Portal de Periódicos da CAPES

Pollen Deposition and Removal by Bees Visiting Two Tree Species in Panama

1987; Wiley; Volume: 19; Issue: 1 Linguagem: Inglês

10.2307/2388460

1744-7429

Allison A. Snow, David W. Roubik,

Plant biochemistry and biosynthesis

In lowland forests of Panama, Cochlospermum vitifolium and Cassia reticulata are nectarless, buzz-pollinated trees that require insect visitation to set seed; Cassia is self-compatible. Large anthophorid bees (Centris and Xylocopo spp.) were the primary pollinators of both species. Single visits almost always resulted in pollen deposition, and losses to pollen thieves were minimal. Bees did not deposit heterospecific pollen, despite obligate nectar foraging at other species. Bee species differed in the amount of pollen deposited and removed per visit; at Cassia flowers Centris longimana deposited more pollen per amount removed than did Centris inermis. These differences in pollination effectiveness may be relatively unimportant in terms of plant reproductive success, however. Seed production did not appear to be limited by pollinator service, and Cassia often received surplus pollen during single visits. MANY RECENT POLLINATION STUDIES have attempted to identify legitimate pollinator species from an array of flower visitors (e.g., Augspurger 1980, Motten et al. 1981, Tepedino 1981, Schemske & Horvitz 1984). Except in cases of obvious nectar or pollen robbing (McDade & Kinsman 1980, Roubik 1982, Renner 1983, Roubik et al. 1985), the relative contributions of different flower visitors to seed production are difficult to determine. Two major components of pollinator service are visitation frequency and effectiveness of pollen transfer during each visit (Motten et al. 1981, Schemske & Horvitz 1984). In this paper we compare pollen transfer during single visits by a range of potential pollinator species. Effective pollination may be evaluated by measuring the amount of pollen deposited or subsequent seed set after visits by specific pollinators. Both methods entail problems, such as uncertain pollen quality or loss of sam- ples during fruit development, but they are superior to indirect measures based solely on pollen load or visitation frequency of different pollinators (e.g., Macior 1969, Feinsinger 1978, Kevan & Baker 1982, Paton & Ford 1983). Two advantages of examining stigmas directly are that heterospecific pollen can often be detected (Waser 1978a, Brown & Kodric-Brown 1979, Feinsinger et al., 1986) and conspecific grains can be counted. The amount of pollen removed from a flower is usually impossible to quantify (but see Strickler 1979, Bertin & Willson 1980). It has been assumed that removal is positively correlated with pollen deposition (Spears 1983). Here we compare pollen transfer by bees visiting flowers of two trees, Cochlospermum vitifolium (Cochlo- spermaceae) and Cassia reticulata (Leguminosae: CaesalI Received 29 April 1985, revision accepted 12 September 1985. 2 Present address: Smithsonian Environmental Research Center, Box 28, Edgewater, MD 21037. pinioideae). Both have poricidal anthers and must be buzz- pollinated. As in other such species, pollen is released from the anthers when a visiting bee shivers its indirect flight muscles (Eickwort & Ginsberg 1980, Buchmann 1983). In Cochlospermum and Cassia, pollen is the only reward; solitary bees that collect it must obtain nectar elsewhere. We addressed the following questions using one or both species: Do flower visitors differ in the amounts of con- specific and heterospecific pollen deposited per visit? Do flower visitors remove different amounts of pollen, and if so, how does this relate to pollen deposition? How do observed differences in pollen transfer influence fruit and seed set? THE PLANTS.-Cochlospermum vitifolium is common in ear- ly successional habitats in Central America, particularly where annual burning suppresses other woody vegetation (Croat 1978, pers. obs.). This species is self-incompatible (Bawa 1974). The 10-12 cm wide yellow flowers have numerous anthers, a 4-5 cm long J-shaped style, and 1000-2500 ovules within a superior ovary (Frankie et al. 1983, Roubik et al. 1982). Flowers last 1-2 days and are borne on terminal inflorescences of trees ranging from 2 to 12 m tall. Most trees display less than 25 flowers per day during the flowering period (Roubik et al. 1982). In Panama leaves are shed during the early dry season (December-January) as trees begin to flower and set fruit. Many young fruits abort, but predispersal seed predation is not apparent (Bawa & Frankie 1983). The mature fruit is an egg-shaped capsule 6-8 cm long, usually containing 100-300 seeds (Roubik et al. 1982). Capsules remain on the trees and release their wind-dispersed seeds during the latter part of the dry season (March-April). In contrast to Cochlospermum, Cassia reticulata is self- compatible (see below), has about 40 ovules per flower, and often occurs in marshy areas that are seasonally flood- ed. Its growth form is that of a small, sometimes prostrate BIOTROPICA 19(1): 57-63 1987 57 This content downloaded from 207.46.13.127 on Fri, 14 Oct 2016 04:25:22 UTC All use subject to http://about.jstor.org/terms tree or large shrub. Cassia typically flowers throughout the dry season, overlapping with Cochlospermum in time but not in habitat (Croat 1978). Flowers are borne on showy spikes of golden-orange, bract-enclosed buds. Buds at the base of the inflorescence open first, and the flowers (3-4 per day) last a single day. Most pollen is contained in two 8-10 mm long anthers flanking 8 anthers that are very reduced in size. The style curves toward the left or right side of the inflorescence (enantiomorphy) and pre- sents the stigma a few millimeters away from the center of the flower. Carpels of pollinated flowers enlarge rapidly following anthesis, but at our study site many young fruits were completely or partially eaten by unidentified insect herbivores. Mature fruits are 20-30 cm long and contain an average of 36 seeds (see below). As the fruits dry, they dehisce along both sides; the seeds remain at- tached by shrivelled funiculi until they fall to the ground.