Limnol. Oceanogr., 44(2), 1999, 393–402

The effects of fish kairomones, crowding chemicals, and day length on the life-history traits of a set of 16 Daphnia magna clones, derived from four populations that differ in fish-predation pressure, were studied. Significant among-population differences were observed, the differences being in concordance with the hypothesis of local adaptation. The among-population genetic differences were not mediated through a change in response to fish kairomones, but through an overall smaller body size, smaller eggs, and a higher number of eggs in clones derived from habitats in which fish are present. Using a model, we show that the observed changes in life-history characteristics may lead to differences in fitness under different predation regimes, such that populations from habitats with fish have highest fitness under fish-predation regimes and populations without a fish background have higher fitness values under invertebrate predation regimes. Local genetic differentiation and local adaptation have been observed in a variety of species and habitats (Endler 1986) and have led to the introduction of the ecotype concept as far back as 1922, when Turesson (1922) realized that differences between organisms from different locations must have a genetic basis and must have resulted from evolutionary differentiation within species according to habitat. Several studies have shown that also for aquatic organisms (mainly Daphnia species), differences in genetic structure between habitats exist (e.g., Hebert 1974; Boileau and Taylor 1994). If this genetic differentiation results from an adaptive response to local selective forces rather than stochastic events, organisms should have highest relative fitness in their site of origin, compared to other sites. The straightforward way to test this would be by means of reciprocal transplantation experiments (Ayre 1985; van Tienderen and van der Toorn 1991; Via 1994). As this is not always possible, organisms of different populations could be cultured in common environments (mimicking environmental conditions of different backgrounds), and among-population differences can be quantified. From these experiments, one can infer that local adaptation has occurred if there are significant amongpopulation genetic differences in the traits and if there is within-population genetic variability allowing for a microevolutionary response of the trait value. This allows for the 1 Permanent address: Laboratory of Ecology and Aquaculture, Katholieke Universiteit Leuven, Naamsestraat 59, B-3000 Leuven, Belgium. 2 Present address: Department of Limnology, EAWAG/ETH, Überlandstrasse 133, CH-8600 Dübendorf, Switzerland.