Open access journal: http://periodicos.uefs.br/ojs/index.php/sociobiology ISSN: 0361-6525 DOI: 10.13102/sociobiology.v60i3.323-328Sociobiology 60(3): 323-328 (2013) research article - Bees implications of the Floral herbivory on Malpighiacea Plant Fitness: Visual aspect of the Flower affects the attractiveness to Pollinators CA Ferreira, hM torezan-silingardi Introduction Interactions among organisms have occurred since the origin of life on Earth, and all species are involved in this process (Thompson, 2010). Among the relationships that occur between plants and animals, pollination is essential be- cause it increases the adaptive value of plants in relation to the environment, provides resources to animals, and occurs in 87.5% of angiosperms (Ollerton et al., 2011). Insects, es- pecially winged social Hymenoptera (bees and wasps), are among the main pollinators in nature (Gullan & Cranston, 2006). Conversely, herbivory is one of the most important antagonistic relationships because it harms angiosperms and reduces their fitness (Strauss et al., 1996, 1997; Price et al., 2011). Herbivores can consume photosynthetic tissues, storage organs, and reproductive structures (Romero & Vas- concellos-Neto, 2007; Del-Claro et al., 2013), thereby com- Abstract the Malpighiaceae family is species-rich and is abundant in Brazil. Malpighiaceae flowers provide oil and pollen to pollinating bees and serve as food for herbivorous insects, which damage the floral structures. although common in the cerrado, florivory is still poorly studied. in the present study, we evaluated the effect of florivory in one of the most common genera of Malpighiaceae in the cerrado (Banisteriopsis) and the impact of florivory on fruiting. the florivory rate was quantified in flowers of B. ma- lifolia belonging to two morphotypes and in flowers of B. variabilis. additionally, a petal-removal experiment was performed, which simulated the presence of damage in the flowers. the manipulation involved a control group with intact flowers, a group without the standard petal and a group of flowers without common petals. the flori- vory in the petals (floral area lost) differed between the species, and B. malifolia was the most damaged. the experimental manipulation revealed that intact flowers had a higher fruiting rate compared with the remaining flowers. these results reinforce the concept that florivory renders flowers less attractive to pollinating bees, which negatively affects the fruiting rate and the reproductive success of plants. We suggest that basic studies (such as the present investigation) be extended to further elucidate the effect of interactions between pollinators, plants, and herbivores on the general structure of communities. Sociobiology An international journal on social insects Universidade Federal de Uberlândia, Minas Gerais, Brazil. Article History edited by Gilberto M M santos, UeFs, Brazil received 15 July 2013 initial acceptance 05 aug 2013 Final acceptance 22 aug 2013 Keywords experimental manipulation, Florivory, Fruiting Corresponding author helena Maura torezan-silingardi lab. de ecologia comportamental e de interações Universidade Federal de Uberlândia Uberlândia, Minas Gerais, Brazil 38400-902 e-Mail: torezan@inbio.ufu.br promising the growth, reproduction, and survival of plants (Torezan-Silingardi, 2011). Florivory, or the consumption of floral structures, negatively affects pollination (Krupnick et al., 1999; McCall & Irwin, 2006), either by destroying the internal structures of the buds and/or by reducing the attrac- tiveness of flowers to the pollinators. For example, flower damage recorded for Sanicula arctopoides causes a reduction in the number and weight of the seeds produced (Lowenberg, 1994), illustrating reproductive losses for the species. There- fore, florivory may affect the maintenance of plants in natural environments (Torezan-Silingardi, 2007, 2011). The signaling of the production and provision of re- sources, by coloration, size, scent, shape, and texture (Gum- bert, 2000; Miller et al., 2011), serves as a cue for floral visi- tors. Several studies have demonstrated that bees respond to these signals by visiting mainly the flowers that offer cues of more profitable rewards such as a greater amount of pollen, ca Ferreira, hM torezan-silingardi - effects of florivory in Banisteriopsis fitness324 at the moment of anthesis and in the intensity of the scent: one type has flowers with an intense pink coloration and a strong scent (B. mali IP - morphotype 1), and the other type produces light pink flowers and a mild scent (B. mali LP - morphotype 2). B. variabilis produces only shrubs with white flowers and no scent. nectar, scent, or inflorescences that are larger, more conspi- cuous, or more numerous (Waser & Price, 1981; Krupnick et al., 1999; Irwin & Strauss, 2005). Similarly, these signals are attractive for other types of visitors and directly affect the reproduction of plants (Torezan-Silingardi, 2007; Alves- Silva et al., 2012). The family Malpighiaceae is very common in the Brazilian Cerrado (Ratter et al., 1997; Giulietti et al., 2005), and the genus Banisteriopsis, one of the largest genera of the family, comprises approximately one half of the species that occur in this biome (Mabberley, 1997). Banisteriopsis flowers supply oil and pollen for pollinating bees (Anderson, 1979, 1990; Vogel, 1974, 1990; Sigrist & Sazima, 2004; Ga- glianone, 2005). However, its flowers, fruit, and leaves are also sought by animals from various taxa (Torezan-Silingar- di, 2011; Oki, 2005; Diniz et al., 2000; Alves-Silva et al., 2013). Therefore, the main objective of the present study was to investigate the hypothesis that florivory in Banisteriopsis significantly reduces the reproductive potential (i.e., fruit for- mation). The specific objectives were (1) to determine the proportion of damage by floral herbivory in the species B. malifolia and B. variabilis and (2) to describe the effect of florivory on the attractiveness of flowers and on the repro- ductive success of B. malifolia. Materials and Methods Study site The study was performed in an area of cerrado strictu senso (a savanna-like vegetation that is the main physiog- nomy of the Cerrado biome) at the legal reserve area of the Clube de Caça e Pesca Itororó de Uberlândia (CCPIU), loca- ted at 18° 59’ S and 48° 18’ W, with an altitude of 863 m and a total area of 640 ha. According to the Koppen classification, the climate of the region is type AW, with two well-defined seasons: the dry season from May to September and the rainy season from October to April, with a mean annual temperatu- re of 22°C and a mean rainfall of 1,500 mm (Bachtold et al., 2012). The vegetation in the reserve includes the follo- wing physiognomies: grassland (campo limpo), shrubland (campo sujo), cerrado strictu senso, woodland (cerradão), palm swamps (veredas), and small patches of mesophyllous forest (see Réu & Del-Claro, 2005, for details on the area). Species B. malifolia variety malifolia (Ness & Martius) G. Gates and B. variabilis B. Gates are extremely similar spe- cies (Fig. 1). Their flowers are complete and conspicuous, with a slightly zygomorphic symmetry and five fimbriated petals, the standard petal has a size and insertion angle that is different from the remaining four petals. B. malifolia exhibits two types of shrubs that vary in the coloration of the flowers Fig. 1. Flowers of Banisteriopsis malifolia morphotypes intense pink (A) and light pink (B) and of Banisteriopsis variabilis (C) present at the reserve area of the Clube de Caça e Pesca Itororó de Uberlândia, state of Minas Gerais, Brazil. Data collection B. malifolia and B. variabilis shrubs of similar size and phenological state were monitored throughout the flowering period, particularly between April and July. For quantifying the floral herbivory, the flowers of 15 individuals of B. ma- lifolia morphotype 1 (B. mali IP), B. malifolia morphotype 2 (B. mali LP), and B. variabilis were collected, photogra- phed, and measured. In total, 45 flowers were analyzed with the Photoshop CS6 software, by measuring the total floral area and the floral area lost. For the experiment of simula- ted florivory, the inflorescences of 30 shrubs of B. malifolia (morphotype 2: light pink), divided in three groups of 10 in- dividuals, were isolated at the pre-anthesis period and were subsequently manipulated using gloves and fine-tip tweezers. The manipulations were always performed before 07:00 to avoid the timepoint at the beginning of pollinator activity. In total, 150 flowers were manipulated (five per shrub), and each individual was submitted to one of the following treat- ments: (1) control, the flowers remained intact without the simulation of herbivory; (2) herbivory was simulated by re- moving only the standard petal, with the common petals in- tact; (3) herbivory was simulated by removing four common petals, with the standard petal intact. Statistical analysis After determining the normality of the data, a one-way analysis of variance (ANOVA) was used to evaluate whether Sociobiology 60(3): 323-328 (2013) 325 the florivory differed between the B. malifolia and B. variabi- lis species and morphotypes. We also assessed whether there was variation in the fruiting rate among the different treat- ments of the damage-simulation experiment. The analyses were performed with the Systat 10.1 software. Results The flowering period occurred between the months of April and July 2012 (Fig. 2). Banisteriopsis malifolia was the first species to exhibit flowers (at the beginning of April), followed by B. variabilis, which had its first flowers recorded at the end of the same month and in the beginning of May. The flowering overlap encompassed approximately 20 days and resulted in a high number of available flowers, which attracted different groups of floral visitors. light-pink morphotype (df 2,42=14.307; P<0.001) (Fig. 3). Of the B. malifolia morphotypes, the light-pink one exhibited a higher number of damaged flowers; however, intensely pink flowers suffered a higher loss of floral tissue area. The flo- rivory percentage differed significantly between the species (0.03±0.13; N= 45). B. malifolia suffered a greater loss of floral area than did B. variabilis (F2,42=4.86; P=0.013) (Fig. 4), and there was no significant difference in these values between the B. malifolia morphotypes. Regarding the site of the damage, both species had damaged flowers with signs of herbivory in the central area of the petals, the margin of the petals and in both parts. The results of the simulated florivory revealed that the visual aspect of the flower affects the attractiveness to polli- nators because there was a significant difference among the fruiting rates in the different treatments (2,93±1,38; N= 30). The group of intact flowers (control) had the highest fruiting rate of all three treatments (F 2,27= 35,413; P<0.001) (Fig. 5). The flowers began the process of anthesis between 06:15 and 10:00 in the morning, with a small delay on clou- dy and rainy days. During this period, pollinators and other groups such as herbivores, predators, and parasitoids were recorded exploring the resources offered by both species. The signs of floral herbivory were easily recognized, and herbivores were found mainly at the onset of flowering, be- tween the months of April and May. Herbivores belonging to the orders Coleoptera, Lepidoptera, Hemiptera, Orthoptera, Hymenoptera, and Thysanoptera were recorded in flowers of both species studied - the most common ones being the Curculionidae beetles (present mainly at the bud stage) and Lycaenidae butterflies and small Thysanoptera (in the buds and flowers). The results demonstrated variations in the floral area, number of damaged flowers, and florivory rate. The num- ber of damaged flowers was higher in B. malifolia than in B. variabilis (Table 1 and 2). The floral area was higher in B. malifolia (2,95±0.76; N= 45), mainly in flowers of the Fig. 2. Flowering phenology of the species of Banisteriopsis va- riabilis and Banisteriopsis malifolia from the reserve area of the Clube de Caça e Pesca Itororó de Uberlândia, state of Minas Gerais, Brazil. Species /Damage Number of flowers Total area of the petal (cm²) Number of damaged flowers (%) Damaged area in cm² (%) B. malifolia IP 80 42.75 32 (40%) 2.80 (6.5%) B. malifolia LP 70 53.90 36 (51%) 2.08 (3.7%) B. variabilis 15 36.42 03 (20%) 0.24 (0.6%) Discussion The results of the floral-manipulation experiments and quantification of the natural florivory and the fruiting rate revealed that florivory negatively affects the reproduc- tion of plants by the genus Banisteriopsis in the Cerrado. Table 1. Estimate of the floral damage and its percentage recorded for the study species and the different morphotypes of Banisteriopsis malifolia. Fig. 3. Floral area of Banisteriopsis variabilis and of Banisteriopsis malifolia and its morphotypes from the reserve area of the Clube de Caça e Pesca Itororó de Uberlândia, state of Minas Gerais, Brazil. IP and LP refer to the two morphotypes of Banisteriopsis malifolia: intense pink and light pink, respectively. ca Ferreira, hM torezan-silingardi - effects of florivory in Banisteriopsis fitness326 These findings thus corroborate the central hypothesis of the present study. Plants have developed several strategies to at- tract the service of pollinators, and this investment favors the maintenance of the species in the ecosystem (Stout, 2000). However, florivory may directly reduce the quality and quan- tity of the available resources (Karban & Strauss, 1993) by alterations in the floral traits, such as shape and size of the petals (Fraze & Marquis, 1994), and by reductions of the vi- sual and olfactory cues (Krupnick & Weis, 1999; Krupnick et al., 1999). The damage recorded in the flowers of both spe- cies of Banisteriopsis is directly related to the use of buds for egg-laying by endophytic florivores (Torezan-Silingardi, 2011) or to the use of petals for feeding by exophytic her- bivores (Alves-Silva et al., 2013), both in the juvenile and adult stages (Torezan-Silingardi, 2007). The use of petals and of the remaining internal bud structures has been associated with a reduction in the overall attractiveness of flowers to their effective pollinators and to the consequent decrease in fruiting success (Leavitt & Robertson, 2006). The flowers of B. malifolia displayed a more exten- sive floral area, a higher number of flowers attacked, and a greater proportion of damage. The herbivores responsible for the reported damage might be attracted by the same cues identified by pollinators to find the flowers, such as the co- loration, intensity of scent, size of the flowers and density of flowers per area, which are more evident in B. malifolia than in B. variabilis. Therefore, the coloration and size of the flower appear to be important characteristics for explaining the preference of floral herbivores. Irwin and Strauss (2005), Frey (2004), and Salomão et al., (2006) discuss similar re- sults in which the attraction cues (such as size, color, shape, and scent) that are recognized by the pollinators and can also be identified by floral herbivores. Table 2. Contingency data based on the presence and absence of floral damage recorded for both morphotypes of Banisteriopsis malifolia and for both species of Banisteriopsis. Fig. 4. Percentage of damage by florivory in the flowers of Banis- teriopsis variabilis and of Banisteriopsis malifolia and its morpho- types from the reserve area of the Clube de Caça e Pesca Itororó de Uberlândia, state of Minas Gerais, Brazil. IP and LP refer to the two morphotypes of B. malifolia: intense pink and light pink, res- pectively. Fig. 5. Mean number of fruits produced during the simulated-flo- rivory experiment in Banisteriopsis malifolia (morphotype 2: light pink) at the reserve area of the Clube de Caça e Pesca Itororó de Uberlândia, state of Minas Gerais, Brazil. The results obtained in the experiment of simulated florivory reinforce the importance of floral attractiveness in the selection of resources by the pollinator because a reduced number of petals negatively affected the production of fruits. Flowers with damaged structures or eaten petals may indicate a deficit or absence of resources, causing the bees to refuse them. Furthermore, the olfactory cues that act together on the floral attraction of pollinators may influence the process of selection. Salomão et al. (2006) studied the effects of the floral-area size and of the floral herbivory in Trichogoniopsis adenantha (Asteraceae) and reported that large bushes and flowers undamaged by herbivory produce a higher propor- tion of seeds. The authors also noted that, just as pollinators prefer to visit larger bushes, herbivores may also respond in the same manner and cause greater damage in larger flowers, thereby reducing their attractiveness to pollinators. Sober et al. (2010) studied the floral herbivory in Verbascum nigrum and reported results similar to the present findings, in which the visitation rate was negatively correlated to florivory. The presence of herbivorous insects on the Baniste- riopsis flowers studied was responsible for the change in the floral traits, and such damage negatively affected the flower- pollinator interactions, causing a significant reduction in the fruit and seed production. This lower fruiting rate may be Presence absence total B. malifolia IP 36 44 80 B. malifolia LP 32 38 70 total 68 82 150 Presence absence total B. malifolia 68 82 150 B. variabilis 3 12 15 total 71 9 165 Sociobiology 60(3): 323-328 (2013) 327 the first sign of a reduction in fitness, considering that the majority of interactions with herbivores cause similar losses. Hendrix and Trapp (1989) tested the hypothesis that the res- ponses to floral herbivory in Pastinaca sativa L. (Apiaceae) may compensate or even increase the fitness of the indivi- duals attacked. The authors found that, despite the compen- satory responses generated, the reduced fitness is a result of damage by herbivory, and the species cannot compensate for this reduction. Therefore, future investigations (e.g., Byk & Del-Claro, 2010) will be important for describing the respon- ses of plant species to floral damage over time. Studies of the interactions that occur between orga- nisms are critical for the survival and reproduction of the species involved. Furthermore, the different forms of inte- ractions generate ecological diversity and enable the forma- tion of rich ecosystems, which are composed of species that have coevolved (Thompson, 2012). We can conclude that the species B. malifolia and B. variabilis are damaged by floral herbivores in different proportions and that this phenomenon directly affects the floral attractiveness, resulting in lower fruit production; furthermore, the higher the damage that is caused, the lower are the visitation rate and the reproduc- tive success. The study of these effects in the interactions between floral herbivores, flowers, and pollinators elucidates how these species relate to the environment and may support future investigations regarding the selection of areas for con- servation of the local biodiversity. Acknowledgments We thank Dr. Maria Cândida Mamede for identifying the species of Banisteriopsis and Dr. Kleber Del-Claro for the suggestions regarding the manuscript. CAF was granted a scholarship by the Brazilian Office for the Advancement of Higher Education (Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES). References Alves-Silva, E. Barônio, G.J., Torezan-Silingardi H.M. & Del-Claro, K. (2012) Foraging behavior of Brachygastra lecheguana (Hymenoptera: Vespidae) on Banisteriopsis mal- ifolia (Malpighiaceae): Extrafloral nectar consumption and herbivore predation in a tending ant system. Entomol. Sci. 16(2):162-169. doi:10.1111/ens.12004. Alves-silva, E., Bächtold, A., Barônio, G.J. & Del- Claro, K. (2013). Influence of Camponotus blan- dus (Formicinae) and flower buds on the occurrence of Parrhasius polibetes (Lepidoptera: Lycaenidae) in Ban- isteriopsis malifolia (Malpighiaceae). Sociobiology, 60(1): 30-34. doi: 10.13102/sociobiology.v60i1.30-34 Anderson, W.R. (1979). Floral conservatism in neotropical Malpighiaceae. Biotropica, 11: 219-223. Anderson, W.R. (1990). The origin of the Malpighiaceae - the evidence from morphology. Mem. N. Y. Bot. Gard., 64: 210-224. Bachtold A., Del Claro, K., Kaminski, L.A., Freitas, A.V.L. & Oliveira, P.S. (2012). Natural history of an ant-plant but-Natural history of an ant-plant but- terfly interaction in a neotropical savanna. J. Nat. Hist., 46: 15-16. Byk, J. & Del-Claro, K. (2010). Nectar- and pollen-gather- ing Cephalotes ants provide no protection against herbivory: a new manipulative experiment to test ant protective capa- bilities. Acta Ethol., 13: 33-38. Del-Claro, K., Guillermo-Ferreira, R., Almeida, E.M., Zardi- ni, H. & Torezan-Silingardi, H.M. (2013). Ants Visiting the Post-Floral Secretions of Pericarpial Nectaries in Palicou- rea rigida (Rubiaceae) Provide Protection against Leaf Her- bivores but not against Seed Parasites. Sociobiology, 60(3): 219-223. doi:10.13102/sociobiology.v60i3.217-221. Diniz, I.R., Morais, H.C. & Hay, J.D. (2000). Natural history of herbivores feeding on Byrsonima species. Braz. J. Ecol., 2: 49-54. Fraze J.E & Marquis R.J. (1994) Environmental contribution to floral trait variation in Chamaecrista fasciculata (Fabace- ae: Caesalpinoideae). Am. J. Bot. 81: 206-215. Frey, F.M. (2004). Opposing natural selection from herbivores and pathogens may maintain floral-color variation in Clayto- nia virginica (Portulacaceae). Evolution, 58: 2426-2437. Gaglianone, M.C. (2005). Abelhas coletoras de óleos e flores de Malpighiaceae. In V.R. Gates, B. (1983). Banisteriopsis, Diplopterys (Malpighiaceae). Flora Neotrop., 30: 1-236. Giulietti, A.M., Harley, R.M., Queiroz, L.P., Wanderley, M.G. & Van Den Berg, C. (2005). Biodiversidade e conservação das plantas no Brasil. Megadiversidade, 1: 52-61. Gullan, P.J & Cranston, P.S. (2006). The insects: an outline of entolomogy. 4th edition. Wiley-Blackwell Science, 565 p. Gumbert, A. (2000). Color choices by bumble bees (Bombus terrestris): Innate preferences and generalization after lear- ning. Behav. Ecol. Socio., 48: 36-43. Hendrix, S.D. & Trapp, E.J. (1989). Floral herbivory in Pas- tinaca sativa: Do compensatory responses offset reductions in fitness? Evolution. 43: 891-895. Irwin, R.E. & Strauss, S.Y. (2005). Flower color microevo- lution in wild radish: Evolutionary response to pollinator- mediated selection. Am. Nat., 165: 225-237. Karban R. & Strauss S.Y. (1993) Effects of herbivores on growth and production of their perennial host, Erigeron glau- cus. Ecology 74: 39-46. Krupnick, G.A. & Weis, A.E. (1999).The effect of floral her- bivory on male and female reproductive success in Isomeris ca Ferreira, hM torezan-silingardi - effects of florivory in Banisteriopsis fitness328 arborea. Ecology, 80:135-1149 Krupnick, G.A., Weis, A.E. & Campbell, D.R. (1999). The consequences of floral herbivory for pollinator service to Iso- meris arborea. Ecology, 80: 125-134. Leavitt, H. & Robertson, I.C. (2006). Petal herbivory by chrysomelid beetles (Phyllotretasp.) is detrimental to polli- nation and seed production in Lepidium papilliferum (Brassi- caceae). Ecol. Entomol., 31: 657-660. Lowenberg, G. J. (1994). "Effects of Floral Herbivory on Maternal Reproduction in Sanicula Arctopoides (Apiaceae)." Ecology 75: 359-369. Mabberley, D.J. (1997). The plant-book. Bath: Cambridge Univ. Press. Mccall, A.C. & Irwin, R.E. (2006). Florivory: The intersec- tion of pollination and herbivory. Ecol. Lett., 9: 1351-1365. Miller, R., Owens, S.J. & Rorslett, B. (2011). Plants and colours: Flowers and pollination. Opt. Laser Technol., 43: 282-294. Oki, Y. (2005). Interações entre larvas de Lepidoptera e as espécies de Malpighiaceae em dois fragmentos de Cerrado do Estado de São Paulo. Tese de doutorado, Universidade de São, Ribeirão Preto, Brasil. Ollerton, J.,Winfree, R. &Tarrant, S. (2011). How many flowering plants are pollinated by animals? Oikos, 120: 321-326. Price, P.W. Denno, R.F. Eubanks, M.D. Finke, D.L. & Ka- plan, I. (2011). Insect Ecology: Behavior, Populations and Communities. Cambridge University. Press, 801 p. Ratter, J.A., Ribeiro, J.F. & Bridgewater, S. (1997). The Bra- zilian Cerrado vegetation and threats to its biodiversity. Ann. Bot., 80: 223-230. Reu, W. F. & Del-Claro, K. (2005). Natural History and Biol- ogy of Chlamisus minax Lacordaire (Chrysomelidae: Chla- misinae). Neotrop. Entomol., 34: 357-362. Romero, G.Q. & Vasconcellos-Neto, J. (2007). Aranhas so- bre plantas: dos comportamentos de forrageamento às asso- ciações específicas. In M.O. Gonzaga, A.J. Santos & H.F. Japyassú (Eds.), Ecologia e comportamento de aranhas (pp. 68-87). Rio de Janeiro: Interciência. Salomão, A.T., Martins, L.F., Ribeiro, R.S. & Romero, G.Q. (2006). Effects of patch size and floral herbivory on seed in Trichogoniopsis adenantha (Asteraceae) in Southeastern Brazil. Biotropica, 38: 272-275. Sigrist, M.R. & Sazima, M. (2004). Pollination and reproduc- tive biology of twelve species of neotropical Malpighiaceae: Stigma morphology and its implications for the breeding sys- tem. Ann. Bot., 94: 33-41. Sober, V., Teder, T. & Moora, M. (2010). Contrasting effects of plant population size on florivory and pollination. Basic Appl. Ecol., 10: 737-744. Stout, J. C. (2000). Does size matter? Bumblebee behaviour and the pollination of Cytisuss coparius L. (Fabaceae).Api- dologie, 31:129-139. Strauss, S.Y. (1997). Floral characters link herbivores, polli- nators, and plant fitness. Ecology, 78: 1640-1645. Strauss, S.Y., Conner, J.K. & Rush, S.L. (1996). Foliar herbi- vory affects floral characters and plant attractiveness to polli- nators: Implications for male and female plant fit- ness. Am. Nat., 147: 1098-1107. Thompson, J.D. (2010). How do visitation patterns vary among pollinators in relation to floral display and floral de- sign in a generalist pollination system? Oecologia, 126: 386- 394. doi: 10.1007/s004420000531. Thompson, J.N (2012). O Futuro dos estudos em interações plantas-animais. In K. Del-Claro & H.M. Torezan-Silingardi (Eds.), Ecologia das interações Plantas-animais: uma abor- dagem ecológico-evolutiva (pp. 307-318). Rio de Janeiro: Technical books. Torezan-Silingardi H.M. (2011). Predatory behavior of Pa- chodynerus brevithorax (Hymenoptera: Vespidae, Eumeni- nae) on endophytic herbivore beetles in the Brazilian tropical savanna. Sociobiology, 57: 181-190. Torezan-Silingardi, H.M. (2007). A influência dos herbívo- ros florais, dos polinizadores e das características fenológicas sobre a frutificação de espécies da família Malpighiaceae em um Cerrado de Minas Gerais. Tese de doutorado, Universida- de de São Paulo, Ribeirão Preto, Brasil. Vogel, S. (1974).Ölblumen und ölsammelndeBienen. Akade- mie der Wissenchaften und der Literatur. Tropische und sub- tropischePflanzenwelt 7. Franz. Steiner Verlag. Wiesbaden. J. Bot. Tax. Geo., 88: 136-137. doi: 10.1002/fedr.19770880110 Vogel, S. (1990). History of the Malpighiaceae in the light of pollination ecology. Mem. N. Y. Bot. Gard., 55: 130-142. Waser, N.M. &. Price, M.V. (1981). Pollinator choice and stabilizing selection for flower color in Delphinium nelsonii. Evolution, 35:376–390.