Effect of nectar pillaging by native stingless bees (Hymenoptera: Apidae) in the abscission of flowers of Bougainvillea spectabilis Willd. (Nyctaginaceae)/Efeito da pilhagem de nectar por abelhas nativas sem ferrao (Hymenoptera: Apidae) na abscisao floral de Bougainvillea spectabilis Willd. (Nyctaginaceae).
The hypothesis of specialization in the evolution of pollination prevailed for a long time, but recently it is believed that pollination systems vary from a generalist form to highly specialized ones, beginning with the idea that a plant tends--to specialize for the most effective pollinator, inversely to what happens in the generalist trend the effectiveness is unexpected. In other words, the characteristics related with efficiency do not promote fidelity in pollination (BARRETO; FREITAS, 2007).
In general, flowers are attractive, with petals fused to form a floral tube, threads fused with the corolla, making possible the access to the nectar produced in the base of the flower nectary disk, allowing the floral visitors that present mouth parts with the length related to the size of the nectary chamber, to reach the nectar (CARVALHO et al., 2007b).
Many flowering plants need a pollinator agent to accomplish the exchange of pollen grains, guaranteeing genetic variability. They present available attractive resources, such as pollen, nectar and oils and important feeding sources for these agents (MALOOF; INOUYE, 2000; ROUBIK, 1982). These resources might attract several species that have no pollination, stealing the resources offered--the robbers (MALOOF, 2001). The pillage causes significant damages in the flower, causing premature abscission, and influencing the reproductive success of the species (ROUBIK, 1982; MOTHERSHEAD; MARQUIS, 2000).
Floral damages caused by opportunist species that do not accomplish pollination, among them native stingless bees, have been considered harmful in commercial species and are studied due to their economical importance by presenting a considerable decrease in productivity (BOICA JUNIOR et al., 2004). Pillaging rates by bees on native species such as Bignoniaceae Juss. and Rubiaceae Juss. can be considerable, and result in lower attractiveness to pollinating species (CARVALHO et al., 2007a e b).
Nyctaginaceae Juss. has three pollinator types: 1 Hymenoptera, visitors of Abronia Juss and Boerhavia L. flowers. The latter have typically melittophilous flowers, with coloration going from white to purple, and with nectaries four millimeters from the reproductive organs (AGUIAR; SANTOS, 2007; GONZALEZ; LOPEZ, 2004); 2--Lepidoptera family--Hesperiidae that pollinates Bougainvillea Choisy. This plant has psychophile flowers, with red coloration, going through yellow and purple (the individuals of Hesperiidae also act as pollination agents in Allionia L. and Abronia Juss). Still belonging to the order Lepidoptera, Sphingidae is the second group of pollinator agents in importance for Mirabilis L. (LEAL et al., 2001; GONZALEZ; LOPEZ, 2004); and 3--Trochilidae (hummingbirds) in Mirabilis froebelii (Behr) Greene that exhibits a combined system of pollination, in which the hummingbirds visit their flowers until the evening, when moths start visiting (LEAL et al., 2001; GONZALEZ; LOPEZ, 2004). The involucre of Mirabilis sp. is an effective barrier of nectar protection against bees (GONZALEZ; LOPEZ, 2004).
Bougainvillea (Nyctaginaceae), have 14 species designated usually as "bougainvilleas", which are native to South America and known as "primavera" or "flores de papel". The flowers are surrounded by bracts, which have various functions, such as insect attraction. When the fruits mature, the bracts lose their color (KOBAYASHI et al., 2007).
Bougianvillea spectabilis Willd. is a woody bush, native to the Northeast of Brazil, cultivated as an ornamental plant worldwide (LORENZI; SOUZA, 2001), presents small flowers of cream color surrounded by bracts, can reach, when adult, 5 to 10 meters in length, has a reproductive cycle at the end of the dry station and is pollinated by night moths, but the availability of nectar in its nectar ducts during the day propitiates a rich alimentary source for other insects (KOBAYASHI et al., 2007).
Carvalho et al. (2007a) recorded a great percentage of flowers that suffered pillage in Tocoyena formosa (Cham. & Schltdl.) K. Schum. ("jenipapo-bravo"), but in other native species these relationships are still not well understood, especially in B. spectabilis, which presents a lack of phenological, reproduction and floral development studies. The present work has as objective to evaluate the floral bee visitors and analyze whether the activity of nectar pillage by native bees influences the floral abscission of B. spectabilis.
Material and methods
The collections were performed in ten individuals of B. spectabilis in the city of Campo Grande, center-west Brazil. Four of the ten studied plants were located at the campus of the Federal University Mato Grosso do Sul (UFMS), which presents a remainder of savannah and riparian forest (20[degrees]27' S and 54[degrees]37' W, altitude of 530 m). The two areas belong to the Natural Reservation Private Patrimony of UFMS (RPPN/UFMS), a total of 36.5 ha, and the vegetation consists of Savannah (EITEN, 1979). The other six plants were located in the vicinity of the UFMS campus, in residential neighborhoods. The climate of the area is of Rainy Tropical Savannah (subtype Aw) (KOPPEN, 1948) with dry winters (May to August) and rainy summers (December to March). The annual mean precipitation is 1,532 mm and annual mean temperatures are between 20 and 22[degrees]C (EMBRAPA, 1985).
Each individual of B. spectabilis was considered as a sampled unit. Ten inflorescences were collected (Figure 1) on the ground and 10 in the plant, between May 4th and June 1st, 2009, with three-day intervals between each collection, totaling 10 repetitions. The total of closed flowers was evaluated (floral buttons) (FF), open flowers (FA), robbed closed flowers (FFP), flowers closed not robbed (FFNP), robbed open flowers (FAP) and opened flowers not robbed (FANP) on the ground and in the plant. For the statistical analyses, Student's t-test was used (ZAR, 1996) to verify whether there was a significant difference among the characteristics and among the flowers on the ground and the flowers of the plant. A simple linear regression was used to verify whether there was a relationship between the closed flowers and robbed closed flowers; open flowers and robbed open flowers, found on the ground and closed flowers and robbed closed flowers; open flowers and robbed open flowers, found in the plant. The statistical program STATISTICA for Windows was used (PETRERE-JUNIOR, 1993).
To verify the floral visitors and pillage activity, two individuals of B. spectabilis located in the UFMS campus were chosen. The activity and frequency of bee visitors was recorded during 12 hours of focal observation, and in each hour, 20 minutes were used to collect the bees with an entomological net (puca) and 40 minutes for observation of the visitors' behavior (robber or pollinator). The captured visitors were killed by ethyl acetate, set in entomological pins, and later identified through specialized bibliographical reference (SILVEIRA et al., 2002). The collected specimens were deposited in the Zoological Collection at UFMS, Campo Grande, Mato Grosso do Sul State, Brazil.
[FIGURE 1 OMITTED]
Results and discussion
Six species of bees were recorded visiting flowers of B. spectabilis, on June 6th, 2009 at the UFMS campus. The mean temperature was 20[degrees]C ([+ or -] 2[degrees]C), relative humidity of the air 70%, and mean wind speed 1.9 m [s.sup.-1]. Trigona spinipes (54.25%) and Paratrigona lineata (39.75%) were the most frequent species, and the least frequent were Augochlora sp. (2.4%), Tetragonisca angustula, Scaptotrigona sp. and Exomalopsis sp. (1.2% each) (Figure 2).
[FIGURE 2 OMITTED]
Prevalence of pillaging activity was observed for the more frequent species of bees, Trigona spinipes and Paratrigona lineata (Table 1). These species of native bees were recorded stealing the nectar by the floral base, opening holes or using the opened holes (Figure 3A, C and D). Augochlora sp. was recorded collecting pollen from open flowers, but not necessarily accomplishing pollination.
We recorded 2,874 flowers on the ground and 2,895 flowers in the plant, totaling 5,769 flowers. The robbed flowers showed damages in the floral base (Figure 1), where the nectary disks are.
There were significant differences relating to all the variables measured between the soil and the plant being the closed flowers, or Floral Buttons (FF) ([chi square] = 770, d. f. = 9, p = 0.001), Open Flowers (FA) ([chi square] = 824.25, d. f. = 9, p = 0.001), Open Flowers Robbed (FAP) ([chi square] = 490.25, d. f. = 9, p = 0.001), Open Flowers Not Robbed (FANP) ([chi square] = 59, d. f. = 9, p = 0.001), Closed Flowers Robbed (FFP) ([chi square] = 1831, d. f. = 9, p = 0.001) and Closed Flowers Not Robbed (FFNP) ([chi square]= 327, d. f. = 9, p = 0.001).
In relation to the flowers on the ground, FF represented 97.29% and the FA 2.71%; FFP 79.54% and FFNP 17.75%; FAP 2.47% and FANP 0.24% of the total. In relation to the flowers of the plant, FF represented 70.16% and the FA 29.84%; FFP 37.06% and FFNP 33.09%; FAP 21.21% and FANP 8.64% (Figure 4). The great amount of closed flowers robbed in the period of the collections must have significantly reduced the reproductive success, reducing the rates of produced seeds.
[FIGURE 3 OMITTED]
Verifying the relationship between the closed flowers and the closed flowers robbed found on the ground, a strong correlation was found (r = 0.68948, n = 100, p = 0.001) (Figure 5A). The correlation between the open flowers and open flowers robbed found on the ground was also highly significant (r = 0.98472, n = 100, p = 0.001) (Figure 5B). The relationship between open flowers and open flowers robbed in the plant showed high significance in the correlation test (r = 0.92404, n = 100, p = 0,001) (Figure 5C). The relationship between the closed flowers and closed flowers robbed found in the plant was also significant (r = 0.64755, n = 100, p = 0.001) (Figure 5D).
The relationship between the closed flowers and the closed flowers robbed found on the ground indicates that the premature floral abscission happens before flower opening, due to the nectar pillaging by native bees. The correlation between the open flowers and open flowers robbed found on the ground shows that the flowers that have not been robbed as floral buttons (FF), suffered pillaging after anthesis, resulting either way in floral abscission.
The relationship between open flowers and open flowers robbed in the plant indicates that the flowers that were not robbed before the anthesis suffer pillaging from native bees afterwards, corroborating the correlation between open flowers and open flowers found robbed on the ground. The relationship between the closed flowers and closed flowers robbed corroborates the correlation between the closed flowers and closed flowers robbed found on the ground, showing that the closed flowers in the plant that suffer pillaging fall to the ground. Trigona spinipes and Tetragonisca angustula present a wide spectrum of floral resources that can be used (SIMIONI et al., 2007), often not being direct pollinators as the case of B. spectabilis. The frequency of visitation by Apis mellifera was not recorded in the day of study, but the presence of this exotic species was observed during the collections of the flowers, being recorded that this species exercises great influence on the use of resources for native species (NOGUEIRA-FERREIRA; AUGUSTO, 2007), thus being able to come to induce pillaging in native bees as a compensatory alternative for obtaining nectar.
Other orders of insects were observed benefiting from the openings made by the bees, such as as thumb-tacks (Hemipterida), ants of several species (Formicidae) and beetles of several species (Coleoptera).
The recorded floral damages should influence the reproductive success of B. spectabilis as well as recorded in other species (CARVALHO et al., 2007a e b). Irwin (2003) suggests that nectar pillaging results in a significant reduction in the visitation in Ipomopsis aggregate (Pursh) V.E. Grant, as well as the pollen dispersion in the population, not generating variability and reducing the production of seeds. Such reductions in the dispersion and pollination should affect in the same way the reproductive biology of B. spectabilis.
[FIGURE 5 OMITTED]
Evolutionary answers to minimize the action of pillaging in flowers are adopted, such as the concentration of alkaloid substances in the nectar that significantly reduce the number of pillaged flowers and the time of visitation (ADLER; IRWIN, 2005). In B. spectabilis, there are no records on the composition of the nectar and possible adaptations to minimize the effects of pillaging.
In Mirabilis jalapa L. (Nyctaginaceae), the pollen diversity, size and accomplish act directly in the induced abortion and performance of the seeds, and the decrease in the attraction of the pollinator caused by robbing influences negatively in the reproductive process (NIESENBAUM, 1999). Still in M. jalapa, the reproductive biology is associated to the self-fecundation of the flowers, in spite of being recorded for the south area of Brazil the presence of a type of moth that accomplishes pollination, besides floral robbers like Xylocopa, Halictidae (Hymenoptera), thrips (Thysanoptera) and Diabrotica (Coleoptera), which obtain nectar through the perforation in the base of the chalice (LEAL et al., 2001).
In B. spectabilis, the reproductive alternatives are still uncertain; however, the great part of the reproduction is accomplished through stakes, as B. spectabilis is considered a commercial species with great ornamental interest, studies related to the floral biology, reproduction and their floral visitors should be motivated to maintain genetic variability, once the reproduction for stakes does allow such characteristics.
The pillaging activity by native bees positively influences the precocious floral abscission in Bougainvillea spectabilis, presenting significant differences among the floral variables measured on the ground and in the plant, indicating a strong correlation between the pillaging activity for native stingless bees and the premature fall of the inflorescences as floral buttons, developed buttons and flowers already open. Evaluations of the capacity and reproductive success in B. specitabilis are necessary for a better understanding of the pollination interactions and resources offered by the plant in its relationships with the floral insect visitors, pollinators, eventual visitors and robbing opportunists.
To Dr. Gustavo Graciolli (UFMS) for the comments and suggestions, Dr. Maria Rosangela Sigrist (UFMS) for authorizing the identification of bee species in her private collection, and Camila Freitas and Carina Jank for English suggestions.
ADLER, L. S.; IRWIN, R. E. Ecological costs and benefits of defense in nectar. Ecology, v. 86, n. 11, p. 2968-2978, 2005.
AGUIAR, C. M. L.; SANTOS, G. M. M. Compartilhamento de Recursos Florais por Vespas Sociais (Hymenoptera: Vespidae) e Abelhas (Hymenoptera: Apoidea) em uma Area de Caatinga. Neotropical Entomology, v. 36, n. 6, p. 836-842, 2007.
BARRETO, A. A.; FREITAS, L. Atributos florais em um sistema de polinizacao especializado: Calathea cylindrica (Roscoe) K. Schum. (Marantaceae) e abelhas Euglossini. Revista Brasileira de Botanica, v. 30, n. 3, p. 421-431, 2007.
BOICA JUNIOR, A. L.; SANTOS, T. M.; PASSILONGO, J. Trigona spinipes (Fabr.) (Hymenoptera: Apidae) em especies de maracujazeiro: flutuacao populacional, horario de visitacao e danos as flores. Neotropical Entomology, v. 33, n. 2, p. 135-139, 2004.
CARVALHO, F. A.; COSTA, N. A.; GARCIA, P. T.; REIS, R. A.; MATA, T. F. F.; RAMO, S. V. N.; MORAIS, H. C. Frequencia de danos por pilhadores de nectar em Tocoyena formosa (Cham. & Schitdl.) K. Schum. Revista Brasileira de Biociencias, v. 5, n. 1, p. 519-521, 2007a.
CARVALHO, A. T.; SANTOS-ANDRADE, F. G.; SCHLINDWEIN, C. Baixo sucesso reprodutivo em Anemopaegma laeve (Bignoniaceae) no parque Nacional do Catimbua, Pernambuco. Revista Brasileira de Biociencias, v. 5, n. 1, p. 102-104, 2007b.
EITEN, G. Formas fisionomicas de cerrado. Revista Brasileira de Botanica, v. 2, n. 2, p. 139-148, 1979. EMBRAPA-Empresa Brasileira de Pesquisa Agropecuaria. Boletim agrometeorologico. Campo Grande: Embrapa CNPGC, 1985.
GONZALEZ, C.; LOPEZ, E. Caryophyllidae Nyctaginaceae. In: FERBER, E. L. C. (Ed.) Guia de Consultas Botanica II. Argentina: Facultad de Ciencias Exactas y Naturales y Agrimensura (UNNE), 2004. p. 135-139.
IRWIN, R. E. Impact of nectar robbing on estimates of pollen flow: Conceptual predictions and empirical outcomes. Ecology, v. 84, n. 2, p. 485-495, 2003.
KOBAYASHI, K. D.; McCONNELL, J.; GRIFFIS, J. Boungainvillea. Ornamentals and Flowers, v. 38, n. 1, p. 1-12, 2007.
KOPPEN, W. Climatologia com un estudio de los climas de la tierra. Ciudad de Mexico: Fondo Cultura Economica, 1948.
LEAL, A. A.; TERADA, Y.; MACHADO, M. F. P. S. Floral biology of a population of Mirabilis jalapa L. (Nyctaginaceae) from Southern Brasil. Acta Scientiarum. Biological Sciences, v. 23, n. 2, p. 587-591, 2001.
LORENZI, H.; SOUZA, H. M. Plantas Ornamentais no Brasil. Arbustivas, herbaceas e trepadeiras. 3. ed. Nova Odessa: Instituto Plantarum, 2001.
MALOOF, J. E. The effects of a bumble bee nectar robbing on plant reproductive sucess and pollinator behavior. American Journal of Botany, v. 88, n. 11, p. 1960-1965, 2001.
MALOOF, J. E.; INOUYE, D. W. Are nectar robber caracters or mutualistis? Ecology, v. 81, n. 10, p. 2651-2661, 2000.
MOTHERSHEAD, K.; MARQUIS, R. Fitness impacts of herbivory through indirect effects on plant-pollinator interactions in Oenothera macrocarpa. Ecology, v. 81, n. 1, p. 30-40, 2000.
NIESENBAUM, R. A. The effects of pollen load size and donor diversity on pollen performance, selective abortion and progeny vigor in Mirabilis jalapa. American Journal of Botany, v. 86, n. 2, p. 261-268, 1999.
NOGUEIRA-FERREIRA, F. H.; AUGUSTO, S. C. Amplitude do nicho e similaridade no uso de recursos florais por abelhas eussociais em uma mata de Cerrado. Bioscience Journal, v. 23, p. 45-51, 2007.
PETRERE-JUNIOR, M. Statistica for Windows. Release 4.3 D. Tulsa, 1993.
ROUBIK, D. W. The ecological impact of nectar-robbing bess and pollinating hummingbirds on a tropical shrub. Ecology, v. 63, n. 2, p. 354-360, 1982.
SILVEIRA, F. A.; MELO, G. A. R.; ALMEIDA, E. A. B. Abelhas brasileiras: sistematica e identificacao. Belo Horizonte: (impresso pelos autores), 2002. v. 1.
SIMIONI, L. C.; D'APOLITO-JUNIOR, C.; RECH, A. R.; BALESTIERI, J. B. P.; MANENTEBALESTIERI, F. C. L. Especies de plantas visitadas por Tetragonisca angustula (Hymenoptera: Meliponinae), em Corumba, Mato Grosso do Sul. In: CONGRESSO DE
ECOLOGIA DO BRASIL, 8., 2007, Caxambu. Anais... Caxambu. Available on: <http://www.seb ecologia.org.br/viiiceb/pdf/1321.pdf>. Access on: 15 Jun. 2007.
ZAR, J. H. Biostatistical analysis. Englewood Cliffs: Prentice-Hall, 1996.
Received on September 9, 2009.
Accepted on March 17, 2010.
License information: This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Rodrigo Aranda *, Gisele Catian, Paulo Alexandre Bogiani and Igor Inforzato
Programa de Pos-graduacao em Ecologia e Conservacao, Universidade Federal de Mato Grosso do Sul, Cidade Universitaria, s/n, Cx. Postal 549, 79070-900, Campo Grande, Mato Grosso do Sul, Brazil. * Author for correspondence. E-mail: firstname.lastname@example.org
Table 1. Frequency of the predominant species of bees in Bougainvillea spectabilis in relation to the pillaging activity and pollination collected in the UFMS campus. Observed activity and frequencies Bee species Robber % Pollination % Trigona spinipes 106 56.10 0 0 Plebeia sp. 78 41.27 1 0.52 Auglochlora sp. 3 1.59 1 0.52 Figure 4. Characteristic of the Flowers (FF--Closed Flowers, FA --Open Flowers, FFP--Closed Flowers Robbed, FFNP--Closed Flowers Not Robbed, FAP--Open Flowers Robbed, FANP--Flowers Opened Not Robbed) sampled in soil and plant in the period between May 4th and June 1st, 2009, with their respective records and percentages. Appraied characteristics Soil plant FF 97.9% 70.16% FA 2.71% 29.89% FAN 0.24% 8.64% FAP 2.47% 21.21% FFN 17.75% 33.09% FFP 79.58% 37.06% Note: Table made from bar graph.