Original article

Interactions on floral resources between the Africanized honey bee Apis mellifera L and the native bee community ( : Apoidea) in a natural "cerrado" ecosystem in southeast Brazil

SR de Menezes Pedro, JMF de Camargo

Depto de Biologia, Faculdade de Filosofia, Ciências e Letras de Ribeirão Preto - USP 14049, Ribeirão Preto, São Paulo, Brazil

(Received 26 September 1990; accepted 15 May 1991)

Summary — Interactions between Apis mellifera and the native bee community, in 1 hectare of a re- growth "cerrado" with typical open-scrub vegetation and tropical climate, at Cajurú, São Paulo state, Brazil, were analysed for relative abundance, phenology and flower visits. Bees on flowers were net- ted every 14 d for a year, totalling 624 h of sampling. 4 086 individuals, representing 192 species and 6 families of Apoidea, visited 140 of the 184 plant species sampled in the area. A mellifera, the second most abundant bee species, was present in a small proportion of the floral sources visited by bees, some of them not primarily melittophilous. Most of the Anthophoridae, Halictidae and Mega- chilidae exploited plants not visited by A mellifera. Even the Meliponinae, the most similar in morpho- logical and behavioral attributes to Apis mellifera overlapped with Apis on few plant species. Africanized honeybee / Apoidea / Meliponinae / food competition / bee plant

INTRODUCTION and Ramalho, 1987; Knoll et al, 1987; Gottsberger et al, 1988; Roubik and More- After the introduction of African honey no, 1990). Surveys on flower visitors, pol- bees to South America in 1956 (Nogueira- len analyses and data from museum col- Neto, 1972), studies of their influence on lections have revealed the "generalist" native bee species have been published habits of the highly social bees in different (eg Roubik 1978, 1979, 1980, 1981, regions of America (Sakagami and Laro- 1983, 1988; Someijer et al, 1983; Posey ca, 1971; Roubik, 1979, 1988; Engel and and Camargo, 1985; Roubik et al, 1986; Dingemans-Bakels, 1980; Sommeijer et Imperatriz-Fonseca et al, 1987; Boreham al, 1983; Absy et al, 1984; Knoll et al, and Roubik, 1987; Cortopassi-Laurino 1987). In French and Guyana, Trigona (s lato) and Apis mellifera showed the MATERIALS AND METHODS highest generalization index for floral "preference" and the highest overlap with The study area was located at the Santa Carlota several other potential guilds (Roubik, Farm, Cajurú, São Paulo State, Brazil (21°18’- 1979). In Trinidad, West Indies, a palyno- 21°27’ S and 47°12’-47°20’ W), 700 m alt. The logical analysis of A mellifera and various original vegetation of this site is spread among several = stingless bees colonies placed in a resi- patches, amounting to 2 000 ha, and the areas 3 000 are dential area with second growth vegetation remaining (ca ha) occupied mainly by sugar cane, coffee and pastures. The revealed a wide and considera- spectrum study site has a typical open-scrub "cerrado" ble in resources for these overlap pollen vegetation, with scarce trees 3-5 m high and a bees (Sommeijer et al, 1983). The inter- predominance of shrubs and grasses. Until specific differences should be to some de- 1960, the site was used as pasture, after which gree representative and, despite the gen- it has been preserved, recovering naturally eralist nature of eusocial bee foraging, through seeding from adjacent areas. These "cerrados" are at the these differences may reflect a form of lim- peripherical disjunctions ited It has been assumed southern limits of the "core" area (plateau of specialization. central Brazil). that highly social bees are competitively Within ca 600 ha of a continuous "cerrado" superior to "less social" or solitary species range, 1 ha (400 x 25 m) was plotted and subdi- in the and harvest of floral re- discovery vided into 8 sub-areas (100.0 x 12.5 m) along and that extensive sources, competition each side of a pre-existent path. between A mellifera and bees for stingless The local climate is "savanna" tropical, with nectar and is since pollen likely, they 4 defined seasons. In the winter (May-August) present a high degree of generalization the daily mean temperature is 18.6° C; the mini- and overlap in flower species visited, time mum drops down to 0° C, and the monthly rain- and place of foraging (Roubik 1978, 1979, fall is 30.5 mm. During the warmest months (De- 1980, 1988). According to Roubik et al cember-February) the daily mean temperature is 24.6° the maximum reaches ≈ 36-38° C on (1986) extinction of some C, the warmest and the rainfall is colonies must occur as a result of days, monthly honey 247.7 mm. Annual mean rainfall for the last 30 bee The best record on im- competition. yr is 1 528.0 mm. Rainfall data for the region of Africanized bees has been pact (AHB) was provided by the Usina Amália Station and provided by the Kayapó Indians, who re- photoperiod data by the Instituto Agronômico de port quite exactly the arrival date of the Af- Campinas (fig 1). moon in rican honey bee (during full Feb- Samples were taken every 14 d for 1 yr, from ruary, 1966) in the region of Gorotire, 5-6 May 1988 to 20-21 April 1989 (fig 1). Each Pará, Brazil (Posey and Camargo, 1985). sampling was subdivided into 2 periods of 6 h: These bees began to attack and pillage 12.00-18.00 on the first day and 06.00-12.00 the nests of the Meliponinae, and other the second day, totalling 624 h in 26 sampling units. The was modified bees at flowers and water sources, but the sampling procedure from that of et al Two collec- of the Africanized bee is Sakagami (1967). aggressiveness tors alternately visited each sub-area for 30 min, said to have the na- diminished, allowing walking randomly and catching any bees on any tive bees to gather pollen and nectar (Pos- kind of flower, and staying ca 3 min at every ey and Camargo, 1985). flowering plant, which was examined regularly 2 h. Bees were or in This paper provides information about every captured individually net and were the interactions of Africanized A mellifera groups by separately preserved according to flower species and time. Plants with the native bee community in terms of were collected for identification and notes were relative abundance, phenology and flower taken on flowering time and resource available visits in a natural "cerrado" ecosystem. (nectar, pollen, oil). Air temperature and relative humidity were measured 2.4 m above the species that occur in southeastern Brazil, ground every 2 h. Nest numbers of A mellifera all have been observed in and near the and stingless bees were estimated, in and close study site, although only 19 were collected to the study site (within 50 m). Voucher speci- during this survey. mens of the plant and bee species were deposit- ed at the Department of Biology, University of Throughout the year, the relative abun- São Paulo, Faculdade de Filosofia, Ciências e dance of A mellifera changed from 0% to Letras de Riberão Preto. 61.4%, showing fluctuations unlike the general trends of the other bee (fig 1). Three abundance peaks were observed for RESULTS Apis, in September, October and April. The other bees, however, were most abundant in August and January, when the Meliponi- Relative abundance and phenology nae were predominant (87.7% of the bees sampled on 11 and 12 August 1988 and The total numbers of bees (species and in- 70.9% on 12 and 13 January 1989). S de- dividuals) per family of Apoidea collected pilis was especially abundant in August at Cajuru are summarized in table I. (35.6%) and T spinipes in January The predominant bee species (67.4%). Bee activity was reduced during were: Trigona spinipes (Fabricius), 697, the coldest months (June and July) and A 17.1%; A mellifera, 656, 16.0%; mellifera was absent, though over 100 indi- clavipes (Fabricius), 356, 8.7%; Paratrigo- viduals of other species were caught at na lineata (Lepeletier), 259, 6.3%; Scaptot- times (2-3 June 1988). Absence of Apis rigona depilis (Moure), 197, 4.8%; Tetrago- activity was also observed in late February nisca angustula angustula (Latreille), 157, and March, although during the same peri- 3.8%; Trigona hyalinata (Lepeletier), 109, od the activity of the other bees was only 2.7% (cf table II). Of the = 38 Meliponinae reduced.

Honey bees collected resources on Apis bees (29.6% of all bees collected); flowers more uniformly during the day, mostly Anthophoridae, Halictidae and Meg- even early in the morning (06.00-08.00 achilidae (55.6%, 52.2% and 64.6% of the and between 16.00 and 18.00 h. The other total for each family, respectively). were most abundant be- bees, however, A mellifera showed a preference for tween 10.00 and 14.00 h, with few individ- Compositae (251; 38.3%), Myrtaceae (87; uals foraging at 16.00-18.00. The great 13.3%), Lythraceae (57; 8.7%) and Rubia- number of at 06.00- Meliponinae foraging ceae (49; 7.5%; cf table III); it was the only 08.00 consisted of T col- mainly spinipes species with a preference for Myrtaceae. on notatum flowers lecting Paspalum Lythraceae and Rubiaceae also did not re- (table IV). ceive many visits from the other bee spe- Seven honey bee nests were counted cies. Anthophoridae, the second most at the study area, 3 inside the plot sur- abundant family in individual numbers, veyed, and 4 within 50 m of the site. showed a preference for Malpighiaceae Swarms in transit were observed in late and Compositae, Halictidae for Composi- August and early September, and the 3 tae, Labiatae and Sterculiaceae, Megachil- nests inside the area were initiated during idae for Compositae, Leguminosae and this same period, occupying armadillo Labiatae, Colletidae for Sapindaceae, La- holes within termite mounds. On 1 Decem- biatae and Lythraceae, and Andrenidae for ber 1988, two swarms had left the termite Labiatae and Solanaceae (table III). Com- mounds. Only one Meliponinae nest (P lin- positae, the most visited plant family, was eata, an underground nesting species) the most abundant in the area, both in was discovered in the plot. Nests of other terms of species and individuals. were located includ- species nearby. They Not all flowers visited by Apis are. ed: 2 nests of S depilis, 1 of Plebeia dror- melittophilous. Some of them present yana (Friese), 2 of T angustula and 1 of characteristics of myophily/cantharophily, testaceicornis Nannotrigona (Lepeletier). (Diplusodon virgatus, Myrcia spp, Cam- Nests of T spinipes, T hyalinata, Trigona pomanesia cambessediana), ornithophily truculenta Almeida and T clavipes were (Ananas ananassoides, Helicteres brevi- not located. One nest of Plebeia remota spira), psychophily (Lippia lasiocalycina, (Holmberg) and one of the necrophagous Alibertia sessilis) and anemophily (P nota- stingless bee Trigona hypogea Silvestri tum and Echinolaena inflexa). Of 13 plant were found, although they were not col- species with anthers with apical poricidal lected on flowers. dehiscence sampled in the area, Apis (3 individuals) visited only 2, Cambessedesia illicifolia and Cochlospermum regium. Oil Flower visits flowers also received visits from honey bees. Among 15 Malpighiaceae species, Of the 184 plant species collected at the Apis (22 individuals) visited 4, Banisteriop- sis Mas- study site, bees were sampled on 140 laevifolia, Byrsonima intermedia, plants belonging to 40 families. Of these, cagnia cordifolia and Tetrapteris sp (table only 47 (33%) were visited by A mellifera, II). ≈ 50% (24 spp) by < 5 individuals (table II). The nectar plants preferentially visited The other 93 plant species, most of them by Apis were Hyptis marruboides, Gochna- typically melittophilous (Pedro, unpub- tia barrosii, Rudgea viburnoides and Ver- lished data), were visited by 1 210 non- nonia spp, and pollen plants D virgatus,

Myrcia albartomentosa, Paspalum nota- Pollen flowers (P notatum, D virgatus) tum and Campomanesia cambessediana were primarily visited in the morning, while (table II). nectar flowers were visited in the after- Non-Apidae bees exhibited only a few noon. A mellifera foraged more uniformly floral "preferences" in common with Apis: throughout the day. On V ferruginea, T hya- Anthophoridae on Vernonia rubriramea, linata foraged especially between 08.00- and Halictidae on Hyptis marruboides and 12.00, S depilis between 10.00-14.00 and P notatum. Apidae other than Apis and T clavipes, 14.00-16.00 h (table IV). Meliponinae included Bombus spp, which were most abundant on H marruboides and V rubriramea (table II), and Euglossi- DISCUSSION nae (Pedro, unpublished data). Meliponinae were most abundant on 6 Relative abundance and phenology species preferentially visited by A mellifera (> 20 individuals; table II, fig 2): H marru- boides (P lineata, T angustula), Gochnatia A colony of T spinipes on average consists barrosii (T angustula, S depilis), D virgatus of 20 000-30 000 (JMFC, unpublished ob- (T spinipes), Vernonia ferruginea (T cla- servations) and an A mellifera colony of vipes, P lineata, S depilis, T hyalinata), V 2 000 > 50 000 individuals (Boreham and a nest inside a rubriramea (T clavipes), and P notatum (T Roubik, 1987). Thus, single area can an enormous num- clavipes, T spihipes); on the latter, 86% of given provide ber of on flowers, near the bees were T spinipes and only 6% foragers especially the nest et al, The were A mellifera. Other plants such as (Sakagami 1967). range A mellifera and T is Mascagnia cordifolia, Waltheria cf comu- exploited by spinipes the nis, Didymopanax vinosum, A ananas- quite large, considering large flight soides and E inflexa had high frequencies range of their workers (2 350 and 840 m, of Meliponinae, but Apis was an occasion- respectively; Kerr, 1959). al visitor (table II). The flowering periods of During the coldest period (June, July), the plants shared by Apis and Meliponinae there was a decrease in bee activity, in- and the foraging records are given in fig- cluding Apis and the Meliponinae, although ure 2. G barrosii, V ferruginea, and V rubri- Sakagami et al (1967) and Sakagami and ramea flowered during the same period. H Laroca (1971) observed that these 2 marruboides, V rubriramea and D virgatus, groups are relatively independent of climat- also had overlapping flowering periods. ic seasonal change. The increase in A mel- Abundance peaks of T angustula, T hyalin- lifera activity in September and October ata, T clavipes and S depilis were coinci- was probably related to swarm movements dent with the flowering of G barrosii, V fer- and the blooming of M alba-tomentosa and ruginea and V rubriramea. P lineata was in April to the flowering period of D virga- more constant during the year, being tus. Abundance of A mellifera and Melipon- abundant mainly when H marruboides inae was not correlated to variation in num- flowered. T spinipes showed a remarkable bers of flowering plant species (fig 1). abundance peak in January coinciding with the blooming of P notatum. A mellife- ra was most abundant when Myrcia alba- Floral preference tomentosa (6 and 7 October 1988), H mar- ruboides and D virgatus flowered (20 and In spite of its "generalist" habits in relation 21 April 1989). to floral resources, only a small share of all resources available in the study plot was As a whole, Meliponinae seem to be used by A mellifera. Although a large de- quite ecletic in terms of foraging styles: gree of overlap has been observed by small bees like Nannotrigona testaceicor- Roubik (1979), our data show that Apis nis perilampoides (Cresson), Tetragonisca does not share floral resources with most jatl (= angustula), Frieseomelitta nigra other bees. Apis was concentrated on flor- (Cresson), Scaura latitarsis (Friese) and S al types not attractive to the other Apoidea, longula (Lepeletier), Tetragona clavipes including M alba-tomentosa and D virga- and Plebeia sp (cf Wille, 1963; Laroca and tus, resources which are not primarily me- Lauer, 1973; Roubik, 1979), Leurotrigona littophilous. Nectar flowers like H marru- muelleri (Friese) and P lineata (personal boides and V rubriramea that were observation) harvest pollen grains left by bee intensively frequented by several spe- other bees on petals and leaves. Others cies of different families were the most like Trigona spp, are able to perforate the and distributed abundant, largely through- bases of long corolla flowers to obtain nec- out the area. Stingless bees, the group tar (Giorgini and Gusman, 1972; Roubik, most similar in morphological and behav- 1979,1982), and small Meliponinae (Trigo- ioral attributes to A mellifera, also exhibited nisca and Plebeia) have been observed some floral preferences in common with exploiting the perforations (Roubik, 1982). the latter. T spinipes overlapped with Apis They also bite poricidal anthers to collect in pollen foraging of D virgatus and P nota- pollen (Wille, 1963). Even "buzzing" behav- tum, the latter was occasion- although only ior is present in the genus Melipona (Buch- used as a source for A mellifera. ally pollen mann, 1983). Therefore, interference of Because of its size and flight pattern, T spi- Apis in the food niche of the Meliponinae can be a more efficient collector on nipes must be minimal. P notatum. V ferruginea was intensively used by the Meliponinae, particularly T We conclude that in the study area, A mellifera a small share of the hyalinata, T clavipes, S depilis; and some- occupies what by A mellifera. In spite of the overlap available resource (33%), half of them vis- among A mellifera, T angustula, T clavipes ited only occasionally (< 5 visitors), despite and S depilis, on G barrosii, V rubriramea the abundance of these plants. Further- and H marruboides, competitive interac- more, most of the bee plants present in the tions could not be assessed, since most of area have attributes that do not permit effi- the factors that reveal the occurrence of cient exploitation by Apis (eg oil flowers, this process were not observed. Although flowers with apical poricidal anthers, etc). aggressive interactions between T hyalina- Some of the pollen flowers primarily visited ta and T spinipes have been observed sev- by A mellifera are "peripheral" for the other eral times on different plant species and in bees. In this sense, the interference of the different places (JMFC, unpublished data) Africanized honey bee on food niche of the this was not observed in the study area. native bees must be minimal.

ACKNOWLEDGMENTS plantes sur les 184 échantillonnées dans la région. A mellifera, qui vient en 2e posi- We thank the technicians of the Depto de Biolo- tion pour l’abondance, est présente sur gia, USP - Ribeirão Preto, for the field and labor- 33% des fleurs visitées par l’ensemble des atory work; M Mazucato for the bee survey and abeilles, certaines d’entre elles n’étant pas JA Tavares Filho for the and climatic data. plant principalement mellifères. Ses sources The were identified MCH plant species by de sont vir- Mamed (Malpighiaceae), T Sendulski (Grami- principales pollen Diplusodon neae) and CFL Muniz (Cyperacerae), K Yama- gatus et Myrcia albatomentosa, celles de moto and H Leitão Filho, V Mantovani, and Ce- nectar sont Hyptis marruboides et Gochna- sar de (other plant families). We also thank D tia barrosii (tableau II, fig 2). La plupart des De Jong for revising the language and style, and Anthophoridae, des Halictidae et des Me- giving some suggestions for improving the gachilidae exploitent des plantes qui ne manuscript. Climatic data of the region were pro- vided by Usina Amália and photoperiod by the sont pas visitées par A mellifera (tableaux Instituto Agronômica de Campinas. We are es- II et III). Les spectres des plantes visitées pecially grateful to J de Sampaio Moreira Netto par A mellifera et les Meliponinae, qui ont and S de Sampaio Moreira Júnior, owners of the pourtant les caractères morphologiques et Santa Carlota Farm, Cajurú, SP, for providing in- éthologiques les plus proches d’A mellife- stallations and logistic support; and especially ra, ne se chevauchent sur for preserving the natural areas, a family tradi- que quelques tion which has been maintained for the past 100 espèces: Trigona spinipes, sur D virgatus years. This project received financial support et Paspalum notatum; Scaptotrigona depi- from the Financiadora de Estudos e Projetos Fl- lis, Trigona hyalinata, Tetragona clavipes conv 4388023300. SRM Pedro received a NEP, et lineata, sur Vernonia ferru- from the Conselho Nacional de Desenvol- Paratrigona grant ginea; S depilis et Tetragonisca angustula, vimento Cientifico e Tecnológico (CNPq) JMFC, research associate, CNPQ, Ref 300014/84-8/ sur G barrosii; P lineata et T angustula, sur Z0/FV H marruboides (fig 2). Une partie de la niche alimentaire occupée par A mellifera dans cet écosystème est «périphérique» Résumé — Interactions entre l’abeille pour les autres espèces, et les interféren- africanisée (Apis mellifera L) et les ces avec la communauté des abeilles non abeilles sauvages (Hymenoptera. Apoi- Apidae sont probablement minimes. dea) au niveau des ressources florales dans un écosystème naturel de «cerra- abeille africanisée / Apoidea / Meliponi- do», dans le Sud-Est du Brésil. On a étu- nae / compétition alimentaire / plante dié, du point de vue de l’abondance relati- mellifère / plante pollinifère ve, de la phénologie et des visites de fleurs, les interactions entre l’abeille africa- nisée (Apis mellifera) et la communauté Zusammenfassung — Wechselwirkun- d’abeilles sauvages sur une végétation ty- gen beim Blütenangebot zwischen afri- pique de brousse ouverte («cerrado»), à kanisierten Honigbienen (Apis mellifera Cajuru dans l’état de São Paulo (climat tro- L) und der heimischen Bienengemein- pical). Sur une parcelle d’un hectare, les schaft (Hymenopteren: Apoiden) in abeilles présentes sur les fleurs ont été einem natürlichen "cerrado" Ökosy- capturées au filet toutes les 2 semaines stem in Südostbrasilien. Die Wechselwir- durant une année. Ces 624 heures kungen zwischen Apis mellifera und den d’échantillonnage ont fourni 4 086 indivi- heimischen Wildbienen in einem "cerrado" dus représentant 192 espèces et 6 familles Sekundärbewuchs mit einer typischen offe- d’Apoïdes ayant visité 140 espèces de nen Buschvegetation und tropischem Klima wurden in Cajuru, Staat Säo Paulo, Meliponinae (Hymenoptera; Apoidea), para Brasilien, in Hinblick auf relative Häufig- coleta de pólen na região do médio Amazo- keit, Phänologie und Blütenbesuch analy- nas. Rev Bras Biol 44 (2), 227-237 siert. Ein ganzes Jahr hindurch wurden die Boreham MM, Roubik DW (1987) Population blütenbesuchenden Bienen auf einer change and control of Africanized honey bees in the Panama Fläche von 1 ha mit einem Netz (Hymenoptera: Apidae) abgefan- Canal area. Bull Entomol Soc Am 33, 34-39 gen. Das bedeutete 624 Stun- insgesamt Buchmann SL Buzz in den Probenfang. Auf 140 der insgesamt (1983) pollination angio- sperms. In: Handbook of Experimental Polli- 184 einbezogenen Pflanzenarten wurden nation Biology (Jones CE, ed) RJ Little, Von 4086 Einzeltiere aus 192 Arten und 6 Fa- Nostrand Reinhold NY, 73-113 milien der Apoidea gefangen. A mellifera, Cortopassi-Laurino M, Ramalho M (1987) On die zweithäufigste Bienenart, war nur auf the pollen harvest by Africanized Apis mellife- einem kleinen Teil (33%) der von Bienen ra and Trigona (Trigona) spinipes in São besuchten Blüten zu finden, zum Teil auf Paulo. In: Proc 10th Int IUSSI Congr: Chem- solchen, die nicht primär zu Bienenpflan- istry and Biology of the Social (Eder Rembold VJ zen zählen. Ihre waren J, H, eds) Peperny, Munich, Hauptpollenquellen 653-654 Diplusodon virgatus und Myrcia albato- mentosa, wichtigste Nektarquellen Hyptis Engel MS, Dingemans-Bakels F (1980) Nectar and resources for bees marruboides und Gochnatia barrosil pollen stingless (Meli- (Ta- poninae, Hymenoptera) in Surinam (South belle II, Abb 2). Die meisten Anthophori- America). Apidologie 11 (4), 341-350 dae, Halictidae und sammel- Megachilidae Giorgini JF, Gusman AB (1972) A importancia ten auf die von A mellifera nicht Pflanzen, das abelhas na polinizacão. In: Manual de besucht wurden (Tabelle II, III). Sogar die Apicultura (de Camargo JMF, ed) Editora Ag- meliponinen, in Gestalt und Verhalten am ronômica "Ceres" Ltda, São Paulo 155-214 ähnlichsten, überlappten mit Apis nur an Gottsberger G, Camargo JMF, Silberbauer- wenigen Pflanzenarten: Trigona spinipes Gottsberger I (1988) A bee-pollinated tropical auf D virgatus und Paspalum notatum; community: the beach dune vegetation of Scaptotrigona depilis, Trigona hyalinata, Ilha de São Luis, Maranhão, Brazil. Bot Tetragona clavipes und Paratrigona linea- Jahrb Syst 109 (4), 469-500 ta auf Vernonia ferruginea; S depilis, Te- Imperatriz-Fonseca VL, Kleinert-Giovannini A, tragonisca angustula auf G barrosil; P line- Ramalho M (1987) Stingless bees and Afri- ata und T auf H marruboides canized honey bees - resource sharing. In: angustula Proc 10th Int IUSSI and Der von A mellifera besetzte Teil Congr: Chemistry (Abb 2). Biology of the Social Insects (Eder J, Rem- der Futternische ist in diesem Ökosystem bold H, eds) VJ Peperny, Munich, 700-701 offensichtlich für andere Arten nur von pe- Kerr WE of As- und die (1959) Bionomy Meliponids.VI. ripherer Bedeutung Auswirkung pects of food gathering and processing in auf die Nicht-Apiden Gemeinschaft wahr- some stingless bees. In: Symp Food Gather- scheinlich minimal. ing Behav of Hymenoptera. Cornell Universi- ty, Ithaca, NY, 24-31 Afrikanisierte Biene / Apoidae / Melipo- Knoll FRN, Bego LR, Imperatriz-Fonseca VL ninae / Nahrungskonkurrenz / Bienen- (1987) Relative abundance and phenology of pflanze bees (Hymenoptera, Apoidea) in São Paulo. Brazil. In: Proc 10th Int IUSSI Congr: Chem- istry and Biology of the Social Insects (Eder REFERENCES J, Rembold H, eds) VJ Peperny, Munich, 702-703 Absy ML, Camargo JMF, Kerr WE, Miranda IPA Laroca S, Lauer S (1973) Adaptacão comporta- (1984) Espécies de plantas visitadas por mental de Scaura latitarsis para coleta de pólen (Hymenoptera, Apoidea). Acta Biol Par Roubik DW (1988) An overview of Africanized Curitiba 2 (1, 2, 3, 4) 147-152 honey-bee populations: reproduction, diet, and In: Proc Int Conf on African- Nogueira-Neto P (1972) Notas sobre a hist6ria competition. da apicultura brasileira. In: Manual de Apicul- ized Honey Bees and Bee Mites (Needham tura (Camargo JMF, ed) Editora Agronômica G, Delfinado-Baker M, Page R, Bowman C, Ceres Ltda, São Paulo, 17-32 eds) Ellis Horwood Ltd, Chichester, UK, 45- 54 Posey DA, Camargo JMF (1985) Additional Roubik DW, Moreno JE Social bees and notes on the classification and knowledge of (1990) stingless bees by the Kayap6 indians of palm trees: what do pollen diets tell us? In: Gorotire, Pará, Brazil. Ann Mus 54, Proc 11th Int Congr IUSSI: Social Insects and Carnegie Vi- 247-274 the Environment (Veeresh GK, Mallik B, raktamath CA, eds) Bangalore, India, 427-428 Roubik DW interactions be- (1978) Competitive Roubik Moreno Wittmann D tween and Africanized DW, JE, Vergara C, neotropical pollinators food with the Af- honey bees. Science 201, 1030-1032 (1986) Sporadic competition rican honey bee: projected impact on neo- Roubik DW (1979) Africanized honey bees, tropical social bees. J Trop Ecol 2, 97-111 and the structure of stingless bees, tropical Laroca S Relative abun- communities. In: Proc IVth Int Sakagami SF, (1971) plant-pollinator dance, and flower visits of on Pollination. Md Sta phenology apid Symp Agric Exp 1, bees in Eastern southern Brazil 403-417 Parana, (Hy- menoptera, Apidae). Kontyû 39 (3), 217-230 Roubik DW behavior of (1980) Foraging compet- SF, Laroca Moure JSM africanized and Sakagami S, (1967) ing honeybees stingless Wild bee biocenotics in São José dos Pinhais bees. (4), 836-845 Ecology 61 (PR), South Brazil. Preliminary report. J Roubik DW (1981) Comparative foraging behav- Fac Sci Hokkaido Univ Ser VI, Zool 16, 253- ior of Apis mellilera and Trigona corvina (Hy- 291 on Baltimora recta menoptera: Apidae) (com- Sommeijer MJ, De Rooy GA, Punt W, De Bruijn positae). Rev Biol Trop 29 (2), 177-183 LLM (1983) A comparative study of foraging Roubik DW (1982) The ecological impact of nec- behavior and pollen resources of various tar robbing bees and pollinating humming- stingless bees (Hymenoptera, Meliponinae) birds on a tropical shrub. Ecology 63 (2), and honeybees (Hymenoptera, Apinae) in 354-360 Trinidad, West Indies. Apidologie 14 (3), 205- Roubik DW (1983) Experimental community 224 studies: time-series tests of competition be- Wille A (1963) Behavioral adaptations of bees tween African and Neotropical bees. Ecology for pollen collecting from Cassia flowers. Rev 64 (5), 971-978 Biol Tiop 11 (2), 205-210