Collection
Sixteen seed batches of the four Papaver taxa were collected (Table 1). For each seed batch, mature capsules of approx. 30 plants were used (in two cases where plants were scarce, approx. ten), and the seeds were well mixed. Two batches per taxon were collected in summer and two in autumn, using the same sites for the two occasions, with the exception of P. dubium ssp. lecoqii, which is rare in Sweden (Jonsell, 2001) and for which we could not find two sites with ripe seeds in summer. Papaver dubium ssp. dubium and P. dubium ssp. lecoqii were distinguished by colour of the dried latex: brown to black and red, respectively (Kadereit, 1989). All sites were within agricultural areas in southern Sweden, and for each taxon there were at least 4 km between the sites used. Seeds were kept indoors (approx. 20 °C, 35 % relative humidity) for 1 week before the onset of experimentation.
Germination experimentFor each treatment (described below), five Petri dishes (5 cm diameter) with 40 seeds in each were used. Ten millilitres of quartz sand (0·35 mm grain size; Baskarpsand 35, AB Baskarpsand, Habo, Sweden) and 3·8 mL deionized water were poured into each dish. The dishes were put on a vibrator to smooth the surface before seeds were placed on the substrate. Petri dishes were sealed with Parafilm. Dishes used for dark treatments were individually wrapped in aluminium foil, directly after seeds had been put on the substrate, and not opened before the end of the experimental period. During experiments, additional water was added when necessary to dishes in treatments with light.
Seeds were subjected to annual temperature cycles, representing three different climates: cold, intermediate and warm (Table 2). Seven temperature environments were used: three constant temperatures, –12 °C (LabRum AB, Stockholm, Sweden), 0 °C (Gram, Denmark) and 5 °C (ADU 200, Styrprojektering AB, Sweden), and four daily alternating, 15/5, 20/10, 25/15 and 30/20 °C day/night with 2 h linear transference between the maximum and minimum temperatures (Rubarth Apparatebau, Laatzen, Germany). All environments had light for 12 h d–1, coinciding with the higher temperature in daily alternating environments [daily alternating and 5 °C: 32–66 µmol m–2 s–1 with R : FR ratio 3·5–4·4; 0 °C and –12 °C: 29–44 µmol m–2 s–1 with R : FR ratio 1·6–2·1 (SKP 200, sensor SKP 215 and SKR 100, sensor SKR 110, Skye Instruments Ltd, Llandrindod Wells, UK)]. Dishes were randomly placed on shelves during incubation, and were rearranged randomly when checked. Each seed batch was subjected to one treatment simulating summer dispersal (beginning with the last 60 d of summer) and another simulating autumn dispersal (beginning with the middle 30 d of autumn) in each climate (Table 2). Experiments ran so that they included 30 d of the third winter after dispersal, i.e. the length of an experiment was 810 or 900 d for seeds beginning cycles with autumn or summer, respectively. Seeds were also subjected to the same environment for 900 d, namely 5/5, 15/5, 20/10, 25/15 or 30/20 °C.
There was one set of five dishes with light during daytime and five sets with continuous darkness in each annual cycle and each continuous temperature. Dishes subjected to light were checked for seedlings once a week over the first 2 months, every second week until day 510, and then every fourth week until the end of the experiment. Five dishes within the dark under constant temperature regimes were opened after each of 60, 270, 480, 690 and 900 d, and in annual cycles after each summer and winter. Seeds treated with light were regarded as having germinated upon root protrusion. Germination in the dark had frequently to be calculated from the number of empty seed coats. Some dishes with long incubation time in darkness at higher temperatures had dried out when opened; such dishes were directly discarded and not included in any evaluation.
Viability
Viability was checked at the beginning of the study by subjecting seeds to incubation on sand as above but with gibberellic acid solution (GA3, 1000 mg L–1; BDH Electran®, VWR International Ltd, Lutterworth, UK) instead of deionized water. Temperatures used were 15/5 or 25/15 °C, depending on the taxon, with light during daytime. Incubation continued until germination ceased. Three dishes with 50 seeds each were used for each seed batch.
During experiments, apparently dead seeds (moist and/or overgrown with mould) were counted and removed. At the end of the study, all dishes containing ungerminated seeds were placed without cover at room temperature for about 1 week, causing them to dry out completely. Thereafter, 3·8 mL GA3 was added to each dish. Dishes were transferred several times between higher and lower temperatures (the number of times depending on response rates) for about 6 months, when possible germination was regarded as having been achieved. Seedlings were removed as they emerged during incubation. Remaining ungerminated seeds were considered dead.
Phenology
Seeds were sown outside in Ledberg, close to one site used for collection of P. argemone (Table 1), 1 week after collection. For each seed batch, five pots, each with 50 seeds, were used. Plastic pots (11 cm diameter) were filled with 0·4 L moist quartz sand (Baskarpsand 35) and the seeds were placed on the surface. A hollow in the ground, 0·15 m deep, was filled with ceramic clay pellets 2–6 mm in diameter (AB Svenska Leca, Linköping, Sweden) and the pots were buried to such a depth that the surface of the sand was at the same level as the surface of the surroundings. The pots were protected from direct wind, rain and sunshine, but the sand was kept moist during the experiments. Temperature was measured with TinytagPlus (Intab; Stenkullen, Sweden) every hour. The pots were checked for seedlings at least twice a month (if snow-free) for about 2·5 years, until 4 December, 2004.
Calculations and analysis
In the germination study, seeds scored as dead during the study and seeds that did not germinate during a GA3 test at the end of the experiments were excluded from calculations of germination. For the phenology study, the numbers of emerged seedlings are reported as fractions of the number of seeds sown.
Analyses of variance of results from continuous temperature regimes and annual cycles were performed with Statistica (StatSoft Inc., 2002) on arcsine-transformed data. One seed batch was regarded as one replicate. Categorical predictors were: taxon, temperature, light condition and collection time for continuous regimes, and taxon, climate, light condition, collection time and starting point for annual cycles. Time was treated as a continuous variable with five measured points: after 60, 270, 480, 690 and 900 d for continuous regimes, and each summer and winter for annual cycles. For each light treatment, the five dishes used were randomly distributed to one point in time each before analysis. For interpretation, special attention was paid to the factor ‘taxon’ and interactions including ‘taxon’, to evaluate possible differences between taxa. Papaver dubium ssp. lecoqii, which was not replicated during summer collection (Table 1), was omitted from ANOVA.
Answers to all your Biology Questions
