Pollination biology of Opuntia imbricata (Cactaceae) in southern Colorado JONNENE D. MCFARLAND Pueblo Zoological Society at Pueblo Zoo Erlrrur~ce, P.O. Box 1245, Pueblo, CO 81002, U.S.A. PETER G. KEVAN' Departmerzt of Erzvirorzrnental Biology, Urziversity of Guelph, Guelph, Orzt., Canada NIG 2W1 AND MEREDITH A. LANE Departrnent of Environmental Populatiorz and Organismic Biology, University of Colorado, Boulder, CO 80309, U.S.A. Received October 27, 1987 MCFARLAND, J. D., KEVAN, P. G., and LANE, M. A. 1989. Pollination biology of Opl~rztia irnbricata (Cactaceae) in southern Colorado. Can. J. Bot. 67: 24-28. The floral phenology, compatibility system, and pollinator fauna of Opuntia irnbricata (Haw.) DC. in southern Colorado were studied. The plants bloom for approximately 4 weeks in June and July. Op~inria irnbricata sets numerous seeds when it is open-pollinated or experimentally cross-pollinated; it is self-incompatible and not apomictic. The most effective pollina- tors were found to be medium to large bees of the genera Diadasia and Lirhurge; beetles are ubiquitous on flowers of 0. irnbricata, but probably do not effect pollination because they rarely move from flower to flower and usually restrict their activities to the stamens and base of the inside of the flowers. MCFARLAND, J. D., KEVAN, P. G., et LANE, M. A. 1989. Pollination biology of Opuntia imbricata (Cactaceae) in southern Colorado. Can. J. Bot. 67 : 24-28. La phCnologie florale, le systkme de compatibilitk et la faune pollinisatrice du Op~lnria irnbricata (Haw.) DC. dans le sud du Colorado ont CtC CtudiCs. La floraison dure environ 4 semaines en juin et juillet. Le Opurztia imbricata forme de nom- breuses graines lorsque la pollinisation est effectuCe par allogamie naturelle ou expkrimentale; il est auto-incompatible et non- apomictique. Les pollinisateurs les plus efficaces Ctaient les abeilies de taille moyenne ii grande des genres Diadasia et Lithurge, les colCoptenes omniprCsents sur les fleurs du 0. irnbricata n'ont probablement aucun effet sur la pollinisation car ils vont rarement d'une fleur ii une autre et limitent d'habitude leur activitt aux Ctamines et ii la base interne de la fleur. [Traduit par la revue] Introduction Opuntia imbricata (Haw.) DC. is a shrubby cylindropuntia that occurs in vast stands across semiarid grasslands in the southwestern United States. It produces large rose-pink flowers at the ends of branches (Benson 1969; Kinraide 1978). At the northern edge of its range, in southeastern Colorado, the blooming period is from early June to midJuly. Each flower has numerous intergrading sepaloid and petaloid parts, numerous thigmotropic stamens attached to the epigynous hypanthium, and a lobed stigma that is elevated above the stamens (Benson 1969; Weniger 1978). Although the pollination biology of several platyopuntia species has been studied (Daumann 1930; Beutelspacher 1971; Barrows et al. 1976; Grant et al. 1979; Grant and Hurd 1979; Grant and Grant 1979; Parfitt and Pickett 1980; Spears 1987; Osborn et al. 1988), the pollinators and breeding system of 0 . imbricata have not been reported previously. Grant and Hurd (1979) suggested that this species is bee-pollinated; how- ever, there is only one published record of a bee visitor, Lithurge apicalis, to the flowers of this plant (Cockerel1 1300). The purpose of this study was to investigate the pollina- tion biology of 0. imbricata by (i) determining whether it is self-compatible; (ii) examining floral features with respect to pollination and pollinators; (iii) identifying flower visitors; (iv) observing the visitors' behavior; and (v) typifying their roles in pollination. Materials and methods A population of ca. 100 plants of Opctnria imbricara located on the 'Author to whom reprint requests should be sent. campus of the University of Southern Colorado, Pueblo, CO, was studied during the summers of 1981 and 1982. Development of individual flowers was recorded photographically from 14 June to 16 June 1981. Nectar was analysed for sugars and amino acids (Baker and Baker 1976, 1983) and pollen for starch, lipids, amino acids, and protein (Baker and Baker 1979; Simpson and Neff 1983). The period of stigma receptivity was determined by using hydrogen peroxide that bubbles in the presence of peroxidase liberated from the receptive stigma (Zeisler 1938; see also Galen and Plowright 1987). The stamens in each of 10 flowers, each from a different plant, were counted. Within the population, six plants with 59 flowers were chosen. Of these, 52 flowers were bagged (with nonwoven plastic pollination bags by Duraweld Ltd., England) before anthesis and then subjected to four experimental treatments as follows: (i) otherwise untreated (n = 21 on six plants); (ii) cross-pollinated by hand (rz = 10 on four plants); (iii) self-pollinated by hand (n = 21 on four plants); (iv) stigmas were removed from an additional seven flowers on three plants before anthesis and these flowers were not bagged. Control flowers (n = 26 on six of the same or different plants) were also chosen and marked as buds. Ripe fruits were collected from treated and control plants and seeds per fruit were counted. Control and experimental groups were compared for significant differences by Student's t-test performed on square root transformed data. Cut seeds were tested for viability using the tetrazolium method (Lakon 1949; Kozlowski 1972; Hartman 1981). Bee visitors to the flowers were observed and visitation frequency per flower per hour throughout the day was calculated by recording visitations for 10 min each h from 0900 until 1500 h throughout the blooming period of the individual flowers under investigation. Representatives of each species of visitor were collected using a stan- dard entomological collecting net, captured directly in glass vials, and by aspirator. Pollen was gently scraped from the ventral abdominal and thoracic regions of each collected insect onto a glass microscope slide. Prinlcd in Canada / Ilnprin~C au Canada