Nectar properties of the sunbird-pollinated plant Impatiens sakeriana: A comparison with six other co-flowering species

This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/copyright

Author's personal copy

Nectar properties of the sunbird-pollinated plant Impatiens sakeriana:A comparison with six other co-flowering species

M. Bartoš a, b,⁎, Š. Janeček b, E. Padyšáková a, b, E. Patáčová b, J. Altman a, b, M. Pešata b,J. Kantorová b, R. Tropek a, c

a Faculty of Science, University of South Bohemia, Branišovská 31, CZ-370 05 České Budějovice, Czech Republicb Institute of Botany, Academy of Sciences of the Czech Republic, Dukelská 135, CZ-379 82 Třeboň, Czech Republic

c Institute of Entomology, Biology Centre, Academy of Sciences of the Czech Republic, Branišovská 31, CZ-370 05 České Budějovice, Czech Republic

Received 20 December 2010; received in revised form 9 May 2011; accepted 11 May 2011

Abstract

Adaptations of the nectar traits in bird-pollinated flowers are amongst the most discussed aspects of floral evolution. In the case of sunbird-pollinated plants, data on nectar traits originate almost exclusively from the South African region and are very scarce for tropical Africa, whereparadoxically the highest sunbird diversity occurs. Here we present a study on the nectar properties of a sunbird-pollinated plant, Impatienssakeriana, growing in the West African mountains, including the nectar production, diurnal changes in the nectar standing crop, the nectarconcentrations, the nectar volumes, total sugar amounts and sugar composition. Moreover we compare the nectar traits of I. sakeriana with sixother co-flowering insect-visited plant species.

Our results showed that many nectar properties, including high volume (approx. 38 μL in flowers unvisited by sunbirds), low sugarconcentration (approx. 30% w/w) and high sucrose content (95%), are specific to I. sakeriana, compared to the insect-visited plants. These are inaccordance with the most recent theory that nectar properties of the sunbird-pollinated plants are similar to those pollinated by hummingbirds.© 2011 SAAB. Published by Elsevier B.V. All rights reserved.

Keywords: Cameroon; Impatiens sakeriana; Nectar; Pollination; Sunbird

1. Introduction

Nectarivory in birds is a widespread phenomenon, especiallyin the tropical and subtropical areas with long floweringseasons. It has been estimated that around 10% of all birdspecies may use nectar as a resource (Wolf and Gill, 1986). Themost famous nectar feeders in the New World are humming-birds, representing the most specialised bird pollinators (Stiles,1978; Arizmendi and Ornelas, 1990; Stiles and Freeman, 1993;Schuchmann, 1999). In the Old World the most specialisednectarivorous birds are sugarbirds, flowerpeckers, sunbirds andspiderhunters (Cheke et al., 2001).

Plants pollinated by birds are expected to produce highervolumes of more diluted nectar than insect-pollinated plants(Bolten and Feinsinger, 1978; Nicolson and Fleming, 2003a;Goldblatt and Manning, 2006). The higher nectar amount isbelieved to be related to the high energetic requirements of thebirds (Heinrich, 1981). The function of low sugar concentrationin the bird-pollinated flowers is much more debatable. Twohypotheses have been offered to explain the low concentrationsin regard to the evolution of the most appropriate nectarproperties for the birds. Baker (1975) noticed that nectar with alow sugar concentration has also a low viscosity, whichfacilitates more efficient extraction from flowers. Calder(1979) suggested that the dilute nectars can support bird waterrequirements in warm to hot environments, but at the same timehe pointed to possible problems with excessive water inputunder lower temperatures. More recent studies show that just

⁎ Corresponding author at: Tel.: +420 384721156; fax: +420 384721136.E-mail address: [email protected] (M. Bartoš).

0254-6299/$ - see front matter © 2011 SAAB. Published by Elsevier B.V. All rights reserved.doi:10.1016/j.sajb.2011.05.015

Available online at www.sciencedirect.com

South African Journal of Botany 78 (2012) 63–74www.elsevier.com/locate/sajb

Author's personal copy

the excess water from dilute nectars represents the prevailingosmoregulatory challenge to nectar-feeding birds. As aconsequence, the potential advantages related with more dilutednectars of bird-pollinated plants are debatable (Martinez del Rioet al., 2001). The daily water intake of birds feeding on dilutenectar can be several times higher than their body weight(McWhorter and Martinez del Rio, 1999; Lotz and Nicolson,1999). The nectar-feeding birds are well adapted for eliminationof surplus water. A South American hummingbird, Sephanoidessephanoides, can regulate the redundant water by decreasingwater reabsorption in its kidneys (Hartman Bakken and Sabat,2006). Another studied nectarivorous bird, the Palestine sunbird(Nectarinia osea), is moreover able to regulate intestinal waterabsorption (McWhorter et al., 2003; McWhorter et al., 2004).The higher ability of the Whitebellied Sunbird (Nectariniatalatala) to produce highly diluted cloacal fluid when feedingon a dilute sucrose solution was manifested by Fleming andNicolson (2003). Besides avoiding over-hydration during thefeeding on more diluted nectars, the nectar-feeding birds facedehydration during times of fasting (naturally during the night)(Hartman Bakken et al., 2004; Hartman Bakken and Sabat,2006; Fleming et al., 2004a). Different bird groups solve thisdilemma in different ways: whereas sunbirds are able to produceconcentrated cloacal fluid (Fleming and Nicolson, 2003),hummingbirds conserve water balance only by remarkablereduction of their glomerular filtration rate (Hartman Bakkenet al., 2004; Hartman Bakken and Sabat, 2006). Dehydrationduring the day activity is rather improbable as the birds canconsume supplementary water (Nicolson and Fleming, 2003b).

Nevertheless the plants produce nectar to increase their fitnessand not to satisfy altruistically birds' requirements (Pyke, 1981;Pyke and Waser, 1981; Martinez del Rio et al., 2001). Inconsequence, we need to consider the nectar properties as acompromise between plant and bird concerns. This could be thereason why the plants specialised to opportunistic nectarivoresproduce relatively less concentrated nectar (8–12%) than plantsspecialised to specialist nectarivores (15–25%), as they need tosatisfy the highly different requirements of their pollinators(Johnson and Nicolson, 2008; Botes et al., 2008; Symes et al.,2008; Brown et al., 2010a,b; Odendaal et al., 2010; Symes et al.,2010). On the other hand, it has been shown that plants pollinatedby specialised birds produce a bit more diluted nectar thanexpected from the birds' preferences (Tamm and Gass, 1986;Roberts, 1996; Blem et al., 1997; Blem et al., 2000; Johnson et al.,2006). This disproportion can be explained considering that adifferent force other than ornithocentricity can modify thenectar properties. The plant aims to reduce the costs relatedwith nectar production and hence to produce less nectar volumeand/or lower nectar concentrations (Bronstein, 2001). Bolten andFeinsinger (1978) proposed that diluted nectar in the humming-bird-accessible flowers evolved not to attract hummingbirds butto avoid attracting bees. Other ecologists explain the relativelylow nectar concentrations of the bird-pollinated flowers byspecific nectar composition and secretion patterns (Nicolson,2002; Nicolson and Fleming, 2003a). Alternatively, the dilutenectars can be also a solely secondary consequence of deep tubularflowers of the bird-pollinated plants, which minimise water

evaporation (Plowright, 1987). Pyke and Waser (1981) speculatedon the hypothesis that nectar properties evolved to affect pollinatorforaging behaviour.

The relation of the nectar sugar composition to the pollinatorclass has also been repeatedly questioned (e.g. Galetto andBernardello, 2004; Chalcoff et al., 2006; Wolff, 2006; Schmidt-Lebuhn et al., 2007). A long-standing paradigm has been thathummingbirds and passerine birds differ in preferences ofnectar sugar composition in the plants pollinated by them.Hummingbirds were generally regarded as sucrose-dominantnectar consumers in comparison with the Old World nectar-ivorous birds, which prefer hexose-dominant nectar (Bruneau,1997; Baker et al., 1998; Nicolson and Fleming, 2003a;Schmidt-Lebuhn et al., 2007). The hypothesis on thisdichotomy is however controversial, as there are large groupsof sunbird-pollinated plants in the Old World which producesucrose-rich nectar (Vos et al., 1994; Barnes et al., 1995). Thebroader synthesis of Nicolson and Fleming (2003a) showed thatthe passerine-pollinated plants embody a bimodal pattern with ahigh number of plants with high sucrose content and manyspecies with hexose-rich nectar. Moreover, the dichotomy wasalso put into doubt by several experimental studies showing thatthe specialised Old World passerines are able to absorb sucroseeffectively (Lotz and Nicolson, 1996; Downs, 1997; Jacksonet al., 1998) and did not prefer hexoses (or in extremely dilutedsolutions only) using equicaloric (Fleming et al., 2004b; Brownet al., 2008) or equiweight (Lotz and Nicolson, 1996; Brownet al., 2008) solutions. Recently, Johnson and Nicolson (2008)have proclaimed this dichotomy as false, and suggested a moreuseful distinction between specialised (i.e. including justhummingbirds or sunbirds) and generalised (i.e. includingbulbuls, weavers, orioles and others) bird pollination systems.Moreover, recent experiments show that the generalist aviannectarivores prefer hexose solutions of concentrations similar asthose found in the plants adapted to these birds (Fleming et al.,2008; Odendaal et al., 2010; Brown et al., 2010a,b).

The theories on the nectar properties of the Old World plantspollinated by specialised sunbirds can be nevertheless stronglyaffected by unbalanced geographical data acquisition. In mostof the synthesising studies the data predominates from the SouthAfrican region (Baker et al., 1998; Johnson and Nicolson,2008), which represents just a marginal area of sunbirddistribution (Cheke et al., 2001). Only a few studies havebeen done on the plant nectar properties in tropical Africa (Voset al., 1994; Burd, 1995; Evans, 1996; Johnson and Brown,2004; Ley and Claßen-Bockhoff, 2009), where paradoxicallythe highest diversity of sunbirds occurs (Cheke et al., 2001).

Moreover, the scarce data from the Afrotropical areas seems tobe quite inconsistent. It has been shown that sunbird-pollinatedLobelia telekii growing onMt. Kenya have nectar concentrations ofaround 60% (Evans, 1996), whereas Lobelia deckenii, growing onMt. Kilimanjaro and visited by the sunbird Nectarinia johnstoniand the mountain chat Cercomela sordida, have only around 8%nectar (Burd, 1995). However, the study of Burd (1995) did notdeterminate the pollination effectiveness of both birds and if thegeneralist chat is the more effective pollinator, the extremelydilute nectar in L. deckenii corresponds to the Johnson and

64 M. Bartoš et al. / South African Journal of Botany 78 (2012) 63–74

Author's personal copy

Nicolson (2008) conclusions, that plants adapted to generalisedbird pollinators have very diluted nectars. Other studied specieshave nectar concentrationsmore typical for the sunbird-pollinatedplants (approx. 15–25%w/w; Johnson andNicolson, 2008). Suchconcentrations have been also demonstrated for several otherspecialised plants, such as several Maranthaceae species fromGabon (Ley and Claßen-Bockhoff, 2009), South-African, East-African and pantropical species of the genus Leonotus (Vos et al.,1994), and a sunbird-pollinated orchid, Disa satyriopsis, fromMalawi (Johnson and Brown, 2004).

It this study, we focused on the nectar properties of Impatienssakeriana,which grows in montane forest and along streams in theCameroonian mountains, and is fully dependent upon sunbirdpollination (Janeček et al., 2011). Moreover, we analysed nectarproperties of six other co-flowering plants visited by insects toreveal if these properties are unique in the wider communitycontext.

2. Materials and methods

2.1. Study area

Fieldwork was carried out in the Mendong Buo area, near theBig Babanki village, the Bamenda Highlands, North-WestProvince, Cameroon (06°05′26″ N, 10°18′09″ E; 2200 m a.s.l.)from November 2007 to January 2008. The study areaexperiences a single wet season from March/April to mid-November, with the precipitation ranging from 1780 to2290 mm/year (Cheek et al., 2000). Work started at thebeginning of the dry season, with the start of the target plants'flowering peaks. The studied area was composed of a mosaic ofAfromontane forest remnants, extensive and intensive pastures,forest clearings and abandoned pastures dominated by Pteridiumaquilinum, and scrubby stream mantle vegetation (for moredetails see Reif et al., 2007; Tropek andKonvicka, 2010; Hořák etal., 2010). The studied plant species were concentrated mainlywithin the forest edges and along streams.

2.2. Study plant species and their visitors

The nomenclature and known species characteristics followCheek et al. (2000). Information on the sunbird visitorsoriginates from our previous studies (Riegert, 2011; Janeček

et al. unpublished results); the insect visitors are covered byTable 1 (for more details see the next paragraph).

I. sakeriana Hook.f. (Balsaminaceae) is a 3 m erect herb ofthe mountain forest or its edge. Deeply red zygomorphicflowers, presented year-round, are long pediculate. Flowerlongevity is about six days. I. sakeriana is pollinated only bytwo sunbird species (Cyanomitra oritis, Cinnyris reichenowi);no insect visitors were observed (Janeček et al., 2011)(Fig. 1a).

Lobelia columnaris Hook.f. (Campanulaceae) is a 3 m erectherb of the mountain forest edges and the mountain grasslands.Blue or pink-purple flowers are comprised to the terminalpyramidal panicle. The flowers live about ten days and arepresented in the dry season. L. columnaris is visited by manyinsect functional groups and also by sunbirds (Cynnyrisbouvieri, C. reichenowi). This species forms the monophyleticgroup with the pachycaul lobelias of East Africa, which arestrongly adapted to bird pollinators (Evans, 1996; Antonelli,2009) (Fig. 1b).

Hypoestes aristata (Vahl) Roem. & Schult. (Acanthaceae) isa 1 m erect herb of the mountain forest or its edge. Pale mauveflowers with darker markings are clustered to the whorls. Theflowers live about five days and are presented in the dry season.H. aristata is visited by many insect functional groups and alsoby sunbirds, especially C. reichenowii (Fig. 1c).

Hypericum revolutum Vahl. (Gutifereae) is an up to 12 mshrub or tree of the mountain forest edges and the streammantels. Its flat yellow flowers develop solitarily on shootapices. Flower longevity is about two days. H. revolutum isvisited by many insect functional groups and also by allpresented sunbird species, which, however, contribute little toits pollination (Janeček et al., 2007) (Fig. 1d).

Brillantaisia lamium (Nees) Benth. (Acanthaceae) is a 1.5 merect herb of the mountain forest edges. Flowers are purple toblue-coloured and form the lax panicle. Flower longevity isabout three days. It is visited just by a few specialised insectfunctional groups (Fig. 1e).

Pycnostachys eminii Gürke (Lamiaceae) is an up to 3 m herbof the mountain forest edges and the mountain grasslands. Paleblue flowers are conglobated into cylindrical spikes. Flowerlongevity is about two days. The flowers are visited by bees andsporadically by C. bouvieri and C. reichenowi, but the birdsshowed negative selection to the plant and their visits are justaccidental (Janeček et al., unpublished results) (Fig. 1f).

Table 1Summary of visits of the morphotaxonomical functional insect groups in the individual plant species.

Col-s Col-n Hym-t Hym-e Hym-s Hym-p Thy Lep Dip-s Dip-n Het Auc Nr. guilds

Impatiens sakeriana 0Hypoestes aristata 69 75 14 9 5 4 6Lobelia columnaris 8 13 48 5 4Hypericum revolutum 795 235 32 507 53 14 129 6 11 9Pycnostachys eminii 6 4 2Brillantaisia lamium 5 4 4 3Virectaria major 5 8 8 32 7 5 6 7

COL-S— highly specialised beetles; COL-N— other nectarivorous and pollenivorous beetles; HYM-T— highly specialised bees with long tongue; HYM-E— bees withlarge societies; HYM-S— bees with small societies and solitary species; THY— thrips; LEP— butterflies and hawk moths; DIP-S— specialised flies; DIP-N— otherflies; HYM-P — parasitoid hymenopterans; HET— true bugs; AUC — leafhoppers.

65M. Bartoš et al. / South African Journal of Botany 78 (2012) 63–74

Author's personal copy

Virectaria major (K. Schum.) Verds. (Rubiaceae) is a 2 mweak-stemmed shrub of the mountain forest edges and themountain grasslands. Pale purple flowers form the erect terminalclusters. Flower longevity is about two days. It is visited by manyinsect functional groups (Fig. 1g).

2.3. Insect visitors—the pilot study

To explore the spectrum of insects visiting the target plants weperformed a pilot-study where we observed individual functional

insect groups. The visitation of the target plants by individual insectfunctional groups was recorded in sixteen 15 m transects of thestream edge vegetation. Each plant species was observed 5 min pereach transect (if present) and visit (if actually flowering). Theobservations were equally distributed within the day and the studyperiod. The recording was limited from 09 h00 to 16 h00, theactivity peak of the most insect pollinators, and to suitable weather(at least partly cloudy). We observed each plant species from 4 to18 h, depending on its abundance in the studied community and itsphenology. In spite of the relatively different sampling effort, the

Fig. 1. The studied plant species. Scale bars=1 cm; (a) Impatiens sakeriana; (b) Lobelia columnaris; (c)Hypoestes aristata; (d)Hypericum revolutum; (e) Brillantaisialamium; (f) Pycnostachys eminii; (g) Virectaria major.

66 M. Bartoš et al. / South African Journal of Botany 78 (2012) 63–74

Author's personal copy

insect functional group number is saturated for all studied plants,which allowed us to compare the individual plant species (seeSupplementary Fig. 1).

Each specimen was classified into one of the 12 morphotax-onomical functional groups according to their relation to thepollination process (modified after Williams et al., 2001 andFenster et al., 2004): highly specialised beetles (Coleoptera:Lycidae); the other nectarivorous and pollenivorous beetles(Coleoptera); highly specialised bees with long tongues(Hymenoptera: Apinae); bees with large societies (Hymenop-tera: Apinae); bees with small societies and solitary species(Hymenoptera: Apinae); thrips (Thysanoptera); butterflies andhawk moths (Lepidoptera); specialised flies (Diptera: Syrphidaeand Bombyliidae); other flies (Diptera); parasitoid hymenop-terans (Hymenoptera: Parasitica); true bugs (Heteroptera); andleafhoppers (Auchenorrhyncha). To avoid accidental visits, weconsidered the plant-visitor group relationship only if the groupwas recorded more than three times in the plant species.Consequently, some of the potential groups (e.g. wasps, ants,spiders or carnivorous beetles) were so occasional that theywere not considered at all. See Table 1 for frequencies of theinsect functional groups visiting the target plant species.

2.4. Nectar production in bagged flowers

To measure nectar production, we followed the methods ofTorres and Galetto (1998). We established the flower sets of 16bagged flowers for individual plant species. The number of setswas species specific to cover the whole flower lifetime assessedduring our pilot studies. The flowers were bagged beforeanthesis. The samples were collected from the flowers ofdifferent ages (flower age classes) at the same time of a day —06 h30 and 16 h30 for the long flowering species (I. sakeriana,L. columnaris) or 06 h30, 11 h30 and 16 h30 for the shortflowering species (H. revolutum, H. aristata, B. lamium,V. major, P. eminii). Because of the small amounts of nectarin the flowers, the nectar samples of P. eminii werecumulatively collected from more flowers in one inflorescenceand later recounted for one flower.

2.5. Nectar standing crop

The nectar samples for the standing crop evaluation werecollected from flowers which were fully exposed to pollinatorsand other visitors. The samples were collected from randomlyselected plants. From each selected plant just one flower wasanalysed (except for P. eminii where more flowers wereanalysed as mentioned above). The samples were collected in 5series in one week intervals. In each series we sampled 12flowers at three different times of the day (6 h30, 11 h30,16 h30).

2.6. Nectar traits

To gather a total nectar amount, all species were sampleddestructively using 5, 10, 25 μL microcapillaries or Hamiltonsyringes (appropriate to flower size and nectar volume). The

sugar concentration was measured using a Pal-1 (Atago co.)pocket refractometer. Small amounts of nectar (mostly highlyconcentrated and highly viscous) were diluted by distilled wateron the refractometer and the concentration was calculated forthe original amount. Total amount of sugar per flower wascalculated from sugar concentration per unit volume (mg/μL)and sugar volume (Bolten et al., 1979). For calculation of sugaramount per μL from w/w concentration (the concentrationmeasured by refractometer) we used exponential regressionequation (Galetto and Bernardello, 2005).

2.7. Nectar sugar composition

For assessment of the nectar sugar composition, 25 randomlyselected flowers of each studied species were sampled. Thenectar samples were carried over to a Whatman filter paper andquickly dried and stored with silica gel in small plastic bags. Foranalyses the nectar samples were washed out from the filterpaper by distilled water.

Sugars were analysed and their relative masses quantified byhigh-performance liquid chromatography (HPLC) using theICS-3000 system (Dionex), with an electrochemical detectorand CarboPac PA 1 column.

2.8. Statistical analyses

All statistical tests were performed using STATISTICA 7.0(Anon, 1996). The data on the sugar content, nectar volume andsucrose/hexose ratio were log-transformed and percentage dataon nectar concentration and nectar residues in non-baggedflowers were arcsin square-root transformed before the analysesto improve normality.

3. Results

3.1. Nectar production in bagged flowers

The accumulation of nectar in the flowers of individualspecies is presented in Fig. 2. We observed two stages of nectarvolume accumulation in the flowers of I. sakeriana, H. aristata,L. columnaris and P. eminii. The first stage was associated withthe volume increase and the second stage with the nectarvolume stagnation. The pattern considering the sugar amountwas similar but an increasing stage was observed additionallyfor H. revolutum. The accumulation stage seems to beinterconnected with the flower longevity. Longer accumulationstages were detected for L. columnaris (approx. four days) andI. sakeriana (approx. 3 days). The single specific pattern ofnectar accumulation for the sunbird specialised plantI. sakeriana was the stability of nectar concentrations. Thehighest fluctuations of nectar concentration were detected inflowers of H. revolutum.

3.2. Nectar standing crop

I. sakeriana had a relatively stable diurnal nectar volume andsugar amount, considering our standing crop data (Fig. 3).

67M. Bartoš et al. / South African Journal of Botany 78 (2012) 63–74

Author's personal copy

68 M. Bartoš et al. / South African Journal of Botany 78 (2012) 63–74

Author's personal copy

Nevertheless, a similar pattern was observed also for the insect-visited plants L. columnaris, P. eminii and B. lamium. Threespecies (H. aristata, V. major and H. revolutum) had muchhigher nectar volume and sugar amounts in the morning,compared with the volumes at midday and in the evening. Forall studied plants, the highest nectar concentrations weredetected at midday.

3.3. Nectar traits

3.3.1. Sugar concentrationThe studied plant species differed in nectar concentrations:

nectars from the bagged flowers were more concentrated thanfrom the flowers fully exposed to any visitors; this decrease ofthe concentration was species specific (Factorial ANOVA;Plant species: d.f. =6; F=63.6, pb0.01; Type of nectarcollection (bagged flowers x standing crop): d.f.=1, F=194.2,pb0.01; Interaction: d.f.=6; F=10.0; pb0.01; Fig. 4). Themean concentration of the cumulative production samplesranged from 31% to 63%. The lowest mean concentration wasrecorded in the flowers of the strictly bird specialisedI. sakeriana, the highest in the flowers of the insect-visitedB. lamium.

The mean concentration of the standing crop samples rangedfrom 19% to 49%. The lowest mean concentration was recordedin the flowers of H. revolutum, the highest in the flowers ofP. eminii.

The mean concentrations of the bagged flowers differedbetween the bird specialised I. sakeriana and the insect-visitedplants (single-sample t-test, d.f.=5, t=5.01, pb0.01) but werenon-significant considering the standing crop data (single-sample t-test, d.f.=5, t=1.56; p=0.18).

3.3.2. Nectar volumeThe volume from the bagged flowers was higher than the

volume from the fully exposed flowers; this decrease of thevolume was species specific (Factorial ANOVA; Plant species:d.f.=6; F=550.6, pb0.01; Type of nectar collection (baggedflowers x standing crop): d.f.=1; F=282.7, pb0.01; Interac-tion: d.f.=6; F=69; pb0.01; Table 2). The nectar residue in thefully exposed flowers compared with the volume in the baggedflowers was the lowest for I. sakeriana (9%) and the highest forB. lamium (61%) (Fig. 4).

The bagged flower nectar crop volume ranged from 0.3 μL(P. eminii) to 38 μL (I. sakeriana). The bird specialisedI. sakeriana had a higher volume than the insect-visited plants(single-sample t-test, d.f.=5, t=2.5875, pb0.01).

The standing crop nectar volumes ranged from 0.06 μL(P. eminii) to 6.95 μL (H. revolutum). The volumes fromI. sakeriana did not differ from the other species (single-samplet-test, d.f.=5, t=−1.756, p=0.14). The nectar volume residue(percentage of nectar volume residue in the flowers accessibleto floral visitors (standing crop) when mean cumulative nectarproduction was considered as 100%) was lowest for

I. sakeriana (single-sample t-test, d.f.=5, t=2.99, pb0.05,Table 2).

3.3.3. Total amount of sugarThe total sugar amounts in the bagged flowers were higher

than the amounts in the flowers fully exposed to any visitors,and this decrease of the amount was species specific (FactorialANOVA; Plant species: d.f.=6; F=801.1, pb0.01; Type ofnectar collection (bagged flowers x standing crop): d.f.=1;F=246.6, pb0.01; Interaction: d.f. =6; F=85.6; pb0.01;Table 2). Considering our nectar production data, the highestsugar amounts per flower were recorded in the flowers of thesunbird-pollinated plant I. sakeriana, and the lowest in theflowers of P. eminii.

The mean sugar amount was higher in the bird specialisedI. sakeriana flowers, compared to the insect-visited plantspecies (single-sample t-test, d.f.=5, t=−3.73, pb0.05). Incontrast, considering the standing crop data, only a statisticallymarginal difference between I. sakeriana and the insect-visitedplants was revealed (single-sample t-test, d.f.=5, t=2.52,p=0.053). Sugar amount residue (percentage of nectar volumeresidue in the flowers accessible to floral visitors (standingcrop) when mean cumulative nectar production was consideredas 100%) was the lowest (but non-significantly) for I. sakeriana(single-sample t-test, d.f.=5, t=2.11, p=0.09, Table 2).

3.4. Nectar sugar composition

The nectar sugar composition is shown in a ternary diagram(Fig. 5). Nectar of the four species was strongly sucrosedominant: H. revolutum and I. sakeriana had very similarcompositions, with 95% and 94% sucrose, as well as V. majorand B. lamium, with 82% and 80% sucrose. H. aristataproduced hexose dominant nectar, with only 3.5% sucrose.L. columnaris produced nectar somewhat higher in sucrose(26%), but still decidedly hexose dominant. The sucrose/hexoseratio differed between species (ANOVA, F=136, pb0.01), butno differences were found between I. sakeriana and the insect-visited species (single-sample t-test, d.f.=5, t=0.847, p=0.44).

4. Discussion

We offer a unique dataset of the nectar properties of severalplant species of the West African mountains. Our data alsoshowed that the highly specialised I. sakeriana have differentnectar properties compared with the insect-visited co-floweringplants.

The properties of the nectar produced by flowers ofI. sakeriana correspond with the recent ideas of Johnson andNicolson (2008), based mainly on South African plants andsupported by further studies (Botes et al., 2008; Symes andNicolson, 2008; Brown et al., 2010b,c). Johnson and Nicolson(2008) suggested that plants adapted to specialised pollinatorsproduce a high volume of nectar with a high sucrose content and

Fig. 2. Nectar production in bagged flowers of different ages, the numbers on x-axes represent individual sampling cohorts (from youngest to oldest cohort). Meansand SE are shown.

69M. Bartoš et al. / South African Journal of Botany 78 (2012) 63–74

Author's personal copy

Fig. 3. Diurnal changes in the nectar standing crops. Means and SE are shown.

70 M. Bartoš et al. / South African Journal of Botany 78 (2012) 63–74

Author's personal copy

sugar concentrations (approx. 15–25% w/w) in contrast toplants adapted to generalised omnivorous birds which producehexose-rich nectar of lower concentrations (8–12% w/w). Wetherefore support the above-mentioned general rejection of theprevious passerine — non-passerine dichotomy in nectarproperties and the related hypothesis that the plants in the OldWorld pollinated by passerine birds have hexose-rich nectars(Bruneau, 1997; Baker et al., 1998; Nicolson and Fleming,2003a; Schmidt-Lebuhn et al., 2007). The high proportion ofsucrose in the nectar of I. sakeriana can be seen as a co-adaptation between I. sakeriana and a specialised nectarivorousCameroon Sunbird (C. oritis). As several experimental studieshave showed, the highly specialised birds prefer sucrosesolutions when the sugar concentrations are similar to thosefound in bird-pollinated flowers (Schonube and Martinez delRio, 2003; Fleming et al., 2004b; Fleming et al., 2008; Brown etal., 2008; Brown et al., 2010c). What is interesting is the slightlyhigher nectar concentration in the flowers of I. sakeriana(30.88% w/w in bagged flowers) than the nectar concentrationrange common in sunbird and hummingbird flowers (15–25%w/w). The concentration of I. sakeriana is nevertheless nearlyidentical as the 31% w/w concentration suggested by Nicolson

and Fleming (2003b) to be preferred by the WhitebelliedSunbird (N. talatala) as the optimal concentration for balancedintake of energy and water.

Nectar volumes and total sugar amounts in unbagged flowersare much lower due to foraging of flower visitors; the decreasein these parameters is even higher in flowers of the sunbirdspecialist I. sakeriana. On the other hand, we assume that thenectar properties of unvisited flowers much better reflect thereal plant adaptations.

Our results show considerable differences in the nectar sugarconcentration between the flowers with excluded visitors(higher concentrations) and the fully exposed flowers (lowerconcentrations). This effect was obvious mainly in insect-visited species. Nevertheless, as we did not study underlyingmechanisms, we can only speculate on the cause of this pattern.As some recent studies (De Vega et al., 2009; Herrera et al.,2008, 2009) have indicated, the nectar sugar concentrationcould be strongly affected by yeast presence. Herrera et al.(2008) observed the influence of increasing yeast density on thereduction of nectar sugar concentration and energetic value ofnectar in three Spanish plant species. De Vega et al. (2009)provide a survey of the frequency and abundance of yeast in

Fig. 4. Nectar concentrations using the nectar accumulation data (filled bars) and the standing crop data (empty bars). Means and SE are shown.

Table 2Nectar volume, total sugar amount and nectar residue of the studied plant species.

Plant species Nectar volume per flower Total amount of sugar per flower

Cumulative nectarproduction (μL)

Standingcrop (μL)

Nectarresidue (%)

Cumulative nectarproduction (mg)

Standingcrop (mg)

Nectarresidue (%)

Impatiens sakeriana 38.42±1.98a 3.51±0.60a 9.14 14.02±0.76a 1.14±0.19a 8.13Hypoestes aristata 1.27±0.13b 0.22±0.03b 17.32 0.98±0.10b 0.08±0.01bc 8.16Lobelia columnaris 24.46±2.10c 6.56±0.71c 26.82 11.34±0.87c 1.68±0.17d 14.81Hypericum revolutum 19.46±2.19c 6.95±0.93c 35.71 6.73±0.69d 1.24±0.17a 18.42Pycnostachys eminii 0.29±0.02d 0.06±0.01b 20.69 0.24±0.02e 0.04±0.005c 16.67Brillantaisia lamium 0.78±0.09bd 0.47±0.05b 60.26 0.55±0.06be 0.26±0.03b 47.27Virectaria major 2.13±0.52b 0.21±0.04b 9.86 0.99±0.28be 0.09±0.02bc 9.09

Nectar volume, total sugar amount and nectar residue of the studied plant species.The same superscripts indicate non-significant differences between individual plant species (posthoc Tukey HSD test), means and SE are shown, Nectar residue —percentage of nectar residue in the flowers accessible to floral visitors (standing crop) when mean cumulative nectar production was considered as 100%.

71M. Bartoš et al. / South African Journal of Botany 78 (2012) 63–74

Author's personal copy

floral nectar from 40 taxonomically diverse South African plantspecies. Variation in yeast incidence amongst plant species wasrelated to differences in pollinator type; the highest proportionof flowers with yeasts was found in the bird-pollinated plantspecies, whilst the lowest values were in the plants visited onlyby Hymenopterans. They moreover showed that nectarconcentration is negatively related with yeast cell density innectar samples of bird-pollinatedWatsonia pillansii. A differenttendency of nectar concentration was found in South AfricanKniphofia caulescens, pollinated by short-billed opportunisticavian nectarivores (Brown et al., 2009). The mean nectarconcentration in flowers bagged for 24 h was 8.5%, whilst themean concentration from standing crops was 10.6%. However,the higher concentrations of nectar sugar in the bagged flowerscould be caused by artificial conditions within the bag (Galettoand Bernardello, 2005). Higher temperature under the bags canincrease evaporation and in consequence nectar concentration.Naturally, the nectar volumes of some open flowers (such asH. revolutum flowers) could be diluted by morning watercondensation in wet environments or the water could moreeasily evaporate from the nectar in hot and dry conditionsduring the day.

Although I. sakeriana represents the only studied plantproducing nectar which is consumed exclusively by sunbirds, itwas observed that sunbirds in the studied locality feed on awider spectrum of available plants (Riegert,, 2011; Janeček etal., unpublished results). It was observed that all three sunbirdspecies occurring in the target area (C. oritis, Cinnyris bouvieri,C. reichenowi) feed frequently on unspecialised H. revolutum,which is however not effectively pollinated by them (Janeček etal., 2007). Additionally, C. oritis feed often on H. sakeriana,C. bouvieri on L. columnaris and C. reichenowi on H. aristata.Nevertheless only I. sakeriana seems to be fully specialised onbird-pollination (Janeček et al., 2011), as the other species arefrequently visited by various insects (see Table 1). Althoughadditional experiments on the pollination effectiveness areneeded, these observations together with this study are inagreement with 1) the often-observed nectarivorous birds

feeding on plants which do not have traits related to the bird-pollination syndrome (Brown and Hopkins, 1995; Franklin andNoske, 2000; Cheke et al., 2001), 2) the idea of a commonoccurrence of asymmetric specialisation in plant-pollinatorrelationships (Vázquez and Aizen, 2004) and 3) studiesshowing that sunbirds are able to feed on a wide range ofnectar concentrations (Lotz and Nicolson, 1999; Nicolson andFleming, 2003b).

Acknowledgements

We would like to thank J. Riegert, O. Sedláček, D. Hořák, J.Reif, D. Fainová, M. Antczak, V. Mikeš and Ernest Vunan forhelp during the field work, and M. Sweney for Englishproofreading. The research was supported by the Czech ScienceFoundation (206/08/H044), the Ministry of Education of theCzech Republic (LC 06073), the Grant Agency of the CzechAcademy of Sciences (IAA601410709), Grant Agency of theCzech Republic (P505/11/1617), GAJU (136/2010/P) and bythe project AV0Z60050516.

References

Anon, 1996. STATISTICA for Windows. Computer program manual Stat Soft,Tulsa, OK, USA.

Antonelli, A., 2009. Have giant lobelias evolved several times independently?Life form shifts and historical biogeography of the cosmopolitan and highlydiverse subfamily Lobelioideae (Campanulaceae). BMC Biology 7, 82.

Arizmendi, M.C., Ornelas, J.F., 1990. Hummingbirds and their floral resourcesin a tropical dry forest in Mexico. Biotropica 22, 172–180.

Baker, H.G., 1975. Sugar concentrations in nectar from hummingbird flowers.Biotropica 7, 37–41.

Baker, H.G., Baker, I., Hodges, S.A., 1998. Sugar composition of nectar andfruits consumed by birds and bats in the tropics and subtropics. Biotropica30, 559–586.

Barnes, K., Nicolson, S.W., Van Wyk, B.E., 1995. Nectar sugar composition inErica. Biochemical Systematics and Ecology 23, 419–423.

Blem, C.R., Blem, L.B., Cosgrove, C.C., 1997. Field studies of rufoushummingbird sucrose preference: does source height affect test results?Journal of Field Ornithology 68, 245–253.

Blem, C.R., Blem, L.B., Felix, J., Van Gelder, J., 2000. Rufous hummingbirdsucrose preferenceNprecision of selection varies with concentration. TheCondor 102, 235–238.

Bolten, A.B.S., Feinsinger, P., 1978. Why do hummingbirds flowers secretedilute nectar? Biotropica 10, 306–309.

Bolten, A.B.S., Feinsinger, P., Baker, H.G., Baker, I., 1979. On the calculationof sugar concentration in flower nectar. Oecologia 41, 301–304.

Botes, C., Johnson, S.D., Cowling, R.M., 2008. Coexistence of succulent treealoes: partitioning of bird pollinators by floral traits and floweringphenology. Oikos 117, 875–882.

Bronstein, J.L., 2001. The cost of mutualism. American Zoologist 41, 825–839.Brown, E.D., Hopkins, M.J.G., 1995. A test of pollinator specificity and

morphological convergence between nectarivorous birds and rainforest treeflowers in New Guinea. Oecologia 103, 89–100.

Brown, M., Downs, T., Johnson, D., 2008. Sugar preferences of nectar feedingbirds— a comparison of experimental techniques. Journal of Avian Biology39, 479–483.

Brown, M., Downs, C.T., Johnson, S.D., 2009. Pollination of the red-hot pokerKniphofia caulescens by short-billed opportunistic avian nectarivores. SouthAfrican Journal of Botany 75, 707–712.

Brown, M., Downs, C.T., Johnson, S.D., 2010a. Sugar preferences of ageneralist nonpasserine flower visitor, the African speckled mousebird(Colius striatus). The Auk 127, 781–786.

Fig. 5. The nectar composition. ImSa — Impatiens sakeriana, HyAr —Hypoestes aristata, LoCo — Lobelia columnaris, HyRe — Hypericumrevolutum, PyEm — Pycnostachys eminii, BrLa — Brillantaisia lamium,ViMa — Virectaria major.

72 M. Bartoš et al. / South African Journal of Botany 78 (2012) 63–74

Author's personal copy

Brown, M., Downs, C.T., Johnson, S.D., 2010b. Sugar preferences anddigestive efficiency in an opportunistic avian nectarivore, the Dark-cappedBulbul Pycnonotus tricolor. Journal of Ornithology 151, 637–643.

Brown, M., Downs, C.T., Johnson, S.D., 2010c. Concentration-dependent sugarpreferences of the malachite sunbird (Nectarinia famosa). The Auk 127,151–155.

Bruneau, A., 1997. Evolution and homology of bird pollination syndromes inErythrina (Leguminosae). American Journal of Botany 84, 54–71.

Burd, M., 1995. Pollinator behavioural responses to reward size in Lobelia deckenii:no escape from pollen limitation of seed set. Journal of Ecology 83, 865–872.

Calder, W.A., 1979. On the temperature-dependency of optimal nectarconcentrations for birds. Journal of Theoretical Biology 78, 185–196.

Chalcoff, V.R., Aizen, M.A., Galetto, L., 2006. Nectar concentration andcomposition of 26 species from the temperate forest of South America.Annals of Botany 97, 413–421.

Cheek, M., Onana, J.M., Pollard, B.P., 2000. The Plants of Mount Oku and theIjim Ridge. Cameroon, Royal Botanic Gardens Kew.

Cheke, R.A., Mann, C.F., Allen, R., 2001. Sunbirds: A Guide to the Sunbirds,Flowerpeckers, Spiderhunters and Sugarbirds of the World. ChristopherHelm, London.

De Vega, C., Herrera, C.M., Johnson, S.D., 2009. Yeasts in floral nectar of someSouth African plants: quantification and associations with pollinator type.South African Journal of Botany 75, 798–806.

Downs, C.T., 1997. Sugar digestion efficiencies of Gurney's Sugarbirds,Malachite Sunbirds, and Black Sunbirds. Physiological Zoology 70, 93–99.

Evans, M.R., 1996. Nectar and flower production of Lobelia telekiiinflorescences, and their influence on territorial behaviour of the scarlet-tufted malachite sunbird (Nectarinia johnstoni). Biological Journal ofLinnean Society 57, 89–105.

Fenster, C.B., Armbruster, W.S., Wilson, P., Dudash, M.R., Thomson, J.D.,2004. Pollination syndromes and floral specialization. Annual Review ofEcology, Evolution, and Systematics 35, 375–403.

Fleming, P.A., Nicolson, S.W., 2003. Osmoregulation in an avian nectarivore,the whitebellied sunbird Nectarinia talatala: response to extremes of dietconcentration. The Journal of Experimental Biology 206, 1845–1854.

Fleming, P.A., Gray, D.A., Nicolson, S.W., 2004a. Circadian rhythm of waterbalance and aldosterone excretion in the whitebellied sunbird Nectariniatalatala. Journal of Comparative Physiology B. 174, 341–346.

Fleming, P.A., Bakken, B.H., Lotz, C.N., Nicolson, S.W., 2004b. Concentrationand temperature effects on sugar intake and preferences in a sunbird and ahummingbird. Functional Ecology 18, 223–232.

Fleming, P.A., Xie, S., Napier, K., McWhorter, T.J., Nicolson, S.W., 2008.Nectar concentration affects sugar preferences in two Australian honeyeatersand a lorikeet. Functional Ecology 22, 599–605.

Franklin, D.C., Noske, R.A., 2000. Nectar sources used by birds inmonsoonal north-western Australia: a regional survey. Australian Journal of Botany 48, 461–474.

Galetto, L., Bernardello, G., 2004. Floral nectaries, nectar production dynamicsand chemical composition in six Ipomoea species (Convolvulaceae) inrelation to pollinators. Annals of Botany 94, 269–280.

Galetto, L., Bernardello, G., 2005. Nectar. In: Dafni, A., Kevan, P.G., Husband, B.C.(Eds.), Practical Pollination Biology. Enviroquest Ltd., Cambridge, pp. 261–313.

Goldblatt, P., Manning, J.C., 2006. Radiation of pollination systems in theIridaceae of sub-Saharan Africa. Annals of Botany 97, 317–344.

Hartman Bakken, B., McWhorter, T.J., Tsahar, E., Martinez del Rio, C., 2004.Hummingbirds arrest their kidneys at night: diel variation in glomerularfiltration rate in Selasphorus platycercus. The Journal of ExperimentalBiology 207, 4383–4391.

Hartman Bakken, B., Sabat, P., 2006. Gastrointestinal and renal responses towater intake in the green-backed firecrown (Sephanoides sephanoides), aSouth American hummingbird. The American Journal of Physiology —Regulatory, Integrative and Comparative Physiology 291, R830–R836.

Herrera, C.M., García, I.M., Pérez, R., 2008. Invisible floral larcenies: microbialcommunities degrade floral nectar of bumble bee-pollinated plants. Ecology89, 2369–2376.

Herrera, C.M., De Vega, C., Canto, A., Pozo, M.I., 2009. Yeasts in floral nectar:a quantitative survey. Annals of Botany 103, 1415–1423.

Heinrich, B., 1981. The energetics of pollination. Annals of Missouri BotanicalGarden 68, 370–378.

Hořák, D., Sedláček, O., Reif, J., Riegert, J., Pešata, M., 2010. When savannahencroaches on the forest: thresholds in bird–habitat associations in theBamenda Highlands, Cameroon. African Journal of Ecology 48, 822–827.

Jackson, S., Nicolson, S.W., Van Wyk, B.-E., 1998. Apparent absorptionefficiencies of nectar sugars in the Cape Sugarbird, with a comparison ofmethods. Physiological Zoology 71, 106–115.

Janeček, Š., Hrázský, Z., Bartoš, M., Brom, J., Reif, J., Hořák, D., Bystřická, D.,Riegert, J., Sedláček, O., Pešata, M., 2007. Importance of big pollinators forthe reproduction of two Hypericum species in Cameroon, West Africa.African Journal of Ecology 45, 607–613.

Janeček, Š., Patáčová, E., Bartoš, M., Padyšáková, E., Spitzer, L., Tropek, R.,2011. Hovering sunbirds in the Old World: occasional behaviour orevolutionary trend? Oikos 120, 178–183.

Johnson, S.D., Brown, M., 2004. Transfer of pollinaria on birds' feet: a newpollination systems in orchids. Plant Systematics and Evolution 244,181–188.

Johnson, S.D., Hargreaves, A.L., Brown, M., 2006. Dark, bitter-tasting nectarfunctions as a filter of flower visitors in a bird-pollinated plant. Ecology 87,2709–2716.

Johnson, S.D., Nicolson, S.W., 2008. Evolutionary associations between nectarproperties and specificity in bird pollination systems. Biology Letters 4,49–52.

Ley, A.C., Claßen-Bockhoff, R., 2009. Pollination syndromes in AfricanMaranthaceae. Annals of Botany 104, 41–56.

Lotz, C.N., Nicolson, S.W., 1996. Sugar preferences of a nectarivorouspasserine bird, the Lesser Double-collared Sunbird (Nectarinia chalybea).Functional Ecology 10, 360–365.

Lotz, C.N., Nicolson, S.W., 1999. Energy and water balance in the lesserdouble-collared sunbird (Nectarinia chalybea) feeding on different nectarconcentrations. Journal of Comparative Physiology B. 169, 200–206.

Martinez del Rio, C., Schondube, J.E., McWhorter, T.J., Herrera, L.G., 2001.Intake responses in nectar feeding birds: digestive and metabolic causes,osmoregulatory consequences, and coevolutionary effects. AmericanZoologist 41, 902–915.

McWhorter, T.J., Martinez del Rio, C., 1999. Food ingestion and water turnoverin hummingbirds: how much dietary water is absorbed? The Journal ofExperimental Biology 202, 2851–2858.

McWhorter, T.J., Martinez del Rio, C., Pinshow, B., 2003. Modulation ofingested water absorption by Palestine sunbirds: evidence for adaptiveregulation. The Journal of Experimental Biology 206, 659–666.

McWhorter, T.J., Martinez del Rio, C., Pinshow, B., Roxburgh, L., 2004. Renalfunction in Palestine sunbirds: elimination of excess water does notconstrain energy intake. The Journal of Experimental Biology 207,3391–3398.

Nicolson, S.W., 2002. Pollination by passerine birds: why are the nectars sodilute? Comparative Biochemistry and Physiology Part B. 131, 645–652.

Nicolson, S.W., Fleming, P.A., 2003a. Nectar as food for birds: thephysiological consequences of drinking dilute sugar solutions. PlantSystematics and Evolution 238, 139–153.

Nicolson, S.W., Fleming, P.A., 2003b. Energy balance in the WhitebelliedSunbird Nectarinia talatala: constraints on compensatory feeding, andconsumption of supplementary water. Functional Ecology 17, 3–9.

Odendaal, T.C., Brown, M., Downs, C.T., Johnson, S.D., 2010. Sugar preferencesand digestive efficiency of the village weaver: a generalist avian pollinator ofAfrican plants. The Journal of Experimental Biology 213, 2531–2535.

Plowright, R.C., 1987. Corolla depth and nectar concentration: an experimentalstudy. Canadian Journal of Botany 65, 1011–1013.

Pyke, G.H., 1981. Optimal nectar production in a hummingbird pollinated plant.Theoretical Population Biology 20, 326–343.

Pyke, G.H., Waser, N.M., 1981. The production of dilute nectars byhummingbird and honeyeater flowers. Biotropica 13, 260–270.

Riegert, J., Fainová, D., Antzak, M., Sedláček, O., Hořák, D., Reif, J., Pešata,M., 2011. Food niche differentiation in two syntopic sunbird species: acase study from the Cameroon Mountains. Journal of Ornithology 152,819–825.

Reif, J., Sedláček, O., Hořák, D., Riegert, J., Pešata, M., Hrázský, Z., Janeček,Š., 2007. Habitat preferences of birds in a montane forest mosaic in theBamenda Highlands, Cameroon. Ostrich 78, 31–36.

73M. Bartoš et al. / South African Journal of Botany 78 (2012) 63–74

Author's personal copy

Roberts, W.M., 1996. Hummingbirds' nectar concentration preferences at lowvolume: the importance of time scale. Animal Behaviour 52, 361–370.

Schmidt-Lebuhn, A.N., Schwerdtfager, M., Kessler, M., Kessler, M., 2007.Phylogenetic constraints vs. ecology in the nectar composition ofAcanthaceae. Flora 202, 62–69.

Schonube, J.E., Martinez del Rio, C., 2003. Concentration-dependent sugarpreferences in nectar-feeding birds: mechanisms and consequences.Functional Ecology 17, 445–453.

Schuchmann, K.L., 1999. Family Trochilidae (hummingbirds). In: Hoyo, del, J.,Elliot, A., Sargatal, J. (Eds.), Handbook of the Birds of the World, Vol. 5.Lynx Editions, Barcelona, pp. 468–680.

Stiles, F.G., 1978. Ecological and evolutionary implications of bird pollination.American Zoologist 18, 715–727.

Stiles, F.G., Freeman, C.E., 1993. Patterns in floral nectar characteristics ofsome bird-visited plant species from Costa Rica. Biotropica 25, 191–205.

Symes, C.T., Nicolson, S.W., 2008. Production of copious dilute nectar in thebird-pollinated African succulent Aloe marlothii (Asphodelaceae). SouthAfrican Journal of Botany 74, 598–605.

Symes, C.T., Nicolson, S.W., McKechnie, A.E., 2008. Response of aviannectarivores to the flowering of Aloe marlothii: a nectar oasis during drySouth African winters. Journal of Ornithology 149, 13–22.

Symes, C.T., Mckechnie, A.E., Nicolson, S.W., Woodborne, S.M., 2010. Thenutritional significance of a winter-flowering succulent for opportunisticavian nectarivores. Ibis 1, 110–121.

Tamm, S., Gass, C.L., 1986. Rates of nectar concentration preferences byhummingbirds. Oecologia 70, 20–23.

Torres, C., Galetto, L., 1998. Patterns and implications of floral nectar secretion,chemical composition, removal effects and standing crop in Madevillapentlandiana (Apocianaceae). Botanical Journal of the Linean Society 127,207–223.

Tropek, R., Konvicka, M., 2010. Forest eternal? Endemic butterflies of theBamenda Highlands, Cameroon, avoid close-canopy forest. African Journalof Ecology 48, 428–437.

Vázquez, D.P., Aizen, M.A., 2004. Asymmetric specialization: a pervasivefeature of plant–pollinator interactions. Ecology 85, 1251–1257.

Vos, W., Edwards, T.J., Van Staden, J., 1994. Pollination biology of annual andperennial Leonotis species (Lamiaceae). Plant Systematics and Evolution192, 1–9.

Williams, N.M., Minckley, R.L., Silveira, F.A., 2001. Variation in native beefaunas and its implications for detecting community changes. ConservationEcology 5, 7.

Wolf, L.L., Gill, F.B., 1986. Physiological and ecological adaptations of highmontane sunbirds and hummingbirds. In: Vuilleumier, F., Monasteio, M.(Eds.), High Altitude Tropical Biogeography. Oxford University Press &American Museum of Natural History, Oxford, pp. 103–119.

Wolff, D., 2006. Nectar sugar composition and volumes of 47 species ofGentianales from a southern Ecuadorian montane forest. Annals of Botany97, 767–777.

Edited by A Pauw

74 M. Bartoš et al. / South African Journal of Botany 78 (2012) 63–74