1229 Botanical Origin of Protein Sources Used by Honeybees (Apis mellifera) in an Atlantic Forest by Talita Antonia da Silveira1*; Maria Emilene Correia-Oliveira1; Augusta C. C. C. Moreti2; Ivani Pozar Otsuk2; Luís Carlos Marchini1 ABSTRACT Productive and reproductive traits of beehives are influenced by climate and food availability in the region where the bees are reared or maintained, thus honey and pollen storage, egg-laying conditions of the queen as well as comb occupation are subject to seasonal variations. The present study was conducted in the apiary of the Department of Entomology and Acarology, College of Agriculture Luiz de Queiróz, ESALQ/USP, in the municipality of Piracicaba, in an area containing fruit trees, ornamental plants and a fragment of a native forest. The objective was to identify protein sources used by honeybees (Apis mellifera) over a whole year (2010-2011) in remnants of the Atlantic forest, information that can be used in the conservation and restoration of degraded areas. For sample preparation, the acetolysis method was adopted (Eredtman 1952) and the quantitative analysis was performed by counting successive samples of 900 grains per sample which were grouped by botanical species and/or pol- len types. The results show that the bees used various plant types in the area, including ruderal species, to maintain their colonies. Apis mellifera seeks food sources in all plants in the surroundings of the apiary, including herbaceous, shrubs, trees, native or introduced. Eucalyptus sp. played an important role as a food source in all seasons due to its wide availability around the apiary and its high flower production. The most frequent pollen types (greater than 10% of the sample) were Anadenanthera sp., Acacia sp, Miconia sp. and Eucalyptus sp. in winter; Philodendron sp., Mikania cordifolia, Parthenium and Eucalyptus sp. in spring ; Alternanthera ficoidea, Chamissoa altissima and Eucalyptus sp. in summer; Philodendron sp., Raphanus sp. and Eucalyptus sp. in autumn. Key words: pallinolog y, pollen, beekeeping. 1Programa de Pós-graduação em Entomologia e Acarologia da Escola Superior de Agricultura “Luiz de Queiroz”, Universidade de São Paulo, 13418-900, Piracicaba, São Paulo, Brasil. 2Instituto de Zootecnia, Secretaria de Agricultura e Abastecimento, 13460-000, Nova Odessa, São Paulo, Brasil *Corresponding author: tali_aband@hotmail.com 1230 Sociobiolog y Vol. 59, No. 4, 2012 INTRODUCTION Pollen is the male gamete of flowers of angiosperms, necessary for egg fertilization and, therefore, for seed formation (Raven 2001). It also serves as a reward for floral pollinators (Schlindwein et al. 2005), mainly bees, which use pollen as a protein source of their diets (Zerbo et al. 2001). Honeybees, Apis mellifera, use pollen as a food source in all developmental stages of the colony (Basim et al. 2006), therefore, protein content and availability the pollen are essential factors for the colony (Zerbo et al. 2001). Pollen is the sole protein source (Roubik 1989) necessary for the proper functioning of the hypopharyngeal gland in worker bees (Crailsheim 1990). Pollen grains vary in shape, size, color, appearance, morpholog y and can, thus, be used to identify plant gender and species (Schmidt & Buchmann 1993; Almeida-Muradian et al. 2005). Pollen grain identification is made through its outer wall, chemically stable and morphologically varied, which allows the identification of various pollen types through the pollen analysis (Salgado-Labouriau 1973, Barth 2004). This identification can be used as indicator of the geographical and botanical origin of honey (Barth 2004), since the pollen types found in these products are considered “fingerprints” of bee foraging habits (Wittmann & Schlindwein 1995). Knowledge of bee foraging habits can be an important tool for beekeep- ing development (Pearson & Braiden 1990), determining the site of apiary implantation and honey production (Ashman et al. 2004; Biesmeijer et al. 2006). This type of study can also be used for reforestation and conservation of areas and for the establishment of ecological corridors (Modro 2009). This work aimed to identify the plant species used as protein sources by honeybees (Apis mellifera) in remnants of the Atlantic forest, information that can be used in conservation and restoration programs of degraded areas. MATERIALS AND METHODS Experimental area Samples were collected at the beginning and end of each of the four sea- sons during a whole year between 2010 and 2011, at the apiary located in the Department of Entomolog y and Acarolog y the College of Agriculture “Luiz de Queiroz”, near a fragment of the Atlantic forest, in the municipal- 1231 Silveira, T.A. et al. — Protein Sources Used by Honeybees in Atlantic Forest ity of Piracicaba (22º52’33”S, 47º38’30”W, altitude 546m), São Paulo state, Brazil. Collection of samples Five hives were selected under similar population conditions and in the first and last weeks of each season (spring, summer, autumn and winter), we installed pollen collectors that allowed the entrance of the bees, but inter- cepted the pollen collected, which fell into a tray and remained until it was removed. Pollen Analysis We used the acetolysis method (Erdtman 1952) for identification of pollen types. The method forces the exit of the cellular content of the pollen grain, leaving only the cell wall. Afterwards, slides were mounted on glycerinated gelatin. The grain characterization was made by comparison with a reference slide collection of the Department of Entomolog y and Acarolog y of the College of Agriculture “Luiz de Queiroz” (ESALQ/USP) and the specialized literature (Barth 1970 a, b, c, Moreti et al. 2002, Roubik & Moreno 1991). After the grain identification, we carried out a quantitative analysis by counting 300 grains per slide, 900 per sample (analysis made in triplicate). The quantitative analysis aims to determine the proportion of contribution of each pollen type, identifying and grouping it according to its occurrence as dominant pollen (DP), when it accounts for more than 45% of total pollen grains counted; accessory pollen (AP, 16-45%) and isolated pollen (IP, up to 15%). The IP was subdivided into significant isolated pollen (SIP, 3-15%) and occasional isolated pollen (OIP, less than 3% of the grains observed) (Louveaux et al. 1978). The main pollen types were photomicrographed through a Zeiss micro- scope coupled to a digital camera. The experimental design was completely randomized with five replicates (represented by hives). Data were analyzed using the MIXED procedure, SAS (Statistical Analysis System), to determine the structure of variance and covariance matrix. The significance level for the variance analysis was 5%. The data were transformed to Log (X+1) to achieve homoscedasticity in the variance analysis. 1232 Sociobiolog y Vol. 59, No. 4, 2012 RESULTS AND DISCUSSION In the study period, 68 pollen types belonging to 19 botanical taxa were found (Table 1). Modro (2011) conducted a study in the same place and found similar results with regard to plant families, however, the author observed a total of 81 taxa. This difference can be explained due to the seasonality of some plant species and the presence of cropped areas in the region of collection. The main pollen types classified in terms of statistical significance were: in spring, Mikania cordifolia and Eucalyptus sp; in summer, Alternanthera ficoidea (dominant); in fall, Eucalyptus sp as accessory pollen, and in winter there was no statistical difference among the pollen types (Table 1). According to Renner (1968), most plant species have flowers that do not produce pollen or nectar throughout the day, but only at certain hours. Thus, bee activity, food type and increased collection time, according to Butler (1945); Moffett & Paker (1953), Bennett & Renner (1961), depend on the following characteristics: genetic trait of the hive, the amount of nectar avail- able, sugar content in flowers, time of day, environmental factors and plant species. Therefore, variations in collection sites greatly affect the collection of floral resources by bees. The families Fabaceae, Asteraceae and Arecaceae showed the greatest num- ber of species in winter, Arecaceae and Asteraceae in spring and Asteraceae in autumn. The main pollen species were: Acacia sp. and Anadenanthera sp. (Fabaceae), Alternanthera ficoidea and Chamissoa altissima (Amaranthaceae), Philodendron sp. (Araceae), Eucalyptus sp. (Myrtaceae), Mikania cordifolia and Parthenium sp. (Asteraceae) and Raphanus sp. (Brassicaceae). The difference in numbers of pollen types along the seasons shows the generalist habit of these bees, however, there seems to have a preference for some plant species that appear in all seasons such as Eucalyptus, Raphanus, Astrocaryum and Philodendron, although none of them are dominant. The preference of Apis mellifera for Eucalyptus pollen has been observed by other authors, whenever Eucalyptus was present in the vicinity of the apiary (Almeida-Anacleto 2007 and Carvalho et al. 1999, in Piracicaba, São Paulo State; Modro 2006 and Barreto 1999, in Viçosa, Minas Gerais State; Luz et al. 2007 in Engenheiro Paulo de Frontin, Rio de Janeiro State, and Moreti et al. 2011 in Pindamonhangaba, São Paulo State – all municipalities in Brazil ). 1233 Silveira, T.A. et al. — Protein Sources Used by Honeybees in Atlantic Forest Table 1 - Average percentage of pollen types collected by Apis mellifera in Piracicaba, São Paulo state, Brazil, in the four seasons in 2010-2011. Means followed by different capital letters in the columns differ by Tukey-Kramer (P <0.05). 1Means in parentheses are averages transformed to Log (X+1) # statistical data on produced pollen. DP -dominant pollen; AP – assistant pollen, SIP – significant isolated pollen, OIP - occasional isolated pollen (X -no collection). Family Species Season Spring Summer Autumn Winter Amaranthaceae Amaranthus sp. X X X 1.46±8.54 OIP (0.36±0.24)A Alternanthera ficoidea 3.70±14.76 SIP (0.27±0.42)AB 59.89±6.61 DP (1.63±0.18)A 4.39±6.08 SIP (0.69±0.17)B X Chamissoa altissima X 16.88±10.61 AP (1.24±0.19)AB X X Anacardiaceae Tapirira sp. X X 2.13±14.78 OIP (0.49±0.42)AB X Araceae type1 2.10±8.53 OIP (0.42±0.24)AB X 1.71±8.53 OIP (0.43±0.24)AB 1.77±8.67 OIP (0.41±0.24)A Philodendron sp. 20.04±6.61 AP (1.01±0.19)AB 5.55±6.69 SIP (0.68±0.19)AB 14.97±8.65 SIP (0.89±0.24)AB 9.85±6.72 SIP (0.72±0.19)A Arecaceae type 1 1.00±10.45 OIP (0.28±0.30)AB X 5.31±8.94 SIP (0.67±0.24)AB 2.34±8.66 OIP (0.47±0.24)A Astrocaryum sp. 4.46±8.53 SIP (0.59±0.24)AB 2.99±10.44 OIP (0.54±0.30)AB 6.78±6.12 SIP (0.66±0.17)AB 1.91±6.73 OIP (0.33±0.19)A Asteraceae Bidens pilosa X X 0.87±6.74 PIO (0.26±0.19)B 0.25±14.77 PIO (0.08±0.42)A Mikania cordifolia 34.10±7.56 AP (1.48±0.21)A X 5.22±8.53 SIP (0.59±0.24)AB 1.05±10.71 OIP (0.31±0.29)A Parthenium sp. 30.96±10.45 AP (1.05±0.30)AB X 4.70±6.13 SIP (0.64±0.17)AB 0.15±10.45 OIP (0.04±0.30)A Type 1 X X 3.04±10.71 SIP (0.60±0.29)AB X Sonchus oleraceus X X 0.55±8.67 OIP (0.18±0.24)B X Brassicaceae Raphanus sp. 1.62±10.44 OIP (0.36±0.30)AB 1.69±10.44 OIP (0.36±0.30)B 21.62±5.03 AP (1.07±0.14)AB 0.87±7.46 OIP (0.37±0.21)A Cucurbitaceae Momordica sp. X X 1.54±8.67 OIP (0.39±0.24)AB 0.42±10.71 OIP (0.14±0.29)A Fabaceae Acacia sp. X X 0.70±10.69 OIP (0.06±0.29)B 17.59±8.54 AP (1.11±0.24)A Anadenanthera sp. X X X 16.59±8.54 AP (0.12±0.24)A Caesalpinia peltophoroides x X X 1.22±8.54 OIP (0.34±0.24)A Mimosa caesalpinifolia 8.40±7.39 SIP (0.77±0.21)AB X X 0.04±10.45 OIP (0.04±0.30)A 1234 Sociobiolog y Vol. 59, No. 4, 2012 Quantitatively, the Eucalyptus sp. grains showed some frequency in almost all seasons. This may be associated with the availability of these resources in the collection site and the high pollen production of this plant species that makes this type of resource more easily collected by bees. This fact is consistent with Bawa (1983) and Castro (1994) who state that the attractiveness of flowering can be influenced by the amount of pollen produced, the concentration and abundance of flowers, number of competing insects, the distance between the flowering sites and nest, and innate preference for the species. However, Table 1 (continued). Family Species Season Spring Summer Autumn Winter Fabaceae type1 x X 1.43±10.45 OIP (0.38±0.30)AB X Lythraceae Lagerstroemia indica 2.32±8.54 OIP (0.51±0.24)AB X X X Malvaceae Paquira sp. X X X 0.15±10.45 OIP (0.06±0.30)A Dombeya sp. X 0.27±10.45 OIP (0.11±0.30)B 0.72±8.53 OIP (0.20±0.24)B X Melastomataceae Miconia sp. X X X 12.39±10.45 SIP (1.13±0.30)A Morus nigra 9.24±8.53 SIP (0.98±0.24)AB 6.08±7.39 SIP (0.63±0.21)AB 1.29±10.44 OIP (0.30±0.30)AB Myrtaceae Eucalyptus sp. 31.34±5.03 AP (1.12±0.14)A 14.83±4.79 PII (0.87±0.13)AB 37.52±4.79 AP (1.53±0.13)A 18.79±5.66 AP (0.99±0.16)A Piperaceae Piper sp. 2.32±7.47 OIP (0.42±0.21)AB 1.89±7.47 OIP (0.40±0.21)B X 6.02±8.65 SIP (0.75±0.24)A Poaceae Type 2 0.51±6.13 OIP (0.16±0.11)B X X X Oryza sp. X X 2.11±10.71 OIP (0.46±0.29)AB X Rutaceae Citrus sp. X X 1.03±10.45 OIP (0.28±0.30)AB X Prunus persica X 8.14±8.54 SIP (0.87±0.24)AB X X Sapindaceae x X X 0.29±10.71 OIP (0.11±0.29)B X Solanaceae Type Solanaceae 0.96±10.45 OIP (0.28±0.30)AB 4.54±10.45 SIP (0.74±0.30)AB X X Scrophulariaceae Scopania sp. X 5.48±10.45 SIP (0.57±0.30)AB X X 1235 Silveira, T.A. et al. — Protein Sources Used by Honeybees in Atlantic Forest these factors are still subject to variation such as flower size, relative humidity, soil moisture, temperature, altitude, time and daylight duration. This generalist habit can occur due to the protein need of the colony, since the protein content may vary with the botanical origin. According to Rouston et al. (2000), protein variation may not be directly related to the attractiveness of pollinators, since the pollen of zoophilic species is not richer in proteins than in anemophilous species. According to Schmidt & Buchmann (1993) bees collect pollen from many plant species and thus have good nutritional balance and high dilution of toxic potential of alkaloids and other toxins. The summer was the only season that showed dominance of a plant spe- cies, despite the 12 different types found. Bees sought preferably the species Alternanthera ficoidea, which is an herbaceous plant originating from the Americas (Mears 1977), showing the importance of plants often considered weeds for honey production (Modro 2011). It is interesting to observe that this same plant appears as SIP in spring and summer, which indicates that it could become an important bee forager in other seasons. The preference for this pollen type, which, at times, appears in small amounts changing to dominant in others, may be related to the energ y bees spend to seek other food sources, since this species occurs in abundance in the surroundings of the apiary (Modro, et al. 2007; Modro, et. al. 2011). Pollen types that had a frequency lower than 10% may be considered sec- ondary, because Ramalho et al., (1985) believe that sources between 1 and 10% of pollen have little attractiveness as food source. These pollen types, however, may be useful as a suplement to food needs of the colony and become important for maintaining balance at other times of year. CONCLUSION Apis mellifera bees seek food sources in all plants in the surroundings of the apiary, regardless of the plant type, whether herbaceous, a shrub, a tree, native or introduced. The Eucalyptus sp played an important role as bee forager in all seasons due to its wide availability in the vicinity of the apiary and its large produc- tion of flowers. The most frequent pollen types (greater than 10% of the sample) were: Anadenanthera sp, Acacia sp, Miconia sp and Eucalyptus sp during winter; 1236 Sociobiolog y Vol. 59, No. 4, 2012 Philodendron sp, Mikania cordifolia, Parthenium sp and Eucalyptus sp in spring ; Alternanthera ficoidea, Chamissoa altissima and Eucalyptus sp in summer; Philodendron sp, Raphanus sp and Eucalyptus sp in autumn. ACKNOWLEDGMENT We wish to thank FAPESP and CAPES for the financial support to the research project. REFERENCES Almeida-Anacleto, D. de. 2007. Recursos alimentares, desenvolvimento das colônias e características físico químicas, microbiológicas e polínicas de mel e cargas de pólen de meliponíneos, do município de Piracicaba, Estado de São Paulo. 133 p. Tese (Doutorado em Entomologia) - Escola Superior de Agricultura “Luiz de Queiroz”, Universidade de São Paulo, Piracicaba. Almeida-Muradian, L.B., L.C. Pamplona, S. Coimbra & O.M. Barth 2005. Chemical composition and botanical evaluation of dried bee pollen pellets. Journal of Food Composition and Analysis 18: 105-111. Ashman, T.L., T.M. Knight, J.A. Steets, P. Amarasekare, M. Burd, D.R. Campbell, M.R. Dudash, M.O. Johnston, S.J. Mazer, R.J. Mitchell, M.T. Morgan, W.G. Wilson 2004. Pollen limitation of plant reproduction: ecological and evolutionary causes and consequences. Ecolog y 85: 2408–2421. Barth, O.M. 2004. Melissopalynolog y in Brazil: a review of pollen analysis of honeys, propolis and pollen loads of bees. Scientia Agricola 61: 342-350. Barth, O.M. 1970a. Análise microscópica de algumas amostras de mel. 1 Pólen Dominante. Anais Academia Brasileira de Ciências 42, 2: 351-366. Barth, O.M. 1970b. Análise microscópica de algumas amostras de mel. 2 Pólen Acessório. Anais Academia Brasileira de Ciências 42, 3: 571-590. Barth, O.M. 1970c. Análise microscópica de algumas amostras de mel. 3 Pólen isolado. Anais Academia Brasileira de Ciências 42, 4: 748-772. Barreto, L.M.R.C. 1999. Levantamento florístico e polínico e estudo melissopalinológico durante a principal safra da microrregião homogênea da Zona da Mata de Viçosa, MG. 87p. Dissertação (Mestrado em Entomologia) – Universidade Federal de Viçosa, Viçosa. Basim, E., H. Basim, M. Özcan 2006. Antibacterial activities of Turkish pollen and propolis extracts against plant bacterial pathogens. Journal of Food Engineering, Oxon, 77, 4: 992-996. Bawa, K.S. 1983. Patterns of flowering in tropical plants. In.: JONES, G.E.;LITTLE,R.J. (Ed.). Handbook of experimental and pollination biolog y. New York: Van Nostrand Reinhold 394-410. 1237 Silveira, T.A. et al. — Protein Sources Used by Honeybees in Atlantic Forest Bennett, M.F. & M. Renner 1961. Variations in the colleting activity of honeybees under laboratory conditions. American Zoolog y Journal 1: 185. Biesmeijer, J.C, B. Van Marwijk, K. Van Oeursen, W. Punt & M.J. Sommeijer 1992. Pollen sources for Apis mellifera L. (Hymoptera, Apidae) in Surinam, based on pollen grain volume. Apidologie 23: 245-256. Butler, C.G. 1945. The influence of various physical and biological factors of the environment on honeybee activity. An examination of the relationship between activity and nectar concentration and abundance. Journal of Exprerimental Biolog y 21: 5-12. Carvalho, C.A.L., L.C.M. Marchini, & P.B. ROS 1999. Fontes de pólen utilizados por Apis mellifera L. e algumas espécies de Trigonini (Apidae) em Piracicaba (SP). Bragantia 58, 1: 49-56. Castro, M.S. 1994. Plantas apícolas – identificação e caracterização. In.: BRANDÃO, A.L.S.; M.A.C. Boaretto (Coord.). Apicultura atual: diversificação de produtos. Vitória da Conquista: UEFS, DFZ, 21-31. Crailsheim, K. 1990. The protein balance of the honey bee worker. Apidologie 21: 417- 429. Erdtman, G. 1952. Pollen morpholog y and plant taxonomy – Angiosperms. Stockholm: Almqvist & Wiksell, 539p. Louveaux, J., A. Maurizio, & G. Vorwohl 1978. Methods of melissopalynolog y. Bee World, Gerrards Cross 59: 4: 139-157. Luz, C.F.P., M.L. Thomé, & O.M. Barth 2007. Recursos tróficos de Apis mellifera L. (Hymenoptera: Apidae) na região de Morro Azul do Tinguá, Estado do Rio de Janeiro. Revista Brasileira de Botânica 30, 1: 29-36. Mears, J.A. 1977. The Nomenclature and Type Collections of the Widespread Taxa of Alternanthera (Amaranthaceae). Academy of Natural Sciences of Philadelphia 129: 1-21. Moffett, J.O. & E.R.L. Parker 1953. Relation of weather factores to nectarflow in honey production. Tech. Bull. Kans. Agric.Exp. Stn 74, 1: 1-27. Modro, A.F.H., L.C. Marchini & A.C.C.C. Moreti 2011. Origem botânica de cargas de pólen de colmeias de abelhas africanizadas em Piracicaba, SP. Ciência Rural, Santa Maria 41, 11: 1944-1951. Modro, A.F.H., D. Message, C.F.P. Luz, & J.A.A.M. Neto 2011. Flora de importância polinífera para Apis mellifera (L.) na região de Viçosa, MG.Revista Árvore, Viçosa-MG, 35, 5: 1145-1153. Modro, A.F.H., I.C. Silva, C.F.P. Luz, D. Message 2009. Analysis of pollen load based on color, physicochemical composition and botanical source. Anais da Academia Brasileira de Ciências 81, 2: 281-285. Modro, A.F.H., D. Message, C.F.P. Luz & J.A.A. Meira Neto 2007. Composição e qualidade do pólen apícola coletado em Minas Gerais, Brasil. Pesquisa agropecuária Brasileira 42, 8: 1057-1065. 1238 Sociobiolog y Vol. 59, No. 4, 2012 Modro, A.F.H. 2006. Flora e caracterização polinífera para abelhas Apis mellifera L. na região de Viçosa, MG. 2006. Dissertação (M. S. em Entomologia) – Universidade Federal de Viçosa, Viçosa. Moreti, A.C.C.C., L.C. Marchini, R.R. Rodrigues, & V.C. Souza 2002. Atlas do pólen de plantas apícolas. Rio de Janeiro: Papel virtual. Moreti, A.C.C.C, E.W. Teixeira, M.L.T.M.F. Alves, K.C.L. Oliveira, & L.B. Almeida-Muradian 2011. Pollen Sources for Apis mellifera (L.) in Pindamonhangaba County, State of São Paulo, Brazil. Sociobiolog y 58, 3:1-12. Pearson, W.D. & V. Braiden. 1990. Seasonal pollen collection honeybees from grass/ shrub highlands in Canterbury, New Zealand. Journal of Apicultural Research 29: 206–213. RAMALHO, M., V.L. Imperatriz-Fonseca, A. Kleinert-Giovannini, & M. Cortopassi- Laurino 1985. Exploitation of floral resources by Plebeia remota Holmberg (Apidae, Meliponinae). Apidologie 16, 3: 307-330. Raven, P.H. Biologia Vegetal. 2001. 2 ed. Rio de Janeiro: Guanabara Dois S.A. 724 pp. Renner, M. 1968. Le sens temps chez l’abeille. In: Chauvin, R.Traité de biologie de l’abeille. Paris: Masson 2: cap.4: 506-532. Roubik, D.W. 1989. Ecolog y and natural history of tropical bees, Cambridge, Cambridge University Press, 514 p. Roubik, D.W. & J.E.P. Moreno 1991. Pollen and spore of barro colorado island. , St. Louis: Missouri Botanical Garden, (Monographs in Systematics Botany, 36) 268 p. Roulston, T.H., J.H. Cane & S.L. Buchmann 2000.What governs protein content of pollen: pollinator preferences, pollen- pistil interactions, or phylogeny ? Ecological Monographs, Bucarest 70, 4: 617 – 627. Salgado-Labouriau, M.L. 1973. Contribuição à palinologia dos cerrados. Academia Brasileira de Ciências, Rio de Janeiro. Schmidt, J.O. & S.L. Buchmann 1993. Other products of the hive.In: Grahan, J.M. (Ed.). The hive and the honeybee. Hamilton: Dadant & Sons, 927-988. Schlindwein, C., D. Wittmann, C.F. Martins, A. Hamm, J.A. Siqueira, D. Schiffler, D. & I.C. Machado 2005. Pollination of Campanula rapunculus L.(Campanulaceae): How much pollen flows into pollination and into reproduction of oligolectic pollinators? Plant Systematics and Evolution 250: 147-156. Wittmann, D. & C. Schlindwein 1995. Melittophilus Plants, their pollen and flower visiting bees in Southern Brazil.1. Loasaceae. Biociências 3: 19-34. Zerbo, A.C., R.L.M.S. Moraes, & M.R. Brochetto-Braga 2001. Protein requirements in larvae and adults of Scaptotrigona postica (Hymenoptera: Apidia, Meliponinae): midgut proteolytic activity and pollen digestion. Comp Biochem Physiol 129: 139–147.