7764 - Laserfiche WebLink

Home Browse Search

Toxicity of carbaryl and malathion 105 -2 Control 1 10 100 1,000 CARBARYL CONCENTRATION (ug/L) 20 10,000 Fig. I. Estimated linear-plateau regression lines for Colorado squawfish exposed to technical carbaryl. Length of exposure was 32 d. Each point represents a mean of two replicates. trout (0, mykiss), Atlantic salmon (Salmo salar), and brook trout (Salvelinusjontinalis). They were two to 10 times more sensitive than fathead minnow (Pimephales promelas), chan- nel catfish (Ictalurus punctatus), and bluegill (Lepomis mac- rochirus). In contrast, Colorado squawfish and bonytail were one to two orders of magnitude less sensitive to malathion than cutthroat trout, rainbow trout, and bluegill; and ap- proximately as sensitive to malathion as fathead minnow and channel catfish. These comparisons provide some basis for discussing relative toxicity of carbaryl and malathion to Col- orado squawfish and bonytail, but two factors that strongly influence toxicity of carbaryl and malathion should be con- sidered. The first is related to size and life stage of test or- ganisms. The relation of toxicant sensitivity to body size is inconsistent [22]; however, early life stages are generally more sensitive to toxicants [23]. The acute toxicity tests summa- rized by Mayer and Ellersieck [6] involved "fingerling fish" weighing from 200 to 1,500 mg [24]. Therefore, we expected Colorado squawfish and bonytail to be relatively sensitive to carbaryl and malathion as a result of a life-stage effect. This prediction was supported for toxicity of carbaryl but not for malathion. However, the effect of a second factor that mod- ifies toxicity, that of pH, should be considered, Carbaryl and malathion hydrolyze rapidly in waters hav- ing pH > 7 [25,26]. The pH of test solutions in our renewal- acute studies ranged from 7.9 to 8.6, and concentrations of carbaryl and malathion declined by 58 and 37070, respectively, in 24 h. Under these conditions our estimates of median le- thal concentrations may have been underestimated by rela- tively short exposure to target concentrations. However, breakdown products of carbaryl and malathion have been shown to be more toxic than their respective parent com- pounds [27,28]. Thus the net toxic effect of rapid hydroly- sis of carbaryl and malathion is uncertain. Use of renewal-acute test results to compare interspecific sensitivity is confounded by the effects of pH, life-stage dif- ferences, and body-size differences, These factors also com- plicate reliable prediction of the effects of carbaryl or malathion on Colorado squawfish and bonytail in natural conditions. Fortunately, estimation of median lethal concen- trations was only a preliminary step in our assessment of tox- icity of carbaryl and malathion to Colorado squaw fish and bony tail. Flow-through ELS tests provided more sensitive measures of effects and permitted more reliable comparisons of interspecific sensitivity. ELS tests Interspecific comparisons of toxicant sensitivity based on ELS tests are of greater value than those based on renewal- acute tests because the design of ELS tests minimizes poten- tial confounding effects. For example, accumulation of toxicant breakdown products was precluded by the flow- through nature of ELS tests. In addition, ELS tests measured a sensitive endpoint (i.e., reduced growth), thus minimizing effects due to life-stage and size differences. Colorado squawfish and bony tail were approximately as sensitive to carbaryl as the fathead minnow [29]. The NOEC and LOEC for fathead minnow were 210 and 680 ILg/L, a range that overlaps that estimated for both endangered fishes. For fathead minnow exposed to malathion, the NOEC and LOEC were 200 and 580 ILg/L [30], These effect concen- trations were approximately five to eight times lower than those estimated for Colorado squawfish and bonytail. How- ever, toxicity to Colorado squawfish to bony tail may have been relatively underestimated for carbaryl and malathion because the exposure period was short (32 d compared to 9 and 10 months for fathead minnow) and reproductive effects were not studied. Long-term toxicity of malathion to two others species, flagfish (Jordanellafloridae) and bluegill, has been studied [31,32]. In a 30-d exposure, flag fish were more sensitive to malathion than Colorado squawfish or bony tail and had a LOEC of 24.7 and 10.9 ILg/L for survival and growth, respectively. Bluegill were also relatively sensitive to malathion: Eaton [32] estimated a LOEC of 7 AlLg/L on the basis of development of spinal deformities in adults. Con- sidering the range of species sensitivity encompassed by ex- isting long-term exposure studies, Colorado squaw fish and bony tail fall near the tolerant end of the spectrum and may be roughly equivalent to fathead minnow in sensitivity to AChE-inhibiting pesticides. Regression analysis vs. hypothesis testing Although threshold concentrations estimated by linear- plateau regression were consistently lower than NOECs from hypothesis testing, the two statistical methods did not lead to vastly different conclusions. Six of eight calculated thresh- old concentrations were within a factor of two ofthe NOECs; the remaining estimates were within a factor of four; and, in two cases, the upper limit of the C.!. for a threshold con- centration contained the NOEC. The tendency of linear-plateau regression to produce lower estimates of effect concentrations may suggest that the procedure underestimated the concentration at which toxic effects began to accrue; however, inspection of Figure 1 shows that fitted regression models and estimated thresholds were appropriate. A more likely explanation for the discrep- ancy between regression and hypothesis-testing estimates is related to experimental design, A weakness of our study was