Evolutionary changes in nectar sugar composition associated with

Functional Ecology 2004 18, 670– 676

Evolutionary changes in nectar sugar composition associated with switches between bird and insect pollination: the Canarian bird-flower element revisited Blackwell Publishing, Ltd.

Y. L. DUPONT,†‡ D. M. HANSEN,*‡ J. T. RASMUSSEN§ and J. M. OLESEN‡ ‡Department of Ecology and Genetics, University of Aarhus, Ny Munkegade building 540, 8000 Aarhus C, Denmark, and §Department of Molecular Biology, University of Aarhus, Gustav Wieds Vej 10, 8000 Aarhus C, Denmark

Summary 1. The bird-flower element of the Canary Islands is a group of endemic plants having traits characteristic of bird pollination, and some are visited by opportunistically nectar-feeding passerine birds. 2. We investigated evolutionary changes in nectar sugar composition in seven Canarian lineages of ornithophilous plant species and their entomophilous relatives. 3. We hypothesized that nectar sugar composition evolved in response to the main pollinator group of a plant. Specialist nectarivores can assimilate sucrose, whereas some opportunistic nectar-feeders digest only the simple hexoses. 4. Sugar composition of nectars was analysed using high pH anion exchange chromatography. 5. Evolution of nectar type was correlated with mode of pollination. Generally, sucrose nectars were associated with insect visitation and hexose nectars with bird visitation. Nectar sugar composition was an evolutionary labile trait within a lineage. Hence, nectar characteristics may have evolved readily, perhaps in response to opportunistically nectarivorous birds living in the Canary Islands. Key-words: Canary Islands, entomophily/ornithophily, neoendemism /relictism, pollination syndrome Functional Ecology (2004) 18, 670 – 676

Introduction Pollination syndromes are groups of floral traits such as colour, morphology, scent and nectar characteristics, which are thought to be associated with certain groups of pollinators. For instance, bird-attracting flowers are typically large, showy and robust flowers or inflorescences having red, orange or bright yellow colours, no scent and copious amounts of dilute nectar (Faegri & van der Pijl 1966; Proctor, Yeo & Lack 1996). Recently, the existence of clear–cut syndromes has been challenged (Waser et al. 1996). However, a strong correlation between major pollinator classes and nectar sugar chemistry remains (Baker & Baker 1983; Baker, Baker & Hodges 1998). This association may be due to physiological constraints imposed by the digestive system of the flowervisiting animals. Specialist nectar feeders, such as bees and hummingbirds, are able to digest the composite sugar, sucrose. In contrast, opportunistically nectar-

© 2004 British Ecological Society

†Author to whom correspondence should be addressed. E-mail: [email protected] *Present address: Institute of Environmental Sciences, University of Zurich, Winterthurerstrasse 190, 8057 Zurich, Switzerland.

feeding passerine birds may be able to exploit only simple hexose sugars, such as glucose and fructose (Martinez del Rio 1990; Martinez del Rio, Baker & Baker 1992). Most nectars can be classified as either sucrose-rich or hexose-rich with few intermediates (Freeman, Worthington & Jackson 1991; Nicolson & van Wyk 1998; Nicolson & Fleming 2003). Bird pollination or ornithophily is known from a variety of plant species (e.g. Ford 1985; Proctor et al. 1996), and nectarivory or nectar-feeding is found in many bird families (Nicolson & Fleming 2003). Pollination systems are generally labile (Grant & Grant 1965; Johnson, Linder & Steiner 1998). Thus, ornithophily probably evolved repeatedly (Crisp 1994; Nicolson & Fleming 2003). However, nectar composition, in particular sugar constituents, is considered a conservative trait (Baker & Baker 1983; Baker et al. 1998; Nicolson & van Wyk 1998). Flower-visitation by birds is known from most parts of the world outside Europe and temperate Asia, from tropical to temperate latitudes and from coastal to highaltitude regions (Proctor et al. 1996). Furthermore, ornithophily is observed in many insular endemic plants (e.g. Anderson 2003). The insect fauna of remote oceanic islands is typically depauperate (Barrett 1996). 670

Table 671 1. Floral characteristics and main flower visitors of the Canarian bird-flower element and some of its close relatives

Evolution of nectar Clade sugar composition

Species

Corolla colour

Flower morphology

Flower visitorsa

Boraginaceae

Echium auberianum Webb & Berth. E. decaisnei Webb & Berth. E. pininana Webb & Berth. E. virescens DC E. wildpretii ssp. wildpretii Pearson ex Hook fil. E. wildpretii ssp. trichosiphon Svent. Canarina canariensis (L.) Vatke Campanula trachelium L. Lotus berthelotii Masf. L. campylocladus Webb & Berth. L. emeroides R. P. Murray L. eremiticus Santos L. maculatus Breitf. L. pedunculatus Cav. L. pyranthus P. Peréz Teucrium heterophyllum L’Hér. T. rotundifolium Schreb. Lavatera acerifolia Cav. L. phoenicea Vent. Digitalis parviflora Jacq. D. grandiflora Miller Isoplexis canariensis (L.) Loud. I. chalcantha Svent. & O’Shan. I. isabelliana (Webb & Berth.) Masf. Scrophularia calliantha Webb & Berth. S. smithii Hornem.

Dark blue White-light blue Light blue White-light blue Red Pink Red-orange Blue-purple Red Yellow Yellow Orange Yellow-orange Yellow Orange Red-orange White-pink Pink Pink-orange Orange-brown Pale yellow Orange Orange Orange Red-orange Dark purple

Medium funnel-shaped Medium funnel-shaped Medium funnel-shaped Small funnel-shaped Medium funnel-shaped Medium funnel-shaped Large campanulate Medium campanulate Large leguminose Small leguminose Small leguminose Large leguminose Large leguminose Small leguminose Large leguminose Medium labiate Small labiate Large actinomorphic Large actinomorphic Small tube-shaped labiate Large campanulate-labiate Large labiate Large labiate Large labiate Medium labiate Small labiate

I (1) I (1) I (1) I (1) B,I (2) I (3) B (4, 5) I (3) B? (4?, 5?) I (3) I (3) B? (4?,5?) B? (4?, 5?) I (3) B? (4?, 5?) I (3), B? (5?) I (6) I (6, 7) B,I (3, 6, 7) I (3) I (8) B,L (3, 4, 5, 6) B (4?, 5?, 9) B (4?, 5?, 9) B? (5?) I (3)

Campanulaceae Fabaceae

Lamiaceae Malvaceae Scrophulariaceae 1

Scrophulariaceae 2

a

Main flower-visitors: B = birds, I = insects, L = lizards. References of flower-visitor observations (? indicates hypothesized flower-visitor): (1) Dupont & Skov 2004, (2) Valido et al. (2002), (3) present study, (4) Olesen (1985), (5) Vogel et al. (1984), (6) E. Portellano, personal communication, (7) FuertesAguilar et al. (2002), (8) Kugler (1969), (9) A. Valido, personal communication.

© 2004 British Ecological Society, Functional Ecology, 18, 670–676

This may be the reason why nectar and fruit are often included in the diet of insect-eating vertebrates on islands (Olesen & Valido 2003). A different or supplementary explanation for the origin of bird pollination on islands is relictualism. Some insular species are thought to be remnants of once widespread species, which have gone extinct on the mainland, but survived in refuges on islands (Engler 1879). A group of plants, which is hypothesized to be relict, is the bird-flower element of the Macaronesian Islands in the Atlantic Ocean (sensu Vogel et al. 1984). Flowers of these plants have several traits characteristic of a bird pollination syndrome. The ancient pollinators of these species may have been sunbirds (Nectariniidae), an important group of pollinators in continental Africa (Vogel et al. 1984; Olesen 1985; Fuertes-Aguilar et al. 2002). However, recent molecular phylogenies do not support a prevalence of relict species on islands (Carlquist 1995). Furthermore, to date no fossil remains of Nectariniidae have been found in Macaronesia, although in general the fossil record of the islands is scanty (J. C. Rando, personal communication). In this study, we focus on the Canary archipelago, which is the centre of diversity for the Macaronesian ornithophilous species (12 of the 17 species listed by Vogel et al. 1984 are Canarian). Our aim is to trace evolutionary transitions in main pollinator type and nectar sugar composition in plant lineages comprising

the Canarian bird-flower element and some of their close entomophilous relatives. Although many workers have investigated nectar sugar composition (see References in Baker & Baker 1983; Baker et al. 1998), comparative approaches are rare (but see Bruneau 1997; Nicolson & van Wyk 1998). We address the questions: Are nectar traits evolutionarily conservative or labile? Are evolutionary switches between pollination modes associated with changes in nectar sugar composition or in flower morphology and colours? Furthermore, can nectar characteristics have evolved recently in response to opportunistically nectar-feeding birds, and/or can relictualism explain the presence of bird-flowers in the Canary Islands?

Materials and methods   The study comprised 23 plant species (including one species with two subspecies) representing seven lineages. These included 10 of 12 Canarian ornithophilous and putatively ornithophilous species (Tables 1, 2). Molecular phylogenies now exist for five of the seven lineages (Fig. 1). Although some of these phylogenies are not fully resolved, the clades including ornithophilous species and their close relatives are relatively well supported. Floral characters and the main observed,

Table 672 2. Study species and their nectar sugar composition

Y. L. Dupont et al. Clade

Speciesa

Code

Localityb

Nc

Glucose (%)

Fructose (%)

Sucrose (%)

Nectar typed

Echium

Echium auberianum E. decaisnei E. pininana

Ea Ed Ep

E. virescens E. wildpretii ssp. wildpretii E. wildpretii ssp. trichosiphon Canarina canariensis

Ev Eww Ewt Cc

Campanula trachelium Lotus berthelotii L. campylocladus L. emeroides L. eremiticus L. maculatus L. pedunculatus Teucrium heterophyllum T. rotundifolium Lavatera acerifolia L. phoenicea

Ct Lob Loc Lor Loe Lom Lop Th Tr Laa Lap

Digitalis parviflora D. grandiflora Isoplexis canariensis

Dp Dg Ica

I. chalcantha I. isabelliana

Ich Iis

Scrophularia smithii

Ss

Las Cañadas, TF La Oliva, FV Barlovento, LP Los Tilos, LP Vilaflor, TF Las Cañadas, TF Caldera Taburiente, LP Anaga, TF Greenhouse, AU Greenhouse, AU Bco. Rio, TF Chasna, TF Vallehermoso Casa Forestal, LP Genovés, TF Hammerum, DK Mazo, LP Granada, ES Bco. Guayadeque, GC El Baran, TF Chamorga, TF Greenhouse, AU Greenhouse, AU Greenhouse, AU La Esperanza, TF Los Tilos, GC Greenhouse, AU Llano de la Pez, GC Greenhouse, AU

2* 5 1* 4* 3 5 5 1 7 1 5 1* 5 5 5 2 5* 5* 5 3 3 6 2 5 8 5 5 5 5*

24·2 52·7 ± 2·1 16·5 16·1 ± 3·6 28·4 ± 2·0 45·3 ± 2·5 49·5 ± 1·8 50·3 54·3 ± 3·7 18·1 52·3 ± 0·8 14·8 31·4 ± 3·9 49·2 ± 0·8 52·4 ± 0·6 18·9 10·4 ± 2·2 12·2 ± 2·0 51·3 ± 4·3 54·1 ± 0·2 54·8 ± 0·8 3·7 ± 0·8 Trace 63·4 ± 8·7 55·7 ± 1·7 57·7 ± 1·7 56·7 ± 2·0 57·3 ± 4·5 Trace

20·7 47·3 ± 2·1 13·4 13·8 ± 3·4 23·3 ± 1·5 40·3 ± 2·4 46·9 ± 3·7 49·7 45·7 ± 3·7 22·4 47·7 ± 0·8 10·1 24·0 ± 2·4 44·7 ± 1·3 47·6 ± 0·6 19·5 10·7 ± 2·1 12·6 ± 1·9 48·7 ± 4·3 45·9 ± 0·2 45·2 ± 0·8 15·4 ± 1·6 21·8 36·6 ± 8·7 42·9 ± 2·0 42·3 ± 1·7 43·0 ± 2·5 41·6 ± 3·6 21·2

55·1 Trace 70·1 70·1 ± 6·8 48·3 ± 3·5 14·5 ± 4·8 3·6 ± 5·1 Trace ND 59·6 Trace 75·0 44·6 ± 5·9 6·2 ± 1·9 Trace 61·6 78·9 ± 4·3 75·2 ± 3·8 Trace ND ND 80·9 ± 0·9 78·2 Trace Trace Trace Trace Trace 78·5

S H S S S H H H H S H S S H H S S S H H H S S H H H H H S

Campanulaceae

Lotus

Teucrium Lavatera

Scrophulariaceae 1

Scrophulariaceae 2 a

Ornithophilous and putatively ornithophilous species are indicated in bold face. Locality or source of nectar. Abbreviations: University of Aarhus (AU), Denmark (DK), Spain (ES), Fuerteventura (FV), Gran Canaria (GC), La Palma (LP), Tenerife (TF). The latter four are islands of the Canary archipelago. c N = number of samples from the population; *indicates that nectar from several different flowers was pooled in the samples. d Nectar type: S = sucrose nectar (% sucrose = 33·3%), H = hexose nectar (% sucrose < 33·3%). ND = not detectable. b

and /or hypothesized, flower-visitors are summarized in Table 1.

  


Most nectar samples were collected in the field, although a few were collected from greenhouse specimens (Table 2). Nectar was extracted from flowers using microcapillary tubes and spotted on Whatman No. 1 filter paper. For each species, five samples of 5 µl, preferably from different individuals, were collected and analysed separately. When 5 µl could not be obtained from one flower, either a smaller sample was used, or nectars from several flowers were pooled in one sample (Table 2). Prior to analysis, the filter paper with the nectar sample was suspended for 5 –8 min in 1 ml of filtered (0·22 µm) water. Nectar sugar constituents (glucose, fructose and sucrose) were separated and quantified using high pH anion-exchange chromatography (HPAEC) combined with pulsed electrochemical-detection (PED) on a P200 Spectra-Physics chromatography system (Spectra-Physics, Fremont, CA). The mono- and

disaccharides were eluted isocratically with 160 m NaOH (1 ml min−1) from a Dionex CarboPac PA1 column and detected with a Dionex PED-2 (Dionex, Sunnyvale, CA). Each day of analysis, calibration curves of known sugar amounts were constructed using standard solutions of 0·5 mg ml−1 or 1 mg ml−1 of each of the sugars glucose, fructose and sucrose (working range 2·5–250 µg). Collected data were integrated using the BORWIN v1·21 program (JMBS Developments, Grenoble, France) and peak areas were regressed against sugar weight. If a peak was less than 1% of the summed peak area it was registered as ‘trace’. If no peak was observed it was registered as ‘not detectable’.

  For each sample, the percentages of glucose, fructose and sucrose of the total sugar weight were calculated. Based on sugar composition, nectars were assigned to one of two classes: (1) sucrose nectars when sucrose content was = 33·3% (corresponding to the categories ‘sucrose-rich’ and ‘sucrose-dominant’ of Baker & Baker

673 Evolution of nectar sugar composition

Fig. 2. Ternary plot showing sugar composition of species belonging to the Canarian bird-flower element (open symbols) and their entomophilous relatives (closed symbols). Samples were pooled within species to obtain species averages. Broken line indicates the boundary between sucrose nectars (above) and hexose nectars (below). For species codes, see Table 2. Fig. 1. Phylogeny of study species. Ornithophilous and putatively ornithophilous species are shown in bold. The tree is modified from: Echium (Böhle, Hilger & Martin 1996), Lotus (Allan et al. 2004), Lavatera (Fuertes-Aguilar et al. 2002), Isoplexis (Carvalho & Culham 1998), Canarina/Campanula (J. M. Olesen & B. K. Ehlers, unpublished data), phylogenetic relations of plant families (Stevens 2001 onwards).


1983) or (2) hexose nectars when sucrose content was < 33·3% (corresponding to ‘hexose-rich’ and ‘hexosedominant’ categories of Baker & Baker 1983). We tested for correlated evolution between the two binary characters, nectar sugar composition (hexose vs sucrose nectars) and main pollinator type (birds vs insects), using Pagel’s (1994) discrete variable method. The method applies continuous time Markov transition probabilities to model evolutionary change of the two characters along the branches of a phylogeny (Pagel 1994). Given a phylogenetic tree (Fig. 1, see below), two models are fitted to the data: (1) a null model (H0), which assumes independent changes of nectar and pollinator traits, and (2) an alternative model, which assumes correlated evolution of the two traits (H1). The log likelihoods of the two models are calculated and compared by the likelihood ratio statistic LR = −2 ln(H0 / H1 ). The statistical significance of LR was tested against a null distribution generated by Monte Carlo randomization (1000 runs). We ran two separate analyses, one including Teucrium heterophyllum among the ornithophilous species (as hypothesized by Vogel et al. 1984), and one including it in the entomophilous group (as suggested by our field observations). The general topology of the phylogenetic tree of our study species was derived from available molecular phylogenies (Fig. 1). Because branch lengths between the different lineages are unknown, all branches of the tree were set arbitrarily to 1. Polytomies in the phylogeny

were resolved to bifurcations by collapsing clades of species with the same nectar and pollination mode characters (Lotus berthelotii, L. eremiticus, L. maculatus and Isoplexis canariensis, I. chalchanta, I. isabelliana, respectively). Collapsing unresolved groups containing species with similar character combinations does not affect the outcome of the analysis substantially (Pagel 1994).

Results Nectars were clearly divided into hexose and sucrose nectars (Fig. 2, Table 2). Nectar sugar composition varied only slightly between individuals and populations within species. Furthermore, greenhouse and field samples were similar in composition (Table 2). Thus, each species had a relatively constant sugar composition, which was not strongly influenced by environmental conditions. A variety of floral forms and colours was present among our study species. Most of the (putatively) ornithophilous flowers were coloured in hues of red, orange or pink, while entomophilous relatives were white, yellow, blue, pink and brown (Table 1). Flower morphology varied more among lineages, than between ornithophilous and entomophilous species within a lineage (Table 1). Although bird-pollinated flowers were mostly larger than insect-pollinated flowers, this trend was not consistent for all lineages (Table 1). With a few exceptions (Echium decaisnei, E. wildpretii ssp. trichosiphon and Lavatera acerifolia), nectar sugar composition differed clearly between the two groups: generally, ornithophilous species had hexose nectars, and the entomophilous species had sucrose nectars (Fig. 2). The comparative analysis showed correlated

674 Y. L. Dupont et al.

evolution between nectar sugar composition and main flower-visitor type (when Teucrium heterophyllum was treated as an entomophilous species: LR = 5·94, P < 0·001; when it was treated as an ornithophilous species: LR = 3·48, P = 0·028).

Discussion     Our results confirmed that most nectars can be classified as either sucrose or hexose nectars. The association of hexose nectars with ornithophily, and sucrose nectars with entomophily, was consistent for both basal and derived species. By contrast, we did not find any consistent differences in floral size or morphology between ornithophilous species and their entomophilous relatives. This is contrary to the traditional view, that nectar sugar composition is a conservative trait, while morphology may change more rapidly in response to pollinators (van Wyk 1993; Baker et al. 1998; Perret et al. 2001; but see Bruneau 1997). Syndromes imply an evolutionary convergence of floral traits in different lineages in response to selection imposed by similar primary pollinators. The strong association of nectar and pollination type combined with evolutionary lability of nectar sugar composition may indicate that nectar phenotype can be influenced by pollinator-mediated selection. Nectar characteristics may evolve independently of floral appearance (Baker et al. 1998). Non-visible factors, such as energetic quality of the nectar, certain nectar constituents, or taste, may be important factors in the foraging choice of flower-visitors (Baker & Baker 1983). Thus, flowervisiting animals may exert stronger selection on floral reward traits than on visual or olfactory attractants.

  


Sugar is the main energy source in nectar. However, bird taxa differ in their ability to digest different types of sugars. In general, bird species that are specialized or semi-specialized for nectarivory can assimilate sucrose (e.g. hummingbirds, honeyeaters, Cape White-Eyes, sugarbirds and sunbirds, see References in Jackson, Nicolson & van Wyk 1998). Less is known about specific sugar absorption efficiencies and preferences in opportunistically nectar-feeding birds (but see Brugger 1992). Some studies indicate that insectivorous, granivorous and frugivorous species may assimilate sucrose less efficiently than hexoses, and thus prefer the latter (Martinez del Rio et al. 1988; Brugger 1992; Malcarney, Martinez del Rio & Apanius 1994). The ability to digest sucrose appears to have a strong phylogenetic component (Martinez del Rio et al. 1992; Nicolson & Fleming 2003). However, physiological traits of the digestive system and feeding choices have been investigated in only a limited number of species (reviewed in Jackson et al. 1998).

       Phylloscopus collybita Vieillot, Parus caeruleus L., Serinus canarius L., Sylvia melanocephala Gmelin, S. conspicillata Temminck and S. atricapilla L. are frequent nectar-feeding bird species in the Canary Islands (Vogel et al. 1984; Olesen 1985; Valido, Dupont & Hansen 2002). These are generalist passerine species, which consume arthropods, seeds and fruits in addition to nectar (Martín & Lorenzo 2001). The passerines have been observed visiting many plant species belonging to the Macaronesian bird-flower element and a number of exotic garden plants (Vogel et al. 1984; Olesen 1985; Trujillo 1992; Valido et al. 2002). Furthermore, bird visits have been observed on some occasions in the mainly entomophilous Echium decaisnei (Trujillo 1992; A. Valido, personal communication), which had hexose-dominated nectar. Thus, the Canarian generalist nectarivorous bird species appear to prefer hexose to sucrose nectars. Additional field observations may reveal bird visits to the other two deviating species, the ‘entomophilous’ hexose-dominated Echium wildpretii ssp. trichosiphon and Lavatera acerifolia. The distribution ranges of these bird species (except S. canarius) extend to continental Europe, where they are mainly insectivorous or granivorous (Bannerman 1965). Nectarivory is occasionally observed in mainland habitats, but is generally considered a rare behaviour (Ford 1985; Westerkamp 1996). This leads to the questions: Why are these widespread bird species frequent nectar-feeders in island habitats? Furthermore, are these generalist birds likely to impose strong enough selection to drive the evolution of nectar traits in the Canarian plants?

    Low species densities on islands may cause relaxation of interspecific competition, which in turn may lead to niche expansion and a subsequent increase in abundance of resident species (MacArthur 1972). Such density compensation appears to have occurred in Phylloscopus collybita and Parus caeruleus in laurel and pine forests of the Canary Islands (A. Valido, unpublished data). Furthermore, P. collybita occupies a larger range of habitats in the Canary Islands compared with continental areas (Volsøe 1951). Nectar-feeding may be a result of niche broadening associated with density compensation and a general scarcity of insects on islands. A similar evolutionary scenario has been suggested to account for nectarivory in lizards on islands (Olesen & Valido 2003). The biotic selection regime of islands is different from that of the mainland. In mainland areas, a plant species may be subject to conflicting selection imposed by many different pollinator species. On islands, fewer, but in some cases more abundant, pollinator species are found (Barrett 1996). Thus, in the island scenario, flowers may evolve

675 Evolution of nectar sugar composition

directionally in response to opportunistically flowervisiting birds, even though these pollinators are not thought to drive evolution in mainland areas. Indeed, birds have been suggested to be more reliable pollen vectors in systems where insect activity is low, such as in high-altitude ecosystems (e.g. Cruden 1972), among winter-flowering species (e.g. Castro & Robertson 1997) and on islands (e.g. Feinsinger, Wolfe & Swarm 1982). The occurrence of typical ornithophilous flowers in the native flora of the Canary Islands has been taken as prima facie evidence that these plant species were once pollinated by specialist flower-visiting birds (Vogel et al. 1984; Olesen 1985). A former presence of Nectariniidae in the Canary Islands cannot be excluded, considering the incomplete nature of the fossil record in general and small birds in particular (J. C. Rando, personal communication). However, bird pollination is found in both basal and derived plant groups. Thus, ornithophily in itself is not an ancestral mode of pollination in the Canarian bird-flower element. The general dominance of hexose-nectars among the Canarian ornithophilous species does not preclude ancient pollination by sunbirds, because extant African sunbirds feed on both sucrose and hexose nectars (Nicolson & Fleming 2003). More interesting, however, is the apparent common evolutionary shifts between sucrose and hexose nectars in many of the Canarian plant groups, which include ornithophilous species. Nectar sugar chemistry in these lineages seems to have evolved readily, perhaps even in response to pollination by non-specialist nectarivorous birds.

Acknowledgements We thank Alfredo Valido and Félix Medina for logistic support, Elena Portellano, Alfredo Valido and Javier Fuertes-Aguilar for sending us nectar samples, José García Casanova, Ángel Bañares, Félix Medina and Ángel Palomares for research permissions, and Lone T. Pallesen for help in the laboratory. Furthermore, we acknowledge two referees for constructive comments. This study received financial support from Mr and Mrs Fiedler’s grant (YLD), Augustinus Foundation (YLD), Gad’s Foundation (YLD), Julie von Müllens Foundation (YLD), Villum Kann Rasmussen’s Foundation (YLD), Carlsberg Foundation (JMO), and the Danish National Science Research Council (JMO).

References


Allan, G.J., Francisco-Ortega, J., Santos-Guerra, A., Boerner, E. & Zimmer, E.A. (2004) Molecular phylogenetic evidence for the geographic origin and classification of Canary Island Lotus (Fabaceae: Loteae). Molecular Phylogenetics and Evolution 32, 123–138. Anderson, S.H. (2003) The relative importance of birds and insects as pollinators of the New Zealand flora. New Zealand Journal of Ecology 27, 83–94. Baker, H.G. & Baker, I. (1983) Floral nectar sugar constituents in relation to pollinator type. Handbook of Experimental

Pollination Ecology (eds C.E. Jones & R.L. Little), pp. 117– 141. Van Nostrand Reinholdt, New York. Baker, H.G., Baker, I. & Hodges, S.A. (1998) Sugar composition of nectars and fruits consumed by birds and bats in the tropics and subtropics. Biotropica 30, 559–586. Bannerman, D.A. (1965) Birds of the Atlantic Islands, Vol. 2. Oliver and Boyd, Edinburgh. Barrett, S.C.H. (1996) The reproductive biology and genetics of island plants. Philosophical Transactions of the Royal Society of London B 351, 725–733. Böhle, U.-R., Hilger, H.H. & Martin, W.F. (1996) Island colonization and evolution of the insular woody habit in Echium L. (Boraginaceae). Proceedings of the National Academy of Sciences of the USA 93, 11740–11745. Brugger, K.E. (1992) Repellency of sucrose to captive American robins. Journal of Wildlife Management 56, 794– 799. Bruneau, A. (1997) Evolution and homology of bird pollination syndromes in Erythrina (Leguminosae). American Journal of Botany 84, 54–71. Carlquist, S. (1995) Introduction. Hawaiian Biogeography. Evolution on a Hot Spot Archipelago (eds W.L. Warren & V.A. Funk), pp. 1–13. Smithonian Institution Press, Washington & London. Carvalho, J.A. & Culham, A. (1998) Conservation status and preliminar results on the phylogenetics of Isoplexis (Lindl.) Benth. (Scrophulariaceae) an endemic Macaronesian genus. Boletim do Museu Municipal do Funchal Sup. 5, 109–127. Castro, I. & Robertson, A.W. (1997) Honeyeaters and the New Zealand forest flora: the utilisation and profitability of small flowers. New Zealand Journal of Ecology 21, 169–179. Crisp, M.D. (1994) Evolution of bird-pollination in some Australian legumes (Fabaceae). Phylogenetics and Ecology (eds P. Eggleton & R.I. Vane-Wright), pp. 281–309. Academic Press, London. Cruden, R.W. (1972) Pollinators in high-elevation ecosystems: relative effectiveness of birds and bees. Science 176, 1439–1440. Dupont, Y.L. & Skov, C. (2004) Influence of geographical distribution and floral traits on species richness of bees (Hymenoptera: Apoidea) visiting Echium species (Boraginaceae) in the Canary Islands. International Journal of Plant Sciences 165, 377–386. Engler, A. (1879) Versuch einer Entwicklungsgeschichte der Pflanzenwelt, insbesondere der Florengebiete seit der Tertiärperiode. 1. Die Extratropischen Gebiete der Nördlichen Hemisphäre. W. Engelmann, Leipzig. Faegri, K. & van der Pijl, L. (1966) The Principles of Pollination Ecology. Pergamon Press, Oxford. Feinsinger, P., Wolfe, J.A. & Swarm, L.A. (1982) Island ecology: reduced hummingbird diversity and the pollination biology of plants, Trinidad and Tobago, West Indies. Ecology 63, 494 –506. Ford, H.A. (1985) Nectarivory and pollination by birds in southern Australia and Europe. Oikos 44, 127–131. Freeman, C.E., Worthington, R.D. & Jackson, M.S. (1991) Floral nectar sugar compositions of some South and Southeast Asian species. Biotropica 23, 568–574. Fuertes-Aguilar, J., Ray, M.F., Francisco-Ortega, J., SantosGuerra, A. & Jansen, R.K. (2002) Molecular evidence from chloroplast and nuclear markers for multiple colonizations of Lavatera (Malvaceae) in the Canary Islands. Systematic Botany 27, 74–83. Grant, V. & Grant, K.A. (1965) Flower Pollination in the Phlox Family. Columbia University Press, New York & London. Jackson, S., Nicolson, S.W. & van Wyk, B.-E. (1998) Apparent absorption efficiencies of nectar sugars in the Cape sugarbird, with a comparison of methods. Physiological Zoology 71, 106–115.

676 Y. L. Dupont et al.


Johnson, S.D., Linder, H.P. & Steiner, K.E. (1998) Phylogeny and radiation of pollination systems in Disa (Orchidaceae). American Journal of Botany 85, 420 – 411. Kugler, H. (1969) Blütenökologie. Gustav Fischer Verlag, Stuttgart. MacArthur, R. (1972) Density compensation in island faunas. Ecology 53, 330–342. Malcarney, H.L., Martinez del Rio, C. & Apanius, V. (1994) Sucrose intolerance in birds: simple nonlethal diagnostic methods and consequences for assimilation of complex carbohydrates. Auk 111, 170–177. Martín, A. & Lorenzo, J.A. (2001) Aves del Archipiélago Canario. Francisco Lemus, La Laguna, Tenerife. Martinez del Rio, C. (1990) Dietary, phylogenetic, and ecological correlates of intestinal sucrase and maltase activity in birds. Physiological Zoology 63, 987–1011. Martinez del Rio, C., Stevens, B.R., Daneke, D.E. & Andreadis, P.T. (1988) Physiological correlates of preference and aversion for sugars in three species of birds. Physiological Zoology 61, 222–229. Martinez del Rio, C., Baker, H.G. & Baker, I. (1992) Ecological and evolutionary implications of digestive processes: bird preferences and the sugar constituents of floral nectar and fruit pulp. Experientia 48, 544 –551. Nicolson, S.W. & Fleming, P.A. (2003) Nectar as food for birds: the physiological consequences of drinking dilute sugar solutions. Plant Systematics and Evolution 238, 139– 153. Nicolson, S.W. & van Wyk, B.-E. (1998) Nectar sugars in Proteaceae: patterns and processes. Australian Journal of Botany 46, 489–504. Olesen, J.M. (1985) The Macaronesian bird-flower element and its relation to bird and bee opportunists. Botanical Journal of the Linnean Society 91, 395– 414. Olesen, J.M. & Valido, A. (2003) Lizards as pollinators and seed dispersers: an island phenomenon. Trends in Ecology and Evolution 18, 177–181.

Pagel, M. (1994) Detecting correlated evolution on phylogenies: a general method for the comparative analysis of discrete characters. Proceedings of the Royal Society of London B 255, 37– 45. Perret, M., Chautems, A., Spichiger, R., Peixoto, M. & Savolainen, V. (2001) Nectar sugar composition in relation to pollination syndromes in Sinningieae (Gesneriaceae). Annals of Botany 87, 267–273. Proctor, M., Yeo, P. & Lack, A. (1996) The Natural History of Pollination. Harper Collins Publishers, London. Stevens, P.F. (2001–) Angiosperm Phylogeny Website, Version 4, May 2003. http://www.mobot.org / MOBOT/research/ APweb/. Trujillo, O. (1992) Los Sílvidos en Gran Canaria. Contribución al estudio de la avifauna canaria. Excmo. Cabildo Insular de Gran Canaria, Las Palmas de Gran Canaria. Valido, A., Dupont, Y.L. & Hansen, D.M. (2002) Native birds and insects, and introduced honey bees visiting Echium wildpretii (Boraginaceae) in the Canary Islands. Acta Oecologica 23, 413– 419. Vogel, S., Westerkamp, C., Theil, B. & Gessner, K. (1984) Ornithophilie auf den Canarischen Inseln. Plant Systematics and Evolution 146, 225–248. Volsøe, H. (1951) The breeding birds of the Canary Islands 1. Introduction and synopsis of the species. Videnskabelige Meddelelser Fra Dansk Naturhistorisk Forening 113, 1–153. Waser, N.M., Chittka, L., Price, M.V., Williams, N.M. & Ollerton, J. (1996) Generalization in pollination systems, and why it matters. Ecology 77, 1043 –1060. Westerkamp, C. (1996) Heimische Blumenvögel. Palmengarten 60, 17–24. van Wyk, B.-E. (1993) Nectar sugar composition in southern African Papilionoideae (Fabaceae). Biochemical Systematics and Ecology 21, 271–277. Received 9 December 2003; revised 18 March 2004; accepted 30 March 2004

Evolutionary changes in nectar sugar composition associated with

Recommend Documents