Original article

Resources foraged by Euglossa atroveneta (Apidae:Euglossinae) at Union Juárez, Chiapas, Mexico.

A palynological study of larval feeding

Elia Ramírez Arriaga Enrique Martínez Hernández

Instituto de Geología, Dpto. de Paleontología, Lab. de Palinología, Universidad NacionalAutônoma de México, Apartado Postal 70-296, 04510 Coyoacân, Mexico

(Received 25 August 1997; accepted 5 March 1998)

Abstract - Weekly larval feeding samples belonging to 51 nests of Euglossa atroveneta were ana-lyzed palynologically at Union Juárez, Chiapas, Mexico. It was determined that E. atroveneta foraged74 plant species throughout 1 year, from 31 May 1990 to 29 May 1991. The most important resourcesof larval feeding were classified as: a) polliniferous: Antirhea sp., Cassia sp., Combretum sp., Com-melina sp., Cordyline terminalis, Dendropanax sp., Leguminosae type 4, Lycianthes sp., Saurauia sp.,Solanum aff. torvum, Solanum aff. tuerckheimii, Tibouchinia longifolia; and b) nectaropollinife-rous: Drymonia aff. strigosa, Kohleria elegans, Lonchocarpus sp. and Nautilocalyx aff. panamensis.Through the year E. atroveneta displayed a heterogeneous foraging behavior with a diversity indexaverage of 0.83, and little overlap for resources between the solitary and parasocial nests studied.© Inra/DIB/AGIB/Elsevier, Paris

Euglossa atroveneta / larval feeding / palynology / Neotropics / orchid bees / throphic resources

1. INTRODUCTION

The Euglossinae are neotropical bees.They are solitary, communal and parasocialbee species, and they comprise five genera:Eulaema, Eufriesea, Euglossa, Exaerete andAglae. The Euglossa genus is the most abun-dant and diverse in the neotropics [17].

This group of bees is well known because

the male bees collect fragrances from orchidflowers and at the same time they carry outthe pollination, often with specific pollini-zers for each orchid species. These rela-tionships are considered to constituteexamples of coevolution [2, 5, 6, 16, 29,32].

* Correspondence and reprintsE-mail: elia@servidor.unam.mx

The Euglossinae have a broad flight pat-tern and sometimes fly as far as 23 km [8].Due to this capacity to travel for collectingpollen and nectar, they are one of the mainagents for cross pollination in tropical rainforest.

These bees visit plants with wide distri-bution and long flowering periods [1, 8],plants that offer abundant food reward (nec-tar or pollen), or that provide resin that isused in nest construction. Records relatedto plant species of feeding importance inEuglossinae have been based exclusivelyon direct field observations [1, 8, 15, 31].However, many flowering plants visited byEuglossine bees are unknown to science. Itis important to increase knowledge of howthe Euglossinae exploit the angiosperms andto find out what taxa are most relevant intheir life cycle. In this way, it will be pos-sible to better understand the role of female

euglossine bees in tropical ecosystems.The utilization of floral resources may

be assessed by direct observation or by paly-nological analysis of larval feeding and mea-suring ecological parameters such as pol-len diversity, evenness and overlapping offeeding resources [4, 9, 12-14, 20-22, 28].

For this research, a nesting area was esta-blished for Euglossa atroveneta in a distur-bed zone on the hillside of Tacaná volcano,in Chiapas State. It was possible to assess theannual foraging behavior of female beesthrough palynological analysis of larvalfeeding and to complement these data bymeans of field observations. Finally, theresults were compared with those cited forstingless bees (Apidae; Meliponinae) stu-died in the same area, where the overlap inresource utilization was also determined.

2. MATERIALS AND METHODS

The studied nests of E. atroveneta are locatedat Unión Juárez in Chiapas (15°5’N, 92°3’W),at an altitude of 1 400 m. In this region, secondaryvegetation of tropical rain forest is found in

patches and the landscape is dominated by cropsof various coffee varieties (Coffea arabica).

The larval feeding samples were collectedfrom open cells in 51 nests of E. atroveneta main-tained in wooden boxes [23], to know whichresources were currently being exploited. During1 year, from 31 May 1990 to 27 May 1991, atotal of 143 weekly samples were collected: insolitary nests, one sample per nest and in para-social nests, from 1 to 5 weekly samples per nest.To remove the larval food a wooden stick wasused and each sample was put in individual tubes,sealed carefully and labeled with date, nest andcell number.

All larval feeding samples were acetolyzed,following Erdtman’s [7] method, and the slidesobtained deposited in the palynological collectionof the Institute of Geology, at the National Uni-versity of Mexico (IGLUNAM).

The pollen grains of each sample were des-cribed and identified, consulting palynologicalreference collections (IGLUNAM) from Chia-pas and Los Tuxtlas, Veracruz, and by means ofMexican catalogues of tropical pollen [10, 11,18, 24] and following floristic lists [3]. For eachsample, the frequencies of the plant taxa werebased on 1 200 pollen grains counted randomly[30].

The important species were those present in10 % or higher percentages [21]. The flower mor-phology of important taxa was reviewed. Somefield observations of plants visited by the beeswere made also.

Based on palynological results, three ecolo-gical parameters were calculated monthly: theShannon-Weaver index [26] (H’), Pielou’s for-mula [19] (J’) and the Schöener index [25] (PS’).

The Shannon-Weaver Index (H’) expressesthe palynological diversity of the samples, consi-dering the total number of plant species and therelative frequency of each taxa in the sample

where H’ is the diversity index, pi is the propor-tion of each pollen type found in samples of nestin question during the month considered, and inthe natural logarithm.

Pielou’s formula (J’) was used to determinethe uniformity of pollen species at each sample.The J’ value can vary from 0 to 1, and is a mea-sure of whether a sample is heterogeneous orhomogeneous

where J’ is the evenness, H’ is the diversity index,and H’ max is the natural logarithm of the totalnumber of pollen types present in the samples.

Schöener (PS) index was calculated to deter-mine the overlapping of resources among the E.atroveneta nests. This index can vary from 0 to1, when two samples are different or similar, res-pectively. The overlap in feeding sources wascalculated monthly

where PS is the resource overlap, phi the pro-portion of pollen from species h collected byE. atroveneta at nest i, and phj is the respectivevalue for the nest j.

The palynological analysis of larval feeding,the study of the flower morphology and the fieldobservation were used to show the collecting stra-tegies of E. atroveneta at Unión Juárez. The eco-logical parameters employed (pollinic diversity, theevenness and overlapping of feeding resources), are also important tools to evaluate foraging beha-vior in the Apidae [12-14, 20-22, 27].

3. RESULTS

During the year of this research, 51 nestsof E. atroveneta were studied. Through paly-nological analysis, 67 plant species belong-ing to 42 families and seven undeterminedtaxa were registered. Of the total taxa obser-ved, only 22 were of importance (≥ 10 %) inthe E. atroveneta diet at the Unión Juárezarea (table I).

The total number of plant species foundfor each nest varied from 4 to 25. The majornumber of taxa (from 20 to 25) was regis-tered only in five parasocial nests.

Generally, only a few species wereexploited with high intensity in solitary andparasocial nests. The species with low fre-quencies could be considered as alternativeresources (table I).

In the case of solitary nests, i.e. nestsconstructed by only one female (n = 42),from one to six species of importance were

registered, with an average of three impor-tant taxa per sample. In parasocial nests(n = 9), three to nine important taxa wereobserved, with an average of five importantspecies per sample.

In this species, some female bees werepolylectic (figure 1), while some others hadan oligolectic behavior (figure 2).

The important resources (> = 10 %) forE. atroveneta were exploited during shorttime periods from 1 to 3 months and consis-ted of Dendropanax sp., Commelina sp.,Kohleria elegans, Cordyline terminalis,Antirhea sp., Rondeletia sp., Solanum aff.

torvum, Solanum aff. tuerckheimii, Cassiasp. and four undetermined species. Otherresources registered during medium timeperiods, from 4 to 6 months were Saurauiasp., Begonia biserrata, Combretum sp.,Drymonia aff. strigosa, Nautilocalyx aff.panamensis and Leguminosae type 4. Thespecies, visited during long flowering per-iods from 7 to 11 months, were Loncho-carpus sp., Tibouchinia longifolia andLycianthes sp.

The monthly analyses showed from nineto 39 plant species, but only from two to tenimportant species per month were recorded.The major number of taxa was registered inJune, July, September, October, January andFebruary (table I).

The flower morphology of these taxawas reviewed and nectaropolliniferous spe-cies with tubular flowers were detected

(Drymonia aff. strigosa, Kohleria elegans,Nautilocalyx aff. panamensis, Rondeletiasp.) and a Fabaceae (Lonchocarpus sp.)(figure 3). Also, two genera had been asso-ciated with buzz pollination for Euglossinaebees: Cassia and Solanum [15, 31].

The important polliniferous species wereSaurauia sp., Dendropanax sp., Begoniabiserrata, Combretum sp., Commelina sp.,Cassia sp., Leguminosae type 4, Cordylineterminalis, Tibouchinia longifolia, Antirheasp., Lycianthes sp., Solanum aff. torvum,S. aff. tuerckheimii and four undeterminedtaxa.

The resources of feeding importancewere from the secondary vegetation of thisarea, most of them constituted the arborealand bushy strata, and few plant speciesbelong to the herbaceous stratum.

Some important plants, such as Begoniabiserrata, Commelina sp., Cordyline termi-

nalis, Saurauia sp. and Solanum torvumwere within a few meters around E. atro-

veneta nests. Other taxa visited exclusivelyto collect nectar and which were near the

nesting area included Elettaria cardamo-mum (cultivated species) and Salvia purpu-rea.

3.1. Foraging behavior

3.1.1. Pollinic diversity (H’)

In the population studied, the diversityindex by nest (n = 51), varied from 0.12 to2.04, with a population average of 1.12 ±

0.19. However, most of the data were be-tween 0.83 and 1.3 (figure 4).

3.1.2. Evenness (J’)

Regarding the evenness (n = 51), thevalues varied from 0.07 to 0.75, with an ave-

rage of 0.48 ± 0.02, which means that E. atro-veneta had heterogeneous collecting sources(figure 5). The major record numbers obtai-ned were registered from 0.4 to 0.6.

3.1.3. Overlapping of feeding resources (PS)

The overlapping values (n = 300) variedfrom 0.002 to 0.95 with a mean of 0.38

± 0.06. On the other hand, the main recordswere grouped in the 0.19-0.57 interval.

In general, when the E. atroveneta popu-lation was increased, the major number ofplants and the major number of high over-lapping indexes were registered. Theseoccurred during June, July, September, Octo-ber, January, February and May (figure 6).

The overlapping behavior occurred whenE. atroveneta females foraged over Sau-rauia sp., Dendropanax sp., Chamaedoreasp., Impatiens wallerana, Begonia biser-

rata, Alnus sp., Betulaceae type 1, Cheno-

podiaceae-Amaranthaceae, Combretum sp.,Evolvulus sp., Dioscorea sp., Drimonia aff.

strigosa, Kohleria elegans, Nautilocalyx aff.panamensis.

4. DISCUSSION

The present study is the first palynolo-gical research of larval feeding carried out insolitary and parasocial E. atroveneta nests.

The E. atroveneta population studied waspolylectic. All the plant species with valuesof importance for E. atroveneta were exclu-sively polliniferous and nectaropollinife-rous from the arboreal and herbaceous strata

belonging to secondary vegetation. Despitethe abundance of Coffea arabica in the sur-roundings, this species was not an impor-tant feeding resource.

Some important genera for E. atroveneta,such as Solanum, Cassia, Drymonia andTibouchinia had been previously cited forother Euglossinae through field observa-tions.

With respect to resource availability, theimportant species Begonia biserrata, Com-melina sp., Cordyline terminalis, Saurauiasp. and Solanum torvum were near the nests

although, these plants were not in abun-dance. This forage behavior differs fromthat cited by Janzen [8], who mentioned thatEuglossinae moves long distances in the

Costa Rican tropical forest, to obtainresources and it can take up to 2 h even fora single trip. It is possible that Euglossa atUnión Juárez visited nearby plants that wereattractive because their exploitation involvedless energy expenditure. It is possible thatthey do travel long distances during periodsof scarcity. This is indicated by pollen col-lecting times that vary from 5 min to 2 h[23].

The differences in the foraging behaviorobserved between the Costa Rica Euglos-sinae and that of E. atroveneta at Unión Juá-

rez, Chiapas, could be related to differencesin the types of vegetation. Hence, Janzencarried out his observations in an undistur-bed tropical rain forest, while the presentstudy was made in a perturbed tropical com-munity. Janzen [8] cited that the importantfood plants for ’orchid bees’ produce a smallnumber of flowers over long periods of time.However, E. atroveneta visited not onlyspecies with long flowering periods, but alsoforaged plants that flowered for mediumand short time periods.

Female bees that constructed their nestsat the same time had similar forage behav-ior and visited the same taxa, although theintensity of exploitation could change. Thiswas reflected in the frequencies registered inlarval feeding samples. There were also dif-ferences in female behavior, possibly relatedto the individual preferences for some flow-ers. This indicates a level of plasticity in thediet of these bees.

Differences were also observed in for-

aging behavior between solitary and para-social nests (i.e. figures I and 2). The differ-ences in the higher number of taxa ofimportance may reflect that parasocial nestsexisted for a longer period of time and it ispossible that during this interval, the plantsregistered at the beginning finished flow-ering and then other plant species started tobloom and were visited by Euglossafemales. The previous statement could betested with the following regression analy-sis with the equation:

where Y is the number of plant species(dependent variable), and X the time of nestconstruction (independent variable).

The result gave a value for R-squared of35.4, which means that 35.4 % of plant spe-cies variation depends on the time of nestconstruction.

Also, differences between solitary andparasocial nests may be due to the particu-lar preferences of each female bee. In para-social nests two or more bees cohabited, butthere was no communication among themon the location of the important resources, asin social bees. Then, it is possible that eachfemale had different preferences at the sametime and thus increased the species rich-ness. This could be proved with a regres-sion where 26.7 % of sample richness is dueto the variation of number of female beesin the same nest. The regression equationwas:

where Y is the number of plant species(dependent variable), and X the number offemale bees by nest (independent variable).

Comparing the results obtained in thepresent study with those from other studiesfor stingless bees (eusocial bees) in the samearea, it was observed that E. atroveneta visi-ted more plant species than stingless bees.This may be due to E. atroveneta displace-ment ability that permits it to forage overmore polliniferous and nectariferousresources in a given time.

In addition, the E. atroveneta larval feed-ing analysis showed that the important tro-phic species are polliniferous and pollini-ferous-nectariferous. Similar results wereobtained for Melipona and Trigonini [12,14, 22, 27], indicating that both bee groupsfollow the same patterns for foraging to opti-mize their resources.

It is worth noting that there were impor-tant common taxa visited by Melipona, Tri-gonini and E. atroveneta in Unión Juárezwhich were: Saurauia sp., Commelina sp.,

Lonchocarpus sp., Cordyline terminalis,Tibouchinia longifolia, Rondeletia sp. andSolanum sp. [ 12, 14, 22, 27]. Particular flo-ral preferences were registered in the E. atro-veneta population, where female beesexploited the following plant species: Den-dropanax sp., Begonia biserrata, Combre-tum sp., Drimonia aff. strigosa, Kohleriaelegans, Nautilocalyx aff. panamensis, Cas-sia sp., Leguminosae type 4, Antirhea sp.,Lycianthes sp., Solanum aff. torvum andSolanum aff. tuerckheimii.

So far, it is evident that a resource par-tition exists among the different Apidaemembers that had been studied at the samezone. Although there is overlapping in feed-ing resources, there are differences in inter-specific and intraspecific forage strategies.

4.1. Ecological aspects

According to the palynological diversityregistered in the samples, the H’ values werehigh throughout the year of study. Theseresults indicated that E. atroveneta collectedits resources heterogeneously. The hetero-geneous foraging in space and time deter-mined that euglossas have different prefe-rences to collect their resources. Nevertheless,they exploited more intensely only a fewplants, and even E. atroveneta had very spe-cific foraging.

In general, the values of PS show a lowamount of overlapping, proving that

although Euglossa females visited the sameblooming resources, these were exploitedin a different way. Higher levels of over-lapping occurred only during the rainy sea-son and at the beginning of the year, coin-ciding with an increase in the E. atrovenetapopulation [23].

If the H’ values of E. atroveneta are com-

pared with those cited for maximum andminimum H’ and J’ values for Tetragoniscaangustula [27] in the same area, the majordiversity in larval feeding samples had beenregistered in both Tetragonisca and E. atro-

veneta.

With respect to the evenness (J’), thehighest index was registered in T. Angus-tula [27], meaning that this bee forageshomogeneously for its resources, and E.atroveneta follows in importance. Incontrast, the lowest evenness data were

registered in E. atroveneta, indicating thatthis bee can collect its resources heteroge-neously, when it has specific floral prefe-rences.

ACKNOWLEDGMENTS

This research was supported by CONACyT(reg. 55888). The authors are grateful to DrMichener and Dr Dressler for their importanttaxonomical contributions as well as MagdalenaAlcayde for reviewing the English version and toM. en C. Lourdes Flores Delgadillo for her sta-tistical advice.

Résumé - Plantes butinées par Euglossaatroveneta (Apidae : Euglossinae) à UniónJuárez, Chiapas, Mexique. Une étudepalynologique de la nourriture larvaire.Des échantillons de nourriture larvaire ontété prélevés chaque semaine durant uneannée, du 31 mai 1990 au 29 mai 1991, dans51 nids d’Euglossa atroveneta à Union Jua-rez, dans le Chiapas (Mexique). Leur ana-lyse palynologique a montré que E. atrove-neta butinait 74 espèces de plantes (tableauI). On a observé des différences dans le com-portement de butinage entre nids solitaires etnids parasociaux (figures 1 et 2), différencesqui sont liés à la période de nidification etaux préférences de chaque femelle. Aucours de l’année E. atroveneta a présenté unindice moyen de diversité de 0,83 (figure3) et un comportement de butinage hétéro-gène (J’= 0,45) (figure 4) avec un faiblerecouvrement des ressources (PS moyen =0,38) entre les nids solitaires et les nids para-sociaux étudiés (figure 5). On a enregistré leplus grand nombre d’espèces butinées et leplus grand indice de recouvrement pour lespopulations les plus nombreuses. Les res-

sources les plus importantes pour la nourri-ture larvaire sont au nombre de 22 et ontété classées en : a) espèces pollinifères :quatre taxons indéterminés, Anthirhea sp.,Begonia biserrata, Cassia sp., Combretumsp., Commelina sp., Cordyline terminalis,Dendropanax sp., Légumineuses type 4,Lycianthes sp., Rondeletia sp., Saurauia sp.,Solanum aff. torvum, Solanum aff. tuerck-

heimii, Tibouchinia longifolia et b) necta-ropollinifères: Drymonia aff. strigosa, Koh-leria elegans, Lonchocarpus sp. et

Nautilocalyx aff. panamensis et toutes lesGesneriaceae à fleurs tubulaires. Certainesde ces espèces sont associées à la pollinisationvibratile (Cassia et Solanum). © Inra/-DIB/AGIB/Elsevier, Paris

Euglossa atroveneta / ressource alimen-taire / nourriture larvaire / analyse pol-linique

Zusammenfassung - Nahrungsressour-cen von Euglossa atroveneta (Apidae:Euglossinae) in Unión Juárez, Chiapas,Mexiko. Eine palynologische Untersu-chung der Larvenernahrung. Bei 51 Nes-tern von Euglossa atroveneta wurdenwöchentlich Proben von der Larvennahrungentnommen und palynologisch untersucht.Es zeigte sich, daß E. atroveneta im Ver-lauf eines Jahres, vom 31 Mai ( 1990) biszum 29 Mai (1991) 74 Pflanzenarten besam-melte, von denen 22 Arten vermehrt beflo-gen wurden. Dabei wurden Unterschiedeim Sammelverhalten zwischen Einzel- und

parasozialen Nestern beobachtet (Abb. 1und 2), die mit der Zeit des Nestbaus undbesonders mit individuellen Bevorzugun-gen einzelner Weibchen zusammenhingen.Während des Jahres zeigte E. atrovenetaeinen durchschnittlichen Diversitätsindexvon 0,83 (Abb. 3) und ein heterogenes (J’=0,45) Sammelverhalten (Abb. 4). Zwischenden einzelnen und den parasozialen Nesternergab sich eine durchschnittliche Über-schneidung der Nahrungsressourcen von PS= 0,38 (Abb. 5). Bei zunehmender Popula-tionsdichte von E. atroveneta war die

Anzahl der besammelten Arten am größtenund der Wert des Überlappungsindex amhöchsten. Als für die Larven wichtigste Nah-rungsressourcen wurden folgende Artenbestimmt: a) die Pollenerzeuger: Antirheasp., Begonia biserrata, Cassia sp., Com-bretum sp., Commelina sp., Cordyline ter-minalis, Dendropanax sp., der Legumino-sen Typ 4, Lycianthes sp., Sauria sp.,Soanum aff. torvum, Solanum aff. tuerck-heimii, Tibouchinia longifolia, und b) dieNektar- und Pollenerzeuger: Drvmonia aff.strigosa, Kohleria elegans, Lonchocarpussp. and Nautilocalvx aff. panamensis. AlleGesneriaceae hatten Röhrenblüten. © Inra/-

DIB/AGIB/Elsevier, Paris

Euglossa atroveneta / Larrennahrung /Palynologie / neotropisch / Orchideen-bienen / Nahrungsressourcen

REFERENCES

[1] Ackerman J.D., Euglossine bees and their nectarhosts, in: D’Arcy W.G., Correa A.M.D. (Eds),The Botany and Natural History of Panamá, Mis-souri Botanical Garden, St. Louis, 1985, pp.225-233.

[2] Ackerman J.D., Geographic and seasonal varia-tion in fragrance choices and preferences of maleEuglossine bees, Biotropica 21 (4) (1989) 340-347.

[3] Breedlove D.E., Listados Florísticos de México.Flora de Chiapas, Instituto de Biología, IV, Uni-versidad Nacional Autónoma de México, 1986,246 p.

[4] Cortopassi-Laurino M., Ramalho M., Pollen har-vest by africanized Apis mellifera and Trigonaspinipes in Sao Paulo: botanical and ecologicalview, Apidologie 19 (1) (1988) 1-24.

[5] Dressler R.L., The Orchids. Natural History anClassification, MA. Harvard University Press,Cambridge, 1981, 332 p.

[6] Dressler R.L., Biology of the orchid bees (Euglos-sini), Annu. Rev. Ecol. Syst. 13 (1982) 373-394.

[7] Erdtman G., The acetolisis method. A revised

description, Svensk Bot. Tidskr. 54 (4) (1960)561-564

[8] Janzen D.H., Euglossine bees as long distancepollinators of tropical plants, Science 171 (1971)203-205.

[9] Kleinert-Giovannini A., Imperatriz-Fonseca V.L.,Aspects of the trophic niche of Melipona margi-nata marginata Lepeletier (Apidae, Meliponi-nae), Apidologie 18 (1) ( 1987) 68-100.

[10] Lozano-García M.S., Martínez-Hernández E.,Palinología de los Tuxtlas: especies arbóreas, Ins-tituto de Biología, Publicaciones especiales 3,Universidad Nacional Autónoma de México,1990, 61 p.

[11] Martínez-Hernández E., Cuadriello-Aguilar J.I.,Ramírez-Arriga E., Medina C.M., Sosa N.M.S.,Melchor S.J.E., Atlas de las plantas y el polenutilizados por las cinco especies principales deabejas productoras de miel, en la region delTacaná, Chiapas, México, Instituto de Geología,Universidad Nacional Autónoma de México,1993, 105 p.

[12] Medina C.M., Explotación de recursos floralespor Nannotrigona testaceicornis (Apidae) en doszonas con diferente altitud y vegetación en elSoconusco, Chiapas, Universidad Nacional Autó-noma de México, Facultad de Ciencias, Tesis deLicenciatura, 1989, 125 p.

[13] Medina C.M., Contribución al conocimiento dealgunos aspectos ecológicos en relación a la floraapícola explotada por abejas europeas (Apis mel-lifera ligustica, Spinola), abejas africanizadas(Apis mellifera scutellata, Lepeletier),e híbridosen el Soconusco, Chiapas. Universidad NacionalAutónoma de México, Facultad de Ciencias, Tesisde Maestría en Biología, 1992, 154 p.

[14] Melchor S.J.E., Explotación de recursos floralespor Scaptotrigona pachysoma (Apidae) en doszonas con diferente altitud y vegetación en elSoconusco, Chiapas, Universidad Nacional Autó-noma de México, Facultad de Ciencias, Tesis deLicenciatura, 1991, 231p.

[15] Michener C.D., An interesting method of pollencollecting by bees from flowers with tubularanthers, Rev. Biol. Trop. 10 (2) (1962) 167-175.

[16] Michener C.D., The Social Behavior of the Bees.A Comparative Study, Harvard University Press,Cambridge, Massachusetts, 1974, pp. 257-260.

[17] Michener C.D., Classification of the Apidae(Hymenoptera), Univ. Kans. Sci. Bull. 54 (4) 9(1990) 75-164.

[18] Palacios-Chavez R., Ludlow-Wiechers B., Vil-lanueva G.R., Flora palinológica de la reserva dela biosfera de Sian Ka’an, Quintana Roo, México,Centro de Investigaciones de Quintana Roo, Che-tumal, 1992, 321 p.

[19] Pielou E.C., Mathematical Ecology, 2nd ed.,Willey-Interscience Publication, John Wiley &

Sons, New York, 1977, 385 p.

[20] Quiroz G.D.L., Patrones Estacionales de utiliza-ción de recursos florales por Scaptotrigona hell-wegeri en la Estación de Biología Chamela,Jalisco, Universidad Nacional Autónoma de México,Tesis de Maestría en Biología, 1993, 148 p.

[21] Ramalho M., Imperatriz-Fonseca V.L., Kleinert-Giovannini A., Cortopassi-Laurino M., Exploi-tation of floral resources by Plebeia remota Holm-berg (Apidae, Meliponinae), Apidologie 16 (3)(1985) 307-330.

[22] Ramírez A.E., Explotación de recursos floralespor Plebeia sp. (Apidae) en dos zonas con dife-rente altitud y vegetación en el Soconusco, Chia-pas, Universidad Nacional Autónoma de Mexico,Facultad de Ciencias, Tesis de Licenciatura, 1989,159 p.

[23] Ramírez A.E., Cuaddriello-Aguilar J.I., Martí-nez-Hernández E., Nest structure and parasite ofEuglossa atroveneta Dressler (Apidae: Bombi-nae: Euglossini) at Unión Juárez, Chiapas,México, J. Kans. Entomol. Soc. 69 (2) (1996)144-152.

[24] Roubik D.W., Moreno P.J.E., Pollen and sporesof Barro Colorado island, Missouri Bot. Gard.,1991, 268 p.

[25] Schöener T.W., The Anolis lizards of Bimini:resource partitioning in a complex fauna, Eco-logy 49 (4) (1968) 704-726.

[26] Shannon C.E., Weaver W., The MathematicalTheory of Communication, Univeristy Of Illi-nois Press, Urbano, 1949, 1 17 p.

[27] Sosa N.M.S., Explotación de recursos floralespor Tetragona jaty en dos zonas con diferentealtitud y vegetación en el Soconusco, Chis, Uni-versidad Nacional Autónoma de Mexico, Facultadde Ciencias, Tesis de Licenciatura, 1991, 98 p.

[28] Sosa N.M.S., Martínez-Hernández E., Lozano-Garcia M.S., Cuadriello-Aguilar J.I., Nectaro-polliniferous sources used by Trigona (Tetrago-nisca) angustula in Chiapas, southern México,Grana 33 (1993) 225-230.

[29] van der Pijl L., Dodson C.H., Orchid Flowers:their Pollination and Evolution, University MiamiPress, Coral Gables, Florida, 1969.

[30] Vergeron Ph., Interprétation statistique des résul-tats en matière d’analyse pollinique des miels,Ann. Abeille 7 (4) (1964) 349-364.

[31] Wille A., Behavioral adaptations of bees for pol-len collecting from Cassia flowers, Rev. Biol.Trop. 11 (2) (1963) 205-206.

[32] Williams N.H., Whitten W.M., Orchid floral fra-gances and male euglossine bees: methods andadvances in the last sesquidecade, Biol. Bull. 164(1983) 355-385.