The component parts of modular organisms often show interspecific variation in their longevity. In plants, the flower is an example of such a structure. Models are developed in this paper to predict optimal floral longevity (the optimal length of time that flowers should remain open and functional) under a variety of conditions. A tradeoff involving allocation of resources to floral construction versus floral maintenance is assumed. The main model variables are the rate at which pollen and seed fitness accrue over time (fitness-accrual rates) and the daily cost of maintaining an existing flower relative to the cost of constructing a new one (floral maintenance cost). Long-lived flowers are selected when fitness-accrual rates and floral maintenance costs are low, whereas short-lived flowers are selected when fitness-accrual rates and floral maintenance costs are high. Dichogamy favors longer-lived flowers relative to homogamy, whereas nonindependence among flowers in their attractiveness to pollinators (attraction to flower clusters) selects for shorter-lived flowers. Reduction in floral maintenance costs later on in the flower's life favors longer-lived flowers. Observations on the dissemination and receipt of pollen in individual flowers over time, together with measurements of corolla respiration and nectar sugar production rate are required to test the model quantitatively. The parameters important to the evolution of optimal floral longevity (i.e., maintenance and construction costs, and fitness-accrual rates) may be general features of evolution of optimal longevities of other repeated structures.
Keywords: Female fitness; male fitness; plant reproduction; pollination biology.
© 1995 The Society for the Study of Evolution.