I germinated seeds exposed to three moisture regimes (flooded, intermittently flooded, or control) in a controlled growth chamber to see how they would respond to flooding as a simulation of fluctuating lake levels. Seeds were chosen based on their likelihood to germinate, meaning very small or flat seeds were excluded from the experiment. Only seeds from 2015 were used, since germinability declines after 2 yr in storage (Hamzé and Jolls 2000). I selected 51 seeds from each of four genotypes (denoted as genotypes 2 through 5) and measured mass of each to 0.001 mg. Seeds were then blocked (17 seeds from each genotype for a total of 68 per treatment group). Seed size has not shown to be significant in determining emergence; however, larger seeds develop larger root systems, which could improve survival (Chen and Maun 1999). Seeds were placed into 29.6 mL clear plastic cups and placed on large baking trays to separate treatment groups, with one tray per treatment group. Due to time constraints and relative ineffectiveness, cold stratification, (the natural method for removing the seed coat) was not used (Chen and Maun 1998). Instead, the radicles of each seed were nicked with a scalpel to allow the seed to imbibe water.
After nicking, seeds were removed one at a time from their cups and 9.05 〖cm〗^3 of dry autoclaved Lake Michigan beach sand was poured into the cup. Seeds were then gently placed on the sand and covered with another 9.05 〖cm〗^3 of sand, the equivalent of 1 cm burial depth. This served the additional purpose of keeping the seeds dark. These conditions result in higher germination (Hamzé and Jolls 2000). A growth chamber was set with a 15/22℃ 10/14 hr thermoperiod, with the lights set to the same schedule. The baking trays were placed inside, one on each shelf.
Each watering regime was designed to simulate how C. pitcheri habitat could change if lake levels fluctuated to new extremes and created new swales, ephemeral or permanent. The flooding regimes were expected to create oxygen stress, since water does not allow for efficient gas exchange (Armstrong et al. 1994). For the constantly flooded treatment, each cup of sand was filled with 9.05 mL of distilled water to completely submerge the sand; cups were refilled as necessary. Seeds in cups in the intermittently flooded treatment were also filled with 9.05 mL of water, but allowed to evaporate and stay dry for 3 days before being refilled. The control group experienced constant moist sand. I added distilled water until the sand was dark down to where the seeds were buried. Distilled water was added as needed to keep moisture level constant. The trays were rotated or moved to a different shelf daily to remove directional effects. Seeds were checked daily for emergence and the time to emergence was noted for every seed. After 17 days, seeds were removed from the growth chamber.