Eukaryotic phytoplankton exhibit an enormous species richness, displaying a range of phylogenetic, morphological and physiological diversity. Yet, until recently, very little was known about the diversity, genetic variation and evolutionary processes within species and populations. An approach to explore this diversity and to understand evolution of phytoplankton is to use population genetics as a conceptual framework and methodology. Here, we discuss the patterns, processes and questions that population genetic studies have revealed in eukaryotic phytoplankton. First, we describe the main biological processes generating genetic variation. We specifically discuss the importance of life-cycle complexity for genetic and phenotypic diversity and consider how such diversity can be maintained during blooms when rapid asexual proliferation dominates. Next, we explore how genetic diversity is partitioned over time and space, with a focus on the processes shaping this structure, in particular selection and genetic exchange. Our aim is also to show how population genetics can be used to make inferences about realized dispersal and sexual recombination, as these processes are so difficult to study directly. Finally, we highlight important open questions and suggest promising avenues for future studies that will be made possible by new sequencing technologies.

Summary: Here we investigated whether there is evidence of local adaptation in strains of an ancestrally marine dinoflagellate to the lacustrine environment they now inhabit (optimal genotypes) and/or if they have evolved phenotypic plasticity (a range of phenotypes). Eleven strains of P olarella glacialis were isolated and cultured from three different environments: the polar seas, a hyposaline and a hypersaline Antarctic lake. Local adaptation was tested by comparing growth rates of lacustrine and marine strains at their own and reciprocal site conditions. To determine phenotypic plasticity, we measured the reaction norm for salinity. We found evidence of both, limited local adaptation and higher phenotypic plasticity in lacustrine strains when compared with marine ancestors. At extreme high salinities, local lake strains outperformed other strains, and at extreme low salinities, strains from the hyposaline lake outperformed all other strains. The data suggest that lake populations may have evolved higher phenotypic plasticity in the lake habitats compared with the sea, presumably due to the high temporal variability in salinity in the lacustrine systems. Moreover, the interval of salinity tolerance differed between strains from the hyposaline and hypersaline lakes, indicating local adaptation promoted by different salinity.

A fundamental question in ecology is whether microorganisms follow the same patterns as multicellular organisms when it comes to population structure and levels of genetic diversity. Enormous population sizes, predominately asexual reproduction and presumably high dispersal because of small body size could have profound implications on their genetic diversity and population structure. Here, we have analysed the population genetic structure in a lake‐dwelling microbial eukaryote (dinoflagellate) and tested the hypothesis that there is population genetic differentiation among nearby lake subpopulations. This dinoflagellate occurs in the marine‐derived saline lakes of the Vestfold Hills, Antarctica, which are ice‐covered most of the year. Clonal strains were isolated from four different lakes and were genotyped using amplified fragment length polymorphism (AFLP). Our results show high genetic differentiation among lake populations despite their close geographic proximity (<9 km). Moreover, genotype diversity was high within populations. Gene flow in this system is clearly limited, either because of physical or biological barriers. Our results discard the null hypothesis that there is free gene flow among protist lake populations. Instead, limnetic protist populations may differentiate genetically, and lakes act as ecological islands even on the microbial scale.

Latency is a common strategy in a wide range of viral lineages, but its prevalence in giant viruses remains unknown. In this work, we describe a 617–kilo–base pairs integrated giant viral element in the model green alga Chlamydomonas reinhardtii . We resolved the integrated viral genome using long-read sequencing, identified a putative polintovirus-like integrase, and show that viral particles accumulate primarily during the stationary growth phase. A diverse array of viral-encoded selfish genetic elements is expressed during viral activity, including several Fanzor nuclease–encoding transposable elements. In addition, we show that field isolates of Chlamydomonas spp. harbor signatures of endogenous giant viruses related to the C. reinhardtii virus that exhibit similar infection dynamics, suggesting that giant virus latency is prevalent in natural host communities. Our work describes an unusually large temperate virus of a unicellular eukaryote, substantially expanding the scope of cryptic viral infections in the virosphere.

Species invasion and range expansion are currently under scrutiny due to increasing anthropogenic impact on the natural environment. This is also true for harmful algal blooms, which have been reported to have increased in frequency. However, this research is challenging due to the ephemeral nature, small size and mostly low concentrations of microalgae in the environment. One such species is the nuisance microalga Gonyostomum semen (Raphidophyceae), which has increased in occurrence in northern Europe in recent decades. The question of whether the species has expanded its habitat range or if it was already present in the lakes but was too rare to be detected remains unanswered. The aim of the present study was to determine the genetic structure and dispersal pathways of G. semen using RAD (restriction‐site‐associated DNA) tag sequencing. For G. semen, which has a huge genome (32 Gbp), we faced particular challenges, but were nevertheless able to recover over 1000 single nucleotide polymorphisms at high coverage. Our data revealed a distinct population genetic structure, demonstrating a divide of western and eastern populations that probably represent different lineages. Despite significant genetic differentiation among lakes, we found only limited isolation‐by‐distance. While we had expected a pattern of recent expansion northwards, the data demonstrated gene flow from the northeast/east towards the southwest/west. This genetic signature suggests that the observed gene flow may be due to dispersal by autumn migratory birds, which act as dispersal vectors of resistant resting propagules that form at the end of the G. semen blooms.

Diel vertical migration (DVM) is the most common behavioral phenomenon in zooplankton, and numerous studies have evaluated DVM under strong seasonality at higher latitudes. Yet, our understanding of the environmental drivers of DVM at low latitudes, where seasonal variation is less pronounced, remains limited. Therefore, we here examined patterns of vertical distribution in copepods in six subtropical Bahamian blue holes with different food web structure and tested the role of several key environmental variables potentially affecting this behavior. Day and night samplings showed that copepods generally performed DVM, characterized by downward migration to deeper depths during the day and upward migration to surface waters at night. Across all blue holes, the daytime vertical depth distribution of calanoid copepods correlated positively with both predation risk and depth of food resources (Chlorophylla), but was less affected by ultraviolet radiation (UVR). A potential explanation is that since UVR is a continuous threat across seasons, zooplankton have established photoprotective pigmentation making them less vulnerable to this threat. The copepods also showed a size-structured depth segregation, where larger individuals were found at deeper depths during the day, which further strengthens the suggestion that predation is a major driver of DVM in these systems. Hence, in contrast to studies performed at higher latitudes, we show that despite the constant exposure to UVR, predator avoidance and food availability are the most pronounced drivers of copepod DVM at those low latitudes, suggesting that the main driver of DVM may vary among systems, but also systematically by latitude.