn, Virginia, by C. fluminea in 1975 was the
result of the species crossing the Appalachian Mountains
from North Carolina into Virginia. In view of C. fluminea' s long
history in the Kanawha River, and that streams confluence
with the New River in West Virginia, it is more likely that in-
festation resulted from an upstream migration either naturally
or anthropogenically mediated. There is little doubt that the
demonstrated rapidity of C. fluminea' s dispersal across the
United States argues against the proposal that large
physiographic features are dispersal barriers of any conse-
quence in the extension of the bivalve’s geographic range.
Neither the Mojave Desert nor the Rocky Mountains appeared
to present any significant barrier to dispersal.

Water temperature would appear to offer the most
significant barrier to northward dispersal of Corbicula
fluminea. Northernmost records for the species (Table 1) are
artificial situations in which C. fluminea lives in water warmed
by thermal effluent from industrial or power generating
facilities. The population in Wisconsin (Cummings and Jones,
1978; Fuller, 1978b; Mathiak, 1979) does not live in such in-
dustrially warmed waters. No further reports or collections
of the St. Croix River population have been made.

Water temperature has been implicated in mass mor-
talities of Corbicula fluminea populations. Sinclair and Isom
(1963) reported one in a Tennessee River population at Wolfe
Island (RM 195.2) in April 1961, and another in the
Cumberland River (RM 100) during this same period. Horn-
ing and Keup (1964) found a decline in the population in the
Cincinnati Reach of the Ohio River between 1962 and 1963
that they attributed to severe winters in which the river was
ice-covered for several days. However, this population has
since recovered and the severe winters of 1977 and 1978
have not significantly reduced their numbers (Taylor, personal
communication, 1979).

Sinclair and Isom (1963) reported that mass mortality
of Corbicula fluminea populations in the Tennessee River
blocked the intake pipe screens of the Chattanooga, Ten-
nessee, water treatment plant. This occurred during late
August and early September 1 962 and suggested that some
physicochemical factor, other than low temperature alone,
may have been responsible for the deaths. Britton and Mor-
ton (1979) reported four instances of mass mortalities in the
Trinity River, Texas, which they attributed to flooding.

The infestation of streams by Corbicula fluminea has
been implicated as a contributing factor in the demise of
native unionid mussels (Gardner et a/., 1976; Cherry et a!.,
1980). C. fluminea has been observed physically dislodging
unionid mussels from the substratum in the Savannah River
of Georgia and South Carolina (Fuller and Richardson, 1 977).
It is doubtful, however, that these activities play a significant

role in competition for space and hence a decline of native
unionid species. Instead, it would appear that C. fluminea is
able to adapt to a wide variety of habitats, especially those
stressed by pollution, r-selection regimes (MacArthur and
Wilson, 1967), while native bivalve species are more adapted
to stable, k-selection habitats.

Britton and Morton (1979) discussed r- versus k-
selection in Corbicula fluminea and found that these bivalves
have both r- and k-selective features but, in the majority of
cases, r-selection seems to be dominant. Sickel (1979) also
discussed what he believed to be a shift from r- to k-selection
strategy in the Altamaha River, Georgia, population.

While Britton and Morton (1979) concede that no species is
probably wholly r- or k-selected, it does seem likely that the
rapid growth of C. fluminea populations and the concomit-
tant decline of unionid populations is a function of r-selected
species (C. fluminea) cohabitating with k-selected species
(unionids) in waters that have become polluted. Since unionid
bivalves have a parasitic larva that must attach to a host fish
in order to metamorphose into an adult, polluted waters may
cause these fish to leave and the unionids, unable to com-
plete their life cycle, become moribund. Conversely, C.
fluminea’s life cycle requires no intermediate host species
and is therefore independent of fish for successful recruit-
ment of new individuals into the population. Sickel (1979)
hypothesized that a lack of one year-old C. fluminea in the
Altamaha River in 1976 may have been due to r-selective
pressures that favored individuals who diverted energy into
growth rather than reporduction. His alternative hypothesis
stated that growth rates of large cohorts in 1973 - 1974 was
decreased by crowding pressures, i.e. intense competition
for both food and space, thereby diverting energy into those
activities from reproductive processes (Sickel, 1979).

Kraemer (1979) noted that in an altered habitat, such
as the Arkansas River, physicochemical factors seemed to
be far more important to the success of Corbicula fluminea
than they would in an unaltered habitat. In an unaltered
habitat, such as the Buffalo River, Arkansas, interspecific
competition between unionids and C. fluminea may shift bet-
ween r- and k-selective strategies from one season of the year
to another.

In either case, an r-selective regime would operate
against the generally k-selected unionid mussels and allow
the generally r-selected Corbicula fluminea to successfully in-
vade a new habitat. The r-, k-selection scheme seems to of-
fer the best explanation for the phenominal success of C.
fluminea' s invasion of North American waters.

Man has unquestionably been primarily responsible
for the rapid transcontinental dispersal of Corbicula fluminea
in the United States. They may be transported long distances
in sand and gravel dredged for making concrete (Sinclair and
Isom (1963). They are harvested in California to be sold as
bait to sport fisherman (Fox, 1970) or in pet shops (Abbott,
1975). A clam purchased from a bait shop or pet store could
easily be discarded in a local stream. The shock of hitting
the water’s surface, when used as bait, could easily stimulate
gravid C. fluminea to release brooded veliger larvae resulting
in the infestation of a previously uninfested stream (Clarke,

CORBICULA SYMPOSIUM

Personal communication, 1982). The common habit of
fishermen of throwing unused bait overboard after fishing
would also be sufficient to infest a stream or lake.

The use of Corbicula fluminea by man has surely made
man the vetor responsible for the current zoogeography of
the species in North America.

ACKNOWLEDGEMENTS

I would like to thank the following curators and curatorial
assistants who generously allowed me to examine their collections
or provided me with information on their institutional holdings: Drs.
George M. Davis and Robert Robertston (ANSP), Dr. Eugene V. Coan
(CAS, SU), Dr. R. A. Davis (CMNH), Dr. Karl A. Hoehn (MNHD), Mr.
Russell Jensen (DMNH), Dr. Alan Solem and Mr. Kenneth Ember-
ton (FMNH), Mr. Kurt Auffenburg (FSM), Dr. William J. Voss (FWM),
Mrs. Constance E. Boone (HMNS), Dr. Everett D. Cashatt (ISM), Dr.
Ralph W. Taylor (MUMC), Mr. Paul D. Hartfield (MMNS), Dr. Ruth
D. Turner (MCZ), Dr. C. O. Minckley (MNA), Mrs. Muriel F. I. Smith
(NMNS), Dr. Rowland M. Shelley (NCSM), Dr. David H. Stansbery
(OSUM), Dr. Willard D. Hartman (PMNH), Mr. Anthony D’Attilio
(SDMNH), Carey R. Smith (SBMNH), Mr. Mark B. DuBois (SBSK),
Dr. Sieved Rohwer (TBWSM), Drs. Arthur H. Clarke and Joseph
Rosewater (USNM ), Ms. Danita Brandt (UCGM), and Dr. William
D. Shepard (UOM). Thanks are also due C. Dale Snow (ODFW), Alan
C. Buchanan and Ron D. Oesch (MDC), and Robert Singleton
(ADPC). I would also like to thank Mr. Gary A. Coovert, Dayton, Ohio;
Mr. David Metty, Cincinnati, Ohio; Dr. R. Tucker Abbott, Melbourne,
Florida; Steven L. Coon, University of Maryland; Dr. Richard L.
Reeder, University of Tulsa; Dr. Robert Bullock, University of Rhode
Island; and Mr. James J. Hall, Duke Power Company, for sharing
their zoogeographic records with me.

Thanks are also due Dr. G. L, Mackie, University of Guelph,
Ontario, and Dr. A. H. Clarke, USNM, for their comments and obser-
vations. I would also like to thank Dr. Louise R. Kraemer, University
of Arkansas, and Dr. Joseph C. Britton, Texas Christian University,
for their efforts to have this work presented at the Symposium.

Special thanks are due Mr. John R. Casadevall for his com-
puter expertise.

This research was supported in part by a grant from the Karl
P. Schmidt Fund, Field Museum of Natural History, Chicago. This
research is a portion of a doctoral dissertation presented to the Facul-
ty of the College of Marine Studies of the University of Delaware.

LITERATURE CITED

Abbott, R. T. 1975. American Seashells, 2nd. Ed. Van Nostrand —
Reinhold Co. (New York). 633 pp.

Abbott, R. T. 1975. Beware the Asiatic freshwater clam. Tropical Fish
Hobbyist 23:15.

Aldridge, D. W. and R. F. McMahon. 1978. Growth, fecundity and
bioenergetics in a natural population of the Asiatic freshwater
clam, Corbicula manilensis Philippi, from north central Texas.
Journal of Molluscan Studies 44:49-70.

Anderson, K. B., C. M. Thompson, R. E. Sparks and A. A. Paparo.
1976. Effects of potassium on adult Asiatic clams, Corbicula
manilensis. Illinois Natural History Survey, Biological Notes No.
98. 7 pp.

Baker, J. R. 1978. Extensions of the known range of Corbicula
manilensis (Pelecypoda: Corbiculidae) in Texas. Southwestern
Naturalist 23(3):536.

Bates, J. M. 1962a. The impact of impoundment on the mussel fauna

of Kentucky Reservoir, Tennessee River. American Midland
Naturalist 68(1):232-236.

Bates, J. M. 1962b. Extension of the known range of Corbicula
fluminea within the Ohio drainage. Nautilus 76(2):35-36.

Behrens, D. W. 1975. Use of disposable beverage containers by the
freshwater clam Corbicula. Nautilus 89(1): 13.

Bequaert, J. C. and W. B. Miller. 1973. The Mollusks of the Arid
Southwest with an Arizona Check List. University of Arizona
Press (Tuscon). xvi + 271 pp.

Bickel, D. 1966. Ecology of Corbicula manilensis Philippi in the Ohio
River at Louisville, Kentucky. Sterkiana 23:19-24.

Blankenship, S. and D. R. Crockett. 1972. Changes in the freshwater
mussel fauna of the Rockcastle River at Livingston, Kentucky.
Transactions of the Kentucky Academy of Science 33:37-39.

Boozer, A. C. and P. E. Mirkes. 1979. Observations on the finger-
nail clam, Musculum partumeium (Pisidiidae), and its associa-
tion with the introduced Asiatic clam, Corbicula fluminea.
Nautilus 93(2):73-83.

Branson, B. A. and D. L. Batch. 1969. Notes on exotic mollusks in
Kentucky. Nautilus 82(3): 102-1 06.

Branson, B. A. and D. L. Batch. 1971. Annotated distribution records
for Kentucky Mollusca. Sterkiana 43:1-9.

Britton, J. C. 1982. Biogeography and ecology of the Asiatic clam,
Corbicula, in Texas. IN: Proceedings of the Symposium on Re-
cent Benthological Investigations in Texas and Adjacent States.
Aquatic Sciences Section. J. R. Davis, ed. pp. 21-31. Texas
Academy of Science.

Britton, J. C. and B. Morton. 1979. Corbicula in North America: the
evidence reviewed and evaluated. Proceedings of the First In-
ternational Corbicula Symposium. J. C. Britton, ed. pp.
250-287. Texas Christian University Research Foundation
(Ft. Worth).

Britton, J. C. and C. E. Murphy. 1977. New records and ecological
notes for Corbicula manilensis in Texas. Nautilus 91(1):20-22.

Burch, B. L. 1978. Asian clam, Corbicula, threatens Hawaii. Nautilus
92(1):54-55.

Burch, J. Q. 1944. Checklist of west American Mollusks. Minutes,
Conchological Club Southern California 38:18.

Carlton, J. T. 1973. Corbicula in San Francisco, California. Nautilus
87(3):87.

Clarke, A. H. 1981. Corbicula fluminea in Lake Erie. Nautilus
95(2):83-84.

Clench, W. J. 1970. Corbicula manilensis Philippi in lower Flordia.
Nautilus 84(1): 36.

Clench, W. J. 1971. Corbicula manilensis (Philippi) in Oklahoma.
Nautilus 85(4): 145.

Clench, W. J. and D. H. Stansbery. 1969. Corbicula manilensis Philip-
pi in the Nolichucky River. Nautilus 82(4): 146.

Coney, C. C., G. C. DeCamp, and M. E. Harrelson. 1983. The Bivalvia
of the Great Pee Dee River in Florence County, South
Carolina. Bulletin of the South Carolina Academy of Science
45:87.

Cooper, C. M. and V. W. Johnson. 1980. Bivalve Mollusca of the
Yalobusha River, Mississippi. Nautilus 94(1):22-24.

Counts, C. L., III. 1981a. Corbicula fluminea (Bivalvia: Corbiculidae)
in British Columbia. Nautilus 95(1): 12-1 3.

Counts, C. L., III. 1982b. Corbicula fluminea (Muller) on the Delmar-
va Peninsula. Veliger 24(2): 1 87-1 88.

Crumb, S. E. 1977. Macrobenthos of the tidal Delaware River be-
tween Trenton and Burlington, New Jersey. Chesapeake
Science 18(3):253-265.

Cummings, S. E. and J. A. Jones. 1978. Occurrence of Corbicula
manilensis Phillipi (sic) in the lower Minnesota River. Journal

COUNTS: CORBICULA FLUMINEA IN THE UNITED STATES

of the Minnesota Academy of Science 34(3): 13-1 4.

Diaz, R. J. 1974. Asiatic clam, Corbicula manilensis (Philippi) in the
tidal James River, Virginia. Chesapeake Science 15
(2):1 18-120.

Dresler, R. J. and R. L. Cory. 1980. The Asiatic clam, Corbicula
fluminea (Muller), in the tidal Potomac River, Maryland.
Estuaries 3:150-151.

Dundee, D. S. 1974. Catalog of introduced molluscs of eastern North
America (north of Mexico). Sterkiana 55:1-37.

Dundee, D. S. and H. A. Dundee. 1958. Extension of known ranges
of four Mollusca. Nautilus 72:51-53.

Dundee, D. S. and W. J. Harman. 1963. Corbicula fluminea (Muller)
in Louisiana. Nautilus 77(1 ):30.

Eagleson, K. and E. L. Morgan. 1977. Growth rate of Corbicula
manilensis in the Clinch River, Tennessee. Associaton of
Southeastern Biologists, Bulletin 24(2): 48.

Eckbald, J. W. 1975. The Asiatic clam, Corbicula, in the upper
Mississippi River. Nautilus 89(1 ):4.

Eng, L. L. 1975. Biological studies of the Delta-Mendota Canal,
Central Valley Project, California II. California Academy of
Sciences (San Francisco). 178 pp.

Eng, L. L. 1 976. A note on the occurrence of a symbiotic oligochaete,
Chaetogaster limnaei, in the mantle cavity of the Asiatic clam,
Corbicula fluminea. Veiiger 19(4):208.

Eng, L. L. 1979. Population dynamics of the Asiatic clam, Corbicula
fluminea (Muller) in the concrete-lined Delta-Mendota
Canal of central California. IN: Proceedings of the First Inter-
national Corbicula Symposium. J. C. Britton, ed. pp. 39-68.
Texas Christian University Research Foundation (Ft. Worth).

Evans, L. P., C. E. Murphy, J. C. Britton and L. W. Newiand. 1979.
Salinity relationships in Corbicula fluminea (Muller, 1774). IN:
Proceedings of the First International Corbicula Symposium,
J. C. Britton, ed. pp. 193-214. Texas Christian University
Research Foundation (Ft. Worth).

Fast, A. W. 1971. The invasion and distribution of the Asiatic clam
(Corbicula manilensis) in a southern California reservoir.
Bulletin of the Southern California Academy of Science
70(3):91-98.

Fetchner, F. K. 1962. Corbicula fluminea (Muller) from the Ohio River.
Nautilus 75(3): 126.

Fisk, L. O. and T. R. Doyle. 1962. Sacramento River Water Pollution
Survey. Appendix D. Benthic Biology. Bulletin No. Ill, Cali-
fornia Department of Water Resources. 78 pp.

Fontanier, C. E. 1982. The distribution of Corbicula (Bivalvia: Cor-
biculidae) in the Brazos River system. Texas Journal of Science
34(1 ):5-1 6.

Fox, R. 0. 1969. The Corbicula story: a progress report. Second An-
nual Meeting, Western Society of Malacologists. 1 1 pp.

Fox, R. O. 1970a. The Corbicula story: chapter two. Third Annual
Meeting, Western Society of Malacologists. 10 pp.

Fox, R. O. 1970b. Corbicula in Baja California. Nautilus 83(4): 145.

Fox, R. O. 1971. The Corbicula story: chapter three. Fourth Annual
Meeting, Western Society of Malacologists. 5 pp.

Fox, R. O. 1972. The Corbicula story: chapter four. Fifth Annual
Meeting, Western Society of Malacologists. 5 pp.

Fuller, S. L. H. 1974. Macroinvertebrates and quantitative macroin-
vertebrate samples. IN: Cooper River Survey 1973 for the E.
I. duPont de Nemours & Company. Department of Limnology,
Academy of Natural Sciences of Philadelphia, pp. 49-69,
143-145.

Fuller, S. L. H. 1976. IX. Biological, ecological, and environmental
characteristics of the site. IN: Final Report submitted to the
National Commission on Water Quality. J. A. Hendrickson, Jr.,

L. F. Berseth, T. E. Walton III and S. L. H. Fuller, eds. Depart-
ment of Limnology, Academy of Natural Sciences of
Philadelphia.

Fuller, S. L. H. 1978a. Fresh-water Mollusks. IN: An Annotated
Checklist of the Biota of the Coastal Zone of South Carolina.
R. G. Zongmark, ed. pp. 136-152. University of South Carolina
Press (Columbia).

Fuller, S. L. H. 1978b. Freshwater mussels (Unionidae) of the Upper
Mississippi River. Academy of Natural Sciences of
Philadelphia. 401 pp.

Fuller, S. L. H. and M. J. Imlay. 1976. Spatial competition between
Corbicula manilensis (Philippi), the Chinese clam (Corbiculi-
dae), and freshwater mussels (Unionidae) in the Waccamaw
River Basin of the Carolinas (Mollusca: Bivalvia). Association
of Southeastern Biologists, Bulletin 23(2):60.

Fuller, S. L. H. and C. E. Powell. 1973. Range extension of Corbicula
manilensis (Philippi) in the Atlantic drainage of the United
States. Nautilus 87(2):59.

Fuller, S. L. H. and J. W. Richardson. 1977. Amensalistic competi-
tion between Corbicula manilensis (Philippi) and the freshwater
mussels (Unionidae) in the Savannah River of Georgia and
South Carolina (Mollusca: Bivalvia). Association of
Southeastern Biologists, Bulletin 24(2):52.

Gainey, L. F., Jr. 1 978a. The response of the Corbiculidae (Mollusca:
Bivalvia) to osmotic stress: the organismal response.
Physiological Zoology 51(1):68-78.

Gainey, L. F., Jr. 1978b. The response of the Corbiculidae (Mollusca:
Bivalvia) to osmotic stress: the cellular response. Physiological
Zoology 51(1 ):79-91 .

Gardner, J. A., W. R. Woodall, Jr., A. A. Staats, Jr. and J. F. Napoli.
1976. The invasion of the Asiatic clam (Corbicula manilensis)
in the Altamaha River, Georgia. Nautilus 90(3): 1 17-125.

Gifford, C. A. 1975. The Asiatic clam, Corbicula sinensis, in Lake
Oklawaha, Florida. Association of Southeastern Biologists,
Bulletin 21(2):56.

Goss, L. B. and C. Cain, Jr. 1975. Power plant condenser and ser-
vice water system fouling by Corbicula, the Asiatic clam. IN:
Biofouling Control Procedure Workshop. Johns Hopkins Univer-
sity, Electric Power Research Institute, Ecological Analysts,
Inc., and Maryland Power Plant Siting Program (Baltimore).

pp. 11-22.

Grantham, B. J. 1967. The Asiatic clam in Mississippi. Mississippi
Water Resources Conference Proceedings, pp. 81-85.

Gunning, G. E. and R. D. Suttkus. 1966. Occurrence and distribu-
tion of the Asiatic clam, Corbicula leana, in the Pearl River,
Louisiana. Nautilus 79(4): 11 3-1 16.

Hanna, G. D. 1966. Introduced mollusks of western North America.
California Academy of Sciences Occasional Papers 48: 1 -1 08.

Hartfield, P. D. and C. M. Cooper. 1982. Notes on the distribution
of the Asiatic clam (Corbicula manilensis) in Mississippi. Journal
of the Mississippi Academy of Science 27:63.

Hartfield, P. D. and C. M. Cooper. 1983. Distribution of Corbicula
fluminea, the Asiatic clam, in Mississippi. Nautilus 97(2):66-68.

Harvey, R. S. 1981 . Recolonization of reactor cooling water system
by the Asiatic clam Corbicula fluminea. Nautilus 95(3): 131-1 36.

Hazel, C. R. and D. W. Kelley. 1 966. Zoobenthos of the Sacramen-
to - San Joaquin Delta. IN: Ecological Studies of the Sacra-
mento-San Joaquin Estuary. Part I. D. W. Kelley, ed. California
Department of Fish and Game, Fish Bulletin 133:113-133.

Heard, W. H. 1964. Corbicula fluminea in Florida. Nautilus
77(3): 105-1 07.

Heard, W. H. 1966. Further records of Corbicula fluminea (Muller)
in the southern United States. Nautilus 79(4): 142-143.

CORBICULA SYMPOSIUM

Heinsohn, G. E. 1 958. Life history and ecology of the freshwater clam,
Corbicula fluminea. M. A. Thesis, University of California,
Berkeley, 64 pp.

Hillis, D. M. and J. C. Patton. 1982. Morphological and electrophoretic
evidence for two species of Corbicula (Bivalvia: Corbiculidae)
in North America. American Midland Naturalist 108(1):74-80.

Horne, F. R. and S. McIntosh. 1979. Factors influencing distribu-
tion of mussels in the Blanco River of central Texas. Nautilus
94(4): 11 9-1 33.

Horning, W. B. and L. Keup. 1964. Decline of the Asiatic clam in
Ohio River. Nautilus 78(1):29-30.

Hubricht, L. 1963. Corbicula fluminea in the Mobile River. Nautilus
77(1 ):31.

Hubricht, L. 1965. Corbicula manilensis in the Alabama River, Nautilus
78(3): 106.

Hubricht, L. 1966. Corbicula manilensis (Philippi) in the Alabama River
system. Nautilus 80(1)32-33.

Ingram, W. H. 1948. The larger freshwater clams of California,
Oregon and Washington. Journal of Entomology and Zoology
40:72-92.

Ingram, W. H. 1957. Use and value of biological indicators of pollu-
tion: fresh-water clams and snails. IN: Biological Problems in
Water Pollution, pp. 94-135.

Ingram, W. H. 1959. Asiatic clams as potential pests in California
water supplies. Journal of the American Water Works Associa-
tion 51 :363-37Q.

Isom, B. G. 1971 . Mussel fauna in Fort Loudoun Reservoir Tennessee
River, Knox County, Tennessee, in December 1970.
Malacological Review 4:127-130.

Isom, B. F. 1972. Mussels in the unique Nickajack Dam construc-
tion site, Tennesse River, 1965. Malacological Review 5:4-6.

Isom, B. G. 1974. Mussels of the Green River, Kentucky. Transac-
tions of the Kentucky Academy of Science 35(1-2):55-57.

Isom, B. G. and P. G. Yockley. 1 968. The mussel fauna of Duck River
in Tennessee, 1965. American Midland Naturalist 80(1):34-42.

Isom, B. G., P. G. Yockley, and C. H. Gooch. 1973. Mussels of the
Elk River basin in Alabama and Tennessee - 1965-1967.
American Midland Naturalist 89(2):437-442.

Jenkinson, J. J. 1979. The occurrence and spread of Corbicula
manilensis in east-central Alabama. Nautilus 94(4): 149-1 53.

Joy, J. E. and L. E. McCoy. 1975. Comparisons of shell dimensions
and viscera mass weights in Corbicula manilensis (Philippi,
1844). Nautilus 89(2):51-54.

Keup, R. H., W. B. Horning and W. M. Ingram. 1963. Extension of
the range of the Asiatic clam to Cincinnati Reach of the Ohio
River. Nautilus 77(1):18-24.

King, R. H. and A. C. Miller. 1982. Proposed man-made lotic habitat
in the Tombigbee River near Columbus, MS. Journal of the
Mississippi Academy of Science 27:63.

Klippel, W. E. and P. W. Parmalee. 1979. The naiad fauna of Lake
Springfield, Illinois: an assessment after two decades. Nautilus
94(4): 189-1 97.

Kool, S. P., C. C. Coney, D. H. Stanbery, and A. E. Bogan. 1981.
Fresh-water mollusks of the Little Pee Dee River in South
Carolina. Bulletin of the South Carolina Academy of Science
43:93.

Kraemer, L. R. 1 976. An evaluation of the effects of dredging within
the Arkansas River Navigation System, V. IV., The effects
upon the benthic associations. Arkansas WRRC Publication
No. 46. pp. 1-357.

Kraemer, L. R. 1978a. Corbicula fluminea (Bivalvia: Sphaeriacea);
the functional morphology of its hermaphroditism. Bulletin of
the American Malacological Union 1978:40-49.

Kraemer, L. R. 1978b. Discovery of a “new” kind of statocyst in the
Asian clam, Corbicula: histological analysis and behavioral
evaluation. American Zoologist 18(3):647.

Kraemer, L. R. 1979. Corbicula (Bivalvia: Sphaeriacea) vs. indigenous
mussels (Bivalvia: Unionacea) in U. S. rivers: a hard case for
interspecific competition? American Zoologist 19:1085-1096.

Kraemer, L. R. 1980. Neuroanatomical correlates of sexual reproduc-
tion in monoecious and dioecious bivalved mollusks. American
Zoologist 20(4): 891.

Kraemer, L. R. and S. Lott. 1977. Microscopic anatomy of the visceral
mass of Corbicula (Bivalvia: Sphaeriacea). Bulletin of the
American Malacological Union 1977:48-56.

Kraemer, L. R. and M. Gordon. 1980. Size and success: Corbicula
(Mollusca: Bivalvia) secret? American Zoologist 20(4):882.

Kuckyr, R. J. and M. F. Vidrine. 1975. Some clams (Mollusca:
Bivalvia) from the Tensas River in Madison Parish, Louisiana.
Association of Southeastern Biologists, Bulletin 22(2):61 .

Leard, R. L., B. J. Grantham and G. F. Pessoney. 1980. Use of
selected freshwater bivalves for monitoring organo-chlorine
pesticide residues in major Mississippi stream systems.
Pesticide Monitoring Journal 14(2):47-52.

Lewis, R. B. and J. R. Brice. 1980. A comparison of the past and
present mussel fauna of the Kankakee River of Illinois. Natural
History Miscellania, Chicago Academy of Science, No. 211.
7 pp.

MacArthur, R. H. and E. O. Wilson. 1967. The Theory of Island
Biogeography. Princeton University Press, Princeton, New
Jersey, xi + 203 pp.

Markham, S. L., C. H. Hocutt and J. R. Stauffer, Jr. 1980. The crayfish
(Decapoda: Astacididae and Gambaridae) and the freshwater
mussels (Mollusca: Pelecypoda) of the lower New River,
Virginia and West Virginia. Natural History Miscellania, Chicago
Academy of Science, No. 208. 1 1 pp.

Mathiak, H. A. 1979. A river survey of the unionid mussels of Wisconsin
1973-1977. Sand Shell Press. Horicon, Wisconsin. 75 pp.

McLeod, M. J. and D. M. Sailstad. 1980. An electrophoretic study
of Corbicula fluminea (Bivalvia: Corbiculidae) in the Catawba
River. Bulletin of the American Malacological Union 1980:17-19.

McMahon, R. F. 1977. Shell size-frequency distribution of Corbicula
manilensis Philippi from a clam-fouled steam condenser.
Nautilus 91(2):54-59.

McMahon, R. F. 1 979. Response to temperature and hypoxia in the
oxygen consumption of the introduced Asiatic freshwater clam
Corbicula fluminea (Muller). Comparative Biochemistry and
Physology 63 A: 383-388.

McMahon, R. F. 1982. The occurrence and spread of the introduced
Asiatic freshwater clam, Corbicula fluminea (Muller) in North
America: 1924 - 1982. Nautilus 96(4): 134-1 41 .

Metcalf, A. L. 1 966. Corbicula manilensis in the Mesilla Valley of Texas
and New Mexico. Nautilus 80(1): 16-20.

Minckiey, W. J., J. E. Johnson, R. N. Rinne and S. E. Willoughby.
1970. Foods of buffalofishes, genus Ictiobus, in central
Arizona. Transactions of the American Fisheries Society
99:333-342.

Morris, J. S. and R. W. Taylor. 1978. A survey of the freshwater

mussels (Bivalvia: Unionidae) of the Kanawha River of West
Virginia. Nautilus 92(4): 153-1 55.

Murray, H. D. 1971a. New records of Corbicula manilensis (Philippi)
in Texas. Nautilus 85(1):35-36.

Murray, H. D. 1971b. Fresh-water mussels of Lake L. B. J., Texas.
Bulletin of the American Malacological Union 1971:36-37.

Murray, H. D. 1978. Freshwater mussels of Lake Corpus Christi,
Texas. Bulletin of the American Malacological Union 1 978:5-6.

COUNTS: GORBICULA FLUMiNEA IN THE UNITED STATES

Nichols, B. L. and R. B. Domermuth. 1981. Appearance of the Asiatic
clam, Corbicula fluminea, in the Susquehanna River. Pro-
ceedings of the Pennsylvania Academy of Science 55: 1 81 -1 82.

O’Kane, K. D., J. C. Britton and D. R. Coldiron. 1977. New distribu-
tional records of Corbicula manilensis (Corbiculidae) in the
south central United States. Southwestern Naturalist
22:397-400.

ORSANCO. 1966. Eighteenth Yearbook. Ohio River Valley Water
Sanitation Commission (Cincinnati).

Parmalee, P. W. 1965. The Asiatic clam (Corbicula) in Illinois. Tran-
sactions of the Illinois Academy of Science 58:38-45.

Pojecta, J. 1964. Notes on the extension of the known range of the
Asiatic clam Corbicula fluminea (Muller) in Ohio River. Ohio
Journal of Science 64(6):428.

Pool, D. and J. D. McCullogh. 1979. The Asiatic clam, Corbicula
manilensis, from two reservoirs in eastern Texas. Nautilus
93(1 ):37.

Prokopovich, N. 1969. Deposition of clastic sediments by clams. Jour-
nal of Sedimentary Petrology 39 :891-901.

Prokopovich, N. 1970. Organic life in the Delta - Mendota Canal,
California. Proceedings of the Symposium on Hydrobiology,
No. 8:191-203.

Prokopovich, N. and D. J. Hebert. 1965. Sedimentation in the Delta
- Mendota Canal. Journal of the American Water Works
Association 57:375-382.

Richardson, W. M., J. A. St. Amant, L. J. Bottroff and W. L. Parker.
1970. Introduction of blue catfish into California. California Fish
and Game Bulletin 56(4):31 1-312.

Rinne, J. N. 1974. The introduced Asiatic clam, Corbicula, in cen-
tral Arizona reservoirs. Nautilus 88(2):56-61.

Rodgers, J. H., Jr., D. S. Cherry, J. R. Clark, K. L. Dickson and J.
Cairns, Jr. 1977. The invasion of the Asiatic clam, Corbicula
manilensis in the New River, Virginia. Nautilus 91(2):43-46.

Rodgers, J. H., Jr., D. S. Cherry, K. L. Dickson and J. Cairns, Jr.
1978. Influence of thermal effluent on population dynamics
of Asiatic clam (Corbicula manilensis) in the New River, Virginia
(Mollusca: Bivalvia). Association of Southeastern Biologists,
Bulletin 25(2):48.

Schneider, R. F. 1967. Range of the Asiatic clam in Florida. Nautilus
81(2):68-69.

Scott-Wasilk, J., G. G. Downing and J. S. Lietzow. 1982. Occurrence
of the Asiatic clam Corbicula fluminea in the Maumee River
and western Lake Erie. Journal of Great Lakes Research
9(1 ):9-1 3.

Shealy, R. M. 1976. The natural history of the Alabama map turtle,
Graptemys pulchra Baur, in Alabama. Bulletin of the Florida
State Museum, Biological Sciences 21(2):47-1 11.

Sickel, J. B. 1973. New record of Corbicula manilensis (Philippi) in
the southern Atlantic slope region of Georgia. Nautilus
87(1 ):1 1-12.

Sickel, J. B. 1976. Population growth and reproductivity of Corbicula
manilensis (Philippi) in the Altamaha River, Georgia (Bivalvia:
Corbiculidae). Association of Southeastern Biologists, Bulletin
23(2):96.

Sickel, J. B. 1979. Population dynamics of Corbicula in the Altamaha
River, Georgia. IN: Proceedings of the First International Cor-
bicula Symposium. J. C. Britton, ed. pp. 69-80. Texas Christian
University Research Foundation (Ft. Worth).

Sickel, J. B. and C. C. Chambers. 1982. Commercial mussel and
Asiatic clam fishery evaluation. Kentucky Department of Fish
and Wildlife Research, N.O.A.A. National Marine Fisheries
Service Project No. 2-367-R. Final Report, vii + 77 pp.

Sickel, J. B., D. W. Johnson, G. T. Rice, M. W. Heyn and P. K.

Wellner. 1981. Asiatic clam and commerical fishery evalua-
tion. Kentucky Department of Fish and Wildlife, N.O.A.A., Na-
tional Marine Fisheries Service Project No. 2-344-R. vii + 83 pp.

Sickel, J. B. and M. B. Lyles. 1981 . Chaetogaster limnaei
(Oligochaeta: Naididae) inhabiting the mantle cavity of the
Asiatic clam, Corbicula fluminea, in Barkley Lake, Kentucky.
Veliger 23(4):361-362.

Siegfried, C. A., M. E. Kopache and A. W. Knight. 1980. The ben-
thos of a portion of the Sacramento River San Francisco Bay
estuary) during a dry year. Estuaries 3(4):296-307.

Sinclair, R. M. 1963. The effects of an introduced clam (Corbicula)
on water quality in the Tennessee River Valley. IN: Proceedings
of the Second Industrial Waste Conference, Vanderbilt Univer-
sity and Tennessee Department of Public Health, Tennessee
Stream Pollution Control Board. 24 pp.

Sinclair, R. M. 1964. Clam pests in Tennessee water supplies. Journal
of the American Water Works Association 56:592-599.

Sinclair, R. M. and W. M. Ingram. 1961 . A new record for the Asiatic
clam in the United States, the Tennessee River. Nautilus
74(3):1 14-118.

Sinclair, R. M. and B. G. Isom. 1 961 . A preliminary report on the in-
troduced Asiatic clam Corbicula in Tennessee. Tennessee
Stream Pollution Control Board, Tennessee Department of
Public Health. 31 pp.

Sinclair, R. M. and B. G. Isom. 1963. Further studies on the introduced
Asiatic clam (Corbicula) in Tennessee. Ten nessee Stream Pollu-
tion Control Board, Tennessee Department of Public Health,
v + 79 pp.

Stein, C. B. 1962. An extension of the known range of the Asiatic
clam, Corbicula fluminea (Muller), in the Ohio and Mississip-
pi Rivers. Ohio Journal of Science 62(6):326-327.

Stotts, V. D., K. D’Loughy and D. B. Stotts. 1977. Waterfowl habitat.
Maryland Department of Natural Resources Wildlife Ad-
ministration Report 77-2. 6 pp.

Taylor, R. W. 1980a. Freshwater bivalves of Tygart Creek, north-
eastern Kentucky. Nautilus 94(2):89-91.

Taylor, R. W. 1980b. A survey of the freshwater mussels of the Ohio
River from Greenup Locks and Dam to Pittsburgh, Pa. U. S.
Army Corps of Engineers (Huntington District), v + 71 pp.

Taylor, R. W. 1982. The freshwater mussels (naiads) of Big Indian
Creek, a small tributary of the Ohio River (Bivalvia: Unionidae).
Nautilus 6(2):66-68.

Taylor, R. W. and C. L. Counts, III. 1977. The Asiatic clam, Corbicula
manilensis, as a food of the Northern Raccoon, Procyon lotor.
Nautilus 91(1):34.

Thomas, N. A. and K. M. MacKenthum. 1964. Asiatic clam infesta-
tion at Charleston, West Virginia. Nautilus 78(1):28-29.

Thomerson, J. E. and D. G. Myer. 1970. Corbicula manilensis: range
extension upstream in the Mississippi River. Sterkiana 37:29.

Thompson, C. M. and R. E. Sparks. 1977a. The Asiatic clam, Cor-
bicula manilensis, in the Illinois River. Nautilus 91(1 ):34-36.

Thompson, C. M. and R. E. Sparks. 1977b. Improbability of disper-
sal of adult Asiatic clams, Corbicula manilensis, via the in-
testinal tract of migratory waterfowl. American Midland
Naturalist 98:219-223.

Tilly, L. C., J. C. Corey and N. E. Bibler. 1978. Response of the Asiatic
clam, Corbicula fluminea to gamma radiation. Health Physics
35:704-708.

Trama, F. B. 1982. Occurrence of the Asiatic clam, Corbicula fluminea
in the Raritan River, New Jersey. Nautilus 96(1):6-8.

van der Schalie, H. 1973. The mollusks of the Duck River drainage
in central Tennessee. Sterkiana 52:45-55.

Vidrine, M. F. and M. S. DeRouen. 1976. Note on the bivalves

CORBICULA SYMPOSIUM

(Sphaeriacea: Cobiculidae and Sphaeriidae and Unionacea:
Margaratiferidae) of the Bayou Teche system in Louisiana.
Association of Southeastern Biologists, Bulletin 23(2): 103.
White, D. S. 1979. The effect of lake-level fluctuation on Corbicula
and other pelecypods in Lake Texoma, Texas and Oklahoma.
IN: Proceedings of the First International Corbicula Symposium.
J. C. Britton, ed. pp. 81-88. Texas Christian University Research
Foundation (Ft. Worth).

White, D. S and S. J. White. 1977. The effect of reservoir fluctua-

tions on populations of Corbicula manilensis (Pelecypoda: Cor-
biculidae). Proceedings of the Oklahoma Academy of Science
57:106-109.

Williams, J. C. 1969. Mussel fishery investigation: Tennessee, Ohio,
and Green rivers: Final Report. Kentucky Department of Fish
and Wildlife Research Project No. 4-1 9-R (Frankfort). 106 pp.

Zeto, M. A. 1982. Notes on freshwater mussels (Unionidae) of the
upper Monongahela River basin, West Virginia. Nautilus
96(4):127-129.

COUNTS: CORBICULA FLUMINEA IN THE UNITED STATES

Table 1. Chronology of invasion of United States waters by Corbicula fluminea. Figures 26-33 depict the course of the invasion. All records
are for only the first documented detection of C. fluminea in a body of water in a particular state.

1924

British Columbia

Vancouver Island near Nanaimo/
USNM 363020

1938

Washington

Columbia River/ Knapton/ CAS 32360

1945

California

Sacramento River/ N of Pittsburg/
CAS 42926

1946

California

Mayberry Cut/ CAS 32369

Mokelumne Aqueduct/ Middle River/

CAS 32237

Potatoe Slough/ Near Lodi/

CAS 37271

San Joaquin River/ Canal E fo Los
Banos/ SBMNH 32378

1950

California

Delta-Mendota Canal/ Near Tracy/
Eng, 1975

1952

California

Tolumne River/ CAS 43671

Washington

Snake River/ USNM 595265

1953

California

All American Canal/ Imperial Valley/

CAS

Colorado Aqueduct/ Mecca/ USNM
613968

1956

Arizona

Salt River/ Papgo Park, Phoenix/
Dundee and Dundee, 1958

1957

Kentucky

Ohio River/ W of Paducah/ Sinclair
and Isom, 1961

1958

California

Salton Seal Imperial Co./ Sinclair
and Isom, 1961

1959

Idaho

Snake River/ Idaho - Washington
State line/ Ingram, 1959

Nevada

Colorado River/ Lake Meade/

Ingram, 1959

Oregon

Smith River/ Douglas Co./ ODFW
Williamette River/ Multnomah Co./
Ingram, 1959

Tennessee

Tennessee River/ Below Pickwick
Dam/ USNM 636118

1960

California

San Jacinto River/ San Jacinto
Reservoir/ Fox, 1970

Stanislaus River/ SE of Ripon/

OSUM 23646

Illinois

Ohio River/ Metropolis/ FMNH

103678

Florida

Escambia River/Near Century/

Schneider, 1967

1961

Alabama

Tennessee River/ Wheeler Reservoir/
NMNS 20569

Arizona

Colorado River/ Lake Martinez/
SBMNH 4456

California

Russian River/ Somona Co./ CAS
37639

Florida

Appalachicola River/ Near
Apalachicola/ Schneider, 1967

Kentucky

Green River/ Above Paradise/ Bates,
1962b

Louisiana

Calcasieu River/ RM 66/ Dundee

and Harman, 1963

Mississippi River/ Unknown/ OSUM

4416

Tennessee

Cumberland River/ Near Stone River
/ Sinclair and Isom, 1963

Alabama

Escambia River/ Near Century/
Hubricht, 1963

Mobile River/ Hubricht, 1963

Arizona

Agua Fria River/ NE of Rock Springs
/Bequaert and Miller, 1973

California

South Bay Aqueduct/ Alamenda Co./
Prokopovich, 1968

Florida

Withlacoochee River/ Inglis/ MCZ
237952

Louisiana

Bayou Magasille/ Assumption Par./
Dundee and Harman, 1963

Bayou Sorrel/ Iberville Par./ Dundee
and Harman, 1963

Ohio

Ohio River/ Pojeta, 1966

Kentucky

Mississippi River/ Wickliffe/ FMNH
123601

Louisiana

Pearl River/ Wilson’s Slough/
Gunning and Suttkus, 1966

Mississippi

Yazoo River/ Vicksburg/ FMNH

1 37777

Texas

Rio Grande/ El Paso/ Britton, 1982

West Virginia

Kanawha River/ Chelyan/ Thomas
and Mackenthum, 1964

Alabama

Alabama River/ Hubricht, 1965

Big Nance Creek/ Near Leighton/
OSUM 11488

Indian Creek/ SW of Huntsville/
OSUM 12848

Arkansas

Black River/ Pocahontas/ MCZ
260919

White River/ S of Clarendon/ OSUM
13992

California

El Capitan Reservoir/ E of San

Diego/ Fast, 1971

Florida

Chipola River/ E of Clarksville/

Heard, 1964

Illinois

Wabash River/ E of Rising Sun/

ISM

Indiana

Ohio River/ Mt. Vernon/ OSUM

14399

Kentucky

Tennessee River/ Below Kentucky
Dam/ MCZ 268647

Missouri

Castor River/ Between Dexter and
Sikeston/ MCZ 268300

New Mexico

Rio Grande/ West Drain, Mesilla
Valley/ Metcalf, 1966

Tennessee

Sequatchie River/ NNE of Whitwell
/OSUM 24223

Alabama

Cahaba River/ Hubricht, 1966
Suncanochee Creek/ Hubricht, 1966

California

Dyer Canal/ Alamenda Co./
Prokopovich, 1968

Livermore Canal/ Alameda Co./
Prokopovich, 1968

Year State

Body of Water/Locality/Reference

Year State

Body of Water/Locality/Reference

CORBICULA SYMPOSIUM

Table 1. (continued)

Year State

Florida
Tennessee
Virginia
1966 Alabama

Arkansas

Kentucky

Mississippi

Tennessee

1967 Alabama

Florida

Mississippi

Missouri

Tennessee

1968 Alabama
Arkansas
California

Body of Water/Locality/Reference Year

Ochlocknee River/ NW of
Tallahassee/ Heard, 1966
South Chickamauga Creek/

Chattanooga/ OSUM 24146
Clinch River/ North Tazewell/

MCZ 268583

Black Warrior River/ Below Lock 16

Dam / OSUM 19084

Coosa River/ Below Logan-Martin

Dam/ OSUM 9005

Locust Fork/ N of Cleveland/ NMNS

65722

Town Creek/ NE of Leighton/ OSUM
22086

St. Francis River/ S of Marked Tree/ 1QfiQ

OSUM 20142

Cumberland River/ Kuttawa/ FMNH
179981

Silver Creek/ At Kentucky Reservoir/

Branson and Batch, 1969
Chickasawhay River/ Near Merril/

Grantham, 1967
Coldwateer River/ Cohoma Co./

Heard, 1966

Leaf River/ McClaine/ Heard, 1966
Harpeth River/ NE of Forest Home/

OSUM 22078

Richland Creek/ W of Pulaski/

OSUM 22087

Cypress Creek/ Near Florence/

USNM 756753

Limestone Creek/ E of Peels Corner/

OSUM 42261

Paint Rock River/ Above Trenton/

DMNH 30382

Escambia River/ E of Century/

OSUM 23450

Suwanee River/ NW of Bell/ FSM
Kentucky River/ Camp Daniel Boone/

Branson and Batch, 1969
Red River/ At Kentucky River/

Branson and Batch, 1969
Rockcastle River/ Livingston/ OSUM
22251

Bear Creek/ Tishomingo State Park/

MMNS 1565

Black River/ Hendrick/ FMNH
156605

Duck River/ ESE of Shelbyville/ MCZ
280464

Elk River/ S of Estill Springs/ MCZ

271671 1970

Terrapin Creek/ Ellisville/ OSUM

28041

Ouachita River/ Camden/ OSUM
20409

Columbia River/ Imperial Co./ CAS
38784

State

Georgia

Indiana

Kentucky

Mississippi

Missouri

Tennessee

Alabama

Arizona

Arkansas

California

Florida

Kentucky

Mississippi

Missouri

Oklahoma

Tennessee

Texas

West Virginia

Alabama

Arkansas

Body of Water/Locality/Reference

Oostanula River/ E of Amuchee/
OSUM 28050

Wabash River/ N of Newport/ OSUM
39346

Nolichucky River/ SE of
of Warrensburg/ MCZ 276636
Mississippi River/ W of Gunnison/
MMNS 1642

Little River Canal/ SE of Gideon/
MCZ 268205

Buffalo River/ N of Napier/ OSUM
34272

Nolichucky River/ SE of Warrenburg/
OSUM 23398

Elk River/ NNW of Elkmont/ FSM
Flint River/ Madison Co./ FSM
Mud Creek/ NW of Hollywood/ FSM
Verde River/ W of Camp Verde/

MNA 24.397

Little River/ Near Mississippi River/
MCZ 280465

Lake Jennings/ Near San Diego/
Richardson et at., 1970
Santa Ana River/ Riverside/ ANSP
342789

Lake Hippochee/ Glades Co./ OSUM

25210

Gasper River/ WNW of Bowling

Green/ OSUM 23060

Bouge Phalia River/ E of Areola/

FSM

Sunflower River/ NE of Indianola/

FSM

Tombigbee River/ Leard et at., 1969
Meramec River/ Times Beach/ MDC
3850

Mississippi River/ New Madrid/
OSUM 26398

St. Francis River/ Dunklin Co./ MDC
6300

North Canadian River/ Lake
Overholser/ OSUM 35789
Red River/ NE of Adams/ OSUM
23079

Nueces River/ Lake Corpus Christi/
Murray, 1971a

Elk River/ E of Big Chimney/ OSUM

23118

New River/ Gauley Bridge/ OSUM
23425

Conecuh River/ FSM
Gantt Lake/ N of Andalusia/ FSM
Santa Bouge Creek/ NW of Franklin/
FSM

Sepulga River/ S of Brooklyn/ FSM
Arkansas River/ RM 50 - 283/
Kraemer, 1971

COUNTS: CORBICULA FLUMINEA IN THE UNITED STATES

Table 1. (continued)

Year State

Florida

Georgia

Illinois

Kentucky

Mississippi

Missouri
North Carolina

Tennessee

1971 Florida
Georgia

Mississippi

New Jersey

Ohio

Washington
1972 Alabama

California

Florida

Georgia

Body of Water/Locality/Reference Year

Lake Okeechobee/ Port Myaca/

DMNH 29055

Yellow River/ E of Millington/ FSM
Chickasawahatchie River/ W of
Newton/ FSM

Lake Allatoona/ Cherokee Co./

OSUM 36761

Mississippi River/ Granite City /

Thomerson and Myer, 1970
Dix River/ Near High Bridge/

Branson and Batch, 1971 1973

Buckatunna Creek/ SE of

Buckatunna/ FSM

Chunky Creek/ NNW of Enterprise/

FSM

Okatibee Creek/ Meridian/ USNM
707708

Gasconade River/ Below Fredrickson
/MDC 2300

Catawba River/ Lake Wylie/ Fox,

1971

Stones River/ E of Nashville/ OSUM
27086

Indian Prairie Canal/ Glades Co./
DMNH 47308

Altamaha River/ RM 116/ Sickel,

1973

Coahulla Creek/ Prater Mill/ OSUM
27377

Consanga River/ Beaverdale/ OSUM
27379

Flint River/ Sickel, 1973

Little Ocmulgee River/ E of Reynolds

/OSUM 39966

Ocmulgee River/ At Oconee River/
Sickel, 1973

Pearl River/ Ross Barnett Dam/
OSUM 27113

Delaware River/ Trenton/ Fuller and
Powell, 1973

Licking River/ S of St. Louisville/
OSUM 26860

Muskingum River/ Lowell/ DMNH
51690

Chehalis River/ Raymond/ DMNH
56395

Choctawahatchee River/ NE of

Geneva/ FSM

Little Uchee Creek/ NW of Ft.
Mitchell/ OSUM 41538
Pea River/ E of Perry/ FSM
Owens River/ Inyo Co./ Fox, 1972
Hoimes Creek/ WSW of New Hope/
FSM

St. Joseph Bay / SE of Port St. Joe/

FSM

Savannah River/ NNE of Milhaven/
Fuller and Powell, 1973

State

Missouri

Ohio

Oregon

South Carolina
Tennessee
Texas
Alabama

Arkansas

California

Florida

Mississippi
New Mexico
South Carolina

Tennessee

Texas

Virginia

Alabama

Arkansas

Florida

Georgia

Body of Water/Locality/Reference

Osage River/ Miller Co./ MDC 5550
Olentangy River/ Delaware Reservoir
/OSUM 33900

John Day River/ Grant Co./ ODFW
Pee Dee River/ SE of Society Hill/
Fuller and Powell, 1973
Fall Creek/ Anchor Mill/ OSUM
33791

Trinity River/ Lake Grapevine/ TCU
357

Little Cypress Creek/ NNW of

Jackson/ OSUM 34860

Peckerwood Creek/ N of Marble

Valley/ OSUM 36004

Buffalo River/ Buffalo River State

Park/ OSUM 41571

Stow Lake/ Golden Gate Park, San

Francisco/ Carlton, 1973

Lake Tsala/ Hernando/ ANSP

332593

Wekiva River/ Seminole Co./ RCB
Yalobusha River/ Grenada Reservoir/
Cooper and Johnson, 1980
Pecos River/ Eddy Co./ USNM
709229

Intracoastal Waterway/ At
Waccamaw River/ Fuller and Powell,
1973

Waccamaw River/ Horry Co./ Fuller

and Powell, 1973

Clinch River/ ESE of Tazewell/

ANSP 335725

Halston River/ Knoxville/ ANSP
335735

Obey River/ Dale Hollow Reservoir/
OSUM 35788

Paint Rock River/ Swaim/ OSUM
335720

Guadelupe River/ Canyon Lake/
Britton and Murphy, 1977
Appomattox River/ Near Hopewell/
USNM 711390

James River/ Near Hopewell/

SBMNH 43037
Big Cedar Creek/ NE of
Shepardsville/ OSUM 35450
Cedar Creek/ NE of Furman/ OSUM
35221

North River/ E of Samantha/ OSUM
36723

Bouef River/ W of Lake Village/
OSUM 35977

Spring River/ S of Ravenden/ OSUM
36642

Main Canal/ Palm Beach Co./ FSM
Ogeechee River/ SW of Halcyon
Dale/ FSM

Potatoe Creek/ W of Thomston/
OSUM 35161

CORBICULA SYMPOSIUM

Table 1. (continued)

Year State

Body of Water/Locality/Reference

Year State

Body of Water/Locality/Reference

Illinois

Illinois River/ Kampsville/ Thompson
and Sparks, 1977

Texas

Red River/ Lake Texoma/ Britton and
Murphy, 1977

Iowa

Mississippi River/ Lansing/ Eckbald,

1975

San Jacinto River/ Lake Houston/
MCZ 293569

Mississippi

Buttahatchie River/ Lowndes Co./

OSUM 36251

Virginia

New River/ Glen Lyn/ Rodgers et at.,
1977

Tennessee

Emory River/ NE of Harriman/

OSUM 36769

West Virginia

Mud River/ Cabell Co./ Joy and
McCoy, 1975

Little Tennessee River/ Near Telico/
SBSK 988

Mississippi River/ Fulton/ SBSK 990

1976 Alabama

Neely Henry Lake/ Britton and
Morton, 1979

Tubbs Creek/ SW of New Mt. Hope/

Texas

Angelina River/ Sam Rayburn Lake/

FMNH 197886

TCU 1177

Brazos River/ Pecan Plantation/ TCU
2983

Arkansas

Caddo River/ Near Amity/ ADPC 155
Chamagnoll Creek/ Calion/ TCU

3004

Colorado River/ Marble Falls/ Britton
and Murphy, 1977

California

Cachuma Lake/ Santa Barbara Co./
SBMNH 48084

Virginia

Chickahominy River/ Lanexa/ DMNH
98701

Salinas River/ Monterey Co./ TCU
3018

1975 Alabama

Dauphin Island/ USNM 76536

Drivers Branch/ SE of Talladega/

FSM

Second Creek/ W of Rogersville/

San Luis Reservoir/ N of Basalt Hill/
TCU 3017

Shasta Lake/ Shasta Co./ CAS

57374

OSUM 36350

Indiana

Salt Creek/ Monroe Reservoir/

Arkansas

Coon Bayou/ W of Winchester/

OSUM 48532

OSUM 39680

Maniece Bayou/ NNW of Bradley/

Louisiana

Bayou Cocodrie/ Terrenbonne Par./
Vidrine and DeRouen, 1976

OSUM 39682

Mckinney Bayou/ NW of Garland

City / OSUM 39676

Mississippi River/ ESE of Lake

Mississippi

Noxubee River/ S of Macon/

FMNH 197887

Tibbee Creek/ N of Tibbee/ FMNH
197860

Village/ OSUM 39666

Red River/ E of Bradley Lake/

Missouri

Big Creek/ Sam Baker State Park /
MDC 157

OSUM 39958

North Carolina

Little River/ Town Creek Mound/

California

Lake Murray/ San Diego/ TCU 3028

NCSM P21 1

Florida

Lake Jackson/ Sebring/ OSUM

36827

Uwaharrie River/ SE of Albemarle/
MCZ 280461

Mayakka River/ Mayakka State Park/
OSUM 37965

Ohio

Miami River/ Dayton/ OSUM 38475
Stillwater River/Dayton/ GAC 762

Illinois

Saline River/ E of Equality/ OSUM

Oregon

Siuslaw River/ Lane Co./ ODFW

36832

South Carolina

Cooper River/ Lake Marion/

Kentucky

Buck Creek/ NW of Ula/ OSUM

TCU 2999

38072

Floyd's Fork/ E of Brooks/ DMNH

106659

Tennessee

Barren Fork River/ McMinnville/
OSUM 40922

Big Bigby Creek/ Canaan/ OSUM

Louisiana

Tensas River/ Madison Par./ Kuckyr
and Vidrine, 1975

40726

Big Rock Creek/ Verona/ OSUM

Maryland

Chesapeake Bay / Susquehanna

Flats/ Stotts et a/., 1977

40747

Big Swan Creek/ SE of Centerville/

Missouri

White Water River/ Bollinger Mill/

MDC 7601

OSUM 40727

East Rock Creek/ NNE of Verona/

North Carolina

Waccamaw River/ Lake Waccamaw/
Fuller and Richardson, 1976

OSUM 41552

Flat Creek/ SW of Rally Hill / OSUM

Ohio

Scioto River/ S of Delaware/ FMNH
171650

40741

Fountain Creek/ SE of Columbia/

Oklahoma

Red River/ Lake Texoma/ TCU 1580

OSUM 40740

South Carolina

Santee River/ Lake Marion/ OSUM

36568

Garrison River/ WSW of Bugscuffle/
OSUM 40913

COUNTS: CORBICLJLA FLUMINEA IN THE UNITED STATES

Table 1. (continued)

Year State

Body of Water/Locality/Reference

Year State

Body of Water/Locality/Reference

Tennessee (con’t)

Greenlick Creek/ SE of Williamsport/
OSUM 40719

Lick Creek J Branton Ford/ OSUM

40731

Missouri

Big River/ NW of House Springs/
OSUM 41133

Bourbeuse River/ Noser Hill/ OSUM
42666

Liepers Creek/ NE of Williamsport/
OSUM 40723

North Carolina

Richardson Creek/ Union Co./ NCSM
P256

Little River/ N of Maryville/ OSUM

40931

Little Duck River/ Manchester/

OSUM 40920

North Fork Creek/ SE of Unionville/
OSUM 40911

Notchy Creek/ SE of Madisonville/
OSUM 40928

Ohio

Hocking River/ NE of Stewart/

OSUM 41395

Little Muskingum River/ ENE of
Marietta/ OSUM 39712

Meigs Creek/ NW of Beverly/ OSUM
40198

Olive Green Creek/ NW of Beverly/
OSUM 40260

Piney River/ NW of Centerville/

OSUM 40733

Rich Creek/ S of Wilhoit Mills/

OSUM 40893

Sinking Creek/ S of Halls Mill/

OSUM 40898

Texas

Big Cypress Creek/ Lake of the

Pines/ Pool and McCullough, 1979
Concho River/ Lake Nasworthy/
Baker, 1978

Sabine River/ Murvaul Reservior/

Pool and McCullough, 1979

Sugar Creek/ S of Shelbyville/

OSUM 40915

West Virginia

Beech Fork Reservoir/ Cabell Co./
MUMC 857

Texas

Blanco River/ At San Marcos River

UOM

Johnson Creek/ S of Ozona / Britton,

Wisconsin

St. Croix River/ Near Hudson/ Fuller,
1978b

1982

Pecos River/ N of Rio Grande/

Britton, 1982

1978 Arkansas

Bayou Bartholomew/ Near Jones
River/ ADPC 51

L’Anguille River/ NE of Mariana/

1977 Alabama

Burnt Corn Creek/ Brewton/ OSUM
42021

Murder Creek/ Evergreen/ OSUM

42259

Piney Creek/ W of Peets Corner/

OSUM 41492

Madison-Mariana Diversion Canal/
ENE of Tuni/ OSUM 43050

Strawberry River/ S of Smithville/
ADPC 108

OSUM 42100

Florida

Lake Oklawaha/ Rodman Dam/ FSM

Florida

Aucilla River/ Taylor Co./ OSUM

26512

42021

Kissimmee River/ SE of Okechobee/

Georgia

Chattahoochee River/ Near

Columbus/ USNM 79558

FSM-U

Oklawaha River/ E of Silver Springs/
OSUM 41184

St. Johns River/ SE of Geneva/

OSUM 45832

Indiana

Big Indian Creek/ Near Crandall/
MUMC 1559

Blue River/ N of Wyandotte/ OSUM
45780

White River/ E of Mendora/ DM 562

Steinhatchee River/ Dixie-Taylor

Co. Line/ DMNH 125738

Maryland

Potomac River/ Washington, D. C.
area / Dresler and Cory, 1980

Georgia

Ohoopee River/ SSW of Reidsville/

FSM

Minnesota

Minnesota River/ Near Burnsville/
Cummings and Jones, 1978

Withlacoochee River/ Brookes-
Lowndes Co. Line/ FSM

Mississippi

Allan Branch/ N of Enterprise/ FMNH

201527

Illinois

Kaskaskia River/ Near Baldwin/
Thompson and Sparks, 1977

Sangamon River/ Lake Sangchris/
Thomas and Sparks, 1977

Moss Creek/ SE of Caichae!/ FMNH
201525

Shubuta Creek/ NW of Shubuta/
FMNH 201514

Kentucky

Mud River/ NE of Beechland/ OSUM

41495

Tygarts Creek/ Below Cascade Cave
/ MUMC 848

Souinlovey Creek/ N of Pachuta/
FMNH 201521

Woodward Creek/ E of Cooksville/
FMNH 201524

Mississippi

Big Black River/ NW of Edwards/

FNMH 198384

Missouri

Cane Creek/ Butler Co./ MDC 1325
Little Black River/ Butler Co./ ACB

CORBICULA SYMPOSIUM

Table 1. (continued)
Year State

Missouri (con’t)

North Carolina

Oklahoma

Tennessee

Texas

West Virginia

1979 Alabama

Arkansas

California

Florida

Georgia

Indiana

Kentucky

Maryland

Missouri

North Carolina
Pennsylvania

Tennessee

Texas

Body of Water/Locality/Reference

Year

Moreau River/ Jefferson City/ MDC
5222

Rocky River/SE of Oakboro/ OSUM
42192

Caddo Creek/ N of Ardmore/ UOM
Collins River/ E of McMinnville/
OSUM 43806

Llano River/ Llano Co./ UOM
Big Seven Mile Creek/ Cabell Co./
MUMC 1144

Guyandotte River/ Midkiff/ MUMC
1221

1980

State

Arkansas

Florida

Illinois

Chattahoochee River/ Below Uchee
Creek/ Jenkinson, 1979
Saugahatchee Creek/ NW of Auburn/
Jenkinson, 1979

Tallapoosa River/ Upstream of Lake

Martin/ Jenkinson, 1979

Uchee Creek/ At Chattahoochee

River/ Jenkinson, 1979

Saline River/ W of Owinsville/ FSM

29553

Lake Casitas/ Near Ojai / SLC
Lake Piru / Near Filmore/ SLC
Merced River/ Merced/ SBSK 1777
Lake Lucy/ Groveland/ FSM 26687
Lake Palatlakaha/ SW of Cleremont/
FSM 26772

Lake Talquin/ W of Tallahassee/

FSM 26932

Mosquito Creek/ E of Chattahoochee
/ USNM 809640

Spring Creek/ SE of Mariana/ FSM
30097

Waccasassa River/ Levy Co./ FSM
Pound Creek/ Lake Meriwether/
OSUM 45570

Stoney Creek/ SE of Noblesville/
OSUM 43620

Coal River/ Boyd Co./ MUMC 1526
Salt River/ SSE of Waterford/ OSUM
44764

Wicomico River/ Salisbury/ Counts,
1981b

Thomas Hill Reservoir/ Macon Co./
MDC 7537

Cape Fear River/ Lee Co./ RS
Ohio River/ Pittsburgh/ Taylor,

1980b

Big Hickory Creek/ SW of
McMinnville/ OSUM 39044
Shoal Creek/ SSW of Pulaski/ USNM
795588

Weakly Creek/ SE of Unionville/
OSUM 40908

Pedernales River/ Near Johnson
City / CEB

San Antonio River/ Karnes Co./ CEB

Kentucky

Michigan

Mississippi

Missouri
North Carolina

South Carolina

Texas

Virginia
West Virginia

1981 Florida

Body of Water/Locality/Reference

Spring Creek/ NNW of West Field/
OSUM 45887

La Grue Bayou/ S of Dry Lake Dam/
FSM 29141

Ichetucknee River/ Ichetucknee State
Park / FSM 26905

Rocky Creek/ N of Sink Creek/ FSM
28350

Santa Fe River/ At Olustee Creek/
FSM 27817

Kankakee River/ Custer Park/ Lewis
and Brice, 1980

Eagle Creek/ Sparta Bridge/ MUMC
1666

Little River/ S of Hopkinsville/ OSUM
49287

Lake Erie/ ENE of Monroe/ OSUM
49999

Big Black Creek/ Jackson Co./

MMNS 1054

Pascagoula River/ Three Rivers/
MMNS 1078

Steel Bayou/ Near Fitler/ Leard et at.
1980

Tallahala Creek/ Forest Co./ MMNS
981

Yockanookany River/ Leake Co./
MMNS 1295

Bryant Creek/' Ozark Co./ MDC 1175
Missouri River/ Merman/ RDO
Eden River/ Near Winston-Salem/

JJH

Mountain Island Lake/ Near Lucia/
USNM 809473

Hartwell Reservoir/ Anderson-
Oconee Cos./ JJH

Lake Keowee/ Oconee-Pickens Cos./
JJH

Little Brazos River/ SE of Law/

OSUM 48170

White River/ White River Lake/
Fontanier, 1982

Potomac River/ Alexandria/ Dresler
and Cory, 1980

Monongahela River/ Morgantown/
Zeto, 1982

Pocatalico River/ Sissonville/ MUMC

1586

Twelve Pole Creek/ Shoals/ MUMC

1587

West Fork River/ NE of West Milford/
Zeto, 1982

Cypress Creek Canal/ NE of North
Lauderdale/ OSUM 49853
Lake Buena Vista/ Orange Co./ RJ
Middle River Canal/ Lauderdale
Lakes/ OSUM 49849

COUNTS: CORBICULA FLUMINEA IN THE UNITED STATES

Table 1. (continued)

Year State

Body of Water/Locality/Reference

Year State

Body of Water/Locality/Reference

Florida (con’t)

Sky Lake/ North Miami Beach/

OSUM 49910

RLR

Little River/ N of Goodwater/ RLR

Georgia

Ghickamauga Creek/ Ringgold/

OSUM 50521

Towaliga River/ Downstream of High
Falls Lake/ Walker, 1982

South Carolina

Edisto River/ ENE of Canadys/

OSUM 48840

Salkahatchie River/ NE of Yemassee
/ OSUM 48840

Kentucky

Nolin River/ White Mills/ MUMC

3029

Tennessee

Hatchie River/ NE of Rialto/

OSUM 50462

Slate Creek/ E of Owingsville/ MUMC
3106

Texas

San Gabriel River/ Circleville/ Hillis
and Patton, 1982

Tradewater River/ S of Sullivan/

OSUM 50808

West Virginia

Hughes River/ E of Cisco/ MUMC
3184

Maryland

Susquehanna River/ Conowingo

1982 Delaware

Nanticoke River/ Near Seaford/ SH

Dam / Nichols and Domermuth, 1981

Kentucky

Elkhorn Creek/ S of Stamping

Mississippi

Amite River/ Amite Co./ MMNS

Ground/ MUMC 3224

1450

Maryland

Nassawango Creek/ Near Snow Hill/

New Jersey

Raritan River/ Near New Brunswick/

USNM 804416

Trama, 1982

Ohio

Maumee River/ Toledo/ Scott-Wasilk

Oklahoma

Arkansas River/ Ft. Gibson Dam/

et a!., 1983

BIOFOULING OF POWER PLANT
SERVICE SYSTEMS BY CORBICULA

T. L. PAGE, D. A. NEITZEL, M. A. SIMMONS
PACIFIC NORTHWEST LABORATORY
RICHLAND, WASHINGTON 99352, U.S.A.
and

P. F. HAYES

ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, DC 20555, U.S.A.

ABSTRACT

Corbicula sp. foul the service water systems at nuclear power plants because the environment
within these systems is compatible with the ecological requirements of the species. To reduce fouling
by Corbicula, components of service water systems and operating procedures that enhance the potential
for fouling need to be identified. Factors important in mediating biofouling of service water systems
appear to be screening potential, minimum and maximum velocities and the operational procedures
employed during power plant biofoulant control and downtime. These conclusions are based on the
results of a categorical model we used to correlate information from power plants with that on Cor-
bicula life history. Power plant parameters in the model include temperature, dissolved oxygen con-
centration, screen and strainer size, maximum and minimum velocities, and elements of the biofoul-
ing control procedures. Parameters for Corbicula include tolerances to temperature, dissolved oxygen,
biofouling control chemicals, velocity preferences, and optimal temperatures for each life stage and
behavior.

The freshwater clam, Corbicula sp., has effectively foul-
ed service water systems in nuclear power plants. After the
Arkansas Nuclear One, Unit 2, power plant was shut down
because clams blocked flow through the containment air
coolers the Nuclear Regulatory Commission (NRC) issued
a bulletin (l&E Bulletin 81-03) requiring power plant operators
to examine for the presence of Corbicula in their power plants
and in the environment near the plants. Information was col-
lected from 87 sites, which included 151 units; 73 of these
units were operating and 78 were planned or under
construction.

Data on the distribution of Corbicula near the plants
was collected as 1) present in plant, 2) present in vicinity,
3) present in waterbody, or 4) not present. The NRC staff
defined “present in plant’’ as live organisms, shells or shell
fragments in or having been found in the circulating water
system, service water system, or fire protection system. The
intake structure was not considered part of the plant. “Pre-
sent in vicinity” was defined as in or having occurred in the
areas near the plant that are subjected to plant related aquatic
biological monitoring programs. “Present in the waterbody”
was defined as in or having occurred in the water from which

the plant obtains cooling water or a connecting body of water
from which colonization of the area in which the plant occurs
is probable. “Not present” was defined as not reported in
the waterbody and the probability of future colonization
unlikely.

A review of the responses to l&E Bulletin 81-03 show-
ed Corbicula was reported present in the plant at 10 sites,
which included 17 operating units (Masnik, unpubl. ms. NRC,
Wash. D.C.) Corbicula was reported in the vicinity but not in
the plant of 5 sites, including 10 operating units. Two sites,
with three units, reported Corbicula in the waterbody but not
in the plant vicinity. Corbicula was reported not present in
the plant, vicinity or waterbody at 33 sites with 43 operating
units.

For plants planned or under construction, Corbicula
was reported as present in the vicinity of 27 sites including
41 units and present in the waterbody at 31 sites with 51 units.
The organism was reported not present for 17 sites with 27
units.

Corbicula was present in 23% of the currently licens-
ed units in the United States. However, when the percent oc-
currence was calculated based on Corbicula occurrence in

American Malacological Bulletin, Special Edition No. 2(1986): 41-45

CORBICULA SYMPOSIUM

the vicinity and waterbody, in-plant occurrence increased to
570/0.

These data illustrate 1) the effectiveness of Corbicula
to invade and foul the service water systems of power plants
if found in the environment around the power plant and 2)
the environment within these systems is compatible with the
environmental requirements of Corbicula. In order to obviate
Corbicula fouling in service water systems of power plants,
Corbicula must be 1) prevented from entering the system or
2) the environment of the system must be made incompatible
with the ecological requirements for growth and survival of
Corbicula. The purpose of this review is to detail Corbicula
characteristics that enhance their ability to service water
systems. The methods and techniques to survey for and con-
trol the Corbicula fouling are not complete as of this review
and we do not suggest that total prevention of Corbicula foul-
ing is possible.

CORBICULA CHARACTERISTICS AND POWER
PLANT CONDITIONS

We used a categorical model to correlate information
from power plants with that of Corbicula life history. Power
plant parameters in this model include temperature, dissolv-
ed oxygen concentration, size of screens and strainers, max-
imum and minimum velocities and elements of biofouling con-
trol and operational prodedures. Paramenters for Corbicula
include tolerance to temperature, dissolved oxygen and
biofouling control chemicals, velocity preferences, optimal
temperatures for each life stage and behavior. The correla-
tions between Corbicula and power plant service water
systems that appear to promote or allow Corbicula fouling
are 1) larva size and size of screens and strainers, 2) larvae
settling and in-plant flow velocities and patterns, 3) larva-
substrate requirements and silt buildup, 4) growth potential
and water temperature, and 5) avoidance behavior and in-
termittent antifoulant application.

LARVA SIZE AND SIZE OF SCREENS AND
STRAINERS

The offspring of Corbicula are retained in the brood
chamber of the adult from the egg to the juvenile stage, at
which time they are released (Britton and Morton, 1982). Dur-
ing the breeding season, an adult can contain thousands of
pre-release juveniles, releasing 300-400 juveniles per day
(Britton, 1982; Aldridge and McMahon, 1978). When releas-
ed the larvae are not typical molluscan veligers, but rather
pediveligers more adapted for crawling than swimming (Brit-
ton, 1982). However, pediveligers can be carried by ambient
currents some distance downstream of the adult population
(Britton, 1982; Eng, 1979).

Figure 1 illustrates the relationships between larvae
size and screening and strainer capability that must occur
to prevent Corbicula from entering the service water systems
of the power plant. When breeding populations of Corbicula
occur in or near the intake structure of a power plant, the
water intake system can entrain Corbicula into the plant. The

CHARACTERISTIC (SIZE)

220/r 3c"1

Corbicula LIFE STAGE

PEDIVELIGER LARVA M

ADULT 1 11 1 hi — H

POWER PLANT SYSTEMS

HEAT EXCHANGERS I—

FIRE PROTECTION I .1

SCREENS AND STRAINERS 1 ' I

LOW FLOW VELOCITY AREAS I—

Fig. 1. Categorical model illustrating correlation between Corbicula
larva size and power plant system screening potential.

obvious control measure is screening or straining. However,
to maintain the cooling requirements of a power plant, large
volumes of water are required. Screen mesh-size is general-
ly limited to no less than 0.3 cm and in-system strainers are
too large to restrict the passage of Corbicula pediveligers or
small juveniles (Goss and Cain, 1977). The approximate size
of the Corbicula pediveliger in 220m (Sinclair and Isom, 1961 ;
Aldridge, 1976; Aldridge and McMahon, 1978; Britton and
Morton, 1982). Corbicula pediveligers in the intake water of
a power plant wil not be screened at the intake or by strainers
incorporated into the service water system. The use of smaller
intake screens and strainers of present design is probably
not practical because of the large volumes of water required
for operation of nuclear power plants.

LARVAE SETTLING AND IN-PUT FLOW VELOCITIES
AND PATTERNS

Juveniles, released from the adult, quickly settle onto
the surrounding substratum (Britton, 1982) unless they are
moved by ambient currents. High velocity flows may prevent
juveniles from settling or may dislodge juveniles before they
are large enough to survive in currents that can move them
(Eng, 1979).

We were not able to locate data correlating the ability
of Corbicula larvae to settle with ambient flow velocities. Eng

CHARACTERISTIC (FLOW)

0 3 m/s

Corbicula LIFE STAGE

PEDIVELIGER LARVA I "

ADULT >■

POWER PLANT SYSTEMS

HEAT EXCHANGERS I — 1 -H

FIRE PROTECTION I— I

SCREENS AND STRAINERS I

LOW FLOW VELOCITY AREAS I

Fig. 2. Categorial model illustrating correlation between Corbicula
flow requirements and in-plant flow velocities.

PAGE ETAL POWER PLANT BIOFOULING

(1979) studied Corbicula populations in the Delta-Mendota
Canal in California. Flow velocities in the canal averaged 1 .2
mps. Sickel (1976) observed substratum preference of Cor-
bicula pediveligers. The preferred velocity for Corbicula lar-
vae settlement on the substrata tested was about 0.3 mps,
however larval settlement may occur in currents less than this.

Using these observations (Eng, 1979; Sickel, 1976)
and the fact that fouling potential in power plant service water
uptake varies, we assumed there are preferred and upper
flow limits for settlement of Corbicula larvae. Figure 2 il-
lustrates the correlation between settling requirements and
in-plant flow velocities that allow settlement of Corbicula
pediveligers in the service water system. Data from operating
power plants indicate that the pediveligers settle in low-flow
areas. Then, as the populations grow, some of the organisms
may move from low-flow to higher-flow areas and be carried

CHARACTERISTIC (SUBSTRATE)

METAL PVC CONCRETE SILT

Corbicula LIFE STAGE
PEDIVELIGER LARVA
ADULT

POWER PLANT SYSTEMS

HEAT EXCHANGERS Tin™ — ' nr . |

FIRE PROTECTION I

SCREENS AND STRAINERS >' ' i

LOW FLOW VELOCITY AREAS H ■ — - 1

Fig. 3. Categorical model illustrating correlation between suitable
substrate for Corbicula and substrates that occur in power plant ser-
vice water systems.

with the water to other locations in the plant where larger in-
dividuals may block condenser tubes or smaller-diameter in-
take pipes in the service water system.

LARVA-SUBSTRATUM REQUIREMENT AND SILT
BUILDUP.

Corbicula seem to prefer sandy or gravel substrata,
but are also found in larger rock or in mud or silt (Britton,
1982). Corbicula densities in a waterbody will vary with the
substratum type (Eng, 1979). In service water systems, Cor-
bicula appear to be associated with accumulations of silt or
corrosion products. It is not clear, at this time, if this associa-
tion is causitive or incidental. The larvae may initially settle
preferentially in areas of silt, or their settling may result in
siltation. Another explanation is that larvae and silt may set-
tle at the same location independent of one another. Materials
found in power plant service water systems are not a prefer-
red substratum for Corbicula settlement or growth, but there
is no indication that larvae cannot settle on the kinds of
materials found in the service water system, in the absence
of silt. However, if suitable substratum for Corbicula
pediveliger settlement does not occur in the plant, then the
incidence of Corbicula in the plant may be reduced.

Figure 3 illustrates the correlation between suitable

substrate for Corbicula and available suitable substratum in
service water systems. Most service water “plumbing” is
constructed with metal or concrete; however, low flow areas
or “dead legs” provide areas were silt and larvae can ac-
cumulate. These silted areas can serve as substrata for Cor-
bicula to initially settle and grow. Additionally, pitted surfaces
on service water systems may promote or allow Corbicula
settlement.

GROWTH POTENTIAL AND WATER
TEMPERATURES

Corbicula juveniles grow rapidly after they are re-
leased from the adult. Depending on the water temperature
when the pediveliger is released, juveniles can grow to 10
to 18 mm shell length in a few months (Britton, 1982; Aldridge
and McMahon, 1978). This rapid growth characteristic may
contribute to the effectiveness of Corbicula as a biofouling
organism. Corbicula become sexually mature very young and
are fecund. Estimates for size at maturation range from 6 mm
to 13 mm shell length (Heinsohn, 1958; Sinclair and Isom,
1963; Alderidge and McMahon, 1978). If the clams inside the
power plant are able to grow at the same rate as those in
nature, then it is possible for clams that have infested a power
plant as a result of the spring spawning season to begin
reproducing in the fall. Clams also reach reproductive size
before they are large enough to become trapped in the
smaller-diameter pipes in the service water system.
Therefore, potentially reproducing populations may become
widely distribution within the plant in a short period.

Corbicula can tolerate a wide range of water
temperatures. Mattice and Dye (1976) reported that for con-
tinuous exposures the upper tolerance limits for 50% of the
clam tested was between 24°C and 34°C when acclimation
temperatures ranged from 5°C to 30°C. Lower tolerance limits
were between 2°C and 12°C for acclimation temperatures
ranging from 15°C to 30°C. Goss etal. (1979) reported similar
tolerance, with variation dependant on acclimation
temperatures and Corbicula size. Figure 4 illustrates the cor-
relation between temperature tolerance of Corbicula and the
occurrence of warm water in service water systems. Water
temperatures in the service water systems are conducive to

CHARACTERISTIC (WATER TEMPERATURE)

0°c 35 - °C

Corbicula LIFE STAGE

PEDIVELIGER LARVA I ■ ■ l

ADULT | ■ 1 " — H

POWER PLANT SYSTEMS

HEAT EXCHANGERS ■».. — «««

FIRE PROTECTION *•*

SCREENS AND STRAINERS — ■ ■ ■

LOW FLOW VELOCITY AREAS ...

Fig. 4. Categorical model illustrating correlation between temperature
tolerances of Corbicula and water temperatures in power plant ser-
vice water systems.

CORBICULA SYMPOSIUM

CHARACTERISTIC (CHLORINE TOLERANCE)

< 0 1 ppm 10 ppm

Corbicula LIFE STAGE

PEDIVELIGER LARVA | ]

ADULT i |

POWER PLANT SYSTEMS ppm

HEAT EXCHANGERS " ■ |

FIRE PROTECTION '■ " ' j

SCREENS AND STRAINERS

LOW FLOW VELOCITY AREAS » "■ |

Fig. 5. Categorical model illustrating the correlation between chlorine
tolerance of Corbicula and chlorine exposures (concentration, time
and frequency) expected in power plant service water systems.

the growth and reproduction of Corbicula and probably
enhance the potential for Corbicula occurrence and fouling
at nuclear power plants.

Power plant service water systems include areas
where the circulating raw water is warmed by plant oper-
ations. The water temperatures are usually within the range
of tolerance for Corbicula. The warmed water can increase
the growth potential of pediveligers and juveniles that have
invaded the plant. This growth over a short period of time
would increase the fouling problems associated with
Corbicula.

AVOIDANCE BEHAVIOR AND INTERMITTENT
ANTIFOULANT CONTROL

A characteristic of mollusk bivalves is their ability to
“clam-up” in response to environmental stimuli. Bivalves
avoid inimical environmental conditions by retreating into their
shells and respiring anaerobically for extended periods of
time. This behavior allows adult Corbicula to avoid antifoulant
control measures such as chlorination. Mattice et at. (1982)
reported that, given the U.S. Environmental Protection Agen-
cy’s regulation limits on chlorine in power plant discharges
(USEPA 1980), the use of chlorine has proven to be ineffec-
tive in controlling clam fouling at power plants. They reported
adult Corbicula can tolerate target concentrations of 10 mg/L
total residual chlorine for up to 30 min. Chlorination practices
at Tennessee Valley Authority plants that include continuous
chlorination during the clam breeding season have been
somewhat successful (Goss and Cain, 1977). However,
residual levels are difficult to maintain in static systems, like
fire protection systems. Burial in silt may provide some addi-
tional protection from intermittently chlorinated water. Figure
5 illustrates the correlation between chlorine tolerance of Cor-
bicula (including their ability to avoid chlorinated water for
extended periods) and the chlorine levels expected in power
plant service watersystems. Tolerance of Corbicula to chlorine
is a function of both concentrations and exposure time. The
tolerance levels illustrated in Figure 5 and generally consis-
tent with exposure resulting from chlorination schedules
generally used at freshwater cooled power plants, e.g. usually

less than 2 hrs during a 24-hr period.

It is unlikely that standard chlorination practices alone
will control Corbicula fouling because adults can avoid short-
term exposure to toxic levels of chlorine, discharge regula-
tions prevent use of chlorine at concentrations and for periods
of time sufficient for effective control and during the larval
release season, chlorination in the plant must be continuous
to control infestation by larvae.

CONCLUSIONS

The freshwater clam Corbicula sp. is common in the
aquatic environments near nuclear power plants in the United
States. These clams readily move from the ambient environ-
ment to the environment of the service water system in the
power plant as pediveligers and smali juveniles. Portions of
the service water system environment are compatible with
the ecological requirements of Corbicula. Once in place in
the power plant, the ability of Corbicula to grow and pot-
entially reproduce and to avoid control measures for short
periods allow Corbicula to effecively foul these plant systems.
Effective control will require changing the environment of the
service water system so Corbicula pediveligers either can-
not enter or become established, grow and reproduce in the
power plant service water system. Environmental changes
may include dewatering of redundant systems during
maintenance schedules, ensuring that systems on stand-by
are filled with antifoulant treated water, antifoulant treatment
of the entire service water system during Corbicula spawn-
ing, or enhanced flow through all service water system com-
ponents. Control will probably be plant specific, and may even
require different controls for different system components
within the same plant.

ACKNOWLEDGMENTS

We thank the following individuals who contributed to the
development of this manuscript. R. W. Hanf helped with the literature
search. Ginny Woodcock typed the manuscript. Carolynn Novich
edited the manuscript. The study was supported by the U. S. Nuclear
Regulatory Commission under a related services agreement, with
the U. S. Department of Energy (Contract No. DE-AC06-76RLO 1830).

LITERATURE CITED

Aldridge, D. W. 1976. Growth, reproduction and bioenergetics in a
natural population of the Asiatic freshwater clam Corbicula
manilensis Philippi. M. A. Thesis, the University of Texas at
Arlington. 97 pp. Aldridge, D. W. and R. F. McMahon. 1978.
Growth, fecundity, and bioenergetics in a natural population
of the Asiatic freshwater clam, Corbicula manilensis Philippi,
from North Central Texas. Journal Molluscan Studies 44:49-70
Britton, J. C. 1982. Biogeography and ecology of the Asiatic clam,
Corbicula in Texas. IN: Proceedings of the Symposium on Re-
cent Benthological Investigations in Texas and Adjacent States ,
pp. 21-31.

Britton, J. C. and B. Morton. 1982. A Dissection Guide, Field and
Laboratory Manual for the Introduced Bivalve Corbicula
fluminea. Malacological Review , Supplement 3. 82 pp.

Eng, L. L. 1979. Population dynamics of the Asiatic clam, Corbicula
fluminea (Muller), in the concrete-lined Delta-Mendota Canal

PAGE ETAL.: POWER PLANT BIOFOULING

of central California. IN: Proceedings, First international Cor-
bicula Symposium, Texas Christian University Research Foun-
dation, Fort Worth, Texas, pp. 39-68

Goss, L. B. and C. Cain, Jr. 1977. Power plant condenser and ser-
vice water system fouling by Corbicula, the Asiatic clam. IN:
L. D. Jensen, ed., Biofouling Control Procedures. Marcel Dek-
ker, Inc., New York. pp. 11-17.

Goss, L. B., J. M. Jackson, H. B. Flora, B. G. Isom, C. Gooch, S.
A. Murray, C. G. Burton and W. S. Bain. 1979. Control studies
on Corbicula for steam-electric generating plants. IN: J. C. Brit-
ton, ed., Proceedings, First International Corbicula Symposium.
Texas Christian University Research Foundation, Fort Worth,
Texas, pp. 139-151.

Heinsohn, G. E. 1958. Life history and ecology of the freshwater clam,
Corbicula fluminea. M. S. Thesis, University of California,
Berkeley, 64 pp.

Mattice, J. S. and L. L. Dye. 1976. Thermal tolerance of the adult
Asiatic clam. In: G. W. Esch and R. W. McFarlane (eds.) Ther-
mal Ecology II. ERDA Symposium Series, CONF-750425, Na-
tional Technical Information Service, Springfield, Virginia, pp.
130-135.

Mattice, J. S., R. B. McLean, and M. B. Burch. 1982. Evaluation of

short-term exposure to heated water and chlorine for control
of the Asiatic clam ( Corbicula fluminea). ORNL/TM-7808.
Prepared for the U. S. Department of Energy of Oak Ridge
National Laboratory, Oak Ridge, Tennessee.

Sickel, J. B. 1976. Population, growth and productivity of Corbicula
maniiensis (Philippi) in the Altamaha River, Georgia (Bivalva:
Corbiculidai). ASB Bulletin 23:96.

Sinclair, R. M. and B. G. Isom. 1961. A preliminary report in the in-
troduced Asiatic clam Corbicula in Tennessee. Stream Pollu-
tion Control Board, Tennessee Department of Public Health.
Nashville 75 pp.

Sinclair, R. M. and B. G. Isom. 1963. Further studies on the introduc-
ed Asiatic clam (Corbicula) in Tennessee. Tennessee Stream
Pollution Conrol Board, Tennessee Department of Public
Health, Cordell Hull Building, Nashville, Tennessee. 75 pp.
U.S. Environmental Protection Agency (USEPA). 1980. Ef-
fluent limitation guidelines, pretreatment standards and new
source performance standards under clean water act; steam
electrical power generating point source category. 40 CGR
Parts 125 and 423. Federal Register 45 (200):68328-68337
(October 14, 1980).

ENGINEERING FACTORS INFLUENCING CORBICULA
FOULING IN NUCLEAR SERVICE WATER SYSTEMS

K. I. JOHNSON
C. H. HENAGER
T. L. PAGE
P. F. HAYES1

PACIFIC NORTHWEST LABORATORY
RICHLAND, WASHINGTON 99342, U.S.A.
and

HJ.S. NUCLEAR REGULATORY COMMISSION
WASHINGTON, D.C. 20555, U.S.A.

ABSTRACT

Corbicula fouling is a persistent problem in the service water systems of nuclear power plants.
An understanding of the biological characteristics of Corbicula and the engineering characteristics
of service water systems is important in developing effective detection and control methods. A data
base of Corbicula fouling was compiled from nuclear and non-nuclear power stations and other in-
dustrial users of large volumes of raw water. The data base was analyzed to identify engineering fac-
tors common to service water systems that are conducive to fouling by Corbicula. Bounds on several
engineering parameters such as flow velocity and water temperature which support Corbicula growth
are given. Service water systems found in BWR and PWR reactors are listed and those that show
fouling are identified. Systems that have fouled include residual heat removal heat exchangers, con-
tainment coolers and turbine-bearing lube oil coolers. Possible safety implications of Corbicula foul-
ing are discussed for specific service water systems. Recommendations are given on how to minimize
the potential for Corbicula fouling.

Fouling of service water systems due to the presence
of the asiatic clam, Corbicula, is a persistent problem in the
nuclear and non-nuclear power industries. Fouling of nuclear
service water systems is especially critical because many ser-
vice water cooling loops are required for safe shutdown of the
reactor. This paper identifies engineering factors that have
commonly occurred where Corbicula were found in nuclear
service water system piping. The safety implications of ser-
vice water system fouling are discussed and recommenda-
tions are given to reduce the potential for Corbicula fouling.

Many of the factors which influence fouling were iden-
tified from first-hand accounts given by utility personnel who
have witnessed fouling incidents involving Corbicula. A sec-
ond source of information was utility responses to IE Bulletin
81-03, “Flow Blockage of Cooling Water to Safety System
Components by Corbicula sp. (Asiatic Clam) and Mytiius sp.
(mussel).” This Bulletin was issued by the Office of Inspec-
tion and Enforcement of the U.S. Nuclear Regulatory Com-
mission to all operating plants and plants under construction.
Other information was obtained from the published literature
on Corbicula.

The conclusions presented in this report are not

necessarily those of the U.S. Nuclear Regulatory Commission
and are not intended to infer any regulatory position on the
part of the NCR.

NUCLEAR SERVICE WATER SYSTEMS

Nuclear service water systems are designed to pro-
vide cooling water to reactor and auxiliary system com-
ponents during both normal and accident conditions. The
water source is raw water taken directly from a river, lake,
or ocean (Haried, 1982). The circulation water system, which
cools the main turbine condensers, is considered separate
from the service water systems.

Service water systems of both boiling water reactors
(BWRs) and pressurized water reactors (PWPs) are divided
into two general categories -- essential and nonessential
systems. Each system handles a different type of cooling load.

Generally, the essential service water system cools
components within the reactor and auxiliary buildings that
are nuclear-related and are required for safe shutdown. The
essential service water system may also be referred to as the
emergency equipment cooling water (EECW) system, the ser-
vice water (SW) system, or the essential raw cooling water

American Malacologies) Bulletin, Special Edition No. 2(1986):47-52

CORBICULA SYMPOSIUM

(ERCW) system. Cooling loops served by the essential ser-
vice water system are classified as safety related.

The nonessential service water system, in general,
cools components within the turbine-generator building that
are non-nuclear related and are not required for safe shut-
down of the reactor. In some plants this system is referred
to as the raw cooling water (RCW) system or the auxiliary cool-
ing water system. Cooling loops served by the nonessential
service water system are classified as nonsafety related.

The fire protection system is a non-nuclear safety
system that often draws its water from the service water system
or from the service water intake bay. Because Corbicula foul-
ing has occurred in the fire protection system, this system will
be discussed along with the service water systems.

Initial fouling control for both BWRs and PWRs occurs
in the service water intake structure. Large chunks of floating
debris such as driftwood and ice are removed by the trash
racks at the opening of the intake structure. After entering the
intake, water passes through self-cleaning traveling screens
which remove debris greater than 13 mm in diameter.
Downstream from the service water pumps, the water passes
through basket strainers which remove particles greater than
3 mm in diameter. Therefore, Corbicula up to 32 mm in
diameter which have been found inside service water heat
exchangers (Goss and Cain, 1976) have come in as larvae
and have found suitable conditions for growth inside service
water systems piping. Thus, an effective means of controll-
ing Corbicula inside service water systems must address con-
trol of Corbicula larvae in the service water.

ENGINEERING FACTORS INFLUENCING COR-
BICULA FOULING

The engineering factors discussed here have occur-
red commonly where Corbicula have been found in service
water system piping. These factors often interact to emulate
environmental conditions known to support Corbicula growth
in their natural environment. The factors are: 1) flow velocity,
2) water temperature, 3) silt and corrosion products, 4) system
redundancy and intermittent use, 5) valve leaks, 6) chlorina-
tion effectiveness and system reliability, and 7) component
size.

FLOW VELOCITY

Low-velocity flow appears to be a major factor suppor-
ting the settlement and growth of Corbicula larvae in service
water systems. Power plant personnel indicate that velocities
up to 0.30 mps may be sufficiently low to allow Corbicula lar-
vae to settle. In addition, Corbicula may attach to piping by
secreting a byssal thread (Sinclair, 1963). Once the larvae set-
tle and attach, minor increases in velocity will not detach them.
It is possible that settled larvae may create eddies which result
in silt deposition, thus compounding the fouling problem.

Low-velocity flow also provides an ideal environment for
the deposition of silt and other suspended particles. Fluid
velocities in municipal water systems are typically kept above
1 mps to prevent silting. At one nuclear plant, levels of
suspended solids in the service water reached as high as

10,000 ppm during peak run-off periods. Silt deposits provide
an environment in which Corbicula may grow, and the silt layer
protects the young Corbicula larvae from chlorine levels that
would be toxic if in direct contact with the larvae.

Unlike stagnant water conditions, low-velocity flow pro-
vides a continuous supply of food and dissolved oxygen to
clams in the piping system. Thus, low-velocity flow not only
allows clams to settle and attach to internal surfaces of the
service waste system, it also provides food and oxygen need-
ed for their growth.

Eddies and backwater conditions occur at or near
abrupt changes in flow path geometry in the service water
system and cause low-velocity flow. Low-velocity flow condi-
tions exist in service water inlet structures (Fig. 1), at inlets
to heat exchanger waterboxes, and where there are sudden
changes in pipe diameter. Low velocity may also occur in lines
with leaking or partially open valves.

CLAM CONCENTRATIONS

Fig. 1 . A Typical Inlet Structure Design Showing Areas Where Clams
and Silt have Deposited (Smithson, 1981).

WATER TEMPERATURE

Water temperature is a major factor that determines
whether service water cooling loops will support Corbicula
growth. Although thermal tolerance limits of Corbicula are
dependent on acclimation temperature and life stage, the up-
per limit appears to be between 31 and 35°C, and the lower
limit between 2 and 4°C (Mattice et a!., 1982). Optimum
temperatures for Corbicula growth are in the mid 20°C range
(Mattice et at., 1982). At many plants, the water temperature
of the service water source is above 2°C for most of the year.
Also, the retention period of water held in redundant heat ex-
changers or systems which see intermittent use is generally

JOHNSON ETAL: ENGINEERING FACTORS INFLUENCING FOULING

long enough for the service water to reach room temperature
(approximately 20°C).

Seasonal temperature extremes affect the population
dynamics of Corbicula in the service water source. Low water
temperatures have been known to cause severe winter kills
in Corbicula populations (Bickel, 1966). The greatest popula-
tion increases and most severe fouling problems have oc-
curred in the southern United States where winter
temperatures of the water source typically remain above 2°C.

SILT AND CORROSION PRODUCTS

Fouling caused by silt and corrosion products (primarily
iron-oxide) is often found in conjunction with clams and relic
shells in carbon steel service water piping. Although these
fouling mechanisms may exacerbate clam fouling, similar en-
vironmental conditions are known to promote all three types
of fouling independently. Two utilities have replaced portions
of their carbon steel service water and fire protection system
piping with stainless steel piping to minimize corrosion. No
further corrosion problems have been reported from them to
date.

As stated previously, low velocity flow conditions that
allow settlement of Corbicula also allow deposition of other
suspended particles such as silt. Silt and mud deposits pro-
vide a natural substratum for Corbicula growth, and can act
as a buffer between clam larvae and chlorinated service
water. Silt deposits exhibit a chemical demand for free
chlorine, and a layer of silt covering clams may also reduce
the rate of chlorine diffusion into the layer where the clams
are located. Therefore, the residual chlorine level in the water
may be several times higher than that to which the buried
clams are exposed. Because of the differences in exposure,
residual chlorine levels known to kill clam larvae when in
direct contact may prove ineffective in controlling clam lar-
vae protected by a layer of silt.

At least two methods can be used to remove silt from
piping systems. The Vermont Yankee nuclear plant, for ex-
ample, has used a chemical dispersant to flush mud and silt
from its circulating water system (Electric Light and Power,
1978). The dispersant increases the wettability of mud and
silt and allows normal water turbulence to keep the particles
suspended so they can be flushed away. The use of some
dispersants may, however, be restricted to closed cycle cool-
ing water systems due to the discharge limits imposed on
these chemicals. High velocity flushing is another method of
removing silt from piping systems.

Corrosion products are often found along with deposits
of silt and Corbicula. Although the interaction between corro-
sion products, silt and Corbicula is not completely understood,
there are definite correlations between the presence of silt
and corrosion products as well as the correlations between
silt and Corbicula deposits, as previously discussed.

Two mechanisms believed to cause corrosion of car-
bon steel in the presence of silt deposits are electrochemical
reactions and the presence of sulfides. Electrochemical cor-
rosion in carbon steel piping results from a nonuniform
distribution of dissolved oxygen (Bacon, 1978). This renders

the area exposed to low oxygen concentration anodic with
respect to areas of higher oxygen concentration. Thus, areas
where silt has deposited may become oxygen deficient cells
where electrochemical corrosion can occur. One utility noted
that pitting and corrosion in their fire protection system is more
prevalent on the bottom, inside surface of piping where silt
and organic matter has deposited. Chemical analysis of the
corrosion product revealed the presence of sulfides, the se-
cond corrosion mechanism, which are known to cause ac-
celerated pitting corrosion in carbon steel piping. Sulfides
could result from decomposition of organic matter by sulfate
reducing bacteria in fire protection and cooling water systems.

Other effects of corrosion are reduced flow area and
increased surface roughness, both of which restrict the flow-
carrying capacity of piping. Increased surface roughness, in
particular, may provide a more suitable surface for attach-
ment by clam larvae and increase the thickness of the boun-
dary layer which further promotes the settlement of silt and
clams. Thus, silting, corrosion products, and Corbicula all con-
tribute to degraded flow conditions in the service water
system.

SYSTEM REDUNDANCY AND INTERMITTENT USE

Many components in redundant systems are used in-
termittently and often exhibit low flow and/or stagnant con-
ditions. Several utilities indicated that fouling typically occurs
in systems with low flow, intermittent flow, or stagnant con-
ditons for extended periods of time.

Redundant cooling loops are provided in the essen-
tial service water system and in some nonessential service
water cooling loops to ensure continuous cooling in the event
that one of the redundant coolers fails. Typical cooling loops
with redundant heat exchangers are the containment cool-
ing units, component cooling (or closed cooling) units, and
turbine-bearing lube oil coolers. Systems used intermittent-
ly include those which provide cooling or service water on
demand only. Examples of such systems are containment
cooling units, residual heat removal heat exchangers
(or decay heat removal heat exchangers), and the fire pro-
tection system. Containment cooling units, for example, are
only in service when the temperature inside the reactor con-
tainment vessel exceeds a specified temperature. Many heat
exchangers only receive flow during scheduled testing per-
formed on a weekly, monthly, or even yearly basis.

Both redundant systems and intermittently used
systems are typically maintained full of service water and in
a standby condition. Plant technical specifications call for
periodic flow testing to ensure the operability of these
systems. Several utilities have increased the frequency of
their flow tests after finding Corbicula in the systems. Although
more frequent flow testing may work to flush the system of
silt and small clams, it is possible that increased flow testing
provides a fresh supply of food and water to the clams more
frequently, thus providing a more habitable environment for
Corbicula trapped in protected areas of the system.

An apparently effective means of controlling Corbicula
in redundant and intermittent-use systems is to schedule flow

CORBICULA SYMPOSIUM

tests coincident with service water chlorination. Thus, when
flow testing is completed, the systems are filled with
chlorinated service water and returned to standby condition.
Because service water flow bypasses systems in the stand-
by mode, failure to chlorinate during flow testing means that
systems which would benefit most from chlorination may
never be chlorinated. Since finding Corbicula in their redun-
dant and intermittent use systems, several plants have im-
plemented such schedules and have noted success.

VALVE LEAKS

Valve leaks are another cause of low-velocity, con-
tinuous flow. Although these leaks may be minor from an
engineering standpoint, the flow may be great enough to pro-
vide clams with a continuous supply of food and oxygen.
There is evidence that Corbicula may be less tolerant of reduc-
ed oxygen levels than other fresh water molluscs (McMahon,
1979). As oxygen levels fall below saturation, oxygen uptake
by Corbicula rapidly decreases to approximately ten percent
of that at saturation. Factors such as Clam respiration, the
oxygen demand of bacteria, and formation of corrosion pro-
ducts reduce dissolved oxygen levels in stagnant service
water. Leaking valves, supplying food and dissolved oxygen,
appear to be a primary cause of Corbicula growth in redun-
dant and intermittent use systems which are assumed to be
stagnant.

Two basic types of valve leaks are actual valve
malfunctions, and leaks within the design specifications of
the valve. Valve malfunctions may be corrected with increas-
ed maintenance, but design allowable leaks are a result of
manufacturing tolerances. Valves in the service water system
may normally allow leaks of up to 10% of the design flow
when they are in the closed position. Of the different valve
types, butterfly valves appear to have the highest potential
for leakage.

At one plant the combination of an open inlet valve
and a closed but leaking outlet valve allowed Corbicula and
silt to deposit in the inlet waterbox of a redundant turbine-
bearing lube oil heat exchanger. The open inlet valve allow-
ed silt and clams to enter the heat exchanger and settle in
the waterbox. The leaking outlet valve provided a continuous
flow (approximately 4 L per minute) of fresh water to the clams
and allowed further deposition of clams and silt. The inlet
water temperature was approximately 16°C. The combin-
ation of a continuous low velocity flow of warm service water
and the accumulation of silt provided conditions which per-
mitted Corbicula growth. The heat exchanger was in stand-
by condition for approximately 9 months during which time
clams and silt accumulated to a depth of 7 to 10 cm. This
fouling incident was discovered during a scheduled, visual
inspection of the turbine-bearing lube oil heat exchanger. Dur-
ing this inspection the on-line turbine-bearing lube oil heat
exchanger was found to be completely free of Corbicula and
silt deposits. The high velocity flow through the on-line heat
exchanger kept silt and young clams from settling there. Plant
personnel speculated that the weekly chlorination (30 minutes
at 1 ppm residual chlorine) was ineffective in controlling

clams which entered in the larval stage because of the pro-
tective layer of silt.

CHLORINATION EFFECTIVENESS AND SYSTEM
RELIABILITY

Chlorination effectiveness and the reliability of
chlorination systems are important factors in controlling Cor-
bicula larvae. Chlorination has been shown to be one of the
most effective means of controlling Corbicula larvae (B.G.
Isom, unpub. manuscript, TVA). However, if chlorination is
not properly scheduled or if residual chlorine levels are not
high enough to kill clam larvae, Corbicula may not be kept
to a nuisance level. Also, mechanically unreliable chlorina-
tion systems can effectively halt all chlorination while the
system is down for repair. During this downtime, Corbicula
larvae can enter the service water system and settle in pro-
tected areas.

As mentioned previously, chlorination may be most ef-
fective when scheduled to coincide with clam spawning
seasons (Goss and Cain, 1976) and flow testing or flushing.
Continuous chlorination at 0.5 to 1.0 ppm total residual
chlorine for one or two 3 week periods during the spawning
season is required to control clam larvae (B.G. Isom, unpub.
manuscript, TVA). Corbicula spawning periods are greatest
in the spring and fall, although they are dictated somewhat
by environmental conditions (primarily water temperature).
Continuous chlorination, however, may be an unreasonable
control method given the current ERA regulations on chlorine
discharge from power plants (Mattice etal., 1982). This con-
clusion assumes that the service water effluent is not
dechlorinated before returning to the source waterbody.

The Tennessee Valley Authority (TVA) has im-
plemented a program of continuous chlorination of essential
service water systems which are in service during the clam
spawning seasons. The TVA program calls for chlorinating the
service water to a total residual chlorine level of 0.6 to 0.8
ppm. The program also requires the nonessential service
water systems to be chlorinated to the same level for two 3
week periods corresponding to the beginning and end of the
clam spawning season. During these periods a small continuous
flow of chlorinated service water is established through all
main fire system headers normally exposed to raw service
water. This ensures that when chlorination has been com-
pleted, the fire protection system will remain in standby con-
dition, filled with chlorinated service water. The TVA clam con-
trol program also includes straining all raw water through 0.8
mm mesh screens.

Correct measurement of free residual chlorine levels
is also a factor which impacts the effectiveness of a chlorina-
tion system. As mentioned previously, silt and other suspend-
ed particles in service water have a chemical demand for
chlorine. This factor makes free residua! chlorine levels both
time and space dependent. Free residual chlorine levels
which are measured near the point of injection will be
unrealistically high in comparison to levels measured at ser-
vice water components farther down stream. For this reason
free residual chlorine should be measured downstream from

JOHNSON ETAL.: ENGINEERING FACTORS INFLUENCING FOULING

all components where chlorination is required.

Unreliable chlorination systems can also be a major
factor in allowing larval Corbicula to become established in
service water system piping. Although plant technical
specifications may call for chlorination at specified times dur-
ing plant operation, the chlorination system is not mechanical-
ly “required” for safe operation. Utility personnel indicate that
often chlorination systems do not receive the same level of
maintenance attention as do other systems more critical to
plant operation. Thus, plants have remained in operation for
several months with the chlorination system out of service.
One severe fouling incident related to chlorination system
reliability, although not involving Corbicula, occurred in a salt
water cooled plant. Severe fouling of the residual heat
removal (RHR) heat exchangers by oysters was directly at-
tributed to the chlorination system being out of service for
an extended period.

One utility noted a correlation between the amount of
maintenance required by their diaphram type chlorination
pumps and whether hypochlorite solution is injected upstream
or downstream from the main service water pumps. They
noted that chlorination systems which inject hypochlorite solu-
tion downstream from the service water pumps have a higher
incidence of pump diaphram failure than similar pumps in
systems where hypochlorite solution is injected directly into
the service water intake structure. This difference has been
attributed to the fact that injection downstream of the service
water pumps requires pumping against a back pressure of
approximately 345 kPa. This pressure, while not unusual for
raw water systems, is high enough to shorten the operating
life of these particular diaphram-type injection pumps.

COMPONENT SIZE

Corbicula fouling most often manifests itself in small
diameter components in the service water and fire protec-
tion systems. One utility indicated that fouling from Corbicula,
silt, and corrosion products is most prevalent in pipes of 100
mm and smaller diameter, with chronic fouling occurring in
pipes 50 mm and smaller. An example is fouling of the 76
mm supply lines to their reactor building cooling units. This
utility has replaced much of its small diameter carbon steel
piping with stainless steel piping. They have also replaced
service water system piping less than 25 mm in diameter with
25 mm stainless steel piping.

Several utilities have indicated that heat exchangers
with tube diameters of 13 mm and less foul more readily than
heat exchangers with larger diameter tubes. Pump motor
room coolers have frequently fouled with buildups of silt, cor-
rosion, and Corbicula. These coolers typically have supply
piping less than 100-mm in diameter.

There is some question as to whether Corbicula ac-
tually settle and grow in these small diameter components
or whether these are simply the locations where adult clams
and relic shells accumulate after being carried into the
system. One utility noted that fouling in heat exchangers con-
sisted largely of relic shells and speculated that the clams
had grown elsewhere in the system and after dying had been

flushed into the heat exchangers. There is evidence that as
the number of clams in low velocity areas increases, some
of the clams are forced into high flow areas and are carried
through the system until they lodge in constricted areas (J.S.
Mattice, unpub. manuscript, EPRI). Areas where clams are
typically found are on heat exchanger tube sheets and behind
inlet valves to intermittent use systems in standby mode.

NUCLEAR SAFETY IMPLICATIONS OF CORBICULA
FOULING

Corbicula fouling of the essential and nonessential ser-
vice water systems both directly and indirectly affects the
overall safe operation of nuclear power plants. Fouling of the
essential service water system directly affects reactor safety
because when essential cooling is interrupted during reac-
tor shutdown, an alternate emergency cooling path must be
established. Similarly, fouling of certain nonessential service
water cooling loops may indirectly affect reactor safety by
causing an unscheduled reactor shutdown and thus requir-
ing cooling from safety related essential service water cool-
ing loops.

An example of essential service water heat exchangers
which have been fouled by Corbicula and which may direct-
ly impact reactor safety are the containment fan cooling units
(Sometimes called containment cooling units or reactor
building cooling units) which are common to PWR plants. The
containment fan cooling units are designed to remove heat
from the containment building during both normal and acci-
dent conditions. In Westinghouse PWRs there are a total of
five units which operate in parallel (Masche, 1971). During
normal operation a maximum of four units are required to
remove the design heat load. Therefore, during normal opera-
tion one of two cooling units are in standby mode. Other PWR
designs also have four to five units with one or two on stand-
by during normal operation.

During normal operation, if containment cooling re-
quirements are not met because of flow blockages, reactor
power would have to be reduced to bring the containment
temperature down. During accident conditions, severe
fouling of the coolers would require that alternate contain-
ment cooling be established. Fouling of containment cool-
ing units with Corbicula has, in fact, forced the shutdown of
a nuclear plant while the coolers were cleaned and restored
to their design capacity.

Turbine-bearing lube oil coolers are an example of
nonessential service water heat exchangers that have foul-
ed with Corbicula. As the name implies, turbine-bearing lube
oil coolers provide cooling to the turbine-bearing lubricating
oil. Turbine-bearing lube oil typically begins to lose its
lubricating ability at temperatures above 150°C. The
temperature of the turbine-bearing lube oil is therefore
monitored and if flow blockage of both the online and backup
lube oil coolers causes it to exceed the allowable temperature
(somewhat below the 150°C maximum), a turbine trip would
be initiated. If the cause of the turbine trip was not readily
apparent or if it were not possible to clean these heat ex-
changers while the reactor was on line, a reactor shutdown
would follow.

CORBICULA SYMPOSIUM

Upon initiation of shutdown, initial reactor cooling in
a PWR is achieved by dissipating heat through the steam
generators and discharging steam to the condensers by
means of the turbine steam bypass system. The residual heat
removal system (RHR) begins removing heat from the re-
actor when the reactor water temperature and pressure have
dropped to approximately 177°C and 2.75 MPa, respective-
ly. The RHR heat exchangers are cooled by the component
cooling loop which is in turn cooled by the component cool-
ing water heat exchangers. The component cooling water
heat exchangers are cooled by service water and are part
of the essential service water system. Therefore, although
the turbine-bearing lube oil coolers are not safety related, their
fouling could cause a reactor shutdown which relies on safety
related service water systems for cooling.

ENGINEERING RECOMMENDATIONS TO MINIMIZE
CORBICULA FOULING

Many actions can be taken to reduce Corbicula foul-
ing to a nuisance level. Some methods such as low-level, con-
tinuous chlorination during Corbicula spawning seasons could
be used to control establishment of clam larvae in the plant.
Continuous chlorination, however, may not be possible given
the current EPA regulations on chlorine discharge from power
plants (Mattice et a!., 1982). Adult Corbicula are best con-
trolled by physically removing them from the intake structure
and internal surfaces of the service water system.

Conditions that promote Corbicula settlement and
growth are costly and should be avoided. Fouling can be
monitored and controlled in several ways. Systems in stand-
by condition during plant operation should be chlorinated
during flow tests, and visual inspections of the system inter-
nals should be performed during outages. Areas such as in-
take structures, heat exchanger waterboxes and other low-
velocity flow areas should be visually inspected during
outages for accumulations of Corbicula. Also, many heat ex-
changers in nuclear service water systems have no individual
flow metering devices. Installation of such devices would pro-
vide early detection of fouling problems during operation
without visual off-line inspections.

During extended outages, service water cooling loops
that are not required for removing residual heat from the reac-
tor or other essential cooling could be dewatered. Buildup
of silt and corrosion products in the service water and fire
protection system should be minimized. Leaky valves or those
that do not operate correctly should be repaired to avoid low-
velocity flow conditions. The installation of fine mesh
strainers, either at the service water pump discharge or at
the inlets to heat exchangers, can greatly reduce the occur-
rence of flow blockages due to Corbicula fouling. Finally, the
chlorination system should be maintained with the same level

of care as the essential service water system, because ef-
fective chlorination indirectly affects the safety related cool-
ing functions of the essential service water system.

ACKNOWLEDGEMENTS

This study was supported by DOE under a Related Services
Agreement (Contract DE-AC06-76RL0 1830) with the Siting and En-
vironmental Branch of the Nuclear Regulatory Commission. Pacific
Northwest Laboratory is operated by Battelle Memorial Institute for
the U.S. Department of Energy.

Utilities and nuclear plants that provided information are not
identified unless requested by the utility. This is to honor the request
of many utilities that they remain anonymous.

The authors would like to acknowledge the contributions of
Linda Krumbah in typing the manuscript, and Carolyn Novich and
Duane Neitzel in providing editorial comments.

LITERATURE CITED

Bacon, H. E. 1978. Corrosion. IN: Mark’s Standard Handbook for
Mechanical Engineers. 9th ed. McGraw Hill, New York, pp. 6-106
to 6-117.

Bickel, E. 1966. Ecology of Corbicula manilensis Philippi in the Ohio
River at Louisville, Kentucky. Sterkiana 23:19-24.

Electric Light and Power. 1978. Mud-flushing additive hikes con-
denser heat transfer. Electric Light and Power (Boston), 56:7.
25 pp.

Goss, L. B. and C. Cain. 1976. Condenser and raw water system
fouling by Corbicula. IN: L. D. Jensen, ed. Biofouling Control
Procedures. Marcel Dekker, Inc. New York, pp. 11-22.
Haried, J. A. 1982. Evaluation of events involving service-water
systems in nuclear power plants. NUREG/CR-2797, U.S.
Nuclear Regulatory Commission, Washington, D.C. 62 pp.
Masche, G. 1971. Systems summary of a Westinghouse pressuriz-
ed water reactor nuclear power plant. Westinghouse Nuclear
Energy Systems, Pittsburgh, Pennsylvania. 238 pp.
Mattice, J. S. 1979. Interactions of Corbicula sp. with power plants.
IN: J. C. Britton, J. S. Mattice, C. E. Murphy and L. W. Newland,
eds. Proceedings of the First International Corbicula Sym-
posium. Texas Christian University Research Foundation, Fort
Worth, Texas, pp. 119-138.

Mattice, J. S., R. B. McLean and M. B. Burch. 1982. Evaluation of
short-term exposure to heated water and chlorine for control
of the Asiatic clam ( Corbicula fluminea). ORNL/TM-7808,
Environmental Sciences Division, Oak Ridge National
Laboratory, Oak Ridge, Tennessee. 34 pp.

McMahon, R. F. 1979. Response to temperature and hypoxia in the
oxygen consumption of the introduced Asiatic freshwater clam
Corbicula fluminea (Muller). Comparative Biochemistry and
Physiology 63A:383-388.

Sinclair, R. M. 1963. Clam pests in Tennessee water supplies.
Kentucky-Tennessee Section Meeting of the Tennessee
Stream Pollution Control Board, Nashville, Tennessee. 8 pp.
Smithson, J. A. 1981. Control and treatment of Asiatic clams in power
plant intakes. IN: Proceedings, American Power Conference.
American Power Conference, Chicago, Illinois, Vol. 43, pp.
1146-1151.

CORBICULA CONTROL AT THE POTOMAC RIVER STEAM

ELECTRIC STATION
ALEXANDRIA, VIRGINIA

JEANNE MILES POTTER AND LAWRENCE H. LIDEN
WATER AND LAND USE DEPARTMENT
POTOMAC ELECTRIC POWER COMPANY
1900 PENNSYLVANIA AVE., N.W.
WASHINGTON, DC 20068, U.S.A.

ABSTRACT

The successful colonization of the freshwater tidal Potomac River by the Asiatic clam has resulted
in severe macrofouling problems at the Potomac Electric Power Company’s (PEPCO) Alexandria,
Virginia Station. Planktonic veligers and small clams entered the plant through the traveling screens,
settled, and attached in slow velocity areas. Fouling problems occurred when clams died and their
shells clogged condenser tubes.

Studies were conducted to ascertain seasonal growth rates of Corbicula in relation to station
fouling and to monitor the effects of thermal effluent on Corbicula populations. Growth rates, as deter-
mined by increases in shell height, were inversely related to size. Mean cumulative increases in shell
height were 9.8 and 7.9 mm for size class III and IV clams, respectively. Corbicula growth rates were
not significantly different among control and thermally-influenced stations. Based upon traveling screen
mesh size of 10 mm, clams < 10 mm could enter the plant and grow, within 1 season, to a size where
they could clog condenser tubes. However, on the basis of plant operational data and in-plant sam-
pling, it was concluded that biennial physical removal was a successful control method.

Corbicula sp., the Asian clam has increased rapidly
in ecological and economic significance since its first sighting
in the United States in 1938 (Burch, 1944 as cited in Mat-
tice, 1983). Since its introduction, the range of the Asiatic clam
has expanded to include most of the freshwater drainage
basins in the United States (Mattice and Dye, 1976). Its high
fecundity, incubatory egg, and planktonic veliger’s ability to
infest raw water supplies has resulted in considerable biofoul-
ing problems for the power industry (Sinclair and Isom, 1963;
McMahon, 1977).

The United States Geological Survey first found Cor-
bicula concentrations in the tidal freshwater portions of the
Potomac River in 1977 (Dressier and Cory, 1980). Within two
years, clams were found fouling intake and condenser areas
of the Potomac Electric Power Company’s Potomac River
Steam Electric Station (SES) in Alexandria, Virginia.
Planktonic veligers and small dams (< 10 mm) apparently
entered the plant through the traveling screens, settled, and
attached in slow velocity areas behind the traveling screens.
Fouling problems occurred when the clams died and their
shells wedged within condenser tubes. Losses at the Potomac
River SES due to Corbicula fouling were divided into three
areas: (1) the efficiency loss due to higher condenser
backpressures; (2) the megawatt losses due to inability to
achieve a full load; and (3) the outage losses associated with

cleaning the condensers. Physical removal of clams during
scheduled overhauls was implemented to control Corbicula
fouling.

The Potomac River SES was inspected several times
during 1979 for Corbicula infestations. Clams were found in
the intake area behind the traveling screens, at the base of
the circulating water pumps, and lodged in tube openings
in the condenser waterboxes. Size distribution analyses of
the clams showed most were smaller in diameter than the
condenser tube openings. The Raw Service Water lines were
inspected in 1981 and no Corbicula were found.

Studies were conducted by PEPCO’s Water and Land
Use Department during 1980 and 1981 to provide suppor-
ting data for Corbicula control methodology. The objectives
of these studies were: (1) to ascertain the seasonal growth
rate of Corbicula in relation to the potential for power plant
fouling; and (2) to monitor the effects of the thermal effluent
on Corbicula populations in the vicinity of the Potomac River
SES.

METHODS AND MATERIALS

The Potomac River SES, located in Alexandria,
Virginia, has five coal-fired steam turbine generators with a
net capacity of 508 Megawatts. Water is withdrawn from the

American Malacological Bulletin, Special Edition No. 2(1986):53-58

CORBICULA SYMPOSIUM

Fig. 1. Location of Potomac River SES in Alexandria, Virginia, and 1981 Corbicula study locations.

POTTER AND LIDEN: CORBICULA CONTROL

Fig- 2. Underwater platforms used to hold Corbicula for growth studies
in the Potomac River, May through October, 1981.

Potomac River for once-through condenser cooling. Max-
imum plant cooling water flow is 19.8 m3/sec. The cooling
water is intermittently chlorinated 3 times daily maintaining
in-plant total residual chlorine concentration of 0.20 mg// for
15 min. The heated effluent is discharged approximately 300
m downstream of the intake. Partial recirculation of discharge
water occurs as a result of tidal action.

The Potomac River at Alexandria is relatively shallow
(0-4 m deep). Net freshwater flow averages 1 ,980 m3/sec and
ranges from 19.8 m3/sec to 2,547 m3/sec.

Growth and survival of tray-held Corbicula were
monitored monthly at 5 stations in the vicinity of the Potomac
River SES, May through October 1981 (Fig. 1). Station 1
was located 3 km upstream of the thermal discharge; Sta-
tions 3 and 4 were directly in the thermal discharge; Station
5 was 4 km downstream of the thermal discharge. Station
2 was located in the intake area and was subject to thermal
recirculation.

Corbicula were placed in oyster trays secured to under-
water platforms (Fig. 2). The underwater platforms were
constructed of polyurethane-coated steel angle and rod stock
and measured 46 x 46 x 36 cm. The platforms were con-
nected with stainless steel cable to a 30 x 30 x 30 cm con-
crete block which was, in turn, attached to an identifying buoy.
The trays were constructed of vinyl-coated stainless steel wire
with hinged tops and measured 45 x 41 x 13 cm. Each tray
was divided into two compartments and lined with ex-
panded polyethylene mesh for separation of Corbicula size
classes.

Corbicula were dredged from the Potomac River below
the confluence of the Potomac and Monacacy Rivers, near

Dickerson, Maryland. Clams were randomly selected,
measured, and sorted into size classes. Size class determina-
tions were based on shell length (after Sickel, 1973, as cited
in Gardner et al. , 1976; see Table 1, Gardner et al., 1976).
Only size class III (13.5-18.5 mm) and IV (>18.5 mm) clams
were used because of insufficient numbers of smaller
Corbicula.

Twenty-five class III and one-hundred size class IV Cor-
bicula were placed in each compartment. Corbicula shell
height and length were measured monthly to determine
growth. Shell dimensions were determined according to
McMahon (1977). Shell height to length ratios averaged 0.97
± 0.05 (n = 500); therefore only height was presented.

Percent mortality was calculated by dividing the
number of dead clams by the total number of dead and live
clams.

Temperature was recorded continuously at each sta-
tion using Endeco Type 109 Recording Thermographs. Sur-
face and bottom dissolved oxygen concentrations were deter-
mined monthly using the azide modification of the Winkler
titration (APHA, 1980). Conductivity (Beckman RS-5 meter)
and turbidity (FI. F. Instruments, model PRT-15) also were
measured monthly at each station.

Means and standard deviations were calculated us-
ing the small sample size approximation in Sokal and Rohlf
(1969). Analysis of variance (ANOVA) was used to test for
station or time effects on shell height for size classes III and IV.

Fig. 3. Changes in mean shell heights of tray clams in the vicinity of
the Potomac River SES, May through October, 1981. Clam height data
from all stations are combined.

CORBICULA SYMPOSIUM

RESULTS AND DISCUSSION

Corbicula growth, as determined by increases in shell
height, generally was continuous throughout the study for size
class III and IV clams (Fig. 3). Cumulative increases in shell
height among size class III Corbicula ranged from 39% at Sta-
tion 5 and 74% at Station 3 with an overall mean increase
of 61% (Table 1). Shell height for size class III clams increas-
ed an average of 9.8 mm from May through October, 1981
(Table 2). Mean monthly increase was 2.0 mm. Minimum and
maximum increases in shell height for size class III Corbicula
occurred in June and July, respectively. Decreases in mean
shell height were noted in June at Station 1 and September
at Station 5. The June decrease at Station 1, in addition to
smaller increases in shell height at other stations, probably
resulted from field measurement errors, since little clam mor-
tality occurred. However, September decreases at Station 5
were attributed to reduced sample size resulting from
cumulative mortality.

Cumulative increases in shell height among size class
IV Corbicula ranged from 31 % at Station 4 to 41 % at Station
5, with a mean increase of 35% (Table 1). Shell height in-
creased an average of 7.9 mm among size class IV clams
at all stations throughout the study (Table 2). Mean monthly
increase was 1 .6 mm. Smallest and largest increases in shell
height occurred in June and September, respectively.
Decreases in shell height were observed in June at Station
3, August at Station 5, and in October at Station 4. The June
decrease in shell height, in addition to the low rate of growth
at other stations during this month, probably resulted from
field measurement errors, since little mortality occurred.
However, decreases during August and October were at-
tributed to reduced sample size resulting from cumulative
mortality.

Growth studies conducted at the Potomac River SES
from June through October, 1980 yielded results similar to
those obtained in this study (PEPCO, 1981). Mean cumulative
shell height increases for size class III Corbicula were 7.6 mm
for 4 months in 1 980 and 9.8 mm for 5 months in 1 981 . Mean
monthly increases in shell height were 1 .9 and 2.0 mm dur-
ing 1980 and 1981, respectively. No data were obtained for
size class IV Corbicula in 1980. Other researchers also have
reported similar growth rates. Auerbach et al. (1978)
monitored Corbicula growth at the Kingston Steam Electric
Plant, Tennessee from May through October, 1976.
Cumulative growth of cage-confined clams was 6.5 mm and
7.5 mm at the station intake and discharge, respectively. Brit-
ton et al. (1979) reported Corbicula growth rates of 0.67, 0.58,
and 0.47 mm per month for 3 clam groups ranging from 14-21
mm, 21-24 mm, and 24-32 mm, respectively.

Analyses of Variance were performed to determine the
effects of location and time on Corbicula shell height for both
size classes (Table 3). Monthly changes in mean shell height
were significantly different (p = 0.05) for size classes III and
IV clams. However, station effects were not significantly dif-
ferent (p = 0.05) for either size class. Station-related thermal
effects were not a source of variance. Temperatures between
stations varied less than 2°C throughout the study.

Table 1. Cumulative percent change in mean Corbicula shell height
in the vicinity of the Potomac River SES, May through October, 1 981 .

Size

Class

Measure-

ment Date
(Month)

Station

Average

III

Jun

-2.1

6.3

2.8

9.4

4.8

4.2

Jul

20.5

24.4

26.7

32.7

16.1

24.1

Aug

37.1

45.3

43.3

46.3

23.2

39.0

Sep

60.4

58.9

64.8

56.8

18.2

51.8

Oct

67.6

62.0

74.4

64.2

39.4

61.5

Jun

3.1

1.3

-6.0

1.6

4.6

0.9

Jul

9.7

4.2

-0.7

17.7

14.2

9.0

Aug

14.2

16.1

11.2

24.3

11.7

15.5

Sep

34.0

30.0

25.9

34.8

40.8

33.1

Oct

34.5

33.9

30.5

40.8

34.7

Table 2. Monthly changes in mean shell height (mm) of Corbicula
in the vicinity of the Potomac River SES, May through October, 1981.

Size

Class

Measure-
ment Date
(Month)

Station

Average

III

Jun

-0.38

1.00

0.44

1.54

0.8

Jul

3.68

2.90

3.83

3.82

1.85

Aug

2.68

3.34

2.65

2.24

1.18

Sep

3.74

2.19

3.44

1.71

-0.83

Oct

0.36

0.49

1.55

1.22

3.50

TOTAL

10.08

9.92

11.91

10.53

6.50

9.79

MEAN

2.02

2.08

2.38

2.01

1.30

1.96

Jun

0.70

0.29

-1.37

0.35

0.97

Jul

1.50

0.65

1.21

3.51

2.06

Aug

1.01

2.64

2.73

1.44

-0.53

Sep

4.47

4.12

3.36

2.28

6.20

Oct

0.11

0.86

1.84

-0.94

0.00

TOTAL

7.79

8.56

7.77

6.64

8.70

7.88

MEAN

1.56

1.71

1.55

1.33

1.74

1.58

Table 3. Analysis of variance for mean monthly changes in Corbicula
shell height in the vicinity of the Potomac River SES, May through
October, 1981.

Source

Date

Station

Error

18.60534

3.19342

23.89350

SIZE

III HEIGHT

4.65133

0.79835

1.49334

3.11

0.53

P>F

0.044 *
0.7123 NS

Source

SIZE

IV HEIGHT

P>F

Date

48.64721

12.16180

7.09

0.0018 *

Station

0.53838

0.13459

0.08

0.9878 NS

Error

27.44674

1.71542

* Significant at P=0.05.

POTTER AND LIDEN: CORBICULA CONTROL

Table 4. Corbicula survival in the vicinity of the Potomac River SES, May through October, 1981.

Station

Size

Class

Month

Live

Dead

Live

Dead

Live

Dead

Live

Dead

Live

Dead

III

May

Jun

Jul

Aug

Sep

Oct

May

100

103

Jun

103

Jul

Aug

Sep

Oct

Corbicula survival was similar for size classes III and

IV until August (Table 4). Large mortalities were observed
among both size classes in August; moreover, those for size
class IV were substantially greater. The large number of Cor-
bicula mortalities paralleled observations of Water and Land
Use personnel conducting ichthyoplankton sampling within
the study area. Numerous Corbicula viscera were observed
in the water column during July and August. These occur-
rences of mortality coincided with maximum ambient water
temperatures (30°C) in the Potomac River, as indicated by
Endeco temperature data. However, this temperature-
mortality relationship was not supported by reported upper
thermal tolerance limits of 34°C for Corbicula (Mattice and
Dye, 1976).

Cumulative survival for size class III and IV clams
averaged 38% and 18%, respectively, at all stations (Table
5). Survival at Stations 1, 2, 3, and 4 was not substantially
different among size classes. However, cumulative survival
at Station 5 was only 8% (Table 5).

CONCLUSIONS AND SUMMARY

Corbicula growth, determined by increases in shell
height, generally was continuous throughout the 6 month
study. Increases in clam shell height were inversely related
to size; smaller clams grew faster. Mean cumulative increases
in shell height for size classes III and IV were 9.8 and 7.9
mm, respectively. Corbicula growth was not influenced by sta-
tion location. Growth at control and thermally influenced sta-
tions was not significantly different (p = 0.05).

Survival of experimental, as well as natural, Corbicula
populations in the Potomac River was poor during the 1981
study .The period of greatest clam mortality coincided with
the period of maximum river temperature. However, this
temperature-mortality relationship was not supported by
reported thermal tolerance data for Corbicula.

Similar trends for growth and survival were found for

Table 5. Cumulative survival of Corbicula in the vicinity of the
Potomac River SES, May through October, 1981.

Size

Class

Measure-
ment Date
(Month)

Station

Average

III

May

100

Jun

100

104

100

Jul

100

Aug

Sep

Oct

37.8

May

100

Jun

100

Jul

Aug

Sep

Oct

18.2

1 980 and 1981. Mean monthly growth rates were 1.9 mm and
2.0 mm for size class III Corbicula during 1980 and 1981,
respectively. Large clam mortalities were observed in August
during both years in test populations.

Although Corbicula spp. has successfully inhabited the
Potomac River in the vicinity of the PEPCO Potomac River
SES since 1977, the results of this study indicated that the
plant’s thermal effluent has not influenced clam growth and
survival during the spring, summer, or fall. In addition, it ap-
peared unlikely, on the basis of the small Delta T discharged
by the plant (1-2°C), that clam populations in the vicinity
have been enriched and maintained by increasing
temperatures to within the clam’s lower thermal tolerance

CORBICULA SYMPOSIUM

range during the winter.

Clams < 10 mm potentially could enter the plant
through the traveling screens (mesh size of 10 mm) and grow
8-10 mm, within 1 season, to a size at which they could clog
condenser tubes (17-19 mm). However, on the basis of plant
operating data and in-plant sampling, it was concluded that
biennial physical removal (every 18 mos) was a successful
control method. Size frequency distributions of Corbicula in
the plant indicated that the clams do not grow to a sufficient
size to cause fouling problems. Plant operating data also
shows that condenser efficiency has not been affected by Cor-
bicula fouling.

ACKNOWLEDGEMENTS

The authors extend their appreciation to Mr. David Bailey for
contributions throughout the study. Ms. Phyllis Frere and Mr. Paul
Willenborg assisted in field collections. Ms. Janice McCarren typed
the final manuscript. Special thanks is given to Ms. Phyllis Frere for
preparing the graphics. This study was supported by the Potomac
Electric Power Company.

LITERATURE CITED

American Public Health Association, American Water Works Assoc-
iation and Water Pollution Control Federation. 1980. Standard
methods for examination of water and wastewater, 15th ed.
American Public Health Association, Washington, D.C. 1181
PP-

Auerbach, S. I., D. J. Nelson, and E. G. Struxness. 1978. Environmen-
tal Sciences Division annual progress report for the period
ending 30 September 1977. ORNL-5365. Oak Ridge National
Laboratory, Oak Ridge, TN.

Britton, J. C., D. R. Coldiron, L. P. Evans, Jr., C. Golightly, K. D.
O’Kane, and J. R. TenEyck. 1979. Reevaluation of the growth

pattern in Corbicula fluminea (Muller) IN: J. C. Britton, ed., Pro-
ceedings, First International Corbicula Symposium. The Texas
Christian University Research Foundation, Fort Worth, TX. pp.
177-192.

Burch, J. Q. 1944. Checklist of West American mollusks, Family Cor-
biculidae. Minutes of the Conchological Club of Southern
California 36:18.

Dressier, P. V., and R. L. Cory. 1980. The Asiatic clam, Corbicula

fluminea (Muller), in the tidal Potomac River, Maryland.
Estuaries 3:150-152.

Gardner, J. A., Jr., W. R. Woodall, Jr., A. A. Staats, Jr., and J. F.
Napoli. 1976. The invasion of the Asiatic clam ( Corbicula
manilensis Philippi) in the Altamaha River, Georgia. The
Nautilus 90:117-125.

Mattice, J. S. 1983. Freshwater macrofouling and control with em-
phasis on Corbicula. Paper presented at Symposium on Con-
denser Macrofouling Control Technologies— The State of the
Art, June 1-3, 1983, Hyannis, MA.

Mattice, J. S., and L. L. Dye. 1976. Thermal tolerance of the adult
Asiatic clam. Pages 130-135 in G. W. Esch and R. W. Mac-
Farlane, eds., Thermal Ecology II. CONF-750425. National
Technical Information Service, Springfield, VA.

McMahon, R. F. 1977. Shell size-frequency distributions of Corbicula
manilensis Philippi from a clam-fouled steam condenser. The
Nautilus 91 : 54-59.

PEPCO. 1981 . Asian clam studies at the Potomac River Steam Elec-
tric Station. Washington, D.C. 22 pp.

Sickel, J. B. 1973. A new record of Corbicula manilensis (Philippi)
in the southern Atlantic slope region of Georgia. The Nautilus
87:11-12.

Sinclair, R. M., and B. G. Isom. 1963. Further studies on the introduced
Asiatic clam (Corbicula) in Tennessee. Tennessee Stream
Pollution Control Board, Nashville, TN. 77 pp.

Sokal, R. R., and F. J. Rohlf. 1969. Biometry. W. H. Freeman and
Co., San Francisco, CA. 797 pp.

A MECHANICAL STRAINER DESIGN FOR CORBICULA FOULING
PREVENTION IN THE SERVICE WATER SYSTEM AT
ARKANSAS NUCLEAR ONE, UNIT 2

D. DAVID MACPHEE

ARKANSAS POWER & LIGHT COMPANY
P.O. BOX 551

LITTLE ROCK, ARKANSAS 72203, U.S.A.

ABSTRACT

Reduced flow in small safety grade lube oil coolers was found to be caused by accumulations
of the Asian clam Corbicula sp. in the service water supply lines. Installation of conventional, nuclear-
grade strainers was the preferred solution, but the twelve month delivery schedule on vendor de-
signed nuclear components forced consideration of other means of obtaining strainers.

In response, Arkansas Power and Light Company contracted qualified consultants and welders
to design and fabricate “clam traps” from standard piping components. Strainers were completed
within nine months and installed during the next refueling outage. These strainers were designed
with provisions for periodic flushing and cleaning as well as passive removal of clams and debris.
This design with plant specific modifications may be applied to other utilities with similar problems.

Arkansas Power & Light Company (AP&L) operates
a two unit nuclear power plant, Arkansas Nuclear One (ANO),
on Dardanelle Reservoir near Russellville, Arkansas. Reser-
voir water is the primary cooling source for ANO’s service
water system. This system is designed to stringent nuclear
standards to assure safe plant operation in the event of
postulated design accidents such as a loss of reactor coolant,
earthquake, floods, etc.

Water is withdrawn from the reservoir for once through
cooling during normal plant operation. Traveling screens (9.5
mm mesh) and basket strainers (4.7 mm mesh) are provid-
ed in the intake structure to remove large debris. Historical-
ly, shock chlorination was used to control biological fouling.
However, because of chlorination equipment reliability pro-
blems and the method used, control of organisms and
biologicaly induced corrosion was not totally effective.

Many heat exchangers in a nuclear power plant are
used only during emergencies or to allow safe plant shut-
down. Except during monthly surveillance flow tests, supply
piping to these exchangers had low flows of fresh water due
to leaking isolation valves, and thus provided an ideal environ-
ment for growth of Corbicula and deposition of river silt
(Neitzel et al., 1984) In the past Corbicula passed through
the intake screens in the larval stage and burrowed into silt
deposits. Buried in these deposits, the clams were largely
immune to intermittent chlorination, and were free to grow
and multiply. Adult clams and clam debris subsequently broke
loose and clogged various coolers and inlet headers, which
required plant shutdown to remove the debris.

Seven small pump oil coolers were found to be most
susceptible to plugging by clams, silt and other debris. Be-
ing at the lowest elevation of the service water system and
in a low flow area, a large amount of debris accumulated in
the water supply piping. In January, 1982, the “B” Low
Pressure Safety Injection (LPSI) pump oil cooler was found
clogged with clams and debris. It was determined that suscep-
tibility to clam plugging was greatest in the small diameter
0.5-0.75 inch (12.7-19.0 mm) supply tubing on the cooler. As
part of an overall upgrade program which included piping
replacement, chemical cleaning, and additional chlorination,
AP&L decided to install straining devices in the supply lines
to “B” LPSI cooler and six other similar coolers as soon as
possible. The target date for installation was September 1982,
the next refueling outage.

DESIGN APPROACH

Initially, efforts were made to buy nuclear grade
simplex or duplex type strainers. However, procurement of
vendor-design nuclear grade strainers was estimated to re-
quire a minimum of twelve months. Since strainers were
needed within six months for installation during the next
refueling outage, AP&L was forced to consider other means
of obtaining them.

The next option considered was purchasing strainers
fabricated for another utility with a cancelled nuclear plant.
However, due to site specific requirements of nuclear plants
and type of strainers available, this option was not feasible.

American Malacological Bulletin, Special Edition No. 2(1986):59-61

CORBICULA SYMPOSIUM

It was determined that nuclear grade strainers similar
to “clam traps” used by the Tennessee Valley Authority in
fossil fueled power plants (Goss and Cain, 1975) could be
fabricated using standard nuclear grade piping, valves, and
fittings. A conceptual design prepared by AP&L was qualified
to meet nuclear requirements by a consultant. Nuclear
qualified welders available in-house fabricated the strainers.
This option proved to be cost-effective, feasible and
achievable within the time limitations.

STRAINER DESIGN REQUIREMENTS

Strainers were designed to remove debris larger than
0.25 inch (6.4 mm) diameter which could clog cooler inlet pip-
ing. The strainer open area was five times the inlet area to
prevent excessive pressure drop. Design temperature and
pressure were 130°F (55°C) and 150 PSIG (1034 KPa), re-
spectively. Stresses due to internal pressure, thermal expan-
sion, deadweight, and seismic forces were considered. Pro-

visions for on-line flushing of the strainer element, and off-
line cleaning were provided. The design also included provi-
sions for differential pressure measurement across the
strainer.

STRAINER DESIGN AND OPERATION

The strainer (Fig. 1) is basically an arrangement of
standard piping components positioned to form an inner pipe
and an outer pipe aligned along their centerlines (Paulsen,
et a/., 1982).

The inner pipe is drilled with a staggered arrangement
of 0.25 inch (6.4 mm) holes to serve as the strainer element.
The outer pipe serves as the shell and pressure boundary.
The inlet end of the shell is sealed with an assembly of a stan-
dard butt-weld pipe cap and an integrally welded socket weld
coupling. The opposite end is a bolted flange connection with
slip-on and blind flanges. The strainer element penetrates
the blind flange, is welded on both sides, and connects on

MACPHEE: MECHANICAL STRAINER DESIGN

the outside to a flush valve. A “sockolet” is welded to the
side to provide the outlet connection. An additional sockolet
connection is provided on the shell to allow for pressure drop
measurements.

In operation, water laden with clam shells and other
debris flows through the inlet into the inner strainer tube
assembly. Water flows circumferentially into the outer plenum
while the debris is retained by the strainer element. The
strained water then exits through the outlet nozzle in the shell.
Periodically, based on pressure drop through the component,
the flush valve is opened and system pressure forces the ac-
cumulated debris to drain through the valve. For cleaning and
maintenance, the flush end assembly including the blind
flange, flush valve, and strainer element is removable in one
piece. Blind flanges are maintained as spare parts for use
if work on the strainer element is required.

ASME BOILER AND PRESSURE VESSEL CODE
REQUIREMENTS

Certain sections of the ASME Boiler and Pressure
Vessel Code were used for design of the strainers (ASME,
1980). Strainers were designed in accordance with Section
III, Nuclear Components. Materials for the strainers were in
accordance with Section II. Welding requirements were in ac-
cordance with Section IX. Fabrication and installation of the
strainers were in accordance with Section XI.

Code symbol stamping is normally required for ven-
dor supplied equipment (i.e. “N”, “NPT”). However, since
components were fabricated by AP&L per Section XI re-
quirements, the code symbol (“NA”) was not required to be
stamped on the strainers.

To avoid classification as a pressure vessel under the
ASME Code, Section VIII, 5 inch (13 cm) schedule 40 pipe
was used for the strainer shell. Under the ASME Code, the
strainers are considered to be piping components. In this
fashion, the design of the strainers did not have to meet Sec-
tion VIII rules and requirements, and thus did not require the
“U” code symbol stamp.

INSTALLATION

Strainers were installed in the September 1 982 refuel-
ing outage, some nine months after project inception . Due
to extensive system measurements and checking for fit, in-
stallation of strainers was accomplished with only minor
clearance problems.

RESULTS

To date (August 1985), installed strainers have served
their intended purpose. As well as preventing clam fouling
of associated coolers, the effectiveness of other control
measures (piping replacement, chemical cleaning, con-
tinuous chlorination) may be determined from the frequen-

cy, type, and amount of biological material collected. In ad-
dition, several leaking isolation valves have been or will be
replaced in the service water system, removing the low water
flows necessary for clam survival in normally stagnant water
lines.

The total cost for the study, design, fabrication and
installation of seven strainers and associated instrumenta-
tion was $254,000 for a total installed cost of $36,300 per
strainer in 1983 dollars. The total includes a material cost of
$25,400. Comparable strainers purchased from a vendor
would be in the range of $20,000 to $25,000, each.

CONCLUSION

A mechanical clam fouling prevention system has been
designed, fabricated and installed to protect small safety
grade lube oil coolers at ANO-2. The strainer is readily adap-
table to any nuclear utility, and can be much cheaper and
obtained faster than strainers purchased from vendors. The
design approach may be used for other nuclear - grade com-
ponents as well.

ACKNOWLEDGEMENTS

I wish to thank the following for their help in the preparation
of this article. Bob West provided biological details and encourage-
ment. Rick Lane and Bill Eaton checked technical details and proof-
read the manuscript. To the many others who fabricated and installed
the strainers, you have my deep appreciation.

LITERATURE CITED

ASME Boiler and Pressure Vessel Committee, 1980. ASME Boiler
and Pressure Vessel Code. Including all addenda up to and
including the Winter, 1981 addenda. American Society of
Mechanical Engineers, New York. Section II, “Material
Specifications”, 2087 pp. ; Section III, “Rules for Construc-
tion of Nuclear Power Plant Components”, Division 1 , Subsec-
tion ND, Class 3 components, 295 pp.; Section VIII, “Rules
for Construction of Pressure Vessels, Division 1”, 611 pp.;
Section IX, “Welding and Brazing Qualifications”, 267 pp.;
Section XI, “Rules For Inservice Inspection of Nuclear Power
Plant Components”, 593 pp.

Goss, L. B. and C. Cain, Jr., 1975. Power plant condenser and ser-
vice water fouling by Corbicula, the Asiatic clam. Presented
at Biofouling Workshop, Electric Power Research Institute and
Maryland Power Plant Siting Program. Johns Hopkins Univer-
sity, Baltimore, Maryland. June 16-17. 7 pp. (unpublished).
Neitzel, D. A., K. I. Johnson, T. L. Page, J. S. Young, and P. M.
Daling. 1984. Bivalve Fouling of Nuclear Power Plant Service
Water Systems, Volume 1, Correlation of Bivalve Biological
Characteristics and Raw Water System Design.
NUREG/CR-4070, PNL-5300. U. S. Nuclear Regulatory Com-
mission, Washington, D C. 119 pp.

Paulsen, W. A., G. A. Wiederstein, S. L. Weiland. 1982. Stress and
Hydraulic Analysis for Service Water Fabricated Pipe Strainers
- Arkansas Nuclear One - Unit 2. Design Report. Nutech
Engineers, San Jose, California. 93 pp (unpublished).

DEVELOPMENT OF A CORBICULA CONTROL TREATMENT
AT THE BALDWIN POWER STATION

JAMES A. SMITHSON
ILLINOIS POWER COMPANY
CLINTON POWER STATION
P. O. BOX 678

CLINTON, ILLINOIS 61727, U.S.A.

ABSTRACT

A successful treatment was developed to prevent condenser fouling by Corbicula at the Baldwin
Power Station. Initially, sodium-meta-bisulfite (Na2S205) was utilized to create anoxic conditions in
the intake basins during scheduled outages. Results were favorable, however, Na2S2C>5 may not have
been solely responsible for Corbicula mortalities since mortality rates could not be replicated in con-
trolled laboratory experiments. It was hypothesized that hydrogen sulfide formed during anoxic con-
ditions in the intake basin may have contributed to Corbicula mortalities. Treatments which utilized
Na2S2C>5 and H2S gas proved to be a successful means of Corbicula control. Aeration of the intake
basin effectively neutralized the treatment chemicals prior to discharge. A permanent chemical distribu-
tion system was installed in the intake basins and annual treatment costs were less than $1,000 per
unit. Condenser fouling was eliminated since the annual treatments prevented clams from growing
larger than the inside diameter of the condenser tubes. Growth and settling rates, and shell length
frequencies of Corbicula were examined during the development and evaluations of the treatments.
Observations on the distribution of Corbicula in the intake basins have some implications on future
designs and operations of intake structure which may reduce Corbicula accumulations.

Asiatic clams [Corbicula fluminea (Muller)] began caus-
ing problems for electric utilities in the 1950’s (Goss et al.,
1979). In the next two decades the number of power plants
plagued by Corbicula increased as Corbicula became more
widely distributed. The small size (200 /*m) and abundance
of the larvae make it difficult to screen Corbicula from the
intakes of power plants or other large water users. Larvae
are capable of growing to 25 mm shell size within a year
(Dreier and T ranquilli, 1981), and clams of this size can cause
blockages in heat exchangers and small pipes. Controls are
further complicated by the clam’s resistance to biocides which
control other fouling organisms.

Corbicula control programs may differ depending upon
the type water system affected and which life stage of Cor-
bicula causes problems. Treatments of affected water
systems will vary depending upon the volume and velocity
of water, pipe size, and possible regulations upon the final
discharge of treated water. The differences in physical size
and chemical tolerances of various life stages of Corbicula
must also be considered in developing control programs. It
is unlikely that a single control program would be effective
for all water systems and all life stages of Corbicula.

A treatment to control condenser fouling problems
resulting from an accumulation of Corbicula within the intake
basin was developed by Illinois Power Company at the

Baldwin Power Station. This treatment may be effective in
other power plants with similar condenser cooling systems
and problems. The treatment methodology was developed
through a series of experiments conducted in the intake basin.
This paper describes the stages in development of the treat-
ment and observations on distribution, growth, and settling
rates of Corbicula.

DESCRIPTION OF CORBICULA PROBLEM
The Baldwin Power Station is located in southwestern
Illinois and is owned by Illinois Power Company. The station
consists of three 600 MW coal-fired generating units. Cool-
ing water is provided by a 810 hectare perched pond which
receives make-up water from the Kaskaskia River. Each
generating unit has a separate intake basin with 9.5 mm mesh
traveling screens. Pumps for the condenser cooling water,
service water and fire protection systems are located in the
intake basin. In 1975 the severity of condenser fouling by Cor-
bicula increased in all three units. The condenser problems
resulted from chronic levels of tube fouling as well as short
episodes of severe fouling (Smithson, 1981). The source of
Corbicula was identified as the intake basin where shells had
accumulated to a depth of one meter.

STRATEGY FOR CONTROL TREATMENTS
The 9.5 mm mesh traveling screens should have

American Malacological Bulletin, Special Edition No. 2(1986):63-67

CORBICULA SYMPOSIUM

prevented larger clams from entering the itake basin, so the
accumulation was assumed to have resulted from larvae or
juveniles settling and growing in the intake basin. The large
volume of water entering the intake basin and the small size
of the larvae precluded continual chemical treatment or
screening. Since entry of larvae could not be prevented, em-
phasis was directed toward preventing them from growing
large enough to plug the 7/8” condenser tubes.

Each generating unit and corresponding intake basin
were taken out of service at least once each year for sched-
uled maintenance. This period offered an opportunity to treat
the intake basin under static water conditions since no pumps
were in operation. A treatment applied during the scheduled
outages would be effective if it killed recently recruited clams
before shells reached a size capable of fouling condenser
tubes. The treatment had to be capable of being neutralized
prior to being discharged to the reservoir.

Fast (1971) reported that Corbicula were restricted to
shallow depths in a California reservoir by hypolimnetic
stagnation. Laboratory experiments by Paparo (1976, per-
sonal communication to W. S. Brenneman, Illinois Power
Company) also suggested Corbicula was susceptible to ox-
ygen depletion. Thus, creation of anoxic conditions in the in-
take basin during scheduled outages appeared to offer an
effective treatment. Sodium-meta-bisulfite (^28205) was the
oxygen scavenger selected to create anoxic conditions. This
chemical is non-toxic and used in preservation of human food.
The anoxic effects are neutralized by aeration.

DEVELOPMENT OF THE TREATMENT

The recommended treatment evolved from a series of
treatments which extended over four years. This section
discusses some of the events and observations which lead
to the recommended treatment.

Six months prior to the first treatment divers removed
most of the Corbicula from the intake basins with a large
submersible trash pump. The first experiment with the
Na2S2C>5 treatment was made during a scheduled outage
in October 1976. SCUBA was used to inspect the intake basin
for the distribution and abundance of Corbicula prior to this
and all following treatments. The pretreatment inspection
revealed densities up to 1600/m2 in low velocity areas. A
saturated solution of water and sodium-meta-bisulfite
(Na2S205) was siphoned to the bottom of the intake basin
and distributed by a diver. A cage containing 200 adult Cor-
bicula was used to assess the treatment effectiveness.
Dissolved oxygen (DO) was depleted from the bottom one
meter of the intake basin within minutes of the treatment.
Twenty-four hours later the DO was still depleted and a post-
treatment inspection was made. Most of the clams were alive
but several appeared stressed and had tightly closed their
shells on their foot. A similar response was observed by
Anderson et al. (1976) when Corbicula was exposed to
potassium at concentrations greater than 120 mg/I. At the
end of the week all the caged Corbicula were dead.

Based upon the success of the first treatment, another
unit was treated during an outage in December 1976. The

treatment was identical to the first, but the Na2S205 did not
deplete the DO. Additional Na2S205 was added, but DO was
still in excess of 4 mg/I. This treatment was probably unsuc-
cessful because the colder water contained higher levels of
DO and lowered the metabolic rates of Corbicula.

Variable effectiveness in the next of treatments promp-
ted laboratory duplication of the treatments. In the first in-
take basin treatment, enough Na2S20s (87 mg/I) had been
added to theoretically deplete oxygen from the entire water
volume. This treatment was successful but similar concen-
trations of Na2S205 in one liter beakers did not cause any
mortalities. However, there was a major difference in the ac-
tual concentration Corbicula were exposed to in the intake
basin as compared to the well-mixed laboratory beaker. In
the intake basin the ^28265 mixture was siphoned to the
bottom and its greater density kept it near the bottom. Dis-
solved oxygen profiles indicated the effects were limited to
the bottom one meter. If all the Na2S205 remained in the bot-
tom one meter, the concentration would have been approx-
imately 900 mg/I, rather than 87 mg/I which would have
resulted from complete mixing.

Concentrations of Na2S20s from 1 ,650 to 3,330 mg/I
produced high mortalities of Corbicula in laboratory
treatments. This suggested that the toxic effect of high con-
centrations of Na2S20s, or the various oxides of sulfur
formed from it, may have contributed to mortalities as much
as oxygen depletion alone. This was supported by another
laboratory treatment in which Corbicula survived DO levels
of 0.4 mg/I or less for five days. The intake basins were treated
with 180 to 400 kg of Na2S20s which would have resulted
in a concentration of 900 to 2000 mg/I if all the Na2S20s re-
mained in the bottom one meter. At these concentrations,
the combination of oxygen depletion and ^28205, or the
sulfur oxides formed from it, were fatal to the clams.

A hydrogen sulfide (H2S) smell was noted during the
inspections following several successful treatments. The
anoxic layer at the bottom of the intake basin created condi-
tions where the H2S could occur. It was not determined if
the H2S resulted from the sulfur bacteria acting upon the ex-
cess sulfates or if it was formed directly from the N 828205.
Since H2S is highly toxic to most life forms, adding more H2S
to the anoxic layer offered a potential for increasing the treat-
ment effectiveness. The H2S could also be easily neutral-
ized by aeration prior to discharging the treatment water.

Hydrogen sulfide was added in the next experimental
treatment after the anoxic zone had been created by the
Na2S205. A gas diffuser was placed in the siphon hose and
the flow of additional Na2S205 treated water dispersed the
H2S. This treatment resulted in over 95% mortality within 48
hours. The addition of H2S to later experiments increased
the effectiveness in cooler water and reduced the time the
intake basin had to remain undisturbed in warmer weather.

RECOMMENDED TREATMENT

The following recommended treatment resulted from
the four years of experimental treatments. The treatment uses
1 58 to 21 0 g/m3 of ^28205 and 2.4 to 3.7 g/m3 of H2S. The

SMITHSON: CORBICULA CONTROL TREATMENT

effectiveness of Na2S2Qs as an oxygen scavenger can be

increased by adding 0.26 g/m3 of cobait chloride as a catalyst.
After one half of the ^28265 has been injected into the bot-
tom of the intake basin, H2S is mixed with the remaining
amount. Water in the intake basin should remain undistrub-
ed for 60 to 72 hours after the treatment. If the temperature
is below 21°C, then the larger amounts of N 328265 and H2S
should be used, and the intake basin should stand undisturb-
ed for an additional 36 hours.

Plastic pipes were permanently installed on the floor
of the intake basin to distribute the chemicals and provide
aeration to neutralize the treatment prior to discharge. The
annual cost to treat each unit was less than $1 ,000. A more
complete description of the treatment methodology is pro-
vided by Smithson (1981).

EVALUATIONS OF TREATMENTS

Treatments effectiveness was evaluated from samples
of clams collected in the intake basin and by comparing con-
denser fouling problems of treated and untreated units. The
percentage of live clams in pretreatment samples were com-
pared with recently killed clams from posttreatment samples
and provided an immediate indication of treatment effec-
tiveness. The size of live clams collected prior to a treatment
indicated the success of the previous treatment and the
growth of newly recruited clams. If the previous treatment was
successful, live Corbicuia in the intake basin should all be
newly recruited and relatively small. An abundance of larger
shells would indicate that the previous treatment was inef-
fective or the period between treatments was too long and

Fig. 1 . Percent abundance of Corbicuia by shell lengths from a treated
and untreated cribhouse.

Fig. 2. Typical distribution of Corbicuia and silt on floor of cribhouse.

recruited clams had grown large enough to plug condenser
tubes. Comparisons of shell sizes from treated and untreated
intake basins indicated that shells from the treated intakes
were not large enough to plug main condenser tubes, but
in untreated intakes 78% could plug condenser tubes (Fig. 1).

The pretreatment inspections also provided an oppor-
tunity to assess the abundance of Corbicuia in the intake
basin. Prior to any treatments, Corbicuia had accumulated
to a depth of one meter, but after several years of treatments
the accumulations were usually less than 7 cm and were com-
posed mostly of relic shells.

Condenser fouling problems were reduced in treated
units. The unit with the longest history of treatments has had
only one minor problem of condenser fouling since treatments
began in 1976. An untreated unit experienced episodes of
severe condenser pluggage each summer until treatments
were started.

DISTRIBUTIONS WITHIN THE INTAKE BASIN

Distribution of Corbicuia in intake basins was noted
during sample collections and has some implications on
design and operating practices of intake basins which could
reduce problems with Corbicuia. The distribution of of Cor-
bicuia and silt was not uniform, but was strongly associated
with low velocity areas (Fig. 2). Clam abundance in the
intake basin appeared to result from larvae or juvenile clams
settling in low velocity areas and growing. When these clams
reached maturity they released additional larvae and ac-
cumulation was compounded. The increased abundance of
larvae in the intake basin also presented a greater threat to
service water and fire protection systems which draw water
from the intake basin. Attachment by byssal threads was
never observed in the intake basin, circulating water piping,
or in condensers.

The uneven distribution of Corbicuia illustrates the im-

CORBICULA SYMPOSIUM

portance of eliminating low velocity areas in designing new
intake structures. Rounding the corners of intake basins and
creating a 45° slope where the walls meet the floor might
reduce the areas where Corbicula accumulate. Installation
of high pressure water nozzles to periodically flush areas
where silt and clams accumulate could be retrofitted into ex-
isting intake basins or incorporated in new disigns. Water
flushed through the chemical distribution lines (Smithson,
1981) reduced the silt accumulation in the intake basin.

Episodes of condenser fouling may result from
changes in Corbicula distribution which subject them to pump
entrainment. McMahon (1979) cited an example where Cor-
bicula within an intake basin may have moved out of the
sediments in response to environmental stress and were
pumped into the condensers. An untreated intake basin at
the Baldwin Power Station experienced severe episodes of
condenser pluggage by Corbicula in early summer each year.
The episodes generally occurred when water temperature first
exceeded 32°C that year. These higher temperatures may
have stimulated Corbicula to seek cooler areas and their
movement exposed them to higher velocities areas where
they were entrained by circulating water pumps.

Episodes of severe condenser fouling could also result
from changes in the number of circulating water pumps be-
ing used. Intake basins usually contain several circulating
water pumps. The number of pumps in operation may vary
with inlet water temperature and the number of generating
units in operation. When a pump is taken out-of-service the
velocity in that area is reduced and may provide an area
where Corbicula can accumulate. If these pumps are out-of-
service for an extended period, they may entrain the ac-
cumulated Corbicula when restarted. This problem also ap-
plies to any infrequently used pump located in an area where
Corbicula settles and accumulates.

SETTLING, GROWTH AND PREDATION

The treatment strategy assumed that Corbicula or
juvenile clams were entering, settling, and growing in the in-
take basin and annual treatments would kill clams before they
grew large enough to plug condenser tubes. In conjunction
with the development of a treatment, a study was conducted
to assess the growth and settling rates of larvae and juveniles
and the effects of predation on small Corbicula.

Trays (38 x 34 x 15 cm) filled with a mud-gravel
substrate were placed at the bottom of the intake canal. Trays
were covered with a wire screen (12 mm aperture) which form-
ed an inverted V-shaped roof over the tray. The wire screen
simulated the intake screens which early life stages pass
through to settle in the intake basin. Two of the four trays
were completely enclosed by the wire mesh, and two trays
had the mesh removed from the ends to allow predator ac-
cess. The trays were 15 cm above the canal bottom to pre-
vent entry by larger clams and were in the canal from April
until November.

The mean density of Corbicula in November was
270/m2 in the completely covered trays and 50/m2 in open-
ended trays. The lower density in the open-ended trays may
have been due to fish predation on the small clams. Minckley

et al (1970) found buffalo fishes consumed large numbers
of Corbicula. Sule et al. (1 981 ) found Corbicula were regular-
ly consumed by bluegill (Lepomis machrochirus), freshwater
drum (Apiodinotus grunniens), and channel catfish (Ictalurus
punctatus). These species are abundant in Baldwin Cooling
Pond and may have accounted for the lower densities in the
incompletely covered trays. Predator absence in intake basins
may contribute in part to the higher densities of Corbicula
observed there.

The height of the sides of the tray above the lake bot-
tom and the 12 mm mesh should have precluded the entry
of larger clams; thus; Corbicula in the trays must have been
recruited as larvae and grew. Britton et al (1979) found con-
siderable recruitment of juvenile Corbicula through 10 mm
opening in the lids of 19 liter buckets. The size range of the
Corbicula in the trays also indicated they had settled as lar-
bae and grew. The shell lengths in November ranged from
5 to 22 mm with a mean length of 16.8 mm. The range and
mean shell length were similar to those reported for caged
young-of-year Corbicula from the intake canal of nearby Lake
Sangchris (Dreier and Tranuilli, 1981). The shell lengths of
the Lake Sangchris clams ranged from 9 to 25 mm with a
mean length of 18.1 mm.

The growth rate of larvae in the trays was useful in
determining the timing of treatments of prevent Corbicula from
reaching a size which wold plug condenser tubes. The
maximum size obtained by November was 22 mm, and bas-
ed upon this growth rate, an annual fall treatment should pre-
vent incoming larvae from reaching a problem size. Outages
of fossil-fueled generating units in power plants are typically
scheduled during the spring and fall. Fall treatments are the
most effective since they occur after the reproductive season.
Recruitment of Corbicula to intake basins should remain
relatively low until larvae are released during the following
spring.

SUMMARY

A variety of measures may be needed to control Cor-
bicula fouling in the various water systems in power plants.
The control treatment described in this paper was successful
in controlling condenser fouling problems at the Baldwin
Power Station and has been patented by Illinois Power Com-
pany. The treatment consisted of the addition of an oxygen
scavenger (Na2S2C>5) followed by an injection of H2S into the
intake basin during the annual scheduled outage of the
generating unit.

Corbicula entered the intake basin through the travel-
ing screens as larvae and settled in low velocity areas where
they accumulated and grew. Annual fall treatments were suf-
ficient to prevent the incoming larvae from becoming large
enough to plug 7/8” condenser tubes between treatments.

ACKNOWLEDGEMENTS

I would like to thank Thomas V. Clevenger for his helpful sug-
gestions on the manuscript.

LITERATURE CITED

Aldridge, D. W. and R. F. McMahon. 1978. Growth, fecundity, and

SMITHSON: CORBICULA CONTROL TREATMENT

bioenergetics in a natural population of the Asiatic freshwater
clam, Corbicula manilensis, Philippi from north central Texas.
Journal of Molluscan Studies 44:49-70.

Anderson, K. B., C. M. Thompson, R. E. Sparks and A. A. Paparo.
1976. Effects of potassium on adult Asiatic clams, Corbicula
manilensis. Biological Notes No. 98, Illinois Natural History
Survey.

Britton, J. C., D. R. Coldiron, L. P. Evans, Jr., C. Golightly, K. D.
O’Kane, and J. R.TenEyck. 1979. Reevaluation of the growth
pattern in Corbicula fluminea (Muller). IN: J. C. Britton, ed., First
International Corbicula Symposium Proceedings. Texas Chris-
tian University, Fort Worth, Texas, pp. 177-192.

Dreier, H. and J. A. Tranquilli. 1981. Reproduction, growth, distribu-
tion, and abundance of Corbicula in an Illinois cooling lake.
Illinois Natural History Survey Bulletin 32(4):378-392
Fast, A. W. 1971. The invasion and distribution of the Asiatic clam
(■ Corbicula manilensis) in a southern California reservior.
Bulletin of Southern California Academy of Science 70(2):91-98.
Goss, L. B., J. M. Jackson, H. B. Flora, B. G. Isom, C. Gooch, S.

A. Murray, C. G. Burton, and W. S. Bain. 1979. Control studies
on Corbicula for steam-electric generating plants. IN: J. C. Brit-
ton, ed., First International Corbicula Symposium Proceedings.
Texas Christian University, Fort Worth, Texas, 140-151.

McMahon, R. F. 1979. Response to temperature and hypoxia in the
oxygen consuption of the introduced Asiatic freshwater clam
Corbicula fluminea (Muller). Comparative Biochemistry and
Physiology. 63(A): 383-388.

Minckley, W. L., J. E. Johnson, J. N. Rinne, and S. E. Willoughby.
1970. Foods of buffalo fishes, genus Ictiobus, in central
Arizona reservoirs. Transactions of the American Fisheries
Society 99:333-342.

Smithson, J. A. 1981. Control and treatment of Asiatic clams in power
plant intakes. Proceedings of the American Power Conference
43:1146-1151.

Sule, M. J., J. M. McNurney, and D. R. Halffield, Jr. 1981 . Food habits
of some common fishes from heated and unheated areas of
Lake Sangchris. Illinois Natural History Survey Bulletin
32(4):500-519.

CORBICULA FOULING AND CONTROL MEASURES
AT THE CELCO PLANT, VIRGINIA

DONALD S. CHERRY1, ROB L. ROY2, RICHARD A. LECHLEITNER1,

PATRICIA A. DUNHARDT1, GREGORY T. PETERS1 AND JOHN CAIRNS, Jr.1

UNIVERSITY CENTER FOR ENVIRONMENTAL STUDIES AND BIOLOGY DEPARTMENT
VIRGINIA POLYTECHNIC INSTITUTE AND STATE UNIVERSITY
BLACKSBURG, VIRGINIA 24061, U.S.A.

AND

2CELCO PLANT, CELANESE FIBERS CORPORATION
NARROWS, VIRGINIA 24124, U.S.A.

ABSTRACT

Corbicula fluminea infestation was initialiy reported in May 1981 and increased substantially
by the fall at the Celco Plant, Celanese Corporation, Narrows, Virginia, as clam numbers at the river
pumphouse station increased from 2,529/m2 on 7/13/81 to 269,105/m2 by 9/29/81. Infestation in the
plant increased throughout 1982 as clam numbers at the pumphouse increased from 2,465 to 23,

869/m2 from 4/13/82 to 1 1/29/82. Spring spawning was documented on 6/8/82 and 5/25/83 at river
temperatures of 17-22 and 14-17 C, respectively. Spawning was continuous throughout the summer
and fall of 1982 and 1983 although a major fall spawn was observed at river temperatures of 8-22
C (10/11 to 11/29/82) and 16-24 C (9/29/83). In-plant continuous chlorination (^ 0.50 mg/L) for four
weeks at the major spring and fall spawning periods was implemented in 1 982 and continued in 1 983
to reduce the incidents of clogged air conditioning condensers, oil cooling heat exchangers, industrial
condensing-recovery complexes and power generating units. Other anti-fouling efforts used included
the removal of clam-laden sediment at the pumphouse and periodic flushing of water holding towers
and dead pipe spaces within the plant. Benthic macroinvertebrate collections were made in the New
River upstream, adjacent to and downstream of the plant before continuous chlorination was im-
plemented and during the two years of its use. No discernible impact of continuous chlorination for
clam control was observed on the invertebrate populations sampled in the river. Literature data sug-
gested that fish populations in the New River area influenced by chlorination would avoid these
discharges. Continuous chlorination had a positive effect in reducing the number of clam fouling in-
cidents within the plant in 1 983 since only six of the seven incidents reported were due to clam shells
and not live clams.

The incidence of the Asiatic clam [Corbicula fluminea
(Muller)] as a biofouling pest in power plant and other in-
dustrial installations has increased alarmingly in the
Southeastern United States during the past decade. After Cor-
bicula was first discovered in 1 938 in the Columbia River near
Knapton, Oregon (Burch, 1944), the Asiatic clam was found
in the Ohio River near Paducah, Kentucky in 1957 (Sinclair
and Isom, 1963). By 1961, it was found in the Upper Ohio
and Kanawha River Drainage at Chelyan, West Virginia
(Thomas and MacKenthun, 1964). A subsequent collection
was made in 1 973 by Joy and McCoy (1 975) in the Kanawha
River, and Rodgers et al. (1977) calculated an upstream in-
vasion of Corbicula in the New River to Glen Lyn, Virginia
at -v 15 km/year. Corbicula and other mollusk fouling occur-
rences in the Southeastern United States were initially

reported and reviewed by Sinclair and Isom, 1963; Sinclair,
1964 and 1971; Isom, 1971; McMahon, 1977, 1982; Cherry
etal., 1980. Corbicula were reported in the Savannah River,
South Carolina by 1 972 (Fuller and Powell, 1 973) and resulted
in fouling disturbances thereafter in power generating facilities
at the Savannah River Project (R. S. Harvey, personal
communication).

The initial documentation of Corbicula in the New River
was reported at the Glen Lyn Plant in October, 1976, with
densities in uninfluenced and thermally influenced areas of
20-30 clams/m2 of river substratum (Rodgers et al., 1977).
Since then, proliferation of Corbicula densities had reached
a high of 11,522/m2 in the thermally influenced area of the
Glen Lyn Plant in February, 1978 (Cherry et at., 1980) Cor-
bicula fouling has been controlled by routine physical

American Malacological Bulletin, Special Edition No. 2(1986):69-81

CORBICULA SYMPOSIUM

maintenance of condenser systems. The diameter of the con-
denser tubes (-VI9 mm) allows most adult clams to pass
through unimpeded. In addition, the Glen Lyn Plant does not
warm the intake of the travelling screens at the pumphouse
for deicing purposes which inhibits successful overwintering
of clams at this location. Consequently, Corbicula population
densities do not reach unusually high numbers in non-
thermally influenced areas (e.g. at the pumphouse) during
the winter when New River temperatures approach 0 C.

The Celco Plant, located 'vl 1 km upstream from the
Glen Lyn Plant, was probably initially inundated with Corbicula
from the massive, 1978 spawn resulting in fouling incidents
in 1 981 . From shell sizes up to 31 mm, clams were calculated
to be 'vS years old in condenser systems fouled with Corbicula
by spring 1981. After the initial fouling occurrence in May
1981 in raw water-fed air conditioning units at the Celco Plant,
numerous (e.g., two to three per month) fouling occurrences

of Corbicula had occurred in plant processing condenser
systems (i.e. power production turbines, oil and air coolers,
acetone recovery, calcium chloride refrigeration, and produc-
tion of distilled water). As a result of these problems, a study
was designed to evaluate the clam populations in and around
the Celco Plant beginning in May 1981 and ending in
November 1983.

The objectives of this study were to: (1 ) measure Cor-
bicula numbers and gravid condition of adults at the pum-
phouse and the thermal discharge stations in the New River
Celco Plant; (2) evaluate the removal of river sediment at the
plant pumphouse as a clam control procedure; (3) determine
the effect of continuous chlorination upon Corbicula within
the plant for a four-week period during peak spawning periods
in spring and fall; (4) identify clam infestation areas within
the plant and provide maintenance for clam control; and (5)
evaluate potential environmental effects of chlorinated

Fig. 1. Sampling stations for Corbicula at the pumphouse and thermal discharge stations of the Celco Plant and chlorine sampling stations
(denoted by *) within the plant. Numbers within each building represent the identification of each production unit while thick lines to and
from each building are raw river pipe lines. Invertebrate sampling stations in the New River are designated as 1-4.

CHERRY ETAL: CORBICULA FOULING AND CONTROL

discharges in the New River from a field study of benthic,
macro-invertebrates and literature review of fish.

MATERIALS AND METHODS

PLANT SITE

The Celco Plant is located along the New River at Nar-
rows, Virginia. New River water is used to: (1) air condition
offices, process control rooms and some production areas;
(2) recover acid and acetone using river water for condenser
cooling; (3) provide fire control; and (4) cool turbine generators
in power production, calcium chloride chillers, and stills for
distilled water. River water is pumped from the pumphouse
through 0.92 and 0.61 -m lines into the plant by six centrifugal
pumps at 188,000 L/min. The chlorination system is approx-
imately 30 m from the pumphouse, and chlorinated water
flows throughout the plant (Fig. 1).

CORBICULA SAMPLING

Asiatic clams were sampled by Ponar dredge at five
locations: 150 m upstream from the plant; at the pumphouse;
at the thermal discharge station (120 m downstream from the
pumphouse); and 75 and 150 m downstream from the
discharge on June 24, 1981 . Thereafter, clam sampling was
emphasized at the pumphouse and thermal discharge sta-
tions from July 13, 1981 through November 29, 1983 once
each month or at two to three-month intervals during the
spring to fall (Fig. 1). At these stations, Corbicula samples
were taken four times during 1981 and five times each dur-
ing 1982 and 1983. Clam samples were taken ^5 m away
from the pumphouse from one end to the other. Clams at the
discharge station were collected by wading and removing
0.25 m2 areas of sediment. All samples were reported as per
m2 of clam numbers.

GRAVID CONDITION OF CORBICULA

Fifty adult clams (^ 10 mm) were evaluated
microscopically using a dissecting microscope for develop-
ing larvae in the gill marsupia during each field sampling ef-
fort. Data were evaluated as the percent of clams sampled
with incubating larvae in the marsupia of gills.

REMOVAL OF RIVER SEDIMENT (MUCKING PROCESS)

In mid-June, 1982, much of the clam-infested
sediments at the pumphouse of the Celco Plant was siphon-
ed or removed by a suction (bilge-“mucking”) pump and emp-
tied into a dumpster in order to reduce the accumulated sedi-
ment and clam population at the pumphouse. “Pre-mucking”
data for clam densities at the pumphouse refer to river sedi-
ment conditions prior to sediment removal while “post-
mucking” data indicate clam density measurements after this
operation.

MONITORING OF CHLORINE CONCENTRATIONS

Several stations in the Celco Plant were monitored for

chlorine residuals (Fig. 1). These included: Buildings 10 and
12 (375 and 275 m away from the chlorinator, respectively),
followed by Buildings 1 and 32 (590 and 460 m, respective-
ly), and the Office-Computer Complex Building which was far-
thest away (790 m). During continuous four-week chlorina-
tion periods, chlorine residuals were measured by
amperometric titration at each station. Chlorine measurements
were taken daily during the continuous, four-week chlorina-
tion periods at the discharge channel. The New River water
depth and the volume of effluent water released from the plant
in the discharge station were measured daily due to a special
NPDES (National Pollutant Discharge Elimination System)
variance permit obtained from the State Water Control Board
in Richmond, Virginia. The following dilution factors were in-
corporated into calculations of the daily chlorinated water
discharge which resulted in an overall calculated chlorine
residual of ^ 0.01 mg/L TRC (total residual chlorine) into the
New River:

Plant Outfall Flow (MGD) X Measured TRC mg/L in Plant Outfall

Daily New River Flow (MGD as millions of gallons/day
< 0.01 (TRC)

IDENTIFICATION OF CLAM INFESTATION AREAS WITHIN
THE PLANT

Dead water pipe lines were identified within the plant
that had become obsolete and were removed. Water towers
used for emergency fire control or production were drained
and sediment was analyzed for potential juvenile and adult
clam habitation each spring and fall during 1982 and 1983.

BENTHIC MACRO-INVERTEBRATE SAMPLING IN
THE NEW RIVER

Environmental effects of chlorine were determined by
establishing sampling stations in similar riffle habitats at the
following locations using a 0.092 m2 Surber Sampler (Fig. 1):

Station 1 : above the Celco Plant ^0.5 km above the U.S.
460 bridge or 2 km above Station 2;

Station 2: just below ('vO.I km) the waste water effluent
of the Celco Plant;

Station 3: just below the fly ash effluent of the Celco Plant,
^0.5 km below Station 2;

Station 4: %1 .0 km below Station 3.

On each of the ten sampling dates (September and
October 1981, and May 1982 before the first four-week con-
tinuous chlorination; and on August, September, October
1982, and four times in 1983 - after four, 28-day continuous
chlorination treatments for Corbicula control), three
macroinvertebrate samples were collected at each of the
above stations. River water depth was less than 0.3 m dur-
ing each sampling effort. Samples were immediately preserv-
ed in 70% ethanol and taken to the laboratory for identifica-
tion. Organisms were identified to genus level, counted and
recorded by sample replicate, and later evaluated in terms of
taxon diversity (Shannon-Wiener) per sample and station, per-

CORBICULA SYMPOSIUM

cent similarity between stations, and major taxonomic groups
by station.

REVIEW OF LITERATURE ON CHLORINE TOXICITY AND
AVOIDANCE RESPONSES OF FISH

To evaluate the potential impact of continuous
chlorination discharges upon the fish community in the New
River, daily TRC concentrations from the thermal discharge
station were compared to chlorine toxicity data in the literature
for known New River fish species. Forty-eight fish species
have been identified and evaluated in the New River relative
to thermal discharges at the Glen Lyn Plant (Stauffer et a/.,
1976). Potential fish avoidance and toxicological conse-
quences of chlorine from studies at the Glen Lyn Plant (Cherry
et at., 1977; 1982; Cherry and Cairns, 1982; Giattina et al.,
1981) were compared to other data (e.g., Heath, 1977;
Seegert et al., 1979).

RESULTS

PRE-MUCKING OPERATION-1981 TO SPRING, 1982
PRELIMINARY SAMPLING

After the initial incident of Corbicula fouling at the Celco
Plant on May 28, 1981, preliminary sampling showed clam

numbers of 21 5/m2 at 1 50 m upstream of the plant, 2,529/m2
at the pumphouse, 272/m2 at the thermal discharge station,
1 ,204/m2 75m downstream of the thermal discharge, and 1/m2
1 50 m downstream of the thermal discharge. Temperatures
at these stations were 25, 26, 38, 29, 29 C, respectively.
Oldest clams (->31 mm) were considered to be recruits from
the Glen Lyn Power Plant in 1978; no clams were found to
be less than 9 mm in shell length. Clam densities were highest
at the pumphouse and at 75 m downstream from the ther-
mal discharge from the plant.

CORBICULA NUMBERS AND GRAVID CONDITION AT THE
PUMPHOUSE

Numbers (per m2) at the pumphouse varied from 2,529
in July to 9,742 in August to a high of 269,1 05 in September,
and then declined to 41,788 in November, 1981 (Table 1).
The high densities in August and September, 1981 were due
primarily to the large number of young clams ^ 1 mm in shell
length (84 and 78% of the samples, respectively). The per-
cent of adult clams with eggs in the gonadal tissue was high
for both months (96 and 76%, respectively) while the percent
with incubating larvae in the marsupia was low (14%). No
young clams ^ 1 mm in shell length were found in the other
three sampling efforts. The relatively high number of clams

Table 1. Number (per m2) of Corbicula fluminea in juvenile and adult clam shell length classes (early juveniles S 1 mm, juveniles to early
adults > 1 to S 10 mm, and adults >10 mm) at the pumphouse station of the Celco Plant, Narrows, Virginia. The occurrence of early juvenile
clams sampled from the river sediment are compared to New River temperature (mean and ranges in C) of two weeks at and prior to each
of the sampling efforts along with the gravid condition (larvae in gills, n = 50) of adult clams.

Water Temperatu

re (C)

Shell Length of clams in

% Adults

Sampling

Date

Mean

Range

> 1 to =10

>10 to 31

Total

w/ larvae
in Gills

7/13/81

23.5

21-27

Pre-Mucking Period3
0

184

2,345

2,529

8/5/81

25.2

24-27b

8,242

321

1,179

9,742

9/29/81

18.7

1 7-22c

207,030

59,374

2,701

269,105

11/11/81

12.6

10-17

40,248

1,540

41 ,788

4/13/82

9.3

6-12

2,408

2,465

6/8/82

19.9

17-22

545

6,680

215

7,440

7/19/82

24.3

23-25b

Post-Mucking Period3
7,479

1,800

570

9,849

10/11/82

20.0

1 8-22c

13,741

1,344

1,375

16,460

11/29/82

9.9

8-1 2C

21,993

640

1,236

23,869

5/25/83

16.5

14-17

127

1,401

684

2,212

6/14/83

20.0

1 9-23b

3,486

1,597

2,250

7,333

8/10/83

26.0

24-27

2,326

1,046

3,397

9/29/83

19.4

1 6-24c

7,225

3,581

830

1 1 ,636

11/29/83

9.0

7-11

2,649

2,339

5,058

aPre-Mucking period occurred before New River sediment was removed by a bilge-pump mucking process at the pumphouse station while
post-mucking occurred after removal of a substantial amount of the sediments.

indicates the mean and range (previous two weeks prior to each sample) of New River daily temperatures during the major spring spawning
season.

indicates the mean and range (previous two weeks prior to each sample) of New River daily temperatures during the major fall spawning
season.

CHERRY ETAL: CORBICULA FOULING AND CONTROL

Table 2. Number (per m2) of Corbicula fluminea in juvenile and adult clam shell length classes (early juveniles = 1 mm, juveniles to early adults
> 1 to = 10 mm, and adults > 10 mm) at the thermal discharge station of the Celco Plant, Narrows, Virginia. The occurrence of early juvenile
clams sampled from the river sediment is compared to effluent temperatures (mean and ranges in C) of two weeks at and prior to each sampl-
ing effort along with the gravid condition (larvae in gills, n = 2-50) of adult clams.

Sampling

Date

Water Temperature (C)

Shell Length of Clams in

Total

% Adults
w / larvae
in Gills

Mean

Range

>1 to Sio

>10 to 31

Pre-Mucking Period3

7/13/81

36.6

31-40

177

272

8/5/81

36.7

31-40

608

659

9/29/81

30.9

31-28a

2,199

1,255

1,096

4,530

11/11/81

29.2

25-33

1,217

317

1,584

4/13/82

28.0

22-30

520

418

938

6/8/82°

—

Post-Mucking Period

7/19/82

38.5

37-41 b

976

741

469

2,186

10/11/82

33.2

28-39

387

424

811

1 1/29/82c

—

5/25/83c

—

6/14/83

34.0

32-37a

317

285

652

8/10/83

38.6

36-41

9/29/83

31.7

31-34

11/23/83

32.0

29-34

indicates the mean and range (previous two weeks prior to each sample) of thermal discharge temperatures during the major spring
spawning week.

indicates the mean and range of thermal discharge temperatures during the major fall spawning season.
cHigh river water levels prevented sampling of clams.

in November was attributed to the clams spawned in
September. Clams from 1-10 mm in shell length in November
were 40,248/m2 which represented 96% of the total. The
gravid condition of clams at the pumphouse station was low
in the November sample (15 and 0% of adults with eggs in
gonadal tissue or larvae in marsupia, respectively). Mean
water temperatures (two weeks prior to and during sampl-
ing) at peak spawning collections in 1981 were 25.2 and 18.7
C, respectively.

On April 13, 1982, clam numbers at the pumphouse
station were 2,465/m2 and increased to 7,440/m2 by June 6,
1982 (Table 1). Clams S 1 mm in length were not found in
the April sample but represented 7.4% of the sample in June.
The gravid condition of adult clams (eggs and larvae, respec-
tively) sampled in 1982 ranged from 75 and 0% in April to
100 and 50% in June. The release of juvenile clams as deter-
mined in the sediment, was initally identified when water
temperature reached 17-22 C (x = 19.9 C).

CORBICULA NUMBERS AND GRAVID CONDITION IN THE
THERMAL DISCHARGE

Numbers (per m2) in the thermal discharge station in-
creased from 272 in July to a high of 4,530 in September and
declined to 1,584 in November, 1981 (Table 2). Spawning
resulted in the collection of clams S 1 mm in length in
September, 1981, when 2,199/m2 (48.5% of the total sam-

ple) was obtained. This was the only time in 1 981 when young
clams Is 1 mm in length were sampled as mean, two-week
water temperature was 30.9 C. The gravid condition in the
November sample was low (6% in gonads and 0% in gills).

No spring spawning activity in the thermal effluent was
recorded in June 1982 since high water conditions prevented
adequate sampling (Table 2). Discharge temperatures rang-
ed from 22 to 30 C during the April and June sampling ef-
forts. The gravid condition of clams in the thermal effluent
was low in the April sample (61 and 0% for eggs and larvae,
respectively) showing no potential of larval release from
adults.

POST-MUCKING OPERATION - SUMMER 1982 TO
FALL 1983

CORBICULA NUMBERS AND GRAVID CONDITION AT THE
PUMPHOUSE

After removal of the sediment at the pumphouse and
thermal discharge stations in 1982, Corbicula numbers at the
pumphouse ranged from 9,849 in July to 16,460 in October
to 23,869 in November (Table 1). Early summer spawning
of Corbicula had occurred by July 19, 1982 when 7,479/m2
of - 1 mm in length were sampled in pumphouse sediment.
This amount represented 76% of the total July sample as
newly spawned individuals. River water temperatures
ranged from 23-25 C. The gravid condition of clams in July

CORBICULA SYMPOSIUM

1982 was 90% (eggs) and 30% (larvae) in adults. A major
spawning of Corbicula was observed from October 1 1 through
November 29, 1 982 in which 1 3,741 /m2 and 21 ,993/m2 clams
S 1 mm in size were sampled from sediments. This abun-
dance of clams ^ 1 mm represented 83 to 92% of the overall
sediment sample taken in October and November, 1982,
respectively. The gravid condition (eggs and larvae, respec-
tively) of adult clams ranged from 100 and 30% in October
to 30 and 0% in November, 1982. During fall spawning, river
water temperatures ranged from 18-22 C in October to as
low as 8.0 C in November 1982.

In 1983, clam numbers ranged from a low 2,212/m2
in May to a high of 11,636/m2 in September (Table 1). The
highest number of clams sampled corresponded to the
release of juvenile clams (M mm) in June (3,486/m2) and
September (7,225/m2). The percent of adults with larvae in
the gills was also high (65 to 80%) during these months of
high spawning activity. Water temperature in June ranged
from 19-23 C and in September, 16-24 C. Spawning was
essentially over by November 29 when river temperatures
were 7-1 1 C.

CORBICULA NUMBERS AND GRAVID CONDITION IN THE
THERMAL DISCHARGE

Corbicula numbers (per m2) in the thermal discharge
station ranged from 2,186 in July to 811 in October, 1982
(Table 2). River conditions were too high to sample in
November. Slight spawning in the thermal discharge station
was observed at a temperature of 37-41 C (976/m2 for clams
^ 1 mm in length) during the July sampling effort but could
not be compared to the June sampling effort due to high river
conditions. The gravid condition of Corbicula sampled in July
1982 was 90 and 10% (eggs in gonads and larvae in mar-
supia). In 1983, the number of clams sampled markedly
decreased from June (652/m2) to November (2/m2). Slight
spawning was observed in June when water temperature was
32-37 C.

OBSERVATIONS BETWEEN PRE- AND POST-MUCKING
SAMPLING

Sediment removal process at the Celco Plant pump-
house was approximately 70% effective from patches of
sediments to bare rock substrate observed from the upper
floor of the pumphouse. A four-fold decline in juvenile and
adult clams ( ^ 1-31 mm) was observed immediately after sedi-
ment removal in July 1982 when compared to June 1982
(Table 1). Adult clams (^10 mm), however, rapidly increas-
ed after July 1982 to numbers by June 1983 which were
similar to those obtained in July to September, 1981.

CHLORINE RESIDUALS MEASURED IN THE CELCO
PLANT AND THERMAL DISCHARGE

Biocidal effectiveness was a function of the rate of
chlorine gas application, water temperature, and distance of
sampling stations from the chlorinator (Table 3). Measurement
of TRC was always lowest at the thermal discharge station

Table 3. Summary of mean total residual chlorine (TRC)
measurements in mg/L at selected sampling stations in the Celco
Plant at the chlorinator rate of 200 to 525 lbs per day in the 0.61
and 0.92-m water lines of the plant. Data were taken from several
sampling periods on August 26, 1982 to November 1, 1983. Numbers
in parentheses represent the water temperature measured at each
sampling station in C.

Total Residual Chlorine (TRC) in mg/L at Sampling Stations

Chlorine Thermal
(lbs/day)Discharge

Bldg 12 Bldg 10

Bldg 32

Bldg 1

Office

Bldg

AUGUST 26,

1982

200

0.02

(33)

400

0.06

0.30

(33)

(27)

(24)

(23)

(26)

525

0.12

0.85

0.77d

0.52°

(34.5)

(27)

(24)

(27)

SEPTEMBER 21, 1982

525

0.18

0.90

0.59

0.76

(29)

(22)

(18)

NOVEMBER 1, 1982

250

0.10

0.34

0.37

0.30

0.16

(28.5)

(17.0)

(16.0)

(17.5)

(15.0)

(15.5)

450

0.24

0.65

0.75

0.60

0.66

0.38

(27.5)

(17.5)

(16.0)

(17.5)

(15.0)

(16.0)

500

0.27

(28.0)

NOVEMBER 16, 1982

200

0.02

0.46

0.11

0.26

0.16

0.13

(21)

(13)

(9)

(10)

350

0.23

1.22

0.29

0.60

0.46

0.36

(21)

(13)

(10)

(9)

(10)

300

0.06

(21)

JULY 17, 1983

325

0.10

0.21

0.25

0.20

0.21

0.13

(33)

(27)

(23)

(33)

(23)

500

0.18

0.55

0.36

0.47

0.45

0.27

(33)

(27)

(23)

JULY 22, 1983

325

0.08

0.35

0.34

0.33

0.32

0.25

(37)

(30)

(26)

(25)

500

0.21

0.67

0.60

0.55

0.56

0.36

(37)

(30)

(26)

(25)

NOVEMBER 1, 1983

325

0.06

0.58

0.06

0.30

0.27

0.15

(39)

(18)

(15)

500

0.20

0.12

0.19

0.63

0.61

0.38

(29)

(18)

(15)

aTRC measurements were not taken at this station.
bWater line was not flushed adequately to obtain a TRC measure-
ment.

cChlorinated river water was not passing through the condenser
when sample was taken.

dWater line was flushed for 30-60 min prior to TRC measurement.

CHERRY ETAL: CORBICULA FOULING AND CONTROL

which was farthest from the chlorinator. When TRC was 0
mg/L, problems in river water circulation or lack of condenser
use were the reason for no chlorine detection. As the chlorina-
tion rate was increased from 200 to 525 Ibs/day during this
period, biocidal effectiveness increased. However, a greater
chlorine application rate was needed in August and
September 1982 (maximum rate of 525 Ibs/day) to achieve
the same biocidal effect in November 1982 (350-450 lbs). In
1983, chlorine application varied from 325 to 500 Ibs/day
depending upon the river flow rate. In most sampling stations
within the plant, TRC concentrations were attempted to be
^ 0.50 mg/L. The office building, which was farthest from the
chlorine application point, had the lowest TRC concentration
(^0.38 mg/L).

No violation of the special variance permit occurred
in 1982-1983. As river flow rates declined to less than 1200
MGD (million gallons/day) chlorine application was reduced
to 350 Ibs/day. By measuring the TRC in the thermal
discharge or plant outfall, and applying it to the plant outfall
rate and flow rate of the river, calculated TRC had to be ^ 0.01
mg/L. For example, if the plant outfall was 60 MGD the
measured TRC was 0.18 mg/L when the New River flow was
1200 MGD, then:

60 MGD X 0.18 mg/L TRC

= 0.009 mg/L TRC

1200 MGD

When comparing chlorine residuals at Bldgs 12, 10,
32 and 1 in the August-September 1982 samples with those
in November 1982, effective biocidal exposures of 0.90 to 0.52
mg/L were obtained at 525 Ibs/day (Table 3). Similar TRC
levels (0.75-0.60 mg/L, were obtained in the same stations
at 450 Ibs/day on November 1 with even higher TRC levels
at 350 Ibs/day (1 .22-0.46 mg/L) on November 16, 1982. The
high and low discrepancies on November 16, 1982, between
Bldgs 1 2 and 1 0 (1 .22 and 0.29 mg/L, respectively) were due
to shifts in chlorinator flow between the 0.61 and 0.92-m lines
that fed these sampling stations.

The mean temperature at Bldgs 12, 10, 32 and 1
decreased steadily at the four sampling dates from August
26 to November 16 (25, 19, 16.4, and 10.5 C, respectively).
The decline in New River water temperature in the plant pro-
cessing stations from summer to fall conditions should have
resulted in lower chlorine usage during the fall to provide the
same degree of biocidal activity as that needed during the
summer. However, less chlorine (325 Ibs/day) was used daily
during the summer due to low New River flow. Similar daily
adjustments in the chlorination from 325 to 500 Ibs/day
were needed in the 1983 chlorination schedules due to river
fluctuations. Maximum chlorination output (500 Ibs/day) on-
ly occurred for 35% of each 28-day period. Therefore, TRC
released into the thermal effluent varied daily from ^0.06
to 0.02 mg/L. This maintained permit limitations of Uo.01
mg/L TRC released into the thermal discharge station.

Chlorination was highly effective in controlling Cor-
bicula infestation. Where dozens of fouling incidents occur-
red in 1982, only seven were reported in 1983. Only one was
due to clogging by live clams; all others were the result of

clam shells.

IDENTIFICATION OF CLAM INFESTATION AREAS WITHIN
THE PLANT

Several obsolete water pipe lines were identified and
removed. The water holding tower for emergency fire con-
trol was drained and sediment was removed. On the roof of
Bldg 10, three holding towers used for production purposes
were drained after each major spring and fall spawning period
determined from clam analysis at the pumphouse station.
During May 1 983, numerous juvenile clams (-1-2 mm) were
found in the tower sediment.

AQUATIC BENTHIC MACRO-INVERTEBRATE COMMUNITY
STUDIES

Each of the four sampling stations appeared to be
comprised of “healthy” aquatic macro-invertebrate com-
munities for all six sampling periods. That is, diversity of
organisms was consistently high at all stations. The mean
number of taxa per station was relatively high for the
1981-1983 combined samples, ranging from 21 .5 at Station
1 to 23.1 at Station 2 (waste water outfall) and 22.1 to 21.4
at Stations 3 and 4, respectively (Table 4). The mean

Table 4. Total number of invertebrate taxa from 3 replications per
station and averages before and after 4-week continuous chlorina-
tion and for all years combined (1981-1983). The rank assigned to
each station represents the degree of taxa abundance by assigning
a number from 1 to 4 to each station relative to the lowest and highest
number of taxa, respectively.

Date of

STATION

2 3

Sample

Before Four-Week Continuous Chlorination

September 1981

October 1981

May 1982

AVERAGE

26.7

27.3

28.7

28.0

After Four-Week Continuous Chlorination

August 1982

September 1982

October 1982

July 1983

September 1983

October 1983

November 1983

AVERAGE

19.3

21.3

19.3

18.6

Combined Average
Before and After
Chlorination

21.5

23.1

22.1

21.4

Range

16-33

17-35

17-34

14-29

Average Rank

2.05

3.10

2.35

2.50

CORBICULA SYMPOSIUM

Table 5. Shannon Wiener diversity of invertebrates per sampling sta-
tion and average diversity before and after continuous chlorination
and for all years combined (1981-1 983). The rank assigned to each
station represents the degree of diversity between stations by assign-
ing a number from 1 to 4 for each station relative to the lowest to
highest diversity, respectively.

Date of

STATION

2 3

Sample

Before Four-Week Continuous Chlorination

September 1981

3.495

3.040

3.000

2.868

October 1981

3.420

3.280

3.000

3.570

May 1982

2.695

3.380

2.716

2.449

AVERAGE

3.203

3.233

2.905

2.962

After Four-Week Continuous Chlorination

August 1982

2.888

3.662

3.134

3.333

September 1982

3.359

3.537

0.031

1.717

October 1982

3.339

3.738

2.387

1.665

July 1983

2.994

2.705

2.520

2.919

September 1983

3.005

3.296

2.985

3.384

October 1983

3.220

3.325

2.529

3.415

November 1983

3.132

3.013

2.291

2.457

AVERAGE

3.134

3.325

2.697

2.698

Combined Average
Before and After
Chlorination

3.155

3.298

2.759

2.778

Range

2.695-

2.705-

2.291-

1.665-

3.495

3.662

3.134

3.570

Average Rank

2.8

3.2

1.6

2.4

Shannon-Weiner diversity index values for the 1981-1982
combined samples were also high, ranging from 2.767 at Sta-
tion 4 to 3.440 at Station 2 (Table 5). In view of these trends,
the high number of taxa and diversity indices observed in-
dicated that the macroinvertebrate assemblages upstream,
adjacent to and downstream from the Celco Plant were
diverse.

When comparing mean taxa and diversity of organisms
before and after four-week continuous chlorination, the follow-
ing trends were observed. Stations 2 and 3, which were poten-
tially influenced by TRC, had an average of 23.1 and 22.1
taxa, respectively, after chlorination compared to 27.3 and
28.7 taxa before chlorination (Table 4). A similar decline,
however, was also seen in Stations 1 and 4 that had no TRC
influence. Mean taxa (before and after chlorination, respec-
tively) declined from 26.7 to 21 .5 in Station 1 and 28.0 to 21 .4
in Station 4. After chlorination, diversity at Station 2 ranged
from 2.705 to 3.738 with an average for 1982-1983 of 3.325
which was higher than the other stations (Table 5). These
high values were probably the result of organic enrichment
from the waste water discharge. Stations 2 and 3 were closest
to chlorinated influence but showed little difference relative
to the total number of taxa, rank of taxa between stations,
and diversity of invertebrates in the uninfluenced Stations 1
and 4.

No major differences were observed between chlorine
influenced Stations 2 and 3 with regard to the percent com-
position of invertebrates by order (T able 6). The greatest per-
cent abundance values (calculated as the mean percent
abundance value per station by groups) were associated with
Ephemeroptera (15.6), Trichoptera (26.1), and Gastropoda
(21 .5). This was consistent with the fact that the New River,
a wide shallow river with relatively high primary productivity,
is composed of benthic feeders that graze on periphyton or
filter seston.

The Ephemeroptera in the New River were primarily
collector-gatherers and scrapers, which feed on attached
plants. These numbers were relatively high in Stations 1 , 2,
and 3 with mean percent abundance values of 1 7.0, 1 7.6 and
18.0, respectively, from 1981-1983 (Table 6). Ephemeroptera
at Station 4 had an average percent abundance of 9.9.

Gastropods, which also feed on attached algae, com-
prised the second highest average percent abundance values
of 24.2, 28.3, 35.4 and 1 6.4 for Stations 1 , 2, 3, and 4, respec-
tively (Table 6). The percent abundance for Stations 2 and
3 was somewhat lower than those for the uninfluenced sta-
tions; these differences may be due to the replacement of
the functional (trophic) category that gastropods occupy at
Stations 1 and 4 and by mayflies at Stations 2 and 3.

Trichoptera in the New River are principally filter-
feeders, feeding on organic particles and smaller organisms
drifting with the current. Their abundance comprised an
average of 26.1% of all 4 stations, with a range from 16.4
at Station 4 to 28.3 at Station 2 (Table 6).

When comparing changes in the percent abundance
of insect orders before and after chlorination, the following
results were observed. The decline in ephemeropterans and
trichopterans after chlorination was coincidental to the in-
crease in Corbicula densities, even in Station 1 upstream from
the Celco plant. At Stations 2 and 3, Corbicula and gastropods
increased after chlorination while ephemeropterans and
trichopterans declined. Dipteran densities declined in all sta-
tions in mid to latter 1982 but increased in 1983 (Table 6).

DISCUSSION

CORBICULA NUMBERS AT THE PUMPHOUSE

The potential for Corbicula infestation at the New River
pumphouse station and fouling within the Celco Plant was
realized from May 28, 1 981 through August 1 982, as evidenc-
ed by field population numbers and by many in-plant fouling
occurrences. From the initial occurrence of clam fouling in
the air conditioning units of the process control room in Bldg
10, disruptions from clogged condensers occurred through
1981 and progressively increased in spring and early sum-
mer of 1982 in units in addition to air conditioning (e.g., con-
denser systems in turbine generators, circulating air and oil
coolers, distilled water systems in Stokes Stills, carrier
calcium chloride chillers, acetone recovery units).

The source of Corbicula fouling was identified at the
Celco pumphouse where clam populations developed to
unusually high numbers. Part of this proliferation at the New
River pumphouse station may have been due to deicing of

CHERRY ETAL: CORBICULA FOULING AND CONTROL

Table 6. Average percent abundance of selected insect orders by station before and after four-week continuous chlorination and for
1981-1983 combined.

Date of

Sample

Sampling

Station

Percent Composition

Ephemerop-

tera

Trichop-

tera

Coleop-

tera

Diptera

Gastropoda

Pelecypoda

Other

Before Four-Week Continuous Chlorination

1981-1982

23.6

24.4

6.7

8.0

30.6

0.2

13.2

Average

14.4

31.1

11.2

18.4

11.3

9.3

15.5

21.5

32.3

5.3

16.5

11.0

8.2

10.5

18.6

20.3

9.1

11.0

29.2

12.5

8.4

After Four-Week Continuous Chlorination

1982

10.5

25.5

15.5

1.8

29.8

11.1

21.3

Average

26.7

9.2

4.0

16.1

13.1

13.4

13.5

22.6

8.0

2.1

28.4

19.9

13.5

7.0

7.1

5.8

1.0

33.9

43.3

7.1

1983

17.2

21.9

16.5

2.8

25.7

11.2

5.5

Average

13.2

27.3

12.7

6.8

23.5

14.0

2.6

18.6

48.4

5.0

8.8

12.9

5.0

1.5

9.3

19.7

11.1

6.7

27.1

21.9

4.4

Overall Average

14.3

22.6

16.1

2.4

27.4

10.7

5.7

After Chlorination

19.0

27.0

11.2

5.6

19.5

13.6

3.3

(1982-1983)

16.7

37.3

6.3

5.8

19.5

11.4

3.2

8.3

14.5

8.8

4.2

30.0

31.1

3.1

Average Percent Abundance by Station for 1981-1983

17.0

24.2

13.2

4.1

28.4

21.0

5.8

17.6

28.3

11.1

17.5

13.1

3.6

18.0

35.4

6.0

9.0

17.0

12.7

3.9

9.9

16.4

8.8

6.2

23.1

28.1

3.6

Average Percent Abundance for Each Taxonomic Group for All Stations Considered Collectively

15.6

26.1

9.8

7.2

21.5

18.7

4.2

the travelling screens during the winter months with heated
water which was released into the immediate New River sedi-
ment (4-8 C). This practice could have enhanced Asiatic clam
proliferation in the New River sediments when seasonally am-
bient winter temperatures reached O C which could have sup-
pressed their development. Mattice and Dye (1976) reported
2 C as the lower temperature where Corbicula could not be
acclimated. Cherry et al. (1 980) reported that clams residing
in stations outside the thermal discharge of the Glen Lyn
Power Plant (several miles downstream of the Celco Plant)
were eradicated during the winter when river temperature
dropped below 2 C. The semi-fine, granular sediments, which
were well aerated from the New River flow and continuous
Celco Plant pumping activities, served to harbor and promote
clam development in water temperatures several degrees
above freezing in the immediate vicinity of the pumphouse.

Clam numbers were highest in the immediate vicinity
of the Celco Plant pumphouse station, probably due to heated
water released into the station from the travelling screen deic-
ing procedures in the winter. Although adult clam numbers
were reduced by 4 and 2-fold, respectively, over the winter

to following spring of 1981-1982 and 1982-1983, sufficient
numbers were available each spring to allow for a great
amount of larval production during the spring spawning
season. The clam population at the Celco Plant was extremely
high regarding clam numbers in comparison to clams sampl-
ed downstream at the Glen Lyn Plant (Cherry et al., 1980).
The highest number of clams collected at the Glen Lyn Plant
in the thermally discharged channel in February 1978 was
11,522/m2 which was much less than the highest obtained
in this study (269,105/m2).

CORBICULA SPAWNING ACTIVITY

Corbicula spawning has been reported during the
spring and fall or continual in some localities (Britton, in
press). Eng (1979) found spawning to be biannual from mid-
April through May and from mid-August through September
in the San Joaquin River system, California, while Sickel
(1979) found Corbicula to have a strong spring spawn and
a weaker fall spawn with lesser but continuous spawning in
between.

CORBICULA SYMPOSIUM

Spawning of clams in the Savannah River, South
Carolina appears to be continuous from April to November
(R. S. Harvey, personal communication). In 1981, we noted
major spawning activity by August 5, 1981, since 84.6% of
the young clams in the sediment were ^ 1 mm in size and
were not found in the previous sampling effort of July 1 3, 1 981
(Table 1). By September 29, 1981, spawned larvae 1 mm)
comprised 76.9% of the sample or 207,030/m2 of sediment
at the pumphouse station. Between August 5, 1981, and
September 29, 1981, clams ^ 1 mm in size had increased
by 25-fold in the pumphouse sediment (8,242 vs 207,030/m2,
respectively). Spawning in the fall of 1 981 was essentially over
by November 11, 1981, after peak spawning had probably
occurred by September 29, 1981 or soon thereafter.

Clam spawning was generally continuous from June
8, 1982 to October 29, 1982 and May 25, 1983 to November
29, 1983 (Table 1). Initial spawning activity generally occur-
red at 17-22 C and 14-17 C in the spring of 1982 and 1983,
respectively. The greatest number of newly spawned clams
(^ 1 mm) in the sediments occurred from July 19 through
November 29, 1982 which suggested that no two peak spring
and fall spawns had occurred. In 1983, however, two major
spawns were evident during June 14 and September 29.
These data have made it difficult to gauge major spawning
conditions with river temprature. Determining the major fall
spawning activity was even more difficult since highly
abundant juveniles in the sediment were obtained at 17-22
C in 1 981 , 8-1 2 C in 1 982 and 1 6-24 C in 1 983. At the Celco
Plant, monthly monitoring of juvenile clams in the sediments
has been established in order to decide when to commence
with a 28-day continuous chlorination after a major output
of juvenile clams has been determined during spring and fall
months.

Determination of spring and fall spawning seasons of
Corbicula in the thermal discharge station cannot be iden-
tified. The number of juvenile clams sampled in the sediment
was low during the three months when larvae ^ 1 mm were
found in 1981-1983 (Table 2). Water temperature ranges were
high during the previous two weeks prior to sampling (e.g.,

31- 38 C in September, 1981, 37-41 C in July, 1982 and

32- 37 C in June, 1983). Most likely, young juveniles were
either transported from areas of lower temperature at the
pumphouse located above this station or from infested sites
within the plant allowing juveniles to grow in the heated
effluent.

PRE- AND POST-MUCKING SEDIMENT PROCESS

Removal of sediments at the Celco pumphouse
station appeared to have a positive effect in reducing adult
clam densities in the New River. It appeared that the
removal of sediment disrupted many clams but for only
a short period of time. In general, a four-fold decrease in adult
clam (sMO mm) numbers was noted between sampling of July
1981 (pre) to July 1982 (post-mucking) (Table 1). Following
4 months after sediment removal adult clam numbers were
almost the same between pre- and post-mucking activities
when comparing data from November 11, 1981 and
November 1982 (1,540 and 1,236/m, respectively). Although

the overall amount of sediment removed at the pumphouse
was ~ 70% effective, sampling of clams in sediment patches
where mucking efforts had missed showed a recovery of adult
densities by Novemer 1982 relative to the pre-mucking
samples. Since the sediment removal process was only tem-
porary it is not recommended as a future clam control pro-
cedure at other industrial installations.

The regularity of sediment removal at the pumphouse
has not been decided. The overall mucking procedure is a
time consuming process (e.g., two weeks) since sediments
were removed in an area along the length of the pumphouse
(45 m) to a distance of 30 m outward into the river. Clam
numbers were highest within the first 10 m adjacent to the
pumphouse with numbers diminishing rapidly thereafter (e.g.,
335-2,345/m2 at 10 m to 1 34-201 /m 2 at 30 m). A repeat of
sediment removal on a two-year basis is not being considered
unless a return of high sediment deposition adjacent to the
pumphouse becomes evident.

COMPARISON OF THERMAL DISCHARGE TRC CONCEN-
TRATION TO FISH AVOIDANCE AND TOXICITY

Biocidal effective levels of total residual chlorine for
a four-week continuous chlorination period have been assum-
ed to be *5 0.50 to 1 .0 mg/L (B. G. Isom, personal communica-
tion). It was difficult to maintain a chlorinated residual of 0.50
mg/L TRC throughout the plant even when the New River
was high for appropriate dilution capabilities. At buildings
nearest the chlorinator, Bldg 12 (275 m away), TRC was as
high as 1.22 mg/L while at the farthest building (Office-
Computer Building, 790 m away) TRC dropped to 0.36 mg/L
(Table 3). Clam fouling problems were most frequent at the
Office-Computer Building of Celco where chlorinated
residuals were lowest. The maximum residual measured in
the thermal discharge station was 0.27 mg/L TRC with a
calculated - dilution factor ^ 0.01 mg/L TRC entering the New
River. TRC measurements 50 m below the effluent in the New
River were not detectable due to the rapid dissipation of the
chlorinated discharge with the assimilative capacity of the
New River. Chlorine measurements through the 130-m
distance of the thermal discharge channel to the New River
confluence showed ~ 50% reduction in TRC.

The effects of TRC during continuous 28-day applica-
tion were assumed to have minimal if any effects upon the
fish populations in the New River near the Celco Plant
because of rapid chlorine dissipation. The highest concen-
trations of TRC measured at the Celco Plant in the thermal
discharge (0.27 mg/L and within the New River (^ 0.01 mg/L

- after calculated dilution factors) downstream of the plant
can be avoided by fish. A majority of the fish species tested
at the Glen Lyn Laboratory have avoided TRC between 0.05

- 0.20 mg/L TRC (Cherry and Cairns, 1982). Rainbow trout
(Salmo gairdneri) have been reported to have a 48-hr LC50
response of 0.09-0.16 mg/L TRC to intermittent chlorination
(Heath, 1977). However, no rainbow trout are found in this
part of the New River or at the Glen Lyn Plant (Stauffer et
al., 1976; Hocutt, 1974). Besides, rainbow trout have been
reported to avoid “continuous” chlorination exposures for
one-hour periods in the laboratory at 0.05-0.10 mg/L TRC

CHERRY ET AL: CORBICULA FOULING AND CONTROL

(Cherry et al., 1982). Even though TRC was continuously
released daily into the thermal discharge station, chlorine con-
centrations would usually vary from one day to the next from
^ 0.06 to 0.27 mg/L. Some days, TRC was not detectable
when the chlorinator malfunctioned. Since there was con-
siderable manipulation of the Celco chlorinator during the
four-week dosing period, the release of TRC into the New
River was not at a constant rate, and at most times fluctuated
on a daily or semi-daily basis depending upon the river flow.

For selected fish species inhabiting the New River, in-
termittent chlorine, acute toxicity values (TRC with high com-
bined residual in mg/L) ranged from 1.50-2.37 for carp
( Cyprinus carpio ), 0.41-0.65 for spotfin shiner ( Notropis
spilopterus), 1.23-3.00 for bluegili (Lepomis macrochirus),
1.15-2.87 for white bass (Morone chrysops), 0.65-0.78 for
channel catfish (Ictalurus punctatus), 0.36-1.09 for white
sucker (Catostomus commersoni), to 1.26-2.03 for mos-
quitofish (Gambusia affinis) (Seegert et al., 1979; Cherry et
al., 1982). Other reviews by Mattice and Zittel (1976) and
Turner and Thayer (1979) provided toxicity data on con-
tinuous, 96-hour chlorine exposures which are not environ-
mentally realistic to the fluctuating TRC released at the Celco
Plant. The above toxicity concentrations of TRC, along with
avoidance capability indicate that no harmful effects should
occur for fish from the Celco Plant TRC effluent that ap-
proached the 0.01 mg/L calculated level in the immediate
New River receiving system. In addition, no documentation
of fish kills have been reported from the 28-day continuous
procedures during 1982-1983.

Acute effects of chlorinated residuals to aquatic, ben-
thic invertebrates ranged from 0.009 mg/L for isonychia, 0.396
mg/L for Hydropsyche, to 0.502 mg/L for Stenonema (Gregg,
1974). Acute toxicity of chlorine to snails ( Nitocris and Physa,
respectively) ranged from 0.044 and 0.258 mg/L (Mattice and
Zittel, 1976). Since the chlorination residuals were not detec-
table at Stations 2, 3, and 4, it is assumed that the continuous
TRC released has no measurable impact upon these in-
vertebrate communities, especially when diversity was
highest at the two stations closest to potential chlorine
influence.

BENTHIC INVERTEBRATE SAMPLING

Benthic invertebrate sampling in the New River, three
times before continuous chlorination and seven thereafter
during 1981-1983, showed no observable impact of residual
chlorine on the invertebrate communities in the stations
sampled immediately within the Celco Plant (Tables 4-6). Sta-
tion 2 was closest to the potential influence of chlorinated
discharge; however, diversity of aquatic invertebrates was
higher (3.325) after chlorination than before (3.233). Taxon
diversity indices between pre- and post-chlorination were
similar at Station 3 (e.g., 2.905 versus 2.697, respectively).
Diversity values at Station 4 ( - 1 .0 km below Station 3) were
as comparable before and after chlorination (2.962 to
2.698, respectively); however, chlorinated effluent had been
completely diluted within the first 50 m of the river before
reaching this station. The average abundance of insect orders

and other invertebrates had changed due to seasonal dif-
ferences after chlorination or due to high numbers of Cor-
bicula having been collected downstream in Station 4 which
influenced low diversity indices (Tables 5-6). It is difficult to
determine whether the reduction in diversity at this station
was due to natural clam invasion or from Celco Plant
discharges. Since Corbicula have only recently invaded this
river system, the shift to lower diversity downstream for
selected insect taxa may have been attributed to the increase
in Corbicula densities. In all stations sampled upstream,
within and downstream of the Celco Plant, a corresponding
increase in clams resulted in a decline in many major insect
groups especially at Station 4 during 1982. During 1983,
however, clam densities in the New River declined
downstream at Station 4 from 43.3% in 1982 to 21.9% in
1983. Insect orders (Ephemeroptera, Trichoptera, Coleoptera,
Diptera) showed a major increase in abundance in 1983.

DISCUSSION OF OTHER APPROACHES FOR CORBICULA
CONTROL

Use of mollusk biocides may have detrimental effects
upon fish and invertebrate fauna in the New River receiving
system below the thermal discharge station and so was
dismissed. Bayer 73 (2-hydroxy-5, 2’ dechlor-4' nitro ben-
zanilide) has been reported to produce a 24-hr TLm of 0.18
mg/L to bluegili sunfish (Sinclair and Isom, 1963). The abili-
ty of Corbicula to “clam up” and thus “avoid” the toxic con-
sequences of these molluscides may result in far greater
ecological hazards to the other biota in the receiving system.

The potential of heated water backflushing through the
Celco Plant was not attempted due to production protocol
and the concern for clogging of production systems with dead
clam shells. Use of industrial anti-flocculants to remove sedi-
ment accumulation was not attempted. We assumed that
Asiatic clam infestation occurs in dead pipe spaces where
the clam can congregate into sediments, not within rapidly
flowing pipe lines. Other industrial biocidal chemicals
(hypobromous acid in addition to reduced chlorination) are
available and are effective for slime (e.g., bacterial, fungal,
algal accumulations) control, but their usefulness for Cor-
bicula biocidal effectiveness is unknown. The use of a
counter-current pump system has merit in reducing sediment
and detrital (and potential clam larvae and shells) material
at the initial pumphouse installations, but the cost in im-
plementing and maintaining this pumping system relative
to Corbicula control may be too high due to the high daily
water volume used by the plant. More research in the above
arenas is recommended before any can be utilized as sup-
portive measures of chlorine in controlling the Asiatic clam.

A major concern at the Celco Plant is that no holding
ponds exist between the thermal discharge and the river. If
ponds were available, they could dilute the chlorinated
residual before it enters the New River. For this plant, a
biocidal agent needed is one with highly ephemeral
characteristics that can be diluted, absorbed, precipitated or
assimilated as a consequence of the assimilative capacity
of the New River. Total residual chlorine has been selected

CORBICULA SYMPOSIUM

as the most appropriate biocidal agent currently available
since it meets the above specifications. The use of chlorine
as a biocidal agent for Corbicula control is suggested for in-
dustrial installations that lack holding ponds to the receiving
river system if a data base of chlorine toxicity with avoidance
behavior of fish and/or concurrent invertebrate field surveys
are available.

SUMMARY

Proliferation of Corbicula in the New River at the Celco
Plant pumphouse station, Celanese Corporation, Narrows,
Virginia, resulted in a major infestation of larval clams with
biofouling disruptions in production and air conditioning con-
denser systems during 1981-1982. Production procedures
such as deicing practices of travelling screens at the pump-
house probably allowed Corbicula to survive adequately dur-
ing the winter and propagate in spring. Dead pipe spaces
and reserve river water holding towers within the plant en-
abled the clam to proliferate further resulting in biofouling in-
cidents in production and air conditioning condenser systems.
These systems either have been removed or periodically
flushed twice each year to remove sediment and juvenile clam
presence. During 1983, six of the seven major clam fouling
incidents were due to dead clam shells which showed that
continuous 28-day chlorination had a positive effect in
minimizing Corbicula infestation after four treatments in
1982-1983. Control measures for Corbicula carried out in
1981-1983 included documentation of spawning occurrences
in the New River sediment coupled with four-week continuous
chlorination at a target, in-plant concentration of ^0.50 mg/L
TRC to minimize larval clam infestation. Much of the sedi-
ment accumulation at the river pumphouse had been re-
moved by a mucking-bilge pump operation in order to reduce
the dense, resident clam population from releasing larvae into
the plant. However, this procedure had a short-term effect
of four months in reducing the adult clam population in the
river sediment. The overall approach by the Celco Plant for
Corbicula control is both traditional and different. Continuous
chlorination for four weeks at anticipated peak, biannual
spawning seasons normally has been used to ensure clam
biocidal effectiveness. Concommittent field sampling of
macro-invertebrate populations found upstream, adjacent to
and downstream from the plant, to ensure that residual
chlorine released from the plant was not harmful to the biota
in the New River is unique. Review of fish responses to
chlorinated residuals (e.g., avoidance behavior relative to tox-
icity concentrations of chlorine) in the literature has shown
no effect (e.g., documented fish kills) from the use of chlorine
to control Corbicula at the Celco Plant other than potential
avoidance reactions.

ACKNOWLEDGMENTS

This research was supported by the Celco Plant, Celanese
Corporation, Narrows, Virginia, 24124.

LITERATURE CITED

Britton, J. C. (In Press). Biogeography and ecology of the Asiatic clam,

Corbicula, in Texas. Texas Journal of Science.

Burch, J. Q. 1944. Checklist of West American mollusks, family Cor-
biculidae. Minutes of the Conch Club of Southern California
36:18.

Cherry, D. S. and J. Cairns, Jr. 1982. Biological monitoring, part V:
preference and avoidance studies. Water Research
16:262-301.

Cherry, D. S., S. R. Larrick, K. L. Dickson, J. Cairns, Jr., and R. C.
Hoehn. 1977. Significance of hypochlorous acid in free
residual chlorine to the avoidance response of spotted bass
( Micropterus punctulatus) and rosyface shiner ( Notropis
rubellus). Journal of the Fisheries Research Board of Canada
34:1365-1372.

Cherry, D. S., S. R. Larrick, J. D. Giattina, J. Cairns, Jr. and J. H.
VanHassel. 1 982. Influence of temperature selection upon the
chlorine avoidance of cold and warmwater fish. Canadian Jour-
nal of Fisheries and Aquatic Sciences 39:162-173.

Cherry, D. S., J. H. Rodgers, Jr., R. L. Graney and J. Cairns, Jr.
1980. Dynamics and control of the Asiatic clam in the New
River, Virginia. Virginia Water Resources Research Center
Bulletin 123, 72 pp.

Eng, L. L. 1979. Population dynamics of the Asiatic Clam, Corbicula
fluminea (Muller), in the concrete-lined Delta-Mendota Canal
of Central California. Proceedings of the First International Cor-
bicula Symposium, J. C. Britton, ed. J. S. Mattice, C. E. Mur-
phy, and L. W. Newland, Associate eds. Texas Christian
University Research Foundation Publication, Fort Worth,
Texas, pp. 39-68.

Fuller, S. L. H. and C. E. Powell. 1973. Range extension of Corbicula
manilensis Phillippi in the Atlantic drainage of the United
States. The Nautilus 87:59.

Giattina, J. D., D. S. Cherry, S. R. Larrick and J. Cairns, Jr. 1981.
Comparison of laboratory and field avoidance behavior of fish
in heated, chlorinated water. Transactions of the American
Fisheries Society 110:526-535.

Gregg, B. C. 1974. The effects of chlorine and heat on selected
stream invertebrates. Ph.D. Dissertation, Virginia Polytechnic
Institute and State University, Blacksburg, Virginia. 154 pp.

Harvey, R. S. Personal Communication. E. I. du Pont, Building
777-1 0A, Savannah River Plant, Aiken, South Carolina 29801 .

Heath, A. G. 1977. Toxicity of intermittent chlorination to freshwater
fish. Influence of temperature and chlorine form. Hydrobiologia
36:39-47.

Hocutt, C. H. 1974. Methodology developed for the assessment of
fish and macroinvertebrate communities in the New River,
Virginia. Ph.D. Dissertation, Department of Biology, VPI&SU,
Blacksburg, Virginia. 174 pp.

Isom, B. G. Personal Communication. TVA Division of Environmen-
tal Planning, Water Quality and Ecology Branch, Mussels
Shoals, Alabama 35660.

Isom, B. G. 1971. Evaluation and control of macroinvertebrate
nuisance organisms in freshwater industrial supply systems.
Paper presented at 19th Annual Meeting of Midwest Ben-
thological Society, March 24-26, 1971.

Joy, J. E. and L. E. McCoy. 1975. Comparisons of shell dimensions
and visceral mass weights in Corbicula manilensis (Philippi,
1844). The Nautilus 89:51-54.

Mattice, J. S. and L. L. Dye. 1976. Thermal tolerance of the adult
Asiatic clams. In, Thermal Ecology II, G. W. Esch and R. W.
McFarlane, editors. National Technical Information Service,
Springfield, Virginia 22161. pp. 130-135.

Mattice, J. S. and H. E. Zittei. 1976. Site-specific evaluation of power
plant chlorination. Journal Water Pollution Control Federation
48:2284-2308.

CHERRY ETAL: CORBICULA FOULING AND CONTROL

McMahon, R. F. 1977. Shell size-frequency distribution of Corbicula
manilensis from a clam-fouled stream condenser. The Nautilus
91 :54-59.

McMahon, R.F. 1982. The occurrence and spread of the introduced
Asiatic clam Corbicula fluminea (Muller) in North America
1924-1982. The Nautilus 96:134-141.

Rodgers, J. H., Jr., D. S. Cherry, J. R. Clark, K. L. Dickson and J.
Cairns, Jr. 1977. The invasion of Asiatic clam, Corbicula
manilensis in the New River, Virginia. The Nautilus 91 :43-46.

Seegert, G. L, A. S. Brooks, J. R. Vande Caste and K. Gradali. 1979.
The effects of monochloramine on selected riverine fishes.
Transactions of the American Fisheries Society 108:88-96.

Sickel, J. B. 1979. Population Dynamics of Corbicula in the Altamaha
River, Georgia. Proceedings of the First International Corbicula
Symposium, J. C. Britton, ed., J. S. Mattice, C. E. Murphy,
and L. W. Newland, Association eds. Texas Christian
University Research Foundation Publication, Fort Worth,
Texas, pp. 69-88.

Sinclair, R. M. 1964. Clam pests in Tennessee water supplies. Journal
American Water Works Association 56:592-599.

Sinclair, R. M. 1971. Annotated bibliography on the exotic bivalve Cor-
bicula in North America, 1900-1971. Sterkiana 43:11-18.

Sinclair, R. M. and B. G. Isom. 1963. Further studies on the introduc-
ed clam ( Corbicula ) in Tennessee Stream Pollution Control
Board, Tennessee Department of Public Health. 76 pp.

Stauffer, J. R., K. L. Dickson, J. Cairns, Jr. and D. S. Cherry. 1976.
The potential and realized influences of temperature on the
distribution of fishes in the New River, Glen Lyn, Virginia.
Wildlife Monographs 50:1-40.

Thomas, N. A. and K. M. MacKenthun. 1964. Asiatic clam infesta-
tion at Charleston, West Virginia. The Nautilus 78:28-31.

Turner, A. and T. A. Thayer. 1979. Chlorine toxicity in freshwater
ecosystems. An evaluation of the Mattice and Zittel model for
deriving toxicity thresholds and a proposed alternative model.
Prepared for Edison Electric Institute, Washington, D C. by
Envirosphere Company, New York, New York 10006. 51 pp.

ASIATIC CLAM CONTROL BY
MECHANICAL STRAINING AND ORGANOTIN TOXICANTS

YUSUF G. MUSSALL!

STONE & WEBSTER ENGINEERING CORPORATION
245 SUMMER ST.

BOSTON, MASSACHUSETTS 02107, U.S.A.

I. A. DIAZ-TOUS

ELECTRIC POWER RESEARCH INSTITUTE
3412 HILLVIEW AVE.

PALO ALTO, CALIFORNIA 94303, U.S.A.
JAMES B. SICKEL
MURRAY STATE UNIVERSITY
MURRAY, KENTUCKY 42071, U.S.A.

ABSTRACT

Corbicula (Asiatic Clams) control technologies include mechanical/physical, chemical, and
biological controls. Large in-line strainers have been used successfully to filter adult clams from cir-
culating water systems. Fine mesh screens with 0.5 mm mesh have recently been evaluated in the
United States for the protection of fish eggs and larvae. Screens with 0.2 mm mesh may be feasible
to screen clam larvae for small flow rates. Very fine mesh screens require very low flow velocities
resulting in large intake structures. The Electric Power Research Institute (EPRI) has investigated
the effectiveness of tributyl tin fluoride (TBTF) pellets to control clam larvae, juveniles, and adults
by conducting preliminary tests at Murray State University. The test results show TBTF to be highly
toxic to all stages of Corbicula.

Asiatic clams ( Corbicula ) are spreading rapidly through
the United States, causing flow blockage and reliability pro-
blems in many freshwater circulating and service water
systems. The Nuclear Regulatory Commision (NRC) has
recently requested plant operators to determine whether the
Asiatic clams are present, to measure the flow rates in
systems that have clams, and to outline future preventative
and detection methods. The clams in once-through cooling
systems are usually found at the bottom of the intake struc-
ture, in the intake tunnels, at the condenser water boxes, in
the condenser tubes, in fire-protection lines, and in the
discharge canals. In closed-cycle systems, clams often ac-
cumulate in cooling tower basins and cooling ponds.

removal has not proven to be very effective and should be
used in combination with other mechanical and chemical con-
trols. Thermal backwash is one of the most promising physical
controls. However, existing freshwater power plants are
usually not equipped with thermal backwash capabilities
because of the recent advent of the Asiatic clams problem.

The Electric Power Research Institute (EPRI), under
Project RP1 689-9, is monitoring the performance of several
mechanical straining systems at existing power plants and
has investigated the use of tributyl tin fluoride (TBTF) pellets
to locally control Asiatic clams in pipelines or intake bays.

This paper discusses the results of these investiga-
tions in order to transfer these technologies to power plant
application.

Control techniques include mechanical/physical,
chemical, and biological controls (Mattice 1983). Biological
control by predators have not yet proven to be effective.
Chemicals such as chlorine have been used with some suc-
cess. Oxygen scavengers such as sodium-bisulfite and
hydrogen sulfide have been used to control clams at intakes.
Chemicals have not been widely used because of en-
vironmental regulations. At some plants, clams are removed
physically on a periodic schedule. However, physical

To exclude clam larvae from entering cooling water
systems, mesh openings as small as 0.2 mm (200 ^m) are
required. Very fine mesh screens with openings as small as
0.001 mm (1 /iim) have been used in tertiary wastewater treat-
ment plants. The flow velocities associated with these very
fine mesh screens are about 0.3 cms (0.01 fps) with the

MECHANICAL STRAINING

American Malacological Bulletin, Special Edition No. 2(1986):83-88

CORBICULA SYMPOSIUM

screen traveling continuously at about 15 mpm (50 fpm). Very
fine mesh screens are not practical to use in screening cir-
culating water systems because they would require many
screens and very large intake structures. However, for
makeup or service water systems, fine mesh (0.2 mm) screen-
ing is operationally feasible, although costly.

Recently, a fine mesh screen (Mussalli et a!., 1981)
with mesh of 0.5 mm (500 /im) and 50 percent open area has
been tested successfully for the protection of fish eggs and
larvae at the Big Bend Station of Tampa Electric Company.
The screen operated continuously with flow velocities of 15
cms (0.5 fps) and 30 cms (1 fps) and at speeds ranging from
2 mps (7 fps) to 8 mps (28 fps). The encouraging test results
have resulted in the incorporation of screens with 0.5 mm
mesh for fish eggs and larvae protection at the Big Bend Sta-
tion in Florida and at the Prairie Island Nuclear Station on
the Mississippi River. Another traveling screen of a center
flow design with 1 mm mesh (1000 nm) has also been
operating successfully at the Barney Davis Station of Cen-
tral Power and Light in Corpus Christi, Texas. The mesh is
nylon with about 50 percent open area.

Continuously traveling screens with 0.2 mm mesh
either of metallic or synthetic fabric and designed with low
approach velocities of about 3 to 6 cms (0.1 to 0.2 fps) could
be operationally feasible for small flow rates. This concept,
however, should be tested in situ prior to prototype applica-
tion and would result in large intake structures requiring five
to ten times more traveling screens than conventionally used.

In-line strainers located between the pump and the
condenser or heat exchangers also can be used to control
juvenile and adult clams (Mussalli and Diaz-Tous, 1983). This
technology is of German and French origin and has recently
begun to be used in the United States. Currently, there are

Fig. 1. Cutaway drawing of a typical amertap debris filter.

CARTRIDGE

WATER (WITH DEBRIS)

Fig. 2. Shellfish strainer. Source: Beaudrey/COGENEL.

Fig. 3. Integrated filter strainer design at the Paluel Nuclear Station

(France).

large strainers in ten U.S. power plant cooling water pipelines.
For a large diameter pipe (up to 2.8 m; 1 10 in.) with a capacity
of 15.5 m3/s (245,000 gpm), a debris filter of a perforated
stainless steel screen (mesh size ranges from 2.8 mm (1/9
in.) to 9.5 mm (3/8 in.) concentrically aligned within a steel
casing as shown on Figure 1 can be used. The filters can
be installed inside the turbine building at the pump house
or in the yard between the pump and the condenser employ-
ing a variety of mounting positions.

The ability of these filters to remove entrained debris
and biogrowth (such as mussels, barnacles, and Asiatic
clams) has been demonstrated at several power plant loca-
tions including Southern Califorina, North Carolina, Ohio, and
Connecticut.

The French also have used self-cleaning strainers that
can be installed at the entrance to the water box. The strainer
has a volute-shaped shell as shown in Figure 2. One feature
of this design is the absence of moving parts.

The integrated filter/strainer design to supplement in-
take screening as it is used at the Electricite de France Paluel

MUSSALLI ETAL: ASIATIC CLAM CONTROL

(1300 MW PWR) Nuclear Station is shown in Figure 3.

One advantage of the filter/strainer is that it improves
condenser performance when used in conjunction with a tube
cleaning system.

Public Service of Indiana at its Cayuga Station has
devised an in-line angled strainer to divert adult clams to a
blowoff bypass, as shown in Figure 4. This design is simple
and has been operating successfully. It is adaptable, however,
to clean circulating water flows such as in closed-cycle
systems where debris is a minimum.

The Electric Power Research Institute is monitoring
the performance of several large filters of German design at
the South Bay Station of San Diego Gas & Electric Corpora-
tion, at Sammis Staion of Ohio Edison Company, at the
Nueces Bay Station of Central Power and Light, and at the
Jack Watson Station of Mississippi Power Company. Another
filter of French design used to filter seaweed and grasses
is also being monitored in Puerto Rico to determine any
operational problems and solutions.

OGRANOTIN TOXICANTS

Tributyl tin fluoride (TBTF) in parts per billion or parts
per trillion concentration may provide an economical
mechanism for controlling clam larvae; in higher concentra-
tions, TBFT may kill adult clams in intake structures and
makeup ponds. The TBTF pellets can be located in low veloci-
ty areas of intake structures where clams congregate, in ser-
vice water pipe laterals, in fire protection pipes, and can be
dispersed in cooling ponds. The organotins will leach out
slowly to locally treat clams in the flow. The advantage of a

controlled release mechanism lies in its simplicity of applica-
tion and low maintenance cost. The effects on the environ-
ment will be minimized, because only a small localized
volume of the flow will be exposed to organotin.

Under Project RP-1 689-9, EPRI sponsored preliminary
tests to determine the effectiveness of TBTF pellets on Cor-
bicula (Asiatic clams) larvae and adults in flowing and static
conditions (EPR1 1 984). Continuous flow and static tests were
conducted at Murray State University in 1983. The compound
tested in this EPRI study was a porous plastic matrix con-
taining 30 percent by weight of TBTF and designated
ECOPRO 1330-S by Environmental Chemicals, Inc. The
solubility of TBTF in water at 20°C is only 0.4 mg/I, but within
the plastic matrix the TBTF hydrolyzes to tributyl tin oxide
(TBTO), which has a solubility of about 10-40 mg/I (Himel
1983).

The ECOPRO 1330-S contained ferric oxide to give it
a density slightly greater than that of water. It was provided
as cut or ground pellets of irregular shape, ranging from 3
to 6 mm in diameter. According to Himel (Personal Com-
munication), the release rate of TBTO from freshly prepared
pellets when placed in water would decrease for several
weeks, then increase slightly to a constant rate that would
be maintained for approximately 2 years. This constant
release rate under continuous flow, unsaturated conditions,
was estimated to be 0.285 ^g/min/g pellet. Calculated con-
centrations of TBTO used in this EPRI study were based on
this estimated release rate.

CONTINUOUS FLOW TEST. This test exposed clam
larvae and adults to water that was briefly in contact with

CORBICULA SYMPOSIUM

TBTF pellets in a continuous flow system to simulate a water
intake or cooling water system in a power plant. The objec-
tive was to determine the lethal concentration at which 50
percent of test animal die (LC50) of larvae and adult
Corbicula.

In these tests filtered lake water was pumped at a con-
stant rate through containers with different amounts (by
weight) of TBTF pellets and allowed to flow into 6-liter glass
tanks containing adult Corbicula. Concurrently, a small frac-
tion of the treated water was diverted into 10 ml tissue culture
dishes containing juvenile clams ranging in age from 1 to
several weeks from the time of release. The continuous flow
of water provided a constant concentration of TBTO under
uniform conditions at 20°C (±0.5°).

The continuous flow system with the chemical con-
tainers is shown in Figure 5. Six containers were used with
combinations of the following weights of pellets: 27.7 g,
15.6 g, 8.76 g, 4.92 g, 2.77 g, 0.0876 g and a control with no
pellets. Flow rates were adjusted to either 250 ml/min or 300
ml/min. The calculated concentrations of TBTO in each test
based on expected release rate and volume of low are
presented in Table 1 . Typical continuous flow test results are
given in Tables 2 and 3.

Water supplied for all tests was obtained from Ken-
tucky Lake, Tennessee River, from where it was pumped
about 50 m into the Hancock Biological Station of Murray
State University where the tests were conducted. The water
was filtered to remove large zooplankton and sediment, but
small phytoplankton remained to provide food for the clams.
A constant level reservoir tanks of 285 I (75 gal) was used
to maintain a constant head for the gravity flow system. A
heater in the tank maintained the temperature at 20°C
( ± 0.05°) Aeration maintained the dissolved oxygen concen-
tration near saturation.

Larval and early juvenile Corbicula are microscopic,
ranging in length from 0.225 to 0.25 mm. Because of
their small size and mobility they require special handling

GLASS TANK ADULT CLAMS

Fig. 5. Continuous flow system with chemical container and exposure
tank for adult clams.

Table 1. Weights and calculated concentrations of TBTF pellets at
each flow rate used in tests.

Volume of Flow

250 ml/min 300 ml/min

Concentration

Weight(g)

ECOPRO

Mg/I

iogOtg/i)

AAQ/I

log/(ng/l)

27.7

31.6

1.50

26.3

1.42

15.6

17.8

1.25

14.8

1.17

8.76

10.0

1.00

8.33

0.92

4.92

5.62

0.75

4.67

0.67

2.77

3.16

0.50

2.63

0.42

0.867

1.00

0.00

—

0.0867

0.10

-1.00

—

0.0

—

techniques for bioassay studies. Generally, larvae or juveniles
must be kept in containers separate from adults because
adults may release additional larvae during the test and these
might become confused with original test individuals at a
similar development state. A convenient method of expos-
ing the juveniles was developed for the tests. Treated water
was diverted at a T-connector (as shown in Figure 5) through
tubing, and pumped by Buchler peristatic pumps into 10 ml
tissue culture dishes at a rate sufficient to maintain good mix-
ing but not so great as to wash out the juvenile clams.

Initially, each test was planned to be conducted for 96
hours, and the 96 hour LC-50 and 95 percent fudicial limits
calculated using Probit Analysis (Finney 1964). However, as
the study progressed it became evident that the high variabili-
ty in mortality made it necessary to extend some of the tests
in an attempt to obtain 100 percent mortality at the highest
concentration. Statistical analysis was accomplished by us-
ing the SAS program (SAS 1989) at the Murray State Univer-
sity Computing Center. Where only one intermediate concen-
tration occurred between 100- and 0-percent mortality, the
data could not be analyzed by the probit method. Tables 2

Table 2. Adult clam mortality during continuous exposure to TBTO
from September 20 to October 7, 1983.

Number of Dead Clams

Calculated
Dose (/*g/1)

Number
of Clams

24 hrs

48 hrs

72 hrs

96 hrs

192 hrs

31.6

10.0

3.2

1.0

0.1

0.0

Note: 8 day LC-50: 2.4 ^g/l; 8 day LC-99: 60.5 /tg/l.

MUSSALLI ETAL: ASIATIC CLAM CONTROL

and 3 present the typical results of the continuous flow tests.
Where data were sufficient for statistical analysis, the LC-50
and LC-99 are given at the bottom of the table.

The time required for adult clams to succumb is longer
than for juveniles, but low concentrations in the range for 5
to 20 hqI\ appear to be lethal. These data indicate a high tox-
icity of TBTO to Corhicula.

STATIC TEST. This test exposed clam larvae and
adults to water in contact with TBTF pellets under static con-
ditions, to simulate a situation in which a bag or canister of
pellets is placed in a tank, screenwell, crib-house, fire pro-
tection piping, etc, and filled with water. The objective of this
test was to estimate a diffusion rate of TBTF from pellets us-
ing information on toxicity determined in the continuous flow
testing.

Five 6-liter glass tanks were each filled with 4 liters
of lake water. In Tanks 1 and 2, 0.5 g of ECOPRO 1330-S
was placed in a section of PVC pipe with both ends opened
and covered with nylon mesh which allowed water to circulate
through the pipes while retaining the ECOPRO pellets. In
Tanks 3 and 4, one end of a “U” shaped PVC was open,
and the other end closed with a PVC cap with 0.5 g of

Table 3. Juvenile clam mortality during continuous exposure to TBTO
from September 29 through October 3, 1983.

Number of Dead Clams

Calculated
Dose (/tg/l)

Number
of Clams

24 hrs

48 hrs

72 hrs

96 hrs

31.6

10.0

3.2

1.0

0.1

0.0

Note: 4 day LC-50: 3.5 ^g/l; 4 day LC-99: 17.4 Mg/I.

95% fiducial limits: LC-50: 2.6-4.8; LC-99: 10.4-51.3

Table 4. Adult clam mortality during exposure to TBTO under static
conditions.

Number of Dead Clams

Days of Exposure

Tank Numbers

No. of Clams 1 2

1 10 0 0

2 10 0 0

3 10 0 0

4 10 0 0

Control 10 0 0

4 5 7 22 27 40

0 10 — — — —

0 0 0 0 0 0

0 0 0 0 2 2

0 0 0 0 0 0

Note: Tanks 1 and 2 with 0.5 g ECOPRO in open 5-cm long PVC
pipe

Tanks 3 and 4 with 0.5 g ECOPRO in 40 cm-long “U” shaped
PVC pipe closed at one end where the chemical was located.

ECOPRO 1330-S held in the capped end by a mesh nylon
screen. In this configuration the TBTO would have to diffuse
40 cm through stagnant water before contacting the water
in the tank holding the clams. Tank 5 was used as a control
with no TBTO pellets.

Ten adult clams were placed in each tank, and gentle
aeration was provided to maintain sufficient dissolved oxygen
and circulation. Juvenile clams were placed in 10 ml tissue
culture dishes which were placed in 50 ml Pyrex beakers and
carefully submerged in the tanks.

In the typical static test, 15 juveniles (1 to 2 weeks old)
and 10 adults were placed in each tank. Clam mortality oc-
curred only in Tanks 1 and 2. After 24 hours most of the
veliger larvae in Tanks 1 and 2 were dead and development
of the surviving larvae had stopped. One juvenile was dead
in Tank 2. In 48 hours all of the veligers were dead in both
tanks, and two juveniles were dead in Tank 2. After 96 hours
two juveniles were dead in Tank 1 , and six in Tank 2. Adult
clams were moribund and had an accumulation of thick
mucus around the edge of the shells. After 1 20 hours all the
adults and juveniles were dead in Tanks 1 and 2.

No deaths occurred in Tanks 3 and 4 even after 31
days. At that time the “U” stoped pipe in Tank 4 was tilted
emptying some of the accumulated TBTO into the tank. Five
days later two adult clams were dead. The undisturbed Tank
3 showed no signs of toxicity after 40 days (Table 4).

These static tests demonstrate that circulation is re-
quired to dissolve and distribute TBTO from ECOPRO
1330-S. Diffusion alone would not be sufficient to treat a stag-
nant body of water with TBTO from ECOPRO. Some water
movement is necessary.

ORGANOTIN TOXICANTS TEST SUMMARY. The
observed response of Corbicula to TBTO indicates certain
characteristics which make TBTO a favorable molluscicide.
In general, if a toxic substance is also irritating, such as
chlorine, the clams tend to close and avoid contact with the
substance for as long as possible. This results in a long treat-
ment period before the clams begin accumulating a toxic
dose. In the case of TBTO, however, at low concentrations
of 1 to 1 0 ngl\ the clams remained open and continued siphon-
ing water. Perhaps this allows an accumulation of a lethal
dose more rapidly. In fact, increasing the concentration above
30 Atg/I might not decrease the time for 100-percent mortali-
ty because clams would remain closed much of the time.

The lake water used in the tests contained many small
rotifers and protozoa. These were rapidly killed at concen-
trations estimated to be greater than 1 /*g/l. At concentra-
tions near 0.1 ^g/l the activities of the microfauna appeared
normal.

Tributyl tin oxide is highly toxic to Corbicula adults,
juveniles, and larvae. At low concentrations (1 to 10 ^g/l) it
apparently causes little or no avoidance response. This allows
the clams to accumulate a lethal dose rapidly.

The variability in the results, small sample size, and
apparent decline in release rate of TBTO from ECOPRO
1330-S during the tests precluded a precise determination
of the concentrations required to control Corbicula. If the
estimated release rates used to calculate the concentration

CORBICULA SYMPOSIUM

of TBTO were reasonably correct, then a continuous exposure
to 1 0 to 30 /ig/l for 1 to 2 weeks would kill 1 00 percent of adult
and juvenile Corbicula. The juveniles would succumb more
quickly than the adults. Larval development and transforma-
tion was stopped at concentrations below 10 ngl\.

TBTO from ECOPRO 1 330-S requires some water cir-
culation to be effective in controlling Corbicula. The circula-
tion speeds solution and distribution of TBTO. The diffusion
rate from ECOPRO pellets in static water was not sufficient
for lethal concentration of TBTO to travel 40 cm in 40 days.

The EPRI study demonstrated the extreme toxicity of
tributyl tin oxide to Corbicula, which warrants further investiga-
tions of the potential of TBTO as an agent for controlling Cor-
bicula in industrial and electric power facility water systems.
It was recommended to EPRI that a more extensive study
be conducted before or concurrently with an actual trial ap-
plication of ECOPRO 1 330-S in a power plant water system.

LITERATURE CITED

EPRI (Electric Power Research Institute). June 1984. Condenser
macrofouling control technologies. Topical Report. EPRI
CS-3550. Project 1689-9

Finney, D. J., 1964. Statistical Method in Biological Assay. Second
Edition, Griffin Press, London.

Mattice, J., June 1983. Freshwater macrofouling and control with
emphasis on Corbicula. Proceedings of EPRI Symposium on
Condenser Macrofouling Control Technologies, The State-of-
the-Art, Hyannis, Massachusetts.

Mussalli, Y. G., J. Williams, and J. Hockman, April 1981. Engineer-
ing evaluation of a dual-flow fine-mesh traveling screen. Pro-
ceedings of the Workshop on Advanced Intake Technology,
San Diego, California, pp. 169-183.

Mussalli, Y. G. and I. A. Diaz-Tous, April 1983. New developments
and evaluations of condenser fouling controls. Proceedings
of the American Power Conference.

Statistical Analysis System. 1979 SAS Institute, Inc. Raleigh, North
Carolina.

CORBICULA POPULATION MORTALITIES: FACTORS
INFLUENCING POPULATION CONTROL

JAMES B. SICKEL
HANCOCK BIOLOGICAL STATION
DEPARTMENT OF BIOLOGY
MURRAY STATE UNIVERSITY
MURRAY, KENTUCKY 42071, U.S.A.

ABSTRACT

Several factors potentially affecting population density in Corbicula fluminea are reviewed. These
include thermal and oxygen tolerance, silt loads, acidic waters, pollution, bacterial and viral infec-
tions, parasites, predators, interspecific competition, genetic changes, and intraspecific competition.
All of these factors may influence clam densities and population structure in certain cases. However,
it is hypothesized that overpopulation with its attendant strain on energy supplies and stress on in-
dividuals is the major cause of the commonly observed mortalities while genetic change through the
process of selection may be a major factor in establishing different life history characteristics in dif-
ferent populations.

Since the discovery of Corbicula fluminea (Muller) in
the United States in 1938 (Burch, 1944), there have been
numerous reports of its range extension and population in-
creases. Concomitantly there have been reports of high mor-
talities during which many dead clams were observed floating
ashore or being caught on screens of industrial water intakes
(Sinclair and Isom, 1963).

Several explanations are generally given to account
for the rapid population growth observed when Corbicula in-
vades a new region. These include high fecundity of
monoecious individuals, free-living planktonic larvae, absence
of naturally adapted predators, and the ability to exploit a
variety of habitats. As one would expect in an organism with
these attributes, Corbicula frequently becomes the dominant
benthic species shortly after invasion, increasing to densities
of thousands per m2 in only a few years (Gardner etal., 1976;
Eng, 1979; Sickel, 1979). This rapid population growth ob-
viously cannot continue indefinitely, and the subsequent
population may demonstrate varying degrees of success in
terms of density, size distribution and ability to survive as
native species adapt to its presence. The focus of this paper
is on the period of adaptation which follows the initial inva-
sion by several years or decades. A review of some of the
possible factors of population control is presented with an
emphasis on mass mortalities. Several case histories are
presented from the Tennessee and Cumberland Rivers and
their respective reservoirs Kentucky and Barkley Lakes.

ETIOLOGY OF CORBICULA MORTALITIES

Various factors have been proposed to account for the

regulation of Corbicula populations and the reported mass
mortalities at different times and in different situations. Pro-
posed factors include excessively high or low temperature,
low dissolved oxygen, silt, acidic waters, pollution, bacterial
or viral infections, parasites, predators, competition, and
genetic changes.

THERMAL AND OXYGEN TOLERANCE

Generally, high temperatures contribute to low dis-
solved oxygen levels in bottom waters of rivers and lakes
because of the lowered solubility of oxygen and the increased
respiratory demands. McMahon (1979) demonstrated that
Corbicula fluminea is poorly adapted to high temperature and
low oxygen. In his experiments, clams were severely stressed
above 25 - 30°C, and Mattice and Dye (1975) indicated a long-
term thermal tolerance of 34°C. They reported a lower
temperature limit of below 2°C. Both high temperature with
reduced oxygen and low temperature may be related to re-
cent mortalities in the Tennessee, Cumberland, and Ohio
Rivers. Horning and Keup (1964) reported a decline of Cor-
bicula from 290/m2 to from 0 - 10/m2 in 1963 in the Ohio River
at Cincinnati. They speculated that the cold winter during
which the Ohio River was frozen over for 7 days was the
cause of the die-off.

The winter of 1 977 in western Kentucky was unusual-
ly cold, and ice up to 7.5 cm thick covered the embayments
of Kentucky Lake. During the week prior to February 5, 1977,
the lake level was lowered about .6 m. This broke up much
of the ice and exposed the shoreline to freezing temperatures.
On 5 February at the Murray State University Hancock

American Malacological Bulletin, Special Edition No. 2(1986):89-94

CORBICULA SYMPOSIUM

Biological Station, many dead Corbicula bodies were ob-
served washing ashore. The tissue appeared fresh, and ex-
amination revealed completely intact bodies with no eggs or
larvae in the gills, no developing eggs in the gonads, and no
crystalline styles. Some gill water tubes were filled with
clumps of clay. It was assumed that these washed in as a
result of the churning action of the waves along shore rather
than entering the living clam and contributing to death. The
cause of death was probably exposure to freezing
temperatures, and the bodies were probably beaten from the
shells by wave action although no shells were observed.

On 4 April 1978 an extensive bed of Corbicula was
discovered in the channel at a depth of 20 m at Cumberland
River mile 67.0 in Barkley Lake, Kentucky. This bed consisted
of mostly large clams 35 - 45 mm in length at a density of
400/m2. About 10% of the clams were recently dead with
tissue just beginning to decay, and about 20% of the total
were old empty shells. Less than 10% were juveniles under
a year old. A continuous decline in numbers occurred with
extensive mortalities observed until November 1981 at which
time there were no more live clams (Table 1).

Table 1. Corbicula at Cumberland River mile 67.0, Barkley Lake,
Kentucky.

Date

Density Alive
(No./m2)

% Alive

4 April 1978

400

16 July 1978

333

4 March 1979

160-200

18 August 1979

172

6 March 1980

200

21 May 1980

25 October 1980

16 November 1981

On 18 August 1979 and again in July 1980, numerous
“floaters” were observed in Barkley Lake in the vicinity of
the bed at CRM 67.0. These were dead Corbicula 35 - 40 mm
in length. The shells were tightly closed with decomposing
tissue producing gas to make them buoyant. Some were so
recently dead that they had no unpleasant odor. Oxygen
measurements revealed D.O. levels of about 5 mg/I, but
temperatures were near 30°C. Sufficient monitoring was not
conducted to determine if D.O. levels may have been
significantly lower at night. For the past several years near
anoxic conditions have been recorded during summer months
at other locations in the deepest channels in both Barkley
and Kentucky Lakes. The combination of low D.O. and high
temperature may have contributed to the demise of the
population.

SILTATION

Bickel (1966) reported annual spring mortalities of all
age classes of Corbicula in the Ohio River at Louisville, Ken-

tucky, during the month of March. He attributed those die-
offs to suspended silt accompanying spring floods. Siltation
in some of the channel regions of Kentucky and Barkley Lakes
has reduced the habitat available to Corbicula as well as other
mollusks. Where currents have been eliminated by altered
hydrodynamics resulting from Kentucky and Barkley Dams,
an extremely fine, soft, and sometimes flocculent sediment
has accumulated which apparently does not support heavy
shelled mollusks.

ACIDIC WATERS

Kat (1982) reported a low resistance to shell dissolu-
tion by acidic waters of Corbicula compared to unionid
mussels. According to Kat, this difference results from the
presence of conchionlin layers in the unionid shells. Without
these layers in the Corbicula shell, once the periostracum
wears off, usually in the region of the umbo, the shell
dissolves rapidly in acidic waters. When a hole breaks through
the shell Kat assumed that death would result from the inva-
sion of microorganisms. Kat indicated that shell dissolution
might be a major source of mortality for Corbicula over about
3 years old in Mosquito Creek, Florida, which had a pH of
5.6. In the Tennessee and Cumberland Rivers the pH is near
neutral, and little shell dissolution has been observed.

BACTERIAL, VIRAL, AND PARASITIC INFECTIONS

On 9 May 1980, Tennessee Valley Authority biologists
investigated a reported mussel kill in the Tennessee River
in the vicinity of miles 407-413, Jackson County, Alabama
(TVA 1 6 May 1 980 memorandum from Robert T. Joyce to Gor-
don E. Hall). They found Corbicula bodies floating throughout
the 8.8 km section of the river. A fisherman indicated that
dead clams had been floating by for a week. Divers examined
mussel beds, and none of the native unionid mussels ap-
peared affected. The TVA biologists did not determine the
cause of death, but they speculated that it was a form of
bacterial or viral infection specific for Corbicula. They in-
dicated that periodic die-offs of Corbicula are reported once
or twice a year.

Al Scott (1980 personal communication) at the Auburn
University Fisheries Laboratory examined recently dead Cor-
bicula and concluded that 1) bacteria had not caused death,
2) histological sections indicated no parasites, and 3) elec-
tron microscopy revealed no viral inclusions.

In May and June 1980 many of the Corbicula from
Barkley Lake CRM 67.0 were infested with Chaetogaster lim-
naei, a naidid oligochaete (Sickel and Lyles, 1981).
Chaetogaster occurred in over 80% of the Corbicula with the
highest intensity in one clam being 1 67 worms. By 1 5 August
no more Chaetogaster could be found. The effect of
Chaetogaster on Corbicula is uncertain. Eng (1976) first
reported Chaetogaster limnaei in Corbicula from California and
indicated that infestation was seasonal with the highest
prevalence (87%) occurring from March through May. He
noted no evidence of parasitism and reported a low intensity
of only several worms per clam.

In the study by Sickel and Lyles (1981), the symptoms

S1CKEL: CORBICULA POPULATION MORTALITIES

of disease observed in the dying clams, both those with and
without Chaetogaster, included the following:

1. Loss of tissue mass, clams appeared emaciated.

2. Soft, watery tissue, lack of normal firmness, and
greater tissue transparency.

3. Thin, transparent mantle had secreted rough
nodules on inner shell surface.

Samples of dying Corbicula were collected in June
1980 from Barkley Lake (CRM 67.0) and sent to the Shellfish
Disease Laboratory, National Marine Fisheries Service, Ox-
ford, Maryland. The clams were examined by C. Austin Farley
and Fred Kern (1980 personal communication). They reported
much tissue necrosis and secondary bacterial decay. No in-
dication of viral infection was found, and they reported no
parasitism. The only unusual findings were large concretions
of amorphous material in the intestinal tubule and gonads.

POLLUTION

Evidence of eutrophication in Kentucky and Barkley
Lakes is seen as increasing areas of anaerobic water in the
deep channel. The sources of nutrients contributing to this
condition are diffuse. Contributing factors are probably hous-
ing developments, agricultural runoff, and municipal waste.
This eutrophication and the associated anaerobic water have
probably contributed to the demise of the Corbicula beds in
Barkley Lake.

Pollution in other regions of the world has been
detrimental to Corbicula. Chen (1976) reported that industrial
pollution has eliminated Corbicula fluminea from many lakes
and streams in Taiwan where the clam is used for food.

PREDATORS

The major predators of young and adult Corbicula are
fish. Britton and Murphy (1977) reported shells up to 5 mm
in length from a redear sunfish, Lepomis microlophus ; 3 mm
from a spotted sucker, Minytrema meianops ; and 3-5 mm from
a drum, Aplodinotus grunniens, from north Texas. Sickel et
al. (1981) reported that drum, blue catfish, ictalurus furcates,
and carp, Cyprinus carpio, in the Tennessee and Cumberland
Rivers consumed large numbers of Corbicula as well as young
mussels. Minckley etai. (1970) reported consumption of Cor-
bicula by smallmouth and largemouth buffalofishes, Ictiobus
bubalus and I. niger. In addition to drum, Dreier et al. (1981)
indicated that bluegill, Lepomis macrochirus, and channel cat-
fish, Ictalurus punctatus, regularly consumed Corbicula. They
presented evidence suggesting that fish predation was a ma-
jor cause of mortality and reduced clam density in Lake
Sangchris, Illinois. Areas of gravel seemed to provide some
protection for young clams from predation.

Although fish may play a minor role in regulating clam
populations, it is difficult to believe that fish could eliminate
a population of Corbicula. However, Corbicula may contribute
significantly to the nutrition of mollusk-eating fish.

Other predators include birds, raccoons, crayfish and
flatworms. Sanderson and Anderson (1 981) reported that the
gizzards of 36% of the hunter harvested waterfowl from Lake
Sangchris contained Corbicula shells. They listed 13 species

of ducks that ingested Corbicula. Taylor and Counts (1977)
reported finding Corbicula shell fragments in Raccoon, Pro-
cyon lotor, scats along with other evidence that raccoons had
been eating Corbicula on the banks of the Ohio River in West
Virginia. In laboratory experiments, Covich et al. (1981) found
that the crayfish, Procambarus clarkii, readily consumed Cor-
bicula under 6 mm in length and Cambarus bartonii ate clams
under 9 mm in length. Sinclair and Isom (1963) suggested
that the flatworm, Dugesia tigrina, potentially could be a
predator of Corbicula in Kentucky Lake. However, juvenile
Corbicula offered to D. tigrina from Kentucky Lake by the
author were never consumed. Another flatworm,
Macrostomum sp., from the Tennessee River was observed
to consume Corbicula larvae and juveniles under 0.25 mm
in length.

INTERSPECIFIC COMPETITION

Although Corbicula frequently produces dense popula-
tions in newly invaded habitats, there is conflicting evidence
regarding its ability to displace native unionid mussels with
which it competes for space and food. Gardner et al. (1976)
indicated a concurrent decline in unionid mussels during the
rapid growth phase of Corbicula in the Altamaha River,
Georgia, from 1971 through 1975. However, deteriorating
water quality and over-harvest of unionids probably con-
tributed to their decline. Sickel (1976) indicated that few Cor-
bicula were found among dense populations of adult
Unionidae in the Altamaha River, but that fewer juvenile
unionids were found on sandbars where Corbicula was abun-
dant. He concluded that Corbicula could not displace
established adult mussels but might interfere with their
reproductive success.

Fuller and Imlay (1976) reported abundant Corbicula
among dead mussel shells in lower reaches of the Waccamaw
River, South Carolina, which had been extensively altered
by human activities. Upstream in undisturbed reaches of the
river, they found a sparse population of Corbicula in areas
heavily populated with mussels. They suggested that “Cor-
bicula does not (and perhaps cannot) dominate indigenous
bivalves in nearly or quite natural habitats, at least in slowly
moving, soft-bottom Coastal Plain streams of the Atlantic
drainage.” Kraemer (1979) supported these observations in
her discourse discribing how Corbicula was able to exploit
habitats unfavorable to unionid mussels, but was not as suc-
cessful in undisturbed areas favorable to mussels. Particular-
ly, Kraemer stated that in chronically disturbed rivers such
as the Arkansas River in which management practices in-
cluding dredging and controlled discharge have reduced the
coarse sediments favored by indigenous mussels, Corbicula
has flourished while mussels declined. In the relatively un-
disturbed Buffalo River, Arkansas, Corbicula competes with
a healthy indigenous mussel community with only moderate
success. In this undisturbed river, Kraemer reported Corbicula
locally abundant only in fine sediments normally not occupied
by mussels.

In a conflicting report, Fuller and Richardson (1977)
stated that Corbicula could be an active “amensalistic com-

CORBICULA SYMPOSIUM

petitor” with mussels even in undisturbed areas. They
reported finding mussels being “uprooted” by Corbicula in
the Savannah River, Georgia and South Carolina. However,
although the Savannah River is not as extensively managed
as the Arkansas River, it is not an "undisturbed” river. Altera-
tions to the river caused by pollution, agricultural and other
human activities and somewhat controlled discharge may
have stressed the indigenous mussels thus providing Cor-
bicula a competitive advantage.

In the tailwaters of Kentucky Dam, Tennessee River,
Kentucky, there is an extensive unionid mussel community
in the gravel substrate, and Corbicula has coexisted with the
mussels since the early 1960’s (Williams, 1969). During
periods of heavy commerical harvesting of mussels, Corbicula
became the numerically dominant bivalve, constituting 99.4%
of the community (Williams, 1969). Even though commerical
harvesting by mussel brail continued to disrupt the mussel
community, Corbicula declined after the massive mortality of
1977 (Sickel and Heyn, 1980).

GENETIC CHANGES

Demonstrating genetic changes in populations is not
easy especially in a species which is reported to have little
or no genetic variability (Smith et ah, 1 979). However, Chitty
(1977) believes that natural selection and the resultant genetic
changes may play a major role in regulating numbers within
a population.

Differences in size and growth rates of Corbicula in
various populations have been considered to be evidence of
the effects of different environments. Growth rates and max-
imum size vary geographically. Eng (1 979) predicted a max-
imum length for Corbicula in the Delta-Mendota Canal, Cali-
fornia, of 37.75 mm. Morton (1977) predicted a maximum
length of 35 mm for Corbicula in Plover Cove, Hong Kong.
The maximum size that Sickel (1979) found in the Altamaha
River, Georgia, was 30 mm in 1 974 and 35 mm in 1 977. These
lengths are similar and might be expected to be characteristic
for the species. However, other populations show quite dif-
ferent shell lengths.

For about a decade prior to the summer of 1977, the
Tennessee River in Kentucky commonly produced large Cor-
bicula over 60 mm in length. They were so abundant and easy
to catch on mussel brails or with long handled rakes that a
prosperous fishbait industry developed. Commerical clam-
mers received 2<P per clam and wholesale bait dealers sold
them for 40 each. In August 1977, essentially all of the large
adult Corbicula in the Kentucky Dam tailwaters died from
unknown causes (Sickel and Heyn, 1980). Since that time
the population has been increasing in density, up to 1800/m2
in 1983, but no individuals have been found in the main river
greater than 12.3 mm in length (Table 2).

The evidence points to a shift in life history traits which
can only be explained by a genetic change. This shift has
resulted in a population with a high fecundity, rapid matura-
tion, and short life span similar to a newly invading population.

INTRASPECIFIC COMPETITION

Recent visual observations with the aid of SCUBA

Table 2. Density, mean shell length and range of Corbicula in the
Tennessee River downstream from Kentucky Dam.

Date

Number
of Grabs

Density

No./m2

Mean Shell
Length (mm)

Range

(mm)

10/18/78

4.55

2.4-10.3

11/10/78

3.58

1.5-8. 5

12/1/78

4.27

1. 7-8.7

6/15/79

5.45

2.9-12,3

6/9/83

1825

5.35

2.4-12.0

have suggested yet another cause, and in certain cases
perhaps the most plausible, for the occurrence of mass mor-
talities of Corbicula. A region of the Tennessee River in the
vicinity of river mile 13.8 was being surveyed for unionid
mussels. This region had been brailed extensively by com-
merical musselers for a number of years thereby reducing
the unionid density, disturbing the substratum, and creating
a habitat more favorable for invasion by Corbicula. The author
observed the entire substratum along a 50 m transect to con-
sist of living Corbicula among freshly dead Corbicula shells
and a few scattered unionids. The bottom literally was creep-
ing with Corbicula. One individual clam could not remain mo-
tionless for more than a few minutes because several
neighbors would move, jockeying for better position, and
climb over or dislodge it. This constant movement must re-
quire an unusually high energy output and, perhaps, con-
tribute to a high mortality.

The density in a I m2 sample was 1600 live Corbicula,
3000 dead shells of Corbicula, and 3 unionid mussels. Over
99% of the Corbicula were 3 years old with an average length
of 33.6 mm and a mean tissue dry weight of 656.8 mg/clam.
This calculates to a dry weight biomass of 10,500 kg/ha. The
greatest tissue dry weight biomass observed in the Altamaha
River at the peak of the Corbicula invasion was 314 kg/ha
(Sickel, 1979). Clearly, a biomass as large as that at Ten-
nessee River mile 13.8 cannot be sustained, and a high mor-
tality must ensue.

During July, August, and early September of 1984 and
1985, numerous dead bodies of Corbicula were seen floating
down the Tennessee River. The primary source of the dead
clams was the bed at mile 13.8. The occurrence of the mor-
tality during late summer might suggest a temperature rela-
tionship. However, the temperature never exceeded 29°C
which would not by itself stress the clams. A more complete
explanation for the timing of the die-off is that the intraspecific
competition is greatest during the summer when the high
temperature causes high metabolic rates and a greater
energy requirement, and the increased activity causes more
frequent disturbance to neighbors and a higher energy out-
put to remain in a competitive feeding position.

Another observation on this population suggests an
explanation for the nearly single age distribution that has been
seen numerous times. At the high density observed, and the
high mobility of the adult clams, juveniles would be hurried
quickly and repeatedly by the adults and the feces and

SICKEL: CORBICULA POPULATION MORTALITIES

pseudofeces of the adults. The sediment between the shells
consisted of these feces and pseudofeces. Since Corbicula
only live 3 to 5 years, within the next year or two the mortali-
ty in this population will peak, reducing the density, and young
clams will once again find a favorable habitat. These obser-
vations may explain the observed dynamics of Corbicula
populations in certain regions where little interspecific com-
petition exists and periodic overpopulation occurs.

CONCLUSIONS

Mortalities in which all age classes are affected are
probably the result of toxic substances, excessive cold, high
temperatures and low dissolved oxygen, or some other en-
vironmental insult. However, many of the mortalities that have
been reported are not a direct result of some environmental
change, but are a natural phenomenon of death in a dense
population in which a large, overpopulated cohort reaches
its age limit or exceeds the biomass capable of being sus-
tained by the environment. The die-off generally occurs dur-
ing the summer when metabolic rates are high and competi-
tion for space and food place excessive energy demands on
the individuals. The result is high mortality made obvious by
the many dead clam bodies floating past fishermen or drawn
into water intake structures. One would not question a report
of millions of dead adult mayflies. Similarly, it should be ac-
cepted that in dense Corbicula populations there will be
periodic massive die-offs. These should be predictable once
a population has been studied and its age structure and age
limit have been determined. Predicting the timing and inten-
sity of massive die-offs will allow suitable precautions to be
taken to avoid problems with large numbers of decompos-
ing clam bodies. However, one must not feel too complacent
with such data because a genetic change under intense
selection might alter the population age structure and life
history.

If generalizations can be made to other species, gains
will have been made toward understanding how populations
are regulated in nature. However, Corbicula may be an ex-
treme case in which mechanisms of selection are exag-
gerated because it is new to this continent and has yet to
attain a balance with other species.

Even though this investigation is incomplete, it is of
heuristic value if it stimulates further investigations or sug-
gests different approaches for studies which might lead to
a better understanding of Corbicula dynamics.

ACKNOWLEDGMENTS

Support for projects that made this report possible came from
the Committee on Institutional Studies and Research, Murray State
University, and a contract funded jointly by the Kentucky Department
of Fish and Wildlife Resources and the National Marine Fisheries
Service.

LITERATURE CITED

Bickel, D. 1966. Ecology of Corbicula maniiensis Philippi in the

Ohio River at Louisville, Kentucky. Sterkiana 23:19-24.

Britton, J. C. and C. E. Murphy, 1977. New records and ecological
notes for Corbicula maniiensis in Texas. Nautilus 91(1 ):20-23.

Burch, J. Q. 1944. Checklist of west American mollusks, Family Cor-
biculidae. Minutes of the Conchological Club of Southern
Califonia, 36:18.

Chen, T. P. 1976. Culture of the freshwater clam, Corbicula fluminea.
pp. 107-110. IN: Agricultural Practices in Taiwan, Page Bros.
Norwich, Ltd. 161 pp.

Chitty, D. 1977. Natural selection and the regulation of density in
cyclic and non-cyclic populations, pp. 27-32. IN: B. Stonehouse
and C. Perrins, eds., Evolutionary Ecology. University Park
Press, London. 310 pp.

Covich, A. P., L. L. Dye, and J. S. Mattice. 1981 . Crayfish predation
on Corbicula under laboratory conditions. American Midland
Naturalist. 105(1): 181 -188.

Dreier, H. and J. A. Tranquilli. 1981. Reproduction, growth, distribu-
tion, and abundance of Corbicula in an Illinois cooling lake.
Illinois Natural History Survey Bulletin, 32(4):378-393.

Eng, L. L. 1 976. A note on the occurrence of a symbiotic oligochaete,
Chaetogaster limnaei in the mantle cavity of the Asiatic clam,
Corbicula maniiensis. The Veliger 19(2):208.

Eng, L. L. 1 979. Population dynamics of the Asiatic clam, Corbicula
fluminea (Muller), in the concrete-lined Delta-Mendota Canal
of Central California. Proceedings First International Corbicula
Symposium. Texas Christian University Research Foundation,
pp. 40-68.

Fuller, S. L. H. and M. J. Imlay. 1976. Spatial competition between
Corbicula maniiensis (Philippi), the Chinese clam (Cor-
biculidae), and freshwater mussels (Unionidae) in the Wac-
camaw River Basin of the Carolinas (Mollusca: Bivalvia).
Bulletin Association Southeastern Biologists 23:60.

Fuller, S. L. H. and J. W. Richardson. 1977. Amensalistic competi-
tion between Corbicula maniiensis (Philippi), the Asiatic clam
(Corbiculidae), and freshwater mussels (Unionidae) in the
Savannah River of Georgia and South Carolina (Mollusca:
Bivalvia). Bulletin of the Association of Southeastern Biologists
24:52.

Gardner, J. A., Jr., W. R. Woodall, Jr., A. A. Statts, Jr., and J. F.
Napoli. 1976. The invasion of the Asiatic clam ( Corbicula
maniiensis Philippi) in the Altamaha River, Georgia. Nautilus
90(3) :1 17-125.

Horning, W. B. and L. Keup. 1964. Decline of the Asiatic clam in
Ohio River. Nautilus 78:29-30.

Kat, P. 1982. Shell dissolution as a significant cause of mortality for
Corbicula fluminea (Bivalvia: Corbiculidae) inhabiting acidic
waters. Malacological Review 15:129-134.

Kraemer, L. R. 1979. Corbicula{ Bivalvia: Sphaeriacea) vs. indigenous
mussels (Bivalvia: Unionacea) in U S. rivers: a hard case for
interspecific competition? American Zoologist 19:1085-1096.

Mattice, J. S. and L. L Dye. 1975. Thermal tolerance of the adult
Asiatic clam. IN: G. W. Esch and R. W. McFarlane, eds., Ther-
mal Ecology II. CONF-750425. National Technical Informa-
tion Service, Springfield, VA. pp. 130-135.

McMahon, R. F. 1979. Response to temperature and hypoxia in the
oxygen consumption of the introduced Asiatic freshwater clam
Corbicula fluminea (Muller). Comparative Biochemical
Physiology 63:383-388.

Minckley, W. L., J. E. Johnson, J. N. Rinne, and S. E. Willoughby.
1970. Foods of buffalofishes, genus Ictiobus, in central Arizona
reservoirs. Transactions of the American Fisheries Society
99:333-342.

Morton, B. 1977. The population dynamics of Corbicula fluminea
(Bivalvia: Corbiculacea) in Plover Cove Reservoir, Hong Kong.

CORBICULA SYMPOSIUM

Journal of Zoology of London 181:21-42.

Sanderson, G. C. and W. L. Anderson. 1981. Waterfowl studies at
Lake Sangchris, 1973-1977. Illinois Natural History Sun/ey
Bulletin 32(4):656-690.

Sickel, J. B. 1976. An ecological study of the Asiatic clam, Corbicula
manilensis (Philippi, 1841), in the Altamaha River, Georgia,
with emphasis on population dynamics, productivity and con-
trol methods. Ph.D. Dissertation, Emory University, Atlanta,
Georgia. 126 pp.

Sickel, J. B. 1979. Population dynamics of Corbicula in the Altamaha
River, Georgia. Proceedings First International Corbicula Sym-
posium. Texas Christian University Research Foundation, pp.
69-80.

Sickel, J. B. and M. W. Heyn. 1 980. Decline of the Asiatic clam, Cor-
bicula fluminea, in the lower Tennessee and Cumberland
Rivers. Bulletin of the American Malacological Union for
1980: 24-26.

Sickel, J. B., D. W. Johnson, G. T. Rice, M. W. Heyn and P. K.
Wellner. 1981. Asiatic clam and commerical fishery evalua-
tion. Kentucky Department of Fish and Wildlife Resources, Pro-

ject No. 2-344-R. 83 pp.

Sickel, J. B. and M. B. Lyles. 1981. Chaeiogasier lirnnaei (Oligochaeta:
Naididae) inhabiting the mantle cavity of the Asiatic clam, Cor-
bicula fluminea, in Barkley Lake, Kentucky. The Veliger
23(4):361-362.

Sinclair, R. M. and B. G. Isom. 1963. Further studies on the intro-
duced Asiatic clam Corbicula in Tennessee. Tennessee Pollu-
tion Control Board, Tennessee Department of Public Health,
Nashville, Tennessee. 76 pp.

Smith, M. H., J. Britton, P. Burke, R. K. Chesser, M. W. Smith and
J. Hagen 1979. Genetic variability in Corbicula, an invading
species. Proceedings First International Corbicula Symposium.
Texas Christian University Research Foundation, pp. 243-248.

Taylor, R. W. and C. L. Counts, III. 1977. The Asiatic clam, Corbicula
manilensis, as a food of the northern raccoon, Procyon lotor.
Nautilus 91(1 ):34.

Williams, J. C. 1969. Mussel fishery investigation, Tennessee, Ohio
and Green Rivers. Kentucky Dept. Fish and Wildlife Resources,
Proj. No. 4-19-4, Frankfort, Kentucky. 106 pp.

CONTROLLING CORBICULA (ASIATIC CLAMS) IN COMPLEX POWER PLANT

AND INDUSTRIAL WATER SYSTEMS

BILLY G. ISOM, CHARLES F. BOWMAN,

JOSEPH T. JOHNSON AND ELIZABETH B. RODGERS
TENNESSEE VALLEY AUTHORITY
MUSCLE SHOALS, ALABAMA 35660 U.S.A.

ABSTRACT

A plan for controlling Asiatic clams, Corbicula sp., was developed by an interdisciplinary team
within TVA during the 1970’s. This plan, which is now in place, has proven very effective in controll-
ing Asiatic clams over the past three years. Basis for the plan include knowledge of the life history
of Corbicula, including size of benthic veligers at spawning and timing of spawning events. A com-
bination of straining with a 1/32-inch (0.8 mm) media, chemical injection, and general “housekeep-
ing” has practically eliminated clam problems. Perhaps even this success could be enhanced and
made more economical with more research on optimization/minimization of chemical concentration
and selection of period(s) for application of controls.

Goss et al. (1979) reported preliminary experience of
the Tennessee Valley Authority (TVA) in control of Corbicula
in steam-electric generating plants up to that period, including
some power generating systems that were just beginning or
about to begin operation. Since 1979, TVA has had good ex-
perience controlling Corbicula, except where mechanical or
operational problems were experienced that interrupted
chlorination. The following “raw water” systems are the ones
usually fouled in fossil and/or nuclear steam-electric
generating plants: Condenser circulating water (CCW)
system; raw service water (RSW) system; essential raw cool-
ing water (ERCW) system; and the raw service water/high-
pressure fire protection (RSW-HPFP) system (Goss et at.,
1979).

The TVA is currently recommending the following
methods for controlling Asiatic clams in its nuclear plants:
All incoming water to the raw water systems should be strain-
ed. Straining is performed by automatic backwash type strain-
ing units located immediately upstream or downstream of the
main pumping units of the system (i.e., at the source).
Strainers have a 1/32-inch (0.8 mm) medium and are design-
ed for periodic or continuous backwashing.

Chlorine is the only chemical currently approved for
mollusc (macroinvertebrate) control (Federal Regulation
47[224]:52293, November 19, 1982) in steam-electric power
generating plants. In order to eliminate the safety considera-
tions necessary when using large storage tanks for gaseous
chlorine, TVA has elected to use sodium hypochlorite that
is generated onsite as needed.

METHODOLOGY

Chlorine is injected as close as is practical to the water

system inlet. Secondary water sources (such as jockey
pumps, normally open interconnections with other water
systems, etc.) are also chlorinated. If the incoming water has
already been chlorinated, no additional injection is necessary.
System design must be considered in conjunction with plant
chemical discharge limits in defining the exact location of
chemical injection.

Except as otherwise noted, the chlorine level
throughout the raw water systems and at the system
discharge is maintained at a total available chlorine residual
of 0.6 to 0.8 ppm during the entire clam spawing period. In
actual practice an 0.6 ppm residual is desirable since 0.8 is
the maximum concentration allowed by NPDES permit, and
you don’t want to exceed that concentration. The clam spaw-
ing period as defined here is that period of time when the
system inlet temperature normally exceeds 60°F (15.6°C). It
should be noted that this is a very conservative approach to
clam control, but is warranted since costs of failure to con-
trol Corbicula fouling properly are extremely high. One system
that is not normally chlorinated is the CCW system which ef-
fects the main cycle heat rejection. It should be remembered
that the raw water systems which are chlorinated constitute
only a small flow/volume in comparison with the CCW system
and since the discharge of these systems is mixed with CCW
before being discharged it can be shown by engineering
calculations that the most stringent standard, 0.01 mg/L or
less at the edge of the mixing zone, can be met without dif-
ficulty (Federal Regulation 47[224]:52293, November 19,
1982).

Provisions are made for periodic chlorine residual
sampling near the discharge of normally-flowing raw water
systems. Provisions are also made in both normally-flowing

American Malacological Bulletin, Special Edition No. 2(1986):95-98

CORBICULA SYMPOSIUM

Fig. 1. Simplified water use diagram for a typical nuclear plant.

ISOM ETAL.: CONTROLLING CORBICULA

and stagnant water systems to sample residual levels
periodically in undrained, normally-isolated system com-
ponents which may have experienced occasional use dur-
ing the clam spawning season. If an inadequate residual level
is found in any isolated area, that area is opened for a period
of time sufficient to allow replacement by chlorinated water.

Initially, sampling to ensure chlorine residual
maintenance should be conducted on a weekly basis for those
systems being chlorinated. A longer time interval may be
found adequate after samples have been analyzed. Design
provisions are made to flush isolated lines periodically in order
to maintain chlorine residual levels.

RESULTS AND DISCUSSION

Goss et ai. (1979) noted that initial fouling problems
at a nuclear plant probably resulted from flooding the system
at least two years prior to unit startup, which provided an en-
vironment for Corbicula growth. Therefore, TVA utilized the
following methods to prevent primary colonization and foul-
ing by Corbicula, even in temporary construction situations:
Unchlorinated water is not allowed to lie stagnant in any raw
water system at any time. Therefore, the systems are de-
signed to allow drainage after initial testing (if practical). If
a water system is going to be used regularly or if draining
is not feasible, temporary provisions are made to inject some
form of chlorine into the system in quantities sufficient to yield
the required residuals. Filling of the system is accompanied
by chlorine injection regardless of the inlet water temperature.

These measures during construction are applied to all
raw water systems except the CCW system. The design per-
mits temporary provisions for chlorinating the CCW system
during the initial filling period. The system should be
drained, cleaned, and refilled with chlorinated water prior to
plant startup.

The following are exceptions to the standard control
measures described above which are applicable to individual
systems.

Essential Raw Cooling Water System (ERCW) - Areas
of the system which are normally stagnant during normal
operation are provided with small mini-flow lines which pro-
vide for sufficient flow through that part of the system to main-
tain the required chlorine residual (Fig. 1).

High Pressure Fire Protection System (HPFP) - Provi-
sions should be made to flush the main supply line headers
periodically.

Raw Service Water System (RSW) - The RSW system
is continuously chlorinated during the entire clam spawning
period only if the RSW system is interconnected with the
HPFP system. Otherwise, chlorination for two 3-week periods,
once at the beginning of the clam spawning period and again
at the end of the clam spawning period, can be followed. Con-
tinuous chlorination during the entire clam spawning season
may be required later if clam problems develop with this
reduced chlorination schedule. These design provisions are
not intended to imply that an automatic timer for chemical
injections is recommended (Fig. 1).

Raw Cooling Water System (RCW) - The RCW system
is chlorinated for two 3-week periods a year as described for
RSW above. If the RCW system is supplied water by a
closed cycle CCW system, the two 3-week periods should
be established by the river temperature rather than the
temperature in the CCW system. Continuous chlorination dur-
ing the spawning period may be required if operating ex-
perience so dictates.

Condenser Circulating Water System (CCW) - The
CCW systems have complete drainage capability for clean-
ing of the system if an excessive population of clams develops
(Fig. 1). In addition, for closed cycle CCW systems, provisions
are made to strain the incoming makeup water and ensure
that it passes through the condenser prior to entering the cool-
ing tower basin. Preliminary test results have shown that
clams in the 1.5 mm size range cannot withstand the high
temperatures (43°C) found at main condenser discharges dur-
ing the summer months. Directing the strained makeup water
to the cooling tower discharge flume, rather than to the cool-
ing tower basin, will quickly subject the incoming clam lar-
vae to a lethal thermal stress (Isom, 1971 and 1976; Goss
et ai. 1979).

In response to NRC “Office of Inspection and Enforce-
ment Bulletin 81-03,” TVA conducted extensive inspections
of safety-related raw water systems beginning in 1981 . These
inspections afford a meaningful comparison between in-
cidents where these procedures were followed and where
they were not followed. The results of these inspections are
as follows:

Bellefonte Nuclear Plant is located at Tennessee River
Mile (TRM) 391 .5. Inspections of the ERCW system were con-
ducted between April 1 982 and January 1 983 with eight ma-
jor components and fourteen valves inspected. In addition,
during September and October 1982, the 36-inch supply
headers were drained, opened at intervals, and cleaned in
conjunction with the cement mortar lining of these headers
for control of corrosion. No evidence of Asiatic clams was
found in any of these inspections.

Inspections of the combined RSW-HPFP system were
conducted during November and December 1982. A total of
eight valves were inspected with no evidence of Asiatic clams.

Watts Bar Nuclear Plant is located at TRM 528. Inspec-
tion of the ERCW system were performed in June and July
1981 and three components of the system were examined.
In addition, during 1982 the supply headers were also
drained, opened at intervals, examined, and cleaned for ce-
ment mortar lining to prevent corrosion. No evidence of
Asiatic clams was found in any of these examinations.

In May 1981, approximately one-half of the fire pro-
tection nozzles at the station service transformer became
clogged with small rocks and Asiatic clam shells. It was be-
lieved that this debris was left from the construction phase.
A valve in the combined RSW-HPFP system was inspected
for Asiatic clams in July 1 981 , but none were found. However,
in December 1982 and again in January 1983, Asiatic clams
were discovered in two different fire hose rack valves. Seven
additional valves were inspected with no further indications
of Asiatic clam infestation. The frequency of HPFP system

CORBICULA SYMPOSIUM

flushing was increased from once each six months to once
each three months. This occurrence is significant in that it
demonstrates that Asiatic clams are, indeed, a threat at WBN
and in that it illustrates the need to sample and flush nor-
mally stagnant systems.

Sequoyah Nuclear Plant is located at TRM 485. In May
and July 1980, three major components of the ERWC system
were examined and headers were flushed and examined for
Asiatic clams. No Asiatic clams were discovered. Between
February and April 1 981 , sections of piping were removed and
inspected for corrosion products, no Asiatic clams were
discovered. Then, in March 1982, while conducting a
surveillance test of the system, a flow decrease from 100 to
31 percent of rated capacity to the containment spary heat
exchanger was noted. The pipe was opened upstream of the
manual inline strainer revealing approximately 15 gallons (57
L) of clam shells which were restricting the flow.

Under normal operating conditions, the 18-inch header
supplying the heat exchanger is stagnant except for a 1-inch
miniflow line around the heat exchanger. The miniflow line
was found to be clogged. Combined with the fact that the
ERCW was strained but not chlorinated the summer of 1 981 ,
except sporadically, this condition was conducive to clam
growth. Subsequently, steps were taken to ensure flow
through the miniflow line and chlorination by repairing the
hypochlorite injection systems. Continuous chlorination as
described above was practiced in 1982 with complete con-
trol of Corbicula.

In conclusion, a combination of straining and screen-
ing makeup water to 1 /32-inch (0.8 mm), chlorine injections
(0.6-0. 8 ppm) during the spawning season, and improved
“housekeeping” has practically eliminated Asiatic clam foul-
ing problems in TVA power plants. We are still conducting
research on optimization of chemical controls and selection
of period(s) for application of controls.

LITERATURE CITED

Isom, B. G. 1971. Evaluation and control of macroinvertebrate
nuisance organisms in freshwater industrial supply systems.
Abstract presented at the 19th Annual Meeting of the Midwest
Benthological Society, March 24-26, unpublished manuscript,
14 pp. Tennessee Valley Authority, Division of Environmen-
tal Research and Development, Muscle Shoals, AL 35660.
Isom, B. G. 1976. State-of-the-art in controlling Asiatic clams (Cor-
bicula manilensis Philippi) and other nuisance organisms at
power plants. Abstracts presented at the North American Ben-
thologica Society, 24th Annual Meeting, March 24-26, un-
published manuscript, 13 pp. Tennessee Valley Authority. En-
vironmental Biology Branch, Division of Environmental plan-
ning. Muscle Shoals, AL 35660.

Goss, L. B., J. M. Jackson, H. B. Flora, B. G. Isom, C. Gooch, S.
A. Murray, G. G. Burton and W. S. Bain. 1977. Control studies
on Corbicula for steam-electric generating plants, pp. 139-151.
IN: Proceedings, First International Corbicula Symposium, J. C.
Britton, ed., Texas Christian University.

POWER STATION ENTRAINMENT OF CORBICULA FLUMINEA
(MULLER) IN RELATION TO POPULATION DYNAMICS,
REPRODUCTIVE CYCLE AND BIOTIC AND ABIOTIC

VARIABLES1

CAROL J. WILLIAMS2 AND ROBERT F. MCMAHON3
DEPARTMENT OF BIOLOGY
BOX 19498

THE UNIVERSITY OF TEXAS AT ARLINGTON
ARLINGTON, TEXAS 76019, U.S.A.

ABSTRACT

The passive downstream dispersal of specimens of the introduced Asian clam, Corbicula fluminea,
was studied in the intake canal of a stream-electric power station on Lake Arlington, Texas, from 29
June 1981 through 6 December 1982. Downstream dispersal was monitored by a zooplankton net
and clam trap periodically placed in the intake canal and by collection of clams from traveling screens
in front of pump embayments. The population dynamics and reproductive cycle of the inlet canal C.
fluminea population were monitored along with water quality parameters and phytoplankton densities.

The inlet canal C. fluminea population had a biannual pattern of reproduction, marked by in-
cubation of developmental stages in the inner demibranchs from late March through early August leading
to an “early” generation and again from late August through early November leading to a “late” gener-
ation. The growth rates of all generations were maximal during the warm summer months. Maximum
densities were associated with recruitment of new generations. Mortality rates remained high throughout
life leaving the majority of individuals as immatures (shell length (SL) < 5 mm). There was a distinct
annual gonadal cycle in which periods of gonad maturation alternated with periods of larval incuba-
tion marked by gonad depletion.

Passive downstream dispersal on water currents was recorded in all size classes, but the ma-
jority were recently spawned juveniles (SL < 1 mm). Peak juvenile transport was associated with either
reproductive periods or low winter water temperatures. Downstream dispersal of immature (SL = 1-7
mm) and adult clams (SL > 7 mm) occurred just prior to the advent of larval incubation. No correl-
ations were found between passive downstream transport rates and phytoplankton densities or water
quality parameters, suggesting that dispersal in this species is closely associated with the reproduc-
tive cycle, with the single exception of juvenile downstream dispersal induced by low temperatures.

Since its introduction to North America from Asia in
the early 1900’s Corbicula fluminea has become a biofouling
pest species of major economic importance (McMahon,
1983). It has been reported to reduce flow in irrigation canals
(Prokopovich and Hebert, 1965; Prokopovich, 1969; Eng,
1979) and underground pipes (Fitch, 1953; Ingram, 1959).
C. fluminea also fouls the water lines and centrifugal pumps

’This research was supported by a grant from the Texas Electric Ser-
vice Company to R. F. McMahon.

2Present Address: Department of Microbiology, The University of
Texas Health Science Center, 5323 Harry Hines Boul., Dallas, Texas,
75325, U.S.A.

3Address reprint requests to R. F. McMahon.

of water treatment plants and causes unpleasant taste and
odor in drinking water (Ingram, 1959; Ray, 1962; Sinclair,
1964; Smith, et at., 1979).

However, the most serious aspect of biofouling by this
species is its ability to enter and foul the raw water systems
of industrial facilities, including steam-electric and nuclear
power plants where steam condensers, service water systems
and auxiliary water systems are occluded by accumulations
of living clams and dead shells (McMahon, 1977; Goss and
Cain, 1977; Goss, etal., 1979; Smith, etal., 1979). In nuclear
power plants both primary and backup systems can be foul-
ed simultaneously leading to major shut-downs for repairs
(Henager, etal., 1985). Incidents of C. fluminea biofouling at
a number of nuclear power stations have caused the United
States Nuclear Regulatory Commission (1981) to issue a

American Malacological Bulletin, Special Edition No. 2(1 986):99-1 1 1

100

CORBICULA SYMPOSIUM

Fig. 1. Lake Arlington, Tarrant County, Texas. Small insert shows locations of collecting site, zooplankton net, clam trap and traveling screens
in the intake inlet and intake canal of the Handley Steam-Electric Power Station.

WILLIAMS AND MCMAHON: CORBICULA ENTRAINMENT IN RAW WATER SYSTEMS 101

bulletin instructing all US nuclear power stations to inspect
their operations for fouling by C. fluminea.

The capacity of C. fluminea for biofouling appears to
depend on its ability to be passively transported on intake
water currents into these systems where they settle, grow and
reproduce (McMahon, 1977; Goss and Cain, 1977; Goss, et
al., 1979). The capacity for passive downstream dispersal is
characteristic of natural populations of C. fluminea (Heinshon,
1958; Morton, 1977a; Aldridge and McMahon, 1978; Eng,
1979; Sickel, 1979; McMahon and Williams 1986a), accoun-
ting, in part, for its highly invasive nature (McMahon, 1982,
1983). Downstream dispersal in C. fluminea has recently been
associated with its ability to produce mucus “draglines” from
the exhalent siphon that act to buoy individuals in the water
column (Prezant and Chalermwat, 1984).

Passive downstream dispersal appears to be unique-
ly characteristic of C. fluminea. It has not been reported to
commonly occur in either unioniid or pisidiid species. In spite
of its important implications to biofouling by this species, this
singular mode of dispersal has received little attention with
the exception of reports of seasonal variation in the densities
of juvenile clams in the water column (Eng, 1979; Sickel,
1979). This report describes an extensive field investigation
of the annual pattern of passive hydrological dispersal of
juvenile, subadult and adult individuals of C. fluminea in the
intake canal of a steam-electric power station in relation to
the population dynamics and reproductive cycle of the source
population, to phytoplankton density and to water quality
parameters. Analysis of the results indicate the biological
basis and adaptive significance of downstream dispersal in
this species. The results of this study also provide data useful
in the prediction of major episodes of impingement and
subsequent biofouling by C. fluminea of industrial and power
station raw water systems.

METHODS

A population of C. fluminea , composed entirely of the
“white” shell morph of Hillis and Patton (1982) was quan-
titatively sampled bimonthly from 29 June 1981 through 6
December 1982. This population occurred in the intake in-
let of the Handley Power Station of the Texas Electric Ser-
vice Company on the northwest shore of Lake Arlington in
Tarrant County, Texas. The power station drew water for its
condenser and service water systems from the inlet (0.7 km
long) through an 0.19 km long intake canal (Fig. 1). The
Handley Power station consisted of five gas-fired, steam-
electric generating units with a combined generating capacity
of 1471 MW and a maximum effluent discharge rate of 4.7
(106) / day1.

The inlet C. fluminea population was sampled near the
inlet’s north shore, 2 km from its opening into the lake pro-
per (Fig. 1). The substratum at this site was 20% gravel (par-
ticle diameter > 1 .8 mm), 77.4% sand (0.1-1 .8 mm) and 2.6%
silt (< 0.1 mm) by dry weight and had an organic content
of 3.5% of total dry weight (Williams, 1985).

The inlet population was quantitatively sampled with
an Ekman dredge (sampling area = 0.052 m2). Qualitative

samples were also taken with a heavy steel box dredge with
a 1 mm mesh collecting basket (for a description see Williams,
1985). This dredge was towed behind a jon boat with a 7.5
hp outboard motor. The box dredge bit deeply into the
substratum and was capable of removing even large unionid
bivalves such as Quadrula quadrula (Rafinesque) which bur-
row to much greater depths than C. fluminea. Clams were
removed from the sediments by passing dredged material
through a 1 mm mesh sieve.

Adult specimens of C. fluminea (shell length > 10 mm)
transported passively on intake water currents were collected
from two traveling screens located in front of pump em-
bayments from which condenser and service water was
drawn for the power station’s no. 3 generating unit (maximum
pumping rate = 1254 (106) / day1). The traveling screens
had a mesh size of 1 cm, and functioned to remove any
material potentially large enough to foul the steam condenser
tubes. These screens remain stationary in the water column
and are periodically rotated vertically past a high-powered
water jet which propels lodged material (including clams) in-
to a trough emptying into a large diameter outlet pipe that
opened into the discharge canal (for a complete description
of traveling screen operation see Bates, 1969). The travel-
ing screens were rotated and cleaned of lodged debris at ap-
proximately 10:00 and 17:00 hrs each day. A steel bucket
with a 1 cm x 2.5 cm diamond steel mesh bottom was placed
in the outlet pipe to collect all clams washed from the screen
at the 10:00 hr rotation on each collection date.

Juvenile clams carried in the water column were col-
lected with an 0.5 mm mesh zooplankton net (50 cm diameter
x 140 cm long). The zooplankton net was fitted with a cur-
rent meter to record the volume of water passing through it.
Clams carried along the bottom by currents were collected
by a clam trap constructed of aluminum bar stock with a fron-
tal opening 1 m wide by 0.5 m high. The clam trap was 1
m long, covered with 1 mm nylon mesh screen overlaid by
a 5.0 mm wire mesh screen and fitted with a removable plastic
jar at the back of the trap in which clams were collected (see,
Williams, 1985, for a more complete description of the clam
trap). Both the zooplankton net and clam trap were held in
intake currents at the head of the power station’s intake canal
by attaching them to a rope secured across the canal (Fig.
1). The clam trap was weighted to remain on the substratum
while the zooplankton net was maintained by a weighted lead
line 30 cm above the substratum. At each collection the clam
trap was submerged for 1-7 days and the zooplankton net
for 0.8-3. 3 hr.

At each collection ambient air temperature, surface
water temperature and ph were recorded at the inlet collec-
tion site. Conductivity and dissolved oxygen (Hellige, Model
342-DO) were determined for water samples taken just above
the substratum with a Kemmerer water sampler. Total water
hardness was determined by EDTA titration on selected col-
lection dates. Turbidity was measured by secchi disk. Daily
power station water pumping rates through the intake canal
were supplied by the Texas Electric Service Company.

At each collection phytoplankton cell density and
chlorophyll concentrations were determined from water

102

CORBICULA SYMPOSIUM

samples taken just above the substratum at the inlet collec-
ting site. Three 0.5 / water samples were fixed and stained
with 1 ml of Lugol’s iodine. Within two days of collection a
5 ml subsample of each water sample was placed in a 2.54
cm diameter settlement chamber and allowed to stand over-
night. The phytoplankton in 16 random 0.0625 mm2 fields
were counted at 400x under an inverted microscope (Olym-
pus model IMT). Phytoplankton counts were divided into three
major divisions: Chlorophyta; Cyanophyta; and Chrysophyta.

To determine chlorophyll contents, three 1 / water
samples taken just above the substratum were filtered
through a Whatman 934-AH glass microfiber filter with an ef-
fective retention size of 1.5 /xm. The chorophyll of algae re-
tained on the filter was extracted by grinding it in 90% acetone
(by volume with H20) and diluting the extract to a volume of
15 ml with 90% acetone. The absorption values of the ex-
tracted chlorophyll solution were then read at 630, 645, 665,
and 750 nm before and after addition of 0.05 ml of 10% HC /
to the cuvette sample (Parsons and Strickland, 1963).
Trichromatic calculations based on absorption values before
and after HC/ addition gave the Chlorophyll a,b,c and
phaeopigment concentrations (Strickland and Parsons, 1972).

The shells lengths (SL, the greatest anterior-posterior
dimension across the valves) of each individual in the col-
lections of the inlet population, from the traveling screens and
from selected clam trap and zooplankton samples were
measured to the nearest 0.1 mm. The SL of individuals >
5 mm was measured with a dial caliper while that of those
< 5 mm was measured with an ocular micrometer in a
binocular dissecting microscope.

For each inlet collection all individuals with an SL >
5 mm were divided into 0.1 mm size classes, and the fre-
quency of individuals in each size class expressed as a per-
centage fo the total sample size. These values were then plot-
ted as frequency histograms for each collection against
sampling date allowing direct visual separation of samples
into separate generations characterized by distinct size group-
ings of different modal shell lengths and ranges of SL. Mean
SL and a standard deviation were then computed for each
generation in each population sample (Aldridge and
McMahon, 1978). Individuals with an SL < 5 mm were pre-
sent in the collections of the inlet population throughout the
sampling period. As these individuals proved to be impossi-
ble to assign to specific generations they were excluded from
the analysis of population size distributions.

At each collection of the inlet C. fluminea population
the reproductive condition of 10-40 mature individuals (SL
> 7.7 mm) was determined by opening the valves and ex-
amining the gonad for the presense of mature eggs and the
inner demibranchs for incubated eggs, embryos, or juvenile
clams.

RESULTS

Mean monthly power station intake water flow rates
were 2545.8 (106) / day1. Flow rates were generally max-
imal in summer and minimal in late fall and early winter. Dai-

ly flow rates were much more variable than monthly averages
and ranged from a high of 4623.5 (106) / day1 on 29 June
1981 to a minimum of 760.9 (1 06) / day"1 on 16 November
1981 (Fig. 2A). Mean secchi depth was 0.9 m. indicating that
inlet water was relatively turbid. Secchi depth values
displayed a seasonal turbidity cycle in which turbity was
greatest (low secchi depth values) during the fall, winter and

1981 1982

Fig* 2. Abiotic parameters recorded at the inlet Corbicula fluminea
collecting site on Lake Arlington, Tarrant County, Texas, over the
duration of the collecting period. A. Intake water flow pumping rates
in 106 //day. Open circles connected by solid lines are daily water
flow rates, solid circles connected by dashed lines are monthly
average pumping rates. B. Secchi disk depth readings in m. C.
Dissolved oxygen concentrations in ppm 02 (mg 02/l). D. Conduc-
tivity in ftmho/cm2. E. Ambient pH values. F. Ambient air (solid circles
connected by dashed lines) and surface water temperatures (open
circles connected by solid lines) in °C.

WILLIAMS AND MCMAHON: CORBICULA ENTRAINMENT IN RAW WATER SYSTEMS 103

□ -•□CM a Phoeopigmant ratio

1981 1982

Fig. 3. Biotic parameters recorded at the inlet Corbicula fluminea
collecting site on Lake Arlington, Tarrant County, Texas. The horizon-
tal axis for both figures is months of the year over the collecting
period. A. Annual variation in phytoplankton cell density. The ver-
tical axis is phytoplankton cell density in cells (1 03) / ml for the total
phytoplankton (open circles), chlorphytes (solid circles), cyanophytes
(open triangles) and chrysophytes (solid triangles). B. Chlorophyll
concentration and chlorophyll a : phaeopigment concentration ratios.
The left vertical axis is chlorophyll concentration in ^g/l for total
chlorophyll (solid circles), chlorophyll a (open circles), chlorophyll
b (solid triangles) and chlorophyll c (open triangles). The right ver-
tical axis is the ratio of chlorophyll a concentration : phaeopigment
concentration (open squares).

spring, and least during the summer (high secchi depth
values) (Fig. 2B). Dissolved oxygen levels remained close to
air saturation values indicating that the inlet C. fluminea
population was not experiencing significant hypoxia (Fig. 2C).
Mean conductivity over the collection period was 270 /xmho
cm2 (Fig. 2D) and mean total hardness, 124.3 mg Ca /_1 which
are both characteristic of waters with moderately high levels

of dissolved minerals. Inlet waters were generally alkaline; am-
bient pH ranged from 6.36 to 8.72 and was less than 7.0 on-
ly during November 1981 and January 1982 (Fig. 2E). Am-
bient water temperature ranged from 6.7°C (7 December
1982) to 33.0°C (13 and 28 July 1981) over the collection
period. Ambient air temperature was generally within a few
°C of water temperature (Fig. 2F).

Phytoplankton cell densities were maximal in July and
August and minimal in November and December (Fig. 3A).
A maximum total phytoplankton cell density (21977 cells ml'1)
occurred on 18 August 1982, and a minimum (1605 cells ml'1)
on 8 December 1981 (Fig. 3A). Mean phytoplankton cell den-
sity over the entire study period was 8314 cells ml'1 (s.d. =
±5377, s.e. = ±1055, n = 27). Mean chlorophyte cell den-
sity was 3491 cells ml'1, mean cyanophyte cell density, 3856
cells mh1 and mean chrysophyte cell density, 2315 cells ml'1.
The cell densities of planktonic cyanophytes were greater
than those of chlorophytes from May through October. The
cell densities of chrysophtes were greater than that of
cyanophytes from 23 November 1981 through 8 February
1982. Only on 23 November and 8 December 1981 did the
ceil densities of chrysophytes exceed those of chlorophytes
(Fig. 3A).

Total phytoplankton chlorophyll concentration was
depressed in mid-summer and in late autumn through early
winter of both 1981 and 1982 (Fig. 3B). Maximum total
chlorophyll concentrations occurred in early autumn and
spring. This seasonal pattern in total chlorophyll concentra-
tion was reflected in the seasonal variation of the concentra-
tions chlorophylls a,b, and c (Fig. 3B). Mean chlorophyll con-
centrations over the collection period were: Chlorophyll a,
7.607 /xg/'1; chlorophyll b, 0.438 /xg/'1; and chlorophyll c,
3.744 n g/'1. With the exception of a single collection on 8
March 1982, phaeopigment concentration was always greater
than that of chlorophyll a (chlorophyll a phaeopigment con-
centration ratio < 1) (Fig. 3B), indicating that a high propor-
tion of the planktonic algal community was senescent
throughout most of the year (Bastardo, 1980).

The inlet C. fluminea population displayed a biannual
reproductive cycle characterized by the incubation of develop-
mental stages in the inner demibranchs and the release of
juvenile clams in the spring through mid-summer (“early”
reproductive period) and again from late summer through ear-
ly winter (“late” reproductive period) giving rise to distinct
early and late generations (Fig. 4). Corresponding to the bian-
nual reproductive cycle was a distinct variation in the percen-
tage of adults with ripe gonads containing mature eggs.
Gonads became depleted of mature eggs during incubation
and juvenile release periods and ripened (characterized by
higher percentages of individuals with gonads containing
mature eggs) during non-incubatory periods. Thus, the
highest percentages of individuals with ripe gonads were
recorded from November through March and August through
September, between the early (mid-March through mid-July)
and late reproductive periods (mid-August through late
November) when the lowest frequencies of individuals with
ripe gonads were recorded (Fig. 4). Such data indicate that
an annual cycle of gonad maturation and subsequent incuba-

MEAN SHELL LENGTH in mm

104

CORBICULA SYMPOSIUM

tion of developmental stages occurs in the inlet C. fluminea
population with the proportion of individuals with gonads con-
taining large numbers of mature eggs approaching peak
values just prior the onset of periods of ctenidial incubation
of developmental stages and juvenile release, during which the
number of adults in the population with ripe gonads marked-
ly declines (Fig. 4). Decline of the number individuals with
ripe gonads during incubatory periods suggests that incuba-
tion is associated with a suppression oogenesis in this
species. Indeed, diversion of energy stores into gamete pro-
duction and gonad maturation after the early reproductive
period may be the fundamental cause of the characteristic
mid-summer cessation of incubation and juvenile release
reported for the majority of C. fluminea populations
(McMahon, 1983).

The early and late reproductive periods give rise to cor-
responding “early” and “late” generations which appeared

as distinct new size classes in the inlet population. The late

reproductive period from 27 August through 12 November
1981, gave rise to a late generation (designated as L-81 in
Fig. 4) which first appeared with a mean SL of 6.0 mm on 26
October 1 981 . Similarly the early reproductive period exten-
ding from 24 March 1982 through 18 August 1982 gave rise
to an early generation (designated E-82 in Fig. 4) which first
appeared in the inlet C. fluminea population with a mean SL
of 5.2 mm on 31 May 1982. A second late reproductive period
occurred from 18 August 1982 through 30 November 1982
giving rise to a second late generation (designed as L-82 in
Fig. 4) first appearing in the collections with a mean SL of
5.5 mm on 12 October 1 982. Other distinct size classes pre-
sent in the collections of the inlet C. fluminea population
represented the early and late generations resulting from
reproductive periods that occurred prior to the initiation of
sampling and included the E-81, L-80, E-80, L-79 and corn-

1981 1982

Fig. 4. Generation shell lengths and reproductive condition in the inlet Corbicula fluminea population in Lake Arlington, Tarrant County, Texas.
The horizontal axis is months of the year over the collecting period. The left vertical axis is mean shell length (SL) in mm. The open circles
connected by solid lines represent the mean SL of individual generations in each sample. The vertical bars about each mean indicate
standard deviations. Individual generations are designated by the reproductive period which produced them [i.e., E-81 , a generation resulting
from the early (E) reproductive period in 1981 (81) or L-80, a generation resulting from the late reproductive period (L) in 1980 (80)]. L-78
+ E-79 indicates the mean SL of combined generations produced from the late reproductive period of 1978 and the early reproductive period
of 1979. The right vertical axis is the numbers of adult individuals in each sample incubating fertilized eggs, embryonic stages and juveniles
in the inner demibranchs (solid triangles connected by solid lines) or with mature eggs in ripened gonads (open triangles connected by
dashed lines) expressed as percentage of the total sample size.

SAMPLE

CLAMS / m CLAMS /

WILLIAMS AND MCMAHON: CORBICULA ENTRAINMENT IN RAW WATER SYSTEMS 105

1981 1982

Fig. 5. Seasonal variation in the density of the inlet Corbicula fluminea population in Lake Arlington, Tarrant County, Texas. The horizontal
axis for both figures is months of the year over the collection period and the vertical axis density in clams/m2. A. Density of the total population
(open circles) and of juvenile clams with shell lengths less than 5.0 mm (solid circles). B. Densities of specific generations. Densities of specific
generations are designated by the reproductive period that gave rise to that generation (for further explanation see caption to Fig. 4) as follows:
L-78 + E-79 (solid diamonds); L-79 (solid squares); E-80 (solid circles); L-79 + E-80 (solid triangles); L-80 (open squares); E-81 (open circles);
L-80 + E-81 (open triangles); L-81 (open diamonds); E-82 (solid crosses); and L-82 (open crosses). The dashed lines represent points at
which density estimates were intitiated for combined pairs of adjacent late and early generations.

bined L-78 + E-79 generation. These older generations were
all present in the initial collections (Fig. 4).

Maximum growth rates for all generations were sus-
tained from mid-May to early November when ambient water
temperatures were above 15°C (Figs. 2F and 4). During the
summer period of rapid growth the shell lengths of individuals
of the E-81 generation became indistinguishable from the
L-80 generation and, therefore, they were thereafter combin-
ed as a single size class into the L-80 + E-81 generation (Fig.
4). Similarly, the E-80 generation became indistinguishable
from the L-79 generation during rapid summer growth in 1981

and were, thereafter combined into a single L-79 + E-80
generation which eventually disappeared from population
samples in December 1982 (Fig. 4). A third grouped genera-
tion was present in the initial sample and was considered to
represent the combined L-78 + E-79 generations. This group
disappeared from the population samples in late July 1982. The
presence of four annual pairs of late and early generations
in the inlet C. fluminea population in both 1981 and 1982
strongly indicates that the maximum life-span of individuals
in this population was approximately 3-3 V2 years (Fig. 4).

The mean total density of the inlet C. fluminea popula-

106

CORBICULA SYMPOSIUM

tion was 168.2 clams nr2 (s.d. = ± 111.4, s.e. = ± 20.3, n
= 31) over the study period. Peaks in total density were
recorded on 28 July 1981 (320 clam rrr2), 26 October 1981
(426 clams rrr2), 3 May 1982 (320 clams nr2), and 29
September 1982 (274 clams rrr2) (Fig. 5A). These density
peaks were clearly associated with the recruitment of new
individuals (SL <5 mm) to the population from early and late
reproductive periods producing the E-81, L-81, E-82, and L-82
generations, respectively (Fig. 4). Another density peak oc-
curred on 24 March 1982 (525 clams nr2) (Fig. 5A). This peak
was not associated with juvenile recruitment and most likely
represented an extensive sampling error (ie. , collection at a
locally restricted site of exceptionally high juvenile clam den-
sity). The large decline in juvenile density following their recruit-
ment to the population (Fig. 5A) reflects the high annual mor-
tality rate of young clams in the inlet population.

After initial recruitment the densities of all generations
in the inlet C. fluminea population declined steadily throughout
the study period (Fig. 5B) suggesting that each generation
is subject to a relatively constant mortality rate throughout
its life span. During June and July 1982, there were four pairs
of late and early generations in the population samples (Fig.
4), including the L-78 + E-79, L-79 + E-80, L-80 + E-81 and
L-81 + E-82 generations. The densities of these pairs of
generations were utilized to estimate annual mortality rates
by expressing the difference in densiy between two adjacent
yearly generation pairs as a percentage of the density of the
most recent pair of early and late generations. These
estimates of annual mortality rates for the inlet C. fluminea
population were approximately 74% in the first year of life,
59% in the second and 93% in the third.

There was a distinct annual cycle of impingement of
adult specimens of C. fluminea onto the powers station’s
traveling screens. Maximum impingement rates occurred in
mid-April 1982 (127 clams day1), and in late July 1982 (105
clams day1) (Fig 6A), just prior to the onset of the early and
late reproductive and incubatory periods, respectively (Fig.
4). There was no significant impingement of adult clams on
the traveling screens during other times of the year (Fig. 6A).

The annual cycle of adult impingement on the travel-
ing screens was reflected in the cycle of retention of smaller
individuals (SL = 1-7 mm) in the clam trap. Clams were taken
in the trap at low levels throughout the study period, indicating
that some passive downstream transport was always occur-
ing in these size classes. However, a distinct peak of max-
imum retention of individuals by the clam trap (161.6 clams
day1; mean SL of trapped clams = 2.6 mm, s.d. = ± 0.75,
n = 5857) occurred in early April 1982 (Fig. 6B). This peak
of retention of subadult clams in the clam trap correspond-
ed directly to the spring peak of adult clam impingement on
the traveling screens (Fig. 6A) just prior to the initiation of
the early reproductive and incubatory period (Fig. 4).

The density of juvenile C. fluminea (SL < 2 mm)
passively suspended in the water column as estimated from
numbers taken in the zooplankton net was highly correlated
with juvenile release by adult clams during reproductive
periods. Peak juvenile densities in the water column were 760
clams 1 00 rrr3 on 21 May 1 981 and 21 1 clams 1 00 nr3 on

Fig. 6. Seasonal variation in the rates of passive downstream disper-
sal on water currents by juvenile, subadult and adult specimens of
Corbicula fluminea in the intake canal of a power station on Lake
Arlington, Tarrant County, Texas, as estimated from adult impinge-
ment on traveling screens, retention of subadults in a clam trap, and
the density of juveniles suspended in the water column estimated
by retention in a zooplankton net. The horizontal axis for all figures
is months of the year over the collection period. A. Rate of impinge-
ment of adult individuals (shell length (SL) > 10 mm) on two traveling
screens in front of the intake embayments of the no. 3 generating
unit in clams impinging the traveling screens per day (open circles).
B. The rate of retention of subadult clams (SL = 1-7 mm) in a clam
trap on the substratum of the intake canal in clams per day (open
circles). C. The density of juvenile clams (SL < 2.0 mm) suspended
in the water column and the entrainment rate of juvenile clams into
the power station’s raw water systems as estimated from the reten-
tion of clams in a zooplankton net held 30 cm above the substratum
of the intake canal. The left vertical axis is a logarithmic scale of the
numbers of juvenile clams entrained through the power plant’s raw
water systems per day (open circles connected by solid lines). The
right vertical axis is the density of juvenile clams suspended in in-
take canal waters in clams / 100 m3 (solid circles connected by dash-
ed lines).

28 June 1982 during the early 1981 and 1982 reproductive
periods, respectively (Fig. 6C). During the 1981

WILLIAMS AND MCMAHON: CORBICULA ENTRAINMENT IN RAW WATER SYSTEMS 107

and 1982 late reproductive periods peak juvenile densities
in the water column were 77 clams 100 nr3 on 1 1 November
1981 and 332.8 clams 100 nr 3 on 27 October 1982, respec-
tively (Fig. 6C). Surprisingly, maximum suspension of juvenile
clams in the water column (9154.4 clams 100 rrr3) occurred
during a period of prolonged low ambient water temperature
from January through February 1982 (mean water
temperature = 10.5°C) (Figs. 2F and 6C).

Values of juvenile density in the water column
multiplied by daily intake water flow rates yielded juvenile
clam entrainment rates through the Handley Power Station’s
raw water systems. Entraiment rates of juveniles on intake
waters were relatively high throughout the study period (Fig.
6C). Peak levels of juvenile entrainment were clearly
associated with the early and late reproductive seasons when
large numbers of juveniles were suspended in the water col-
umn (Fig. 6C). However, maximum entrainment rates (2.5
(1 08) clams day1) occurred on 22 February (Fig. 6C) and were
associated with maximal densities of juveniles in the water
column during a period of low winter water temperatures (see
above).

Least squares linear regression analysis was utilized
to determine if any direct relationships exist between the
various abiotic and biotic parameters recorded during the
study (ie., water temperature, pH, conductivity, dissolved ox-
ygen, turbidity, water flow rate, total algal cell concentration,
total chlorophyll content, and chlorophyll a: phaeopigment
ratio) and the rate of passive downstream transport of C.
fluminea as represented separately by impingement of adults
on the traveling screens, retention of subadults in the clam
trap and suspension of juveniles in the water column
(measured by retention in the zooplankton net). No signifi-
cant linear relationships could be found between any of these
parameters and passive downstream dispersal of adult,
subadult and juvenile clams (P > 0.05). However, there ap-
peared to be a tendency for adult clam impingement rates
onto traveling screens to increase with decreasing chlorophyll
a: phaeopigment ratio representative of increasingly senes-
cent phytoplankton populations (P < 0.1, r = -0.45, n = 19),
and for densities of juvenile clams in the water column to in-
crease with declining oxygen concentration (P < 0.1, r =
-0.357, n = 30).

DISCUSSION

The Lake Arlington inlet C. fluminea population had
a life cycle characterized by a biannual pattern of reproduction.
An “early” period of egg and developmental stage incubation
and subsequent juvenile release extending from mid-spring
to mid-summer was separated from a “late” reproductive
period extending from late summer to early winter by a non-
reproductive, non-incubatory period in mid-summer. Early
generations from the early reproductive periods appeared in
the population in May or June, grew to a mean SL ranging
from 11 mm (E-82) to 16.5 mm (E-81) by the following
December and, subsequently achieved a mean SL of 18.8
mm (E-81) to 18.9 mm (E-80) by the following June after one
year of growth (Fig. 4). The late generations arising from late

reproductive periods grew little through the winter and subse-
quently initiated rapid growth the following spring to reach
a mean SL ranging from 19 mm (L-80) to 21 mm (L-81) after
the first year of life (Fig. 4). Two year old clams in the Lake
Arlington inlet population reached a mean SL of 35.7 mm
(L-79 + E-80 generations) and in the third and terminal year
of life a mean SL of 46.0 mm was attained (L-78 + E-79 genera-
tions) (Fig. 4).

A very similar pattern of life-cycle has been reported
for a natural iotic population of C. fluminea in the Clear Fork
of the Trinity River in north central Texas (McMahon and
Williams, 1986b). Like the Lake Arlington population this
population had two generations per year and a maximum life
span of three years. The early generation reached a mean
SL of 22.5 to 23.4 mm and the late generation, 20.6 to 24.3
mm in the first year of life. A mean SL of 35.6 mm was achiev-
ed after two years and individuals at the end of the third and
terminal year of life reached a mean SL of 41.0 mm. Other
detailed studies of the reproductive and life-cycles of C.
fluminea populations also report two annual reproductive
periods, attenuated life spans of 1.5 to 4 years and shell
growth rates ranging from 16 mm to 33 mm in the first year
of life in Texas (O’Kane, 1976; Aldridge and McMahon, 1978;
McMahon and Williams, 1986b), central California (Heinsohn,
1958; Eng, 1979), Asia (Morton, 1977a) and Africa (Leveque,
1973). Thus, a biannual reproductive pattern, high growth
rates and relatively short life spans appear to be characteristic
of C. fluminea throughout its world-wide range. The minor dif-
ference in life-cycle, growth rates and life spans of
geographically separated populations of C. fluminea may be
attributable to environmentally induced ecophentypic vari-
ation. Certainly, as this species has been reported to have
relatively variable generation growth rates and life spans from
year to year within a single population (McMahon and
Williams, 1986b), environmentally induced geographic varia-
tion in growth rates and life spans is not unexpected (see
McMahon, 1983 and McMahon and Williams, 1986b for a
review of growth rates in C. fluminea).

Surprisingly, the growth rates and life spans of the in-
let C. fluminea population were quite different from those
reported in an earlier study of another population of this
species in the same lake carried out from late 1974 through
the end of 1975 (Adridge and McMahon, 1978). In this earlier
study both the early and late generations had shorter life
spans (1 .5-2 years) and much higher shell growth rates (mean
SL = 30-33 mm in the first year of life). A possible explana-
tion for these temporal differences in life cycle and growth rate
may lie in a general decline in the phytoplankton production
of Lake Arlington. Mean annual phytoplankton densities in
Lake Arlington during 1971 were 18.3 (103) cells ml-1 (Carr,
1973). In 1979 they were 38.4 (103) cells ml'1 with a mean
chlorophyll a concentration of 13.6 ng /- L (Peeler, 1980). Dur-
ing the present study (1981-1982) mean algal density declined
to 8.3 (103) cells ml'1 and mean chlorophyll a concentration to
7.6 ^g /- L These levels represent a 54% to 78% decline
in mean phytoplankton density and a 44% decline in mean
chlorophyll a concentration between the period during which
Aldridge and McMahon (1 978) completed their observations

108

CORBICULA SYMPOSIUM

on C. fluminea in Lake Arlington and the sampling period of
the present investigation. As phytoplankton appears to be a
major food source for C. fluminea (Foe and Knight, 1985,
1986; Lauritsen, 1986), as it is for most lamellibranch bivalves
(Owen, 1966), it is not unexpected that such a major decline
in phytoplankton productivity would be associated with
decreased growth rates and correspondingly increased life
spans in this species. Certainly, it appears that variation in
both physical factors as temperature and catastrophic climatic
events (Horning and Keup, 1964; Bickel, 1966; White and
White, 1977; White, 1979; Cherry, et al . , 1980; Dreier and
Tranquilli, 1981; McMahon and Williams, 1986b) and biotic
factors such as the level of primary productivity may have
significant impacts on the population dynamics of this
species. Such environmentally induced ecophenotypic varia-
tions appear to account, in great part, for the geographic
variations in growth rates, life spans and maximum shell
lengths reported for C. fluminea in North America (McMahon,
1983).

The inlet C. fluminea population displayed a distinct
seasonal alternation between gonad maturation and incuba-
tion of fertilized eggs, developmental stages and juvenile
clams in the interlamellar spaces of the inner demibranchs.
Gonads were observed to become depleted of mature eggs
during incubatory periods and to produce mature eggs and
ripen during non-incubatory periods (Fig. 4). In contrast, Eng
(1979) and Kraemer, et al. (1986) have reported that mature
eggs occurred throughout the year in the gonads of Califor-
nia and Arkansas populations of C. fluminea, while sper-
matogenesis occurred only during reproductive periods. This
pattern was distinctly different from the Lake Arlington C.
fluminea population in which the gonads of the majority of
specimens became degenerate and appeared to be depleted
of sperm and eggs during the latter portions of incubatory
and juvenile release periods. A similar decline in gonad con-
dition has been reported for an Asian lentic population of C.
fluminea (Morton, 1977a). This inhibition of gametogenesis
during incubatory periods may account for the marked mid-
summer cessation of incubation and juvenile release reported
for the vast majority of C. fluminea populations (Heinsohn,
1958; Morton, 1977a; Aldridge and McMahon, 1978; Eng,
1979; Sickel, 1979; Dreier and Tranquilli, 1981 ; McMahon and
Williams, 1986b). It is possible that incubation places con-
siderable metabolic demands on adult clams, effectively
diverting metabolites from incorporation into developing
gametes. Such metabolic demands may be associated with
the reduction of the filtering efficiency of the inner demi-
branches when they are distended with developing embryos,
effectively reducing the assimilated energy available for
gamate production. Alternately, Morton (1977b, 1982) has
suggested that incubated embryonic stages may recieve
nourishment from hypertrophied epithelial cells lining the in-
terlameller spaces of the inner demibranch of adult C.
fluminea. Such diversion of metabolites to incubated
developmental stages could place a considerable pressure
on the energy reserves of adult individuals preventing their
utilization for gamete production during incubatory periods.

While the ability of C. fluminea to foul industrial water

systems has been well documented (see McMahon, 1983,
for a review of biofouling problems with C. fluminea), few at-
tempts have been made to study the relationships between
this species’ biology and its nature as a biofouling pest
species. Such studies are of great importance to the even-
tual development of rational and effective biofouling control
procedures for C. fluminea. Two major biofouling problems oc-
cur with C. fluminea in the raw water systems of steam-
electric and nuclear power stations. The first involves the
passive transport of adults into turbine steam condensers
where they lodge at slight constrictions in the condenser tube
walls (McMahon, 1977). The second problem concerns the
passive hydrological transport of juvenile and subadult clams
on intake water currents into service and auxiliary raw water
systems utilized for cooling and other purposes. Transported
clams settle in low flow areas of these systems to grow,
reproduce, accumulate and eventually occlude water flow to
levels that seriously compromise system operations (Goss and
Cain, 1977; Goss, eta., 1979; Smith, et al., 1979; Cherry, et
al., 1980; Henager, et al., 1985).

Presently, the only effective control measures for ser-
vice and auxiliary water systems involve periodic chlorina-
tion to eliminate impinging juveniles and subadults (Sinclair
and Isom, 1963; Goss and Cain, 1977; Goss, et al., 1979;
Smith, et al., 1979; Mattice, et al., 1982). Several reports have
suggested that chlorination to control juvenile impingement
of service water systems need ony be applied during high
risk periods of juvenile impingement associated with the early
and late reproductive periods whose onset and duration could
be determined by monitoring the reproductive condition of
adult clams in the source population and intake waters for the
presence of newly released juveniles (Ingram, 1959; Cherry,
et al., 1980; Smith, et al., 1979). The results of this study in-
dicate that while high levels of juvenile transport on intake
waters are certainly associated with reproductive periods,
significantly high levels of entrainment also occurred in non-
reproductive periods, particulay during periods of low winter
water temperatures (< 10°C) when the density of juveniles
in the water column was 12-50 times greater than at any other
time of the year (Fig. 6C). The reasons for high levels of
juvenile suspension in the water column during low water
temperatures are presently unknown. It may be associated
with a low temperature inhibition of juvenile byssal thread for-
mation (see Kraemer, 1979, for a description of the juvenile
byssus in C. fluminea) or a reduced capacity for burrowing,
either of which would greatly increase the susceptibility of
juveniles to passively enter the water column. The level of
entrainment of juvenile C. fluminea through the raw water
systems of the Handley Power Station was quite remarkable,
often surpassing 107 individuals day1 (Fig. 6C). These high
levels of downstream dispersal allowed the thermal effluent
discharge canal of the power station to be recolonized at rates
of 352 clam nr2 day1 and 522 clams nr2 day1 in the falls of
1981 and 1982, respectively, after the discharge canal popula-
tion had been completely eliminated during the previous sum-
mers by lethally high ambient water temperatures (McMahon
and Williams, 1986a).

As high levels of juvenile entrainment through power

WILLIAMS AND MCMAHON: CORBICULA ENTRAINMENT IN RAW WATER SYSTEMS 109

station raw water systems are not restricted to reproductive
periods (Fig. 6C) and as immature clams (SL = 1-7 mm) are
carried downstream continually on water currents as reflected
by their retention in the clam trap throughout the year (Fig.
6B), chlorination procedures to control biofouling by C.
fluminea will almost certainly have to be applied continuously
throughout the year to be effective. If chlorination is not con-
tinuous small individuals suspended in the water column dur-
ing non-reproductive periods may settle in service and aux-
iliary water systems and rapidly grow to chlorination resis-
tant sizes (Mattice, 1979; Mattice, et al., 1982) especially dur-
ing the winter months when growth would be stimulated by
the warmer water temperatures of service water systems
(McMahon and Williams, 1986b).

Immature specimens of C. fluminea ranging in SL from
1 to 7 mm were retained in the clam trap which rested directly
on the substratum. In contrast, only juvenile clams with a max-
imum SL of 2.0 mm were taken in the zooplankton net which
was suspended 30 cm off the substratum. As no clams with
an SL greater than 2.0 mm were taken in the zooplankton net,
individuals with greater shell lengths do not appear to become
suspended in the water column. Instead, they must be mainly
transported downstream by being carried over the substratum
surface by water currents (“rolling”). A recent study has in-
icated that specimens of C. fluminea with an SL much greater
than 2.0 mm can enter the water column by producing a
“dragline” composed of mucus threads from the exhalent
siphon under laboratory conditions (Prezant and Chalerm-
wat, 1984). However, no individuals with an SL greater than
2.0 mm were trapped in the water column of the intake canal
even though many thousands of individuals were taken
throughout the course of the study. Therefore, mucus
draglines do not appear to function to suspend larger clams
in the water column. Rather, they appear to be involved with
transport of larger individuals over the surface of the
substratum, even in the very high water current velocities of
the intake canal. Certainly, as the vast majority of individuals
of C. fluminea dispersed on water currents are juveniles, adult
hydrological transport appears to be of little real significance
to the downstream dispersal of this species.

There was an apparent tendency for the number of
juvenile C. fluminea suspended in the water column to in-
crease directly with decrease in dissolved oxygen concen-
tration (P < 0.1). The O2 consumption of adult C. fluminea
is severely inhibited by even relatively minor levels of hypoxia
(McMahon, 1979). If the metobolic rates of juvenile individuals
are similarly depressed by hypoxic conditions, they may
become stressed and unable to burrow and/or maintain a
byssal connection to the substratum making them much more
susceptible to passive hydrological transport.

Subadult and adult specimens of C. fluminea were
maximally retained in the clam trap and traveling screens,
respectively, in mid-April and late July just prior to the onset
of the early and late reproductive periods (Fig. 6A and B).
Adult individuals taken from the traveling screens appeared
to be too dense to be carried in the water column. Instead,
they appeared to be carried by intake water currents
downstream over the substratum surface (unpublished obser-

vations). While there was an apparent tendency (P < 0.1)
for adult impingement on the traveling screens to increase
with increasing senescence of the phytoplankton communi-
ty (associated with a decrease in food quality marked by
chlorophyll a: phaeopigment ratios < 1), the pronounced in-
creases in the numbers of clams impinging the travel screens
and clam trap just prior to incubatory periods suggests the
majority of this phenomenon is associated with the reproduc-
tive cycle. Certainly, it is tempting to speculate that the
passive downstream transport of gravid individuals may repre-
sent a sort of pre-reproductive dispersal which would be of
obvious adaptive significance to an invasive species such as
C. fluminea which inhabits unstable aquatic environments
(McMahon, 1983; McMahon and Williams, 1986a and b).
However, this study and that of McMahon and Williams
(1986a) indicate that adult downstream dispersal is of little
consequence compared to the massive dispersal of juveniles
in this species. Recently, it has been shown that adult clams
impinging the traveling screens of the Handley Power Sta-
tion prior to reproduction have reduced tissue weights,
decreased tissue total organic content: nitrogen content ratios
and decreased molar ratios of oxygen consumed: nitrogen
excreted compared to individuals in the inlet source popula-
tion, which indicated that dispersing adults were showing
symptoms of reduced energy assimilation and starvation
(Williams, 1985; Williams and McMahon, 1985). As such,
dispersing adults appear to be in poor reproductive con-
dition. Therefore, leaving the substratum to be carried
downstream on water currents to microhabitats more favorable
to the acquisition of food resources to support gamate pro-
duction and embryo incubation may be a highly adaptive
behavior in C. fluminea. Such a hypothesis is supported by
the observation that adults only disperse in high numbers just
prior to the onset of reproductive periods (Fig. 6C).

Power plant intake pump embayments behind travel-
ing screens may harbor very dense populations of adult C.
fluminea (McMahon, 1977; Dreier and Tranquilli, 1981;
Harvey, 1981; Smithson, 1981). These populations appear
to be the main source of adults impinging and fouling tur-
bine steam condensers (McMahon, 1977; Smithson, 1981).
If adults in these embayments are subject to the same
seasonal patterns of passive downstream dispersal as those
recorded for the Lake Arlington inlet C. fluminea population,
then major episodes of steam condenser biofouling by C.
fluminea will be most likely to occur just prior to the early and
late reproductive periods, in early spring and mid-summer,
respectively. In this regard, steam condenser biofouling con-
trol procedures involving either periodic removal (Goss and
Cain, 1977; Goss, et al., 1979; Smith, et al., 1979; Harvey,
1981) or chemical treatment (Smithson, 1981) of embayment
populations would be most effective if they were applied just
prior to these major pre-reproductive episodes of passive
adult downstream dispersal.

ACKNOWLEDGEMENTS

The authors wish to thank David Bible, Juan Ibarra, Colette

110

CORBICULA SYMPOSIUM

O'Bryne-McMahon, Ralph Williams, Joseph Gilly, and Wesley Truitt
for assistance with the field collections. Colleen C. Bronstad provid-
ed technical assistance with the laboratory. Dr. Craig D. Sandgren
assisted with phytoplankton cell counts and chorophyll concen-
tration determinations. Special appreciation is extended to Mark
Spiegal and William Hoerster of the Texas Electric Service Company
for providing technical assistance and records of discharge volumes
of the Handley Power Station. James Schmulen, Environmental
Scientist for the Texas Electric Service Company provided support
and advice over the course of the study. This research was supported
by a grant from the Texas Electric Service Company to R. F.
McMahon.

LITERATURE CITED

Aldridge, D. W. and R. F. McMahon. 1978. Growth, fecundity, and
bioenergetics in a natual population of the freshwater clam,
Corbicula manilensis Philippi, from north central Texas. Jour-
nal of Molluscan Studies 44: 49-70
Bastardo, H. 1980. The chlorophyll a: phaeopigment ratio as
an indicator of the process of decomposition in some
freshwater plants. Acta Biologica Venezuelica 10: 241-253.
Bates, D. W. 1969. The horizontal traveling screen. IN: Engineering
Aspects of Thermal Pollution, F. L. Parker and P. A. Krenkel,
eds., pp. 225-242. Vanderbilt University Press, Nashville,
Tennessee.

Bickel, D. 1966. Ecology of Corbicula manilensis Philippi in the Ohio
River at Louisville, Kentucky. Sterkiana 23: 19-24.

Carr, L. M. 1973. Diversity and density of phytoplankton populations
in a reservoir receiving a heated effluent. M. S. Thesis, The
University of Texas at Arlington. 123 pp.

Cherry, D. S., J. H. Rodgers, Jr., R. L. Garney and J. Cairns, Jr.
1980. Dynamics and control of the Asiatic clam in the New
River, Virginia. Bull. no. 123, Virginia Water Resources
Research Center, Virginia Polytechnic Institute and State
University, Blacksburg, Virginia. 72 pp.

Dowell, C. L. and D. S. Breeding. 1966. Dams and reservoirs in
Texas. Texas Water Development Board 48: 1-267.

Dreier, H. and J. A. Tranquilli, 1981. Reproduction, growth, distribu-
tion and abundance of Corbicula in an Illinois cooling lake.
Illinois Natural Histroy Survey Bulletin 32: 378-393.

Eng. L. L. 1 979. Population dynamics of the Asiatic clam, Corbicula
fluminea (Muller), in the concrete-lined Delta-Mendota Canal
of central California. IN: Proceedings, First International Cor-
bicula Symposium, J. C. Britton, ed., pp. 39-63. Texas Chris-
tian University Research Foundation, Fort Worth, Texas.
Fitch, J. E. 1953. Corbicula fluminea in the Imperial Valley. Minutes
of the Conchology Club of Southern California 130: 9-10.
Foe, C. and A. Knight. 1985. The effect of phytoplankton and
suspended sediment on the growth of Corbicula fluminea
(Bivalvia). Hydrobiologia 127: 105-115.

Foe, C. and A. Knight. 1986. A thermal energy budget for juvenile
Corbicula fluminea. IN: Proceedings, Second International Cor-
bicula Symposium, J. C. Britton, ed. American Malacological
Bulletin Special Edition No. 2: 143-150.

Goss, L. B. and C. Cain, Jr. 1977. Power plant condenser and ser-
vice water system fouling by Corbicula, the Asiatic clam. IN:
Biofouling Control Procedures, L. D. Jensen, ed., pp. 11-17.
Marcel Dekker, Inc., New York.

Goss, L. B., J. M. Jackson, H. B. Flora, B. G. Isom, G. Gooch, S.
A. Murray, C. G. Burton and W. S. Bain. 1979. Control studies
on Corbicula for steam-electric generating plants. IN: Pro-
ceedings, First International Corbicula Symposium, J. C. Brit-
ton, ed., pp. 139-151. Texas Christian University Research

Foundation, Fort Worth, Texas.

Harvey, R. S. 1981 . Recolonization of a reactor cooling water system
by the clam Corbicula fluminea. Nautilus 95: 131-136.

Heinsohn, G. E. 1958. Life history and ecology of the freshwater clam,
Corbicula fluminea. M. S. Thesis, University of California, Santa
Barbara. 64 pp.

Henager, C. H., Sr., P. M. Daling and K. I. Johnson. 1985. Bivalve
fouling of nuclear power plant service-water systems.
NUREG/CR-4070, PNL-5300, Vol. 3, Division of Radiation Pro-
grams and Earth Sciences, Office of Nuclear Regulatory
Research, U. S. Nuclear Regulatory Commision, Washington,
D. C. National Technical Information Service, Springfield,
Virginia. 51 pp.

Hillis, D. M. and J. C. Patton. 1982. Morphological and electrophoretic
evidence for two species of Corbicula (Bivalvia: Corbiculidae)
in North America. American Midland Naturalist 108: 74-80.

Horning, W. B. and L. Kemp. 1964. Decline of the Asiatic clam popula-
tion in the Ohio River. Nautilus 78: 29-30.

Ingram, W. M. 1959. Asiatic clams as potential pests in California
water supplies. Journal of the American Water Works Associa-
tion 51: 363-370.

Kraemer, L. R. 1978. Corbicula fluminea (Bivalvia: Sphaeriacea): The
functional anatomy of its hermaphroditism. Bulletin of the
American Malacological Union 1978: 40-49.

Kraemer, L. R. 1979. Juvenile Corbicula: Their distribution in the
Arkansas River benthos. IN: Proceedings, First International
Corbicula Symposium, J. C. Britton, ed., pp. 89-97. Texas
Christian University Research Foundation, Fort Worth, Texas.

Kraemer, L. R., C. Swanson, M. Galloway and R. Kraemer. 1986.
Biological basis of behavior in Corbicula fluminea. II. Func-
tional morphology of reproduction and development and
review of evidence for self-fertilization. IN: Proceedings,
Second International Corbicula Symposium, J. C. Britton, ed.
American Malacological Bulletin Special Edition No. 2: 1 93-201 .

Lauritsen, D. D. 1986. Assimilation of radiolabeled algae by Corbicula.
IN: Proceedings, Second International Corbicula Symposium,
J. C. Britton, ed. American Malacological Bulletin Special Edi-
tion No. 2: 219-222.

Leveque, C. 1973. Dynamiquedes peuplements biologie et estimation
de la production des mollusques benthiques du Larc Tchad.
Cahiers O.R.S.T.O.M. Serie Hydrobiologie 7: 117-147.

Mattice, J. S. 1979 Interactions of Corbicula sp. with power plants.
IN: Proceedings, First International Corbicula Symposium, J.
C. Britton, ed., pp. 119-138. Texas Christian University
Research Foundation, Fort Worth, Texas.

Mattice, J. S., R. B. McLean and M. B. Burch. 1982. Evaluation of
short-term exposure to heated water and chlorine for the con-
trol of the Asiatic clam. ORNL/TM-78Q8, Publication no. 1748,
Oak Ridge National Laboratory, Environmental Sciences Divi-
sion, National Technical Information Service, Springfield,
Virginia. 33 pp.

McMahon, R. F. 1977. Shell size-frequency distributions of Corbicula
manilensis Philippi from a clam-fouled steam condenser.
Nautilus 91: 54-59.

McMahon, R. F. 1979. Response to temperature and hypoxia in the
oxygen consumption of the introduced Asiatic freshwater clam
Corbicula fluminea (Muller). Comparative Biochemistry and
Physiology 63A: 383-388.

McMahon, R. F. 1982. The occurence and spread of the introduced
Asiatic freshwater clam, Corbicula fluminea (Muller) in North
America: 1924-1982. Nautilus 96: 134-141.

McMahon, R. F. 1983. Ecology of an invasive pest bivalve, Corbicula.
IN: The Mollusca, Vol. 6, Ecology, W. D. Russell-Hunter, ed.,
pp. 505-561. Academic Press, San Diego.

WILLIAMS AND MCMAHON: CORBICULA ENTRAINMENT IN RAW WATER SYSTEMS 1 1 1

McMahon, R. F. and C. J. Williams. 1986a. Growth, life cycle, up-
per thermal limit and downstream colonization rates in a
natural population of the freshwater bivalve mollusc, Corbicula
fluminea (Muller), receiving thermal effluents. IN: Proceedings,
Second International Corbicula Symposium, J. C. Britton, ed.
American Malacological Bulletin Special Edition No. 2: 231-239.

McMahon, R. F. and C. J. Williams. 1986b. A reassessment of growth
rate, life span, life cycle and population dynamics in a natural
population and caged individuals of Corbicula fluminea (Muller)
(Bivalvia: Corbiculacea). IN: Proceedings, Second International
Corbicula Symposium, J. C. Britton, ed., American
Malacological Bulletin Special Edition No. 2: 151-166.

Morton, B. 1977a. The population dynamics of Corbicula fluminea
(Bivalvia: Corbiculacea) in Plover Cove Reservoir, Hong Kong.
Journal of Zoology, London 181: 21-42.

Morton, B. 1977b. The occurrence of inflammatory granulomas in
the ctendial marsupium of Corbicula fluminea (Mollusca:
Bivalvia): A consquence of larval incubation. Journal of In-
vertebrate Pathology 30: 5-1 4.

Morton, B. 1982. Some aspects of the population structure and sexual
strategy of Corbicula cf. fluminalis (Bivalvia: Corbiculacea) from
the Pearl River, Peoples Republic of China. Journal of
Molluscan Studies 48: 1-23.

O’Kane, K. D. 1976. A population study of the exotic bivalve Cor-
bicula manilensis (Philippi, 1841) in selected Texas reservoirs.
M. S. Thesis, Texas Christian University. 134 pp.

Owen, G. 1966. Feeding. IN: Physiology of Mollusca, Vol. II, K. M.
Wilbur and C. M. Yonge, eds., pp. 1-51 . Academic Press, New
York.

Parsons, T. R. and J. D. H. Strickland. 1963. Discussion of spec-
trophotometric determination of marine-plant pigments, with
revised equations for ascertaining chlorophylls and carote-
noids. Journal of Marine Research 21: 155-163.

Peeler, J. S. 1980. Primary productivity, standing crop and produc-
tion intensity of phytoplankton in Lake Arlington, Texas -- A
reservoir receiving a heated effluent. M. S. Thesis, The Univer-
sity of Texas at Arlington. 120 pp.

Prezant, R. S. and K. Chalermwat. 1984. Flotation of the bivalve Cor-
bicula fluminea as a means of dispersal. Science 225:
1491-1493.

Prokopovich, N. P. 1969. Deposition of clastic sediments by clams.
Sedimentry Petrology 39: 891-901.

Prokopovich, N. P. and D. J. Hebert. 1965. Sedimentation in the
Delta-Mendota Canal. Journal of the American Water Works
Association 57: 375-382.

Ray, H. C. 1962. Chokage of filtered-water pipe systems by freshwater
molluscs. In: Proceedings of the All-India Congress of Zoology,
Part 2, pages 20-23.

Sickel, J. B. 1979. Population dynamics of Corbicula in the Altamaha

River, Georgia. IN: Proceedings, First International Corbicula
Symposium, J. C. Britton, ed., pp. 69-80. Texas Christian
University Research Foundation, Fort Worth, Texas.

Sinclair, R. M. 1964. Clam pests in Tennessee water supplies. Journal
of the American Water Works Association 56: 592-599.

Sinclair, R. M. 1974. Effects of an introduced clam (Corbicula) on
water quality in the Tennessee River Valley. IN: Proceedings
of the Second Industrial Waste Conference, pp. 43-50. Ten-
nessee Stream Pollution Control Board, Tennessee Depart-
ment of Public Health, Nashville, Tennessee.

Sinclair, R. M. and B. G. Isom. 1963. Futher studies on the introduced
Asiatic clam (Corbicula) in Tennessee. Tennessee Stream
Pollution Control Board, Tennessee Department of Public
Health, Nashville, Tennessee, 76 pp.

Smith, A. L., A. Muia, J. P. Farkas and D. O. Bassett. 1979. Clams
- A growing threat to inplant water systems. Plant Engineer-
ing 1979: 165-167.

Smithson, J. A. 1981. Control and treatment of Asiatic clams in power
plant intakes. Proceedings of the American Power Conference
43: 1146-1151.

Strickland, J. D. H. and T. R. Parsons. 1972. A Practical Handbook
of Sea Water Analysis. Fisheries Research Board of Canada,
Ottawa, 310 pp.

United States Nuclear Reguatory Commission. 1981. Flow blockage
of cooling water to safety system components by Corbicula
sp (Asiatic clam) and Mytilus sp (Mussel). IE Bulletin no. 81-03,
Office of Inspection and Enforcement, United States Nuclear
Regulatory Commission, Washington, D.C. 6 pp.

White, D. S. 1979. The effect of lake-level fluctuations on Corbicula
and other pelecypods in Lake Texoma, Texas and Oklahoma.
IN: Proceedings, First International Corbicula Symposium, J.
C. Britton, ed., pp. 82-88. Texas Christian University Research
Foundation, Fort Worth, Texas.

White, D. S. and J. S. White. 1977. The effect of reservoir fluctuaions
on populations of Corbicula manilensis (Pelecypoda: Cor-
biculidae). Proceedings of the Oklahoma Academy of Science
57: 106-109.

Williams, C. J. 1985. The population biology and physiological
ecology of Corbicula fluminea (Muller) in relation to downstream
dispersal and clam impingement on power station raw water
systems. M. S. Thesis, The University of Texas at Arlington.
152 pp.

Williams, C. J and R. F. McMahon. 1985. Seasonal variation in ox-
ygen consumption rates, nitrogen excretion rates and tissue
organic carbon: nitrogen ratios in the introduced Asian
freshwater bivalve, Corbicula fluminea (Muller) (Lamellibran-
chia: Corbiculacea). American Malacological Bulletin 3:
267-268.

CORBICULA IN ASIA - AN UPDATED SYNTHESIS

BRIAN MORTON
DEPARTMENT OF ZOOLOGY
THE UNIVERSITY OF HONG KONG
HONG KONG

ABSTRACT

A previous review of Corbicula in Asia (Morton, 1979b) came to the tentative conclusion that there
are but two species. This resulted from analysis of early conchological assessments of the genus.
This review is of contemporary conchological literature but relies principally on the researches of
biologists who have collectively defined the species concerned. It is reasserted that two species are
predominant.

C. fluminalis (Muller) occurs in estuaries and tolerates salinities of up to 50%. It is dioecious
with a trend towards protogyny. Breeding occurs over a single winter season, when temperatures
are low. Eggs are not incubated, fertilization being external; strangely, however, typical incubatory
glands develop in the inner demibranchs and it seems possible that under certain extreme condi-
tions eggs may be incubated. A single growth ring/annum is produced, the species reaching a max-
imum theoretical shell length in southern China of 54 mm and living for up to 10 years.

C. fluminea (Muller) is fresh water with only a limited tolerance of low (1 5%) salinities. This smaller
species (a shell length of up to 35 mm in southern China) is dioecious with a high percentage of her-
maphrodites in lentic waters but hermaphrodite with an equal % of females (no males) in lotic systems.
Possibly other sexual strategies occur under extreme conditions. The species incubates fertilized eggs
within the inner demibranchs. These are released as crawling pediveligers at a shell length of 220
/im. There are two peaks of reproduction, one in early, the other in late summer when temperatures
are high. Two growth rings are thus produced/annum, the species living for approximately three years.

C. fluminea has been introduced into N. America.

In an earlier review of the status of Corbicula in Asia,
the tentative conclusion was reached, despite the plethora
of available species names, that but two highly variable, wide-
ly distributed, species occur (Morton, 1979b). Following
research upon the biology, ecology, reproduction and mor-
phology of representatives of these two species in Hong Kong
(Morton, 1973; 1977a; 1982; 1983) it was concluded (Morton,
1982) that these could be correlated with the types of Cor-
bicula fluminea and C. fluminalis, though since the latter
species has never before been recorded from China, this
judgement was qualified.

The earlier review concentrated, largely, on old con-
chological studies particular attention being paid to the works
of Prashad (1924; 1928a; b; 1929; 1930). At that time there
was very little biological information on the genus that would
enable a more meaningful analysis to be made. New species
of Corbicula are still being described (Ray, 1967; Tem-
charoen, 1971; Brandt, 1974; Djajasasmita, 1977b), despite
the fact that there is already a ridiculous nomenclatorial load,
and, it is further clear, there is immense variability in shell
form, size and colour in representatives of Corbicula.

Recently, species of Corbicula have been introduced
into areas outside the usual range (Asia) and the artificial but
established range in N. America. These are France and Por-

Fig.l . The distribution of Corbicula in Africa and Asia, (after Zhadin,
1948; 1965) and in N. America (after Britton and Morton, 1982). Sites
of recent introductions into Portugal, France and Argentina are also
recorded. Figures in circles refer to the numbers of species of Cor-
bicula presently recorded from various regions of Africa and Asia.

American Malacological Bulletin, Special Edition No. 2(1986):1 13-124

113

114

CORBICULA SYMPOSIUM

tugal (Mouthon, 1981) and Argentina (Ituarte, 1981). It seems
very clear that Corbicula is going to be spread yet further.
Where Corbicula has spread outside its native range, it has
become an important biofouling pest, especially in N. America
(Morton, 1979a). It may equally become a pest organism in
newly occupied areas. It is thus necessary for responsible
decisions to be made now with regard to the identity of the
introduced species on the basis of reliable characters. The
present taxonomic chaos that surrounds Corbicula in Asia
results from the quest of conchologists to erect names for
shells barely “different from”, “somewhat similar to” or
somehow “related to” other highly dubious “species”. Here
the taxonomic status of Corbicula in Asia is re-examined, prin-
cipally on the characteristics of each species, other than the
shell. The researches of modern biologists in Asia are
discussed and form the basis for the decisions made here.
However, I have also tried to ascribe names to what I believe
to be the extant species based upon (to me) reasonable tax-
onomic criteria.

In light of the considerable biofouling problems that
are ascribed to the introduced Corbicula in N. America, I also
assess the biofouling potential of the species in its home
range.

CORBICULA IN ASIA

TAXONOMY AND BASIC BIOLOGY
U.S.S.R.

This discussion of Coroicula in Asia commences with
a review of those species occurring in the fresh waters of the
U.S.S.R. It begins here for the single important reason that
from this vast region, Zhadin (1965) records but two species,
namely C. fluminalis (Muller 1774) and C. fluminea (Muller
1 774). The former (apparently) has a middle Asian, the latter
an east and southern Asian distribution. A point of anomaly
with Morton (1977a; 1982), is that Zhadin records maximum
shell lengths of 21mm (C. fluminalis) and 37mm (C. fluminea),
whereas Morton records maximum theoretical lengths for
these species of 54mm and 35mm respectively. However,
Alimov (1974) has shown that the growth constant of the von
Bertalanffy equation is functionally dependant upon prevail-
ing environmental factors and that the maximum length of
the adult animal increases with an increase in the sum of the
effective habitat temperatures. Thus both species may have
a highly variable form and attain different maximum sizes in
the various components of their wide ranges.

Zhadin (1948) maps the range of Corbicula,
demonstrating for both species an essentially Asian, tropical,
distribution (Fig. 1). On the basis of Zhadin’s researches
Sinclair & Isom (1963) suggested that C. fluminalis might
prevail in west Asia while C. fluminea might prevail in the
south and east.

Mandryka (1981) records C. japonica Prime from
brackish water lakes on the shore of the Sea of Japan; from
two of these, populations of the bivalve were thought to be
heterogenous with length/height relationships suggesting two
morphological groupings. Perhaps, however, two species

were coexisting here. Yaroslavtseva, Pavlenko & Fedoseeva
(1981) also record C. fluminea from the U.S.S.R.

Issatullaev (1980) describes the Corbiculidae of Cen-
tral Asia. C. tibetensis (Prashad) and C. ferghanensis (Kur-
salova and Starobogatov) are said to be ovoviviparous while
C. cor (Lamarck), C. purpurea (Prime) and C. fluminalis are
said to be oviparous.

AFRICA

Counts (1980) has examined the zoogeographic
records of museums around the world for species of Corbicula
collected in Africa. Seventeen species were recorded (in-
cluding C. fluminea) as follows:

C. aegyptica Bogart, C. africana Kiawis, C. agrensis
Kurr, C. artini Pallary, C. astartina Martens, C. australis
(Muller), C. cummingtoni Smith, C. fischeri Germain,
C. fluminea (MUIIer), C. kirkii Prime, C. lamarckiana
Prime, C. oliphantensis Craven, C. pusilla (Philippi), C.
radiata Hanley, C. sikarae Ancey, C. subradiata Kurr,
and C. tanganyicensis Crosse.

Records for C. agrensis and C. australis from Ghana and
South Africa were considered doubtful as these species have
an Indian and Australian distribution respectively. Most
African species were confined between longitudes 26°E and
34°E and between elevations of 0 and 1500m.

Kenmuir (1980) records only C. africana Krass from
Lake Kariba. Most significantly, Mandahl-Barth (1954) records
that but two species occur in the White Nile, i.e., C. africana
in the middle reaches of the river and C. fluminalis in most
of its lower reaches.

ISRAEL

Tchernov (1975) reports that C. fluminalis is the only
corbiculid found in the Sea of Galilee.

INDIA

Lomte (1971) records two species of Corbicula from
the Marathwada region of India. Akhatar (1978) similarly
records two species from Lahore, i.e., C. striatella Deshayes
and C. regularis. Mudkhede and Nagabhushanam (1977)
report upon the heat tolerance of C. regularis from
Marathwada.

Ray (1967) records a new species of Corbicula, C.
krishnaea, from Maharashtra, India.

LAOS

Temcharoen (1971) describes a new species of Cor-
bicula - C.crocea - from Laos. It differs from all other species
of this genus “by its reddish-yellowish colour” and is known
from the type locality only,

CAMBODIA

Mizuno and Mori (1970) record C. noetlingi Martens
and C. petiti Clessin from Lakes in Cambodia.

MORTON: CORBICULA IN ASIA

115

MALAYSIA

Berry (1974) sexed a sample of C. malaccensis from
Malaysia and showed that of those individuals greater than
8mm shell length, 40 were hermaphrodites, 19 were female
and two were male. Fertilized eggs are brooded in the inner
demibranchs of both females and hermaphrodites and are
released as juveniles. This author also records C. javanica
from Malaysia.

THAILAND

Mizuno and Mori (1970) have made an ecological
survey of Asian lakes and record the following species of Cor-
bicula from Thailand: C. noetlingi von Martens, C. siamensis
Prashad, C. pef/f/ Clessin, C. iamarckiana Prime, C. larnaudieri
Prime and C. ligidiana ( = C. lydigiana Prime) from Thailand.

A major review of the Corbiculidae of Thailand by
Brandt (1974) recorded the following 23 species, including
C. fluminea: C. arata (Sowerby), C. blandiana Prime, C.
bocourti (Morelet), C. javanica (Mousson), C. Iamarckiana
Prime, C. lydigiana Prime, C. pisidiformis Prime, C. castanea
Morelet, C. cyreniformis Prime, C. tenuis Clessin, C. fluminea
(Muller), C. noetlingi von Martens, C. regia Clessin, C. gusta-
viana von Martens, C. moreietiana Prime, C. siamensis
Prashad, C. erosa Prime, C. iravadica Hanley & Theoblad,
C. baudoni Morelet, C. gubernatoria Prime, C. leviuscuia
Prime, C. solidula Prime, C. messaged Bavay and
Dautzenberg.

Not content with this, however, Brandt also describes
5 new species: C. virescens, C. pingensis, C. occidentiformis,
C. vokesi and C. heardi. Virtually all of the earlier described
species reported upon by Brandt can be allied to C. fluminea
(Morton, 1979), many of them being so synonymised by
Prashad (1928b) anyway.

INDONESIA

Djajasasmita (1975; 1977a) reviewed the species of
Corbicula occurring throughout Indonesia. Of 35 previously
described species, 16 were considered valid. These are:-

Sumatra: C. gustaviana von Martens, C. moltkiana
Prime, C. sumatrana Clessin, C. tobae von
Martens, C. tumida Deshayes.

Java: C. javanica (Mousson), C. pulchella (Mousson),
C. rivalis (Philippi).

Borneo: C. bitruncata von Martens, C. pullata Philippi.

Celebes: C. lindoensis Bollinger, C. loehensis Kruimel,
C. matannensis Sarasin and Sarasin, C. sub-
planata von Martens.

Timor: C. australis (Lamarck).

New Guinea: C. debilis (Gould).

A Philippine species, C. squalida Deshayes, was also added
as a new record. Subsequently, Djajasasmita (1977b) describ-
ed a new species (C. lacunae) from 21 specimens collected
from two freshwater marshes in E. Java. The new species
apparently shows ‘relationship’ with C. loehensis from
Celebes.

Bentham-Jutting (1953) had earlier reviewed the Cor-
biculidae of Java and also recorded only C. javanica, C. rivalis

u. s. Nat. Bins. / z 2. yqrp

Lea Coll.

!'2

!§

Plate 1 . The holotype and label of C. leana Prime (USNM. 122429).

and C. pulchella from this region.

PHILIPPINES

Villadolid and Del Rosario (1930) report that Corbicula
manillensis Philippi is the only corbiculid recorded from
Laguna de Bay and its tributaries. This name is a junior
synonym of C. fluminea (Morton, 1979). In Laguna de Bay
the species attains a length of some 30mm, breeds from
March through to July, incubates larvae in modified inner
demibranchs and the population comprises approximately 4
size classes.

CHINA (PEOPLE’S REPUBLIC OF)

From China, Liu and Huang (1964) recorded C. nitens
(Philippi) and Corbicula sp. from the San-Men-Hsia Reser-
voir of the Yellow River. Tchang, Li and Liu (1965) recorded
C. fluminea, C. aurea (Heude), C. largillierti (Philippi) and C.
nitens as occurring in Tung-ting Lake and its surrounding
waters, Hunan Province. Tchang and Li (1965) record but C.
fluminea and C. largillierti from Poyang Lake and its surroun-
ding waters, Kiangsi Province. Chen (1979), however, only
records C. fluminea from Lake Hwama, Hubei Province.

Liu, Zhang, Wang and Wang (1979) have reviewed the
Corbiculidae from China and record the following species
- C. fluminea, C. largillierti and C. nitens, the latter two hav-
ing a restricted distribution, with C. largillierti endemic to the
lakes of the Yangtze River.

Miller and McClure (1931) and Morton (1973) reported
upon a species called C. manilensis from the Pearl River,
southern China. This name, however, is but a junior synonym
of C. fluminea (Morton, 1979b) and Morton (1982) has subse-
quently suggested that this species is, in fact, comparable
with C. fluminalis, a species hitherto not recorded from the
Chinese mainland.

HONG KONG

Morton (1977a; 1983), Brandt (1980) and Dudgeon

116

CORBICULA SYMPOSIUM

(1980; 1982) record only C. fluminea from Hong Kong, though
as noted above C. fluminalis occurs in the Pearl River estuary
close to Hong Kong.

TAIWAN

A major review of Taiwanese literature by Wu (1980)
concludes that only C. fluminea occurs in Taiwan. The same
author (Wu, 1979) provides morphometric data for this
species. Similarly, Chen (1976) records that the common
cultivated species in Taiwan is C. fluminea but is reported
to be dioecious with external fertilization and planktonic larvae!

KOREA

Oyama (1943) reviewed the species of Korean Cor-
bicula, recording 7 species but including, significantly, C.
fluminea and C. japonica. The other five species were C.
suifunensis Lindholm, C. colorata von Martens, C. felnouilliana
Heude, C. vicina Heude and C. papyracea Heude.

Lee and Park (1974) further recorded C. orientalis
Lamarck (considered by Johnson (1959) to be C. japonica)

while Lee and Heo (1980) discuss C.elatior von Martens (a
species which Oyama considered but a form of C. japonica).

Lee and Chung (1980) believe the common ‘marsh’
clam of Korea to be C. fluminea. Thus, contemporary Korean
authors seem to be reporting upon two species referrable to
C. fluminea and C. japonica.

JAPAN

Perhaps the most authoritative accounts of Corbicula
in Asia come from Japan where there is a long history of con-
chological and maiacological research. Kuroda (1938) record-
ed 20 species from Japan but, significantly, concluded that
all of these could be assigned to two species groups one
generally found in saline waters, the other in freshwater.
Kurashige (1945a;b) reports upon the distribution, ecology
and shell shape of C. felnouilliana and C. fluminea. The most
recent review of Japanese Corbicula by Habe (1977) con-
cludes that there are 4 species, i.e., C. japonica Prime, C. san-
dai Reinhardt, C. fluminea Muller and C. leana Prime.

The holotype of C. leana is shown in Plate 1 and can

Plate 2. The holotypes and labels of C. fluminea Muller) and C. fluviatilis (Muller). (Universitetets Zoologiske Museum, Copenhagen).

MORTON: CORBICULA IN ASIA

117

Plate 3. The hoiotype and label of C. japonica Prime (MCZ. 15904).

be compared with the holotypes of C. fluminea and C. fluviatilis
(Plate 2) (the latter is by general consent (Prashad, 1929;
Morton, 1979b) considered synonymous with C. fluminea).
Bearing in mind that the type of C. fluminea is a juvenile, there
can be little doubt that C. leana is virtually indistinguishable
from C. fluminea (or C. fluviatilis). Indeed even Prashad (1924)
who thought there were 69 valid species of Corbicula in Asia,
considered C. leana to be an “insular form” of C. fluminea.

Plates 3 and 4 illustrate the hoiotype and paratype of
C. japonica and C. sandai respectively (note that C. sandai
was originally described as but a variety of C. japonica). When
compared with the hoiotype of C. fluminalis (Plate 5) they are
virtually indistinguishable, showing the same conical shell
with narrow growth lines.

On conchological grounds therefore it is suggested
that in Japan but two species occur; C. fluminalis (= C.
japonica) and C. fluminea ( = C. leana). It is, however, con-
ceded that C. sandai may be either a lake morphological form
of the otherwise brackish water C. fluminalis or a species
endemic to the ancient Lake Biwa (Hayashi, 1972; Mori,

118

CORBICULA SYMPOSIUM

Plate 5. The holotype of C. fluminalis (Muller) (Universitetets Zoologiske Museum, Copenhagen).

MORTON: CORBICULA IN ASIA

119

JAPAN

SOUTHERN CHINA

Corbicula japonica

Corbicula fluminalis

Brackish waters

Asahina, 1941

Hayashi, 1956

Kado & Murata, 1974
Matsushima, 1980

Maru, 1981

Brackish waters

Miller & McClure, 1931
Morton, 1973; 1982

Dioecious

Utoh, 1981

Maru, 1981

Dioecious with a trend
towards protogyny

Morton, 1982

Non incubatory

Asahina, 1941

Non incubatory

Morton, 1982

Winter breeding season

Utoh, 1981

Winter breeding season

Morton, 1982

Late summer breeding

season

Maru, 1981

Life span of 8-9 years

Utoh, 1981

Life span of up to 10 years

Morton, 1982

Growth rings:

1/annum

Utoh, 1981

Maru, 1981

Growth rings:

1/annum

Morton, 1973;

1982

JAPAN

SOUTHERN CHINA

Corbicula leana

Corbicula fluminea

Fresh water

Asahina, 1941

Matsushima, 1980

Kado & Murata, 1974

Fresh water

Morton, 1977a

Hermaphrodite

Fuziwara, 1979

Lentic waters:

dioecious + hermaphrodite
Lotic waters:
female + hermaphrodite

Morton, 1983

Incubatory
(inner demibranchs)

Fuziwara, 1977

Incubatory
(inner demibranchs)

Morton, 1977a; b; c

Matures at 10mm

Fuziwara, 1977; 1979

Matures at 10mm (approx.)

Morton, 1977a

Breeding season:
May-August
May-November
Summer

early summer

April-October

May & Juiy-September

Tamura, 1959

Ikematzu & Kammakura, 1975
Ikematzu &

Yamane, 1977

Fuziwara, 1975;

1977; 1978

Fuziwara, 1979

Kaurajiri, 1948

Breeding season:

(2 peaks/annum):

April-October

Morton, 1977a

Larvae released
as D-shaped crawling
juveniles (20(tym)

Fuziwara, 1977

Tamura, 1959

Larvae released
as D-shaped crawling
juveniles (200 /un)

Morton, 1977a

Growth rings:

2/annum

Fuziwara, 1978

Growth rings:

2/annum

Morton, 1977a

Life-span: ? *

Fuziwara, 1978

Life-span: 3 years

Morton, 1977a

*N.B. Fuji (1957) describes a population of C. japonica as comprising 3 age classes — was he actually investigating C. leana?

Table 1 . A comparison of characterizing features of C. japonica and C. leana from Japan with C. fluminalis and C. fluminea from southern China.

1978), though Maru (1981), and Nakao (1982) also recorded
this “species” from other lakes and lagoons in Japan and
Itasaka, Sugita and Hori (1980) record it from the Seta River.
Similarly, C. japonica is riverine but also occurs in lakes and
lagoons (Gose, 1965; Fuji, 1979).

I have also reviewed the modern literature on
Japanese Corbicula by biologists. It is significant that most
contemporary Japanese scientists (e.g. Kado and Mat-
sushima, 1976a; b) only refer to two species, namely C.
japonica and C. leana, though some ecologists e.g. Hayashi

120

CORBICULA SYMPOSIUM

(1972), Mori (1978), Maru (1981) and Nakao(1982), also report
on C. sandai. Biological data on these two species are com-
pared in Table 1 with data on C. fluminalis and C. fluminea
from southern China largely derived from Morton (1973;
1977a; 1982; 1983).

Thus, C. japonica (like C. fluminalis in the Pearl River,
southern China) is a brackish water species with the ability
to tolerate saline (70% sea water) conditions (Asahina, 1941 ;
Kado and Murata, 1974; Matsushima, 1980). C. leana (like
C. fluminea ) on the other hand is essentially a fresh water
species with a very much reduced capability of tolerating
saline conditions up to 15% (Kado and Murata, 1974; Mat-
sushima, 1980).

C. japonica apparently, even after up to 9 years growth
rarely attains a shell length in excess of 30mm (Utoh, 1981),
whereas C. fluminalis in China can attain a maximum shell
length of 54mm. Such discrepancies are not critical, however,
Alimov (1974), as pointed out earlier, having shown that
growth is dependent upon temperature. Thus in higher
latitudes C. fluminalis (and C. japonica) attains a smaller max-
imum size (Zhadin, 1965) than C. fluminalis in southern China
(Morton, 1982). The same is true of C. fluminea in N. America
(Britton and Morton, 1979).

C. fluminalis (Morton, 1982) and C. japonica (Utoh,
1981) live for over 8 years, possibly up to 10; they both pro-
duce 1 growth ring/annum, correlated with a single breeding
season in the colder months of the year. Maru (1981) reports
that in Lake Abashiri, C. japonica matures at a length of
15mm, three years after hatching. C. fluminalis and C.
japonica are essentially dioecious, the former (Morton, 1982)
possessing a small % of hermaphrodites as is typical of many
fresh water, otherwise dioecious, bivalves, e.g., Dreissena
polymorpha (Antheunisse, 1963), Geloina (Morton, 1985) and
Anodonta (Dudgeon and Morton, 1983). In C. fluminalis , Mor-
ton (1982) has detected a trend towards protogyny so that
a greater % of young animals are female, and a greater %
of older individuals are male. Neither C. fluminalis nor C.
japonica incubate larvae in the ctenidia, though Morton (1982)
has shown that in C. fluminalis , ctenidial glands typical of
those of the incubatory C. fluminea (Morton, 1977a; b; c) do
develop, but for some as yet unknown reason. Possibly under
extreme environmental conditions, larvae can be retained.
It is (only) possibly significant that Miyazaki (1936) has sug-
gested that C. sandai (from L. Biwa) is non-incubatory and
yet produces non-swimming larvae. This anomaly has been
commented upon before (Morton, 1979). Is it possible that
in a lake environment C. fluminalis is incubatory? The pro-
blem of C. sandai has to be resolved, though this may be dif-
ficult as the “species” is apparently being replaced by C.
leana (C. fluminea) with progressive eutrophication of Lake
Biwa (Itasaka, Sugita, Okumura and Hori, 1980; Mori, 1978).

C. fluminea and C. leana similarly possess essentially
the same characteristics. Both are smaller species, reaching
in Hong Kong a maximum shell length of 35mm and in Japan
between 26mm (Ikematsu and Yamane, 1977) and 40mm
(Fuziwara, 1978). The species breeds in summer (Tamura,
1959; Ikematzu and Kammakura, 1975; Ikematzu and
Yamane, 1977), typically in two peaks in early and late sum-

mer (Kaurajiri, 1948) though Fuziwara (1975; 1977; 1978;
1979) has shown that almost continuous breeding is possi-
ble when temperatures exceed 19°C. It seems possible that
temperature is critical in determining the length of the
breeding season of this species there being minimum and
maximum temperatures below and above which reproduc-
tion is possible. Morton (1977a) working on a reservoir popula-
tion of this species suggested that it is a protandric her-
maphrodite in contrast to other workers who considered it
either dioecious (Lee and Chung, 1980) or a simultaneous
hermaphrodite (Ikematsu and Yamane, 1977; Kraemer and
Lott, 1977; Kraemer, 1978). This matter has now at least been
partly resolved (Morton, 1983). In Hong Kong, the species
is dioecious in lakes with a large (30%) percentage of her-
maphrodites. Thus, since juvenile males attain maturity before
juvenile females, gonad smears to determine sex would sug-
gest protandry (Morton, 1977a). This is not so, however. In
streams, no males occur in the population and the species
can in these situations be described as hermaphrodite with
an equal percentage of females. It seems possible that C.
fluminea has an extremely variable sexuality, enabling it to
survive a wide range of environmental conditions. This has
been noted by both Morton (1983) and Ikematsu and
Yamane (1977) and Fuziwara (1979) for C. fluminea and C.
leana respectively. However, one thing is clear, C. fluminea
(and C. leana) broods fertilized eggs in the inner demibranchs
to a D stage crawling pediveliger. Such juveniles are releas-
ed at a characteristic length of 200/tm in both C. fluminea
(Morton, 1977a) and C. leana (Fuziwara, 1977). Glands in the
inner demibranchs possibly serve to nourish the juveniles;
unreleased, dead juveniles result in the formation of cysts
to encapsulate them (Morton, 1977c; Britton, Barcellona,
LaGrone and Hagan, 1981). These in turn are autotomised
from the gill. C. fluminea produces two growth rings/annum,
reaches maturity at a shell length of approximately 10mm and
lives for but up to 3 years. Age is not reported upon for C.
leana in Japan, but Fuji (1957) working on C. japonica showed
that the population of this species had three age classes.
However, other workers (Utoh, 1981) have shown that C.
japonica lives for 8 or 9 years. Is it possible that Fuji was in
fact working on C. leana , the taxonomy of these species be-
ing at that time completely confused anyway? If so, then it
is clear that C. fluminea and C. leana share almost identical
characters.

Thus on biological grounds (as with conchological
criteria) it is suggested that in Japan but two species of Gor-
bicula occur, i.e., C. fluminalis and C. fluminea, which are
equivalent to what are at present called C. japonica and C.
leana. Much contemporary biological literature from Africa
and Asia similarly indicates that two species are present
throughout. These too can be allied to C. fluminalis and C.
fluminea. Figure 1 indicates the numbers of species presently
thought to occur in each country here discussed.

BIOFOULING OF CORBICULA IN ASIA

There are no records in the literature of Corbicula caus-
ing problems of biofouling in its natural range. To the con-

MORTON: CORBICULA IN ASIA

121

trary, in its introduced range of N. America, Corbicula is a
serious pest, these problems being reviewed by Sinclair and
Isom (1963) and Morton (1979a).

Two reasons, possibly, account for this. First, in its
natural range, Corbicula will be subject to the natural checks
of disease, parasitism and predation that maintain a popula-
tion balance. Only in the introduced range, free of these con-
straints can the population balance be overturned such that
the invasive species undergoes a population “explosion”.
Second, the major centres of urban population in Asia are
at river mouths. Power stations erected here would not be
invaded by C. fluminalis since this estuarine species does not
possess the attributes necessary for fouling existence i.e. , it
is long-lived and dioecious with external fertilization. C.
fluminea on the other hand though possessing all the advan-
tages for fouling i.e., a short life span, rapid development,
hermaphrodite and the release of brooded pediveligers is
similarly excluded because of a low salinity tolerance. The
widespread use of salt or estuarine waters for industrial cool-
ing purposes thus effectively excludes Corbicula fluminea as
a significant biofouling agent in Asia.

INTRODUCED CORBICULA

To date, species of Corbicula have been introduced
into three significant areas (Fig. 1).

1. North America. Britton and Morton (1979; 1982) have
shown that the species of Corbicula introduced into North
America is C. fluminea. Previously called by a number of other
names, e.g., C. leana and C. manilensis (both of which are con-
sidered junior synonyms of C. fluminea) (Morton, 1979b) this
species matches in every respect the biological characters
of C. fluminea (Table 1). It has become a very important
biofouling pest throughout its range in North America.

2. South America. Counts (1980) records 20 species
of Corbicula from South America, though generally speak-
ing the continent is not within the range of Corbicula s.s. and
the family is here represented by Neocorbicula Fischer, 1 887
(Parodiz and Hennings, 1965). Ituarte (1981) has reported the
introduction of species of Corbicula into Argentina. This author
considers that two species have been introduced, namely C.
fluminea and C. leana. It is here concluded that C. leana is
no more than the Japanese “form” of C. fluminea thereby
suggesting that either C. fluminea in Argentina exists as two
morphological forms or that C. fluminalis as well as C. fluminea
has been introduced. More research on Corbicula in the La
Plata River is required. Native S. American Corbicula are not
known to be biofouling pests, but the potential problems of
the introduced species have yet to be determined.

3. France and Portugal. Mouthon (1981) reports that
a species of Corbicula has recently been introduced into the
estuaries of the Dordogne, France and the River Tagus in
Portugal, accompanied by various species of Pisidium.
Mouthon concludes that this species is C. fluminalis. The two
populations exhibit physiological and morphological dif-
ferences, however, and thus more research is required to
demonstrate whether either or both C. fluminea and C.
fluminalis have been introduced into either or both sites. In

this area, Corbicula can be considered to be reoccupying its
old range (Zhadin, 1948; 1965).

A careful watch should be maintained for other Cor-
bicula introductions elsewhere.

CONCLUSIONS

This review reaches but one conclusion. Throughout
the generic range, i.e., Africa and Asia, there are two
predominat species of Corbicula. These can be named C.
fluminalis (Muller, 1 774) and C. fluminea (Muller, 1 774). A vast
array of other Corbicula species names have been erected
by conchologists on shell characters alone. I consider all of
these invalid until such time as each and every one of them
can be shown to be biologically different from C. fluminalis
or C. fluminea which are now distinguished by a range of good
characters not based on the shell, a feature clearly extreme-
ly variable in both species.

Two areas requiring caution are identified. First, it
seems possible that both species can occur in the same river
system e.g., the White Nile, (Mandal-Barth, 1954); C.
fluminalis at the mouth, C. fluminea at the head. Since both
are capable of some degree of salt tolerance, the former more
so than the latter, it seems clear that in the middle reaches,
the two species may overlap. Investigation of this mixed
population could lead to erroneous conclusions. Some
anomalies already exist, e.g., Mandryka (1981) reports two
morphological groupings for C. japonica in coastal lakes of
the Sea of Japan. Fuji (1957) reported that C. japonica lives
for but 3 years whereas contemporary authors (Utoh, 1981)
report that this species lives for over 8 years.

Second, it is clear that C. fluminea , as in Hong Kong
(Morton, 1977a; 1983), can occur in lentic and lotic systems
where sexual strategies are different. Possibly, C. fluminalis
can do likewise in which case river and lake populations of
a long lived, dioecious winter breeding species of Corbicula
need careful assessment.

Possibly the species concerned is C. fluminalis, or at
least a morphological form of that species, or it may be an
isolated endemic species, as are supposed to be C. sandai
from L. Biwa (Japan) (Mori, 1978) and C. largillierti from the
lakes of the Yangtze River, China (Liu, Zhang, Wang and
Wang, 1979).

Such populations require urgent attention to test the
2 species model, here argued for Corbicula in Asia.

ACKNOWLEDGEMENTS

I am grateful to the following for the loan of type material: Dr.
K. J. Boss, Museum of Comparative Zoology, Harvard University;
Dr. R. S. Houbrick, Smithsonian Institution, Washington; Dr. J.
Janssen, Forschungsinstitut Senckenberg, Frankfurt; Dr. Jorgen
Knudsen, Universitetets Zoologiske Museum, Copenhagen.

I am also, especially, grateful to Prof. J. C. Britton (Texas Chris-
tian University) for critically reading the first draft of the manuscrip*
of this paper and for, over many years of friendship, stimulating
arguments regarding Corbicula.

122

CORBICULA SYMPOSIUM

LITERATURE CITED

Akhatar, S. 1978. On a collection of fresh water mollusks from Lahore.

Biologia 24: 437-447.

Alimov, A. F. 1974. Growth regularities in freshwater bivalve mollusks.
Zhurnal Obshchei Biologii 35: 576-589.

Antheunisse, L. J. 1963. Neurosecretory phenomena in the Zebra
mussel, Dreissena polymorpha. Archives neerlandaises de
Zoologie. 15: 237-314.

Asahina, E. 1941 . An ecological study of Corbicula japonica Group,
the brackish water bivalve, with special reference to the en-
vironmental factors of its habitat in Hokkaido. Bulletin of the
Japanese Society of Scientific Fisheries 10: 143-152.

Bentham-Jutting, W.S.S. Van. 1953. Systematic studies on the non-
marine Mollusca of the Indo-Australian Archipelago. IV. Critical
revision of the fresh water bivalves of Java. Treubia 22: 1 9-73.

Berry, A. J. 1974. Fresh water bivalves of peninsular Malaysia with
special reference to sex and breeding. Malaysian Naturalist
Journal 27: 99-110.

Brandt, A. J. 1980. An annotated checklist of the non-marine molluscs
of Hong Kong. In: Proceedings of the First International
Workshop on the Malacofauna of Hong Kong and southern
China, B.S. Morton, ed. pp. 101-108. Hong Kong University
Press.

Brandt, R. A. M. 1974. The non-marine Mollusca of Thailand. Ar-
chiv fur Molluskenkunde 105: 1-423.

Britton, J. C., W. J. Barcellona, M. LaGrone and J. Hagan. 1981.
Ctendial autotomy in Corbicula fluminea in response to massive
granulomas. Science 212: 551-553.

Britton, J. C. and B. Morton. 1979. Corbicula in North America: the
evidence reviewed and evaluated. In, Britton, J. C., J. S. Mat-
tice, C. E. Murphy, and L. W. Newland, eds. pp. 15-38 Pro-
ceedings of the First International Corbicula Symposium, Texas
Christian University Research Foundation, Ft. Worth, Tx.

Britton, J. C. and B. Morton. 1982. A dissection guide, field and
laboratory manual to the introduced bivalve, Corbicula
fluminea. Malacological Review, Supplement 3, 82 pp.

Chen, Q.Y. 1979. A report on Mollusca of Lake Hwama, Hubei Pro-
vince. Oceanologia et Limnologia Sinica 10: 46-66.

Chen, T. P. 1976. Culture of the freshwater clam, Corbicula fluminea,
In: Aquaculture Practices in Taiwan. Page Brothers (Norwich,
U. K.) Ltd., 107-110.

Counts, C. L. III. 1980. The genus Corbicula (Bivalvia: Corbiculidae)
in Africa and South America: zoogeographic and taxonomic
problems. Bulletin of the American Malacological Union, 46:
71-72.

Djajasasmita, M. 1975. On the species of the genus Corbicula from
Celebes, Indonesia (Mollusca: Corbiculidae). Bulletin of
Zoologisch Museum, Universiteit van Amstredam 4: 83-86.

Djajasasmita, M. 1977a. A new species of freshwater clam from Java,
Indonesia (Bivalvia: Corbiculaidae). The Veliger 19: 425-426.

Djajasasmita, M. 1977b. An annotated list of the species of the genus
Corbicula from Indonesia (Mollusca: Corbiculidae). Bulletin of
Zoologisch Museum, Universiteit van Amsterdam 6: 1-9.

Dudgeon, D. 1980. Some aspects of the biology of Cristaria
(Pletholophus) disoidea (Bivalvia: Unionacea) in Plover Cove
Reservoir, Hong Kong. In: Proceedings of the First International
Workshop on the Malacofauna of Hong Kong and southern
China, B. S. Morton, ed. pp. 181-120, Hong Kong University
Press.

Dudgeon, D. 1982. Aspects of the desiccation tolerance of four
species of benthic Mollusca from Plover Cove Reservoir, Hong
Kong. The Veliger. 24: 267-271.

Dudgeon, D. and B. S. Morton. 1983. The population dynamics and
sexual strategy of Anodonta woodiana (Bivalvia: Unionacea)
in Plover Cove Reservoir, Hong Kong. Journal of Zoology, Lon-
don 201: 161-183.

Fuji, A. 1957. Growth and breeding season of the brackish water
bivalve, Corbicula japonica, in Zyusan-gata Inlet. Bulletin of
the Faculty of Fisheries, Hokkaido University 8: 178-184.

Fuji, A. 1979. Phosphorus budget in a natural population of Corbicula
japonica Prime in a poikilohaline lagoon Zyusan-ko, Japan.
Bulletin of the Faculty of Fisheries, Hokkaido University 30:
34-49.

Fuziwara, T. 1975. On the reproduction of Corbicula leana Prime.
Japanese Journal of Malacology (Venus) 34: 54-56.

Fuziwara, T. 1977. On the growth of young shell of Corbicula leana
Prime. Japanese Journal of Malacology (Venus) 36: 19-24.

Fuziwara, T. 1978. On the ovulation of Corbicula leana Prime.
Japanese Journal of Malacology (Venus) 37: 22-28.

Fuziwara, T. 1979. A hypothesis on the distribution of Corbicula leana
Prime. Japanese Journal of Malacology (Venus) 38: 68-69.

Gose, K. 1965. The shell zone of Lake Kitaura. Japanese Journal of
Ecology 15: 228-232.

Habe, T. 1977. Systematics of Mollusca in Japan: Bivalvia and
Scaphopoda. Hokuryu-kan Publishing Co. Ltd., Tokyo, pp. 372.

Hayashi, K. 1972. Ecological studies on the useful Mollusca in Lake
Biwa (First Part). Japanese Journal of Malacology (Venus) 19:
9-34.

Hayashi, Y. 1956. On the variation of Corbicula due to environmen-
tal factors Japanese Journal of Malacology (Venus) 19: 54-60.

Ikematzu, W. and M. Kammakura. 1975. Ecological studies of Cor-
bicula leana Prime. I. On the reproductive season and growth.
Bulletin of the Faculty of Agriculture, University of Miyazake
22: 185-195.

Ikematzu, W. and S. Yamane. 1977. Ecological studies of Corbicula
leana Prime. III. On spawning throughout the year and fertili-
zation in the gonad. Bulletin of the Japanese Society of Scien-
tific Fisheries 43: 1139-1146.

Issatullaev, Z. 1980. Bivalve mollusks of the family Corbiculidae in
central Asia. Zoologicheskii Zhurnal 59: 1130-1136.

Itasaka, O., M. Sugita and T. Hori. 1980. Distribution of some shellfish
in the Seta River, Japan. Memoirs of the Faculty of Educa-
tion, Shiga University: Natural Sciences 30: 9-24.

Itasaka, O., M. Sugita, Y. Okamura and T. Hori. 1980. On the water
quality of Lake Biwa-ko and the inhabitability of the bivalve
Corbicula sandai in the Seta River. Memoirs of the Faculty of
Education, Shiga University: Natural Sciences 20: 18-34.

Ituarte, C. F. 1 981 . Primera noticia de la introduccion de pelecipodos
asiaticos en el area rioplatense (Mollusca: Corbiculidae).
Neotropica 27: 79-82.

Johnson, R. I. 1959. The types of Corbiculidae and Sphaeriidae
(Mollusca: Pelecypoda) in the Museum of Comparative
Zoology and bio-bibliographic sketch of Temple Prime, an ear-
ly specialist of the group. Bulletin of the Museum of Com-
parative Zoology, Harvard University 120: 431-479, pis. 1-8.

Kado, Y. and O. Matsushima. 1976a. Some properties of alkaline
phosphatase in tissues of brackish and freshwater clams Cor-
bicula japonica and Corbicula leana. Journal of Science,
Hiroshima University, Series B, Division 1 (Zoology) 26: 139-
148.

Kado, Y. and O. Matsushima. 1976b. Separation of alkaline
phosphatase isoenzymes in digestive diverticula of brackish
and freshwater clams Corbicula japonica and Corbicula leana.
Journal of Science, Hiroshima University, Series B, Division
1 (Zoology). 26: 149-158.

MORTON: CORBICULA IN ASIA

123

Kado, Y. and H. Murata. 1974. Responses of brackish and freshwater
clams, Corbicula japonica and C. leana, to variations in salinity.
Journal of Science, Hiroshima University, Series B Division
1 (Zoology) 25: 217-224.

Kaurajiri, M. 1948. On the young of Mashijimi, Corbicula leana. Nip-
pon Suisan Gakkaishi 14: 17-22.

Kenmuir, D. H. S. 1980. The mussel resources of Lake Kariba, Zim-
babwe. Transactions of the Zimbabe Scientific Association 60:
7-10.

Kraemer, L. R. 1978. Corbicula fluminea (Bivalvia: Sphaeriacea): the
functional morphology of its hermaphroditism, Bulletin of the
American Malacological Union for 1978, pp. 40-49.

Kraemer, L. R. and S. Lott. 1977. Microscopic anatomy of the visceral
mass of Corbicula (Bivalvia: Sphaeriacea). Bulletin of the
American Malacological Union for 1977, pp. 48-56.

Kurashige, E. 1945a. Ecology and morphological study of Korean
Corbicula. Part 4. Distribution, ecology and shell shape of the
Kiiro Shijimi Corbicula felnouilliana Heude. Venus 14: 73-78.

Kurashige, E. 1945b. Ecology and morphological study Korean Cor-
bicula. Part 5. Distribution, ecology and shell shape of the Kiiro
Shijimi Corbicula fluminea (Muller) Venus 14: 87-96.

Kuroda, T. 1938. Studies on Japanese Corbicula. Venus 8: 21-36.

Lee, E. H. and W. D. Heo. 1980. Taste compounds of Corbicula
elatior. Bulletin of National Fisheries, University of Busan
(Natural Sciences) 29: 31-46.

Lee, K. Y. and S. C. Park. 1974. Studies on LDI1 isoenzymes, RNase
and RNase inhibitor activities in varying animals including
parasites and cancer tissues. Seoul Journal of Medicine 14:
224-234.

Lee, T. Y. and E. Y Chung. 1980. Reproductive cycle of the marsh
clam Corbicula fluminea (V. Muller). Publications of the Institute
of Marine Science, National Fisheries, University of Busan 12:
47-54.

Liu, Y. Y. and Y. Y. Huang. 1964. Notes on freshwater mollusks of
the San-Men-Hsia reservoir on the Yellow River and its
neighbouring regions, China. Acta Zoologies Sinica 16:
429-439.

Liu, Y. Y„ W. Z. Zhang, Y. X. Wang and E. Y. Wang. 1979. An En-
cyclopedia of the Economically Important Animals of China -
Freshwater Molluscs. Peking: Scientific Press, p. 134 (in
Chinese).

Lomte, V. S. 1971 . Studies on the freshwater mussels of Marathwada
region. Marathwada University Journal of Science, Section B,
Biological Sciences 10: 143-154.

Mandahl-Barth, G. 1954. The freshwater Mollusca of Uganda and
adjacent territories, Annales du Musse Royal du Congo Beige,
Tervuren, Belgique. Series 8, Science Zoologiques 32:
121-192.

Mandryka, O. N. 1981 . Populations of the bivalve Corbicula japonica
from brackish water lakes on the shore of the Sea of Japan.
Vestn. Leningrad University (Biology). 3: 18-25. (In Russian)

Maru, K. 1 981 . Reproductive cycle of the brackish water bivalve Cor-
bicula japonica in Lake Abashiri, Japanese Scientific Reports
of the Hokkaido Fish Experimental Station 23: 83-96.

Matsushima, O. 1980. The efflux and tissue of ninhydrin positive
substances in brackish and freshwater clams, Corbicula
japonica and C. leana, with special reference to osmotic
responses. Annotationes Zoologices Japonenses 53: 77-88.

Miller, R. C. and F. A. McClure. 1 931 . The freshwater clam industry
of the Pearl River. Lingnan Science Journal 10: 307-322.

Miyazaki, I. 1936. On the development of bivalves belonging to the
genus Corbicula. Bulletin of the Japanese Society of Scientific
Fisheries 5: 249-254.

Mizuno, T. and S. Mori. 1970. Preliminary hydrobiological survey
of some southeast Asian inland waters. Biological Journal of
the Linnean Society of London 2: 77-117.

Mori, S. 1978. Seventh report of the regular limnological survey of
Lake Biwa (1973). III. Benthos. Memoirs of the college of
Science, Kyoto University, Series B 7: 11-29.

Morton, B. S. 1973. Analysis of a sample of Corbicula manilensis
Philippi from the Pearl River, China. Malacological Review 6:
35-37.

Morton, B. S. 1977a. The population dynamics of Corbicula fluminea
(Bivalvia: Corbiculacea) in Plover Cove Reservoir, Hong Kong.
Journal of Zoology, London 181: 21-42.

Morton, B. S. 1977b. The hypobranchial gland in the Bivalvia. Cana-
dian Journal of Zoology 55: 1225-1234.

Morton, B. S. 1977c. The occurrence of inflammatory granulomas
in the ctenidial marsupium of Corbicula fluminea (Mollusca:
Bivalvia); a consequence of larval incubation. Journal of In-
vertebrate Pathology 30: 5-14.

Morton, B. S. 1979a. Freshwater fouling bivalves. In, Britton, J. C.,
J. S. Mattice, C. E. Murphy and L. W. Newland, eds. pp. 1-14.
Proceedings of the First International Corbicula Symposium,
Texas Christian University Research Foundation, Fort Worth,
TX.

Morton, B. S. 1979b. Corbicula in Asia. In, Britton, J. C., J. S. Mat-
tice, C. E. Murphy, and L. W. Newland, eds. pp. 15-38. Pro-
ceedings of the First International Corbicula Symposium, Texas
Christian University Research Foundation, Fort Worth, Tx.

Morton, B. S. 1982. Some aspects of the population structure and
sexual strategy of Corbicula cf. fluminalis (Bivalvia: Cor-
biculacea) from the Pearl River, People’s Republic of China.
Journal of Molluscan Studies 48: 1-23.

Morton, B. S. 1983. The sexuality of Corbicula fluminea in lentic
and lotic waters in Hong Kong. Journal of Molluscan Studies
49 (1): 81-83.

Morton, B. S. 1985. The reproductive strategy of the mangrove
bivalve Polymesoda (Geloina) erosa (Bivalvia: Corbiculacea)
in Hong Kong. Malacological Review. 18: 83-89.

Mouthon, J. 1981 . Sur la presence en France et an Portugal de Cor-
bicula (Bivalvia: Corbiculidae) Originaire d’Asie. Basteria
45: 109-116.

Mudkhede, L. M. and R. Nagabushanam. 1977. The heat tolerance
of the freshwater clam Corbicula regularis. Marathwada
University Journal of Science 16: 151-154.

Nakao, S. 1982. Community structure of the macrobenthos in the
shallow waters in northern Japan. Memoirs of the Faculty of
Fisheries, Hokkaido University 28: 225-304.

Oyama, K. 1943. Study on east Asian Corbicula family. 2. On Korean
Corbicula Japanese Journal of Malacology (Venus) 1 2: 1 50-1 58.

Parodiz, J. J. and L. Hennings. 1965. The Neocorbicula (Mollusca,
Pelecypoda) of the Parana - Uruguay basin, South America.
Annals of the Carnegie Museum 38: 69-96.

Prashad, B. 1924. Zoological results of a tour in the Far East. Revi-
sions of the Japanese species of the genus Corbicula. Memoirs
of the Asiatic Society of Bengal 4: 522-529.

Prashad, B. 1928a. Revision of the Asiatic species of the genus Cor-
bicula. I. The Indian species of Corbicula. Memoirs of the In-
dian Museum 9: 13-27.

Prashad, B. 1928b. Revision of the Asiatic species of the genus Cor-
bicula. II. The species of the genus Corbicula. Memoirs of the
Indian Museum 9: 29-48.

Prashad, B. 1929. Revision of the Asiatic species of the genus Cor-
bicula. III. The Indo-Chinese species of the genus Corbicula
from China south-eastern Russia, Tibet, Formosa and the

124

CORBICULA SYMPOSIUM

Philippine Islands. Memoirs of the Indian Museum 9: 49-68.

Prashad, B. 1930. Revision of the Asiatic species of the genus Cor-
bicula. IV. The species of the genus Corbicula from the Sun-
da Islands, the Celebes and New Guinea. Memoirs of the In-
dian Museum 9: 193-203.

Ray, H. C. 1967. A new species of freshwater clam C. krishneae from
Maharashtra India. Archiv fur Molluskenkunde 96: 191-193.

Sinclair, R. M. and B. G. Isom. 1963. Further studies on the introduc-
ed Asiatic clam Corbicula in Tennessee. Tennessee pollution
Control Board, Nashville, 76 pp.

Tamura, T. 1959. Propogation of Corbicula. In, Science of Aqua-
culture 4th Edition, Chapter 21.

Tchang, S. and S. C. Li. 1965. Bivalves (Mollusca) of the Poyang
Lake and its surrounding waters, Kiangsi Province, China, with
description of a new species. Acta Zoologica Sinica 17:
309-318.

Tchang, S., S. C. Li and Y. Y. Liu. 1965. Bivalves (Mollusca) of Tung-
Ting Lake and its surrounding waters, Hunan Province, China.
Acta Zoologica Sinica 17: 197-213.

Tchernov, E. 1975. The Molluscs of the Sea of Galilee. Malacologia
15: 147-184.

Temcharoen, P. 1971. New aquatic molluscs from Laos. Archiv fur
Molluskenkunde 101: 91-109.

Utoh, U. 1981. Growth of the brackish water bivalve Corbicula
japonica in Lake Abashiri, Japan. Scientific Reports of the Hok-
kaido Fishery Experimental Station 23: 65-82.

Villadolid, D. V. and F. G. Del Rosario. 1930. Some studies on the
biology of Tulla {Corbicula maniilensis Philippi), a common food
clam of Laguna de Bay and its tributaries. Philippine
Agriculturist 19: 355-382.

Wu, W. L. 1979. Morphometric studies on Corbicula fluminea (Muller
1774) (Bivalvia: Corbiculidae). Bulletin of Malacology, Republic
of China 6: 1-12.

Wu, W. L. 1980. The list of Taiwan bivalve fauna. Quarterly Journal
of the Taiwan Museum 33: 55-208.

Yaroslavtseva, L. M., V. A. Pavlenko and S. V. Fedoseeva. 1981.
Correlation between cellular resistance of some marine
mollusks to dilution of seawater and their ability of salinity ac-
climation. Biologiya Morya (Vladivostock) 1: 54-60.

Zhadin, V. I. 1948. On the origin of the fauna and biocenoses of
U.S.S.R. inland waters. In: Sbornik Pamyati Akademika S.A.
Zernova. Izdatel’stvo AN SSSR. pp. 56-71.

Zhadin, V. I. 1965. Mollusks of fresh and brackish waters of the
U.S.S.R. Academy of Sciences of the Union of Soviet Socialist
Republics.

ELECTROPHORETIC VARIATION
IN NORTH AMERICAN CORBICULA

MICHAEL J. MCLEOD
BIOLOGY DEPARTMENT
BELMONT ABBEY COLLEGE
BELMONT, NORTH CAROLINA 28012, U.S.A.

ABSTRACT

An electrophoretic study of 16 populations of Corbicula has revealed genetic variation in all
but two populations. This study used starch gel electrophoresis to analyze 14 enzymes encoded by
21 loci. There was a high degree of similarity between most populations at most loci. There were,
however, some loci, notably CAT, where eastern and western populations had a different allele at
greatest frequency. There were also unique alleles, present at relatively high frequencies, in single
populations. These data suggest 1) the possibility of multiple introductions of Corbicula into this country
and, 2) the existence of several genetic races within any given population. There is also evidence
for two species of Corbicula in this country. These data have possible importance to the success of
control measures.

Corbicula is of interest to population geneticists and
systematists for several reasons. It is a relatively recent in-
troduction to North America, the earliest recorded collections
being 1924 in British Columbia (Counts, 1981) and in 1938
in Washington state (McMahon, 1982). Corbicula is now
almost universally found throughout the major river systems
south of 40° latitude. It has, then, in roughly 45 years migrated
across the United States. In doing so, these clams have had
to successfully conform to a large variety of environmental
conditions. The assumption has been that both the initial in-
troduction and the subsequent migrations represent founder
events (Smith et al., 1979). The genetic structure of popula-
tions immediately following the bottleneck induced by a
founder event and the rate at which variation is reintroduced
into such populations are both important to our overall
understanding of colonization as well as population genetic
theory. Corbicula is an ideal system for investigating the
genetics of founder events, partly because the timing of its
spread has been so well documented (McMahon, 1982).

We do not know the reproductive strategy employed
by Corbicula in this country. Recently, based on histological
evidence, Kraemer (1979), Kraemer and Lott (1978), and Mor-
ton (1 982) have suggested that North American Corbicula are
simultaneous hermaphrodites and at least potentially self-
fertilizing. If true, this would be interesting since there have
been very few studies of founder events involving self-
fertilizing species. In any event, this clam has a high fecun-
dity (Aldridge and McMahon, 1978) and can quickly establish
itself in favorable habitats. It is a weed species ( sensu Harlan,
1965) and a highly opportunistic organism. One objective of
the study described in this paper was to explore the amount
and pattern of genetic variation in populations of North

American Corbicula.

A second major question that was addressed here con-
cerns the number of species of Corbicula in North America.
The conclusion of the First International Corbicula Symposium
was that only one species had been introduced (Britton and
Morton, 1979). This conclusion has recently been challenged
based on both electrophoretic and morphological evidence
(Hillis and Patton, 1982).

There have been three previous published elec-
trophoretic studies of North American Corbicula. The first
(Smith et al., 1979) surveyed five populations in the U.S. as
well as five Asian populations. They reported no variation
within or among U.S. populations although they did find some
variation in clams in Asia. Hillis and Patton (1982) surveyed
populations from the Brazos River, Texas, and likewise found
no variations within each species, although they suggested
that two species were present. McLeod and Sailstad (1980)
collected a single population from the Catawba River, NC,
monthly for one year. They reported genetic variation in the
population at three of the seven loci examined.

MATERIALS AND METHODS

Clams from 15 populations were examined using
horizontal starch gel electrophoresis. The populations were
as follows: TOL = Lake Erie (Toledo), Ohio; DAY = Great
Miami River (Dayton), Ohio; CHA = Catawba River
(Charlotte), NC; GTF = Wateree (Catawba) River (Great
Falls), SC; SAN = Wateree (Catawba) River (Santee), SC;
WIL = Lake Waccamaw (Wilmington), NC; PUG =
Caloosahatchee River (Punta Gorda), FL; CAD = Little River
(Cadiz), KY; DGL = DeGray Lake, AR; LOP = Lake of the

American Malacological Bulletin, Special Edition No. 2(1 986): 125-1 32

125

126

CORBICULA SYMPOSIUM

Pines, TX; LFF = Lake Fairfield, TX; AUS = Colorado River
(Austin), TX; DR-1, DR-2 = Pinto Creek (Del Rio), TX; VVA
= Verde Valley, AZ; RVC = Sacramento River (Rio Vista),
California. Sample size ranged from 18-43, depending mostly
on survival during transit. There was no indication of differen-
tial survival rates between the two morphs. Individuals from
the Charlotte, NC population were included on every gel for
reference.

Clams (whole bodies) were homogenized in an equal
volume of cold 0.5 M tris HC1, pH 7.1 buffer (Hornbach et
al., 1980) and the samples stored at -45C until run (usually
not more than 24 hours). Twenty-one loci were resolved us-
ing the methods of Selander et al. (1971) and Ayala et al.
(1972), except for octopine dehydrogenase. The stain for oc-
topine dehydrogenase was 20ml 0.2 tris HC1, pH 8, 30 mg
octopine, 2ml NAD, 2ml MTT, 0.5 ml PMS. The loci and buf-
fer systems employed were as follows: discontinuous
borate/tris-citrate buffer (Poulik, 1957), phosphoglucose
isomerase (GI-1, -2), phosphoglucose mutase (PGM), oc-
topine dehydrogenase (OCT-1 , -2); tris-borate buffer, pH 9.4
(Ayala et al., 1972), malic enzyme (ME), glutamate ox-

aioacetate transaminase (GOT-1 , -2), total protein (TP-1 , -2,
-3), leucine aminopeptidase (Lap -1, -2), 6-phosphoglucose
dehydrogenase (6-PGDH); tris-maleate EDTA buffer
(Selander et al., 1971), malate dehydrogenase (MDH-1 , -2),
isocitrate dehydrogenase (IDH-1, -2), catalase (CAT), xanthine
dehydrogenase (XDH), x-glycerophosphate dehydrogenase
(X-GPDH).

Estimate of genetic distance, D, (Nei, 1972; 1978) and
a cluster analysis (unweighted pair group method, Sneath
and Sokal, 1973), were calculated using the BIOSYS-1 com-
puter program of Swofford and Selander, (1981).

RESULTS

The generally accepted methods of interpreting gels
were followed in this study. The general methodology and
assumptions involved in interpreting banding patterns on gels
and distinguishing monomeric and dimeric proteins, as well
as identifying homozygous and heterozygous individuals have
been extensively discussed (Scandalios, 1969; Manwell and
Baker, 1970; Tracey et al., 1975; Crawford and Wilson, 1977).

Table 1. Allele frequencies at each of the polymorphic loci. Populations are listed in the same order as in Table 2 and abbreviations are
explained in the methods section. The n at PGI-1 designates a null allele and indicates that the locus was not resolved.

POPULATION

Locus1

allele

TOL

DAY

CHA

GTF

SAN

WIL

PUG

CAD

DGL

LOP

LFF

AUS

RVC

DR-1

DR-22 VVA2

PGM

0.10

1.00

0.30

1.00

0.90

1.00

0.57

1.00

0.70

1.00

0.43

LAP-1

0.07

0.24

1.00

0.93

1.00

0.52

1.00

0.76

1.00

0.48

LAP-2

0.05

0.10

0.78

1.00

0.95

1.00

0.90

1.00

0.12

1.00

GOT-I

0.05

0.10

0.11

0.09

1.00

0.95

0.90

0.89

1.00

0.91

0.16

0.84

GOT-2

1.00

0.03

1.00

0.97

PGI-1

1.00

0.97

1.00

PGI-2

1.00

0.97

1.00

0.03

1.00

CAT

0.68

1.00

0.97

1.00

0.93

1.00

0.24

0.10

0.32

0.03

0.07

1.00

0.76

1.00

0.97

1.00

0.90

0.82

0.03

0.18

1.00

6-PGDH

1.00

0.38

1.00

0.62

1.00

0.97

1.00

0.03

XDH

0.18

0.82

1.00

1 Locus designations are PGM = phosphogulcomutase, LAP = leucine aminopeptidase, GOT = glutamate oxaloacetate transaminase, PGI
- phosphoglucose isomerase, CAT = catalase, 6-PGDH = 6 phosphoglucose dehydrogenase, ME = malic enzyme (NADP-dependent
malate dehydrogenase), XDH = xanthine dehydrogenase.

2These two populations are the purple morph. All other populations are the white morph.

MCLEOD: ELECTROPHORETIC VARIATION IN CORBICULA

127

Fig. 1. Geographic variation in allele frequencies at the catalase (CAT) locus in Corbicula. The CAT-a allele is predominant in the eastern
population and the CAT-b allele is the major allele in the west.

Therefore it is not necessary to repeat those discussions here.

Zones of activity on a gel, which varied independent-
ly of other such zones of activity, were considered to be en-
coded by a single gene locus. The fastest anodaily migrating
zone of activity was arbitrarily designated as the first locus
encoding a particular enzyme, the next fastest as the second
locus, and so on. Within a zone of activity, the most anodal
band was designated the a allele, the next fastest the b allele,
and so on until all bands in that zone were identified.

The quaternary structure of the proteins examined in
Corbicula, at least where it could be determined by banding
patterns of heterozygotes, was what would be expected for
those molecules. There were very few heterozygotes found
and only a limited number of loci were heterozygous.
Specifically, heterozygotes were only found at LAP-1 , LAP-2,
CAT, PGM and PGI-2 and the banding patterns for
heterozygotes at each locus was consistent with what has
been reported in the literature (Scandalios, 1969; Selander
eta!., 1971; Ferguson, 1980). For example, PGM had the dou-
ble banded pattern expected of a monomeric molecule. The
only exception to the expected patterns was at CAT, where
the heterozygotes were double banded although CAT has
been shown in mice to be a tetramer (Hoffman and
Grieshaber, 1976).

There were distinct morphological types examined in

this study, corresponding to the white and purple forms of
Hillis and Patton (1982) and Fontanier (1982). The purple form
was found as a distinct population in Verde Valley, Arizona,
and was sympatric with the white morph at Del Rio, TX. No
microhabitat difference such as was reported by Hillis and
Patton (1982) was observed in the Del Rio population. The
other 13 populations were exclusively the white morph. The
white morph is the form that has been considered C. fluminea
and I will refer to it as such. A total of 21 loci (10 monomor-
phic and 1 1 polymorphic) were resolved in the purple morph.
The PGI-1 locus was never resolved in specimens of the white
C. fluminea. Because of this lack of activity, even when white
morph specimens were run on the same gel as purple morph
individuals which stained for PGI-1, the PGI-1 locus in the
white morph was considered to be coded for by a null allele
(Manwell and Baker, 1970; Ferguson, 1980). PGI-1 provid-
ed a marker between the two forms. There was also a fixed
difference at ME, with the purple form being monomorphic
for the ME-a allele and C. fluminea monomorphic for the ME-b
allele. There was a frequency difference at PGM between the
two morphs (Table 1).

Electrophoretic variation was present in every popula-
tion except Cadiz, KY, and Punta Gorda, FL. The variant loci
were not identical in all populations (Table 1). For example,
all populations were monomorphic for the PGI-1 a allele ex-

128

CORBICULA SYMPOSIUM

cept Lake Fairfield which was fixed for PGI-2b allele. Most
populations were monomorphic for GOT-b. Three Catawba
(Wateree) River populations, however, had the GOT-1 a allele
at a frequency of about 0.1 . The purple morph from Del Rio,
TX, also had the GOT-la allele. There was one locus, CAT,
where the two major alleles were segregated geographically
into eastern and western populations (Figure 1). The western
populations mostly had the CAT-b allele, while the eastern
populations had the CAT-a allele at greatest frequency.

There was, actually, very little variation in any single
population. The percent of polymorphic loci (99% criterion)
per population ranged from 0 to 19% in C. fluminea and up

Table 2. Genetic variation in each population of Corbicula . Del Rio-1
is C. fluminea, Del Rio-2 and Verde Valley are the purple morph.

Population

% Poly-
morphism1

% Hetero-
zygosity

Mean number
of alleles/
locus

Toledo

9.5

0.37

1.09

Dayton

4.8

1.05

Charlotte

19.0

0.13

1.19

Great Falls

14.3

0.25

1.14

Santee

4.8

1.05

Wilmington

4.8

1.05

Punta Gorda

1.00

Cadiz

1.00

DeGray Lake

4.8

1.05

Lake of Pines

4.8

0.25

1.05

Lake Fairfield

4.8

1.05

Austin

9.5

1.09

Rio Vista

14.3

1.14

Del Rio-1

4.8

1.05

Del Rio-2

22.7

0.25

1.23

Verde Valley

4.8

1.05

1 99% criterion

to 22.8% in the purple morph (Table 2). The percen-
tage of polymorphic loci did not appear to be biased by sam-
ple size (Table 2). The mean number of individuals
heterozygous at a locus (averaged over all loci) in a popula-
tion was extremely low. Only five populations had any
heterozygous individuals (Table 2), and heterozygosity in
those varied from 0.13% to 0.37%. Heterozygosity at a par-
ticular locus (b) is defined as h = 1 -E Xp, where Xj is the
frequency of the / th allele, and average heterozygosity (H,
the value reported in this paper) is the mean of h over all loci
in a population.

Nei’s (1972; 1978) standard genetic distance is
generally accepted as representing the average number of
codon substitutions per gene, detected electrophortically,
since two populations diverged (Ayala ef a/., 1975; Nei, 1976;
Thorpe, 1982). The genetic distance, D, between populations
of C. fluminea (Table 3), indicate that these populations have
diverged from each other to varying extents (range of D =
0.001 - 0.185). The C. fluminea populations are, however,
closer to each other than to the two populations of the pur-
ple morph (Table 3). There is also a rough separation of the
C. fluminea populations into two geographic groups; an
eastern group and a west-southwest group. This relationship
is evident in the dendrogram (Fig. 2) drawn from the genetic
distance between populations.

DISCUSSION

In the first electrophoretic survey of Corbicula ,
Smith ef a/. (1979) found no variation in the LJ.S. popu-
lations but did see variation in Asian populations. I used
somewhat different techniques and also surveyed for some
different enzymes and did find low levels of electrophoretic
variation in most U.S. populations. The different results are
not necessarily surprising in that different chemical condi-
tions, such as changes in pH, can influence protein mobility
on an electrophoretic gel. Similarities on gels may not be real,

Table 3. Nei's Genetic distance between populations of Corbicula. Numbers at top of Table correspond to populations in the same order
as at the left. Del Rio-2 and Verde Valley are populations of the purple morph.

Population

Dayton

Toledo

0.030

Charlotte

0.071

0.056

Santee

0.089

0.073

0.060

Wilmington

0.069

0.056

0.001

0.060

Punta Gorda

0.069

0.056

0.001

0.060

0.001

DeGray L.

0.123

0.072

0.047

0.113

0.049

L. of Pines

0.071

0.028

0.034

0.081

0.035

0.009

L. Fairfield

0.176

0.126

0.100

0.170

0.101

0.050

0.060

Austin

0.173

0.121

0.094

0.163

0.095

0.046

0.055

0.098

Del Rio-1

0.090

0.119

0.089

0.158

0.090

0.050

0.057

0.097

0.098

Del Rio-2

0.185

0.143

0.224

0.131

0.230

0.180

0.150

0.244

0.240

0.220

Great Falls

0.024

0.006

0.050

0.065

0.050

0.101

0.049

0.154

0.151

0.145

0.168

Rio Vista

0.160

0.107

0.079

0.149

0.082

0.030

0.040

0.082

0.079

0.083

0.213

Verde Valley

0.284

0.246

0.191

0.147

0.196

0.197

0.257

0.249

0.251

0.134

0.237

MCLEOD: ELECTROPHORETIC VARIATION IN CORBICULA

129

although differences almost always are so and thus elec-
trophoresis tends to underestimate variation (Ayala, 1982).
Differences in results between Smith et at. (1979) and this
present study may also be a reflection of the number of loci
sampled and also different number of populations sampled
in the two studies.

Variation was observed at different loci in different
populations. For example, the XDH-a allele was found only
in the Toledo population and was present there at a frequency
of 0.18 (Table 1). There are two possible explanations to ac-
count for these apparently unique alleles. The first is that
these alleles really exist in a number of populations at very
low frequency and have only been detected where random
drift has acted to increase the frequency of a particular allele.
This hypothesis requires that the original bottleneck (introduc-
tion to the U.S.) was relatively large and non-restrictive.
Genetic drift has been frequently cited as a mechanism to
explain differences in gene frequencies between populations
(Spiess, 1977; Beaumont, 1977; Beaumont, 1982; Lieb etal.,

1983). A second possibility is that these apparently unique
alleles arose in the particular population in which they were
found by mutations. Most of the populations surveyed had
been established for less than 15 years when they were
sampled. While it can be argued that the short period of time
that the populations have been in existence necessitates a
high de novo mutation rate for Corbicula, this is not the case.
If one assumes a similar mutation rate for allozymes in Cor-
bicula as has been observed in Drosophila (1.28x 10'6,
Voelker et al., 1980), and makes reasonable assumption of
genome size (104 genes), then the number of mutations per
individual is equal to 0.026. Given both the fecundity and the
large population sizes of which Corbicula is capable and the
possibility of 0.026 mutations/individual, the detection of one
new allele in a population which has been extant for 10-15
years (20-30 generations) does not seem unreasonable.
Therefore, an abnormally high mutation rate is not necessary
to account for the unique alleles. However, either relaxation
of selection, such as that proposed by Carson (1975) in his

Fig. 2. Dendogram constructed from standard genetic distance estimates (Nei, 1972) using an unweighted pair-group method. Note that
the purple morph populations (marked with an *) are distinct from the white morph populations.

130

CORBICULA SYMPOSIUM

founder flush-crash speciation theory or strong selection
favoring the allele would seem to be important to allow the
mutations to increase to the frequencies reported here (Table
1).

The pattern of variation (low to moderate polymor-
phism and little or no heterozygosity) is the same as been
observed in facultative self-fertilizing species (Selander and
Hudson, 1976; McLeod etal., 1981). In these species a series
of monomorphic races become established in a population
with only infrequent cross-fertilization between races. These
monomorphic races allow a population to maintain both
genetic variability in case of an environmental perturbation
and also gave large numbers of individuals that are highly
adapted to the current conditions. Hybridization between
races (reflected by the low level of heterozygosity) is both rare
and a chance occurrence. Smith etal. (1979) suggested, and
the data presented here also indicate, that Corbicula
possesses generalist alleles that allow for wide phenotypic
responses to environmental conditions. The fact that Cor-
bicula, with a limited amount of genetic diversity, has been
able to invade a number of different habitats should indicate
that controlling the occurrence of this clam would be difficult.
Corbicula would seem to be so phenotypically plastic that it
can respond physiologically to many control measures. If the
unique alleles do represent mutations then the potential ability
of Corbicula to respond to control measures, as well as the
potential to colonize new areas, increase. This is not meant
to imply that mutations necessarily increase the homeostatic
ability of Corbicula. As is frequently mentioned in textbooks,
most mutations are deleterious (Dobzhansky et al. 1977).
However, deleterious does not mean lethal but instead im-
plies a reduction in fitness from an ideal genotype (Spiess,
1977). It is conceivable that a rare mutation would change
the kinetic properties of a critical enzyme so that the enzyme
could function in the new conditions presented by control ef-
forts or a range extension and the animal would survive.

Control could potentially be even more difficult if the
genetic variation reported here allows for differential response
in a population to specific control measures. If having different
alleles at a particular locus allow a few individuals to survive
and continue the population, then the effect of control has
been to select for a population resistant to that control
measure. This resistance is what has occurred in insect
populations treated with pesticides (Dobzhansky etal., 1977).
Correlations have been found between environmental com-
ponents and allele frequencies at a specific locus in
Drosophila (Steiner, 1979).

In the context of one theme of the symposium (con-
trol of Corbicula ), the question of how many species of Cor-
bicula exist in the U.S. is not entirely academic. There are,
however, several problems which make answering that ques-
tion difficult. One problem is the definition of a species when
dealing with an organism that is capable of self-fertilization
and apparently has a limited amount of outcrossing. The
biological species concept (Mayr, 1970) depends on the ability
or inability of organisms to interbreed. No truly satisfactory
and accepted definition which can be applied to organisms
such as Corbicula has been advanced. All attempts to

distinguish species, whether they are based on comparative
morphology, karyotype, interbreeding ability, behavioral or
physiological differences, or electrophoretic similarity, have
as a basic premise an assumption of underlying genetic dif-
ferentiation. Electrophoresis has been shown to be efficacious
in demonstrating genetic relationships and divergence (Ayala,
1972, 1982; Avise, 1974).

There are a number of studies which compare elec-
trophoretic and morphological similarities between and
among populations or species. As one might expect, in some
studies isozymes and morphology are in close cor-
respondence (Grudzien and Turner, 1983; Bryant, 1984), and
in other studies there is no congruence (Gould et at., 1974;
Hornbach et al., 1980). Examples of convergence in mor-
phology, but distinctness in isozymes between species are
known (McLeod etal., 1980; Zimmerman and Nejtek, 1977).
Enzyme electrophoresis differs from morphology in terms of
providing systematic data. Morphological characters are often
controlled by several to many genes and alleles at these loci
may influence the phenotype in the same way so that a large
number of genotypes can result in the same phenotype (Gott-
lieb, 1977). The phenotype in electrophoresis is represented
by colored bands on a gel that indicate areas where an en-
zyme has catalyzed a particular reaction. Difference in mobili-
ty on gels are the result of changes in the gene coding for
the polypeptide and so the mobility differences are a result
of genetic differences (Ferguson, 1980). While it is true that
electrophoresis has a number of limitations (see Ayala, 1982),
the relationship between genotype and phenotype is reason-
ably straight-forward in electrophoresis especially when com-
pared with morphology (Ferguson, 1980).

The presence of two species was suggested by Hillis
and Patton (1982) based on both morphology and on the
presence of fixed differences at six loci. The data presented
in this study also suggests that there are two species pre-
sent. Fixed differences were found at two loci (PGI-1 and ME)
as well as major frequency difference at PGM. The enzymes
considered here were not completely the same as those used
by Hillis and Patton (1982) and so between their study and
this one fixed differences have been found at eight loci. Nei’s
genetic distance (Table 3) indicates that the two populations
of the purple morph were relatively closely related to each
other (D = 0.135) and were distinct from C. fluminea (D =
0.21). While the presence of heterozygotes in some popula-
tions indicated that some outcrossing does occur, albeit in-
frequently, there were no heterozygotes in the Del Rio white
morph population at PGI-1 or ME. There were heterozygotes
within the purple morph population. If these were a single
species it would not be unrealistic to expect to find hybrids
in a sympatric population where heterozygotes do exist in one
component of the population. There was information
presented at the symposium by Britton, and Schofield and
Britton (see paper in this volume) that indicates that the pur-
ple and white morphs have different juvenile growth rates,
differential physiological responses to potassium and to
sodium thiosulfate, and some segregation into habitats of dif-
ferent water quality. Hillis and Patton (1982) also found dif-
ferences in shell length, width, height, and weight, as well

MCLEOD: ELECTROPHORETIC VARIATION IN CORBICULA

131

as shell color and number of sulcations. Shell morphology,
including color, is notoriously poor for delineating species in
molluscs. In oysters, shell color, size, and individual shell
dimensions are greatly influenced by local environmental fac-
tors (Galtsoff, 1964), and do not reflect electrophoretic rela-
tionship (Groue and Lester, 1982). However, the combina-
tion for morphologic, electrophoretic, physiological and
ecological differences taken together seem substantial. It
does seem pointless and even foolhardy to attach a species
name to the purple morph. Since Corbicula is not endemic
to the U.S., and was introduced from Asia, a much more
thorough survey of Asian species is necessary before tax-
onomic relationships to Asian species can be established.

There are indications in the data that the east coast
and Ohio Valley populations are distinct from the more
western ones (Figs. 1 and 2). This suggests that the eastern
populations have been isolated from the western ones and
that most have probably orginated from other eastern popula-
tions. McMahon (1982) has suggested that east coast popula-
tions were founded from an Ohio River population (at
Paducah, KY) via southern migration along the Mississippi
River. It is possible that the Paducah population was formed
from a California population (McMahon, 1982; Britton and
Morton, 1982) and was isolated for a sufficiently long time
to evolve the frequency differences now seen. It is equally
possible that there were two separate introductions to this
country; one on the west coast and a second into the Ohio
River. Without a more complete survey of Asian populations
it is impossible to choose between these two options. If the
eastern populations do represent a second successful in-
troduction then the presence of western population alleles
at loci like CAT (Fig. 2) may represent recruitment from those
populations through migration. It may also represent con-
vergence through mutation and genetic drift.

There is, then, genetic variation in most populations
of Corbicula examined in this study and, by extrapolation, in
North America. There may also have been two successful
introductions of C. fluminea, thus increasing the potential
gene pool of U.S. Corbicula, as well as an introduction of a
second species (possibly C. fluminalis ?, Morton, 1977). The
presence of genetic races in a population, as well as the ex-
istence of two species, increases the problem of controlling
the occurrence of these organisms.

ACKNOWLEDGEMENTS

This study was supported by a Cottrell College Science Grant
from Research Corporation. The following people supplied clams
used in this study: Joe Britton, Chris Foe, Robert Rutter, Jennifer
Scott-Wasilk, and Jim Sickel. The help of Mike Romano in the field
and Jenny Rowland and Warren Murray in the laboratory is gratefully
acknowledged. Dr. Sheldon Guttman ran the computer analysis of
the data.

LITERATURE CITED

Aldrich, D. W. and R. F. McMahon. 1978 Growth, fecundity, and
bioenergetics in a natural population of the Asiatic freshwater
clam, Corbicula manilensis Philippi, from north central Texas.
Journal of Molluscan Studies 44: 49-70.

Avise, J. C. 1974. Systematic value of electrophoretic data.
Systematic Zoology 23: 465-481.

Ayala, F. J. 1975. Genetic differentiation during the speciation pro-
cess. Evolutionary Biology 8: 1-78.

Ayala, F. J. 1982. The genetic structure of species, pp. 60-82, IN:
R. Milkman, ed., Perspectives on Evolution, Sinauer
Associates Sunderland, MA.

Ayala, F. J., J. R. Powell, M. L. Tracey, C. A. Mourao and S. Perez-
Salas. 1972. Enzyme variability in the Drosophila willistoni
group. IV. Genic variation in natural populations of Drosophila
willistoni. Genetics 70: 113-139.

Ayala, F. J., M. L. Tracey, D. Hedgecock and R. C. Richmond. 1975.
Genetic change during the speciation process in Drosophila.
Evolution 28: 576-592.

Beaumont, A. R. 1982. Geographic variation in allele frequencies
at three loci in Chlamys opercularis from Norway to the Brit-
tany coast. Journal of the Marine Biological Association of the
United Kingdom 62: 243-261.

Britton, J. C. and B. Morton. 1979. Corbicula in North America: The
evidence reviewed and evaluated, pp. 249-287, IN: J. C. Brit-
ton, ed., Proceedings of the First International Corbicula Sym-
posium. Texas Christian University Research Foundation, Fort
Worth, TX.

Britton, J. C. and B. Morton. 1982. A Dissection guide, field and
laboratory manual for the introduced bivalve Corbicula
fluminea. Malacological Review, Supplement 3: 1-82.

Bryant, E. H. 1984. A comparison of electrophoretic and mor-
phometric variability in the face fly, Musca autumnalis. Evolu-
tion 38: 455-458.

Carson, H. L. 1975. The genetics of speciation at the diploid level.
The American Naturalist 109:83-92.

Counts, C. 1., III. 1981. Corbicula fluminea (Bivalvia: Sphaeriacea)
in British Columbia. Nautilus 95: 12-13.

Crawford, D. J. and H. D. Wilson. 1977. Allozyme variation in
Chenopodium fremontii. Systematic Botany 2: 180-190.

Dobzhansky, T., F. J. Ayala, G. L. Stebbins, and J. W. Valentine.
1977. Evolution. W. H. Freeman, San Francisco.

Ferguson, A. 1980. Biochemical Systematics and Evolution. Blackie
and Son, Glasgow, U.K.

Fontanier, C. E. 1982. The distribution of Corbicula (Bivalvia: Cor-
biculidae) in the Brazos River system, Texas, 25 August-12
November, 1980. Texas Journal of Science 34: 5-15.

Galtsoff, P. S. 1964. The American Oyster Crassostrea virginica
Gmelin U.S. Fish and Wildlife Service Fisheries Bulletin 64:
1-480.

Gottlieb, L. D. 1977. Electrophoretic evidence and plant systematics.
Animals of the Missouri Botonical Garden 64: 161-180.

Gould, S. J., D. S. Woodruff and J. P. Martin. 1974. Genetics and
morphometries of Cerion at Pongo Carpet: A new systematic
approach to this enigmatic land snail. Systematic Zoology 23:
518-535.

Groue, K. J. and L. J. Lester. 1982. A morphological and genetic
analysis of geographic variation among oysters in the Gulf of
Mexico. The Veliger 24: 331-335.

Grudzien, T. A. and B. J. Turner. 1983. Biochemical systematics of
Allodontichthys: A genus of goodeid fishes. Biochemical
Systematics and Ecology 11: 383-388.

Harlan, J. R. 1965. The possible role of weed species in the evolu-
tion of cultivated plants. Euphytica 14: 173-176.

Hillis, D. M. and J. C. Patton. 1982. Morphological and electorphoretic
evidence for two species of Corbicula (Bivalvia: Corbiculidae)
in North America. American Midland Naturalist 108: 74-80.

Hoffman, H. A., and C. K. Grieshaber. 1976. Genetic studies of

132

CORBICULA SYMPOSIUM

Murine catalase: Regulation of multiple molecular forms of
Kidney catalase. Biochemical Genetics 14: 59-66.

Hornbach, D. J., M. J. McLeod, S. I. Guttman and S. K. Seilkop.
1980. Genetic and morphological variation in the freshwater
clam, Sphaerium (Bivalvia: Sphaeriidae). Journal of Molluscan
Studies 46: 158-170.

Kraemer, L. R. 1979. Corbicula fluminea (Bivalvia: Sphaeriacea): The
functional morphology of its hermaphroditism. Bulletin of the
American Malacological Union For 1979: 40-49.

Kraemer, L. R. and S. Lott. 1978. Microscopic anatomy of the viscera!
mass of Corbicula (Bivalvia: Sphaeriaceae). Bulletin of the
American Malacological Union For 1977: 48-56.

Lieb, C. S., D. G. Buth, and G, C. Gorman. 1983. Genetic differen-
tiation in Anolis sagrai: A comparison of Cuban and introduced
Florida populations. Journal of Herpetology 17: 90-94.

Manwell, C., and C. M. A. Baker. 1970. Molecular Biology and the
Origin of Species: Heterosis, Protein Polymorphism and Animal
Breeding. University of Washington Press, Seattle.

Mayr, E. 1970. Populations, Species and Evolution. Belknap Press,
Cambridge, MA.

McLeod, M. J. and D. J. Sailstad. 1980. An electrophoretic study
of Corbicula fluminea (Bivalvia: Corbiculacea) in the Catawba
River. Bulletin of American Malocological Union For 1980:
17-19.

McLeod M. J., D. C. Wynes and S. I. Guttman. 1980. Lack of
biochemical evidence for hybridization between two species
of Darters. Comparative Biochemistry and Physiology, B, 67:
323-325.

McLeod, M. J., and D. J. Hornback, S. I. Guttman, C. M. Way and
A. J. Burky. 1981. Environmental heterogeneity, genetic
polymorphism, and reproductive strategies. American
Naturalist 118: 129-134.

McMahon, R. F. 1982. The occurrence and spread of the introduced
Asiatic freshwater clam, Corbicula fluminea (Muller), in North
America: 1924-1982. Nautilus 96: 134-141.

Morton, B. 1977. The population dynamics of Corbicula fluminea
(Bivalvia: Corbiculacea) in Plover Cover Reservoir, Hong Kong.
Journal of Zoology, London, 181: 21-42.

Morton, B. 1982. Some aspects of the population structure and sexual
strategy of Corbicula cf. fluminalis (Bivalvia: Corbiculacea) from
the Pearl River, People’s Republic of China. Journal of
Molluscan Studies 48: 1-23.

Nei, M. 1972. Genetic distance between populations. American
Naturalist 106: 283-292.

Nei, M. 1976. Mathematical models of speciation and genetic
distance, pp. 723-765, IN: S. Karlin and E. Nevo, eds., Popula-

tion Genetics and Ecology. Academic Press, NY.

Nei, M. 1978. Estimation of average heterozygosity and genetic
distance from a small number of individuals. Genetics 89:
583-590.

Poulik, M. D. 1957. Starch gel electrophoresis in a discontinous
system of buffers. Nature 180: 1477-1479.

Scandalios, J. G. 1969. Genetic control of multiple molecular forms
of enzymes in plants: a review. Biochemical Genetics 3: 37-79.

Selander, R. K., M. H. Smith, S. Y. Yang, W. E. Johnson, and J.
B. Gentry. 1971. Biochemical polymorphism and systematics
in the genus Peromyscus I. Variation in the old-field mouse
(Peromyscus polionotus). University of Texas Publication 7103:
49-90.

Selander, R. K. and R. O. Hudson. 1976. Animal population struc-
ture under close inbreeding: The land snail Rumina in southern
France. American Naturalist 110: 695-718.

Smith, M. H., J. C. Britton, P. Burke, R. K. Chesser, M. W. Smith
and J. Hagen. 1979. Genetic variability in Corbicula, and in-
vading species, pp. 244-248, IN: J. C. Britton, ed., Pro-
ceedings, First International Corbicula Symposium. Texas
Christian University Research Foundation, Fort Worth, TX.

Sneath, P. H. A. and R. R. Sokal. 1973. Numerical Taxonomy. W.
H. Freeman, San Francisco.

Spiess, E. B. 1977. Genes in Populations. John Wiley and Sons, New
York.

Steiner, W. W. M. 1979. Genetic variation in Hawaiian Drosophila.
VIII. Heterozygosity and genetic changes in isolated popula-
tions of D. engyochracea. Biochemical Genetics 17: 645-664.

Swofford, D. C. and R. B. Selander. 1981. BIOSYS-1: A FORTRAN
program for the comprehensive analysis of electrophoretic
data in population genetics and systematics. Journal of Heredi-
ty 72: 281-283.

Thorpe, J. P. 1982. The molecular clock hypothesis: Biochemical
evolution, genetic differentiation and systematics. Annual
Review of Ecology and Systematics 13: 139-168.

Tracey, M. L., K. Nelson, D. Hedgecock, R. A. Shleser, and M. L.
Pressnick. 1975. Biochemical genetics of lobsters; genetic
variation and the structure of American lobster ( Homarus
americanus) populations. Journal of the Fisheries Research
Board of Canada 32: 2091-2101.

Voelker, R. A., H. E. Schaffer, and T. Mukai. 1980. Spontaneous
allozyme mutations in Drosophila melanogaster: Rate of oc-
currence and nature of the mutants. Genetics 94: 961-968.

Zimmerman, E. G. and M. E. Nejtek. 1977. Genetics and speciation
of three semispecies of Neotoma. Journal of Mammalogy 58:
391-401.

A METHOD FOR EVALUATING THE SUBLETHAL IMPACT OF
STRESS EMPLOYING CORBICULA FLUMINEA

CHRISTOPHER FOE AND ALLEN KNIGHT
LAND, AIR AND WATER RESOURCES
UNIVERSITY OF CALIFORNIA, DAVIS
DAVIS, CALIFORNIA 95616, U.S.A.

ABSTRACT

The feasibility of using the Asiatic clam, Corbicula fluminea, for monitoring point source discharges
has been evaluated. Clams were transplanted into cages at several locations around the outfall of
a proposed waste treatment plant in the Sacramento-San Joaquin Delta, California. Reproduction,
shell growth, condition index (ratio of tissue weight to shell length), survival, and copper and zinc
tissue concentration were evaluated for ten months. In addition, at one test site, the reproduction,
growth, and condition index of a wild clam population was also determined. Comparison of sublethal
parameters between the wild and caged clams demonstrated no significant difference (P>0.10).

Initiation of the operation of the proposed waste treatment facility was delayed; therefore, our
study includes only predischarge data. However, in June, there was a strong settlement of the bar-
nacle, Balanus improvisus, on the monitoring cages and clams. Barnacle fouling appeared to pro-
duce a clam stress gradient as we recorded a simultaneous decrease in clam reproduction, condition
index (P <0.05), and survival (P <0.05), but not growth (P >0.1), with increased fouling. The seasonal
pattern in the estuary for Corbicula’s copper and zinc tissue concentration was also determined. Metal
concentration increased in the summer independent of both the clam’s gut content and alterations
in its biomass. In conclusion, we believe our technique may represent, with more work, a promising
method for simultaneously monitoring alterations in clam tissue concentration and sublethal impacts.

The pollution of aquatic systems is of increasing con-
cern. One aspect receiving considerable attention is the
evaluation of toxic materials and their sublethal impact on
aquatic invertebrates. A major problem is the development
of methods for measuring alterations in the life history
characteristics of aquatic organisms under field conditions.
The Marine Mussel Watch (Goldberg et a/., 1978) and the
Coastal Environmental Assessment Stations Program (Phelps
and Galloway, 1980) were implemented to assess the impact
of marine pollution using the bay mussel, Mytilus edulis, as
a sentinel organism. The Marine Mussel Watch program at-
tempts to identify areas with pollution problems by measur-
ing alterations in tissue burden levels while the Coastal En-
vironmental Assessment Stations Program attempts to
develop methods for assessing the sublethal impact of these
contaminants. Together, these programs appear to be mak-
ing progress in identifying the impact of pollution on marine
communities at the population level. Unfortunately, no com-
parable techniques are presently available for freshwater.

The objective of the present study was to begin the
development of methods to ascertain the impact of pollution
on freshwater communities by using the Asiatic clam, Cor-
bicula fluminea (Muller). Specific objectives were to develop

methods for simultaneously measuring alterations in Cor-
bicula life history characteristics and tissue-burden levels.
Cause and effect relationships may be more apparent if
changes in both functions are measured simultaneously. In
general, the methods employed in this study have been
modelled after those of the Marine Mussel Watch and the
Coastal Environmental Assessment Stations Program.

The freshwater clam was chosen as it appears a pro-
mising bioindicator candidate. First, as a filterfeeder, it is
known to bioaccumulate both heavy metals and synthetic
organic compounds (Burress and Chandler, 1976; Woodard,
1979; McCleneghan and Rectenwald, 1979; Leard et at.,
1980; Rodgers ef a/., 1980; Cory and Dresler, 1981; Graney
et a!., 1983; Hayes and Phillips, 1985). Second, although
recently introduced, it has spread rapidly across the United
States and is now an important benthic invertebrate in many
major drainage basins. Consequently, the clam is easy to col-
lect for study and alterations in its population dynamics may
be of ecological significance. Third, the clam is relatively im-
mobile and long-lived (often > 3 years, Eng, 1979) and,
therefore, an excellent long-term water quality monitoring
organism. Fourth, methods for spawning and culturing Cor-
bicula in the laboratory are under development (Foe, 1983;

American Maiacological Bulletin, Special Edition No. 2(1 986): 133-1 42

133

134

CORBICULA SYMPOSIUM

Foe and Knight, 1985; 1986b). Considerable other basic
physiological information, relative to other freshwater in-
vertebrates, is also available (Mattice and Dye, 1975;
McMahon, 1979a,b; Gainey, 1978a; Foe and Knight, 1986a).
This information could be valuable in the future for interpreting
monitoring observations. Finally, Corbicula has a salinity
tolerance of 0 to, at least, 3 °/00 (Gainey, 1978a,b). No other
freshwater bioindicator candidate has this wide a salinity
distribution. This trait is essential for evaluating toxicity in the
upper portion of estuarine systems, often the area most heavi-
ly impacted by pollution (Forstner and Wittman, 1979).

METHODS AND MATERIALS

Location. The clams evaluated in the present study were col-
lected in early February, 1981, from a clean, subtidal, sand
beach off the main channel of the Sacramento River near
Sherman Lake, California (Fig. 1). This population has been
identified electrophoretically as Hillis and Patton’s (1982)
white Corbicula morph (McLeod, 1986). The clams were
transferred to cages, which were suspended for ten months
around the Contra Costa Sanitation District outfall located in
New York Slough. Five stations (number 1 through 5) were
established around the outfall in subtidal areas characterized
by continuous water movement but minimal wave action.
Temperature and salinity were recorded weekly at each site.
Differences in temperature between stations were compared
by Analysis of variance (ANOVA) and Duncan multiple range
test.

Cages Units. Seven cages (66 clams/cage) were placed at
each station. In March the cages containing clams for the
growth evaluation were lost from station 4. The remainder
of the cages lasted until December when all cages from sta-

Fig. 1 . Map of the Western Sacramento-San Joaquin Delta, Califor-
nia showing the monitoring stations (1-5), the waste outfill, and
the clam collection site.

tion 3 were lost resulting in the termination of the experiment.

Cage units were constructed of plastic fluorescent light
egg-crate-type panelling and from polyethylene screen with
a 3 mm X 4 mm mesh (Fig. 2a, b). Cage sections were bound
together with surgical rubber tubing and stainless steel bolts.
Individual cage compartments measured 41mm X 41mm X
22mm and were each numbered to aid in identifying individual
clams. During field exposure, each cage unit was secured
about half a meter off the bottom in an upright position by
metal rebar stakes driven into the substrate.

The cages were colonized during the study by several
invertebrates including juvenile Corbicula, the barnacle
Balanus improvisus, the amphipods Corophium spinicorne
and C. Stimpsone, and the hydroid Cordylophora lacustris.
These epizooites restricted water flow through the cages and,
undoubtedly, competed with clams for suspended food. Foul-
ing was minimized by cleaning the cages (but not the clams)
monthly with a coarse nylon scrubbrush.

SUBLETHAL INDICES

1 . Reproduction. Clam reproduction was estimated by a com-
bination of laboratory and field experiments exploiting the fact
that Corbicula incubates its young on its inner gill
demibranchs (Eng, 1979). The development rate of the mar-
supial larvae was estimated by inducing adults to spawn in
the laboratory by means of thermal shock (Loosanoff and
Davis, 1963) and recording daily the percentage of adults with
young on their gills. Simultaneously, replicate sets of petri-
dishes placed on the aquarium bottom were monitored to
determine when larvae first settled. The results of these ex-
periments were used to establish the sampling frequency
necessary for estimating field reproduction.

Field reproduction was determined by transplanting
400 clams (30 to 35 mm shell length) into four cages at each
station. Previous work demonstrated that this size class has
the highest fecundity (Foe and Knight, 1981). Ten individuals
were collected weekly between 1 April and 30 November from
each station and preserved in 5% Formalin for subsequent
analysis. In addition, concurrent reproduction samples were
taken from the wild population at station 5 to compare the
reproduction of caged and natural clam populations. Cages
were sampled sequentially to minimize clam disturbance.

Corbicula’ s reproductive output was estimated quan-
titatively only for the caged and wild clam population at sta-
tion 5. Here, the number of incubated larvae in one ran-
domly selected gill from each of the five clams was counted
by extrusion onto a microscope slide followed by resuspen-
sion in 100 ml of saturated sugar solution. After vigorous
agitation, a 5 ml subsample was drawn and passed through
a 60-/tm mesh screen. The larvae were enumerated using
a dissecting microscope at 30 X power magnification. This
process was repeated until either 100 young or the entire
sample was processed. Repetitive larval counts of gill extracts
with known numbers of larvae present demonstrated that the
procedure had an accuracy of about 90%. To improve our
estimate of Corbicula’ s reproduction, an additional five clams
were examined during periods of intense clam reproduction.
Reproductive output was averaged for each week and

FOE AND KNIGHT: CORBICULA MONITORING METHOD

135

Fig. 2a. Cage before the addition of clams. 2B. Cage with clams after several months of field exposure.

136

CORBICULA SYMPOSIUM

summed over the entire season to estimate the clam’s an-
nual fecundity.

At the remaining four sites, a more qualitative estimate
of reproduction was employed because of the large effort in-
volved in direct counts of marsupial young. These estimates
were made by inspecting the gills from five animals weekly
and assigning them a score from 1 to 5 based on a subjec-
tive estimate of the number of larvae present. Scores of 1 ,
2, 3, 4, and 5 represented 0-1000, 1000-10,000,
10,000-20,000, 20,000-30,000, 30,000-40,000 young per
adult, respectively. These scores were averaged weekly for
each station and summed over the entire season to estimate
annual reproduction.

2. Shell Growth. Seventy-six individuals (2 clams in each 1
mm size class between 3 and 40 mm) were deployed in cages
at each station and their shell growth determined every 60
days using vernier calipers. After ten months exposure,
growth was averaged for clams in each initial 5 mm shell size
class, and the difference in growth between stations com-
pared using ANOVA and Duncan multiple range test.

At station 5, the growth of caged Corbicula was also
compared with that of transplanted tethered individuals to
ascertain the effect, if any, of caging on clam growth. Previous
work (Foe and Knight, unpubl.) has demonstrated that the
growth of tethered clams is not statistically different from that
of the wild population. The tether method consisted of
cementing a small plastic disk with an identifying number and
a short length of 5 lb monofilament fishing line (20-30 mm)
to each clam shell with fast-drying Duro Super glue® . The
monofilament fishing lines were attached to a heavier nylon
line with stainless steel fishing swivels, and the unit staked
taut along the bottom. The method has the advantage of
allowing the clams to burrow and move about naturally in the

sediment. Also, the tether unit is easily retrieved for measure-
ment of individual clam growth.

One hundred individuals (four in each 1 mm shell size
class between 15 and 40 mm) were tethered, and their growth
was measured every 60 days. Differences in growth between
tethered and caged clams at station 5 was analyzed by a
paired t-test using the recorded growth rate of each 5 mm
shell length size class.

3. Condition Index. Condition index is defined as the ratio of
dry tissue weight to shell length. A decrease in the ratio is
indicative of a deterioration in health of the bivalve popula-
tion (Bayne et al. , 1 976). Condition indices were determined
by transplanting at each station six groups of 25 animals (3
individuals in each 5 mm shell length size class between 3
and 40 mm) and randomly selecting one group every 60 days
for analysis. In addition, twenty-five wild clams from station
5 were also collected on each occasion to compare their con-
dition index with that of caged clams. Condition indices were
calculated by measuring individual clam shell lengths with
vernier calipers and determining ash free dry weight from the
difference in clam tissue weight dried at 60°C and ashed at
480°C. For each group, log weight was regressed against log
shell length. Differences in the slope and intercept of the
regressions from each of the five stations were compared
bimonthly using an analysis of covariance, if a difference was
detected, then a pairwise comparison of the individual lines
was conducted to determine which lines were different after
correcting the overall significance level of the test with a
Bonferroni adjustment (Neter and Wasserman, 1974). Lines
which were not different were combined to calculate a com-
mon regression equation.

4. Mortality. Clam mortality was assessed every two months
in the cages by recording the number of clams dying during

Fig. 3. Comparison of barnacle infestation on clams during August from station 1 (lower row) and station 5 (upper row).

FOE AND KNIGHT: CORBICULA MONITORING METHOD

137

the interval as determined from the number of empty shells
or missing animals and dividing this by the number alive at
the start of the period.

HEAVY METAL TISSUE CONCENTRATIONS
Field Procedures. Seasonal changes in copper and zinc tissue
concentration were monitored by transplanting 160 clams (30
to 35 mm shell length) into cages at each station. In addi-
tion, at station 3, animals were attached to tethering units
adjacent to the cages to determine if the metal content of
clams buried in the sediment differed from those of clams
held in cages in the water column. Twenty transplanted clams
from each group were collected every 60 days and frozen
at -20°C in acid-washed ziplock plastic bags for future
analysis. Whenever possible, wild clams were also collected
from station 3. Comparisons of metal concentration between
wild and transplanted individuals provide an indication of how
representative the metal content of transplanted clams are
of the natural population. Finally, an additional 20 animals
were harvested from the cages placed at station 3 in August
and their stomach and intestinal tracts dissected out with
stainless steel scalpel and forceps. Their tissue concentra-
tion was compared with that of ungutted caged clams to deter-
mine whether increases in metal concentration reflected an
increase in the metal content in the animals’ gut or an ac-
tual increase in clam tissue concentration.

LABORATORY PROCEDURES. Mean copper and zinc tissue
concentration was determined bimonthly from 5 samples of
3 homogenized clams each using the procedures outlined
in the California Marine Mussel Watch (Stephenson et a!.,
1979). At least five procedural blanks were employed during
each assay to detect contamination arising during the diges-
tion or analysis phase. No detectable amount of either metal
was ever reported for the blanks. Metal concentrations were
determined by the flame method on a model 751 Instrumen-
tation Laboratories atomic adsorption spectrophotometer and
reported as ^g metal/gm dry tissue weight (ppm). Differences
between sample means were analyzed with ANOVA and Dun-
can multiple range test.

RESULTS AND DISCUSSION

To date, only background information on Corbicula
sublethal indices and its heavy-metal concentration have
been obtained for the New York Slough area. A delay in in-
itiating the operation of the treatment plant provided us little
opportunity to evaluate our monitoring method in a waste
discharge. However, in June, the settlement of the barnacle,
Balanus improvises , provided an opportunity to evaluate our
method in another stress gradient. Greatest barnacle recruit-
ment occurred at station 1 and decreased rapidly eastward
with no settlement at site 5 (Fig. 3). The juvenile barnacles
were removed from the cages but left on the clams, as we
did not wish to unduly disturb the test animals. However, by
August, the barnacles had grown considerably and appeared
to be stressing Corbicula by preventing the complete closure
of the clam’s valves. All barnacles were cleaned off the clams;
however, it appeared that the barnacles had already stressed

1981

Fig. 4. Seasonal pattern of temperature (°C) and specific conduc-
tance (millimho/cm) at station 5.

many Corbicula. Although this was not the kind of stress we
had originally intended to measure, we reason that our
sublethal indices should exhibit a response in any stress gra-
dient. Therefore, we have analyzed, instead, Corbicula’ s
sublethal response to Balanus settlement.

Ambient water temperature was similar at all five sta-
tions on each occasion measured (Fig. 4, P>0.25, ANOVA).
Water temperature rose rapidly after January, peaked at 25°C
in July, and declined quickly in November and December.
Electrical conductivity increased at the more seaward sta-
tions. However, the largest difference recorded between sta-
tions 1 and 5 was only 7%. At no time did any value exceed
3 °/( jo salinity, well within Corbicula’s tolerance (Evans et al.,
1979).

SUBLETHAL INDICES

1 . Reproduction. Laboratory experiments demonstrated that
the development time from egg to pediveliger larva was be-
tween 3 and 8 days. The first eggs appeared in the gills on
day 3 and were all released by day 1 1 (solid line, Fig. 5). This
established an upper limit of eight days for maturation. The
first pediveligers were observed settling in petri-dishes on the
aquaria bottom on day 6 (broken line, figure 5). The dif-
ferences in time between when eggs were first observed in
the gill pouches and young on the aquaria bottom provided
the lower estimate of 3 days for larval development. These
laboratory estimates of Corbicula’ s marsupial development
rate were used to establish the seven day sampling periodicity
used subsequently for collecting field reproduction samples.

The wild clam population at station 5 spawned twice
in 1981 (Fig. 6). The first spawning occurred from mid-April
to May and the second from August to September. Sixty per-
cent of the larvae (36,521 larvae per adult) were produced
in the first reproductive periods and forty percent in the se-
cond (24,660 larvae per adult). This was different from 1980
at station 5 when 38% of the fecundity occurred in the first

Vo ADULTS INCUBATING YOUNG

138

CORBICULA SYMPOSIUM

Fig. 5. Percentage of adult clams incubating young and the number
of settled pediveliger larvae petri-dish-1 on consecutive days after
thermal induction of spawning.

period and 62% occurred in the second (Foe and Knight, un-
publ.). Also, total reproductive output was somewhat greater
in 1980 at 74,810 larvae per adult (Foe and Knight, 1981).
However, both fecundity estimates appear similar to those
reported for Corbicula from north central Texas (Aldridge and
McMahon, 1978).

Comparison of reproduction rates for wild and caged
clams at station 5 (Fig. 6) demonstrate a greater overall
reproduction output in the wild population. This was primari-
ly due to a failure of the caged clams to spawn as intensively
as the wild population during the second reproductive period.
This is, tentatively, interpreted as being due to cage induced
stress. Estimates of reproductive activity for clams from the
remaining four stations indicated that spawning was also
strong and similar at all sites during the first reproductive
period (30,000-40,000 young adult -1). However, clam
reproduction appeared to decrease progessively at more
westward stations in the estuary during the second spawn-
ing period. Fall reproduction was 0, 0, 5-12,000, 20-30,000

Fig. 6. Comparison of reproduction for caged and wild clams at sta-
tion 5.

and 24,660 young per adult at sites 1 , 2, 3, 4 and 5, respec-
tively. The decrease in reproductive activity is attributed to
the progressive increase in Balanus fouling.

2. Shell Growth. Caged Corbicula began to grow in April, when
the water temperature rose above 15°C, and continued
through November. Growth was greatest for smaller clams
and decreased as shell size increased. Shell formation ap-
peared normal for clams of all size classes, except some in-
dividuals of a 1 5 to 20 mm shell length who exhibited an ab-
normally concave growth form. This deformation did not ap-
pear to affect their subsequent growth.

The growth of caged and tethered clams was com-
pared at station 5 to determine the effect of caging on clam
growth. Previous experiments have demonstrated that the
growth of tethered clams is similar (P>0.25) to that of the
surrounding wild population (Foe and Knight, unpublished).
The growth of caged clams was slightly less than that of
tethered ones, however, not significantly so (P>0.1, paired
t-test). Previous experiments have demonstrated that the
growth rate of small clams (4 to 9 mm shell length) is unaf-
fected by caging (Foe and Knight, 1985). Therefore, it
was concluded that holding Corbicula in cages did not
significantly alter their growth pattern. So, we feel justified
in presenting the growth rates derived from caged individuals
as representative of the natural population.

Comparisons of growth between stations revealed a
decrease at sites 1 and 2 in both August and October. This
decrease is attributed to the Balanus infestation. However,
the difference in growth was not significant when evaluated
over the entire transplant period (7 February to 1 5 December,
P>0.1, ANOVA). Therefore, an annual size-specific growth
curve (Fig. 7) for Corbicula has been calculated for the
Sacramento-San Joaquin Delta in 1981 by averaging the
growth of all clams in each 1 mm shell size class using cage

INITIAL SHELL LENGTH (mm)

Fig. 7. Annual size-specific growth of Corbicula caged in the Western
Sacramento-San Joaquin Delta, California during 1981. Data points
are the mean of 5 to 10 clams; the dotted lines are ± standard
deviation.

FOE AND KNIGHT: CORBICULA MONITORING METHOD

139

Fig. 8. Zinc concentration (^g mg-1) for caged, tethered and wild clam
populations at station 3.

200

180

160

140

£, 120
Cn

* 100
■D

E 80

3 60

Fig. 9. Copper concentration (ng mg-1) for caged, tethered and wild
dam populations at station 3.

• Caged
□ Tethered
▼ Wild

Feb April June Aug Oct Dec

1981

growth from stations 1 , 2, and 5 and tether data from site 5
(the growth cages for station 3 were lost in November, and
from station 4 in March). The resulting growth pattern is
similar to that report for clams from the Sacramento-San Joa-
quin Delta during 1980 (Foe and Knight, 1981).

3. Condition Index. Comparison of the condition index regres-
sion equations for caged and wild clams at station 5
demonstrated that both had similar length-weight ratios (‘b’
regression coefficients) throughout the year (Table 1). This
indicates that caging had little effect on Corbicuia’s condi-
tion index. However, the data does reveal a strong seasonal
alteration in the weight of a ‘standard’ 30 mm clam at sta-
tion 5 (Table 1). The decrease in weight appeared greatest
during the second reproductive period (August through
September). Decreases in clam weight were also recorded
in December. We have noticed a similar phenomenon dur-
ing other years and now tentatively attribute this to a lack of

food during the winter months (Foe and Knight, 1985).

Comparison of the regression equations between sta-
tions revealed no significant difference in clam condition in-
dices through June (P > 0.1 , analysis of covariance, Table 1 ).
However, in August there was a significant decrease in weight
at the more westerly stations (P<0.05). This decrease oc-
curred simultaneous to the Balanus fouling, suggesting that
the barnacles imposed the stress on Corbicula.

4. Mortality. Mortality through June averaged 1% (Table 2).
Between August and October there was a significant increase
in mortality at the more westward stations (P < 0.05, ANOVA)
which corresponded with the gradient of Balanus fouling. Mor-
tality decreased again in December after barnacle removal.
SUBLETHAL INDEX SUMMARY. The biological indices pro-
vide a coherent picture of Corbicula stress. All four indices
decreased during barnacle fouling at the more heavily in-
fested stations. Differences in condition index and mortality

Table 1. Summary of condition index regression equations for Cor-
bicula's ash-free dry weight (gms) regressed against shell length (mm)
and predicted weight derived from those equations for a ‘standard’
30 mm clam.

Month

(1981)

Station

Regression
equation^/
b a

Predicted

weight

(mg)

April

1 ,2,3,4,5,5W?/

3.00

4.93

317

June

1 ,2,3,4,5,5W

2.66

4.93

280

August

1,2?/

2.91

4.83

294

3,4

2.91

4.81

310

5,5W

2,95

4.83

334

October

1,2

2.99

5.00

261

3,4,5,5W

2.84

4.75

286

December

2.95

5.00

225

2,3,4,5,5W

2.50

4.32

238

II log weight (gms) = b (iog shell length (mm)) - a.

?/ Wild population at Station 5.

?/ Differences between stations in regression coefficients during
the same sampling period indicate statistical differences in at
least one of the regression values at the 5% level.

Table 2. Mortality (%) of caged Corbicula in New York Slough. Values
are for the percentage of deaths occurring during the previous two
month period.

Month

(1981)

STATION

April

0.5

0.8

0.9

1.1

0.9

June

0.5

2.6

2.5

0.5

1.1

August

0.4

1.5

1.2

0.0

0.7

October

61.0

28.0

15.5

6.7

0.6

December

6.6

4.9

—

2.8

1.6

140

CORBICULA SYMPOSIUM

Table 3. Copper tissue concentrations (^g/gm) for caged Corbicula in New York Slough.

Month

(1981)

STATIONS

February

76.6 a 2.2./

76.6 a

(pretransplant)

±19.0

April

78.4 a

77.5 a

75.3 a

85.5 a

94.5 a

±15.2

±7.3

±9.0

±13.6

±6.9

June

81.5 a

145.9 b c

146.2 b c

239.9 d e

118.0 a b

±6.4

±16.7

±15.0

± 19.2

± 14.0

August

114.4 a b

148.6 b c

146.7 b c

275.4 e

120.4 a b

±17.8

±17.3

± 10.0

±19.4

±18.6

October

103.3 a b

107.5 a b

67.7 a

172.8 d

150. a b

±9.6

±24.2

±4.4

±14.8

±16.4

December

89.7 a

95.0 a

—

219.2 d

66.4 a

± 10.5

±12.8

—

±40.0

±7.1

Mean ± standard error.

Values with the same letter are not statistically different at the 5% level.

were significant (P<0.05, Tables 1 and 2). We believe this

hibited similar zinc concentrations. The differences between

provides a good example of the clam’s response in a stress

the April samples is interpreted as being due to the fact that

gradient. In the future, fouling problems such as those en-

transplanted clams initially had a lower zinc level and that

countered in this study can be avoided by cleaning both clams

the rate of metal uptake required between four and six months

and cages.

before the two concentrations could become equal. However,

METAL TISSUE CONCENTRATIONS

once similar, the metal dynamics of both groups remained

Comparison of metal concentrations in tethered, caged and

the same. All three groups subsequently demonstrated a

wild clams- Zinc. Zinc tissue concentrations were always

significant increase in zinc concentration in the summer and

similar for caged and tethered clams at station 3 (Fig. 8).

depuration in the fall and winter (P<0.05, ANOVA, Fig. 8).

However, April zinc levels were greater in the wild popula-

Copper. The copper concentration of the tethered and wild

tion than in either set

of transplanted clams (P<0.05,

clam population was similar on all occasions (Fig. 9, P>0.1,

ANOVA). Thereafter, both wild and transplanted clams ex-

ANOVA). In contrast, the copper concentration of the caged

Table 4. Zinc tissue concentrations (/ig/gm) for caged Corbicula in New York Slough.

STATIONS

Month

(1981)

February

143.4 a I-?/

143.4 a

±29.5

April

178.6 b c

171.6 b c

155.8 a b

159.9 a b

173.7 b c

± 13.1

±9.3

±6.9

±5.5

±10.1

June

152.8 a b

168.9 a b c

189.8 cd

221.2 e

134.9 a

±5.5

±6.6

±11.5

± 17.4

±10.1

August

232.5 e

194.7 cd

227.7 e

219.0 d e

200.2 c d e

±13.3

±11.5

±14.2

±9.9

±14.8

October

169.2 a b c

194.2 c d

171.2 b c

159.1 a b

211.9 d e

±2.9

±8.2

±4.6

±14.4

±9.8

December

186.9 c d

179.5 be

—

188.0 c d

179.6 b c

±7.5

±9.1

—

±19.9

±14,2

II Mean ± standard error.

?/ Values with the same letter are not statistically different at the 5% level (see text for details).

FOE AND KNIGHT: CORBICULA MONITORING METHOD

141

Table 5. Summary of reported copper and zinc tissue concentra-
tions for Corbicula.

Tissue Concentration

Zn 11

Cull

Location

Author

126?/

44 ?/

California

Woodward,

(72-288)

(18.6-81.0)

inland rivers
(1976-1978)

1979;

McCleneghan
and Rectenwald,
1979.

421

43.5

Glen Lyn

Cherry ef a/.,

(313-522)

(33.5-108.2)

Power Plant,
New River, VA.

1980

173.7

94.5

New York

Present study

(134-232.5)

(66.4-275.4)

Slough and San

Joaquin River,

CA.

II Median and range in ppm dry weight.

?/ Wet weight converted to dry weight (x6).

clams appeared higher in June and August than that of either
the wild or tethered population. However, this difference was
not significant (P>0.05, ANOVA). In conclusion, therefore,
comparisons of copper and zinc tissue concentrations be-
tween wild and transplanted clams seem to demonstrate that
transplanted individuals can be employed to monitor metal
concentrations of the natural population.

SEASONAL CLAM TISSUE CONCENTRATION IN NEW
YORK SLOUGH

Copper and zinc tissue concentrations in caged clams
along New York Slough and the San Joaquin River increased
during the summer at all stations (Tables 3 and 4). Peak
values often occurred in August. Copper concentration was
significantly higher during the summer at stations 2, 3 and

4 (Table 3), whereas zinc was greater at station 1 , 3, 4 and

5 (Table 4, ail at P<0.05, ANOVA).

After April, the copper concentration in clams at sta-
tion 4 was significantly greater (Table 3) than that at any other
site (P < 0.05, ANOVA). This result was unexpected, as there
was no apparent source for the metal. We speculate that it
may have eminated locally from the sediment.

Rapid fluctuations in biomass have been documented
to produce an impression of a rapid fluctuation in metal tissue
concentrations if the data is analyzed on a dry weight basis
(Boyden, 1974; Strong and Luoma, 1981). Corbicula’ s
biomass did fluctuate significantly during the study both as
a function of reproduction and the Balanus settlement (Table
1). However, all our conclusions concerning clam metal tissue
concentrations remain the same when the data was
reevaluated in terms of the total metal content of a standard
30 to 35 mm clam.

Flegal and Martin (1977) have cautioned that er-
roneously high metal tissue concentrations can result from
including sediment bound metals in gut tissues. This could

be particularly important for suspension-feeding bivalves such
as Corbicula, which are known to ingest large amounts of in-
organic material (Foe and Knight, 1985). Therefore, the
metal concentrations of gutted and ungutted clams at sta-
tion 3 were compared in August to determine whether the
increase in clam metal concentration represented an actual
increase in tissue concentration or a transient increase in in-
gested sediment-bound metal. The copper and zinc concen-
tration of gutted clams was 1 52.4 ± 1 6.5 ppm and 21 6.4 ±
12.6 ppm. respectively (mean ± 1 standard error). These dif-
ferences were not significantly different from those reported
for ungutted clams (P > 0.25, student t-test, Tables 3 and 4).
Therefore, the seasonal increase in clam metal concentra-
tion appears to represent an actual increase in tissue
concentration.

No other metal tissue concentration data has been
reported for an invertebrate from the Sacramento-San Joa-
quin Delta. However, Siegfried etal., (1980) has reported an
increase in sediment metal concentrations during the sum-
mer in the same general area. The metal appeared to be
bound to the silt and organic fractions of the sediment. The
reported increase in Corbicula' s metal concentration may,
therefore, represent the bioaccumulation of this metal into
the filterfeeding portion of the food chain.

Finally, we have found no evidence of a deleterious
sublethal impact on Corbicula of the high summer metal con-
centrations. However, our results are confounded by the
simultaneous settlement of Balanus and by the loss of the
growth cages at station 4.

The State of California monitored between 1976 and
1979 metal concentrations in selected fish and invertebrates
from California’s inland waters, but not from the Sacramento-
San Joaquin Delta (Table 5). When available, Corbicula was
employed as their benthic bioindicator species. Our zinc
levels are somewhat higher but still comparable to those
measured by the State (Table 5); however, our copper values
are consistently greater than theirs. For example, at station
4 our copper concentrations ranged between 1 and 3.5 times
greater than those reported by the state monitoring program.
The high metal tissue concentrations in Corbicula may result
either from the extensive industrial activity along the shores
of the western delta or from the natural tendency of metals
to concentrate in the freshwater portion of the estuary
(Forstner and Wittman, 1979). Regardless, the headwaters
of the Sacramento-San Joaquin Delta are one of the most
productive areas of the entire estuary (Ball and Arthur, 1 979)
and the presence of high metal concentrations here in both
the sediment and some fauna deserve additional study.

ACKNOWLEDGEMENTS

This study was supported by grants from DOW Chemical Com-
pany and from the Environmental Protection Agency (grant No.
440344-22870). We thank Barry Votaw for assistance in the field and
Doug Howell for reading an early version of the manuscript.

LITERATURE CITED

Aldridge, D. W. and J. F. McMahon. 1978. Growth, fecundity and

142

CORBICULA SYMPOSIUM

bioenergetics in a natural population of the Asiatic freshwater
clam Corbicula fluminea Philippi from North Central Texas.
Journal of Molluscan Studies 44:49-70.

Ball, M. D. and J. F. Arthur. 1979. Planktonic chlorophyll dynamics
in the Northern San Francisco Bay and Delta, pp. 267-285.
IN: T. J. Conomos, ed., San Francisco Bay: The Urbanized
Estuary. Pacific Division, American Association for the Ad-
vancement of Science, San Francisco, California.

Bayne, B. L., J. Widdows, and P. J. Thompson. 1976. Physiological
Integrations, pp. 261-293. IN: B. L. Bayne, ed., Marine
Mussels. Cambridge University Press, Cambridge.

Boyden, C. R. 1974. Trace element content and body size in mollusc.
Nature 251:311-314.

Burres, R. M. and J. H. Chandler. 1 976. Use of the Asiatic clam Cor-
bicula leana in toxicity tests. Progressive Fish Culturist 38:1 0.

Cherry, D. S., J. H. Rodgers, R. L. Graney, and J. Cairns. 1980.
Dynamics and control of the Asiatic clam in the New River,
Virginia. Virginia Water Resources Research Center, Bulletin
123, 72 pp.

Cory, R. 1. and P. V. Dresler. 1981. Thermal plumes, trace metals
and Asiatic clams of the Potomac River and Estuary. Estuaries
4(3):296.

Eng, L. L. 1979. Population dynamics of the Asiatic clam Corbicula
fluminea (Muller) in the concrete lined Delta-Mendota Canal
of Central California, pp. 40-68. IN: J. C. Britton, ed., Pro-
ceedings of the First International Corbicula Symposium. Texas
Christian University Research Foundation Publication, Texas.

Evans, L. P., C. E. Murphy, J. C. Britton and L. W. Newland. 1979.
Salinity relationships in Corbicula fluminea (Muller), pp.
193-214 IN: J. C. Britton, ed., Proceedings of the First Inter-
national Corbicula Symposium. Texas Christian University
Research Foundation Publication, Texas.

Flegal, A. R. and J. M. Martin. 1977. Contamination of biological
samples by ingested sediment. Marine Pollution Bulletin
8(4):90-92.

Foe, C. G. and A. W. Knight. 1981. Growth and Reproductive pat-
tern of the Asiatic clam Corbicula fluminea in the lower
Sacramento-San Joaquin Delta. Estuaries 4(3):280.

Foe, C. G. 1983. Flow and Substrate requirements of juvenile Cor-
bicula fluminea (Bivalvia). pp. 27-42. In: (A. W. Knight, ed),
The evaluation of contamination sediment employing selected
benthic freshwater organisms. University of California, Depart-
ment of Land, Air and Water Resources. Water Science and
Engineering paper No. 4516.

Foe, C. G. and A. W. Knight. 1985. The effect of phytoplankton and
suspended sediment on the growth of Corbicula fluminea
(Bivalvia). Hydrobiologia. 127, 105-115.

Foe, C. G. and A. W. Knight. 1986a. A thermal energy budget for
juvenile Corbicula fluminea. IN: J. C. Britton, ed., Proceedings
of the Second International Corbicula Symposium. American
Malacological Bulletin Special Edition No. 2: 143-150.

Foe, C. G. and A. W. Knight a. 1986b. Growth of Corbicula fluminea
on algal and artificial diets. Hydrobiologia. 133, 155-164.

Forstner, U. and G. T. W. Wittmann. 1979. Metal pollution in the
aquatic environment. Springer-Verlag, New York. 486 pp.

Gainey, L. F. Jr. 1978a. The response of theCorbiculidae (Mollusca:
Bivalvia) to osmotic stress: The organismal response.
Physiological Zoology 51:68-78.

Goldberg, E. D., V. T. Bowen, J. W. Farrington, G. Harvey, J. H. Mar-
tin, P. L. Parker, R. W. Risebrough, W. Robertson, E.
Schneider and E. Gamble. 1978. The Mussel Watch. En-
vironmental Conservation. 5(2): 1-1 5.

Graney, R. L., D. S. Cherry, and J. Cairns. 1983. Heavy metal in-

dicator potential of the Asiatic clam (Corbicula fluminea ) in ar-
tificial stream systems. Hydrobiologia 102:81-88.

Hayes, S. P. and P. T. Phillips. 1985. California State Mussel Watch
Marine Quality Monitoring Program. 1983-84. Water Quality
Monitoring report No 85-2WQ. State of California Water
Resources Control Board 95 pp.

Hillis, P. M. and J. C. Patton. 1982. Morphological and Electrophoretic
evidence for two species of Corbicula (Bivalvia: Corbiculidae)
in North America. American Midland Naturalist 108(1):74-80.

Leard, R. L., B. J. Grantham and G. F. Pessoney. 1980. Use of
selected freshwater bivalves for monitoring organochlorine
residues in major Mississippi stream systems 1972-1973.
Pesticide Monitoring Journal 14(2):47-52.

Loosanoff, V. L. and H. C. Davis. 1963. Rearing of bivalve mollusks.
Advances in Marine Biology. 1 :2-1 30.

Mattice, J. S. and L. L. Dye. 1975. Thermal tolerance of the adult
Asiatic clam. pp. 130-135. In: (G. W. Esch and R. W.
McFarlane, eds.), Thermal Ecology II. Conferance 750425 Na-
tional Technical Information Service. Springfield, Virginia.

McCleneghan, K. and H. J. Rectenwald. 1979. Toxic substances
Monitoring program 1978. California State Water Resources
Control Board. Water Quality Monitoring Report No. 79-25.

McLeod, M. J. 1986. Electrophoretic variation in U. S. Corbicula. IN:
J. C. Britton, ed., Proceedings of the Second International
Corbicula Symposium. American Malacological Bulletin Special
Edition No. 2:125-132.

McMahon, R. F. 1979a. Response to temperature and hypoxia in
the oxygen consumption of the introduced Asiatic freshwater
clam, Corbicula fluminea (Muller). Comparative Biochemistry
and Physiology 63A:383-388.

McMahon, R. F. 1979b. Tolerance of Aerial exposure in the Asiatic
freshwater clam, Corbicula fluminea (Muller), pp. 227-242. IN:
J. C. Britton, ed., Proceedings of the First International Cor-
bicula Symposium. Texas Christian University Research Foun-
dation Publication, Texas.

Neter, J. and W. Wasserman. 1974. Applied Linear Statistical
Methods. Richard d. Levin Inc. Homewood. III.

Phelps, D. K. and W. B. Galloway. 1980. A report on the Coastal
Environment Assessment Stations (CEAS) program, pp. 76-81 .
In: (A. D. McIntyre and J. B. Pearce, eds.), Biological Effects
of Marine Pollution and the Problems of Monitoring. Rapports
et Proces-Verbaux des Reunions. Vol. 1979.

Rodgers, J. H., D. S. Cherry, R. L. Graney, K. L. Dickson and J.
Cairns. 1980. Comparisons of heavy metal interactions in
acute and artificial streams: Bioassay techniques for the
Asiatic clam (Corbicula fluminea ) pp. 260-280. IN: Eaton, Par-
rish and Hendricks, eds., Aquatic Toxicology, ASTM STO 707,
American Society for Testing and Materials.

Siegfried, C. A., M. E. Kopache and A. W. Knight. 1980. The ben-
thos of a portion of the Sacremento-San Francisco Bay
estuary, California, during a dry year. Estuaries 3(4):296-307.

Stephenson, M. S., M. Martin, S. E. Lange, A. R. Flegal and J. H.
Martin. 1979. California Marine Mussel Watch 1977-1978. Vol.
II. Trace metal concentrations in the California mussel Mytilus
californianus. Water Quality Monitoring Report No. 79-22.

Strong, C. R. and S. N. Luoma. 1981. Variations in the correlation
of body size with concentration of copper and silver in the
bivalve Macoma balthica. Canadian Journal of Fisheries and
Aquatic Science. 38:1059-1064.

Woodward, R. 1979. Toxic substances monitoring program 1976-
1977. California State Water Resources Control Board. Wafer
Quality Monitoring Report No. 79-20.

A THERMAL ENERGY BUDGET FOR JUVENILE CORBICULA FLUMINEA

CHRISTOPHER FOE AND ALLEN KNIGHT
DEPARTMENT OF LAND, AIR AND WATER RESOURCES
UNIVERSITY OF CALIFORNIA, DAVIS, CALIFORNIA 95616, U.S.A.

ABSTRACT

A thermal energy budget was developed for immature Corbicula fluminea (Muller) at 16, 20,
24 and 30°C. Water filtration rate at these temperatures was 2.80, 3.85, 5.33, and 13.39 m//mg dry
wt/hr. Respiration rates were 0.39, 0.56, 0.71 , and 0.85 ^/02/mg dry wt/hr. Assimilation efficiency was
48, 51, 36, and 13%, and activity levels were 98.9, 90.1, 95.1, and 8.0%, respectively. These rates
have been combined to form a scope for growth model which predicts optimal clam growth near 20°C,
negative growth above 29°C, and high tissue growth at temperatures as low as 16°C. Growth was
measured in the laboratory at two-degree increments between 16 and 32°C. Tissue growth at each
temperature was 6.2, 6.8, 6.9, 5.8, 4.9, 3.0, 1 .4, and -0.52 mg dry wt/month, respectively. All experimental
animals died when evaluated at 32°C. These results are discussed in terms of our current knowledge
about Corbicula' s thermal tolerance and life history.

The Asiatic clam, Corbicula fluminea (Muller), was first
reported in the Pacific Northwest at the turn of the century
(Burch, 1 944; Counts, 1981a) and has since spread eastward,
becoming a dominant benthic invertebrate (in biomass) in
many of the major drainage basins in the United States (Mat-
tice, 1979). This expansion is still in progress today with
reports in 1980 of the successful colonization of additional
rivers in Maryland, Pennsylvania, New Jersey and a section
of Lake Erie along the Ohio Shoreline (Counts, 1981b; Trana,
1982; Clarke, 1981). In some locations, Corbicula now occurs
in sufficient numbers to clog water transportation systems
and power plant intakes and may even competitively exclude
other benthic forms (Goss and Cain, 1975; McMahon, 1977;
Prokopovich, 1969; Morton, 1979; Gardner ef a/., 1976;
Boozer and Murkes, 1 979; Kraemer, 1 979). As a result there
is interest in predicting the ultimate geographic distribution
of this pest organism.

Temperature is a major environmental factor that in-
fluences invertebrate physiology and is important in deter-
mining geographic distributions (Kinne, 1970). There have
been several studies of Corbicula' s thermal tolerance (Mat-
tice and Dye, 1975; Mattice, 1979; McMahon, 1979).
However, no one has investigated the influence of
temperature on Corbicula' s growth and attempted to explain
the pattern in terms of the underlying physiological processes.

Energy budgets follow the flow of energy through
organisms by measuring caloric intake and losses. If the net
energetic balance is positive the animal will grow, if it is
negative, it will be forced to utilize its own tissues in order
to survive. Therefore, energy budgets represent an integra-
tion of all the physiological processes occurring in an
organism and provide an indication of the “whole animal”
response (Bayne ef a/., 1976a).

The purpose of this study was to develop a laboratory
thermal energy budget for Corbicula. We were particularly
interested in determining the optimal temperature for growth
and the extremes where stress occurs. Our results are
discussed in terms of what is presently known about Cor-
bie u la's thermal tolerance and employed to speculate about
the clam’s ultimate geographic distribution in the United
States.

METHODS

The formula for a bivalve energy budget is (Crisp,

1971):

C = R + F + P (1)

upon rearrangement this reduces to

C-F = A = R + P (2)

P = A - R (3)

where C is the amount of energy ingested, R is the energy
expended in metabolism, F the energy lost through urea and
fecal production, A the caloric value of the food absorbed
across the intestinal wall, and P the energy value of somatic
and gametic tissue growth.

Equation (3) has been termed the “scope for growth”
of an animal (Warren and Davis, 1967; Bayne ef a/., 1976a)
and represents an index of energy available for growth and
reproduction. Immature clams were used in this study and,
therefore, alterations in Corbicula’ s energy balance only af-
fect its capacity for somatic growth.

The various physiological parameters needed to
calculate the energy budget were estimated as follows (Wid-
dows, 1978):

C (cal/month) = filtration rate (m//month) x activity
level (%) x algal concentration (cells/

American Malacological Bulletin, Special Edition No. 2 (1986):143-1 50

143

144

CORBICULA SYMPOSIUM

m /) x algal caloric content (cal/cell).

A (cal/month) = C (cal/month) x assimilation effi-
ciency (%).

R (cal/month) = metabolic rate (m/02/month) x oxy-
caloric constant (4.86 cal/m/02).

P (cal/month) = tissue growth (mg/month) x caloric
content of tissue (cal/mg).

Each of these rates was estimated as a function of
temperature, converted to caloric equivalents per month, and
combined to estimate Corbicula’s scope for growth at 16, 20,
24, and 30°C. The model was validated by comparing the
predicted growth with actual measured laboratory tissue
growth over a 30-day period.

Immature clams (8.0-1 1 .0 mm shell length) were col-
lected from a subtidal population at Sherman Island in the
Sacramento River near Rio Vista, California. This population
has been identified electrophoretically (McLeod, 1986) as the
white Corbicula morph (Hillis and Patton, 1982). Clams
employed in the growth, filtration and respiration evaluations
were checked for reproductive activity at the end of the
respective experiments. No evidence of reproduction was
ever obtained. Shell lengths were determined to the nearest
0.1 mm with an ocular micrometer. Tissue weights were
estimated from a log-log regression of shell length (mm)
against ash-free dry weight (mg) of animals dried at 60°C and
ashed at 450°C. The equation for this relationship was:

log weight = 3.45 log length - 2.24 (R2 > 0.99) (4)

Clams were collected during July and August at an
ambient water temperature of 22°C and acclimated to the ap-
propriate experimental temperature at the rate of 1°C per day.
Once at the proper temperature, they were held an additional
four days prior to use. Specimens for the growth and fil-
tration experiments were fed during acclimation while those
used in the respiration studies were not. Culture water was
obtained from a nonchlorinated deep-well system with the
general chemical composition listed in Foe and Knight (1986).
The difference between means for each physiological rate
measured was analyzed with analysis of variance and
Newman-Keul mean separation tests.

FILTRATION

Filtration was measured in a static system consisting
of a series of 30 ml glass funnels partially submerged in a
water bath to control temperature (SE<0.08°C for each ex-
periment, N = 9). The stem of each funnel was connected
to an air supply to insure a gentle source of mixing. One
animal was held per funnel in a small aluminum cup
cemented to the inside wall. Clams were fed during the ther-
mal acclimation period on a trialgal culture of
Chlamydomonas, Chlorella, and Ankistrodesmus each at a
concentration of 3 x 10s cells/m/. Filtration experiments were
conducted in a similar culture at an initial algal concentra-
tion of 106 cells/m/. The procedure consisted of placing
animals in the funnels and observing when they opened and
commenced filtering. After an hour of continuous feeding,
a water sample was drawn and the cell concentration deter-
mined with a hemacytometer. Filtration was calculated from

the formula of Fox et al. (1937):

F _ In Ct - In C2 v
W T

where F is the filtration rate (m//mg-hr), Ci the mean algal
density of three control funnels maintained without clams after
60 minutes of continuous aeration (cells/m/), C2 the algal den-
sity after 60 minutes of clam filtration (cells/m/), V the funnel
water volume (ml), T 60 minutes, and W clam ash-free dry
weight.

ASSIMILATION

Assimilation efficiency was determined by the Conover
method (Conover, 1966):

A = (f ' E) x 100

(1 - E) (f)

where f is the ash-free dry weight ratio of phytoplankton prior
to ingestion and E is a similar ratio for the feces. Corbicula' s
fecel pellets were easily recognized among the pseudofecal
material as compact string-like extrusions. They were col-
lected from the bottom of the culture vessels every fourth day
with a micropipette and filtered onto tared precombusted
0.45-/im size glass fiber filters. These were dried at 60°C and
ashed at 450°C. Water samples were simultaneously taken
from each treatment and processed in a similar fashion to
estimate a valve for f. A bacterial control was conducted by
holding feces an additional four days at 24°C in order to deter-
mine if microbial action significantly altered E. A paired one-
tailed T-test indicated no significant decrease in this value
(p>0.25, N = 6).

METABOLIC RATE

Oxygen consumption was measured with a Gilson dif-
ferential respirometer. The procedure consisted of random-
ly selecting groups of four clams and placing them in sterilized
sediment in the respirometer flasks for at least one hour to
adjust to test conditions. Oxygen consumption was then
determined at 10 minute intervals for three consecutive hours.
A least squares regression was used to calculate the mean
oxygen consumption rate for each set of one hour
measurements. These values were averaged, corrected to
standard temperature and pressure, and divided by the
estimated weight of the animals to calculate each oxygen con-
sumption replicate in ^/02/mg dry wt/hr.

GROWTH

Growth was evaluated at 2° increments between 16
and 32°C using 10 / aquaria. Water temperature was con-
trolled to within 0.5°C with an American Instrument Company
supersensitive relay system. Clams were held in these
aquaria in small dishes without sediment. Previous ex-
periments have demonstrated that the laboratory growth of
Corbicula is independent of both flow and substrate (Foe,
1983). Water temperature, nitrate, nitrite, ammonia, pH and
dissolved oxygen were monitored regularly. Hydrogen ion

FOE AND KNIGHT: THERMAL ENERGY BUDGET

145

concentration was measured with a Corning model 61 OA pH
meter, dissolved oxygen with a model 51 B YSI oxygen
probe, and nitrite and nitrate concentration with a DR-EL/1
Hach kit. Ammonia concentration was determined by the
phenolhypo-chloride method of Solorzano (1969). The con-
centration of toxic unionized ammonia was calculated from
pH and total ammonia (Armstrong et al., 1978).

Clams were fed on a trialgal diet consisting of even
proportions of Ankistrodesmus, Chloreila, and Chlamy-
domonas at a total concentration of 105 cell/m/. In previ-
ous feeding experiments with Corbicula this defined algal
diet has been found to produce consistent, positive clam
growth (Foe and Knight a, 1986). We assume for this ex-
periment that this algal concentration represented an ad
libidum food level for Corbicula as pseudofeces, which are
thought to indicate a food saturated condition (Winter, 1969,
1978; Schulte, 1975), regularly formed at all temperatures
and repeated microscopic examinations of the pseudofecal
material demonstrated that all three algal species were pre-
sent in about similar proportion. Stock cultures of
Ankistrodesmus and Chloreila were purchased from the
Carolina Biological Supply Company. Chlamydomonas was
obtained from the University of Texas. All were grown as
monocultures under greenhouse conditions in aerated bot-
tles immersed in a temperature controlled water bath
(20-25°C). Ankistrodesmus and Chloreila were cultured on
Woods Hole media (Nichols, 1973) while Chlamydomonas
was grown on Hunter’s media (Starr, 1978).

The algal concentration in each clam growth treatment
was restored daily to maintain a constant algal density. These
concentrations were estimated by in vivo fluorescence with
a model III Turner Fluorometer from a least square regres-
sion of algal fluorescence against cell number (Strickland and
Parsons, 1969). Cell densities for each regression were deter-
mined with a hemacytometer. Repetitive microscopic ex-
aminations of the clam growth cultures demonstrated that all
algal species were always present in similar proportions.
There was no evidence of either a differential algal filtration
rate by the clams or of differential algal settlement from the
aquaria water column. Algal dry weight was estimated by
filtering water of a known cell concentration through a fared
0.45 /zm glass fiber filter (mean weight = 6 x 10"2mg/106
cells, SE = 0.01, N = 6). A known amount of seston was

scrapped from these filters, compacted into a pellet and ox-
idized in a Phillipson microbomb calorimeter (mean caloric
value = 2.27 cal/mg, SE = 0.25, N = 6). From this the
average caloric value of the trialgal culture was determined
to be 1.36 x 10"2 cal/m/ of water.

Clam shell and tissue growth was determined by
numbering both valves of each individual with a Sharpie
pen® . Shell length was determined to the nearest 0.1 mm
at the beginning and end of the experiment with an ocular
micrometer. Initial weights were estimated from the log-log
regression of shell length (mm) against ash-free dry weight
(mg) of animals collected at the beginning of the experiment
(Eq. 4). Final organic weights were determined as the dif-
ference in weight of shelled animals dried at 60 and ashed
at 450°C. Tissue growth was estimated from the difference
between initial and final weight. Relative tissue growth was
calculated by dividing the estimated tissue growth of each
individual by its initial weight and multiplying by 100. The
caloric value of the tissue was determined by gutting juvenile
clams and drying the tissue at 60°C before oxidation in a
Phillipson microbomb calorimeter. The mean caloric value
was 5.14 cal/mg (SE = 0.41, N = 8).

ACTIVITY LEVELS

The proportion of time clams spend actively filtering
is defined as their activity level. This figure was determined
for each temperature by inspecting the growth treatments
twice daily to establish the number of individuals open and
with siphons extended. This posture is presumed to indicate
active feeding.

RESULTS

FILTRATION

Filtration rates for Corbicula are presented in Table 1 .
Filtration was not statistically different at 16 and 20°C but in-
creased rapidly thereafter with rising temperature (p < 0.05).

Several authors have reported filtration data for Cor-
bicula. The values measured by Haines (1 979), Prokopovich
(1 969) and Habel (1970) are lower than those reported in the
present study. However, both Haines' and Prokopovich’s
rates may represent stressed animals as the authors report
some mortality during experiments. In contrast, most of our
values are comparable to those reported by Mattice (1979).

Table 1. Respiration, filtration, and assimilation rates for Corbicula as a function of temperature.

Temp

(°C)

Metabolic Rate1
Ot/02/mg-hr)

Filtration Rate1
(m//mg-hr)

Assimilation1

(0/0)

0.39 a

0.03

2.80 a

0.37

48 a

8.1

0.56 a b

0.03

3.85 a

0.41

51 a

6.5

0.71 be

0.03

5.33 b

0.22

36 a

7.6

0.85 c

0.01

13.39 c

0.99

13 b

2.9

'Values with the same letter are not statistically different at the 5% level. (See text for statistical details.)

146

CORBICULA SYMPOSIUM

An important difference though, is that Mattice reports his
rates to be greatest at 24°C while we found that Corbicula’s
potential filtration rate increased steadily through 30°C.
However, actual clam ingestion rates (filtration rate x activi-
ty level) decreased at the higher temperature in the grow-
out experiment as activity levels fell faster than filtration rates
rose.

ASSIMILATION

Assimilation rates for Corbicula are listed in Table 1.
Assimilation was constant over a range of temperature from
16 to 24°C. However, the efficiency dropped significantly at
30°C (p < 0.01 ), which we interpret as the result of the clam
entering a zone of thermal stress. Lauritsen (1986) reports
similar assimilation values for Corbicula feeding on monoalgal
diets of Chlorella (33%) and Scenedesmus (45.4%).

METABOLIC RATE

Metabolic rates (Table 1) increased with rising
temperature as the Newman-Keul multiple comparison test
revealed statistical differences between each nonadjacent
pair. Q10 values for differences in the rate of oxygen con-
sumption between 16-20, 20-24, and 24-30°C were 2.47, 1.81,
and 1 .35, respectively. The decrease in Q10 between 24 and
30°C may indicate the onset of thermal stress.

Reported literature respiration data for Corbicula in-
dicate that our values are somewhat higher than previous
estimates (Habel, 1970; McMahon, 1979). Some of this dif-
ference may be explained by the fact that other studies us-
ed larger animals and weight specific metabolism is known
to be inversely proportional to body size (Zeuthen, 1947). It
is interesting that McMahon’s data indicates a decrease in
metabolic rate at 30°C for both “acclimated” and “nonac-
climated” individuals. This decrease may indicate thermal
stress. Our rate data did not show a similar trend (Table 1),
however, the Q10 data does suggest some thermal inhibition
at the higher temperatures.

ACTIVITY LEVELS AND MORTALITY

Table 2 summarizes mortality and activity rates for the

Table 2. Temperature dependent activity and mortality rates for
Corbicula.

Temperature

(°C)

Activity level (o/o)1

Mortality (%)

98.9 a

2.2

96.3 a

4.1

90.1 a

3.3

97.3 a

6.9

95.1 a

4.6

98.2 a

7.1

74.1 b

5.2

8.0 c

1.8

—

100

Walues with the same letter are not statistically different at the 1%
level. (See text for statistical details.)

Table 3. The caloric value of the physiological rates used in the com-
putation of Corbicula scope for growth* (Fig. 1).

Temp.

A-R

(°C)

(cal/mo)

74.83

35.92

3.77

32.15

31.87

93.65

47.76

5.40

42.36

35.47

136.89

43.79

6.84

36.95

25.19

28.95

3.76

8.13

-4.37

-2.67

*See the methods section for computational details.

clam growth experiment. No mortalities were reported below
30°C while three clams died at this temperature and all
animals died in two weeks at 32°C. Differences in activity were
analyzed with a one-way ANOVA and a Duncan multiple
range test. The results indicate no difference in activity below
28°C which suggests that Corbicula filters continuously at
these tempertures. Activity falls at 28 and again at 30°C as
clams spend a greater portion of their time in the closed mode
(P< 0.01).

ENERGY BUDGET

The laboratory rates for Corbicula’ s filtration, assimila-
tion, respiration, and activity levels have been converted to
caloric equivalents per month in Table 3 and integrated into
a scope for growth model (Eq. 3) in Fig. 1 . The figure includes
two curves. The area bounded by the upper dashed line and
the abscissa is an estimate of Corbicula’ s caloric intake (A).
The area between the lower solid line and the X-axis is the
energy expended in respiration (R). The stippled area bet-
ween (A-R), represents the amount of energy available for
Corbicula’ s growth as a function of temperature.

Fig. 1. Predicted scope for growth model for Corbicula at
temperatures between 16°C and 30°C. Curve “A” is calculated from
the number of calories assimilated at each temperature, and “R"
from the number lost through respiration.

FOE AND KNIGHT: THERMAL ENERGY BUDGET

147

This modei allows us to make several important predic-
tions. First, optimal Corbicula growth for the “white” morph
should occur around 20°C. This is the temperature where Cor-
bicula has the greatest amount of net available energy. Sec-
ond, the model predicts negative growth to occur at
temperature above 29°C as more energy is being expended
at this temperature than can be obtained by filtration. Third,
the model predicts a high potential growth rate at
temperatures as low as 16°C.

GROWTH

Water quality parameters were measured during the
30-day growth experiment. Test temperatures varied less than
0.2°C in each treatment. Dissolved oxygen was always close
to saturation. Hydrogen-ion concentration appeared to be
strongly influenced by the photosynthetic activity of the algae
and rose and fell with the diel light cycle (values ranged be-
tween 8.3 and 8.8). Algal concentration ranged between 0.4
and 1.5 x 105 cells/m/ with a mean concentration for each
treatment of about 105 cells/m/. Nitrogen levels were always
low. Nitrite and nitrate concentrations were undetectable.
Mean total ammonia concentration ranged between 0.02 and
0.06 mg//. The concentration of toxic unionized ammonia was
consistently less than the 20 ^gll value which EPA considers
safe (EPA, 1975).

Clam growth is summarized in Table 4. Growth was
positive at temperatures below 30°C. Relative tissue growth
was particularly high at the lower temperatures evaluated.
For example, clams nearly tripled in weight during the month
at 20°C. Differences in tissue growth were analyzed by a one-
way ANOVA and a Duncan multiple range test. Tissue growth
appears optimal between 18 and 20°C and thereafter
decreases steadily with increasing temperature (p<0.05).
Growth was negative at 30°C. Shell growth was analyzed in
a similar fashion with like results. Shell growth was greatest
between 18 and 20°C and decreased rapidly at higher
temperatures (p<0.01). A slight amount of positive growth

was recorded at 30°C. This suggests that the processes of
shell and tissue growth may not occur simultaneously.

An advantage of using energy budgets is that the
animal’s caloric intake must balance its energy loss. The
degree to which these do not balance is a measure of the
errors included in the measurement of the various
physiological parameters. Figure 2 includes a comparison
between Corbicula' s predicted tissue growth derived from the
scope for growth model, and actual laboratory growth. Cor-
bicula’s predicted growth was always greater than that ac-
tually measured. However, the 95% laboratory growth con-
fidence limits include the predicted scope for growth value
at three or the four temperatures tested. This indicates that
the various physiological rates used in the calculation of the
scope for growth model are approximately correct. The data
also substantiates, at least for immature Corbicula, the use
of scope for growth (Widdows, 1978) to estimate instan-
taneous growth rates. We have no reasonable explanation
for the large differences between the predicted and measured
growth values for 24°C.

There are several hypotheses as to why the scope for
growth model overestimates Corbicula1 s actual growth. First,
the model assumes no pseudofecal production. Observations
show that Corbicula begins producing pseudofeces at algal
densities between 105 and 106 cells/m/. Small amounts of
pseudofeces were regularly seen intermixed with the fecal
pellets and were carefully separated out before estimating
assimilation. No attempt was made to calculate their caloric
value and subtract this from the ingestion rate. Consequent-
ly, we may have overestimated the animals’ net caloric in-
take. Second, respiration was estimated using starved
animals. Therefore, our respiration values approximate the
standard metabolism of Bayne et at. (1976b). Unpublished
data (ours) demonstrates that Corbicula’ s respiration almost
doubles when phytoplankton is introduced into the respira-
tion chamber. Presumably the increased metabolic costs are
the result of filtration and digestion of the algae. The rates

Table 4. Laboratory growth of Corbicula as a function of temperature.

Temp.

(°C)

Shell growth1’2
(mm/moth)

Tissue growth1’2
(mg/month)

Relative

tissue

growth

(0/0)

1.38 a

0.05

6.2 a b c

0.20

164.7

1.73 b

0.06

6.8 a b

0.19

181.6

1.74 b

0.16

6.9 b

0.21

186.7

1.03 c

0.08

5.8 c

0.17

151.4

0.96 c d

0.09

4.9 d

0.30

125.6

0.91 d

0.08

3.0 e

0.25

74.6

0.70 e

0.10

1.4 f

0.24

39,5

0.153 f

0.04

-0.52 g

0.21

-17.8

—

'Values with the same letter are not statistically different for shell growth at the 1% level and tissue growth

at 5% level.

2No growth is reported at 32°C as ail animals in the treatment died during the experiment.

148

CORBICULA SYMPOSIUM

14 16 18 20 22 24 26 28 30 32

TEMPERATURE

Fig. 2. Comparison of Corbicula' s predicted growth derived from the
scope for growth model and actual measured laboratory growth. The
vertical bars indicate the laboratory 95% growth confidence limits.

we report, therefore, may underestimate Corbicula’ s actual
daily metabolic costs. Finally we did not determine the amount
of energy lost through ammonia production. The marine
mussel Mytilus edulis is estimated to lose between 0.1 and
4% of its ingested energy as excreta (Bayne and Widdows,
1978; Widdows et a!., 1981).

Net production efficiency (K2) is defined as the amount
of energy an organism expends on growth and reproduction
(P) divided by the amount assimilated (A). The parameter is
a measure of the efficiency with which the ingested energy
is utilized for growth and reproduction. K2 values calculated
from Table 3 for Corbicula at 16, 20 and 24°C are 88.7, 74.3
and 57.5%, respectively. These values are similar to those
measured for an Asiatic clam population from Lake Arlington,
Texas (66-77%; Aldridge and McMahon, 1978). As noted by
these authors, Corbicula’ s K2 values are among the highest
ever recorded for a freshwater mollusc. Such a high K2 effi-
ciency should confer an exploitative-type competitive advan-
tage for Corbicula relative to other freshwater invertebrates.
This energetic advantage may, in part, help explain how the
recently introduced Asiatic clam so rapidly become a domi-
nant benthic invertebrate in many American water systems.

Our laboratory clam growth data (Table 4) is difficult
to compare with that of other researchers because of dif-
ferences in animal size and season of measurement.
However, our growth rates appear roughly comparable to the
average summer growth rates of 1 .60-2.50 mm/month
reported for small clams in field studies by O’Kane (1976),
Mattice (1979), and Eng (1979). We have also measured (in
the laboratory) the shell and tissue growth of another group
of somewhat smaller (5-8 mm shell length) clams fed on the
same trialgal diet (Foe and Knight, 1986). Their shell
and tissue growth was 0.39 ± 0.09 mm/month and 2.54 ±
0.75 mg/month (mean ± 1 standard error). The growth rate
of the latter group of clams was similar to that of other in-

dividuals of the same size caged in the Sacramento-San Joa-
quin Delta (2.43 ± 0.1 8 mg/month). The reason for the large
difference in growth rate between laboratory studies is not
well understood. However, the slower growth of the smaller
clams is undoubtedly, in part, due to the fact that Corbicula’ s
growth increases with body size until about 15-18 mm shell
length (Foe and Knight, in preparation). Also, the two groups
of clams were collected at different times and many,
therefore, not have been in similar physiological condition.
In other laboratory studies, we have obtained shell and tissue
growth rates of 2.78 mm/month and 9.25 mg/month for 6-9
mm clams fed on algae from water collected from the
Sacramento-San Joaquin Delta, spiked with nitrogen and
phosphorus and incubated in a green house for 4 days to
induce a phytoplankton bloom before being further enriched
with 106 cells/m/ Ankistrodesmus (Foe and Knight, 1985).
The enhanced clam growth in this latter study appears to
result primarily from the higher assimilation rate of estuarine
diatoms (-95%). Unfortunately, this latter, more successful,
culture method is not suitable for the present type of study
in which a defined algal diet with known caloric value is
needed. These reported variations in clam laboratory growth
do not negate the conclusions of the present study. However,
they do suggest that the various physiological rates which
together determine a clam’s energy budget may vary substan-
tially with animal size, season, ration type, and previous en-
vironmental history. Also, obviously, our reported energy
budget applies only for the “white” Corbicula morph. An in-
teresting future research topic might be to ascertain how the
energetics of the “purple” morph differ from that of the
“white” one. Such information could be valuable for predic-
ting how the two morphs will eventually partition their niche.

DISCUSSION

Temperature is a major environmental factor influenc-
ing the geographic distribution of aquatic invertebrates
(Kinne, 1970). Low temperatures may set the northern limit
of Corbicula' s distribution as clams were unable to survive
winter temperatures below 0°C in the New River, Virginia in
1975, while in contrast, population density remained stable
in an adjacent thermal outfall (Gainey et at, 1980). This
agrees well with laboratory studies which have shown that
the ultimate lower incipient lethal temperature for Corbicula
is 2°C (Mattice and Dye, 1975). By definition this is the lowest
temperature to which clams can be acclimated without
temperature related mortality.

Reproductive studies demonstrate that spawning is in-
duced biannually as water temperature passes through 16°C
(Eng, 1979; Heinsohn, 1958; Sickel, 1979). In northern states,
reproduction may be delayed for several months or even
reduced to a single period because of prolonged low winter
temperatures (Eng, 1979). We propose that the 2°C lower in-
cipient lethal temperature and the 16°C temperature depen-
dent reproductive cycle may constitute the critical lower
temperatures for Corbicula. Together they may define the nor-
thern range of the Asiatic clam.

Field observations suggest that Corbicula' s upper ther-
mal limit lies between 29 and 35°C. This temperature range

FOE AND KNIGHT: THERMAL ENERGY BUDGET

149

may determine the dam’s southern distribution. Haines
(1979) reported the complete mortality of animals transplanted
into sewage treatment ponds at St. Croix, Virgin Islands when
temperatures ranged between 25 and 35°C. Busch (1974)
observed high mortality in ponds where temperatures fre-
quently exceeded 32°C. Habel (1970) recorded almost com-
plete mortality in catfish enclosures when the temperature
rose to 35°C. Mattice and Dye (1 976) indicate that the upper
incipient lethal laboratory temperature for Corbicula is 32°C.
Our scope for growth model predicts a somewhat low upper
temperature limit as growth was negative above 29°C. Our
laboratory growth and mortality data (Tables 2 and 4) sub-
santiate this conclusion by showing adverse temperature ef-
fects at 30°C.

Field studies indicate that Corbicula has a positive
growth potential between 15 and 30°C. However, it is difficult
to predict the optimal growth temperature from these studies.
For example, in northern California growth is inhibited in the
winter at temperature below 14°C and in the summer at
temperatures above 25 and 30°C (Heinshon, 1958; Eng,
1979). Unpublished data (ours) from California’s Sacramento-
San Joaquin Delta show that the growth of caged clams cease
at temperatures below 15°C while optimal growth occurs bet-
ween 20 and 23°C (the maximum recorded ambient water
temperature). Sickle (1979) reports negligible growth below
15°C in the Altahama River, Georgia. Maximal growth occur-
red somewhere between 18 and 28°C. Mattice (1979) reported
maximum growth at 24°C for clams caged at the Glen Lynn
thermal outfall in Tennessee. Our laboratory culture ex-
periments demonstrate that Corbicula’ s growth is maximal
at 20°C. Analysis of the scope for growth model reveals that
this is not the result of the dominant performance of any one
physiological process but rather results from small adjust-
ments in each rate. As a result, Corbicula exhibits no evidence
of a thermal compensatory adjustment enabling it to main-
tain a stable scope for growth over a wide thermal range as
has been reported for several marine interidal invertebrates
including the bivalve Mytilus edulis (Bayne ef a/., 1976b).
Laboratory growth experiments (Table 4) confirm a single nar-
row growth maximum between 18-20°C. We conclude,
therefore, that Corbicula is best adapted to grow in en-
vironments with ambient temperatures near 20°C.

Finally, the laboratory based scope for growth model
predicts a high potential growth rate at temperatures as low
as 16°C. This is in contradiction to field work (Heinsohn, 1958;
Sickle, 1979; Eng, 1979) which has reported that growth
ceases at this temperature. This contradiction implies that
factors other than temperature may control Corbicula’ s growth
at least in the lower portion of its temperature range. This
conclusion led us to investigate the effects of ration size on
Asiatic clam growth. We have found that Corbicula is food
limited during most of its growing season in California’s rather
eutrophic Sacramento-San Joaquin Delta (Foe and Knight,
1985). This result leads us to conclude that in most other
systems the growth dynamics of Corbicula are also probably
being determined to a large extent by the amount of available
food and not by the ambient water temperature. Hence, we
recommend that the optimal thermal growth conclusions of

this study be applied to field situations with caution.

ACKNOWLEDGMENTS

The study was supported by Grant No. 440344-22870 from

the Environmental Protection Agency. We thank Carol DiGiorgio for

help with the algal culture and clam growth experiments and Barry

Votaw for assistance with the respiration portion of the study.

LITERATURE CITED

Aldridge, D. W. and R. F. McMahon. 1978. Growth, fecundity and
bioenergetics in a natural population of the Asiatic freshwater
clam, Corbicula manilensis, Philippi, from North Central Texas.
Journal of Molluscan Studies 44: 49-70.

Armstrong, D. A., D Chippendale, A. W. Knight and J. Colt. 1978.
Interaction of ionized and unionized ammonia on short-term
survival and growth of prawn larvae, Macrobrachrium
rosenbergii. Biological Bulletin 154, 15-31.

Bayne, B. L., J. Widdows and P. J. Thompson. 1976(a). Physiological
Integrations, pp. 261-293. In: (B. L. Bayne, ed.), Marine
Mussels, Cambridge University Press, Cambridge.

Bayne, B. L., R. L. Thompson, J. Widdows. 1976b. Physiology I. pp.
121-159. In: (B. L. Bayne, ed.), Marine Mussels, Cambridge
University Press, Cambridge.

Bayne, B. L., and J. Widdows. 1978. The physiological ecology of
two populations of Mytilus edulis L. Oecologia 37:137-162.

Boozer, A. C., and P. E. Murkes. 1979. Observations on the finger-
nail clam Musculum partumeum (Pisidiidae) and its associa-
tion with the introduced Asiatic clam Corbicula fluminea.
Nautilus 93:73-83.

Burch, J. Q. 1944. Checklist of west American mollusks, family Cor-
biculidae. Minutes of the Conchological Club of Southern
California 36:18.

Busch, R. L. 1974. Asiatic clams, Corbicula manilensis, Philippi as
biological filters in channel catfish, Ictalurus punctatus
(Rafinesque) cultures. M. S. Thesis Auburn University, Auburn,
Alabama. 84 pp.

Clarke, A. H. 1981 . Corbicula fluminea in Lake Erie. Nautilus 95:83-84.

Conover, R. J. 1966. Assimilation of organic matter by zooplankton.
Limnology and Oceanography 1 1 :338-354.

Counts, C. L. III. 1981(a). Corbicula fluminea (Bivalvia: Sphaeriacea)
in British Columbia. Nautilus 95(1): 12-1 3.

Counts, C. L. III. 1981(b). Corbicula fluminea on the Delmarva Penin-
sula Maryland. Veliger 24:187-188.

Crisp, D. J. 1971. Energy flow measurements, pp. 197-279. In: N.
A. Holmes and A. D. McIntyre (eds.), Methods for the Study
of Marine Benthos, (IBP Handbook No. 16). Oxford: Blackwell
Scientific Publications.

Eng, L. L. 1979. Populations dynamics of the Asiatic clam Corbicula
fluminea (Muller) in the concrete-lined Delta-Mendota Canal
of Central California, pp. 39-68. In: J. C. Britton (ed.), Pro-
ceedings of the First International Corbicula Symposium. Texas
Christian University Research Foundation Publication, Texas.

Environmental Protection Agency (EPA). 1975. Methods of Acute Tox-
icity Tests with Fish, Macroinvertebrates and Amphibians. U.S.
Environmental Protection Agency Ecological Research Series.
EPA 660 3-75-009: 61 p.

Foe, C. G. 1983. Flow and substrate requirements of juvenile Cor-
bicula fluminea In: A. W. Knight (ed.), The Evaluation of Con-
taminated Sediment Employing Selected Benthic Freshwater
Organisms. University of California, Department of Land, Air
and Water Resources, Water Sciences and Engineering Paper

150

CORBICULA SYMPOSIUM

No. 4516: 27-42.

Foe, C. G. and A. W. Knight a. 1986. Growth of Corbicula fluminea
fed algal and artificial diets. Hydrobiologia. 133, 155-164.

Foe, C. G. and A. W. Knight b. 1985. The effect of phytoplankton
and suspended sediment on the growth of Corbicula fluminea
(Bivalvia). Hydrobiologia. 127, 105-115.

Fox, D. L., H. U. Sverdrup and J. P. Cunningham. 1937. The rate
of water propulsion by the California mussel. Biological Bulletin
Marine Biological Laboratory, Woods Hole, Massachusetts 72:
417-38.

Gardner, J. A., W. R. Woodall, A. A. Staats and J. F. Napoli. 1976.
The invasion of the Asiatic clam ( Corbicula manilensis) in the
Altahama River, Georgia. Nautilus 90: 117-125.

Goss, L. B., and C. Cain. 1975. Condenser and raw water system
fouling by Corbicula. In An assessment of technical and
ecological effects of biofouling control procedures at thermal
power cooling water systems, Workshop sponsored by Power
research institute RP-49 and Maryland power plant siting pro-
gram, John Hopkins, University of Baltimore, Maryland, June
16-17. Abstract in Corbicula Newsletter 1(2):6.

Graney, R. L., D. S. Cherry, J. H. Rodgers and J. Caines. 1980. The
influence of thermal discharge and substrate composition on
the population structure and distribution of the Asiatic clam
Corbicula fluminea in the New River, Virginia. Nautilus 94(4):
130-135.

Habel, M. L. 1970. Oxygen consumption, temperature tolerance and
filtration rate of the introduced Asiatic Clam Corbicula manilen-
sis from the Tennessee River. M. S. Thesis, Auburn Univer-
sity, Auburn, Alabama. 91 pp.

Haines, K. C. 1979. The use of Corbicula as a clarifying agent in ex-
perimental tertiary sewage treatment process on St. Croix,
U. S. Virgin Islands, pp. 165-177. In: J. C. Britton (Ed.), Pro-
ceedings of the First International Corbicula Symposium, Texas
Christian University Research Foundation Publication, Texas.

Heinsohn, G. E. 1958. Life History and Ecology of the Freshwater
clam Corbicula fluminea. M. S. Thesis, University of Califor-
nia Berkeley. 64 p.

Hillis, D. M. and J. C. Patton. 1982. Morphological and electrophoretic
evidence for two species of Corbicula (Bivalvia: Corbiculidae)
in North America. American Midland Naturalist 108(1):74-80.

Kinne, O. 1970. Temperature, invertebrates, pp. 407-514. In: O Kinne
(ed.) Marine Ecology, Vol. 1. Wiley Interscience, London.

Kraemer, L. L. 1979. Corbicula (Bivalvia: Sphaeriacea) vs indigenous
mussels (Bivalvia: Unionacea) in U.S. rivers: A hard case for
interspecific competition? American Zoologist 19:1085-1096.

Lauritsen, D. D. 1986. Assimilation of radiolabelled algae by Cor-
bicula. IN: J. C. Britton ed., Proceedings of the Second Inter-
national Corbicula Symposium. American Malacoiogical Bulletin
Special Edition No. 2: 219-222.

Mattice, J. S. 1979. Interactions of Corbicula sp. with power plants,
pp. 119-139. In: J. C. Britton (ed.), Proceedings of the First
International Corbicula Symposium, Texas Christian Univer-
sity Research Foundation Publication, Texas.

Mattice, J. S. and L. L.Dye. 1975. Thermal tolerance of the adult
Asiatic clam. pp. 130-135. In: G. W. Esch and G. W. McFarlane
(eds.) Thermal Ecology II. Conference 750425 National
Technical Information Service, Springfield, Virginia.

McLeod, M. J. 1986. Electrophoretic variation in U.S. Corbicula. IN:
J. C. Britton (ed.), Proceedings of the Second International Cor-
bicula Symposium. American Malacoiogical Bulletin Special

Edition No. 2: 125-132.

McMahon, R. F. 1 979. Response to temperature and hypoxia in the
oxygen consumption of the Introduced Asiatic freshwater clam.
Corbicula fluminea (Muller) Comparative Biochemistry and
Physiology 63A:383-388.

McMahon, R. F. 1977. Shell size frequency distributions of Corbicula
manilenis (Philippi) from a clam fouled steam condenser.
Nautilus 91:54-59.

Morton, B. 1979. Freshwater fouling Bivalves, pp. 11-15. In: J. C.
Britton (ed.), Proceedings of the First International Corbicula
Symposium. Texas Christian University Research Publication,
Texas.

Nichols, H. W. 1973. Growth media, freshwater, pp. 7-25. In: J. R.
Stein (ed.), Phycological Methods I. Cambridge University
Press.

O’Kane, K. D. 1 976. A population study of the exotic clam Corbicula
manilensis (Philippi, 1841) in selected Texas reservoirs. M.
S. Thesis, Texas Christian University, Fort Worth, TX.

Prokopovich , N . P. 1 969. Deposition of clastic sediments by clams.
Journal of Sedimentary Petrology 39:891-901.

Schulte, E. H. 1975. Influence of algal concentration and temperature
on the filtration rate of Mytilus edulis. Marine Biology
30:331-341.

Sickel, J. B. 1979. Population dynamics of Corbicula in the Altamaha
River Georgia, pp. 69-88. In: J. C. Britton (ed.), Proceedings
of the First International Corbicula Symposium, Texas Chris-
tian University Research Foundation Publication, Texas.

Solorzano, L. 1969. Determination of ammonia in natural water by
the phenolhypochlorite method. Limnology and Oceanography
14:799-801.

Standard Methods for the Examination of Water and Wastewater.
1975. 14th Edition, American Public Health Association,
American Water Works Association, and Water Pollution Con-
trol Federation, Washington, D. C. 1193 p.

Starr, R. C. 1978. Culture collection of algae. Journal of Phycology
14(4):92-96.

Trana, F. B. 1982. Occurence of the Asiatic clam Corbicula fluminea
in the Raritan River, New Jersey. Nautilus 96:6-8.

Warren, C. E., and G. E. Davis. 1967. Laboratory studies on the
feeding, bioenergetics and growth of fish. pp. 175-214. In: (S.
D. Gerking, ed.), The Biological Basis of Freshwater Fish Pro-
duction, Blackwell Scientific Publications Oxford.

Widdows, J. 1978. Physiological indicies of stress in Mytilus edulis.
Journal of the Marine Biological Association of the United
Kingdom 58:125-142.

Widdows, J., D. K. Phelps, and W. Galloway. 1981. Measurement
of physiological conditions of mussels transplanted along a
pollution gradient in Narragansett Bay. Marine Environmen-
tal Research 4:181-194.

Winter, J. E. 1978. A review of the knowledge of suspension feedings
in the lamellibranchiate bivalves with special reference to ar-
tificial aquaculture systems. Aquaculture 13:1-33.

Winter, J. E. 1969. On the influence of food concentration and other
factors on filtration rate and food utilization in the mussel Arc-
tica islandica and Modiolus modiolus Marine Biology A. Q7 A 35.

Zenthen, E. 1947. Body size and metabolic rate in the animal kingdom
with special regard to the marine microfauna. Comptes Rendus
des Travaux du laboratoire Carlsberg, Serie Chimique
26: 17-161.

A REASSESSMENT OF GROWTH RATE, LIFE SPAN, LIFE CYCLES AND
POPULATION DYNAMICS IN A NATURAL POPULATION AND FIELD CAGED
INDIVIDUALS OF CORBICULA FLUMINEA (MULLER) (BIVALVIA:

CORBICULACEA)1

ROBERT F. MCMAHON AND CAROL J. WILLIAMS2
SECTION OF COMPARATIVE PHYSIOLOGY
DEPARTMENT OF BIOLOGY
BOX 19498

THE UNIVERSITY OF TEXAS AT ARLINGTON
ARLINGTON, TEXAS 76019, U.S.A.

ABSTRACT

A population of Corbicula fluminea in the Clear Fork of the Trinity River, Tarrant County, Texas,
was sampled bimonthly from 10 September 1980 through 20 December 1982. Growth rates of caged
and free living marked individuals were also monitored. The population had a distinct biannual pattern
of reproduction, incubation of eggs and embryonic stages occurring from late March through July
and again from late August through November, giving rise to distinct "early” and “late” generations,
respectively. Growth rates, life spans and reproductive patterns for individual generations were variable.
The late generation was not successfully produced in the fall of 1982. Population density ranged from
305 to 16198 clams nrr2. The high reproductive capacity of the population allowed quick recovery of
density after catastrophic population declines.

Shell growth of caged and freeliving marked individuals closely paralleled that of individual
generations. Growth rates were maximal during the warmest periods of the year and were minimal
during the colder winter months. Degrowth [decreases in shell length (SI)] occurred in larger individuals
during the winter and catastrophic declines in population density. Growth rates of caged individuals
were negatively correlated with size and positively correlated with temperature allowing the develop-
ment of a growth rate model based on these two parameters. Maximum growth rates ranged from
5.4 mm SL 30 days*1 to 0.8 mm SL 30 days*1 for individuals with a SL of 5 mm and 30 mm, respectively.

The high growth rates, attenuated life spans, high fecundity, high proportion of immature in-
dividuals in the population, and ability to rapidly recover from catastrophic declines in density
characteristic of the Clear Fork C. fluminea population are all life history traits associated with op-
timalization of reproduction and survival to maturity in highly unstable habitats. Such characteristics
also account for this species’ rapid spread through North American drainage systems and its nature
as a biofouler of industrial raw water systems.

The growth, reproduction and life cycle of North
American populations of the Asian freshwater bivalve, Cor-
bicula fluminea (Muller), have been a matter of considerable
confusion and controversy since the first reports of its in-
troduction into the United States from Southeast Asia (Burch,
1944) (for reviews of the population dynamics of C. fluminea

’This research was supported by a grant from the Organized
Research Funds of the University of Texas at Arlington to R. F.
McMahon.

2Present Address: Department of Microbiology, The University of
Texas Health Science Center, 5323 Harry Hines Boul., Dallas, Texas,
75235, U.S.A.

see Britton and Morton, 1979; McMahon, 1983). While the
majority of investigators now agree that C. fluminea has two
annual reproductive periods in which juveniles are released
(mid-spring through mid-summer and late summer through
early to late fall) in both North America (Aldridge and
McMahon, 1978; Boozer and Mirkes, 1979; Eng, 1979; Dreier
and Tranquilli, 1981; McMahon and Williams, 1986; Sickle,
1973, 1979; Williams and McMahon, 1986) and endemic
Asian populations (Morton, 1977), there continues to be
significant disagreement in the literature regarding this
species’ growth rate, population age structure and life span.

The earliest investigations of growth rates and life span
in C. fluminea were based on size distributions of one time

American Malacologies! Bulletin, Special Edition No. 2(1986): 151-166

151

152

CORBICULA SYMPOSIUM

only population samples and/or on growth rings or annuli in
the periostracum and/or mineral portions of the shell. These
initial estimates appeared to be biased toward the previous-
ly known, relatively slow growth rates of unionid (Negus, 1966)
and pisidiid bivalves (Avolizi, 1976, Hornback et al., 1980,
Way et al., 1980, and references therein) and were often bas-
ed on the erroneous assumption that C. fluminea had only
one annual reproductive period. Later studies described a
biannual reproductive pattern which produced two distinct
generation size classes each year (Aldridge and McMahon,
1978; Britton et al., 1979; Coldiron, 1975; Eng, 1979; Hein-
sohn, 1958; McMahon and Williams, 1986; Morton, 1977;
Williams and McMahon, 1986). In addition, shell annuli
numbers are not closely associated with age in C. fluminea
(Dudgeon, 1980; Heard, 1964).

Utilization of size class-frequency and shell annuli data
from single collections to interpret population dynamics based
on the assumption of a single annual reproductive period led
to initial estimates of the life span of C. fluminea that were
often up to twice actual values and, therefore, to under-
estimations of shell growth rates (Diaz, 1974; Gardner, In-
gram, et al., 1976; Gunning and Suttkus, 1966; Hubright,
1966; et al., 1964; Keup, et al., 1963; Sickel, 1973; Sinclair
and Isom, 1963). These early estimates of growth rates
ranged from 2 mm to 12 mm in the first year of life. In con-
trast, studies recognizing that this species produced two
distinct annual generations estimated growth rates to range
from 16 mm to 30 mm in the first year of life (for a review
of growth and reproduction in C. flumina see McMahon, 1983).

Some early published estimates also suggested that
growth rates increased with size and age in C. fluminea, a
very atypical pattern among molluscs (for a review of estima-
tions of growth rate see Britton et al., 1979). Recent studies
have indicated that C. fluminea both grows more rapidly and
has a much shorter life span than previously suspected.
These studies indicated that growth rate declined with in-
creasing size and age and that maximum life span was ap-
proximately 1.5 to 4 years in North American (Aldridge and
McMahon, 1978; Britton, ef al., 1979; Eng, 1979; Heinsohn,
1958; Williams and McMahon, 1986), Asian (Morton, 1977)
and African populations (Leveque, 1973).

Reported growth rates of enclosed individuals of C.
fluminea ranging from 2.0 mm / 30 days to 6.5 mm / 30 days,
closely correspond to those determined by iterative population
sampling. In addition, they have provided strong collaborative
evidence that growth rate is negatively correlated with size
and age in C. fluminea (Britton et al., 1979; Buttner and
Heidinger, 1980; Dreierand Tranquilli, 1981; Mattice, 1979;
O’Kane, 1976).

In spite of the apparent general concensus in the re-
cent literature on the population dynamics of C. fluminea, ac-
counts of North American populations with reduced growth
rates and life spans extended beyond three to four years still
appear in the literature, particularly for populations in the
southeastern United States (Mattice and Wright, 1985). Fur-
thermore, even though C. fluminea appears to be most highly
adapted for life in lotic habitats (Kraemer, 1979; McMahon,
1 983; Morton, 1 982), there have been no examinations of the

population dynamics of this species in a lotic habitat. Nor have
there been any concurrent estimations of growth rate from
both the iterative analysis of population size-frequency data
and from field enclosed individuals. In order to more rigorous-
ly reassess the population dynamics of this species in a
natural lotic habitat, an intensive, two year study of growth,
reproduction and life cycle was carried out on a river popula-
tion of C. fluminea in North Central Texas. This study reports
the results of the analysis of the age-size structure of iterative
bimonthly population samples and densities of that popula-
tion. In addition, concurrent determinations were made of the
growth rates of marked, released, recaptured individuals and
of individuals maintained in a field enclosure. The data from
these studies allowed development of a simple model of in-
dividual growth rate in C. fluminea based on regression
against shell length and ambient water temperature.

METHODS

Specimens of C. fluminea were collected bimonthly
from 10 September 1980, to 20 December 1982, from a
population in the Clear Fork of the Trinity River approximately
one mile downstream from the outfall of Lake Benbrook, an
artificial impoundment of the Clear Fork, in Tarrant County,
Texas. The Clear Fork population consisted entirely of in-
dividuals of the “white shell morph” (characterized by white
to light purple or light blue internal shell color and widely
spaced shell sulcations) as described by Hillis and Patton
(1982). The collection site was an area of the river approx-
imately 12 m wide, 15 m long and 0.25-1 .5 m in depth at nor-
mal river levels. The substratum was course limestone gravel
mixed with larger limestone boulders, sand and silt. Generally
the site was characterized by relatively moderate current flow
(*4m / min) with the exception of isolated periods of high
runoff.

Specimens were quantitatively sampled with an Ekman
bottom dredge (sampling area = 0.19 m2) forced by hand
deeply into the substratum. Clams were removed from dredg-
ed material by sieving it through a 1 .0 mm mesh. Sampling
continued until a sample size of at least 100 individuals was
taken. Average sample size over all collections was 341 in-
dividuals (s.d. = ± 261, range = 104-1243, n = 47). All in-
dividuals were returned to the laboratory within two hours of
collection and immediately fixed in 12% (by volume with
H20 neutralized) formaldehyde. Large individuals (Length >
8 mm) were maintained in water at 70°C for approximately
15 min to induce shell gaping prior to fixation.

The shell length (SL, hereafter referred to as “length”,
the greatest anterior-posterior dimension across the shell
valves) of each individual in the sample was measured to the
nearest 0.1 mm. The SL of larger individuals (SL -^-6 mm) was
measured with a dial caliper and that of smaller individuals
(SL ^ 6 mm) with an eye piece micrometer at lOx under a

binocular dissecting microscope.

The individuals in each sample were then divided in-
to consecutive 0.2 mm size classes and each size class was
expressed as a percentage of sample size. These values were
then plotted as frequency histograms against sampling date

MCMAHON AND WILLIAMS: CORBICULA GROWTH RATE

153

over the entire coiiection period, individuals in each sample
were assigned to separate generations by visual separation
of the size-frequency histograms into distinct size class group-
ings with separate modal SL values (after the method of
Aldridge and McMahon, 1978). Mean SL, standard deviations
and ranges were computed for each generation in each
sample.

The reproductive condition of a subsample of 1 0 to 1 5
adult clams (SL > 10 mm, Aldridge and McMahon, 1978)
was assessed for each collection. The tissues of each in-
dividual in the subsample were excised and the inner
demibranchs of both ctenidia (gills) examined under a
binocular microscope for the presence of fertilized eggs,
developmental stages and/or juveniles being incubated in
their interlamellar cavities (for a discussion of reproduction
in C. fluminea see Kraemer, 1977, 1978; Kraemer and Lott,
1977).

Throughout the collecting period, individually marked
specimens of C. fluminea were maintained at the collecting
site in an enclosure constructed of 5 mm mesh galvanized
hardware cloth that was 45 cm long, 30 cm wide and 7 cm
high. This cage was anchored directly on the substratum at
a depth of 0.7 m by covering it with an arched cement tile
that allowed water circulation through the top and sides of
the cage. Water current velocity at the cage site was 9 m
min'1. On 10 September 1980 the cage was initially stocked
with 66 individuals. The shell lengths of caged individuals
were measured to the nearest 0.1 mm with a dial caliper and
a permanent identifying number was scraped on the shell with
a scapel blade. The SL of caged individuals which ranged
from < 7.0 mm to > 40.0 mm, were selected to be represen-
tative of the SL range of the natural population. At every col-
lection the SL of each caged individuals was measured to
the nearest 0.1mm with a dial caliper. Dead individuals were
removed from the cages and new individuals were recruited
to the cage periodically to maintain a size range represen-
tative of the natural population. The number of caged in-
dividuals remained at 44-70 throughout the sampling period.
Mortality of caged individuals was low and generally never
accounted for more than 10% of caged individuals between
adjacent collections.

To determine the field growth rates of freeliving in-
dividuals, 994 clams were individually marked and their SL
measured to the nearest 0.1 mm. These were released at the
collecting site in an area (2m x 2m) immediately downstream
from the cage on 10 September 1980. The area of release
was sampled at each collection and the lengths of any recap-
tured individuals measured to the nearest 0.1 mm. After
measurement captured individuals were returned to the col-
lecting site. On 1 7 May 1 981 , and 26 May 1981, another 215
and 77 marked individuals were released at the collecting
site, respectively. The sampling of released individuals con-
tinued until 13 September 1981.

Physical measurements of ambient air temperature,
ambient water temperature, pH, conductivity and dissolved
oxygen (utilizing a modified Winkler assay: Helliage, Model
342-DO) were determined at each collection. Total calcium
and magnesium water hardness were determined by EDTA

titration on 23 July, 1982, and 5 August, 1982. All collections
were carried out between 1100 and 1500 hours, a period
when ambient water temperatures, pH and dissolved oxygen
values approach their daily maxima.

RESULTS

Dissolved oxygen concentration at the collection site
remained close to air saturation levels (Fig. 1A). Mean
dissolved 02 concentration was 6.8 mg O2/'1 (s.d. = ± 2.7,
n = 36) over the course of collection. With the exception of
an unusually high value of 898 ^mho cm-2 recorded on 1
October 1 980, and an unusually low value of 1 1 9 ^mho cm-2
on 5 August 1 982, conductivity ranged from 200 to 450 ^mho
cm-2 (mean conductivity = 31 4.2 ^mho cm’2, s.d. = ± 103.4,
n = 48)(Fig. IB). Total water hardness was 102 mg Ca /-1
and calcium hardness, 82 mg Ca /-1. Such conductivity and

ASOND JFMAMJ JAS ONDJFMAMJJ ASON DJ
1980 1981 1982

Fig. 1 . Physical parameters at the Corbicula fluminea collecting site
on the Clear Fork of the Trinity River, Texas. The horizontal axis is
months of the year. A Dissolved oxygen concentration (mg 02//).
B. Conductivity (/xmho / m2). C. pH. D. Ambient air (open triangles
connected by dashed lines) and water temperatures (open circles,
solid lines) in °C.

154

CORBICLJLA SYMPOSIUM

hardness values are characteristic of the moderately hard
waters that generally occur in North Central Texas (Aldridge
and McMahon, 1978; McMahon, 1975, 1976). Ambient pH
ranged between 5.90 and 8.46. A pH of less than 7.0 was
recorded on only 7 of 48 collections, indicating that the Clear
Fork was generally alkaline (Fig. 1C). Ambient air temperature
was more variable than water temperature, ranging from
2. 6°C to 34. 2°C (mean air temperature = 22.0°C, s.d. = ±
7.8, n = 48) (Fig. ID). A minimum ambient water temperature
of 4.8°C occurred on 20 December 1982, and a maximum
of 29°C, on 7 June, 19 July and 2 August, 1981 (Fig. ID).

Mid-summer water temperatures in the Clear Fork of the Trini-
ty River were significantly higher from June through August
in 1981 (mean water temperature = 27.9°C, s.d. = ±1.2,
range = 26.2-29.0, n = 7) than in 1982 (mean water
temperature = 24.0°C, s.d. = ± 1.6, range = 22.0-25.8,
n = 4) (Fig. ID).

In 1982 the Clear Fork C. fluminea population had two
distinct reproductive periods during which fertilized eggs,
developmental stages and juvenile individuals were incubated
in the interlamellar spaces of the inner demibranches followed
by the spawning of juveniles (Fig. 2). An “early” reproduc-

1980 1981 1982

Fig. 2. Mean generation shell lengths (SL) in bimonthly samples of the Clear Fork Corbicula fluminea population. The horizontal axis is months
of the year over the collection period. Open circles connected by solid lines represent the shell length of separate generations. Vertical bars
are standard deviations. Points without standard deviation bars are means of samples with n < 7. Specific generations are designated by
time (E = early or L = late) and year of recruitment to the population, ie., E-80 = a generation recruited from the early reproductive period
of 1980, L-81 = a generation recruited from the late reproductive period of 1981. The solid bars above the horizontal axis indicate periods
when adult individuals were observed to incubate fertilized eggs and developmental stages in the inner demibranchs.

MCMAHON AND WILLIAMS: CORBICULA GROWTH RATE

155

tive period extended from late March through the end of Ju-
ly 1981, and the “late” reproductive period extended from
the end of August through the end of November 1 981 . Dur-
ing 1 982 the only significant reproductive effort observed was
the “early” reproductive period from late April through early
September (Fig. 2).

The early reproductive periods gave rise to early
generations designated as either E-81 or E-82, which first ap-
peared in the population as distinct size classes with a mean
SL of 2.6 mm on 19 July 1981 and 3.0 mm on 23 July 1982,
respectively (Fig. 2). The 1981 late reproductive period
similarly gave rise to a late generation designated L-81 , first
appearing as distinct size grouping with a mean SL of 4.1
mm on 10 November 1981 (Fig. 2). The E-81 generation grew
rapidly through the summer and fall, reaching a mean SL of
1 8.0 mm by December 1981. The E-82 generation grew more
slowly, reaching a mean SL of only 8.6 mm by December
1982 (Fig. 2). The L-81 generation displayed only a small in-
crease in mean SL during the winter of 1981-1 982, but, subse-
quently grew rapidly through the following spring, summer
and fall to reach a mean SL of 20.6 mm by December 1982.
The E-81 generation disappeared from the collections at a
mean SL of 23.8 mm on 30 September 1982, after 1 .5 years
of life (Fig. 2). In contrast, some individuals of generations
spawned prior to 1 981 , appeared to survive for longer periods.
These generations were represented by distinctly larger size
classes in the samples and were designated as the E-80 and
L-80 generations (Fig. 2). In January 1981 these generations
had a mean length of 23.0 mm and 16.7 mm, respectively.
They could no longer be separated as distinct size classes
by July 1981, and were, therefore, combined into a single
size class designated as the E-81 + L-81 generation. These
combined generations reached a mean length of 28.5 mm
by the January 1 982, and disappeared from the samples as
two year old individuals on 10 June 1982, after achieving a
mean length of 35.6 mm (Fig. 2).

Initially, two other distinct size classes of larger clams
appeared in the samples. One had a mean SL of 32.2 mm
and represented the combined E-79 + L-79 generations. This
group grew to a mean length of 41.0 mm by 19 April 1982,
and then disappeared from subsequent samples after three
years of life (Fig. 2). The largest distinct size grouping in the
initial samples had a mean SL of 40.4 mm and represented
the combined E-78 + L-78 generations (Fig. 2). This group
reached a mean length of 42.2 mm on 21 June 1981 , before
disappearing from the samples after three years of life (Fig. 2).

Maximum density values in the Clear Fork C. fluminea
population were clearly associated with the recruitment of new
individuals from the early and late generations (Figs. 3A, D
and E). For example, maximum density in 1981 was 9445
clams nr2 on 3 December. Of this value, 8714 clams nr2 or
92% were newly recruited individuals of the E-81 and L-81
generations (Figs. 3A and D). Similarly, the E-82 generation
accounted for 1 5524 clams nrr2 or 96% of the maximum den-
sity value of 16198 clams nr2 recorded on 27 August 1982
(Figs. 3A and E).

Annual mortality rates were relatively high in all age
classes. Within the first year of life the density of the E-81

generation declined from a maximum of 8714 clams rrr2 on
3 December 1981, to 96 clams nr2 on 3 December 1982,
yielding an annual mortality rate of 98% (Fig. 3D). In the sam-
ple of 14 December 1980, the generation densities were as
follows: 1.5 year old individuals of the E-80 + L-80 genera-
tions, 2250 clams nr2 (Fig. 5C); 2.5 year old individuals of
the E-79 + L-79 generations, 693 clams nrr2 (Fig. 5B); and

Fig. 3. Seasonal variation in the density of the Clear Fork Corbicula
fluminea population. The horizontal axes for all figures are months
of the year, and the vertical axes, population density in clams (1 03)
/ m2. Densities of specific generations are indicated by time (E or
L) and year of generation recruitment (see caption of Fig. 3 for fur-
ther explanation). A. Density of the total population. B. Densities of
the combined E-78 + L-78 generations and the combined E-79 +
L-79 generations. C. Densities of the E-80 (open circles), L-80 (open
triangles) and subsequently during collections of these generations
combined (E-80 + L-80) (solid circles). D. Densities of the E-81 (open
circles) and L-81 generations (open triangles). E. Density of the E-82
generation. The vertical arrow indicates the date of a hail storm and
tornado associated with a catastrophic density decline.

156

CORBICULA SYMPOSIUM

3.5 year old individuals of the E-78 + L-78 generations, 19
clams nr2 (Fig. 3B). These data yield the following mortality
rates: 69% in the second year of life (2250 - 693/2250 x 100);
and 97% in the third year of life (693-19/693 x 100).

There were large year-to-year variations in the relative
success of early and late generations. The density of the E-80
generation was nearly equal to that of L-80 in 1 981 , indicating
nearly equal recruitment rates (Fig. 3G). In contrast, the den-
sity of the L-81 generation was only 2.7% that of the E-81
generation from early November 1981 through mid-June
1982, suggestive of extremely poor recruitment of the L-81
generation to the population (Fig. 3D). This trend towards poor
recruitment of the late generation in the Clear Fork C. fluminea
population culminated in 1982 when there was no recruitment
of a late generation to the population (Fig. 3E).

A catastrophic density decline of the Clear Fork C.
fluminea population occurred in the spring of 1981. Total den-
sity declined from 2655 clams nr2 on 26 April 1981, to 725
adult clams nr2 on 1 0 May 1981. Density subsequently declin-
ed to 305 clams nr2 by 21 June 1981 (Figs. 3A, B and C),
yielding an 89% mortality over a 25 day period. This massive
population decline was associated with an extensive hail
storm and the passage of a tornado directly over the col-
lection site on the night of 9 May 1981. The very large hail
stones (diameter < 4 cm) and cold rain appeared to have

rapidly reduced water temperatures at the site. The following
morning water temperature was 19°C, 4-5°C below normal
temperatures for that period (26 April 1 981 = 23.5°C; 26 May
1981 = 26.1 °C) (Fig. ID). This initial cold shock appeared
to have induced a major, instantaneous mortality in the C.
fluminea population. The decomposing clam tissues caused
hypoxic conditions, resulting in a subsequent near elimina-
tion of both the remaining C. fluminea and fish populations
for many miles downstream. During this period the E-78 +
L-78 and E-79 + L-79 generations were completely eliminated
from the population, while the densities of the E-80 and L-80
generations were severely reduced (Figs. 3B and C). In ad-
dition, recruitment of the E-81 generation was delayed by two
to three weeks (Fig. 3D). In spite of this catastrophic reduc-
tion in density, subsequent recruitment of the E-81 and L-81
generations allowed recovery of the population to normal den-
sities of approximately 4000 clams nr2 by the winter of
1981-1982 (Figs. 3A and D). During the spring of 1982 den-
sity reached 14,000 clams nrr2, following the highly successful
recruitment of the E-82 generation (Figs 3A and E).

The shell growth rates of caged individuals appeared
to closely parallel those of marked individuals released into
the natural population when they were visually compared over
the same time period (October 1980-August 1981) (Fig. 4).
The equivalency of growth rates of caged and freeliving in-

1980 1981

Fig. 4. Growth of selected individuals of Corbicula fluminea marked, released and recaptured in the field (solid circles connected by solid
lines) compared with that of individuals maintained in field enclosures (open circles connected by dashed lines). The horizontal axis is months
of the year, the vertical axis, shell length in mm.

MCMAHON AND WILLIAMS: CORBICULA GROWTH RATE

157

dividuals indicated that the method of enclosure utilized did
not inhibit shell growth as reported for other enclosure ex-
periments with C. fluminea (Britton et al., 1979).

The growth rates of caged clams had a strong negative
linear correlation with SL. Forty six separate least squares
linear regressions of the daily rate of increase in SL versus
length were computed from the progressive change in the

SL of caged individuals over sequential measurement periods
(Table 1). Of these 46 growth rate regressions 36 were signifi-
cant at P < 0.1 and 26 at P < 0.001 (Table 1). The relation-
ship between growth rate and SL was not generally signifi-
cant during winter periods of little or no observable growth
when temperature fell below 11°C or during periods of en-
vironmental stress associated with population declines (Figs.

Table 1. Slopes (b), y intercepts (a), coefficients of determination (r), probability levels (P), and sample size (n) of linear regression equations
relating daily growth rate to shell length (SL) [Growth Rate in mm SL/day = a + b (SL in mm)] over periods between adjacent collections
at indicated average ambient water temperatures (°C).

Date

°C

22 Oct.,

1980

0.0003

-0.00001

-0.095

>0.20

22.0

7 Nov.,

1980

0.0048

-0.0001

-0.015

>0.20

20.3

22 Nov.,

1980

0.0310

-0.0010

-0.537

<0.001*

15.1

14 Dec.,

1980

0.0322

-0.0011

-0.935

<0.001*

12.3

10 Jan.,

1981

0.0160

-0.0004

-0.682

<0.001*

11.0

1 Feb.,

1981

0.0046

-0.0001

-0.461

<0.001*

9.0

25 Feb.,

1981

0.0024

-0.00001

-0.313

<0.01*

12.0

15 March,

1981

0.0221

-0.0007

-0.526

<0.001*

16.0

5 Apr.,

1981

0.0236

-0.0008

-0.526

<0.001*

19.0

26 Apr.,

1981

0.0690

-0.0021

-0.816

<0.001*

22.3

10 May,

1981

0.2212

-0.0082

-0.394

<0.001*

21.3

16 May,

1981

-0.0085

0.0004

-0.152

>0.20

19.5

26 May,

1981

0.0091

-0.0006

-0.213

<0.05*

23.1

7 June,

1981

0.0675

-0.0003

-0.492

<0.001*

26.2

21 June,

1981

0.0482

-0.0015

-0.371

<0.005*

27.6

5 July,

1981

0.1058

-0.0033

-0.467

<0.002*

28.0

19 July,

1981

0.0888

-0.0025

-0.447

<0.002*

28.0

2 Aug.,

1981

0.1984

-0.0065

-0.293

<0.05*

29.0

16 Aug.,

1981

0.0965

-0.0029

-0.535

<0.001*

28.6

30 Aug.,

1981

0.0814

-0.0024

-0.607

<0.001*

27.6

13 Sept.,

1981

0.0870

-0.0024

-0.619

<0.001*

26.5

27 Sept.,

1981

0.1335

-0.0042

-0.885

<0.001*

25.3

12 Oct.,

1981

0.1041

-0.0032

-0.735

<0.001*

25.3

10 Nov.,

1981

0.0458

-0.0012

-0.602

<0.001*

22.0

3 Dec.,

1981

0.0319

-0.0009

-0.743

<0.001*

17.0

5 Jan.,

1982

0.0128

-0.0003

-0.648

<0.001*

12.5

21 Jan.,

1982

0.0013

0.00002

0.024

>0.50

9.2

12 Feb.,

1982

-0.0005

0.00007

0.169

>0.10

8.2

3 Mar.,

1982

0.0020

-0.00006

-0.209

<0.10*

8.8

5 Apr.,

1982

0.0082

-0.0003

-0.447

<0.001*

16.9

19 Apr.,

1982

0.0295

-0.0008

-0.647

<0.001*

19.0

3 May,

1982

0.0154

-0.0003

-0.306

<0.02*

20.5

18 May,

1982

0.0702

-0.0018

-0.708

<0.001*

21.8

31 May,

1982

0.0952

-0.0025

-0.751

<0.001*

21.8

10 June,

1982

0.0740

-0.0020

-0.646

<0.001*

23.4

23 July,

1982

0.1665

-0.0053

-0.700

<0.001*

23.9

5 Aug.,

1982

0.0677

-0.0021

-0.337

>0.10

23.0

27 Aug.,

1982

0.0047

0.0003

0.006

>0.50

24.0

3 Sept.,

1982

0.0788

-0.0029

-0.385

<0.002*

22.0

17 Sept.,

1982

0.1007

-0.0036

-0.543

<0.001*

19.9

30 Sept.,

1982

0.0802

-0.0032

-0.717

<0.001*

19.4

15 Oct.,

1982

0.1226

-0.0046

-0.781

<0.001*

16.9

29 Oct.,

1982

0.0757

-0.0028

-0.655

<0.001*

15.2

15 Nov.,

1982

0.0271

-0.0006

-0.090

>0.20

10.9

30 Nov.,

1982

0.0108

-0.0004

-0.223

>0.10

7.1

20 Dec.,

1982

0.0028

-0.0001

-0.036

>0.50

6.4

'Indicates a significant linear relationship between shell length in mm and shell growth rate in mm SL/day at P < 0.1.

158

CORBICULA SYMPOSIUM

3A, B and E) as occurred on 16 May 1981, and 27 August
1982 (Figs. 2 and 5).

Sequential regressions of shell growth rate versus SL
of caged animals (Table 1) were utilized to estimate the
growth rates of standard size individuals with shell lengths
of 5, 10, 20 and 30 mm over the entire collection period (Fig.
5). Such determinations indicated that growth was maximal
in all size classes from late April through late October. Growth
rate subsequently declined in November and essentially
ceased in mid-winter (Fig. 5). This annual growth cycle of
caged individuals was also reflected in the growth of individual
generations in the natural population (Fig. 2). The shell growth
rate of smaller caged individuals was always greater than that

of larger clams (Fig. 5). Similarly, in the natural population
the mean SL of younger generations composed of smaller
individuals increased at a greater rate through time than that
of older generations consisting of larger individuals (Fig. 2).
Maximum estimated shell growth rates of standard size in-
dividuals were: 5 mm SL = 0.181 mm day1 or 5.40 mm 30
days'1; 10 mm SL = 0.139 mm SL day1 or 4.17 mm SL 30
days"1; 20 mm SL = 0.069 mm SL day1 or 2.07 mm SL 30
days'1; and 30 mm SL = 0.025 mm SL day1 or 0.75 mm
SL 30 days'1 (Fig. 5). No substantial growth was recorded
in individuals over 40 mm SL throughout the study period.
Moreover, during periods of both environmental stress
(marked by major declines in population density) and over-

6.0

5.0

4.0

3.0

2.0

1.0

-1.0

c n
x
m

— I
X

Fig. 5. Shell growth rates of standard sized individuals of Corbicula fluminea as estimated from least squares linear regressions of growth
rates versus shell length (SL) computed from individuals maintained in an enclosure in the Clear Fork of the Trinity River, Texas. The horizon-
tal axis is months of the year. The left and right vertical axes are shell growth rate in mm additional SL / day and mm SL / 30 days, respective-
ly. The solid bars above the horizontal axis indicate the duration of reproductive and spawning periods. The solid vertical arrow indicates
the date of a catastrophic reduction in population density associated with a hail storm and tornado. Note that this reduction in density was
associated with a distinct, short-term cessation of growth.

MCMAHON AND WILLIAMS: CORBICULA GROWTH RATE

159

wintering the SL of larger caged individuals (SL > 20 mm)
decreased, yielding the negative growth rate estimates for
larger size classes from the growth rate regressions computed
for these periods (Table 1, Fig. 5).

The annual cycle of shell growth in caged individuals
of C. fluminea was closely related to field ambient water
temperature, increasing with increasing temperature (Figs.
1 D and 5). Linear least squares regressions of the common
logarithms of daily shell growth rate values (predicted from
the growth rate regressions in Table 1) versus ambient water
temperature for standard size individuals were found to be
significant (P < 0.1) in specimens < 30 mm SL. These
regressions for standard individuals with an SL of 5 mm, 10
mm, 20 mm, 30 mm and 40 mm are as follows:

5 mm SL,

Log10 mm :

day1 ;

■3.

152

077

(°C),

= 44,

r =

624, P

.001;

SL,

Log10

SL day1

-3.

032

.069

(°C),

= 44,

r =

658, P

.001;

SL,

Log10

SL day*1

—

-3.

,249

.068

(°C),

= 44,

r =

697, P

.001;

SL,

Log10

SL day1

-3.

,345

.041

(°C),

= 33,

r =

333, P

.1; and

SL,

Log io

SL day1

-3.

,800

.050

(°C),

= 10,

r =

310, P

.2.

These regression equations were then utilized to predict the
growth rates of standard individuals of 5, 10 and 20 mm SL
over the normal ambient water temperature range occurring
at the Clear Fork collecting site (4.8°C - 29°C) (Fig. 6). For
all three standard individuals growth rate increased exponen-
tially above 15°C, and was greatly inhibited below that
temperature (Fig. 6). When these growth rate regressions are
plotted on a log-io scale against ambient water temperature
it became apparent that relatively high levels of growth were
only sustained by individuals with an SL < 20 mm, while
those of clams > 30 mm SL were greatly depressed at all
temperatures. Indeed, the growth rates of individuals > 40
mm SL were detectable only above 1 5°C (Fig. 7).

Since the growth of caged specimens was the same
as that of marked freeliving individuals (Fig. 4) it was also
assumed to be equivalent to that clams in the natural popula-
tion. This assumption was tested by comparing the increase
in the mean SL of individual generations through time with
that predicted from sequential regressions of the shell growth
rate of caged individuals versus SL (Table 1). Generation
growth rates were predicted from these regression equations
by utilizing the mean SL of generations with relatively high
field densities in January 1981 (E-81, L-80, and E-80) and
1982 (E-82, L-81 , E-81 and E-80 + L-80) as an initial SL value
and iteratively estimating increases in SL between successive
collection dates over an annua! growth cycle from the ap-
propriate sequential growth rate regression equations. This
predicted annual growth pattern was then compared visual-
ly to that estimated from the mean SL of generation size classes
in sequential population samples (Figs. 8A and B). In nearly
all cases the annual pattern of increase in SL predicted by

Fig. 6. Effect of ambient water temperature on the shell growth rate
of standard individuals of Corbicula fluminea in the Clear Fork of the
Trinity River, Texas. The horizontal axis is ambient water temperature
in °C. The vertical axis is shell growth rate (mm additional shell length
/ day) estimated for individuals with standard lengths of 5 mm (open
circles), 10 mm (solid triangles) and 20 mm (open squares) from
regressions of growth rate versus shell length of caged specimens
(SL) (Table 1). The solid lines represent best fits of least squares
linear regression equations of the common logarithm of estimated
shell growth rates versus ambient water temperature for standard
individuals as follows: 5 mm SL, log10 mm SL day*1 = -3.152 + 0.077
(°C); 10 mm SL, log10 mm day*1 = -3.032 + 0.069 (°C); and 20 mm
SL, log10 mm day*1 = -3.249 + .068 (°C).

the growth rate regression equations almost exactly coincided
with the actual annual increase in SL of specific generations
(Figs. 8A and B). The exceptions were the L-81 and combined
E-80 + L-80 generations in 1982 (Fig. 8B). These two genera-
tions had relatively low densities (Fig. 3C); therefore, dif-
ferences between predicted and observed shell growth pat-
terns may have resulted from random field sampling errors
in the determination of both initial and subsequent mean
lengths of these generations.

In order to incorporate both the effects of ambient
water temperature and size into a more general model of shell
growth in C. fluminea the per day growth rates of standard
individuals of 5 mm, 10 mm, 20 mm, 30 mm and 40 mm SL
were computed from the sequential growth rate versus SL
linear regression equations for each collection date (Table
1). These estimated growth rate values were then transformed

160

CORBICULA SYMPOSIUM

TEMPERATURE in °C

15 0

1.5

0.30

0 15

€

X
>
— I
m

if)

Fig. 7. Least squares fits of shell growth rate versus ambient water
temperature of individuals of Corbicula fluminea held in field
enclosures in the Clear Fork of the Trinity River, Texas. The horizontal
axis is field ambient water temperature (°C). The left and right ver-
tical axes are logarithmic scales of shell growth rate as mm of addi-
tional shell length per day (mm SL / day) and mm SL / 30 days,
respectively. The solid lines represent best fits of least squares linear
regressions of the common logarithm of shell growth rate in mm SL
/ day versus ambient water temperature [logio mm SL / day = a +
b(°C)], for individuals with a standard shell lengths of 5 mm, 10 mm,
20 mm, 30 mm and 40 mm. Regression parameters “a" (intercept)
and “b” (slope) are indicated above the appropriate regression line.

into common logarithms and fitted to a least squares multi-
ple linear regression versus both SL and average ambient
water temperature between sequential SL measurements of
caged individuals. This model incorporates both the negative
linear relationship between shell growth rate and size, and
the positive exponential relationship between shell growth
rate and temperature. However, it cannot predict the decrease
in SL that occurred in larger specimens (SL < 30 mm) dur-
ing the colder winter months. As shell degrowth occurred at
very low rates over relatively short durations, the inability of
the model to predict it appears to be of little real significance.
This multiple linear model of shell growth rate for the Clear
Fork C. fluminea population is:

Log io mm SL day*1 = -2.621 - 0.034 (mm SL) + 0.065 (°C),
r = 0.691, n = 174, P < 0.001.

1982

Fig. 8. Comparisons of the increase in the mean shell length (SL)
of individual generations of the Clear Fork Corbicula fluminea popula-
tion as estimated independently by the visual analysis of size-
frequency distributions of bimonthly samples and by computation
from corresponding regressions of shell growth rate versus shell
length derived from the growth of individuals held in a field enclosure
(Table 1). A. Comparisons of the increase in the mean SL of specific
generations estimated from sample size frequency analysis and in-
dividual growth in a field enclosure during 1981. B. Comparisons
of the increase in mean SL of specific generations estimated from
sample size-frequency analysis and individual growth in a field
enclosure during 1982. For both figures the horizontal axis is months
of the year and the vertical axis, SL in mm. The open triangles con-
nected by dashed lines are the mean SL of specific generations
estimated from distinct size-frequency groupings in field collections.
The open circles connected by solid lines are the increase in shell
SL of the same generations independently predicted by starting at
the same initial SL as that of a specific generation size-frequency
grouping in the earliest January sample and estimating subsequent
SL increases from sequential least squares linear regressions of the
shell growth rate of field enclosed specimens versus SL (Table 1).

Best fits of this model at temperatures spanning the
normal ambient range (5°-30°C) over shell lengths ranging

from 5 mm to 40 mm (Fig. 9A) and for standard individuals
ranging in SL from 5 mm to 40 mm over an ambient

MCMAHON AND WILLIAMS: CORBICULA GROWTH RATE

161

TEMPERATURE in °C

Fig. 9. Least squares fits of multiple linear regression model of the
shell growth rate of Corbicula fluminea in relation to individual shell
length (SL) and ambient water temperature (°C) based on the growth
rates of individuals held in a field enclosure in the Clear Fork of the
Trinity River, Texas. The shell growth rate model utilized was: Log10
shell growth rate as additional mm SL day1 = -2.621 - 0.0342 (mm
SL) + 0.0645 (°C); r = 0.69; n = 174; and P < 0.0001. For both
figures the left and right axes are shell growth rates in additional
mm SL / day and mm SL / 30 days, respectively. A. Least square
best fits of the shell growth rate at natural ambient water temperatures
over a range of SL extending from < 5 mm to > 35 mm. The horizon-
tal axis is SL in mm. The solid lines represent best fits of the above
regression equation at ambient water temperataures of 5°C, 10°C,
15°C, 20°C, 25°C and 30°C. B, Least squares best fits of the shell
growth rate of standard sized individuals over a temperature range
of < 5°C to > 30°C. The horizontal axis is ambient water temperature
in °C. The solid lines represent best fits of the above growth rate
mode! for standard individuals with shell lengths of 5 mm, 10 mm,
20 mm, 30 mm, and 40 mm.

temperature range of 0°C to 35°C (Fig. 9B) clearly
demonstrate the stimulatory effects of increasing ambient
temperature and inhibitory effects of increasing size on the
shell growth rates of C. fluminea. Such curves allow rapid
visual estimation of shell growth rate at any particular size-
temperature combination within the normal range of SL and
ambient water temperatures encountered in North American
C. fluminea populations.

DISCUSSION

The growth rate of caged individuals of C. fluminea was

equivalent to that of marked freeliving individuals and,
therefore, to that of the natural population (Fig. 4). Since the
growth rates estimated from sequential growth rate regres-
sion equations of caged individuals correspond closely to con-
current generation growth rate estimates based on the size-
frequency analysis of sequential population samples (Fig. 8),
such analysis appears to be a reliable methodology to
o estimate the growth rates and life spans of C. fluminea popula-

| tions. Size-frequency analysis of repetitive population

^ samples has been utilized to evaluate the growth and life cy-
cle of a number of C. fluminea populations (Aldridge and
> McMahon, 1978; Eng, 1979; Heinsohn, 1958; Leveque, 1973;
McMahon and Williams, 1986; Morton, 1977; Williams, 1985;
Williams and McMahon, 1986). These studies have indicated
3 that C. fluminea populations have two reproductive and
spawning periods per year and a variable life span as short
g as 1.5 years (Aldridge and McMahon, 1978; Heinsohn, 1958),
but never extending beyond three (Leveque, 1973; Morton,
1977; McMahon and Williams, 1986; Williams, 1985; Williams
and McMahon, 1986) or four years (Eng, 1979). In the past,
such analyses have been questioned because the interpreta-
tion of shell size-class frequency distributions could reflect
the biases of individual investigators. However, this study has
demonstrated that the directly measured growth of caged in-
dividuals is essentially equivalent to that estimated for distinct
generation cohorts from seqential sample size-frequency
analysis. Therefore, the latter should now be accepted as a
reasonably accurate methodology for the analysis of popula-
tion growth and age structure in this species.

The life cycle, life span and population dynamics of
the Clear Fork C. fluminea population displayed a remarkable
year-to-year variation. Individuals representing four different
biannual reproductive and spawning periods (the E-78 +
L-78, E-79 + L-79, E-80 + L-80, and E-81 + L-81 genera-
tions) were present in the 1981 samples, indicative of a life
span of slightly greater than three years. Yet, the 1982
samples had representatives of only three biannual reproduc-
tive periods (the E-80 + L-80, E-81 + L-81 , and E-82 genera-
tions) (Fig. 2), indicative of a 1 .5 to 2.5 year life span. In ad-
dition, while there were distinct early and late reproductive
periods in 1980 and 1981 only the early reproductive and
spawning period was successful in 1982.

Generation growth rates in the Clear Fork population
were also highly variable from year to year. The E-81 genera-
tion reached mean SL of 1 7.9 mm by December 1 981 , while
the E-82 generation achieved a mean SL of only 10.3 mm
by that time (Fig. 2). Differences in ambient water temperature
may account for this growth rate variation. Summer (June -
end of August) ambient water temperatures averaged 4°C
lower in 1 982 (24.0°C) than in 1 981 (27.9°C). The growth rate
model described in the Results predicts that the growth of
individuals with an SL of 5-15 mm would be 78% greater in
the warmer temperatures of 1 981 than in 1 982 (Figs. 9A and
B). In December the mean SL of the E-81 generation was
74% larger than that of the E-82 after the first summer and
fall of growth (Fig. 2), suggesting that the observed difference
in growth rate was wholely attributable to interannual
temperature differences. Indeed, the reduced growth of the

162

CORBICULA SYMPOSIUM

E-82 generation prevented it from reaching sexual maturity
(SL > 10 mm, Aldridge and McMahon, 1978) in time to par-
ticipate in the late reproductive and spawning period of 1982.
As older generations were extinct by this time (Fig. 3), no ef-
fective late reproductive and spawning period occurred.
Therefore, differences in ambient water temperature appear
to have accounted for most of the observed interannual varia-
tion in life cycle and reproduction of the Clear Fork C. fluminea
population. Temperature variation may also partially account
for the geographic, ecophentypic variations reported in the
growth, reproduction and population dynamics of this species
in North America and Asia (see McMahon, 1983 and papers
published in this symposium for a review of growth and life
cycle in C. fluminea).

Certainly, the high levels of environmentally induced,
year-to-year variation in the life-history parameters of the
Clear Fork C. fluminea population, may reflect the even
greater levels of variation reported for geographically
separated populations in North America. Such interpopula-
tion and intrapopulation variation could be partially respon-
sible for the apparent confusion regarding this species’
growth, reproduction and life cycle. However, this study along
with those of Aldridge and McMahon (1978), Britton ef a/.
(1979), O’Kane (1976) and Williams and McMahon (1986) all
indicate that C. fluminea populations in Texan freshwaters
have maximum life spans of two to three years and a bian-
nual reproductive pattern. Data for Texan populations closely
corresponds to that of Heinsohn (1958) and Eng (1979) for
Californian populations and those of Morton (1977) and
Leveque (1973) for native Asian and African populations,
respectively.

The maximum growth rates of caged individuals in the
Clear Fork population (5 mm SL = 5.4 mm SL 30 days"1 to
30 mm SL = 0.75 mm SL 30 days"1) fall well within those
reported for other enclosure experiments with C. fluminea.
Growth rate estimates for caged specimens with a SL < 10
mm have ranged from 2.0 to 2.5 mm 30 days"1 (Mattice, 1979;
O’ Kane, 1976) to 6.5 mm 30 days"1 (Dreier, 1977; Dreier and
Tranquilli, 1981). For larger specimens (SL > 10 mm), re-
duced growth rates have been reported (Britton, ef a/., 1979;
Buttner and Heidinger, 1980) which were similar to those
recorded in this study.

The growth rate of caged individuals in the Clear Fork
had a highly significant negative linear relationship with SL
(Table 1). Negative correlations between growth rate and size
have been reported a number of times for C. fluminea (Brit-
ton, ef a/., 1979; Dreier and Tranquilli, 1981; Joy, 1985,
O'Kane, 1976; Mattice, 1979; Mattice and Wright, 1985; Pool
and Tilly, 1977). The majority of these studies have suggested
that the relationship between growth rate and size is linear
while Britton ef a/., (1979) indicated that an exponential model
may be more appropriate. While exponential models are
generally appropriate to descirbe the relationship between
growth and size in this species they cannot predict the
negative growth rates (measured decrease in SL) observed
to occur in larger caged specimens during the winter or
periods of environmental stress (Fig. 5). In order to account
for such negative growth 48 separate linear models of growth

rate versus SL have been presented, each associated with
a specific set of environmental temperature conditions (Table
1).

Of the environmental factors that affected the growth
rate of the Clear Fork C. fluminea population temperature was,
by far, the most important. Our data and that of others (Brit-
ton ef a/., 1979; Buttner and Heidinger, 1980; Dreier and Tran-
quilli, 1981; Mattice, 1979; Mattice and Wright, 1985; O’Kane,
1976; Pool and Tilly, 1977), indicate that increasing
temperature stimulates growth in this species. Therefore, no
universal model of growth in C. fluminea can be valid unless
it incorporates both size and temperature effects, as does
the model presented herein (Figs. 9A and B). As our growth
rate model incorporates both size and temperature effects
it may allow biologists and engineers concerned with con-
trol of this species to predict the time required for impinging
juveniles to reach sizes that occlude heat exchangers in ser-
vice and auxiliary water systems.

The high capacity for growth of C. fluminea may be
associated with its unusually high filtration rates compared
to other freshwater species. Filtration rates for C. fluminea
are estimated to range from 250 ml clam"1 hr-1 to > 1000
ml clam"1 hr1 (Buttner and Heidinger, 1982; Foe and Knight,
1986; Mattice, 1979). Such high ingestion rates are
associated with elevated assimilation efficiencies (Foe and
Knight, 1986; Lauritsen, 1986) and net production efficien-
cies (> 70%, Lauritsen, 1986, Aldridge and McMahon, 1978)
in this species, supporting rapid tissue growth.

Unlike reports from Asia, Africa and the Western
United States which all suggest that C. fluminea populations
have roughly similar growth rates and life spans, United
States populations east of the Mississippi River have been
reported to have lower growth rates and longer life spans of
5 to 8 years (Gardner ef a/., 1976; Keupefa/., 1963; Mattice,
1979; Mattice and Wright, 1985; Sickel, 1973; Sinclair and
Isom, 1963). While some of these estimates may have
resulted from assuming one generation per year for this
species, others are based on the growth of caged individuals
(Mattice, 1979; Mattice and Wright, 1985). The growth rate
model developed by Mattice and Wright (1985) predicts a life
span of at least 6 years to reach an SL equivalent to the
largest individuals in the population. However, this study and
others have demonstrated high levels of interannual, in-
trapopulation variation in the growth rates of C. fluminea
populations (Mattice and Wright, 1985; Williams and
McMahon, 1986). Large interannual variations in growth are
associated with phytoplankton availability (Williams and
McMahon, 1986) or with differences in temperature regime
(this study). McMahon (1983) suggested that maximum size
is directly correlated with growth rate in this species.
Therefore, the presence of very large specimens in a popula-
tion may not be indicative of individuals with long life spans
but rather, of generations previously experiencing excep-
tionally good conditions for growth. This is the case in the
Clear Fork C. fluminea population where individuals of the
E-78 + L-78 generation attained lengths > 45 mm at the end
of a three year life span, while members of the E-79 + L-79
generation did not have shell lengths much in excess of 35

MCMAHON AND WILLIAMS: CQRBICULA GROWTH RATE

163

mm in their third and terminal year of life (Fig. 2). Indeed,
when growth is computed from our model over the excep-
tionally low annual temperature cycle that occurred in 1982
(Fig. 1 D), an early generation would require 8 years to reach
an SL of 30 mm, while only four years would be required to
reach that size over the warmer temperature cycle of 1981
(Fig. ID). At the even warmer temperatures recorded for a
C. fluminea population in Lake Arlington, Texas, from March,
1981 to March, 1982 (Williams and McMahon, 1986), our
growth model predicts only three years to reach an SL of 30
mm. The summer of 1980, prior to our initial collections of
the Clear Fork population, had been among the warmest on
record in Texas, with maximum daily ambient air and water
temperatures averaging 32°C from June through August 1980
(National Oceanic and Atmospheric Administration, 1980).
Water temperatures in this range exponentially increase the
growth rate of C. fluminea and stimulate new growth in larger
specimens (SL > 30 mm) (Figs. 6 and 9). Therefore, the very
warm water temperatures of 1980 may have allowed two and
three year old individuals to grow rapidly to the very large
sizes observed in our initial collections (Fig. 2). Indeed, con-
tinual monitoring through the spring of 1986 has indicated
that although a three year life span has been maintained in
the Clear Fork population, such large size classes have not
reoccurred since 1981 (McMahon, unpublished observations).
As such, at least in Texas, exceptionally large individuals in
C. fluminea populations may result from environmental con-
ditions that support elevated growth rates in specific genera-
tions, but do not result from extended life spans.

There is biochemical evidence that C. fluminea popu-
lations east of the Mississippi River have gene pools distinct
from those west of the Mississippi River (McLeod, 1986).
Therefore, these populations may represent genetically
distinct “physiological races” characterized by longer life
spans than reported for this species throughout the rest of
its world-wide range (McMahon, 1983). However, such
populations should be subjected to long-term concurrent
studies of both population age-size structure variations and
individual growth rates in field enclosures before the general
acceptence of extended life spans (> 6 years) for this species
in the southeastern United States. Recently, specimens of
C. fluminea were reported to grow from a mean SL of 13 mm
to 26 mm when held in an enclosure in the Kanawha River,
West Virginia, for 38 weeks (Joy, 1985). This growth rate is
very similar to those recorded for the E-81 and L-81 genera-
tions in the Clear Fork population. Such new data strongly
suggest that the growth and life span of C. fluminea popula-
tions in the eastern United States are well within the ranges
recorded for this species in other geographical areas of its
range.

In the Clear Fork population the majority of growth oc-
curred above 15°C. Similarly, a 14°C limit for growth was
reported for a C. fluminea population in the Delta Mendota
Canal, California (Eng, 1979). This is also the approximate
temperature at which reproductive activity is initiated in both
the Clear Fork and other populations (13°-19°C) (Aldridge and
McMahon, 1978; Dreier and Tranquilli, 1981; Eng, 1979; Mor-
ton, 1977; Williams and McMahon, 1986). In addition, the

filtration rate of C. fluminea is reduced by > 50% below 20°C
(Mattice, 1979). Therefore, fundamental physiological and
metabolic changes must occur which allow this species to
switch from a slow growing, non-reproductive, rather inac-
tive state, to a fast growing, reproductive, highly active state
as temperatures rise above 15°-18°C.

Temperatures above 24°-25°C are reported to inhibit
growth in laboratory acclimated specimens of C. fluminea (Foe
and Knight, 1986; Mattice 1979; Mattice and Wright, 1985)
and to reduce filtration rate (Mattice, 1979) and ventilation
and oxygen consumption rates (McMahon, 1979a). It was
somewhat surprising then that no suppression of growth rate
was observed in caged individuals in the Clear Fork popu-
lation at temperatures up to 30°C (Figs. 6 and 7). Indeed,
other studies indicate that population growth rates are main-
tained in C. fluminea at field water temperatures as high as
33°C (Aldridge and McMahon 1978; Williams and McMahon,
1986). A recent study of a steam-electric power plant ther-
mal discharge C. fluminea population reported nc inhibition
of growth at temperatures approaching 36°C, the apparent
long-term upper lethal limit of this species (McMahon and
Williams, 1986). Further, oxygen consumption rates were not
suppressed at temperatures as high as 33°C in specimens
of C. fluminea experiencing those temperatures in the field
(Williams, 1985). The conflicting data from laboratory ac-
climated and field-conditioned individuals suggest that this
species is capable of long-term (seasonal) physiological
temperature compensation under ambient field conditions,
not revealed in shorter-term laboratory temperature acclima-
tion experiments.

There were distinct declines in the growth rate of caged
individuals in May 1981 and June 1982, associated with
periods of catastrophic reductions in population densites (Fig.
3). During these and overwintering periods larger individuals
exhibited shell “degrowth”, characterized by a directly
measured slow decrease in SL. Bivalves are reported to buffer
hydrogen ion produced during anaerobic respiration with
carbonate released from dissolution of the shell. Therefore,
long-term anaerobiosis leads to a reduction in shell mineral
content (Akberali et al., 1983; Goddard and Martin, 1966).
During overwintering periods and periods of environmental
stress C. fluminea may close its valves and become partially
or completely anaerobic, leading to shell dissolution, erosion
of the shell edge and degrowth. Exposure of C. fluminea to
environmental stress causes changes in the internal shell
microstructure (Prezant and Chalermwat, 1983). Dissolution
of the shell in specimens exhibiting degrowth was evi-
denced by the presence of greater amounts of uncalcified
shell maxtrix and periostracum at the shell edge.

This study also provided an opportunity to observe the
recovery of a C. fluminea population after a catastrophic den-
sity decline. Major declines of population density in this
species have been associated with reproduction (Ingram,
1959), low winter water temperatures (< 2°C)(Bickel, 1966;
Cherry ef al., 1980; Dreier and Tranquilli, 1981 ; Horning and
Keup, 1964; Mattice and Dye, 1976; Rodgers et al., 1979),
and exposure to air by receding water levels (McMahon,
1979b; White, 1979; White and White, 1977). C. fluminea can

164

CORBICULA SYMPOSIUM

rapidly reestablish populations after severe density reduc-
tions. A population in the New River, Virginia, recovered to
1000 clams nr2 within five months of nearly complete exter-
mination by low winter water temperatures (Rodgers et at.,
1979). Similarly, irrigation canal populations of C. fluminea
have been reported to recover to extremely high densities
within one year of canal dewatering and nearly complete
removal of resident adult populations (Eng, 1979; Pro-
kopovich, 1969; Prokopivch and Hebert, 1965). In the Clear
Fork population density declined from 2655 clams nr2 on 26
April 1981, to 305 clams nr2 on 21 June 1981 (Fig. 3A).
Reproduction by the relatively few surviving adults allowed
recovery of 1980 density levels through recruitment of the
E-81 and L-81 generations by the spring of 1982 (Figs. 3A and
D). Reproduction by these two generations and subsequent
recruitment of the resulting E-82 generation increased popula-
tion density to 12000 clams nr2 by the winter of 1982 (Figs.
3A and E). This astonishing capacity for rapid population
recovery is associated with the high fecundity of C. fluminea ,
reported to average 68,678 juveniles per adult in Texas
populations (Aldridge and McMahon, 1978). The high
reproductive capacity, high growth rate, small size at maturity
(SL < 10 mm, Aldridge and McMahon, 1978), and attenuated
life span (Aldridge and McMahon, 1978; Eng, 1979; Hein-
sohn, 1958; Leveque, 1973; Mattice and Wright, 1985; Mor-
ton, 1977; Williams and McMahon, 1986; this study) of C.
fluminea appear to be adaptations that optimize the pro-
duction and survival to maturity of offspring in highly unstable
environments where such life history traits favor rapid popula-
tion growth and expansion and, therefore, are of high selec-
tive value (Stearns, 1976, 1977; for a review of life history
traits in C. fluminea see McMahon, 1983). It is the exceptional
ability of C. fluminea to successfully invade and survive in
highly disturbed habitats that not only accounts for its rapid
spread in North American fresh waters (McMahon, 1982), but,
also for its ability to invade, survive, grow and reproduce in
industrial raw water cooling and service water systems.

ACKNOWLEDGEMENTS

The authors wish to express their gratitude to Juan Ibarra,
Ralph Williams, Joe Gilly, Colette O’Bryne-McMahon, John Stamen,
and Colleen Bronstad for their assistance with the field collections;
to John Stamen, Bruce Whitehead and Colleen Bronstad for
assistance with sample measurements and data analysis and to
Majbritt Angarano and Roger Byrne for their assistance in the
preparation of the manuscript. The authors also wish to thank Dr.
Peter Calow, Dr. Joseph C. Britton, and Dr. David S. White for their
critical reviews of the manuscript. This research was supported by
a grant from Organized Research Funds of the University of Texas
to R. F. McMahon.

LITERATURE CITED

Akberali, H. B., K. Brear and J. D. Currey. 1983. Mechanical and
morphological properties of the shell of Scrobicularia plana
(Da Costa) under normal and stress conditions. Journal of
Molluscan Studies 49: 93-97.

Aldridge, D. W. and R. F. McMahon. 1978. Growth, fecundity, and
bioenergetics in a natural population of the Asiatic freshwater

clam, Corbicula manilensis Philippi, from North Central Texas.
Journal of Molluscan Studies 44: 49-70.

Avolizi, R. J. 1976. Biomass turnover in populations of viviparous
sphaeriid clams: Comparisons of growth, fecundity, mortali-
ty and biomass production. Hydrobiologia 51: 163-180.

Boozer, A. C. and P. E. Mirkes. 1979. Obervations on the fingernail
clam, Musculium partumeium (Pisidiidae), and its association
with the introduced Asiatic clam, Corbicula fluminea. Nautilus
93: 73-83.

Bickel, D. 1966. Ecology of Corbicula manilensis Philippi in the Ohio
River at Louisville, Kentucky. Sterkiana 23: 19-24.

Britton, J. C. and B. Morton. 1979. Corbicula in North America: The
evidence reviewed and evaluated. In: Proceedings, First In-
ternational Corbicula Symposium, J. C. Britton, ed., pp.
250-287. Texas Christian University Research Foundation,
Fort Worth, Texas.

Britton, J. C., D. R. Coldiron, L. P. Evans, Jr., C. Golightly, K. D.
O’Kane and J.R. TenEyck. 1979. Reevaluation of the growth
pattern in Corbicula fluminea (Muller). In: Proceedings, First
International Corbicula Symposium, J. C. Britton, ed., pp.
177-192. Texas Christian University Research Foundation,
Fort Worth, Texas.

Burch, J. Q. 1944. Check list of West North American marine
mollusca from San Diego, Calif., to Alaska. Minutes of the Con-
chology Club of Southern California 38: 11-18.

Buttner, J. K. and R. C. Heidinger. 1980. Seasonal variations in
growth of the Asiatic clam, Corbicula fluminea (Bivalvia: Cor-
biculidae) in a Southern Illinois fish pond. Nautilus 94: 8-10.

Buttner, J. K. and R. C. Heidinger. 1 982. Rate of filtration in the Asiatic
Clam, Corbicula fluminea. Transactions of the Illinois State
Academy of Science. 74: 13-17.

Cherry, D. S., J. H. Rodgers, Jr., R. L. Graney and J. Cairns, Jr.
1980. Dynamics and control of the Asiatic clam in the New
River, Virginia. Bulletin 123, Virginia Water Resources
Research Center, Virginia Polytechnic Institute and State
University, Blacksburg, Virginia. 72 p.

Coldiron, D. R. 1975. Some aspects of the biology of the exotic
mollusk Corbicula (Bivalvia: Corbiculidae). M. S. Thesis, Texas
Christian University. 92 p.

Diaz, R. J. 1974. Asiatic clam, Corbicula manilensis (Philippi) in the
tidal James River, Virginia. Chesapeake Science 1 5: 1 1 8-1 20.

Dreier, H. 1977. Study of Corbicula manilensis at Lake Sangchris.
In: The Annual Report for Fiscal Year 1976. Lake Sangchris
Project, pp 7.1-7.52. Illinois Natural History Survey, Urbana,
Illinois.

Dreier, H. and J. A. Tranquilli. 1981. Reproduction, growth, distribu-
tion and abundance of Corbicula in an Illinois cooling lake.
Illinois Natural History Survey Bulletin 32: 378-393.

Dudgeon, D. 1980. A comparative study of the Corbiculidae of
southern China In: Proceedings, First International Workshop
on the Malacofouna of Hong Kong and Southern China, B. Mor-
ton, ed., pp. 37-60. Hong Kong University Press, Hong Kong.

Eng, L. L. 1 979. Population dynamics of the Asiatic clam, Corbicula
fluminea (Muller), in the concrete-lined Delta-Mendota Canal
of Central California. In: Proceedings, First International Cor-
bicula Symposium, J. C. Britton, ed., pp. 39-68. Texas Chris-
tian University Research Foundation, Fort Worth, Texas.

Foe, C. and A. Knight. 1986. A thermal energy budget for juvenile
Corbicula fluminea. In: Proceedings, Second International Cor-
bicula Symposium, J. C. Britton, ed. American Malacological
Bulletin, Special Edition No. 2:143-150.

Gardner, J. A., Jr., W. R. Woodall, Jr., A. A. Staats, Jr. and J. F.
Napoli 1976. The invasion of the Asiatic clam (Corbicula

MCMAHON AND WILLIAMS: CORBIGLJLA GROWTH RATE

165

manilensis Philippi) in the Altamaha River, Georgia. Nautilus
90: 117-125.

Goddard, C. K. and A. W. Martin. 1966. Carbohydrate metabolism.
IN: Physiology of Mollusca, Vol. II, K. M. Wilbur and C. M.
Yonge, eds., pp. 273-308. Academic Press, New York.

Gunning, G. E. and R. D. Suttkus. 1966. Occurrence and distribu-
tion of Asiatic clam, Corbicula leana, in the Pearl River, Loui-
siana. Nautilus 79: 113-116.

Heard, W. H. 1958. Corbicula fluminea in Florida. Nautilus 77:
105-107.

Heinsohn, G. E. 1958. Life history and ecology of the freshwater clam,
Corbicula fluminea. M. S. Thesis. University of California, Santa
Barbara. 64 p.

Hillis, D. M. and J. C. Patton. 1982. Morphological and electrophoretic
evidence for two species of Corbicula (Bivalvia: Corbiculidae)
in North America. American Midland Naturalist 108: 74-80.

Hornbach, D. J., C. M. Way and A. J. Burky. 1980. Reproductive
strategies in a freshwater sphaeriid clam, Musculium par-
tumeium (Say), from a permanent and temporary pond.
Oecologia (Berl.) 44: 164-170.

Horning, W. B. and L. Keup. 1964. Decline of the Asiatic clam in
the Ohio River. Nautilus 78: 29-30.

Hubricht, L. 1966. Corbicula manilensis (Philippi) in the Alabama River

System. Nautilus 80: 32-33.

Ingram, W. M. 1959. Asiatic clams as potential pests in California
water supplies. Journal of the American Water Works Associa-
tion 51 : 363-370.

Ingram, W. M., L. Keup and C. Henderson. 1964. Asiatic clams at
Parker, Arizona. Nautilus 77: 121-125.

Joy, J. E. 1985. A 40-week study on growth of the Asian clam, Cor-
bicula fluminea (Muller), in the Kanawha River, West Virginia.
Nautilus 99: 110-116.

Keup, L., W. B. Horning and W. M. Ingram. 1963. Extension of range
of Asiatic clam to Cincinnati Reach of the Ohio River. Nautilus
77: 18-21.

Kraemer, L. R. 1977. Aspects of the funtional morphology of the man-
tle/shell and mantle/gill complex of Corbicula (Bivalvia:
Sphaeriacea: Corbiculidae) Bulletin of the American
Malacological Union, Inc. 1977: 25-31.

Kraemer, L. R. 1978. Corbicula fluminea (Bivalvia: Sphaeriacea): The
functional morphology of its hermaphroditism. Bulletin of the
American Malacological Union, Inc. 1978: 40-49.

Kraemer, L. R. 1979. Corbicula (Bivalvia: Sphaeriacea) vs. indigenous
mussels (Bivalvia: Unionacea) in U.S rivers: A hard case for
interspecific competition? American Zoologist 19: 1085-1096.

Kraemer, L. R. and S. Lott. 1977. Microscopic anatomy of the visceral
mass of Corbicula (Bivalvia: Sphaeriacea). Bulletin of the
American Malacological Union, Inc. 1977: 48-56.

Lauritsen, D. D. 1986. Assimilation of radiolabeled algae by Corbicula.
In: Proceedings, Second International Corbicula Symposium,
J. C. Britton, ed. American Malacological Bulletin, Special Edi-
tion No. 2: 219-222.

Leveque, C. 1973. Dynamiquedes peuplements biologie et estimation
de la production des mollusques benthiques du Lac Tchad.
Cahiers O.R.S.T.O.M. Serie Hydrobiologie 7: 117-147.

Mattice, J. S. 1979. Interactions of Corbicula sp. with power plants.
In: Proceedings, First International Corbicula Symposium, J.
C. Britton, ed., pp. 119-138. Texas Christian University
Research Foundation, Fort Worth, Texas.

Mattice, J. S. and L. L. Dye. 1976. Thermal tolerance of adult Asiatic
clam. In: Thermal Ecology II, G. W. Esch and R. W. McFarlane,
eds., pp. 130-135. GONF-750425, United States Energy
Research and Development Administration Symposium
Series. National Technical Information Service, U.S. Depart-

ment of Commerce, Springfield, Virginia.

Mattice, J. S. and L. L Wright. 1985. Aspects of growth of Corbicula
fluminea with supporting field and laboratory growth data.
Publication No. 2427, Environmental Sciences Division, Oak
Ridge National Laboratory, Oak Ridge, Tennessee.
ORNL/TM-8910, National Technical Information Service, U.S.
Department of Commerce, Springfield, Virginia. 64 pp.

McLeod, M. J. 1986. Electrophoretic variation in United States Cor-
bicula. In: Proceedings, Second International Corbicula Sym-
posium, J. C. Britton, ed. American Malacological Bulletin,
Special Edition No. 2: 125-132.

McMahon, R. F. 1975. Effects of artificially elevated water
temperatures on the growth, reproduction and life cycle of a
natural population of Physa virgata Gould. Ecology 56:
1167-1175.

McMahon, R. F. 1976. Growth, reproduction and life cycle in six Tex-
an populations of two species of fresh-water limpets. American
Midland Naturalist 95: 174-185.

McMahon, R. F. 1979a. Response to temperature and hypoxia in
the oxygen consumption of the introduced Asiatic freshwater
clam Corbicula fluminea (Muller). Comparative Biochemistry
and Physiology 63A: 383-388.

McMahon, R. F. 1979b. Tolerance of aerial exposure in the Asiatic
freshwater clam, Corbicula fluminea (Muller). In: Proceedings,
First International Corbicula Symposium, J. C. Britton, ed., pp.
227-241. Texas Christian University Research Foundation,
Fort Worth, Texas.

McMahon, R. F. 1982. The occurrence and spread of the intro-
duced Asiatic freshwater clam, Corbicula fluminea (Muller),
in North America: 1924-1982. Nautilus 96: 134-141.

McMahon, R. F. 1983. Ecology of an invasion pest bivalve, Corbicula.
In: The Mollusca, Vol. 6, Ecology, W. D. Russell-Hunter, ed.,
pp. 505-561. Academic Press, Inc., San Diego.

McMahon, R. F. and C. J. Williams. 1986. Growth, life cycle, upper
thermal limit and downstream colonization rates in a natural
population of the freshwater bivalve mollusc, Corbicula
fluminea (Muller), receiving thermal effluents. In: Proceedings,
Second International Corbicula Symposium, J. C. Britton, ed.
American Malacological Bulletin, Special Edition No. 2:
231-239.

Morton, B. 1977. The population dynamics of Corbicula fluminea
(Bivalvia: Corbiculacea) in Plover Cove Reservoir, Hong Kong.
Journal of Zoology, London 181: 21-42.

Negus, C. L. 1966. A quantitative study of growth and production
of unionid mussels in the River Thames at Reading. Journal
of Animal Ecology 35: 513-532.

National Oceanic and Atmospheric Administration. 1980. Local
Climatological Data, Dallas-Fort Worth, Texas, Monthly Sum-
maries for 1980. Environmental Data and Information Service,
Asheville, North Carolina.

O’Kane, K. D. 1976. A population study of the exotic bivalve Cor-
bicula manilensis (Philippi, 1841) in selected Texas reservoirs.
M. S. Thesis, Texas Christian University. 134 pp.

Pool, A. C. and L. J. Tilly. 1977. A model to determine growth rate
of Corbicula. CONF-771043-1 , U.S. Energy Research and
Development Administration, National Technical Information
Service, U.S. Department of Commerce, Springfield, Virginia.

11 pp.

Prezant, R. S. and K. Chalermwat. 1983. Environmentally induced

166

CORBICULA SYMPOSIUM

changes in shell microstructure of the Asiatic clam Corbicula.
American Zoologist 23: 914.

Prokopovich, N. P. 1969. Deposition of clastic sediments by clams.
Sedimentary Petrology 39: 891-901.

Prokopovich, N. P. and D. J. Hebert. 1965. Sedimentation of the
Delta-Mendota Canal. Journal of the American Water Works
Association. 57: 375-382.

Rodgers, J. H., Jr., D. S. Cherry, K. L. Dickson and J. Cairns, Jr.
1979. Invasion, population dynamics and elemental accumula-
tion of Corbicula fluminea in the New River at Glen Lyn,
Virginia. In: Proceedings, First International Corbicula Sym-
posium, J. C. Britton, ed., pp. 99-110. Texas Christian Univer-
sity Research Foundation, Fort Worth.

Sickel, J. B. 1973. A new record of Corbicula manilensis (Philippi)
in the Southern Atlantic Slope Region of Georgia. Nautilus 87:
11-12.

Sickel, J. B. 1979. Population dynamics of Corbicula in the Altamaha
River, Georgia. In: Proceedings, First International Corbicula
Symposium, J. C. Britton, ed., pp. 69-80. Texas Christian
University Research Foundation, Fort Worth, Texas.

Sinclair, R. M., and B. G. Isom. 1963. Further studies on the intro-
duced Asiatic clam ( Corbicula ) in Tennessee. Tennessee
Stream Pollution Control Board, Tennessee Department of
Public Health, Nashville, Tennessee. 76 pp.

Stearns, S. C. 1976. Life-history tactics: A review of the ideas. Quarter-
ly Review of Biology 51: 3-47.

Stearns, S. C. 1977. The evolution of life history tactics: A critique

of the theory and a review of the data. Annual Review of
Ecology and Systematics 8: 145-171.

Way, C. M., D. J. Hornbach and A. J. Burky. 1980. Comparative life
history tactics of the sphaeriid clam, Musculium partumiem
(Say) from a permanent and temporary pond. American
Midland Naturalist 104: 319-327.

White, D. S. 1979. The effect of lake-level fluctuations on Corbicula
and other pelecypods in Lake-Texoma, Texas and Oklahoma.
In: Proceedings, First International Corbicula Symposium, J.
C. Britton, ed., pp. 82-88. Texas Christian University Research
Foundation, Fort Worth, Texas.

White, D. S. and S. J. White. 1977. The effect of reservoir fluctua-
tions on populations of Corbicula manilensis (Peiecypoda: Cor-
biculidae). Proceedings of the Oklahoma Academy of Science
57: 106-109.

Williams, C. J. and R. F. McMahon. 1986. Power station entrainment
of Corbicula fluminea in relation to population dynamics,
reproductive cycle and biotic and abiotic variables. IN: Pro-
ceedings, Second International Corbicula Symposium, J. C.
Britton, ed., American Malacological Bulletin, Special Edition,
No. 2: 99-111.

ASPECTS OF GROWTH OF CORBICULA FLUMINEA

J, S. MATTIGE1 AND L. L. WRIGHT
ENVIRONMENTAL SCIENCES DIVISION
OAK RIDGE NATIONAL LABORATORY
OAK RIDGE, TENNESSEE 37831, U.S.A.

ABSTRACT

Studies of caged Asiatic clams (Corbicula fluminea) in Watts Bar Reservoir (Clinch River, Ten-
nessee) indicate that interaction of intrinsic and extrinsic factors affects the rate of increase in shell
length (SL). Measurements of marked and unmarked clams were made at two sites differing by about
5°C in both 1976 and 1977, and at lower temperature sites in 1978 and early 1979. The growth season
in Watts Bar began in April and continued through October. From November through March growth
virtually ceased. Increase in SL was an inverse linear function of initial SL; i.e., small clams grew
faster than large ones. Rate of shell length increase differed between the two sites. Growth rates also
varied during the growth season. A comparison of these rates for each site during the year indicated
that temperature plays a major role in growth rate determination. Type of algae and food (seston con-
centration) were not correlated with growth rate. A decline in growth rate in spring, at about the time
of peak reproduction, indicates that intrinsic factors are also important. Nevertheless both laboratory
and field studies suggest that the temperature for optimum growth occurs in the mid-twenty °C range.
Because of this relationship between growth and temperature, estimates of both growth and life span
may vary depending on the natural annual temperature regime. For example, life span of Corbicula
in the Clinch River System is more than four years and could be as long as six (or more) years, which
is substantially longer than for populations in some locations. Such differences could influence deci-
sions regarding initiation of control procedures at industrial facilities.

Corbicula fluminea* has become the primary focus of
macrofouling control for industrial facilities sited on freshwater
bodies in the United States (Mattice, 1983). With the recent
report of Corbicula fluminea in Lake Erie (Clarke, 1981), this
species has now achieved expansion into all of the major
drainage basins in the United States. This has taken place
in roughly 40 years, an invasion that appears unprecedented
for aquatic invertebrates. Although no comprehensive list of
industries impacted by this expansion has compiled, a large
variety of facilities have been specifically mentioned. It is pro-
bably safe to say that a majority of the industries in Corbicula’ s

*We have chosen to use the species name fluminea here because
of evidence (Smith et ai, 1979) that clams from the population we
studied are not different from those in Texas that Morton (1979) has
described as Corbicula fluminea. However, some controversy does
exist regarding the proper name (Sinclair and Isom, 1963) suggesting
the need for further comprehensive study.

’Present Address: Ecological Studies Program, Electric Power
Research Institute, 3412 Hillview Avenue, Palo Alto, CA 94303, U.S.A.

geographic range that require supplies of raw water have had
problems with Corbicula fouling. The expense of fouling in-
cidents has varied greatly, but has at least reached the
multimillion dollar level in some cases (pers. comm., Robert
West, Arkansas Power Co., Little Rock, Ark.). It is clear that
some method of controlling fouling of Corbicula is sorely need-
ed to prevent increasing societal costs of generation and
manufacturing.

At present, cost effective control of Corbicula fouling
requires information about when the clams reproduce and
how fast they grow. Despite the burgeoning interest in con-
trol of Corbicula fouling, the only procedure that has proven
universally effective has been physical removal (Mattice,
1983). Thus, it is important to know when to schedule clam
removal to prevent the clams from reaching a size large
enough to occlude water lines or condenser tubes. Reproduc-
tion and rate of growth play important roles in determining
these schedules.

This study examined growth of Corbicula in a
river/reservoir system known to present Corbicula fouling to
industrial facilities (Sinclair and Isom, 1963; Goss and Cain,

American Malacological Bulletin, Special Edition No. 2(1 986): 167-1 78

167

168

CORBICULA SYMPOSIUM

1976; Goss et a!., 1979). Comparisons between and within
years at sites differing in water quality were made to deter-
mine intrinsic and extrinsic factors that affect growth of the
clams. Only data summaries are reported here; the complete
data set is included in Mattice and Wright, (1985).

SITE DESCRIPTION

Growth of Corbicula fluminea, was investigated from
1975 to 1979 in the intake and discharge areas of the
Kingston Steam Plant, a coal-fired electricity generating sta-
tion near Kingston, Tennessee (84°31’W, 35°54’N). Intake
source water varied seasonally depending on relative flows
of the Emory, Clinch, and Tennessee Rivers. Flows of the
latter two rivers depended on operation of upstream dams.
However, the three sites where clams were held (Fig. 1) were
dominated by through-plant water flow (3671 m3 per minute
when all nine units are operating) so that water qualities at
all sites could be assumed to be similar except for changes
induced by plant operation. These changes included a
discharge-to-intake water temperature differential of 7.5 to
8. CPC and intermittent application of chlorine ranging from
499 to 2722 kg per day depending on time of year. Chlorina-

Fig.l. Schematic diagram of the study site indicating the locations
of cages (■ = study of winter growth of small clams, ◄ = all other
growth studies), Tennessee Valley Authority temperature probes (•),
and phytoplankton (P) and seston (S) sample sites.

tion is conducted 5 days/week with each unit receiving ap-
plication for 30 minutes to control condenser fouling. Chlorine
is measured daily at the discharge of each unit during
chlorination to ensure that limits of 0.5 mg/I total residual and
0.2 mg/I free residual chlorine are not exceeded; no chlorine
can be detected (amperometric titration) more than 3-15
meters beyond the combined discharge of all units (personal
communication, Alex Riddings, Public Safety Office, Kingston
Steam Plant, Kingston, Tenn.).

MATERIALS AND METHODS

Clams used in these studies were collected from areas
near the Kingston Steam Plant using a venturi suction
sampler (Mattice and Bosworth, 1 979). After transport of the
samples to the laboratory, clams were separated from the

substrate and acclimated in a flowing water system to within
+ /- 2°C of the intake or discharge temperatures at < 1 °C per
day. Clams greater than 9.0 mm (maximum shell length) were
marked with a number using a small dental grinder; clams
smaller than 9.0 mm were not marked. Clams in the laboratory
were fed a mixture of ground trout chow (Ralston Purina Co.,
St. Louis, Mo.) and Staple Flake food for Tropical Fish (Hartz
Mountain Corporation, Harrison, N.J.) twice a day except on
weekends. Clams maintained on this regimen did grow. Once
acclimated, the shell length (SL) of each clam was measured
with a dial caliper to the nearest 0.1 mm, and clams were
transferred to field cages (described below) weighted to rest
on the substrate at water depths of about 3 m. At various in-
tervals thereafter, the cages were brought to the surface and
the shell lengths of the clams remeasured. Individual clams
were not out of water for more than one-half minute at a time
during any of these procedures.

Three types of cages were used in these studies.
Clams greater than 15 mm SL were held in 0.5 x 0.5 x 0.33
m cages of ~4 mm mesh Vexar plastic netting attached to
a redwood frame. Clams less than 1 5 mm SL were generally
held in 2 liter cages made of covered plastic tubs. Each of
these covered tubs had four holes (2.5 x 2.5 cm) in the sides
and one larger hole in the top covered with 1 mm mesh nitex
screening to permit flow of water and food into the cage. The
screened tubs were held inside the larger cages described
above. The third type of cage was used only for studies of
winter growth of small (3. 0-6. 5 mm SL) clams. These cages
consisted of an envelope (~8 x 15 cm) of 1 mm mesh
stainless steel screen held together with staples. Cages were
attached to a weighted line.

Several water quality parameters were also measured
at various times during these studies. Water temperatures
were measured daily either as continuous recordings at the
cage sites (Ryan recording thermometer, Ryan instruments,
Inc., Kirkland, WA 98033) or at hourly intervals (Tennessee
Valley Authority temperature recorders). See Figure 1 for
positions relative to cage positions. Mean daily temperatures
were estimated by eye from the 30-day strip charts after the
recordings were retrieved from the Ryan recording ther-
mometer. Hourly temperatures were averaged from the TVA
data. In some cases, interconversions between temperature
readings were made to supply missing data (Mattice and
Wright, 1985). Phytoplankton complement and relative com-
position were estimated by examining water samples taken
with a 2 liter Kemmerer bottle and filtered through a plankton
net. Samples were collected monthly from October 1975
through November 1976 except for the months of April, Ju-
ly, and September. From October 1975 through May 1976,
samples were filtered through a relatively coarse (80 ^m)
mesh Wisconsin-type plankton net. From June 1976 through
November 1976, samples were filtered through a 10 mesh
synthetic net. At each station on each date, two samples were
filtered, one near the surface and one near the bottom of the
water column. Algal identity and relative composition were
determined microscopically. On numerous occasions seston
concentration was measured in one or both coves by filter-
ing 1 liter of water through a tared Whatman GF/C glass fiber

MATTICE AND WRIGHT: GROWTH OF CORBICULA FLU MINE A

169

Fig. 2. Intake and discharge water temperatures (°C) during cage growth studies in 1 976-1 979. Dotted lines indicate periods when temperature
in the discharge was estimated based on data from TVA’s closest temperature probe with a correction factor applied.

filter (-0.45 equivalent pore size). Following drying for
12-24 h at 105°C the filter was reweighed using a balance
accurate to 10_4g to obtain seston concentration (mg/1).

Coincident with these field studies, a short-term
laboratory study was conducted to examine growth of ten Cor -
bicula at each of several temperatures in the laboratory.
Clams were collected and acclimated as above to
temperatures of 15, 20, 25, and 30°C and marked individually.
The clams (four size groups) were taken placed on stainless
steel screening in 195-1 flow-through aquaria. The influent
water of each aquarium contained a mixture of ground and
slurried flakefood and trout chow from a head tank. The slurry
of food was added to the head tank twice per day and the
flow from the head tank was controlled by a metering pump
to produce a calculated maximum seston concentration of
6 mg/I (measured as above) in each aquarium. Between food
additions to the head tank, the concentration available to the
clams varied from 2-7 mg/I. Seston was kept in suspension
in the head tank and aquaria by air bubblers. At weekly in-
tervals, all clams were remeasured and growth estimated as
the difference between final shell length (FSL) and initial shell
length (ISL).

Several samples of clams in the 15 to 40 mm SL size
range that had been fixed in 12% neutral formalin after col-
lection for other purposes were examined for presence of in-
cubating young. The adult clams were opened and the inner
demibranch removed to a petri dish. The demibranch was
teased apart with needles and examined at 30x using a
dissecting microscope. Numbers of incubating young per
clam were counted according to the following categories: 0,
1-19, 20-99 and >100.

Various factors were examined as determinants of
growth. Effect of initial clam size on growth was determined
by linear regression. Length data were not log transformed
because variance did not appear related to size. In addition,
growth rates for most intervals and years were fit better by
the linear model than by an exponential model (see Mattice
and Wright, 1985). Based on these results, growth rate com-
parisons between sites and between years were made by F-
test comparisons of linear regressions of shell length increase
on initial shell length (ISL). Examination of effect of

temperature and seston on growth of clams in the field was
based on comparison of growth by a standard-size clam dur-
ing the various intervals between measurements. Before con-
ducting these analyses we examined the growth rate data
for serial correlation by regressing growth rate in each inter-
val (except the first) on growth rate in the previous interval
for the summer of 1977. Regressions for clams held in the
discharge (r2 = 0.078; p = 0.47) and intake r2 = 0.073; p
= 0.46) were not significant so we assumed that growth rate
in each interval was independent of that in the other inter-
vals. A 20 mm SL clam was chosen to be the “standard”
clam, because growth at this size was relatively rapid and
would thus easily demonstrate seasonal differences, and
because use of this size did not involve extrapolation from
measured size groups.

A typical life cycle was estimated by assuming that
mean release of young occurred on April 21 and in late

1976

Fig. 3. Comparison of mean shell lengths of clams held in cages
in the intake and discharage areas of the Kingston Steam Plant during
1976. Intake cage I was lost between March and June so a new cage
(Intake II) was positioned in April. The dotted line is a hypothetical
construction of clam growth in Intake I during the period between
actual measurements.

170

CORBICULA SYMPOSIUM

August, that release size was 0.2 mm SL, and that growth
to the mean maximum size found in the natural population
in the discharge area would approximate that which occur-
red in 1977. The April 21 date was chosen based on reproduc-
tive data presented below, on the appearance of recruits in
the cages, and on earlier reports (Heinsohn, 1958; Isom,
1971 ; Aldridge, 1976; Aldridge and McMahon, 1978) relating
appearance of young to water temperature. The late August
date was chosen by back-calculation from the size of recruits,
which could not have been from the spring generation, in the
field growth cages. For each subsequent interval the change
in SL for the appropriate size of clam was calculated from
the linear regression of increase in SL on initial shell length
(ISL) for that interval. (Equations may be found in Mattice and
Wright, 1 985, Appendix D.) This increase in SL was then add-
ed to the initial SL to obtain the initial SL for the next inter-
val. This procedure was continued until the growth curve
became asymptotic and the time for this occurrence was then
chosen to indicate the mean length of life. Determination of
the asymptote is somewhat arbitrary, but comparison of the
plot with a standard Walford (1946) plot helps provide support.

RESULTS AND DISCUSSION*

FIELD WATER TEMPERATURES

Water temperatures at the cage sites followed a typical
yearly cycle (Fig. 2) for the period of all studies. Diurnal
temperature variation was small ( < 1 °C) except during periods
of rapid temperature change in spring and fall (<2°C). The
discharge temperature at the cage site averaged about 5°C
higher than that in the intake, the difference tending to be
largest at the winter and summer extremes.

FIELD STUDIES - SEASONAL PATTERN OF GROWTH
Growth (mean shell length increase) of Corbicula
fluminea from December 1975 through December 1976 was
distinctly seasonal (Fig. 3). The clams (ISL ranged from
10.7-16.4 mm) did not grow significantly in either the
discharge or intake cages from December 5, 1975, through
February 25, 1976. Water temperatures during this period
did not exceed 16.0°C (m = 10.9°C) or 10.5°C (m = 7.5°C)
in the discharge and intake, respectively. Discharge clams
had grown significantly by the next time shell lengths were
measured on April 6. Water temperatures during this period
ranged from 1 6-1 8°C (m = 1 6.5°C). The first intake cage was
lost between February 25 and June 1 1 , but clams in the se-
cond cage did grow in the period from April 29 to May 21,
suggesting that the hypothesized relation (Fig. 3) is not
unreasonable. Growth of clams in the intake and discharge
continued at a fairly rapid rate from April through September
or October, then rapidly decreased. Significant shell length
increases were not observed in November or December when
temperatures averaged 14.8°C in the discharge and 8.9°C
in the intake.

'Summary data are presented here; the complete data set is included
in Mattice and Wright (1985).

Table 1. Mean growth (mm) of clams1 held in the intake and
discharge areas of Kingston Steam Plant during winter and early
spring, 1978-1979.

Date

INTAKE

DISCHARGE

Water

Temp.

(°C)

Number

Clams

Mean

Growth2

(mm)

Water

Temp.

(°C)

Number

Clams

Mean

Growth2

(mm)

11/21/78

16.5

12/1/78

—

18.9

—

1/11/79

2.4

0.12*

5.9

0.23

1/26/79

2.4

0.01

5.1

0.02*

2/16/79

1.0

0.00

4.6

0.00

3/9/79

8.4

0.01

6.8

0.07*

3/30/79

10.2

0.05*

14.6

0.26*

4/27/79

16.1

0.09

20.4

—

1 1nitial shell lengths ranged from 4.2 to 9.1 mm.
2Since previous measurement.

'Significantly different from zero (t test, p = 0.05).

Absence of winter growth also applied to clams in the
4 to 9 mm SL range (Table 1). Growth was essentially nil in
the intake for the whole period from November 21 through
April 27, 1979, and in the discharge for the period from
January 11 through March 9, 1979 (Table 1). When growth
did occur, in a statistical sense, discharge clams grew more
than those in the intake. Biologically significant growth ap-
pears limited to a 7 to 9 month period in the Clinch River
system.

Despite the appearance of the growth curves in
Figure 3, growth, expressed as mean shell length (SL) in-
crease per day, of Corbicula was not constant over the sum-
mer months (March through August 1976) in either the in-
take or discharge (Fig. 4). In both areas, growth increased

1976

Fig. 4. Mean growth rates (and 95% confidence limits) of clams in
the intake and discharge areas of Kingston Steam Plant during
various periods of 1976. Growth rate for a period is plotted at the
midpoint of that period.

MATTICE AND WRIGHT: GROWTH OF CORBICULA FLUMINEA

171

INITIAL SHELL LENQTH (mm)

Fig. 5. Relation between shell length increase and initial shell length
for clams held in the intake and discharge areas in 1976. The In-
take I and Discharge results are for the period February 25 through
October 18, while those for Intake II are for the period April 19 through
November 19.

to a peak in May then dropped radically in early June before
again increasing in late June. A second peak in growth in
late June or July was again followed by a decline. Growth
rates in the discharge generally exceeded those in the in-
take except during June and iate August through September.
Because of the relationship between ISL and growth (see
below), caution must be used in comparing the above rates
between intervals, as they ignore the fact that the clams grew
progressively larger over the period of measurement.

FIELD STUDIES - RELATIONSHIP BETWEEN GROWTH
AND INITIAL CLAM SIZE

Growth of Corbicula in 1 976 was negatively correlated
with ISL (Fig. 5). The correlation coefficents (r2) for each group
were 0.29, 0.41 , and 0.37 for clams in cages Discharge, In-
take I, and Intake II, respectively. Each of the r2 values was
significantly different from zero (F test, p < 0.05) despite the
relatively small range of initial shell lengths for each group.
Direct comparison of growth in all three groups was obviated
by the unequal periods of measurements, however, Intake
I and Discharge data were collected over the same period
(February 25 to October 18, 1976). Growth in the discharge
area generally exceeded that in the intake area in 1976 (F
test; p <0.05). The relationship between growth and ISL in-
dicates that valid comparisons of growth between areas or
over time must be carried out for clams of similar size.

Clam growth in 1977 was also inversely related to ISL
(Fig. 6). Both linear and exponential models were examined
for fit to the relationship between SL and ISL. Based on com-

parison or r2 and p values and residual sums of squares, the
linear model provided a better fit both for the whole period
the clams were in the field and for a great majority of the in-
divdual observation periods (Mattice and Wright, 1985). For
reasons sited in Mattice and Wright (1985), data for clams
in the 5 mm ISL size class were not included in the analysis.
This did not affect the choice of the linear model. Coefficients
of correlation for clams 10 mm SL and larger were 0.96 and
0.83 for intake and discharge clams, respectively. Britton et
al. (1979) reported that an exponential model provided a bet-
ter fit to their data, but this conclusion was probably influenc-
ed by inclusion of data for unequal time intervals and different
initiation dates in the analysis. McMahon (this volume), on
the other hand, found that a linear model provided the best
fit for growth of the clams that he studied and data reported
by Dreier and Tranquilli (1981) also appear to support choice
of a linear model.

FIELD STUDIES - GROWTH COMPARISONS BETWEEN
SITES AND YEARS

In 1977, clams in the intake grew more rapidly than
clams in the discharge, except for the largest clams (Fig. 6).
The intercepts and slopes are significantly different (p
<0.001); Ftest). The relationship between growth rate in mm
SL/day (G) and initial shell length in mm (ISL) was G =
-0.0013 ISL + 0.064 for the intake clams and G = -0.001 1

Fig. 6. Growth rates of clams held in the intake (•) and discharge
(O) areas of the Kingston Steam Plant in 1 977 as a function of initial
size. Clams were held from April 21 to November 23, 1977. Solid
lines are least squares fits to data for all clams excluding the 5 mm
SL size group. Dashed lines are least squares fit to data for all clams.

172

CORBICULA SYMPOSIUM

ISL + 0.049 for the discharge clams. The relationship bet-
ween growth in the intake and discharge (intake < discharge)
is the opposite of that found in 1976. This will be discussed
later, but it is clear that clam growth can differ between sites.

Rate of SL increase per day also can differ between
years at the same site (Fig. 7). The total interval over which
growth was compared, differed somewhat between 1977
(April 21 -October 4 or 166 days) and 1978 (April 4-September
27 or 175 days), but the longer time in 1978, all within the
growth period for the clams, would tend to increase the 1978
growth rate in comparison to that for 1977. Nevertheless,
growth in the intake in 1977 exceeded that in 1978 (F test;
p< 0.05). The relationship between growth rate (G) and in-
itial shell length (ISL) for 1977 was G = -0.0016 ISL + 0.079
(r2 = 0.96) and for 1978 was G = 0.0016 ISL + 0.070 (r2
= 0.93). The slopes did not differ significantly (F test; p =
0.50), but the intercepts were significantly different (p<0.05).

The growth rates determined in our studies are well
within the range of those reported by other investigators. Cor-
bicula near the Kingston Steam Plant grow significantly faster
than two populations in Hong Kong (Morton, 1977), at about
the same rate as those in Lake Benbrook, Texas (O’Kane,
1 976; Britton et al., 1 979), and substantially slower than those
in Lake Arlington, Texas (Aldridge, 1976; Aldridge and
McMahon, 1978). The reasons for these differences are
unknown.

Fig. 7. Growth rates as a function of initial shell lengths of clams
held in the intake of the Kingston Steam Plant in 1977 (O) and 1978
( • ). The smaller groups of clams were not individually marked, and
thus were assigned the mean initial shell length for the groups.

o.to

0.08

0.06

0.04

0.02

la)

20.4

25.1

24.4

DISCHARGE

27.8

29.4

28.6

27.9

28.4

26.4

20.6

4/19 5/13 6/3 6/28 7/12 7/20 8/11 8/25 9/16 10/4

Fig. 8. Growth rates for a “standard" 20 mm SL during various in-
tervals from April through November 1977, in the intake (a) and
discharge (b) of the Kingston Steam Plant. Mean temperature (°C)
for each interval is shown above the growth rate for that interval.

FIELD STUDIES - GROWTH DURING SUCCESSIVE
INTERVALS

Growth of clams in 1977, normalized for initial size,
was not constant during the yearly growth period (Figures
8a and 8b). Normalization was achieved by calculating growth
of a 20 mm SL clam for each period from a regression of
growth for the period on size at the beginning of the period
(for equations see Mattice and Wright, 1985). In the discharge,
growth in the first period was at about the median rate for
all periods. This was followed by a distinct decrease for the
May 13 and June 3 period. Growth in the discharge then rose
to its maximum which included most of the month of June.
Mean temperature during this period was 25.1°C. Growth
from July through November declined roughly linearly through
a period of increasing then decreasing temperature. In the
intake, growth from April to June was low, followed im-
mediately by growth at the maximal rate observed for the
year. Growth was then relatively constant through late August.
Temperatures also were relatively constant during this period.
A second period of maximal growth occurred during the
August 25 to September 1 6 period, after which growth declin-
ed rapidly as the temperature also declined. Growth in the
discharge was higher than in the intake from mid-April
through late June, but this relationship was reversed for the
rest of the growth period. The higher growth in the intake dur-
ing this latter (and longer) period accounted for the higher
overall growth of intake clams over the whole time the clams
were in the field.

RELATIONSHIP BETWEEN FOOD AVAILABILITY AND
GROWTH

We examined both algal type and seston concentra-
tion as potential causes of some of the differences that we

MATTICE AND WRIGHT: GROWTH OF CORBICULA FLUMINEA

173

observed in growth rates. Davis and Guillard (1958) found
that growth of larvae of the oyster, Crassostrea virginica
declined when they were fed high concentrations of the algae,
Monochrysus, but the relationship of seston concentration or
algal type and growth of clams has not been examined.
However, a number of investigators have demonstrated ef-
fects of algal concentration or types on filtration, the primary
food gathering mechanism of most bivalves.

Diatoms were the dominant phytoplankton in samples
collected in the intake and discharge areas of Kingston Steam
Plant between October 7, 1975, and November 9, 1976. A
total of 48 genera were identified (Mattice and Wright, 1 985),
but diatoms always constituted 95% or more of the
phytoplankton in terms of both cell number and cell volume.
The change in mesh size of nets used before and after the
June sampling period did not affect conclusions with regard
to dominant classes or genera (Mattice and Wright, 1985).
Percent distribution of the genera of Bacillariophyceae and
Chrysophyceae was quite uniform throughout the year.
Melosira was the most prominent genus, making up greater
than 90% of total cell number except in May, when
Asterionella and Fragilaria each accounted for 5 to 10% of
total cell number, and on June 24, when Melosira accounted
for about 50% of the total number and Asterionella, Fragilaria,
Stephanodiscus, and Synedra made up about 45% of total
number.

The other phytoplankton taxa were represented by
more species in summer than winter, but were never domi-
nant. Chlorophycean diversity was high especially in sum-
mer, but at no time did the total come to more than about
3% of the total of all phytoplankton, Cyanophyceae,
Dinophyceae, and Euglenophyceae combined never ac-
counted for as much as 2% of total numbers, and generally
were found sparingly. Phytoplankton were present sparsely
in January, February, and March, but had increased greatly
by May in both intake and discharge samples. Intake and
discharge samples were generally similar except on February
10 and March 16. On the former date, only empty frustules
of Melosira and Fragilaria were found in the discharge
although cells in the intake samples appeared viable. On
March 1 6, numbers of both genera and cells of phytoplankton
in the discharge were substantially lower than in the intake.

Although the phytoplankton populations in the study
area were diverse, it is unlikely that changes in the algal com-
plement could account for any differences observed in growth
of Corbicula. Diatoms were dominant throughout the year.
All the dominant genera are large, which may have affected
the maximum filtration rate of the clams (Morton, 1971), but
it seems doubtful that the variation in size that does occur
would be responsible for seasonal effects. Furthermore, con-
centrations of potentially toxic phytoplankton such as Gym-
nodinium veneficum (Ballatine and Morton, 1956), Chlorella
(Davids, 1964), and Ceratium hirundinella (Stanczykowska et
ai, 1976), which have been shown to inhibit bivalve filtration,
either were not present or were rare. Thus, it seems unlikely
that the type of algae in the seston had any effect on Cor-
bie u la growth.

Seston concentration did vary significantly during the

year, ranging from about 7 to 23 mg/I from March to October
in 1977 and 1978 (Figure 9). Seston levels in the intake and
discharge frequently were not measured on the same date.
On dates where comparable data were taken, seston levels
in the discharge were slightly lower, but we assumed that this
difference was too small to be significant. We therefore

1977

Fig. 9. Seston concentrations measured in the intake and discharge
of the Kingston Steam Plant from March through October 1977 and
May through November in 1978. Line connects values for consecutive
dates in each year regardless of site.

combined the data from the two areas for seasonal com-
parisons. In 1977, concentration rose from early March to a
peak in mid-April, then declined through about mid-May
before rising to the seasonal maximum in early June. Seston
dropped to the lowest seasonal concentration in early July,
then, except for one low value in late August, remained bet-
ween 15 and 20 mg/1 through mid-October. For most of 1978,
seston concentrations were lower than in 1977 (Figure 9). The
major differences were the lack of a May-June peak in seston
concentration in 1978 and the sustained low concentrations
in August and early September. By late September concen-
trations rose rapidly to reach 1977 levels in mid-October.

Growth differences at the intake and discharge sites
(1 each) for periods within 1977 (Figure 8) did not correlate
with seston concentration. This type of analysis is permitted,
because growth in successive intervals was not correlated
(see Materials and Methods). Correlation coefficients (r2) for
the relationship between mean seston concentration and
growth rate for each interval were 0.041 (N = 9; p = 0.61)
and 0.001 (N = 9; p = 0.98) for intake and discharge areas,
respectively. In addition, correlation coefficients were not
significantly increased by assuming that growth in an inter-
val was related to food availability in the prior interval (r2 =
0.04 and 0.02 for intake and discharge areas, respectively).
Unless seston concentration was completely independent of
the amount of usable food available to Corbicula, food sup-
ply did not appear to be related to growth for this year. Thus,
we conclude that changes in seston concentration were not
responsible for the differences in growth that we observed.

This conclusion seems counterintuitive. Mattice (1979)
has shown that filtration rate (volume of water pumped/time)
of Corbicula is not related to seston concentration over the
range from 7 to 24 mg/I. This would suggest that the higher

174

CORBICULA SYMPOSIUM

the seston concentration the more food that Corbicula could
remove from the water to support growth. There are obvious
limits to this expectation as indicated by results of earlier
workers (Loosanoff and Engle, 1947; Morton, 1971; Thomp-
son and Bayne, 1974), who found that much higher concen-
trations of inert particles or monoalgal cultures caused
decreases in filtration rate of mussels, oysters, and the zebra
clam, Dreissena polymorpha. However, filtration is only
the first step in the feeding process, sorting and ciliamediated
transport to the mouth also being necessary for ingestion.
During normal feeding large particles are rejected prior to
reaching the mouth and ejected from the mantle cavity as
pseudofeces. At high concentrations of seston, ingestion
declines and may actually cease even though filtration con-
tinues. We have no hard data to indicate whether this might
cause the lack of correlation between seston concentration
and growth, but substantial amount of pseudofeces were pro-
duced by Corbicula during field studies of filtration (Mattice,
1979; Mattice and Wright, unpublished data). Further
possibilities are that growth of Corbicula in the Clinch River
system is dependent on the presence of the smaller algal
species or that a factor other than food, e.g., calcium, is the
limiting factor in growth. Clearly, further studies of relation-
ships between nutrition and growth are needed.

EFFECTS OF TEMPERATURE ON GROWTH

We examined the relationship between temperature
and Corbicula growth in the laboratory. As in the field studies,

Fig. 10. Growth rates and 95% confidence intervals for clams with
initial shell lengths of 5-10 mm and 10-20 mm at 15, 20, 25, and 30°C.
Optimum growth temperature appears to be about 25°C.

growth rates of clams in the laboratory varied with initial size
of the clams. All but 4 of the 130 clams examined grew over
the four week period but growth of clams >20 mm ISL ap-
peared to be limited by total nutritional intake and these data

Fig. 11. Growth rates of clams during each interval in 1976 and 1977 in the discharge (a) and intake (b) areas plotted against the mean
temperature of the interval.

MATTICE AND WRIGHT: GROWTH OF CORBICULA FLUMINEA

175

are not presented here (see Mattice and Wright, 1985).
Growth rates of clams in the two smaller size groups followed
the usual relationship found between physiological rates and
temperature (Fig. 10). Growth rates of clams was low at 15°C.
Growth rates increased with temperature through about 25°C,
then decreased at 30°C. Optimum growth temperature ap-
pears to be about 25°C, although rates at 20°C and 30°C were
not significantly lower. Extrapolation from these data to lower
temperatures suggested 1 1 to 1 2°C as a lower threshold for
growth, which is within the range (8 to 16°C) found during
our various field studies.

We also considered the relationship between
temperature and growth rate using field data collected in 1976
and 1977 (Fig. Ila-d). The validity of this type of evaluation
is based on the lack of correlation between growth rates of
clams during consecutive sampling periods. Gur logic is
somewhat circular, but it appears that the relationship bet-
ween growth rate and temperature in the field is similar in
form to that found in the laboratory (growth rate is highest
at water temperatures about 25°C) except during May and
early June and during the late summer-autumn period after
occurrence of the highest water temperatures in both the in-
take and discharge areas. A similar relationship was found
when data from Aldridge (1976) were plotted in this way sug-
gesting that these periods are important for understanding
the seasonal shifts in growth of Corbicula. The mid-20°C op-
timum for growth, which is supported by both laboratory and
field data on growth, may reflect optima for filter feeding
(Mattice, 1979) and/or oxygen consumption (McMahon and
Aldridge, 1976).

CLAM REPRODUCTION

Growth rates appear to decline during the period of
incubation. The small percentage of adults incubating young
and the predominance of tanned shells of the young on May
26, 1977, followed by the total absence of young in the adults

Table 2. Occurrence of young Corbicula in the incubatory gills of
adults.

Date3

Areab

Number

Glams

Percent of Clams Containing
Indicated Number of Larvae in
Single Gill

<100

20-99

1-19

5/26/77

22c

6/1/77

—

100

9/27/77

33d

13c

10/11/77

40°

—

10/19/77

2QC

—

10/19/77

—

100

10/20/78

—

11/21/78

—

100

aln day order, regardless of year collected.
bD = discharge area; I = intake area.
cShells of most young tanned (= older stages).
dShells of most young clear (= younger stages).

collected June 1 suggest the end of a spring incubation period
(Table 2). A similar situation (declining occurrence of in-
cubating young) is also indicated for the period from
September 27 through November 21 even though samples
were taken in two areas and on two different years. Although
further work is required for confirmation, these data suggest
a bimodal reproductive period with peaks in release of young
occurring before late May and sometime between early June
and late September (probably closer to the latter) such as
has been found for Corbicula populations in some other areas
(Heinsohn, 1958; Coldiron, 1975; Aldridge and McMahon,
1 978). Growth rates of clams during both 1 977 and 1 978 ap-
pear to decline at about the same time as the spring peak
of larval release, but the coincidence of the late summer
growth decline and peak of larval release is not clear. Decline
at the time of incubation is not unexpected because the gills
become distended as the larval clams grow and water flow
is likely inhibited.

LONGEVITY AND MAXIMUM SIZE

Corbicula in the Clinch River system appear to have
life spans that can approach six or more years, but almost

GROWTH SEASON

Fig. 12. Hypothetical mean growth over time of Corbicula from release
(~0.2 mm) through death. The growth curve is based on data col-
lected in the intake of the Kingston Steam Plant in 1977 and assumes
April 21 and late August release dates. Growth season refers to
growth within a 12 month period except for the first growth season.
Not shown are the cessations of growth during each winter period.

certainly exceed four years (Figure 1 2). The smoothed curves
obscure seasonal shifts in growth (Figure 8), but are conve-
nient. The equation used to estimate growth over each 12
month period was FSL = 13.86 + 0.7IISL, where FSL = final
shell length and ISL = initial shell length, both in millimeters.
This relationship was derived from measured growth of clams
ranging from 4.8 to 48.4 mm SL that were observed in the
intake in 1977 between April 21 and November 23. This in-
terval encompasses the total growth period in the intake cove.
Equations derived for shorter intervals (Mattice and Wright,
1985) were used to estimate growth during the first growing
season for both spring and fall generations. Year to year shifts
are not considered since only data from one year were used

176

CORBICULA SYMPOSIUM

to generate the curve. Regardless, it seems unlikely that the
conclusions would be invalidated.

Life spans reported for populations of Corbicula have
ranged from 14-17 months (Aldridge and McMahon, 1978)
to 5-7 years (Sinclair and Isom, 1963). Britton et al. (1979)
presented arguments regarding interpretation of length-
frequency distributions and concluded that the typical life cy-
cle of Corbicula fluminea was three or fewer years. They at-
tributed the longer estimates to the failure of earlier in-
vestigators to recognize that young clams were released dur-
ing two rather distinct periods each year. Their argument is
compelling, but the same criticism does not apply to our
results or to those of Dreier and Tranquilli (1981), both of
which support life cycle estimates of four or more years. Cage
studies (Dreier and Tranquilli, 1981; this study) are subject
to the criticism of confinement or cage effect, but Britton et
al. (1 979) showed that such criticism was largely unwarranted
even for their cages which had smaller and fewer openings
and thus would have inhibited water flow substantially more
than in our study and that of Dreier and Tranquilli (1981). Fur-
ther substantiation for our longer (>4 years) estimate of
longevity is evidenced by the fact that only one of the ten
clams of >40 mm SL in our cages died during the period from
April 21, 1977, to September 27, 1978. Even if we assume
the minimum times to reach the 40 mm size class from
Aldridge and McMahon (1978) or Britton etal. (1979), our con-
clusions regarding longevity based on the subsequent 1.4
years survivorship would differ from theirs. We agree that the
three- or fewer-year life cycle estimate is reasonable for the
populations they studied, but not for those in Illinois (Dreier
and Tranquilli, 1981), Tennessee, and perhaps California
(Eng, 1979). It appears that there is no typical C. fluminea
life cycle, though reasons for this are yet obscure.

Plasticity also appears to characterize maximum adult
size of different populations. The von Bertalanffy (1938)
growth model appears to provide a good fit to data on Cor-
bicula populations (Morton, 1973 and 1977; Alimov, 1974).
This is true for our growth data as well (displayed as a Walford
plot in Figure 13) and provides an estimate of maximum
theoretical shell length of 47.6 mm. As stated by Britton and
Morton (1979), however, maximum theoretical length appears
to vary from population to population, a conclusion that is sup-
ported by observed maximal lengths ranging from 25 mm
(Lauritsen, 1982) to the 75-mm behemoth that Billy Isom (Ten-
nessee Valley Authority, Muscle Shoals, Ala.) displayed at
this Symposium. Analysis using a Walford (1946) plot may
prove to be of limited use even within a single population
because in the early 1970s it was not unusual to find clams
in the 50- to 60-mm size range (maximum size ~65 mm) in
the discharge area near the site of our cage studies (Mattice
and Wright, unpublished observations). Perhaps these larger
clams are characteristic of an earlier stage of population
development, because clams >50-mm SL have become rare
in our collections. Again, however, the variability of response
of Corbicula in different environments is obvious, a
characteristic that would appear to allow the clam to exploit
a broad range of environments and spread as rapidly as it
has through the United States.

Fig. 13. A Walford plot for Corbicula fluminea based on data
presented in Figure 12. Shell length on April 21 for one year (Lt + -|)
was plotted as a function of shell length on the same date as a year
before (Lt). Initial size for young at the time of release was assumed
to be 0.2 mm. The maximum theoretical length was calculated as
47.6 mm SL.

This plasticity does not augur well for development of
standardized schedules of Corbicula control at industrial
facilities. Most fouling occurs after the clams, which have
entered the plant as larvae or young, grow to fouling size.
In Tennessee, control would appear to be effectively applied
once per year because clams would not be likely to grow to
fouling size in the interim. Flowever, at least in some areas
of Texas fouling could be a problem unless control strategies
are applied twice per year. Further investigation of the intrinsic
and extrinsic factors that control clam growth will be required
to determine control startegies at specific industrial sites.

SUMMARY

Growth in shell length (SL) of Corbicula fluminea varies
during periods within a year and between years and is related
to both intrinsic and extrinsic factors. During the periods of
the year when ambient water temperature is low, clams do
not grow. The threshold for growth seems to be about
1 1-1 2°C; the range of estimates from the different studies sug-
gest that some other environmental factor such as food supply
may play modifying role. As temperatures begin to rise in
spring, growth begins and for most periods is linearly related
to clam size: smaller clams grow faster than larger clams.
Growth rate appears to increase as temperature increases
until May or early June when there is a rapid decline in growth,
apparently related to the peak period of incubation of the
Spring generation of young clams. This decline seems likely
to be related to both shifts in energy use and inhibition of
water flow for feeding by the young clams in the inner, in-

MATTICE AND WRIGHT: GROWTH OF CORBICULA FLUMINEA

177

cubatory, half of the gill. Growth rate then increases again,
the level of growth attained being consistent with a hyper-
bolic relation between growth and temperature, with a
temperature about 25°C being the optimal growth
temperature. Food (measured as seston concentration) does
not appear to be related to growth, but it is doubtful that this
conclusion is generally applicable to all environments where
Corbicula is found. Growth from mid to late June through mid-
August or early September appears to follow temperature,
again suggesting a 25°C optimal growth temperature.
However, growth rate in the subsequent period, which also
includes the second yearly peak of incubation and release
of young, is substantially less than would be expected based
on temperature. Growth continues at a low rate through
October or November, depending on temperature, then
essentially ceases for the winter period (temperatures less
than 1 1 or 12°C). Growth rate also varies from year to year
at the same site, but the reason is less clear.

The life span of Corbicula in the Clinch River system
is more than four years and could be as long as 6 + years.
Differences between observations of file span and maximum
size of Corbicula in populations as widely dispersed as Hong
Kong, Tennessee, Texas and Illinois suggest that Corbicula
has a plastic rather than a predetermined physiology, which
allows populations to efficiently exploit a wide range of en-
vironmental conditions. This plasticity is undoubtedly respon-
sible for the rapid and unprecedented spread of Corbicula
in the United States and its success in infesting industrial
facilities of all kinds.

ACKNOWLEDGMENTS

We thank Dr. Hollings Andrews, Tennessee Technological
University, Cookeville, Tenn., for providing analyses of phytoplankton
populations and B.G. Blaylock and R. M. Cushman for providing
helpful comments on the draft of the manuscript. Mr. L. B. Kennedy
Plant Superintendent of the Kingston Steam Plant, Kingston, Tenn.,
graciously allowed us to conduct our studies within the confines of
plant grounds and the Delta Management Section of the Tennessee
Valley Authority supplied printouts of water temperature data at sta-
tions near the plant.

Research sponsored by the Office of Health and Environmental
Research, U. S. Department of Energy, under contract DE-
AC05-840R21400 with Martin Marietta Energy Systems, Inc., En-
vironmental Sciences Division Publication 2389.

LITERATURE CITED

Aldridge, D. W. 1976. Growth, reproduction, and bioenergetics in
a natural population of the Asiatic freshwater clam Corbicula
manilensis Philippi. M. S. Thesis, the University of Texas at
Arlington. 97 pp.

Aldridge, D. W. and R. F. McMahon 1978. Growth, fecundity, and
bioenergetics in a natural population of the Asiatic freshwater
clam, Corbicula manilensis Philippi from north central Texas.
Journal of Moliuscan Studies 44: 49-70.

Alimov, A. F. 1974. Growth regularities in fresh water bivalve
mollusks. Zhurnal Obshehai Biologii 35: 576-589.

Ballantine, D. and J. E. Morton 1956. Filtering, feeding, and diges-
tion in the Lamellibranch Lasaea rubra. Journal of the Marine
Biological Association U. K. 35: 241-274.

Britton, J. C., D. R. Coldiron, L. P. Evans, C. Golightly, K. D. O'Kane
and J. R. Ten Eyck 1979. Reevaluation of the growth pattern
in Corbicula fluminea (Muller), pp. 177-192. In: J. C. Britton,
J. S. Mattice, C. E. Murphy and L. W. Newland (eds.), Pro-
ceedings of the First International Corbicula Symposium. TCU
Research Foundation, Fort Worth, Texas.

Britton, J. C. and B. Morton. 1979. Corbicula in North America: the
evidence reviewed and evaluated, pp. 249-287. In: J. C. Brit-
ton, J. S. Mattice, C. E. Murphy and L. W. Newland (eds.),
Proceedings of the First International Corbicula Symposium.
TCU Research Foundation, Fort Worth, Texas.

Clarke, A. H. 1 981 . Corbicula fluminea in Lake Erie. Nautilus 95: 83-84.

Coldiron, D. R. 1975. Some aspects of the biology of the exotic
mollusk Corbicula (Bivalvia: Corbiculidae). M. S. Thesis, Texas
Christian University, Fort Worth, Texas. 92 pp.

Davids, C. 1964. The influence of different suspensions of
microorganisms of different concentrations on the pumping
and retention of food by the mussel Mytilus edulis (L.).
Netherlands Journal of Sea Research 2: 233-249.

Davis, H. C. and R. R. Guillard 1958. Relative value of ten genera
of microorganisms as foods for oysters and clam larvae.
Fishery Bulletin, U.S. Fish and Wildlife Service 58: 293-304.

Dreier, H. and J. A. Tranquilli 1981. Reproductive, growth, distribu-
tion, and abundance of Corbicula in an Illinois cooling lake.
In: R. W. Larimore and J. A. Tranquilli (eds.) The Lake
Sangchris Study: Case History of an Illinois Cooling Lake. Il-
linois Natural History Survey Bulletin 32: 378-393.

Eng, L. L. 1 979. Population dynamics of the Asiatic clam; Corbicula
fluminea (Muller), in the concrete-lined Delta-Mendota Canal
of Central California, pp. 39-68. In J. C. Britton, J. S. Mattice,
C. E. Murphy, and L. W. Newland (eds.), Proceedings of the
First International Corbicula Symposium. TCU Research Foun-
dation, Fort Worth, Texas.

Goss, L. B. and C. Cain 1976. Condenser and raw water system foul-
ing by Corbicula. pp. 11-22. In L. D. Jensen (ed ), Biofouling
Control Procedures. Marcel Dekker, Inc., New York.

Goss, L. B., J. M. Jackson, H. B. Flora, B. G. Isom, C. Gooch, S.
A. Murray, C. G. Burton, and W. S. Bain 1979. Control studies
on Corbicula for steam-electric generating plants, pp. 139-151 .
In J. C. Britton, J. S. Mattice, C. E. Murphy and L. W. Newland
(eds.), Proceedings of the First International Corbicula Sym-
posium. TCU Research Foundation, Fort Worth, Texas.

Heinsohn, G. E. 1958. Life history and ecology of the freshwater clam,
Corbicula fluminea. M. S. Thesis, University of California,
Berkeley, 64 pp.

Isom, B. G. 1971. Evaluation and control of macroinvertebrate
nuisance organisms in freshwater industrial supply systems.
Titles and Abstracts of the 19th Annual Meeting of the Midwest
Benthological Society, March 24-26, South Bend, Indiana.

Lauritsen, D. D. 1982. Biological interactions between mussels and
the Asiatic clam Corbicula fluminea in Lake Waccamaw, North
Carolina. Titles and Abstracts of the Thirtieth Annual Meeting
of the North American Benthological Society Meeting, May
18-21, 1982 at the University of Michigan, Ann Arbor, Ml.

Loosanoff, V. L. and J. B. Engle 1947. Effect of different concentra-
tions of microorganisms on the feeding of oysters (O. virginica).
Fishery Bulletin, U. S. Fish and Wildlife Service 51: 31-57.

Mattice, J. S. 1979. Interactions of Corbicula with power plants, pp.
119-138. In J. C. Britton, J. S. Mattice, C. E. Murphy and L.
W. Newland (eds.), Proceedings of the First International Cor-
bicula Symposium. TCU Research Foundation, Fort Worth,
Texas.

Mattice, J. S. 1983. Freshwater fouling with emphasis on Corbicula.
Chapter 4: pp. 1-30. 1. A. Diaz-Tous, M. J. Miller and Y. Musalli

178

CORBICULA SYMPOSIUM

(eds.), Condenser Macrofouling Control Technologies-The State
of the Art. EPRI CS3343. Electric Power Research Institute,
3412 Hillview Ave., Palo Alto, CA 94303.

Mattice, J. S. and W. Bosworth 1979. A modified venturi suction
sampler for collecting Corbicula. Progressive Fish Culturist 41 :
121-123.

Mattice, J. S. and L. L. Wright 1985. Aspects of Growth of Corbicula
fluminea. ORNL/TM-8910. Environmental Sciences Division,
Oak Ridge National Laboratory, Oak Ridge, IN 37830. 62 pp.

McMahon, R. F. and D. W. Aldridge 1976. Respiratory response to
temperature and low oxygen tension in Corbicula manilensis
Philippi. Abstracts of the thirty-ninth Annual Meeting of the
American Society of Limnology and Oceanography, June
21-24, 1976. Skidaway Institute of Oceanography, Savannah,
GA.

Morton, B. 1971. Studies on the biology of Dreissena polymorhpa
Pall. V. Some aspects of filter-feeding and the effect of
microorganisms upon the rate of filtration. Proceedings of the
Malacological Society of London 39: 289-301 .

Morton, B. 1973. Some aspects of the biology and functional mor-
phology of the organs of feedings and digestion of Limnoperna
fortunei (Dunker) (Bivalvia: Mytilacea). Malacologia 12: 265-281.

Morton, B. 1977. The population dynamics of Corbicula fluminea
(Bivalvia: Corbiculacea) in Plover Cove Reservoir, Hong Kong.
Journal of Zoology of London 181: 21-42.

Morton, B. 1979. Corbicula in Asia. pp. 15-38. In J. C. Britton, J. S.
Mattice, C. E. Murphy and L. W. Newland (eds.), Proceedings

of the First International Corbicula Symposium. T.C.U.
Research Foundation, Fort Worth, Texas.

O’Kane, K. D. 1976. A population study of the exotic bivalve Cor-
bicula manilensis (Philippi, 1841) in selected Texas reservoirs.
M. S. Thesis, Texas Christian University. Department of
Biology, Fort Worth, Texas. 134 pp.

Sinclair, R. M. and B. G. Isom 1963. Further studies on the introduced
Asiatic clam Corbicula in Tennessee. Tennessee Pollution
Control Board, Tennessee Department of Public Health, Cor-
dell Hull Building, Nashville, Tenn. 76 pp.

Smith, M. H., J. Britton, P. Burke, R. K. Chesser, M. W. Smith and
J. Hagen 1979. Genetic variability in Corbicula, an invading
species, pp. 243-248. In J. C. Britton, J. S. Mattice, C. E.
Murphy and L. W. Newland (eds.), Proceedings of the First
International Corbicula Symposium. T.C.U. Research Foun-
dation, Fort Worth, Texas.

Stanczykowska, A., W. Lawacz, J. S. Mattice and L. Lewandowski
1976. Bivalves as a factor effecting circulation of matter in
Lake Mikolajskie (Poland). Limnologica (Berlin) 10(2): 347-352.

Thompson, R. J. and B. L. Bayne 1974. Some relationships between
growth, metabolism, and food in the mussel Mytilus edulis.
Marine Biology 27: 317-326.

von Bertalanffy, L. 1938. A quantitative theory of organic growth.
Human Biology 10: 181-213.

Walford, L. A. 1946. A new graphic method of describing growth of
animals. Biological Bulletin 90: 141-147.

CORBICULA IN PUBLIC RECREATION WATERS OF TEXAS: HABITAT
SPECTRUM AND CLAM-HUMAN INTERACTIONS

RAYMOND W. NECK

TEXAS PARKS AND WILDLIFE DEPARTMENT
4200 SMITH SCHOOL ROAD
AUSTIN, TEXAS 78744, U.S.A.

ABSTRACT

Examination of water bodies associated with Texas State Parks and other selected recreational
areas revealed presence of Corbicula in 41 of 92 surveyed sites. Absence of Corbicula is related to
presence of salt or brackish waters, extreme winter temperatures, periodic water shortages in small
streams and lakes, and undefined ecological of historical factors which have delayed invasion into
certain waters. Prime microhabitats for Corbicula in Texas include sandy-bottomed streams in eastern
Texas, loose-gravel-bottomed pools in the Texas Hill Country, and moderate-energy reservoir lake
shores. Construction of recreational facilities may either enhance or reduce the local habitat for Cor-
bicula. No harmful effects of Corbicula upon park facilities were discovered.

The Asiatic dam, Corbicula fluminea (Muller, 1774),
has spread over most of the United States (McMahon, 1 982)
following its initial introduction into the northwestern United
States in the early twentieth century (Burch, 1944; Counts,
1981). The first Texas populations were discovered near El
Paso in 1964 (Metcalf, 1966), followed by reports from the
lower Rio Grande by 1969 (Murray, 1971). East Texa popula-
tions probably invaded from Louisiana (Britton, 1982;
McMahon, 1982). By the late 1970’s Corbicula had been
found in most of Texas except the Brazos River (Britton and
Murphy, 1977). Fontanier (1982) later documented the ap-
pearance and spread of Corbicula in the Brazos River. The
most up-to-date distribution map for Corbicula in Texas can
be found in Britton (1982).

Two phenotypes of Corbicula have been found in
Texas waters: the white morph and the purple morph. Pro-
per taxonomic treatment of these two forms is unclear at this
time. Fontanier (1982) found both forms at the same locali-
ty, but these forms do not appear to be inter-breeding as
revealed by allozyme studies (Hillis and Patton, 1982).
Previous studies have revealed color variation in Corbicula
due to genetic, ontogenetic and environmental factors
(Sinclair and Isom, 1963). In this report, Asiatic clam popula-
tions in Texas would be referred to by generic name only.
All Corbicula observed in this study are referable to the white
morph.

Corbicula has been the focus of many studies due to
its economic importance when it blocks condensers of elec-
trical generating plants (Mattice, 1979). Other economic ef-

fects of Corbicula have been discussed (Sinclair and Isom,
1963). Corbicula is also significant as a food item for fish
(Minckley et at., 1970), utilization as a clarifying agent in
sewage treatment ponds (Dinges, 1976; Haines, 1979) and
as a bio-assay organism in pollution studies (Clark et al.,
1979). Corbicula has been implicated in the decline of native
freshwater mussels (Gardner et al., 1976) especially in heavily
managed waterways (Kraemer, 1979).

Apparently, no discussion of the relationship between
Corbicula and outdoor human recreational facilities has ap-
peared in the vast literature on Corbicula (see Mattice et al.,
1979). The outdoor recreation industry has grown steadily in
the past few decades, coincidentally or not, during the same
time period as the expansion of Corbicula. The purpose of
this study was to investigate the occurrence of Corbicula in
public recreational waters of Texas in order to determine
human impact upon this clam and any effects of the clams
upon humans. Concurrently, optimum microhabitats for Cor-
bicula in Texas could be characterized.

METHODS AND MATERIALS

For this survey, various bodies of water were sampled
from June 1978 to March 1983. Most sites were sampled
more than once. Particular effort was made to visit localities
during low flow and low lake level periods. Bottom sampling
was accomplished manually as well as utilizing an 8mm mesh
net and an Ekman dredge (utilized from bridges, piers and
boats). Water line accumulations and flood debris piles were

American Malacological Bulletin, Special Edition No. 2(1 986): 179-1 84

179

180

CORBICULA SYMPOSIUM

examined if no living Corbicula were found in initial surveys.
If shells were located in these latter sites, sampling for live
clams was continued. Peripheral localities were sampled
upstream and downstream or on adjacent lake shores as
appropriate.

Texas State Parks with water bodies were visited for
sampling. Several other public recreational areas were in-
vestigated and those which produced significant data con-
cerning Corbicula are discussed below. Particular attention
was given to areas with direct clam-human interaction. Notes
on microhabitat of viable populations of Corbicula were
recorded. Operational personnel of Texas Parks and Wildlife
Department were questioned regarding the impact of Cor-
bicula upon park operations. References to mud, sand,
gravel, etc. refer to standard geological size classifications.

RESULTS

STATE PARK SYSTEM SURVEY

The Texas State Park system presently consists of 1 1 6
units which are located from the Pineywoods Region of east
Texas westward to the Chihuahuan Desert of El Paso and
from the High Plains and canyons of the Panhandle
southward to the subtropical alluvial plain of the Rio Grande
delta region. Elevational limits range from sea level along the
Gulf of Mexico to a maximum of 2180 m in Franklin Moun-
tains State Park. Area of park units varies from 6638 hec-
tares for Palo Duro Canyon State Park in the Panhandle to
.0024 hectares for Action State Park.

A total of 92 units were judged to have surface water
present (2 parks were counted twice because of separate
water systems of different classifications). Notes on
microhabitats containing Corbicula were taken in order to
characterize the preferred habitat in Texas. Presence/
absence of Corbicula was related to type of freshwater habitat,
east Texas vs. west Texas, status of park (open vs. closed
to public) and type of park facilities.

Initially, the 94 park units with surface water were sub-
divided as to occurrence of fresh, brackish and salt water
(Table 1). Corbicula was absent from all brackish and
saltwater bodies and those freshwater bodies occurring in
parks which also have saltwater. The freshwater systems in

Table 1 . Relative occurrence of Corbicula in state park units of Texas
subdivided as to class of water present. Probability value is measure
of odds of obtaining observed distribution by chance.

Water Class

Corbicula

Present

Absent

Total

Freshwater only

Fresh and saltwater

Brackish water only

Saltwater only

Total

x2(3) = 10.64; p = .014

Table 2. Relative occurrence of Corbicula in state park units of Texas
subdivided as to type of freshwater present.

Corbicula

Water Class

Present

Absent

Total

Isolated

Enclosed lake

Reservoir

Stream

Total

x2(3) = 12.286; p = .006.

these coastal parks are generally small ponds which
periodically dessicate and have very limited inflow streams,
if any, which provide additional colonization routes. Of those
parks with only freshwater present, exactly 50 percent con-
tained populations of Corbicula.

The 86 parks with freshwater (including 4 also with
saltwater) were characterized as follows: 1) isolated water
system within a park, 2) small lake enclosed by the park, 3)
park on reservoir, and 4) units with a stream along the edge
or through the park (Table 2). Corbicula was found to be ab-
sent from parks with isolated water systems (ponds likely to
dessicate) and rare in parks with enclosed lakes (which
typically are fed by intermittent or small spring-fed streams).
Corbicula was found most often in reservoirs and streams
which have abundant colonization opportunities.

In order to determine geographical variations in Cor-
bicula occurrence, these same 86 parks were divided into
those in the eastern and western halves of Texas (Table 3).
The eastern parks receive more precipitation and are almost
all on the Coastal Plain. Western parks include a few coastal

Table 3. Relative occurrence of Corbicula in state park units of Texas
with freshwater subdivided into eastern and western Texas.

Corbicula

Location

Present

Absent

Total

Eastern

Western

Total

x2(i) = 1.325; p = .25

Table 4. Relative occurrence of Corbicula in state park units of Texas
with freshwater subdivided into open and closed parks.

Corbicula

Park

Present

Absent

Total

Open

Closed

Total

x2(i) = 0.56; p = .454

NECK: CORBICULA IN PUBLIC RECREATION WATERS

181

plain sites, but most units are in piedmont, hilly, upland plains
and mountainous areas. The eastern half of Texas exhibits
a greater dependability of water flow due to greater annual
average precipitation and more even distribution throughout
the year. No significant difference existed in percentage
presence of Corbicula between eastern and western parks.

Further analyses of Corbicula distribution patterns
were made to determine the correlation, if any, of public
utilization of water bodies and presence of Corbicula (Table
4). The 86 park units were first divided into sites open to the
public and those still closed pending planning and develop-
ment procedures. Corbicula was just as likely to be present
in closed parks as open parks.

The 66 open parks were characterized as to presence
of fishing as a recreational resource. Corbicula was more like-
ly to be present (slightly short of significance) in parks with
fishing activities than ones without such facilities (Table 5).
The 50 parks with fishing were then divided into those with
and without boat ramps (Table 6). The presence of a boat
ramp appeared to be irrelevant to the occurrence of Corbicula.

Table 5. Relative occurrence of Corbicula in open state park units
of Texas with freshwater subdivided into parks with and without
fishing facilities.

Corbicula

Facility

Present

Abent

Total

Fishing

No Fishing

Total

x2(-|) = 3.564; p -

: .06

Table 6. Relative occurrence of Corbicula in open state park units
of Texas with freshwater fishing facilities subdivided as to occurrence

of boat ramp.

Corbicula

Boat Ramp

Present

Absent

Total

Present

Absent

Total

x2(-|) = .002; p =

.963

ECOLOGICAL PREFERENCES

In Texas Corbicula is restricted to non-saline waters.
Corbicula is more abundant in waterbodies with a substan-
tial fraction of sand with moving water.

In larger reservoirs optimal microhabitat appears to be
a moderate-energy shore. Not only is the substrate likely to
be of a coarse nature, but the oxygen level is presumed high
and accumulation of metabolites is not likely. Corbicula has
been found in non-stagnant cat-tail marsh along the margins
of Cedar Creek Reservoir, Henderson County. At Twin Buttes

Reservoir (Concho River), near San Angelo, Corbicula was
abundant in sandy substrata that typify this reservoir.
However, Corbicula also occurs in gravelly mud and rock
riprap on the face of the dam where wave action was
moderate (Twin Buttes Reservoir; Brady Reservoir on Brady
Creek). In the Rio Grande immediately below Anzalduas Dam,
Corbicula is found among and under small boulders (20-25
cm diameter) present in midstream areas and in sand banks
along the shore.

Reservoirs with mud bottoms are not conducive to
development of dense populations of Corbicula. Dead shells
were found uncommonly in Lewisville Lake, Denton County;
these shells are believed to be from low density populations
in flowing creeks which drain into the reservoir. However, Cor-
bicula was abundant in small-gravel substratum in the stream
(Elm Fork Trinity River) immediately below the dam.

One clearly defined optimal microhabitat in the Cen-
tral Texas Hill Country is among gravel (up to 100 mm in
length) scattered in shallow pools below riffles (Fig. 1).
Occurrence of Corbicula in such areas was observed
at Guadalupe River State Park and South Llano State Park.
Such substrates occur at depths between 10 cm and 1 .5 m.
Corbicula occurred on a packed clay/rock bottom at the same
level as larger non-attached gravel rocks, but these clams
had not burrowed into the substratum. Clams are not buf-
feted because they are within the boundary layer created by
the larger gravel rocks where the current pressure is minimal.
Limited number of small clams (6-8 mm) are present. Cor-
bicula is found among small gravel in potholes of the
limestone bottom of Onion Creek, McKinney Falls State Park.

River reaches below dams are also “favored” areas.
Tremendous bars composed of live Corbicula are exposed
during the fall months in the Colorado River in Travis Coun-
ty. Warm season flows are maintained because of demands
for irrigation water downstream rice farmers. When demand
for irrigation water diminishes in autumn, water flow is greatly
reduced unless heavy rainfall occurs upstream. Large ex-
panses of riverbottom consisting of hydraulic accumulations
(see Eng, 1979) of Corbicula are exposed. Death occurs
quickly with smaller clams expiring before the larger clams.

One of the major causes of mortality in Corbicula in

Fig. 1. Occurrence of Corbicula in gravel-bottomed pools of rivers
in Hill Country of central Texas. Size of arrows directly correlated
with relative water velocity.

182

CORBICULA SYMPOSIUM

many areas of Texas is alteration in water depth and current,
both increased and decreased, due to natural and anthropo-
genic causes. Numerous clams were present in pools
below riffles in Guadalupe River State Park, Comal Co., on
23 July 1 978. A torrential flood with river rises of 1 5-1 8 m on
1-2 August resulted in massive flood damage (Schroeder et
al., 1979). Examination of these pools on 9 August 1978
revealed no living Corbicula. Subsequently, Corbicula has
recolonized these areas. Corbicula in the sand banks below
Anzalduas Dam, particularly the smaller ones, die from
overheating following water level fall, even if they can reach
water-saturated sand below the surface.

CLAM-HUMAN INTERACTIONS

The effect of human activities upon Corbicula was
observed at Huntsville State Park, in eastern Texas in
December 1 977. Shells were present in most parts of the lake,
but articulated shells were much less common in the swim-
ming area (4.9 pairs/sq. m.) than in a nearby, otherwise
similar, portion of shallow lake (25.2 pairs/sq. m.).

The distribution of Corbicula in an artificial impound-
ment (Lake Long) in Austin was investigated. A small area
on Lake Long has been developed for picnic sites and swim-
ming. Natural bottom of this lake is black clay, but a small
area has been enhanced for swimming by placement of sand.
Examination of the clay bottom portion (40 m beach length
x 10 m distance from beach) of the lake edge revealed no
Corbicula , while the sandy beach area supported a thriving
population. A more intensive survey utilizing detailed hand
sampling was conducted on 31 July 1981. Numbers of clams
per square meter were recorded at intervals of 0.2 m depth
(Fig. 2). Very small clams (2-5 mm) were found only under
and attached to small gravel (up to 8 cm length). An average
of 1.1 young clams were found attached to each piece of
gravel (22 rocks per square meter).

Only seven state parks obtain water directly from sur-
face sources. Of these seven parks, six contain Corbicula in
the relevant body of water; the seventh park is located in the

WATER DEPTH CM]

Fig. 2. Depth distribution of size classes of Corbicula at Lake Long,
Austin, Texas. Numbers of data points indicate number of Corbicula
per square meter.

Panhandle in the large area of western Texas in which Cor-
bicula has yet to be discovered. None of the six parks have
reported any problems associated with Corbicula. A typical
water procurement system utilizes a 3-inch intake pipe with
a one-eighth inch mesh screen. Water is pumped into a set-
tling basin where it is chlorinated, filtered and rechlorinated.

DISCUSSION

This survey revealed that Corbicula was present in less
than half the freshwater sites surveyed in the Texas Park
system. Several parks which contain aquatic habitats suitable
for Corbicula revealed no individuals. Some of these “absent”
results may be due to sampling error at sites where individuals
of Corbicula were too sparse or localized to detect. However,
some of the “absent” results may reflect the highly dynamic
nature of Corbicula populations in Texas. Following attain-
ment of high densities, Corbicula populations often crash;
some of these population declines may progress to local ex-
tirpation. The absence of Corbicula from irrigation water
storage reservoirs at long distances from the Rio Grande
coupled with the occurrence of populations in reservoirs close
to the Rio Grande indicates a dispersal distance limit in silt-
filled irrigation canals (Neck and Metcalf, in press). Corbicula
shells from eastern Texas tend to be smaller than those from
other portions of Texas with alkaline waters (indicating a
shorter life span).

Unopened parks tend to be environmentally similar to
those open parks which support populations of Corbicula.
New parks tend to be on large reservoirs or streams with 13
of the 20 new sites located in eastern Texas. Only two of these
new sites have an isolated system or an enclosed lake. Ad-
ditionally, many of these new parks are located on lakes or
streams which have other points of public access.

Corbicula is not a problem in park water systems
because few ever arrive in the settling basin, and these in-
dividuals may be killed by the initial chlorination treatment.
Sinclair and Isom (1963) discussed the success of chlorina-
tion in controlling Corbicula in an industrial water supply.
Ingram (1959) recounted problems with Corbicula in
underground canals and pumping stations in agricultural and
municipal water supply systems, but did not report any oc-
currences in municipal water distribution systems. Such oc-
currences are known for aquatic snails and other species of
clams as reviewed by Ingram (1956).

Certain operational techniques in various state parks
may impact Corbicula populations. The water level of Lake
Raven in Huntsville State Park is lowered each winter to con-
trol aquatic vegetation with chemical sprays. The winter
drawdown of Lake Raven also functions as a management
technique to control Corbicula , but this effect is entirely for-
tuitous. Such a lowering (in the cooler months) will cause mor-
tality of most individuals in the shallow margins of the lake.
The susceptibility of Corbicula to water level declines results
from reduced migratory abilities (White, 1979) and intolerance
to aerial exposure (McMahon, 1979).

There are three major microhabitat classifications
which appear to be optimal for Corbicula in Texas. Areas of

NECK: CORBICULA IN PUBLIC RECREATION WATERS

183

sandy or rock-bottomed streams of intermediate flow probably
represent the “best” habitat for Corbicula. Greater numbers
of Corbicula on sand than silt was observed in Florida pop-
ulations (Gottfried and Osborne, 1982). Another prime
microhabitat occurs among loose gravel substrata in
shallow pools between riffles in streams of the Texas Hill
Country in the central part of the state west of the Bal-
cones Fault Zone (Fig. 1). In such locations, Corbicula is
actually an epifaunal bivalve. The third favored microhabitat
is a moderate energy lakeshore where wave action is suffi-
cient to remove most or all silt and clay particles, but not
strong enough to allow frequent disturbance of the
substratum. These three classification types must contain
non-saline, relatively unpolluted water in the warmer portion
of the state.

The absence of Corbicula from brackish and saltwater
habitats probably is due to physiological stress (Gainey,
1978a;b) which appears to be inversely proportional to period
of acclimation (Evans et ai. , 1979). Fontanier (1982) reported
Corbicula in the Brazos River no further downstream than
Farm Road 1462, Brazoria County. Absence of Corbicula from
coastal rivers under tidal influence has been noted in
Mississippi (Hartfield and Cooper, 1983). However, Corbicula
occurs in tidal portions of the Potomac River, Maryland
(Dresler and Cory, 1980). Valves of Corbicula found on Gulf
beaches in Texas (O’Kane et ai , 1977) undoubtedly repre-
sent river drift material that has been redeposited in the surf
zone. Britton (1982) was unaware of any records of live Cor-
bicula from Texas tidewaters.

Absence of Corbicula from areas of Texas which ex-
perience the most severe winter weather may not be totally
due to temperature effects, although Corbicula is not tolerant
of long-term subfreezing conditions (Horning and Keup,
1964). The Panhandle and western Texas lack permanent
streams due to reduced precipitation levels. Chances for in-
troduction may be somewhat reduced, but the absence of
Corbicula in Lake Theo (Caprock Canyons State Park, Briscoe
Co.), contrasts with the occurrence of introduced populations
of the bullfrog, Rana catesbeiana (Neck, 1980) and the ex-
tralimital unionid, Anodonta grandis (Neck, 1982), in this lake.

Since size of an individual indicates length of growth
period and/or rate of growth, size of Corbicula valves also
indicates suitability of habitat. The largest valves recovered
during this study were 52.9 mm long and were from Llano
Grande Lake, Hidalgo County. Britton (1982) reported a
specimen from Benbrook Lake with a length of 60 mm.

Environmental reasons for the size-class/depth pat-
terns observed at Lake Long are manifold. While the
shallower water has a sandy bottom as opposed to a more
mixed sandy ciay substratum at greater depths, an area of
shallow water with a similar mixed sand/clay substratum
revealed no clams. Human activity patterns in the beach area
both inadvertently disturb and deliberately remove clams.
Clams in shallow water are continually buried by trampling
action of humans and a substantial number are removed, par-
ticularly by adolescent humans. Substratum is significant in
that the gravel rocks are important, and possibly essential,
for survival of recently metamorphosed Corbicula in this area.

ACKNOWLEDGEMENTS

I thank C. J. Adair, D. W. Buchanan and D. C. Dewitt for in-
formation concerning park water systems. T. B. Samsell III drafted

the figures. Two anonymous reviewers made suggestions that greatly

improved the manuscript.

LITERATURE CITED

Britton, J. C. 1982. Biogeography and ecology of the Asiatic clam,
Corbicula, in Texas, p. 21-31 IN: Proceedings of the Symposium
on Recent Benthological Investigations in Texas and adjacent
States, J. R. Davis, ed., Texas Academy of Science, Austin,
278 p.

Britton, J. C. and C. E. Murphy. 1977. New records and ecological
notes for Corbicula manilensis in Texas. The Nautilus.
91:20-23.

Burch, J. Q. 1944. Checklist of West North American Marine Mollusca
from San Diego, Calif., to Alaska. Minutes of the Conchological
Club of Southern California. 38:11-18.

Clark, J. H., A. N. Clark, D. J. Wilson and J. J. Friauf. 1979. On the
use of Corbicula fluminea as indicators of heavy metal con-
tamination. pp. 153-163, IN: Proceedings, First International
bicula Symposium, J. C. Britton, ed., Texas Christian Univer-
sity Research Foundation, Fort Worth.

Counts, C. L. III. 1981. Corbicula fluminea (Bivalvia: Sphaeriacea)
in British Columbia. The Nautilus. 95:12-13.

Dinges, R. 1976. A proposed integrated biological wastewater treat-
ment system. P. 225-230, in Biological Control of Water Pollu-
tion. J. Gourbier and R. W. Pierson, Jr., ed., U. Pennsylvania
Press Philadelphia, 340 p.

Dresler, P. V. and R. L. Cory. 1980. The Asiatic clam, Corbicula
fluminea (Muller), in the tidal Potomac River, Maryland.
Estuaries. 3:150-151.

Eng, L. L. 1979. Population dynamics of the Asiatic clam, Corbicula
fluminea (Muller), in the concrete-lined Delta-Mendota Canal
of central California. Pp. 39-68, in Proceedings, First Interna-
tional Corbicula Symposium. J. C. Britton, ed., Texas Chris-
tian University Research Foundation, Fort Worth.

Evans, L. P. Jr., C. E. Murphy, J. C. Britton and L. W. Newland. 1979.
Salinity relationship in Corbicula fluminea (Muller, 1974). Pp.
193-214, IN: Proceedings, First International Corbicula Sym-
posium. J. C. Britton, ed., Texas Christian University Research
Foundation, Fort Worth.

Fontanier, C. E. 1982. The distribution of Corbicula (Bivalvia: Cor-
biculidae) in the Brazos River system, Texas, 25 August - 12
November 1980. Texas Journal of Science. 34:5-15.

Gainey, L. F. Jr. 1978a. The response of the Corbiculidae
(Mollusca:Bivalvia) to osmotic stress: The organismal
response. Physiological Zoology. 51:68-78.

Gainey, L. F. Jr., 1978b. The response of the Corbiculidae
(Mollusca:Bivalvia) to osmotic stress: The cellular response.
Physiological Zoology. 51:79-91.

Gardner, J. A. Jr., W. R. Woodall, Jr., A. A. Staats, Jr. and J. F.
Napoli. 1976. The invasion of the Asiatic clam ( Corbicula
manilensis Philippi) in the Altamaha River, Georgia. The
Nautilus. 90:117-125.

Gottfried, P. K. and J. A. Osborne. 1982. Distribution, abundance,
and size of Corbicula manilensis (Philippi) in a spring-fed cen-
tral Florida stream. Florida Scientist. 45:178-188.

Haines, K. C. 1979. The use of Corbicula as a clarifying agent in ex-
perimental tertiary sewage treatment process on St. Croix,
U.S. Virgin Island. Pp. 165-175, in Proceedings, First Interna-
tional Corbicula Symposium. J. C. Britton, ed., Texas Chris-
tian University Research Foundation, Fort Worth.

184

CORBICULA SYMPOSIUM

Hartfield, P. D. and C. M. Cooper. 1983. Distribution of Corbicula
fluminea, the Asiatic clam, in Mississippi. The Nautilus.
97:66-68.

Hillis, D. M. and J. C. Patton. 1982. Morphological and electrophoretic
evidence for two species of Corbicula (Bivalvia: Corbiculidae)
in North America. American Midland Naturalist. 108:74-80.

Horning, W. B. and L. Keup. 1964. Decline of Asiatic clam ( Corbicula
fluminea) in Ohio River. The Nautilus. 78:29-30.

Ingram, W. M. 1956. Snail and clam infestations of drinking-water
supplies. Journal of American Waterworks Association.
48:258-268.

Ingram, W. M. 1959. Asiatic clams as potential pests in California
water supplies. Journal of American Waterworks Association.
51:363-370.

Kraemer, L. R. 1979. Corbicula (Bivalvia:Sphaeriacea) vs indigenous
mussels (Bivalvia:Unionocea) in U.S. rivers: A hard case for
inter-specific competition? American Zoologist. 19:1085-1096.

Mattice, J. C. 1979. Interactions of Corbicula sp. with power plants,
pp. 119-138, IN: Proceedings, First International Corbicula Sym-
posium. J. C. Britton, ed., Texas Christian University Research
Foundation, Fort Worth.

Mattice, J. C., L. L. Eng and B. N. Collier. 1979. Corbicula 1979: A
bibliography, pp. 289-313, IN: Proceedings, First International
Corbicula Symposium. J. C. Britton, ed., Texas Christian
University Research Foundation, Fort Worth.

McMahon, R. F. 1979. Tolerance of aerial exposure in the Asiatic
freshwater clam, Corbicula fluminea (Muller), pp. 227-241 , IN:
Proceedings, First International Corbicula Symposium. J. C.
Britton, ed., Texas Christian University Research Foundation,
Fort Worth.

McMahon, R. F. 1982. The occurrence and spread of the introduced
Asiatic freshwater clam, Corbicula fluminea (Muller), in

North America: 1924-1982. The Nautilus. 96:134-141.

Metcalf, A. L. 1966. Corbicula manilensis in the Mesilla Valley of Texas
and New Mexico. The Nautilus. 80:16-20.

Minckley, W. L., J. E. Johnson, J. N. Rinne and S. E. Willoughby.
1970. Foods of buffalofishes, genus Ictobius, in central Arizona
reservoirs. Transactions of the American Fisheries Society.
99:333-342.

Murray, H. D. 1971. New record of Corbicula manilensis (Philippi)
in Texas. The Nautilus. 58:35-36.

Neck, R. W. 1980. Geographical distribution. Rana catesbeiana
(Bullfrog). Herpetological Review. 11:38.

Neck, R. W. 1982. Occurrence of Anodonta grandis (Say) in Lake
Theo, Briscoe Co., Texas. Texas Conchologist. 18:49-52.

Neck, R. W. and A. L. Metcalf. In press. Freshwater mussels
(Unionacea) of the Lower Rio Grande. Sterkiana.

O’Kane, K. D., J. C. Britton and O. R. Coldiron. 1977. New distribu-
tional records of Corbicula manilensis (Pelecypoda:Cor-
biculidae) in the south central United States. Southwestern
Naturalist. 22:397-400.

Schroeder, E. E., B. C. Massey and K. M. Waddell. 1979. Floods
in central Texas, August 1978. United States Geological Survey
Open-File Report. 79-682: 1-121.

Sinclair, R. M. and B. G. Isom. 1963. A preliminary Report on the
Introduced Asiatic Clam Corbicula in Tennessee. Tennessee
Stream Pollution Control Board, Tennessee Department of
Public Health, Nashville, 31 p.

White, D. S. 1979. The effects of lake-level fluctuations on Corbicula
and other pelecypods in Lake Texoma, Texas and Oklahoma,
pp. 81-88, IN: Proceedings, First International Corbicula Sym-
posium. J. C. Britton, ed., Texas Christian University Research
Foundation, Fort Worth.

THE ASIATIC CLAM IN LAKE ERIE

JENNIFER SCOTT-WASILK
JEFFREY S. LIETZOW
GARY G. DOWNING
and

KELLY L. (CLAYTON) NASH

THE TOLEDO EDISON COMPANY
300 MADISON AVENUE
TOLEDO, OHIO 43652, U.S.A.

In 1 981 and 1 982, the thermal plume areas of four power plants along the southern
shore of Lake Erie were sampled for Corbicula fluminea. Corbicula were found in only two
of those four locations: the thermal plumes of the Toledo Edison Acme and Bay Shore
Generating Stations. Both power plants are coal-fired with once-through condenser cool-
ing systems. Acme is located in Toledo on the Maumee River. Bay Shore is located east
of Toledo on the southern shore of Maumee Bay.

The two power plants at which no Corbicula were found are the Toledo Edison Davis-
Besse Nuclear Power Station located near the mouth of the Toussaint River and the
Cleveland Electric Illuminating Company Eastlake Power Plant on the Central Basin east
of Cleveland. Davis-Besse has a closed-cycle natural draft cooling tower, and hence no
significant thermal plume. Eastlake has a once-through condenser cooling system and a
large thermal plume.

In 1982, specimen length varied from 4 mm to 35 mm. The majority of the specimens
were collected from the Bay Shore thermal plume. The substrates in which the clams were
found were predominately clay and muck.

The 1981 and 1982 findings were consistent in indicating that Corbicula have not
spread beyond the confines of the thermal plumes of the Acme and Bay Shore Generating
Stations.

American Malacological Bulletin, Special Edition No. 2(1986): 185

185

BIOLOGICAL BASIS OF BEHAVIOR IN CORBICULA FLUMINEA, I
FUNCTIONAL MORPHOLOGY OF SOME TROPHIC ACTIVITIES

LOUISE RUSSERT KRAEMER
UNIVERSITY OF ARKANSAS
FAYETTEVILLE, ARKANSAS 72701 .U.S.A.

ABSTRACT

Understanding the functional morphology of trophic activity of Corbicula fluminea (Muller) pro-
vides a useful basis upon which to design appropriate control protocols for the clams. Accordingly,
this paper reports results of pertinent research by the author. Characteristic, rapid locomotion is ac-
counted for at least in part by (1) the unusual (for a freshwater bivalved mollusk) structural autonomy
of the adductor muscles and the “suturing” of the mantle lobes so as to provide a pallial foramen
for those muscles; and (2) the recently discovered, conjoined statocysts near the pedal ganglion. Agile
locomotion of juvenile clams is produced by (1) precocious differentiation of the statocysts; (2) well
developed retractor muscles; and (3) telescoping “laminae” of the juvenile foot, all recently interpreted
with videotaping and scanning electron microscopy (SEM). Putative sensory cilia discovered on the
lip of the excurrent siphon help account for the extreme sensitivity of that tissue to mechanical stimuli.
Location and interpretation of the paired sense organs, the osphradia, on the ventral surface of the
visceral ganglion above the dorsal shelf of the excurrent chamber, indicate function different from
that of gastropod osphradia, perhaps a light sensor function— and certainly needing further study.

As a consequence of the First Corbicula Symposium
in 1977 (Britton, ed., 1979), consensus was reached that Cor-
bicula fluminea Muller is hermaphroditic (Britton and Morton,
1979; Kraemer, 1979a), and that its young are shed into the
environment primarily as juveniles (Kraemer, 1979a). It was
also reported there that the juveniles develop a byssal thread
that is used as an anchor to the substratum, and is thereby
associated with rapid downstream disbursement and local
establishment of the young clams (Kraemer, 1979a). In the
literature reviewed for the preparation of this paper little fur-
ther work on the functional morphology of C. fluminea other
than that by the present author was to be found (Britton and
Morton, 1982). During the interim since 1977 some in-
vestigators have been concerned with careful analysis of the
taxonomic position of C. fluminea (e.g. McCleod, 1983), with
life history and distribution of C. fluminea (Counts, 1981 ,1983;
McMahon, 1982; Hall, 1983), with some physiological traits
of C. fluminea (McMahon, 1982; McCleod, 1983) and with
diagnostic shell microarchitecture (Counts and Prezant, 1979;
Prezant and Chalermwat, 1983).

Primary focus in the Second International Corbicula
Symposium, held in Little Rock, Arkansas, in 1983, was on
the currently researched level of understanding of C. fluminea
as a serious macrofouling organism in U.S. rivers, and on
presently available technical means for bringing the clams
under control. In that context it is appropriate to review results
of research since 1 977 which further elucidate the biological
characteristics of C. fluminea. With thorough biological evalua-
tion, technical protocols for control of the clams can be

rigorously evaluated and the future role of C. fluminea can
be assessed.

In this paper results of recent studies on the functional
morphology of some trophic activities of C. fluminea (e.g.
locomotion, siphoning) are reviewed and evaluated as a basis
for estimating the likely efficacy of control procedures for the

clams.

MATERIALS AND METHODS

Three-dimensional visualization of the microarchitec-
ture of tissues and organs involved in trophic activities of C.
fluminea was done by means of analysis of several thousand
serial sections of whole clams. Ultrastructure study of some
structures (e.g. motor and sensory cilia) was done with scan-
ning electron microscopy (SEM). Developmental sequencing
and the function of certain embryonic structures (e.g. the
juvenile foot) was determined with the aid of a Panasonic
Videocamera attached to an AO-Microstar compound
microscope. Additional details regarding materials and
methods used are included in the subsequent article
(Kraemer et al., 1985).

RESULTS

FUNCTIONAL MORPHOLOGY OF LOCOMOTION

Many workers have observed the remarkable rapidity
with which C. fluminea spreads through great expanses of

American Malacologies! Bulletin, Special Edition No. 2(1986): 187-191

187

188

CORBICULA SYMPOSIUM

river bottoms in the United States. It was found, for exam-
ple, that C. fluminea had not only spread through more than
240 miles of the Arkansas River in less than 10 years, but
that it had become by far the most abundant species in the
benthos (Kraemer, 1 975, 1 977). It seemed evident that many
juvenile clams were transported downstream, perhaps at-
tached to sand grains by means of their byssal thread.
Recently Prezant and Chalermwat (1984) have presented
evidence from which they argue that C. fluminea may achieve
downstream distribution by drifting attached to mucus
strands. The very active movements of C. fluminea may ac-
count for some aspects of its distribution patterns, especial-

Fig. 1 a. Drawing of posterior region of C. fluminea from a specimen
that was relaxed in Nembutal. The left shell valve was removed and
the left mantle lobe was reflected to show underlying tissues, such
as outer gill. The fusion of the mantle lobes not only reduces the
pedal gape, but also creates a foramen around the posterior adduc-
tor muscle, etc. Horizontal field width = 33 mm. b. Photomicrograph
of posterior region of relaxed specimen that was preserved and
removed from the shell valves. The posterior adductor muscle has
fallen out, leaving a foramen clearly visible in the mantle lobes.
Horizontal field width = 6.5 mm. CT, cardinal tooth; MF, mantle
foramen; OG, outer gill; PA, posterior adductor muscle.

ly upstream. Unlike the indigenous river mussels (Unionacea)
that exhibit slow, ponderous foot movement, C. fluminea has
a foot regularly engaged in rapid backward and forward, and
side-to-side movement. Locomotion by a large mussel is
seldom more than 20 cm/hr, while the much smaller C.
fluminea has been clocked at up to 250 cm/hr (Kraemer,
1977). Pyramidal shape of the shell valves of the clam and
the unhampered movement of the adductor muscles have
been shown to allow for more autonomous movement of the
shell valves than is possbile for mussels.

In dissecting carefully preserved specimens of C.
fluminea it was repeatedly observed that adductor muscles
of the clam and especially the posterior adductor muscle,
would simply fall out of the mantle when the muscle was
separated from shell valves. A hole or “pallial foramen”
thereby became visible (Fig. 1a,b). Frontal sections made from
this region of the clam reveal a peculiar histological “sutur-
ing” of the right and left mantle lobes in the region of the
pallial foramen (see Kraemer, 1977).

Corbicula fluminea, unlike any other bivalve similary
studied, has a pair of statocysts (putative balance organs)
located just above the pedal ganglion in the midventral por-
tion of the visceral mass, and which are conjoined by a
slender, hollow tube (Fig. 2a) (Kraemer, 1978). In contrast,
the slow-moving, indigenous mussels have a pair of
statocysts, each of which is slung by its own statocyst nerve
from the cislateral cerebral ganglion, and each is thus wide-
ly separated from the other (Kraemer, 1978, 1984). It seems
likely that the conjoined statocysts provide a neurological
basis for the clam’s rapid foot movements. While these are
the first conjoined statocysts to be found for any bivalved
mollusk, it seems probable that such statocysts will also be
found in other bivalves exhibiting similar locomotion.

Juvenile locomotion. Recent studies have revealed that
juvenile G. fluminea have well differentiated, disproportionate-
ly large statocysts, even though the young clams measure
less than 200 nm in length at this stage. Further, it has been
found (Kraemer, 1984; Kraemer and Galloway, in press) that
the juvenile foot has accordion-like laminae along its length
which facilitate rapid, agile movement, allowing the young
clam to crawl under surface water film or over minute bits
of detritus suspended in the water (Fig. 2b, c). These findings
provide a morphological basis for repeated observations
(West, personal communication) of young C. fluminea
“floating” in the water column.

SIPHON MOVEMENTS

Siphons of several species of Corbicula have been
found to be taxonomically distinctive (Britton and Morton,
1979,1982). Siphons of C. fluminea , like those of many of its
indigenous relatives, the freshwater pill clams (Pisidium) and
fingernail clams (Sphaerium), are slender, muscular, fused
tubes. The siphons are extremely mobile, and are not par-
ticularly sensitive to light (as are the siphons of the mussels,
Unionacea). The siphons are especially responsive to tac-
tile stimuli or to movements of the water column, however.
Recent studies using scanning electron microscopy (SEM)

KRAEMER: CORBICULA FUNCTIONAL MORPHOLOGY

189

Fig. 2 a. Photomicrograph of cross-section of C. fluminea in region
of pedal ganglion, showing unusual conjoined statocysts above the
pedal ganglion, (from Kraemer, 1978) Horizontal field width = 3.2
mm. b. Scanning electron micrograph of juvenile C. fluminea taken
from ovisac of marsupial gill in adult specimen. Note conspicous
laminae of the large foot of the young clam. Horizontal field width
= 225 fim. c. Scanning electron micrograph showing detail of distal
end of foot in juvenile clam. Note the conspicuous laminae and the
apical cilia. Horizontal field width = 73 /tm. AC. apical cilia; JF,
juvenile foot; L, lamina; M, mantle; SL, statolith; ST, statocyst; T,
hollow tube which joins the statocysts.

made possible the comparative study of several kinds of ef-
fector cilia on the gills, gonopore lips and labial palps of C.
fluminea. It was thus possible to evaluate newly discovered,
minute clumps of cilia on the distal lips of the excurrent siphon
of C. fluminea, as putative sensory cilia (Kraemer, 1 983). The
latter may very well constitute mechanoreceptors which ac-
count for the extreme tactile sensitivity of the siphonal lips
(Fig. 3). Other authors seem to have found similar organelles
with comparable function in some echinoderms (Whitfield
and Emson, 1983).

OSPHRADIA: CHEMICAL OR PHYSICAL SENSORS IN C.
FLUMINEA ?

The osphradium occurs in the roof of the incurrent
siphon of gastropod mollusks, where its function as a
chemoreceptor or mechanoreceptor has been experimentally
implicated. However, previous studies of (marine and
estuarine) bivalve osphradia have been inconclusive as to
location, orientation or function of the organ. Analysis of
transverse and sagittal serial sections of C. fluminea reveal
that the organ is paired, and that it is adjacent to the ventral
surface of the large visceral ganglion. The osphradium of C.
fluminea is extensively innervated by neuronal fibers from
the dorsally situated, visceral ganglion. Ventrally, many
delicate, unmyelinated fibers of the osphradia innervate a
patch of modified epithelium on the roof of the excurrent canal
(see Kraemer, 1981).

Recent studies of mollusk osphradia ultrastructure
(Haszprunar, 1983) indicate that: (1) findings of the microar-
chitecture of osphradia in C. fluminea appear to be within the
norm for bivalves; and (2) that function of these well differen-
tiated sense organs in C. fluminea and in bivalves as a whole,
is much in need of further study. It has been suggested
(Kraemer, 1981) that their innervation, microarchitecture and
location may even indicate a pineal body-like, light sensor
function for the osphradia of C. fluminea.

SUMMARY AND DISCUSSION

Functional morphological studies of C. fluminea reveal
that: (1) Rapid locomotion of the clams is aided by suturing
of the mantle lobes around the adductor (especially posterior)
muscles, which allows for more effective adduction of the
heavy shell valves. (2) Locomotion is further aided by con-
joined statocysts, the first such statocysts found in any
mollusk species. Organization of the statocysts is peculiarly
and necessarily implicated in their back-and-forth, side-to-side
foot movements. (3) Recently discovered horizontal laminae
of the foot of juvenile clams facilitate the agile, rapid exten-
sions, turnings and withdrawals of the foot in young clams.
(4) Videomicroscopy has recently allowed clear visualization
of the proportionately large, well differentiated statocysts of
juvenile C. fluminea and shows that the statocysts may func-
tion as sensors implicated in the complex movements of the
clams. (5) The mobile, fused siphons of C. fluminea are
equipped with patches of recently discovered, putative sen-
sory cilia, which seem to provide the morphological basis for

190

CORBICULA SYMPOSIUM

Fig. 3 a. Scanning electron micrograph of siphons of C. fluminea. Arrow indicates lip of excurrent siphon upon which the very small ciliary
tufts shown in b were seen. Horizontal field width = 9.4 mm. b. Scanning electron micrograph of surface of lip of excurrent siphon, showing
isolated ciliary tufts of putative sensory function, (after Kraemer, 1981). Horizontal field width = 55 /urn. ES, excurrent siphon; C, ciliary tuft;
IS, incurrent siphon.

the tactile sensitivity of the siphonal edges. (6) Histological
and neuroanatomical details of the osphradia of C. fluminea
offer a new view of possible physical or chemical sensory
function for these sense organs, not only in C. fluminea but
in other bivalves as well.

It seems plausible to argue that the foregoing infor-
mation on the functional morphological basis of certain trophic
activities of C. fluminea provides useful background for
evaluating the C. fluminea populations in U.S. rivers, and the
available means for their control. An immediate effect of the
rapid spread and establishment of a great biomass of C.
fluminea in managed U.S. rivers called attention of in-
vestigators to the apparent replacement of the indigenous
freshwater mussel fauna in those stream bottoms by an “in-
vasive” species. How had the swift faunal exchange taken
place? Decline of the great U.S. mussel fauna has been con-
sidered at length (Clarke, 1970).

Was C. fluminea “taking over” the river bottoms as
a superior competitor? It was argued that ecological “crunch”
in those rivers created a far different interspecific contact than
that due to true competition (Wiens, 1977). Evidence ac-
cumulated that C. fluminea was far more apt to attain a large
biomass in benthos of “managed” rivers (e.g. dredged,
dammed) than in less disturbed streams. In the latter sites,
Asian clams could live along with mussels, but without the
evident success of the mussels (Kraemer, 1979b). It was
argued further that the great size range of C. fluminea ,
unusual for a freshwater benthic species, allows it: (1) to
establish large populations of small, reproductively active
clams, that equal the size of freshwater gastropods, insects

and many crustaceans, and (2) to establish populations of
large (reproductively active) benthic animals which approach
the size of freshwater mussels (Kraemer and Gordon, 1980).
These authors suggested that C. fluminea may be meeting
criteria for “the size range of success” in establishing its
significant presence in U.S. rivers (Kraemer and Gordon,
1980).

Why have the indigenous relatives of C. fluminea, the
pill clams (Pisidium) and the fingernail clams (e.g. Sphaerium ),
not exploited damaged river bottom habitat with runaway
biomass as C. fluminea has? C. fluminea certainly has a much
heavier shell, a longer life span and a much greater size
range. Important cues to the effectiveness of the trophic ac-
tivities of C. fluminea are cited in functional morphological
characteristics reported in this paper. Of equal or greater
importance, however, are comparable considerations of the
functional morphology of reproduction and development of
C. fluminea which are analyzed in the paper following this
one (Kraemer et al., 1985).

ACKNOWLEDGEMENT

Funding from Arkansas Power and Light Company supported
part of this study. Thanks are also due to several anonymous
reviewers of the manuscript. I should further like to thank Charles
M. Swanson and Marvin L. Galloway for their skillful assistance.

LITERATURE CITED

Britton, J. C. (ed.) 1979. Proceedings of the First International

KRAEMER: CORBICULA FUNCTIONAL MORPHOLOGY

191

Corbicula Symposium. Texas Christian University, Fort Worth,
Texas. October 13-15, 1977.

Britton, J. C. and B. Morton. 1979. Corbicula in North America: the
evidence reviewed and evaluated. IN: J. C. Britton, ed., Pro-
ceedings of the First International Corbicula Symposium. Texas
Christian University, Fort Worth, Texas. October, 1977:
249-287.

Britton, J. C. and B. Morton. 1982. A Dissection Guide, Field and
Laboratory Manual for the Introduced Bivalve, Corbicula
fluminea. Malacological Review, Supplement 3: 1-82.

Clarke, A. H. (Ed.) 1970. Proceedings of a Symposium on rare and
endangered mollusks. Malacologia 10(1): 7-56.

Counts, C. L., III. 1981. Corbicula fluminea (Bivalvia: Sphaeriacea)
in British Columbia. Nautilus 95(1): 12-13.

Counts, C. L., III. 1983. Chronology of the invasion of North America
by Corbicula fluminea (Bivalvia: Corbiculidae). American
Malacological Bulletin 1: 100.

Counts, C. L., Ill and R. S. Prezant. 1979. Shell structure and
histochemistry of the mantle of Corbicula leana Prime, 1864.
American Zoologist 19(3): 1007.

Hall, J. L. 1983. The life history and production of immature Cor-
bicula fluminea (Mollusca: Bivalvia) in Lake Norman, North
Carolina. American Malacological Bulletin 1: 96.

Haszprunar, G. 1983. Comparative ultrastructural investigations of
the osphradial sense organs of Gastropoda: its bearing on
ecology, systematics and function. Abstracts of the Eighth In-
ternational Malacological Congress, Budapest, Hungary,
August, 1983: 51.

Kraemer, L. Russert. 1976. An evaluation of the effects of dredging
within the Arkansas River Navigation System, IV. The effects
upon the benthic associations. Arkansas WRRC Publication
No. 46: 1-357.

Kraemer, L. Russert. 1977. Aspects of the functional morphology
of the mantle/shell and mantle/gill complex of Corbicula
(Bivalvia: Sphaeriacea: Corbiculidae). Bulletin of the American
Malacological Union For 1977: 25-31.

Kraemer, L. Russert. 1978. Discovery of two distinct kinds of
statocysts in freshwater bivalved mollusks: some behavioral
implications. Bulletin of the American Malacological Union For
1978: 24-28.

Kraemer, L. Russert. 1979a. Juvenile Corbicula: their distribution in
the Arkansas River benthos. IN: J. C. Britton, ed., Proceedings
of the First International Corbicula Symposium. Texas Chris-
tian University, Fort Worth Texas. October, 1977: 89-97.

Kraemer, L. Russert. 1979b. Corbicula (Bivalvia: Sphaeriacea) vs.
indigenous mussels (Bivalvia: Unionacea) in U.S. rivers: a hard
case for interspecific competition? American Zoologist 19:

1085-1096.

Kraemer, L. Russert. 1981 . The osphradial complex of two freshwater
bivalves: histological evaluation and functional context.
Malacologia 20(2): 205-216.

Kraemer, L. Russert. 1983. Comparative functional morphology of
cilia of Corbicula (Bivalvia: Corbiculidae): possible criteria for
effector and putative sensory types. American Malacological
Bulletin 1: 13-20.

Kraemer, L. Russert. 1984. Aspects of the functional morphology
of some freshwater bivalve nervous systems: effects on
reproductive processes and adaptation of sensory
mechanisms in the Sphaeriacea and Unionacea. Malacologia
25(1): 221-239.

Kraemer, L. Russert and M. L. Galloway. 1986. Larval development
of Corbicula fluminea (Muller) (Bivalvia: Corbiculacea): an ap-
praisal of its heterochrony. American Malacological Bulletin:
4(1):61-79.

Kraemer, L. Russert and M. E. Gordon. 1980. Comparison of mollusks
retrieved by crowfoot dredge and ponar grab sampler from the
White River at St. Charles, Arkansas with comment on popula-
tion structure of Corbicula fluminea (Bivalvia: Sphaeriacea).
Bulletin of the American Malacological Union For 1 980: 63-67.

Kraemer, L. Russert, C. Swanson, M. Galloway and R. R. Kraemer.
1986. Biological basis of behavior in Corbicula fluminea, II.
Functional morphology of reproduction and development and
review of evidence for self-fertilization. IN: J. C. Britton, ed.,
Proceedings of the Second International Corbicula Symposium,
American Malacological Bulletin Special Edition, No.
2:193-201.

McCleod, Michael J. 1986. Electrophoretic variation in U.S. Corbicula.
IN: J. C. Britton, ed., Proceedings of the Second International
Corbicula Symposium, American Malacological Bulletin Special
Edition No. 2: 125-132.

McMahon, R. F. 1982. The occurrence and spread of the introduced
Asiatic freshwater bivalve, Corbicula fluminea (Muller), in North
America: 1924-1982. Nautilus 96(4): 134-141.

Prezant, R. S. and K. Chalermwat. 1983. Environmentally induced
changes in shell microstructure of the Asiatic bivalve, Cor-
bicula. American Zoologist 23(4): 914.

Prezant, R. S. and K. Chalermwat. 1984. Flotation of the bivalve Cor-
bicula fluminea as a means of dispersal. Science 225:
1491-1493.

Whitfield, P. J. and R. H. Emson. 1983. Presumptive ciliated recep-
tors associated with the fibrillar glands of the echinoderm, Am-
phipolis squamata. Cell and Tissue Research 232: 609-624.

Wiens, J. A. 1977. On competition and variable environments.
American Scientist 65: 590-597.

BiOLOGICAL BASIS OF BEHAVIOR IN CORBICULA FLUMINEA, II.
FUNCTIONAL MORPHOLOGY OF REPRODUCTION AND
DEVELOPMENT AND REVIEW OF EVIDENCE FOR
SELF-FERTILIZATION

LOUISE RUSSERT KRAEMER, CHARLES SWANSON, MARVIN GALLOWAY

AND ROBERT KRAEMER
UNIVERSITY OF ARKANSAS,

FAYETTEVILLE, ARKANSAS 72701, U.S.A.

ABSTRACT

Results reported in this study of the functional morphology or reproduction and development
are based on findings from northwest Arkansas populations of Corbicula fluminea (Muller). A captive
population of the clams was maintained by AP&L personnel in the intake bays of the Arkansas Nuclear
One facility of the Arkansas Power and Light Company on Lake Dardanelle, an impoundment of the
Arkansas River at Russellville, Arkansas. The captive population and other “natural” populations of
C. fluminea in the Buffalo River and White River in Northwest Arkansas, and the Llano River in Llano
County, Texas, were subjected to long term study. Serially sectioned C. fluminea at various stages
of development, fresh-tissue dissections and scanning electron microscopy (SEM) were used in this
study. Results include (1 ) verification of the proto-oogamous development of the reproductive system
in C. fluminea and determination of the role of early innervation of the gonopores and development
of follicular “ganglia” which accompany later stages of spermatogenesis; (2) verification of the
developmental sequence in oogenesis and the sequential, changing appearance of the oogenic follicles
in the visceral mass; (3) determination of characteristics of the biflagellate sperm of C. fluminea as
well as evidence that three “kinds” of sperm are not polymorphic sperm but are quite likely several
stages in spermiogenesis; (4) additional evidence of intrafollicular, self-fertilization of eggs in the visceral
mass of C. fluminea, from sightings of intrafollicular embryos in fixed and in fresh tissues; (5) deline-
ation of the entire developmental sequence in C. fluminea, along with evidence that the embryonic
stages most frequently shed into the environment are the early to late juveniles.

Early reports in the United States regarding reproduc-
tion and life history of Corbicula fluminea (Muller) asserted
a hermaphroditic habit and spawn of various developmental
stages, especially the trochophore (Sinclair and Isom, 1963;
Sinclair, 1971). These assertions were based largely on
bibliographic evaluations assembled by earlier workers in the
face of dramatic, sudden change in the malacofauna of U.S.
rivers. While the 1977 Corbicula Symposium had produced
consensus that C. fluminea is probably a hermaphrodite,
definitive supporting evidence was not available. Consequent-
ly, later study by Kraemer and Lott (1977) and by Kraemer
(1978) on histological differentiation of the reproductive
tissues and related nervous tissue (Kraemer, 1983a, 1984)
were apparently the first in the U.S. to report anatomical
details of the reproductive process in C. fluminea.

In this paper we review results of our continuing study
of the biological basis of reproduction in C. fluminea. A
number of our findings stem from studies done with the sup-
port of the Arkansas Power and Light Company, whose per-

sonnel have maintained a captive population of C. fluminea
at the Arkansas Nuclear One site at Russellville, Arkansas
on the Arkansas River, since 1981. Since a costly “clam clog”
at the site in September, 1980, AP&L has maintained a
substantial logistical and administrative interest in the
reproductive cycling of C. fluminea.

MATERIALS AND METHODS

Histological serial sections of approximately 6 animals
each in the size ranges 2mm, 4mm, 8mm, and 20mm of C.
fluminea, made as described elsewhere (Kraemer and Lott,
1 977; Kraemer, 1 978; 1 979b) were used to work out the three
dimensional structure and to describe: (1) the anatomical site
and histological context within which differentiation of the
gonadal follicles occurs; (2) intrafollicular development se-
quence of oogenic and spermatogenic follicles; (3) develop-
ment and later innervation of the paired gonopores; (4) ap-
pearance and differentiation of the follicular “ganglia;” (5)

American Malacological Bulletin, Special Edition No. 2 (1986): 193-201

193

194

CORBICULA SYMPOSIUM

histological evidence of intrafollicular fertilization and cross-
fertilization; (6) mode of passage of gametes and (evidently
self-fertilized) embryos from the gonopores into the anterior
chambers of the inner gills.

Living specimens were obtained from several “wild”
populations in the Buffalo River and the White River of north-
west Arkansas, from the Llano River in Llano, Texas. Living
clams were obtained also from some captive populations
maintained in the intake bays at Arkansas Nuclear One, a
facility of Arkansas Power & Light Company located at
Russellville, Arkansas near Lake Dardanelle on the Arkan-
sas River. Collections of living material from Russellville were
made at monthly, bimonthly and (during reproductive
seasons) at weekly and daily intervals. Fresh tissue dis-
sections were made to determine: (1) the extent of growth
of the oogenic follicles; (2) state of intrafollicular development

of oogenic follicles; (3) presence or absence of spermatogenic
follicles (which appear peripherally and mostly near the sur-
face of the visceral mass); (4) kinds of spermatogenic cells
present in the follicles; (5) kinds of sperm and sperm mo-
tility; (6) presence, developmental stages, numbers and
distribution of embryos within the gills; (7) presence of em-
bryos and their developmental stage, within the follicular
chambers of the visceral mass. Results of the foregoing
studies are reviewed in this paper and will be discussed in
detail elsewhere (Kraemer and Galloway, 1986; Kraemer and
Swanson, in prep.).

All stages of embryonic development and many stages
of oogenic and spermatogenic development were monitored
with a Panasonic VHS Minivision, Color Video Camera,
mounted on an AO 110 Microstar microscope with phase
optics.

Fig. 1. Drawing of Corbicula fluminea, left valve removed and portion of left mantle lobe reflected to show location of main nerve ganglia
and main connective nerves. AA, anterior adductor muscle; BSH, branchia shelf; CG, cerebral ganglion; CVC, cerbrovisceral connective;
F, foot; LP, labial palp; PA, posterior adductor muscle; PM, pedal muscle; VG, visceral ganglion; VM, visceral mass.

KRAEMER ET AL.\ CORBICULA REPRODUCTIVE FUNCTIONAL MORPHOLOGY

195

Fig. 2. a. Scanning electron micrograph showing gonopore of Cor-
bicula fluminea. GP, lip of gonopore; GVC piece of cerebrovisceral
connective; CC, ciliary cluster, peculiar to the gonopore lips. Horizon-
tal field width = 373 /xm. b. Photomicrograph of sagittal section of
posterodorsal, lateral region of visceral mass, showing a gonopore
and some of its innervation by means of fibers from the
cerebrovisceral connective nerve. CVC, cerebrovisceral connective;
GD, gonoduct; GP, gonopore. (From Kraemer, 1978). Horizontal field
width = 663 ^m.

Details of spermatogenesis, spermiogenesis and em-
bryogenesis were elucidated with preparation of tissues as
described elsewhere (Kraemer, 1983b) for viewing with an
ISI-60 Scanning Electron Microscope (SEM) at 30 Kv and a
working distance of 15 nm.

RESULTS

DEVELOPMENT AND INNERVATION OF THE
GONOPORES. When young clams attain a length of 3-4 mm,
serial sections reveal that, before there is any histological in-
dication of gonad development, there are a pair of well-
differentiated gonopores. The gonopores are located, one on
either side of the posterior dorsal surface of the visceral mass,
where the latter forms a juncture with the kidneys. This is
also the site where large cerebrovisceral nerve connectives
emanate from the visceral mass to course posteriorly and join
the prominent, fused visceral ganglion of the clam’s central
nervous system (Kraemer, 1978) (Fig. 1).

The gonopores exhibit conspicuous lips composed of
tall, ciliated columnar epithelium. The cilia are large, evidently
effector organelles that manifest a peculiar clumped array
(Kraemer, 1983b) (Fig. 2a). Only when oogenesis is initiated
in young clams, do the cerebrovisceral connectives “sprout”
nerve fibers that innervate the epithelium of the gonopore lips,
(Fig. 2b) (Kraemer, 1978). Function of this highly innervated
gonopore suface is not understood, but certainly merits fur-
ther investigation because of several likely roles the gonopore
opening may play in fertilization or embryogenesis.

PROTO-OOGAMY, OOGENESIS AND DEVELOP-
MENT OF OOGENIC FOLLICLES. It was initially assumed
that C. fluminea was protandrous. Basis for this assumption
rested on the finding that hermaphroditic bivalved mollusks
tend to be protandrous (Fretter and Graham, 1964) and
secondly that the indigenous, thin-shelled relatives of C.
fluminea, the fingernail clams (Sphaerium) and the pill clams
(, Pisidium ) were protandrous (Heard, 1977). Nevertheless, the
first histological indication of gonadal development in C.
fluminea occurs when oogenic follicles differentiate next to
the basement membranes of the mucosa of the gut wall or
of the digestive glands, (Kraemer, 1978). As slender tubes
in close association with digestive tissues, the initial oogenic
follicles enlarge, branch and ramify through the stroma of the
visceral mass. The stroma itself undergoes substantial
change from a loose collagenous tissue to a compact cellular
tissue. Epithelium of the digestive glands also changes from
low cuboidal to tall columnar epithelium (Kraemer, 1978).

As the oogenic follicles enlarge and increase in
number, their contents undergo conspicuous histological
change as well. At first the young oogenic follicles contain
small oocytes of various sizes attached to the inner surface
of the follicular membranes. The oocytes then enlarge and
come to occlude the lumen of the oogenic follicles. Next the
enlarged oocytes appear stalked and elongate. Finally the
oogenic follicles appear quite empty of oocytes but may con-
tain occasional embryos (discussed further below), as the
mature oocytes are evidently discharged, (Kraemer, 1978).

The foregoing developmental sequence was worked

196

CORBICULA SYMPOSIUM

Fig. 3. a. Photomicrograph of section of visceral mass of C. fluminea showing “follicular ganglion” such as appear late in oogenic develop-
ment at regions of confluence with spermatogenic follicles. N, nerve which attaches to follicular ganglion to either the pedal ganglion or one
of the cerebral ganglia; NP, apparent neuropile; SF, spermatogenic cells in spermatogenic follicle; VMS, visceral mass stroma. (After Kraemer,
1978). Horizontal field width = 414/tm. b. Similar section of visceral mass, showing confluence of well developed oogenic and spermatogenic
follicles. DG, digestive gland; LSF, lumen of seminiferous follice; 00, oocyte; SF, seminiferous follicle; SS, sperm sphere (comprised of mature
sperm; VMS, visceral mass stroma. Horizontal width = 1968 /^m.

Fig. 4. Scanning electron micrograph of mature sperm of C. fluminea
showing most of its biflagellate tail. BT, biflagellate tail; SH, sperm
head. Horizontal field width = 32 ^m.

out from study of thousands of serial sections of young clams
(Kraemer, 1978). In extensive study of hundreds of fresh
tissue dissections since 1981, however, it has been pos-
sible for us to verify all of these stages in oogenesis as a
seasonal sequence of oogenesis as well, with the exception
of the initial follicle appearance. Evidently, once C. fluminea
has achieved sexual maturity, although there is much
seasonal growth and resorption of oogenic follicles, some
oogenic follicles are present at all seasons of the year.

DEVELOPMENT OF THE SPERMATOGENIC FOL-
LICLES, FOLLICULAR GANGLIA, SPERMATOGENESIS,
SPERMIOGENESIS AND SPERM MOTILITY. Only when
oogenic follicles are well differentiated and when oogenesis
within these follicles is advanced do spermatogenic follicles
appear. The foregoing sequence is true not only developmen-
tally but seasonally in the life history of C. fluminea. Sper-
matogenic follicles appear peripheral to the oogenic follicles.
At the confluence of oogenic and spermatogenic follicles, in
at least four paired locations in the visceral mass, clusters
of what appear to be neuronal cell bodies appear during the
reproductive maturation of the young clams. The cell bodies
surround a feltwork of evident nerve fibers. Each of these
structures, designated “follicular ganglia” (Kraemer, 1978,
1979b; 1984), is clearly (Fig. 3a) innervated by a nerve from
either the cislateral cerebral ganglion or the pedal ganglion.
The follicular space surrounding each “follicular ganglion”
is typically filled with spheres of mature sperm, some oocytes,
and occasionally what appear to be embryos (Fig 3b, 5a). The
foregoing observations have been made repeatedly in studies
of serial sections of C. fluminea. It is the intrafoilicular loca-
tion of both the “follicular ganglia” and their proximity to the
intrafoilicular embryos which allow the conclusion that the
“ganglia” may orchestrate sperm maturation and in-
trafollicular, self-fertilization as well.

Certain details of spermatogenesis, including sper-
miogenesis, have been worked out (Kraemer, 1983b;
Kraemer and Swanson, in prep.). Mature sperm are all
biflagellate, large cells. Their tapering heads cluster in
spheres in the follicular lumen or in flattened spheres against
the follicular wall. What initially appears to be polyspermy in

KRAEMER ET AL.\ CORBICULA REPRODUCTIVE FUNCTIONAL MORPHOLOGY

197

Fig. 5. a. Photomicrograph of mature gametogenic follicle of C. fluminea , showing presence of evidently self-fertilized embryos. Horizontal
field width = 750 fim. b. Section of gonoduct, showing evidently self-fertilized embryo in the duct. Horizontal field width = 490 /*m. c. Section
of gametogenic follicle containing section of evidently self-fertilized embryo in veliger stage. Horizontal field width = 183 ^m. d. Section of
marsupial gill also showing section of veliger, evidently the result of cross-fertilization. Horizontal field width = 258 ^m. E, embryo; GD, gonoduct;
GP, gonopore; LOF, lumen of gametogenic follicle; LOG, lumen of marsupial gill; SVM, stroma of visceral mass; V, velum.

198

CORBICULA SYMPOSIUM

our studies of C. fluminea, now is found to be different stages
of spermiogenesis. Both light microscopy of living sperm and
SEM studies have allowed us to conclude a spermiogenic
sequence in C. fluminea from “round-headed” to “wide-
headed” to “slender-headed” (mature) sperm, all of which
are biflagellate (Kraemer, 1983c; Kraemer and Swanson, in
prep.). These are the first biflagellate sperm known to be
reported for any bivalved mollusk, (Fig. 4).

Motile sperm are not commonly observed. In fresh
tissue dissections, the “wide-headed” sperm often exhibit
a characteristic, “twitching” movement in which one of the
flagella trails at an acute angle from the sperm cell, and the
other flagellum produces locomotor waves of varying
amplitude which begin at the flagellum base and move to its
distal tip, thereby producing the twitching movement of the
large sperm head. Rarely are rapidly swimming, mature
sperm seen in fresh preparations of living spermatogenic
follicles. Mature sperm cells may readily separate from their
spherical clumps, aided by water currents and by the lashing
independent movements of one of the sperm cell’s two
flagella.

While our cumulative data indicate that oogenesis con-
tinues throughout the year, though it slows in January and
February, our data also indicate (Kraemer and Galloway,
1986) that spermatogenesis is seasonal and responsive to
water temperature change. After 7-10 days of water
temperature between 17-19°C in April, spermatogenic follicles
become numerous in the visceral mass, and they are well
developed, containing many spheres of mature sperm. Sper-
matogenesis will continue through the spring and summer,
though in a fairly rhythmic pattern of approximately three
week intervals, until the water temperature reaches 32-34°C.
In late fall (usually early November) spermatogenic follicles
are much reduced in size, number and spermatogenic ac-
tivity, once the water temperature has fallen and remains
below 17°C.

CROSS FERTILIZATION. On several occasions pe-
culiar mucous strands have been observed in our laboratory
(in shallow water, mid-summer), trailing from the siphons of
one clam in a population to the siphons of another. When
examined under the microscope, the mucous strands have
been found to contain many, “twitching” sperm. It is
hypothesized that the connecting mucous strands may ef-
fect cross-fertilization in these hermaphroditic animals. Some
support for the foregoing contention lies in the fact that a
senior malacologist from western China reports (Brian Mor-
ton, personal comm.) that commercial cultivators of Corbicula
have known for years that within several weeks after such
mucous strands appear in their clam cultures, young clams
will appear in those cultures!

SELF FERTILIZATION. As noted above, serial sec-
tions of C. fluminea earlier revealed that some of the
gametogenic follicles contained young embryos (Kraemer,
1978). While embryos have been found within oogenic
follicles, they are characteristically seen within follicles near
the “follicular ganglia” described above (Fig. 5a, c). Embryos
have also been seen in the gonoduct and in the region of
the gonopore (Fig. 5b). Some of the intrafollicular embryos

appearing in serial sections of the visceral mass prove to be
veligers (Fig. 5b) indistinguishable from veligers which ap-
pear in serial sections of marsupial gills (Fig. 5d). It seems
quite parsimonious to reason that embryos have been found
within the gametogenic follicles in the visceral mass because
they have been produced there as a consequence of self
fertilization.

Recently, the opportunity to make hundreds of fresh
tissue dissections of the visceral mass of C. fluminea has led
to the discovery in at least four instances of large numbers
of young embryos within the gametogenic follicles of the
visceral mass (Kraemer and Galloway, 1986). The dissections
were made with care to assure that no contamination of the
visceral mass with marsupial gill tissue had occurred. It was
thereby possible to ascertain that several clams thus
dissected did indeed house hundreds of blastula-like embryos
within the follicles of the visceral mass (Fig. 6a). These were
often surrounded by sperm (Fig 6a, S). All of the aforemen-
tioned clams were identified during the fall reproductive pulse,
after spermatogenesis had apparently ceased.

During the 1983 Second International Corbicula Sym-
posium Kennedy (Kennedy, et al., in press) reported on a very
painstaking effort to rear C. fluminea isolates, in order to deter-
mine the likelihood of self fertilization. Results of this careful
work were mixed. Self fertilization does seem highly likely
to occur in C. fluminea however, for the following reasons:

(1) young embryos have been repeatedly found in the visceral
mass in serial sections of the clams (Kraemer, 1978, 1984);

(2) the embryos have invariably been located within the
gametogenic follicles or within the gonoduct; (3) intrafollicular
embryos have often been seen in the region of the “follicular
ganglia” where sperm and eggs are in close proximity; (4)
since spermatogenic and oogenic follicles are contiguous in
C. fluminea, the mature gametes have ready access to each
other, a situation which obtains in other molluscan bivalves
known to self fertilize (Fretter and Graham, 1964); (5) young
embryos have been found within the visceral mass on several
occasions in our fresh tissue dissections, when precautions
have been taken to avoid contamination with gill tissue; (6)
the instances of self fertilization seem to occur primarily dur-
ing the fall reproductive pulse when falling water temperature
seems to effect a “closing down” of spermatogenesis in most
members of a clam population; (7) it seems that the fall pulse
is more likely to be associated with “clam clogs” than is the
spring pulse, according to some members of the nuclear
power industry (Bob West, personal comm.). Such an obser-
vation may be due to the unusual spate of young clams
generated in the fall pulse, both as a consequence of self
fertilization and of cross fertilization; and (8) though rearing
of isolate clams by other workers has not produced definitive
results, it has developed some indication that self fertiliza-
tion probably occurs in C. fluminea. In summary, we are con-
vinced that self fertilization is a regular occurrence in C.
fluminea although the process may be limited to a period late
in the fall reproductive pulse.

REVIEW OF EMBRYOGENESIS. Unlike its in-
digenous relatives (the pill clams, Pisidium, and the finger-
nail clams, Sphaerium, etc.) that show repression both of the

KRAEMER ET AL.\ CORBICULA REPRODUCTIVE FUNCTIONAL MORPHOLOGY

199

I I

' • - \\ n

v m
1 $

ft ■ L \ •*4

I * - t ' '■

III .. ■

Fig. 6. a. Photomicrograph of a living blastula, taken from a gametogenic follicle in the visceral mass of an adult specimen of C. fluminea,
and thus evidently self fertilized. Note the many sperm surrounding the blastula. Horizontal field width = 268 ^m. b. Living trochopore, taken
from the marsupial gill. Horizontal field width = 435 fim. c. Living veliger, taken from the marsupial gill. Horizontal field width = 310 fim.
d. Living, straight-hinged juvenile, taken from the marsupial gill. Horizontal field width = 360 /un. A, apical ciliary tuft; B, blastocoel; H, straight
hinge; S. larval shell valve; V, velum.

200

CORBICULA SYMPOSIUM

numbers of embryos developed and of the developmental
stages which appear, C. fluminea characteristically produces
thousands of juvenile clams during each reproductive pulse.
In our laboratory we have determined that fertilization (cross
fertilization and self fertilization) is followed by cleavage which
produces a distinct blastula. After the blastula stage (Fig. 6a)
a gastrula with pyramidal shape and a conspicuous
blastopore near the vegetal pole, develops. Still within the
parent’s tissues, the gastrula develops into a barrel-shaped
trochophore, replete with apical ciliary tuft, (Fig. 6b). In fresh
tissue dissections, the trochopore larvae found in the gills
exhibit varied behavior, swimming in circular or longitudinal
paths, but always with apical cilia “forward”.

Within the marsupial gills, metamorphosis of the
trochophore into a veliger larva takes place, as a cilia-covered,
flange-like membrane extends laterally from the surface of
the larva, posterior to the apical ciliary tuft. Next, rudimen-
tary valves of the clam appear, gradually enlarging to enclose
more and more of the veliger. The velum and apical ciliary
tuft remain conspicuously apparent, however (Fig. 6c). Still
within the marsupial gill, the veliger develops into a
pediveliger. With the apical ciliary tuft still “anterior”, a mass
of tissue just posterior to the velum begins to grow and dif-
ferentiate, finally producing the small, increasingly active foot
of the pediveliger.

Next, both the apical ciliary tuft and the velum disap-
pear, and the juvenile stage is established as the foot grows,
lengthens, changes its contour (Fig. 6d). At this stage the con-
joined statocysts described earlier (Kraemer, 1984) are also
well differentiated and the gills enlarge their ciliary surface.
The gut now exhibits a long, algae-filled loop and a twirling
crystalline style, as the young clam is busily feeding. Our data
indicate that it is chiefly in the juvenile stage that C. fluminea
is spawned, typically when the shell valves have reached a
length of about 200 /xm. The foregoing information is treated
in detail elsewhere (Kraemer and Galloway, 1986).

SUMMARY AND DISCUSSION

We have reviewed findings from our continuing study
of the biological basis of reproduction in C. fluminea. We have
reported on oogenesis (1) that well defined gonopores con-
stitute the first histological sign of development of reproduc-
tive structures in the young clam; (2) that the gonopores
become innervated only when gonadal development begins;
(3) that oogenesis occurs first and in association with base-
ment membranes of gut wall or digestive glands; (4) that
oogenesis is accompanied by changes in the visceral stroma
and in the digestive gland epithelium; (5) that oogenesis, once
initiated in the young clam, probably continues throughout
the life of the clam, though more sluggishly in January and
February; (6) that development of the oogenic follicles follows
a predictable developmental and seasonal sequence.

Concerning spermatogenesis we have reported that:
(1) spermatogensis occurs only after oogenesis is well ad-
vanced; (2) spermatogenesis occurs at intervals during the
year, and is evidently quite susceptible to water temperature
change; (3) spermatogenesis occurs in follicles which develop

peripheral to the oogenic follicles; (4) at the confluence of
oogenic and spermatogenic follicles, “follicular ganglia” ap-
pear which may affect maturation of sperm and/or self fer-
tilization; (5) spermiogenesis involves development of a se-
quence of “round-headed,” “wide-headed” and “slender
headed” sperm, rather than polyspermy; and (6) all sperm
in C. fluminea are biflageilate sperm, the only molluscan
biflagellate sperm known to us.

Concerning fertilization we have reported that: (1)
cross fertilization may be the norm in C. fluminea, as repeated
observations of movement of spheres of mature sperm and
a number of observations of sperm-laded mucous strands
connecting siphons of neighboring clams, suggest apparent
mechanisms by means of which cross fertilization takes place;
(2) seif fertilization, we are convinced, also takes place in C.
fluminea, though perhaps on a less regular basis, perhaps
primarily in the fall and in association with seasonal
temperature drop; (3) self fertilization is evident from our serial
section study which has repeatedly located embryos within
the gametogenic follicles of the visceral mass, and from find-
ing on at least four carefully controlled occasions, during fresh
tissue dissections of the visceral mass, numerous young em-
bryos within the visceral mass (Fig. 6a).

We have noted contrasting reproductive features of
C. fluminea and its indigenous freshwater relatives, the pill
clams and fingernail clams, and have reported that: (1) unlike
the latter, C. fluminea does not exhibit extensive repression
of embryo development and repression of developmental
stages to produce relatively few, large, mature young; (2) on
the contrary, C. fluminea not only produces from hundreds
to thousands of tiny, 200 /xm long, juvenile clams during each
reproductive pulse, but also each of the juvenile clams has
progressed rapidly through clearly evident blastula, gastrula,
trochophore, veliger and pediveliger stages, before reaching
juvenile status. It has been reported elsewhere (Kraemer,
1979a) that byssus development in C. fluminea is a post-
spawning feature to aid the young clam as a holdfast for the
substratum. The pill clams and fingernail clams, in contrast,
develop a “placental” byssus that serves as a holdfast for
the embryo within the marsupial gill (Mackie, 1979).

A continuing puzzie from our findings lies in the fact
that there are some reproductive differences between the in-
digenous pill clams and fingernail clams and C. fluminea
which are very striking: (1) protandry for the former, proto-
oogamy for C. fluminea] and (2) uniflagellate sperm for the
former, biflagellate sperm for C. fluminea. Other differences
between the indigenous and the introduced clams seem to
be those of timing of embryonic events — timing which allows
a very different life style for the two kinds of clams.

In the indigenous pill dams and fingernail clams,
development bypasses the trochophore, veliger and the
pediveliger stages, all of which are evident in C. fluminea.
In the indigenous pill clams and fingernail clams, byssus ap-
pearance is early, in C. fluminea it is late, as noted above.
In the indigenous clams few young are produced with each
brood, some of which attain nearly parental size and even
undergo sexual maturation while still in the parental gill. Way,
et al. (1980) examined comparative life history tactics of a

KRAEMER ET AL: CORBICULA REPRODUCTIVE FUNCTIONAL MORPHOLOGY

201

sphaeriid clam, Musculium partumeium (Say) from a per-
manent and from a temporary pond and found significant in-
traspecific life history variations between the two clam popula-
tions. Interspecific variation in life history tactics has also been
evaluated, by Stearns (1976, 1977).

It has been suggested elsewhere (Kraemer and Gallo-
way, 1986) that C. fluminea may be able to vary its life history
tactics from those of the indigenous clams by employing al-
ternative timing of developmental events. Such a change in
life history timing has been characterized at length by Gould
(1977) as heterochrony. By speeding differentiation of many,
many embryos, and by retaining a series of developmental
stages which are repressed in the indigenous clams, C.
fluminea can greatly increase its reproductive potential. By
retarding its byssal development until after spawning in the
juvenile stage, the byssus of C. fluminea can perform a
substantially different function than is manifested by the
“placental” byssus of the indigenous clams. Indeed it may
well be that the C. fluminea presence in U. S. river systems
is largely due to the evident heterochrony which, our studies
have revealed, exists between the indigenous, thin shelled
fingernail clams and pill clams, and the introduced Asian clam
Corbicula fluminea.

ACKNOWLEDGEMENTS

It is a pleasure to acknowledge critical funding for this study
provided by Arkansas Power and Light Company of Little Rock,

Arkansas. The authors wish to thank Robert M. West who ad-
ministered the funds and skillfully supervised the project, and Charles
Adams and Dennis Calloway who maintained the captive popula-
tion of C. fluminea in the intake bays at Arkansas Nuclear One at
Russellville, Arkansas and who were very helpful and effective in
providing both organisms and information for the study. We would
also like to thank the anonymous reviewers of this paper for their
helpful suggestions.

LITERATURE CITED

Fretter, V. and A. Graham, 1964. Reproduction. In: Wilbur, K. M. and
C. M. Yonge, Physiology of Mollusca , I. Academic Press, New
York. pp. 127-164.

Gould, Stephen Jay. 1977. Ontogeny and Phytogeny . Belknap Press
of Harvard University Press, Cambridge, Massachusetts. 501

pp.

Heard, W. H. 1977. Reproduction of fingernail clams (Sphaeriidae:

Sphaerium and Musculium). Malacologia 16: 421-455.
Kennedy, V. S., L. van Heukelem and W. F. van Heukelem (in press).
Experiments on self-fertilization in the Asiatic clam, Corbicula

sp., (Bivalvia: Corbiculidae).

Kraemer, Louise Russert. 1978. Corbicula fluminea (Bivalvia:
Sphaericea): the functional morphology of its her-
maphroditism. Bulletin of the American Malacological Union
For 1978: 40-49.

Kraemer, Louise Russert. 1979a. Juvenile Corbicula: their distribu-
tion in the Arkansas River benthos. In: Britton, J. (ed.), Pro-
ceedings of the First International Corbicula Symposium. Texas
Christian University, Fort Worth, Texas, October 13-15, 1977.
pp. 89-97.

Kraemer, Louise Russert. 1979b. Corbicula (Bivalvia: Sphaeriacea)
vs. indigenous mussels (Bivalvia: Unionacea) in U.S. rivers:
a hard case of interspecific competition? American Zoologist
19: 1085-1096.

Kraemer, Louise Russert. 1983a. Comparative functional morphology
of cilia of Corbicula fluminea (Bivalvia: Corbiculidae): possi-
ble criteria for effector and putative sensory types. American
Malacological Bulletin 1 : 1 3-20.

Kraemer, Louise Russert. 1983b. Ontogenetic aspects of biflagellate
sperm in Corbicula fluminea (Muller) (Bivalvia: Sphaeriacea).
Transactions of the American Microscopical Society 102: 88.

Kraemer, Louise Russert. 1984. Aspects of the functional morphology
of some fresh-water bivalve nervous systems: effects on
reproductive processes and adaptation of sensory
mechanisms in the Sphaeriacea and Unionacea. Malacologia
25(1): 221-239.

Kraemer, Louise Russert and M. K. Galloway. (1986). Larval develop-
ment of Corbicula fluminea (Muller): and appraisal of its
heterochrony. American Malacological Bulletin 4(1):61-79.

Kraemer, Louise Russert and Susy Lott. 1977. Microscopic anatomy
of the visceral mass of Corbicula (Bivalvia: Sphaeriacea).
Bulletin of the American Malacological Union For 1977: 48-55.

Mackie, G. L. 1979. Growth dynamics in natural populations of
Sphaeriidae clams (Sphaerium, Musculium, Pisidium). Cana-
dian Journal of Zoology 57: 441-456.

Sinclair, R. M. 1971. Annotated bibliography on the exotic bivalve
Corbicula in North America, 1900-1971 . Sterkiana, 43: 11-18.

Sinclair, R. M. and B. G. Isom. 1963. Further studies on the intro-
duced Asiatic clam ( Corbicula ) in Tennessee. Tennessee
Stream Pollution Board, Tennessee Department of Public
Health, Tennessee.

Stearns, S. C. 1976. Life-history tactics: a review of the ideas. Quarter-
ly Review of Biology, 51: 3-47.

Stearns, S. C. 1977. The evolution of life history traits: a critique of
the theory and a review of the data. Annual Review of Ecology
and Systematics 8: 145-171.

Way, C. M., D. J. Hornbach and A. J. Burky, 1980. Comparative life
history tactics of the sphaeriid clam, Musculium partumeium
(Say) from a permanent and a temporary pond. American
Midland Naturalist 104(2): 319-327.

UNSOLVED PROBLEMS AND PROMISING APPROACHES IN THE

STUDY OF CORBICULA

K. ELAINE HOAGLAND

CENTER FOR MARINE AND ENVIRONMENTAL STUDIES
LEHIGH UNIVERSITY

BETHLEHEM, PENNSYLVANIA 18015 U.S.A.*

and

ACADEMY OF NATURAL SCIENCES
19TH AND THE PARKWAY
PHILADELPHIA, PENNSYLVANIA 19103 U.S.A.

ABSTRACT

Research on the introduction of Corbicula to the United States and subsequent biofouling prob-
lems has been fragmented by the need for immediate answers in special situations. The problems
should be examined in the more general context of other introduced species problems and species
outbreaks.

It is clear that basic research on Corbicula is badly hampered by confusion in systematics of
the genus. Shell data alone are inadequate for species determination until allozyme and anatomical
data establish the limits to taxa. Once limits are known, we can determine the number of species
in the U.S., their distributions, and their affinities to Asian taxa. Topotype populations in Asia must
be compared to U.S. Corbicula sp(p). Then, past conflicting research on life history, reproduction,
historical distribution patterns, and physiology can be properly interpreted and extended. Finally,
physiological and life history data can be applied to formulate general and local control strategies.

Examples of the use of allozyme data to solve problems in systematics and zoogeography,
leading to clarification of physiological and life-history bases of species outbreaks, are presented.
These include the cases of the polychaete Capitella capitata, the bivalve Teredo bartschi, and the
gastropod Crepidula fornicata, in addition to Corbicula. Data available as of 1984 indicate that Cor-
bicula consists of two species in North America.

Local control strategies depend upon knowledge of natural environments, reservoir popula-
tions, and the artificial environment of industrial plants. The importance of local biologists’ knowledge
of population dynamics in proposing cost-efficient solutions is obvious. Insufficient information on lar-
val physiology and behavior still hampers biologists in formulating the best possible solutions.

Data that plant operators should accummulate and make available to biologists include exact
location of living clams vs. shells, effectiveness studies of mechanical devices to eliminate clams,
and data to be acquired whenever clams are removed from a plant, such as number and sizes. Final-
ly, biologists can only present useful solutions if they are aware of economic and engineering aspects
of potential control strategies.

Many kinds of professional scientists and engineers have
had to deal with Corbicula as a biofouling agent. This group
includes managers and government regulators, as well as
general in-house and consulting biologists and chemists, and
academic specialists in ecology or malacology. My remarks
are intended to communicate with these people on several
levels about the kinds of data needed to understand Corbicula
in the U.S. I hope to illustrate to those eager for immediate
answers the need for some research into fundamental ques-
tions such as systematics and ecology of natural populations.

'Mailing Address

I first present a hierarchy of interrelated questions that
must be answered, optimally by coordinated research efforts
of many types of scientists, in order to find optimal control
strategies. I show how answers to some of these general
questions have worked to bring understanding of other cases
of introduced species and/or population outbreaks. Then I
discuss site-specific data needed to adapt general strategies
to local control problems.

METHODS

Most of this paper involves general discussion of the

American Malacological Bulletin, Special Edition No. 2(1986):203-209

203

204

CORBICULA SYMPOSIUM

Table 1. Ecological and Reproductive Characteristics of Some Corbicula.

SPECIES

HABITAT

WHERE STUDIED

LIFE HISTORY

SEXUALITY

C. fluminea
fide Morton, 1982

Streams

S. China

Breeds twice a year.

Broods young to 200 ^m.
Lives 3 years.

Dioecious + hermaph-
rodites?

C. cf fluminea

Streams,

impoundments

N. America

Same as above. Some re-
ported to release veligers.

Simultaneous hermaphrodite;
may self-fertilize

C. cf fluminalis
fide Morton, 1982

Upper estuary

Pearl River,

Canton area

Spawns annually; cool
waters. May live 10 years.

Most are dioecious.

C. leana fide

Fuziwara, 1975, 1979

Streams

Japan

Ovovivi parous; spawns
twice a year; warm water.

Hermaphrodite

literature. The literature on Corbicula is not exhaustively
reviewed; this is not a review paper perse. Literature on other
introduced species problems and on related subjects are
discussed in the context of Corbicula.

UNSOLVED PROBLEMS

SYSTEMATICS AND BIOLOGY

The literature of North American Corbicula is filled with
taxonomic confusion. Although the name Corbicula fluminea
(Muller) was used in the 1960’s (e.g., Hubricht, 1963), C.
manilensis Philippi was used frequently during the 1960’s and
1970’s (e.g., McMahon, 1977). Corbicula leana Prime, C.
fluminalis Muller, and C. sinensis (e.g., Gunning and Suttkus,
1966; Gifford, 1974) have also been used to refer to the in-
troduced Corbicula in North America. While some workers
suggest that there is more than one species in North America
(e.g., Hillis and Patton, 1982), others, recognizing taxonomic
confusion, have attempted to standardize the usage of C.
fluminea as the single species in North America (Britton, 1979;
Britton and Morton, 1979).

These are not merely academic issues. The species
question is critical to control issues especially as we consider
more sophisticated and less environmentally-damaging
chemical treatments, and treatments based on population
dynamics and reproductive biology. Each species is unique
in its range, habitat, physiology, life history, and mode of
reproduction. The potential for spread to new waters may be
species-specific. While the systematics even in its native Asia
is by no means resolved, we do know that there are several
species with differences in habitat. For example, Corbicula
fluminea is said to prefer lotic environments (streams), and
C. cf fluminalis of Morton (1 982) prefers lentic upper estuaries.
The species also differ in tolerance to saltwater. Table 1 sum-
marizes some of the major ecological and reproductive dif-
ferences between Asian species sometimes thought to have
been species introduced to North America, as described in
the literature.

There are relatively few shell characters to separate
the species, and limits to intraspecific shell variation are poor-

ly understood. Geographic variation in physiology, sex deter-
mination, and reproduction are undefined. There are
references in the literature to a single species (C. fluminea)
possessing different sexual strategies (e.g., protandry, pro-
togyny, separate sexes) in different parts of its range (Mor-
ton, 1982). Morton uses evidence from other taxa such as
Sphaeriacea (Mackie, 1973) and Unionacea (Bloomer, 1939)
to support his claim that C. fluminea is protandric and a
simultaneous hermaphrodite in different parts of its range.
These other taxa, however, do not have both protandry and
simultaneous hermaphroditism within a single species. In the
Bloomer paper, one species was an asynchronous her-
maphrodite; two others in a different genus had separate
sexes. Evidence for Corbicula is circular, since we do not
know from genetic or anatomical evidence if the different
allopatric forms with supposedly different sexual strategies
are the same species. In most cases in the Mollusca where
a species was once thought to have more than one form of
sexuality or reproduction, we now know sibling species were
involved (e.g., Gallardo, 1977). In fact, in the mollusks, there
are no documented cases showing protandry and protogyny
in the same species. The data presented by Morton (1982)
really suggest alternating sexuality, an asynchronous
development of eggs and sperm, in a hermaphroditic species
in Asia (see Hoagland, 1984b, for definitions of terms describ-
ing sexuality in mollusks). Even if it is shown that sex deter-
mination or reproduction is plastic in one species, we can-
not extend such attributes to other species.

Those interested in control of Corbicula can proceed
most efficiently by knowing, first of all, how many species
there are and their distribution(s), both in North America and
in Asia, and possibly also in South America. Historical pat-
terns of introduction and spread can be clarified, and at-
tributes of introduced populations can be compared with
native populations. Then one can place the existing literature
on physiology and reproduction in its proper context. Other-
wise, past work, especially that done in Asia, is of little use.
Since physiological factors such as limits to temperature,
salinity, heavy metals, and oxygen are used in control
strategies, the variance between and within species, and any

HOAGLAND: PROBLEMS AND APPROACHES IN CORBICULA STUDY

205

adaptive changes between American populations and their
Asian relatives, are significant.

The best data available for American Corbicula
systematics are detailed anatomy (e.g., Kraemer, 1977, 1978,
1983; Kraemer and Lott, 1977), embryology (Kraemer etal.,
1985), and lately, electrophoretic analysis of allozymes.
Although Smith et al. (1979) demonstrated genetic uniform-
ity in 5 populations from California to S. Carolina suggesting
one genetically impoverished species in North America, Hillis
and Patton (1982) had different results. They demonstrated
quite convincingly that there are two non-interbreeding stocks
of Corbicula living at times sympatrically in Texas, fixed for
alternate alleles at 6 of 26 genetic loci, yet both expressing
genetic uniformity (lack of heterozygosity). Shell color,
sculpture, shape parameters, and ecological differences
sorted out perfectly with the two genetic types. Hence, they
concluded there are two species of Corbicula in Texas, a
“white” and a “purple” one. Their data are strengthened by
those of McLeod (1986), who reports additional elec-
trophoretic differences. Schofield and Britton (paper
presented at the 2nd International Corbicula symposium)
show some very suggestive physiological differences and
microhabitat differences, correlated with the two shell types.
The data suggest that a relatively rare purple species may
exist that can extend into low pH, high calcium waters
closed to the white species. The white taxon tends to live in
sediments of smaller grain size than the purple taxon (Hillis
and Patton, 1982).

The finding of more than one species through elec-
trophoretic techniques has allowed the establishment of limits
to ecological and phenotypic variation in other taxonomic
groups. Chambers (1978) detected two sympatric species of
the freshwater snail Goniobasis electrophoretically, one
preferring vegetation and one preferring rocks. Grassle and
Grassle (1 976) could sort major reproductive differences and
otherwise-overlooked morphological differences in 6 species
of the pollution-indicating polychaete species complex
Capitella capitata, once the species were determined elec-
trophoretically. This work has forced re-evaluation of applied
ecological studies, because each member of the species com-
plex has its own life history and physiological tolerances. Yet
many biologists still do not attempt to identify Capitella to the
species level.

Other examples of electrophoretic separation of
species followed by morphological delimiting of taxa are in
the freshwater unionid clam group. Davis (1983) was able to sort
the genus Uniomerus into three species electrophoretically,
and once this was done, seeming confusion in shell
phenotype variation was resolved and the species can now
be identified morphologically. Elliptio lanceolata likewise has
been found to be a complex of at least 6 species that sort
electrophoretically and in terms of shell phenotype (Davis,
1984). Coney (in prep.) has found some important anatomical
differences between E. /anceo/afa-group species as well. In
slight variation of this sequence of scientific progress, I
recognized sibling species of the marine gastropod Crepidula
by noting major differences in larval development, and con-
firmed that these differences represented unique taxa by elec-

trophoretic analysis of allozymes (Hoagland, 1984a).

A discussion of the uses of allozymes in systematics
can be found in Ayala (1976, 1983). Use of electrophoretic
techniques to identify allozyme variation provides discrete
phenotypic characters that are often easily correlated with
their genetic counterparts. On the other hand, complex
phenotypic characters such as growth and reproductive pat-
terns involve the interaction of many separate gene loci. Elec-
trophoretic studies provide as many specific characters as
loci can be resolved. Usually 20-30 consistently scorable loci
can be achieved for mollusks with the common starch-gel
technology, and more with more elaborate procedures.
Starch-gel electrophoresis is conservative for systematic work
at the genus level in that some closely migrating enzyme
forms (allelomorphs) cannot be resolved as different.
Therefore, genetic differences between closely-related taxa
are usually underestimated.

If the species under investigation can be bred, the
genetic basis of an allozyme pattern can be determined direct-
ly (e.g., Lassen, 1979). If not, it can usually be inferred from
studies of other closely-related taxa and from the molecular
structure of the particular enzyme. One value of elec-
trophoresis is that the data so derived are independent of
other data sets, and provide strong corroboration of tax-
onomic decisions based on other types of data, when the data
sets converge (Davis, 1983).

It is easy to control for non-genetic aspects to enzyme
patterns by avoiding use of food-containing organs, and by
doing control studies of a known species of a particular genus.
Controls can be done for age, sex, season, and food. Several
populations of the known species from different environments
can be electrophoresed. In reality, such control experiments
have been done for many organisms, and environmental in-
duction of allozymes has been found to be a rare exception
rather than the rule (e.g., Livingstone, 1981). Problems have
occurred only in a few cases, particularly with food-induced
allozymes of non-specific digestive enzymes such as
esterases (Oxford, 1975). Yet the genetic basis for esterase
patterns has also been demonstrated (Saul et.ai, 1978).
Environmentally-induced enzyme patterns would not be ex-
pected to correlate with characters such as shell sculpture,
shell shape, or resistance to toxicants, nor would they be
stable over time and space, as was found for Corbicula by
Hillis and Patton (1982) and McLeod (1986). Therefore, the
pattern observed by Hillis and Patton in which electrophoretic
patterns at several loci consistently matched a set of shell
phenotypic characters, with no intermediates, in a 3-meter2
area where water quality and food availability are relatively
uniform, makes implausible the argument that all these dif-
ferences could be due to where the clams were living.

The electrophoretic characters are far more resistant
to environmental change (i.e., have higher heritability) than
are morphological shell characters such as color and distance
between sulcations. While both of these shell characters un-
doubtedly have genetic components, they also have large en-
vironmental components. For example, the purple pigment
highlights in the white taxon can be reduced or eliminated
by stressing the animal (Prezant and Chalermwat, 1984). Ex-

206

CORBICULA SYMPOSIUM

periments that are designed to test the hypothesis that the
purple and white taxa are ecophenotypes must take the
heritability of the traits used in testing the theory into account.
It is more likely that an individual genetically competent to
produce a purple shell can be stressed so as to lose the pur-
ple pigment, than it is for a white individual to suddenly begin
to produce purple pigment when placed in a new environ-
ment. Switches in phenotypic expression of a shell trait do
not address the question of whether or not the animals belong
to the same gene pool.

The biological species concept states that two popula-
tions are separate species if they do not share a common
gene pool. The only direct tests are to try to find evidence
of interbreeding; e.g., to look for allozyme patterns
demonstrating reproductive isolation of sympatric popula-
tions. The finding of fixed alternate alleles at several loci in-
cluding Krebs cycle enzymes in mobile aquatic mollusks such
as Corbicula living in a single creek or water system is in-
deed a conclusive demonstration of genetic (reproductive)
isolation, assuming the electrophoretic data are sound. This
type of finding has confirmed the species status of 5 often-
sympatric sibling species of the gastropod Littorina in the
British Isles (Wilkins and O’Regan, 1980). Experimental cross-
breeding of individuals could also be attempted, but is am-
biguous because animals sometimes have natural barriers
to reproduction that can be circumvented in the lab.

Immediate work ahead is to delineate complete ranges
of the two electrophoretically-delineated taxa. So far, the pur-
ple taxon is positively identified from Texas, Arizona, and
California. Areas that need to be examined include Oklahoma,
New Mexico, and Mexico. The electrophoretic data must be
correlated with complete comparative anatomical, shell mor-
phological, and reproductive characters such as presence
of brooded larvae. Then, past work on physiology, ecology,
and life history in the U.S. and Asia must be evaluated and
assigned to the correct taxon, based on studies of topotypes
in Asia. Characters such as shell pigment or globosity can-
not be used alone to delineate species, but once species are
defined on the basis of multiple data bases, these characters
may be useful in species identification, especially in the field.
Until species assignment is sure, conservative researchers
will identify their data according to whether the purple or white
taxon was studied, to avoid accidentally confounding the data
for more than one species and to allow others to compare
data on similar populations.

Positive indentification will require collection and com-
parative study of topotype material in Asia. This is because
originally, most Asian taxa were described by shell characters
alone, and the Asian names must be given biological reality
in terms of assignment of the proper mode of sexuality,
reproduction, and anatomical pattern. American researchers
will certainly be cooperating in the future with their counter-
parts in Asia and in southern Europe and South America
where Corbicula also exists, first to correlate species iden-
tifications and then to compare data on the biology and con-
trol of Corbicula species.

The allozyme studies themselves can be used to
understand population structure and genetic variability of the

introduced populations. Such information provides clues as
to the reproductive mode of the populations. For example,
polymorphism yet low individual heterozygosity as found by
McLeod (1986) suggests that self-fertilization occurs at least
some of the time in North American Corbicula. Recent
anatomical and embryological work with the white taxon of
Corbicula support the possibility of both cross- and self-
fertilization. If self-fertilization is common in North American
comparative anatomy of the purple taxon, relative to the white
taxon, to find if it too might have the capability of self-
fertilization. If self-fertilization is common in North America
Corbicula, the development of genetic races with their own
physiological characteristics is at least possible. This could
make control more site-specific. The definitive experiment on
self-fertilization, isolation of juveniles that eventually mature
and produce offspring, has yet to be done.

Comparison of intra- and interspecific variation also
provides information on adaptability of Corbicula. Studies of
the shipworm Teredo bartschi introduced from Florida to New
Jersey and Connecticut showed that the species was naturally
highly monomorphic and that introduced populations were
even more so, due to founder effects and bottle-necking.
Parallel physiological studies revealed that, despite low en-
zyme variability, the native and introduced populations had
broad physiological tolerances (Hoagland, 1983). I also found
low polymorphism in the native and introduced populations
of the marine gastropod Crepidula fornicata (Hoagland,
1984a). Other workers have found low genetic variability in
a variety of invertebrate species that retain high powers of
migration and are frequently introduced to new areas (Price
and Jain, 1981; Selander and Hudson, 1976; Selander and
Kaufman, 1975).

It is interesting that the work done so far on genetics
of Corbicula also indicates low genetic variability at the
population level (Smith et al., 1979; McLeod and Sailstad,
1980; Mcleod, 1986). These findings violate the common
wisdom that genetic variability of allozymes should be high
to allow adaptation to a wide variety of habitats. In fact,
restricted genetic variability with concommitant high
phenotypic variability may be characteristic of organisms that
are capable of being successful introduced species (Price
and Jain, 1981). On the other hand, McLeod (1986) finds
evidence of accumulation of new mutations in American Cor-
bicula. Nevo, et al. (1977) have shown genetically-based ther-
mal adaptation in marine fouling organisms living in thermal
effluents. An important new line of research is in enzyme
kinetics, especially thermal properties, of the forms of en-
zymes discovered in environmentally-distinct populations of
Corbicula. Such work has been done in fishes (Powers et al.,
1979) and other organisms. The correlation of electrophoretic
results with physiology and environmental parameters will in-
crease our understanding of the potential for Corbicula to
spread to both warmer and colder waters.

One confusing aspect of Corbicula life history is the
report that the clam releases pediveliger larvae in many
localities, but that veiigers are released at other times and
places (e.g. Aldridge and McMahon, 1978). Sickel (pers.
comm.) reports that clams from the Little River released

HOAGLAND: PROBLEMS AND APPROACHES IN CORBICULA STUDY

207

pediveligers in the lab, while those from the Tennessee River
released veligers. Growth patterns of adults were also dif-
ferent. It is very unusual for a single species of mollusk to
release different stages of larve. It could be that early release
of veligers is an abnormality due to the stress of transport
to the laboratory, yet reports suggest that the veligers are
healthy. The question certainly requires more careful in-
vestigation, since stage and size of released larvae determine
their ability to enter water systems. Planktonic larvae are
highly unusual in riverine environments.

LOCAL POPULATIONS AND CONTROL

Assuming that the species of Corbicula and its general
biological features are known as discussed in the above sec-
tion, local factors of importance to control require on-site
research. One must understand the local “natural” environ-
ment, in a river, lake, or impoundment, as well as the man-
made environment. The river is the reservoir for future inva-
sions. If Corbicula is also at high density in the natural river
or if re-invasion is easy, any ameliorating action taken will
be temporary. If the artificial environment has a much higher
density than the river itself, one might ask why. What is it
about plant design, especially water velocity, that provides
an ideal habitat? A comparison of existing and new en-
vironmental data such as water chemistry, flow rates,
sedimentation, and substratum for local man-made and natural
areas with and without Corbicula is in order. Graney et ai.
(1980), for example, analyzed the influence of substratum and
temperature on population dynamics of Corbicula near a
plant, but such information correlating plant activities and the
natural population dynamics is rare. The physical constraints
on settlement and survival of pediveligers and post-
metamorphosed juveniles are still not defined. Seasonal pat-
terns of settlement, growth, and migration in the natural area
are particularly important to correlate with control activities.

Corbicula in the United States does occur in many
types of waters in terms of chemical composition, physical
properties, and temperature regimes. In fact, one important
piece of research would be to pick a geographical area and
systematically collate existing information on water proper-
ties (and variations therein) where Corbicula exists and where
it does not, but is expected, zoogeographically. Such work
has begun, starting with the more general problem of water
quality inside and outside the range of Corbicula, as reported
elsewhere in this volume (Counts, 1986).

An example of this approach, where natural and out-
break populations of a nuisance organism were compared,
is a study of hookworm in India. It was found that in West
Bengal, two species of hookworm were endemic and a high
proportion of people were infected, yet worm burden (densi-
ty of hookworm in people) was low. Comparison with other
areas with high worm burden suggested that social and
agricultural habits played a role. Conversion of land to grow-
ing non-edible products such as mulberry had the potential
to turn a mild disease into an outbreak situation because peo-
ple would then have greater contact with the disease (Schad,
1971; Schad et ai, 1975). Analysis only of outbreak areas
is insufficient to pick out essential differences in ecology that

cause the outbreak. Similar reasoning suggests that we
should study carefully the places where Corbicula was in-
troduced, then died back. A disease organism as was sug-
gested in a preliminary, inconclusive report by Sickel and
Lyles (1 981 ) or a change in water quality are possible findings
with implications for local control. Few parasitologists or in-
vertebrate pathologists have been involved in Corbicula
research.

It would seem to be valuable also to compare local
populations of Corbicula in North America with those in Asia
that are natural and are not nuisance populations, to see how
the natural populations are regulated. However, the Asian
populations are usually reduced by heavy human predation.
This solution does not seem to be possible in North America,
as American Asians seem no longer attracted to this food
source! I observed Corbicula in Lake Er Hai, Yunnan Pro-
vince, People’s Republic of China, to be very abundant from
the lake shore to water of at least 10 meters, despite heavy
fishing with nets and shoreline gathering. Mounds of shells
were the result of years of human predation. Corbicula was
only one of three mollusks consumed on regular basis; one
other remained extremely abundant in the lake as well. The
balance of invertebrates in the lake that would exist without
interference by man could not be determined. Whether Cor-
bicula is controlled in Asia by some disease, natural predator,
or human predators in unknown. There is little hope that a
simple introduction of another Asian invertebrate, a predator,
to the United States would bring Corbicula under control. It
would quite possibly create more problems, as has happen-
ed with many other introductions of predators, such as in-
troduction of predatory snails in Hawaii, which have destroyed
the native fauna.

Several kinds of data are needed for understanding
Corbicula inside the man-made environment. First of all, ex-
actly where are the clams living, and where are only dead
shells accumulating? What are the water velocities in the af-
fected areas, relative to lab-determined values for settlement
and growth of Corbicula ? What other fouling exists, e.g., slime
and accumulation of silt? What food for Corbicula exists? One
would expect that Corbicula has catholic taste, but local nutri-
tional studies should be done to identify food organisms. If
screens, traps, backflushing, or other physical means of con-
trol have been used, their effectiveness should be compared
quantitatively to the situation before control was attempted,
Records should be kept on the numbers of shells, number
alive, size distribution, and time of year each time physical
removals are done. These data should be given to biologists
along with cost/benefit analysis on options the biologists feel
are available for control.

A coordination of in-house biological work with plant
management decisions can sometimes be facilitated by out-
side expertise. Such expertise can also effectively focus areas
of research. It is necessary to combine general off-site
laboratory studies such as flume studies or physiological and
genetic analyses with on-site work on local population and
environmental parameters. Whether on- or off-site, biologists
need enough engineering and economic information to make
reasonable suggestions for control of Corbicula. Optimally,

208

CORBICULA SYMPOSIUM

biologists should work with experts in plant design to minimize
clam habitats in new and redesigned plants. Scientists should
broaden their literature searches to include works on other
introduced and/or nuisance species. Finally it is essential that
our work be published rather than buried in private or govern-
mental reports, a fate too common in environmental biology.

ACKNOWLEDGEMENTS

I thank L. Kraemer for inspiring me to write this paper and
for reviewing the manuscript. G. M. Davis also read and criticized
the manuscript. Funds for on-site study of Corbicula were provided
by the Potomac Electric Power Company (PEPCO). Lehigh Univer-
sity’s Centers for Energy Research and for Marine and Environmental
Studies provided assistance in the production of the manuscript and
attendance at the symposium of which this manuscript is a part.

LITERATURE CITED

Aldridge, D. W. and R. F. McMahon. 1978. Growth, fecundity, and
bioenergetics in a natural population of the Asiatic freshwater
clam, Corbicula manilensis Philippi, from central Texas. Journal
of Molluscan Studies 44:49-70.

Ayala, F. J., ed. 1976. Molecular Evolution. Sinauer Associates,
Sunderland Mass. 277 pp.

Ayala, F. J. 1983. Review: Enzymes as taxonomic characters. Pp
3-26 IN: Systematics Association (Great Britain) special Volume
24, Protein Polymorphism: Adaptive and Taxonomic
Significance. G. S. Oxford and D. Rollinson, eds. Proceedings
of a Symposium, 13-15 July 1982, York, England.

Bloomer, H. H. 1939. A note on the sex of Pseudanodonta
Bourguignat and Anodonta Lamarck. Proceedings of the
Malacological Society of London 23:285-297.

Britton, J. C. 1979. Editor’s introduction. Proceedings, First Interna-
tional Corbicula Symposium. Ft. Worth, Texas, Texas Chris-
tian University Research Foundation, pp. v-vi.

Britton, J. C. and B. Morton. 1979. Corbicula in North America: the
evidence reviewed and evaluated, pp. 249-287 IN: Pro-
ceedings, First International Corbicula Symposium. J. C. Brit-
ton, ed. Ft. Worth, Texas, Texas Christian University Research
Foundation.

Chambers, S. M. 1978. An electrophoretically-detected sibling
species of “ Goniobasis floridensis" (Mesogastropoda; Pleur-
oceridae). Malacologia 17:157-162.

Counts, C. L. 1986. Zoogeography and chronology of the invasion
of the United States by Corbicula fluminea. IN: J. C. Britton,
ed. Proceedings, Second International Corbicula Symposium.
American Malacological Bulletin Special Edition No. 2. 7-39.
Davis, G. M. 1983. Relative roles of molecular genetics, anatomy,
morphometries and ecology in assessing relationships among
North American Unionidae (Bivalvia). Pp. 193-222. IN
Systematics Association (Great Britain) Special Volume 24,
Protein Polymorphism: Adaptive and Taxonomic Significance.
G. S. Oxford and D. Rollinson, eds. Proceedings of a Sym-
posium, 13-15 July 1982, York, England.

Davis, G. M. 1984. Genetic relationships among some North Ameri-
ican Unionidae (Bivalvia): Sibling species, convergence, cla-
distic relationships. Malacologia 25(2): 629-648.

Fuziwara, T. 1975. On the reproduction of Corbicula leana Prime.
Venus 34:54-56.

Fuziwara, T. 1979. A hypothesis on the distribution of Corbicula leana
Prime. Venus 38(1):68-69.

Gallardo, C. S. 1977. Two modes of development in the mor-
phospecies Crepidula dilatata (Gastropoda: Calyptraeidae)

from Southern Chile. Marine Biology 39:241-251.

Gifford, C. A. 1974. The Asiatic clam, Corbicula sinensis, in Lake
Ochlawaha, Florida. Association of Southeastern Biologists
Bulletin 21(2):56.

Graney, R. L., D. S. Cherry, J. H. Rodgers, and J. Cairns, Jr. 1980.
The influence of thermal discharge and substrate composi-
tion on the population structure and distribution of the Asian
clam, Corbicula fluminea, in the New River, Virginia. Nautilus
94: 130-134.

Grassle, J. P. and J. F. Grassle. 1976. Sibling species in the marine
pollution indicator Capitella (Polychaeta). Science 192:567-569.

Gunning, G. E. and R. D. Suttkus. 1966. Occurrence and distribu-
tion of Asiatic clam, Corbicula leana in Pearl River, Louisiana.
Nautilus 79:113-117.

Hillis, D. M. and J. C. Patton. 1982. Morphological and electrophoretic
evidence for two species of Corbicula (Bivalvia:Corbiculae) in
North America. American Midland Naturalist 108(1): 74-80.

Hoagland, K. E. 1983. Life history characteristics and physiological
tolerances of Teredo bartschi, ashipworm introduced into two
temperate-zone nuclear power plant effluents. Pp. 609-662
IN: Sengupta and Lee, eds. Waste heat utilization and Manage-
ment. Hemisphere Publishing Corp., Washington, D. C.

Hoagland, K. E. 1984a. Use of molecular genetics to distinguish
species of the gastropod genus Crepidula. Malacologia 25(2):
607-628.

Hoagland, K. E. 1984b. Use of the terms protandry, protogyny, and
hermaphroditism in Malacology. American Malacological
Bulletin 3(1):85-88.

Hubricht, L. 1963. Corbicula fluminea in the Mobile River. Nautilus
77:31.

Kraemer, L. R. 1977. Aspects of the functional morphology of the
mantle shell and mantle gill complex of Corbicula (Bivalvia:
Sphaeriacea: Corbiculidae). Bulletin of the American
Malacological Union For 1977: 25-31.

Kraemer, L. R. 1978. Corbicula fluminea (Bivalvia: Sphaeriacea): the
functional morphology of its hermaphroditism. Bulletin of the
American Malacological Union For 1978: 40-49.

Kraemer, L. R. 1983. Aspects of the functional morphology of some
freshwater bivalve nervous systems: effects on reproductive
process and adaptation of sensory mechanisms in the
Sphaeriacea and the Unionacea. Malacologia 25 (1): 221-239.

Kraemer, L. R. and Lott, S. 1977. Microscopic anatomy of the visceral
mass of Corbicula (Bivalvia: Sphaeriacea). Bulletin of the
American Malacological Union For 1977: 48-56.

Kraemer, L. R., C. Swanson, M. Galloway and R. Kraemer. 1986.
Biological basis of behavior in Corbicula fluminea. II. Functional
morphology of reproduction and development and review of
evidence of self-fertilization, pp. 193-202. Proceedings, Second
International Malacological Bulletin, Special Edition No
2:193-202.

Lassen, H. H. 1979. Electrophoretic enzyme patterns and breeding
experiments in Danish mudsnails (Hydrobiidae). Ophelia 18:
83-88.

Livingstone, D. R. 1981. Induction of enzymes as a mechanism for
the seasonal control of metabolism in marine invertebrates:
glucose-6-phosphate dehydrogenase from the mantle and
hepatopancreas of the common mussel Mytilus edulis L. Com-
parative Biochemistry and Physiology 69B: 147-156.

Mackie, G. L. 1973. Biology of Musculium securis (Pelecypoda,
Sphaeriidae) in temporary forest ponds, a permanent stream
near Ottawa, Canada. Ph.D. Dissertation, U. of Ottawa, On-
tario, Canada. 175 pp.

McLeod, M. 1986. Electrophoretic variation in U.S. Corbicula. IN: J.
C. Britton, ed. Proceedings, Second International Corbicula

HOAGLAND: PROBLEMS AND APPROACHES IN CORBICULA STUDY

209

Symposium. American Malacological Bulletin Special Edition
No. 2:125-132.

Mcleod, M. J. and D. M. Sailstad. 1980. An electrophoretic study
of Corbicula fluminea (Bivalvia: Corbiculacea) in the Catawba
River. Bulletin of the American Malacological Union For 1980:
17-19.

McMahon, R. F. 1977. Shell size-frequency distribution of Corbicula
manilensis Philippi from a clam-fouled steam condenser.
Nautilus 91:54-59.

Morton, B. 1982. Some aspects of the population structure and sexual
strategy of Corbicula cf fluminalis (Bivalvia: Corbiculacae) from
the Pearl River, People’s Republic of China. The Journal of
Molluscan Studies 48(1): 1-23.

Nevo, E., T. Shimony, and M. Libni. 1977. Thermal selection of
allozyme polymorphism in barnacles. Nature 267: 699-701.

Oxford, G. S. 1975. Food induced esterase phenocopies in the snail
Cepaea nemoralis. Heredity 35: 361-370.

Powers, D. A., G. S. Greavey, and A. R. Place. 1979. Physiological
correlation between lactate dehydrogenase genotype and
haemoglobin function in killifish. Nature 277: 240-241.

Prezant, R. S. and K. Chalermwat. 1984. Induction of color forms
in Corbicula. American Malacological Bulletin 2:87 (Abstract).

Price, S. C. and S. K. Jain. 1981 . Are inbreeders better colonizers?
Oecologia 49: 283-286.

Saul, H., M. J. Sinsko, P. R. Grimstad, and G. B. Craig, Jr. 1978.
Population genetics of the mosquito Aedes triseriatus:
genetic-ecological correlations at the esterase locus. American

Naturalist 112: 333-339.

Schad, G. 1971. The ecology of interacting populations of man and
hookworms in rural West Bengal. U.S. Dept, of Health, Educa-
tion, and Welfare NIH Final Progress Report for 1970: 53 p.

Schad, G., E. S. L. Soulsby, A. B. Chowdhury, and H. M. Gilles. 1975.
Epidemiological and serological studies of hookworm infec-
tion in endemic areas in India and West Africa. Nuclear Techni-
ques in Helminthology Research, International Atomic Energy
Agency, Vienna: 41-54.

Selander, R. K. and R. O. Hudson. 1976. Animal population struc-
ture under close inbreeding: the land snail Rumina in southern
France. American Naturalist 110:695-718.

Selander, R. K. and D. W. Kaufman. 1975. Genetic population struc-
ture and breeding systems, pp. 27-48 IN: C. L. Market ed.,
Isozymes IV: Genetics and Evolution, Academic Press, N.Y.

Sickel, J. B. and M. B. Lyles. 1981. Chaetogaster limnaei
(Oligochaeta: Naididae) inhabiting the mantle cavity of the
Asiatic clam, Corbicula fluminea, in Barkley Lake, Kentucky.
Veliger 23(4): 361-362.

Smith, M. H., J. Britton, P. Burke, R. K. Chesser, M. W. Smith, and
J. Hagen. 1979. Genetic variability in Corbicula, an invading
species, pp. 243-248 IN: J. C. Britton, ed., Proceedings, First
International Corbicula Symposium. Fort Worth, Texas, Texas
Christian University Research Foundation.

Wilkins, N.P. and D. O’Reagan. 1980. Genetic variation in sibling
species of Littorina. Veliger 22:355-359.

BIOLOGY OF CORBICULA IN CATFISH REARING PONDS

JOSEPH K. BLJTTNER1

DEPARTMENT OF ZOOLOGY AND FISHERIES RESEARCH LABORATORY
SOUTHERN ILLINOIS UNIVERSITY
CARBONDALE, ILLINOIS 02901, U.S.A.

ABSTRACT

Corbicula were stocked with channel catfish, Ictalurus punctatus , in two of four 0.06 ha ponds.
In 1977 clams were placed in cages suspended in the water of two ponds at 828 and 1010 kg/ha.
In 1979 clams were stocked on the substratum of two ponds at 1717 and 1222 kg/ha. In both years
two similar ponds received no clams and served as control ponds. All ponds were mechanically cir-
culated and stocked with 300 channel catfish fingerlings in 1 977. None of the ponds were circulated
in 1979 and each pond received 302 catfish fingerlings. Corbicula survived and reproduced in both
1977 and 1979, but survival and reproduction were insufficiet to maintain stock density. The clam
had no significant (P>0.05) beneficial or detrimental effect on catfish survival, growth or feed conver-
sion, but was associated with improved water quality. Level of dissolved oxygen, rate of net produc-
tion, and rate of net respiration were greater and turbidity was lower in ponds with Corbicula. Signifi-
cant (P<0.05) improvement of water quality occurred only in 1977. At densities stocked, Corbicula
had a variable effect on total abundance, relative abundance, and diversity of plankton. I conclude
that Corbicula, which is commonly viewed as a nuisance, may be an unrealized asset. The clam has
potential as an aquaculture organism and biological filter.

The Asiatic clam, Corbicula fluminea, (Muller) is wide-
ly established throughout the United States (Clarke, 1981;
Counts, 1981; Nichols and Domermuth, 1981; McMahon,
1982; and Scott-Wasilk et a/., 1983). Although the clam is
tolerant of a wide variety of habitats it has been studied most
intensively in large lakes and reservoirs (Dreier and Tranquilli,
1980; Sickel et a!., 1981) and flowing waters (Sinclair and
Isom, 1963; Sickel, 1976, 1979). Few studies have examined
the biology of Corbicula in small bodies of water (Carlton,
1973) and with the exception of Chen (1976) in Taiwan,
growth studies by Buttner and Heidinger (1980) in southern
Illinois, and unpublished theses by Mabel (1970) and Busch
(1974) none have examined the biology of Corbicula in
aquaculture systems. I investigated the biology of Corbicula
in channel catfish, Ictalurus punctatus, rearing ponds and ex-
amined effects of the clam on pond biota and water quality.

MATERIALS AND METHODS

Between 1 June and 20 September 1977 and between
1 July and 4 October 1979 Corbicula were stocked with chan-
nel catfish in two of four 0.06 ha ponds. In 1977 water in all
ponds was mechanically circulated at approximately 86
//sec; in 1979 none of the ponds were circulated. Number

’Present Address: Department of Biological Sciences
State University of New York
College at Brockport
Brockport, New York 14420, U.S.A.

of clams stocked (Table 1) was determined by availability and
filtration rate, which averaged 347 ml/h/clam (Buttner and
Heidinger, 1981). Clam density was always sufficient to filter
one volume of pond water ( ~ 530 m3) each week (once every
168 h). In 1977 Corbicula were placed in cages suspended
in the water of ponds 1 and 16 at 828 and 1010 kg/ha, respec-
tively. In 1 979 clams were stocked directly on the substratum
of ponds 1 and 16 at 1717 and 1222 kg/ha, respectively. In
1977 additional clams were introduced; on 4 September 93.6
kg were obtained and stocked in cage D in pond 16 (Table
1). In both years ponds 2 and 17 received no clams and ser-
ved as controls. On 17 April 1977 each of the four ponds were
stocked with 300 channel catfish fingerlings that weighed ap-
proximately 49 g each and on 2 May 1 979 each pond receiv-
ed 302 channa! catfish that weighed approximately 60 g each.
Catfish were fed Purina Trout Chow #6 at 2% of their body
weight daily, six days per week. Feeding rates were adjusted
weekly and all ponds received equal quantities of feed.

Corbicula survival, distribution, and reproductive suc-
cess were monitored. Corbicula survival was determined by
weighing all clams retrieved from cages between 28
September and 8 October 1977, except for cage D added
to pond 16 on 4 September 1977. Five 2000 g samples of
clams were collected from cage D, the proportion of live clams
determined, and percentage survival computed. Between 7-9
October 1979 clam survival was determined by sampling the
area stocked with clams. Clam distribution was not random
so the sampling effort was weighted and areas of greatest
clam density received more effort. Total area sampled in pond

American Malacological Bulletin, Special Edition No. 2(1 986):21 1-218

211

212

CORBICULA SYMPOSIUM

Table 1. Corbicula stock, harvest, and survival data for clams in 0.06 ha channel catfish rearing ponds in 1977 and 1979.

Year Pond Stock Weight Turn3 Harvest Weight Survival Turn3

no. date stocked over date harvested (%) over

(kg.) (hrs.) (kg.) (hrs.)

1977

1-2 June

Cage

28 Sept.

Cage

A 21.9

A 7.0

32.0

B 23.3

B 9.5

40.8

C 4.5

C 2.3

51.1

Total 49.7

Total 18,8

Avg. 37.8

158

Cage

A 24.4

A 6.4

26.2

B 16.3

B 5.4

33.1

C 19.6

C 9.1

45.7

Subtotal 60.6

Subtotal 20.1

Avg. 34.5

141

4 Sept.

D 93.6

D 83.4

89.1

Total 104.3 28

1979b 1 27 June- 103.0 29 7-9 Oct. 82.9 80.5 36

16 1 July 73.3 41 56.1 76.5 77

Approximate time required by clams to filter water volume of each pond at stock and harvest densities.

bln 1979 all clams were stocked directly on the substratum. At harvest all Corbicula from 16.2% and 9.0% of the area with clams in ponds
1 and 16, respectively, were collected and used to compute harvest weight and percentage survival.

1 was 33.6 m2 (16.2% of the area with clams) and in pond
16 was 10.5 m2 (9.0% of the area with clams).

Distribution of Corbicula on the substratum was deter-
mined in 1979. Effect of water temperature, clam distribu-
tion as number/m2 and clam depth in the substratum on clam
survival were examined.

Reproductive success of Corbicula was determined by
periodic sampling of the substratum and water for immature
clams and by the presence of young-of-the-year clams at time
of catfish harvest.

Effect of Corbicula on catfish survival, growth and
feed conversion (weight of feed presented to catfish div-
ided by weight gain of catfish) was determined at harvest.

Table 2. Channel catfish stock, survival, and growth data for ponds 1

Dissolved oxygen (D.O.), diurnal production and noc-
turnal respiration rates, water temperature, pH, turbidity,
alkalinity and level of nitrogenous wastes were monitored
throughout the study to describe the environmental situation
and to determine if presence of Corbicula affected water quali-
ty. D.O., water temperature and pH were monitored daily just
below the surface and at 1 .0 m depth at dawn and dusk in 1977.
In 1979 D.O. and water temperature at surface, 0.5 m and
1.0 m were measured 3 times per week at 0700-0900 and
1 600-1 800; pH was determined weekly from a single sample
collected at 0.5 m. Reported values are the average of all
depths sampled. Sampling frequency increased if D.O. fell
below 3.0 mg/I. Net production and respiration rates were

2, 16 and 17 in 1977 and 1979.

Year

Food fed
per pond
(kg)

Pond

number

Clams

present

No. catfish
stocked

No. catfish
harvested

Survival

(%)

Aver. wt.

stocked

(g)

Aver. wt.
harvested

(g)

Feed

conversion

Yield
(kg /ha)

1977

111

Yes

300

288

96.0

49.0

387.8

1.14

1861

300

266

88.7

49.0

394.1

1.21

1747

Yes

300

296

98.7

49.0

366.2

1.18

1807

300

288

96.0

49.0

395.0

1.11

1896

Yes

302

293

97.0

302

296

98.0

Yes

302

295

97.7

302

296

98.0

60.0

342.2

1.37

1671

60.0

345.8

1.34

1706

60.0

345.4

1.34

1968

60.0

388.2

1.16

1915

1979

113

BUTTNER: BIOLOGY OF CORBICULA IN CATFISH PONDS

213

computed from level of D.O. observed at dawn and at dusk.
Turbidity, alkalinity and nitrogenous wastes (total ammonia-
nitrogen, TAN; nitrite-nitrogen, N02 -N; and nitrate-nitrogen,
N03 -N) were measured weekly in both years by a single water
sample collected at 0.5 m. A polarographic meter was used
to measure D.O. and temperature; pH was determined with
a pH meter, turbidity with a turbidimeter, TAN by colorimetry
in 1977 and by an ion specific electrode in 1977 and 1979,
N02-N and N03-N colorimetrically, and alkalinity
potentiometrically.

Plankton were quantitatively sampled each week.

Table 3. Summary of descriptive water quality parameters measured
in 1977 and 1979.

Parameter

Minimum

value

Maximum

value

Alkalinity

130

(mg/I CaC03)

Turbidity

122

(JTU)

Dawn

7.06

8.58

Dusk

7.24

8.94

Temperature (C)

Dawn

15.0

31.5

Dusk

17.6

34.0

Oxygen (mg/I)

Dawn

1.09

9.59

Dusk

1.47

13.22

Phytoplankton were collected with an integrated column
sampler (Buttner, 1981), preserved in Lugols solution (20 ml
KI, 10 g 1, 20 ml glacial acetic acid, and 200 ml distilled water),
identified to the generic level and counted as plants of each
genus (Vollenweider, 1969). Zooplankton were collected by
vertical tow from substratum to surface with a plankton net
(a 35 /xm mesh in 1977 and a 140 /xm mesh in 1979).
Zooplankton were preserved in 70% ethanol, identified to the
species level, and counted. Plankton diversity was computed
using the formula developed by Gleason (1922) and modified
by Margalef (1958):

Ln(N)

where H is taxa diversity, S is number of taxa, and Ln(N) is
the natural log of the total number of plankton. Effect of Cor-
bicula on plankton abundance, composition, and diversity was
examined statistically.

All statistical analyses were conducted with the
Statistical Analysis System computer package (SAS Institute,
1979). Effects of Corbicula on channel catfish were examin-
ed by ANOVA. Effects of Corbicula on plankton and water
quality were examined by multiple regression analyses with
the linear and quadratic effect of water temperature and week
sampled as covariates.

RESULTS

Corbicula survived and reproduced in the catfish rear-
ing ponds, both when suspended in the water and when
stocked on the substratum. Survival of clams suspended in
circulated ponds was poor and averaged 36.0% (excluding

Table 4. Average dissolved oxygen (D.O.), net diurnal production and net nocturnal respiration rates, and turbidity for four 0.06 ha channel
catfish rearing ponds with and without Corbicula in 1977 and 1979. Numbers in parentheses are standard deviations.

Year

Corbicula3

No. days

D.O.

Net diurnal

Net nocturnal

Turbidity

present

dawn D.O.

dawn

dusk

02 produced

02 consumed

(JTU)

below 3.0

(mg/I)

(g/m2/h)

1977

Yes

5.29x

7.33x

0.1 48x

0,201x

60. 4X

1 June-

(1.31)

(1.92)

(0.099)

(0.127)

(27.0)

20 Sept.)

4.81

6.30

0.110

0.144

79.4

(1.34)

(1.73)

(0.077)

(0.089)

(17.9)

Sample size b

448

444

1979

Yes

4.31

6.80

0.301

0.166

59.5

1 July-

(1.54)

(1.97)

(0.165)

(0.106)

(17.0)

4 Oct.)

3.99

6.35

0.291

0.153

66.6

(1.53)

(2.40)

(0.177)

(0.101)

(28.8)

Sample size b,c

148

108

a 'Corbicula stocked in ponds 1 and 16 on 1-2 June 1977 and 27 June- 1 July 1979. Clams were absent from ponds 2 and 17.
DUnless stated otherwise sample size is the sum of an equal number of observations for ponds with and without Corbicula.
cln 1979 sample size for net nocturnal oxygen consumption was 30 and 29 for ponds with and without Corbicula, respectively,
indicates a significant (P < 0.05) difference between ponds with and without Corbicula.

214

CORBICULA SYMPOSIUM

cage D). Clam survival on the substratum averaged 78.8%
(Table 1).

The presence of many immature clams, approximately
0.20 mm in shell length (SL), identified from preserved

Table 5. Average level of nitrogenous wastes for 0.06 ha channel
catfish rearing ponds with and without Corbicula in 1977 and 1979.
Numbers in parentheses are standard deviations.

Year

Corbiculaa

present

Total
ammonia
NH3 + 4-N
(mg/I)

Nitrite

no2-n

(mg/I)

Nitrate

N03-N

(mg/I)

1977

Yes

0.233x

0.070

0.645

(1 June-

(0.200)

(0.059)

(0.536)

20 Sept.)

0.060

0.072

0.591

(0.055)

(0.051)

(0.482)

Sample sizeb,c

1979

Yes

0.772

0.042x

0.592

(1 July-

(0.500)

(0.027)

(0.402)

4 Oct.)

1.175

0.029

0.492

(0.936)

(0.011)

(0.212)

Sample size

aCorbicula stocked in ponds 1 and 16 on 1-2 June 1977 and 27 June-
1 July 1979. Clams were absent from ponds 2 and 17.
b Unless stated otherwise sample size is the sum of an equal num-
ber of observations for ponds with and without Corbicula.
cSample size for total ammonia in 1977 was 14 and 13 for ponds
with and without Corbicula, respectively,
indicates a significant (P < 0.05) difference between ponds with
and without Corbicula.

zooplankton samples collected in 1979 indicate Corbicula
were fertile and spawned successfully in the catfish rearing
ponds. Several clams less than 10 mm SL (minimum size
stocked) were collected when ponds were drained. The
presence of these clams confirms that some recruitment of
Corbicula occurred in the catfish rearing ponds.

Corbicula usually clumped together, both when placed
in the cages and when stocked on the substratum. In cages
Corbicula formed dense masses of many clams. On the
substratum clams frequently clumped together in tight
masses of 5 to 20 clams/50 cm2 or in loose clusters of 50
to 100 clams/3000 cm2. Clams were usually found in the
substratum rather than upon it. An inverse relationship
(P < 0.05) was observed between clam size and depth in the
substratum; smaller clams were located deeper in the
substratum. Most clams were collected from the upper 5 cm
of the substratum.

Channel catfish survival, growth, and feed conversion
were not affected (P>0.05) by the presence of Corbicula
(Table 2). Catfish survival for all ponds summed averaged
96.3% (S.D. = 3.2%) and average weight of catfish at harvest
was 371 g (S.D. = 23 g). Catfish feed conversion (weight of
feed presented to catfish divided by weight gain of catfish)
averaged 1.23 (S.D. = 0.10).

Water quality parameters were monitored primarily to
describe the system in which the clam was studied (Table
3), but the presence of Corbicula did influence water quality
(Tables 4 and 5). Circulation of ponds in 1977 increased tur-
bidity, broke up thermal stratification, and increased average
temperature of the water by 2 to 4°C.

In 1977 D.O. was significantly (P<0.05) greater in
ponds with Corbicula stocked in cages suspended in the water
of circulated ponds than in ponds without the clam (Table 4).
In 1979 the presence of Corbicula stocked upon the
substratum of uncirculated ponds had no significant (P> 0.05)
effect on D.O. (Table 4). Incidence of D.O. at or below critical

Table 6. Total abundance of phytoplankton and zooplankton and average phytoplankton generic and zooplankton species diversity in 0.06
ha channel catfish rearing ponds with and without Corbicula in 1977 and 1979. Numbers in parentheses are standard deviations.

Year

Corbiculaa

Sample

Phytoplankton

Zooplankton

present

size

abundance

generic

abundance

species

(No.//x106)

diversity

(No.//x102)

diversity

1977

Yes

14.9X

2.4

23. 4X

2.3

(1 June-

(13.4)

(0.5)

(27.1)

(0.5)

20 Sept.)

9.6

2.5

10.8

2.4

(5.8)

(0.5)

(6.0)

(0.6)

1979

Yes

31.4

2.6X

4.6

2.4X

(1 July-

(37.3)

(0.5)

(3.0)

(0.5)

4 Oct.)

31.8

2.9

3.6

2.0

(48.6)

(0.5)

(1.9)

(0.5)

aCorbicula stocked in ponds 1 and 16 on 1-2 June 1977 and 27 June-1 July 1979. Clams were absent from ponds 2 and 17.
indicates a significant (P < 0.05) difference between ponds with and without Corbicula.

BUTTNER: BIOLOGY OF CORBICULA IN CATFISH PONDS

215

levels (3.0 mg/I ~ 40% saturation) at dawn was less frequent
in ponds with Corbicula than in ponds without the clam (Table
4). The reduction in frequency of critical oxygen levels at dawn
in the presence of Corbicula was significant (P<0.05) only
in 1977.

Average rates of net diurnal production and net noc-
turnal respiration were computed (Table 4). Net diurnal pro-
duction and net nocturnal respiration rates, as determined
by changes in D.O., were greater in ponds with Corbicula,
but the differences were significant (P<0.05) only in 1977.

Turbidity was reduced in ponds with Corbicula (Table
4). Observed reduction in turbidity was significant (P < 0.05)
only in 1977.

Levels of nitrogenous wastes were occasionally
greater in ponds with Corbicula than in ponds without the clam
(Table 5). TAN was significantly (P<0.05) greater in ponds
with Corbicula in 1977, but not in 1979. N02 -N levels were
similar in all ponds in 1977, but significantly greater (P<0.05)
in ponds with Corbicula only in 1979. Presence of Corbicula was
not associated (P>0.05) with changes in level of N03 -N.

Seventy-one genera of phytoplankton and 69 species
of zooplankton were collected and identified. Phytoplankton
abundance ranged from 2.6x1 06 to 2.7x1 08 plants per liter
and zooplankton numbers ranged from 76 to 1 .2x1 04 per liter.
Phytoplankton populations were dominated by green algae,
but blue-green algae and diatoms occasionally became abun-
dant. In 1977 zooplankton populations were dominated by
rotifers and cladocerans; in 1979 calanoid copepods and
cladocerans dominated. Phytoplankton generic and
zooplankton species diversity ranged from 2.4 to 2.9.

Presence of Corbicula was correlated with changes in
plankton abundance, composition, and diversity (Table 6).
In 1977 Corbicula was associated with significantly (P<0.05)
greater numbers of phytoplankton and zooplankton, but in
1979 no significant (P>0.05) difference in plankton abun-
dance was observed between ponds with and without Cor-
bicuia. Presence of Corbicula was not correlated (P>0.05)
with changes in composition of phytoplankton, and the only
significant (P< 0.05) effect on composition of zooplankton was
an apparent increase in proportion of cyclopoid copepods in
ponds with Corbicula in 1979. In 1979 phytoplankton generic
diversity was significantly (P < 0.05) lower and zooplankton
species diversity was significantly (P < 0.05) higher in ponds
with Corbicula ; in 1977 plankton diversity was smaller in all
ponds.

DISCUSSION

SURVIVAL, GROWTH AND REPRODUCTIVE SUCCESS OF
CORBICULA

Survival of Corbicula stocked at 828 and 1010 kg/ha
in cages suspended in the water of mechanically circulated
0.06 ha channel catfish rearing ponds averaged 36%. Sur-
vival of clams stocked at 1222 and 1717 kg/ha on the
substratum of uncirculated ponds averaged 79%. These
values are superior to the 2% survival observed by Habel
(1 970) for Corbicula stocked at 6880 to 40,860 kg/ha in cages
and on the substratum of 0.0007 ha catfish rearing pools. The

79% survival rate exceeds the 62% observed by Busch (1974)
for clams stocked on the substratum, but the 36% survival
rate is lower than the 88% observed by Busch (1974) for
clams suspended in the water. Busch used the same system
and approximate clam density as Habel (1970). Apparently
Corbicula can survive in channel catfish rearing systems, but
substantial mortality occurs.

I attribute Corbicula mortality to high temperatures
(>33°C), reduced ability to uptake oxygen at intermediate
temperatures (25 to 30°C), and relatively poor tolerance to
low levels of oxygen. Mattice and Dye (1976) and Mattice
(1979) found Corbicula survived indefinitely at 32°C, while a
30 min. exposure at 33°C produced mortalities. Mortality in-
creased with temperature until 100% mortality occurred after
a 30 min. exposure at 43°C. Cherry et at., (1980) observed
temperature induced mortalities at 36°C. McMahon and
Aldridge (1 976) found oxygen uptake increases to 25°C, but
decreases greatly between 25 to 30°C. Above 30°C oxygen
uptake increases slightly until thermal stress becomes fatal.
Compared with other freshwater mollusks Corbicula is con-
siderably less tolerant to low levels of oxygen (McMahon,
1979). Oxygen uptake rate of clams at 70% oxygen satura-
tion is 1/2 their rate at 100% saturation (McMahon and
Aldridge, 1976). In my study water temperatures in excess
of 33°C commonly occurred in the summer between 1200
and 1800 h and contributed to Corbicula mortality.
Temperatures above 25°C were frequently associated with
reduced levels of D.O. (<40% saturation) at dawn in channel
catfish rearing ponds and probably contributed to clam
mortalities.

Since some Corbicula survived in channel catfish rear-
ing ponds at least some of the clams were capable of
withstanding or avoiding stressful temperatures and low
oxygen. Caged clams could not avoid stressful temperatures
and the dense clumps of Corbicula in cages also restricted
water flow and promoted a localized area of reduced oxygen
that contributed to the high rate of clam mortality observed.
Corbicula stocked on the substratum were distributed in less
dense clusters than clams in cages and frequently burrowed
to a depth of 2 to 5 cm, which was 1 to 2°C cooler than the
water immediately above it. Superior clam survival observed
in 1979 (uncirculated ponds) was probably due to water
temperatures 2 to 4°C cooler than those in circulated ponds
(1977) and because the uncaged clams were able to disperse
and avoid localized oxygen depletions. It is possible that Cor-
bicula from the extreme southern United States may be more
tolerant of conditions in channel' catfish rearing ponds than
were clams from southern Illinois.

Growth of Corbicula in catfish rearing ponds was
previously reported (Buttner and Heidinger, 1980). Summer
and fall growth rates were similar to rates observed by other
investigators (O’Kane, 1976; Sickel, 1976; Britton etal., 1979;
Dreier and Tranquilli, 1980). Winter growth rates were not
comparable with values reported from other studies con-
ducted in the deep south where water temperatures in winter
often exceed the average 3.0°C observed by Buttner and
Heidinger (1980).

Corbicula reproduced in catfish rearing ponds, but

216

CORBICULA SYMPOSIUM

recruitment was insufficient to maintain stocking densities.
Collection of several 8-1 0 mm SL and many 0.20 mm SL Cor-
bicula in the late summer and early fall document the occur-
rence of at least two spawns. Late summer or early fall spawn
of Corbicula in 0.0007 ha catfish rearing pools was also
reported by Busch (1974) in Alabama.

Although Corbicula spawns continually, biannual
spawning peaks exist that are correlated with water
temperature (Heinsohn, 1958; Sickel, 1976; Eng, 1979; Mat-
tice, 1979). Spawning is inhibited by temperatures less than
12-16°C or greater than 24-26°C (Heinsohn, 1958; Britton et
al. , 1979; Eng, 1979). Temperatures above 32°C prohibit
spawning (Aldridge and McMahon, 1978). Optimum spawn-
ing temperature is between 22 and 24°C (Britton etai, 1979;
Dreier and T ranquilli, 1 980). Temperatures suitable for spawn-
ing occurred immediately after clam introduction in June and
at the end of August or in early September. However,
throughout most of the study period temperatures common-
ly exceeded the 26°C known to inhibit Corbicula spawning.

Recruitment of spawned Corbicula was poor and few
clams survived to 8-10 mm SL. High water temperatures
(commonly >26 C) and low D.O. (frequently ^=40% satura-
tion) inhibited spawning and may have killed immature clams.
Predation by crayfish, Orconectes immunis, possibly re-
duced abundance of juvenile clams. Other crayfish species,
Procambarus clarkii and Cambarus bartoni, prey on juvenile
Corbicula (Auerbach and Reichle, 1980; Covich et al., 1981).
Ingestion of juvenile Corbicula by channel catfish was not
documented, although more than 50 gut samples were ex-
amined. Busch (1974) also found channel catfish predation
upon Corbicula in 0.0007 ha pools negligible.

EFFECT OF CORBICULA ON CHANNEL CATFISH.

Channel catfish survival, growth and feed conversion
compared favorably with values reported for commercial
operations (Tiemeier and Deyoe, 1973; Gray, 1978; Piper et
al., 1982). Although Corbicula had no observable effect on
catfish, its presence was correlated with improved water qual-
ity. Since catfish survival, growth and feed conversion de-
pend upon the environmental situation (Thurston et al., 1979;
Allen and Kinney 1981; Piper et al., 1982), they may be
enhanced only if Corbicula promotes water quality.

EFFECT OF CORBICULA ON WATER QUALITY

Corbicula reduced the biological oxygen demand by
cropping detritus, bacteria, and phytoplankton. Corbicula
cropping of the phytoplankton possibly stimulated produc-
tion and respiration by promoting an active vigorous algal
population. Busch (1974) found Corbicula stocked in excess
of 6850 kg/ha decreased the incidence of critically low D.O.
at dawn, but had no consistent effect on average D.O. at
dawn.

Photosynthesis was promoted by turbidity reduction,
which increased the depth of the euphotic zone. The limited
reduction of turbidity associated with Corbicula (Table 4) was
most likely due to pond circulation in 1977, to the activity of
the abundant crayfish, Orconectes immunis, and to the
relatively low density of clams stocked. Densities of Corbicula

greater than 6800 kg/ha apparently are more effective in
reducing turbidty than the lower densities used in my study
(Habel, 1970; Busch, 1974; Haines, 1979).

Nitrogenous wastes were occasionally higher in ponds
with Corbicula than in ponds without the clam; significant ef-
fects were observed in 1977 for TAN and in 1979 for N02-N.
However, the increased level of nitrogenous wastes is likely
independent of the clam. In laboratory studies that I con-
ducted with Corbicula isolated in 50 ml of pond water for 24 h,
no buildup of TAN was observed. Further, the only substan-
tial source of nitrogenous waste unique to ponds with Cor-
bicula was decaying clam tissue. Given the weight and com-
position of putrefying clam tissue the maximum daily increase
in TAN would be approximately 0.0006 mg/I. This amount is
insufficient, by an order of 2 to 3 magnitudes, to account for
differences observed between ponds with and without Cor-
bicula. At no time did nitrogenous wastes in any of the ponds
attain toxic levels (Colt and Armstrong, 1981).

EFFECTS OF CORBICULA ON PLANKTON

Plankton populations were monitored each week by
a single sample obtained from each pond. Samples so col-
lected contain much inherent variability (Verduin 1959;
Wetzel, 1975). However, the sampling regime would permit
documentation of trends and differences between ponds with
and without Corbicula. Unfortunately another variable existed,
planktivorous fish. The fish, mainly gizzard shad, Dorosoma
cepedianum, and sunfish, Lepomis spp., were accidently in-
troduced into most experimental ponds through the water
supply. Contaminant fishes were planktophagic and could
reduce total abundance, alter composition and affect diver-
sity of plankton populations. Effects of these fish on plankton
populations were examined statistically. Weight of contami-
nant fish and abundance of plankton, both zooplankton and
phytoplankton, was positively correlated (ANOVA, P<0.05)
in 1977. In 1979 a significant (ANOVA, P<0.05) negative cor-
relation existed between weight of introduced fish and abun-
dance of zooplankton. Plankton most dramatically affected
were the crustaceans.

Total abundance, composition, and diversity of
plankton were similar to values previously reported for
eutrophic waters (Pennak, 1946; Margalef, 1958; Meyer and
McCormick, 1971; Wetzel, 1975). Phytoplankton abundance
was greater in 1 979 than in 1 977, probably due to the absence
of mechanical circulation which resuspended silt and main-
tained increased turbidity in 1977. Green algae and blue-
green algae dominated the phytoplankton populations in all
ponds as they normally do in fertile waters during the sum-
mer (Phillips and Whitford, 1959; Ewing and Dorris, 1970;
Boyd, 1973). Zooplankton samples were dominated by rotifers
and cladocerans in 1977, but not in 1979. This is probably
an artifact related to differences in sampling technique. A 35
nm mesh net that retained rotifers was used in 1977 but the
140 fim mesh net used in 1 979 permitted passage of rotifers.

Although invertebrates and fish have suppressed
numbers of plankton (Pennington, 1941; Dunseth, 1977;
Porter, 1973; Zaret, 1980), Corbicula did not consistently alter
plankton abundance or composition. Algal populations were

BUTTNER: BIOLOGY OF CORBICULA IN CATFISH PONDS

217

not reduced by Corbicula, probably due to the low number
of clams stocked, low turnover rate (ponds water volume
filtered once every 2 to 7 days) and the relatively rapid doubl-
ing time of algae. In 1 977 the statistically significant (P < 0.05)
increase in plankton abundance observed in ponds with Cor-
bicula and the significant (P<0.05) increase in abundance
of cyclopoid copepods in 1979 were probably independent
of the presence of Corbicula. Contaminant fish and aquatic
insects (Notonectidae) were present in all ponds, particular-
ly those with Corbicula in 1979. These organisms prey on
zooplankton and can reduce their abundance and change
the composition of plankton (Brooks and Dodson, 1965;
Brooks, 1968; Applegate and Mullan, 1969; Zaret, 1980).

The reason for the significant decrease in
phytoplankton generic diversity and increase in zooplankton
species diversity observed in 1979 for ponds with Corbicula
is unknown. Perhaps the greater density and increased sur-
vival of Corbicula and the large number of zooplankton pre-
sent eliminated or reduced certain green algae and diatoms.
Increased zooplankton species diversity may be related to
the absence of contaminant fish in pond 16 and presence
of relatively large numbers of contaminant fish in pond 1
(ponds 2 and 17 had approximately the same type and
number of contaminant fish). Such a difference would pro-
mote the development of additional niches and increase
diversity.

CONCLUSION

Corbicula has been viewed commonly as a nuisance or pro-
blem species. However, the attributes that promote this
characterization indicate it may be an unrealized asset. Cor-
bicula has potential as an aquaculture organism and
biological filter in fish rearing ponds (Buttner, submitted). The
clam grows rapidly, is low on food chain, effectively removes
suspended materials, is tolerant of a wide variety of en-
vironmental conditions, does not compete with channel cat-
fish, and possibly promotes superior water quality. Commer-
cial markets of Corbicula as a bait, in the aquarium trade,
as a food for domesticated animals, and for human consump-
tion already exist (Fox, 1971; Chen, 1976; Sickel eta!., 1981;
Britton and Sickel, 1982).

ACKNOWLEDGMENTS

I would like to acknowledge the assistance provided for this
investigation by the Department of Zoology and the Fisheries
Research Laboratory at Southern Illinois University at Carbondale.
The study was supported, in part, by a grant from Sigma Xi, the
Scientific Research Society.

LITERATURE CITED

Aldridge, D. W. and R. F. McMahon. 1978. Growth, fecundity and
bioenergetics in a natural population of the Asiatic freshwater
clam, Corbicula martilensis Philippi, from north central Texas.
Journal of Molluscan Studies 44: 49-70.

Allen, L. J. and E. C. Kinney (eds). 1981. Proceedings of the Bio-
Engineering Symposium for Fish Culture. American Fisheries
Society. Fish Culture Section Publication Number 1. 307 pp.

Applegate, R. L. and J. W. Mullan. 1969. Ecology of Daphnia in Bull
Shoals Reservoir. Bureau of Sport Fisheries and Wildlife
Research. Report Number 74. 23 pp.

Auerbach, S. I. and D. E. Reichle. 1980. Environmental Sciences
Division Annual Progress Report. September 30, 1979. Oak
Ridge National Laboratory Publication No. 5620. 188 pp.

Boyd, C. E. 1973. Summer algal communities and primary produc-
tivity in fish ponds. Hydrobiologia. 41: 357-390.

Britton, J. C., D. R. Coldiron, L. P. Evans, Jr., C. Golightly, K. D.
O’Kane, and J. R. TenEyck. 1979. Reevaluation of the growth
pattern in Corbicula fluminea (Muller) pp. 177-192 In: J. C. Brit-
ton (ed.), First International Corbicula Symposium. Texas Chris-
tian University. Fort Worth, Texas. 313 pp.

Britton, J. C. and J. B. Sickel. 1982. Corbicula Newsletter. 7:2.

Brooks, J. L. 1968. The effects of prey size selection by lake
planktivores. Systematic Zoology 17:273-291.

Brooks, J. L. and S. I. Dodson. 1 965. Predation, body size, and com-
position of plankton. Science 150:28-35.

Busch, J. L. 1974. Asiatic clams Corbicula manilensis Philippi as
biological filters in channel catfish Ictalurus punctatus (Rafin-
esque) cultures. M.Sc. thesis. Auburn University. Auburn, AL.
95 pp.

Buttner, J. K. 1981. Asiatic Clam in Channel Catfish Rearing Ponds:
Its Biology and its Effect on Water Quality. Ph.D. dissertation.
Southern Illinois University. Carbondale, IL. 236 pp.

Buttner, J. K. Submitted. Potential of the Asiatic clam Corbicula
fluminea , as a biological filter and aquaculture organism in
fish rearing ponds. Progressive Fish-Culturist. In press.

Buttner, J. K. and R. C. Heidinger. 1980. Seasonal variations in
growth of the Asiatic clam, Corbicula fluminea, in a southern
Illinois fish pond. Nautilus. 94:8-10.

Buttner, J. K. and R. C. Heidinger. 1981. Filtration rate of the Asiatic
clam, Corbicula fluminea. Transactions Illinois Academy of
Science. 74:13-17.

Carlton, J. T. 1973. Corbicula in San Francisco, California. Nautilus.
87:87.

Chen, T. P. 1976. Culture of the freshwater clam, Corbicula fluminea.
Pp. 107-110 in Aquacultural Practises in Taiwan. Page
Brothers Limited. Norwich, England. 161 pp.

Cherry, D. S., J. H. Rodgers, Jr., R. L. Graney, and J. Cairns, Jr.
1980. Dynamics and control of the Asiatic clam in New River,
Virginia. VPI- Virginia Water Resources Center Bulletin 127.
72 pp.

Clarke, A. H. 1981 . Corbicula fluminea in Lake Erie. Nautilus 95:83-84.

Colt, J. E. and D. A. Armstrong. 1981. Nitrogen toxicity to crusta-
ceans, fish and molluscs. Pp. 34-47 In: L. J. Allen and E. C.
Kinney (eds.), Proceedings of the American Fisheries Society
Bio-Engineering Symposium for Fish Culture. American
Fisheries Society. Fish Culture Section Publication Number
1 . 307 pp.

Counts, C. L. III. 1981. Corbicula fluminea (Muller) on the Delmarva
Peninsula, Maryland. Veliger. 24: 187-188.

Covich, A. P., L. L. Dye, and J. S. Mattice. 1981. Crayfish predation
on Corbicula under laboratory conditions. American Midland
Naturalist 105:181-188.

Dreier, H. and J. A. Tranquilli. 1980. Reproduction, growth, distribu-
tion, and abundance of Corbicula in an Illinois cooling lake.
In: R. W. Lawrimore and J. A. Tranquilli (eds.), The Lake San-
crhis Study: Case History of an Illinois cooling lake. Bulletin
Illinois Natural History Survey. Number 32.

Dunseth, D. R. 1977. Polyculture of channel catfish, Ictalurus punc-
tatus. silver carp, Hypophthalmichthys molitrix, and three all-
male tilapias Sarotherodon sp. Ph.D. dissertation. Auburn
University. Auburn, AL. 61 pp.

218

CORBICULA SYMPOSIUM

Eng, L. L. 1979. Population dynamics of the Asiatic clam, Corbicula
fluminea (Muller), in the concrete-lined Delta-Mendota canal
of central California. Pp. 40-68 in J. C. Britton (ed.), Pro-
ceedings First International Corbicula Symposium. Texas Chris-,
tian University. Fort Worth, Texas. 313 pp.

Ewing, M. S. and T. C. Dorris. 1970. Algal Community and struc-
ture in artificial ponds subjected to organic enrichment.
American Midland Naturalist 83: 565-582.

Fox, R. O. 1971. Have you met Corbicula - the freshwater invader?
Tabulata 1:3-5.

Gleason, H. A. 1922. On the relation between species and area.
Ecology 3:152-162.

Gray, D. C. 1976. The Biology of Channel Catfish Production. Ex-
tension Service Circular Number 535. University of Arkansas.
14 pp.

Habel, M. L. 1970. Oxygen consumption, temperature tolerance, and
filtration rate of the introduced Asiatic clam Corbicula manilen-
sis from the Tennessee River. M.Sc. thesis. Auburn Univer-
sity. Auburn, AL. 91 pp.

Haines, K. C. 1979. The use of Corbicula as a clarifying agent in ex-
perimental tertiary sewage treatment process on St. Croix,
U.S. Virgin Islands. Pp. 165-176 In: J. C. Britton (ed.), Pro-
ceedings First International Corbicula Symposium. Texas Chris-
tian University. Fort Worth, Texas. 313 pp.

Heinsohn, G. E. 1958. Life history and ecology of the freshwater clam,
Corbicula fluminea. M. A. thesis. University of California. Santa
Barbara, California. 64 pp.

Margalef, K. 1958. Temporal succession and spatial heterogeneity
in phytoplankton. Pp. 323-349 In: A. A. Buzzati-Traverso (ed.),
Perspectives in Marine Biology. University of California Press.
Berkeley, California. 621 pp.

Mattice, J. S. 1979. Interactions of Corbicula sp. with power plants.
Pp. 119-138. In: J. C. Britton (ed.), Proceedings First Interna-
tional Corbicula Symposium. Texas Christian University. Fort
Worth, Texas. 313 pp.

Mattice, J. S. and C. C. Dye. 1976. Thermal tolerance of the adult
Asiatic clam. Abstract in Corbicula Newsletter 2:8.

McMahon, R. F. 1979. Response to temperature and hypoxia in the
oxygen consumption of the introduced Asiatic clam Corbicula
fluminea (Muller). Comparative Biochemistry and Physiology
63 A: 383-388.

McMahon, R. F. 1982. The occurrence and spread of the introduced
Asiatic freshwater clam Corbicula fluminea (Muller), in North
America: 1924-1982. Nautilus 96:134-141.

McMahon, R. F. and D. W. Aldridge. 1976. Respiratory response
to temperature and low oxygen tension in Corbicula manilensis ,
Philippi. Abstract in Corbicula Newsletter 1:6.

Meyer, K. A. and J. F. McCormick. 1971. Seasonal fluctuation of
phytoplankton composition, diversity, and production in a
freshwater lake. Journal Elisha Mitchell Scientific Society 87:
127-138.

Nichols, B. L. and R. B. Domermuth. 1981. Appearance of the Asiatic
clam, Corbicula fluminea, in the Susquehana River, Penn-
sylvania. Pennsylvania Academy of Science 55:181-182.

O’Kane, K. D. 1976. A population study of the exotic bivalve Cor-
bicula manilensis (Philippi, 1841) in selected Texas reservoirs.
M.Sc. thesis. Texas Christian University. Fort Worth, Texas.
134 pp.

Pennak, R. W. 1946. The dynamics of freshwater plankton popula-
tions. Ecological Monographs 16:339-355.

Pennington, W. 1941. The control of the numbers of freshwater
phytoplankton by small invertebrate animals. Journal of

Ecology 29:204-211.

Phillips, R. C. and L. A. Whitford. 1959. A quantitative study of the
summer phytoplankton in four central North Carolina ponds.
Hydrobiologia 14: 147-154.

Piper, R. C., I. B. McElwain, L. E. Orme, J. P. McCraren, L. G. Fowler
and J. R. Leonard. 1982. Fish Hatchery Management. U. S.
Department of the Interior. Washington, D. C. 517 pp.
Porter, K. G. 1973. Selective grazing and differential digestion of
algae by zooplankton. Nature 244:1979-180.

SAS (Statistical Analysis System) Institute. 1979. SAS User’s Guide.

SAS Institute. Raleigh, N.C. 494 pp.

Scott-Wasilk, J., G. G. Downing and J. S. Lietzow. 1983. Occurrence
of Asiatic clam Corbicula fluminea in the Maumee River and
Western Lake Erie. Journal of Great Lakes Research 9:9-13.
Sickel, J. B. 1976. An ecological study of the Asiatic clam, Corbicula
manilensis (Philippi, 1841), in the Altamaha River, Georgia,
with emphasis on population dynamics. Ph.D. thesis. Emory
University. Atlanta, GA. 140 pp.

Sickel, J. B. 1979. Population dynamics of Corbicula in the Altamaha
River, Georgia. Pp. 69-80 In: J. C. Britton (ed.), Proceedings
First International Corbicula Symposium. Texas Christian
University. Fort Worth, Texas. 313 pp.

Sickel, J. B., D. W. Johnson, G. T. Rice, M. W. Heyn, and P. K.
Wellner. 1981. Asiatic clam and commericial fishery evalua-
tion. Abstract Corbicula Newsletter 6:15-16.

Sinclair, R. M. and B. G. Isom. 1963. Further Studies on the Intro-
duced Asiatic Clam (Corbicula). Tennessee Department of
Public Health. 78 pp.

Thurston, R. V., R. C. Russo, C. M. Fetterolf, Jr., T. A. Edsall, and
Y. M. Barber, Jr. (eds.). 1979. A review of the EPA Red Book:
Quality Criteria for Water. Water Quality Section. American
Fisheries Society Bethesda, Maryland. 313 pp.

Tiemeier, O. W. and C. W. Deyoe. 1973. Producing Channel Cat-
fish. Agriculture Experimental Station. Bulletin 576. Kansas
State University. 25 pp.

Verdiun, J. 1959. Photosynthesis by aquatic communities in north-
western Ohio. Ecology 40:377-383.

Vollenweider, R. A. 1969. A Manual on Methods for Measuring Primary
Production in Aquatic Environments. IBP Handbook No. 12.
Blackwell Scientific Publications. Oxford and Edinburg. 212

pp.

Wetzel, R. G. 1975. Limnology. W. B. Saunders Company.

Philadelphia, Pennsylvania. 743 pp.

Zaret, T. M. 1980. Predation and Freshwater Communities. Yale
University Press. New Haven, CN. 187 pp.

ASSIMILATION OF RADIOLABELED ALGAE BY CORBICULA

DIANE D. LAURITSEN
DEPARTMENT OF ZOOLOGY
NORTH CAROLINA STATE UNIVERSITY
RALEIGH, NORTH CAROLINA, 27695-7017, U.S.A.

ABSTRACT

Despite its abundance in many aquatic systems, little is known about the feeding ecology of
the Asiatic clam, Corbicula fluminea. Because Corbicula is now found in several coastal North Carolina
rivers that have periodic summer blue-green cyanobacterial blooms, this study compared the poten-
tial for the clams to use 14C labeled cultures of the filmentous blue-green Anabaena oscillarioides with
that of the green algae Chlorella vulgaris and Ankistrodesmus sp. Although there were significant dif-
ferences in the use of the algal foods by the clams, assimilation and net production efficiencies of
clams fed the blue-green were not significantly different from efficiencies of the two green algae. The
significance was a result of the lower assimilation of Ankistrodesmus (47%, compared to 58% for Ana-
baena and 56% for Chlorella). Also, net production efficiencies were significantly higher for
Ankistrodesmus (78%) than for the other foods (61 % for Anabaena and 59% for Chlorella). Any poten-
tial harm from blue-green “blooms" to populations of Corbicula is likely due to clogging of clam gills,
causing most of the potential food material to be rejected as pseudofeces, and the reduction in dissolved
oxygen concentrations in bottom waters as dead algae sink to the bottom and decay.

To date, little is known of the physiology of feeding in
Corbicula fluminea (Muller), although previous work indicates
that the clam can filter a wide range of particle sizes (Wallace
et al., 1977) at fairly high rates (e.g., Mattice, 1979). In par-
ticular, knowledge is lacking on the utilization of different food
sources by the clams. This kind of information is needed to
develop Corbicula aquaculture techniques, as well as to
assess the impact of variations in food quality and quantity
on the growth of these clams in more natural habitats.

C. fluminea has invaded most of the large coastal rivers
of North Carolina within the past 5-8 years. Several of these
rivers (such as the Chowan River in the Northeast) have
become eutrophic, with periodic summer blooms of blue-
green cyanobacteria. The abundance of blue-greens in
eutrophic waters suggest that they do not readily enter food
chains (Porter and Orcutt, 1980); the reasons for this are
unclear and may be a combination of nutritional inadequacy,
toxicity, unmanagability, and/or buoyancy of the algae.

The work reported here is a preliminary attempt to
assess differences between food sources for Corbicula and
in particular to compare blue-green to green algae. This has
been done in the laboratory by measuring assimilation and
net production efficiencies of Corbicula fed monocultures of
algae that have been labeled with 14C. This method has also
facilitated comparison of filtration rates of the different algae,
which included the filamentous blue-green cyanobacterium
Anabaena oscillarioides, the small, sphaerical, unicellular
green alga Chlorella vulgaris, and the sickle-shaped
unicellular green alga Ankistrodesmus sp.

MATERIALS AND METHODS

Algal species used in feeding experiments were main-
tained in continuous culture at 20°C, with constant light pro-
vided by cool white fluorescent lamps. Cultures were gently
agitated with a stirring bar and plate, and were bubbled with
air to maintain a constant pH. Algal growth media and
glassware were autoclaved before use and efforts were made
to keep bacterial contamination at a minimum. The green
algae Ankistrodesmus sp. and Chlorella vulgaris (obtained
from Carolina Biological Supply, Burlington, North Carolina)
were grown with a modified ASM media (Lauritsen and
Mozley, 1983), while the nitrogen-fixing blue-green Anabaena
oscillarioides (isolated from the Chowan River, North Carolina)
was grown with Chu-10 media (Chu, 1942) containing no
nitrogen.

Aliquots of each algal species were inoculated with 2.5
n Ci of 14C sodium bicarbonate 6-18 hr before feeding ex-
periments, then placed in cool white light. After light incuba-
tion, green algal cultures were then centrifuged and algal cells
resuspended in filtered (Whatman GF/C glass fiber filters) lake
water. Cell volume and density of resuspended cultures was
determined with a Coulter Counter Model T/\|| with a 100/*
aperture.

Collections of live Corbicula were made in Lake
Waccamaw (Columbus Co., North Carolina) and the Chowan
River near Winton, N. C. The clams were maintained in
aerated aquaria at 20°C for at least a week prior to feeding
experiments, and during holding were fed mixed cultures of

American Malacological Bulletin, Special Edition No. 2(1 986):21 9-222

219

220

CORBICULA SYMPOSIUM

DOSING CHAMBER FEEDING CHAMBER

Fig.l. Schematic of continuous-flow system for feeding experiments.
Algae are added to dosing chamber, then circulate to the feeding
chamber, where the clam filters out the algae.

Ankistrodesmus, Scenedesmus and Chlorella. Clams were
always fed 36-48 hours before feeding experiments were
begun, then 24 hours before the start of feeding experiments
clams of similar sizes were isolated in beakers of filtered lake
water to empty their guts. The range in shell length of Cor -
bicula used in experiments was 20-23.3 mm.

Individual clams were placed in feeding chambers with
1 .5 / of circulating, filtered lake water 1 hour before exposure
to labeled algae to aliow them to acclimate. A schematic of
a feeding chamber is illustrated in Figure 1; they were a
modification of a design by Peirson (1983) and Riisgard and
Mohlenberg (1979). Specific volumes (1 mm3) of labeled food
were added to the dosing reservoirs of the chambers and the
clams were allowed to feed for 1 hour. Clams were then re-
moved and the contents of the feeding chambers were filtered
through 4.25 Whatman GF/C filters and radioassayed. Eight
clams were tested for each algal food.

Clams removed from the feeding chambers were
placed in aerated 1 / side-arm Erlenmeyer flasks with 500
ml of filtered lake water. Flask sidearms were connected to
impinger traps containing 25 ml of a mixture of ethylene glycol
monomethyl ether and ethanolamine to trap 14C02 (Peirson,
1983). Feces were collected daily by filtering the flask water
through Whatman GF/C filters and radioassaying each day’s
sample separately. After two days, the clams were removed
from the flasks and frozen. The water in the filter flasks was
acidified with 1 ml of cone HC /, then aerated for 30 min with
the impinger traps still attached to drive off any remaining
14C02. Aliquots of the flask water were then assayed to deter-
mine the amount of 14C in soluble form. One ml of methanol
was added to samples of the mixture in the C02 traps before
radioassay to facilitate mixing of this liquid with the scintilla-
tion cocktail.

Frozen clam tissue and liquor was removed from
shells, macerated, and placed in scintillation vials. Wet tissue
weight was determined, then a mixture of Scintigest and water
were added according to manufacturer’s instructions. Tissues
were then digested 12-24 hr in a 50°C water bath.

To measure 14C incorporation in clam shells, individual

shells were placed in side-arm Erlenmeyer flasks containing
100 ml of 10% HC/. The flasks were aerated and C02 im-
pinger traps were attached to the sidearms to collect any
14C02 evolved during disintegration of the shells. Samples
of the trap liquid were counted for radioactivity after addition
of menthanol and scintillation cocktail.

All samples were counted on a Beckman LS 700 li-
quid scintillation counter, using a toluene 14C external
standard.

In radiotracer experiments, carbon assimilation is
determined by adding the radioactivity retained in the animal
(without gut contents) to the complete metabolic losses
(respiration) of tracer during the experimental period. Animals
were held in respiration chambers for a 2-day period after
feeding experiments so that unassimilated material would be
voided while at the same time measuring 14C2 evolved.
Because the clams did not filter out all of the labeled algal
cells after 1 hr in the feeding chambers, assimilation was
determined from the percentage of tracer ingested:

% Assim. Eff. = 14G in tissues + 14C respired

14C ingested

Net production efficiency (energy available for growth
and reproduction, e.g., Russell-Hunter, 1972) for clams fed
each algal species were determined as:

% Net Prod. Eff. = 14C in tissues

14G tissues + 14C respired

The following equation was used to determine filtra-
tion rate for individual clams (Goughian 1969):

filtration rate = volume (ml) loge / initial cone \
time \ final cone /

The concentration ratio was determined as the total amount
of algal radioactivity dosed divided by the algal radioactivity
that remained in the chamber at the end of the feeding period.
Because the water in feeding chambers was continuously cir-
culating, algal settling was assumed to be negligible.

The Fmax test (Sokal and Rohlf, 1981) was used to test
for homogeneity of variances of assimilation efficiencies, net
production efficiencies, and filtration rates. Assimilation and
net production efficiencies were arcsine transformed, then
analysis of variance tests were performed to determine if there
were significant differences between treatments (algal
species). Analysis of variance was calculated on untrans-
formed filtration rates.

RESULTS

No significant difference was found in variances of
assimilation and net production efficiencies, so standard er-
rors of retransformed means for these efficiencies are ex-
pressed as the square root of the retransformed mean square
error of the ANOVA for each (Table 1). There was a signifi-
cant difference (F = 4.66, P < .05) between transformed
assimilation efficiencies of the different algae, due to the
significantly lower assimilation of Ankistrodesmus (47%) com-
pared to Anabaena (58%) and Chlorella (56%) (F = 9.19, p

LAURITSEN: CORBICULA FEEDING

221

Table 1 . Mean carbon assimilation (14C utilized, as a % of what was
ingested, or A/I) and net production efficiencies (represents the
energy available for growth and reproduction as a % of what was
assimilated, or P/A), and mean filtration rates (in ml-tr1, with stan-
dard errors in parenthesis) for Corbicula fluminea fed equivalent
volumes of 14C labeled algae.

A/I

P/A

Filtration

Rate

(mi-tr1)

Anabaena

57.7

61.4

587.4

Ankistrodesmus

47.4

78.2

(90.1)

765.6

Chlorella

56.3

59.4

(18.2)

770.0

3.01

9.63

(37.1)

Table 2. Mean percentage of total radioactivity filtered by Corbicula
found in each kind of sample, with standard errors indicated in paren-
thesis. No label was recovered in clam shells.

FECES

D014C

14C02

TISSUES

Anabaena

12.37

30.09

21.14

36.38

(2.19)

(31.5)

(3.58)

(5.36)

Ankistrodesmus

17.70

34.86

10.64

36.77

(4.07)

(3.03)

(2.74)

(4.32)

Chlorella

37.08

6.68

22.70

33.52

(3.86)

(1.38)

(.97)

(2.94)

< .01).

Net production efficiencies were also significantly dif-
ferent (F - 4.55, p < .05) between algal species. But while
assimilation of Ankistrodesmus was low compared to the other
algae, mean net production efficiency was significantly higher
(78%) than Anabaena (61%) and Chlorella (59%) (F = 9.02,
p < .01, Table 1). The high net production values of
Ankistrodesmus were due to the lower percentage of
assimilated carbon respired as 14C02 (Table 2).

A large proportion of feces were produced by clams
fed Chlorella (Table 2), and microscopic examination indicated
that the feces were composed primarily of clumps of green
cells. A smaller proportion of intact cells of Ankistrodesmus
were observed in clam feces, while no Anabaena filaments
were found in feces. However, Anabaena was the only algal
food that caused Corbicula to produce pseudofeces, or
clumps of filaments that would be ejected before they were
ingested.

Filtration rates were fairly constant over all algal food
treatments, with means ranging from 709-770 ml-tr1 (Table
1), resulting in no significant difference between them (F =
1.03, p > .25).

DISCUSSION

Most feeding studies of bivalve molluscs have been
done under laboratory conditions, and the advantages of this

method include the ability to control and manipulate specific
variables such as temperature, food quantity and quality.
There are, however, certain disadvantages which must be
taken into consideration when interpreting their results and
applying them to natural systems. For example, bivalves can
be very sensitive to disturbances and test conditions, resulting
in significant deviations in important physiological functions.
This is evidenced by the wide fluctuations in filtration rates
reported for Corbicula fluminea : 20-150 ml-tr1 (Prokopovich,
1969); a mean of 11 ml-fr1 (Habel, 1970); a mean of 816 ml-fr1
(Auerbach et al., 1977); 500-600 ml-fr1 (Mattice, 1979).
Although Mattice (1979) mentions that feeding rate in Cor-
bicula can be affected by food “quality”, I found very similar
filtration rates for each algae used, indicating that the clams
were not selecting against any algal food on the basis of
“taste”. Filtration rates of Corbicula are thus much higher
than other freshwater bivalves such as Dreissena poly-
morpha (Stanczykowska et al., 1976; Walz, 1978) and
Sphaerium striatinum (Hornbach et al., 1984).

Assimilation efficiencies are a measure of how much
energy is utilized as a proportion of what is ingested. About
half of the 14C ingested by clams fed each of the three algal
foods was expelled as wastes (feces and dissolved organic
14C; Table 2), and so assimilation efficiencies were fairly
similar. Other researchers (e.g., Peirson, 1983) have noted
that Chlorella is not well assimilated by filter-feeding bivalves
and have concluded that the thick cell wall effectively prevents
lysing of the ceils. This is probably the case with Corbicula
as well, since the feces consisted primarily of mucus-bound
clumps of cells. To a lesser extent the same also seems
to hold true for the other green alga, Ankistrodesmus. It may
be that freshwater bivalves are not particularly efficient at
utilizing what they filter from the water; assimilation ef-
ficiencies for the freshwater mussel Dreissena polymorpha
average about 40% (Stanczykowska ef al., 1975; Walz, 1978).
The important difference with Corbicula is that because fil-
tration rates are so much higher, the total amount of material
assimilated over any given time will be much higher.

Net production efficiencies are a measure of energy
available for growth and reproduction, determined as a
percentage of assimilation. The results from my laboratory
experiments are similar to values calculated from carbon
budgets of population of Corbicula in Lake Arlington, Texas
(mean of 71%, Aldredge and McMahon, 1979). Such high
net production efficiencies have been reported only for one
other freshwater bivalve (60-80% for Pisidium conventus,
Holopainen and Hanski, 1979). These high net production
efficiencies are possible because Corbicula, a relatively
sedentary organism, expends little energy on respiration (e.g.,
McMahon, 1979).

Growth rates of C. fluminea are usually much higher
than rates reported for other freshwater bivalves (e.g., Horn-
bach ef al., 1980; Negus, 1966). Filtration rates and net pro-
duction efficiencies reported here indicate that the clams are
able to remove a relatively large proportion of potential food
material from the water, and are very efficient at utilizing what
food is assimilated, thereby providing the energy needed for
these rapid growth rates.

222

CORBICULA SYMPOSIUM

Any substantial differences in utilization of different
kinds of algal would indicate that the diet could affect in situ
growth rate of the clams (or otherwise be detrimental— cf. In-
troduction). Corbicula appears to show no substantial dif-
ference in its ability to utilize the blue-green Anabaena or the
two green algae Chlorella and Ankistrodesmus. However,
dense assemblages of blue-greens such as the filamentous
Anabaena or clump-forming species such as Microcystis,
which develop during nuisance blooms in summer, may have
a negative impact on Corbicula by clogging their gills and
causing most of the potential food they filter to be rejected
as pseudofeces. But since many blue-greens float on or near
the water surface, they may remain relatively “unavailable”
to the clams in deeper water. In situ experiments would be
needed to determine the actual effects of blue-green blooms
on clam filtration.

Perhaps the most significant impact of blue-green
blooms is in lowering dissolved oxygen concentrations near
the sediment surface as dying algae sink to the bottom and
decay. Corbicula is intolerant of low oxygen conditions
(McMahon, 1979), so that blue-green blooms could wipe out
existing populations of the clams (I have observed this in at
least one large coastal river in North Carolina) and/or pre-
vent the clams from becoming established.

ACKNOWLEDGEMENTS

I would like to thank W. M. Peirson, N. C. State University,
for allowing me to use some of his equipment for feeding experiments,
and Bob McMahon and Sam Mozley for reviewing earlier drafts of
this manuscript. This research was supported by the Water
Resources Research Institute, University of North Carolina, Project
No. A-124-NC.

LITERATURE CITED

Aldredge, D. W. and R. F. McMahon. 1978. Growth, fecundity, and
bioenergetics in a natural population of the Asiatic freshwater
clam Corbicula manilensis Philippi, from North Central Texas.
Journal of Molluscan Studies 44: 49-70.

Auerbach, S. I., D. E. Reichleand E. C. Struxness. 1977. Environmen-
tal Sciences Division Annual Progress Report ORNL 5257.
Oak Ridge National Laboratory, Oak Ridge, Tenn. 295 pp.
Chu, S. P. 1942. The influence of the mineral composition of the
medium on the growth of planktonic algae. Journal of Ecology
30: 284-325.

Goughian, J. 1969. The estimation of filtering rate from the clearance
of suspensions. Marine Biology 2: 356-358.

Habel, M. L. 1970. Oxygen consumption, temperature tolerance, and
filtration rate of the introduced Asiatic clam Corbicula manilen-
sis from the Tennessee River. M. S. thesis, Auburn Univer-
sity, Auburn. 91 pp.

Holopainen, I. J. and I. Hanski. 1979. Annual energy flow in popu-
lations of two Pisidium species (Bivalvia, Sphaeriidae), with
discussion on possible competition between them. Archiv fur
Hydrobiologie 86: 338-354.

Hornbach, D. J., C. M. Way and A. J. Burky. 1980. Reproductive
strategies in a freshwater sphaeriid clam, Musculium par-
tumeium (Say), from a permanent and a temporary pond.
Oecologia 44: 164-170.

Hornbach, D. J., C. M. Way, T. E. Wissing and A. J. Burky. 1984.
Effects of particle concentration and season on the filtration
rates of the freshwater clam, Sphaerium striatinum Lamark
(Bivalvia: Pisidiidae). Hydrobiologia 108: 83-96.

Lauritsen, D. D. and S. C. Mozley. 1983. The freshwater Asian clam
Corbicula fluminea as a factor affecting nutrient cycling in the
Chowan River, N. C. Water Resources Research Institute
Rept. No. 192, University of North Carolina, Raleigh, N. C.

60 pp.

Mattice, J. S. 1979. Interactions of Corbicula sp. with power plants.
In: J. C. Britton (ed.), Proceedings of the First International
Corbicula Symposium, Texas Christian University, Oct. 13-15,
1977. p. 119-138.

Negus, C. L. 1966. A quantitative study of growth and production
of unionid mussels in the River Thames at Reading. Journal
of Animal Ecology 35: 513-532.

Peirson, W. M. 1983. Utilization of eight algal species by the bay
scallop, Argopecten irradians concentricus (Say). Journal of
Experimental Marine Biology and Ecology 68: 1-11.

Porter, K. G. and J. D. Orcutt, dr. 1980. Nutritional adequacy,
manageability, and toxicity as factors that determine the food
quality of green and blue-green algae for Daphnia. In: W. C.
Kerfoot (ed.), Evolution and Ecology of Zooplankton Com-
munities. ASLO Special Symposium Vol. 3. Hanover: Univer-
sity Press of New England, p. 268-281.

Prokopovich, N. P. 1969. Deposition of clastic sediments by clams.
Journal of Sedimentary Petrology 39: 891-901.

Riisgard, H. U. and F. Mohlenberg. 1979. An improved automatic
recording apparatus for determining the filtration rate of Mytilus
edulis as a function of size and algal concentration. Marine
Biology 52: 61-67.

Russell-Hunter, W. D. 1970. Aquatic Productivity. London: Macmillan
Co. 306 pp.

Sokal, R. R. and F. J. Rohlf. 1981. Biometry. San Francisco: W. H.
Freeman and Co. 859 pp.

Stanczykowska, A., W. Lawacz and J. Mattice. 1975. Use of field
measurements of consumption and assimilation in evaluation
of the role of Dreissena polymorpha in a lake ecosystem. Polish
Archiv fur Hydrobiologie 22: 509-520.

Stanczykowska, A., W. Lawacz, J. S. Mattice and K. Lewandowski.
1976. Bivalves as a factor affecting circulation of matter in
Lake Mikolajskie (Poland). Limnologica (Berlin) 10:347-352.

Wallace, J. B., J. R. Webster and W. R. Woodall. 1977. The role of
filter feeders in flowing waters. Archiv fur Hydrobiologie 79:
506-532.

Walz, N. 1978. The energy balance of the freshwater mussel
Dreissena polymorpha Pallas in laboratory experiments and
in Lake Constance. I. Pattern of activity, feeding and
assimilation efficiency. Archiv fur Hydrobiologie Suppliment
55: 83-105.

ADAPTATIONS OF PISIDIIDAE (HETERODONTA: CORBICULACEA)

TO FRESHWATER HABITATS

GERALD L. MACKIE
DEPARTMENT OF ZOOLOGY
UNIVERSITY OF GUELPH
GUELPH, ONTARIO NIG 2W1

ABSTRACT

The Pisidiidae have representatives in virtually all types of freshwater habitats, from temporary
ponds and roadside ditches to deep, cold profundal zones of oligotrophic and eutrophic lakes. A review
of the studies reported to date suggests that there are peculiarities in the structure, composition and
morphology of the shell, and in the anatomy of the gills, siphons and byssal apparatus, as well as
in the physiologies and ecologies of many species that have made the Pisidiidae one of the most
common and widely distributed groups of freshwater invertebrates in all parts of the world.

At present, 39 living species belonging to five genera
of freshwater corbiculacean clams are recognized in North
America. Corbicula and Eupera are monospecific in North
America with C. fluminea (Muller) and E. cubensis (Prime) the
only species reported to date (Britton and Morton, 1979;
Mackie and Huggins, 1976). The greatest number of species
are in the genus Pisidium (26). Sphaerium has 8 species and
Musculium has 4.

Out of 62 states, provinces and territories in North
America, Pisidium casertanum (Poli) occurs in 61 and is the
most cosmopolitan of all corbiculacean clams (Mackie, 1981).
Corbicula is found in 35 and Eupera in 7 states, provinces
and territories (McMahon, 1982; Mackie, 1981). The rarest
species is Pisidium ultramontanum Prime (Mackie, 1981).

All corbiculacean clams are ovoviviparous and relative-
ly small in size. The largest species is C. fluminea reaching
6 cm in length (Britton and Morton, 1979). The smallest
species is Pisidium punctatum Sterki (usually less than 1 .5
mm long). Most pisidiids have shell lengths less than 8 mm;
only Sphaerium simile (Say) and Musculium transversum (Say)
grow as large as 2 cm in shell length (Mackie and Huggins,
1983).

ADAPTATIONS IN SHELL STRUCTURE AND SHAPE

The shell of Corbicula is thick and heavy. Pisidiids have
relatively thin shells. Aragonite crystals, present in all cor-
biculacean clams, is laid down in a complex crossed-lamellar
structure (Mackie, 1978). The laminated sheets of aragonite
that characterize the nacre of Unionidae is absent in cor-
biculacean clams. The complex crossed-lamellar structure
is considered to be an adaptation to high resistance to abra-
sion (Carter, 1980).

The effect of environment on shell shape and the rela-
tion between fecundity and shell shape are well documented.
By transferring Sphaerium corneum (Linnaeus) from organical-

enriched water to clean water or by experimentally increas-
ing the dissolved oxygen content of the water, Thiel (1926)
was able to demonstrate a change in roundness or thickness
of the shells. Since this change was also accompanied by
an increase in natality, Thiel (1926) suggested that the clean
water with lower food production was a less favorable environ-
ment for S. corneum and consequently the survival of the
population required more effective reproduction. The higher
number of embryos was suggested to need more room and
to cause change in the direction of growth. Similar relation-
ships between shell thickness (width) and number of embryos
was demonstrated by Holopainen and Kuiper (1982) for P.
casertanum and Mackie and Flippance (1983a) for Sphaerium
rhomboideum (Say).

Changes in shell size (as well as weight and calcium
content) have also been shown to be related to other en-
vironmental variables. Mackie and Flippance (1983b), using
canonical correlation analyses on species collected from 53
habitats with total alkalinity ranging from 0 to 280 mg
CaCC^L-, showed that a decrease in acid-neutralizing capaci-
ty relative to the noncarbonate anion content was accom-
panied by short but heavily calcified shells in Pisidium variabile
Prime, P. casertanum, S. simile and S. striatinum (Lamarck).
Waters with increasing acid-neutralizing capacity relative to
calcium hardness are accompanied by longer shells with in-
creases in tissue calcium relative to total weight in Musculium
securis (Prime) and Pisidium compressum Prime. All six
species of pisidiids that were analyzed for canonical correla-
tions showed more than one significant canonical variate,
suggesting that acid deposition would not be a factor in their
disappearance from water with pH greater than 5.50 (Mackie
and Flippance, 1983b).

Of the 13 species of pisidiids collected from the 53
habitats, Mackie and Flippance (1983c) were able to derive
correlation coefficients between calcium content of the

American Malacological Bulletin, Special Edition No. 2 (1986): 223-229

223

224

CORBICULA SYMPOSIUM

whole individual and pH, total alkalinity, total hardness and
calcium hardness of the water for only seven species (the
sample size being too small for the remaining six species).
Significant correlations (P <0.05) were found for only five
of the seven species. Two species (Sphaerium rhomboideum
(Say) and S. simile) showed negative correlations, while three
(P. casertanum, P. compressum, and S. striatinum) showed
positive correlations between calcium content of individuals
and environmental calcium content. Numerous other studies
also demonstrated strong correlations between environmental
calcium content and molluscan (including pisidiids) calcium
content (Russell-Hunter et al., 1967; Lee and Wilson, 1969,
1974; Burky et al, 1979), distribution (Boycott, 1936; Macan,
1950; McKillop and Harrison, 1972), abundance (Dussart,
1976, 1979a; Williams, 1970a,b), physiology (Harrison, 1968;
Hunter and Lull, 1977), and life history (Dussart, 1979b;
Thomas et al., 1974), to mention only a few.

Russell-Hunter et al. (1981) describe an irregular
distribution of four shell types in gastropods with respect to
water characteristics: (i) a direct relationship between the
amount of shell calcification and water hardness, (ii) a con-
stant ratio of shell calcium to whole animal dry weight
throughout growth in a wide range of calcium concentrations,
(iii) a positive relationship between shell mass and trophic
conditions, and (iv) no relationship between the amount of
shell calcium, organic carbon or nitrogen, and the water
characteristics. A similar distribution of shell types is evident
in Pisidiidae (Mackie and Flippance, 1983c), with even a fifth
type, an inverse relationship between shell calcification and
dissolved calcium, as reported by Burky et al. (1979) for
Sphaerium striatinum.

Although many significant (P < 0.05) correlations were
found by relating size, weight and calcium content of pisidiids
to the “buffer variables” (i.e. pH, alkalinity, total hardness,
calcium hardness) of the water, these correlations do not
necessarily imply cause/effect relationships. Indeed, it was
indicated earlier that reproduction may cause changes in
shape. Nevertheless, the studies do show that the changes
in size, weight and calcium content can be related to an en-

Table 1. Calcium carbonate and carbon content of shells in com-
mon species of Pisidiidae in the study area. The species are arranged
in order of decreasing calcium carbonate content.

Species

Shell CaC03
as % of
total dry wt.
±95% C.l.

C mg-1

shell

±95% C.l.

Sphaerium striatinum

92.2 ± 1.69

5.33 ± 0.68

Sphaerium simile

90.7 ± 2.53

ND2

Pisidium compressum

90.3 ± 2.53

Musculium securis

80.0 ± 3.21

8.32 ± 1.57

Pisidium casertanum

65.8 ± 1.66

10.18 ± 2.77

1C.I. = Confidence Interval

ZND = Not determined for species in waters with > 45 mg CaC03
L_1 total alkalinity.

vironmental variable or set of environmental variables and
these changes may be of adaptive significance. As an adap-
tation, changes in calcium content are easier to explain than
changes in size and weight of pisidiids. Particularly easy to
explain are positive correlations between calcium contents
of pisidiids and of the environment. Decreases in shell
calcium content are usually accompanied by increases in car-
bon content (Burkey et al., 1979). Analyses of shell carbon
content in species from acidified lakes indicate very large pro-
portions of organic matter (probably conchiolin) in relation
to calcium carbonate content (Table 1) suggesting that high
organic content in shells may be an adaptation to resist ero-
sion from acidified waters. Burky et al. (1 979) discusses the
adaptive significance of the inverse relationship between the
amount of shell CaCC>3 and shell organic carbon and nitrogen
in pisidiid clams; they concluded that this inverse relation-
ship is convincing evidence that pisidiid shells are consistently
built from some base level adaptive need of mechanical pro-
tection. Burky et al., (1979) also suggested that pisidiid clams
have a common strategy for shell secretion which is under
genetic control. If this is true then the thick periostracum of
P. casertanum in acidifying lakes may be merely reflecting
the erosion of calcareous components, leaving only the
periostracum which is more resistant to the corrosive
effects of hydrogen ions. In this event, the inverse relation-
ship between the amount of organic material and calcium con-
tents in the shell may be a demonstration of the corrosive
effects of hydrogen ions on shell calcification, rather than a
thickening of the periostracum as an adaptation to resist shell
erosion. Nevertheless, P. casertanum has a thicker
periostracum than most other species of Pisidiidae, as Figure
1 in Mackie (1978) strongly indicates, and would explain the
relative success of this species in corrosive environments.
Correlations between organic content of the shell and pH,
and trophic status of the environment are also present in
unionids (Agrell, 1949; Singer, 1981).

Another interesting mechanism in Pisidiidae for
resisting the corrosive effects of hydrogen ions is by changes
in the morphology of the shell. Mackie and Flippance (1983b)
showed by canonical correlations that in lake acidification (i.e.
decreasing alkalinity) the calcium content of P. casertanum
and Sphaerium striatinum decrease, but a high density of
CaCC>3 is maintained in the shell by forming shorter (therefore
more compact) shells. Hence, the protection offered by the
shell can be maintained in low-alkalinity waters. The canonical
correlation analyses also indicated that long, thin shells, which
would provide less protection in acidifying waters than short,
thick shells, are formed only in waters with increasing alkalini-
ty relative to calcium hardness, as in M. securis.

Other species (e.g. M. securis) show no relationship
in the calcium contents between the animal and environment
(Mackie and Fiippance, 1983c). These species appear to re-
ly on calcium in allochthonous organic material (e.g. leaves
of trees) rather than on bedrock supplies (i.e. limestone) which
probably characterize the calcium content of most watersheds
(Mackie and Flippance, 1983d). This adaptation to organically-
derived calcium may also explain why most species of
Musculium are found in woodland pools.

MACKIE: ADAPTATIONS OF PISIDIIDAE

225

ADAPTATIONS IN SOFT ANATOMY

CTENIDIA. In Corhicula the two pair of ctenidia are well-
developed, but in pisidiids the outer demibranches are re-
duced to small lobes or are absent. As in most freshwater
bivalves, the inner gills of corbiculaceans function as mar-
supia. However, the larvae of Corbicula are released at an
earlier developmental stage (i.e. veligers) than are the lar-
vae of pisidiids. In pisidiids the trochophore and veliger stages
are passed in the egg and the young are released as
miniature adults. Also, the larvae of pisidiids are incubated
in brood sacs (Mackie et al., 1974a) whereas the larvae of
Corbicula are incubated in the water tubes of the inner gill
(Sinclair and Isom 1961).

The incubation of larvae within brood sacs until the
young are capable of an independent existence has con-
siderable survival value (for the population and species).
Mackie (1979a) and McKee and Mackie (1980) demonstrated
that extra-marsupial larvae survive longer periods of desic-
cation than do parents. The resistance of extra-marsupial lar-
vae to desiccation (and/or the protection offered by the
parent’s shell) is an important adaptation for dispersal by
aerial transport on waterfowl (Mackie, 1979a) and for survival
of dry periods in ephemeral ponds (McKee and Mackie, 1980).

Brooding of larvae also has survival value in toxic en-
vironments. Ninety-six hour exposure of Pisidium equilaterale
Prime, gravid with extra-marsupial larvae, down to pH 2.5
killed the parents but the larvae showed little mortality (Mackie
et al., 1983).

SIPHONS. The siphons of Corbicula are relatively complex
compared to those of Pisidiidae. Although the anal and
branchial siphons are merely modifications of the mantle
lobes (Kraemer, 1977), they are ornamented with papillae,
tentacles and pigment (Sinclair and Isom, 1963). The siphons
in Pisidiidae are simple tubes and the branchial siphon is re-
duced to a mere slit in the mantle in Pisidium sp and is ab-
sent in Neopisidium.

The size and development of siphons in Pisidiidae are
related to the size of the animal; the siphons are smallest and
most poorly developed in Pisidium sp. which have smaller
average shell lengths (4mm) than Eupera or Muscuiium (avg.
length = 7 mm; the siphons are fused along their length for
only the basal half) and Sphaerium (avg. length = 9 mm; the
siphons are fused for almost their entire length). When fully
extended, the siphons may be 50-1 00% of the animal’s shell
length.

The short, often slit-like sipons of Pisidium spp. ap-
pears to have some adaptive significance. Since most
Pisidium spp. are small and live in the surficial layers of
sediments (Meier-Brook, 1969) their siphons, even when fully
extended, would rarely penetrate the mud-water interface.
Hence, most of the water that is filtered by Pisidium is de-
rived from the interstitial spaces of the sediments. Indeed long
siphons in these situations would probably be cumbersome
and impede or restrict water flow into the mantle cavity. The
longer siphons in Eupera, Muscuiium and Sphaerium spp
would therefore imply more of an epifaunal habit. An epifaunal
habit has often been described for Eupera (Heard, 1965),

Muscuiium (Boozer and Mirkes, 1979; McKee and Mackie,
1981) and Sphaerium (Hynes, 1972), although infaunal habits
are also well known (Gale 1971, 1973).

BYSSUS. The only corbiculaceans with a functional byssus
in adults are Corbicula (Sinclair and Isom, 1963) and Eupera
(Heard, 1977). A byssal gland is present in larvae of
Muscuiium and Sphaerium corneum and S. occidental
(Mackie et al., 1974b; Heard, 1977) but all other species of
Sphaerium and Pisidium lack a byssal gland in all life stages.
In those species that have a functional larval byssus, the
byssal stalk arises in the foot of the larvae (prodissoconch)
and inserts on a small bulb attached to the descending
lamella of the inner gill (Mackie et al., 1974b). Heard (1977)
suggests that the byssus functions to prevent precocious birth
(i.e. abortion) until the larval gonads are sexually mature.
However, the gonads of Muscuiium species mature during
adult life (Mackie eta!., 1976), so this function seems doubtful.
Yonge (1962) considers the presence of a byssal apparatus
in the adult as representing the persistence of a post-larval
organ (i.e. the animals possessing it are in this respect
neotenous). Boozer and Mirkes (1979) suggested that M. par-
tumeium (Say) has a functional byssus during adult life, but
could not substantiate it. The threads observed by Boozer
and Mirkes (1979) were probably mucous strings since the
byssal gland is absent in adults (Mackie et al., 1974b and
unpublished data).

PHYSIOLOGICAL ADAPTATIONS

Respiratory adaptations of corbiculacean clams to
temperature and oxygen content is well documented (Burky
and Burky, 1976; McMahon, 1979; McKee and Mackie, 1983).
All corbiculaceans appear to have poor respiratory adapta-
tions to high temperature. Only Corbicula has so far been
reported to have poor respiratory adaptation to hypoxia
(McMahon and Aldridge, 1976; Aldridge, 1976; Aldridge and
McMahon, 1978; McMahon, 1979), although Habel (1970) and
Busch (1974) in less thorough studies report tolerance of C.
fluminea to hypoxic conditions. Most pisidiids are more
tolerant of hypoxia and are better regulators of oxygen con-
sumption than are C. fluminea and appear to be able to adapt
to hypoxia at some stage in their life cycles. All species of
Muscuiium are commonly found in nearly anoxic pond waters;
Pisidium idahoense Roper can survive hypoxia for at least
two weeks (Juday, 1908); Pisidium casertanum inhabits nearly
anoxic substrata in the summer (Berg and Jonasson, 1965;
Mackie, 1979b); both Muscuiium securis and Sphaerium oc-
cidentals can survive anaerobic environments but M. securis
is a facultative anaerobe and S. occidentale is an obligate
aerobe during estivation (McKee and Mackie, 1983), while
C. fluminea is largely excluded from reducing substrata.

Pisidiidae also appear to be more efficient at oxygen
uptake during hypoxia than C. fluminea (McMahon, 1979).
Berg et al. (1962) reported good regulation of oxygen uptake
in P. casertanum. Alimov (1 965), Burky and Burky (1 976) and
McKee and Mackie (1983) reported efficient oxygen uptake
in Sphaerium corneum, Pisidium walked Sterki and M. securis
and S. occidentale, respectively, although they attributed

226

CORBICULA SYMPOSIUM

variations in respiration to variations in temperature.

Pisidiids exhibit both the over-compensation and
“reverse” acclimation patterns described by Precht et al.,
(1973). In P. walkeri (Burky and Burky, 1976) and M. securis
(McKee and Mackie, 1983), the over-wintering generation
displays over-compensation (decreasing respiration as
temperature rises) in the spring at 10 and 20°C. Reverse com-
pensation is apparent in S. corneum since respiration rates
at 20°C decrease from summer to fall (Alimov, 1965). Both
patterns occur in the annual life cycle of S. occidentale
(McKee and Mackie, 1983).

In ephemeral habitats, respiratory adaptations are
related to the specific life histories of the pisidiids. Oxygen
uptake rates at field temperature and Q10 values are low in
both S. occidentale and M. securis during hibernation and
estivation, indicating respiratory stability and energy conser-
vation (McKee and Mackie, 1983). These rates and respiratory
coefficients peak in spring during maximum growth and
reproduction. However, respiration in M. securis is more in-
dependent of temperature than in S. occidentale, reflecting
the requirement of M. securis to complete growth and
reproduction more quickly. During estivation, M. securis is
a facultative anaerobe while S. occidentale is an obligate
aerobe. Arousal from estivation, as indicated by oxygen up-
take, is delayed when clams are introduced to pond water.
This adaptation inhibits a premature resumption of activity
and decreases the likelihood of desiccation (McKee and
Mackie, 1983). Similar adaptations have been reported in
some unionid bivalves (Dance, 1958; Dietz, 1974).

Although McMahon (1979) regards C. fluminea as
relatively intolerant of high summer ambient temperatures,
corbiculids appear able to tolerate higher temperatures than
pisidiids. Corbicula is most common in the southern United
States where ambient water temperatures above 30°C oc-
cur frequently, but pisidiids are least common there and are
most common in northern United States and Canada where
ambient water temperatures usually do not exceed 30°C. Cor-
bicula fluminea can tolerate short-term exposure to 43°C (after
acclimating at 30°C) (Mattice and Dye, 1976) but most
pisidiids seem to perish quickly at 30°C. Nevertheless growth
and reproduction in both groups are severely impaired above
25-30°C.

Being relatively poorly adapted to hypoxia and extreme
temperatures, C. fluminea has remained physiologically close-
ly allied to its estuarine ancestors (McMahon, 1979). Such
physiological prerequisites are restricting C. fluminea to
relatively large bodies of temperature-stable and well oxygen-
ated flowing fresh water conditions (McMahon, 1979). Most
pisidiids, on the other hand, appear to be less restricted by
their respiratory physiology and should remain competitive
in the majority of North American freshwater habitats. This
has been confirmed by at least one study (Boozer and Mirkes,
1979).

ECOLOGICAL ADAPTATIONS

Some of the most interesting adaptations of pisidiids
to temporary and permanent aquatic habitats have been

elucidated through studies of growth dynamics and reproduc-
tive habits. By regressing larval shell length on parent shell
length, Mackie (1976b) was able to compare larval growth
rates in relation to that of parents. He found that species with
slow larval growth rates are usually semelparous and
univoltine. These species can be iteroparous by precocious
birth of larvae and multivoltine by accelerated growth of
semelparous individuals. Species that live one year and have
rapid larval growth rates are usually iteroparous because lar-
vae grow faster than parents and (or) there is precocious birth
of larvae.

Studies of growth dynamics in two permanent pond
populations of S. rhomboideum showed that width has a
positive allometric relationship with length and height of the
shell (Mackie and Flippance, 1983a). This results in an in-
crease in interior shell volume that is significantly correlated
with the increase in space required by developing larvae. An
iteroparous reproductive strategy has been shown to be
dependent upon both a positive allometric relationship be-
tween larval and parent shell lengths and birth of a cohort
during or immediately before an active growing period. If birth
of the cohort occurs immediately before a dormant period,
semelparity will occur, no matter what type of allometric rela-
tionship is obtained between larval and adult shell lengths
(Mackie and Flippance, 1983a).

Studies of species (e.g. S. occidentale, M. securis, M.
partumeium) from temporary ponds to date show that all of
them have the potential for iteroparous reproduction (Mackie,
1979b; Way etal., 1980; McKee and Mackie, 1981). In some
instances, semelparity is exhibited but only because the
ponds dry up before a second litter is produced (Mackie et
al., 1978). In temporary ponds, where there may be high mor-
tality during the dry season, an iteroparus reproductive
strategy that produces many young in as short as time as
possible by a single generation will perpetuate the popula-
tion (and species) more effectively than a semelparous
reproductive strategy.

For populations with a one year life span, semelparity
could be considered to be a “luxury” that is afforded only
to species that have low mortalities in a harsh environment
(because the species is very tolerant to a wide range of en-
vironmental stresses). In this case the population can be
maintained by several different parental generations with one
reproduction each (i.e. semelparity) rather than by a single
parent generation with several reproductive periods in its one
year life span (i.e. iteroparity). Hence P. casertanum, a very
cosmopolitan and tolerant species (Clarke, 1979) appears to
have adopted a semelparous strategy (Thut, 1969; Mackie,
1979b; Burky et al., 1981). Species less tolerant (e.g. P.
variable) will adopt an iteroparous strategy when the environ-
ment imposes physiological limits on growth, reproduction,
and maintenance functions (Way and Wissing, 1982).

For species that live for more than one year, very
different reproductive strategies are observed. For example,
Pisidium conventus inhabits only the deep, cold, profundal
zones of oligotrophic lakes (or littoral zones of subarctic lakes)
and has a three year life span. The profundal zones of
oligotrophic lakes have relatively constant environmental con-

MACKIE: ADAPTATIONS OF PISIDIIDAE

227

ditions, including temperature, and in essence lack seasons.
Pisidium conventus Clessin grows very slowly and produces
such few offspring in each litter that it must reproduce 4-5
times in its lifetime in order to maintain the population densi-
ty (Holopainen, 1979). Hence, an iteroparous reproductive
strategy is characteristic of this oligotrophic species.
However, P. casertanum (that lives for more than one year)
is able to adopt either strategy. This versatility in reproduc-
tion strategy (inpart related to life span) is able to afford P.
casertanum a very cosmopolitan distribution.

These data suggest that all species are potentially
iteroparous and semelparity occurs only if a species’ life span
is suddenly truncated. However, this has been shown to be
not the case for M. securis ; Mackie ef a/. (1 976) transplanted
M. securis from a temporary pond into a permanent pond and
a river and in all cases, the semelparous population from the
temporary pond remained semelparous in the permanent
habitats, even though their life spans were extended by two
to three months. Hence, semelparity in this case appears to
be an evolved (i.e. genetic) life history trait. For gastropods,
Calow (1 978) associates the semelparous state with reproduc-
tive recklessness and the iteroparous state with restraint on
the part of the parent; semelparity is considered to have
evolved in association with adaptations that ensure a greater
chance of survival of the offspring. Calow (1978) should be
consulted for an excellent discussion of theories on life-cycle
strategies in gastropods.

Other life history traits also show some adaptive value.
For example, pisidiids may display synchronous (e.g. M.
securis, M. partumeium, P. variable, P. compressum) or asyn-
chronous (e.g. S. rhomboideum) reproduction. Pisidiids which
inhabit temperate, ephemeral ponds are synchronous in their
reproductive patterns and respond to uncertainty in the length
of time water remains in the pond (Way ef a/., 1980; Horn-
bach, Way, ef a/., 1980; McKee and Mackie, 1981; Way and
Wissing, 1982). By being synchronous in their reproductive
peiods, P. variable and P. compressum respond to habitat
optima set by seasonality and the effects of local fluctuations
in habitat features can be reduced by extending their
reproductive period over several months (Way and Wissing,
1982).

In other pisidiids (e.g. S. rhomboideum) reproductive
parents and newborn are present throughout most of the year
(Mackie and Flippance, 1983e). This type of asynchronous
reproduction ensures that the reproductive effort of an
organism is not entirely lost during periods of environmental
fluctuations. Several alternative adaptations to varying
degrees of environmental uncertainty are discussed by
Southwood (1977).

Many attempts (e.g. Mackie ef a/., 1978; Holopainen
and Hanski, 1979; Kraemer, 1979; Way efa/., 1980; Horn-
bach ef a/., 1980, 1982; McKee and Mackie, 1981) have been
made to predict corbiculacean life history from life history
features according to r, k, and stochastic theories (Stearns
1976, 1977, 1980). However, many intraspecific variations
in pisidiids cannot be explained by these theories (Way and
Wissing, 1982; Mackie and Flippance, 1983e). The mix of
reproductive strategies within a species may merely indicate

that the variations in life history patterns (i.e. traits without
coadaption) is more a function of local environmental imposi-
tions than of evolved (genetic) life history strategies (i.e.
coadapted traits). Often environmental stability is defined on
the basis of life history patterns; rather, life history should
be predictable on the basis of environmental stability, as
originally implied by McArthur and Wilson (1967). Calow
(1978) discusses the possible consequences of “r” and “k”
selection on gastropod life cycles.

LITERATURE CITED

Agrell, I. 1949. The shell morphology of some Swedish unionidesas
affected by ecological conditions. Arkiv fur Zoologi 41 A: 1 -30.
Aldridge, D. W. 1976. Growth, reproduction and bioenergetics in a
natural population of the Asiatic freshwater clam, Corbicula
manilensis Philippi. Master’s Thesis, The University of Texas
at Arlington, 97 pp.

Aldridge, D. W. and R. F. McMahon, 1978. Growth reproduction and
bioenergetics in the Asiatic freshwater clam, Corbicula
manilensis Philippi. Journal of Molluscan Studies 44, 49-70.
Alimov, A. F. 1965. Oxygen consumption of the freshwater mollusc,
Sphaerium corneum (L.). Zoologicheskii Zhurnal 44: 1 558-1 562.
Berg, K. and P. M. Jonasson. 1965. Oxygen consumption of pro-
fundal lake animals at low oxygen content of the water.
Hydrobiologia 26: 131-143.

Berg, K., P. M. Jonasson, K. W. Ockelmann. 1962. The respiration
of some animals from the profundal zone of a lake.
Hydrobiologia 19: 1-39.

Boozer, A. C. and P. E. Mirkes. 1979. Observations on the finger-
nail clam Musculium partumeium (Pisidiidae), and its associ-
ation with the introduced Asiatic clam, Corbicula fluminea. The
Nautilus 93: 73-83.

Boycott, A. E. 1936. The habitats of freshwater Mollusca in Britain.

Journal of Animal Ecology 5: 116-186.

Britton, J. C. and B. Morton. 1979. Corbicula in North America: The
evidence reviewed and evaluated. IN: J. C. Britton ed., Pro-
ceedings, First International Corbicula Symposium, pp.
250-287. Texas Christian University Research Foundation,
Forth Worth, Texas.

Burky, A. J., M. E. Benjamin, D. M. Catalano and D. J. Hornbach.
1979. The ratio of calcareous and organic shell components
of freshwater sphaeriid clams in relation to water hardness
and trophic conditions. Journal of Molluscan Studies 45:
312-321.

Burky, A. J. and K. A. Burky. 1976. Seasonal respiratory variation
and acclimation in the pea clam, Pisidium walked Sterki. Com-
parative Biochemistry and Physiology 55A: 109-1 14.

Burky, A. J., D. J. Hornbach and C. M. Way. 1981 . Growth of Pisidium
casertanum (Poli) in West Central Ohio. Ohio Academy of
Science 81: 41-44.

Busch, R. L. 1974. Asiatic clams Corbicula manilensis Philippi as
biological filters in channel catfish Ictalurus punctatus (Rafines-
que) cultures. Master’s Thesis, Auburn University, 84 pp.
Calow, P. 1978. The evolution of life-cycle strategies in freshwater
gastropods. Malacologia 17: 351-364.

Carter, J. G. 1980. Environmental and biological controls of bivalve
shell mineralogy and microstructure. IN: D. C. Rhoads and
R. A. Lutz eds., Skeletal Growth of Aquatic Organisms, pp.
69-114. Plenum Press, N.Y.

Clarke, A. H. 1979. Sphaeriidae as indicators of trophic lake stages
The Nautilus 84: 178-184.

Dance, S. P. 1958. Drought resistance in an African freshwater

228

CORBICULA SYMPOSIUM

bivalve. Journal of Conchology 24: 281-283.

Dietz, T.H. 1974. Body fluid composition and aerial oxygen consump-
tion in the freshwater mussel, Ligumia subrostrata (Say): Ef-
fects of dehydration and anoxic stress. Biological Bulletin 147:
560-572.

Dussart, G. B. J. 1976. The ecology of freshwater molluscs in north-
west England in relation to water chemistry. Journal of
Molluscan Studies 42: 181-198.

Dussart, G. B. J. 1979a. Sphaerium corneum (L.) and Pisidium spp.
Pfeiffer - The ecology of freshwater bivalve molluscs in relation
to water chemistry. Journal of Molluscan Studies 45: 19-34.

Dussart, G. B. J. 1979b. Life cycles and distribution of the aquatic
gastropod molluscs Bithynia tentaculata (L.), Gyraulus albus
(Muller), Planorbis planorbis (L.) and Lymnaea peregra (Muller)
in relation to water chemistry. Hydrobiologia 67: 223-239.

Gale, W. F. 1971 An experiment to determine substrate preference
of the fingernail clam, Sphaerium transversum (Say). Ecology
52: 367-370.

Gale, W. F. 1973. Substrate preference of the fingernail clam,
Sphaerium striatinum (Lamarck) (Sphaeriidae). Southwestern
Naturalist 18: 31-37.

Habel, M. L. 1970. Oxygen consumption, temperature tolerance, and
filtration rate of the introduced Asiatic clam Corbicula manilen-
sis from the Tennessee River. Master’s Thesis, Auburn Univer-
sity, 91 pp.

Flarrison, A. D. 1968. The effects of calcium bicarbonate concen-
tration on the oxygen consumption of the freshwater snail
Biomphalaria pfeifferi (Pulmonata: Planorbidae). Archiv fur
Hydrobiologie 65: 63-73.

Heard, W. H. 1965. Recent Eupera (Pelecypoda: Sphaeriidae) in the
United States. American Midland Naturalist 74: 309-317.

Heard, W. H. 1977. Reproduction of fingernail clams (Sphaeriidae:
Sphaerium and Musculium). Malacologia. 16: 421-455.

Holopainen, I. J. 1979. Population dynamics and production of
Pisidium species (Bivalvia, Sphaeriidae) in the oligotrophic and
mesohumic Lake Paajarvi, southern Finland. Archiv fur
Hydrobiologie Supplementband 54: 1-23.

Holopainen, I. J. and I. Hanski. 1979. Annual energy flow in popula-
tions of two Pisidium species (Bivalvia, Sphaeriidae), with
discussion of possible competition between them. Archiv fur
Hydrobiologie 86: 338-354.

Holopainen, I. J. and J. G. J. Kuiper. 1982. Notes on the morphometry
and anatomy of some Pisidium and Sphaerium species
(Bivalvia, Sphaeriidae). Annates Zoological Fennici 19: 93-107.

Hornbach, D. J., C. M. Way, and A. J. Burky. 1980. Reproductive
strategies in the freshwater sphaeriid clam, Musculium par-
tumeium (Say), from a permanent and a temporary pond.
Oecologia 44: 164-170.

Hornbach, D. J., T. E. Wissing and A. J. Burky. 1982. Life history
characteristics of a stream population of the freshwater clam,
Sphaerium striatinum Lamark (Bivalvia: Pisidiidae). Canadian
Journal of Zoology 60: 249-260.

Hunter, R. D. and Lull, W.W. 1977. Physiological and environmen-
tal factors influencing the calcium-to-tissue ratio in popula-
tions of three species of pulmonate snails. Oecologia (Berlin)
29: 205-218.

Hynes, H. B. N. 1972. Ecology of Running Waters. University of Toron-
to Press. 55 pp.

Juday, C. 1908. Some aquatic invertebrates that live under anaerobic
conditions. Transactions of the Wisconsin Academy of Science,
Arts, and Letters 16: 10-16.

Kraemer, L. R. 1977. Aspects of functional morphology of the man-

tle/shell and mantle/gill complex of Corbicula (Bivalvia:
Sphaeriacea: Corbiculidae). Bulletin of the American
Malacological Union For 1977: 25-31.

Kraemer, L. R. 1979. Corbicula (Bivalvia: Sphaeriacea) vs. indigenous
mussels (Bivalvia: Unionacea) in U.S. rivers: A hard case of
"interspecific competition?” American Zoologist 19:
1085-1096.

Lee, G. F. and W. Wilson. 1969. Use of chemical composition of
freshwater clamshells as indicators of paleohydrologic con-
ditions. Ecology 50: 990-997.

Lee, G. F. and W. Wilson. 1 974. Studies on the Ca, Mg, and Sr con-
tent of freshwater clamshells. Transactions of the Wisconsin
Academy of Science, Arts, and Letters 62: 173-180.

Macan, T. T. 1950. Ecology of freshwater Mollusca in the English
Lake District. Journal of Animal Ecology 19: 124-146.

MacArthur, R. H. and E. O. Wilson. 1967. The theory of island
biography. Monographs in Population Biology 1. Princeton
Univ. Press, New Jersey.

Mackie, G. L. 1978. Shell structure in freshwater Sphaeriacea
(Bivalvia: Heterodonta). Canadian Journal of Zoology 56: 1-6.

Mackie, G. L. 1979a. Dispersal mechanisms in Sphaeriidae
(Mollusca: Bivalvia). Bulletin of the American Malacological
Union For 1979: 17-21.

Mackie, G. L. 1979b. Growth dynamics in natural populations of
Sphaeriidae clams (Sphaerium, Musculium, Pisidium). Cana-
dian Journal of Zoology 57: 441-456.

Mackie, G. L. 1981 . Nearctic freshwater Sphaeriacea (Bivalvia). Pro-
ceedings of the 2nd American Malacological Union Sym-
posium on the endangered mollusks of North America. Bulletin
of the American Malacological Union For 1981: 49-52.

Mackie, G. L. and L. A. Flippance. 1983a. Growth dynamics in
Sphaerium rhomboideum (Bivalvia: Pisidiidae). Canadian
Journal of Zoology 61 : 868-873.

Mackie, G. L. and L. A. Flippance. 1983b. Relationships between
buffering capacity of water and the size and calcium content
of freshwater mollusks. Freshwater invertebrate Biology 2:
48-55.

Mackie, G. L. and L. A. Flippance. 1983c. Variations in calcium con-
tent of freshwater Mollusca in Ontario in relation to calcium
content of the water. Journal of Molluscan Studies 49: 204-21 1 .

Mackie, G. L. and L. A. Flippance. 1983d. Calcium sources for growth
of Musculium securis (Bivalvia: Pisidiidae). Canadian Journal
of Zoology 61 : 874-878.

Mackie, G. L. and L. A. Flippance. 1983e. Life history variations in
two populations of Sphaerium rhomboideum (Bivalvia:
Pisidiidae). Canadian Journal of Zoology 61: 860-867.

Mackie, G. L. and D. G. Huggins. 1976. Biological notes on Eupera
cubensis (Bivalvia: Sphaeridae) from Kansas. Journal of the
Fisheries Research Board of Canada 33: 1652-1656.

Mackie, G. L. and D. G. Huggins. 1983. Sphaeriacean clams of Kan-
sas. Technical Publication, State Biological Survey 14: 1-92.

Mackie, G. L., S. U. Qadri, and A. H. Clarke, Jr. 1974a. Develop-
ment of brood sacs in Musculium securis (Bivalvia:
Sphaeriidae). The Nautilus 88: 109-111.

Mackie, G. L., S. U. Qadri and A. H. Clarke, Jr. 1974b. Byssus struc-
ture of larval forms of the fingernail clam, Musculium securis
(Prime). Canadian Journal of Zoology 52: 945-946.

Mackie, G. L., J. B. Brooke, and M. R. Servos. 1983. Cause and
effect relationships between Mollusca and the acid-nuetralizing
capacity of acidifying lakes. Unpublished report of National
Research Council of Canada, Ottawa, Ontario.

Mackie, G. L., S. U. Qadri and A. H. Clarke, Jr. 1976. Reproductive
habits of four populations of Musculium securis (Bivalvia:

MACKIE: ADAPTATIONS OF PISIDIIDAE

229

Sphaeriidae) near Ottawa, Canada. The Nautilus 90: 76-86.

Mackie, G. L., S. U. Oadri and R. Reed. 1978. Significance of litter
size in Musculium securis (Bivalvia: Sphaeriidae). Ecology. 59:
1069-1074.

Mattice, J. S. and L. L. Dye. 1976. Thermal tolerance of the adult
Asiatic clam. IN: Thermal Ecology II, G. W. Esch and R. W.
McFarlane, eds., pp. 130-135. United States Energy Research
and Development Administration Symposium Series,
Washington, D.C.

McKee, P. M. and G. L. Mackie. 1980. Desiccation resistance in
Sphaerium occidentale and Musculium securis (Bivalvia:
Sphaeriidae) from a temporary pond. Canadian Journal of
Zoology 58: 1693-1696.

McKee, P. M. and G. L. Mackie. 1981. Life history adaptations of
the fingernail clams Sphaerium occidentale and Musculium
securis to ephemeral habitats. Canadian Journal of Zoology
59: 2219-2229.

McKee, P. M. and G. L Mackie. 1983. Respiratory adaptations of
the fingernail clams Sphaerium occidentale and Musculium
securis to ephemeral habitats. Canadian Journal of Fisheries
and Aquatic Sciences 40: 783-791.

McKillop, W. B. and A. D. Harrison. 1972. Distribution of aquatic
gastropods across an interface between the Canadian Sheild
and limestone formations. Canadian Journal of Zoology 50:
1433-1445.

McMahon, R. F. 1979. Response to temperature and hypoxia in the
oxygen consumption of the introduced Asiatic freshwater clam
Corbicula fluminea (Muller). Comparative Biochemistry and
Physiology 63 A: 383-388.

McMahon, R. F. 1982. The occurrence and spread of the intro-
duced Asiatic freshwater clam, Corbicula fluminea (Muller), in
North America: 1925-1982. The Nautilus 96: 134-141.

McMahon, R. F. and D. W. Aldridge. 1976. Respiratory response
to temperature and low oxygen tension in Corbicula manilen-
sis Philippi. Abstract of a paper presented at 39th Annual
Meeting of the American Society of Limnology and
Oceanography, Savannah, Georgia.

Meier-Brook, C. 1969. Substrate relations in some Pisidium species
(Eulamellibranchiata: Sphaeriidae). Malacologia 9: 121-125.

Precht, H., J. Christopherson, H. Hensel and W. Larcher. 1973.
Temperature and Life. New York: Springer-Verlag, 779 pp.

Russell-Hunter, W. D., M. L. Apley, A. J. Burky and R. J. Meadows.
1967. Interpopulation variation in calcium metabolism in the
stream limpet, Ferrissia rivularis (Say). Science 1 55: 338-340.

Russell-Hunter, W. D., A. J. Burky and R. D. Hunter. 1981. Inter-
population variation in calcareous and proteinaceous shell
components in the stream limpet, Ferrissia rivularis.
Malacologia 20: 255-266.

Sinclair, R. M. and B. G. Isom. 1961. A preliminary report on the

introduced Asiatic clam Corbicula in Tennessee. Tennessee
Stream Pollution Control Board, Tennessee Department of
Public Health, Nashville, Tennessee, 31 pp.

Sinclair, R. M. and B. G. Isom. 1963. Further studies on the intro-
duced Asiatic clam Corbicula in Tennessee. Tennessee
Stream Pollution Control Board, Tennessee Department of
Public Health, Nashville, Tennessee, 76 pp.

Singer, R. 1 981 . Notes on the use of shells of Anodonta grandis (Say)
(Bivalva: Unionidae) as a paleoecological indicator of trophic
status and pH. IN: Effects of Acid Precipitation on Benthos
R. Singer, ed., pp. 103-1 1 1 . Proceedings of a Symposium on
Acidic Preparation on Benthos, 1980, North American Ben-
thological Society, Hamilton, New York.

Southwood, T. R. E. 1977. Habitat, the templet for ecological
Strategies. Journal of Animal Ecology 46: 337-365.

Stearns, S. C. 1976. Life history tactics: a review of the ideas.
Quarterly Review of Biology 51 : 3-47.

Stearns, S. C. 1977. The evolution of life history traits: a critique of
the theory and a review of the data. Annual Review of Ecology
and Systematics 8:1-1 1 .

Stearns, S. C. 1980. A new view of life histroy evolution. Oikos 35:
266-281 .

Thiel, M. E. 1926. Formwachstumversuche an Sphaerium corneum.
IV. Roux’ Arch. Entwicklungsmechanik 108: 87-137. (Not seen,
cited by Holopainen and Kuiper, 1982).

Thomas, J. D., M. Benjamin, A. Lough and R. H. Aram. 1974. The
effects of calcium in the external environment of the growth
and natality rates of Biomphalaria glabrata (Say). Journal of
Animal Ecology 43: 839-860.

Thut, R. N. 1969. A study of the profundal bottom fauna of Lake
Washington. Ecological Monographs 39: 79-100.

Way, C. M., D. J. Hornbach and A. J. Burky. 1980. Comparative life
history tactics of the sphaeriid clam, Musculium partumeium
(Say), from a permanent and a temporary pond. American
Midland Naturalist 104: 319-327.

Way, C. M. and T. E. Wissing. 1982. Environmental heterogeneity
and life history variability in the freshwater clams, Pisidium
variabile (Prime) and Pisidium compressum (Prime) (Bivalvia:
Pisidiidae). Canadian Journal of Zoology 60: 2841-2857.

Williams, N. V. 1970a. Studies on aquatic pulmonate snails in Cen-
tral Africa. 1. Field distribution relation to water chemistry.
Malacologia 10: 153-164.

Williams, N. V. 1970b. Studies on aquatic pulmonate snails in Cen-
tral Africa. 2. Experimental investigation of field distribution
patterns. Malacologia 10: 165-180.

Yonge, C. M. 1962. On the primitive significance of the byssus in
the Bivalvia and its effects in evolution. Journal of the Marine
Biological Association of the United Kingdom 42: 112-125.

GROWTH, LIFE CYCLE, UPPER THERMAL LIMIT AND
DOWNSTREAM COLONIZATION RATES IN A NATURAL
POPULATION OF THE FRESHWATER BIVALVE MOLLUSC,
CORBICULA FLUMINEA (MULLER) RECEIVING THERMAL

EFFLUENTS1

ROBERT F. MCMAHON AND CAROL J. WILLIAMS2
SECTION OF COMPARATIVE PHYSIOLOGY
DEPARTMENT OF BIOLOGY
BOX 19498

THE UNIVERSITY OF TEXAS AT ARLINGTON

ARLINGTON, TEXAS 76019, U.S.A.

ABSTRACT

Quantitative bimonthly to monthly samples of Corbicula fluminea were collected from March
1981, through December 1982, from a power station’s discharge canal receiving thermal effluents.
Population density peaks occurred in the spring and fall. Maximum density was 21930 clams/m2. The
population was completely eliminated at summer water temperatures > 36°C, the apparent absolute
upper lethal limit of this species. The population became reestablished in the fall by passive hydrological
transport of juveniles through the plant’s steam condensers from an endemic intake population.
Recolonization occurred only after temperatures fell below 30°C, suggesting that higher temperatures
may inhibit successful settlement by young individuals. Recolonization was rapid, ranging from 319
clams nr2 day1 in 1981 to 522 clams nr2 day1 in 1982. Spawning occurred in late spring and again
in early summer leading to “early” and “late” generations, respectively. Both the early and late genera-
tions continued to grow until the upper lethal limit was reached in July. The early generation was ther-
mally eliminated in the summer, but became reestablished in the discharge canal the following fall,
disappearing from the population due to thermal elimination the following spring. Both generations
survived less than one year and never exceeded a shell length of 18 mm. The high capacity of C.
fluminea for rapid downstream dispersal and colonization is an adaptation to unstable, disturbed, lotic
habitats and, in large part, is responsible for this species’ spread through North American drainage
systems. The relatively low upper thermal limit (36°C) of C. fluminea and the apparent inability of juveniles
to successfully settle at temperatures > 30°C have important implications for the control of this species
in industrial raw water systems.

The introduced, Asian freshwater bivalve, Corbicula
fluminea (Muller), is reported to occur in the heated effluent
canals of electrical power stations (Cherry, etal., 1980; Dreier
1977; Dreier and Tranqilli, 1981; Eckbald, 1975; Rodgers, et
al., 1977, 1979; Thomas and MacKenthum, 1964). In
discharge canals mid-summer ambient water temperatures
can surpass the upper lethal limits of resident C. fluminea
populations (Dreier, 1977; Dreier and Tranquilli, 1981). While
long-term upper lethal temperature limits of C. fluminea have
been determined in controlled laboratory studies (Mattice and

'This research was supported by a grant from the Texas Electric
Service Company to R. F. McMahon.

2Present address: Department of Microbiology, The University of
Texas Health Science Center, 5323 Harry Hines Boul., Dallas, Texas,
75235, U.S.A.

Dye, 1976), and anecdotally estimated from populations in
outdoor pools (Busch, 1974; Greer and Ziebell, 1972; Habel,
1970; Haines, 1979), no rigorous attempt has been made to
determine the absolute upper lethal temperature limit for a
population of this species in its preferred natural lotic habitat
(McMahon, 1983). As power station discharge canals close-
ly approximate the preferred riverine habitat of C. fluminea
(Kraemer, 1979; McMahon, 1982, 1983), they provide a
unique opportunity to assess this species’ temperature
tolerance limits under natural conditions. Such canals also
allow the assessment of the rate of passive, current mediated,
downstream dispersal and colonization by juveniles of this
species after resident clam populations are thermally
eliminated in mid-summer.

Information regarding field thermal tolerance limits and
juvenile downstream colonization rates are of obvious

American Malacological Bulletin, Special Edition No. 2(1 986): 23 1-239

231

232

CORBICULA SYMPOSIUM

significance to the future development of biofouling control
measures for this species (for review of biofouling by C.
fluminea see Goss, etal., 1979; McMahon, 1977, 1983). This
paper describes an investigation of a natural population of
C. fluminea inhabiting the thermal effluent canal of a gas-
fired steam-electric power station in north central Texas in
which field estimations of upper thermal limits and
downstream recolonization rates were determined from
bimonthly to monthly samples over a 22 month period from
1981 to 1982.

MATERIALS AND METHODS

Over a period extending from 25 March 1 981 , through
17 December 1982, specimens of C. fluminea were quan-
titatively sampled from a population in the thermal discharge
canal of the Handley Power Station of the Texas Electric Ser-
vice Company on Lake Arlington, Tarrant County, Texas. This
population was composed entirely of the “white” morphotype
of C. fluminea (Hillis and Patton, 1982). The sampling site was
approximately 200 m downstream from the opening of the
effluent discharge pipe. Lake Arlington was formed from an
artificial impoundment of Village Creek, a tributary of the West
Fork of the Trinity River. At capacity it has a surface area
of 920.7 ha and a volume of 56 x 106 m3 H20 (Dowel! and
Breeding, 1966). The Handley Power Station has five gas-
fired generating units with a maximum output of 1471 MW.
The discharge canal is approximately 15 m wide, 3-4 m deep
and 1.5 km long (For a map of Lake Arlington and the
discharge canal see Williams and McMahon, 1986). Max-
imum heated effluent discharge rate was 4716.1 x 10 6 /
day1.

Specimens of C. fluminea were collected bimonthly to
monthly from a substratum of clay and gravel with an Eckman
Dredge (sampling area = 0.19 m2). The dredge was mounted
on the end of streel pole which allowed it to be forced deeply
into the substratum. Dredged material was passed through
a 1 mm mesh sieve and all individuals of C. fluminea removed.
Sampling continued until at least 100 individuals were taken
(sample size range = 100-1586). The sieve retained all in-
dividuals with shell lengths greater than 0.9 mm. Living
specimens were immediately fixed in 12% neutralized for-
maldehyde (by volume). At each collection ambient air and
water temperatures, pH, conductivity, and dissolved oxygen
concentration were recorded. Water hardness values were
also determined on selected collection dates by EDTA titra-
tion (Hach, Model HAC-DT Water Hardness Test Kit). Daily
records of mean and maximum discharge ambient water
temperature and effluent discharge rate were obtained from
the Texas Electric Service Company.

After return to the laboratory the shell lengths (SL, the
greatest anterior-posterior dimension across the shell valves)
of each individual in the collection were measured to the
nearest 0.1 mm. The shell lengths of larger individuals (SL
> 6.0 mm) were measured with a dial caliper while those
of smaller individuals (SL < 6.0 mm) were measured with
an ocular micrometer mounted in a dissecting microscope
under 10X magnification. For each collection the number of

individuals in each 0.2 mm size class were expressed as a
percentage of the total sample size and plotted as frequency
histograms in sets corresponding to collection dates. Visual
examination of size class distribution across sequential col-
lections allowed each sample to be divided into separate
generations characterized by distinctly different shell length
distributions (after the method of Aldridge and McMahon,
1978). A mean SL, standard deviation, and range of SL were
then computed for each generation in each sample.

RESULTS

The mean conductivity of discharge water during the
collection period was 274 ^mho cm-2 (s.d. = ± 35.1, s.e.
= ±5.9, range = 190-320, n = 35). Mean water hardness
was 102 mg Ca /-1 (s.d. = ± 19.5, range = 83-122, n =
3). Both values are indicative of moderately hard waters.
Mean pH (computed from H + concentrations) was 7.78 (s.d.
= ± 0.56, s.e. = ± 0.10, range = 6.69-8.52, n = 35) which
is characteristic of relatively alkaline habitats. Mean ambient
dissolved 02 concentration was 7.6 mg 02 /_1 (s.d. = ± 0.4,
range = 5.5-12, n = 29). The mean difference between
discharge water ambient 02 concentration and that at full air
saturation was -0.7 mg 02 Z*1 (s.d. = ±1.6, s.e. = ± 0.3,
range = ± -3.5-1 .9, n =29) indicating that the resident C.
fluminea population was rarely if ever exposed to biologically
significant levels of hypoxia.

Mean ambient air temperature at collection was
22.4°C (s.d. = ± 7.7, s.e. = ± 1.3, n = 35). Air temperature
reached a maximum of 36°C on 28 July 1 981 , and a minimum
of 6.7°C on 8 February 1982. Mean ambient discharge water
temperature at the time of collection was 26.7°C (s.d. = ± 8.7,
s.e. = ± 1.4, n = 35). A peak discharge water temperature
of 40.8°C occurred on 29 June 1981, and a minimum
temperature of 1 0.5°C on 8 February 1 982 (Fig. 1). Daily mean
discharge water temperatures (monitored continuously by the
Texas Electric Service Company) were averaged over se-
quential three day periods (Figs. 1 and 2). The mean of these
three day average discharge temperatures over the sampl-
ing period was 25.3°C (s.d. = ± 8.1, s.e. = ± 0.6, n = 196).
A maximum three day average temperature of 39.0°C was
recorded on 13-15 August 1982, and a minimum of 11.3°C
on 16-18 January 1982 (Figs. 1 and 2). The mean daily max-
imum discharge water temperature (computed as sequen-
tial three day averages) over the sampling period was 27.1°C
(s.d. = ± 8.5, s.e. = ±0.6, n = 195). The highest three
day average maximum temperature recorded was 41 ,7°C on
13-15 August 1982. A minimum temperature of 12.4°C was
recorded on 13-15 January 1982, 16-18 January 1982, and
7-9 February 1982 (Fig. 2).

The monthly mean of daily discharge flow rates
averaged 2545.8 x 106 / day1 (s.d. = ± 558.9 x 10®, s.e.
= ± 119.2 x 10®, n = 22). The maximum mean monthly
discharge rate of 3628.6 x 10® / day1 occurred in June 1981 ,
while a minimal monthly mean discharge rate of 1 649.4 x 1 0®
Z day1 occurred during November 1981 (Table 1). An ab-
solute maximum discharge rate of 4623.5 x 10® / day1 was
recorded on 8 June 1981 , 24, 25, and 26 July 1982, and 27

MCMAHON AND WILLIAMS: CORBICULA IN THERMAL EFFLUENTS

233

and 30 August 1982. There was no water discharged on 2
December 1981. The mean of the monthly averages of
discharge current flow rate over the course of the study was
3.9 m min"1 (s.d. = ± 0.9, s.e. = ± 0.2, n = 22). Maximum
and minimum average monthly discharge currents were 5.6
m min-1 and 2.5 m min'1 in June 1981 , and November 1981 ,
respectively (Table 1).

The means and standard deviations of shell length (SL)

for each generation in each sample are displayed in Figure

1 . Two generations per year occurred in the Handley Power
Station discharge canal. An "early” generation resulting from
a spring reproductive period appeared in the late spring and
early summer of each year of collection and is designated
in Figure 1 as either E-81 which first appeared as a distinct
size class with a mean SL of 3.6 mm on 20 May 1 981 , or as
E-82 which first appeared in the samples on 7 July 1 982 with
a mean SL of 5.1 mm. A second, or “late” generation
resulting from a second, late summer reproductive period ap-

1981 1982

Fig. 1. Generation mean shell lengths of the Corbicula fluminea population in a power station’s thermal effluent discharge canal on Lake
Arlington, Texas. The horizontal axis for both figures is months over the collection period. A. Variation in the ambient water temperature
of the discharge canal over the collection period. Vertical axis is ambient water temperature in °C, open circles are three day averages of
mean daily discharge water temperatures and solid horizontal bars indicate periods when the C. fluminea population was thermally eliminated
from the discharge canal. B. Generation mean shell lengths for consecutive samples of the discharge C. fluminea population. The horizontal
axis is mean shell length in millimeters, open circles are the mean shell lengths of each generation in each consecutive population sample
over the collecting period, and the vertical bars about each mean are standard deviations. Generation shell length means without standard
deviation bars represent sample sizes of less than six individuals. Circles connected by solid lines indicate the change in mean shell lengths
through time of specific generations identified by the reproductive period that gave rise to that generation (i.e. E-81 indicates the generation
produced during the early reproductive period of 1981 and L-82, that produced during the late reproductive period in 1982). The dashed
line connects the mean shell lengths of the E-81 and E-82 generations across periods during which all individuals were thermally eliminated
from the population. The solid horizontal bars indicate periods when the C. fluminea population was thermally eliminated from the discharge canal.

234

CORBICULA SYMPOSIUM

a a

00ooV

¥.

✓

o©a
^ » °l

o<p

c < °c 4p
i itf,
%

M, <

«P

o %< V

OQA

■ X

MAM

J A
1981

N D

J J
1982

N D

Fig. 2. Seasonal variations in the density of a Corbicula fluminea population in relation to ambient water temperature in the thermal effluent
discharge canal of a power station on Lake Arlington, Texas. The horizontal axis for both figures is months of the year over the collecting
period. A. Seasonal variation in ambient water temperature. Vertical axis is ambient water temperature in °C, open circles are averages of
mean daily water temperatures for consecutive three day periods, open triangles are averages of daily maximum water temperatures for con-
secutive three day periods, solid squares are ambient water temperatures measured at the collection site at the time of collection, and solid
horizontal bars indicate periods when all individuals of C. fluminea were thermally eliminated from the discharge canal. B. Seasonal variations
in the densities of the total C. fluminea population and individual generations over the course of the collection period. The vertical axis is
density in hundreds of clams/m2 on a logarithmic scale. Open circles connected by solid lines are the density of the total population. The
densities of individual generations are indicated as follows: E-80 (resulting from the early reproductive period of 1980), exes connected by
dashed lines; L-80 (resulting from the late reproductive period of 1980), open triangles connected by solid lines; E-81 , solid circles connected
by dashed lines; L-81, open squares connected by solid lines; E-82, solid triangles connected by dashed lines; L-82 open diamonds con-
nected by solid lines. The solid horizontal bars represent periods when all individuals of C. fluminea were thermally eliminated from the population.

MCMAHON AND WILLIAMS: CORBICULA IN THERMAL EFFLUENTS

235

Table 1. Monthly values of mean water discharge rate, standard deviation, standard error, range and mean current flow rate in the heated
effluent discharge canal of the Handley Steam Electric Power Station, Tarrant County, Texas.

Mean Monthly Standard Standard Mean Monthly

Discharge Rate Deviation Error Range Days Flow Rate

Month

/x106/day

Recorded

m/min

Mar.,

1981

1899.0

±572.4

± 102.8

59.1-2434.8

2.9

Apr.,

1981

1859.8

± 533.5

±97.4

759.4-2926.1

2.9

May,

1981

2251.7

±885.4

±159.0

781.3-3519.4

3.5

Jun.,

1981

3628.6

±736.3

± 134.4

2032.0-4623.5

5.6

Jul.,

1981

3375.4

±608.5

±109.3

2075.2-4259.0

5.2

Aug.,

1981

2598.2

±816.1

±151.5

352.5-4432.7

4.0

Sep.,

1981

2424.7

±549.9

± 100.4

1253.0-3425.8

3.7

Oct.,

1981

2411.5

±545.7

±98.0

1253.0-3425.8

3.7

Nov.,

1981

1649.4

±935.7

±170.8

60.6-3324.2

2.5

Dec.,

1981

2248.3

±844.9

±151.7

0.0-4201.8

3.5

Jan.,

1982

2752.8

±1049.8

± 188.6

1293.9-4601.5

4.2

Feb.,

1982

2462.8

±675.8

± 127.7

1321.7-3483.9

3.8

Mar.,

1982

2419.8

±780.1

±140.1

821.5-4201.6

3.7

Apr.,

1982

2308.3

±774.8

±141.5

1312.6-3595.8

3.6

May,

1982

2256.8

±517.8

±93.0

1331.1-2882.7

3.5

Jun.,

1982

2714.5

±781.1

±142.6

1334.4-4220.2

4.2

Jul.,

1982

3448.9

±879.1

±157.9

1500.1-4623.5

5.3

Aug.,

1982

3462.4

±884.0

±158.8

2075.2-4623.5

5.3

Sep.,

1982

3164.0

±707.2

±128.3

2056.7-4242.0

4.9

Oct.,

1982

2538.0

±826.7

± 148.5

1352.9-3618.7

3.9

Nov.,

1982

2246.2

±827.2

± 148.6

58.9-3310.9

3.5

Dec.,

1982

1886.6

±867.8

± 155.9

58.9-3326.2

2.9

peared in early autumn and is designated in Fig. 1 as either
L-81 , first occurring in the samples on 9 November 1981 , at
a mean SL of 3.4 mm or as 1-82, first occurring on 27 Oc-
tober 1982, with a mean SL of 2.3 mm (Fig. 1). Also occurr-
ing during the early portions of the sampling period were the
E-80 and L-80 generations which resulted from respective
spring and fall reproductive periods in 1980 (Fig. 1).

During both 1981 and 1982 two annual peaks of den-
sity occurred in the discharge canal C. fluminea population,
the first in early April (16938 clams/m2 on 3 April 1981 , and
11889 clams/m2 on 5 April 1982) and a second in mid-fall
(7656 clams/m2 on 9 November 1981 , and 21930 clams/m2
on 10 November 1982). These spring and fall density max-
ima were associated with the appearance and rapid ac-
cumulation of new individuals of the “early” and “late”
generations, respectively (Fig. 2) (Williams and McMahon,
1986).

The densities of all generations declined rapidly in both
years of the study in June and early July as water
temperatures rose above 30°C (Fig. 2). All living individuals
were eliminated from the discharge canal by the end of July
in both 1981 and 1982 (Fig. 2). This mid-summer extinction
of all living clams was associated with a rise in daily, average,
ambient water temperature above 36°C and with daily water
temperature maxima generally greater than 40°C (Fig. 2).
Following this apparent mid-summer thermal extinction, no
living individuals of C. fluminea were taken in the discharge
canal until 9 November 1981 or 29 September 1982 (Fig. 2).
This elimination of the Corbicula population from the

discharge canal appeared to be entirely temperature depen-
dent as daily discharge volume and flow rates (which could
possibly carry individuals downstream away from the popula-
tion) remained near mid-summer levels in the fall when the
population displayed a rapid increase in density in both 1981
and 1982 (Table 1, Fig. 2).

In the fall of both 1981 and 1982 recolonization of the
discharge canal by C. fluminea was extremely rapid and
presumably resulted from the passive downstream transport
of juvenile and young clams from a viable population occur-
ing in the power station’s intake canal and its associated lake
inlet (Williams and McMahon, 1986; for a discussion of
passive dispersal of juvenile C. fluminea on water currents
see McMahon, 1982, 1983 and references therein). This per-
manent inlet canal population was a source of juvenile and
young individuals which were entrained through the power
station’s steam condensers and carried into the discharge
canal with condenser effluents (Williams and McMahon,
1986). As such, the rate of reestablishment of the discharge
canal popoulation could provide an estimate of the
downstream colonization rate of C. fluminea by passive
hydrological transport. On 15 October 1981, no living clams
occurred in the discharge canal, while on 9 November 1981,
population density had reached a maximum of 7656 clams/m2
yielding a downstream colonization rate of 319 clams nr2
day1. Correspondingly, only 6 clams/m2 were recorded on
29 September 1982, density thereafter rose to a maximum
of 21930 clams/m2, yielding a downstream colonization rate
of 522 clams nr2 day1. As the approximate substratum sur-

236

CORBICULA SYMPOSIUM

face area of the discharge canal was 6.75 x 104 m2,
downstream dispersal rates in the entire canal can be
estimated as 2.15 x 107 clams/day and 3.52 x 107 clams/day
in the falls of 1981 and 1982, respectively. These high disper-
sal rates are associated with the fall reproductive period and
release of large numbers of juveniles of the late generation
(L-81 in 1981 and L-82 in 1982, Fig. 1) which accounted for
the vast majority of recolonizing individuals (Fig. 2). After fall
recolonization the shell length distribution of the early genera-
tions (E-81 and E-82) increased in size throughout the winter
and reached a maximum of 17-18 mm before disappearing
from the population in the following spring (Fig. 1). The late
generations (L-80 and L-81 ) grew very little during the winter.
Rapid growth occurred the following spring, the late genera-
tions reaching a maximum mean SL of 12 to 15 mm just
before thermal extinction in mid-summer. Therefore, both the
spring and fall generations have highly attenuated life cycles,
each generally surviving one year or less in the discharge
canal (Fig. 1). The analysis of generation growth patterns is
complicated by continual immigration of individuals from in-
take populations into the discharge canal (Williams and
McMahon, 1986). As such, the growth rates of small in-
dividuals (< 10 mm SL) in the discharge canal essentially
reflect those of the population in the lake proper (see the data
of Williams and McMahon, 1986). However, as larger
specimens of C. fluminea (< 15 mm SL), not subject to
passive hydrological dispersal (see below), maintained high
growth rates in the discharge canal up to the point of ther-
mal extinction it appears that temperatures approaching the
upper lethal limit do no inhibit growth in this species (Fig. 1).

Only relatively small individuals of C. fluminea occurred
in the discharge canal population. No living specimens with
an SL of greater than 1 8.3 mm occurred among the 9868 in-
dividuals taken over the collection period. As both the intake
canal (Williams and McMahon, 1986) and intake embayments
behind the traveling screens (McMahon, 1977) harbor popula-
tions of C. fluminea with individuals ranging in SL from 1 mm
to 40 mm and as individuals with SL’s > 20 mm routinely
occur in the discharge side of the condensers (McMahon,
1 977) the lack of large living specimens in the discharge canal
appears to indicate that there may be a maximum size in this
species for successful passive hydrological dispersal and
downstream settlement (Table 1).

In both 1 981 and 1 982, individuals of the late genera-
tions (L-80 and L-81) did not recolonize the discharge canal
after thermal extinction even though individuals of these
generations occurred in relatively high densities in upstream
areas of both the intake canal and intake embayments and
were of an SL range that would readily allow them to pass
through the tubes of the power station’s steam condensers
(McMahon, 1977). As these generations did not reappear in
the collections after thermal elimination (Figs. 1 and 2) they
may have grown beyond a size at which successful
downstream resettlement could normally occur in lotic
habitats, even though the passive hydrological transport of
similar sized individuals has been reported (Williams and
McMahon, 1986; McMahon, 1977; Prezant and Chalermwat,
1984). As the maximum sizes of recolonizing individuals in

the falls of 1981 and 1982, were 14.3 mm and 12.8 mm,
respectively, and as no individuals of the previous years late
generation were recovered after thermal elimination it ap-
pears that only individuals of C. fluminea of less than 15 mm
SL or one year in age are able to successfully disperse to
and colonize downstream habitats.

DISCUSSION

The C. fluminea population disappeared from the
discharge canal of the Handley Power Station after average
ambient water temperatures rose above 36°C in both 1981
and 1982 (Fig. 2), suggesting that the absolute upper lethal
limit of this species lies very near that temperature. While
C. fluminea can tolerate short-term (acute) exposures as high
as 43°-47°C (Mattice, 1979; Mattice and Dye, 1976; Isom, et
al., 1978; McMahon, 1979), it is far less tolerant of long-term
(> 24 h) exposures to elevated temperatures (Mattice and
Dye, 1976). Our field estimate of upper lethal limit at 36°C
corresponds well with the long-term (96 hour exposure) max-
imum upper lethal limit of 34°C reported in a laboratory study
of individuals of C. fluminea acclimated to 32°C (Mattice, 1979;
Mattice and Dye, 1976). Other laboratory determinations of
tolerance of extended exposure to high temperatures place
the upper lethal limit of C. fluminea between 33.5°C and
38.0°C (Cherry, et a!., 1980; Habel, 1970; Mudkhede and
Nagabhushanam, 1977). In artificial outdoor ponds C.
fluminea is reported to have a somewhat lower long-term
temperature tolerance, reported upper lethal limits ranging
from 30°-35°C (Busch, 1973; Greer and Ziebell, 1972; Habel,
1970; Haines, 1979). However, such estimates may be con-
founded by exposure of experimental individuals to severe
hypoxic stress in artificial standing water habitats (Busch,
1974; Habel, 1970; McMahon, 1979).

Another C. fluminea population in a discharge canal
receiving heated effluents has also been reported to have
been eliminated when mid-summer water temperatures
reached 40°C (Dreier, 1977; Dreier and Tranquilli, 1981).
Such data have lead to suggestions that the introduction of
heated discharge water into intake areas either by backflash-
ing through steam condensers (Goss, et al., 1979; Mattice,
1979; Mattice, et al., 1982) or by recirculation of heated ef-
fluents from discharge canals into intake structures (Mattice,
1979; Mattice, et al., 1982) could raise water temperatures
to levels which would eradicate Corbicula populations and,
therefore, eliminate the major sources of juvenile and small
clams impinging a power station’s raw water systems.

Periodic recycling or backflushing of thermal effluents
through steam condensers into intake areas may not be a
feasible eradication procedure for C. fluminea as this species
can withstand short-term exposures (15-30 min) to
temperatures ranging from 43-45°C (Mattice and Dye, 1976;
Isom, et al., 1978; McMahon, 1979). Such temperatures are
rarely achieved in the thermal effluents of most power sta-
tions (Figs. 1 and 2, Dreier and Tranquilli, 1981 ; Bird, 1976;
McMahon, 1975, 1976a; Cherry et al., 1980; Gibbons and
Sharitz, 1974; Esch and McFarlane, 1976). Indeed, the mix-
ing of backflushed heated effluents with cooler intake waters

MCMAHON AND WILLIAMS: CORBICULA IN THERMAL EFFLUENTS

237

would probably reduce their temperatures well below lethal
levels.

While short-term temperature shock appears imprac-
tical as a control measure for Corbicula , the results of our
observations indicate that longer-term recycling of thermal ef-
fluents into intake areas during warmer summer months may
have promise as a control methodology. Summer water
temperatures in Texas lakes and reservoirs (and presumably
in other aquatic habitats in the southern United States)
routinely reach or surpass 30° to 32°C (see the data of Aldridge
and McMahon, 1978; McMahon, 1975, 1976a, 1976b; Bird,
1976; Tommey, 1976). Long-term recirculation of thermal ef-
fluents into intake areas during these periods could allow
maintenance of intake water temperatures above the field up-
per lethal limit of 36°C eliminating C. fluminea populations
from intake canals, embayments and other structures within
14 days (Figs. 1 and 2).

In the summers of 1981 and 1982 the C. fluminea
population in the Handley Power Station’s discharge canal
was reestablished (presumably by individuals hydrologically
transported through the steam condensers from resident in-
take populations) in late September (1982) or early October
(1981). In both of these years reestablishment of the popula-
tion did not occur until average ambient water temperatures
in the discharge canal had fallen below approximately 30°C
(Figs. 1 and 2), a temperature approximately 6°C below the
apparent upper lethal limit. Clams remained absent from the
canal during the periods when water temperatures ranged
from 30-36°C (September through early October) even though
entrainment of young individuals through steam condensers
into the discharge canal remained at relatively high levels ( >
107 juveniles/day, see the data of Williams and McMahon,
1986). The inability of juvenile C. fluminea to become suc-
cessfully established in the substratum at temperatures above
30°C is somewhat surprising as this species has been
reported to survive summer temperatures above 30°C in
natural populations (Aldridge and McMahon, 1978; Dreier and
Tranquilli, 1981; Williams and McMahon, 1986). It is temp-
ting to hypothesize that at temperatures above 30°C small
individuals of C. fluminea may be incapable of successfully
producing or anchoring themselves to the substratum with
the larval mucilagenous byssal thread (Kraemer, 1979;
Sinclair and Isom, 1963) and, therefore, would be incapable
of successful settlement at these temperatures. Indeed, sharp
declines in the density of the C. fluminea population in the
Handley Power Station’s discharge canal occurred in both
1981 and 1982, as spring ambient water temperatures rose
above 30°C (Fig. 2), indicating not only an inhibition of set-
tlement by newly released juveniles, but also the possible
hydrological removal of individuals that had become
established in the substratum the previous fall. The possibility
that temperatures greater than 30°C can inhibit successful
settlement of juvenile Corbicula has very important implica-
tions to the development of control measures for this species
in industrial facilities utilizing large quantities of raw water.
Certainly, the effects of temperature on byssus formation and
successful establishment in the substratum by juvenile C.
fluminea warrants further study.

High sublethal temperatures appeared to have little
or no effects on the generation growth rates of the C. fluminea
population in the Handley Power Station’s discharge canal.
Instead, growth rates remained constant or increased as the
upper thermal limit of 36°C was approached (Fig. 1). Examin-
ation of other published growth rate data for C. fluminea also
seems to indicate a lack of high temperature inhibition of
growth (Aldridge and McMahon, 1978; McMahon and
Williams, 1986; Pool and Tilly, 1977). Rather, increasing
temperatures appear to have direct stimulatory effects on
growth rate (McMahon and Williams, 1986), including
evidence for stimulated growth rates in populations receiv-
ing thermal effluents (Mattice, 1979; Dreier and Tranquilli,
1981).

Juvenile (SL = 0.2 mm) and young specimens (SL <
5.0 mm) of C. fluminea are passively transported in large
numbers on water currents (Goss and Cain, 1977; Goss et
al., 1979; Sickel, 1979; Sinclair, 1964; Sinclair and Isom,
1963). Entrainment of the Handley Power Station’s raw water
systems by such hydrologically transported specimens has
been reported to reach levels on the order of 108 clams/day
(Williams and McMahon, 1986). Such passive downstream
dispersal has been claimed to be responsible for the rapid
spread of this species in North American drainage systems
(McMahon, 1982, 1983). Recently, it has been suggested that
such hydrological transport is associated with the ability of
C. fluminea to produce mucus draglines from the exhalent
siphon that increase it’s susceptibility to passive suspension
in the water column (Prezant and Chalermwat, 1984). Cer-
tainly, passive hydrological transport accounted for the ex-
tremely rapid reestablishment, after thermal elimination, of
the Handley Power Station’s discharge canal population of
C. fluminea as fall temperatures returned to favorable (<
30°C) levels (Fig. 2). The reestablishment of the discharge
canal population was extremely rapid, densities increasing
from 0 to 7656 clams/m2 in 25 days in 1981 and from 6.4 to
21930 clams/m2 in 42 days in 1982 (Fig. 2) yielding coloniza-
tion rates of 319 clams rrr2 day1 and 522 clams nr2 day1,
respectively. A similar mid-summer thermal extinction and
fall reestablishment of a discharge canal C. fluminea popula-
tion has been reported for an electrical power station on Lake
Sangchris, Illinois. This population’s density increased from
3 clams/m2 (below 1 m in depth) on 12 August, 1975 (after
water temperatures had reached 40°C) to 430 clams/m2 in
February, 1976 (Dreier, 1977; Dreier and Tranquilli, 1981)
yielding a recolonization rate of 2.5 clams nr2 day1. After
a mid-winter extinction of a C. fluminea population occurred
in the New River, Virginia, when water temperatures fell below
2°C clam densities recovered from 0.0 clams/m2 in early April
to 1000 clams/m2 in late September, (Cherry et al., 1980;
Rodgers, et al., 1979) yielding a recolonization rate of 5.5
clams nr2 day1. As the vast majority of individuals in this
reestablished population were immature (SL < 7.5 mm)
(Cherry, ef al., 1980; Rodgers, etal., 1979), it must be assum-
ed that they were passively carried downstream on water cur-
rents from a viable upstream population.

The results of this study indicate that only individuals
with an SL of less than 15 mm became reestablished in the

238

CORBICULA SYMPOSIUM

Handley Power Station’s discharge canal after thermal ex-
tinction. The vast majority of these individuals were juveniles
of the most recent late generation with an SL less than 5 mm
(Figs. 1 and 2). Similarly, a portion of the New River in Virginia
was recolonized by individuals of C. fluminea less than 13.5
mm SL following a mid-winter extinction of the endemic
population (see above) (Cherry, eta/., 1980; Rodgers, etal.,
1979), indicating that successful downstream dispersal and
colonization is essentially limited to smaller, mostly immature
specimens. While larger adult specimens are also subject to
passive hydrological transport (Williams and McMahon,
1986), their numbers are so small in relation to passively
transported juvenile and immature specimens as to be of no
real significance to the passive downstream dispersal of this
species.

The extensive capacity of juvenile C. fluminea for
passive hydrological transport is not only associated with its
ability to invade and foul industrial raw water systems (Goss,
et al., 1979), but may also be primarily responsible for this
species’ remarkably rapid disersal in, and colonization of the
downstream portions of the major North American drainage
systems in which it has become established (McMahon, 1982,
1983). The great capacity of C. fluminea for dispersal along
with its reduced age and size at maturity, high growth rates,
elevated fecundity, short generation times, abbreviated life
cycles and hermaphroditic reproductive schesis make this
species highly adapted for reproduction and survival in
disturbed, highly variable, lotic freshwater habitats, particular-
ly those subjected to human interference (McMahon, 1982,
1983). These characteristics, in large measure, also account
for the nature of C. fluminea as a pest species in North
America.

ACKNOWLEDGEMENTS

The authors wish to thank David R. Bible, Juan A. Ibarra,
Ralph Williams, Joe Gilly and Wesley Truitt for field assistance with
the collections. Colleen C. Bronstad provided technical assistance
in laboratory. Special appreciation is extended to Mark Spiegal and
William Hoerster of the Texas Electric Service Company who assisted
with studies within the Handley Power Station and provided daily
discharge volume and temperature data. We also wish to express
our deep appreciation to Jim Schmulen, then Environmental Scien-
tist of the Texas Electric Service Company, for his support and ad-
vice during the course of this study. This research was supported
by a grant to R. F. McMahon from the Texas Electric Service
Company.

LITERATURE CITED

Aldridge, D. W. and R. F. McMahon. 1978. Growth fecundity, and
bioenergetics in a natural population of the Asiatic freshwater
clam, Corblcula manilensis Philippi , from North Central Texas.
Journal of Molluscan Studies 44: 49-70.

Bird, J. L. 1976. Zooplankton densities and dispersion in a shallow
water reservoir receiving thermal effluent. Master’s Thesis, The
University of Texas at Arlington. 69 pp.

Busch, R. L. 1974. Asiatic clams Corbicula manilensis Philippi as
biological filters in channel catfish, Ictalurus punctatus
(Rafinesque) cultures. Master’s Thesis, Auburn University. 84
PP-

Cherry, D. S., J. H. Rodgers, Jr., R. L. Graney and J. Cairns, Jr.
1 980. Dynamics and Control of the Asiatic Clam in the New
River, Virginia. Bulletin 123, Virginia Water Resources
Research Center, Virginia Polytechnic Institiute and State
University, Blacksburg, Virginia. 72pp.

Dowell, C. L. and S. D. Breeding. 1966. Dams and reservoirs in
Texas. Texas Water Development Board 48: 1-267.

Dreier, H. 1977. Study of Corbicula manilensis at Lake Sangchris.
In: The Annual Report for Fiscal Year 1976. Lake Sangchris
Project, pp. 7.1-7.52. Illinois Natural History Survey, Urbana,
Illinois.

Dreier, H. and J. A. Tranquilli. 1981. Reproduction, growth, distribu-
tion and abundance of Corbicula in an Illinois cooling lake.
Illinois Natural Histroy Survey Bulletin 32: 378-393.

Eckbald, J. W. 1975. The Asian clam Corbicula in the Upper
Mississippi River. Nautilus 89: 4.

Esch. G. W. and R. W. McFarlane (eds.). 1976. Thermal Ecology II.
U.S. Energy Research and Development Administration,
CONF-750425, National Technical Information Service, U.S.
Department of Commerce, Springfield, Virginia. 404pp.

Gibbons, J. W. and R. R. Sharitz (eds.). 1974. Thermal Ecology. U.S.
Atomic Energy Commission, CONF-730505, National
Technical Information Service, U.S. Department of Commerce,
Springfield, Virginia. 670 pp.

Goss, L. B. and C. Cain, Jr. 1977. Power plant condenser and ser-
vice water system fouling by Corbicula, the Asiatic clam. In:
L. D. Jensen, ed., Biofouling Control Procedures, pp. 11-17.
Marcel Dekker, Inc., New York.

Goss, L. B., J. M. Jackson, H. B. Flora, B. G. Isom, C. Gooch, S.
A. Murray, C. G. Burton and W. S. Bain. 1979. Control studies
on Corbicula for stream-electric generating plants, In: J. C.
Britton, ed., Proceedings, First International Corbicula Sym-
posium, pp. 139-151. Texas Christian University Research
Foundation, Fort Worth, Texas.

Greer, D. E. and C. D. Ziebell. 1972. Biological removal of phosphates
from water. Journal of the Water Pollution Control Federation
12: 2342-2348.

Haines, K. C. 1979. The use of Corbicula as a clarifiying agent in
experimental tertiary sewage treatment process on St. Croix,
U.S. Virgin Islands. In: J.C. Britton, ed., Proceedings, First Inter-
national Corbicula Symposium, pp. 166-175. Texas Christian
University Research Foundation, Fort Worth, Texas.

Hillis, D. M. and J. C. Patton. 1982. Morphological and electrophoretic
evidence for two species of Corbicula (Bivalvia: Corbiculidae)
in North Central America. American Midland Naturalist 108:
74-80.

Isom, B. G., C. Gooch, L. T. Neill and J. Moses. 1978. Acute ther-
mal effects on Asiatic clams ( Corbicula manilensis Philippi).
Report No. I-WQ-78-12, Tennessee Valley Authority, Division
of Environmental Planning, Special Projects and Research
Program, Water Quality and Ecology Branch, Mussel Shoals,
Alabama. 19 pp.

Kraemer, L. R. 1979. Corbicula (Bivalvia: Sphaeriacea) vs. indigenous
mussels (Bivalvia: Unionacea) in U.S. Rivers: A hard case for
interspecific competition? American Zoologist 19: 1085-1096.

Mattice, J. S. 1979. interactions of Corbicula sp. with power plants.
In: J. C. Britton, ed., Proceedings, First International Corbicula
Symposium, pp. 1 19-138. Texas Christian University Research
Foundation, Fort Worth, Texas.

Mattice, J. S. and L. L. Dye. 1976. Thermal tolerance of adult Asiatic

MCMAHON AND WILLIAMS: CORBICULA IN THERMAL EFFLUENTS

239

clam. In: G. W. Esch and R. W. McFarlane, eds., Thermal
Ecology II, pp. 130-135. U.S. Energy Research and Develop-
ment Administration, CONF-750425, National Technical In-
formation Service, U.S. Department of Commerce, Springfield,
Virginia.

Mattice, J. S., R. 8. McLean and M. J. Burch. 1982. Evaluation of
Short-term Exposure to Heated Water and Chlorine for the Con-
trol of the Asiatic clam. Pub. No. 1748, Oak Ridge National
Laboratory, Environmental Sciences Division, National
Technical Information Service, U.S. Department of Commerce,
Springfield Virginia. 33 pp.

McMahon, R. F. 1975. Effects of artificially elevated water
temperatures on the growth, reproduction and life cycle of a
natural population of Physa virgata Gould. Ecology 56:
1167-1175.

McMahon, R. F. 1976a. Effluent-induced interpopulation variation
in the thermal tolerance of Physa virgata Gould. Comparative
Biochemistry and Physiology 55A: 23-28.

McMahon, R. F. 1976b. Growth, reproduction and life cycle in six
Texan populations of two species of freshwater limpets.
American Midland Naturalist 95: 174-185.

McMahon, R. F. 1977. Shell size-frequency distributions of Corbicula
manilensis Philippi in a clam-fouled stream condenser. Nautilus
91: 54-59.

McMahon, R. F. 1979. Response to temperature and hypoxia in the
oxyen consumption of the introduced Asiatic freshwater clam
Corbicula fluminea (Muller). Comparative Biochemistry and
Physiology 63A: 383-388.

McMahon, R. F. 1982. The occurrence and spread of the introduc-
ed Asiatic freshwater clam, Corbicula fluminea (Muller), in
North America: 1924-1982. Nautilus 96: 134-141.

McMahon, R. F. 1983. Ecology of an invasion pest bivalve, Corbicula.
In: W. D. Russell-Hunter, ed., The Moll u sea, Vol. 6, Ecology,
pp. 505-561. Academic Press, San Diego.

McMahon, R. F. and C. J. Williams. 1986. A reassessment of growth
rate, life-span, life-cycle, and population dynamics in a natural
population and field caged individuals of Corbicula fluminea
(Muller) (Bivalvia: Corbiculacea). In: J. C. Britton, ed. Pro-
ceedings, Second International Corbicula Symposium,
American Malacological Bulletin Special Edition No. 2:
151-166.

Mudkhede, L. M. and R. Nagabhushanam. 1977. Heat tolerance of
the freshwater bivalve clam, Corbicula reqularis. Marathwada
University Journal of Science 16: 151-154.

Pool, A. G. and L. J. Tilly. 1977. A Model to Determine Growth Rate
of Corbicula. CONF-771043-1 , U.S. Energy Research and
Development Administration, National Technical Information
Service, U.S. Department of Commerce, Springfield, Virginia.

11 pp.

Prezant, R. S. and K. Chalermwat. 1984, Flotation of the bivalve Cor-
bicula fluminea as a means of dispersal. Science 225:
1491-1493.

Rodgers, J. H., Jr., D. S. Cherry, J. R. Clark, K. L. Dickson and J.
Cairns, Jr., 1977. The invasion of the Asiatic clam, Corbicula
manilensis in the New River, Virginia. Nautilus 91 : 43-46.

Rodgers, J. H., Jr., D. S. Cherry, K. L. Dickson and J. Cairns, Jr.
1979. Elemental accumulation of Corbicula fluminea in the New
River at Glen Lyn, Virginia. In: J. C. Britton, ed., Proceedings,
First International Corbicula Symposium, pp. 99-110. Texas
Christian University Research Foundation, Fort Worth, Texas.

Sickel, J. B. 1979. Population dynamics of Corbicula in the Altamaha
River, Georgia. In: J. C. Britton, ed., Proceedings, First Inter-
national Corbicula Symposium, pp. 69-80. Texas Christian
University Research Foundation, Fort Worth, Texas.

Sinclair, R. M. 1964. Clam pests in Tennessee water supplies. Journal
of the American Water Works Association 56: 592-599.

Sinclair, R. M. and B. G. Isom. 1963. Further Studies on the Introduced
Asiatic Clam (Corbicula) in Tennessee. Tennessee Stream
Pollution Control Board, Tennessee Department of Public
Health, Nashville, Tennessee. 76 pp.

Thomas, N. A. and K. M. MacKenthum. 1964. Asiatic clam infesta-
tion at Charleston, West Virginia. Nautilus 78: 28-29.

Tommey, W. H. 1976. The fishes of Blundell Reservoir, Monticello,
Texas. Master's Thesis. The University of Texas at Arlington.
58 pp.

Williams, C. J., and R. F. McMahon. 1986. Power station entrain-
ment of Corbicula fluminea (Muller) in relation to population
dynamics, reproduction cycle, and biotic and abiotic variables.
In: J. C. Britton, ed., Proceedings, Second International Cor-
bicula Symposium, American Malacological Bulletin Special
Edition No. 2: 99-111.

Electrophoretic variation in North American Corbicula.

MICHAEL J. MCLEOD 125

A method for evaluating the sublethal impact of stress employing Corbicula fluminea.

CHRISTOPHER FOE and ALLEN KNIGHT 133

A thermal energy budget for juvenile Corbicula fluminea.

CHRISTOPHER FOE and ALLEN KNIGHT 143

A reassessment of growth rate, life span, life cycles and population dynamics in a natural
population and field caged individuals of Corbicula fluminea (Muller)

(Bivalvia: Corbiculacea). ROBERT F. MCMAHON and CAROL J. WILLIAMS 151

Aspects of growth of Corbicula fluminea.

J. S. MATTICE and L. L. WRIGHT 167

Corbicula in public recreation waters of Texas: habitat spectrum and clam-human interactions.

RAYMOND W. NECK 179

The Asiatic clam in Lake Erie.

JENNIFER SCOTT-WASILK, JEFFREY S. LIETZOW, GARY G. DOWNING and

KELLY L. (CLAYTON) NASH 185

Biological basis of behavior in Corbicula fluminea, I. Functional morphology of some trophic activities.

LOUISE RUSSERT KRAEMER 187

Biological basis of behavior in Corbicula fluminea, II. Functional morphology of reproduction and
development and review of evidence for self-fertilization. LOUISE RUSSERT KRAEMER,
CHARLES SWANSON, MARVIN GALLOWAY and ROBERT KRAEMER 193

Unsolved problems and promising approaches in the study of Corbicula.

K. ELAINE HOAGLAND 203

Biology of Corbicula in catfish rearing ponds.

JOSEPH K. BUTTNER 211

Assimilation of radiolabeled algae by Corbicula.

DIANE D. LAURITSEN 219

Adaptations of Pisidiidae (Heterodonta: Corbiculacea) to freshwater habitats.

GERALD L. MACKIE 223

Growth, life cycle, upper thermal limit and downstream colonization rates in a natural

population of the freshwater bivalve mollusc, Corbicula fluminea (Muller) receiving

thermal effluents. ROBERT F. MCMAHON and CAROL J. WILLIAMS 231

73
>

m m X w m

_ c/) “ y> \ z CO

NIAN INSTITUTION NOIlfUllSNI NVIN0SH1MS S3 1 dVd 3 11 LIBRARIES SMITHSONIAN INSTITUTION NOIlfUIJL!
z v.. co z ^ co «? co 2 .w co

^ 1 X 1 ^ 1 1 'W' | X 1 | >

iiws^ssidvaan libraries^smithsonian institution noiiiuiisni nvinoshiiws^sb i avd a nz li b rar i

CO Z CO X _ CO — CO

Q _ X^JiliSV^ Q

NIAN“lNSTITUTI0N^N0linillSNI~,NVIN0SHimSZS3 laVUai^LIBRARI ES^ SMITHSONIAN- INSTITUTION NOIlfUllS

Z t- , Z r- Z 2

O zCsoriTX _ vv CD X^vwTX O O

00
73
>

m m ^

CO _ CO — CO fz CO

dims ssiavaan libraries Smithsonian institution NouniiiSNi nvinoshiiws saiavaan librari

co z , co z co z co z ,

S ^ 2 < zrT^oX E <

> 2 NjJF > ^^225^ 2 " V?.v > ’ ^ 2 ' 2

NIAN INSTITUTION ^NOlinillSNI _NVINOSHimSC° S3 I d Vd a IT LI B RAR I ES^SMITHSONIAN INSTITUTION W NOimilJLi

— CO z \ 00 „ 5: CO — ^ CO

s> &> ^ W A - co ^ ^

q ^ O ' 'n^1 ^ X^Jvav^X o xj>\o ?va^X ~ O

LIIAIS ^S3 I dVda n^UBRAR I ES SMITHSONIAN INSTITUTION “NOIlfUllSNI "WlNOSHlIWS ^S3 I d Vd a H ~L I B RAR I
2 r ^ ^ ^ ^ 2 r 2

o rn - .■ m /SSnS&S. 2 X^Tx - O zCSmTx “

ID XiES*' -■ XOuii52X m

_ co £ co \ £ co

NIAN INSTITUTION NOIlfUllSNI NVIN0SH1IWS SBIdVdSII LIBRARIES SMITHSONIAN INSTITUTION NOIlfUll!
co z _ co z .... co z

.,v/ ' z

'-/ a Jw 0

4W x

__ _ __ Wm- F-

^2 ^ 5 Vo/vdc^ 5 2 > Nj/pjiiVSX 2 ^ >

iiiais s3iavaan_L,BRAR i es Smithsonian __iNSTiTUTi0NwN0iinniSNi_NviN0SHiiws S3iavdan_L|BRARi

51 < >/

oa 5; X'oTTT'X m ’''MW - X>oTfr^X “ ^ cn 5

NIAN- INSTITUTION N0linillSNI_NVIN0SHimS~S3 IdVdail LIBRARI ES^SMITHSOmAN- INSTITUTION _ NOlinillJ

^ 2 | | 5 /^r^\ I #Sk ” i /#^

- h 'ISt^A^y ^

m ^ rn Z m ^ m

HUMS S3ldVaan~ LIBRARIES SMITHS0NIAN~ INSTITUTION^ NOIlfllliSNI ~ NVINOSHIIWS S3ldVdail LIBRARI

CO Z CO z CO Z CO Z ,

< ^ S . s S AidSfcy . <

^ cn

=° /X

Yf- ' ”

CD
70
>

r£/ -

m x^yd^ y,

C o £ co fE ^ —

vinoshiiws S3iavuan libraries Smithsonian institution noiidiiisni nvinoshiiins saiavaan lib

co 2: .... 00 z _ oo z — _ z

dITHSONIAN INSTITUTION NOIlfUllSNI NVIN0SH1IWS S3ldVdail LIBRARIES SMITHSONIAN INSTITUTION NOI.

to — . co ^ <n 5 w

Ld /T^uX Z v ill 77. U XaH50/v7x lu

O xty. pc>z __ x"colsj^ q >\ ~ x^ijgs^x O X£X O

^inoshiiinszS3 lavaan^UBRARi Es:zSMiTHS0NiAN"JiNSTiTUTi0N^N0iinuiSNi-JNViN0SHiiws saiavaan LIB

Z r~ Z r~ z r~ , z r~

o /^jvnr^x “ ZiivtoJs O .. _ XodTJoX °

i I

m x^os*^ m xiftos^jx- ^ XjVAs>vz m vw ~ \£vama^ m

co — £ co £ to £ to

VIITHSONIAN INSTITUTION NOIlfUllSNI NVIN0SH1IWS S3ldVdail LIBRARIES SMITHSONIAN INSTITUTION NOI

co z oo z , co z co

2 . s sMr.. < >* s <

± <"% x'vv ^ ^

_ >• s ^ | ^ !' '

VINOSHimS^Saiavaan^LIBRARIES^SMITHSONIAN INSTITUTION W NOIlfUllSNI NVINOSHimS^ S 3 I d V d 3 1 1 LIB
to — m — co — >co

O X^vosvvjjz _ o

^ _J Z

VIITHSONIAN INSTITUTION NOIlfUllSNI NVIN0SH1MS S3ldVdail LIBRARIES SMITHSONIAN INSTITUTION NOI
- - -y r- Z r- 2

m x^yd^ ^ \ 2 N^TTlsbX m rn ^

viNosHims saiavaan” librari es^smithsonian- institution NoiinuiSNrNviNOSHims s3iavaan~LiB

CO Z c,. CO 2-" ' CO Z ^ CO z

WSovTx 2 . < AS^\ .. 2 Z < Zav'SovX 2 <

VIITHSONIAN INSTITUTION ^NOlinillSNI^NVINOSHlIlMS^ S3 I d Vd a 11^ LI B RAR I ES ^SMITHSON IAN INSTITUTION NOI
5 £ y-c^x z ,.v "» 5 5

o xc'ypv^ _ NiUiwz Q N\ “ XguTIstjX O xg*. ocz Z 'W" O

VIN0SH1IINS S3 I dVda ll^LIBRAR I ES*2 SMITHSONIAN-* INSTITUTION 2 NOIlfllllSNI^NVINOSHlIWS^Sa I dVda 11 LIE

Z r- Z r] z r~ , Z i-

O /in iCX O O y'rTATirTX w O

m x^vosy'^z ^ m x^ostj^z ^ xjvasi^z rn )^y ^ pi

MITHSONIAN INSTITUTION NOIlfUllSNI ~ NVIN0SH1IWS S3 I d Vd 3 1 1 ~ L I B R A R I E S SMITHSONIAN_INSTITUTION NOI

Z 1 CO z CO z , to z » CO

s s ^ s Xssnn^X <

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "Proceedings of the Second International Corbicula Symposium"

See other formats