1. Trophic fractionation was studied in short-term laboratory feeding experiments with larvae of the deposit-feeding midge Chironomus riparius. Larvae were fed food of terrestrial (oats, peat) and aquatic origin (Spirulina, Tetraphyll Ò ). 2. By analysing both whole larvae and isolated gut contents we were able to distinguish between the isotopic signature of recently ingested food and that of assimilated carbon and nitrogen in body tissue. Additionally we studied the effects of microbial conditioning, i.e. the colonisation and growth on food particles of microbes, on the isotopic signal of food resources. 3. Nitrogen fractionation for the different food types ranged from 0.67& to 2.68& between consumer and diet and showed that isotopic fractionation can be much lower than the value of 3.4& that is commonly assumed. 4. Microbial degradation of food particles resulted in an approximate doubling of the d 15 N in 8 days, from 6.24 ± 0.05& to 11.36 ± 0.56&. Values for d 13 C increased only marginally, from )20.66 ± 0.11& to )20.34 ± 0.12&. These results show that microbial conditioning of food may affect dietary isotope signatures (in particular N) and, unless accounted for, could introduce an error in measures of trophic fractionation. Microbial conditioning could well account for some of the variation in fractionation reported in the literature.

During long-term standardized toxicity tests with Chironomus riparius, food additions are a prerequisite for normal development and to avoid false-positive results. Consequently, larvae may selectively feed on added food rather than on contaminated sediment, which may confound toxicity test results. We designed a feeding study and estimated the degree of feeding on different food resources by using stable isotope and fatty acid (FA) analyses. In one treatment, larvae were offered both artificial sediment (peat, kaolin clay, sand, and calcium carbonate) and added food (TetraPhyll), whereas larvae in the two other treatments had access to either one of these potential food items. The highest biomass and survival were found among larvae with access to both artificial sediment and TetraPhyll. Two-source mixing models revealed that larval Chironomus that were offered both TetraPhyll and artificial sediment obtained 94 +/- 6.9% of their carbon and 90 +/- 4.3% of their nitrogen from added TetraPhyll. Larvae with access to only sediment had lower delta13C and delta15N (-23.34 +/- 0.56 per thousand and 0.33 +/- 0.52 per thousand) than those that were offered both sediment and TetraPhyll (-20.95 +/- 0.13 per thousand and 7.45 +/- 0.36 per thousand) or only TetraPhyll (-20.17 +/- 0.20 per thousand and 7.82 +/- 0.15 per thousand). In addition, FA composition of larvae that were offered both artificial sediment and TetraPhyll closely resembled that of those fed exclusively TetraPhyll. These results show that larval Chironumus strongly prefer added food, rather than artificial sediment in long-term toxicity tests. This preferential feeding behavior affects exposure pathways and ultimately toxicity test results.

We quantified sublethal and sex-specific cypermethrin effects in experiments (29 days) with the midge Chironomus riparius at different levels of sediment organic matter content (0, 5, and 20%). We found highly significant effects of cypermethrin concentrations and sediment type on emergence, mean development rate, and adult size. For example, emergence/survival rates were 70-100% below 0.8 μg/l and unaffected by organic matter content. At 3.2 μg/l, however, no larvae survived in sediment without organic matter, but survival successively increased to 26 ± 11% in sediment with 20% organic matter. Mean development rates were always higher for males than for females, and significant differences between sexes occurred consistently in controls and in treatments with our lowest concentration of 0.05 μg/l. Sex-specific differences in mean development rate decreased across the cypermethrin gradient, suggesting that male development was affected more than that of females at similar concentrations. We also found an increase in adult size across the concentration gradient in sediments lacking organic matter and suggest an increased feeding activity due to sublethal toxic stress as a probable causal mechanism. We speculate that the observed sex-specific effects on development rates and adult size can have strong repercussions on emergence timing and fecundity, respectively.

No abstract

Juvenile Baltic salmon, Salmo salar, were fed commercial salmon food contaminated with different concentrations of polychlorinated naphthalenes (PCNs; 0.1, 1, 2, or 10 microg PCN/g food). Among other effects, possible estrogenic impact caused by PCNs were evaluated. Fish were therefore fed a diet contaminated with 17beta-estradiol (E2; 0.94 or 9.4 microg E2/g food), as a positive control. After 8, 13, 17, and 41 weeks, sampling took place. Growth, liver somatic index (LSI), EROD activity, and vitellogenin content in blood plasma were measured along with morphological studies of gonads and chemical analyses to determine the effects caused by PCNs. Exposure to PCNs did not seem to have any effects on body weight gain, since the group fed the high dose followed the growth in the control group during the entirely experimental period. After 41 weeks of exposure the groups fed 2 and 10 microg PCN/g food had significantly lower LSIs compared with the control, indicating liver toxic effects of PCNs. Furthermore, a dose-dependent induction of EROD activity was found. At week 41, the control group had an activity of 4.9 +/- 4.8 pmol/mg prot/min, whereas it was between 69 +/- 21 and 720 +/- 370 pmol/mg prot/min in the exposed groups, respectively. Examination of gonadal morphology showed that PCNs also had negative effects on ovaries in Baltic salmon, including delayed development. The distribution between females and males, gonadal morphology, and vitellogenin content in blood plasma did, however, indicate that PCNs are not capable of causing effects similar to E2 or xenoestrogens. Exposure to both of the concentrations of E2 resulted in decreased body weight gain, increased LSI, and feminization of the gonads. E2 did, however, not induce any EROD activity.