A comparison of freshwater mussels (unionidae) from a late-archaic archeological excavation with recently sampled Verdigris River, Kansas, populations.
Freshwater mussel extirpations and population declines due to anthropogenic activity have spanned the last century. This faunal change has been attributed to catastrophic events (toxic spills, excessive pesticide usage, illegal discharges, mine-spoil runoff) and cumulative changes to environmental suitability (impoundments, dredging, water quality, riparian development) that continue to imperil this faunal group (Williams et al., 1993; Neves, 1997; Haag, 2012). Comparisons of extant mussel assemblages with shells from archeological sites are of interest not only to anthropologists but also biologists considering modern conservation issues. Results of previous studies of shell material from archeological sites range from documenting extinctions and extirpations of species in the Cumberland River, Tennessee (Parmalee et al., 1980) to the relative intactness of mussel fauna over time in the Pomme de Terre River, Missouri (Klippel et al, 1978).
In 2009 and 2010, the Kansas Historical Society (KSHS) directed a salvage archeological survey and excavation at KSHS site 14MY388 (Fig. 1). This site is located along a sharp bend of the Verdigris River, Montgomery County, Kansas within an area prone to severe riverbank erosion and was approved for a streambank stabilization project pending an archeological survey. A preliminary geophysical survey of the construction site was completed using magnetic field gradient and electrical resistance to reveal burned rock features. These techniques resulted in the mapping and subsequent unearthing of a stone hearth cooking site, referred to herein as the hearth site (Tomasic, 2010, 2012). Excavation of the hearth and adjacent area uncovered dense concentrations of mussel valves, as well as a variety of artifacts including pottery, chipped stone tools, and bone fragments. The area near ([less than or equal to] 1 m) the hearth contained relatively few mussel shells, while the highest quantities of mussel shell appeared to be located in a ring-like fashion surrounding the hearth (Tomasic, 2010).
The Verdigris River watershed originates in the tallgrass prairie eco-region and the majority of the upland vegetative cover is native grassland (United States Geological Survey, 2000). There are two United States Army Corps of Engineers controlled impoundments upstream (Fall River and Toronto reservoirs) and one downstream of the study site in Oklahoma (Oologah Reservoir). Limestone and chert gravel streambed deposits form favorable substrate for unionid mussels at riffle/run sites. The Verdigris River currently has a diverse and relatively dense population of unionid mussels (Miller and Lynott, 2006).
It is of archeological and ecological interest to document species of mussels that were recovered at the hearth site and to compare relative abundance from this mussel midden with the mussel assemblage currently found in the Verdigris River. This information could be utilized to guide current management including re- introduction or population augmentation of rare mussel species. Objectives of this study were to: (1) determine age of the mussel midden site and document unionid mussel species that may have been used as a food source by past inhabitants lixing along the Verdigris River, (2) compare relative abundance of unionid mussel species from the hearth site with the assemblage that is currently in the Verdigris River, (3) compare size of intact mussel valves from the hearth site with samples recently taken from nearby Verdigris River sites, and (4) examine intact valves from the hearth site to estimate season of harvest by growth rings.
Methods used for excavation of the hearth site (N37.35193 W95.67545) have been described previously (Tomasic, 2010, 2012). Most (72%) of the area within a radius of 4 m from the hearth site was excavated. During excavation, samples of carbonized material were collected and submitted to the Illinois State Geological Survey for radiocarbon dating. Chipped stone artifacts and pottery fragments were classified based on the typology established by Justice (1987) and Marshall (1972).
Relic unionid valves from the hearth site were retained and bagged at the excavation for later examination. Contents of each bag were sorted into: (1) whole valves, (2) identifiable fragments of valves, and (3) unidentifiable fragments. Weights for each category were tallied and summed. Valves and valve fragments were compared to voucher specimens collected from nearby river sample sites to verify accurate identification. Umbo shape, beak cavity, external sculpture, arrangement of pseudocardinal and lateral teeth, and muscle scars were used to identify valves.
Intact and mostly intact relic valves from the hearth site were measured. Because the posterior end of many relic shells was broken, height measurements rather than length measurements were collected.
Intact valves from the hearth site were also examined in an effort to determine seasonality of the harvest. We examined small (young) individuals because spacing of growth rests is wider in young individuals that are growing rapidly. Western fanshell (Cyprogenia aberti) valves were also closely examined because this species exhibits well-defined ridges on the shell that mark the annual winter growth rests. A growth rest near the shell margin was presumed to indicate that the individual was harvested in winter or early spring. Conversely, if the shell had grown far beyond the last growth rest, this was taken as evidence for a late summer or fall harvest.
To assess the modern mussel fauna, Kansas Department of Wildlife and Parks personnel sampled mussels in 2009-2010 at eight riffle-run sample sites along a 13.2 km reach of the Verdigris River. Sample sites (A-H) were distributed both upstream and downstream from the hearth site and are collectively referred to as the Verdigris River survey (Fig. 1). Verdigris River survey sites were selected because they had been sampled previously, were identified as sites of past commercial harvest, or had stable gravel substrates under riffle/run flow and nearby bedrock exposures. Coordinates for the sample sites are: A: N37.39576 W95.66539, B: N37.39357 W95.66402, C: N37.37444 W95.66943, D: N37.37064 W95.66856, E: N37.36054 W95.67760, F: N37.35092 W95.68629, G: N37.32445 W95.68569, H: N37.31385 W95.69676.
Site sampling in 2009-2010 was conducted during low summer flows (<4.25 cms). All sample sites were 100 m long and 10 m wide, beginning from the shallow (gravel bar) side of the river and 1 m from shore. Forty 1-[m.sup.2] quadrats were searched at each sample site. Coordinates for quadrat placement were drawn from a random number table. Sample size was based on a previous survey (Cope, 1983) and a pilot study that showed 95% of present species were encountered by subsampling 4% of the sample site area. Each quadrat was hand excavated to about 15-20 cm substrate depth or until bedrock, or a compacted layer of gravel, was reached. Collected mussels were identified and measured before they were returned to the substrate.
Relative abundance (percent of total sample) was calculated for each species from the hearth site. Mean relative abundance was calculated from the eight Verdigris River sample sites. A comparison of the overall distribution of species between the hearth site and the Verdigris River survey were compared with a two-tailed chi-square test. If overall relative abundance distribution between the two samples was found to be significantly different (P [less than or equal to] 0.05), individual species were compared using Fisher's Exact Test for contingency tables on only those species that made up >5% of either sample total. Species richness, evenness, and Shannon diversity indices were also calculated for the hearth site and each river sample site (Odum, 1971).
Shell size differences between the hearth site and the Verdigris River survey were compared using a Kruskal-Wallis statistic but only when sample size for any species attained was [greater than or equal to] seven measurable valves from the hearth site.
The stone hearth and the mussel midden were most likely created during the Late Archaic Period and probably in use during the Early Woodland Period (4000-1000 B.P.), based on radiocarbon dating of stone hearth charcoal and styles of artifacts recovered throughout the midden. A charcoal sample from the hearth site generated a date of 3190 [+ or -] 20 B.P. Nearby charcoal samples were aged at 2790, 2320, 2210, 2175, and 1955 B.P. A variety of chipped stone projectile points and pottery fragments were excavated in association with the mussel midden. All projectile points were classified as Late Archaic and Woodland periods. All pottery fragments were identified as Woodland Period pottery belonging to the Cuesta/Cooper phase of Southeast Kansas and Northeast Oklahoma (Tomasic, 2010, 2012).
Shell material from the hearth site totaled 31 kg. These relic valves were divided into three categories by weight: measurable (5.8%), identifiable fragments (56.5%), and unidentifiable fragments (37.7%). Species was determined for 935 valves or valve fragments. Twenty-four species were represented at the hearth site (Table 1). Five species documented at the hearth site were missing from the modern sample. These were deertoe (Truncilla truncata), rabbitsfoot (Quadrula cylindrica), black sandshell (Ligumia recta), flutedshell (Lasmigona costata), and fatmucket (Lampsilis siliquoidea) (Table 1).
Verdigris River survey sites yielded 4567 live specimens. The mean number of mussels per [m.sup.2] quadrat ranged from 7.8 to 47.6 and the overall mean was 14.3 mussels/[m.sup.2] (SE = 4.78). Twenty living species were represented (Table 1). Two species were present in the Verdigris survey but missing from the hearth site. These were fragile papershell (Leptodea fragilis) and fawnsfoot (Truncilla donaciformis).
There were eight species of mussels from the hearth site that had a relative abundance exceeding 5% (Table 1). These were: threeridge (Amblema plicata), pimpleback (Quadrula pustulosa), Ouachita kidneyshell (Ptychobranchus occidentalis), round pigtoe (Pleurobema sintoxia), Western fanshell, Wabash pigtoe (Fusconia flava), bleufer (Potamilus purpuratus), and monkeyface (Quadrula metanevra). In contrast, the Verdigris River survey had six species with a mean relative abundance >5%. They were: Wabash pigtoe, monkeyface, pimpleback, pistolgrip (Tritogonia verrucosa), round pigtoe, and Ouachita kidneyshell.
Relative abundances at the hearth site differed significantly from the Verdigris River survey (chi square = 511, P < 0.01, 8df). Compared to the proportions in the Verdigris River sample (Fisher's Exact Test), there were more than expected threeridge (P < 0.01), Ouachita kidneyshell (P < 0.01), round pigtoe (P < 0.01), Western fanshell (P < 0.01), and bleufer (P < 0.01) at the hearth site. There were fewer than expected Wabash pigtoe (P < 0.01), monkeyface (P < 0.01), and pistolgrip (P < 0.01) at the hearth site. There was no significant difference in proportion of pimpleback (P = 0.45) between the Verdigris River survey and the hearth site. Mussel richness, diversity, and evenness were slightly greater at the hearth site than in the Verdigris River survey (Table 2).
There were five species from the hearth site that had [greater than or equal to] seven valves intact enough to collect height measurements. These five species were threeridge, Ouachita kidneyshell, round pigtoe, Western fanshell, and Wabash pigtoe. In four of these five species, valves found at the hearth site were significantly smaller than those found in the Verdigris River survey (Table 3).
Of the 63 intact valves measured from the hearth site, there were ten that were either Western fanshell or species harvested at a young enough age to show wide growth bands interrupted by rest zones. These included the following species: Western fanshell (6), round pigtoe (2), Wabash pigtoe (1), and Neosho mucket (Lampsilis rafinesqueana; 1). In these ten valves, the growth zone was wide and without any evident rest ridge at harvest. This observation indicates that the mussels were probably harvested during late summer or fall, near the end of the growing season.
Mussel shell is often found in archaeological sites throughout the Central Plains (Wedel, 1986). The use of mussels as food is well documented (e.g., Parmalee and Klippel, 1974; Wedel, 1986; Parmalee and Bogan, 1998; Blakeslee, 2000). The association of the stone hearth and shells at the Verdigris hearth site makes a strong argument for the use of mussels as food by the Late Archaic-Early Woodland period inhabitants. Putative rock hearths are relatively common archeological features in southeast Kansas and northeast Oklahoma and are referred to in the archeological literature as burned rock features (Thies and Witty, 1992), burned rock complexes (Williams, 1988), and rock concentrations (Baldwin, 1969). These features date primarily to the Late Archaic (Thies and Witty, 1992) and there is general agreement that they were used for food processing (Wandsnider, 1997; Williams, 1988; Adair, 2006; Thoms, 2008).
The midden associated with the Verdigris hearth site is extensive and might have accumulated over decades or even centuries. The mussels were probably cooked in a manner similar to modern clambakes, in which a fire is used to heat stones, and the heated stones are then used to steam or bake the clams (Wandsnider, 1997; Parmalee and Bogan, 1998; Blakeslee, 2000). The scatter of shells is similar to patterns of refuse disposal surrounding hearths of contemporary hunter-gatherers. Shells were discarded in a semicircular pattern away from the use area (Tomasic, 2012). Although the site surface had been subjected to plowing, the hearth and mussel midden were apparently not seriously affected. Spatial patterns in the distribution of artifacts are often well-preserved when located beneath the plowzone (Roper, 1976; Logan and Hill, 2000).
The most productive and inviting time to collect mussels in the Verdigris River is when the air temperature is high and the water flow is low. That time is typically mid-Jul. to mid-Sept. in southeast Kansas. From the limited mussel growth analysis of young shells examined, this was also the season of harvest of mussels by the Late Archaic-Woodland period inhabitants that created the mussel midden.
The smaller size of prehistoric valves at the hearth site compared to mussels currently living nearby is a typical observation at Holocene archeological sites in the midwestern and eastern United States (reviewed by Peacock, 2000). Three explanations are probable. First, archaic mussels might have been smaller if growth rates were slower in the past. Growth rates might have changed as a result of water temperature, stream size, or perhaps modern agricultural runoff have enhanced the microbial food base in streams (reviewed by Peacock, 2000). Second, a higher proportion of young individuals might be expected if recruitment was more frequent and consistent in the past than at present. Age distributions of many modern mussel populations appear to be dominated by older age classes because of infrequent recruitment (Haag, 2012). Third, midden shells might be smaller than living mussels because of harvest effects. Native Americans might have considered smaller mussels more palatable, or they might have intentionally left the larger specimens to serve as parental stock. Conversely, high harvest rates might have removed individuals before they could reach a large size. It should be possible to test the competing hypotheses of growth rates versus harvest effects by comparing mussel size at age, if shell preservation permits measurements at growth annuli.
The large number of species represented at the hearth site and their relative abundances argue that prehistoric harvest was largely indiscriminate. Species richness, diversity, and evenness were greater in the hearth site sample than in the Verdigris River survey (Table 2). Only two modern species are missing from the hearth sample (fragile papershell and fawnsfoot). The absence of fawnsfoot, a very small species, at the hearth site is understandable because this species might have been too small to be significant as food. The absence of fragile papershell, which has very thin valves, could be the result of degradation of these fragile shells.
Of the top ten most abundant species in the hearth-site sample, eight are in the top ten of the modern sample (Table 1). However, eight common species differed significantly in relative abundance between the hearth site and the modern assemblage. Species that were more abundant at the hearth site were threeridge, Ouachita kidneyshell, round pigtoe, Western fanshell, and bluefer. Common species that were relatively less abundant at the hearth site were Wabash pigtoe, monkeyface, and pistol grip (Table 1). These differences could have resulted from selective harvest or from differences between the archaic and modern mussel communities. The shells of Ouachita kidneyshell and related species were apparently favored for use as scraping tools (Parmalee, 1988; Barnhart and Ray, 2006). Western fanshell might have been selectively harvested as food, because it was interred, with meat intact, in burial mounds in southwest Missouri, perhaps as a food source for the deceased (Oesch, 1984). Historical, rather than prehistoric harvest explains the rarity of threeridge in the modern assemblage. Threeridge was the most common mussel at the hearth site and it was the most common mussel in the Verdigris as late as 1979. However, the threeridge population in the Verdigris was decimated by heavy harvest for the cultured pearl industry in the late 1980s and early 1990s (Miller and Mosher, 2008).
Besides harvest effects, changes in availability of fish hosts could be responsible for changes in the mussel community. There is no recorded evidence regarding changes in fish abundance in the Verdigris River over time. However, the trends in mussel abundance may reflect general trends in the fish community from prehistoric to modern times.
Western fanshell and Ouachita kidneyshell were more abundant at the hearth site than in the Verdigris River survey. Both of these species are long-term brooders that use darters (Percidae) as hosts and release benthic conglutinates in the early spring (Barnhart et al., 2008). The apparent decline of these species in the Verdigris River might indicate abundance of darters was greater prehistorically than at present. The two most abundant species in the Verdigris River presently are the Wabash pigtoe and monkeyface. Both were much less common in the hearth-site collection. Both of these species are short-term summer brooders that release their glochidia in summer and both use shiners (Cyprinidae) and not darters as hosts. The Wabash pigtoe releases small pelagic conglutinates, while the monkeyface displays a modified excurrent aperture that acts as a lure (Crownhart et al., 2006; Barnhart et al., 2008). The apparent increase in relative abundance of these shiner-dependent species contrasts with the decrease of the darter specialists.
Bleufer was more abundant and pistolgrip was less abundant at the hearth site than the modern assemblage (Table 1). These species are both host specialists but do not utilize either shiners or darters. Bleufer is a long-term brooder that uses freshwater drum (Aplodinotus grunniens) as its sole host, and pistolgrip is a short-term spring brooder that use catfish (Williams et al., 2008). Bleufer is unusual in that it reaches its northern limit of distribution at about 37[degrees]N latitude, approximately the latitude of the study area, and it may be range-limited by cooler temperatures further north (Haag, 2012).
In addition to the presence of suitable hosts, mussels require suitable substrate composition, suspended food, water quality, and water quantity. Significant changes have occurred in the Verdigris River within a few years' time that may affect population density of a single species (Miller and Obermeyer, 1997) or several species (Miller and Lynott, 2006). Excess nutrients and siltation are major problems affecting mussel abundance in Kansas streams (Angelo et al., 2009). For example, the once-rich mussel fauna of the Cottonwood River in Kansas became depanperate following feedlot runoff in the 1960s (Prophet, 1969).
Changes in stream hydrology due to impoundment construction and operation probably influence mussel community structure (Vaughn and Taylor, 1999). Upstream reservoirs have been identified in reducing the effect of short-term droughts by providing a low base flow rate when there might otherwise have been no flow during that time (Miller and Obermeyer, 1997). Poor water clarity from siltation is often exacerbated by long-term high-flow water release from upstream impoundments after high rain-fall events. The resulting silt-laden water would probably inhibit effectiveness of visual fish lures.
The hearth-site midden, this survey of extant mussels, and previous basin-wide mussel surveys (Cope, 1983; Obermeyer et al., 1997; Miller and Lynott, 2006) provide evidence that at least four species have been extirpated from the Verdigris River in Kansas. These are rabbitsfoot, black sandshell, flutedshell, and fatmucket. Deertoe was also missing from the modern collection but is known to still occur in the Verdigris (Angelo et al., 2009). All of these species are widespread in the Mississippi basin and reach western limits of their ranges in eastern Kansas. Each of them was relatively uncommon in the hearth-site sample (Table 1). We hypothesize that these species persisted in the Verdigris River of Kansas until relatively recent anthropogenic alterations of water quantity and quality occurred. The effects of impoundments, row-crop agriculture, pollution, and droughts of the 1930s and 1950s pushed their populations to levels beyond recovery.
For example, rabbitsfoot has declined range-wide and has been proposed for federal status as threatened (Federal Register, 2012). It is a summer brooder that uses cyprinid fishes as hosts (Fobian, 2007). Females seek shallow water during summer before releasing their glochidia, so they may be particularly vulnerable to water level fluctuations. The only known population of rabbitsfoot remaining in the Verdigris River is downstream of Oologah Reservoir in Oklahoma. The reservoir prevents upstream migration of host fishes infested with rabbitsfoot glochidia, thus precluding recolonization of sites in Kansas. Black sandshell was a common species in rivers south of the Kansas River in the early 20th century (Scammon, 1906) and relic shells of black sandshell can still be found at sites throughout the Verdigris and Neosho systems (Obermeyer et al., 1997). However, the species is apparently extirpated in Kansas. The primary hosts of black sandshell are walleye and sauger (Khym and Layzer, 2000). Reduced numbers of these host species in the Verdigris River (Cross and Collins, 1995) may have played a role in the extirpation of black sandshell from Kansas.
Seven other species that were rare at the hearth site persisted in the Verdigris to the present day. Among these is Neosho mucker, which is under review for federal listing as endangered (Federal Register, 2012). Neosho mucket is a long-term brooder and utilizes black basses (Micropterus spp.) as hosts (Barnhart and Roberts, 1997). Although Neosho mucket is a dominant species at some sites in the Spring River in Kansas and Missouri, it was rare both at hearth-site and in the modern Verdigris River samples, suggesting that its low abundance in the Verdigris may be a "normal" condition. Neosho mucket has been lost in the relatively recent past from the majority of sites where it occurred in Kansas, as evidenced by relic shells (Obermeyer et al., 1997; Obermeyer, 1999). If populations were historically small they might have been vulnerable to local extirpation, and the modern loss of metapopulation connectivity as a result of reservoirs and channel modifications makes recolonization difficult or impossible by restricting movement of host fish (Watters, 1996; Dean et al., 2002).
The fact that most of the unionid mussel assemblage has persisted in the Verdigris River over the last few millennia is remarkable and is attributable to their resilience to human impacts and watershed alterations. It also points to the fact that the Verdigris River watershed remains relatively healthy regarding riverine life.
Acknowledgments.--We thank the numerous volunteers who assisted in mussel surveys and archeological excavations. We are grateful to landowners for allowing access to their land, especially Dr. Phil Eastep and the Dan Small family. We thank Tom Mosher, Ben Neely, Lina Miller, the anonymous reviewers and editors who suggested many useful comments to improve this manuscript.
ADAIR, M. J. 2006. Paleoethnobotanical research in Kansas, p. 248-263. In: R. J. Hoard and W. E. Banks (eds.). Kansas Archaeology. University Press of Kansas in association with the Kansas State Historical Society, Lawrence.
ANGELO, R. T., M. S. CRINGAN, E. HAYS, C. A. GOODRICH, E. J. MILLER, M. A. VANSCOYOC, AND B. R. SIMMONS. 2009. Historical changes in the occurrence and distribution of freshwater mussels in Kansas. Great Plains Res., 19:89-126.
BALDWIN, J. 1969. The Lawrence Site, Nw6: A Nonceramic site in Nowata County, Oklahoma, p. 67-118. In: D. G. Wyckoff (ed.). Bulletin of Oklahoma. Anthropological Society. Vol. XVIII. Oklahoma Anthropological Society, Oklahoma City.
BARNHART, M. C. AND A. D. ROBERTS. 1997. Reproduction and fish hosts of unionids from the Ozark Uplifts. In: K. S. Cummings, A. C. Buchanan and L. M. Koch (eds.). Conservation and management of freshwater mussels It. Proceedings of a UMRCC symposium, 16-18 Oct. 1995, St. Louis, Missouri. Upper Mississippi River Conservation Committee, Rock Island, Illinois.
--AND C. H. RAY. 2006. Mussel shells from Henson Cave and Henson Shelter. Appendix 3, p. 226-229. In: J. H. Ray and M. D. Conner (eds.). Archeological investigations at six shelter sites in southern McDonald County, Missouri. Special Publication No. 5. Center for Archeological Research, Missouri State University, Springfield.
--, W. R. HAAG, AND W. N. ROSTON. 2008. Adaptations to host infection and larval parasitism in Unionoida. J. N. Am. Benthol. Soc., 27(2):370-394.
BLAKESLEE, D. J. 2000. Mussels, bison kills, and pots: clarity in the archeological record. Cent. Plains Archeol., 8 (1):5-11.
COPE, C. H. 1983. Kansas freshwater mussel investigation, NMFS Project 2-378-R. Unpublished report, Kansas Fish and Game Commission, Pratt. 39 p.
CROSS, F. B. AND J. T. COLLINS. 1995. Fishes in Kansas (2nd ed.). University of Kansas Natural History Museum, Lawrence. 315 p.
CROWNHART, A., B. SIETMAN, M. HOVE, AND N. RUDH. 2006. Quadrula metanevra glochidia metamorphose on select minnow species. Ellipsaria, 8:6-7. Available at: http://ellipse.inhs.uiuc.edu/fmes/.
DEAN, J., D. EDDS, D. GILLETTE J. HOWARD, S. SHERRADEN, AND J. TIEMANN. 2002. Effects of lowhead dams on freshwater mussels in the Neosho River, Kansas. Trans. Kans. Acad. Sci., 105:232-240.
FEDERAL REGISTER. 2012. Endangered and threatened wildlife and plants; proposed endangered status of the Neosho mucker, threatened status of the rabbitsfoot, and designation of critical habitat for both species. Fed. Reg., 77:63440-63536.
FOBIAN, T. B. 2007. Reproductive biology of the rabbitsfoot mussel (Quadrula cylindrica) (Say, 1817) in the upper Arkansas River system, White River system, and the Red River system. M.S. Thesis, Missouri State University, Springfield. 104 p.
--. 2012. North American freshwater mussels: natural history, ecology and conservation. W. H. Cambridge University Press, Cambridge, U.K. 505 p.
KHYM, J. R. AND J. B. LAYZER. 2000. Host fish suitability for glochidia of Ligumia recta. Am. Midl. Nat., 143:178-184.
KLIPPEL, W. E., G. CELMER, AND J. R. PERDUE. 1978. The Holocene naiad record at Rodgers Shelter in the western Ozark Highland of Missouri. Plains Anthrop., 23(82):257-271.
JUSTICE, N. D. 1987. Stone Age spear and arrow points of the midcontinental and eastern United States. Indiana University Press, Bloomington. 288 p.
LOGAN, B. AND M. HILL. 2000. Spatial analysis of small scale debris from a Late Prehistoric site in the Lower Missouri Valley, Kansas. J. Field Archaeol., 27:241-256.
MARSHALL, J. O. 1972. The Archeology of the Elk City Reservoir: A local archeological sequence in Southeast Kansas. Kansas State Historical Society Anthropological Series Number 6, Topeka. 269 p.
MILLER, E. J. AND B. K. OBERMEVER. 1997. Population increase of Quadrula metanevra in southeast Kansas, p. 30-36. In: K. S. Cummings, A. C. Buchanan, C. A. Mayer and T. J. Naimo (eds.). Conservation and management of freshwater mussels II: Initiatives for the future, Proceedings of 2nd UMRCC Symposium, 16-18 Oct. 1995, St. Louis, Missouri. Upper Mississippi River Conservation Committee, Rock Island, Illinois.
--AND S. T. LYNOTT. 2006. Increase of unionid mussel populations in the Verdigris River, Kansas, from 1991 to 2003. Southeast. Nat., 5(3):383-392.
--AND T. D. MOSHER. 2008. Commercial harvest and status of a freshwater mussel (threeridge Amblema plicata) in Kansas. Trans. Kans. Acad. of Sci., 111:118-124.
NEVES, R. J. 1997. Keynote address: a national strategy for the conservation of native freshwater mussels, p. 1-10. In: K. S. Cummings, A. C. Buchanan, C. A. Mayer and T. J. Naimo (eds.). Conservation and management of freshwater mussels II: Initiatives for the future, Proceedings of 2nd UMRCC Symposium, 16-18 Oct. 1995, St. Louis, Missouri. Upper Mississippi River Conservation Committee, Rock Island, Illinois.
OBERMEVER, B. K., D. R. EDDS, C. W. PROPHET, AND E. J. MILLER. 1997. Freshwater mussels (Bivalvia: Unionidae) in the Verdigris, Neosho, and Spring River basins of Kansas and Missouri, with emphasis on species of concern. Am. Malacol. Bull., 14(1):41-55.
--. 1999. Recovery plan for four freshwater mussels in southeast Kansas: Neosho mucket (Lampsilis rafinesqueana), Ouachita kidneyshell (Ptychobranchus occidentalis), rabbitsfoot (Quadrula cylindrica), and western fanshell (Cyprogenia aberti). Kansas Department of Wildlife & Parks, Pratt. 52 p.
ODUM, E. P. 1971. Fundamentals of ecology (3rd ed.). W.B. Saunders, Co, Philadelphia. 574 p.
OESCH, R. D. 1984. Missouri naiades: A guide to the mussels of Missouri. Missouri Department of Conservation, Jefferson City. 271 p.
PARMALEE, P. W. AND W. E. KLIPPEL. 1974. Freshwater mussels as a prehistoric food resource. Am. Antiq., 39 (3):257-271.
--, --. AND A. E. BOGAN. 1980. Notes on the prehistoric and present status of the naiad fauna of the Middle Cumberland River, Smith County, Tennessee. The Nautilus, 94(3):93-105.
--. 1988. A comparative study of late Prehistoric and Modern molluscan faunas of the Little Pigeon River system, Tennessee. Am. Malacol. Bull., 6(2):165-178.
--AND A. E. BOGAN. 1998. The Freshwater Mussels of Tennessee. The University of Tennessee Press, Knoxville. 328 p.
PEACOCK, EVAN. 2000. Assessing bias in shell assemblages. J. Field Archeo., 27:183-196.
PROPHET, C. W. 1969. River pollution bv feedlot runoff. Proc. Okla. Acad. Sci., 48:207-209.
ROPER, D. C. 1976. Lateral displacement of artifacts due to plowing. Am. Antiq., 41(3):372-375.
SCAMMON, R. E. 1906. The Unionidae of Kansas, Part 1: an illustrated catalogue of the Kansas Unionidae. Univ. Kans. Sci. Bull., 3:279-373.
THIES, R. AND T. WITTY. 1992. The archaic of the central plains. Revista de Arqueologia Americana, 5:137-165.
THOMS, A. V. 2008. The fire stones carry: ethnographic records and archaeological expectations for hot-rock cookery in western North America. J. Anthrop. Archaeo., 27:443-460.
TOMASIC, J. J. 2010. The Eastep site revealed. Kans. Preservation, 32(4):8-14.
--. 2012. Phase IV excavations of 14MY388 at the Eastep riverbank stabilization project, an NRCS undertaking in Montgomery County, Kansas. Report submitted to the Natural Resources Conservation Service, Salina, Kans. Copies available from the Kansas SHPO, Topeka. 142 p.
UNITED STATES GEOLOGICAL SURVEY. 2000. Estimation and comparison of potential runoff-contributing areas in Kansas using topographic, soil, and land-use information. Report No. 00-4177. Lawrence. 55 p.
VAUGHN, C. C. AND C. M. TAYLOR. 1999. Impoundments and the decline of freshwater mussels: A case study of an extinction gradient. Cons. Biol., 13:912-20.
WANDSNIDER, L. 1997. The roasted and the boiled: food composition and heat treatment with special emphasis on pit-hearth cooking. J. Anthrop. Archaeo., 16:1-48.
WATTERS, G. T. 1996. Small dams as barriers to freshwater mussels (Bivalvia, Unionoida) and their hosts. Biol. Cons., 75:79-85.
WEDEL, W. R. 1986. Central plains prehistory: Holocene environments and culture change in the Republican River Basin. University of Nebraska Press, Lincoln. 280 p.
WILLIAMS, B. G. 1988. 14CT347: A burned rock complex of Southeastern Kansas, Kansas State Historical Society Contract Archeology Publication 6. Kansas State Historical Society, Topeka. 123 p.
WILLIAMS, J. D., M. L. WARREN, JR., K. S. CUMMINGS, J. L. HARRIS, AND R. J. NEVES. 1993. Conservation status of freshwater mussels in the United States and Canada. Fisheries, 18(9):6-22.
--, A. E. BOGAN, AND J. Y. GARNER. 2008. Freshwater mussels of Alabama & the Mobile Basin in Georgia, Mississippi & Tennessee. University of Alabama Press, Tuscaloosa. xv + 908 p.
SUBMITTED 1 NOVEMBER 2012
ACCEPTED 30 SEPTEMBER 2013
EDWIN J. MILLER (1)
Kansas Department of Wildlife, Parks, and Tourism, 5089 C.R. 2925, Independence 67301
JOHN J. TOMASIC
Department of Anthropology, University of Kansas, Lawrence 66047
M. CHRISTOPHER BARNHART
Biology Department, Missouri State University, Springfield 65897
(1) Corresponding author: email: email@example.com
TABLE 1.--A comparison of relative abundances between the hearth-site excavation and the 2009 2010 Verdigris River mussel survey (Mean and SD from eight sample sites). Species with relative abundances of >5% were compared using a Fisher's exact test (significant differences are indicated by > or < sign). The ten highest rankings are indicated within parenthesis. Species presumed extirpated are indicated by (X) Number (rank) Hearth Verdigris Species site river Threeridge (Amblema plicata) 157 (1) 74 (10) Pimpleback (Quadrula pustulosa) 125 (2) 563 (3) Ouachita kidneyshell 120 (3) 219 (6) (Ptychobranchus occidentalis) Round pigtoe (PLurobema sintoxia) 95 (4) 310 (5) Western fanshell (Cyprogenia 84 (5) 125 (7) aberti) Wabash pigtoe (Fusconaia (lava) 81 (6) 1336 (1) Bleufer (Potamilus purpuratus) 73 (7) 11 Monkeyface (Quadrula metanevra) 72 (8) 1088 (2) Pistolgrip (Tritogonia verrucosa) 25 (9) 457 (4) Mapleleaf (Quadrula quadrula) 19 (10) 43 Deertoe (Truncilla truncata) 15 0 Wartyback (Quadrula nodulata) 10 98 (8) Rabbitsfoot (Quadrula cylindrica) 10 0 (X) Threehorn wartyback (Obliquaria 9 88 (9) reflexa) Creeper (Strophitus undulatus) 8 14 Washboard (Megalonaias nervosa) 6 31 Butterfly (Ellipsaria lineolata) 5 14 Black sandshell (Ligumia recta) 5 0 (X) Fluted shell (Lasmigona costata) 5 0 (X) Neosho mucket (Lampsilis 4 7 rafinesqueana) Plain pocketbook (Lampsilis 2 38 cardium) Fragile papershell (Leptodea 0 26 fragilis) Fat mucket (Lampsilis siliquoidea) 2 0 (X) White heelsplitter (Lasmigona 2 11 complanata) Yellow sandshell (Lampsilis teres) 1 5 Fawnsfoot (Truncilla donadformis) 0 9 TOTAL 935 4567 Relative abundance (%) Hearth Verdigris river Species site (mean [+ or -] SD) Threeridge (Amblema plicata) 16.8 > 1.3 [+ or -] 0.7 Pimpleback (Quadrula pustulosa) 13.4 ns 11.0 [+ or -] 4.1 Ouachita kidneyshell 12.8 > 6.3 [+ or -] 2.7 (Ptychobranchus occidentalis) Round pigtoe (PLurobema sintoxia) 10.2 > 7.9 [+ or -] 4.9 Western fanshell (Cyprogenia 9.0 > 3.9 [+ or -] 2.8 aberti) Wabash pigtoe (Fusconaia (lava) 8.7 < 20.6 [+ or -] 11.1 Bleufer (Potamilus purpuratus) 7.8 > 0.3 [+ or -] 0.3 Monkeyface (Quadrula metanevra) 7.7 < 26.9 [+ or -] 9.8 Pistolgrip (Tritogonia verrucosa) 2.7 < 12.7 [+ or -] 4.5 Mapleleaf (Quadrula quadrula) 2.0 0.6 [+ or -] 0.5 Deertoe (Truncilla truncata) 1.6 0 Wartyback (Quadrula nodulata) 1.1 2.4 [+ or -] 0.7 Rabbitsfoot (Quadrula cylindrica) 1.1 0 Threehorn wartyback (Obliquaria 1.0 2.3 [+ or -] 1.1 reflexa) Creeper (Strophitus undulatus) 0.9 0.4 [+ or -] 0.2 Washboard (Megalonaias nervosa) 0.6 0.6 [+ or -] 0.6 Butterfly (Ellipsaria lineolata) 0.5 0.3 [+ or -] 0.2 Black sandshell (Ligumia recta) 0.5 0 Fluted shell (Lasmigona costata) 0.5 0 Neosho mucket (Lampsilis 0.4 0.1 [+ or -] 0.1 rafinesqueana) Plain pocketbook (Lampsilis 0.2 1.0 [+ or -] 0.7 cardiurn) Fragile papershell (Leptodea 0 0.7 [+ or -] 0.7 fragilis) Fat mucket (Lampsilis siliquoidea) 0.2 0 White heelsplitter (Lasmigona 0.2 0.3 [+ or -] 0.5 complanata) Yellow sandshell (Lampsilis teres) 0.1 0.1 [+ or -] 0.2 Fawnsfoot (Truncilla donadformis) 0 0.2 [+ or -] 0.2 TOTAL 100 100 TABLE 2.--Mussel community metrics comparing the hearth site and the Verdigris River survey Metric Hearth Verdigris Verdigris sites site total sample mean [+ or -] SD, n = 8 Species Richness 24 20 18 [+ or -] 2.9 Shannon Diversity 2.5 2.1 2.1 [+ or -] 0.2 Index Evenness 0.79 0.68 0.72 [+ or -] 0.07 TABLE 3.--Mussel height comparison between valves from the hearth site and those sampled at eight Verdigris River sites in 2009-2010. Only species from the hearth site with [greater than or equal to] seven measurable specimens were compared Species Hearth site([+ or -] SD) Threeridge 50.3 [+ or -] 9.8 (12) Ouachita kidneyshell 39.4 [+ or -] 6.7 (8) Round pigtoe 52.0 [+ or -] 1.8 (7) Western fanshell 43.5 [+ or -] 6.7 (13) Wahash pigtoe 40.1 [+ or -] 5.6 (7) Species Verdigris river([+ or -] SD) Kruskal-Wallis Threeridge 72.2 [+ or -] 13.5 (74) 19.48 Ouachita kidneyshell 46 [+ or -] 7.5 (219) 5.9 Round pigtoe 57.1 [+ or -] 9.7 (310) 2.57 Western fanshell 53.7 [+ or -] 8.2 (125) 14.43 Wahash pigtoe 52.4 [+ or -] 8.9 (1336) 12.86 Species P-value Threeridge <0.01 Ouachita kidneyshell 0.02 Round pigtoe 0.11 Western fanshell <0.01 Wahash pigtoe <0.01
|Printer friendly Cite/link Email Feedback|
|Author:||Miller, Edwin J.; Tomasic, John J.; Barnhart, M. Christopher|
|Publication:||The American Midland Naturalist|
|Date:||Jan 1, 2014|
|Previous Article:||Population demographics and life history of the round Hickorynut (Obovaria subrotunda) in the Duck River, Tennessee.|
|Next Article:||Historical changes in fish assemblage structure in Midwestern nonwadeable rivers.|