Cyclically parthenogenetic organisms, which reproduce both sexually and asexually, are common in nature, in both the animal and plant kingdoms [1, 2]. This mode of reproduction, especially its asexual (clonal) component, has attracted considerable interests among population geneticists and evolutionary biologists. Asexual phase of the cyclically parthenogenetic reproduction cycle may have profound impact on population structures. For example, in the long-term, clonal reproduction may reduce the number of genetically distinct individuals within a population and hence decrease the effective population size (e.g. [3, 4]). It may also lead to a spatial genetic autocorrelation, which could be falsely attributed to limited propagule dispersal or kin-structured colonization (e.g. ). Finally, as clones differ in fitness under varying environmental conditions , changes in clone frequencies are expected across the growing season . Consequently, clonal selection can result in the strong reduction of clonal diversity [4, 8].
In freshwater habitats, cyclical parthenogenesis is common among many groups of zooplankton . Cladocerans, which reproduce parthenogenetically during favourable conditions and switch to sexual reproduction when conditions deteriorate (e.g. ), are particularly important in these environments, being the main component of aquatic food-webs [10, 11]. The cladoceran genus Daphnia is commonly used as a model system for cyclical parthenogenesis in ecological and evolutionary research. In some Daphnia species complexes, interspecific hybrids may be produced during the sexual part of their reproductive cycle [12, 13]. Hybridization has been documented within several species complexes of Daphnia from Eurasia, North America and Australia [12, 13], but most research has concentrated on the D. longispina complex, inhabiting permanent lakes of the northern temperature zone . In Europe, this complex includes, together with some rarer taxa, the widespread and ecologically important species D. cucullata, D. galeata and D. longispina . These species often form interspecific hybrids which sometimes reach high abundances [13, 15–18]. Once Daphnia hybrids are produced by sexual recombination, they can be maintained by clonal propagation for many generations , as in other cyclical parthenogens (e.g. [2, 19]). In the D. longispina complex, although parental species also reproduce clonally for most of the year, there is evidence that they invest more into sexual reproduction than their F1 hybrids [16, 20].
In previous studies of the D. longispina complex, the relative frequencies of different taxa were compared across time (e.g. [20–23]) and space (e.g. [15, 17, 24]). However, changes in clonal structure have been largely unexplored due to methodological limitations. So far, the most common method for identification of clones in the D. longispina complex has been allozyme electrophoresis (e.g. [23, 25–28]), although RAPD markers were also used occasionally (e.g. ). However, allozyme studies are limited by the few polymorphic loci they provide; in most cases, it is likely that the multilocus genotypes defined by allozymes represented clonal groups . This substantially limits the power to trace the frequencies of single clones and to study clonal structure in general. RAPDs, although more variable, have often poor reproducibility  and, being dominant markers which cannot separate homozygotes from heterozygotes , have limited use in the analyses of population structure. Recently, microsatellite markers have been developed for the D. longispina complex . However, the subsequent studies employing these markers have focused so far on either a description of population state at a single time point [29, 33] or on exploring temporal changes at the taxon level only [18, 21, 34]. In other systems, microsatellites have already been proven to be very powerful in tracing clonal lineages; for example, in the cyclically parthenogenetic aphid  or in bacterial populations .
In the present study, we used 10 microsatellite loci to explore temporal and spatial dynamics in the taxonomic and clonal structure of the D. longispina hybrid complex, in two reservoirs in the Czech Republic. The canyon-shaped morphology of these reservoirs creates longitudinal environmental gradients, which results in a spatial variation in the composition of zooplankton communities including Daphnia [17, 24]. One of the studied reservoirs (Římov) was recently dominated by a single parental species (D. galeata), whereas three parental species (D. galeata, D. longispina, and D. cucullata) as well as their interspecific hybrids coexisted in the second reservoir (Vír) [17, 24]. We screened Daphnia communities in these reservoirs at the end of the growing season, when temperate lakes undergo a major change - a transition from summer stratification to autumn mixing and winter conditions . The goals of the study were to explore dynamics in taxonomic and clonal structure, across both time (generation-to-generation) and space (between sampling stations along the reservoir's longitudinal gradient), during a period of seasonal environmental change. We also tested one particular hypothesis that the clonal diversity is lower in hybrids than in parental species, due to some pre- and postzygotic barriers between parental genomes , resulting in a lower number of newly produced hybrids in comparison to parental clones.