Download relationships between morphology and ecology

COMMUNITY
ORGANIZATION
RELATIONSHIPS
IN HUMMINGBIRDS:
BETWEEN
AND
MORPHOLOGY
ECOLOGY
JAMESH. BROWNAND MICHAEL A. BOWERS
Departmentof EcologyandEvolutionary
Biology,Universityof Arizona,
Tucson, Arizona 85721 USA
AssTRACT.--Hummingbirdspeciesinhabiting restrictedgeographicregions exhibit morphologicalpatternsthat differ significantlyfrom thosepredictedby null modelsin which
speciesare selectedat random from appropriatespeciespools.TemperateNorth American
hummingbirdsare convergent:more similar in bill length, body weight, and wing length
than predictedby severalnull models.Thesetemperatespeciesalsoare more similarto each
otherthantheyare to morecloselyrelated(congeneric)
speciesfromsubtropical
andtropical
habitats.Hummingbirds of the Greater and LesserAntilles show a nonrandomdistribution
of speciesamong islands:all islandsinhabitedby hummingbirdshave at leasttwo species,
and thesefall into two distinctly different size categories.Allometric scalingof bill length
with respectto body massis distinctivein Antillean hummingbirds;bill length increases
more rapidly with increasingbody weight in West Indian hummingbirdsthan in random
samplesof hummingbirdsof the world or in other birds. These morphologicalpatterns
appear to reflect two ecologicalprocesses:interspecificcompetition among hummingbirds
and mutualisticcoevolutionwith flowers.Hummingbird speciesof similar morphologyuse
similar floral resourcesbut rarely coexistin the samelocal areas.Speciesof divergent morphologyexploitdifferentfood resources
and frequentlycoexistlocally.Lengthof the bill,
which influencesaccessto different kinds of flowers,is particularlyimportantin the organization of thesesimple hummingbird associations.Received
28 June1983,accepted
5 July 1984.
lems can be overcome in some casesby reanalyzing information already available for certain groups of organisms. Although these
analysesare no substitute for careful experimental and observationalfield studies,they may
son 1956, Hutchinson 1959, MacArthur 1972). clarify issuesand focus future empirical work.
Hummingbirds, the nonpasserineavian famSuch patterns have played a particularly importantrole in the developmentof currentideas ily Trochilidae,provide an excellentsystemfor
about the role of competition and other inter- addressing these issues. This presumably
specificinteractionsin the organizationof nat- monophyleticgroup, containingmore than 100
ural communities(e.g. MacArthur 1972, Cody genera and 300 species,is widely distributed in
and Diamond 1975). Recently, however, these the New World, attaining its greatestdiversity
ideas have been criticized for two reasons. First,
in tropical latitudes. Hummingbirds are spe-
EVOLUTIONARY
ecology has a long tradition
of inferring dynamicprocesses
of ecologicalinteraction from static patterns in the morphologies and geographic distributions of closely
related species(e.g. Lack 1947,Brown and Wil-
manyof the purportedpatternshavenot been cialized for feeding on floral nectarand share
rigorously tested against appropriate null
models that assumethey could result from random processesrather than deterministic bio-
a suite of characteristicsincluding long, slender bills, tiny feet, and wings adaptedfor hovering flight. Becausethey are easily observed
logical mechanisms(e.g. Strong et al. 1979,
and their
Connor
and Simberloff
1979, Simberloff
food resources
can be monitored
ac-
and
curately,hummingbirdshave been the subjects
Boecklen 1981). Second, the relationship be- of many behavioral and ecologicalstudies.
tween morphologyand ecologyoften is simply
Several investigators have suggestedthat
assumed or inferred from indirect evidence inspeciesin particular geographicareasexhibit
stead of being documented rigorously by ob- morphologicalpatterns that reflect evolutionservation and experiment (e.g. Lack 1947, ary adaptationsboth for competitive interacHutchinson 1959, Findley 1973, Cody 1974, tions with other hummingbirds and for mutuBrown 1975, Karr and James1975).These prob- alistic interactionswith bird-pollinated flowers
251
The Auk 102:251-269. April 1985
252
BI•OWN
ANDBOWEI•S
(e.g. Grant and Grant 1968, Kodric-Brown and
Brown 1978, and Brown and Kodric-Brown 1979
for temperate North America; Lack 1973, 1976,
and Kodric-Brown
et al. 1984 for Caribbean
lands; Snow and Snow 1972, Colwell
is-
1973,
Feinsinger 1976, and Stiles 1975 for tropical
America; and Feinsinger and Chaplin 1975,
Feinsingerand Colwell 1978,and Feinsingeret
al. 1979 for hummingbirds in general). Although these interactionshave been studied
carefully in certain instances,there has been
little attempt to test the purported patterns of
morphologicalorganizationagainstalternative
[Auk, Vol. 102
clusive use of local areasby individuals and
species,and its outcome depends largely on
habitat(includingflowerdensity)and wing disc
loading (the ratio of body weight to the area
swept out by the wings in flight; Feinsinger
and Chaplin 1975, Kodric-Brown and Brown
1978,Feinsingeret al. 1979).
The claim that temperatehummingbirdsare
convergentin morphologyhasnot been tested
rigorouslyagainsta null hypothesis.
If the claim
iscorrect,thentemperatehummingbirdsshould
be more similar to each other than a comparable number of speciesdrawn at random from
models that assume no interspecific interac- an appropriatespeciespool. We testedthis hytions. We performed such testsfor the species pothesisby comparingmeasurementsof body
inhabiting temperateNorth America and the weight, bill (exposedculmen)length,and wing
Antillean islands, and the results are discussed
length for the temperatespecieswith setsof
here in terms of what is known about the pro- eightspecies
randomlydrawnfroma largepool
cessesof ecologicalinteraction.Our analysisis of speciesof hummingbirdsof the world for
basedon three quantitativemorphologicaltraits which measurementswere available (see Apthat are known to be important in humming- pendix). We conductedthree testsof the null
bird ecology:culmen(bill) length,which influ- hypothesisthat the temperatehummingbirds
ences the kinds of flowers from which the birds
are no more similar in size than randomly ascan forage;body weight, which indicatestotal sembled associationsof species.
First, we used computer simulation techenergy requirements;and wing length, which
togetherwith body weight affectsthe aerody- niques to draw 8 speciesat random 1,000 different times from the pool of 123 speciesfor
namicsof flight (Greenewalt 1975).
which
TEMPERATE NORTH
AMERICAN
all
measurements
were
available.
For
eachset of eight specieswe computedthe vari-
ASSOCIATIONS
ance for each of the three measurements, and
Eight species,belongingto four genera,have
breeding rangesthat extend well into temperate North America. Several other species,not
consideredfurther here, are primarily subtrop-
we recorded
ical in distribution
munitieswas greaterthan in the real community for all 1,000draws(Table 1). Clearly, temperate hummingbirdsare much more similar
and
reach
their
northern
limits just across the Mexican-United States
border. The collectivebreeding range of these
temperate speciesextends from the deserts of
the southwestern
United
States and northern
Mexico to southern Alaska and from Florida to
southeasternCanada. All of these speciesare
migratory, wintering in Mexico and Central
America.
Several investigatorshave noted that the
north temperatehummingbirdsare extremely
similarin sizeand shape.Grant and Grant(1968;
see also Brown and Kodric-Brown 1979) suggested that this similarity representsconvergence among speciesto utilize a common set
of widely distributedflower species,which also
have converged in size, shape, and nectar rewards.Hummingbirdscompetefor thesefloral
resourcesprimarily by aggressionand territoriality. This interferenceusuallyresultsin ex-
the number
of cases in which
the
variancefor a null communityexceededthat of
the observed temperate assemblage.For all
three characters, variance in the random com-
in size and shape than a random subsetof the
hummingbirds of the world (Fig. 1).
A randomdraw from the globalspeciespool
may not be an appropriatemethod for evaluating certainnull hypotheses,
however.For one
thing, the temperatespeciesprobablyare more
closelyrelatedto eachother than to certainexclusively tropical groups.Thus, the similarity
could be attributed to failure to diverge from a
recent common
ancestor rather than to conver-
gencein responseto similar ecologicalconditions. Biogeographicand taxonomicrelationshipssuggestthat temperateNorth American
hummingbirdsderived from tropical montane
and subtropicalancestorsthat colonizedfrom
Mexico. The morphology of temperate hum-
mingbirdsalsomightbe constrainedby the de-
April1985]
Hummingbird
Morphology
andEcology
253
TABLE
1. Resultsof simulationstestingthe null hypothesisthat the eight speciesof temperateNorth American hummingbirdsare more similar than expectedfrom a random draw of eight speciessampledfrom
the world speciespool. Simulationscomparedthe actualvariancesof morphologicalcharacters
of temperate
speciesagainstvariancesof the entire world speciespool, the speciespool that inhabitscontinentalNorth
America north of the Isthmusof Tehuantepec(biogeographicconstraints),and speciespool with wing
lengthsand body weightswithin the range observedfor temperatespecies(morphologicalconstraints).
Number
of random
draws with variances
greater than observedvariance
World
Variancein 8 North
Culmen length (mm)
Wing length (mm)
Weight (g)
species
Biogeographic
Morphological
American species
pool
constraints
constraints
1.75
15.21
0.203
1,000
1,000
1,000
996
999
999
1,000
248
523
mands of the fluctuating temperate climate or
by the necessityto migrate from the temperate
zones to winter in tropical and subtropical regions and survive in these habitats (see Feinsinger 1980). Such constraintswould indicate
that ecologicalfactorsinfluence the morphology and communityorganizationof hummingbirds, but would suggestthat other conditions
probablyare more important than biotic interactions with flowers or with other hummingbird specieswithin the temperatezone of North
America.
We then testedtwo null hypothesesthat imposeadditional constraintson the speciespool.
Unfortunately there is no modern, generally
acceptedclassificationof the family Trochilidae that attempts to reconstructphylogenetic
relationshipsamong the genera,so we did not
feel confident restricting our speciespool to
some taxonomicsubgroupwithin the family.
Instead
we drew
random
communities
from
a
speciespool limited by either biogeographicor
morphologicalconsraints.In a secondtest we
limited the speciespool to those 29 species(for
which we have measurementsout of a possible
38 species) that inhabit continental North
America north of the Isthmus of Tehuantepec
in southern Mexico (Peterson and Chalif 1973).
Becausemany biogeographersrecognize the
regionbetweenthe Isthmusof Tehuantepecand
the United
States-Mexican
border
as the area
of subtropicaltransition between tropical South
and Central America and temperate North
America (Brown and Gibson 1983), this procedure defined a speciespool consistingof 8 temperate North American speciesand 21 subtropical Mexican species.In a third test we limited
the speciespool to only those 28 speciesthat
had body weights and wing lengthswithin the
range (2.66-4.25 g and 39.8-49.7 mm) for temperatespecies.Note that both the biogeographically and the morphologically constrained
species pools are also taxonomically constrained, becausethey contain only 18 and 17
genera, respectively, of the 70 genera in the
world speciespool for which all measurements
are available for at least one species.
When 1,000 draws of 8 specieseachare taken
from either the biogeographicallyor morphologically constrainedpools,the variation in the
relevant traits is much greater than for the temperate hummingbirds (Table 1). This is true for
all three charactersin the test with biogeographicconstraints,but of courseonly for culmen lengths in the test with morphological
constraints(since all speciesin the pool were
within the range of variation of the temperate
speciesin both body weight and wing length).
The hypothesis that temperate hummingbirds are similar, not asa result of evolutionary
convergencebut simply becausethey have not
diverged substantiallyfrom a common ancestor, also can be evaluatedby comparingvariation within and between genera. Two of the
four genera of temperate hummingbirds have
other specieswith exclusively tropical distributions. If temperate hummingbirds have not
diverged significantly from a common ancestor, these tropical congeners should also be
similar in size and shapeto their temperate relatives. Clearly this is not the case (Table 2).
Tropical speciesof the genera Selasphorus
and
Calypteare uniformly smaller than their temperate congeners. The temperate representatives of these genera are similar to each other
and to the other two monotypicgeneraof tern-
254
BROWN
ANDBOWERS
TEMPERATE
•
NORTH
[Auk,Vol. 102
AMERICA
GREATER
ANDLESSER
ANTILLES
WORLD
LOG BOOYWEIGHT (g)
LOG CULMEN LENGTH (f'om)
LOG WING LENGTH(ram)
Fig. 1. Frequencydistributionsof body weights(g), culmenlengths(mm), and wing lengths(mm) for
speciesof hummingbirdsfrom temperateNorth America, the Greaterand LesserAntilles, and the world.
Note the smallvariationamongtemperatespeciesand the bimodalpatternfor Antillean speciescompared
to the world speciespool.
peratehummingbirds,especiallyin bill length.
The greatersimilarity amongtemperatespecies
than among congenericspeciesprovidesstrong
evidence that the temperate hummingbirds
have acquired their strikingly similar morphologiesby evolutionaryconvergence.
These analysesdemonstratethat temperate
North Americanhummingbirdsare muchmore
similarin morphology,especiallyin bill length,
than expectedon the basisof several different
null hypotheses.Becausebill length is of primary importance in determining the kinds of
flowersfrom which hummingbirdscan forage
(e.g. Snowand Snow 1972,Colwell 1973,Stiles
1975, Feinsinger and Colwell 1978, KodricBrown et al. 1984), these resultsalso provide
strong support for the idea (Grant and Grant
1968, Brown and Kodric-Brown 1979) that the
birds have convergedto usethe samekinds of
flowers.It seemshighly unlikely that physiological adaptations to climate or migration
would result in more precise convergencein
bill length than in either body size or wing
length.
ANTILLEAN
COMMUNITIES
The tropical islandsof the Greaterand Lesser
Antilles are inhabitedby 14 speciesand 9 gert-
era of hummingbirds(Bond 1971).All but 2 of
these speciesare endemic to the Greater and
LesserAntilles. Other speciesnot considered
in the presentanalysisoccuron islandsaround
the periphery of the CaribbeanSea (the Bahamas, Old Providence, St. Andrew, Trinidad, and
Tobago),but theseislandsdiffer conspicuously
in habitatand geologicalhistory from the Antilles. As noted by Lack (1973, 1976),the Antillean hummingbirds appear to be distributed
among the 36 islands in nonrandom patterns.
Although Mona, a relatively large island between Puerto Rico and Hispaniola, has no
hummingbirds, the remaining 35 have 2-5
species(Table 3). Lack noted that speciesoccurring together on the same island tended
either to differ substantiallyin body size and
overlap widely in local distribution or to be
relatively similar in size but occur in different
habitats,usually at different elevations.Intensive work
on Puerto
Rico and more casual ob-
servationson other islandssuggestthat Antillean hummingbirds have morphologiesthat
reflectboth interspecificcompetitionwith other hummingbirdsand coevolutionwith mutualistic flowers (Kodric-Brown et al. 1984). The
situation is similar to that in temperate North
America, exceptthat in the Antilles there are
two size categoriesof hummingbirdsthat are
April1985]
Hummingbird
Morphology
andEcology
255
TABLE2. Mean (œ),range, and coefficientof variation (CV), with standarddeviation (Soy)
' in parentheses,
for temperatehummingbirdsand temperate,tropical, and combinedspeciesof Selasphorus
and Calypte.
Temperate
hummingbirds
(8 species,4 genera)
TemperateSelasphorus Tropical Selasphorus
and Calypte
and Calypte
(5 species)
(6 species)
All Selasphorus
and Calypte
(11 species)
Culmen length (mm)
œ
17.66
Range
15.54-19.73
CV
0.0747 (+0.0188)
Wing length (mm)
œ
Range
CV
Weight (g)
17.61
11.77
14.72
16.57-19.06
0.0521 (+0.0165)
10.59-12.90
0.0790 (+0.0228)
10.59-19.06
0.2201 (+0.0469)
43.92
45.15
37.86
41.17
39.77-49.68
0.0887 (+0.0222)
39.77-49.68
0.0975 (+0.0308)
31.61-41.57
0.1033 (+0.0298)
31.61-49.68
0.1329 (+0.0283)
œ
3.29
Range
CV
3.46
2.66-4.25
0.1370 (+0.0343)
3.08-4.25
0.1349 (+0.0427)
2.17
2.98
1.90-2.47
0.1315 (+0.0379)
1.90-4.25
0.2578 (+0.0549)
aS• calculated from Sokal and Rohlf (1969).
highly specificpollinatorsof two distinct sets approximately20-50% of the null assemblages
of hummingbird-pollinated
plant species:
small, were more different than the actual Antillean
short-billed hummingbirds can forage only specieswhen both were comparedto the entire
from short-tubed flowers that have low rates of
speciespool (Table 4). This is hardly convincnectar secretion, whereas large, long-billed ing evidence for the nonrandomassemblyof
birds feed almostexclusivelyfrom long-tubed Antillean communities, but note that fewer
flowers that secrete more copious nectar.
simulated
The relationship between Antillean hummingbird morphologyand ecologycan be examined more rigorouslyby comparingthe observed morphologies and distributions of
specieswith thoseexpectedon the basisof null
differencefor culmen length (102 out of 500;
models
that assume
that
insular
communities
are assembledat random, uninfluenced by interspecificinteractions.
Comparison
withtheglobalspecies
pooL--Wefirst
testedwhether the morphologiesof Antillean
hummingbirds differ significantly from a random sample drawn from the entire pool of
hummingbirdsof the world. For each of the
three measurements(body weight, culmen
length, and wing length) we drew at random
from the pool 14 speciesfor which the relevant
data were available. This was repeated 500
times. We quantified the maximum difference
distributions
exceeded
the observed
P = 0.2) than for the other two measurements.
Comparison
with a randomizedAntillean species
pool.--We next testedwhether the distribution
of speciesmorphologiesdiffered from that expectedif the 14 specieswere redistributedrandomly among islandsin the frequenciesthat
they actually occur.There are 35 islandsinhabited by hummingbirds, but the same combinationsof speciesoccurtogetheron severaldifferent islands(Table3). Thus,it is inappropriate
for most tests to consider
each island
as an in-
dependent sample.To be rigorousand conservative we usedonly the nine different combinations to obtain the observed distributions
for
that for the entire world speciespool, and then
each measurement.We generatedan expected
distributionby drawing nine different combinations of speciesat random from the pool of
14 Antillean species(Table 3). This draw was
constrainedto have the samenumber of species
determined
in each combination
between
the actual Antillean
the
number
of
distribution
random
and
assem-
blagesthat were more different than this when
also compared to the world speciespool. We
measured the difference using the D-statistic
from the Kolmogorov-Smirnov
two-sampletest,
which
is the maximum
cumulative
deviation
over correspondingintervalsbetweentwo distributions(Siegel1956).Dependingon the trait,
as the real combinations
(3 two-species,4 three-species,1 four-species,
and 1 five-speciescombination). This procedure was repeated750 times to generatean expected frequency distribution for each morphologicaltrait. Becausethe birds in both the
real and artificial assemblages
differ in absolute
size,and we wanted to evaluatethe hypothesis
256
BROWN
ANDBOWERS
[Auk,Vol. 102
T^BLE3. The distribution of hummingbird specieson 39 islandsof the Greater and LesserAntilles. From
Bond (1971) and Lack (1973).
Grenada
x
x
Carriacou
Grenadines
St. Vincent
Barbados
St. Lucia
x
x
x
x
x
x
x
x
x
x
La Martinique
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
St. Martin
x
x
x
x
x
Dominica
Marie
Galante
La Guadeloupe
D•sirade
Montserrat
Antigua
Barbuda
Nevis
Saba
St. Eustatius
St. Christopher
Anguilla
x
St. Bartholomew
x
Bequia
Anegada
x
x
x
x
St. Croix
x
x
St. Thomas
Puerto Rico
Mona
x
x
x
x
X
X
x
•ispaniola
x
x
Ile de la Gon•ve
x
x
Tortuga Island,
x
Haiti
x
x
Jamaica
x
Cuba
x
x
Isle of Pines
Culebra
x
x
Vieques
St. John
that coexisting speciestend to differ in size
morethan expectedon the basisof chance(Lack
1973,1976),we expressed
the differencesin size
between every pair of speciesin all combinations (both real and randomly assembled)as
the difference in the logarithms for each measurement.
The
observed
and
null
distributions
were
comparedand the significanceevaluatedwith
a Kolmogorov-Smirnov test (Fig. 2). Differences between observed and null are significant for culmen length (P < 0.05), marginally
significant for body weight (0.05 < P < 0.10),
and insignificant for wing length (P > 0.10).
Note that it is particularly with respect to bill
length that real communities represent nonrandom combinationsof morphologies. Also,
the
observed
distributions
for
both
culmen
April 1985]
TABLE 4.
Hummingbird
Morphology
andEcology
Simulation
results
that
tested
whether
257
3
morphologicalcharactersof Antillean hummingbird speciesdiffer significantly from a sample of
14 speciesdrawn at random from the world species
pool
.40.•0'
.gO-
Number
Character
Number
greater
•'
D-statistic
less
or equa!
Z
Wings
0.15
246
254
•
Weights
0.18
364
136
O
Culmens
0.20
398
102
•
h
.10-
.40.30.20-
length and body weight differ conspicuously
from the null assemblagesin having a pronounced peak at a difference in logarithms of
approximately 0.30, which correspondsto a
body size ratio of about two. This is a conservative analysisin the sensethat it incorporates
.10
.30'
all specieswithin islands.When they occuron
the same island, speciesof similar size often
.20-
are segregated into different elevations and
.10-
habitats (Lack 1973, 1975; confirmed on Puerto
Rico by Kodric-Brown et al. 1984).
speciesper island. Becausethere is one island
(Mona) with no hummingbirdsand many with
two species(Table 3), Lack suggestedthat, in
particular,the absenceof islandswith only one
specieswas unlikely to be due to chance.
distribution
i• []Ob
[]
.06
Distribution
of species
amongislands.--Wetested
directly againsta null model Lack's(1973, 1976)
suggestionthat the Antillean assemblages
are
nonrandom with respect to the number of
We tested the observed
Wings
.40-
of num-
bersof speciesamong islandsagainsta Poisson
distribution. This null model predicts the independent distribution of rare events. It is appropriate here becausethe number of species
per island is small and apparently not influenced significantly by island area or interisland distance(Lack 1973, Terborgh 1973). Resuits of this analysis indicate that the actual
distribution differs (P < 0.01) from the predicted null distribution in having too few islandswith only one species,too many islands
with two species,and too few islandswith three
or more species(Fig. 3). This pattern is highly
nonrandom(X2= 27.26, P << 0.01; Fig. 3 shows
the data plotted by number of speciesper island, but some adjacent categories can be
lumped to give sufficientlyhigh expectedvalues to meet the assumptionof the Chi-square
analysis).This pattern suggeststhat any island
.12
.18
DIFFERENCE
.24
.50
Of
.56
.42
Null
.48
.54
LOGARITHMS
Fig. 2. Comparisonsbetween observed distributions of measurements
for Greater
and Lesser Antil-
lean hummingbird speciesand those measurements
predicted from a null model that assumesthat combinations of speciesare assembledat random from
the pool of 14 Anti!lean species.Measurementsare
expressedas differencesin the logarithms for each
pair of species.
that can supportone hummingbirdspeciescan
be invaded successfullyby at least one additional species.The precisereasonsfor this are
obscure,but we suspectthey have to do with
the success
in colonizationby hummingbird-pollinated plant speciesas well as hummingbirds,
and by the coevolutionof the hummingbirdplant interactions within islands (see KodricBrown et al. 1984).
Allometryof billsandbodysize.--A conspicuousfeatureof Fig. 2 is the pronouncedpeak in
observedmeasurementsof both body weights
and culmenlengthsof coexistingcombinations
of speciesthat occurs at a ratio of about 2.0,
whereasthe peak in wing length corresponds
to a ratio of about
1.2. This indicates
unusual
allometric scaling of body parts in Antillean
258
BROWN
ANDBOWERS
hi
Z20
hi
015
]
[]
o
b.
o
Observed
Expecled
io
z
5
c•
b3
o
2
NUMBER
OF
3
SPECIES
4
5
PER
ISLAND
Fig. 3. Comparison between the observed num-
ber of hummingbirdspeciesamong36 islandsof the
Greater and LesserAntilles and that predicted by a
[Auk,Vol.102
ly twice as high as the exponents for either
bills or wings (0.32-0.53) of any of the other
birds(Fig. 4). Statistically,the exponentfor Antillean hummingbirdbills can be readily distinguishedfrom the exponentsof all the other
allometric relationships (P < 0.01), which are
all much more similar to each other (Table 5).
The bills of Antillean hummingbirdsare much
more different in length than would be expected on the basis of body weight. Also of
interest, although it will come as no surprise
to studentsof hummingbirds,is the greatvariation around the allometric relationshipsfor
bills (r2= 0.39) of hummingbirdsin the global
pool,comparedto comparablerelationshipsfor
either Antillean hummingbirds, owls, or corvids (all r2 > 0.79). This indicates that differ-
encesin bill length (and shape)play such an
importantrole in the ecologyof hummingbirds
that selectionhas producedmorphologiesof
many speciesthat deviate greatly from prehummingbirds.Becausemassand volume scale dicted allometric relationships(see also Feinas the cube of linear measurements,body singer and Colwell 1978).
weight ratiosof about 2.0 usuallyare associated
The unusualallometricscalingof bill length
with ratios of linear measurements of 1.2-1.3.
suggestsa final and admittedly ad hoc test of
This typical allometry is observedfor the re- the null hypothesisthat Antillean humminglationshipbetweenbody weight (a volumetric birds representa random sampleof the hummeasurement)and wing length (a linear mea- mingbirds of the world. We used Monte Carlo
surement),but bill lengths of different-sized methodsto draw 14 speciesat random from the
Antillean hummingbirds are more different 123 speciesfor which both culmen length and
than would be expectedon the basisof body body weight data were available. Then we calPoisson
distribution.
weight ratios.
culated
the
correlation
coefficient
for
the
14
These relationshipscan be examined more pairs of log-transformeddata. This procedure
rigorouslyby deriving the allometricequations was repeated 1,000 times. In 996 of 1,000 sets
from regressions
fitted to the data.Theseequa- of speciesassembledrandomly from the world
tions take the form l = CW'" where l is a linear
species pool the correlation coefficient was
dimension(bill length or wing length in mm), lower than the value (0.79) calculated for the
C is a fitted constant, W is a volumetric meareal Antillean fauna, resulting in unequivocal
surement (body weight in g), and rn is the ex- rejectionof the null hypothesis(P = 0.004).
ponent relating the two variables(or the slope
All Antillean hummingbirdsappearto conof the regressionline fitted to log-transformed form to a special relationship between bill
data). In Fig. 4 we have plotted fitted regres- length and body size that is very different from
sion equations for log-transformed measure- comparablerelationshipsboth for hummingmentsof Antillean hummingbirds,humming- birds in general and for other kinds of birds.
birds of the world, North American corvids
Why shouldspeciesthat differ in body size be
(crowsand jays),and North Americanowls.We even more different in bill length than expectincluded the last two groups becausewe be- ed on the basisof the allometric relationship
lieve they are representativeof the generalal- in other hummingbirds, unless speciesof dif1ometricscalingof bill length and wing length ferent sizes tend to occur together and have
in birds. Each of these two families exhibits a
been selectedto diverge in bill length? Both
wide range of body sizesbut doesnot use its Lack's qualitative observationsand our quanbill for foragingin the samespecializedway titative analyses(Figs. 1-3) show that pairs of
that hummingbirdsdo. The exponent(slope) speciesof different size usually occurtogether
for Antillean hummingbirdbills (0.79) is near- on the same islands.Although these distribu-
April 1985]
Hummingbird
Morphology
andEcology
OWLS
HUMMINGBIRDS
z
LLI
259
2.0[
1.55
and
CORVIOS
CORVID$
._1 i 4•
z
o
._1
oO
....
o.•
2.6
2.5
z
o
.:••
o o
t so
12'0
z
2.O
o
1.6
.a
LOG
BODY
I 8
WEIGHT
20
22
2.4
2.6
2.8
3.0
3.2
(g)
Fig.4. Allometricrelationships
of culmenandwing lengthagainstbodyweightfor Antilleanhummingbirds,hummingbirds
of the world,and North Americanowlsand corvids.Note that the slopefor Antillean
hummingbirdculmens(solidcircles)is nearlytwicethat of eitherbills or wingsof any of the otherbirds.
tions may reflect historical and contemporary
biogeographicconstraintsas well as contemporary biotic interactions,thesespeciesappear
to have adaptedto coexistenceby diverging in
body size and especiallyin bill length. This is
supportedby taxonomicbodysizepatterns.For
example, Chlorostilbon
ricordiion Cuba and C.
swainsoniion Hispaniola, where they coexist
with smaller hummingbirds, are much larger
than C. maugaeus
on PuertoRicoand other congenericspecieson the tropicalAmericanmainland (see Appendix).
This interpretationalsois consistentwith the
detailed ecological observations of KodricBrown et al. (1984) on Puerto Rican humming-
Only large hummingbirds with long bills can
legitimately extract nectar from long-tubed
flowers,which producesufficientquantititesof
nectar to make foraging rewarding for large
birds. On the other hand, both large and small
birds can feed from short-tubed flowers, but
these produce such small quantitites of nectar
that foragingis economicalonly for the smaller
species. Thus differences among coexisting
speciesin bill length and body size are functionally interrelated and appear to reflect selection both to diverge in resourceutilization
soas to reduceinterspecificcompetitionand to
coevolve with specificflower speciesso as to
provide effective pollination.
birds. They confirmedLack'sobservationthat
speciesof similarsizeare segregated
by habitat
DISCUSSION
and elevation,whereasspeciesof dissimilar'size
are broadly and locally sympatric.Thesepairs
Our analysesdemonstratethat specificmorof coexistingspeciesfeed from almost com- phological traits characterize the hummingpletely nonoverlappingsetsof flower species, birdsfrom particulargeographicareas.Species
and flower utilization is related to bill length.
from temperate North America differ from
260
BROWN
ANDBOWERS
[Auk,Vol. 102
TABLE
5. Resultsof t-teststhat test the null hypothesisthat slopesbetweenculmenlength or wing length
and body weight are statistically(P < 0.05)equal.All comparisons
were done on log-transformeddata (see
Fig. 3).a
Culmens
Wings
Hum-
Antillean
North
North
minghumming- American American birds of
birds
owls
corvids
Antillean
humming-
North
American
the world
birds
owls
North
American
corvids
Culmens
Antillean hummingbirds
North
North
American
American
owls
corvids
-5.87**
4.63**
-1.25
3.01'*
2.37*
1.27
Antillean hummingbirds
4.37**
1.52
0.25
!.00
North
North
4.75**
3.87**
1.13
2.01
0.13
0.87
1.25
0.36
-0.37
0.50
-1.16
4.63'*
1.26
0.13
1.25
0.25
0.13
Hummingbirds of the
world
Wings
American
American
owls
corvids
Hummingbirdsof the
world
1.05
a, p < 0.01, ** P < 0.001.
those inhabiting the tropical islands of the
Greater and Lesser Antilles, and each of these
groupsrepresentsa selectedsubsampleof the
hummingbirds of the world. These morphological patterns indicate that ecological processesseverely limit the kinds of hummingbirds that comprise communities in different
geographic areas. Ecological constraints, of
which the mostimportant is probablythe number and kinds of floral resources,limit the range
and combinationsof morphologicaltraitsof the
speciesthat inhabit a region. Competitive interactionsbetween hummingbirds, on the other hand, require speciesthat coexist in local
areasto differ in morphologysufficientlyto exploit substantially nonoverlapping food resources.The eight hummingbird species in
temperate North America utilize a large set of
convergentlysimilarfloral resources
(Grantand
different size coexistlocally to form the basic
two-speciescommunity that is found on all islands with hummingbirds (Lack 1973, 1976;
Kodric-Brown et al. 1984).
These results suggestthat it should be possible to predict quite preciselythe morpholog-
icalcharacteristics
of the hummingbird
species
•
occupying different habitats and geographic
regions. Depending on the flowers and perhaps on other environmental conditions,only
specieswith certain combinationsof morphologiesshould be able to coexistto form stable
communities.We have begun to characterize
thesetraits for temperate North American and
Antillean communities.We expectother hummingbird associations exhibiting different
combinationsof morphologiesto inhabit other
areas, especially tropical continental habitats
where several speciescoexistlocally. Available
Grant 1968, Brown and Kodric-Brown 1979).
information suggeststhat tropical mainland
Consequentlythey exhibitlittle variationin bill communitiesoften containmorphologicallydilength and body size, and only one species verse hummingbird assemblages,with species
normally is found within a local habitat (Ko- exhibiting different combinationsof body size,
dric-Brown and Brown 1978, Brown and Kowing length, bill length, and bill shape(Snow
dric-Brown 1979). The Antillean islands are inand Snow 1972, Stiles 1975, Feinsinger 1976,
habited by two body size and bill length Feinsinger and Colwell 1978).
categoriesof hummingbirdsthat have coeFeinsingerand Colwell (1978) alsohave been
volved with two distinct setsof flower species impressed by the close correspondencebe(Kodric-Brownet al. 1984). Hummingbirds of tween morphologyand communityecologyof
similar size tend not to occur on the same ishummingbirds.Basedlargely on their own exland, and when they do are normally found in periencesin continentaltropical America, they
different habitats, whereas hummingbirds of have attemptedto deducethe rules that govern
April 1985]
Hummingbird
Morphology
andEcology
the assemblyof speciesinto communitiesof in-
261
cations.Recently there has been much debate
creasingdiversity.Interestingly,they describe about whether ecologicalcommunitiesexhibit
the basic two-speciestropical community as nonrandom patterns of organization that reconsisting of a territorialist and a low-reward
flect ecologicalprocesses
suchas interspecific
trapliner or generalist,but thesecategoriesdo competition.Severalauthors(e.g. Connor and
not accurately characterize the two-species Simberloff1979, Strong et al. 1979, Simberloff
communities of the Antillean islands, where
and Boecklen1981) have pointed out correctly
the main differencesbetween locally coexist- that apparentpatternsin the morphologies
and
ing speciesare in body size, bill length, and distributionsof speciesseldomhavebeentested
flower specialization.In fact, both large and to determinewhether they differ significantly
small speciestend to be trapliners, at least in
from appropriatenull modelsthat assumethat
the Greater Antilles (Kodric-Brown et al. 1984).
Nevertheless,the ecologicalconstraintsand
opportunities that affect the occurrence of
hummingbirdsin a particularregion obviously
are reflectedin the morphologiesof coexisting
species,suggestingthat it eventually may be
possibleto formulatecommunityassemblyrules
that are both preciseand general.
One general result of almost all of our analysesis that of the three morphologicalcharacteristicsstudied,bill length consistentlyshows
the clearest, most nonrandom patterns. This
may not seemsurprisingin view of the obvious
importance of bill size and shape in the foraging of hummingbirds.Nevertheless,it emphasizes the roles of food resources,interspecificcompetitionfor food, and mutualistic
plant-pollinatorinteractions
in determiningthe
morphologicalattributesand ecologicalroles
of those species that coexist to form natural
communities.Not only do these kinds of eco-
logicalinteractionslimit the morphologicalattributesof the speciesthat occurwithin certain
geographicareas,but they also influence the
evolution
of these traits.
In the case of hum-
mingbirds, mutualistic interactions between
birds and flowersmay be at leastas important
ascompetitionamonghummingbirdspeciesin
the adaptiveradiationof the entire family Trochilidae
and the evolution
of those subsets of
speciesthat inhabit the particulargeographic
regions.In fact,it is probablyunrealisticto try
to assessthe relative importance of these two
kindsof interactionsbecausethey are intimate-
ly interrelated;convergenceor divergencein
bill length and other aspectsof morphology
affectthe relative abilities of different species
to use particular kinds of flowers, and this in
turn influencesboth the extent of interspecific
competitionand the efficiencyof pollination.
The relationship between morphology and
community ecologythat we have documented
for hummingbirdshas severalgeneralimpli-
speciesassociateat random. However, here and
elsewhere (Bowers and Brown 1982, Brown and
Bowers 1984) we have shown that communities
of both rodents and hummingbirdsexhibit
nonrandom organization that apparently reflectsthe role of interspecific
competitiveand
(in the caseof hummingbirds)mutualisticinteractions.In both casesthere already existeda
largebackgroundof field studieson the ecology of thesespeciesthat includedanalysesof
the relationshipbetweenmorphologicaltraits
and processes
of resourceexploitationand aggressiveinterference.This facilitated the formulation and testing of null hypothesesthat
are not only statisticallyrigorousbut alsobiologicallyappropriate.
It is hardlysurprisingthat
a goodknowledgeof the biologyof the organisms concerned
is a valuable
aid in construct-
ing and testingrealisticnull hypotheses
and in
elucidatingthe ecologicalprocesses
that influence community organization.
In investigatingthe relationship between
morphology
andcommunityecology,it is often
usefulto restrictthe analysisto closelyrelated
species.This is not to imply that only closely
related speciesinteract strongly to influence
community organization.On the contrary, distantly related taxa often are involved in all
classesof interspecificinteractions,including
competition.However,their morphologiesmay
be so different that they provide little clue to
their specificecologicalroles.For example,on
most Antillean islandsa passefinebird, the
bananaquit(Coereba
fiaveola),removesnectar
from long-tubedflowersand presumablycompetes significantlywith large hummingbirds
(Kodric-Brownet al. 1984). In part becauseof
its distant phylogeneticrelationship, its morphology is adapted for a different manner of
foraging:it cutsthroughthe floral tube while
perched,ratherthanprobingthroughthe flower opening while hovering as hummingbirds
do. Consequently,
its morphological
character-
262
BI•OWNAND BOWERS
[Auk, Vol. 102
ß & A. KODRIC-BROWN.
1979. Convergence,
istics provide little indication that it interacts
competition and mimicry in a temperate commuch more closelywith hummingbirdsthan
munity of hummingbird-pollinated flowers.
any other Antillean passefine.
Ecology60: 1022-1035.
Closely related species often are so conBROWN, W. L., & E. O. WILSON. 1956. Character disstrained by their common phylogenetic histoplacement. Syst. Zool. 5: 49-64.
ries that they differ substantiallyin only a few CARPENTER,
F. L. 1976. Ecologyand evolution of an
characteristics.Often, especially in sympatric
Andean hummingbird ( Oreotrochilusestella).
species, these differences reflect divergent
Berkeleyand LosAngeles,Univ. CaliforniaPress.
mechanisms of resource utilization in response
to interspecificcompetition.Thereforeß
they tell
more about the role of ecologicalinteractions
in causingcharacterdisplacementand adaptive
radiation within phyletic lineagesthan they do
about the organization of diverse communities
COD¾,M.L. 1974. Competitionand the structureof
bird communities. Princeton, New Jersey,
Princeton
Univ.
Press.
, & J. M. DIAMOND. 1975. Ecology and the
evolution of communities. Cambridge, Massachusetts, Harvard Univ. Press.
COLWELL,
R. K. 1973. Competition and coexistence
in a simple tropical community. Amer. Natur.
containing many distantly related taxa. Analysis of morphologicalpatternsamong closely
107: 737-760.
related speciesis useful becauseit may suggest CONNOR,E. F.ß•: D. SIMBERLOFF.
1979. The assembly
mechanismsof ecologicalinteractionthat affect
of speciescommunities:chance or competition?
communitystructure,but otherapproaches
will
Ecology60: 1132-1140.
be necessaryto explore the consequencesof FEINSINGER,
P. 1976. Organizationof a tropicalguild
of nectivorousbirds. Ecol. Monogr. 46: 257-291.
theseinteractionsamongdistantlyrelatedtaxa.
ACKNOWLEDGMENTS
We thank R. Zusi, R. K. Colwell, and J.B. Dunning
for providing unpublishedmeasurementsof hummingbirds.A. Kodric-Brown,G. S. Byersß
and D. F.
1980. Asynchronousmigration patternsand
coexistenceof tropical hummingbirds. Pp. 411419 in Migrant birds in the Neotropics(A. Keast
and E. S. Morton, Eds.). Washington, D.C.,
Smithsonian
Institution
Press.
, & S. B. CHAPLIN. 1975. On the relationship
the WestIndies,which yieldedvaluableinsightsinto
the ecologyof Antillean hummingbirds.Brown'sresearchwas supportedby NSF GrantsDEB 76-99499
betweenwing disk loading and foragingstrategy in hummingbirds.Amer. Natur. 109:217-224.
ß & R. K. COLWELL.1978. Community organizationamongneotropicalnectar-feedingbirds.
and DEB 80-21535.
Amer.
Gori collaborated with J. H. Brown in field work in
, --,
LITERATURE CITED
AMERICANORNITHOLOGISTS'
UNION. 1983. Check-list
of North American birds, 6th ed. Baltimore,
Zool.
18: 779-796.
J. TERBORGH,
& S. B. CHAPLIN. 1979.
Elevationand the morphologyßflight energeticsß
and foragingecologyof tropicalhummingbirds.
Amer.
Natur.
113: 481-497.
FINDLE¾,
J. S. 1973. Phenetic packing as a measure
Amer. Ornithol.
Union.
of faunal diversity.Amer. Natur. 107:580-584.
BOND,J. 1971. Birds of the West Indies. Bostonß GRANT,K. A., & V. GRANT. 1968. Hummingbirds
and their flowers. New York, Columbia Univ.
Houghton Mifflin Co.
Press.
BOWERSß
M. A., & J. H. BROWN.1982. Bodysize and
coexistence in desert rodents: chance or comGREENEWALT,
C. H. 1975. The flight of birds.Trans.
Amer. Phil. Soc. 65: 1-67.
munitystructure?
Ecology63:391-400.
G.E. 1959. Homage to SantaRosalia,
BROWNß
J. H. 1975. Geographical
ecologyof desert HUTCHINSONß
rodents.Pp. 315-341 in Ecologyand evolutionof
communities(M. L. Cody and J. M. Diamond,
--,
Natur.
93: 145-159.
KARR,J. g., •: F. C. JAMES.1975. Ecomorphological
Press.
configurationsand convergent evolution. Pp.
258-291 in Ecologyand evolution of communi& g. A. BOWERS.1984. Patterns and proties (M. L. Cody and J. M. Diamond, Eds.).Camcessesin three guilds of terrestrialvertebrates.
bridge, Massachusetts,
Harvard Univ. Press.
Pp. 282-296 in Ecologicalcommunities:
concepA., & J. H. BROWN. 1978. Influence
tual issuesand the evidence (D. Simberloff and KODRIC-BROWN,
of economics,interspecificcompetitionand sexD. R. StrongßEds.). PrincetonßNew Jersey,
Princeton Univ. Press.
ual dimorphism on territoriality of migrant Ru& A. C. GInSON. 1983. Biogeography. St.
fous Hummingbirds.Ecology59: 285-296.
--,
--,
G. S. BYERS,& D. F. GORI. 1984. OrLouis, MissourißMosby.
Eds.).Cambridge,Massachusetts,
HarvardUniv.
--,
or why aretheresomanykindsof animals.Amer.
April 1985]
Hummingbird
Morphology
andEcology
ganization of a tropical island community of
hummingbirdsand flowers. Ecology65: 13581947.
Darwin's
finches. London,
Cam-
bridge Univ. Press.
ß 1973. The numbersof hummingbird species
in the West Indies.
Evolution
27: 326-337.
1976. Island biology. Oxford, BlackwellSci-
$OKAL,R. R., & F. J. ROHLF. 1969. Biometry: the
principles and practiceof statisticsin biological
research. San Francisco, W. H. Freeman and Co.
STILES,
F. G. 1973. Foodsupply and annual cycle of
the Anna Hummingbird. Univ. California Publ.
ZooL 97: 1-1110.
entific.
MACARTHUR,
g. $. 1972. Geographicalecology.New
York, Harper and Row.
PETERS,
J. L. 1945. Check-list of birds of the world,
vol. 5. Cambridge, Massachusetts,Harvard Univ.
Press.
PETERSON,
g. t., & E. L. CHALIF. 1973. A field guide
to Mexicanbirds. Boston,Massachusetts,
Houghton Mifflin
Co.
America, part 3. Washington, D.C., U.S. Government Printing Office.
1914.
1975. Ecology,flowering phenology, and
hummingbird pollination of some Costa Rican
Heliconia
species.Ecology56: 285-301.
STRONG,D. g., JR., L. g. $ZYSKA,& D. $IMBERLOFF.
1979. Tests of community-widecharacterdisplacementagainstnull hypotheses.
Evolution33:
897-913.
TER}•ORGH,
J. 1973. Chance,habitat, and dispersalin
RIDGEWAY,R. 1904. The birds of North and Middle
--.
SNOW,B. K., & D. W. SNOW. 1972. Feeding niches
of hummingbirdsin a Trinidad valley. J. Anita.
Ecol. 41: 471-485.
1368.
LACK, D.
--.
263
The Birds of North
and Middle
Amer-
ica, part 4. Washington, D.C., U.S. Government
Printing Office.
SIEGEL,
S. 1956. Nonparametricstatisticsfor the behavioral sciences. New York, McGraw-Hill.
$IMBERLOFF,
D., & W. BOECKLEN. 1981. Santa Rosalia
the distribution
of birds in the West Indies. Evolution 27: 338-349.
WASER,
N.M. 1978. Competitionfor pollination and
sequentialflowering in two Coloradowild flowers.Ecology59: 934-944.
WETMORE,
g. 1968. The birds of the Republic of
Panama,part 2. Washington,D.C., Smithsonian
Press.
reconsidered:size ratios and competition. Evolution
35: 1206-1228.
APPENDIXßWing length (mm), body weight (g), and APPENDIX. Continuedß
culmenlength(ram)of hummingbirds
in the world
Charspeciespool. Taxonomyfollowsthe A.O.U. (1983;
North American species)and Peters (1945). MeaSpecies
acter
surementswere taken from Ridgeway (1904, 1914),
Wing
Wetmore (1968), Stiles (1973), Carpenter (1976), Glaucishirsuta(A)
Weight
Feinsinger (1976), Waser (1978), Brown and Ko-
dric-Brown(1979),and Feinsingeret al. (1979).J.
H. Brown, R. K. Colwell, and R. Zusi contributed
Threnetes
leucurus
unpublisheddata.Temperate(T) and Antillean (A)
hummingbird speciesare designated,as are those
usedin the biogeographically
(B)and morpholog- T. ruckeri
ically (M) constrainedtestsof null hypothesesfor
the temperatefauna.
Char-
Species
acter
Doryferajohannae
Wing
Weight
Culmen
D. ludoviciae
Wing
Weight
Androdon
aequatorialis
œ
SD
53.60 1.00
4.10 0.10
--
--
58.16 0.70
5.72
0.20
n
9
9
35
36.13
--
16
65.37
--
15
8.00
39.77
--
1
--
15
SD
n
59.22
6.82
2.10
0.52
76
96
Culmen
32ß25
Wing
Weight
59.13
5.59
Culmen
29.50
Wing
Weight
56.85
5.92
Culmen
30ß92
--
0.80
0.10
--
-0.70
73
20
42
--
45
11
--
45
Wing
Weight
-6.00
---
-1
Culmen
--
--
--
Wing
Weight
59.85
5.66
Culmen
42.50
40
64
--
20
---
-2
Wing
Weight
P. superciliosus
(B)
Wing
Weight
59.74
6.04
1.34
0.27
268
277
Culmen
37.65
1.17
255
Culmen
-5.70
0.40
0.24
P. syrmatophorus
20
Culmen
Culmen
P. guy
--
Wing
Weight
Phaethornis
yaruqui
œ
--
--
--
264
BROWN
ANt)BOWERS
APPENDIX.
Continued.
APPENDIX.
[Auk,Vol. 102
Continued.
Char-
Species
P. hispidus
Char-
acter
œ
SD
n
Wing
Weight
56.90
4.86
0.60
0.12
22
26
Culmen
--
P. anthophilus
Wing
Weight
57.23
4.60
Culmen
35.30
P. ruber
Wing
--
--
---
23
1
--
23
--
--
--
Weight
2.15
--
Culmen
--
--
--
P. griseogularis
Wing
Weight
•
2.00
---
-1
--
--
--
P. longuemareus
Wing
Culmen
(M)
Eutoxeres
aquila
E. condamini
Weight
49.61 0.30
2.97
0.38
4
85
Culmen
31.19
--
79
71.45
--
29
Weight
10.83
--
3
Culmen
27.09
--
Wing
Weight
68.10 0.80
9.36 0.27
--
--
69.34 -8.87 0.41
51
37
Culmen
23.17
Campylopterus
curvipennis
Wing
C. largipennis
Wing
Weight
Culmen
Culmen
C. hemileucurus
(B)
C. ensipennis
C. falcatus
Eupetomena
macroara
mellivora
Melanotrochilus
f•,lSC•S
--
--
51
---
-7
--
--
0.80
0.21
12
29
--
--
Wing
Weight
72.56
--
---
9
--
--
Culmen
26.20
Wing
Weight
76.87 -10.46 0.50
122
56
Culmen
32.48
--
123
--
--
--
Weight
9.67
--
Culmen
--
--
--
Wing
--
--
--
Weight
7.50
--
Culmen
--
--
--
Wing
Weight
-8.74
---
-29
--
--
Wing
Weight
67.41
7.41
Culmen
19.60
-0.63
--
88
---
-150
--
--
--
Culmen
17.71
C. thalassinus(B)
Wing
Weight
63.79
5.91
Culmen
23.20
-0.67
--
-0.34
--
21.50
Wing
Weight
Anthracothorax
Wing
Weight
Culmen
viridigula
(A)
50
9
64
--
-0.49
n
2
17
--
1
---
-45
--
62.31
7.46
2.61
0.49
--
31
25
Culmen
24.53
1.70
31
Wing
Weight
65.45
6.71
-0.22
109
11
Culmen
26.14
A. nigricollis
Wing
Weight
66.18
7.00
--
-0.55
109
89
51
Culmen
24.18
--
88
A. veraguensis
Wing
Weight
66.75
--
---
8
--
Culmen
24.57
--
A. dominicus
(A)
Wing
Weight
62.81 1.86
5.66 0.41
42
32
Culmen
24.13
0.75
42
A. mango(A)
Wing
Weight
68.31
7.81
-0.67
21
4
Culmen
26.03
Eulampisjugulaffs
(A)
Wing
Weight
73.21
8.67
Culmen
23.59
E. holosericeus
(A)
Wing
Weight
60.00 2.04
5.60 0.40
Culmen
22.74
2.20
91
Chrysolampis
mosquitus
Wing
Weight
54.57
3.88
-0.39
48
20
--
-0.56
--
21
45
11
45
92
21
Culmen
12.39
Wing
Weight
47.34
2.71
1.09
0.18
54
23
Culmen
10.72
1.18
54
Klaisguimeti
Wing
47.77
--
64
Weight
Culmen
2.54
13.24
--
7
Orthorhyncus
cristatus(A, M)
0.20
48
7
--
64
Abeilliaabeillei
Wing
Weight
-2.67
---
-3
Culmen
--
--
--
Stephanoxis
lalandi
Wing
Weight
-4.03
---
-3
Culmen
--
--
--
Lophornis
ornata
Wing
--
--
--
Weight
2.20
--
Culmen
--
--
--
Wing
--
--
--
Weight
2.13
--
Culmen
--
--
--
L. magnifica
8
3
L. delattrei
Wing
Weight
37.54
--
---
25
--
Culmen
10.89
--
25
L. helenae
Wing
Weight
-2.60
---
-4
Culmen
--
--
--
50
65
84
-6.75
SD
A. prevostii(B)
--
88
28
-8.05
70.55
6.66
Culmen
C. serrirostris
3
Culmen
Wing
Weight
75.20
7.55
3
Wing
Weight
Coh•ff delphinae
Wing
Weight
9
Wing
Culmen
Florisuga
--
73.20
8.32
C. coruscans
--
Wing
Weight
Weight
œ
29
22
24
Phaeochroa
cuvierii
-6.40
acter
58
Wing
Culmen
Species
April1985]
APPENDIX.
Hummingbird
Morphology
andEcology
APPENDIX.
Continued.
Continued.
Char-
Char-
Species
L. adorabilis
Popelairia
langsdorffi
D. longicauda
Chlorestes
notatus
œ
SD
n
Wing
Weight
-2.70
---
-4
Culmen
--
--
--
Wing
--
--
--
Weight
3.00
--
--
--
C. gibsoni
C. ricordii(A)
C. maugaeus
(A, M)
Cyanophaia
bicolor(A)
Culmen
13.85
--
20
Wing
Weight
-3.00
---
-1
Culmen
--
--
--
Wing
Weight
-3.61
---
-14
--
--
--
43.33 -2.97 0.50
Culmen
13.40
--
6
-2.72
---
-5
Culmen
--
--
--
Wing
furcata(A)
T. watertonii
51.50
--
4.23
--
22
1
Culmen
17.20
--
22
Wing
Weight
54.51
4.85
---
14
1
Culmen
17.30
--
14
Wing
Weight
47.53
2.93
---
45
45
Culmen
13.62
--
45
Wing
58.58
--
26
4.55
--
8
--
22
Weight
Wing
Weight
Culmen
16.41
52.61 0.50
52
4.35 0.24
72
20.19
--
36
--
Wing
--
--
Weight
4.50
--
2
Culmen
--
--
--
T. glaucopis
Wing
Weight
-5.32
---
-60
Culmen
--
--
--
T. lerchi
Wing
Weight
-2.00
---
-3
--
--
Panterpe
insignis
Wing
Weight
64.47 -5.69 0.39
68
47
Culmen
20.86
68
Culmen
Damophila
julie
(M)
Lepidopyga
coeruleogularis
(M)
L. goudoti
--
51
51
Culmen
15.17
51
Wing
Weight
48.76 -4.20 0.30
21
3
Culmen
19.40
20
Wing
Weight
50.00 -3.95 0.30
1
4
Culmen
19.10
1
--
H. cyanus
Wing
Weight
Goldmania
violiceps
Goethalsia
bella
Trochilus
polytmus
(A)
Leucochloris
albicollis
Polytmus
guainumbi
•?
SD
n
49.30 -3.53 0.18
1
Culmen
17.00
1
Wing
Weight
57.00 -3.50 0.27
--
1
210
Culmen
17.60
--
1
Wing
Weight
48.72
3.60
---
59
13
Culmen
18.24
--
59
Wing
Weight
-4.00
---
-2
Culmen
--
--
--
Wing
Weight
-3.47
---
-8
--
--
Wing
Weight
--
50.28 -3.63 0.20
26
6
Culmen
18.73
--
26
Wing
Weight
52.94
--
---
5
--
Culmen
17.86
--
5
Wing
Weight
53.90
4.10
---
1
Culmen
19.80
--
1
---
-26
Wing
Weight
-6.45
Culmen
--
--
--
Wing
Weight
-4.70
---
-3
Culmen
--
--
--
Leucippus
chionogaster
Wing
Weight
-3.80
---
-7
--
--
--
Talaphorus
tuczanowskii
Wing
Weight
69.90
6.00
---
1
1
Culmen
22.80
--
1
Wing
Weight
49.00
3.74
Culmen
Amazilia
candida(B, M)
-0.43
Culmen
17.60
Wing
Weight
54.40 2.20
4.72 --
Culmen
--
--
--
A. versicolor
Wing
Weight
-4.00
---
-1
--
--
--
A. fimbriata
Wing
Weight
A. lactea
A. amabilis
A. cyaneotiocta
--
1
4.78
0.48
9
Culmen
21.10
54.20
4.55
Wing
Weight
1
3
62
51.60
Wing
Weight
Culmen
--
1
14
A. chionopectus
Culmen
43.46 -3.31 0.15
--
H. sapphirina
--
Wing
Weight
--
H. eliciae(M)
acter
Culmen
6
3
Wing
Weight
Culmen
Thalurania
--
20
5
Weight
C. swainsonii
(A)
3
40.60 -3.00 0.40
Wing
Weight
Hylocharis
xantusii
(B,M)
H. leucotis
(B)
Wing
Weight
Culmen
Chlorostilbon
aureoventris(M)
Species
acter
Culmen
Discosura
conversii
(M)
265
--
52.45
4.32
Culmen
19.20
Wing
Weight
61.10
5.84
Culmen
19.90
--
1.10
0.10
--
-0.55
--
-0.63
--
1
10
--
26
18
26
1
13
1
266
BROWN
ANDBOWERS
APPENDIX.
Continued.
APPENDIX.
[Auk,Vol. 102
Continued.
Char-
Species
Char-
acter
œ
SD
n
Species
acter
œ
SD
n
A. franciae
Wing
Weight
50.60
5.04
---
1
5
Chalyburabuffonii
Wing
Weight
66.65
6.20
-0.40
45
6
Culmen
23.00
--
Culmen
25.39
--
45
A. leucogaster
Wing
--
--
C. melanorrhoa
Weight
4.90
--
Wing
Weight
66.70
--
---
20
--
--
--
Culmen
23.30
--
20
Wing
Weight
66.70
7.00
---
18
1
Culmen
25.54
--
18
---
-30
Culmen
A. cyanocephala
(B)
Wing
Weight
61.10
5.84
-0.63
6
--
1
13
Culmen
19.90
A. beryllina(B)
Wing
Weight
56.50
4.94
Culmen
19.40
A. saucerrottei
Wing
Weight
53.90
4.49
Culmen
23.16
--
120
Wing
47.00
--
1
4.10
0.39
35
A. tobaci(M)
Weight
Culmen
A. viridigaster
Wing
Weight
A. edward
Wing
Weight
Culmen
18.30
--
1
--
-0.40
1
1
7
--
-0.37
--
1
---
-5
--
--
39
Culmen
19.43
Wing
Weight
58.60
4.71
Culmen
23.40
A. yucatanensis
Wing
Weight
A. tzacatl(B)
Wing
Weight
57.28 -5.03 0.28
Culmen
22.56
A. amazilia
Wing
Weight
52.45 -4.32 0.55
26
18
Culmen
19.20
26
A. violiceps
(B)
Wing
Weight
58.40 -5.87 0.40
1
7
Culmen
21.60
--
1
Wing
Weight
62.25
--
---
2
--
Culmen
18.00
--
Wing
Weight
57.32 -4.27 0.30
120
126
Culmen
21.37
--
114
Wing
48.05
--
21
3.00
--
1
Eupherusa
poliocerca
E. eximia(B)
E. nigriventris
(M)
Weight
Elvirachionura(M)
E. cupreiceps
(M)
Microchera
albocoronata
(M)
1
11
--
1
-3.57
---
-7
--
--
--
--
--
--
--
--
--
--
Weight
7.96
--
268
Culmen
--
--
--
clemenciae
Wing
Weight
67.23 -6.73 0.87
Culmen
20.92
L. viridipallens
Wing
Weight
L. hemileucus
Adelomyia
melanogenys
Clytolaema
rubricauda
15.43
Culmen
15.75
--
45
Wing
Weight
46.58
3.00
---
26
13
Culmen
15.12
--
26
Wing
Weight
40.77 -2.71 0.20
17
8
Culmen
12.24
16
--
--
13
---
-11
--
--
61.69
5.60
---
18
1
Culmen
20.68
--
20
40
20
Culmen
21.80
Wing
Weight
50.73 0.50
3.77 0.31
Culmen
16.10
Wing
18
62.15 -5.64 0.27
--
--
--
--
Weight
7.32
--
Culmen
--
--
1
--
5
--
Wing
Weight
Culmen
--
--
--
Heliodoxa
Wing
--
--
--
Weight
7.74
--
Culmen
--
--
rubinoides
58.80 0.80
6.30 0.10
26
35
Polyplancta
aurescens
8
--
Wing
Weight
H. jacula
Wing
Weight
70.91 -8.30 0.40
47
10
Culmen
23.22
47
Wing
Weight
72.61 -6.80 0.46
67
38
Culmen
30.68
Culmen
Eugenes
fulgens
(B)
Sternoclyta
cyanopectus
Topazapella
--
--
--
24
32
--
--
67
Wing
--
--
--
Weight
9.06
--
17
Culmen
--
--
--
--
--
--
Wing
Weight
Culmen
Oreotrochilus
melanogaster
66.40 0.70
7.38 0.18
8
8
H. leadbeateri
21
45
16
-5.42
13
26
Wing
Weight
L. castaneoventris Wing
Weight
1
49.05
2.99
-7.29
L. amethystinus
(B)
36
Wing
Weight
--
--
Wing
--
26
105
Culmen
-0.21
Culmen
Lampornis
Culmen
39
58
A. rutila (B)
Culmen
Wing
Weight
--
52.40 -4.70 0.20
-0.54
Aphantochroa
cirrochloris
1
120
139
-6.70
--
C. urochrysia
12.12
--
---
6
--
Wing
Weight
65.00
4.40
---
1
2
Culmen
17.70
--
i
April1985]
APPENDIX.
Hummingbird
Morphology
andEcology
Continued.
APPENDIX.
Continued.
Char-
Species
O. estella
Patagona
gigas
Aglaeactis
cupripennis
Lafresnaya
lafresnayi
œ
SD
n
Wing
Weight
66.40
8.11
-0.39
1
37
Culmen
18.50
Coeligena
coeligena
C. wilsoni
C. helianthea
C. lutetiae
C. violifer
C. iris
Ensiferaensifera
Sephanoides
sephanoides
Boissonneaua
fiavescens
B. matthewsii
B.jardini
H. exortis
1
10
36.30
80.10 -7.55 0.59
Culmen
20.00
--
1
Wing
Weight
63.60
4.50
---
1
2
25.60
--
Wing
Weight
--
30.50
Wing
Weight
70.76 0.60
6.75 0.10
Culmen
26.00
Wing
Weight
1
23
--
27
34
-7.00
---
-1
--
--
--
Wing
Weight
73.27 0.80
7.13 0.20
19
19
Culmen
31.20
--
1
Wing
Weight
70.10
8.17
---
1
3
Culmen
29.40
--
1
Wing
Weight
78.80
8.13
---
1
3
Culmen
33.00
--
1
Wing
Weight
74.78 1.60
7.44 0.30
Culmen
37.50
--
1
Wing
Weight
77.90
8.00
---
1
2
Culmen
28.00
--
1
Wing
Weight
77.20 -11.83 0.50
6
7
1
3
Culmen
--
--
--
Wing
Weight
-5.00
---
-1
Culmen
--
--
--
Culmen
16.30
-0.90
1
8
--
1
-6.65
---
-4
Culmen
--
--
--
Wing
Weight
-8.20
---
-2
--
--
61.20 0.50
5.27 0.10
Culmen
--
--
--
Wing
Weight
-4.00
---
-1
Culmen
--
--
--
SD
--
--
5.33
--
Culmen
--
--
--
Wing
Weight
-5.00
---
-1
--
--
--
Wing
Weight
E. luciani
Culmen
21.40
--
--
--
Weight
5.80
--
Culmen
--
--
E. mosquera
Haplophaedia
aureliae
n
Weight
Culmen
58.90
5.17
---
Culmen
19.10
--
Wing
Weight
67.35 0.15
6.26 0.16
Wing
Wing
Weight
58.75 0.50
5.46
0.17
3
1
3
1
4
8
1
--
1
--
41
37
20.32
Ocreatus
underwoodii
(M)
Wing
Weight
44.14
3.03
Culmen
13.00
--
1
Lesbia
victoriae
Wing
Weight
59.10
5.10
---
1
1
Culmen
15.00
--
1
Wing
Weight
49.70
3.75
---
1
6
Culrnen
10.40
--
1
Wing
Weight
57.50
5.88
---
1
4
Culmen
19.20
--
1
Wing
Weight
56.80
5.03
---
1
4
Culmen
20.00
L. nuna(M)
Sappho
sparganura
Polyonymus
caroli
Metalluraphoebe
Wing
Weight
Culmen
M. theresiae
M. eupogon
Wing
Weight
-5.35
--
--
--
Culmen
0.50
0.14
1
-2
--
--
---
Culmen
11.80
--
Wing
Weight
59.60 0.80
4.60 0.10
--
Wing
Weight
Culmen
11.80
M. tyrianthina
Wing
Weight
55.24 0.11
3.52 0.33
Chalcostigma
stanleyi
Oxypogon
guerinii
Aglaiocercus
kingi
55.60
4.56
8
19
--
64.10
5.00
--
15
---
M. williami
•
26
28
g
Wing
Culmen
Wing
Weight
Wing
Weight
acter
Eriocnemis
vestitus
1
1
74.00
8.05
H. squamigularis
1
3
6
--
Wing
Weight
H. viola
1
105.70 8.00
11.17 0.50
Culmen
Culmen
Heliangelus
amethysticollis
-1.29
Species
1
Wing
Weight
Culmen
C. torquata
132.30
20.24
--
Culmen
Culmen
Pterophanes
cyanopterus
Char-
acter
Wing
Weight
267
-0.20
--
1
2
1
10
10
--
1
5
1
10
21
Culrnen
10.10
--
1
Wing
Weight
66.70
5.84
---
1
11
Culmen
9.80
--
1
Wing
61.40
--
1
Weight
5.00
--
3
Culmen
7.50
--
1
Wing
Weight
56.85 0.60
4.76 0.58
Culmen
13.70
--
4
17
1
268
BROWN
AND
BOWERS
APPENDIX.
Continued.
APPENDIX.
Continued.
Char-
Species
Char-
acter
œ
SD
A. emmae
Wing
Weight
66.90
5.70
---
Culmen
14.40
--
1
Oreonympha
nobilis
Wing
Weight
-9.00
---
-1
Culmen
--
--
--
Wing
--
--
--
Weight
3.00
--
Augastes
scutatus
n
1
1
Culmen
--
--
--
Wing
Weight
-4.00
---
-2
--
--
--
Schistes
geoffroyi
Wing
Heliothryxbarroti
51.40 1.00
Weight
3.62
Culmen
--
0.22
--
Wing
Weight
68.60
5.43
Culmen
--
--
--
Heliactincornuta
Wing
--
--
--
Weight
7.80
--
4
mirabilis
Heliomaster
constantii(B)
H. longirostris
(B)
Weight
76
9
4
7
--
1
3.00
--
1
14.70
--
1
Wing
68.00
--
1
7.30
--
2
Culmen
41.00
--
1
Wing
59.24
--
16
6.55
0.68
8
Weight
Culmen
--
16
-5.00
---
-1
Culmen
--
--
--
Wing
--
--
--
5.00
--
H. squamosus
Wing
Weight
H. furcifer
Weight
Culmen
35.18
--
--
Culmen
17.40
--
1
Thaumastura
cora
Wing
Weight
36.10
2.00
---
1
1
Culmen
12.80
--
1
Wing
Weight
36.10
2.40
---
-2
Culmen
Philodice
mitchellii
Dorichaenicura
1
4
--
--
Wing
Weight
41.50
3.29
---
8
21
Culmen
19.30
--
8
Wing
37.80
--
1
3.13
0.10
3
Weight
--
19
12
Culmen
13.11
--
19
Calothorax
lucifer
(B)
Wing
Weight
39.28
2.80
---
19
3
Culmen
21.05
--
19
C. pulcher(B)
Wing
Weight
37.08 -2.93 0.30
Culmen
18.07
Archilochus
colubris(T,B, M)
Wing
Weight
41.05
3.19
Calliphlox
Culmen
15.00
--
1
35.26
2.40
---
29
1
Culmen
20.41
--
29
17.96
Wing
44.21
Weight
3.20
19.73
--
-0.30
--
1.19
0.19
0.55
12
3
12
26
35
27
119
76
137
Wing
--
--
--
amethystina
Weight
2.43
--
24
--
--
--
Mellisugaminima
(A)
Wing
Weight
M. helenae
(A)
Culmen
37.10 -2.43 0.10
17
6
Culmen
10.45
--
17
Wing
31.61
--
13
--
--
Weight
--
Culmen
10.76
Calypteanna
(T, B, M)
Wing
Weight
49.56
4.25
Culmen
19.06
0.95
71
C. costae
(T, B, M)
Wing
Weight
44.45
3.08
0.82
0.29
76
58
Culmen
17.71
0.56
Stellulacalliope
(T,B,M)
Wing
Weight
40.31
2.66
0.80
0.32
71
51
Culmen
15.54
0.70
71
Atthis heloisa(B)
Wing
Weight
35.12
--
---
25
--
Culmen
11.77
--
25
Myrtis fanny
Wing
Weight
40.80
2.30
---
Culmen
19.20
--
1
Acestrura
mulsanti
Wing
Weight
39.55
3.75
---
2
6
Culmen
16.20
--
1
A. heliodor
Wing
Weight
32.40
--
---
20
--
Culmen
15.00
--
Selasphorus
platycercus
Wing
Weight
49.68
3.43
-0.31
Culmen
17.51
0.45
145
Wing
42.30 0.84
326
(T,B,M)
Wing
Weight
Culmen
Culmen
--
Wing
Weight
C. bryantae(M)
---
-0.15
1
Rhodopis
vesper
(M)
Calliphlox
evelynae
49.30
3.88
34.33
2.23
--
38.50
Culmen
Weight
Wing
Weight
76
0.30
0.30
--
Tilmaturadupontii
(B)
(T, B, M)
H. aurita
--
n
13
17.98
Wing
SD
--
Culmen
Culmen
œ
A. alexandri
66.15
5.47
--
acter
8
Wing
Weight
Loddigesia
-0.20
Species
1
A. lumachellus
Culmen
[Auk,VoL102
S. rufus(T, B, M)
Weight
S. sasin(T, B, M)
S.fiammula(M)
3.36
--
0.97
0.33
0.31
13
71
79
107
1
1
20
58
86
68
Culmen
16.57
0.67
326
Wing
Weight
39.77
3.16
0.83
0.21
111
38
Culmen
17.20
0.60
111
Wing
41.57
--
2.47
--
3
--
40
Weight
Culmen
12.90
40
April1985]
APPENDIX.
Hummingbird
Morphology
andEcology
Continued.
APPENDIX.
Continued.
Char-
Species
S. torridus
S. simoni
269
Char-
acter
œ
SD
n
Wing
Weight
40.50
! .90
---
8
!
Culmen
!!.89
--
8
Wing
Weight
38.68
--
---
!!
--
Culmen
!0.59
--
!!
Species
acter
œ
SD
n
S. ardens
Wing
Weight
40.17
--
---
19
--
Culmen
!2.54
--
S. scintilla
Wing
Weight
34.60 -2.!5 0.!6
43
39
Culmen
!!.93
43
--
!9
[From "Sexualselectionand the nesting of birds," by J. A. Allen (!885 Auk 2: 129-!39):
"Mr. Wallace, and after him Mr. Dixon and others,
in discusing[sic]the question How do young birds
learn to build
their first nest? claim that 'instinct'
has
nothing to do with the matter,--that they learn by
observationand are guided by memory! Says Mr.
Wallace: 'It has, however, been objectedthat observation, imitation, or memory, can have nothing to do
with a bird's architecturalpowers,becausethe young
birds which in Englandare born in May or June,will
proceedin the following April or May to build a
nests[sic]asperfectand as beautiful as that in which
it was hatched, although it could never have seen
one built. But surely the young birds beforethey left
the nesthad ampleopportunitiesof observingitsform,
its size,its position,the materialsof which it was con-
imitation,memory,and hereditary habit, 'play the minor parts.' 'To credit birds,' he says,'with such marvellous power as blind and infallible instinct in
building their nestswould be to place them far beyond man himself in intelligence,and allot to them
a faculty which is superhuman.... A bird's intellectual powers advance towards maturity much more
quickly than in the human species.A young bird
three or four daysold is capableof considerablepowers of memory and observation,and during the time
that elapsesin which it is in the nest it has ample
opportunityof gaining an insight into the architecture peculiar to its species.It seesthe position of the
nest, it notesthe materials,and when it requires one
for itself,is it sovery extraordinarythat,profitingby
structed, and the manner in which those materials
such experience, it builds one on the same plan?
were arranged. Memory would retain these observationstill the following spring,when the materials
would comein their way during daily searchfor food,
and it seemshighly probable that the older birds
would begin building first, and that those born the
preceding summer would follow their example,
Again, birds often return to the place of their birth
the following season,and possiblyseethe old home
many times ere they want one for themselves.This,
aided by the strong hereditary impulse to build a
nest similar to the one in which they were born,
inherited from their parents,aidsthem in their task.'
This reasoning, I am free to confess,strikes me, to
learningfrom themhow the foundationsof the nest
were laid and the materialsput together. Again we
have no right to assumethat young birds generally
pair together,'etc.Mr. Dixon restatesthe casein much
the sameway. Alluding to 'blind instinct' as a factor
in the case, he says: 'To credit the bird with such
instinct, which because it seems so self-evident is
taken to be matter of fact, is to admit that it possesses
intellectualpowersinfinitely superiorto thoseof man;
whilst the evidencethat can be gatheredon the subject all goesto show that its intellectual powers are
of preciselythe samekind asman's,but someof them,
of course,are infinitely inferior in degree, whilst
othersare unquestionablysuperior.'He assumesthat
say the least,as extraordinary!A degreeof mental
power,at leastof memoryand of imitation,is ascribed
to young birds which is not only 'superhuman,'but
of which there is neitherproof,nor evenpossibility
of proof.Mr. Dixon hasthe 'three or four daysold'
nestlingtaking note of and memorizing its surroundings before,in the caseof the higher Oscines,it has
thepowerto evenopenitseyes!Yet with all this ascribed
precosityand keennessof observation,and this wonderfulpowerof memoryandimitationin youngbirds,
Mr. Dixon finds it neccessary[sic] to call in the aid
of 'a strong hereditary impulse to build a nest similar
to the one in which they were born,' which is more
(continued
on p. 322)