Pollination ecology of the tropical weed Triumfetta semitriloba Jacq. (Tiliaceae), in the South-eastern Brazil

Ecologia da polinização da planta invasora, Triumfetta semitriloba Jacq. (Tiliaceae), no sudeste do Brasil

POLLINATION ECOLOGY OF THE TROPICAL WEED Triumfetta semitriloba Jacq. (TILIACEAE), IN THE SOUTH-EASTERN BRAZIL

ROSANE G. COLLEVATTI,¹ LÚCIO A. O. CAMPOS² and ALEXANDRE F. DA SILVA³

¹Departamento de Biologia Animal, Universidade Federal de Viçosa, CEP 36570-000, Viçosa, MG, Brazil

²Departamento de Biologia Geral, Universidade Federal de Viçosa, CEP 36570-000, Viçosa, MG, Brazil

³Departamento de Biologia Vegetal, Universidade Federal de Viçosa, CEP 36570-000, Viçosa, MG, Brazil

Correspondence to: Rosane G. Collevatti, Departamento de Ecologia, Universidade de Brasília, CEP 70910-900, Brasília, DF, Brazil, e-mail:[e-mail]:[e-mail]

Received Feb 25, 1997 ¾ Accepted May 27, 1998 ¾ Distributed August 28, 1998

(With 2 figures)

ABSTRACT

This work aimed to study the pollination ecology of the tropical weed Triumfetta semitriloba Jacq. (Tiliaceae), in Viçosa, South-eastern Brazil, during the flowering season of 1993 and 1994. Two patches located in pasture (P1 and P2) and one patch in a forest trail (P3) were chosen and ten plants on each patch were sorted. The number of opened flowers were counted during one day, in each flowering month and patch. All observed flower visitors were identified and their behavior while visiting flowers was recorded. Frequency of visits to flowered branches was obtained and some pollinator individuals were captured for analysis of pollen load. Flowers are conspicuously yellow and actinomorphic, with five nectaries around the ovary base, and opened sequentially in the afternoon. Flower phenology followed a modified steady-state Gentry's pattern. The number of opened flowers was higher in P2, but differences between months were not homogeneous between patches. Considering behaviour when collecting pollen or nectar, which permitted impregnation of stigma with pollen, visiting frequency and percent of T. semitriloba pollen on pollen load (100% for all of them, except for Augochlorella michaelis which was 81%) the following species were the mainly pollinators: Augochloropsis cupreola, Augochlorella michaelis, Cressomiella aff. sussurans, Cressomiella sussurans, Cressomiella sp., Pseudocentron paulistana, Ceratinula sp1, Ceratinula sp2 and Ceratinula sp3, Melissodes sexcincta, Apis mellifera, Plebeia cf. nigriceps, Plebeia droryana. Frequency of pollinators visitation was not different between patches and not uniform during anthesis. There was a higher pollinator activity between 15:00 and 17:00 hr.

Key words: pollination, Triumfetta semitriloba, Tiliaceae, tropical bees, weed.

RESUMO

Ecologia da Polinização da Planta Invasora, Triumfetta semitriloba Jacq. (Tiliaceae), no sudeste do Brasil

Este trabalho tem por objetivo estudar a ecologia da polinização da planta invasora Triumfetta semitriloba Jacq. (Tiliaceae), em Viçosa, MG, Brasil, durante a estação de floração de 1993 e 1994. Foram escolhidas duas manchas de plantas localizadas em pastos abandonados (P1 e P2) e uma mancha em clareira de mata (P3). Dez plantas, em cada mancha, foram sorteadas. Durante um dia de cada mês de floração foi contado o número de flores abertas por planta. Todos os visitantes florais foram identificados e registrado o comportamento. Foi obtida a freqüência de visitas aos ramos floridos e alguns polinizadores foram coletados para análise da carga de pólen. As flores de T. semitriloba são actinomórficas com cinco nectários em torno da base do ovário, e abrem, seqüencialmente, no período da tarde. A fenologia floral seguiu o padrão "steady-state" modificado de Gentry. O número de flores abertas foi maior em P2, mas as diferenças entre meses não foram homogêneas entre as manchas. Considerando o comportamento enquanto coletavam pólen ou néctar, a freqüência de visita e a porcentagem de pólen de T. semitriloba na carga polínica (100% para todas as espécies de abelhas, exceto para Augochlorella michaelis que foi igual a 81%) as seguintes abelhas foram consideradas como principais polinizadoras: Augochloropsis cupreola, Augochlorella michaelis, Cressomiella aff. sussurans, Cressomiella sussurans, Cressomiella sp., Pseudocentron paulistana, Ceratinula sp1, Ceratinula sp2 e Ceratinula sp3, Melissodes sexcincta, Apis mellifera, Plebeia cf. nigriceps, Plebeia droryana. A distribuição de freqüência de visitas dos polinizadores não foi diferente entre manchas e não foi uniforme durante a antese. Houve uma maior atividade dos visitantes entre 15 e 17 horas.

Palavras-chave: polinização, Triumfetta semitriloba, Tiliaceae, abelhas tropicais, plantas invasoras.

INTRODUCTION

As the "renaissance botanist" Herbert Baker realized, the study of tropical weeds could provide valuable information for understanding mechanisms of breeding systems and life history evolution (Baker, 1965; Stebbins, 1989).

Breeding system, flower morphology, pollinator traits, biomass allocated to vegetative versus reproductive tissues are all correlated with life history, such as growth form, pattern of development, time to first breeding, number of reproduction events, offspring characteristics (Spira, 1980; Cruden & Miller-Ward, 1981; Queller, 1984; Brunet & Charlesworth, 1995; Damgaard & Abbott, 1995). Weed species, defined as colonizers associated with human disturbance activities (Baker, 1965), are always invading new habitats. Uncertainty of abiotic and biotic conditions found in each habitat, like nutrients and the "pool" of pollinators may select self-compatible and self-pollinated breeding systems, unspecialized flower morphology, a greater investment in seed production improving colonizing ability, a low pollen/ovule ratio, a low biomass investment in attractive structures, like flowers, pollen and nectar (Baker, 1967; Abrahamson, 1975, 1979; Cruden, 1976; Graumann & Gottsberger, 1988).

Weed species are usually visited by many bee species, being the main pollen and nectar sources in highly disturbed areas, such as abandoned pastures, agricultural areas, and road sides. Triumfetta semitriloba Jacq (Tiliaceae) is a perennial facultative autogamous weed shrub (Collevatti et al., 1997), occurring in tropical America. It is commonly found in abandoned pastures and other disturbed areas as roadsides and secondary forest boundaries and is highly resistant to drought stress, loosing almost all leaves, and resprouting when cut down (Collevatti, personal observation). Other study has already shown the importance of this plant species for many bee species in pastures (Cure at al., 1993), although it lacks information about pollination ecology, such as effective pollinators and breeding system.

In this paper we present the study about pollination ecology of T. semitriloba, and specifically address the following issues: (1) flower morphology and anthesis; (2) flower visitors and pollinators.

MATERIAL AND METHODS

Field work was conducted on three patches in Viçosa (20^o45'S, 42^o50'W), in South-eastern Brazil (Minas Gerais), during the flowering season of 1993 and 1994 (March to May), a region characterized by a nutrient poor soil, highly disturbed by agricultural activities. Two patches were pasture areas: P1, with approximately 200 m², with 85 flowering individual; P2, a rectangular shaped area of 100 m^2, with 30 flowering plants of T. semitriloba. The third patch, P3, with approximately 200 m²,was located in a secondary forest trail with 25 flowering plants of T. semitriloba, disposed in two parallel lines of plants, like a very narrow rectangle. All flowering plants of the three patches were marked and numbered.

Fifty flowers were randomly collected for morphological characterization. Flower anthesis was observed during one day, in each flowering month and patch. The number of opened flowers was counted periodically (30 min.), since the beginning of anthesis, at 13:00 hr, until the closure of all flowers (at about 19:00 hr). An analysis of covariance ANCOVA (Zar, 1974) was used to verify the effect of patch, month and time on the number of opened flowers. Mean differences were tested using an a posteriori Tukey test.

Some individuals of each flower visitor species were captured and identified, specially on the flowering season of 1993. Other individuals were eventually collected, on 1994, to confirm visual identification on the field. The activity and behavior of each flower visitor, while visiting flowers, were recorded. As visiting frequency to an individual flower was extremely low, it was observed the frequency of visits to flowering branches, in each month and patch. Five branches per plant were observed, and the frequency of visits was calculated as the number of visits per number of branches observed. A Kolmogorov-Smirnov Two Sample test was used to analyze the difference in visiting frequency between time, patches and flowering months. A Kolmogorov-Smirnov One Sample Test (Zar, 1974) was used to test uniformity of visits during anthesis time.

Percentage of T. semitriloba pollen grains on the pollen load of the most frequent bee species was determinate. Ten up to 15 female bees were captured per species, and pollen cleaned from scopae was stored in glacial acetic acid and counted under stereoscopic microscope.

RESULTS

T.semitriloba flowers are actinomorphic and conspicuous yellow (Fig. 1). Despite the existence of five nectaries on the base of ovary, nectar secretion is rather small, therefore it was not possible to collect nectar for analysis.

Flowers opened sequentially, from 13:00 to 16:00 or 16:30 hr, when the first opened flowers started to close (Fig. 2). The moment of complete closure differed for patches and flowering months. Flowers opened and closed early in P3, the secondary forest trail population. Number of opened flowers varied between patches, months and with time of anthesis (Table 1). Mean number of opened flowers was higher in P2 than in P1 and P3, when we get the three months of flowering (Table 2), but differences between months were not clear-cut.

Sixty three species of insects were registered on flowers of T. semitriloba (Appendix I). Bee species of the family Halictidae were the most abundant visitors (N = 10), followed by Megachilidae (N = 8) and Anthophoridae (N = 7) (Appendix I). Apidae and Colletidae were poorly represented. Concerning the three patches, bee species richness (S) was higher in P1 (S = 17), followed by P2 (S = 14) and P3 (S = 6). None of the bee species were found in the three patches ¾ P1 and P2 shared some species, probably due to habitat resemblance (both were pasture areas). Considering behavior when collecting pollen or nectar, which allowed impregnation of the stigma with pollen, together with visiting frequency and percent of T. semitriloba pollen on pollen load (100% for all of them, except for Augochlorella michaelis which was 81%) the following species were the main pollinators: Augochloropsis cupreola, Augochlorella michaelis, Cressomiella aff. sussurans, Cressomiella sussurans, Cressomiella sp., Pseudocentron paulistana, Ceratinula sp1, Ceratinula sp2 and Ceratinula sp3, Melissodes sexcincta, Apis mellifera, Plebeia cf. nigriceps, Plebeia droryana. The other species were considered as eventual pollinators (low visiting frequency): Augochloropsis sp., A. aurifluens, A. cf. argentina, A. cf. cleopatra, A. electra, Augochlora esox, A. foxiana, Hypanthidium rubiventris, Cressomiella bertonii, Megachile tuberculifera, Pseudocentron cf. aetheria, or opportunists (nectar collection without pollination), Dialictus sp., Xylocopa suspecta, Paratetrapedia sp1 and sp2, Ptiloglossa cf. pretiosa and Ptiloglossa sp.

Frequency of bee visitation, during anthesis, was not different between patches, considering the mainly pollinators (p > 0.10, for all species). Pooling the patches where the species occurred, frequency of bee visitation was not uniform during anthesis (Table 3). Visitation frequency was higher between 15:00 and 17:00 hr. There was no visiting between 13:00 and 14:00 hr, and after 18:30 hr.

DISCUSSION

Triumfettasemitriloba can be considered a facultative autogamous species, ripening fruits and viable seeds even in the absence of pollinators (Collevatti et al., 1997). A low pollen/ovule ratio agreed with this result (Collevatti, 1995), following Cruden (Cruden, 1976; Cruden & Miller-Ward, 1981). Other characteristics agree to the expected by theory for colonizers species: the flower presents polifilic characteristics (Faegri and van der Pijl, 1971), a low investment in attractive structures, such as nectar, pollen and showy petals (Cruden, 1976; Cruden and Miller-Ward, 1981; Graumann and Gottsberger, 1988).

Actinomorph symmetry, with periant free components allow easy access to resources (pollen and nectar), and really attract many species of insects. Although individual flowers are very small to be attractive from a distance, their showy color (yellow), the distribution of individual plants within patches, and flowering synchronism within-plant and patch may result in a conspicuous resource for bee pollinators (Heinrich and Raven, 1972; Augspurger, 1980).

Flowering phenology can be classified as a modified steady state (Gentry, 1974), with a long flowering season (three months), and a high number of flowers opening every day. Variance in flower opening, between individual plants and patches, was too high due to differences in plant size or nutritional and water stress (Weiner, 1988; Wagner 1989). Plants in P1 were submitted to water stress during dry season, and lost their leaves. Patch P2 was situated along a stream and P3 in a forestry gap, and plants did not lose leaves. Besides this, P3 was similar to P1, in mean number of opened flowers, probably because of a higher allocation of resources to vegetative structures, due to a low light incidence inside the forest. It was observed that leaves in P3 were larger than on P1 and P2 (Collevatti, personal observation).

Only Apoidea species were effective pollinators. Therefore, Formicidae species such as Zacryptocerus pusillus and Pseudomyrmex gracilis, that visit the extrafloral nectary of T. semitriloba (Collevatti, 1995), visited flowers to collect nectar. Hence, while passing on the flower, some pollen grain may reach the stigma. Chrysomelidae beetles, Pseudodiabrotica sp1 and Pseudodiabrotica sp2, were the main flower herbivores, damaging the petals, sepals, anthers and stigma. It is possible that some pollen grains carried by these species reach the stigma, but pollination may be counteract by stigma damage. Activity of these species on flowers did not displace bee species, like Augochlorella michaelis, Plebeia cf. nigriceps or Ceratinula spp.

Although pollen was the main resource found in T. semitriloba for bees pollinators, many species visited flowers for nectar. Pollen flowers are specially visited by solitary bees (Heinrich, 1975), that gather pollen for their nest, and nectar requirement is lower than for social bees, that collect nectar for the colony. Nevertheless, flower choice may be correlated with body size, which is correlated with energetic requirements (Heinrich & Raven, 1972; Heinrich, 1975). Bees with larger body size need more energy for metabolism and flight, hence this kind of bee may be less frequent than small sized bees in flowers with low content of nectar. The pollinators of T. semitriloba were mainly solitary or subsocial species (Augochlorini); just three species were eusocial: Apis mellifera, Plebeia droryana and Plebeia cf. nigriceps. The two species of Plebeia foraged in groups of four or five individuals per flower. Bees of larger body size were not common, and three species, Xylocopa suspecta, Ptiloglossa sp. and Ptiloglossa cf pretiosa were not effective pollinators.

The commonest bees Melissodes sexcincta, Pseudocentron paulistana, Cressomiella sussurans, Cressomiella aff. sussurans and Cressomiella sp. (medium sized bees) collected nectar while collecting pollen. There was not a temporal switch in pollen and nectar collection. Two male individuals of Melissodes sexcincta were observed collecting nectar, but visited only a small number of flowers, like a sampling behavior. Even bees that collected pollen and nectar with the same frequency (Augochloropsis cupreola and Ceratinula sp1) did not switch nectar and pollen collection periods. Probably, all these species were collecting nectar in another patch of flower species, if nectar quantity of T. semitriloba was not sufficient (Zimmerman, 1982b; Plowright & Laverty, 1984).

Bees that collected nectar with the same or higher frequency than pollen, Augochlorella michaelis, Ceratinula sp1, Ceratinula sp2, Ceratinula sp3 and Dialictus sp., were bees of small body size and were collected in P3, excepting Dialictus sp.

Although T. semitriloba flowering season matches with other plant species, the main pollinators were constant to its flowers, inferring from pollen load. Flower constancy may be established by resource presentation (attractiveness), manipulation facility, reward quantity or quality (Free, 1970; Heinrich, 1975). These factors provide information about the resources that are associated to some flower characteristics, such as color (specially for bees), forming an "image search", decreasing the cost of flower manipulation or resource recognition (Shettleworth, 1984; Krebs & Kacelnik, 1991). For T. semitriloba pollinators, flower constancy may be conditioned by flowering synchrony (conspicuous and abundant resource) and an easy access to resource (polifilic flower). This may explain why some generalists bee species, such as Apis mellifera and Plebeia spp. presented 100% of T. semitriloba pollen grains on pollen load. Many studies have shown that A. mellifera prefers flower constancy foraging strategy than others (e.g. cost or variance minimization or intake maximization) (Wells & Wells, 1983, 1984, 1986).

Bees visiting frequencies where not uniform but there was not a clear-cut pattern. Frequencies were low, for all species, probably due to flower opening and resource availability patterns. Flowers opened sequentially and the number of flowers increased until 16:00 hr. As flowers were opening a low number of visits depleted all pollen resource in a flower. Therefore, the high number of opened flowers did not reflected a high level of resource, since a great number of flowers were already depleted, showing that number of flowers is not a good predictor of pollen resource level (Tepedino & Staton, 1982; Zimmerman & Pleasants, 1981). Hence, flower opening pattern (sequential) and depletion of pollen by bee visits may cause a patchy distribution of resource, with "hot spot" of just opened or non-visited flowers with pollen and "cold" already visited flowers, and a low rate of visit per flower.

This pattern may have an important effect on pollinators behavior and hence, on gene flow. Optimal foraging theory predicts short distance flight, minimizing costs of flight thus, maximising fitness (Heinrich, 1975; Pyke, 1984). Distances may be shorter for bees collecting pollen (Zimmerman, 1982a). Therefore, autogamy and geitonogamy rate would be higher when bees fly to nearest neighbors, such as T. semitriloba pollinators do (Collevatti, 1995). Plants may increase xenogamy driving pollinators to visit more flowers and fly longer distances - by a low nectar production, a high temporal and spatial variability in nectar production, a low flower production per plant (Heinrich & Raven, 1972; Augspurger, 1980; Zimmerman, 1988; Ott et al., 1985). For T. semitriloba, which produces many flowers per individual plant, inter-plant movement, thus rate of xenogamy, could be increased by a sequential opening and a randomly patch distribution of resource.

Acknowledgements — This study is part of the MSc. Thesis of the first author, and was supported by the Universidade Federal de Viçosa and a grant from CAPES, whose assistance we gratefully acknowledge. We wish to thanks Dr. J. M. Camargo for bee species identification; Dr. P. S. Fiúza for Hemiptera identification; to USDA researches for identification of the following groups: J. M. Kingsolver, Bruchidae; A. L. Norrbom, Diptera; A. S. Menke, Vespidae and Chrysididae; D. R. Smith, Formicidae; R. E. White, Chrysomelidae; Dr. F. S. Lopes, Dr^a M. E. C. Amaral and Dr. J. H. Schoereder for suggestions; M. C. Vieira for field and laboratory assistance.

• ABRAHAMSON, W. G., 1975, Reproductive strategies in dewberries. Ecology, 56: 721-726.
• ABRAHAMSON, W. G., 1979, Patterns of resource allocation in wildflower populations of field and woods. Am. J. Bot., 66: 71-79.
• AUGSPURGER, C. K., 1980, Mass-flowering of a tropical shrub (Hybanthus prunifolius): influence on pollinator attraction and movement. Evolution, 34: 475-488.
• BAKER, H. G., 1965, Characterisitcs and modes of origin of weeds. In: H. G. Baker & G. L. Stebbins (eds.), The Genetics of Colonizing Species Academic Press, New York, pp. 147-172.
• BAKER, H. G., 1967, Support for Baker's Law as a rule. Evolution, 21: 853-856.
• BRUNET, J. & CHARLESWORTH, D., 1995, Floral sex allocation in sequentially blooming plants. Evolution, 49: 70-79.
• COLLEVATTI, R. G., 1995, Comportamento de Forrageamento dos Visitantes florais de Triumfetta semitriloba Jacq. (Tiliaceae) em Viçosa, Minas Gerais Universidade Federal de Viçosa, Viçosa, MG, Brasil (dissertação de Mestrado).
• COLLEVATTI, R. G.; AMARAL, M. E. C. & LOPES, F. S., 1997, Role of pollinators in seed set and a test of pollen limitation hypothesis in the tropical weed Triumfetta semitriloba (Tiliaceae). Rev. Biol. Trop, 45: 1401-1407.
• CRUDEN, R. W., 1976, Intraspecific variation in pollen-ovule ratios and nectar secretion-preliminary evidence of ecotypic adaptation. Ann. Missouri Bot. Gard., 63: 277-289.
• CRUDEN, R. W. & MILLER-WARD, S., 1981, Pollen-ovule ration, pollen-size, and the ratio of stigmatic area to the pollen-bearing area of the pollinator: an hypothesis. Evolution, 35: 964-974.
• CURE, J. R., BASTOS FILHO, G. S., OLIVEIRA, M. J. F. & SILVEIRA, F. A., 1993, Levantamento de abelhas silvestres na Zona da Mata de Minas Gerais. I ¾ Pastagem na região de Viçosa (Hymenoptera: Apoidea). Revista Ceres, 40: 131-161.
• DAMGAARD, C. & ABBOTT, R. J., 1995, Positive correlations between selfing rate and pollen-ovule ratios within plant populations. Evolution, 49: 214-217.
• FAEGRI, K. & van der PIJL, L., 1971, The principles of pollination ecology, Pergamon Press, Oxford. 291 p.
• FREE, J. B., 1970, The flower constancy of bumblebees. J. Anim. Ecol, 39: 395-402.
• GENTRY, A. H., 1974, Flowering phenology and diversity in tropical Bignoniaceae. Biotropica, 6: 64-68.
• GRAUMANN, S. & GOTTSBERGER, G., 1988, Reproductive strategies in allogamous and autogamous Senecio species. Lagascalia, 15: 673-679.
• HEINRICH, B., 1975, Energetics of pollination. Ann. Rev. Ecol. Syst., 6: 139-170.
• HEINRICH, B. & RAVEN, P. H., 1972, Energetics and pollination ecology. Science, 176: 597-602.
• KREBS, J. R. & KACELNIK, A., 1991, Decision-making. In: J. R. Krebs & N. B. Davies (eds.). Behavioural ecology: an evolutionary approach 2^a ed., Blackwell Scientific Publ, Oxford, pp. 91-121.
• OTT, J. R., REAL, L. A. & SILVERFINE, E. M., 1985, The effect of nectar variance on bumblebee patterns of movement and potential gene dispersal. Oikos, 45: 333-340.
• PLOWRIGHT, R. C. & LAVERTY, T. M., 1984, The ecology and sociobiology of bumblebees. Ann. Rev. Entomol., 29: 175-199.
• PYKE, G. H., 1984, Optimal foraging theory: a critical review. Ann. Rev. Ecol. Syst., 15: 523-575.
• QUELLER, D. C., 1984, Pollen-ovule ratios and hermaphrodite sexual allocation strategies. Evolution, 38: 1148-1151.
• SPIRA, T. P., 1980, Floral parameters, breeding systems and pollinator type in Trichostema (Labiatea). Am. J. Bot, 67: 278-284.
• SHETTLEWORTH, S. J., 1984, Learning and behavioural ecology. In: J. R. Krebs & N. B. Davies (eds.), Behavioural ecology: an evolutionary approach, 2^a ed., Blackwell Scientific Publ, Oxford, pp. 170-196.
• STEBBINS, G. L., 1989. Introduction: Herbert Baker: A renaissance botanist. In: J. H. Bock & Y. B. Linhart (eds.), The Evolutionary Ecology of Plants, Westview Press, San Francisco, pp. 1-4.
• TEPEDINO, V.J. & STATON, N.L., 1982, Estimating floral resources and flower visitors in studies of pollinator-plant communities. Oikos, 38: 384-386.
• WAGNER, L.K., 1989, Size dependent reproduction in Poa annua and P. pratensis In: J.H. Bock & Y.B. Linhart (eds.), The Evolutionary Ecology of Plants Westview Press, San Francisco, pp. 273-284.
• WEINER, J., 1988, The influence of competiton on plant reproduction. In: J.L. Doust & L.L. Doust (eds.), Plant Reproductive Ecology: Patterns and Strategies, Oxford University Press, Oxford, pp. 228-245.
• WELLS, H. & WELLS, P. H., 1983, Honey bee foraging ecology: optimal diet, minimal uncertainty and individual constancy? J. Anim. Ecol., 52: 829-836.
• WELLS, H. & WELLS, P. H., 1986, Optimal diet, minimal uncertainty and individual constancy in the foraging of honey bees, Apis mellifera J. Anim. Ecol., 55: 881-891.
• WELLS, P. H. & WELLS, H., 1984, Can honey bees change foraging patterns? Ecol. Ent., 9: 467-473.
• ZAR, J. H., 1974, Biostatistical analysis, Prentice-Hall, Englewood Cliffs.
• ZIMMERMAN, M.,1982a, The effect of nectar production on neighborhood size. Oecologia, 52: 104-108.
• ZIMMERMAN, M., 1982b, Optimal foraging: Random movement by pollen collecting bumblebees. Oecologia, 53: 394-398.
• ZIMMERMAN, M., 1988, Nectar production, flowering phenology, and strategies for pollination. In: J. L. Doust & L. L. Doust (eds.), Plant reproductive ecology: Patterns and strategies, Oxford University Press, New York, pp. 157-178.
• ZIMMERMAN, M. & PLEASANTS, J. M., 1981, Competition among pollinators: quantification of available resources. Oikos, 38: 381-383.

Publication Dates

• Publication in this collection
  05 July 2001
• Date of issue
  Aug 1998

History

• Accepted
  27 May 1998
• Received
  25 Feb 1997