The pond snail Lymnaea stagnalis has been proposed as a standard test organism for aquatic invertebrates for future OECD guidelines. The pond snail has a rather complex reproductive system: as a simultaneous hermaphrodite, it maintains both male and female function at the same time. The decision on how much to invest into the male or female function is generally designated as the `sexual conflict' of the snail. Mating opportunity, amongst other factors, influences both fecundity and egg size. The aim of this dissertation was to i) scrutinize the proposed experimental conditions for their suitability for standardization, ii) explore possibilities of using DEB theory to understand the e ects of di erent stressors and their interactions on the reproductive system of the pond snail, and iii) investigate the potential population-level effects of the observed interactions.
We found that the juvenile pond snails are food limited by lettuce, which is the recommended food source in ecotoxicological investigations with this species. The initial food limitation can lead to an overestimation of the toxicity of compounds if it is not accounted for in the analysis of test results. Since food availability also determines the age at first reproduction, the initial food limitation also has serious impacts for the predictions of population-level effects. Moreover, using lettuce as a food source holds potential for interacting with compounds used in toxicity tests. In a case study with the herbicide diquat, we could demonstrate that the hormetic effect pattern in growth and reproduction was caused by the interaction of the herbicide with the lettuce. Both results are in general applicable for the suitability of food sources in ecotoxicity tests: juveniles commonly have different nutritional needs than adults, and might thus be unintentionally limited. And, many food sources in general hold potential for interaction with toxicity, and we need to be aware of that as this interaction can strongly in uence the interpretation of toxicity tests.
When applying the DEB model, we found that the pond snail seems to undergo a so-called metabolic acceleration, which has been commonly found in molluscs. The acceleration plays a major role in the determination of egg costs in the model, and thus in population-level predictions. The investment per egg is generally also in uenced by food availability that the mother experienced, a phenomenon commonly known as a maternal effect. In an experiment designed for investigating the interaction of the sexual con ict and the maternal effect, we discovered that the pond snail invests relatively more energy into the male function when food is scarce. A similar phenomenon has been observed in many simultaneous hermaphrodites, which suggest a broader applicability of our model. However, there was no clear in uence of food availability on the investment per egg, but the rather small effect as predicted by the DEB model for the pond snail could easily have been missed. The population-level e ects of the observed interactions can be further explored by using DEB-IBM, an individual based modeling framework, which was designed to investigate DEB individuals at the population level.
In conclusions, we can say that lettuce is no optimal food source for ecotoxicity experiments with the pond snail. The juveniles are food limited when fed with lettuce, which interactions with toxicant effects. If no better food source can be found, a DEB model should be used for the interpretation of toxicity tests. Since the investment per offspring is potentially in uenced by many maternal factors (including toxicants), it needs to be determined in toxicity tests. When under stress, pond snails can chance the balance between investing in the male and female function. Such behavior can bias the interpretation of toxic effects on reproduction. Considering the energy budgets at individual level is important to extrapolate effects at the population level.