Laserfiche WebLink
<br />) <br /> <br />rotifer diet will be employed (Figure 3). This experimental design is equivalent to <br />conducting two toxicity tests simultaneously. In one test, the larvae will be exposed to five <br />test waters. In the other, larvae will be exposed to test waters and food organisms that <br />were cultured in the corresponding test water. Tests will be conducted using l-L plastic <br />beakers containing 0.50 L of site water or control water. Contents of each beaker will be <br />renewed every 24 h. The dilution factor will be 0.5 or higher. Test waters and larvae will <br />be randomized to exposure beakers. Cool-white fluorescent lamps will be the only source <br />of illumination, and a 16:8-hour light:dark photoperiod will be maintained. Water-quality <br />will be measured routinely: alkalinity, hardness, pH, and specific conductance will be <br />measured twice weekly; dissolved oxygen will be measured daily; water temperature will <br />be measured continuously. Experimental animals will be acclimated to testing conditions <br />for 48 hours before toxicity tests are initiated. Twenty larvae will be placed in replicate <br />exposure chambers. Rotifers from respective test waters will be introduced into exposure <br />chambers three times a day and at least 2 hours before renewal. Survival oftest animals <br />will be monitored daily. Growth will be quantified at the end of the 28-dayexposure <br />period. <br /> <br />'j <br /> <br />Selenium concentrations in algae, rotifers, and fish will be measured by neutron activation <br />analysis (Environmental Contaminants Research Center, Columbia, Missouri). . <br />Concentrations in algae and rotifers in each site-water culture will be measured at the <br />beginning, middle, and end of the 28-day exposure period. Concentrations in razorback <br />sucker larvae will be measured at the beginning and end of the exposure period. These <br />data will be useful for evaluating if accumulation of selenium in laboratory animals is <br />consistent with field observations. <br /> <br />Agreement of predicted and observed toxicological responses will be assessed by <br />inspection of the data and formal statistical analyses. General agreement will suggest that <br />the predictive equations give reasonable estimates of toxic effects (Figure 2). <br />Disagreement will suggest that the equations are inaccurate or that co-contaminants <br />influenced the outcome of the toxicity tests. Analysis of covariance using logistic and <br />polynomial regression will be used to compare predicted and observed survival and growth <br />in each test water following the methods of Bailer and Oris (1997). Statistical analyses <br />will be conducted using SAS@ statistical software (SAS Institute 1991). <br /> <br />) <br /> <br />Selenium 5 <br />