DANIEL L. GRAF
Museum of Comparative Zoology, Harvard University, Cambridge, Massachusetts 02138 and Department of Biology Northeastern University, Boston, Massachusetts 02115
ABSTRACT.-The chronology of the reinvasion of freshwater pearly mussels (Bivalvia: Unionoidea) into the southern Glacial Lake Agassiz region is reconstructed from the published record of the latest Pleistocene (Wisconsin) deglaciation of the region and known spatial and temporal distributions. The earliest invading unionoidean fauna entered the region above the driftless area of southeastern Minnesota and western Wisconsin between 14,000 and 11,000 years before present; the remaining species colonized the region postglacially. The formation of St. Anthony Falls is correlated with the drainage of Lake Agassiz from the Herman to the Campbell Beach level as an estimate of its date of formation, and a new chronology and mechanism of mussel dispersal into the Red River of the North basin is proposed.
Malacologists, working in the southern Glacial Lake Agassiz region (e.g., Wilson and Danglade, 1914; Dawley, 1947; Cvancara and Cvancara and Cvancara et al., see below; Clarke, 1973, 1981) and elsewhere (e.g., Simpson, 1896; Ortmann, 1924; Johnson, 1970, 1980), recognized the effect of the latest Pleistocene (Wisconsin) deglaciation on the distribution of freshwater pearly mussels (Bivalvia: Unionoidea) in the Canadian Interior and Great Lakes basins. The meltwater connections produced by wasting glaciers allowed the migration of mussels upstream from southern refugia (Johnson, 1980). Once water levels fell after glaciers receded, the basins took on their present configurations, limiting the dispersal of aquatic organisms such as fish-dispersed unionoideans across the divides.
The dispersal chronology of mussels into the region's now disjunct watersheds can be determined by analyzing their past and present distributions in the light of the known sequence of interbasin connections. Dawley (1944, 1947) recognized these aquatic connections, but limited geological information prohibited her from proposing a mechanism or chronology of dispersal. Cvancara (1966, 1967, 1970, 1975, 1976, 1983; Cvancara et al., 1966; 1972; 1976, 1979; Ashworth and Cvancara, 1983) contributed a wealth of information on the Recent and fossil Unionoidea of the southern Lake Agassiz basin but speculated little on the origin of the region's mussel community. Underhill (1989) has previously proposed a dispersal chronology for the reinvasion of the Minnesota's ichthyofauna; this is discussed below.
Present hydrology.-The southern Glacial Lake Agassiz region is composed of the headwaters of three major continental drainage basins (the Great Lakes, Hudson Bay and Mississippi River basins), and was formerly occupied by the early stages of the Glacial Lake Agassiz watershed, It is here divided into five drainage systems: the Lake Superior, Lake of the Woods, Red River, and Upper and Lower Mississippi River systems (Fig. 1).
The Lake Superior System, that portion of the Great Lakes basin within the region, includes Lake Superior and its tributaries in Ontario, Minnesota and Wisconsin. The Hudson Bay basin in the region includes only those tributaries of the Nelson River above Lake Winnipeg. For the purposes of this paper, the Nelson River watershed is subdivided into two drainage systems: the Red River System (the Red River of the North and all of its tributaries) and the Lake of the Woods System (the Winnipeg River with its tributaries).
The remainder of the study area drains to the Gulf of Mexico. This basin is subdivided into the Upper and Lower Mississippi River systems (following Underhill, 1989; also Graf, 1997) based on their divergent histories and disparate mussel communities. The former is that portion of the Mississippi River and all of its tributaries above St. Anthony Falls at Minneapolis, Hennepin County, Minnesota. The Lower Mississippi River System is the river and all of its tributaries between St. Anthony Falls and the mouth of the Wisconsin River (an arbitrary southern limit).
These five drainages were in place by the beginning of the Holocene, but they have not always been isolated. The enormous volume of meltwater left in the wake of wasting glaciers overran many of the region's present drainage divides and connected now separate basins until water levels receded.
Forty-six species of Unionoidea occur within the southern Glacial Lake Agassiz region. They were determined by examination of the holdings of the James Ford Bell Museum of Natural History, University of Minnesota, St. Paul, and a review of the relevant literature (Baker, 1928; Goodrich and van der Schalie, 1932; Cvancara, 1970, 1983; Clarke, 1973; Mathiak, 1979; Cummings and Mayer, 1992; Graf, 1997). The taxonomy follows Williams et al. (1993) and figures of these mollusks can be found in Cummings and Mayer (1992) and Clarke (1973).
Based on their distributions (Table 1), each species is placed into one of four unionoidean faunas associated with the five drainage systems of the southern Glacial Lake Agassiz region. Unionoideans of the same fauna are assumed to have had similar modes and tempos of dispersal via aquatic connections during the history of the study area.
Most species belong to the Lower Mississippi River Fauna (Table 1); these are the unionoideans limited to the Lower Mississippi River System. The Upper Mississippi River Fauna is composed of those unionoideans present in the Upper Mississippi River System and, with the exceptions of Strophitus undulatus and Utterbackia imbecillis, the four other drainage systems of the region. Five mussel species found only in the Lower Mississippi and Red River systems have been united as the Red River Fauna. Elliptio complanata, the lone unionoidean of the Lake Superior Fauna, is found only in the Lake Superior System. The fossil record.-Unlike the fish (Underhill, 1989), there is a fossil record, albeit limited, of the Unionoidea of the region. The Quaternary unionoidean paleofauna of the southern Lake Agassiz basin was reviewed by Ashworth and Cvancara (1983), the territory immediately to the W by Cvancara (1976), and that of eastern North America in general by Bogan and Grady (1991) .
The available literature provides an adequate record of the temporal distributions of unionoideans in the region and establishes that freshwater mussels inhabited environments in contact with glaciers (Clayton, 1961; Tuthill, 1961; Tuthill et al., 1964a). According to the dated or correlatable Lake Agassiz fossil material reported by Ashworth and Cvancara (1983) and Cvancara (1976), only Lampsilis siliquoidea is present in sediments older than 12,000 yr before present (ybp) (Moran et al., 1973). Anodontoides ferussacianus, Lasmigona complanata, Lampsilis cardium and Pyganodon grandis appeared between 12,000 and 8000 ybp (Moran et al., 1971). Although no Lake Agassiz unionoidean material is known for the interval from 8000 to 4000 ybp, in sediments dated between 4000 ybp and the present horizon, all the species of the Upper Mississippi River and Red River North faunas occur except Lasmigona costata, Potamilus alatus and Utterbackia imbecillis (Tuthill, 1964; Cvancara and Harrison, 1966; Cvancara, 1967; Cvancara et al., 1972, 1976). Tuthill (1963b), Tuthill et al., (1964b), and Zoltai (1969) reported only unidentifiable unionoidean fragments. The current Red River basin fauna is listed in Table 1.
Tuthill (1962, 1963a, 1967) erroneously noted numerous species from Pleistocene and Holocene Red River basin collections. These reports are based largely on Tuthill's own records, and they have not been replicated or substantiated by later collectors (i.e., Cvancara, 1970; Clarke, 1973). By excluding these studies (and the works cited within them), no species other than those of the present mussel community have ever been identified from Lake Agassiz sediments. I have dealt with erroneous, nonfossil records from the Lake Agassiz basin of Minnesota elsewhere (Graf, 1997).
Little fossil material exists for the portion of the study region to the E of the Lake Agassiz basin. Cvancara et al. (1979) recorded Pyganodon grandis from Glacial Lake Aitkin sediments dated at 11,500 ybp. Zoltai and Herrington (1966) reported Lampsilis siliquoidea and Obovaria olivaria from Canadian Lake Superior sediments older than 6000 ybp. Although I haven't examined voucher material, the occurrence of the latter in the Lake Superior basin is considered erroneous unless further evidence for its presence becomes available.
RECONSTRUCTION OF MUSSEL INVASIONS
In order to reconstruct the migration of pearly mussels into the southern Glacial Lake Agassiz region based on their distributions and the accepted chronology of interbasin connections, it is necessary to assume that they require aquatic connections for dispersal. Nearly all unionoideans are dispersed by fish (except for Simpsonaias ambigua which parasitizes the mud puppy, Necturus maculatus), and mussels have very specific geographic distributions limited to one or more drainage basins (van der Schalie, 1939). Terrestrial migration has never been proposed for the Unionoidea.
Some earlier authors (e.g., Call, 1878; Kew, 1893; Simpson, 1893) believed it possible for aquatic birds to disperse mussels. According to van der Schalie (1939), "no one has ever established that freshwater mussels have actually been transported from one stream to another in this manner." Even if a bird did carry a unionoidean to a new watershed, it would have to be a gravid female and an appropriate host would need to be present for the species to become established (discussed by Johnson, 1970). Further, high glochidial and juvenile mortality (McMahon, 1991) surely prohibits this. Dawley (1944) noted that it was conceivable that mussels may be distributed via transplanted fish; however, no evidence exists that this has occurred within the study area.
Another necessary assumption is that the drainage systems have remained isolated within the study area (except the Red River and Lake of the Woods systems, which are confluent at Lake Winnipeg) since their establishment at the close of the Pleistocene. Dawley (1947) noted, however, that Lakes Big Stone and Traverse, on the Minnesota-Dakota border, may be connected during times of high water, as they were in 1915. Cvancara (1970, pers. comm.) believed that the proximity of these two lakes probably still allows the transfer of mussels today between the Red and the Lower Mississippi river systems. If such dispersal occurred frequently, however, there would likely be a smaller difference between the number of species in the Red River System (13) and the species found in the Minnesota River basin (39) (Graf, 1997).
The following discussion outlines a proposed chronology for the invasion of each unionoidean fauna into the southern Lake Agassiz region. The proposed chronology in each section is based on Underhill's (1989) hypotheses of the distribution of Minnesota's ichthyofauna, with exceptions as noted. All absolute dates are from Clayton and Moran (1982) unless otherwise cited.
The late glacial invasion of the Upper Mississippi River Fauna.-St. Anthony Falls, because it prevents the upstream migration of fish, is a barrier to unionoidean dispersal. Wilson and Danglade (1914) believed that mussels above the Falls were carried by their fish hosts which were " . . . able to ascend the falls while [other glochidial hosts] could not accomplish it." Underhill (1989), however, concluded that "St. Anthony Falls has been a barrier to fish migrants since the late Pleistocene." Any mussels found in the Upper Mississippi River System either (1) migrated into the basin previous to the formation of the Falls or (2) had some postglacial dispersal route that bypassed the Falls.
St. Anthony Falls was formed by the rapid downcutting in the Minnesota River valley associated with the initial flooding (Glacial River Warren) through the southern outlet of
Glacial Lake Agassiz (Fig. 2) (Ojakangas and Matsch, 1982) but not with the flooding that occurred immediately before the final abandonment of that outlet. The latter flooding was not catastrophic and did not result in the lowering of the southern outlet (Fenton et al., 1983). Thus, the formation of the Falls occurred sometime before 11,000 ybp (the end of the first round of drainage through the southern outlet) but well after 11,700 ybp (the beginning of the formation of a lake in the southern Lake Agassiz basin). This correlates with the drainage of Lake Agassiz from the Herman to the Campbell Beach level (Fig. 2).
Since the Upper Mississippi River System's aquatic connections to other drainages within the region depended on a Lake Agassiz level above that of the Herman Beach (Hobbs, 1983), such connections must have been severed by the time the Falls formed. Thus, the hypothesis that the mussels somehow postglacially skirted the barrier can be rejected.
The Upper Mississippi River Fauna populated the Hudson Bay and Lake Superior drainages via the impounded water that breached drainage divides during the highest water stages of Glacial Lake Agassiz (including Glacial Lakes Aitkin and Upham; Fig. 2). Once those connections were severed with the drop in level of Lake Agassiz (Hobbs, 1983), the Lake of the Woods and Lake Superior systems remained confluent with the Lower Mississippi River System off and on for hundreds of years (Clayton and Moran, 1982). If unionoidean species other than those of the Upper Mississippi River Fauna had been present in the region during this period, it is assumed that they would have colonized the former two basins. Because the mussels of the Upper Mississippi River are the only species above St. Anthony Falls, they were the only unionoideans that migrated into the region before its formation.
The Upper Mississippi River Fauna first migrated from the Lower Mississippi River System into central Minnesota sometime after 14,000 ybp with the wastage of the stagnant Superior Lobe ice in the Twin Cities lowland and dispersed widely throughout the region before around 11,000 ybp (the latest time for the formation of the Falls of St. Anthony). The postglacial invasion of the Red River Fauna.-The ice lobe that held Lake Agassiz at a level high enough to drain through its southern outlet also forced the overflow of water in the Lake Superior basin to the Lower Mississippi River System via the St. Croix River (Clayton, 1983) (Fig. 2). Retreat of this ice not only exposed lower outlets for Lake Agassiz to the Great Lakes but allowed the Lake Superior basin to drain to Lake Michigan as well. Thus, concurrent with a connection between Lake Agassiz and the Mississippi River, a Lake Superior-Mississippi connection also existed.
If the unionoideans were carried along the path proposed previously for both the mussels and fish (Dawley, 1944, 1947; Cvancara, 1970; Clarke, 1973; Underhill, 1989), the Red River Fauna migrated up the Minnesota River valley to invade Lake Agassiz through its southern outlet. This hypothesis, however, fails to explain the absence of the Red River Fauna from the Lake Superior System, even though there would have been access via the St. Croix River.
The Red River Fauna is also absent from the Lake of the Woods System, further challenging this hypothesis. While Lake Agassiz drained through its southern outlet, the Red River and Lake of the Woods systems would have been confluent over a large area. Indeed, they have been confluent for at least the last 9500 yr when Glacial Lake Agassiz permanently abandoned its outlet to the Minnesota River, yet only two members of the Red River Fauna (Lasmigona costata and Potamilus alatus) have reached the Winnipeg River below Lake of the Woods (Clarke, 1973). The Winnipeg River appears to be the site of current mixing of the Red and Upper Mississippi River faunas. However, the numerous waterfalls along that river have surely arrested upstream dispersal since the withdrawal of Lake Agassiz from the region.
An alternative hypothesis is that the Red River Fauna entered the Lake Agassiz Basin postglacially via stream capture. Radke's (1992) statistical comparison of the fish and mussel communities of Red and Minnesota River tributaries indicates the possibility of a "... postglacial riverine connection between the Otter Tail River [a tributary of the Red River] and the Minnesota River. There are six species of fishes . . . which inhabit (or have inhabited) the Otter Tail River and no other waters within the Red River System. [Amia calva and Notropis anogenus] do inhabit lakes, and could have migrated through Galcial [sic] Lake Agassiz to the Otter Tail, but there are four species of fishes [Notropis texanus, Hypentelium nigricans, Etheostoma caeruleum and Percina phoxocephala] which live only in rivers and streams . . . " However, according to Underhill (pers. comm.), this single record of P. phoxocephala was based on a misidentified P maculata. Radke's conclusions were also based on two collections of mussels from the Otter Tail River: a single, dead Actinonaias ligamentina and a pair of Elliptio dilatata, later reidentified as Lampsilis siliquoidea and Ligumia recta, respectively (Graf, 1997).
The proposed mechanism for the postglacial connection between the Red and Lower Mississippi River systems is stream-head capture. Most tributaries of the eastern Red River basin generally trend E to W. The Otter Tail River, however, runs N to S above the mouth of the Pelican River at Fergus Falls, Otter Tail County, Minnesota. Earlier in postglacial time, the Upper Otter Tail River (that portion of the present-day Otter Tail River above the mouth of the Pelican River) continued S across the present continental divide to the Pomme de Terre River, a tributary of the Minnesota River, while the Pelican River drained S to the Mustinka River, another tributary of the Red River of the North System. The Lower Otter Tail River flowed in its present stream channel at that time, beginning at a point to the W of its present junction with the Upper Otter Tail River System. Headward migration of the Lower Otter Tail River eventually led to the capture of the Pelican and Upper Otter Tail rivers and the beheading of the Pomme de Terre River.
According to this hypothesis, with the abandonment of the southern outlet around 9500 ybp and the withdrawal of Lake Agassiz from the study area ca. 8600 ybp, the five drainage systems of the region were isolated much as they are today. Pearly mussels from southern refugia migrated throughout the Lower Mississippi River System. When the Upper Otter Tail River System was captured, the unionoideans present in those streams dispersed downstream into the Red River of the North and its tributaries, and the new species introduced by this mechanism were those species of the Red River Fauna. This hypothesis accounts for the absence of the Red River Fauna from the Lake of the Woods and Lake Superior systems.
The postglacial invasion of the Red River Fauna is supported by the fossil record of the region. From sediments dated from before 12,000 ybp to ca. 8000 years ago, the only mussel species present in Lake Agassiz sediments are widespread species of the Upper Mississippi River Fauna. In sediments dated after 4000 ybp, the species of the Red River Fauna appear (Ashworth and Cvancara, 1983).
The postglacial invasion of Lower Mississippi River and Lake Superior faunas.-The last unionoideans to invade the region via the Mississippi River are those of the Lower Mississippi River Fauna. If the Red River Fauna did migrate into the Red River System in postglacial time, there is no way to know whether members of the Lower Mississippi River Fauna were present in the Lower Mississippi River System at that time or not. The upper reaches of the Pomme de Terre River presently support a mussel community very similar to that of the Red River System (Bright et aL, 1995). Thus, the Red River Fauna may have been part of the larger migration of mussel species from southern refugia in postglacial time; since they were the only species located in the Pomme de Terre River at that time, they were the only species captured.
The most recent species to invade the region is Elliptio complanata of the Lake Superior Fauna. According to historical records (Walker, 1913; Goodrich and van der Schalie, 1932; Dawley, 1947), the Lake Superior Fauna likely migrated into Minnesota portions of Lake Superior from the eastern Great Lakes in the last 80 yr or so.
The distributional hypotheses presented here require further testing. Radke's (1992) stream-capture hypothesis is by far the most challenging new idea presented, and the dispersal chronologies of the fishes and the mussels have been revised because of it. Biogeographical studies of other aquatic taxa (e.g., the freshwater decapods) will undoubtedly further clarify the hypotheses presented through their own reflection of the history of interbasin connections in the region.
Acknowledgments.-This paper is dedicated to Professor Emeritus James C. Underhill, Curator of Ichthyology of the James Ford Bell Museum of Natural History, University of Minnesota. Professor Underhill talked with me at length on the glaciation and biogeography of the aquatic organisms of Minnesota. He and the late Dr. Robert C. Bright provided me with a wealth of resources, information, encouragement, and lunches, and for their contributions to this research, I am greatly indebted. Drafts of this paper were critically reviewed and improved by K J. Boss, A. M. Cvancara, J. M. Hogan, G. S. Jones, R. I. Johnson, E. Ruber, J. C. Underhill and anonymous reviewers. A. E. Bogan and K. E. Kolb provided references that I would not have otherwise been able to acquire. Although many pairs of eyes have scrutinized this work, any errors present are my own.
AsHWoRTH, A. C. AND A. M. CVANCARA. 1983. Paleoecology of the southern part of the Lake Agassiz Basin, p. 133-156. In:J. T. Teller and L. Clayton (eds.). Glacial Lake Agassiz. Geol. Assn. Can. Spec. Pap. 26.
BAKER, F. C. 1928. The fresh water Mollusca of Wisconsin: Part II. Pelecypoda. Bull. Wis. Geol. Nat. Hist. Suru, 70. University of Wisconsin, Madison. 495 p.
BOGAN, A. E. AND F. V. GRADY. 1991. Two Pleistocene molluscan faunas from eastern West Virginia, p. 189-213. In: J. R. Purdue, W. E. Klippel and B. W. Styles (eds.). Beamers, Bobwhites, and Blue-Points: Tributes to the Career of Paul W. Parmalee. Ill. State Mus. Sci. Pap. 23. BRIGHT, R. C., C. GATENBY, R HEISLER, E. PLUMMER, K STRAMER AND W. OSTLIE. 1995. A survey of the mussels of the Pomme de Terre and Chippewa Rivers, Minnesota, 1990. Bell Museum of Natural History, St. Paul, Minn. 131 p.
CALL, R E. 1878. Mode of distribution of fresh-water mussels. Am. Nat., 12: 472-473. CLARKE, A. H. 1973. The freshwater molluscs of the Canadian Interior Basin. Malacologia, 13: 1-509. 1981. The freshwater molluscs of Canada. National Museum of Natural Sciences, National Museums of Canada, Ottawa. 446 p.
CLAYTON, L. 1961. Late Wisconsin Mollusca from ice-contact deposits in Logan County, North Dakota. Proc. N. Dak. Acad. Sa., 15: 11-18.
. 1983. Chronology of Lake Agassiz drainage to Lake Superior, p. 291-307. In: J. T. Teller and L. Clayton (eds.). Glacial Lake Agassiz. Geol. Assn. Can. Spec. Pap. 26. AND S. R. MORAN. 1982. Chronology of Late Wisconsin glaciation in Middle North America. Quat. Sci. Rea, 1: 55-82.
CUMMINGS, K S. AND C. A. MAYER. 1992. Field guide to freshwater mussels of the Midwest. Ill. Nat. Hist. Surv., Manual 5. 194 p.
CVANCARA, A. M. 1966. Distribution and ecology of mussels in the Red River Valley, Grand Forks to Dayton. Proc. N. Dak. Acad. Sci., 20: 156-157 (abstract). 1967. Mussels of the Red River Valley in North Dakota and Minnesota and their use in deciphering drainage history, p. 187-196. In: W. J. Mayer-Oakes (ed.). Life, land and water. University of Manitoba Press, Winnipeg.
1970. Mussels (Unionidae) of the Red River Valley in North Dakota and Minnesota, U.S.A. Malacologia, 10: 57-92.
1975. Aquatic malacogeography of North Dakota. Proc. N. Dak. Acad. Sci., 27: 68-82. 1976. Aquatic mollusks in North Dakota during the last 12 000 years. Can. J. Zool., 54: 1688-1693.
1983. Aquatic Mollusks of North Dakota. Report of investigation no. 78. North Dakota Geol. Surv., Kaye's, Inc., Fargo. 141 p.
AND S. S. HARRISON. 1966. Distribution and ecology of mussels in the Turtle River, North Dakota. Proc. N. Dak. Acad. Sci., 19: 128-146.
R. G. HEETDERKS AND F. J. ILJANA. 1966. Local distribution of mussels, Turtle River, North Dakota. Proc. N. Dak. Acad. Sci., 20: 149-155.
, J. M. ERIKSON AND J. J. DELIMATA. 1972. Present and past mollusks of the Forest River, North Dakota. Proc. N. Dak. Acad. Sci., 25: 55-65.
, R. D. NORBY AND J. B. VAN ALSTINE. 1976. Mollusks of the Sheyenne River, North Dakota, USA: present and past. Malacol. Rev., 9: 25-38.
, K. L. MALMQuisT AND H. C. HOBBS. 1979. Mollusks from Glacial Lake Aitkin sediments, northeastern Minnesota Minn. Acad. Sci., p. 27-28, Abstracts of Pap. 47th Annu. Spring Meeting.
DAWLEY, C. W. 1944. Distribution and growth studies of the Unionidae and aquatic Gastropoda found in Minnesota. Ph.D. Thesis, University of Minnesota, Minneapolis. 307 p. 1947. Distribution of aquatic mollusks in Minnesota. Am. Midl. Nat., 38: 671-697. FENTON, M. M., S. R. MoRAN,J. T. TELLER AND L. CLAYTON. 1983. Quaternary stratigraphy and history in the southern part of the Lake Agassiz Basin, p. 49-74. In:J. T. Teller and L. Clayton (eds.). Glacial Lake Agassiz. Geol. Assn. Can. Spec. Pap. 26.
GOODRICH, C. H. AND H. VAN DER SCHALIE. 1932. I. On an increase in the naiad fauna of Saginaw Bay, Michigan; II. The naiad species of the Great Lakes. Occas. Pap. Mus. Zool. Univ. Mich., 238: 1-14.
GRAF, D. L. 1997. Distribution of unionoid (Bivalvia) faunas in Minnesota, USA. Nautilus, 110: 45-54. HOBBS, H. C. 1983. Drainage relationships of Glacial Lakes Aitkin and Upham and early Lake Agassiz in Northeastern Minnesota, p. 245-260. In: J. T. Teller and L. Clayton (eds.). Glacial Lake Agassiz. Geol. Assn. Can. Spec. Pap. 26.
JOHNSON, R. I. 1970. The systematics and zoogeography of the Unionidae (Mollusca: Bivalvia) of the southern Atlantic Slope region. Bull. Mus. Comp. Zool., 140: 263-450. 1980. Zoogeography of the North American Unionacea (Mollusca:Bivalvia) North of maximum Pleistocene glaciation. Bull. Mus. Comp. Zool., 149: 77-189.
KEw, H. W. 1893. The dispersal of shells: an inquiry into the means of dispersal possessed by fresh
water and land Mollusca. Kegan Paul, Trench, Trubner and Co., London. 291 p. McMAHoN, R. E 1991. Mollusca: Bivalvia, p. 315-399. In:J. T. Thorp and A. P. Covich (eds.). Ecology and classification of North American freshwater invertebrates. Academic Press, New York. MATHIAK, H. A. 1979. A river survey of the Unionid mussels of Wisconsin, 1973-1977. Sand Shell Press, Horicon. Wisconsin. 75 p.
MORAN, S. R., L. CLAYTON AND A. M. CVANCARA. 1971. New sedimental and paleontological evidence for history of Lake Agassiz: Snake Curve Section, Red Lake Co., Minn. Proc. N. Dak. Acad. Sci., 24: 61-73.
, M. W. SCOTT AND L. A. BROPHY. 1973. Catalogue of North Dakota radiocarbon dates. N. Dak. Geol. Surv. Misc. Ser, 53. 51 p.
OJAKANGAS, R. W. AND C. L. MATSCH. 1982. Minnesota's geology. University of Minnesota Press, Minneapolis. 255 p.
ORTMANN, A. E. 1924. Distributional features of naiads in tributaries of Lake Erie. Am. Midl. Nat., 9: 101-117.
RADKE, S. 1992. Otter Tail River fishes and mussels; an old connection. Plan B Master's Thesis, University of Minnesota, Minneapolis. 31 p.
SCHALIE, H. VAN DER. 1939. Distributional studies of the naiads as related to geomorphology. J. Geomorphol., 2: 251-257.
SIMPSON, C. T. 1893. On the relationships and distribution of the North American Unionidae, with notes on the west coast species. Am. Nat., 27: 353-358.
1896. The classification and geographical distribution of the pearly fresh-water mussels. Proc. U.S. Natl. Mus., 18: 295-343.
TUTHILL, S. J. 1961. A molluscan fauna and late Pleistocene climate in southeastern North Dakota. Proc. N. Dak. Acad. Sci., 15: 19-26.
1962. A checklist of North Dakota Pleistocene and Recent Mollusca. Sterkiana, 8: 12-18. 1963a. Corrections and additions to the checklist of North Dakota Pleistocene and Recent Mollusca. Sterkiana, 10: 29-30.
1963b. Molluscan fossils from Upper Glacial Lake Agassiz sediments in Red Lake County, Minnesota. Proc. N. Dak. Acad. Sci., 17: 9S101.
. 1964. Unusually well-preserved Lampsilis luteolus and molluscan fauna in posthypsothermal sediments. Compass, 41: 149-155.
. 1967. Paleo-zoology and molluscan paleontology of the Glacial Lake Agassiz Region. pp. 299-312. In: W. J. Mayer-Oakes (ed.). Life, land and water. University of Manitoba Press, Winnipeg.
L. CLAYTON AND W. M. LAIRD. 1964a. A comparison of a fossil Pleistocene fauna from North Dakota with a Recent molluscan fauna from Minnesota. Am. Midl. Nat., 71: 344-362. , W. M. LAIRD AND R.J. KRESL. 1964b. Fossiliferous marl beneath Lower Campbell (Glacial Lake Agassiz) Beach Sediments. Proc. N. Dak. Acad. Sci., 18: 135-140. UNDERHILL, J. 1989. The distribution of Minnesota fishes and Late Pleistocene glaciation. J. Minn. Acad. Sci., 55: 32-37.
WALKER, B. 1913. The unioned fauna of the Great Lakes. Nautilus, 27: 1823, 29-34, 4047, 56-59. WILLIAMS, J., M. WARREN, K CUMMINGS, J. HARRIS AND R. NEVES. 1993. Conservation status of freshwater mussels of the United States and Canada. Fisheries Bull. Am. Fish. Soc., 18: 6-22. WILSON, C. B. AND E. DANGLADE. 1914. The mussel fauna of central and northern Minnesota. Appendix V to the Report of the U.S. Commissioner of Fisheries, 1913 (Bureau of Fisheries Document No. 803). 26 p.
ZOLTAI, S. C. 1969. Sampling fossil mollusks from Glacial Lake Agassiz sediments. J. Paleontol., 43: 534-537.
AND H. B. HERRINGTON. 1966. Late glacial molluscan fauna north of Lake Superior, Ontario. J Paleontol., 40(2): 439-446.
SUBMITTED 9 OCTOBER 1995 ACCEPTED 31 OCTOBER 1996…