The £owers of most angiosperm species are hermaphroditic. Spatial separation of male and female organs within a £ower (hercogamy) is a common character traditionally interpreted as an adaptation to reduce intra£oral self-fertilization, one potential cost of hermaphroditism. Another possible cost that may lead to selection for hercogamy is physical interference between male and female £oral functions. Here, I present evidence demonstrating the role of a £oral character in reducing female interference with male function. The bi-lobed stigma of the bush monkey£ower closes after receiving pollen, causing increased spatial separation of the anthers and stigma (`movement' hercogamy). Experimental manipulations show that £owers with closed stigmas export more than twice as much pollen to other £owers as those in which the stigma is prevented from closing. However, stigma closure only minimally reduces the potential for intra£oral self-pollination. This study provides the ¢rst experimental evidence that selection to reduce intra£oral male^female interference can be a strong selective force and can drive the evolution of £oral characters usually interpreted as mechanisms to reduce self-fertilization.

The bilobed stigma of many species in the order Scrophulariales closes in response to touch by an animal pollinator. In hummingbird-pollinated bush monkey flower, Mimulus aurantiacus (Scrophulariaceae), closure is rapid, occurring within seconds of tactile stimulus. We investigated the proximate causes of stigma closure and subsequent reopening in M. aurantiacus, as well as potential costs and benefits of stigma closure for female fitness. Stigma closure is elicited by both touch and pollen, but closure in response to pollen is much slower, requiring 0.5-1.5 h. Stigmata reopen within 2.5-4.5 h if touch, but no pollen, is applied. Upon receipt of pollen, most stigmata remain closed for the remaining lifetime of the flower, even if less pollen is received than is needed for full seed set. Those stigmata that do reopen after pollination generally require between 20 and 28 h to do so, much longer than for unpollinated stigmata. Reopening after pollination appears to be a response to low seed set rather than to low pollen load. Natural pollination of stigmata manipulated to prevent closure shows that closure does not increase capture of pollen or seed set. In fact, closure reduces the average pollen load received by flowers. Despite this, there is no evidence that stigma closure has any negative effect on female fitness in terms of seed set or germinability. Hypotheses for the adaptive significance of stigma closure are discussed. Understanding proximate causes of stigma closure and reopening is essential in the evaluation of these hypotheses.

No abstract

No abstract

Wetland rapid assessment methods (RAMs) can provide a cost effective, scientifically defensible estimate of wetland and riparian condition for use in ambient and project monitoring in resource management and regulatory programs. Those who have chosen to develop a RAM to assess wetland and riparian condition are faced with a range of issues and important choices that they must make throughout the development process. This paper is intended as a practical guide to RAM development. Six basic stages in the RAM development process are discussed: (1) organize RAM development by identifying the intended applications, assessment endpoints, and geographic scope of the RAM and forming appropriate teams to advise and review the development process and its products; (2) build a scientific foundation for method development by conducting a literature review, choosing a wetland classification system, building conceptual models, and identifying the major assumptions underlying the model; (3) assemble the method as a system of attributes and metrics that describe a full range of conditions; (4) verify the ability of the method to distinguish between wetlands along a continuum of conditions; (5) calibrate and validate the method against sets of quantitative data representing more intensive measures of wetland condition; and (6) implement the method through outreach and training of the intended users. Important considerations within each of these stages lead to choices in accuracy, precision, robustness, ease of use, and cost. These are identified and the tradeoffs of the various options discussed. Experience with the ongoing development and implementation of the California Rapid Assessment Method (CRAM) is used to illustrate these stages and associated choices in RAM development.