Variation in thermal tolerance among three Mississippi river populations of the zebra mussel, Dreissena polymorpha.

ABSTRACT To investigate the occurrence of thermal adaptation in the zebra mussel, Dreissena polymorpha, inhabiting the lower reaches of the Mississippi River, we compared lethal heat-tolerance among three populations (mussels collected at Lake Pepin, MN; Alton, IL; and Baton Rouge, LA). We determined time-to-death at 32[degrees]C for 160 individuals per site, for mussels collected at a water temperature of 15[degrees]C and then maintained in the laboratory for about 8 wk under uniform conditions. Both shell length and condition index significantly affected survival time and were included as covariates in the analysis for interpopulation differences in heat-tolerance. Zebra mussels from our southernmost location had a higher heat-tolerance than those from the two northern locations. This difference in heat-tolerance among sites may indicate adaptation to local temperature regimes. In addition, in a comparison of heat tolerance within populations, we separated mussels into size classes (where larger mussels have been exposed to local conditions longer) and calculated an adjusted mean time-to-death (TTD). We found a different TTD/size relationship depending on sampling location. Minnesota mussels had decreasing heat tolerance as size increased, where Louisiana mussels had the opposite relationship. These patterns of heat-tolerance within populations indicate a selection pressure for increased heat-tolerance at Louisiana. However, even if the selection pressure is strong at the Louisiana site, it has not (at least not yet) resulted in an adaptation, as high heat-tolerance is not ubiquitous within this population. Zebra mussels may have insufficient genetic variation for heat-tolerance or gene flow may have been too strong for genetic adaptation to occur in the short amount of time that zebra mussels have occurred in the lower Mississippi River.

KEY WORDS: water temperature, heat, adaptation, zebra mussel, Mississippi River

INTRODUCTION

Zebra mussels, Dreissena polymorpha, were introduced into Lake St. Clair in 1988 and most likely arrived there in ballast water of transoceanic shipping (Ludyanskiy et al. 1993). These mussels have a 2-wk planktonic veliger larval stage, whereas adults attach to hard substrate via byssal threads. Currents, boats, and even waterfowl can thus transport veligers or adults into new habitats. Zebra mussels have spread into all of the Laurentian Great Lakes and the Mississippi River drainage in the last 15 y. Because zebra mussel populations are now distributed throughout a large part of North America, they experience a variety of local conditions.

One environmental factor that differs substantially along the North American range of the zebra mussel is water temperature. For example, a large difference in water temperature exists between Lake Superior and the lower reaches of the Mississippi River. The high summer temperatures in the southern US are likely to have an impact on zebra mussels. Temperatures above 25[degrees]C have a variety of negative effects on zebra mussels, including reduced growth (Thorp et al. 1998), reduced foot activity and byssus thread production (Rajagopal et al. 1997), and increased respiration (Alexander et al. 1994). The latter study also showed that zebra mussels have little capacity for metabolic temperature acclimation (MTA). However, the mussels used in the laboratory experiment that yielded this result may have been affected by handling, because MTA is higher for mussels maintained in their natural environment (Stoeckmann & Garton 1997, Stoeckmann 2003). While zebra mussels can survive short periods (a few hours) of exposure to temperatures as high as 39[degrees]C, the long-term lethal limit for zebra mussels appears to be approximately 30[degrees]C (Iwanyzki & McCauley 1993, Spidle et al. 1995, McMahon 1996, Thorp et al. 1998).

Because high water temperature has a negative effect on zebra mussels, it acts as a selective agent and could potentially drive genetic adaptation in zebra mussel populations inhabiting areas with relatively high temperatures. Geographic variation in heat-tolerance has been reported among zebra mussel populations from sites along the Volga River (Smirnova et al. 1992) and among North American populations (Thorp et al. 1998), although results from North America were not conclusive. It also appears that North American zebra mussels have a higher heat-tolerance than their North European conspecifics (Rajagopal et al. 1997). McMahon (1996) reported that heat-tolerance among zebra mussels from different sites was positively correlated with temperature at the mussel collection site and argued that it was most likely a consequence of long-term "acclimatization," or nongenetic differences among populations.

As defined, adaptation differs from acclimatization in that the former is a genetic change over time as a result of selection pressure. As of yet, there is no evidence that the observed differences in heat-tolerance have a genetic basis and are thus due to adaptation. For a set of European populations, Fetisov et al. (1991) showed distinct allozyme differences between zebra mussels at a control site and others in a thermal discharge, but thermal tolerance was not quantified in these populations. In a previous study, we also found a shift in allele frequency at the leucine aminopeptidase (Lap) locus along the latitudinal gradient formed by the Mississippi River (Elderkin et al. 2001). A similar allele shift was found among populations of the marine mussel Mytilus at the Lap locus (Koehn et al. 1980). In later experiments, selection due to different local environmental conditions was implicated for the latter allele frequency shift (Hilbish & Koehn 1985). Identifying selection on a particular allele is difficult and involves many experimental steps (Mitton 1997). One approach commonly used to investigate the genetic nature of environmental tolerances is to determine if the tolerances are still present after maintaining populations for several generations under uniform conditions in the laboratory, thus keeping effects from acclimatization to a minimum (see Klerks & Levinton 1989). So far, this has not been possible for zebra mussels.

The approach used here as a first step for determining whether temperature adaptation has occurred in the most southern part of the zebra mussel range in North America, was a comparison of heat-tolerance among populations from three sites along the Mississippi River. A lack of tolerance differences would exclude the presence of genetic differences in tolerance, whereas the presence of tolerance differences could be followed up at a later stage by research specifically addressing the genetic basis of these differences. To reduce nongenetic variation, we used mussels collected at different latitudes but collected at the same temperature and compared heat-tolerance after mussels had been kept at a common temperature for about 8 wk. The Mississippi River main channel extends along a north-south gradient with resulting differences in water temperature. Over the period 1995 to 2001, surface water temperatures in Louisiana were on average 3[degrees]C higher than in Illinois and 5[degrees]C higher than in Minnesota (Fig. 1). In addition to comparing heat-tolerance among the three populations, we investigated whether heat-tolerance was different among size classes within each population for further insight into the presence of selection pressure for this trait.

[FIGURE 1 OMITTED]

MATERIALS AND METHODS

Collection

Zebra mussels were collected from three locations along the Mississippi River (Fig. 2): Lake Pepin, Minnesota (latitude 44.47 N., longitude 92.29 W.); Alton, Illinois (latitude 38.90 N., longitude 90.15 W.); and Baton Rouge, Louisiana (latitude 30.45 N., longitude 91.13 W). We collected mussels at each location by scraping a basket along the sides of floating docks. They were collected approximately 1-2 m from the surface of the water. We collected mussels when the ambient river temperature reached 15[degrees]C; at this temperature mussels reach their highest body weight prior to spawning (Allen et al. 1999). Actual collection dates in April and May differed by location. Mussels collected from Minnesota and Louisiana represented a wide size range, whereas only a limited size range was available at Illinois. In prior years, a wide size range was available at the Illinois site (personal observation), but in the summer prior to collection the population crashed and only the fall cohort was available for collection.

[FIGURE 2 OMITTED]

Holding Conditions

After collection, mussels were transported immediately to Lafayette, Louisiana. Mussels were maintained there in 40-L aquariums and temperatures were lowered to 5[degrees]C over a period of 2 days (5[degrees]C [day.SUP.-1]). Mussels were then maintained at this low temperature, reducing the metabolic expenditure for the mussels, feeding requirements, and stress from build-up of waste products in the aquariums. Both N[H.sub.3] and N[O.sub.3] were monitored weekly using Lamotte test kits, and water was changed when these nitrogenous compounds reached detectable levels. The mussels were kept at 5[degrees]C for approximately 4 wk.

Feeding

We maintained mussels on a diet of 4-6 mg [mussel.sup.-1] [day.sup.-1] of a commercially available diatom suspension (Diet C, Coast Sea Food Co.), which has been used successfully in other experiments (Stoeckmann & Garton 1997, Madon et al. 1998). We fed the mussels daily, except during times when temperatures were at 5[degrees]C or above 32[degrees]C. Mussels can be maintained at low temperatures with no feeding because they are near their lowest metabolism at this temperature (Schneider 1992).

Quantification of Heat-tolerance

We randomly selected individuals from each site after they had been at 5[degrees]C for about 4 wk. We removed the test individuals, scrubbed their shells to remove any epifauna, and placed them in the test environment described later. When choosing test individuals, we tried to represent an equal number from each shell length size class (grouped into 5-mm intervals). All individuals were between 10-40 mm. However, there was only one size class (10-15 mm) present in the Illinois sample. We allowed mussels to naturally attach themselves to the inside of the container during an acclimation period of 14 days.

We designed a common-garden experiment to test for differences in heat-tolerance among the three Mississippi River populations. We put 80 individuals per site inside each of two aquariums placed in an environmental chamber. We placed batches of 20 individuals from a site in a separate 9 x 9 x 4-cm mesh box (with 3-mm mesh openings). Each aquarium held four boxes from each site, with strong water circulation and aeration ensuring that all containers in an aquarium experienced identical conditions.

We raised the temperature inside the chamber 1[degrees]C [day.sup.-1] until the water temperature reached 22[degrees]C. The mussels remained at this new temperature for 14 days (following McMahon & Ussery 1995) and we then increased the temperature 1[degrees]C [day.sup.-1] until the water temperature reached 32[degrees]C. Prior to the water reaching this temperature, individuals were monitored daily and any individual that expired during acclimation was removed. When the water reached 32[degrees]C, we examined every individual at 3-h intervals until all individuals expired. An individual was considered dead when it could no longer completely close its valves, even with gentle prodding. We immediately removed dead individuals from the aquariums, measured their shell length, and determined their condition index.

Condition Index

Condition index (CI) is a measure of health in bivalves and has been used extensively in the oyster Crassostrea virginica (Hopkins 1949). Under favorable conditions, a bivalve will maximize the amount of tissue inside its shell. Condition index is a ratio of the estimated shell volume to the dry soft tissue mass inside the shell (Hopkins 1949). We had previously noted that zebra mussels from these sites differed consistently in condition index (unpublished data), indicating that the different local environmental conditions may bring about differences in health status (and consequently heat-tolerance) at the time of collection. As we were interested in genetic differences in heat-tolerance rather than nongenetic ones, we quantified the mussels CI and used this as a covariate when analyzing for differences in heat-tolerance. We determined the CI for an individual after it expired by blotting it to remove excess water, determining its shell plus soft tissue mass, removing the tissue from the shell, weighing the shell, drying the tissue overnight at 65[degrees]C, and then determining the mass of the dried tissue. We calculated the CI with the following equation (Hopkins 1949, Lawrence & Scott 1982):

   CI = dw x 100 / (ww-sw),

   where

   dw = dry soft tissue mass (g),
   ww = shell + wet soft tissue mass (g), and
   sw = shell mass (g).

This assumes that the liquid and tissue inside the shell of oysters equals 1 g [cm.sup.-3] and that the mass of the total material inside the shell is an accurate estimate of the volume of that shell (Lawrence & Scott 1982).

Data Analysis

Statistical analyses were conducted using SAS Version 7 or StatView 5.0 (both SAS Institute, Cary, North Carolina). Analysis of variance was used to compare shell length and condition index among mussels from the different collection sites and followed by Bonferroni/Dunn posthoc pairwise comparisons using a Bonferroni--adjusted alpha value of 0.0167. Correlation analysis was used to determine whether condition index and length were related (to assess the extent to which these are independent covariates). Survival analysis was used to determine the presence of differences in time-to-death (TTD) among populations. We used the SAS LIFEREG procedure to analyze the differences in TTD among populations using population, condition index, and length of each mussel as variables. The LIFEREG procedure used an accelerated failure time model to analyze the effect of multiple continuous (CI and shell length) and categorical (population) variables on TTD. We used a Weibull distribution in our model because of our data's fit to a straight line for the log (-log [1-survival]) versus log time estimates, which indicated the appropriateness of that distribution (Dixon & Newman 1991).

In a second analysis, we determined mean TTD separately for (5-mm wide) size classes within our northern (Minnesota) and southern (Louisiana) sampling locations. In this case, we used size class as a surrogate for age, where larger mussels are, on average, older (and have been exposed to heat-stress more often) than smaller mussels. To adjust for differences in condition index among individuals, we calculated each mussel's time-to-death that would be expected if it had a CI of 4, using the model obtained in our previous survival analysis (see Results section) with adjusted TTD = TTD x [e.sup.((4-CI) x .049)]. We then calculated the mean ([+ or -] 95% confidence interval) CI-adjusted TTD for each size class within a population and then compared results among populations. We did not use the Illinois population for this analysis, because only one size class was present here. The 30-35 mm class was present in the Minnesota population, but not in Louisiana.

RESULTS

Condition index and shell length differed among the three populations, with both mean shell length and mean condition index being lowest in the Illinois population and highest for the Minnesota zebra mussels (Fig. 3). Therefore, we included these variables as covariates when comparing heat-tolerance among the three populations. Correlation analysis revealed a significant correlation between these two covariates (r = 0.40, P < 0.0001), showing that these variables were not fully independent. We nevertheless used both variables in the survival analysis model because the correlation was relatively weak, both had a significant effect in the model (Table 1), and because the two covariates did not influence each other's effect in the model.

[FIGURE 3 OMITTED]

Survival analysis of our data from the exposure of mussels to a lethal temperature showed that zebra mussels from our southernmost location (Louisiana) had a significantly higher TTD than mussels from the Minnesota and Illinois locations (Table 1). In the survival analysis as used here (that includes a categorical variable such as collection site), one of the categories is used as the base one and compared with the others. The results in Table 1 used the Louisiana population as the base one and thus yield a comparison of the Louisiana population to the Illinois and Minnesota populations (but does not compare the Illinois and Minnesota populations to each other). Running the accelerated failure time model with a different population serving as the basis for comparison showed that the Minnesota and Illinois populations did not differ from each other in their heat-tolerance ([X.sup.2] = 3.255, P = 0.0712).

CI-adjusted time-to-death differed among size classes and between the Minnesota and Louisiana populations (Fig. 4). As also indicated by the earlier survival analysis, TTD tended to decrease with mussel size and to be higher in the Louisiana population but differences among populations were size-dependent. TTD did not differ among populations for the two smallest size classes (where all 95%-confidence limits overlapped) and was distinctly different among populations only for the 20-25 mm size class. The Minnesota mussels generally showed a decreased TTD with increasing size, whereas the Louisiana population showed a highest TTD at an intermediate size class (the 20-25-mm group).

[FIGURE 4 OMITTED]

DISCUSSION

The survival analysis revealed phenotypic differences in heat-tolerance between the southern-most population studied and two populations further north in the Mississippi River. We have no direct evidence of a genetic component for these resistance differences, thus they may have resulted from either acclimation or adaptation. Because the water temperature for the period 1995 to 2001 averaged 3[degrees]C to 5[degrees]C higher at the Louisiana location than at the other two locations, local thermal adaptation may have occurred. It has been shown previously that a 5[degrees]C difference in water temperature can cause a selection pressure strong enough to result in genetic differentiation (Dahlhoff & Somero 1993, Lin & Somero 1995).

Local adaptation is retarded by gene flow. In this case, where populations are present in a riverine system and where the organism has a planktonic larval stage, gene flow is expected to be substantial (Stoeckel et al. 1997). This is exacerbated by the recruitment situation in rivers (Horvath et al. 1996), where planktonic larvae do not recruit to their own population because offspring are carried downstream by the current. Thus, selection has to be very strong in order for genetic differentiation to be possible. However, a similar situation occurs for open populations of marine mussels with the same reproductive strategy (Gilg & Hilbish 2003) where larvae are carried away via currents. And genetic differentiation has been reported in such situations (e.g., in Mytilus edulus, Hilbish & Koehn 1985). Also, our earlier allozyme survey (Elderkin et al. 2001) showed that genetic differentiation is possible among Mississippi River zebra mussels despite high gene flow.

Additional insight into the occurrence of adaptation and the presence of selective pressure for increased temperature tolerance was obtained by comparing heat-tolerance among size classes within a population. In this comparison, the results indicate that two processes may have been responsible for the size-TTD pattern. Firstly, there seemed to be a general decrease in TTD with increasing size (i.e., bigger individuals were less heat-tolerant). This was evident among the Minnesota individuals and it was the overall pattern identified in the survival analysis. A negative relationship between heat-tolerance and zebra mussel size was also reported by McMahon (1996). Secondly, for the Louisiana population, there appeared to be strong selective pressure by water temperature, with only the most heat-tolerant individuals surviving to the 20-25 mm size class (as evidenced by a high heat-tolerance in this size class). The lack of an increased temperature tolerance in the smaller Louisiana size classes is inconsistent with the occurrence of adaptation; adaptation would have resulted in an increased heat-tolerance in all size classes of the Louisiana population. Therefore, it does seem that there is a strong selection pressure for an increased heat-tolerance in the southern range of the zebra mussel but this has not (at least not yet) resulted in an adaptation. The lack of adaptation at this stage may be due to gene flow being too strong or due to insufficient genetic variation for selection to act on. Our research on the heritability of heat-tolerance indicates that there may indeed be insufficient genetic variation for heat-tolerance in Mississippi River zebra mussels (Elderkin et al. 2004). Our results are consistent with McMahon's view (2002) that the zebra mussel owes its success as an invader to its r-selected nature rather than its inherent tolerance to environmental extremes or an ability to rapidly acclimatize to environmental extremes. This may then mean that marginal habitats, such as the extreme southern part of the Mississippi River, may be maintained by the influx of veligers from more optimal habitats rather than by adaptation to the stressful conditions.

TABLE 1.
Results of the SAS LIFEREG survival analysis procedure for
time-to-death (TTD) of zebra mussels from three locations exposed
to a lethal temperature. This analysis determined the effect of
sampling location on TTD using condition index and size as
covariates. Locations are listed by standard state abbreviation. This
analysis assumed a Weibull distribution.

                         [Beta]     SE of     [chi
Variable          df   estimate *   [Beta]   square]       P

Intercept         1      4.502      0.077    3416.96   <0.0001
Location (all)    2                            16.38    0.0003
LA vs. MN         1     -0.081      0.037       4.80    0.0285
LA vs. IL         1     -0.163      0.042      15.39   -0.0001
Condition Index   1      0.049      0.009      29.26   -0.0001
Length            1     -0.012      0.004      10.11    0.0015
Scale             1      0.284      0.011

* The ([Beta]s are the exponents in the survival analysis regression
model that reflect the influence of each variable, while the
intercept and scale parameters are Weibull distribution parameters.

ACKNOWLEDGMENTS

The authors thank Y. C. Allen, M. L. Beaulne, M. Lemke, J. A. Stoeckel, M. Davis, R. Burdis, Argosy Casino Baton Rouge, and the crew of "The Belle of Alton" for help in collecting mussels; R. Burdis (Minnesota DNR), L. Soeken-Gittenger (INHS), and D. J. Walters (Louisiana District USGS) for providing Mississippi River temperature data; E. Bullard, P. Van Zandt, and P. Leberg for assistance with the statistical analyses. The authors also thank P. Leberg, S. Mopper, L. Deaton, and J. E. Marsden for comments on an earlier version of this manuscript. This research was supported by a grant from US Army Corps of Engineers, Waterways Experiment Station (Contract# DACW39-90-P-0104) to CLE and PLK, a University of Louisiana Fellowship, funding from the University of Louisiana Graduate Student Organization, and a Sigma Xi grant-in-aid of research to CLE.

LITERATURE CITED

Alexander, J. E., J. H. Thorp & R. D. Fell. 1994. Turbidity and temperature effect on oxygen consumption in the zebra mussel (Dreissena polymorpha). Can. J. Fish. Aquat. Sci. 51:179-184.

Allen, Y. C., B. A. Thompson & C. W. Ramcharan. 1999. Growth and mortality rates of the zebra mussel, Dreissena polymorpha, in the Lower Mississippi River. Can. J. Fish. Aquat. Sci. 56:748-759.

Dahlhoff, E. & G. N. Somero. 1993. Kinetic and structural adaptations of cytoplasmic malate dehydrogenases of eastern Pacific abalone (genus Haliotis) from different thermal habitats: Biochemical correlates of biogeographical patterning. J. Exp. Biol. 185:137-150.

Dixon, P. M. & M. C. Newman. 1991. Analyzing toxicity data using statistical models for time to death: An introduction. In: Newman, M. C. & A. McIntosh, editors. Metal ecotoxicology: concepts and applications. Chelsea: Lewis. pp 207-242.

Elderkin, C. L., J. A. Stoeckel, D. J. Berg & P. L. Klerks. 2004. Heritability of heat tolerance in zebra mussel veligers. J. Great Lakes Res. 30:360-366.

Elderkin, C. L., P. L. Klerks & E. Theriot. 2001. Shifts in allele and genotype frequencies in zebra mussels, Dreissena polymorpha, along the latitudinal gradient formed by the Mississippi River. J. N. Am. Benthol. Soc. 20:595-605.

Fetisov, A. N., A. V. Rubanovich, T. S. Slipchenko & V. A. Shevchenko. 1991. Effect of the temperature factor on the genetic structure of populations of Dreissena polymorpha (Bivalvia). Soviet Genetics 26:1159-1163.

Gilg, M. R. & T. J. Hilbish. 2003. The geography of marine larval dispersal: coupling genetics with fine-scale physical oceanography. Ecology 84:2989-2998.

Hilbish, T. J. & R. K. Koehn. 1985. The physiological basis of natural selection at the Lap locus. Evolution 39:1302-1317.

Hopkins, A. E. 1949. Determination of condition of oysters. Science 110: 567-568.

Horvath, T. G., G. A. Lamberti, D. M. Lodge & W. L. Perry. 1996. Zebra mussel dispersal in lake-stream systems: source-sink dynamics? J. N. Am. Benthol. Soc. 15:564-575.

Iwanyzki, S. & R. W. McCauley. 1993. Upper lethal temperature of adult zebra mussels (Dreissena polymorpha). In: T. F. Nalepa & D. W. Schloesser, editors. Zebra mussels, biology, impacts, and control. Boca Raton: Lewis. pp 667-673.

Klerks, P. L. & J. S. Levinton. 1989. Rapid evolution of metal resistance in a benthic oligochaete inhabiting a metal-polluted site. Biol. Bull. 176: 135-141.

Koehn, R. K., R. I. E. Newell & F. Immermann. 1980. Maintenance of an aminopeptidase allele frequency cline by natural selection. Proceedings of the National Academy of Sciences of the United States of America Biological Sciences 77:5385-5389.

Lawrence, D. R. & G. I. Scott. 1982. The determination and use of condition index of oysters. Estuaries 5:23-27.

Lin, J. J. & G. N. Somero. 1995. Thermal adaptation of cytoplasmic malate dehydrogenases of eastern Pacific barracuda (Sphyraena spp): The role of differential isoenzyme expression. J. Exp. Biol. 198:551-560.

Ludyanskiy, M. L., D. McDonald & D. MacNeill. 1993. Impact of the zebra mussel, a bivalve invader. BioScience 43:533-544.

Madon, S. P., D. W. Schneider, J. A. Stoeckel & R. Sparks. 1998. Effects of inorganic sediment and food concentration on energetic processes of the zebra mussel, Dreissena polymorpha: Implication for growth in turbid rivers. Can. J. Fish. Aquat. Sci. 55:401-413.

McMahon, R. F. 1996. The physiological ecology of the zebra mussel, Dreissena polymorpha, in North America and Europe. Am. Zool. 36: 339-363.

McMahon, R. F. 2002. Evolutionary and physiological adaptations of aquatic invasive animals: r selection versus resistance. Can. J. Fish. Aquat. Sci. 59:1235-1244.

McMahon, R. F. & T. A. Ussery. 1995. Thermal tolerance of zebra mussels (Dreissena polymorpha) relative to rate of temperature increase and acclimation temperature. Vicksburg: U. S. Army Engineer Waterways Experiment Station.

Mitton, J. B. 1997. Selection in Natural Populations. Oxford University Press, New York.

Rajagopal, S., G. Van der Velde & H. A. Jenner. 1997. Response of zebra mussel, Dreissena polymorpha, to elevated temperatures in the Netherlands. In: F.M. D'Itri, editor. Zebra mussels and aquatic nuisance species. Chelsea: Ann Arbor Press. pp 257-273.

Schneider, D. W. 1992. A bioenergetics model of zebra mussel, dreissena polymorpha, growth in the Great Lakes. Can. J. Fish. Aquat. Sci. 49:1406-1416.

Smirnova, N. F., G. I. Biochino & G. A. Vinogradov. 1992. Some aspects of the zebra mussel, (Dreissena polymorpha) in the former European USSR with morphological comparisons to Lake Erie. In: T. F. Nalepa & D. W. Schloesser, editors. Zebra mussels: biology, impacts, and control. Boca Raton: Lewis. pp 217-226.

Spidle, A. P., E. L. Mills & B. May. 1995. Limits to tolerance of temperature and salinity in the quagga mussel (Dreissena bugensis) and the zebra mussel (Dreissena polymorpha). Can. J. Fish. Aquat. Sci. 52: 2108-2119.

Stoeckel, J. A., D.W. Schnieder, L. A. Soeken, K. D. Blodgett & R. E. Sparks. 1997. Larval dynamics of a riverine metapopulation: Implications for zebra mussel recruitment, dispersal, and control in a large river system. J. N. Am. Benthol. Soc. 16:586-601.

Stoeckmann, A. M. 2003. Pysiological energetics of Lake Erie dreissenid mussels: A basis for the displacement of Dreissena polymorpha by Dreissena bugensis. Can. J. Fish. Aquat. Sci. 60:126-134.

Stoeckmann, A. M. & D. W. Garton. 1997. A seasonal energy budget for zebra mussels (Dreissena polymorpha) in western Lake Erie. Can. J. Fish. Aquat. Sci. 54:2743-2751.

Thorp, J. H., J. E. Alexander, Jr., B. L. Bukaveckas, G. A. Cobbs & K. L. Bresko. 1998. Responses of Ohio River and Lake Erie dreissenid molluscs to changes in temperature and turbidity. Can. J. Fish. Aquat. Sci. 55:220-229.

CURT L. ELDERKIN (1,2,) * AND PAUL L. KLERKS (1)

(1) University of Louisiana at Lafayette, Department of Biology, P.O. Box 42451, Lafayette, Louisiana 70504; (2)Current address: Miami University, Department of Zoology, 212 Pearson Hall, Oxford, Ohio 45056

* Corresponding author. E-mail: [e-mail]