The role of variable pollination environments in maintaining mixed mating systems is an active area of research. Dimorphic cleistogamy, in which a plant reproduces by both open, facultative outcrossing chasmogamous (CH) flowers and closed, cleistogamous (CL) flowers presents an excellent opportunity to study mixed mating. For example, plastic responses in allocation to an optimal floral type could serve as an adaptive strategy that maintains mixed mating under variable pollination environments. We tested for pollen limitation and plastic responses in allocation to different floral types under manipulated pollination conditions in the dimorphic cleistogamous, mixed mating annual, Triodanis perfoliata. Using a field population, we quantified pollen limitation, auto-fertility and plastic responses in the breeding system by measuring allocation to flower number and seed set of floral types. We found no evidence for pollen limitation for CH flowers, and CH flowers had low efficacy of autonomous selfing. Importantly, we found that T. perfoliata alters floral number following changes in pollination conditions, with pollen-supplemented plants having lower relative CH flower number than non-supplemented plants. Breeding system plasticity may allow for benefits from outcrossing through CH flowers, but also increased overall fitness through relatively cheap CL reproduction. After CH flowers receive pollen, subsequent production of CH flowers was reduced, which may be due to resource limitation. Our findings did not support a theoretical model predicting increased CH flowers with high pollination levels. These results increase our understanding of the role of pollination services and resource allocation in the maintenance of mixed mating systems, which also warrants further investigation.

Plant breeding systems can vary widely among populations, yet few studies have investigated abiotic factors contributing to variation across a broad geographic range. Here we investigate variation in reproductive traits of Triodanis perfoliata (Campanulaceae), a species that exhibits dimorphic cleistogamy, a condition in which individual plants have both closed (selfing: cleistogamous: CL) and open (selfing or outcrossing: chasmogamous: CH) flowers. Chasmogamous production is theorized to be more costly because CH flowers have a larger exposed surface area and thus are more likely to lose more water than CL flowers. We examine relationships between abiotic conditions (temperature, precipitation and soil characteristics) and variation in breeding systems across 14 widespread populations using ordinary least squares models. We found that a large proportion of breeding system variation was described by climate and soil variables (R2 = 0.65–0.92). These results support the hypothesis that variation in the environment drives variation in breeding system allocation. Our broad geographic analyses provide a framework for mechanistic studies of cleistogamy, and employ a novel approach for examining reproductive traits and environmental variation at large scales. Given that two major components of our models were temperature and precipitation, our study further emphasizes the potential for ongoing climate change to alter plant breeding systems.

The maintenance of outcrossing in cleistogamous plants that produce both open, facultatively outcrossing chasmogamous (CH), and closed, obligate selfing cleistogamous (CL) flowers is puzzling because CL reproduction is thought to be more reliable and less costly. A possible explanation for the maintenance of CH flowers is the avoidance of inbreeding depression. However, inbreeding depression for cleistogamous species has rarely been quantified. In this study, we estimate levels of inbreeding depression in plants from three populations of Triodanis perfoliata (L.) Nieuwl., a dimorphic cleistogamous annual, under greenhouse conditions. Estimates of inbreeding depression at multiple life stages in all three populations were low and often not different from zero. Inbreeding depression at specific life stages varied, with two populations showing later-acting inbreeding depression, which is also found in other selfing species. In two of the study populations, selfed CL progeny outperformed selfed CH progeny, indicating a flower-type effect. The low levels of inbreeding depression and the superior fitness of CL compared with selfed CH flowers that we observed make the maintenance of CH flowers in this system surprising, and suggest that other advantages of outcrossing CH flowers are likely responsible for maintaining mixed mating in this species.

No abstract

To teach the most central concepts in evolutionary biology, we present an activity in pollination biology. Students play the role of either pollinator or flower and work through a set of scenarios to maximize plant fitness. This “Pollination Game” facilitates critical and inquiry-based thinking, and we accompany each round of the exercise with a set of discussion questions and answers. We have piloted and fine tuned this exercise with high school students, and improved the exercise with the input of high school teachers at a teaching conference. The activity could easily be adapted for freshman undergraduate students.