SRQ-NERp·3

The Freshwater Bivalve Mollusca(Unionidae, Sphaeriidae, Corbiculidae)

of theSavannah River Plant, South Carolina

byJoseph C. Britton

andSamuel L. H. Fuller

A Publication of the Savannah River Plant

National Environmental Research Park Program

United States Department of Energy

...---------NOTICE---------,

This report was prepared as an account of work sponsored by the UnitedStates Government. Neither the United States nor the United States Depart­mentof Energy.nor any of their contractors, subcontractors,or their employ­ees, makes any warranty. express or implied or assumes any legal liabilityorresponsibilityfor the accuracy, completenessor usefulnessofanyinformation,apparatus, product or process disclosed, or represents that its use would notinfringe privately owned rights.

A PUBLICATION OF DOE'S SAVANNAH RIVER PLANT NATIONAL ENVIRONMENT RESEARCH PARK

NOVEMBER 1980Copies may be obtained from

Savannah River EcologyLaboratory

THE FRESHWATER BIVALVE MOLLUSCA(UNIONIDAE, SPHAERIIDAE, CORBICULIDAEj

OF THE SAVANNAH RIVER PLANT, SOUTH CAROLINA

by

JOSEPH C. BRITTONDepartment of Biology

Texas Christian UniversityFort Worth, Texas 76129

andSAMUEL L. H. FULLER

Academy of Natural Sciences at PhiladelphiaPhiladelphia, Pennsylvania

Prepared Under the Auspicesof

The Savannah River Ecology Laboratoryand

Edited by Michael H. Smith and I. Lehr Brisbin, Jr.

1979

SRO-NERP-3

TABLE OF CONTENTSPage

INTRODUCTION 1STUDY AREA " 1LIST OF BIVALVE MOLLUSKS AT THE SAVANNAH RIVER PLANT............................................ 1ECOLOGICAL CONSIDERATIONS " 2COLLECTING TECHNIQUES 3PRESERVATION OF BIVALVE MOLLUSKS FOR IDENTIFiCATION............................................ 4KEY TO SPECIES OF BIVALVE MOLLUSKS OF THE SAVANNAH RIVER PLANT............................... 5TAXONOMIC CRITERIA " 7DESCRIPTIONS OF THE BiVALViA........................................................................ 7

Family Unionidae 7Subfamily Unioninae 7

Genus Elliptio " 7Elliptio complanata " 7Elliptio icterina " 9Elliptio lanceo/ala 10Elliptio congaraea " 11Elliplio Iratema 12

Genus Uniomerus 12Uniomerus tetralasmus 13

Subfamily Anodontinae 13Genus A/asmidonla " 13

A/asmidonta triangu/ata " 13Genus Anodonta 14

Anodonta imbecillis 14Anodonta couperiana 15Anodonta cataracla 15

Genus Slrophitus . . .. . . . .. .. . . .. . . . . .. .. . . . . . .. . . .. . .. . .. . .. . . . . .. . . . . . .. . . .. . . .. . . . .. . . . . . 15Strophitus undu/atus 16

Subfamily Lampsilinae " 16Genus Carunculina " 16

Carunculina pulla 16Genus Villosa " 16

Villosa aeiumbts 16Villosa vibex 17

Genus Lampsilis 17Lampsilis cariosa 17Lampsilis radiata sp/endida 18

Family Corbiculidae 19Genus Corbicula 19

Corbicu/a fluminea ,..................................... 19Family Sphaeriidae 20

ACKNOWLEDGMENTS 21LITERATURE CiTED 22GLOSSARY 24ILLUSTRATIONS 26PLATE I 28PLATE II 30MAP LEGENDS 32

INTRODUCTION

This is the third in a series of guides to the fauna ofthe National Environmental Research Park of theSavannah River Plant. As with the former work, on thedecapod crustaceans (Hobbs et al., 1976), the mainpurpose of the present paper is to aid investigators withthe identification of a taxonomic group which is oftendifficult for inexperienced investigators to identify - inthis case, the bivalve mollusks that can be expected tobe found at the Savannah River Plant (SRP).

Within the last ten years, two important studies havebeen made of the unionid mussels of the southernAtlantic slope drainage. The first (Johnson, 1970) ismonographic in scope and is an indispensible guide tothe unionid shells from southern Georgia to Virginia andMaryland. The second paper (Fuller, 1971) is morerestricted in scope, dealing primarily with mussels of theSavannah River system, but is important because of theemphasis upon the soft-part morphology of bivalves. Thepresent paper is in no way intended to supplant either ofthese important studies. The reader is encouraged tohave both papers readily available before any extensiveresearch on the Bivalvia of this region is attempted. Youwill find here, however, a handy guide and reference tothe more common and abundant bivalves in the watersof the SRP. A dichotomous taxonomic key is providedto common forms and to unreported species whosegeographic distributions include nearby localities. Everyspecies collected during field reconnaissance in thesummer of 1977 is described, and a discussion of someaspects of its ecology, life history, larval hosts, andother pertinent information is provided.

STUDY AREA

The Savannah River Plant (SRP) occupiesapproximately 200,000 acres in southern South Carolina,occupying portions of Aiken, Barnwell and Allendalecounties, near the cities of Augusta, Georgia, and Aiken,South Carolina. The SRP is located just below the FallLine in the Atlantic Slope Coastal Plain. Five minordrainage systems flow from or through the SRP to theSavannah River, which courses for 22 miles along thesouthwestern border of the plant. A variety of terrestrialand aquatic habitats occur within the SRP boundaries.The former include lowland hardwood forests andcypress-gum swamps, turkey oak and longleaf pinesandhills, upland oak-hickory forests, and large old-fieldareas. The aquatic habitats include the Savannah Riverand adjacent swamplands, several reservoirs includingthe 2,800-acre nuclear reactor cooling reservoir, ParPond, various natural and thermally affected (by severalnuclear production reactors) streams, and a number ofCarolina bays, beaver-dam ponds, and abandoned farmponds. Designated as the first National EnvironmentalResearch Park, the SRP is a unique protected outdoor

laboratory where research can be conducted in bothundisturbed and stressed ecosystems.

During May, June, and July 1977, one of us (JCB)spent several weeks surveying the various aquatichabitats on or near the SRP for bivalve mollusks. A totalof 46 specific localities was visited, of which 26 provedto have at least one species of bivalve. Additionalmaterial was provided (by SLHP) from extensivecollections made during numerous stream surveys forthe Academy of Natural Sciences at Philadelphia(ANSP).

LIST OF BIVALVE MOLLUSKS AT THE SRP

There are several suprageneric classifications of theUnionacea, including those by Ortmann (1910, 1911,1912, 1916, and 1919), Frierson (1927), Modell (1942,1949, 1964), Haas (l969a, 1969b) and Heard and Guckert(1971). Following the similar treatments by Johnson(1970) and Fuller (1971), we employ the Ortmann systemhere. Three families, 14 genera, and 26 species arerecognized with geographic ranges which include or arenear the SRP. They are included in the following list.Twenty-three of these species were collected during thesummer of 1977 andlor during ANSPISRP surveillancesince 1955. The remaining three species may occurwithin the boundary of the SRP andlor in the adjacentportions of the Savannah River. They are indicated byan asterisk in the list below and are included in the key.Otherwise, these three species are not given additionaltreatment herein.

SUPERFAMILY UNIONACEA Thiele 1934FAMILY UNIONIDAE Fleming 1828

SUBFAMILY UNIONINAE Swainson 1840GENUS Fusconaia Ortmann 1912

"Fusconaia masoni (Conrad1834)'

GENUS Uniomerus Conrad 1853Uniomerus tetralasmus (Say1831)

GENUS Elliptio Rafinesque 1819Elliptio congaraea (Lea 1831)E. complanata (Lightfoot 1786)E. fraterna (Lea, 1852)E. lanceolata (Lea 1828)E. icterina (Conrad, 1834)

SUBFAMILY ANODONTINAE Swainson1840

GENUS Lasmigona Rafinesque 1831*Lasmigona (Platynaias)subviridis (Conrad 1835)

GENUS Alasmidonta Say 1818Alasmidonta (Alasmidonta)triangulata (Lea 1858)*A. (Decurambis) varicosa(Lamarck 1819)

~~~~--~- --_.- ---_ ..

GENUS Anodonta (Lamarck 1799)Anodonta (Utterbackia)couperiana Lea 1840A. (U.) imbecillis Say 1829A. (Pyganodon) cataracta Say1817

GENUS Strophitus Rafinesque 1820Strophitus undulatus (Say 1817)

SUBFAMILY LAMPSILINAE von Ihering1901

GENUS Carunculina Baker 1898Carunculina pulla (Conrad 1838)

GENUS Vil/osa Frierson 1927Vil/osa vibex (Conrad 1834)V. delumbis (Conrad 1834)

GENUS Lampsilis Rafinesque 1820Lampsilis radiata splendida (Lea1838)L. cariosa (Say 1817)

SUPERFAMILY CORBICULACEA Gray 1847FAMILY CORBICULIDAE Gray 1847

GENUS Corbicula von Muhlfeld 1811Corbicula (Corbicula) fiuminea(Miiller 1774)

FAMILY PISIDIIDAE Gray 1857GENUS Pisidium Pfeiffer 1821

Pisidium americum (Muller 1774)Pisidium casertanum (Poli 1791)Pisidium compressum Prime 1851Pisidium dubium (Say 1834)

GENUS MusculiumMusculium transversum (Say1829)

GENUS Eupera Bourguignat 1854Eupera cubensis (Prime 1865)

"This is Pleurobema mason; teste Johnson (1970). See Fuller (1973) forextensive discussion.

ECOLOGICAL CONSIDERATIONS

Freshwater bivalve mollusks, especially the unionids ,tend to be most abundant in flowing streams and rivers.The numbers of individuals and the species diversityusually diminish in the still waters of lakes or basins(e.g., Carolina Bays). Regardless of habitat, thedistribution of bivalves is often patchy, clumped, orotherwise irregular. Large numbers of clams may becollected in one area, whereas in a similar environmentonly a few meters away, few or no specimens may befound. The irregular distribution of unionids may berelated to the distribution patterns of freshwater fishwhich are employed by larval clams (glochidia) as hostsfor dispersal (see later discussion). Freshwater mussels

2

are often very abundant in substrates of shallow poolsand protected coves or among reeds where various fishspecies may seek shelter.

There are some conditions where few if any clams ormussels should be expected. Ephemeral ponds orstreams rarely harbor bivalves. Even the largest Carolinabays, which are never completely dry, may be devoid ofclams. Permanent waters with poor circulation or flowrarely produce large clam populations. An abundance ofindividuals of certain species (e.g., Anodonta imbecillisimay occur in Par Pond and other reservoirs at SRP butwe have never found more than four or five bivalvespecies in these lakes. As sessile filter-feeders, mostfreshwater bivalves fare best when nutrient-laden watersflow continuously over them.

SRP streams are not especially rich in suspendedparticulate nutrients (Langley and Marter, 1973), hencethe populations of bivalves in these waters are limited.Clean sand bars in the fast-flowing streams may harbor afew specimens, but most bivalves of SRP streams occuralong banks or in stream eddies where nutrients tend toaccumulate. The Sphaeriidae or fingernail clams are anexception. They can be found in the same localities asthe stream Unionidae , but they are also particularlyabundant in gravel substrates of stream riffies.

The river environment provides a greater nutrient loadthan that of smaller streams; as might be expected, thevariety of bivalve populations is greater there. If onewishes to collect the rarer species of mussels in thevicinity of the SRP, localities in or near the SavannahRiver should be searched first. Mussels and clams aremost abundant on semi-protected shoals or bars of theriver where there is both adequate flow and protectionagainst the catastrophic effects of flooding. They alsooccur in deeper parts of the river, but collecting in theseareas is considerably more difficult.

The recent introduction of Corbicula fiuminea into theSavannah River (or any waterway in this country) raisessome concern about the effects that this expandingpopulation may have on native bivalves. C. fluminea canreach exceptionally high densities just a few years afterbeing introduced into an area (see below), Eventually.shell densities may become so great as to reduce thehabitable space for other clams (e.g., Eng, 1977).Alternately, nutrient and/or chemical resources may bedepleted or otherwise made unavailable to nativespecies. The impact of C. fluminea on other clams is yetto be assessed, but one should be aware that a possiblechange in population densities may occur as a result ofthis invading bivalve.

In addition to nutrients, there are other aspects ofwater chemistry that may also influence the distributionof bivalves. Surface waters of southern South Carolinatypically are slightly acidic and have low conductivity.Of the several ionic components which influenceconductivity, calcium is one of the most important tomollusks. Calcium is required by bivalves for shell

formation. If it is in limited supply, the growth and/orabundance of bivalve species may be similarly limited.This becomes especially critical in slightly acidic waters.If calcium carbonate shells are exposed directly toacidity, they will undergo dissolution. Mollusks whichlive in acidic waters usually develop thick, organicperiostracallayers to protect the underlying, inorganiccarbonate from chemical breakdown-but mechanicalabrasion can wear away the protective periostracum,thereby exposing the calcium carbonate shell to chemicaldissolution. Thus, there is a series of interactions thatensure a turnover of calcium within the aquatic system.In flowing waters, the net effect will be a loss of calciumto the system if there are no local sites of replenishment.In the vicinity of the SRP, several aquifers flow fromcalcium-rich groundwater sources to the surface andprovide constant replenishment (Langley and Marter,1973). Thus, although the calcium content of surfacewaters at the SRP seems low (generally less than 1.5rng/I), this element is in constant supply fromgroundwater sources and is supplemented to a smalldegree by the dissolution of older and/or dead shells.The loss of shell material as a result of dissolution maybe more critical to the individual mollusk. If a clamcannot replace shell material faster than it is beingremoved by mechanical abrasion or by chemicaldissolution, eventually the shell will become perforated.One occasionally finds living specimens with a perforateumbo exposing the underlying mantle tissue to theexternal environment. More often, numerous unhingeddead valves with umbonal perforations are found,attesting to the frequency with which this processoccurs. Although there are no relevant quantitative data,one wonders how important this process might be inlimiting the bivalve populations and especially individualspecies in the SRP area.

All of the bivalves at the SRP are filter-feedingorganisms. Waterborne food enters the mantle cavitythrough a posterior branchial aperture and circulatesamong the gill dernibranchs, on whose surfacesparticulate material is trapped. The water exitsposteriorly through an anal aperture located dorsal to thebranchial aperture. The branchial and anal apertures ofCorbicula and some of the Sphaeriidae are siphonal. Inmost bivalves, the branchial aperture is papillate; thepapillae are usually employed as a screen or filter forvery large particles. The anal aperture mayor may notbe papillate, and the presence or absence of papillaehere is a useful criterion for identification.

Food material and other particulate matter trapped bycilia of the gills are passed either toward the mouth byspecific ciliary tracts or along rejection paths forelimination from the mantle cavity. Rejected materialthat does not pass into the digestive tract is eliminatedas "pseudofeces" through the branchial aperture andbetween the mantle margins anterior of the incurrentaperture. (Bivalves held in aquaria frequently will

3

eliminate considerable quantities of pseudofecalmaterial). Additional details of the mechanics andphsysiology of the filter-feeding process can be foundelsewhere (e.g., Owen, 1966).

The similarities of habitat, feeding behavior, andnumerous other traits are presumably responsible for ahigh degree of convergent evolution with respect to shellmorphology within the Unionacea, Sphaeriidae, andCorbiculacea. There are numerous examples among theUnionacea of distinctly different soft-parts enclosedwithin superficially similar shells. Such is the case withmany of the SRP unionaceans. At first glance, the shellsof several SRP mussel species will appear almostidentical. This makes the task of identification especiallydifficult for those working with the group for the firsttime. Johnson (1970) presented an extensive discussionof zoogeographical and phylogenetic relationships amongAtlantic slope drainage unionids.

The life cycle of freshwater bivalves varies accordingto the taxonomic groups. All bivalves at the SRP broodlarvae in gill marsupia. The Sphaeriidae andCorbiculidae produce brood chambers within the innerdemibranchs of the gill. The former group restrictsmarsupia to the anterior portions of the demibranch, andthe latter employs the entire demibranch. Members ofboth families release the larvae to the environment,where they undergo metamorphosis into juvenile clams.Heard (1977) discusses in considerable detail thereproductive biology of the Sphaeriidae, and additionalremarks on the life cycle of Corbicula are presentedbelow.

The Unionacea may brood larvae in all fourdemibranchs (e.g., Amblema, not found in this area), inonly the outer demibranch (e.g. Elliptio), or in onlyspecific portions of the outer demibranchs (e.g.,Lamps/lis). Upon release from the brood chamber, theunionacean larva (glochidium) usually must undergo anobligatory parasitic encystment on gills or fins offreshwater fishes if development is to proceed. See the"Life History" section of Elliptio camplanata foradditional information on the reproductive behavior andthe larvae of freshwater mussels.

COLLECTING TECHNIQUES

Clams and mussels are most easily collected duringthe drier months, when the water levels and flow ratesof streams or rivers are low. Shoals or bars in largerivers become treacherous or deeply submerged duringhigh water periods, this makes collecting difficult andeven dangerous. In the drier seasons, the shoals oftenbecome partially exposed, easily worked, and oftenhighly productive of mussels and clams. A particularlygood river locality for mussels is in the lee of a sharpbend, i.e., along the shore opposite the cutting bank.

Mussels in smaller streams (such as the many at theSRP) may be found on mid-stream shoals, but more

frequently they lie partially buried in sand, mud orgravel along the stream bank. Sandy, sandy gravel, orsandy mud bottoms seem to be the most productivesubstrates, but some mussels are characteristically foundin abundance among very fine, organically richsediments trapped at the bases of tree roots or in calmareas of the stream. Rapidly shifting, transitory sandbars are usually unsuitable habitats, but one species inthe Altamaha River basin of Georgia iCanthyria spinosa(Lea)) has spines on each valve, presumably to facilitateshell anchorage in this kind of habitat.

The Sphaeriidae, Corbicula, and some Unionidaecommonly occur in the gravel substrates of stream rifflezones. This is a particularly good locality to search forfingernail clams.

If collecting is done during a period of low water andhigh air temperature, a word of caution is in order.Specimens collected for laboratory experiments shouldthen be taken only if they were covered by at leastseveral inches of water. Some clams (Corbiculafiumlnea, in particular) may suffer latent mortalityseveral days or even weeks after air exposure or heatstress in temperatures above 37°C (Coldiron, 1975). Onthe other hand, if specimens are required for museumcollections, locality records, or gross morphologicalstudies, the more easily obtained exposed orsemi-exposed individuals are preferred. The exposurehas likely severely weakened many of these specimens,and collecting them instead of reproductive stocks fromdeeper waters will minimize depletion of the healthybivalve populations.

There is no "best method" to collect bivalvemollusks. Many standard quantitative techniques (e.g.,grab samplers) are ineffective with some bivalve groups,especially the Unionidae. Grab samplers will almostnever collect large mussels in proportion to their actualabundance in a locality. These devices may be useful,however, for population estimates of Corbicula or theSphaeriidae. The most reliable techniques for samplingbivalves are hand methods, despite numerous attemptsby a variety of workers to develop other samplingprocedures. There are several kinds of hand collectingtechniques, all of which should be attempted at a givenlocality.

First and foremost, the most reliable of all methods tocollect larger unionids remains hand-sifting andsearching the substrate. According to the preference ofthe collector, gloves may be worn to minimize abrasionsand minor lacerations. One may "scan" a clear streamfor larger specimens by walking along a bank or in theshallows and looking for the characteristic wedge-shapedposterior region projecting from the substrate in shallowwater. The collector should always walk upstream,thereby ensuring that his view will be in clear,undisturbed water. In more turbid streams, or in lowflow conditions, the scan technique may be impossible.The collector then must rely upon a blind search and

4

must use hands to sift along a sandy spit or among rootsunder a bank. A small rake may be useful in somelocalities but, if there is a large amount of rootedvegetation, it may become frequently tangled and proveineffective.

If one is searching for Sphaeriidae or Corbicula,screening or seiving the substrate often is an efficientmethod of obtaining numerous specimens. Note,however, that screening rarely expedites the search formost of the larger mussels. Dip nets have beensuccessfully employed at SRP for collecting bivalves,and they are especially well suited for obtainingfingernail clams and Corbicula. These clams almostalways lie buried in the substrate, so a dip net should beraked through the streambed and then washed of itssediment burden before the net is searched. Lookcarefully among the organic debris for smallerspecimens.

In shallow riffle zones, a Surber sampler or the PIBS(Portable Invertebrate Box Sampler) device can providereasonably accurate quantitative estimates of theSphaeriidae and/or Corbicula populations. Like otherquantitative samplers, however, these devices aregenerally ineffective for quantitative sampling ofunionids.

Most of the methods described above can be usedalong lake shores and in the shoals of larger rivers, aswell as in the smaller stream localities of the SRP. Fordeeper lake or river waters, a shell dredge (e.g., WildlifeSupply Co., No. 175) may be useful. The effectivenessof dredging is highly dependent upon the nature of thesubstrate. Heavy vegetative cover, such as thatfrequently occurring on SRP lake bottoms, or thenumerous snags or stumps that are common in theSavannah River, tend to reduce the effectiveness of theshell dredge as a bivalve collector. Nevertheless, duringJune, 1977, good results were obtained by dredging insome parts of Pond B and Par Pond in order to obtainAnodonta imbecillis and along deeper shoals in theSavannah River in order to obtain Corbicula fluminea.

Specimens have been collected by use of mask,snorkel, and fins in all localities sampled. The use ofmask and snorkel along shoals in the Savannah Riverwas especially productive. Diving with SCUBA gear wasattempted in Par Pond, with only fair results. The lowvisibility and heavy plant growth on the bottom inhibitedthe diver's collecting effectiveness considerably.

PRESERVATION OF BIVALVE MOLLUSKS FORIDENTIFICATION

Shell features of bivalve mollusks are often used asthe only criteria for identification. Indeed, it is far easierand inexpensive to prepare and maintain a collection ofdry shells than a "wet collection." Yet many Unionaceaand Sphaeriidae are difficult or impossible to identify atthe species level by using shell characters exclusively.

Hence, both "wet" and "dry" collections arerecommended, especially if the purpose of the collectionis to be for later reference. Techniques for both kinds ofcollections will be described.

Tissue preservation is achieved in 5-6% neutral orbuffered formalin. It is recommended that specimens be"pegged" open to facilitate penetration of thepreservative. This can be done by inserting smallwooden wedges between the valves of a naturally gapingclam before it has an opportunity to close. Even withpegging, the soft tissues contract strongly upon exposureto the preservative. To avoid this response, clams andmussels can be relaxed in a dilute solution containingNembutal" (sodium pentobarbital).

Place the specimens to be relaxed in a container justlarge enough to hold them without expanded soft partstouching with other specimens, but small enough tominimize fluid volumes. Add enough natural water to thecontainer barely to cover the bivalves. Then add ten totwenty drops of Nembutal (stock solution containing 50mg sodium pentobarbital per ml fluid carrier) to thewater. Do not disturb the specimens for at least onehour. Then touch the extended foot of a clam with aprobe. If it retracts quickly and/or if no specimens haveexposed soft parts, add to the water a quantity ofNembutal equal to the first dose. Repeat this process,reducing the quantity of Nembutal you add as theresponse becomes progressively lethargic. Eventually,there will be little or no retraction upon tactilestimulation. At this point gradually begin to introducesmall amounts of 5-6% formalin or 70% alcohol into thewater. We have found alcohol more effective at thisstage, as it acts as a further relaxing agent, particularly ifit is added slowly. Eventually, however, the specimensshould be transferred to 5-6% formalin solution forcomplete fixation.

A word of warning: The relaxation techniquedescribed above is a very tedious process and notalways successful. Furthermore, it seems to work moreeffectively in some genera than in others. Be preparedfor several repetitions, if necessary, before achievingsatisfactory results. Fuller (1971) has figured specimensof Lampsilis radiata splendida and L. r. siliquoidea(Lea) that were relaxed by the Nembutal technique andare fine examples of excellent relaxation andpreservation.

Specimens should remain in 5-60/", buffered formalinfor one to several weeks to ensure good penetration ofthe fixative. Eventually, however, they should betransferred to 70% ethanol (ethyl alcohol). Pigmentedareas will rapidly fade in alcohol; so, if notes are to bemade on pigmentation in tissues, this should be doneshortly after fixation or, better still, in life beforepreservation.

Sometimes it is desirable to separate soft parts fromshell material and to leave the shell dry and the softparts in alcohol. The preserved tissue can be removed

5

almost intact from the surrounding shell by carefuldissection. If this is done, identification should bemaintained between shell and soft parts. A fine wirewith an identifying label can be affixed to the muscularfoot tissue, and a similar identifying number can bewritten on the interior surfaces of both valves using apermanent ink.

If one intends to keep only shell material of somespecimens, there are easier methods to obtain cleanedshells. The quickest procedure is to place livingspecimens in boiling water. Within a few minutes, thevalves will gape noticeably. Remove the specimens fromthe boiling water and lift the soft parts away from theshells. Tissues should never be extracted forpreservation using this method, for there is considerableand progressive shrinkage of the soft parts duringboiling.

When freshwater shells are dried, there is usually aproblem with cracking, peeling, or warping ofperiostracal layers. Sometimes this can be retarded bylightly coating the valves with mineral oil. Thin shells.such as those of Anodonta, are also plagued by crackingor fragmentation of carbonate layers when dried. Eventhe application of mineral oil may not prevent this kindof damage.

"Abbott Laboratories, North Chicago, Illinois. This barbituate isavailable only by prescription, and/or federal (FDA) permit for Class Iproscribal substances.

KEY TO THE SPECIES OF BIVALVE MOLLUSKSOF THE SAVANNAH RIVER PLANT

Two excellent keys to the Unionacea of this regionhave already appeared. Johnson (1970), treating theentire Unionacean fauna of the southern Atlantic sloperegion, relied exclusively upon shell morphology in hiskey. Fuller (1971) employed soft-part criteria (especiallythe nature of the mantle margin) in his key of theunionaceans of the Savannah River system. The keyemployed in this paper is more restrictive than either ofthe two previous efforts: only. those species known orreported to occur in the vicinity of the SRP are included.Also, it is organized differently: some species "key out"in associations different from those in Johnson's orFuller's works. This was done purposely in an attemptto enhance the value of all of the keys. Johnson hasexcellent plates of comparative shell material and types.which should be consulted. Fuller has numerousanatomical notes, which are also valuable aids inidentification. There are some cases, however, in whichthe existing keys simply cannot be improved upon. Thereader should not be surprised to find a few coupletssimilar or identical in all three works. A glossary isincluded at the end of this report in order to provide aready reference to the more technical terms used in thiskey.

Some species not reported from the SRP are includedbelow. They are indicated by an asterisk. No treatmentof these species is given in the descriptive section thatfollows.

Ia Posterior portion of mantle fused to form one ortwo elongate siphons 2

Ib Posterior mantle unfused ventral to position ofanus; incurrent (branchial) and anal areas areunfused but differentiated apertures; mantlefusion may occur dorsal to position ofanus U nionacea 3

2a Shell oval or trigonal; umbo usually central;deeply incised broadly spaced concentric ringspresent over entire surface of valve; three cardi­nal teeth in each valve (one in each valve isclosely applied to dorsal surface of hingeplate) (Corbiculidae) Corbicula fiuminea

2b Shell oval or trigonal; umbo usually anterior orposterior to middle; valves may have very fineconcentric lines, but never deeply incised rings ~

two or fewer cardinal teeth in each valve ....Sphaeriidae (See separate section on page 20.)

3a Hinge plate completely edentulous; anal aper-ture without papillae .4

3b Pseudocardinal teeth, lateral teeth, or both pre­sent on hinge plate although sometimes vesti­gial; anal aperture with or without papillae ..6

4a Umbo not elevated above dorsal margin; shellusually less than 120 rnrn in length 5

4b Umbo elevated above dorsal margin; shell often120 mm or larger Anodonta cataracta

5a Ventral margin straight or slightly convex, sub­parallel with dorsal margin; length/height ratiousually greater than 2.0; periostracum withbroad green rays .Anodonta imbecillis

5b Ventral margin noticeably convex, divergentfrom ptane of dorsal margin; length/height ratiousually less than 2.0; periostracum with narrowgreen rays .Anodonta couperiana

6a Mantle margin at anal aperture lacking distinctpapillae (smooth or crenulate) 7

6b Mantle margin at anal aperture with distinct,well formed papillae 11

7a Lateral teeth present on hinge plate 87b Lateral teeth absent from hinge plate 98a Branchial papillae dendritic; periostracum with

satin-like texture, unrayed; pseudocardinal teethchunky, serrated; hinge plate lacking interdentalprojection in left valve Uniomerus

tetralasmus8b Branchial papillae simple (unbranched); per­

iostracum may be smooth, shiny, or slightlyroughened, but not with satin-like texture, oftenwith distinct rays; pseudocardinal teeth thin,lamellate; hinge plate with interdental projectionin left valve *Lasmigona subvirdis

6

9a

9b

lOa

lOb

Ila

II b

12a

12b

l3a

I3b

14a

14b15a

15b

16a

16b

17a

17b

Pseudocardinal teeth small, poorly formed, orvestigial; posterior ridge not sharply angular...................................... 10

A single, distinctly elevated pseudocardinaltooth in each valve; posterior ridge sharply an-gular Alasmidonta triangalataPseudocardinal teeth reduced to slight swellingson hinge plate anterior to umbo; posterior slopeusually without radial wrinkles ... .Strophitas

undulatusPseudocardinal teeth more distinctly formed,single in each valve; posterior slope with radialwrinkles "Alasmidonta varicosaMantle anterior to branchial aperture undiffer­entiated and smooth; marsupium formed alongtotal length of outer demibranchs, inner demi­branchs mayor may not form marsupium9 shelllacking conspicuous sexual dimorphism .... 12Mantle margin anterior to branchial aperturecarunculate, papillose, andlor formed into flapsor thickened pads; marsupium restricted toposterior half of outer demibranchs, neverformed within inner demibranchs; male andfemale shells usually sexually dimorphic ... 17Soft tissues usually pigmented vividly red ororange; all four demibranchs form marsupium;rare in most localities ...."Fusconaia masoniNot as above; marsupium restricted to outerdemibranchs 13Length/height ratio usually less than 1.8; shellsubtriangular or subquadrate; periostracumusually centrally smooth, shiny .Elliptio

congaraeaLength/height ratio usually greater than 1.8;shell lanceolate, rhomboidal, or trapezoidal;periostracum smooth or rough 14Sculpture including weak plications traversingdisk obliquely and weak subradial ridges onposterior slope .Elliptio fraternaNot as above 15Length/height ratio usually greater than 2.5;shell distinctly lanceolate . .Elliptic lanceolataLength/height ratio usually less than 2.5; shellrhomboidal or trapezoidal 16Shell distinctly trapezoidal; periostracum usu­ally quite rough, rarely shiny anywhere onvalve: egg masses in marsupium narrowly elon-gate, slightly broader ventrally Elliptio

complanataShell subrhomboidal; periostracum often withslight sheen medially, but without gloss on an­terior or posterior slopes; egg masses in mar-supium broadly oval Elliptio icterinaMantle margin anterior to branchial aperturecarunculate in female; shell rarely exceeding 35mm in length Carunculina pullaMantle margin anterior to branchial aperture

papillate, flap-like, or forming a thick pad, butnot carunculate; shell frequently exceeding 35mm in length 18

18a Mantle margin anterior to branchial aperturepapillate in female; shell rarely exceeding 70 mmin length 19

18b Mantle margin anterior to branchial apertureflap-like or formed into a thickened pad; shelloften exceeding 70 mm in length 20

19a Periostracum with wide green rays occasionallyinterrupted by growth rests; sexual dimorphismweakly expressed Vil/osa vibex

19b Periostracum with narrow green rays often in-terrupted by broad yellowish background areas;sexual dimorphism well expressed ....Villosa

delumbis20a Adult shell large, heavy, subovate (female) to

subtrigonal (male); periostracum smooth, yel­lowish, usually lacking rays; mantle anterior tobranchial aperture flap-like Lampsilis

cariosa20b Adult shell moderately large, moderately thick­

ened, subrhomboidal; periostracurn rough, some­what satin-like, brownish or reddish, usuallywith faint rays; mantle anterior to branchialaperture forming a very darkly pigmented,thickened pad, and an anterior, ribbon-like flap.............. .Lampsilis radiata splendida.

TAXONOMIC CRITERIA

Most of the terms employed in the descriptions beloware defined in the glossary at the end of this paper andare illustrated in figures following the glossary. In eachsystematic description, two meristic shell ratios aregiven: length/height and height/inflation (= breadth). Inorder to standardize these values to fascilitatecomparison, the dimensions of length, height andinflation were measured at consistent locations on allshells (see Figures 2 and 4). Vernier calipers were usedfor measurements. Length was taken as the maximumanterioposterior elongation of the shell. Height andinflation were measured at the position of the umbocrest, immediately posterior to the umbo origin (seeFigure 4). This does not always correspond withmaximum height or inflation as can be seen in thefigures. Hence, any future measurements must be takenexactly as indicated if comparisons with the presentvalues are desired.

DESCRIPTIONS OF THE BIVALVIA

SUPERFAMILY UNIONACEA Thiele 1934FAMILY UNIONIDAE Fleming 1828

DIAGNOSIS: Freshwater bivalves or clams withequivalve shells usually larger than I em in length; hinge

7

edentulate, or with either pseudocardinal or lateral teethor both; ligament opisthodetic and parvincular;periostracum usually conspicuous; a sheet of tissuealmost completely separates excurrent and incurrentmantle cavity; gills with distinct interlamellar septa thatparallel gill filaments and form watertubes, some or all ofwhich become brood chambers (marsupia) for ova andlarvae; larvae (glochidia) parasitic on vertebrates, usuallyfish.

SUBFAMILY UNIONINAE Swainson 1840

DIAGNOSIS: Marsupium occupying entire outerdemibranchs; demibranchs only moderately distendedwhen filled with ova and/or larvae; anal aperturepapillate although papillae may be very short;pseudocardinal and lateral teeth present; glochidia ovalto subovate , with or without hooks.

GENUS Elliptio Rafinesque 1819

DIAGNOSIS: Shell moderately to greatly elongated(length/height ratio) varying from 1.50 to at least 4.0);longitudinal axis generally straight. Beaks low,inequilateral, anterior, and prosogyrate. Ligamentprominently elevated. Periostracum well developed anddarkly colored in adults, often rayed in juveniles.Dentition consisting of 2 pseudocardinal teeth and 2lateral teeth in left valve, I or 2 pseudocardinal teethand I lateral tooth in right valve. Anal papillae present;simple branchial papillae usually dominant, but dentriticbranchial papillae sometimes present, occasionallynumerous; papillae absent along mantle margin anteriadfrom branchial aperture. Dorsal margins of innerlamellae of inner demibranchs not completely fused tovisceral mass. Eggs 'white; glochidia without hooks,usually expelled from marsupium in compacted masses(ovisacs).

TYPE-SPECIES: Unio nigra Rafinesque 1820(~ Elliptio crassidens (Lamarck 1819».

Elliptic complanata (Lightfoof 1786)(Plate I, Figure I; Map 3)

DIAGNOSIS: Shell up to 140 mm in length; elongate,trapezoidal; mean length/height ratio 1.96 (N ~ 52, S.D.= 0.15, range 1.74 to 2.84); valves usually flattened orslightly inflated, sometimes moderately inflated; meanheight/inflation ratio 1.68 (N = 39, S.D. ~ 0.17, range1.3 to 2.1). Anterior end smoothly rounded; posteriorend higher and trapezoidal, sometimes biangulate;ventral margin straight to noticeably arcuate; dorsalmargin slightly arched but appearing parallel orsubparallel to ventral margin, giving emphasis totrapezoidal shape of shell. Umbos low, uninflated,frequently eroded in larger specimens; beak sculpturewhen not eroded, of low concentric loops subparallel togrowth lines. Posterior ridge faintly double.Periostracum in young specimens brownish, yellowish,or greenish, sometimes with faint rays over entire

surface; periostracum in adults brown to black, usuallywith strongly elevated growth lines over entire shell.Pseudocardinal teeth in left valve small, serrated; singleright pseudocardinal serrated, but a vestigial tooth mayoccur slightly anterior to prominent pseudocardinal;lateral teeth with granulated surfaces. Pallial line oftenloops posteriad to posterior adductor muscle scar withposteriad extension usually less than one-fourth totallength of muscle scar. Nacre purple, copper orange, orpink, rarely salmon or iridescent white.

Branchial and anal papillae mostly unbranched,distinctly elevated, moderately to heavily pigmented.Glochidial masses in marsupium compacted into narrowelongate masses (ovisacs) along most of length andheight of outer demibranch, ovisacs may be expelledfrom marsupium intact as white narrow elongate wafers.Sexes separate, shell without sexual dimorphism.

TYPE LOCALITY: Restricted to Potomac River,Washington, D.C. by Johnson, 1976.

TYPE: Probably lost (Johnson, 1970, fide Dance,1966).

DISTRIBUTION: This is one of the most common ofthe Atlantic Slope unionids ranging from central Georgiato northern Canada and westward to Lake Superior.Matteson (I948a, 1948b) discusses the generaldistribution of the species.

ECOLOG Y: E. complanata seems to prefer sandy orsandy-mud substrates in moderately to fast flowingstreams. However, it can be found in almost anyfreshwater habitat from organically rich pond bottoms toclean sandy bars in sandy rivers. It has even beensuccessfully introduced into new lentic habitats (seeSmrchek, 1971, concerning the introduction of E.complanata in Douglas Lake, Michigan). It can beespecially abundant in some parts of the SavannahRiver. where giant specimens in excess of 140 mm havebeen found. At SRP, E. complanata shares a commonhabitat with E. icterina, where the two speciesfrequently coexist in approximately equal numbers. See,also, Matteson (l948a, pp. 715-716).

LIFE HISTORY: Most of the information presentedhere is from Matteson (l948a), who describes the lifehistory of E. complanata in considerable detail. Thegeneral pattern of development demonstrated by thisspecies is very similar to that of many other unionidsand can serve as a guide for the group. Hence, moreinformation is presented here than in subsequent "LifeHistory" sections. In the latter, those features whichtend to differ from this species will receive the greatestattention.

Sexes are separate in E. complanato as is the typicalcase in the unionids. Matteson (I948a) found viablesperm present in the testes throughout the year, with thesperm count increasing during late spring. Males beginto release sperm into surrounding waters via the analaperture in late April. Release continues throughmid-June. with the peak sperm release occurring in

8

mid-May. Sperm enter the female mantle cavity throughthe incurrent aperture. Simultaneously eggs are movedfrom the ovary via gonoducts to the supra-branchialchamber where fertilization normally occurs. Matteson(1948a) has observed fertilization in gill marsupia, and,in one instance, sperm "in the ovarian ducts leading tothe suptrabranchial chambers." Fertilization occurs atthe primary oocyte stage and eggs fertilized in thesupra-branchial chambers are moved into water tubes' ofgill marsupia in the outer demibranchs of the gills.

Details of cleavage and pre-glochidial development aregiven by Matteson (I948a). Briefly, development fromzygote to mature larvae requires approximately onemonth in the gill marsupium. Each water tube containshundreds of developing larvae which are loosely boundtogether in white "ovisacs." If female clams aredisturbed during the brooding period by suddentemperature changes, toxic substances in the water,rough handling or other factors, they often release theovisacs, The ovisac of E. complanata is distinctivelyshaped, being long, narrow and slightly broaderventrally. This is in direct contrast to the shorter,broader, more oval ovisac of an aberrant Mill Creekpopulation of Elliptio , which we presently refer to as E.icterina (see below). The shape of the aborted ovisacreflects the shape of the water tube cavity and,according to Fuller (1971, 1972), the evolutionary historyof the genus. Matteson (l948a) reports that expulsion ofovisacs represents abortion of larval stages rather thanthe primary mechanism of release. He shows thatovisacs contain larvae in various stages of development.Glochidia are brooded until "mature," when they arereleased individually via the suprabranchial chamber andanal aperture. The actual mechanism of transport to thesuprabranchial chamber is unknown. However, asglochidia are ejected from the anal aperture, they areaccompanied by mucous strands which attach towhatever they contact in "cobweb" structures.Apparently this facilitates encounter with fish hosts.

An individual glochidium of E. complanata is a small,oval, bivalved, hookless larva approximately 200 I-' inlength and slightly less in height. Unionid glochidiausually must attach to a suitable fish host fordevelopment to proceed, although some species (e.g.,Anodonta imbccillis fide van der Schalie, 1970) appear tobe facultative parasites, which may develop directlywithout encystment. Obligatory parasitic glochidia suchas those of E. complanata usually attach to and encyston fish gills. Some species (e.g., Anodonta spp.) havehooks on the larval shell and can attach to the fins offishes. Glochidia are apparently host specific, buthost-glochidia relationships are far from being wellknown. There is considerable controversy concerningthe "parasitic" period of the unionid life cycle,particularly whether the encysted glochidium actuallyderives nutrition from the host fish or merely uses thefish as a dispersal mechanism. Fuller (1974) reviews the

problem and presents a considerable bibliography. Thosefish that serve as glochidial hosts of SRP mussels arelisted in a separate section in each species discussion(see below).

Matteson (l948a) reports the encysted stage in E.complanata to be approximately 18 days. During the endof this period, the glochidia undergo a metamorphosis inwhich there is considerable alteration of internalanatomy. The clam then drops from the fish host tobegin independent development. This period of the lifecycle of most mussels including E. complunata is poorlyunderstood. Hatteson (I948a) was able to hold juvenileclams 35 days after the glochidial stage, with the shellincreasing from 208 I" to 840 1". Most information on thegrowth of unionids concerns specimens of at least I emor more in length. Matteson (I948a) suggests that thegreatest amount of growth in E. complanata occursduring the third growing season and the next greatest inthe second growing season. The amount of shell materialdeposited diminishes with age with little appreciablegrowth of the shell occurring after the twelfth year(Matteson, I948a). Sexual maturity is apparently reachedin the second or third year.

GLOCHIDIAL HOSTS: The yellow perch, Pereafiavescens has been reported to harbor glochidia of E.complanata (Lefevre and Curtis, 1912; Matteson, 1948a)as has the banded killifish, Fundulus diaphunus (Wiles,1975). P. fiavescens is known to occur commonly inwaters throughout the SRP (McFarlane, 1976).

ADDITIONAL REMARKS: Elliptic complanata andE. icterina are the most common species of unionidsencountered in this area. They are also frequentlydifficult to distinguish, especially on the basis of shellmaterial alone. E. icterina tends to be somewhat smallerand more pointed and to have a smoother shell than E.complanata. The latter species usually possesses a roughperiostracum and a more quadrate or rhomboidal shape.None of these differences is consistently reliable. Themost positive means of identification is the nature of themarsupia and the contained ovisacs (see Life Cycle,above). Unfortunately, the ovisacs are formed onlyduring summer months.

Fuller (1971) suggests that the pigmentation of analand branchial apertures of E. complanata is darker thanthat of E. icterina, We have not found this to be areliable criterion for differentiating the species. AlthoughE. complanata tends to have more darkly pigmentedapertural regions, there is considerable variability. ManyE. icterina, including Mill Creek populations (see below)and those from other localities on and off the SRP havebeen observed with very darkly pigmented anal andbranchial apertures.

Electrophoretic techniques can be employed todifferentiate the species (M. Smith, J. Hagan and J.Britton, unpublished data). For example, when theenzyme system Phosphoglucomutase (POM) is examinedusing starch gel electrophorsis on a tris-citrate buffer

9

(pH 8.0), the two species distinctly separate with thehomozygous product of E. complanata migratingconsistently faster than the homozygous product of E.icterlna. There are other differences that can also beshown by electrophoresis.

In summary, although E. complanata and E. icterinaare the most abundant species of the SRP, their absoluteidentification often remains elusive, even for experts (cf.Fuller, 1971). Anyone requiring precise specificidentification should always consult a knowledgeablemalacologist for advice and assistance.

Elliptio icterina (Conrad 1834)(Plate I, Figure 2; Map 2)

DIAGNOSIS: Shell reaching 86 mm in length;elongate, sublanceolate to subtrapezoidal: meanlength/height ratio 2.12 (N = 59, S.D. ~ 0.23, range 1.81to 3.49); values moderately inflated; meanheight/inflation ratio 1.49 (N = 59; S.D. ~ 0.15, range1.00 to 1.81). Anterior end smoothly rounded; posteriorend often pointed, sometimes biangulate; ventral marginslightly convex, rarely arcuate; dorsal margin straight tojunction with posterior margin where the junction isoften noticeably angulate; dorsal and ventral marginsusually not parallel or subparallel for most of theirlengths, so shell rarely presents rhomboidal shape.Umbos low, slightly inflated, frequently eroded in largerspecimens; beak sculpture when not eroded, of severalconcentric elevated loops which are subparallel togrowth lines and slightly tuberculate posteriorly.Posterior ridge slightly to distinctly double. Periostracumin young specimens as in E. romplanata; adultperiostracum often semi-polished on central shell,blackish or brownish black; infrequently with roughlyformed and elevated growth lines. Hinge teeth typical ofgenus, left pseudocardinals small, equal in size; rightpseudocardinals usually double with right anteriorpseudocardinal almost vestigial; laterals withgranulations. Pallial line often curves posteriad toposterior adductor scar, but rarely extends posteriadmore than one-half length of posterior adductor scar.Nacre iridescent white, pink, salmon orange, or purple.

Branchial and anal papillae mostly unbranched, verylightly to moderately pigmented. Olochidial masses inmarsupium inflated and broadly oval in shape, may beexpelled intact from marsupium as white broadlyelliptical wafers. Sexes separate. shell without sexualdimorphism.

TYPE-LOCALITY: Muddy shore, Savannah River,opposite Augusta, Georgia.

TYPE: Holotype, Academy of Natural Sciences ofPhiladelphia, 41381 figured by Johnson, 1970, Plate 9,Figure 3.

DISTRIBUTION: From the Escambia River system ofAlabama and Florida eastward through Florida andnorthward along the eastern Atlantic slope to the WhiteOak River, North Carolina.

ECOLOGY: In the region of sympatry (which includesthe SRP), E. ictcrina and E. complanata frequentlyshare similar habitats and can be collected side by sidein streams, rivers, or lakes. Both species are usuallymost abundant in small fast-flowing streams where theygenerally lie partially buried along stream banks or onbars. Like E. complanata, E. icterina may be veryabundant in some localities in the Savannah River, butrarely attains the size reached by some E. complanatathere. The Mill Creek population (see AdditionalRemarks, below) occurs in a shallow, narrow streamthat is only a few em deep except in occasional "pools"(perhaps 30 em in depth) close to its banks. Mill Creekis an upland SRP stream that drains a small watershedinto Upper Three Runs Creek. Its water presumably isacidic during much or all of the year, but this is thenorm for higher-gradient creeks draining essentiallyconiferous watersheds in the Savannah River basin.

LIFE HISTORY: The life history of E. icterina isprobably very similar to that described for E.complanata although there have been no detailedstudies. Egg masses accumulate in the marsupium duringJune and July. The glochidium is hookless, oval andvery similar in appearance to that of E. complanata. Seethe discussion under E. complanata for differences ofmarsupial structure and shape of ovisacs.

GLOCHIDIAL HOSTS: UnknownADDITIONAL REMARKS: The difficulty of

distinguishing E. icterina from E. complanata has beendiscussed in the remarks of the latter species, whichshould be consulted by the reader.

The specimens of E. icterina inhabiting Mill Creekdisplay some characteristics which may justifyrecognition of this population as (at least) a subspeciesdiffering from the typical icterina. The marsupium of theMill Creek E. icterina is far more obese than ischaracteristic for the species, and it is restrictedprimarily to the median one half of the outerdemibranch.

Most students of Nearctic Naiades have agreed thatSimpson (1900, 1914) and Ortmann (1910. 1911, 1912,1916) were justified in interpreting marsupial type as themost important factor in unionid classification. Indeed,discovery of a novel kind of marsupium has regularlybeen taxonomically recognized by erection of a newgenus. There is precedent, then, for also regarding theunderscribed Mill Creek '"Elliptio '" as the type speciesof an undescribed genus, on account of the restriction ofits marsupium to the median water tubes of the outerdemibranchs. (This character is otherwise known only inthe monotypic genus Hcmistena of the naiadbiogeographic Cumberlandian province (Ortmann, 1918;Neel and Allen, 1964; van der Schalie and van derSchalie, 1950). On the other hand, there is no reason toassume that Elliptio cannot be an exceptional case,where more than one marsupial type may have arisenwithin the same genus.

10

At any rate. introduction of new taxa is inappropriatein a handbook such as the present report. We haveemphasized the peculiarities of "the Mill Creek 'Elliptio'"simply in order to acquaint SRP scientists with thiszoological curiosity in their midst and to encourage itsfurther study and conservation.

This last point is of great importance. The Mill CreekElliptio is proposed as a population of special concern inSouth Carolina. As long as the SRP remains a NationalEnvironmental Research Park, there will be little or nodanger that the public will disturb this population. Thereis a very real potential danger, however, that SRPinvestigators themselves may unwittingly harm it. In aneffort to forestall such mistakes, it would be well tosurvey SRP Elliptio. especially populations in the upperThree Runs drainage, where (at least in Mill Creek)more typical E. icterina populations are believed to besympatric with the Mill Creek form.

Such an investigation would best be conducted inJuly, when the latter is known to be gravid. The valvescan be pried apart, slowly and gently (with thumbnail orweak reverse pliers), just enough to reveal the chargedmarsupium (if, of course, the specimen in hand is afemale). If the marsupium is fully charged and if it isobese and restricted the median halves of the outerdemibranch, the specimen belongs to "the Mill CreekElliptio"; if the specimen is a male or a barren female,this test cannot be performed, and field identification(i.e., without killing the animal) is not practicable unless,of course, shell characters will serve, as often is the casewith E. congaraea (Lea) and E. complanata (Lightfoot).Finally, such surveillance, conducted with utmostcaution, would do much to clarify the taxonomic status(and thus the legal status, if any) of what we are calling"the Mill Creek Etliptio": We realize that this animalmight well prove to be only a local abbe ration of E.icterina, but even in this case the surveillance wouldhave been fruitful in providing knowledge of a creaturethat, under any circumstances, is highly unusual andrequires conservation in the interest of further study andthe genetic richness of the molluscan fauna.

The discussion of E. ictertna by Fuller (1971) refersexclusively to the Mill Creek population. Hence, manyof the statements attributed to E. icterina as a whole byFuller (1971) are restricted exclusively to this singlepopulation, including figures of the marsupial structure.

Fuller (1972) described a new species, Elliptiomarsupiobesa from the Cape Fear River, NorthCarolina, and suggested that this new species is closelyrelated to E. icterina.

Elliptio lanceolata (Lea 1828)(Plate I, Figure 4; Map 5)

DIAGNOSIS: Shell up to 140 mm in length;noticeably longer than high, mean length/height ratio 2.6(N = 26, S.D. = 0.37, range 2.05 to 4.02); valvesmoderately inflated; mean height/inflation ratio 1.66 (N

= 26, S.D. = 0.12, range 1.43 to 1.83). Anterior endsmoothly rounded; posterior end frequently pointed,lanceolate, slightly biangulate in larger specimens;ventral margin straight or slightly arcuate, junction withposterior margin noticeably angular; dorsal margin andligament area elongate, slightly convex, subparallel toventral margin, junction of dorsal and posterior marginsobtusely angular. Umbos low, slightly inflated, distinctlyanteriad , frequently eroded in larger specimens; beaksculpture similar to that of E. icterina. Periostracumgrowth lines finely concentric, color in young specimenssimilar to that described for E. complanatu, brownish orblackish in adults, central area of shell smooth orsubpolished, closely spaced and elevated growth linesmay be produced on periostracum around shellperiphery. Pseudocardinal teeth small, left posterior andright anterior pseudocardinals poorly elevated orvestigial, surface of lateral teeth granulate. Pallial linepasses distinctly posterior adductor scar, then passesdorsally often as much as two-thirds length of posterioradductor scar before joining it.

Branchial and anal papillae mostly unbranched,distinctly elevated, lightly to moderately pigmented.Glochidial masses in marsupium very narrow, consistingof one, two, or occasionally three layers of larvae,extend 'Mh the length of outer dernibranch, being absentin ventral-most portion.

TYPE-LOCALITY: Tar River, Edgecombe County,North Carolina.

TYPE: Holotype, National Museum of NaturalHistory, 85905, figured in Johnson, 1970, Plate II,Figure 2.

DISTRIBUTION: Escambia and Apalachicola Riversystems in Florida and Alabama, but not extending topeninsular Florida, and from the Satilla River system,Georgia, northward to the Susquehanna River system,Pennsylvania.

ECOLOGY: E. lanceoluta is primarily a stream orriver mussel found in many of the same localities as E.icterina or E. complanata. It is only rarely taken fromlakes, apparently preferring flowing water to calmerconditions.

LIFE HIS TOR Y: This species produces marsupial eggmasses during the early summer from late May to earlyJuly. Egg masses are very narrow and extend 5/sth theheight of the dernibranch. These are distinguishingfeatures separating E. lanceo/ata from E. icterina, whichhas a shell sometimes of similar shape. Glochidia arehookless, oval, and typical of the genus.

GLOCHIDIAL HOSTS: Unknown.ADDITIONAL REMARKS: There is considerable

variability in the length of E. lanceolata. Riverspecimens tend to demonstrate more elongate formsthan specimens collected from smaller streams orcreeks. In the latter instance, the shell shape of E.lanceolata approaches that of E. icterina, andidentification can be difficult.

II

There is a closely related species, E. shepardiana(Lea 1834), in the Altamaha River system where it issympatric with E. lanceolata. Considering the largeamount of variability in E. laneeolata in the SavannahRiver drainage, one must question whether E.shepardiana is a distinct species and not an ecotypicvariation. Needless to say, E. shepardiana is quitesingular with respect to shell elongation with alength/height ratio frequently exceeding 4 andapproaching 6.

Elliptio eongaraea (Lea 1831)(Plate I, Figure 3; Map 3)

DIAGNOSIS: Shell up to 101 mm in length; slightlyelongate, sub-triangular, mean length/height ratio 1.7 (N= 29, S.D. = 0.13, range 1.45 to 2.12); valves noticeablyinflated; mean height/inflation ratio 1.5 (N ~ 29, S.D.~O.II, range 1.06 to 1.64). Anterior end smoothlyrounded to subangular; posterior end produced, pointedposterioventrally, ventral margin slightly convex; dorsalmargin and ligament area short, slightly convex, notparallel with ventral margin. Posterior ridge prominent tosubprominent, secondary ridge may occur; posteriorslope frequently with short bars or ridges crossinggrowth lines. Umbos moderately elevated, frequentlyeroded in larger specimens, located in anterior third ofshell. Periostracum smooth, yellowish or brownish andoften with greenish rays in young specimens; becomingyellowish brown to chestnut in larger shells, polished tosubpolished, sometimes with very fine radialcrenulations on central shell area; thin elevatedconcentric growth lines more apparent at anterior endand on posterior slope. Right pseudocardinal teeth large,chunky, serrated, subequal; left posterior pseudocardinallarge, serrated, more prominent than left anteriorpseudocardinal which may sometimes appear vestigial;lateral teeth short, with granular surfaces. Pallial lineusually not produced posteriad to posterior adductorscar. Nacre white or salmon, slightly irridescent.

Simple and dendritic papillae produced at branchialaperture; anal papillae short, lightly to moderatelypigmented. Nature of gravid marsupium unknown.

TYPE-LOCALITY: Congaree River, South Carolina.TYPE: Holotype, National Museum of Natural

History, 85693, figured in Johnson, 1970. Plate 5. FigureI.

DISTRIBUTION: From the Ogeechee River system,Georgia to the Cape Fear River system, North Carolina.

ECOLOGY: Although E. congaraea has beencollected from lower Three Runs Creek at DonoraStation, the species is apparently best adapted for largerrivers. It is commonly found on sand bars along theSavannah River.

LIFE HIS TOR Y: There is apparently no publishedinformation on the life history of E. congaraea. Manyfemales were collected in June, 1977, during field workfor this report, but none were found with larvae in the

gills and only a few were found with eggs in the gonads.Ortmann (1912) reports larvae in the gills of a closelyrelated species, E. crassidens , in mid-June. Possibly E.congaraca may hold larvae in the marsupium only avery short time before release.

GLOCHIDIAL HOSTS: UnknownREMARKS: Patrick et al. (1966) report E. incrassutus

from the Savannah River in the vicinity of the SRP.They are using this junior synonym of E. crassidens(Lamarck 1819) erroneously in reference to E.congaraea.

E. congaraea differs in several ways from the speciesof Elliptio discussed above, even though it presumably isa member of the same subgenus. For example, it has amore inflated, thicker shell. The reproductive period forthis species apparently is not similar to that of E.complanata. E. icterina, or E. lanceolata, The speciesoften displays at least some dendritic branchial papillae,despite Fuller's (1971) indication that dendritic papillaeare absent. We believe that E. congaraea should receivefurther study, particularly comparative soft-part analysiswith other Elliptio species. It and closely relatedcongeners such as E. crassidens (which is the typespecies of Elliptio) may prove to be sufficiently distinctfrom the complanata-icterina-lancrotata group that thelatter series might require subgeneric distinction.

Elliptio frutcrna (Lea 1852)(Figured by Johnson, 1970. Plate 7)

DIAGNOSIS: Shell moderately elongate and laterallycompressed; sculpture consists of weak plicationsrunning obliquely across the disc and of weak subradialridges on the posterior slope; periostracum is brownishyellow.

Incurrent papillae commonly are at least bifid, anddark pigment of mantle margin at incurrent apertureextends conspicuously for a short distance anteriad.

TYPE-LOCALITY: Lea (1852) based his originaldescription of Unio fraternus upon populations fromGeorgia in the Chattahoochee River of the ApalachicolaRiver basin in the naiad Apalachicolan biogeographicregion and from Georgia and South Carolina in theSavannah River of the Savannah River basin in thenaiad Atlantic drainage biogeographic region (see vander Schalie and van der Schalie, 1950). However, aChattahoochee specimen is the figured type. Therefore.conchologically relevant populations of the Savannahand of any other Atlantic drainage basins that maybecome involved cannot unequivocally be attributed toU. fraternus until living (or well preserved)Chattahoochee River specimens become available forcomparison to extant Savannah River materials. As aresult, we can only assume, on the basis ofconchological evidence, that our Savannah Riverspecimens are true Unio fraternus, Johnson (1970) wasthe first to ascribe U. fraternus to Elliptio as E.fraterna.

12

TYPE: The type specimen of Unio fraternus (Elliptiofratema] is the holotype , NMNH 85396.

DISTRIBUTION: The known (or supposed)geographical distribution of Elliptio fraterna is confinedto the Chattahoochee and Savannah Rivers of Georgiaand/or South Carolina.

ECOLOGY: In the mainstem Savannah River, Elliptiofrutema has been taken only from sand bars beneathone or two feet of water.

LIFE HISTORY: Nothing is known about the lifehistory of Elliptio fraterna .

GLOCHIDIAL HOSTS: Nothing is known about thelarval host(s) of Elliptio fraterna .

ADDITIONAL REMARKS: The 1972 discovery byone of us (SLHF) of conchologically at least putativeElliptio fraterna in the Savannah River close to SRP isthe cause of admitting this poorly understood species tothe present report. If, as we believe, this record isauthentic, it is the only recorded capture of living E.fraterna since the species' original description almost130 years ago. These data are sufficient demonstration ofthis species' natural rarity in at least the Atlanticdrainage.

GENUS Uniomerus Conrad 1853

DIAGNOSIS: Shell moderately elongate, rhomboidal,longitudinal axis generally straight. Beaks low,inequilateral, anterior, and prosogyrate. Ligamentelongate. slightly to moderately elevated. Periostracumvery well developed, clothlike , dark in adults, unrayed.Dentition consisting of 2 pseudocardinal teeth and 2lateral teeth in left valve, and one pseudocardinal andone lateral tooth in right valve. Anal papillae short orappearing as crenulations but never completely absent;branchial papillae dendritic; papillae absent along mantlemargin anterior to branchial aperture. Inner surface ofinner demibranchs not fused to visceral mass;marsupium restricted to outer demibranchs. Glochidiawithout hooks, held in marsupium in narrow masses.

TYPE-SPECIES: Unio tetralasmus Say 1831,subsequent designation by Simpson (1900).

REMARKS: The genus is considered monotypic byJohnson (1970 and 1972) and Fuller (1971). There aremarked morphological differences between Atlantic slopeand Interior Basin populations which Johnson ascribesto ecophenotypic variation. Valentine and Stansbery(1971) have a more liberal interpretation ofthe numberof species in this genus, stating that there are at leasttwo and possibly more. Based upon differences in shellmorphology, one is tempted to agree with the latterauthors that at least two species can be recognized, withthe populations of the Savannah River being atypical ofthe specimens from the Gulf Coastal drainages. Until adetailed examination of soft-part morphology is made,however. we are reluctant to differentiate the two shellforms as distinct species, and follow the moreconservative positions of Fuller and Johnson.

Uniomerus tetralasmus (Say 1831)(Plate I, Figure 6; Map 4)

DIAGNOSIS: Shell up to 114 mm in length, elongate,rhomboidal; mean length/height ratio 1.73 (N ~ 12, S.D.~ 0.11, range 1.50 to 1.88); valves moderately inflated;mean height/inflation ratio 1.59 (N ~ 12, S.D. ~ 0.13,range 1.38 to 1.81). Anterior end smoothly rounded;posterior end high, trapezoidal, relatively straight fromnear termination of hinge teeth to subangularposteroventral margin; ventral margin slightly convex,usually not arcuate; dorsal margin almost straight toslightly curved, angular at junction with posteriormargin. Umbos low, uninflated, frequently eroded inlarger specimens; beak sculpture when presentconsisting of 6 to 8 low concentric loops oblique togrowth lines on posterior slope. Posterior ridge broadlyrounded; posterior slope wide with one or two radiatingfinely incised lines. Periostracum usually dark brown orblack, heavy, often with a cloth like sheen.Pseudocardinal teeth in left valve serrated, subtriangularor triangular; single left pseudocardinal slightly larger,serrated; trace of a vestigial pseudocardinal may appearanterior to large left member; laterals typical of genus,sometimes faintly granulate. Pallial line extends slightlyto distinctly posteriad to posterior adductor scar, fromone-eighth to one-third length of scar. Nacre of valveinterior white, bluish lavendar, coppery purple, pink orpurple.

Anal papillae short and crenulate; branchial papillaemostly dendritic, some simple; anal and branchialapertures moderately pigmented; papillae absent alongventral margin below and anterior to branchial aperture,mantle frequency exhibits a faint gray band along ventralmargin. Glochidial masses in marsupium very narrow,elongate rods along most of length and height of outerdemibranch. Sexes separate. sexual dimorphism notclearly indicated in shell.

TYPE-LOCALITY: Bayou St. John, Louisiana.TYPE: Apparently lost, fide Johnson, 1970.DISTRIBUTION: Central and southern United States

from the base of the Rocky Mountains to the Atlanticslope drainages, generally below latitude 40 degrees.Johnson (1970) indicates this species is "noticeablyscarce" in the Carolinas.

ECOLOGY: U. t etralasmns is not especially commonin the waters of the SRP, but it does seem to be presentin ponds as well as streams, and frequently in fine mudsor silts. It has been taken in Upper Three Runs andLower Three Runs Creeks, where it avoids the swifterwaters over sand banks, preferring instead the fineorganics and leaf debris traps created by slowly movingside eddies. It is also present in the muds of Par Pond.

U. tetralasmus is particularly well adapted forsurviving drought or low water conditions for extendedperiods. This ability has prompted Fuller (1971) todescribe the species as having an ' enormous tenacity

13

for life." If stranded by dropping water levels in dryingstreams or ponds, this mussel will bury itself deeply inthe moist substrates to avoid dessication. One of us(JCB) has collected U. tetralasnius from a north Texasfarm pond which had been drained at least one monthprior to the date of collection. The surface of the drainedpond was firm enough to walk upon without sinking intomud, yet numerous living specimens were found at thebottom of deep cracks caused by the drying substrate.Simpson (1893), Baker (1928), and others have publishedsimilar observations.

LIFE HISTORY: Only the outer demibranchs aremarsupial in this species. Female marsupia were bareduring June, 1977, lacking either eggs or larvae, butseveral females were collected with eggs in the gonads.

GLOCHIDIAL HOSTS: Unknown.REMARKS: Valentine and Stansbery (1971) consider

the South Carolina populations of Uniomerus to be U.obesus (Lea 1831). They consider this species a shorter,more inflated form which differs from the type-species,U. tetrulasmus . Detailed examination of soft tissuemorphology or cytochemistry may prove that U. obesusis a valid name.

SUBFAMILY ANODONTINAE Swainson 1840

DIAGNOSIS: Marsupia occupying entire outerdemibranchs, demibranchs greatly distended when filledwith ova and/or larvae; secondary partitioning of thewater tubes present; anal aperture without papillae;lateral teeth, pseudocardinal teeth or both frequentlyreduced or absent; umbonal sculpture usually a series ofraised double loops; glochidia ovate or subtriangularwith hooks.

GENUS Alasmidontu Say 1818

DIAGNOSIS: Shell inequilateral, elliptical orrhomboidal, inflated moderately heavy posterior ridgeusually prominent. Umbos inflated, anterior, andprosogyrate. Ligament short but prominently elevated.Periostracum thin to well developed, darkly colored inadults, often rayed. Pseudocardinal dentition alwayspresent; lateral dentition present, vestigial or absent;when present paired laterals may be in right valve. Innersurface of inner demibranchs mayor may not fuse withvisceral mass epithelium.

TYPE-SPECIES: Unio hcterodon Lea 1830.

A/asmidonta triangulata (Lea 1858)(Plate 11. Figure 1, Map 4)

DIAGNOSIS: Shell elongate, subrhomboidal usuallyless than 60 mm in length; mean length/height ratio 1.49(N ~ 4, S.D. ~0.04, range 1.44 to 1.51); valves thin tomoderately thickened, inflated; mean height/inflationratio 1.25 (N ~ 4, S.D. ~0.002, range 1.2 to 1.3).Anterior end rounded with most pronounced curvaturealong dorsal half of anterior margin; posterior enddistinctly pointed at termination of posterior ridge.

angulate at margin of posterior slope; ventral margingently curving; dorsal margin noticeably arched orsinuate, being most angulate just anterior to the umbo.Umbos prominent, inflated and elevated, crossed byseveral distinct, broadly spaced ridges that extend ontothe uppermost eighth of the shell. Posterior ridgedistinctly elevated, angular; secondary or tertiary ridgesmay occur on posterior slope and form marginalangulations at their termination. Periostracum relativelythin over most of shell exterior, but becoming slightlythickened on posterior slope, light green or yellow inyoung specimens, becoming brown, greenish brown orblack in older animals; green rays often prominent onshell beneath periostracum. Left valve with a small,poorly elevated pseudocardinal tooth with a slightgroove anterior to it and a slightly toothed interdentumto the posterior; left lateral tooth almost absent.

TYPE-LOCALITY: Upper Chattahoochee River,Georgia.

TYPE: Holotype, NMNH, 86249, figures in Johnson,1970, Plate 13, Figure 5.

DISTRIBUTION: Apalachicola River system to thesouth, and the Ogeechee, Savannah, and Santee-CooperRiver systems on the Atlantic slope. The species is notcommon in any part of its range, and thus shouldperhaps be considered a threatened or endangeredspecies.

ECOLOGY: A. triangulata is a rare species with littleknown about its ecology or life cyile. It is usually foundin rivers or larger streams throughout its range, butnever in abundance. During June, 1977, the species wasfound at only one locality on the Savannah River.

GLOCHIDIAL HOSTS:: There have been no reportsof host fish for the glochidia of Alasmidonta triangulata.Morrison (in Clarke and Berg, 1959) reports thatglochidia of A. calceola (Lea 1830) from the Mississippisystem attach to a darter, Etheostoma nigrum,McFarlane (1976) does not report this fish from watersof the SRP, but several other species of the same genusare present.

GENUS Anodonta Lamarck 1799

DIAGNOSIS: Shell slightly to noticeably inequilateral,elongate, variably inflated, usually thin, fragile, posteriorridge may be present but poorly differentiated. Umboslow or high, submedian to noticeably anterior. Ligamentnarrow, low and elongate. Periostracum usually ratherthin, variably colored in adults, often rayed. Hinge teethtotally absent. Inner demibranchs not fused to surface ofvisceral mass along most of their height.

TYPE-SPECIES: My titus cygneus Linnaeus 1758(Opinion 94 and 561, International Commission ofZoological Nomenclature, 1926 and 1959).

REMARKS: Johnson (1970) recognized twosubgenera, Pyganodon Crosse and Fisher 1894(including A. cataractal and Utterbackia F. C. Baker1927 (including A. imbecillis and A. couperiana) in the

14

Savannah River drainage. The two groups can bedifferentiated in the Savannah drainage fauna by thenature of the umbo: Utterbackia has very low, almostflattened beaks; Pyganodon has inflated and elevatedumbos.

SUBGENUS Utterbackia F. C. Baker 1927Anodonta imbecillis Say 1829

(Plate I, Figures 7 and 8; Map 3)

DIAGNOSIS: Shell elongate, subrhomboid, up to 80mm in length; mean length/height ratio 2.27 (N = 30,S.D. = 0.18, range 2.02 to 3.08); valves very thin,fragile, moderately inflated; mean height/inflation ratio1.51 (N ~ 30, S.D. =0.12, range 1.28 to 1.75). Anteriorend smoothly rounded; posterior end subpointed toangular with greatest distention in lower half; ventralmargin slightly curving, subparallel to dorsal margin;dorsal margin almost straight, elongate. Umbos flat, notelevated or inflated; umbonal sculpture consisting of aseries of distinctly raised double loops. Posterior ridgesmoothly rounded; posterior slope somewhat flattened,sometimes raised into a slightly flaring wing.Periostracum relatively thin on all parts of shell, smoothand shiny medially, but slightly rougher on posteriorslope; color tan; yellowish brown, greenish brown orgreen; broad greenish rays common. Hinge thin, narrow,edentulous, ligament thin, narrow, elongate. Nacre ofshell interior bluish white to metallic steel blue.

Anal papillae absent, branchial papillae simple,unbranched, moderately pigmented. Monoecious, outerdemibranchs greatly distended when charged withglochidia.

TYPE-LOCALITY: Wabash River, New Harmony,Indiana (restricted by Clench and Turner, 1956).

TYPE: Original type lost. Neotype, SenickenbergMuseum 4301, selected by Haas, 1930.

DISTRIBUTION: From the eastern base of the RockyMountains to the Atlantic Slope.

ECOLOGY: A. imbecillis is a common pond or lakemussel, apparently preferring the quieter waters. It livesin sand, mud, or fine silts, but is usually absent fromgravels or shifting sands in fast flowing streams. This isthe most abundant species in Par Pond and other pooledbasins of the SRP, being common from shore to depthsof 8 to 10 m. The deeper water populations in Par Pondfrequently produce a prominent "wing" or elevation ofshell material posterior to the umbo (Plate I, Figure 7).

A. imbecillis is very common and easily collected inthe SRP lakes, where it cannot be confused with anyother mussel. The only similar species, A. couperiana,may occasionally coexist with A. imbecillis. Patrick etal. (1966) report that both species were collected from asingle station on the Savannah River near Johnson'sLanding. The species can be distinguished in that theshell of A. couperiana is more inflated and the ventralmargin is noticeably more convex than A. imbecillis ,which has a more elongate, streamlined appearance.

LIFE HISTORY: The life history of A. imbecillisdiffers from that of other mussels in several ways. Thespecies is almost always hermaphroditic (van derSchalie, 1970), although it is uncertain whetherself-fertilization is possible in this species. It may alsoavoid the parasitic larval stage, with larvae developingdirectly in gill marsupia (see below). The larvae arebrooded during the fall and winter months and may beheld in the gills as late as mid-May (Ortmann, 1909), Theglochidia bear hooks, which can be used to facilitateattachment to a fish host.

GLOCHIDIAL HOSTS: Clarke and Berg (1959) report.that glochidia of A. imbecillis have been found inassociation with tissues of the cyprinid Semotitusatromaculatus, the Creek Chub. Tucker (1927, 1928) hasshown that the centrarchid Lepomis cyunellus, theGreen Sunfish, also harbor glochidia. Both fish specieshave been reported from SRP by Mcf-arlane" (1976), butneither is reported from Par Pond, where A. imbecillis isvery common, Thus, it is possible that additional fishspecies act as hosts for the larvae. On the other hand,there are reports that A. imbecillis glochidia are onlyfacultative parasites and are capable of developingdirectly without the parasitic stage (e.g., Howard, 1914).

REMARKS: Clark and Berg (l959)present a discussionof the anatomy of this species.

Anodonta couperiana Lea 1840(see Johnson, 1970, Plate 16 and 17 for figures)

DIAGNOSIS: Similar to A. imbecillis except ventralmargin noticeably convex and divergent from the planeof dorsal margin; length/height ratio usually less than2.0; periostracum with narrow green rays; usuallynoticeably globose.

TYPE-LOCALITY: Hopeton, near Darien, McIntoshCo., Georgia.

TYPE: Lectotype, USNM 86673.DISTRIBUTION: St. Marys River, Florida to Cape

Fear River Drainage, North Carolina.ECOLOGY: Johnson (1970) states that the species

prefers sandy or muddy bottoms of ponds and sluggishstreams.

GLOCHIDIAL HOSTS: Unknown.REMARKS: This species was not collected in 1977,

but it has been recorded in previous collections made byFuller from the Savannah River.

SUBGENUS Pyganodon Crosse and Fisher 1894Anodonta cataracta Say 1817

(Plate I, Figure 5; Map 2)

DIAGNOSIS: Shell elongate, subrhomboidal, large, upto 175 mm in length; mean length/height ratio 1.87 (N =

*McFarlane (personal communication) doesn't believe Green Sunfishhave been reliably reported from SRP. In the reference cited he citesFreeman for the record of this species presence in streams. McFarlanebelieves the record may very well be due to aberrant warmouth basswhich he has seen here.

15

3, S.D. ~ 0.16, range 1.75 to 2.06); valves thin, light inyounger specimens but becoming slightly thicker inolder, larger shells; moderately well inflated; meanheight/inflation ratio 1.28 (N ~ 3, S.D. = 0.07, range1.20 to 1.35). Anterior end smoothly rounded; posteriorend somewhat pointed or wedge-shaped; ventral marginalmost straight or slightly and smoothly curving; dorsalmargin straight, subparallel to ventral margin. Umbosslightly to moderately elevated above hinge line,somewhat inflated, beak sculpture having characteristicdouble-loop ridges of genus. Posterior ridge broadlyrounded; posterior slope broad, slightly depressed.Periostracum smooth, shiny over median region of shell,becoming rough and laminate anteriorly and posteriorly,especially on posterior slope: color yellowish brown,yellowish green, or greenish; broad green rays common.Hinge endentulous, thin, narrow, except below ligament,where it becomes slightly thickened; ligament poorlyelevated, thin, narrow, elongate. Nacre of shell interiorwhite or bluish white, sometimes with a slight pinkishflush.

Anal papillae absent, branchial papillae simple,unbranched, moderately pigmented. Sexes separate.

TYPE-LOCALITY: Deep part of a mill dam nearPhiladelphia, Pennsylvania (locality restricted byJohnson, 1970).

TYPE: Original type lost. No neotype designated.DISTRIBUTION: Alabama to the St. Lawrence River

along coastal drainage systems and possibly westwardalong the St. Lawrence system to Michigan.

ECOLOGY: A. cataracta can be found in a variety ofhabitats. It is equally common in lakes, streams, andrivers, and in substrates from mud to sand. It generallyavoids the swifter waters of streams, seeming to prefermore slowly moving currents in protected areas ofstreams. It has been found in the Savannah River nearthe mouth of Lower Three Runs Creek, at DonoraStation, and in Pond B. It likely also occurs in Par Pondas well. It also seems to be rather abundant in thenearby Clark Hill Reservoir.

LIFE HISTORY: Like many of the Anodontinae, A.cataracta holds larvae in the marsupium for severalmonths during the winter season, beginning in lateSeptember or October (Conner, 1909). Most specimensdischarge the hooked glochidia in May, but Ortmann(1909) reports gravid females rarely can be found as lateas mid-July.

GLOCHIDIAL HOSTS: Lefevre and Curtis (1910)report glochidia of this species from the cyprinidCyprinus carpio, the carp, which is present at SRP(McFarlane, 1976).

REMARKS: This is A. hallenbecki Lea of Patrick etal. (1966).

GENUS Strophitas Rafinesque 1820

DIAGNOSIS: Shell elliptical to rhomboid moderatelythickened, posterior end usually pointed, hinge teeth

rudimentary with usually at least one vestigial tooth ineach valve.

Strophitus undulatus (Say 1817)Figured in Johnson, 1970, Plate 17

DIAGNOSIS: Shell to 90 mm in length, rhomboid,valves moderately inflated, inequilateral; umboswell-expanded, vestiges of one pseudocardinal tooth ineach valve anterior to umbo; lateral teeth absent; nacrebluish or pinkish.

TYPE· LOCALITY: Schuylkill River, near Philadelphia,Philadelphia Co., Pennsylvania (restricted by Johnson,1970).

TYPE: Original type lost. No neotype designated.DISIRIBUTlON: Savannah River, South Carolina to

the St. Lawrence River, Canada and westward to theMississippi and Ohio River drainages.

LIFE HISTORY: This is another species ofAnodontinae which may complete metamorphosis fromglochidium to adult without the need of a vertebrate host(Johnson, 1970).

GLOCHIDIAL HOSTS: The Largemouth Bass,Micropterus salmoides and the Northern Creek Chub,Semotilus atromuculatus have been identified by Baker(1928) as hosts for glochidia in this species.

REMARKS: This species has been collected only oncein the vicinity of the SRP, on the Savannah River.

SUBFAMILY LAMPSILINAE von Ihering1901

DIAGNOSIS: Marsupium usually restricted toposterior half or less of outer two demibranchs, thisregion usually becoming greatly distended when filledwith ova and/or larvae; no secondary partitioning ofwater tubes; lateral teeth always present but sometimesweakly developed; pseudocardinal teeth usually present;glochidia subovate or hatchet-shaped, with or withouthooks.

GENUS Carl/nell/ina Baker 1898

DIAGNOSIS: Shell small, moderately inflated,obovate, moderately thickened, often pointedposteriorly; pseudocardinal and lateral teeth present;inner demibranch partially or completely free of visceralmass; female with well developed caruncle on mantlebelow branchial opening.

Caruncullna pulla (Conrad 1838)Figured in Johnson, 1970, Plate 17

DIAGNOSIS: Shell small, up to 35 mm in length,elongate, solid, inflated; beaks not well elevated; leftvalve with two pseudocardinals and two lateral teeth;right valve with one pseudocardinal and one lateraltooth; nacre usually white; female with pronouncedcaruncle.

TYPE-LOCALITY: Wateree River, South Carolina,restricted by Johnson, 1970.

16

TYPE: Original type lost. No neotype designated.DISTRIBUTION: Ocmulgee River, Georgia to the

Neuse River, North Carolina.GLOCHIDIAL HOSTS: Unknown.REMARKS: Fuller collected this species from the

Savannah River. It was not collected in 1977.

GEN US Vil/osa Frierson 1927

DIAGNOSIS: Shell usually small or moderate in size.thin; greatest height always posterior to center;periostracum often distinctly rayed; umbos in anteriorthird of shell; beak sculpture distinctly of double-loopedridges; posterior ridge indistinct or broadly rounded;pseudocardinals and laterals present; usually narrow andcompressed toward hinge line; mantle margin anterior tobranchial aperture distinctly papillate, but not forming apad or flap.

REMARKS: This is Mieromya Agassiz 1852, ofauthors, a name preoccupied in the Insecta.

Vil/osa delumbis (Conrad 1834)(Plate II, Figures 3 and 4; Map 5)

DIAGNOSIS: Shell with pronounced sexualdimorphism, relatively thin; length up to 70 mm; meanlength/height ratio 1.93 (N = 13, S.D. ~ 0.09, range 1.80to 2.12); valves moderately to greatly inflated, especiallyin females; mean height/inflation ratio 1.48 (N = 13,S.D. =0.14, range 1.33 to 1.75). Anterior marginsmoothly rounded; posterior half of shell considerablytaller in females than in males; posterior margin of malessubangulate with most acute curvature in posteroventralarea; posterior margin of females subtruncate ventrallyand produced dorsally with most acute curvature indorsal third of posterior margin; posteroventral marginof females almost straight and oblique to anteroposterioraxis of shell; ventral margin of male slightly curving;ventral margin of female descending almost straight to apoint about two-thirds of the distance from anterior end,thence turning at an obtuse angle and passing almoststraight to posterior termination; dorsal margin slightlycurved with hinge plate quite thin. Ligament narrow butprominently elevated. Umbos elevated slightly abovehinge line, moderately inflated, double loop sculpturedistinct in those specimens without umbonal shellerosion. Posterior ridge smoothly rounded, obscure,slightly more evident in male than female; marsupiumimposes a distinctive bulge on female shell anteriad fromposterior ridge; posterior slope may bear faint traces ofradial lines. Periostracum thin, smooth, clothlike,sometimes shiny on medial surface, but often laminateon posterior slope; color yellowish or yellowish tan,becoming brown or chestnut in older specimens, shellalways bearing narrow green rays which are usuallyinterrupted along their length by non-pigmented areas.Left valve bearing two thin, erect almost triangularpseudocardinals, which are closely applied to hinge line;separated by a narrow interdentum from two thin,

closely spaced lateral teeth; right anterior pseudocardinalsmall, lying on hinge line; right posterior pseudocardinallarger, more projecting, and more distinctly separatedfrom hinge line; narrow interdentum separatespseudocardinal area from a single right lateral tooth,Anterior adductor muscle scar moderately to deeplyimpressed; internal nacre white or bluish white, slightlyiridescent.

Inner mantle fold anterior to branchial siphondistinctly papillate in female, but may be rudimentary orabsent in male; mantle pigmentation moderately toweakly expressed.

TYPE-LOCALITY: Small stream near Cooper River,South Carolina.

TYPE: Apparently lost (Johnson, 1970).DISTRIBUTION: From the Altamaha River system,

Georgia, to the Neuse River system, North Carolina.ECOLOGY: This species was almost always found in

association with calm eddies in fast flowing streamswhere organic detritus and leaf debris tends toaccumulate. The most reliable method of collecting V.delumbis in these areas is to sift the debris through one'shands, searching for shells. V. vibex, Elliptiocomplanata. E. icterlna, and E. lanceolata are commonassociates with V. dclumbis,

LIFE HISTORY: Ortmann (1912) discusses severalMicromyu (~ viitosai spp., suggesting that the period ofbrooding is during the winter. However, numerousgravid females of this and the following species werecollected during June, 1977. As with all Larnpsilinae ,only the posterior portions of the outer demibranchs aremarsupial.

GLOCHIDIAL HOSTS: Unknown.REMARKS: Although male and female shells are

distinctly and easily differentiated, the dimorphism is notevident in the various computed shell ratios (e.g.,length/height) used in this study. This is largely due tothe location on the shell where measurements are taken.If. for example, height were measured more posteriorly,the dimorphism would appear in all ratios employing theheight dimension.

Villosu vibex (Conrad 1834)(Plate II, Figure 2; Map 5)

DIAGNOSIS: Shell weakly sexually dimorphic, thin;length up to 60 mrn: mean length/height ratio 1.96 (N ~

6, S.D. ~ 0.05, range 1.85 to 2.01); valves moderatelyinflated; mean height/inflation ratio 1.55 (N = 6, S.D.~0.08, range 1047 to 1.64). Anterior margin smoothlyrounded; posterior margin somewhat pointed in males.but more broadly rounded and more elevated in females;ventral margin of males slightly convexly curved inmales but almost straight or slightly arcuate in females;dorsal margin slightly curving, hinge plate very narrow.Ligament narrow and prominently elevated. Umbosslightly inflated and weakly elevated above hinge line,double-loop sculpture present in those specimens

17

without umbonal shell erosion. Posterior ridge smoothlyrounded, obscure; posterior slope may bear faint tracesof thin, slightly elevated or incised radial lines.Periostracum thin, generally smooth, almost glossy onmedial surfaces; color greenish yellow, yellowish,yellowish brown, or tan with broad green radial lineswhich usually are not interrupted along their length. Leftanterior pseudocardinal thin, erect, triangular, spacedsomewhat away from hinge plate; left posteriorpseudocardinal vestigial, two thin, narrow, closelyspaced left lateral teeth. Right anterior pseudocardinallarger, erect, triangular, distinctly projecting andseparated from hinge plate; single, thin, relatively shortright lateral tooth. Anterior adductor muscle scar slightlyto weakly impressed; internal shell nacre bluish white orsteel blue and iridescent.

Inner mantle fold anterior to branchial siphondistinctly papillate in female but may be rudimentary orabsent in male; mantle rather darkly pigmented.

TYPE-LOCALITY: Black Warrior River, south ofBlount's Spring, Alabama.

TYPE: Holotype, ANSP, 56488a, figured by Johnson,1970, Plate 17, Figure 10.

DISTRIBUTION: Pearl River drainage in Mississippieastward along Gulf Coastal drainages to Florida, andfrom the Altamaha River system, Georgia, to the CapeFear River, North Carolina.

ECOLOGY: Similar to that of V. delumbis.LIFE HISTORY: See V. delumbis,GLOCHIDIAL HOSTS: Unknown.

GENUS Lampsilis Rafinesque 1820

DIAGNOSIS: Shell moderate to large in size, solid,subovate to elongate; periostracum with or without rays;position of umbos variable; beak sculpture absent or ofthe double-looped type; sexual dimorphism usuallyprominent with female frequently more inflated andposteriorly truncate; pseudocardinals and lateralspresent, narrow, lamellar; mantle margin extended in aprominent flap-like papillate fold or pad in female;"eyespot" near branchial aperture often present infemale.

Lampsilis curiosa (Say 1817)(Plate II, Figures 5 and 6; Map 4)

DIAGNOSIS: Shell large, heavy with pronouncedsexual dimorphism; length up to 140 mm in males, and120 mm in females; mean length/height ratio of males1.52 (N ~ 27, S.D. ~ 0.07, range lAO to 1.66); meanlength/height ratio of females 1047 (N = 8, S.D. = 0.10,range 1.36 to 1.67); valves greatly inflated, especially infemales; mean height/inflation ratio of males 1.43 (N ~

27, S.D. ~ 0.07, range 1.31 to 1.58; mean height/inflationratio of females 1.28 (N ~ 8, S.D. = 0.06, range 1.22 to1.38; mean length/inflation ratio of males 2.17 (N ~ 27,S.D. = 0.15, range 1.93 to 2.59); mean length/inflationratio of females 1.89 (N ~ 8, S.D. ~ 0.20, range 1.73 to

2.30). Anterior end often bluntly pointed, sometimesslightly flaring, especially in the female. acute curvatureof posterior margin more pronounced in dorsal half withventral portion smoothly curving to join ventral margin:posterior margin extended and slightly to moderatelypointed in male. more abrupt and almost truncated infemale; ventral margin smoothly curving; dorsal marginarched. subangular with apex of arch located aboveinterdetum in left valve and near termination of rightlateral tooth; ligament large. prominently elevated.Umbos inflated. raised above hinge line. frequentlyeroded in large specimens. Posterior ridge usuallyobscure. posterior slope delineated primarily by changein periostracal texture. Periostracum well developed.thick. shiny smooth on anterior and lateral shellsurfaces. but becoming dull and distinctly laminate onposterior slope; color yellow. yellow brown. chestnut orreddish brown. Two serrated pseudocardinal teeth in leftvalve, anterior tooth usually larger, projecting anteriad.posterior tooth erect under umbo; broad interdentum inleft valve with hinge plate arched in posterior third ofinterdentum; two relatively short left lateral teeth withlower lateral having a slightly granular face. Twoserrated right pseudocardinal teeth with the ventral toothlarger and slightly posterior to the dorsal one, dorsalpseudocardinal sometimes sharply pointed; broadinterdentum in right valve; right lateral tooth short.prominently elevated. forming a distinct ledge on hingeplate. Pallial line usually does not extend beyondposterior termination of posterior adductor; anterioradductor scar distinctly impressed. Nacre of valveinterior white or bluish white and frequently flushed withpink.

Distinct. large darkly pigmented "eyespot" on eachmantle lobe of female just ventral to branchial aperture;mantle anterior to eyespot produced into a flap of tissue(not pad-like), lightly pigmented. Ventral margin ofmarsupium darkly pigmented.

TYPE-LOCALITY: Schuylkill River. near Philadelphia.Pennsylvania (restricted by Johnson. 1947a).

TYPE: Original type lost. Neotype , HarvardUniversity Museum of Comparative Zoology. 178839.selected and figured by Johnson. 1947a. Plate 19. FigureI.

DISTRIBUTION: Northern Georgia to Nova Scotiaalong the Atlantic slope drainages, and westward alongthe St. Lawrence River system.

L'COLOG Y: L. curiosa is primarily a large riverspecies. being particularly common along sand bars ofthe Savannah River. Its range apparently overlapsslightly with L. orhroccu , a tide-water species commonnear river mouths. It cannot be confused with L.ochracea near SRP as that species apparently does notoccur here. Johnson (1947a) has additional informationon the ecology and systematics of both species. L.curiosa is only rarely found in smaller streams or creeks.

LIF/:; HISTORY: The life history of L. carioso has

18

never been studied in detail. but it is presumably similarto that of several other closely related Lampsilinae (e.g..L. ventricose. L. ovcuu, etc.). Females of these speciesapparently employ the mantle flaps and "eyespots" onthe mantle margin as fishlures, attracting host species offish for the glochidia larvae. Kraemer (1970) presents adetailed discussion of this interesting behavior. Theeyespots are not photoreceptors. but are pigmentedareas which mimic the eyes of other organisms.

The species has been described as probablybradytictic (Ortmann. 1909; Conner. 1909). Almost allfemales collected in June. 1977. had glochidia or eggs inthe marsupia,

GLOCHIDJAL HOSTS: Unknown.REMARKS: L. cariosa is one of several closely

related species. which include L. ovata (Say 1817). L.dotabrucformis (Lea 1857). and L. satura (Lea 1852).Admittedly, the species listed here cover a very broadgeographic distribution. but certainly no more so thanUniomerus trtralasmus, which is regarded by some(e.g., Johnson. 1970 and 1972; and Fuller. 1971) asconspecific throughout the range, In fact. U. tetralasmusdisplays at least as much variation as all of the abovetaken together.

For example. L. dolabrocfonnis is a speciespresumably endemic to the Altamaha River system andallopatric with L, curiosa, whose range extendssouthward only to the Ogeechee River system. Johnson(1971) frequently compares L. dolabracformis to otherspecies: 1) "Is close to L. cariosa'" 2) "resembles LeXcaVll([{S," 3) "resembles L. binominutns." He pointsout that Simpson's (1914) report of L. dolobracjormisfrom the Ogeechee and Savannah River systems areerroneous records of L curiosa, In another sectionJohnson (1971) suggests that L. curiosa I) "can beconfused with L. dotabraeformis:' 2) "has been mostoften confused with L ochracva.:' and. with respect toL. ovalo, 3) "while the two species are close. they aredistinct." It is not surprising that Cvancara (1963)concluded that three members of the Lumpsilis complexfrom the central U.S .. L, ventricosa, L. ovata, and L.excarata probably represent subspecific taxa. It issurprising. however. that subsequent workers (Johnson1970; Fuller. 1971) do not formalize similar trends in theAtlantic slope Lampsilinae. We can follow Johnson(1970) and Fuller (1971) for the time being. accepting L.curiosa as the nominal designation for the large lampsilidof the Savannah River near the SRP. but we also suggestthat a careful anatomical, cytochemical or biometricstudy be attempted on this large. widely distributedcomplex to resolve some of the problems encounteredby a taxonomy based upon shell morphology alone.

Lampsilis radiata sptendida (Lea 1838)(Plate II. Figure 7; Map 5)

DIAGNOSIS: Shell moderately large. heavy. withslight sexual dimorphism; length up to 120 rnm; mean

length/height ratio 1.89 (N ~ 21. S.D. ~0.10. range 1.72to 2.10); valves well inflated; mean height/inflation ratio1.21 (N = 21, S.D. =0.10, range 0.98 to 1.42) with littledifference in this ratio between males and females.Anterior end smoothly rounded or sometimes slightlyangular in the dorsal portion; posterior margin bluntlypointed in males, smoothly rounded and less producedin females; ventral margin almost straight or slightlycurved; dorsal margin slightly arched; ligament large,broad, elongate, prominently elevated. Umbosmoderately inflated above hinge line, prominent.Posterior ridge rounded but distinctly formed; two orthree additional radial lines may appear on posteriorslope; posterior slope broad. Periostracum welldeveloped, extensively laminate but shiny on medialsurface, texture almost like satin. despite numerouslamellae; color reddish brown, yellowish, yellowishgreen or dark brown in older shells, frequently coveredwith numerous green rays which are irregularly spacedand varying in size from narrow to broad. Two elongatesubequal serrated pseudocardinal teeth in left valveseparated by a deep groove; anterior pseudocardinalsmall or vestigial, posterior tooth triangular, erect; righthinge plate with an elongate, narrow interdentum; leftlateral tooth single, straight, or slightly arched. Pallialline extends to or slightly beyond posterior terminationof posterior adductor scar; anterior adductor scar deeplyimpressed with a distinct shell ridge immediatelyposterior to scar. Nacre of valve interior pink or reddishbronze, often white centrally.

Pad below the branchial aperture of female modifiedinto a broad, darkly pigmented pad, less prominent inmale. Marsupium typical of genus, darkly pigmentedventrally.

TYPE-LOCALITY: Altamaha River, Mcintosh or LongCounty, Georgia.

TYPE: Holotype, NMNH, 84893, figured by Johnson,1970, Plate 22, Figure 5.

DISTRIBUTION: From the Altamaha River system,Georgia, to the Santee-Cooper River system, SouthCarolina.

ECOLOG Y: L. radiata splendida is usually found infast-flowing streams and rivers where it occurs on sandor gravel. It is not particularly abundant in fine-grainedsubstrates. but occasional specimens may be found inlakes (Johnson, 1970). It is most abundant in the largercreeks and river habitats near SRP.

LIFE HISTORY: This is another presumed bradyticticspecies in which almost all specimens collected duringJune, 1977, from the SRP and from the Ohoopee River,Georgia, had eggs in the marsupium.

GLOCHIDIAL HOSTS: Unknown.REMARKS: Although one can detect slight sexual

dimorphism in the shell morphology of L. radiatasplendida, this could not be demonstrated at astatistically significant level for either of the basic shelldimension ratios (length/height and height/inflation)

19

reported for specimens collected during this study. Thisis probably due to the use of a standard location forbiometric measurements.

L. radiata splendida cannot be confused with anyother freshwater mussel at SRP. It is distinctlylampsiline and the very darkly pigmented pad before thebranchial aperture of females is distinctive. This pad ispresent, but developed to a lesser degree in the male.

FAMILY CORBICULIDAE

DIAGNOSIS: Usually brackish to freshwater clamswith oval to trigonal shells from 0.5 to 10 cm in length;hinge with well developed cardinal and lateral teeth;ligament opisthodetic and parivincular: periostracumconspicuous; sculpture of valves usually of very fine tocoarse concentric bands.

GENUS Corbincula von Muhlfeld 1811

DIAGNOSIS: Shell with central or subcentral beaks,oval to trigonal shape; concentric sculpture usually wellimpressed on valves; lateral teeth usually elongate andserrate; inner demibranchs of gills form brood chamberfor pediveliger larvae: some species viviparous but mostdischarge pediveliger larvae which develop directlywithout parasitic stage.

Corbiculo ftnmineu (Muller 1774)(Plate II. Figure 8; Map 2)

DIAGNOSIS: Shell to about 65 mrn in length butusually less than about 40 mrn: mean length/height ratio1.13 (N ~ 200, S.D. ~ 0.07, range 1.00 to 1.30); valvesmoderately inflated; mean height/inflation ratio 1.48 (N~ 200, S.D. = 0.31, range 0.94 to 3.21). Shell shaperoughly equilateral suboval to trigonal; posteroventralmargin sometimes slightly more produced thananteroventral margin; dorsal margin angular; hinge platebroad with well developed dentition. Ligament broad butvery short. extending from beak to near origin ofposterior lateral teeth, Umbos central, well elevatedabove hinge plate. not elongate, often with distinctiveblue or purple blaze in shells less than 10 mrn length;posterior ridge absent. Periostracum yellowish in youngshells but becoming brown, reddish brown, or chestnutbrown in larger specimens. fresh periostracum even inlarge shells often yellowish or light tan. Sculpture onvalve exterior consists of well-impressed concentricrings with 6 to 10 rings per em. Three cardinal teeth ineach valve; left posterior cardinal almost vestigial, rightanterior cardinal closely adjacent to upper hinge plate;one anterior and one posterior lateral tooth in left valve:two anterior and two posterior lateral teeth in rightvalve: lateral teeth serrate; thickened calcareous plateusually underlies ligament area. Pallial line not sinusoidnear posterior adductor muscle scar; internal nacrewhite, lavender, or purple. Distinct anal and branchialsiphons formed; inner demibranchs modified as

marsupia; animals monoecious and brood developmentalstages (trochophore , and pediveliger larvae) in marsupia,

TYPE-LOCALITY: Unspecified locality in China,TYPE: Universitetets Zoologiske Museum,

Copenhagen (no catalog number assigned), figured byMorton, 1977a, Plate I.

DISTRIBUTION: Originally from central andsoutheastern Asia, this clam was first discovered in theUnited States in 1938 in the Columbia River, Oregon. Bythe early 1950's, it had expanded the range along thewest coast to southern California. It has since spreadrapidly throughout the United States and occurs acrossthe country from California to Pennsylvania, New Jerseyand Delaware, primarily south of latitude 42'. Itappeared in the Savannah River drainage between 1965and 1973.

ECOLOGY: An extensive body of literature isdeveloping on this potential pest species. C. flumineahas been responsible for clogging water cooling systems(e.g. cooling condensers) in a variety of industrialoperations, impeding water flow in irrigation and watersupply canals, spoiling gravel beds used for concreteaggregate, and perhaps crowding and competing forliving space with native unionid mussels. Many of theseproblems are associated with the life cycle and highfecundity of this clam (see below). It is beyond thescope of the present paper to describe the numerouspublished ecological studies on this species, For theearlier literature, the reader is referred to Sinclair andIsom (1963) and Sinclair (1971). The Proceedings of theFirst International Corbicula Symposium (1979) containsadditional literature and an extensive bibliography,

During field work at SRP in June, 1977, C. fiumineawas found to be exceedingly abundant at every localitysampled in the Savannah River. Densities in excess of2,000 living specimens/rn- were measured. C. fiumineawas also found in Beaver Dam Creek, but at no otherupland stream locality at SRP. Apparently C. fiumineawas recruited into Beaver Dam Creek via the intakewater for the SRP 400 Production area, with specimensor larvae being transported through the system anddeposited with the effluent entering Beaver Dam Creek.Other stream localities at the SRP should be closelymonitored with the date of first occurrence noted. Inview of the fact that there is exceedingly limited accessto most SRP streams, an assessment of potentialupstream migration by this species may be possible inthis area.

LIFE HISTORY: C. ftumineu in the United Statesappears to be a monoecious species, although Morton(1977a) suggests that a Hong Kong population in PloverCove displays protandry. Larvae are brooded in theinner demibranchs. A glochidium is never formed, andC. fiuminea never relies upon parasitism for dispersal.Rather, the larvae are released as free-livingpediveligers. In most localities larvae are produced twiceyearly, spring and faiL Aldridge and McMahon (1976)

20

estimate the average individual daily fecundity of thisspecies to be 387.0 veligers/clam and 319.8 veligers/clamfor spring and fall reproductive periods, respectively.Brooded larvae have been known to die before releasefrom the gill marsupia, stimulating a pathological tissueresponse (Morton, 1977b). Unreleased larva aresurrounded by amoebocytes which proliferate to form agranuloma,

The metamorphosis to a juvenile clam apparentlyoccurs shortly after release from the marsupium. Growthis exceedingly rapid during the first year, with shelllengths of 20 mm often being attained. Thereafter, thegrowth rate gradually diminishes (Britton, et al., 1979),The majority of a population apparently live two to threeyears, with some individuals surviving perhaps five ormore years.

REMARKS: C. fiuminea has been called C.manilensis, C. leana . and occasionally C. fluminalis inthis country, Consult Britton and Morton (1979) for ajustification of C. fiuminea (Muller 1774) as the validname for American Corbicula populations.

FAMILY SPHAERIIDAE

DIAGNOSIS: Shells thin, often translucent, small tominute, and oval to subtriangular in lateral outline.Ligament rarely external and, if so, usually partially tocompletely depressed. Cardinal and lateral teeth usuallypresent; two or fewer cardinals in each valve. In allthese respects, this family differs from the Unionidae.

Organization of soft tissues in Sphaeriidae similar tothat in Unionidae (discussed above), but there is onedifference readily useful in identification. In Unionidaethere is no fusion of the apposing mantle margins inorder to separate incurrent and excurrent mantleapertures, whereas this occurs in the Sphaeriidae, whoseapertures are separate and sometimes provided withextensible mantle margin tissue so that true siphons areformed.

Again unlike the Unionidac , the Sphaeriidae broodlarvae and young in gills until early adulthood, whenthey are released from the parent into the externalenvironment. where they pursue an independentexistence without intervention of a metamorphic periodof parasitism on another organism. The Sphaeriidaeapparently can forego the parasitic period. which is thesource of vagility for the comparatively ponderousUnionidae, because, as much smaller, lighter, and moreactive animals. they can depend upon mechanicaltransport for extending their geographic range.

REMARKS: We are confident that six species ofSphaeriidae (the pill, pea, and fingernail clams) have inrecent years been identified with dependable accuracyfrom the SRP. These are Pisidium amnicum (Muller1774), P. casertannrn (Poli 1791), P. compressurn (Prime1851), and P. dubium (Say 1834), plus Musculiumtransversurn (Say 1829) and Eupora cubensis (Prime1865),

Lawrence Tilly (SRL) contributed Pisidium amnicumfrom Upper Three Runs (Creek) and an indeterminatecongener from Par Pond. One of us (JCB) found P.amnicum in a Tinker Creek riffle just below KennedyPond. The other (SLHF) and colleagues are responsiblefor the other records, which have accumulated duringthe about 25-year ANSP surveillance of SRP waterquality, especially in Lower Three Runs (Creek) and theSavannah River.

Because of their small sizes, identification of spaeriidclams is difficult, especially in the case of Pisidium,which, unfortunately, is much the most species-richNearctic sphaeriid genus. On the other hand, at least onthe SRP, identification of Eup era cubensis (because ofits characteristic pigmentation of shell), Musuliumtransversum (which has a characteristic shape), anySphaerium that might be encountered (they are few andrefreshingly large), and P. duhium (also large, at leastfor a Pisidium, and thus easily visible!) is easy orrelatively so. However, the problem of the minutepisidia remains. Comprehensive taxonomic keys(Herrington, 1962; Burch, 1972, 1975a) to NearcticSphaeriidae exist and assuredly are vastly superior totheir predecessors. Nevertheless, identification ofPisidium species by shell criteria alone is hampered bythe fact that they seem to be convergent among severalsubgenera of this genus. Consequently, a key whoseinitial couplets dealt with soft-tissue discriminants amongthe subgenera and that only then moved to shellcharacters could provide a new approach to Pisidiumidentification by markedly reducing (or perhaps eveneliminating) the decisions between/amongconchologically similar taxa. Indeed, such a projectcould be the objective of a worker on the staff of (orvisiting) the SRP National Environmental ResearchPark. In any case, the need for improved aids inidentifying Pisidium is evident and great.

Opportunities for studying Pisidium and otherSphaeriidae are prevalent at the SRP. These days,

21

however, the interested investigator must search forthem more assiduously than was required in the earlyseventies, when enormous beds of submerged aquaticvegetation in the Savannah River provided optionalhabitat for scores of macroinvertebrate species,including Sphaeriidae. During that period of obviouseutrophy, P. casertanum and P. compressum wereabundant in the weed beds, and Musculium transversumwas almost inconceivably plentiful; the mobility of theseanimals means that they could exploit a dependablehabitat in the water column above their normal infaunalstation. Eupera cubensis was rare in the weed beds,presumably because its usual habit of byssal attachmentdoes not provide it adequate purchase in vegetation, butfor many years of ANSP investigation at SRP thisspecies has been varyingly common in bank-protectionriprap along the mainstream Savannah. In recent years.however, probably in response to improved treatment oforganic wastes in the upstream vicinity of Augusta,Georgia, the Savannah River has reverted to its earlieroligotrophy, and these extraordinary sphaeriidpopulations have substantially abated with the weedbeds, although all species can still be found.

ACKNOWLEDGEMENTS

Support for this project was provided by the NationalEnvironmental Research Park program of the SavannahRiver Plant Operations Office (US-ERDA). Severalpersons aided in the field collections. Majorcontributions were made by John Hagan, LarryChampagne, and Keith Daum. We are also appreciativeof the field assistance rendered by Michael W. Smith, JillGlanville, Peyton Williams, Tom Graham, and RonBlessing. Jean B. Coleman prepared the distributionmaps. Assistance for the typing of this manuscript wasprovided by the Department of Biology, Texas ChristianUniversity.

------------------------------------------------

LITERATURE CITED

Aldridge, D. W. and R. F. McMahon. 1976. Populationgrowth and reproduction in the life-cycle of Corbiculamanilensis Philippi. Abstract, Thirty-ninth AnnualMeeting, Amer. Soc. Limnology and Oceanogr.(printed in) Corbicula Newsletter, 1(4):3.

Baker, F. C. 1928. The fresh water Mollusca ofWisconsin. Part II. Pelecypoda. Part 2, Wise. Geol.and Nat. Hist. Survey Bull., 70:495 pp.

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 Corbiculafluminea (Muller). In: Britton, Joseph C., (ed.), Proc.First International Corbicula Symposium. TexasChristian University Research Foundation, pp.177-192.

Britton, J. C. and B. Morton. 1979. Corbicula in NorthAmerica: the evidence reviewed and evaluated. In:Britton, Joseph C., (ed.), Proc. First InternationalCorbicula Symposium. Texas Christian UniversityResearch Foundation, pp. 249-287.

Clarke, A. H., Jr. and C. O. Berg. 1959. The freshwatermussels of central New York. Cornell University Agr.Expt. Sta., Mern., 367:1-79.

Clench, W. J. and R. D. Turner. 1956. Freshwatermollusks of Alabama, Georgia, and Florida from theEscambia to the Suwannee River. Bull. Florida StateMus., 1:97-239.

Coldiron, D. R. 1975. Some aspects of the biology of theexotic mollusk Corbicula (Bivalvia: Corbiculidae). M.S. Thesis, Texas Christian University, Fort Worth,Texas. 92 pp.

Connor, C. H. 1909. Supplementary notes on thebreeding seasons of the Unionidae. Nautilus,22(10):111-112.

Cvancara, A. M. 1963. Clines in three species ofLampsilis (Pelecypoda: Unionidae). Malacologia,1:215-225.

Dance, S. P. 1966. Shell Collecting, An IllustratedHistory. Univ. Calif. Press, Berkeley, Calif., 334 pp.

Eng., L. L. 1977. Biology and population dynamics ofthe Asiatic clam, Corbicula manilensis (Philippi,1841), in the Delta-Mendota Canal, San JoaquinValley, California. Doctoral Dissertation, Univ. Calif.,Davis, 175 pp.

Frierson, L. S. 1927. A classified and annotated checklist of the North American naiades. Waco, Texas:Baylor Univ. Press, p. I-Ill.

Fuller, S. L. H. 1971. A brief field guide to thefresh-water mussels (Mollusca:Bivalvia:Unionidae) ofthe Savannah River system. Assoc. Southeastern Biol,Bull., 18(4):137-146.

___ . 1972. Elliptio marsupiobesa, a new fresh-watermussel (Mollusca:Bivalvia:Unionacea) from the CapeFear River, North Carolina. Proc. Acad. Nat. Sci.,Philadelphia, 124(1):1-10.

22

___. 1972. An undescribed structural feature in themarsupium of Elliptio lanceolata (Lea 1828)(Unionidae), Nautilus, 86(2-4):85-86.

___. 1973. Fusconaia masoni (Conrad 1834)(Bivalvia: Unionacea) in the Atlantic drainage of thesoutheastern United States. Malacological Rev.,6:105-117.

___ . 1974. Clams and mussels (Mollusca: Bivalvia).In Hart, C. W. and S. L. H. Fuller (eds.). PollutionEcology of Freshwater Invertebrates. Academic Press,New York, pp. 215-273.

Haas, F. I 969a. Superfamilia Unionacea. In: DasTierreich. Berlin. Lief., 88:I-X, 1-663.

___. 1969b. Superfamily Unionacea. In: Moore, R.C. (ed.). Treatise on Invertebrate Paleontology, PartN, Mollusca 6. Bivalvia 1, N411-N471.

Heard, W. H. 1977. Reproduction of fingernail clams(Sphaeriidae:Sphaerium and Musculium). Malacologia,16(2):421-455.

Heard, W. H. and R. H. Guckert. 1971. Are-evaluationof the recent Unionacea (Pelecypoda) of NorthAmerica. Malacologia, 10(2):333-355.

Hobbs, H. H., III., James H. Thorp and Gilbert E.Anderson. 1976. The freshwater decapod crustaceans(Palaemonidae, Cambaridae) of the Savannah RiverPlant, South Carolina. National EnvironmentalResearch Park, Savannah River Plant, 63 pp.

Howard, A. D. 1914. Some cases of narrowly restrictedparasitism among commercial species of freshwatermussels. Trans. Amer. Fisheries Soc., 44(1):41-44.

Johnson, R. I. 1947a. Lampsilis carrosa Say andLampsilis ochracea Say. ace. Papers Moll., Mus.Compo Zool., I: 145-156.

___. 1947b. The authorship of Elliptio complanatus.Nautilus, 62-36.

___ . 1970. The systematics and zoogeography of theUnionidae (Mollusca:Bivalvia) of the southern Atlanticslope region. Bull. Mus. Compo Zool., 140(6):263-449.

___ . 1972. The Unionidae (Mollusca:Bivalvia) ofpeninsular Florida. Bull. Florida State Mus. BioI. Sci.,16(4):181-249+Errata.

Kraemer, L. R. 1970. The mantle flap in three species ofLampsilis (Pelecypoda: Unionidae). Malacologia,10(1):225-282.

Langley, T. M. and W. L. Marter. 1973. The SavannahRiver Plant Site. U. S. Atomic Energy CommissionReport DP-1323, 175 pp.

Lea, I. 1852. Descriptions of new species of the familyUnionidae. Trans. Amer. Philos. Soc. 10:253-294.

Lefevre, G. and W. C Curtis. 1910. Reproduction andparasitism in the Unionidae. J. Exp. Zool.,9(1):79-115.

___ . 1912. Studies on the reproduction and artificialpropagation of freshwater mussels. Bull. BureauFisheries, 30:105-201.

Matteson, M. P. 1948a. The taxonomic and distributionalhistory of the freshwater mussel Elliptio complanatus(Dillwyn, 1817). Nautilus, 61:127-132; 62:13-17.

___ . 1948b. Life history of Elliptio complanatus(Dillwyn, 1917). Amer. MidI. Nat., 40:690-723.

MacFarlane, R. W. 1976. Fish diversity in adjacentambient, thermal, and post-thermal freshwaterstreams. In Esch, G. W. and R. W. McFarlane (eds.),Thermal Ecology II. ERDA Symposium Series,268-271.

Modell, H. 1942. Das Naturliche system der Najaden. I.Archiv fur Molluskenkunde , 74:161-191.

___ . 1949. Das naturliche system der Najaden. 2.Archiv fur Molluskenkunde, 78:28-48.

___ . 1964. Das naturliche system der Najaden. 3.Archiv fur Molluskenkunde, 93:71-126 (see Englishtranslation of part of this paper by D. H. Stansberyand U. Soehngen, 1964, Sterkiana, 14:1-18).

Morton, B. 1977a. The population dynamics ofCorbicula fiuminea (Bivalvia: Corbiculacea) in PloverCove Reservoir, Hong Kong. J. Zool., London,181:21-42.

___ . 1977b. The occurence of inflammatorygranulomas in the ctenidial marsupium of Corbiculafluminea (Mollusca:Bivalvia): A consequence of larvalincubation. J. Invert. Path., 30(1):5-14.

Ortmann, A. E. 1909. The breeding season of Unionidaein Pennsylvania. Nautilus, 22:91-95, 99-103.

___. 1910. A new system of the Unionidae.Nautilus, 23: 114-120.

___. 191 I. A monograph of the Najades ofPennsylvania. Parts I and 2. Mem. Carnegie Mus.,4:279-347.

___ . 1912. Notes upon the families and genera ofthe Najedes. Ann. Carnegie Mus., 8:222-365.

___ . 1919. A monograph on the naiades ofPennsylvania. Part 3. Systematic account of thegenera and species. Mem. Carnegie Mus., 8:1-384.

Owen, G. 1966. Feeding. In: Wilbur, K. M. and C. M.Yonge (eds.), Physiology of Mollusca. New York,Academic Press, pp. I-51.

23

Patrick, R., J. Cairns and S. Roback. 1966. Anecosystematic study of the fauna and flora of theSavannah River. Proc. Acad. Nat. Sci., Philadelphia,118:109-407.

Simpson, C. T. 1893. On the relationship anddistribution of the North American Unionidae. Amer.Nat. 27:353-358.

___ . 1900. Synopsis of the Najades, or pearlyfreshwater mussels. Proc. U. S. Nat. Mus.,22:501-1044.

Sinclair, R. M. 1971. Annotated bibliography on theexotic bivalve Corbicula in North America, 1900-1971.Sterkiana, 43: I 1-18.

Sinclair, R. M. and B. G. !som. 1963. Further studies onthe introduced clam Corbicula in Tennessee. StreamPollution Control Board, Tenn. Dept. Public Health,75 pp.

Smrchek, J. C. 1971. Present status of Elliptiocomplanatus in Douglas Lake, Michigan. Sterkiana,43:10.

Tucker, M. E. 1927. Morphology of the glochidium andjuvenile of the mussel Anodonta imbecillis. Trans.Amer. Micros. Soc., 46:286-293.

___ . 1928. Studies on the life cycles of two speciesof freshwater mussels belonging to the genusAnodonta. BioI. Bull., 54:117-172.

Vallentine, B. D. and D. H. Stansbery, 1971. Anintroduction to the naiads of Lake Texoma region,Oklahoma, with notes on the Red River fauna(Mollusca: Unionidae). Sterkiana, 42:1-40.

van der Schalie, H. 1970. Hermaphroditism amongNorth American freshwater mussels. Malacologia,10:93-112.

___ , and A. van der Schalie. 1950. The mussels ofthe Mississippi River. Amer. MidI. Nat. 44:448-466.

Wiles, M. 1975. The glochidia of certain Unionidae(Mollusca) in Nova Scotia and their fish hosts.Canadian J. Zool. 53:33-41.

GLOSSARY

anal aperture - The dorsal posterior excurrent mantleopening or siphon through which water and feces exitthe mantle cavity.

ungulate - Having the tendency to form an angle, incontrast to being smoothly rounded.

arcuate - Curved or bent as in the shape of a bow.beak - The umbo, or oldest portion of a bivalve shell.

Located on the dorsal surface, the beaks arefrequently elevated above the surrounding shellmaterial.

biangulate - Composed of two angular components.bifid - Having two parts; usually applied to the

description of hinge teeth.bradytictic - Referring to freshwater mussels which

retain ova and/or glochidia in gill marsupia for verylong periods, usually during all seasons exceptsummer.

branchial aperture - The ventral posterior incurrentopening or siphon through which water enters themantle cavity. Water and pseudofeces can exit themantle cavity through this aperture as a result of briefforceful contractions of various musculature.

cardinal teeth - Short, elevated triangular or slightlyelongate toothlike lamella on the hinge plateimmediately ventral to the umbo.

caruncle - A small fleshy protuberance on the inner edgeof each side of the mantle, directly anterior to thebranchial opening. Characteristic of members of thegenus Carunculina.

clam - A common popular term for various members ofthe molluscan class Bivalvia. The term is notrestricted to any particular group of bivalves, exceptby colloquial usage.

crenulate - With a wrinkled or minutely toothed margin.dendritic - Finger-like branching.demibraneh - One half of a gill of a bivalve mollusk.

Bivalves possess one pair of gills. Each gill is reflectedor folded upon itself; in cross-section it would appearmuch like the letter W. A demibranch is one V of theW. Upon dissection, bivalves appear to have two gillson each side of the body. Actually, these are twodemibranchs of the same gill.

edentulous - Lacking hinge teeth, as in Anodonta.elongate - The length dimension is noticeably greater

than the height.globose - Tending toward a spherical shape.glochidium - The characteristic bivalved parasitic larval

stage of the Unionacea.gravid - When said of bivalves which brood larvae in gill

marsupia, this term refers to the fact that the gillscontain eggs, developmental stages, or larvae.

height - The distance between dorsal and ventralmargins of the shell. See Figure 4, page 83.

height/inflation ratio - The height divided by inflation.hinge plate - The opposing dorsal portions of the valves

24

which include any lateral, pseudocardinal, or cardinaldentition and the interdentum.

inflation - The lateral thickness of a bivalve. See Figure2, page 83.

interdentum - The space on the hinge plate betweencardinal or pseudocardinal dentition and the lateralteeth.

lamellate (lamellar) - Formed in thin sheets or plates.laminate - Formed of thin sheets or plates lying one

upon another.lanceolate - Elongate, lance-shaped.lateral teeth - Elongate tooth-like lamellae on the

posterior half of the hinge plate in Unionacea, oranterior and posterior to the cardinal teeth in othergroups.

left valve - The shell half on the left side of the bodywhen the shell is placed with the hinge up and theanterior end directed away from the observer.

length - The maximum anteroposterior dimension. SeeFigure 4, page 83.

length/height ratio - The shell length divided by shellheight. See Figure 4, page 83.

ligament - The elastic organic structure which joins thetwo valves of a bivalve mollusk along the dorsalsurface.

mantle - An epithelial extension of the dorsal body wallof mollusks which usually secretes the shell andencloses a mantle cavity.

marsupium - The pouch or brood chamber that containsova, developmental stages, or larvae. In bivalves, thegills can serve as marsupia. The location and size ofthe marsupia are often useful taxonomiccharacteristics.

mussels - A common name of two general categories ofbivalve mollusks, I) members of the Unionacea,which are most often freshwater bivalves; or 2)members of the Mytilidae, which have numerousbrackish water or marine representatives.

nacre - The white, pearly, or iridescent inner shell layerof bivalves. It may be variably colored in somebivalves.

naiad - A general term applying to the freshwatermussels of the Superfamily Unionacea.

obovate - Somewhat oval in outline, but one endnarrower than the other.

opisthodetic - The condition in bivalve mollusks inwhich the hinge ligament is posterior to the umbo.

pallial line - An impressed line on the bivalve shellinterior locating the position of attachment of mantlemusculature.

papillate (papillose) - Having numerous finger-likeprojections or papillae.

parvivincular - Ligament with long axis parallel to hingeline and lying posterior.

pe divelig er - A form of shelled molluscan veliger larvae

which is adapted primarily for crawling rather thanswimming.

periostracum - The organic covering on the exterior ofbivalve shells.

plication - A fold or costa involving a major portion ofthe shell thickness.

posterior ridge - An external, slightly to sharply angulateridge on the posterior portion of many bivalve shellsextending from the umbo to the ventral margin. Whenthe ridge is not easily differentiated by an angulationon the shell, its location may often be determined by adistinct change in external sculpture.

posterior slope - The area of a bivalve shell dorsal andposterior to the posterior ridge.

prosogyrate - The beaks of the umbo point anteriorly.protandry - A hermaphroditic condition whereby the

male gonad develops first, degenerates, and isfollowed by the development of the female gonads.

pseudocardinal teeth - In Unionacea bivalves, thesestructures lie on the hinge plate near or anterior to theumbo. They are compressed, often triangular orslightly elongate dentition that have embryonicaffinities closest to anterior lateral dentition, but lie inthe position normally occupied by cardinal teeth.Unionaceans lack true cardinal dentition.

25

ray - A streak or linearly pigmented area usuallyassociated with the periostracum. It may beinterrupted or continuous.

right valve - The valve opposite the left valve, see latter.sinuate - To bend up and down or in and out.subovate - A shape approximating an oval, but slightly

to moderately produced or elongated in at least onedirection.

suhtrigonal - Transitional between trigonal and ovate: orinequilaterally trigonal.

tachytictic - Bivalve mollusks which brood larvae inmarsupia for relatively brief periods of time (i.e.,usually during spring and/or summer months only).

trigonal - Shaped as a triangle; three cornered; usuallyequilateral.

tuberculate - Describing the presence of tubercules,pustules or small globular surface irregularitiesespecially on the exterior of a bivalve shell.

truncate - Cut-off, abbreviated, more-or-Iess squarelyshortened.

umbo - The oldest part of the bivalve shell, the beak.See latter.

valve - One of two opposing halves of the shell of abivalve mollusk, which lie on the lateral surfaces ofthe body and are joined by a ligament.

ILLUSTRATIONS

I. Elliptio complanata, right valve exterior.2. Elliptio complanuta, dorsal aspect with posterior end toward top of page. Solid bar across valves indicates location

where inflation (I) is measured.3. Elliptio comp/anata, left valve interior.4. villose vibex, left valve exterior. Solid bars indicate locations where length (L) and height (H) were measured for this

study: note that the position of measurement is not necessarily the position of greatest shell height (see "height" inGlossary).

5. Schematic representation of dendritic branchial papillae, as expressed in Uniomerus tetralasmus .6. Posterior mantle margins of three lampsiline genera. a. Lampsilis. Note flap-like lobe ventral to branchial aperture

(BA) and prominent eyespot (E). b. Carunculina, Note caruncle (C) (i.e., mass of tissue) below branchial aperture(BA). c. Villosa .

Legend: A, anterior end; AA, anterior adductor muscle scar; AN, anal aperture; BA, branchial (incurrent) aperture; D,dorsal margin; E, eyespot; H, height; I, inflation; ID, interdentum; L, length; LG, ligament; LT, lateral teeth; P,posterior end; PA, posterior adductor muscle scar; PL, pallial line; PR, posterior ridge; PS, pseudocardinalteeth; D, umbo or beak; V, ventral margin.

26

5 .... ' .. .'. .

A

10o

v

27

LT

3

p

LG

-~ '\'P'

.,) )./

, -:--l \ .>

~<->

----H

4

PLATE I

Figure I. Elliptio complanata, male, Collected June 10, 1977, Pen Branch Creek at Highway C Bridge, SRP,Figure 2. Elliptio icterina, male. Collected June I, 1977, Mill Creek, SRP.Figure 3. Elliptio congaraca. male. Collected June II, 1977, on sand bar of Savannah River approximately 2 km

upstream from Johnson's Landing.Figure 4. Elliptio lancrolata, male. Collected June 2, 1977, Tinker Creek at Highway 8-11 crossing, SRP.Figure 5. Anodonta cataracta, Collected June 15, 1977, Lower Three Runs Creek near Donora Station, SRP.Figure 6. Uniomerus tetralasmus, female. Collected June 14, 1977, Lower Three Runs Creek near Donora Station, SRP.Figure 7. Anodonta imbecillis. Collected June 9, 1977, Par Pond, SRP, water depth approximately 8 m.Figure 8. Anodonta imbecillis. Collected June 9, 1977, Par Pond, SRP, water depth less than J m.

28

5~~~- - .~

, ,""'"

7

29

PLATE II

Figure I. Alasmidonta triangulata. Collected June 11, 1977, on sand bar of Savannah River approximately 6 km upstreamfrom Johnson's Landing.

Figure 2. Villosa vibex, female. Collected June 6, 1977, Reedy Branch Creek near Road 8-1, SRP.Figure 3. Villosa delumbis, female. Collected June 27, 1977, Tinker Creek just downstream from Kennedy Pond, SRP.Figure 4. Villosa delumbis, male. Collected June IS, 1977, Lower Three Runs Creek about y, km downstream from

Donora Station, SRP.Figure 5. Lampsilis cariosa, male. Collected June 21, 1977, on sand bar of Savannah River approximately 6 km upstream

from Johnson's Landing.Figure 6. Lampsilis cariosa, female. Collected July 20, 1977, on sand bar of Savannah River approximately 2 km

upstream from Johnson's Landing.Figure 7. Lampsilis radiata splendida, male. Collected June IS, 1977, Lower Three Runs Creek about V3 km downstream

from Donora Station, SRP.Figure 8. Corbicula j1uminea. Collected June 3, 1977, Savannah River, mile 157, shore of midchannel island, SRP.

30

...

MAP LEGENDS

Map I. The Savannah River Plant showing major drainage systems and sampling locations .• Bivalve mollusks collected at these locations;'" Bivalve mollusks not collected or observed at these locations.

THE FOLLOWING ARE ACADEMY OF NATURAL SCIENCES AT PHILADELPHIA COLLECTINGLOCALITIES WHERE ADDITIONAL BIVALVE MATERIAL HAS BEEN COLLECTED.

1 On Upper Three Runs about \.i! mi below SRP Road A, where Elliptio icterina has been collected.

2 On the Savannah River, where Elliptio congaraca. E. icterina, E. complunata. E. lanceolata, Uniomerustetralasmus, Anodonta imbecillis, A. cataracta, Vi/losa delumbis, Lampsilis curiosa, and Corbiculaflumineu have been collected.

3 On the Savannah River, where all species collected at Station 2 were also collected, plus Alasmidontattiangulata. Andonta couperiana, and Strophitus undulatus,

4 On the Savannah River, where all species collected at Station 3 except Strophitus undulatus were collected,plus Carunculina pulla and Elliptio fraterna .

BB On the Savannah River, where Elliptio icterina and Lampsilis radiata splendida were collected.

D Donora Station, where Elliptio icterina. E. complanata, E. lanccotata, Uniomerus tetralasmus, Anodontacouperiana, Villosa delumbis, and Lampsilis curiosa were collected.

L3R On Lower Three Runs Creek at SRP Road A, where Eltiptio icterina, E. lanceolata, Anodonta cataractuand Villosa vibex were collected.

TeR On Lower Three Runs Creek at Tabernacle Road, where Elliptio icterina, E. lanceolata and Vil/osa vibexwere collected.

THE FOLLOWING MAPS INDICATE THE LOCATION OF BIVALVES COLLECTED DURING THE SUMMER,1977.

Map 2. Collecting localities for Elliptio icterina, Anodonta cataracta, and Corbicula fiuminea,Map 3. Collecting localities for Elliptic camp/ana/a, Elliptic congaraea , and Anodonta imbecillis,Map 4. Collecting localities for Uniomerus tetralasmus , Lampsilis cariosa, and Alasmidonta triangulata.Map 5. Collecting localities for Elliptio lanceolata . Lampsilis radiata splendida, Vil/osa vibex, and V. delumbis .

32

33

.."Vl.....

0, , , ! IIki lometers

o I I !

I- miles

N

, I

miles

GEORG IA

• EllilZ.tio icterino--• Anodonla coloracla

• Corbicllia fillmineo

34

..'\ ..'\....

oI ! , , , Ikilometers

oI

N

•••

GEORGIA

£//il!.f/o comp/anofa----E//iptio congaraeaAnodonfa imbeci//us

35

oI ' , , , Ikilometers

o, 1

miles

N

! ,

miles

GEORGIA

• Uniomerus letraloslTlus.. Lamp-sills cariosa

• AlaslTI/donfa tasaostst«

36

oI ! , , , Iki lometers

oI

)..

C.,?

N

N

, ,miles

; :~~' .. 1 ,..

........ 0~e.,' . .: ;{ ~.,.~ IS'

e; ',c-;.~ , c;./ ..7::: Or" '; \ '='

.,J" :.f//;'·~C"'\tt. c-/-' "O"o/. . .$ ~, ' "r./ eCoAr--/ " G)

S' "..,' .,', .: ,.~./ SOUTH : ,,~GEORG IA 41,,, ' ,'. \ ,.c:......1\' '," CAROLINA· I-~~ j

E/liptio !anc8o!afo ~/'vl:~ r: ..{"• Lomp..si!is radio/a §12!endida r... y' ",0""• Vi!!osa de!umbis and 0 '.,'(':

I ! I ! , IV. vibex kilometers:

oI

•

37

NOTES

NOTES

NOTES