Pollination Biology of Eulophia Alta Orchidaceae

8/3/2019 Pollination Biology of Eulophia Alta Orchidaceae

1/16

Pollination biology of Eulophia alta (Orchidaceae) in Amazonia: effects ofpollinator composition on reproductive success in different populations

Andreas Jurgens1,*, Simone R. Bosch2, Antonio C. Webber3, Taina Witt1, Dawn Frame4

and Gerhard Gottsberger2

1School of Biological and Conservation Sciences, University of KwaZulu-Natal, P. Bag X01 Scottsville, Pietermaritzburg 3209,

South Africa,2Botanischer Garten und Herbarium, Universitat Ulm, Hans-Krebs-Weg, D-89081 Ulm, Germany,

3Departamento

de Biologia ICB, Universidade Federal do Amazonas, Estrado do Contorno 3000, BR-69077-000 Manaus, AM, Brazil and4IRD, UMR AMAP, Montpellier, F-34398 France

Received: 2 April 2009 Returned for revision: 4 June 2009 Accepted: 1 July 2009 Published electronically: 8 August 2009

Background and Aims Spatial variation in pollinator composition and abundance is a well-recognized phenom-enon. However, a weakness of many studies claiming specificity of plantpollinator interactions is that they areoften restricted to a single locality. The aim of the present study was to investigate pollinator effectiveness of thedifferent flower visitors to the terrestrial orchid Eulophia alta at three different localities and to analyse whether

differences in pollinator abundance and composition effect this plants reproductive success.Methods Natural pollination was observed in vivo, and manipulative experiments were used to study the pollina-tion biology and breeding system of E. alta at three sites near Manaus, Brazil. To gain a better understanding ofthe underlying mechanisms of pollinator attraction, nectar composition and secretion patterns were also studied,floral scent composition was analysed and a bioassay was conducted.Key Results Flower visitors, pollinator composition, pollinia transfer efficiency of particular pollinator speciesand natural fruit set differed among the investigated populations of E. alta. Flowers were self-compatible, par-tially autogamous and effectively pollinated by five bee species (four Centris species and Xylocopa muscaria).Visiting insects appeared to imbibe small amounts of hexose-rich nectar. Nectar sugar content was highest on thethird day after flower opening. Floral fragrance analyses revealed 42 compounds, of which monoterpenes andbenzenoids predominated. A bioassay using floral parts revealed that only floral tissue from the labellumchamber and labellum tip was attractive to flower visitors.Conclusions The data suggest that observed differences in reproductive success in the three populations cannotbe explained by absolute abundance of pollinators alone. Due to behavioural patterns such as disturbance ofeffective pollinators on flowers by male Centris varia bees defending territory, pollinia transfer efficiencies of

particular pollinator species also vary between study sites and result in differing reproductive success.

Key words: Eulophia alta, Orchidaceae, floral biology, floral volatiles, GC-MS, nectar composition, pollinatorperformance, reproductive success.

I N TR O D U C TI O N

Plant species visited by similar functional pollinator types (pol-linator groups sensu Fenster et al., 2004) exhibit suites of floralcharacters that researchers attribute to convergent floral adap-tations to the behaviour, morphology and physiology of pollina-tors (Johnson and Steiner, 2000; Fenster et al., 2004).Plantpollinator interactions are frequently regarded as tightlycoevolved andhighly specialized, andtherehas been a long-heldbelief that general evolutionary trends in pollination systemstend towards increased specialization; this concept was enun-ciated by Stebbins most effective pollinator principal(MEPP), in which he stated that . . . the characteristics offlowers will be moulded by those pollinators that visit it mostfrequently and effectively (Stebbins, 1970: 318).

However, over the last decade, as part of the debate aboutthe underlying principles of specialization in pollinationsystems, this concept has been re-evaluated (Ollerton, 1996;Waser et al., 1996; Johnson and Steiner, 2000; Aigner,

2001; Fenster et al., 2004; Waser and Ollerton, 2006). Dataanalysis of flower visitors and their occurrence on flowers inplant communities has shown that highly specialized systems(wherein a plant species is associated with a single flowervisitor species) are actually rather rare, and that generalizedrelationships are most common (see Ollerton, 1996; Waseret al., 1996). Running counter to this general phenomenon,most Orchidaceae are considered highly specialized with

respect to their flower visitor interactions (Tremblay, 1992)and coevolution in this family has resulted in many interestingand unusual pollination syndromes (van der Pijl and Dodson,1966; Faegri and van der Pijl, 1979; Johnson et al., 1998;Pauw, 2006). An analysis of published data on orchid pollina-tors by Tremblay (1992) revealed that there are on averageabout three pollinator species per orchid species.

Specialized plant pollinator interactions are particularly vul-nerable to pollen and pollinator limitation, which can affect repro-ductive success (Burd, 1994; Ashman et al., 2004; Burd et al.,2009). In a review of the reproductive success of nectariferousand nectarless orchids, Neiland and Wilcock (1998) noted that* For correspondence. E-mail [email protected]

# The Author 2009. Published by Oxford University Press on behalf of the Annals of Botany Company. All rights reserved.

For Permissions, please email: [email protected]

Annals of Botany 104: 897912, 2009

doi:10.1093/aob/mcp191, available online at www.aob.oxfordjournals.org


2/16

with respect to fruit production, tropical orchid species are onaverage only one-third as successful as their temperate counter-parts, irrespective of whether a species produces nectar or not.Their tropical orchid data set is biased in that it is drawn fromCentral American epiphytic orchids, a natural consequence ofthe meagre data available on tropical terrestrial species. Aware

of this lacuna, Neiland and Wilcock (1998) write that it isunclear if the commonly encountered low reproductive successof tropical orchids is the result of (1) regional or growth habitphenomena, (2) dissimilarities of temperate versus tropicalorchid population structures (as reported by Ackerman 1986),(3) a lack of suitable pollinators in the tropics or (4) increasedcompetition for fewer pollinators. Clearly, it is possible that anycombination of the aforementioned factors may be involved andthat different factors may be involved depending upon thespecies and its biological context. As Neiland and Wilcock(1998) recognize, what is needed are more studies on the repro-ductive biology of tropical terrestrial orchids.

The present study investigates the pollination biology, breed-ing system, nectar production and floral scent composition of

Eulophia alta, a perennial, tropical, terrestrial orchid. Dressler(1981) understood that only medium-sized to large bees couldpollinate the large gullet flowers typical of Eulophia and heconsidered Xylocopa bees to be the usual pollinators of thisgenus. Accordingto theMEPP, floral evolution at thepopulationlevel is driven by the most effective pollinator (Stebbins, 1970).However, as pointed out by Waseret al. (1996), one of theweak-nesses of reproductive biology studies claiming specificity ofplant pollinator interactions is that they are often very restrictedtemporally and spatially. It is common practice for studies tofocus on pollinator interactions of a single population perplant species. However, spatial variation in pollinator compo-sition and abundance is now a well-recognized phenomenon(e.g. Campbell, 1987; Ackerman et al., 1997; Parra-Tabla

et al., 2000), and hence it is important to study a species atmore than one site whenever possible. With this in mind, thepol-lination biology and reproductive success of three E. alta popu-lations in Amazonia were studied: (1) on well-drained ground ina campina sombreada, (2) a steep slope (likewise campinasombreada) and (3) a dry, sandy locality (campina aberta).To characterize the importance of each flower visitor the relativeabundance of the observed flower visitors at each site was eval-uated, and additionally, pollinia transfer efficiency of eachvisitor species was measured by counting the number offlower visits leading to deposition of pollinia on stigmas and/or removal of pollinaria from flowers.

M A TER I A LS A N D M ETH O D S

Plant material

The pantropical genus Eulophia R.Br. ex Lindl. as presentlyconceived belongs to sub-tribe Cyrtopodiinae (tribeCymbidieae) and comprises about 230 species that grow in tro-pical and southern Africa, south-west Arabia, Madagascar, theMascarenes, tropical and subtropical Asia, Southeast Asia,Australasia and tropical America (Thomas, 1998). Thomasfurther points out that no satisfactory treatment of Eulophiaexists; many of the species are widespread and highly variable,and there are even saprophytic forms (Dressler, 1981). A given

Eulophia species can grow in widely different habitats fromswamps, to forests, sandy beaches, mountain grasslands andeven semi-desert regions. Eulophia alta (L.) Fawc. & Rendleis the sole representative of Eulophia in the New World, prob-ably originating in the Old World, where it grows in tropicalsub-Saharan Africa. In the New World it grows from Georgiaand Florida (USA) south through the Caribbean and fromMexico south to Paraguay and Argentina. The plant producesan erect, loose raceme bearing about 80120 medium-sized tolarge distally purple proximally green flowers (Fig. 1). Thestature (erect leaves) of the study plants varied from 0.5 to1.5 m tall and inflorescences were similar in size. Vouchers ofE. alta were deposited in the herbarium of the UniversidadeFederal do Amazonas, Manaus, Brazil, and ULM.

Study sites

The pollination biology of E. alta was studied from April toJuly 1999 in a region 17 km north of Manaus, Brazil. For the

A

B C

F IG . 1 . ( A) Eulophia alta individuals flowering (in front) and fruiting(behind) at habitat sitio. (B,C) Flowers after being fixed in alcohol withdead posed flower visitors, Xylocopa muscaria (B) and Centris spilopoda

(C), to show the size of the insects in relation to flower size.

Jurgens et al. Pollination biology of Eulophia alta (Orchidaceae)898


3/16

comparative study, three different sites on terra firme (non-inundated land) were chosen. E. alta was the only orchidspecies growing at each site. Distances between populationsranged from 3.5 to 5 km. The plant typically grows on well-drained soil principally in campina sombreada (shadedcampina) or in campina aberta (open campina; Braga and

Braga, 1975). Campina is a special, relatively low, woodyAmazonian vegetation; it occurs on white sand patchesnested within Amazonian forest. The edaphic and floristiccomposition of campina is distinct from Amazonian forestand other vegetation types. The first study site was acampina sombreada called sitio, which had been burned1015 years ago. The site covered about 2500 m2 and the veg-etation was dominated by E. alta and shrubby Piper spp.Weekly phenological estimates during 3 months of investi-gation showed that 4313+315 flowers were open daily. Thesecond campina sombreada site was a steep (308)NNW-orientated slope, hereforth called slope. This sitecovered about 2650 m2 and had experienced earth slides andother forms of erosion associated with the laying of a pipe

several years prior to this study. Vegetation composition herewas similar to sitio but less dense, and the area was partlyshaded by Cecropia sp. trees (a plant indicative of disturbedsites). Daily flower number of E. alta was about half of thatat sitio (2263+172 open flowers per day). The third site,here named sand, covered about 2600 m2 and was situatedat a dry, sandy locality (campina aberta), which up toabout 10 years prior to the study had been a sand mine. Atthis site, vegetation cover was only herbaceous and plant diver-sity was low. Although E. alta plants were frequent here, therewere on average only 1596+155 flowers in bloom during thestudy period and therefore flower density was lower than at theother two sites.

Floral morphology and anatomy

Morphometric measurements were made on at least 20 ran-domly collected flowers. Width of the labellum chamberentrance, length and width of the dorsal sepal, lateral sepal,lateral petals and labellum were recorded (see Fig. 2).

Presence and site of osmophores and nectaries were detectedby staining whole flowers with neutral red (according toVogel, 1962). The flower parts were then preserved in 70 %ethanol for further analysis. Preserved material was furtherdehydrated, embedded in paraffin and sections cut using amicrotome. Thin sections were stained with Toluidine O (fol-

lowing Gerlach 1984) and viewed with a light microscope. Forscanning electron microscopy (SEM) study of surface mor-phology, flower parts stored in 70 % ethanol were placed inFDA (formaledhyde dimethyl acetate, 95 : 5) for at least24 h, dehydrated further, critical-point dried, sputter-coatedwith gold and then viewed at 5 kV using a Zeiss DSM 942scanning electron microscope.

Measures of flower longevity and stigma receptivity

Flower longevity is known to be influenced by pollination(e.g. Arditti, 1976; Gori, 1983; Primack, 1985; Proctor andHarder, 1995). In subtribe Cyrtopodiinae, two hard polliniawith a stipe and viscidium form a single pollinarium and the

anther sits like a cap on the column (Dressler, 1981).Normally, the anther caps of E. alta fall off during pollinariaremoval, or in unpollinated flowers they eventually dry andfall off. To evaluate the influence of pollination events onflower longevity in E. alta, the following six treatments wereapplied to each of a set of 15 mature flowers: removal ofanther cap, pollinarium intact (A 2 P ); anther capintact, removal of pollinaria (A P 2 ); anther cap and pol-linaria removed (A 2 P 2 ); anther cap and pollinariaintact (A P ); pollinated (outcrossed), own pollinariaremoved (Poll P 2 ); pollinated (outcrossed), own polli-naria intact (Poll P ). Anther caps and pollinaria wereremoved on the first day of anthesis, and pollination wascarried out on the third day after flower opening. The time

until the resupinate flowers either (1) wilted (indicating unfer-tilized flowers) or (2) became upright as fruits developed (suc-cessful pollination) was recorded. To determine the period ofstigma receptivity, 368 flowers were marked at the beginningof anthesis, and in the morning and at midday on each dayafter the onset of anthesis (here defined as flower opening) atleast eight flowers were manually pollinated with conspecificpollinia from neighbouring plants. Fruit set was recorded.

Evaluation of breeding system

In order to investigate the breeding system of E. alta, fruitset of variously manipulated flowers was counted and com-pared with reproductive success under natural conditions.

When pollinator exclusion was necessary, bags were madefrom nylon stockings pulled over wire cages which were bigenough to prevent contact between flowers and nylon mesh.The cages were placed at appropriate height over the inflores-cences and tied to the inflorescence stems and stabilizingbamboo sticks that were fixed to the ground. All covers wereremoved when flowering was finished and inflorescenceswere tagged with coloured threads until fruit harvest.

To test whether apomixis occurs, 154 flowers (one inflores-cence at each site: 79 flowers at sitio, 37 flowers at slopeand 38 flowers at sand) were emasculated and then bagged.Self-compatibility in E. alta was evaluatedby hand-pollinating

sep

sep

lab

lpet lpet

1cm

F IG . 2. Side-view of an Eulophia alta flower (left) and frontal view of thelabellum structure (right). Abbreviations: sep sepals, lpet lateral petals,

lab tip of labellum.

Jurgens et al. Pollination biology of Eulophia alta (Orchidaceae) 899


4/16

using forceps. At sitio, to test for autogamy, five flowers werehand-pollinated with their own pollinia on each of five baggedinflorescences and, to test for geitonogamy, on another fiveinflorescences five flowers were pollinated with pollinia fromother flowers of the same inflorescences. To test for xenogamy,a similar experimental protocol was carried out with pollinia

from inflorescences of plants at least 5 m distant. The capacityof E. alta for spontaneous auto-pollination was evaluated ateach locality by bagging 12 inflorescences in bud with finemesh nylon.

To determine the degree of pollinator-mediated outcrossing,all open flowers on 12 inflorescences at each site were emascu-lated. These inflorescences were left unbagged until fruitsmatured and then total fruit set was recorded. Additionally,the occurrence of pollinator-mediated geitonogamy was inves-tigated. At each site, 20 flowers on each of eight inflores-cences, which were distinct both spatially and temporally,were emasculated; then, after removal of anther caps, the pol-linaria of all other flowers on these inflorescences were stainedwith a 1 : 10 000 solution of neutral red. Pollen transfer of

stained pollinia to emasculated flowers was monitored andrecorded daily.To determine the reproductive success of E. alta under

natural conditions, fruit set of in total 87 unmanipulated inflor-escences at sitio, slope and sand was counted.Furthermore, in order to measure the degree of possiblepollen limitation, induced cross-pollinations were performedfor comparison with natural fruit set. Twelve inflorescencesbagged in bud stage at sitio were allowed to come intoflower then hand-pollinated, using forceps, with polliniafrom plants at least 5 m distant.

Besides fruit set, other reproductive modalities that mightdiffer due to different pollination modes, such as outcrossingversus auto-pollination, are seed set and seed viability.

However, as there are millions of small seeds per fruit inorchids, accurate counts of seed numbers are difficult to obtain(Proctor and Harder, 1994). Based on data from Proctor andHarder (1994), Neiland and Wilcock (1995), and Nazarov andGerlach (1997), Johnson and Edwards (2000) found that inorchids, ovule number correlates with the number of pollengrains typically deposited onto a stigma after an effective polli-nator visit; therefore, the expected pollenovule ratio should beabout one if entire solid pollinia are deposited on stigmas, asborne out by the study of Nazarov and Gerlachs (1997) forCoryanthes senghasiana. Given this, we assumed that inE. alta, each pollination event resulting in fruit set is sufficientto produce full seed set because whole pollinia are transferred.For this reason, we decided to measure seed quality, here

treated as the number of seeds containing an embryo, of fruitsproduced by different hand-pollination modes. Orchid seedshave no endosperm; they consist mainly of the embryo and atranslucent testa. Therefore, embryos are readily visible whenseeds are viewed under a light microscope (Leitz Ortholux II).Ten flowers on each of ten bagged-in-bud inflorescences weresubjected to the following treatments: hand-pollinated with pol-linia from donor plants (at least 5 m distant), pollinated with pol-linia from other flowers of the same plant, and self-pollinatedwith their own pollinia. For each of the treatments, ten fruits(one per plant) were randomly chosen and from each fruit 500seeds were taken and checked for the presence of an embryo.

The percentage of seeds with an embryo (hereafter embryocontent) was calculated per fruit and averaged per treatment.

Flower position in an inflorescence can be a factor in fruitset and seed number as a result of resource allocation(Corbet, 1998; Medrano et al., 2000; Vallius, 2000). Tostudy the effect of flower position with respect to sexual repro-

duction, fruit set and embryo content of flowers from thelower, middle and upper third of ten inflorescences wereevaluated after hand-pollination (outcrossed, minimum 5 mdistant) of all flowers. Total fruit set was recorded and ten ran-domly chosen fruits of the lower, middle and upper third ofeach inflorescence were examined for embryo content.

Pollinator observations

In order to assess the composition and abundance of flowervisitors a total of 229 h were spent observing the plants at thethree sites combined. Observations were made at single inflor-escences or small groups of inflorescences during 2 4-hperiods at different times of the day, from early morning

(0630 h) until late afternoon (approx. 1800 h). Observedflower-visiting insects were counted and their behaviour wasnoted. Most of the insect species were identified on thewing. Specimens of each species were caught for identifi-cation. All flower visitors collected were sent to specialists atthe Universidade Federal do Parana, Curitiba, Brazil, foridentification, and are deposited in the entomological collec-tions of INPA, Manaus and in the insect collection of theHerbarium ULM at the University of Ulm, Germany.

Relative abundance (RA) is given as the percentage ofobserved flower visits of each insect species at a particularsite (100 % all observed flower visits of all insect species).Pollinia transfer efficiency (PE) of each pollinator species ata particular site is given as the percentage of observed suc-

cessful flower visits (visits with deposition of pollinia onstigmas and/or removal of pollinaria from flowers) dividedby all observed flower visits of that insect species. The result-ing total pollinia transfer effectiveness (TE) of pollinators at aparticular site was calculated as the number of successfulflower visits of a particular pollinator species over a periodof 100 h of observation.

At each site, reproductive success of male and female func-tion was evaluated on randomly chosen inflorescences. Inorder to consider only outcrossing pollinia transfers, pollinariaof a chosen inflorescence were stained with a neutral red sol-ution (1 : 10 000) after removal of the anther cap. FollowingSinger and Cocucci (1997), donor efficiency, i.e. the ratiobetween male and female function, was calculated by dividing

female reproductive success (i.e. percentage of flowers polli-nated with unstained pollinaria) by male reproductivesuccess (i.e. percentage of flowers having stained pollinariaremoved). At each site throughout the flowering season, atotal of eight inflorescences of differing flowering times wereexamined.

Nectar analysis

Flowers of E. alta have no spurs and no large nectar dropletwas visible inside or outside the flowers. However, most flowervisitors were observed to search for something at the labellum



5/16

tip, and traces of fluid were found here, but never inside thechamber. Nectar production and composition over time wereinvestigated by rinsing the labellum tip (the chamber couldnot be rinsed as this caused too much tissue damage) of 96flowers with 70 % ethanol and then collecting the liquid forlater analysis. Samples were taken from bagged 1 12-d-old

flowers. Nectar samples were analysed by high performanceliquid chromatography (HPLC) as described in Witt et al.(1999) using a Waters High Performance CarbohydrateColumn supplemented with a complementary Sentry Guardprecolumn, a 510 pump and a 717 autosampler. Beforeinjection all samples were dried in a vacuum centrifuge(Savant Speed Vac SC 100) and diluted with 200 mL water.Injection volume was 10 mL and samples were eluted withan acetonitrilewater mixture (71 : 28) at a flow rate of1.4 mL min21 and temperature of 35 8C. Glucose, fructoseand sucrose were detected using a Waters 410 refractionindex detector and quantified by means of WatersMillenium software.

Sampling of volatiles

Floral scent was collected following the method of Jurgenset al. (2000). To prevent flowers being visited and pollinatedby insects, five individuals were bagged (as described abovefor evaluation of the breeding system) when flowers werestill in bud. The volatile compounds of a single flower perinflorescence were collected using a headspace technique. Toaccumulate the floral scent and prevent wind drift, eachflower was covered with a glass jar (10 cm in diameter and10 cm long) and the scented air surrounding the flowerwithin the jar was pumped (200 ml min21 for 2 h, using abattery-operated membrane pump, G12/01 EB, GardnerDenver Thomas GmbH, Inc., Puchheim, Germany) into a

narrow glass tube packed with 150 mg adsorbent [TenaxTA

(2,6-diphenyl-p-phenylene oxide), mesh size 6080 andCarbotrapTM, mesh size 2040]; thereafter, the adsorbed vola-tiles were extracted with 1 mL of acetone into glass vials,which were then hermetically sealed. The adsorbent tubeswere preconditioned by washing with acetone and dried at250 8C.

Gas chromatography and mass spectrometry

GC-MS analyses were performed using an ion trap instru-ment wherein both ionization and mass analysis occur in thesame chamber. The GC-MS (Saturn 2000, Varian, WalnutCreek, CA, USA, and 8200 CX auto injector) was equipped

with a fused-silica capillary column (CP-Sil 8CB-MS,0.25 mm film thickness, 30 m long, 0.25 mm inner diameter,Varian). The 1-mL samples were introduced using a 1079injector; an injection split ratio of 20 was applied. The temp-erature of the column was programmed to rise from 60 to260 8C, at a rate of 8 8C min2

1; the temperature of the injector,

transfer line and ion trap was held at 200, 175 and 200 8C,respectively. Settings of the MS were: mass spectra 70 eV(in EI mode) and scan range was 40650 amu at a scan rateof 1 scan min21. The GC-MS data were processed using theSaturn Software package version 5.2.1. Component identifi-cation was processed using the NIST 05 mass spectral database

(NIST algorithm) and confirmed by comparison of retentiontimes with published data (Jennings and Shibamoto, 1980).Identification of some compounds was also confirmed by com-parison of mass spectra and retention times with those of auth-entic standards.

Bioassay to test the attractivity of floral organs

While preparing flower parts for preservation in alcohol (at adistance of at least 20 m from a Eulophia population), 13 bees(three Centris rubella, five C. varia and five unidentified indi-viduals) were observed to be attracted exclusively to the label-lum tissue, especially to parts of the labellum chamber (ten of13 bees). To confirm this chance observation, the labellum tip(Fig. 3A), labellum base (Fig. 3B), column, sepals and petalswere tested separately for their potential attractiveness toflower visitors in the field. Flower parts cut from ten flowerswere put on Petri dishes and the open dishes were placed100 cm above ground at a distance of 40 cm from each otherin a natural population of E. alta. Distance to the nearest

growing Eulophia was 2.6 m and the experiment was carriedout for 2 d between 1100 and 1400 h as prior observation ofthe daily activity of flower visitors had shown that allspecies visited flowers more or less around midday.

A

A

B

B

F IG . 3. Longitudinal section of a flower of Eulophia alta and surface of theupper side of the lip: (A) surface with stomata-like openings at the tip of the

lip, and (B) palisade-like cells in the labellum chamber of the flower.



6/16

Statistical analyses

Data sets were tested for normal distribution and homogen-eity of variances and further tests were applied accordingly.Data on flower longevity and embryo content of seeds pro-duced from different pollination experiments were submittedto ANOVA and least significant difference (l.s.d.) tests for

post-hoc comparison of means. A Kruskal Wallis test wasused to analyse variation of seed embryo content withrespect to position in inflorescences. Comparison of inflores-cence flower numbers among treatments and populations wasdone with ANOVA followed by Tukeys HSD test forunequal numbers. For fruit set, KruskalWallis ANOVA wasused to test for pair-wise differences among treatments andpopulations, followed by a Tukey Kramer test for non-parametric data as a post-hoc test. Male and female functionand the resulting donor efficiencies of inflorescences were ana-lysed with MANOVA and l.s.d. tests for a post-hoc comparisonbetween sites. All statistical analyses were performed withSTATISTICA 5.1.

R ES U LTS

Floral morphology

The erect racemose inflorescences of E. alta bear about 100flowers (average of sample sets ranged from 86 to 123 per

inflorescence, Table 1). The three sepals (1924 78 mm)and two petals (1517 5.57 mm) are nearly equal in sizeand shape. The free labellum (19.522 10.512 mm) is tri-lobate; the central lobe serves as a landing platform for flowervisitors. Together with the two lateral petals, the two laterallobes of the labellum embrace the column, forming a labellum

chamber, into which flower visitors may enter. Inside the label-lum chamber is a narrow pass approx. 8.59 mm wide andapprox. 6 mm high. However, the lip is to some degree flexibleand the weight of flower visitors may depress it, therebyincreasing the size of the labellum chamber entrance. Withthe exception of the labellum, perianth segments are pale pur-plish pink. Most parts of the labellum are also pale pink, butthe tip and base are dark purplish pink. The colouration ofthe column is likewise pale: thus, the labellum contrastssharply with the surrounding, less conspicuous, floral parts.Neutral red staining of flowers revealed no (additional)colour marks on sepals and lateral petals, suggesting thateither flowers of E. alta have no osmophores or osmophoresoccur in the naturally dark-coloured parts of the labellum

and are therefore not apparent by this technique. Anatomicalstudy of the labellum shows that this appendage has three epi-dermis cell types. This is in contrast to the sepals and lateralpetals, which have only one. Inside the labellum chamber atthe base of the lip, besides the commonly encountered flat epi-thelial cells, there are papillose, tube-like cells with thin cell

TA B L E 1. Fruit set of Eulophia alta under natural conditions and in tests for auto-pollination, pollinator-mediated

outcross-pollination and geitonogamy, and maximized fruit set by outcrossing hand-pollination

Sitio Slope SandKruskalWallis ANOVA: differences of

fruit set among sites

Natural fruit set (unmanipulated)

Inflorescences (n) 42 19 16Flowers (n) 3811 1943 1977Flowers per inflorescence 90.7+19.5 102.3+24.0 123.6+20.7* F2,77 49.93Fruit set (%) 10.9+3.0

a,A 19.8+3.6b,D 14.7+2.0

c,G P , 0.001Spontaneous auto-pollination (bagged and intact)

Inflorescences (n) 12 12 12Flowers (n) 1069 1222 1042Fruits (n) 44 59 63Flowers per inflorescence 89.1+24.8 101.8+26.6 86.8+24.6 F2,36 10.03Fruit set (%) 4.4+1.5

d,B 4.9+1.3d,e,E,F 6.1+1.2

e,H P , 0.01Vector-mediated outcrossing (free and emasculated)

Inflorescences (n) 12 12 12Flowers (n) 1037 1150 1136Flowers per inflorescence 86.4+18.9 95.8+25.5 94.7+19.3 F2,36 22.93Fruit set (%) 5.6+2.1

f,B11.6+3.3

g,F4.1+1.7

f,H,JP , 0.001

Vector-mediated geitonogamy (free and emasculated; stained pollinaria)Inflorescences (n) 8 8 8

Flowers (n: emasculated) 160 160 160Flowers (n: stained pollinaria) 632 663 729Fruit set (%) 1.3+2.3h,B 1.9+2.6h,E 0+0h,J n.s.

Maximized fruit set (outcrossing hand-pollination)Inflorescences (n) 12 2 2Flowers (n) 1233 2 2Flowers per inflorescence 102.8+18.1 2 2Fruit set (%) 93.3+8.9

C2 2

KruskalWallis ANOVA: differences of fruitset among treatments

F4,86 68.00;P , 0.001

F3,51 44.29;P , 0.001

F3,48 41.67;P , 0.001

Fruit set: different superscript letters (among treatments, AJ; among sites, ah) indicate significant differences according to the applied tests; n.s. notsignificant; 2 no data. Flower numbers: *Following Tukeys HSD for unequal number only flower number of inflorescences used for measuring naturalfruit at sand varied significantly from all others (among sites: ANOVA, F2,74 14.40, P , 0.001; among treatments: ANOVA, F2,37 11.50, P , 0.001).



7/16

walls (Fig. 3B), which may function as osmophores (compare,for example, Vogel, 1962; Davies and Winters, 1998; Teixeiraet al., 2004). Furthermore, the adaxial surface of the lip iscovered with cone-like, well-cutinized cells. Stomata-likeopenings were found near the vascular bundles at the labellumtip (Fig. 3A); these openings might indicate the presence of

intrafloral nectaries (but see Daumann, 1970).

Floral longevity and stigma receptivity

Unpollinated flowers having an intact pollinarium andanther cap began wilting on the 16th day of anthesis, andlasted at most for 21 d before finally abscising. Experimentalmanipulation of flowers showed that pollinaria removal andpollination had the most significant effect on flower longevity,followed by anther cap removal. Flowers pollinated on thethird day of anthesis, stopped nectar secretion and scent pro-duction within 24 h, and wilting of petals and initiation offruit development were visible by the fifth day, independentof pollinaria removal (Table 2). It took up to 69 d for fruits

to mature. Differences in flower longevity between allgroups were highly significant (ANOVA: F5,84 1398.77,P , 0.001; l.s.d. test: P , 0.001) except between pollinatedflowers with and without their own pollinaria (P 0.42).Stigma receptivity was similar to that of whole flower longev-ity of unpollinated flowers. Stigmas were receptive up to the17th day of anthesis and set 100 % fruits after pollination(284 flowers between 1 and 17 d old). Thereafter, stigmareceptivity decreased steadily to 12.5 % fruit set for flowerspollinated on the 21st day of anthesis (84 flowers between18 and 21 D old).

Breeding system

None of the emasculated bagged flowers set fruit, whereasall flowers that were hand-pollinated with pollinia of thesame flower or flowers of the same inflorescence set fruit, asdid flowers pollinated with foreign conspecific pollinia, indi-cating that our study plants are not apomictic but fully self-compatible. E. alta was found to be partially autogamous.Fruits developed in 4.4 % of the bagged flowers at sitio,4.9 % at slope and 6.1 % at sand, confirming that somefruit production takes place without a vector when anthercaps dry and fall off, and pollinaria bend or fall down onto

the stigma of the same flower; hence, auto-pollination orvectorless self-pollination (Catling, 1990) occurs. Further-more, variation in auto-pollination among sites was statisti-cally significant (Kruskal Wallis ANOVA: H2,36 10.03,P , 0.01).

All hand-pollinated flowers used in experimental manipula-tions (xenogamy, geitonogamy and autogamy) for seed qualitystudies set fruit. Nevertheless, seed quality, as defined by seedembryo content, differed significantly between treatments

(ANOVA: F2,27

132.31, P , 0.001). Differences betweenselfing (autogamy 85.4 %, geitonogamy 86.8 %) and out-crossing (98.2 %) were highly significant (l.s.d. test: P ,0.001). After pollination of all flowers with outcrossed polli-nia, seed quality decreased significantly (Kruskal Wallistest: H2,30 23.39, P , 0.001) in an acropetal direction(Table 3). There was no fruit abortion on the lower third ofthe inflorescence, and the further from the base the greaterthe degree of abortion, culminating in 24 % abortion in theupper third of the inflorescence. Seed size showed a similartrend with seeds in lower fruits being larger than seeds infruits at the top of the inflorescence (data not shown).Furthermore, in the self-compatibility experiment, when only25 flowers of an inflorescence were outcrossed by hand, they

all set fruit (100 % fruit set), whereas when all flowers in 12inflorescences were outcrossed by hand, fruit set was lower(93.3 %, Table 1).

Pollination biology

Eulophia alta is exclusively melittophilous. At the presentsites anthophorid bees are the main flower visitors and mosteffective pollinators of the many-flowered inflorescences.The highest flower visitor frequencies were recorded onapprox. 3 5-d-old flowers. In total, 19 species representingsix Hymenopteran and two Lepidopteran families wereobserved visiting the flowers (Table 4). The relative abundance

of flower visitors and pollinators, and pollinator pollinia trans-fer efficiencies at the different localities are given in Table 4.The bee species Centris inermis and Centris bicornuta couldnot be distinguished from each other by eye in the field, andfor this reason the data in Table 4 give the sum of the twospecies. Species richness of flower visitors at the three siteswas quite different. At sitio 19 species of flower visitorsand pollinators were observed, whereas there were ninespecies at slope and five at sand. Field observationsshowed that flowers are visited until the 14th day of anthesis.Most individuals of insect species recorded as nectar-seeking(Table 4), for example all anthophorid bees, were observed

TA B L E 2. Flower longevity after manipulation of the male and

the female function

Treatment n Longevity (h; mean+ s.d.)

A P 15 490+14a

A 2 P 15 394+13b

A P 2 15 243+11c

A 2 P 2 15 209+22d

POLL P 2 15 111+17e

POLL P 15 106+16e

A anther cap retained, A 2 anther cap removed, P ownpollinarium retained, P 2 own pollinarium removed, Poll pollinated.Different superscript letters indicate significant dissimilarities (ANOVA:F5,84 1398.77, P , 0.001; l.s.d. test: P , 0.001).

TA B L E 3. Influence of flower position within inflorescences on

fruit set and embryo content of seeds after pollination of whole

inflorescences

Position in inflorescenceFruit abortion

(%; mean+ s.d.)Embryo content of seed

(%; mean+ s.d.)

Lower third 0 83.4+9.7Middle 9.7+15.6 47.6+14.3Upper third 24.1+10.6 29.4+7.2



8/16

TA B L E 4. (a) Flower visitors and pollinators of E. alta for the three habitats sitio, slope and sand and their body length [

abundance [RA (%)], pollinia transfer efficiency [PE (%)] and the total number of flower visits with effective pollinia transfers pe

n number of insects observed. (b) Summary of observations.

(a)Sitio Slope

Flower visitorsBL

(mm) Nectar-seeking nA

(100 h21)RA(%)

PE(%)

TE(100 h21)

A(100 h21)

RA(%)

PE(%)

TE(100 h21)

HYMENOPTERAApidae

Centris bicornuta Mocsary & Centrisinermis Friese

14 &15

37 36.3 4.4 17.1 6.2 0 0 0 0

Centris flavifrons Fabricius 23 275 237.3 29.1 0 0 37.9 7.8 0 0 Centris minuta Fabricius 12 138 24.5 3.0 17.1 4.2 73.6 15.1 18.9 13.9 Centris rubella Lepeletier 20 186 87.3 10.7 10.5 9.2 111.5 22.8 12.5 13.9 Centris spilopoda Moure 13 86 26.5 3.2 6.3 1.7 67.8 13.9 15.0 10.2 Centris varia Erichson 13 210 167.6 20.6 11.9 20 35.6 7.3 4.8 1.7 Centris sp. 10 78 35.3 4.3 1.7 0.6 31.0 6.4 8.1 2.5 Ceratina sp. 16 3 2.9 0.4 0 0 0 0 0 0 Euglossa chalybeata Friese 15 2 2 2.0 0.2 0 0 0 0 0 0

Euglossa mandibularis Friese n.d.2

1 1.0 0

.1 0 0 0 0 0 0 Eulaema mocsaryi Friese n.d. 2 1 1.0 0.1 0 0 0 0 0 0

Xylocopa muscaria Fabricius 16 187 92.2 11.3 4.4 4.1 106.9 21.9 15.5 16.6 Halictidae

Augochlora esox Vacahl n.d. 2 6 5.9 0.7 0 0 0 0 0 0 Megachilidae

Megachile sp. 9 107 80.4 9.9 2.4 1.9 20.7 4.2 8.1 1.7 Sphecidae

Bembix sp. 13 2 10 4.9 0.6 0 0 3.4 0.7 0 0 Eumenidae

Eudynerus sp. 16 2 3 2.9 0.4 0 0 0 0 0 0 LEPIDOPTERANymphalidae sp. n.d. 6 5.9 0.7 0 0 0 0 0 0 Lycaenidae

Strymonidia sp. n.d. 2 2.0 0.2 0 0 0 0 0 0 (b)Observed visits/observation hours 832/102 h 425/87 h

Pollinator species/flower visitors 9/19 7/9 Flower visits per 100 h 815.7 488.5 Pollinator visits per 100 h (SA/100) 549.8 447.1 Effective pollinator visits per 100 h(STE/100)

47.8 60.5

Pollinator visits (%) 67.4 91.5 Effective pollinator visits (%) 6.8 10.3

atPolitechnikaWroclawskaonNovember16,2011 http://aob.oxfordjournals.org/ Downloadedfrom
http://aob.oxfordjournals.org/http://aob.oxfordjournals.org/http://aob.oxfordjournals.org/http://aob.oxfordjournals.org/http://aob.oxfordjournals.org/http://aob.oxfordjournals.org/http://aob.oxfordjournals.org/http://aob.oxfordjournals.org/http://aob.oxfordjournals.org/http://aob.oxfordjournals.org/http://aob.oxfordjournals.org/http://aob.oxfordjournals.org/http://aob.oxfordjournals.org/http://aob.oxfordjournals.org/http://aob.oxfordjournals.org/http://aob.oxfordjournals.org/http://aob.oxfordjournals.org/http://aob.oxfordjournals.org/http://aob.oxfordjournals.org/http://aob.oxfordjournals.org/http://aob.oxfordjournals.org/http://aob.oxfordjournals.org/http://aob.oxfordjournals.org/http://aob.oxfordjournals.org/http://aob.oxfordjournals.org/http://aob.oxfordjournals.org/http://aob.oxfordjournals.org/http://aob.oxfordjournals.org/http://aob.oxfordjournals.org/http://aob.oxfordjournals.org/http://aob.oxfordjournals.org/http://aob.oxfordjournals.org/


9/16

extending their proboscis and probing extensively at the label-lum tip. Regularly, anthophorid bees were observed secreting asubstance onto the labellum tip and thereafter imbibing a fluid.However, not all nectar-seeking insects entered the labellumchamber after probing the labellum tip. In contrast to theoften time-consuming probing at the labellum tip, when

insects ventured into the labellum chamber, they remainedthere for only a few seconds. As entering of the labellumchamber is required for effective pollen transfer, insectspecies that were never seen to enter the labellum chamberare treated as flower visitors, but not pollinators. Forexample, C. flavifrons was unable to enter the chamber dueto its relatively large body size and weight, and althoughmembers of the genera Euglossa, Eulaema, Eudynereus andCeratina probed at the labellum tip, they never entered thelabellum chamber. Only insects with an appropriate bodysize (body length between 10 and 20 mm) are able to fit intothe labellum chamber; upon entry they made contact withthe stigma and anther (Figs 1 and 2, Table 4). When appropri-ately sized flower visitors entered the labellum chamber, they

usually were effective pollinators (i.e. deposited pollinia and/or removed pollinaria). In the case of two small bees,Megachile sp. and Centris sp., some visits into the labellumchamber were ineffectual in pollinating the flower. Pollinariadeposition on insects was usually on the posterior portionsof the thorax, especially the scutellum and/or metanotum; pla-cement varied slightly among species depending on body sizeand shape. When pollinators backed out of the labellumchamber, pollinia carried by the bees were transferred fromtheir body to the stigma, and subsequently a pollinariumfrom the anther was attached to the insect. This sequenceensured that pollinator-mediated selfing (autogamy) by depo-sition of pollinia onto the stigma of the same flower during asingle visit did not occur. Moreover, pollinator-mediated geito-

nogamy was rare (see Table 1), yet features known to preventgeitonogamous pollination, such as protandry, bending ofpollen stalk or anther cap retention (see Catling and Catling,1991), were not observed in E. alta.

Relative abundance of flower visitors differed between sites(Table 4); on average, species with the highest relative abun-dance were five Centris species (C. minuta, C. flavifrons,C. rubella, C. varia and C. sp.), and Xylocopa muscaria, fol-lowed by C. spilopoda and Megachile sp. Summing the totalnumber of effective visits (TE/100 h) over all sites,C. minuta proved to be the most important pollinator followedby C. rubella, C. varia, X. muscaria and C. spilopoda. Thestrong territorial behaviour of male C. varia individuals(which also visited flowers effectively) at sitio was striking.

In the early morning, these males established their territoryin parts of the Eulophia population where there was highflower density; during the day, they attacked all flower visitorswithin their territory except for female C. varia bees.Conversely, flower-visiting C. varia females were disturbedby mating attacks by conspecific males. The territorial be-haviour of C. bicornuta and C. inermis was less ferociousand these bees often lost in competitions with C. varia for ter-ritory. Other Centris species such as C. minuta, C. spilopodaand C. sp. were observed to establish their territories at themargins of or outside the Eulophia population, on plants ofE. alta and an unidentified Piper species.

Orchid populations growing in a campina sombreada(sitio and slope) received more flower visits per hourthan those in campina aberta (sand; Table 4). Relative pol-linator abundance, however, was lowest at sitio, where thehighest proportion of non-pollinating insects was found.Number of effective pollinator visits was highest at slope

(60.5 effective flower visits per 100 h) and lowest at sand(33.9 effective flower visits per 100 h). Natural levels of

fruit set showed a significant variation among populations(Kruskal Wallis ANOVA: H2,77 49.93, P , 0.001) with10.9 % at sitio, 19.8 % at slope and 14.7 % at sand.Pollinator-mediated outcrossing reached 11.6 % at slope,and 5.6 and 4.1 %, respectively, at sitio and sand(Kruskal Wallis ANOVA: H2,36 22.93, P , 0.001). Incontrast, experimental outcrossing by hand-pollinations atsitio, the only site studied, produced 93.3 % fruit set(Table 1).

Differences in flower number in inflorescences used forevaluation of donor efficiencies in the three populationsproved to be insignificant (ANOVA: F2,21 1.21, P

0.32). Variation of male and female reproductive successand donor efficiency of tested inflorescences was highly sig-nificant between sites (MANOVA: Wilks l 0.29, RaosR6,38 5.50, P , 0.001). Differences in subsequent testswere found to be significant for male (F2,21 9.58, P ,0.01) and female reproductive success (F2,21 8.34, P ,0.01), but not for the resulting donor efficiency (F2,21 1.45, P 0.26). In the investigated inflorescences, most polli-naria were removed at slope, followed by sitio and sand(Table 5; 24.6, 21.0 and 12.8 %, respectively). Female repro-ductive success showed a similar ranking order, i.e. 5.6 % pol-linated flowers at slope, 3.0 % at sitio and 2.4 % at sand.On average, inflorescences having the highest donor efficien-cies were at slope (0.23); at sand and sitio donor efficien-

cies of inflorescences were lower (0.16). The overall donorefficiency of 0.23 at slope, followed by sand and sitiowith 0.19 and 0.14, respectively, parallels ranking of naturalfruit set. At all three sites, there were clearly more pollinariaremovals than effective pollinations.

TA B L E 5. Male and female reproductive success and donor

efficiency in three populations of Eulophia alta

Sitio Slope Sand

Inflorescences (n) 8 8 8

Flowers (n) 768 722 705Removals (n) 162 178 91Pollinations (n) 23 41 17Donor flowers (%) 21.1 24.7 12.9Pollinated flowers (%) 3.0 5.7 2.4General donor efficiency 0.14 0.23 0.19Mean+ s.d. of inflorescences:Flowers per inflorescence (n) 96.0+13.9 90.3+9.1 88.1+7.2Donor flowers (%) 20.0+7.6

a24.6+4.2

a12.8+4.1

b

Pollinated flowers (%) 3.0+1.1c 5.6+1.8d 2.3+2.0c

Donor efficiency 0.16+0.09e 0.23+0.07

e 0.16+0.11e

Different superscript letters indicate significant dissimilarities among sites(MANOVA: Wilks l 0.29, Raos R6,38 5.50, P , 0.001).



10/16

Nectar production and nectar composition

Concomitant with anthesis, E. alta produces small amountsof nectar on the labellum tip. The nectar seems to dry immedi-ately after secretion and is visible only under a microscopewhere it appears as a thin film over the labellum tip. Toimbibe this film of dried sugar, insects usually lick at the label-

lum tip, or as suggested by the present observations, they firstsecrete a substance onto the surface then remove the liquidwith their proboscis. In bagged flowers, nectar was found upto 11 d from the start of anthesis, with the highest sugarcontent on the third and fourth days of anthesis (Fig. 4).Composition analysis showed that the nectar is dominated byglucose and fructose, and can be classified as hexose-richaccording to Baker and Baker (1983).

Floral scent composition

Flowers emit what was to our nose a sweet/floral scent from

the base of the labellum. Scent emission begins when a floweropens and peaks on the ninth day of anthesis. Even on the 17thday, a weak scent was detectable. A summary of the floralscent chemistry showing the relative amounts of volatiles,and their distribution among the main chemical compoundclasses, i.e. fatty acid derivatives, benzenoids, phenyl propa-noids, isoprenoids and nitrogen-containing compounds, isgiven in Table 6. Floral fragrance analysis of E. alta yielded40 compounds, 33 of which were identified. The floral scentof E. alta is predominantly composed of monoterpenes(40.4 %) and benzenoids (38.7 %). The main volatile com-pounds were jasminaldehyde (16.5 %), followed bycis-b-ocimene (13.4 %) and benzyl acetate (9.8 %).

Attractivity of different floral organs

The bioassay experiment revealed that only floral tissuefrom the labellum chamber and labellum tip was found toattract flower visitors. During a total of 6 h of observation,18 insects were recorded landing on the tissue of the labellumchamber (four X. muscaria, seven C. minuta, six C. varia and

one C. sp.), and three were observed to land on the labellum tip(one X. muscaria, one C. minuta, one C. varia). No insectswere observed to land on tissue from the column, sepals orpetals.

D I S C U S S I O N

Flower visitors and pollinator performance

Throughout its range, Eulophia alta is a weedy, colonizingplant of disturbed environments. Despite being a colonizingspecies, it is functionally specialized and at the sites detailedhere is seemingly dependent on a particular functional pollina-

tor type. Although E. alta was studied at three localities thatvaried in degree of perturbation, they all fall into the generalhabitat variation found in its range. The flower in E. alta hasa form that restricts entry to insects of a distinct body sizeand form. Even so, 19 different insect species belonging toHymenoptera and Lepidoptera were observed visiting E. altaflowers. Of these, only nine bee species had a body size thatallowed them to fit between the column and labellum, andthus potentially to act as pollinators. Except for one megachi-lid species (of minor importance), in E. alta all effective pol-linators were anthophorid bees. With respect to the mostimportant pollinators, at each site slightly differing subsetsof 3 5 anthophorid bee species accounted for more than

Number of days after flower opening

Sugarcontent(gsample1)

0

5

10

15

20

25

30

35

1110987654321

Fructose

Glucose

Sucrose

F I G . 4. Average sugar content and sugar composition of nectar from bagged E. alta flowers sampled between the first and 11th day of flower anthesis.



11/16

90 % of the effective visits. If the combined data from all threesites are averaged, four Centris bee species (C. minuta,C. varia, C. rubella and C. spilopoda) and X. muscariaaccount for more than 80 % of the effective visits over 100 hof observation. A recent phylogenetic interpretation (Strakaand Bogusch, 2007) places Centris and Xylocopa beesin different subfamilies of the Apoidea (Apinae and

Xylocopinae, respectively; see also Michener, 2000),suggesting that these genera are not closely related; they do,however, form a guild united in having a functional typefitting the morphological and behavioural requirements suit-able to the pollination of E. alta. Eulophia is an Old Worldgenus that has its centre of diversity in Africa. The populations

of E. alta in the New World represent the only expansion ofthe genus into the Americas; it is probable that they havespread from Africa as Centris bees are restricted to the NewWorld (Michener, 2000) and together the pollinating speciesoverlap much of the New World range of E. alta. We considerit likely that some of the study Centris species pollinate E. altaoutside of the study area, and possibly other Centris specieselsewhere in the Americas. Unlike Centris, Xylocopa isfound on most continents and achieves maximum speciesdiversity in Africa (Leys et al., 2002) and may pollinateE. alta there. The fact that Centris and Xylocopa pollinateE. alta in the New World is probably related to their morpho-logical similarity to the originally pollinating bees in the OldWorld (likely species of Xylocopa, Anthophora and

Amegila). Moreover, it is thought that African and SouthAmerican subgenera of Xylocopa are not particularly close(Leys et al., 2002), strengthening the argument that therelationship of the pollinating bee genera is one of form andbehaviour rather than close phylogenetic affinity.

Considering effective visits (TE; Table 4), it would seemthat E. alta is rather more specialized with respect to pollina-tors than might be concluded by casual inspection of data oninsect visitors and number of pollinator species. Similarly,Lock and Profita (1975) found that a West African speciesof Eulophia, E. cristata, a common terrestrial orchid, isvisited by butterflies, flies, wasps and bees; however, onlyone of two pollinating Xylocopa species, X. olivaceae, was aregular visitor and appeared to be the usual pollinator.

Few studies on floral biology in Orchidaceae have focusedon more than one population per species (but see, forexample, Borba and Semir, 2001; Brys et al., 2008); we findthat not only local pollinator spectra, but also pollen donorefficiency and pollinia transfer efficiency of particular pollina-tor species (see Table 4; e.g. C. varia, C. spilopoda andX. muscaria) vary between study sites and result in differingreproductive success (Table 1). Due to their abundance andefficiency, the most important pollinators at sand wereC. minuta, C. varia and C. sp., whereas at slope C. minuta,C. rubella, C. spilopoda and X. muscaria played the mostimportant roles, and at sitio the C. bicornuta/inermiscomplex, C. varia, C. rubella, C. minuta, C. spilopoda andX. muscaria were most effective with respect to total pollina

transfer. Our results show that flowers at the sites slopeand sand are visited by a smaller spectrum of insects thanflowers at sitio. The sheltered environment and the morefavourable conditions at this site, which result in a highlydiverse and relatively dense vegetation cover, may be thereason for the generally high abundance and diversity ofinsects, and in turn high species richness of flower visitorsand most effective pollinators observed at this site. Manyinsect species show limited foraging ranges and are sensitiveto changes in vegetation structure, habitat type, size and con-nection to other areas (Janzen, 1971; Powell and Powell,1987; Johnson and Bond, 1992; Parra-Tabla et al., 2000).

TA B L E 6. Average relative amounts of floral volatiles emitted by

Eulophia alta (n 4)

Criteria* RR

Relativeamount

(%)

Fatty acid derivatives3-Hexen-2-one a 372 1.9n-Octanal b 495 trEthyl tetradecanoate a 1518 0.4Octadecyl acetate a 1534 1.73,7,11,15-Tetramethyl-2-hexadecen-1-ol a 1674 2.2BenzenoidsBenzaldehyde c 442 0.12-Phenylethyl alcohol b 663 0.6Benzyl acetate c 734 9.81-Phenylethyl acetate b 771 5.5Benzenepropanol b 835 tr2-Phenylethyl acetate b 866 0.1Benzyl propionate b 870 1.4Phenylacetaldehyde b 1041 0.6Eugenol b 1062 trIsoamyl benzoate b 1115 0.1

Cyclamen aldehyde a 1144 0.1Methylisoeugenol a 1184 2.1Jasminaldehyde a 1370 16.5Benzyl benzoate c 1506 1.2Nitrogen-bearing compoundsMethyl 2-aminobenzoate a 992 2.3Terpenoidsa-Pinene c 399 3.3b-Pinene b 462 trD-Limonene c 536 trcis-b-Ocimene b 637 13.4Monoterpene m/z: 136, 121, 107, 93, 81, 69 a 697 0.4b-Terpineol (p-Menth-8-en-1-ol) b 717 0.6Monoterpene m/z: 149, 135, 121, 108, 93, 79 a 744 tra-Terpineol (p-Menth-1-en-8-ol) a 785 7.7Monoterpene m/z: 136, 121, 107, 93, 79, 69 a 791 0.4Monoterpene m/z: 136, 121, 105, 93, 77, 69 a 853 8.6

cis-Geraniol b 857 1.0Isopulegol a 909 0.2Monoterpene m/z: 136, 119, 105, 93, 85, 77 a 1028 0.1a-Cedrene a 1164 trEthyl linalool a 1217 4.7Miscellaneousg-Nonalactone a 1013 3.2g-Undecalactone a 1279 1.4a-Ethoxynaphthalene a 1230 1.5Unidentifiedm/z: 218, 203, 175, 129, 115, 91 a 1415 2.0m/z: 258, 243, 228, 213, 187, 91 a 1614 2.5Total 97.7

Compounds are listed according to relative retention time order (RR) ineach compound class. trtrace amounts (,0.1 %). Unknowns were includedwhen present with more than 1 % in any sample.

* Compound identification criteria: a comparison of MS with publisheddata; b comparison of MS and retention time with published data; c identity confirmed by comparison of MS and retention time of authenticatedstandard.



12/16

Disturbance may result in the loss of alternative food sources,nesting habitats and refuges against predators for pollinators(Aizen and Feinsinger, 1994). Differences in pollinator avail-ability between populations may lead to variations in reproduc-tive success (Campbell, 1987). It is also possible that differingflower numbers and densities at the three sites (sitio . slope .

sand; see site descriptions) might affect flower visitor frequen-cies per flower and resulting reproductive success. Bullocket al. (1989) showed that due to differences in plant andflower number, abundance and diversity of pollinators differsignificantly between sites separated by no more than 300 m.

If the number of flower visits by pollinator species translateddirectly (or nearly so) into greater fruit set, then we wouldexpect the highest natural fruit set to occur at sitio, where549.8 of the observed flower visits ( per 100 h) were madeby pollinator species; at slope 447.1 visits were from poten-tial pollinators and 197.5 visits at sand. In fact, highestnatural fruit set was found at slope, and lowest in sitio. Itwould appear that relative pollinator abundance, which waslowest at sitio (67.4 %, and where there was the highest pro-

portion of non-pollinating insects) and highest at sand andslope (97.5 and 91.5 %, respectively), also plays a role.With regard to absolute effectiveness, the locality having thehighest number of pollinia transfers was slope (60.5 per100 h). This result accords well with donor efficiency (DE),which is of particular importance in explaining differencesbetween sites as natural fruit set parallels DE ranking ofsites. If we relate for pollinator species the number of effectivevisits per 100 h to the number of observed flower visits per100 h it becomes apparent that pollinators at sand weremost efficient as 15.2 % of their visits (33.9 of 197.5 pollinatorvisits) led to successful pollination. Pollinator species at sitiowere least successful with 47.8 of 549.8 visits (6.8 %) beingeffective for pollinia transfer ( for comparison: slope 60.5

of 447.1 visits

10.3 %). The present data suggest that differ-ences in reproductive success in the investigated E. altapopulations cannot be explained by the absolute abundanceof pollinators alone. From the field observations it was clearthat disturbance of potentially effective pollinators duringflower visits was one reason for differing pollinia transfer effi-ciencies. This was particular the case at sitio, where afterpotential pollinators had landed and fed for a short time onlabellum tip nectar, they often fled from flowers before enter-ing the labellum chamber because of harassment by maleC. varia bees. It is known that male Centris bees defend terri-tories around food sources using pheromones and show aggres-sive behaviour (Frankie et al., 1980; Vinson et al., 1995). Theobserved territorial behaviour of male C. varia individuals at

sitio can explain the relatively low number of effectivevisits (TE per 100 h) of most other pollinating bee species atthis site as compared with the other sites. At the other twosites, C. varia showed no clear territorial behaviour, possiblydue to less favourable conditions, such as less vegetationcover, fewer total flowers and lower flower density, which ren-dered establishment and defence of territories untenable.

Overall, the present results are compatible with the recog-nized trend among tropical orchids to produce little fruit; com-monly less than 50 % of flowers produce fruit (Neiland andWilcock, 1998). We found that after pollination, there was arapid cessation of nectar and flower scent production followed

by wilting; this phenomenon, besides being energeticallyeconomical, may prevent a waste of limited pollinator servicesto already pollinated flowers. Casper and La Pine (1984)propose that when pollen is presented in pollinia, an individualflower benefits little in either receiving or donating pollen by avisit from a second pollinator for the simple reason that the

first pollinator to visit the flower is likely to remove allpollen and deposit sufficient pollen for full seed set.Initially, however, when a potentially effective pollinatorvisits a flower, there is no deposition, only attachment; butas an orchid pollinator usually visits many conspecificflowers, this initial effect is swamped out. The pollination-induced wilting observed in E. alta tends to support thehypothesis of Casper and La Pine (1984) and correlates withfindings in many other orchids (e.g. Luyt and Johnson, 2001,and references therein). The low donor efficiency ratios atthe three study sites indicate that there were distinctly morepollinarium removals than pollinations. Relatively low pollina-tor transfer efficiency and an extremely low donor efficiencyratio help to explain the overall low reproductive success of

E. alta. Pollinaria losses from pollinators can occur in anumber of ways, the most obvious being when bees removethem by grooming (see also Janzen, 1980).

Breeding system and reproductive success

Eulophia alta is highly self-compatible but not apomictic,and auto-pollination as defined by Catling (1990) occurs.This finding confirms the personal communications ofSalazar and Ackerman given in Catling and Catling (1991);auto-pollination in E. alta was described to be by bending ofthe caudicle, stalklett or pollen mass (see Salazar pers.comm. in Catling, 1990). In the present study, the nyloncovers may have reduced the incidence of self-pollination

because they shade the flowers and may retard anther capdrying and abscission. This would explain why combinedexperimental fruit set by autogamous, geitonogamous andxenogamous pollination, especially at sand, were markedlylower than natural fruit set (Table 1). At sand, shadingtrees and shrubs were absent and under natural conditionsthe proportion of auto-pollination at this site might well behigher than that recorded by the simulations; additionally,auto-pollination might contribute to the relatively highnatural fruit set at this site. The auto-pollination modeobserved in E. alta differs from that described byWilliamson (1984) for eight African Eulophia species. In themostly small, inconspicuous-flowered species which hestudied, Williamson found that auto-pollination occurred by

stigma contact with outgrowths of the pollinia.With respect to reproductive success, the present data

strongly suggest that pollinator-mediated outcrossing (11.6 %at slope, and 5.6 and 4.1 % at the other sites) cannot solelyaccount for observed natural fruit set, which was markedlyhigher (19.8 % at slope, 10.9 % at sitio, 14.7 % at sand).Yet, although capable of auto-pollination, E. alta did notseem to be especially adapted to selfing. Fruit set was rela-tively low compared with typically pollinator-independentorchids (see Neiland and Wilcock, 1998, and referencestherein) and the resultant seeds were significantly less viablethan outcrossed seeds (similar results were reported for



13/16

E. cristata by Lock and Profita, 1975). Moreover, fruit set inE. alta following hand-pollination (93.3 %) was much higherthan fruit set under natural conditions (maximum of 19.8 %).

The significance of nectar secretion in E. alta

Although E. alta produces minute amounts of nectar, thequality and quantity is sufficient to be attractive to flower-visiting insects, and to promote pollination. The fact thatthere are many flowers per population may compensate some-what for the poor nectar reward of individual flowers.Moreover, individual plants as well as entire populations area reliable nectar source for months. During the 3 months ofobservation, the E. alta populations studied were in flower con-stantly. Individual inflorescences had about 100 flowers, thelast floral buds opening on average 28 d after the first. Dueto extreme flower longevity single inflorescences presentedopen receptive flowers for about 47 d. Thus, the floweringphenology of E. alta can be classified as steady state(Gentry, 1974), wherein a plant produces a few flowers per

day over an extended period of time (usually a month or more).It has been argued that absence or low levels of floral

rewards in many orchid species might be a means of reducinggeitonogamy (see Johnson and Nilsson, 1999, and referencestherein). The minute nectar reward of individual E. altaflowers in combination with age-based differences in qualityand quantity, therefore, might be a way of avoiding excessivepollinator-mediated geitonogamy by encouraging movementto other inflorescences for optimal nectar rewards while atthe same time maintaining low-level attractiveness of unpolli-nated flowers.

Neiland and Wilcock (1998) reviewed the literature andshowed that average fruit set of rewarding orchids is higherthan that of deceptive ones. Ackerman et al. (1994) have

shown that even a small nectar reward in Comparettia falcataenhances pollinator attraction and fruit set. Neiland andWilcock (1998) suggest that the evolution of nectar productionwithin Orchidaceae has been the most frequent means of escap-ing reproductive limitation of low pollinator frequencies. InEulophia cristata, Lock and Profita (1975) reported flowervisits by nectar-seeking bees and butterflies although theflowers are scentless to the human nose and offer no freenectar in the spur. They assumed that these visits are stimulatedby the promising appearance (colour, shape) of the flowers;notwithstanding this, overall only relatively few flowers arevisited by pollinating bees, because bees quickly learn thatnothing is to be gained from such visits. In E. alta a nectarreward is offered, but the low fruit set of E. alta indicates

that this still does not translate into high fruit set, probablybecause nectar is exposed on the labellum so that imbibingdoes not equate with pollination. In this connection it shouldbe mentioned that Kullenberg (1961, p. 281) discovered thatthe terrestrial African species, E. horsfallii, which exposes nosuperficial nectar, was visited by food-linked Xylocopa beesthat tear holes with their mandibles into the labellum baseand lick. Such observations are highly suggestive and may, inpart, explain the selective forces that have led to the evolutionof superficial nectar in E. alta. It is still unclear why flower visi-tors ofE. alta enter the labellum chamber at all if nectar is onlyoffered on the labellum tip. It is possible that the similarity of

colours (both the tip and the base of the labellum are deep pur-plish pink) and odour emission at the labellum base repeatedlydeceive them into entering the labellum chamber in search ofmore nectar. It seems that pollinators recognize and areattracted to flowers signalling higher nectar content. Therewas a tendency among observed nectar-seeking insect species

to visit flowers most frequently on the third day of anthesis(our unpubl. res), the day when nectar sugar amount inbagged flowers is at a maximum. After the third day theamount of sugar in bagged flowers decreased, indicatingresorbtion of nectar. However, we do not know how nectar con-sumption by visitors affects nectar secretion in E. alta. As inother species, it may well be that visited flowers secretenectar for much longer than bagged flowers. Decreasingnectar sugar content and amount in older flowers may directpollinators to younger flowers, which are less likely to havedonated pollen or received pollinia, or to different inflores-cences. Further studies are needed to clarify how flower visitorsperceive floral signals and if they recognize correspondingreward quality from a distance.

The role of floral scent for pollinator attraction

In many orchids, floral fragrances are a major attractant forpollinators (van der Pijl and Dodson, 1966). Besides being areward in and of itself (for perfume-collecting male euglossinebees; see, for example, Williams and Whitten, 1983; Gerlachand Schill, 1991; Gerlach, 1995), floral fragrances may influ-ence pollinators in several ways: they may attract from a dis-tance, they may trigger search behaviour, and they may actas cues for alighting or probing, or as nectar guides (Faegriand van der Pijl, 1979). Several of these functions may beinvolved in the case of E. alta. As summarized by Dobson(1994) for many bees, visual cues operate primarily at long

range (within several metres; but see Chittka and Raine,2006) whereas flower volatiles are generally considered effec-tive in orientating bees at short range within 1 m or less.However, for bees such as male Euglossines, which cantrace odours over considerable distances (up to 1 km,Dressler, 1982), it is well known that fragrances may bemore important for pollinator attraction over long distancesthan visual cues (Bergstrom, 1978; Williams, 1983). Themany-flowered inflorescences of E. alta are certainly visuallyattractive but the attraction of anthophorid bees to seeminglyinconspicuous cut flower parts at a distance of 20 m fromthe E. alta population strongly suggests that floral scentplays an important role in attraction over long distances. Thefloral scent ofE. alta consists largely of aromatic alcohols, aro-

matic esters, phenyl propanoids and monoterpenes. Many ofthe floral scent compounds found in this species have pre-viously been reported as floral constituents in other orchids(see Knudsen et al., 2006) and many are known fromeuglossine-pollinated species (Williams and Whitten, 1983).Some of the floral scent compounds (alpha-pinene, benzylacetate, benzyl benzoate, eugenol, 2-phenylethyl alcohol)have been shown to elicit an electroantennography responsein experiments with male Euglossine bees (Eltz and Lunau,2005). Most of these compounds are even known to elicit be-havioural responses in Euglossine males (see, for example,Williams and Whitten, 1983). However, in contrast to the



14/16

information that is available for Euglossine bees, odour data orobservations on Centris bee-associated flowers is scarce. Thesweet floral fragrance of Passiflora alata, a species pollinatedby different Centris and Xylocopa species, is dominated byaromatic alcohols (e.g. benzyl alcohol, 4-methoxy phenol, 2-phenylethyl alcohol) and monoterpenes (e.g. nerol, geraniol)

(Varassin et al., 2001). Varassin et al. (2001) observed thatbees visited the flowers 1 h after anthesis, after the beginningof scent emission. This supports the idea that scent alonemight be an important attractant for Centris bees. Moreresearch is needed to understand the role of floral fragrancesin Anthophoridae bee attraction.

Blossom attractiveness, flower longevity and pollen limitation

Floral traits that increase attractiveness to pollinators arepredicted to evolve through selection on male rather thanfemale function (see Vaughton and Ramsey, 1998, and refer-ences therein). Several authors have found a relationshipbetween visitation rate or male fitness curve (MFC) and

flower number or floral display (e.g. Rademaker and DeJong,1998; Ohashi and Yahara, 1998; Vaughton and Ramsey,1998), suggesting that visitation is often proportional to thenumber of flowers per plant. In general, visitation rate orMFC seems to be a decelerating function of floral display. Inthe special case of orchids, Maad (2000) found forPlatanthera bifolia that both male and female fitness werehighest in plants with many flowers and concluded that alarge inflorescence attracts more pollinators. This correlateswith findings of other authors who have found that increasingflower numbers per plant contribute to pollinator attraction,thereby increasing the probability of reproduction (Nilsson,1992, and references therein; Rodriguez-Robles et al., 1992).As E. alta flowers remain receptive for about 16 d and floral

scent of unpollinated flowers increases until the ninth day ofanthesis, they contribute to display and long-distance attractionas well as retain the possibility of becoming fertilized. Thus,increased attractiveness by surplus flowers should increaseboth male and female fitness. This effect together with pollina-tor limitation could be large enough to explain the differencesbetween 93 % fruit set in hand-pollinated inflorescences and amaximum of 19.8 % natural fruit set in E. alta.

Pollen limitation (PL) is a concept derived from sexual-selection theory as applied to plant reproductive ecology andevolution, and refers to the observed phenomenon thatinadequate pollen quantity or quality can translate intoreduced seed quantity or quality, i.e. plant reproductivesuccess (Ashman et al., 2004). The most common interpretation

of PL is that of Burd and his co-workers (Burd, 1994; Ashmanet al., 2004), who deduce that pollination environment variesrandomly, and therefore that stochastic models best describeobserved phenomena. These authors note (Ashman et al.,2004) that stochastic variation among flowers in pollenreceipt is a bet-hedging strategy that commonly leads to lowseed set (Stephenson, 1981) or PL (Burd, 1995). There hasbeen an over-simplification of this concept, in that experimentalsupplemental pollinations are routinely used to test for the pres-ence of PL and whenever the results of theseexceed natural seedset the system is branded pollen-limited or pollinator-limited,and the researchers then attempt to explain the discrepancy. The

theoreticians instrumental in developing and popularizing thisconcept understand that . . . the standard empirical approachfor detecting PL remains problematic for several statistical . . .and biological reasons (Ashman et al., 2004); this last point isoften forgotten. We suggest that in our system there is noreason to suppose that the ideal of 93 % (as realized by sup-

plemental hand-pollination) represents an optimum, often ifever achieved in nature. It is hard to imagine, even under thebest of conditions, that pollinators in natural richness and abun-dance will visit each receptive orchid flowernor that therewill bea nearly one-to-one correspondence between pollinator removalof pollinaria and deposition on receptive flowers. Relativefruit or seed setper se does not determine the survival of a popu-lation or species but rather it is whether the seeds produced aresufficient in quality and quantity to sustain the populationthrough generations. It is possibly less a question of pollinatorlimitation leading to high PL, than the intrinsic fuzziness ofthis orchid system. The widespread distribution of this coloniz-ing terrestrial orchid points to the success ofE. alta despite thefindings here that individuals typically set less than 20 % fruit.

It is misleading to take an unnatural optimum (outcrossed, hand-pollinated flowers) as the standard for a natural system thatdepends upon a subtle interplay of many interrelated factors,and which has a certain amount of built-in play (bet-hedgingof some authors) to accommodate variable (stochastic) con-ditions. Moreover, the performance of a pollinator assemblageat a given locality is more than the sum of the performances ofa single pollinator species. Species-specific pollinator perform-ance is dynamic and influenced at each locality (or in each com-munity) by the dynamics peculiar to that pollinator community.In a recent paper by Burd et al. (2009), they extend their earliertheoretical framework to ovule number and consider pollinationsuccess at the floral level in many plants to be stochastic, theinevitable consequence of the hazards of pollination service

by biotic and abiotic vectors. In turn, this acts as a source ofselection, and they find that there is a positive correlationbetween stochastic variation in ovule fertilization opportunityand ovule number (Burd et al., 2009). According to them, thisexplains the commonly observed phenomenon that flowersreceiving experimental supplemental pollination frequentlyhave elevated seed production. Their analyses lead them to con-clude that . . . on average, flowers contained ovules that wouldnothavebeen fertilized in theabsence of hand pollination (Burdet al., 2009); and incidently, thereby begging the question of theutility of the supplemental pollination standard. Burd et al.(2009) predict that those plants experiencing greater uncertaintyof pollination attributable to whatever factor (e.g. reliance onspecialist pollinators, clumped pollen deposition) will have

many ovules and they cite orchids as a possible example ofthis. This explanation is harmonious with our results and weconcur with Burd et al.s (2009) assessment that overproduc-tion of ovules makes evolutionary sense in the light of stochasticdisparity in mating success. Such a view opens the door to theunderstanding that not every ovule produced is expected to befertilized nor every flower to be pollinated.

C O N C LU S I O N S

We have demonstrated that performance of a pollinator assem-blage at a given locality is more than the sum of the standard



15/16

performances of single pollinator species. Species-specificpollinator performance is not a fixed mathematical expression,but is influenced at each locality (or in each community) by theinteractions peculiar to that pollinator community, which hasits own dynamics. This study has highlighted some of thefactors that need to be considered when studying natural

orchid populations and underlines the need for multiplestudy sites to gain a deeper appreciation of orchid reproductivedynamics. In the case of E. alta, the typical orchid bauplan ofits flowers and their size restricts entry to insects of a distinctbody size, shape and behaviour, and only those visitors whichfulfil these requirements can act as successful pollinators.Eulophia alta flower form varies little throughout its widerange, and hence pollinator type probably remains much thesame although pollinator species may and does vary. Thisunderlines the basic resilience of functional pollinator typeswhereas species, in this case pollinating bees, are more fluidin space and time.

A C K N O W LED G EM EN TSWe thank Professor Pe. J. S. Moure, Curitiba, for the identifi-cation of insects, and two anonymous reviewers for helpful cri-ticisms and suggestions.

LI TER A TU R E C I TED

Ackerman JD. 1986. Mechanisms and evolution of food-deceptive pollinationsystems in orchids. Lindleyana 1: 108113.

Ackerman JD, Rodriguez-Robles JA, Melendez EJ. 1994. A meager nectar-offering by an epiphytic orchid is better than nothing. Biotropica 26:4449.

Ackerman JD, Melendez-Ackerman EJ, Salguero-Faria J. 1997. Variationin pollinator abundance and selection on fragrance phenotypes in an epi-phytic orchid. American Journal of Botany 84: 13831390.

Aigner PA. 2001. Optimality modeling and fitness trade-off: when shouldplants become pollinator specialists? Oikos 95: 177184.

Aizen MA, Feinsinger P. 1994. Forest fragmentation, pollination, and plantreproduction in a chaco dry forest, Argentina. Ecology 75: 320341.

Arditti J. 1976. Post-pollination phenomenon in orchid flowers. The OrchidReview 84: 261268.

Ashman TL, Knight TM, Steets JA, et al. 2004. Pollen limitation of plantreproduction: ecological and evolutionary causes and consequences.Ecology 85: 24082421.

Baker HG, Baker I. 1983. Floral nectar sugar constituents in relation to pol-linator type. In: Jones CE, Little RJ, eds. Handbook of experimental pol-lination biology. New York: Van Nostrand Reinhold, 117141.

Bergstrom G. 1978. The role of volatile chemicals in Ophrys-pollinator inter-actions. In: Harborne JB. ed. Biochemical aspects of plant and animalcoevolution. London: Academic Press, 207230.

Borba EL, Semir J. 2001. Pollinator specifity and convergence in fly-pollinated Pleurothallis (Orchidaceae) species: a multiple populationapproach. Annals of Botany 88: 7588.

Braga MMN, Braga PIS. 1975. Estudos sobre vegetacao das CampinasAmazonicas, IV. Estudos ecologicos na Campina de Reserva Biologicada INPA/SUFRAMA, km 45. Acta Amazonica 5: 4042.

Brys R, Jacquemyn H, Hermy M. 2008. Pollination efficiency and reproduc-tive patterns in relation to local plant density, population size, and floraldisplay in the rewarding Listera ovata (Orchidaceae). Botanical Journalof the Linnean Society 157: 713721.

Bullock SH, Martinez del Rio C, Ayala R. 1989. Bee visitation rates to treesof Prockia crucis differing in flower number. Oecologia 78: 389393.

Burd M. 1994. Batemans principle and plant reproduction: the role of pollenlimitation in fruit and seed set. Botanical Reviews 60: 83139.

Burd M. 1995. Ovule packaging in stochastic pollination and fertilizationenvironments. Evolution 49: 100109.

Burd M, Ashman T-L, Campbell DR, et al. 2009. Ovule number per flowerin a world of unpredictable pollination. American Journal of Botany 96:11591167.

Campbell DR. 1987. Interpopulational variation in fruit production: the roleof pollination limitation in the Olympic mountains. American Journalof Botany 74: 269273.

Casper BB, La Pine TR. 1984. Changes in corolla colour and other floralcharacteristics in Cryptantha humilis (Boraginaceae): cues to discourage

pollinators? Evolution 38: 128141.Catling PM. 1990. Auto-pollination in the Orchidaceae. In: Arditti J. ed

Orchid biology, reviews and perspectives V. Portland, OR: TimberPress, 121 158.

Catling PM, Catling VR. 1991. A synopsis of breeding systems and pollina-tion in North American orchids. Lindleyana 6: 187210.

Chittka L, Raine NE. 2006. Recognition of flowers by pollinators. CurrentOpinion in Plant Biology 9: 428435.

Corbet SA. 1998. Fruit and seed production in relation to pollination andresources in bluebell, Hyacinthoides non-scripta. Oecologia 114:349360.

Daumann E. 1970. Das Blutennektarium der Monocotyledonen unter beson-derer Berucksichtigung seiner systematischen und phylogenetischenBedeutung. Feddes Repertorium 80: 463590.

Davies KL, Winters C. 1998. Ultrastructure of the labellar epidermis inselected Maxillaria species (Orchidaceae). Botanical Journal of the

Linnean Society 126: 349361.

Dobson HEM. 1994. Floral volatiles in insect biology. In: Bernays EA. ed. Insectplant interactions. Boca Raton, FL: CRC Press, 4781.

Dressler RL. 1981. The orchids: natural history and classification.Cambridge, MA: Harvard University Press.

Dressler RL. 1982. Biology of the orchid bees (Euglossini). Annual Review ofEcology and Systematics 13: 373394.

Eltz T, Lunau K. 2005. Antennal response to fragrance compounds in maleorchid bees. Chemoecology 15: 135138.

Faegri K, van der Pijl L. 1979. The principles of pollination ecology, 3rd edn.Oxford: Pergamon Press.

Fenster CB, Armbruster WS, Wilson P, Dudash MR, Thomson JD. 2004.Pollination syndromes and floral specialization. Annual Review ofEcology, Evolution, and Systematics 35: 375403.

Frankie GW, Vinson SB, Coville RE. 1980. Male behavior of Centris adaniin the Costa Rican dry forest, with an emphasis in territoriality(Hymenoptera: Anthophoridae). Journal of the Kansas EntomologicalSociety 53: 837857.

Gentry AH. 1974. Flowering phenology and diversity in tropicalBignoniaceae. Biotropica 6: 6468.

Gerlach D. 1984. Botanische Mikrotechnik. Stuttgart: Thieme.Gerlach G. 1995. Duftanalysen ein Schlussel zum Verstandnis der

Bestaubungsbiologie neotropischer Parfumblumen. Tropenforschung 10:231240.

Gerlach G, Schill R. 1991. Composition of orchid scents attracting euglossinebees. Botanica Acta 104: 379391.

Gori DF. 1983. Post-pollination phenomena and adaptive floral changes. In:Jones CE, Little RJ. eds. Handbook of experimental pollinationbiology. New York: Van Nostrand Reinhold, 3149.

Janzen DH. 1971. Euglossine bees as long-distance pollinators of tropicalplants. Science 171: 203205.

Janzen DH. 1980. Self- and cross-pollination of Encyclia cordigera(Orchidaceae) in Santa Rosa National Park, Costa Rica. Biotropica 12:7274.

Jennings W, Shibamoto T. 1980. Qualitative analysis of flavor and fragrancevolatiles by glass capillary gas chromatography. New York: AcademicPress.

Johnson SD, Bond WJ. 1992. Habitat dependent pollination success in a Capeorchid. Oecologia 91: 455457.

Johnson SD, Edwards TJ. 2000. The structure and function of orchid polli-naria. Plant Systematics and Evolution 222: 243269.

Johnson SD, Nilsson LA. 1999. Pollen carryover, geitonogamy, and the evol-ution of deceptive pollination systems in orchids. Ecology 80:26072619.

Johnson SD, Steiner KE. 2000. Generalization versus specialization in plantpollination systems. Trends in Ecology and Evolution 15: 140143.

Johnson SD, Linder HP, Steiner KE. 1998. Phylogeny and radiation of pol-lination systems in Disa (Orchidaceae). American Journal of Botany 85:402411.



16/16

Ju rgens A, Webber AC, Gottsberger G. 2000. Floral scent compounds ofAmazonian Annonaceae species pollinated by small beetles and thrips.Phytochemisty 55: 551558.

Knudsen JT, Eriksson R, Gershenzon J, Stahl B. 2006. Diversity and distri-bution of floral scent. The Botanical Review 72: 1120.

Kullenberg B. 1961. Studies in Ophrys pollination. Zoologiska Bidrag FranUppsala Vol. 34. Uppsala: Almqvist & Wiksells Boktryckeri AB.

Leys R, Cooper SJB, Schwarz MP. 2002. Molecular phylogeny and historical

biogeography of the large carpenter bees, genus Xylopcopa(Hymenoptera: Apidae). Biological Journal of the Linnean Society 77:249266.

Lock JM, Profita JC. 1975. Pollination of Eulophia cristata (Sw.) Steud.(Orchidaceae) in Southern Ghana. Acta Botanica Neerlandica 24:13

Pollination Biology of Eulophia Alta Orchidaceae

Documents

Transcript of Pollination Biology of Eulophia Alta Orchidaceae