Background and Aims Seed dormancy determines the environmental niche of plants in seasonal environments, and has consequences for plant performance that potentially go far beyond the seed and seedling stages. In this study, we examined the cascading effects of seed dormancy on the expression of subsequent life-history traits and fitness in the annual herb Arabidopsis thaliana. 

Methods We planted seeds of >200 recombinant inbred lines (RIL) derived from a cross between two locally adapted populations (Italy and Sweden), and both parental genotypes at the native site of the Swedish population in three consecutive years. We quantified the relationship between primary seed dormancy and the expression of subsequent life-history traits and fitness in the RIL population with path analysis. To examine effects of differences in dormancy on relative fitness of the two parental genotypes, we planted dormant seeds during the seed dispersal period and non-dormant seeds during the germination period of the local population.

Key Results In the RIL population, strong primary dormancy was associated with high seedling survival, but with low adult survival and fecundity, and the path analysis indicated that this could be explained by effects on germination timing, rosette size, and flowering start. The relationship between primary seed dormancy and germination proportion varied among years, and this was associated with differences in seasonal changes in soil moisture. The planting of dormant and nondormant seeds indicated that the lower primary dormancy of the local Swedish genotype contributed to its higher germination proportion in two years, and to its higher fecundity in one year. 

Conclusions Our results show that seed dormancy affects trait expression and fitness components across the life cycle, and suggest that among-year variation in the incidence of drought during the germination period should be considered when predicting the consequences of climatic change for population growth and evolution.

The data set includes data used in the analyses presented in the paper Postma, F. M., and J. Ågren. 2022. Effects of primary seed dormancy on life-time fitness of Arabidopsis thaliana in the field. Annals of Botany, in press. Methods are described in detail in this paper.

The data are found in four excel files. A brief summary of methods used to obtain the data in each file are given below.

PostmaAgren_ParentalGenotypes.xlsx

In each of 3 years (2012, 2013, and 2014), seeds of an Italian and a Swedish genotype were planted at Rödåsen, north-central Sweden (the site of origin of the Swedish genotype) either as primary dormant seeds at the time of seed dispersal in the local natural population, or as non-dormant seeds at the time of seed germination in the local population. The primary dormant seeds had been produced at the site and were collected at seed maturation 14-19 days prior to sowing. Each genotype was planted in 80 replicates (primary dormant seeds in each of the three years, non-dormant seeds in the 2012 experiment), 60 replicates (non-dormant seeds in the 2013 experiment), or 30 replicates (non-dormant seeds in the 2014 experiment). Each replicate consisted of 40 seeds sown in one cell of a plug tray, except for non-dormant seeds in the 2014 experiment of which 20 seeds were planted in each cell. The excel file includes data on sowing day (day when seeds were planted), germination proportion, seedling and adult survival, fecundity (number of fruits produced by fruit-producing plants), fitness (fruit production per planted viable seed), germination time, rosette size and flowering start.

PostmaAgren_RILs.xlsx

In each of 3 years, primary dormant seeds of 219 (2012), 220 (2013), and 204 (2014) recombinant inbred lines derived from a cross between the Italian and the Swedish genotype were planted at Rödåsen, north-central Sweden (the site of origin of the Swedish parental genotype) during the natural seed dispersal period in early summer. The seeds had been produced at the site and were collected 14-19 days prior to sowing. Each line was sown in up to 10 replicates in the 2012 and 2013 experiments, with each replicate consisting of 40 seeds sown in one cell of a plug tray. In the 2014 experiment, we planted 5 replicates per recombinant inbred line, with 40 seeds per replicate for 180 RILs and between 20-35 seeds per replicate for 40 RILs. Low seed production at the experimental site in 2014 made it necessary to reduce the number of replicates (number of plugs sown), and the number of seeds sown per plug of some lines. The excel file includes the following data: dormancy score of seeds produced at the field site by the recombinant inbred lines (obtained in a previous experiment), and mean phenotypes of each recombinant inbred line in the experiments established in 2012, 2013, and 2014. Phenotypes are mean germination proportion, seedling and adult survival, fecundity (number of fruits produced by fruit-producing plants), germination time, rosette size and flowering start (2013 and 2014 experiments only).

PostmaAgren_NaturalPopulation.xlsx

Number of seedlings with cotyledons only and number of seedlings with true leaves were regularly monitored in 25 squares of 10 x 10 cm in the natural population at the site of the experiment (Rödåsen, north-central Sweden) in 2012, 2013, and 2014. The excel file includes data on number of seedlings with only cotyledons, number of seedlings with true leaves, and total number of seedlings in the plots at each census.

PostmaAgren_SoilConditions.xlsx

Soil temperature and water potential were recorded hourly by sensors placed ca. 1 cm below the soil surface at the experimental site at Rödåsen, north-central Sweden. The excel file includes data on minimum, mean, and maximum daily soil temperature, and mean daily water potential (kPa) in 2012, 2013, and 2014.

Please, see ReadMe file.