Function of Mentzelia nuda (Loasaceae) Postfloral Nectaries in Seed Defense

Mentzelia nuda is a coarse, short-lived perennial of the High Plains of North America. The flowers secrete nectar which serves as an attractant for pollinating bees. Postfloral nectar secretion attracts ants whose presence significantly enhances seed set. This is the first demonstration of advantage to postflowering activity of a floral nectary. RECENT STUDIES of nectaries that lie outside the flower (generally called "extrafloral nectaries") indicate that, like floral nectaries, they are involved in plant-insect mutualism. Whereas floral nectaries facilitate outcrossing by rewarding pollinators, extrafloral nectaries attract insects, especially ants, whose activities decrease herbivore damage to the plant (Bentley, 1977a, b; Deuth, 1977; Keeler, 1977, 1980; Tilman, 1978; Inouye and Taylor, 1979; Koptur, 1979; O'Dowd, 1979; Pickett and Clark, 1979). Floral nectar varies greatly in nectar content (Percival, 1962; Baker and Baker, 1973, 1975), and attracts a diversity of animals, including birds, bats, butterflies, moths, flies, bees, and beetles (Percival, 1965; Faegri and van der Pijl, 1979; Proctor and Yeo, 1973). Extrafloral nectaries have been shown to attract ants, wasps, flies and beetles (Putman, 1963; Keeler, 1978). When floral and extrafloral nectaries are functioning simultaneously on the same plant, visitors to the two nectary types are distinct (Keeler, 1977, 1980). This is probably a result of both structure and nectar quality. The visual and olfactory attractants of the flower are lacking at extrafloral nectaries, while the position of the extrafloral nectaries often excludes potential pollinators, especially vertebrates and large lepidoptera. Floral nectar has been shown to be attractive to ants (Feinsinger and Swarm, 1978; Schubart and Anderson, 1978), but the chemical content of floral and extrafloral nectar appears to differ (Keeler, 1977, 1980; Baker, Opler and Baker, 1978), suggesting differential attractiveness may exist (but see Koptur, 1979). 1 Received for publication 9 February 1980; revision accepted 22 June 1980. I thank I. Baker for identifying the nectar constituents, and W. La Berge, C. O'Brien, J. Ballard, and B. Ratcliffe for identifying the insects. I thank R. Schmidt for references to "post floral nectaries;" R. Kaul and M. Bolick provided moral support and taxonomic clarification; S. Cantwell assisted with data analysis. Nectaries that reward pollinators, but continue to function after the corolla has fallen and the fruit develops, have been described (Daumann, 1932, 1974; Bentley, 1977a; Faegri and van der Pijl, 1979), but their function has not been determined. The presence of a functioning nectary on the developing fruit may confer a selective advantage. This paper reports a study to test this hypothesis, using the floral and postfloral nectaries of Mentzelia nuda. METHODS AND SITE-Mentzelia nuda (Pursh.) T.&G. (Loasaceae) (Fig. 1, 2) is a tall (to 1 m) coarse herb of disturbed sites from eastern Montana and the western Dakotas to west Texas, New Mexico and Arizona. It is a short-lived perennial and blooms from July to September. This study was carried out at the Cedar Point Biological Station, Keith Co., Nebraska, in the summer of 1978 and 1979. Nectar was collected with micropipettes and transferred onto filter paper. Nectar content was determined by I. Baker, Dept. of Botany, University of California, Berkeley. Pollinators and herbivores were observed and collected. Identification was provided by B. Ratcliffe, Nebraska State Museum, Lincoln, and W. E. La Berge, Illinois Natural History Survey, Urbana. Weevils were identified by C. O'Brien, Florida A and M University, Tallahassee. J. Ballard, University of Nebraska, Lincoln, identified the ants. Voucher specimens are in the possession of the individuals who identified the insects and in the Nebraska State Museum. Development of 87 marked fruits was followed to determine length of developmental period and natural levels of seed production. The potential effect of ants was determined by banding with Tanglefoot resin to exclude ants from the plants. At the end of the growing season, ripe pods of experimental and control plants were collected and scored for damage 295 Keeler in American Journal of Botany (February 1981) 68(2). Copyright 1981, Botanical Society of America. Used by permission.