143

SHORT COMMUNICATIONS

ORNITOLOGIA NEOTROPICAL 23: 143–149, 2012© The Neotropical Ornithological Society

FLORAL TRAITS OF PLANTS VISITED BY THE BEE HUMMINGBIRD (MELLISUGA HELENAE)

Bo Dalsgaard1,2, Daniel W. Carstensen3, Arturo Kirkconnell4, Ana M. Martín González1,5, Orestes Martínez García6, Allan Timmermann7, & William J. Sutherland2

1Center for Macroecology, Evolution and Climate, Department of Biology, University of Copenhagen, Universitetsparken 15, DK-2100 Copenhagen, Denmark. E-mail:

bo.dalsgaard@bio.ku.dk2Conservation Science Group, Department of Zoology, University of Cambridge, Downing

Street, Cambridge CB2 3EJ, United Kingdom. 3Ecology and Genetics, Department of Bioscience, Aarhus University, Ny Munkegade 114,

DK-8000 Aarhus, Denmark.4Museo Nacional de Historia Natural de Cuba, Obispo 61, Habana Vieja, Cuba.

5Pacific Ecoinformatics and Computational Ecology Lab, 1604 McGee Avenue, Berkeley, CA 94703, USA.

6Organo CITMA Ciénaga de Zapata, Carretera Playa Larga Km 26, Matanzas, Cuba.7Ecoinformatics and Biodiversity, Department of Bioscience, Aarhus University, Ny

Munkegade 116, DK-8000 Aarhus C, Denmark.

Características florales de las plantas visitadas por el zunzuncito (Mellisuga helenae).

Key words: Bee Hummingbird, Trochilidae, conservation, hummingbird-plant interactions, seasonality,Caribbean, Cuba, West Indies.

INTRODUCTION

Hummingbirds and their food-plants rely to alarge extent on each other for food supplyand pollination service, respectively. Thismutual relationship has co-evolved for mil-lions of years and across the Americas involv-ing over 330 hummingbird species and anestimated nearly 8000 plant species relying onhummingbirds as their principle pollinators

(Feinsinger & Colwell 1978, Schuchmann1999, Nicolson & Fleming 2003, Dalsgaard etal. 2011). Although birds other than hum-mingbirds also act as pollinators (Feinsinger& Colwell 1978, Rocca & Sazima 2008, Dals-gaard 2011, Kissling et al. 2011), humming-bird-plant interactions are one of the moststriking examples of bird-plant co-evolution,and the associated morphological adapta-tions have been studied intensively (e.g., Snow

144

DALSGAARD ET AL.

& Snow 1972, Brown et al. 1978, Feinsinger &Colwell 1978, Stiles 1981, Feinsinger et al.1982, Arizmendi & Ornelas 1990, Cotton1998, Buzato et al. 2000, Temeles & Kress2003, Lara 2006, Dalsgaard et al. 2009, Abra-hamczyk & Kessler 2010). These studies havefound that plants adapted to hummingbird-pollination typically are odorless, have a longand narrow conspicuous orange-red tubularcorolla, and produce extensive dilute nectar -broadly matching the sensory capabilities,morphology and energetic demands of hum-mingbirds (Temeles & Kress 2003, Dalsgaardet al. 2009). Similarly, hummingbirds show anarray of features considered nectar-feedingadaptations, such as the capability to hover, aspecialized tongue, highly specialized andvariable bill morphologies, as well as a minutebody size (Brown et al. 1978).

The smallest of all birds is the Bee Hum-mingbird (Mellisuga helenae) endemic to theCuban archipelago, including the main islandof Cuba and Isle of Youth (formerly Isle ofPines), in the West Indies (AOU 1998, Gar-rido & Kirkconnell 2000). Of the 15 hum-mingbird species endemic to the West Indies(Lack 1973, 1976), the Bee Hummingbird isthe only species considered threatened on theIUCN red list (BirdLife International 2011).Although the Bee Hummingbird is the small-est bird in the World, is categorized as NearThreatened and its population size is clearlydeclining, only very basic information regard-ing its distribution and reproduction has pre-viously been reported (Martínez et al. 1998,Schuchmann 1999, Garrido & Kirkconnell2000). Almost nothing is known about itsfeeding ecology. For instance, apart from therecord of some 15 food plants, nothing isreported about its preferred food plants orthe overlap in flower use with the larger-sizedCuban Emerald (Chlorostilbon ricordii) - theonly other resident Cuban hummingbird(Martínez et al. 1998, Schuchmann 1999,Garrido & Kirkconnell 2000).

We here: 1) identify and describe the floraltraits of several food plants the Bee Hum-mingbird uses; 2) report to what extent theCuban Emerald uses the same flowers as theBee Hummingbird; and 3) compare our dataon flower use with those previously reported,and discuss how our results could directfuture studies beneficial for the conservationof the Bee Hummingbird.

METHODS AND STUDY SITE

Food plants. Our study was conducted nearPlaya Larga (22°17N, 81°12W) in the out-skirts of the Zapata Swamp in Cuba, WestIndies - one of the few remaining core areasfor the Bee Hummingbird (Garrido & Kirk-connell 2000). We focused our fieldwork onmeasuring the floral traits (see below) ofplants known to be visited by the Bee Hum-mingbird. All the included food plants werelocated in the field by OMG and AM (seeacknowledgments), both of whom have over30 years of experience as bird guides locatingthe Bee Hummingbird and its flowers in theZapata Swamp, and have studied the BeeHummingbirds reproduction and use of foodplants in the Zapata Swamp (Martínez et al.1998). We only included plants that OMGand AM knew to be used by the Bee Hum-mingbird and only at localities they knew well,i.e., we were certain that the individual plantsstudied had been visited by the Bee Hum-mingbird (which coincide with Kirkconnell’sfield data). The plants were later identified byRamona Oviedo, a botanist from the Instituteof Ecology and Systematic, Cuba. Althoughduring the fieldwork period we observed theBee Hummingbird visiting some of the iden-tified plants (Pavonia paludicola, Lysiloma lati-siliquum, and Cissus obovata), we did notattempt to estimate how the relative visitationrate varied among plant species. We alsoreport whether the plants were visited by theCuban Emerald; this information was likewise

145

SHORT COMMUNICATIONS

based on the accumulated knowledge byOMG and AM, although during fieldwork wealso observed the Cuban Emerald visitingseveral of the identified plants (Forsteroniacorymbosa, Pavonia paludicola, Lysiloma latisiliq-uum, Turnera ulmifolia, Hamelia patens, and Cissusobovata).

Floral traits. We measured six quantitative andsemi-quantitative floral traits, which havebeen shown previously to be important instructuring West Indian hummingbird-plantcommunities into traditional ornithophilousflowers and more generalistic entomophiloussyndrome flowers (Dalsgaard et al. 2009). Thesix floral traits can be divided into: two“restriction” traits (corolla depth, corollawidth); two “reward” traits (nectar volume,nectar concentration); and two “attraction”traits (color, odor). Corolla depth was mea-sured as the internal distance from the open-ing to the base of the flower. When the nectarwas not protected by a tubular corolla, wemeasured the distance from the start of thefloral restriction to the base of the flower.Likewise, corolla width was measured as theinternal width of the opening of the flower.Both restriction traits were measured with adigital caliper to nearest mm. Nectar volumewas measured using 1, 5, and 10 µl micropi-pettes, while nectar concentration (expressedas % sucrose equivalents) was measured witha refractometer and corrected for tempera-ture. The floral color of each species wasassigned as perceived by the human eye.Where the floral bracts/stems were a majorpart of the color display, as in C. obovata, theywere included in the assignment. Finally, theflower of each species was coded according totheir odor, with (1) representing no odor; (2):little; (3): moderate; and (4): strong odor. Allfloral trait estimates were obtained during 15–27 July 2010 (i.e., in the rainy season outsideof the Bee Hummingbirds breeding season),and are means of 8–15 flowers measured on

un-bagged and fresh flowers sampled withinthe first hour of sunrise. Hence, nectar vol-ume estimates express standing crop availableto hummingbirds and other visitors in theearly morning and not nectar production per se(Cotton 1998, Dalsgaard et al. 2009).

RESULTS AND DISCUSSION

Each of the identified ten plant species visitedby the Bee Hummingbird belongs to differentplant families, of which one species (Antigononleptopus) was introduced to Cuba. The remain-ing nine plant species are either native orendemic to Cuba (Table 1). Most of the tenflowers visited by the Bee Hummingbird - ofwhich nine were also visited by the CubanEmerald, as well as insects, such as bees andbutterflies (pers. observ.) - have floral traitsthat fall outside the traditional ornithophiloussyndrome, noticeably most flowers had littlenectar and a short, white corolla, althoughsome had more vivid colors (Table 1, Fig. 1).This indicates that the Bee Hummingbirduses plants with generalized pollination sys-tems as has also been reported for plants vis-ited by small hummingbirds on Trinidad,Tobago, and other islands in the West Indies(Feinsinger et al. 1982, Dalsgaard et al. 2008,2009).

Conclusions. A total of 15 plant species from 15different genera had previously been reportedto be visited by the Bee Hummingbird, but nostudy had described their floral traits (Mar-tínez et al. 1998, Schuchmann 1999). Of theherein reported ten food plants, only one spe-cies (Hamelia patens) occurs in the 15 generapreviously listed as being visited by the Beehummingbird. Many of the previouslyreported food plants in the Zapata Swampwere observed in March–June during thebreeding season of the Bee Hummingbird(Martínez et al. 1998), whereas the food plantswe here report bloom in the latter part of July,

146

DALSGAARD ET AL.Pl

ant f

amily

Plan

t spe

cies

Plan

t1 or

igin

NCo

rolla

de

pth(

mm

)Co

rolla

w

idth

(mm

)N

ecta

r vo

lum

e (m

l)N

ecta

r co

nc. (

%)

Colo

rO

dor2

Apo

cyna

ceae

Bora

gina

ceae

Ch

ryso

balan

acea

eM

alvac

eae

Mim

osac

eae

Pass

iflor

acea

ePo

lygo

nace

aeRu

biac

eae

Verb

enac

eae

Vita

ceae

Forst

eroni

a cor

ymbo

sa

Tour

nefor

tia hi

rsutis

sima

Chry

sobala

nus i

caco

Pavo

nia

palud

icola

Lysil

oma

latisi

liquu

m Tu

rnera

ulm

ifolia

A

ntigo

non

leptop

us

Ham

elia p

aten

s Cl

erode

ndru

m ac

uleatu

m Ci

ssus o

bova

ta

1 2 2 2 2 2 3 2 2 1

8 15 12 11 9 11 11 8 11 9

2.62

± 0

.44

SD6.

79 ±

0.5

4 SD

4.87

± 0

.44

SD16

.83

± 1

.64

SD1.

03 ±

0.3

4 SD

8.88

± 0

.80

SD3.

29 ±

0.3

8 SD

21.3

3 ±

1.2

7 SD

18.2

0 ±

1.9

7 SD

0.00

± 0

.00

SD

1.97

± 0

.13

SD1.

09 ±

0.2

6 SD

1.81

± 0

.30

SD6.

52 ±

0.9

0 SD

1.12

± 0

.19

SD0.

84 ±

0.1

5 SD

1.73

± 0

.21

SD3.

47 ±

0.4

7 SD

1.53

± 0

.14

SD1.

48 ±

0.1

5 SD

0.48

± 0

.31

SD0.

25 ±

0.1

0 SD

0.82

± 0

.93

SD29

.61

± 2

4.37

SD

0.51

± 0

.41

SD1.

26 ±

1.1

4 SD

1.25

± 0

.66

SD5.

41 ±

5.3

0 SD

0.37

± 0

.30

SD0.

50 ±

0.3

4 SD

40.0

3 ±

5.2

1 SD

15.8

5 ±

2.6

7 SD

27

.19

± 9

.41

SD16

.64

± 2

.03

SD35

.89

± 2

2.14

SD

26.8

9 ±

2.5

1 SD

16.8

8 ±

2.6

5 SD

25.9

1 ±

0.8

9 SD

22.9

1 ±

2.8

8 SD

22.3

5 ±

5.5

5 SD

red

whi

tew

hite

yello

w-g

reen

whi

teye

llow

pink

oran

ge-r

edw

hite

oran

ge-r

ed

1 3 1 2 3 2 2 1 1 1

TABL

E 1

. Flo

ral t

raits

of t

en p

lant s

pecie

s vi

sited

by

the

Bee

Hum

min

gbird

(Mell

isuga

hele

nae).

The

Cub

an E

mer

ald (C

hloro

stilbo

n ric

ordii)

visi

ted

all s

peci

esex

cept

Chr

ysoba

lanus

icac

o. 1 Pl

ant o

rigin

: 1 =

end

emic

; 2 =

nat

ive;

3 =

intro

duce

d. 2 O

dor:

code

d as

1–4

: 1=

no

odor

; 2 =

littl

e; 3

= m

oder

ate;

and

4 =

stro

ngod

or.

147

SHORT COMMUNICATIONS

FIG. 1. The flowers of ten plant species visited by the Bee Hummingbird: a) Forsteronia corymbosa (Apocy-naceae); B) Tournefortia hirsutissima (Boraginaceae); C) Chrysobalanus icaco (Chrysobalanaceae); D) Pavonia palu-dicola (Malvaceae); E) Lysiloma latisiliquum (Mimosaceae); F) Turnera ulmifolia (Passifloraceae); G) Antigononleptopus (Polygonaceae); H) Hamelia patens (Rubiaceae); I) Clerodendrum aculeatum (Verbenaceae); J) Cissus ob-ovata (Vitaceae). Note that the photos of each flower have a different scale. For the actual size of eachflower, see Table 1 (photos: Bo Dalsgaard).

148

DALSGAARD ET AL.

outside the breeding season of the Bee Hum-mingbird. In Cuba, the dry season extendsfrom November to April, whereas the wettestmonths are from June to October (Garrido &Kirkconnell 2000). Therefore, it seems likelythat the flower community largely changesbetween the dry and the rainy season, roughlycoinciding with the termination of the BeeHummingbirds breeding season.

This study is a first step in identifying thefloral niche of the Bee Hummingbird. We rec-ommend that a quantitative field study beundertaken to evaluate the floral overlap inthe context of floral use and availability, andbehavioral interactions with the CubanEmerald (Vaurie 1957) and insect-pollinators,such as the potential negative impact of thenon-native honeybee (Apis mellifera). Compar-ing the herein reported food plants of the BeeHummingbird with the plant species reportedby Martínez et al. (1998), indicates that itwould be essential to examine both dry sea-son and rainy season floral overlap andbehavior. It would also be interesting to studythe effect of introduced plants, forest loss anddegradation on the feeding behavior of theBee Hummingbird (BirdLife International2011). Finally, we recommend that field stud-ies are also conducted in other areas than theZapata Swamp, e.g., in the Bee Humming-birds strongholds in the lowlying Guanaha-cabibes peninsula in western Cuba and in theeastern mountains of Cuchillas del Toa andSierra Cristal. This should prove valuable forthe conservation of the Bee hummingbird, anendemic and threatened species of the Cubanarchipelago, West Indies.

ACKNOWLEDGMENTS

Thanks to Ramona Oviedo for identifying theplants, to Angel Martínez and Lazara ReinaRodriquez Gazmury for help in the field, andto Joseph M. Wunderle, André Weller, Gary F.Stiles, and an anonymous reviewer for provid-

ing useful comments improving the manu-script. BD was supported by The CarlsbergFoundation, The Danish Research Council |Natural Sciences and Weis-Fogh Fund atDepartment of Zoology, University of Cam-bridge, DWC by The Faculty of Science andTechnology at Aarhus University, AMMG bythe Spanish Ministry of Education, and WJSby Arcadia.

REFERENCES

Abrahamczyk, S., & M. Kessler. 2010. Humming-bird diversity, food niche characters, andassemblage composition along a latitudinal pre-cipitation gradient in the Bolivian lowlands. J.Ornithol. 151: 615–625.

AOU. 1998. Check-list of North American Birds.7th ed. American Ornithologists Union, Wash-ington, D.C., USA.

Arizmendi, M. D. C., & J. F. Ornelas. 1990. Hum-mingbirds and their floral resources in a tropi-cal dry forest in Mexico. Biotropica 22: 172–180.

BirdLife International. 2011. IUCN Red List forbirds. Downloaded on 2 August 2011 fromhttp://www.birdlife.org.

Bleiweiss, R. 1998. Origin of hummingbird faunas.Biol. J. Linn. Soc. 65: 77–97.

Brown, J. H., W. A. III Calder, & A. Kodric-Brown.1978. Correlates and consequences of bodysize in nectar-feeding birds. Am. Zool. 18: 687–700.

Brown, J. H., & M. A. Bowers. 1985. Communityorganization in hummingbirds: relationshipbetween morphology and ecology. Auk 102:251–269.

Buzato, S., M. Sazima, & I. Sazima. 2000. Hum-mingbird-pollinated floras at three Atlantic for-est sites. Biotropica 32: 824–841.

Cotton, P. A. 1998. Coevolution in an Amazonianhummingbird-plant community. Ibis 140: 639–646.

Dalsgaard, B., A. M. Martín González, J. M. Ole-sen, A. Timmermann, L. H. Andersen, & J.Ollerton. 2008. Pollination networks and func-tional specialization: a test using Lesser Anti-llean plant-hummingbird assemblages. Oikos

149

SHORT COMMUNICATIONS

117: 789–793. Dalsgaard, B., A. M. Martín González, J. M. Ole-

sen, J. Ollerton, A. Timmermann, L. H. Ander-sen, & A. G. Tossas. 2009. Plant-hummingbirdinteractions in the West Indies: floral specialisa-tion gradients associated with environment andhummingbird size. Oecologia 159: 757–766.

Dalsgaard, B. 2011. Nectar-feeding and pollinationby the Cuban Green Woodpecker (Xiphidiopicuspercussus) in the West Indies. Ornitol. Neotrop.22: 447–451.

Dalsgaard, B., E. Magård, J. Fjeldså, A. M. MartínGonzález, C. Rahbek, J. M. Olesen, J. Ollerton,R. Alarcón, A. C. Araujo, P. A. Cotton, C. Lara,C. G. Machado, I. Sazima, M. Sazima, A. Tim-mermann, S. Watts, B. Sandel, W. J. Sutherland,& J.-C. Svenning. 2011. Specialization in plant-hummingbird networks is associated with spe-cies richness, contemporary precipitation andQuaternary climate-change velocity. PLoSONE 6: e25891.

Feinsinger, P., & R. K. Colwell. 1978. Communityorganization among neotropical nectar-feedingbirds. Am. Zool. 18: 779–796.

Feinsinger, P., J. A. Wolfe, & L. A. Swarm. 1982.Island ecology: reduced hummingbird diversityand the pollination biology of plants, Trinidadand Tobago, West Indies. Ecology 63: 494–506.

Garrido, O. H., & A. Kirkconnell. 2000. Field guideto the birds of Cuba. Cornell Univ. Press, NewYork, New York, USA.

Kissling, D. W., C. H. Sekercioglu, & W. Jetz. 2011.Bird dietary guild richness across latitudes,environments and biogeographic regions. Glo-bal Ecol. Biogeogr. 21: 328–340.

Lack, D. 1973. The number of species of hum-mingbirds in the West Indies. Evolution 27:326–337.

Lack, D. 1976. Island biology illustrated by the landbirds of Jamaica. Blackwell, Oxford, UK.

Lara, C. 2006. Temporal dynamics of flower use byhummingbirds in a highland temperate forest inMexico. Ecoscience 13: 23–29.

Martínez García, O., L. Bacallao Mesa, & E. NievesLorenzo. 1998. Estudio preliminary de la con-ducta reproductive de Mellisuga helenae (Aves,Apodiformes) en condiciones naturals. ElPitirre 11: 102–106.

Nicolson, S. W., & P. A. Fleming. 2003. Nectar asfood for birds: the physiological consequencesof drinking dilute sugar solutions. Plant Syst.Evol. 238: 139–153.

Rocca, M. A., & M. Sazima. 2008. Ornithophilouscanopy species in the Atlantic rain forest ofsoutheastern Brazil. J. Field Ornithol. 79: 130–137.

Schuchmann, K. L. 1999. Family Trochilidae(Hummingbirds). Pp. 468–680 in del Hoyo, J.,A. Elliott, & J. Sargatal (eds). Handbook of thebirds of the world. Volume 5: Barn-owls tohummingbirds. Lynx Edicions, Barcelona,Spain.

Snow, B. K., & D. W. Snow. 1972. Feeding niches ofhummingbirds in a Trinidad valley. J. Anim.Ecol. 41: 471–485.

Stiles, F. G. 1981. Geographical aspects of bird-flower coevolution, with particular reference toCentral America. Ann. Mo. Bot. Gard. 68: 323–351.

Temeles, E. J., & J. Kress. 2003. Adaptation in aplant-hummingbird association. Science 300:630–633.

Vaurie, C. 1957. Field notes on some Cuban birds.Wilson Bull. 69: 301–313.

Accepted 14 March 2012.