Content uploaded by Avery Russell
Author content
All content in this area was uploaded by Avery Russell on Aug 16, 2021
Content may be subject to copyright.
SYMPOSIUM
Bene“fit” Assessment in Pollination Coevolution: Mechanistic
Perspectives on Hummingbird Bill–Flower Matching
Alejandro Rico-Guevara ,*
,†,1
Kristiina J. Hurme,* Rosalee Elting*
,†
and Avery L. Russell
‡
*Department of Biology, University of Washington, 24 Kincaid Hall, Seattle, WA 98105, USA;
†
Division of Ornithology,
Burke Museum of Natural History and Culture, 4300 15th Ave NE, Seattle, WA 98105, USA;
‡
Department of Biology,
Missouri State University, 910 S John Q Hammons Pkwy, Springfield, MO 65897, USA
From the symposium “Physical mechanisms of behavior” presented at the virtual annual meeting of the Society for
Integrative and Comparative Biology, January 3– 7, 2021.
1
E-mail: colibri@uw.edu
Synopsis One of the reasons why flowering plants became the most diverse group of land plants is their association
with animals to reproduce. The earliest examples of this mutualism involved insects foraging for food from plants and,
in the process, pollinating them. Vertebrates are latecomers to these mutualisms, but birds, in particular, present a wide
variety of nectar-feeding clades that have adapted to solve similar challenges. Such challenges include surviving on small
caloric rewards widely scattered across the landscape, matching their foraging strategy to nectar replenishment rate, and
efficiently collecting this liquid food from well-protected chambers deep inside flowers. One particular set of convergent
traits among plants and their bird pollinators has been especially well studied: the match between the shape and size of
bird bills and ornithophilous flowers. Focusing on a highly specialized group, hummingbirds, we examine the expected
benefits from bill–flower matching, with a strong focus on the benefits to the hummingbird and how to quantify them.
Explanations for the coevolution of bill–flower matching include (1) that the evolution of traits by bird-pollinated plants,
such as long and thin corollas, prevents less efficient pollinators (e.g., insects) from accessing the nectar and (2) that
increased matching, as a result of reciprocal adaptation, benefits both the bird (nectar extraction efficiency) and the plant
(pollen transfer). In addition to nectar-feeding, we discuss how interference and exploitative competition also play a
significant role in the evolution and maintenance of trait matching. We present hummingbird–plant interactions as a
model system to understand how trait matching evolves and how pollinator behavior can modify expectations based
solely on morphological matching, and discuss the implications of this behavioral modulation for the maintenance of
specialization. While this perspective piece directly concerns hummingbird–plant interactions, the implications are much
broader. Functional trait matching is likely common in coevolutionary interactions (e.g., in predator–prey interactions),
yet the physical mechanisms underlying trait matching are understudied and rarely quantified. We summarize existing
methods and present novel approaches that can be used to quantify key benefits to interacting partners in a variety of
ecological systems.
Introduction to plant–pollinator
functional trait matching
A fundamental question in ecology is how speciali-
zation evolves and is maintained. Plant–pollinator
mutualisms are a model system for the study of spe-
cialization and both plants and their animal pollina-
tors can vary from being highly generalized to being
strongly specialized. For example, plants with a gen-
eralist pollination strategy might receive pollination
services from a wide range of pollinator taxa (e.g.,
Fig. 1A; Hern
andez-Conrique et al. 2007), while
highly specialized plants often restrict access to a
few pollinator taxa (e.g., Fig. 1B; Sargent and
Vamosi 2008;Soteras et al. 2018). Specialization on
the part of the plant is thought to arise due to the
costs of associating with less efficient visitors and
selection by the most effective pollinator (Pauw et
al. 2020). Animals vary in their effectiveness as
ßThe Author(s) 2021. Published by Oxford University Press on behalf of the Society for Integrative and Comparative Biology.
All rights reserved.
For permissions please email: journals.permissions@oup.com.
Integrative and Comparative Biology
Integrative and Comparative Biology, pp. 1–15
doi:10.1093/icb/icab111 Society for Integrative and Comparative Biology
Downloaded from https://academic.oup.com/icb/advance-article/doi/10.1093/icb/icab111/6288459 by Missouri State University user on 12 August 2021
pollinators and visits by ineffective pollinators can
carry heavy fitness costs for a plant, including exces-
sive pollen loss (the male plant gamete), increased
heterospecific pollen transfer (Ashman and Arceo-
G
omez 2013), consumption of nutritive food
rewards (e.g., nectar) without pollinating (i.e., rob-
bing), and even cause damage to the flower (Rojas-
Nossa et al. 2021). These costs might drive the evo-
lution of floral traits that filter less effective pollina-
tors. Similarly, more effective pollinators (e.g.,
transporting more conspecific pollen) shape selection
for floral traits that further enhance their effective-
ness as pollinators. In some cases, highly specialized
plants are pollinated by animals that have in turn
specialized to feed on that plant’s food rewards
(Brown and Kodric-Brown 1979;Serrano-Serrano
et al. 2017). Such interactions, where both plant
and pollinator are reciprocally specialized, create a
tighter coevolutionary association that frequently
involves the evolution of specialized traits that fur-
ther enhance specialization (Proctor et al. 1996).
Specialized traits that enhance reciprocal speciali-
zation are often matched between plant and pollina-
tor (i.e., “trait matching”). At broad taxonomic
scales, selection favors similarity in traits across
groups of pollinators and plants (Johnson and
Anderson 2010;Medeiros et al. 2018), and trait
matching is frequently studied in the context of pair-
wise interactions between plant and pollinator spe-
cies. For example, the length of hawkmoth tongues
may closely match the length of floral tubes
(Johnson et al. 2017) in a community of hawkmoths
pollinating long-tubed flowers. A classic case of pair-
wise trait matching occurs between Angraecum ses-
quipedalia, a Madagascaran orchid with an extremely
long nectar spur, and its presumed coevolved spe-
cialist pollinator, the long-tongued sphinx moth,
Xanthopan morganii (Darwin 1862;Johnson and
Anderson 2010;Arditti et al. 2012). Trait-matching
is often inferred through comparing the morphology
of plant and pollinator, and is assumed to be func-
tional (Castellanos et al. 2003;Maglianesi et al. 2014;
Weinstein and Graham 2017). That is, trait matching
is predicted to simultaneously improve the precision
of pollen placement on the pollinator and subse-
quent deposition on the stigma and enhance the ex-
traction of floral rewards by pollinators (Castellanos
et al. 2003). In addition, trait-matching may limit
interspecific competition, both for the pollinator,
by limiting which plant species a given pollinator
may profitably visit, and simultaneously for the
plant, by reducing heterospecific pollen transfer
(Waser and Fugate 1986;Ashman and Arceo-
G
omez 2013;Maglianesi et al. 2014;Fonseca et al.
2016). From the plant’s perspective, important fac-
tors to consider include sequence of plants/flowers
visited by the pollinator, the visitation timing and
distance between those flowers, the number of pol-
linators and their morphologies, the adhesive prop-
erties, size, and amount of pollen grains deposited
per floral visit, and the degree of reproductive in-
compatibility of the plant species involved. When a
pollinator visits a flower, it can already be carrying
and could deposit heterospecific pollen (which has
the potential to reduce plant fitness, Waser and
Fugate 1986;Fonseca et al. 2016). Furthermore,
any heterospecific pollen left on the pollinator’s
body can reduce space available for pollen from a
conspecific plant, reducing pollination probability.
Finally, conspecific pollen can be lost and wasted
by pollinator delivery to a heterospecific plant
(Morales and Traveset 2008).
Avian adaptations to collect floral nectar have
been summarized in the appropriately named
“syndrome of anthophily” (Stiles 1981). Features of
this syndrome include a bill that is usually slender
(e.g., Fig. 1), often long and/or curved, and a bifur-
cated tongue tip (grooved, fringed, and/or capable of
rolling into a tube) that is extendable beyond the bill
tip. In contrast with other nectar-feeding animals
(like some bats that store their long tongues deeper
inside their bodies; Muchhala 2006), specialized nec-
tarivorous birds not only have evolved elongated
tongues to reach the nectar, but also long bills to
hold them and allow them to be inserted inside
the nectar chamber with enough precision to collect
the liquid reward efficiently (Stiles 1981). In fact, bill
length and shape are considered key to functional
trait matching, because while tongues may extend
far beyond the bill tip, in most birds, tongues have
motion control only at their base and are too flimsy
to traverse barriers within the flower on their own
(e.g., Rico-Guevara et al. 2019). Here, we focus on
hummingbirds, the most speciose group of anthoph-
ilous vertebrates. Hummingbirds exhibit a wide array
of traits associated with their dependence on nectar-
feeding (Stiles 1981;Fleming and Muchhala 2008),
which allow them to feed on flowers well enough to
make their living out of small volumes of nectar.
Presumed adaptations to nectarivory in humming-
birds range from the traditionally recognized ones
(such as elongated bills) that evolved more than 30
million years ago (Mayr 2004), to specialized tongue
elastic properties described recently (Rico-Guevara et
al. 2015). Hummingbird feeding apparatus present a
variety of minute structures at their tips that are
believed to be adaptations to improve the collection
and transport of nectar (Rico-Guevara and Rubega
2A. Rico-Guevara et al.
Downloaded from https://academic.oup.com/icb/advance-article/doi/10.1093/icb/icab111/6288459 by Missouri State University user on 12 August 2021
2011,2017). We discuss the evolution of presumed
functional trait matching between hummingbird bills
and flowers and propose tools to study this interac-
tion from biomechanical and behavioral perspectives.
Similar to their pollinators, hummingbird-
pollinated plants have repeatedly evolved a suite of
traits that characterize them as throchiliphilous
(plants that use hummingbirds as pollination vec-
tors). Such convergent traits deviate to varying
degrees from the ancestral states of entomophilous
(insect-pollinated) plants and include larger flowers,
increased correspondence with the bills of the hum-
mingbirds that are pollinating those flowers, fine-
tuned placement of reproductive organs to contact
hummingbird surfaces (different areas in the bill,
head, and throat), larger volumes of more dilute
nectar, overall larger caloric rewards, reduced or ab-
sent nectar amino acids, reddish coloration, reduced
scent cues, and in some cases, hanging (pendant)
flowers that are reached via hovering, or inflorescen-
ces shaped to support the weight of clinging birds
(Stiles 1978; Mart
ınez del Rio et al. 2001;Rocca and
Sazima 2010;Pauw 2019). Some species may explore
other food sources when there is intense competi-
tion, for example, recent studies have shown that
hummingbirds include nonthrochiliphilous plants
in their diet (e.g., Bee hummingbirds: Dalsgaard et
al. 2008,2009;Waser et al. 2018;Wessinger et al.
2019), demonstrating the importance of studying be-
havioral plasticity and expanding studies to all the
potential sources of nectar in a given environment.
Throchiliphilous plants (approximately 7,000 spp.)
have likely independently evolved hundreds of times
across 68 families (Thomson and Wilson 2008;
Abrahamczyk and Renner 2015), highlighting the
importance of hummingbirds as pollinators in a va-
riety of ecosystems.
Traits that are likely involved in functional trait
matching, such as bill and floral morphologies are
well characterized. Similarly, the interactions between
a hummingbird and ornithophilous plant species in
pollination networks are also well known. The key
gap in our knowledge is understanding how the
expected benefits from functional trait matching
are achieved. In this perspective piece, we discuss
how hummingbird behavior and morphological
matching between the feeding apparatus of hum-
mingbirds and the floral structures of plants interact
and modulate the benefits for both the plant and
pollinator. From the hummingbird perspective, we
discuss adaptations that enhance the extraction of
floral rewards, focusing on nectar drinking efficiency,
as this is the most prominent link between the evo-
lution of nectarivory in birds and coevolution with
ornithophilous plants. To understand the general
benefits of trait matching, we present novel methods
that could allow researchers to quantify the benefits
of trait matching. The tools we present provide a
window into how the hummingbirds interact with
the flowers at the moment of the floral visit,
highlighting the importance of considering behavior
as an integral part of the study of trait matching
between hummingbirds and the plants they pollinate.
Evolution of trait matching in
hummingbird–plant interactions
Determining whether trait matching is functional
(i.e., benefits both hummingbird and plant) depends
in part on understanding how mutualists may be
driving coevolution. The close physical match be-
tween slender hummingbird bills and the elongated
and narrow-entranced corollas of hummingbird pol-
linated plants is frequently interpreted as sufficient
evidence of reciprocal adaptation between these two
mutualist groups (see Cronk and Ojeda 2008).
Indeed, there is a strong evidence that hummingbird
pollination drives the diversification of
hummingbird-pollinated plants and the evolution
of their floral morphology (Temeles et al. 2002;
Pauw 2019;Wessinger et al. 2019).
A
B
Fig. 1. Strategies in hummingbird–plant coevolutionary systems.
(A) Generalist Indigo-capped hummingbird (Saucerottia cyanifrons)
visiting a flower of a leguminous tree (Calliandra spp.), a plant that
also exhibits a generalist strategy (Hern
andez-Conrique et al.
2007). (B) Specialist Black-throated mango (Anthracothorax nigri-
collis) visiting a specialized throchiliphilous plant (Aphelandra spp.).
Photos taken at the Colibr
ı Gorriazul Research Station, near
Fusagasug
a, Colombia, courtesy of Ricardo Zarate.
Bene“fit” in bill–corolla coevolution 3
Downloaded from https://academic.oup.com/icb/advance-article/doi/10.1093/icb/icab111/6288459 by Missouri State University user on 12 August 2021
However, floral traits that appear to be matched
with hummingbird pollination may also be to some
extent a consequence of historical contingency,
rather than having evolved to match exclusively
within the plant–hummingbird association.
Hummingbird pollination is thought to be rarely an-
cestral and transitions from insect pollination to
hummingbird pollination are the norm (Tripp and
Manos 2008;Serrano-Serrano et al. 2017;Dellinger
et al. 2019;Wessinger et al. 2019). In transitions to
hummingbird pollination, morphological traits asso-
ciated with hummingbird pollinated flowers might
therefore evolve as a function of increasing exclusion
of less efficient insect pollinators (and release from
the interspecific competition via heterospecific pollen
transfer) when hummingbird pollination is assured
(Arcos et al. 2018). Experimental studies provide
substantial support for this interpretation (Mackin
et al. 2021). Elongated and narrow-entranced corol-
las reduce the efficiency of insects as pollinators and
also reduce insect pollinator preference for these
flowers, making them less able to handle the flower
and extract nectar (Castellanos et al. 2004;Gegear
et al. 2017;Arcos et al. 2018). Similarly, transitions
from hummingbird pollinated to insect pollinated or
to mixed pollination systems exhibit shifts toward
shorter and wider-entranced corollas that enhance
insect pollination but have no effect on humming-
bird pollination effectiveness (Tripp and Manos
2008;Arcos et al. 2018).
Altogether, the evidence suggests that the match
between hummingbird bills and corollas may be
driven in large part not by coevolution between
hummingbirds and throchiliphilous plants (in which
both bill and flower shape evolve to match each
other) but by selection on floral shape to filter out
less effective insect pollinators (evolving narrow and
long corollas and other filtering mechanisms).
Hummingbirds are more effective pollinators even
of many insect-pollinated plants (Thomson and
Wilson 2008), and so selection favors floral mor-
phology that excludes insect visitors when more ef-
fective hummingbird pollinators are present. In turn,
these elongated flowers with narrow entrances drive
selection on hummingbird bills to be thinner and
longer to match floral morphology. Once this prom-
inent hummingbird hallmark, the slender bill,
evolved, it became a historical contingency that was
likely perpetuated through “guild coevolution” (see
below). Insect exclusion seems to have driven the
evolution of narrow flowers and subsequent slender
bills; however, it seems unlikely that plants that have
recently transitioned to rely on hummingbird polli-
nators exert strong selection on hummingbird bill
shape. Instead, it is more likely that associated flow-
ers evolve to “fit the bill” (Wilson et al. 2006;Arcos
et al. 2018), although rare cases of strong pairwise
coevolution undoubtedly occur (Stein 1992;
Abrahamczyk et al. 2014;Lagomarsino 2015).
Most hummingbird species have bills likely capa-
ble of extracting nectar from most trochiliphilous
plants, regardless of shared coevolutionary history
(aka guild coevolution) (Cotton 1998). For example,
the Green-backed firecrown (Sephanoides sephanio-
des) a hummingbird only found in South American
latitudes, would be able to feed on the flowers of
plants that originated and coevolved with North
American hummingbird species, due to similar selec-
tive pressures where both hummingbirds and thro-
chiliphilous plants occur. In all hummingbird–plant
assemblages, there are common inefficient pollina-
tors (like short-tongued bees) that need to be re-
stricted from crawling inside the flower to access
the nectar. This is the consensus explanation for
why tubular flowers have narrow entrances (too nar-
row for insect bodies) and elongated corollas (too
long for insect mouthparts to access e.g.,
Rodr
ıguez-Giron
es and Llandres 2008;Dalsgaard
et al. 2009). While there are many other insect pol-
linators that could potentially still reach the nectar,
hummingbird-pollinated plants have additional bar-
riers at the entrance of the nectar chamber that pre-
vent even long insect mouthparts from accessing the
nectar. For example, microtrichia on internal flaps
projecting from the internal corolla walls require suf-
ficient physical strength to be surpassed; humming-
birds, but not insects, can push their mouthparts
through, specifically, their bill tips (Wolf et al.
1972,1975).
In addition, independent appearances of trochi-
liphily among plants (see Introduction to plant–pol-
linator functional trait matching section) are
common (Wessinger et al. 2019). Plant clades that
switch to be hummingbird-pollinated frequently
evolve with local hummingbirds feeding on other
co-occurring trochiliphilous plants (Serrano-Serrano
et al. 2017). These new trochiliphilous clades con-
verge on similar floral traits to take advantage of
existing hummingbird pollinators, thus also resulting
in selection for these hummingbirds to maintain
existing bill morphology to sustain efficient nectar
extraction. This pattern of evolution seems to be
the norm in hummingbird–plant assemblages
(Temeles et al. 2002;Maglianesi et al. 2014;
Weinstein and Graham 2017). Nonetheless, extreme
cases of bill–corolla matching do occasionally occur,
in which hummingbirds have evolved uncommon
bill morphologies that often match the lengths and
4A. Rico-Guevara et al.
Downloaded from https://academic.oup.com/icb/advance-article/doi/10.1093/icb/icab111/6288459 by Missouri State University user on 12 August 2021
curvatures of the flowers they feed on. Examples of
this include the Sword-billed hummingbird (Ensifera
ensifera) with an exceptionally long bill that matches
the elongated corolla of the Passion flowers
(Passiflora spp.) they visit (Lindberg and Olesen
2001;Abrahamczyk et al. 2014), and the Sicklebill
hummingbirds (Eutoxeres spp.) with exceptionally
curved bills that match the curvature of the
Heliconia and Centropogon flowers they feed on
(Stein 1992;Lagomarsino 2015). These extreme cases
of bill–corolla matching are likely cases of runaway
reciprocal exploitation, in which coevolutionary
trends of functional trait matching go beyond what
is strictly required to reduce competition on both
the bird and the plant sides, if the benefits from a
tighter match among the assemblage of species con-
tinue to be advantageous.
Physical components of hummingbird–
plant trait matching and pollination
Our understanding of the selective forces at play
during the evolution of bill–corolla matching is
greatly improved by considering the physical mech-
anisms that determine the benefits of stronger
matching (Box 1 ). While in this section, we primar-
ily consider trait-matching benefits at the level of
isolated bird–plant coevolution, in the next section,
we discuss interference competition and its potential
to influence bill shape evolution (thus also affecting
coevolutionary processes, see Behavioral components
of hummingbird–plant trait matching section). Thus,
the mechanistic principles we describe here can be
transferred to the level of an assemblage of compet-
ing pollinators and plants, providing a more com-
prehensive explanation for how trait matching
evolves.
Part of the rationale behind the bill–corolla
matching coevolution explanation is that humming-
bird individuals of a given species (or sex) should
obtain higher net energy gain while feeding on flow-
ers matching their bill length and shape. One com-
plication of this explanation comes from the fact that
hummingbirds have tongues that can be extended up
to two times their bill length (Rico-Guevara 2017),
and it is unknown to which degree this protrusion
ability varies across species, especially during nectar
feeding. For example, of the two coexisting species in
which one has a bill that is half as long as the other
(or three-fourths as long as in Fig. 1), an individual
from the short-billed species, by extending its
tongue, could potentially reach the nectar in flowers
that “match” the long-bill species better. In fact,
many species of hummingbirds visit flowers with
corollas longer than their bills (Dalsgaard et al.
2021). Which begs the question, is the difference in
bill length sufficient to result in niche partitioning?
Three factors need to be considered here. (1)
Hummingbird tongues are mostly inert structures
that only have motion control at their base (e.g.,
Rico-Guevara et al. 2019); thus, when the tongue is
extended inside the flower, it behaves like a thread
that could get stuck to one of the sides of the co-
rolla, bent among floral reproductive structures, and
other vicissitudes. Hummingbird bills are rigid struc-
tures that guide the tongue through the length of the
corolla. (2) Along the same line, many hummingbird
pollinated flowers have nectar chamber barriers that
need to be passed to access the liquid reward.
Hummingbird tongues are too flimsy to transverse
these obstacles on their own, therefore, bill tips also
provide the final push to access the nectar deep in-
side the flower. Finally, (3) longer bills, by probing
deeper inside corollas, achieve smaller distances be-
tween bill tips and nectar than shorter bills. Smaller
bill tip–nectar distances yield greater licking rates
(Ewald and Williams 1982) because the tongue needs
to travel a shorter distance while reciprocating to
collect the liquid, and this, in turn, results in higher
nectar extraction efficiency (Hainsworth 1973;
Hainsworth and Wolf 1976;Montgomerie 1984;
Grant and Temeles 1992;Temeles and Roberts
1993;Temeles 1996). Under this lens, with every-
thing else being equal, bills with a higher match to
the flower shape are expected to be more efficient in
terms of nectar intake rate (Wolf et al. 1972;Temeles
and Kress 2010). However, tests of this hypothesis
have produced conflicting results. For example, we
would expect that long-billed species are more effi-
cient on long-tubed flowers and short-billed hum-
mingbirds are more efficient on short-tubed
flowers, but experimental results do not support
the second prediction (e.g., Hainsworth 1973;
Montgomerie 1984;Temeles and Roberts 1993;
Temeles 1996). In fact, under experimental condi-
tions, longer-billed birds feed more quickly from
longer flowers than shorter-billed birds, but
shorter-billed birds do not feed more quickly from
shorter flowers than longer-billed ones (Temeles
1996;Temeles et al. 2002,2009). If long bills can
also reach the rewards inside short corollas, mini-
mizing bill tip—nectar distances, and achieving
higher energy intake rates (Montgomerie 1984),
why does not selection always favor longer bills?
Drawbacks associated with longer bills include
that longer-billed hummingbirds make more inser-
tion errors when feeding in short, narrow flowers
compared to shorter-billed hummingbirds (Temeles
Bene“fit” in bill–corolla coevolution 5
Downloaded from https://academic.oup.com/icb/advance-article/doi/10.1093/icb/icab111/6288459 by Missouri State University user on 12 August 2021
1996). Insertion errors would increase the total time
required per floral visit and therefore reduce net en-
ergy gain per visit. Similarly, wielding long bills
might reduce control of fine adjustments that are
required to properly position the bill tip and tongue
to transverse the internal floral barriers (when pre-
sent) that prevent easy access to the nectary and
reaching all folds containing nectar. Additionally,
longer bills may require longer intraoral transport
times given that after offloading the nectar near
the bill tip, the liquid needs to be actively trans-
ported to the throat (Rico-Guevara 2014). Shorter
bills reduce the distance between the body of the
bird and substrate surfaces that could be used for
support (e.g., feeding while clinging or on the
ground). Employing substrates for weight support
greatly reduces energy expenditure (hovering is en-
ergetically expensive), and thus many hummingbirds
regularly perch to perform several consecutive visits,
even from the same perching position (e.g., in
Asteraceae inflorescences Stiles 2008). Finally, by be-
ing closer to the flower, hummingbirds with short
bills can more easily perforate the corolla near the
nectary to reach the nectar and more easily exploit
existing holes in the corolla made by other visitors
(Lara and Ornelas 2001).
Although we have focused on bill length, a suite of
traits are involved in bill–corolla trait matching; for
example, bill curvature shows trait matching with
flower shape (Maglianesi et al. 2014). The extremes
of these trait-matching axes (length and curvature)
are reflected in their morphological diversification
into a novel range of bill shapes (Cooney et al.
2017;Naval
on et al. 2021). To showcase morpholog-
ical extremes, compare the short straight beaks
(7 mm) of Purple-backed thornbills
(Ramphomicron microrhynchum) to the extremely
long (100 mm) and slightly recurved bills (poten-
tially enhancing visits to pendulous flowers, Stiles
2008) of Sword-billed hummingbirds, to the 90de-
curved bills of Sicklebill hummingbirds, matching
their preferred flowers (Stein 1992). While hum-
mingbird assemblages are often composed of species
with more modest differences in bill shape (but see
Rico-Guevara 2008), it is an open question whether
there is typically sufficient morphological diversity
for trait-derived niche partitioning to develop.
Other factors such as phenological and microhabitat
BOX 1
Here we present the parameters that would ideally be measured to evaluate the benefits to the hummingbird from trait matching
via quantification of hummingbird nectar extraction performance in the wild (Fig. 2). When a hummingbird feeds on a flower that matches
its bill shape, it would be expected to obtain a larger net energy gain (calories acquired minus invested) than when feeding on a flower with a
poorer match. For example, the hummingbird might be able to more efficiently insert its bill into the flower, reducing its time to access nectar
(handling time minus licking time), increase extraction efficiency, and collect a larger reward. By measuring the amount of nectar in the
flower and the licking rate, we can calculate the volumetric extraction efficiency (ll/s), and with a range of potential values for nectar
concentration in a particular flower, we can estimate the energy intake rate (cal/s). While it is not possible to measure the precise nectar
concentration consumed in a given floral visit without disturbing the system, the best approximation involves measuring the nectar con-
centration on the same lower but at a different time, or at the same time from adjacent flowers, or by characterizing nectar concentration
variation among flowers and across the relevant temporal scale (e.g. daily fluctuation). Nectar concentration is commonly measured via a
refractometer, following extraction of nectar from ‘bagged’ flowers (preventing visitors from extracting nectar) via microcapillary tubes.
nectar concentration = cal/μl
handling
time
licking rate
μl/s
cal/s
μl
Fig. 2. Variables for quantification of hummingbird drinking per-
formance during visits to wild flowers. Energy intake rate (cal/s)
¼extraction efficiency (ml/s)nectar concentration (cal/ml). See
text for a discussion of licking rate and handling time.
6A. Rico-Guevara et al.
Downloaded from https://academic.oup.com/icb/advance-article/doi/10.1093/icb/icab111/6288459 by Missouri State University user on 12 August 2021
overlap, abundance (e.g., V
azquez et al. 2009), and
type of competitive interactions (see Behavioral com-
ponents of hummingbird–plant trait matching sec-
tion) need to be considered to achieve a full
understanding of the magnitude of the importance
of functional trait-matching in determining hum-
mingbird–plant networks. Similar to bill length, bill
curvature shows trait matching with flower shape
(Maglianesi et al. 2014). Even if corolla and bill
length correspond, if curvature differs, benefits
from the length match will be greatly reduced or
even nullified for plant and hummingbird.
Similarly, benefits will not be fully realized if bill–
corolla curvature, but not lengths, match [see discus-
sion on curvature measurement methods in (Rico-
Guevara and Araya-Salas 2015)]. Finally, the corolla
and nectar chamber entrances restrict access also
based on bill and tongue thickness (see Evolution
of trait matching in hummingbird–plant interactions
section), and the depth to which the bill can pene-
trate the corolla depends on its internal configura-
tion. For example, a thicker bill tip and/or tongue
could prevent deep access inside the flower because
of corollar internal diameter constraints and/or the
presence of reproductive structures (stigma and/or
anthers) that reduce the bill tip mobility (Smith et
al. 1996). The actual bill tip space required for effec-
tive drinking is determined by the bill tip thickness
when the bill is opened to receive the tongue full of
nectar. Therefore, the thicker the tongue, the larger
the opening has to be to allow for access to the fully
loaded tongue (Grant and Temeles 1992). Given the
suite of physical traits that have to be considered to
characterize bill–corolla functional trait matching, a
thorough experimental and quantitative exploration
of the reciprocal benefits determined by matching
hummingbird bill–flower morphospaces is war-
ranted. Imperfect bill–corolla match does not neces-
sarily restrict access to the nectar of throchiliphilous
plants (or even other plants e.g., Waser et al. 2018),
and so hummingbirds can feed on a variety of floral
resources. Characterizing the costs and benefits of
physical trait matching, while considering the behav-
ioral and ecological context will be a key to under-
standing how they might drive coevolution.
While we have focused primarily on the pollina-
tor’s perspective, from the plant’s perspective, the
benefits for an individual plant from increased bill–
flower matching come from enhanced pollen depo-
sition on pollinator surfaces that ultimately contact a
conspecific flower’s stigma (cross-pollination, e.g.,
Betts et al. 2015). In terms of trait matching, there
is strong selection on floral morphology that forces
the pollinator into a position in which pollen is
picked up through contact with the anthers (from
the male perspective) and that effectively deposits
conspecific pollen onto the stigma (from the female
perspective). Reduced bill–flower fit might either re-
sult in lower chances of pollen dispersal because the
flower becomes a less desirable resource for the pol-
linator (if it experiences low net energy gain), or in
nectar extraction by the pollinator without proper
pollen transfer (e.g., robbing or mismatch between
floral reproductive organ surfaces and pollinator sur-
faces, e.g., Betts et al. 2015). The effects on pollen
deposition and actual transport (how much of it can
remain) on a given surface (e.g., forehead) of a pol-
linator, especially in the face of multiple consecutive
deployments, need to be documented to better un-
derstand the influence of foraging circuits (see
Behavioral components of hummingbird–plant trait
matching section) and mechanisms influencing pol-
lination outcomes. There are many open questions
regarding the underlying physical mechanisms in-
volved in successful pollen transfer. For example,
by pressing against the pollinator, does the anther
remove previously deposited pollen? Are different
structural and/or chemical properties of pollen adap-
tive in terms of pollen deposition and transport on
the pollinator and transfer to the stigma? Are partic-
ular pollinator surfaces adapted for pollen dispersal
(e.g., feathers)? Does preening/cleaning remove pol-
len and how often does this behavior occur between
floral visits (e.g., bill rubbing against branches)?
These are just a few of the many possible questions
relating to mechanisms potentially enhancing or dis-
rupting trait matching, but they stress the impor-
tance of quantifying both physical mechanisms and
behavioral components.
Behavioral components of
hummingbird–plant trait matching
While we have considered functional trait matching
mostly as a consequence of physical mechanisms,
animal behavior at multiple scales plays a powerful
role in determining the degree to which physical
components contribute to functional trait matching.
For example, at the level of individual floral visits,
cognitive flexibility and behavioral plasticity in hum-
mingbird lapping rate (e.g., Roberts 1995) and
tongue protrusion distance (Rico-Guevara 2017)
could enable hummingbirds to functionally adjust
their match to different flowers. Thus, even when
bill and flower seem morphologically matched, be-
havioral plasticity could expand the range of matches
to some degree. In addition, learning and cognition
generally (recently reviewed by Gonz
alez-G
omez and
Bene“fit” in bill–corolla coevolution 7
Downloaded from https://academic.oup.com/icb/advance-article/doi/10.1093/icb/icab111/6288459 by Missouri State University user on 12 August 2021
Araya-Salas 2019), preference (related to particular
behavioral strategies), and competition (related to
abundance of both resources and competitors e.g.,
Simmons et al. 2019) are potentially powerful behav-
ioral components of functional trait matching evo-
lution. Here, we focus on behavioral strategies at the
level of landscape (defense and movement among
floral resources) and its consequences for the evolu-
tion of bill–flower matching in hummingbird–plant
assemblages. Optimal foraging theory predicts that
hummingbirds should attempt to balance costs and
benefits associated with finding and extracting better
quality nectar rewards (Heinrich 1975;Stiles 1975;
Pyke 2019;Blem et al. 2000). However, to achieve
a net positive energy gain, hummingbirds might con-
ceivably visit flowers that are a poor functional
match to their bills. As long as the hummingbird
achieves a net positive energy gain, a hummingbird
may, for example, prefer to visit flowers that are
close together to reduce costs associated with search-
ing, even if the bill–corolla match for some of those
flowers is poor. In addition, trochiliphilous flowers
frequently conceal their nectar deep within the nec-
tary, lacking visual, olfactory, or electrostatic cues
associated with the reward (Rocca and Sazima
2010;Lunau et al. 2020;Pauw et al. 2020), unlike,
for example, bee-pollinated plants (Clarke et al.
2013;Russell et al. 2018). Consequently, achieving
a net positive energy gain from flowers is a proba-
bilistic game that has multiple successful behavioral
strategies (Feinsinger and Colwell 1978).
Here, we provide new terms to focus on two
widely observed and presumably mutually exclusive
behavioral strategies thought to maximize net energy
gain, and which have consequences for plant repro-
ductive success: (1) stationary interference (formerly
“territoriality”), in which individuals stay within a
resource patch and use aggressive behaviors to inter-
fere with attempts of different nectarivores to access
the patch and (2) traveling exploitation (formerly
“traplining”), in which individuals forage on resour-
ces scattered across the landscape, traveling among
foraging areas in a particular sequence (e.g., Stiles
1975;Feinsinger and Colwell 1978;Tello-Ramos et
al. 2015, see Kamath and Wesner 2020;Sargent et al.
2021) for a discussion on the controversies and
semantic problems with the prior terms). These
new terms describe both the predominant
landscape range and the type of competition
associated with these ends of the spectrum of
behavioral strategies. A stationary interference
strategy is expected to be used when tolerating
other nectarivores is costly, such as when floral
nectar is readily accessible to a diverse assemblage
of pollinators (Stiles 1975). This is because
interfering with the foraging of competitors is most
useful when competitors can access the resource as
easily as the interferer, resulting in agonistic
interactions collectively known as interference
competition (Rico-Guevara et al. 2019). Conversely,
when foraging on a specific plant species carries a
net positive benefit for only well-matched nectari-
vores, the focal pollinator is released from the need
to defend nectar resources from all possible nectar-
ivores (see Sargent et al. 2021). The net energy gain
is thus maximized by visiting only flowers with a
good match, regardless of their spatial proximity
and thus a traveling exploitation strategy may be
favored, in which pollinators visit plants scattered
across a broad range (Ohashi and Thomson 2009;
Buatois and Lihoreau 2016). Note that this is differ-
ent from how the term traplining has been applied
to foraging circuits independent of the scale at which
they occur (Tello-Ramos et al. 2015). From this per-
spective, traplining can also be performed by a sta-
tionary interferer while visiting the flowers in a given
patch, while also defending it (Tello-Ramos et al.
2015). In our new terminology, a traveling exploiter,
by definition, does not stay in an area to defend the
resources within.
When hummingbirds restrict access to resource
patches via stationary interference, this strategy can
shift the realized hummingbird–plant interactions
from what would be expected given the fundamental
niche distribution based only upon bill–corolla
matching. Stationary interference is linked to aggres-
sive behavior, which can also be a strong selective
force on hummingbird bill morphology, and thus
indirectly drive selection on flower shape and other
aspects of a plant’s reproductive strategy via coevo-
lution. Indeed, agonistic behaviors are associated
with at least some bill traits in the context of intra-
sexual competition and evolution of intrasexually se-
lected weapons (Rico-Guevara and Araya-Salas 2015;
Rico-Guevara and Hurme 2019), and have been pro-
posed to be associated with interspecific competition
(Rico-Guevara et al. 2019). For example, physical
confrontations in which the bill is used to contact
the opponent could select for (1) longer beaks that
provide increased reach to stab an opponent before
getting stabbed, (2) straighter beaks that transmit the
force from bill base to tip better, (3) thicker beaks
that can withstand higher axial loadings, (4) sharper
bill tips that can pierce with less applied force, and
(5) stronger bill tips that resist bending forces while
biting (e.g., to pluck feathers or physically displace
opponents).
8A. Rico-Guevara et al.
Downloaded from https://academic.oup.com/icb/advance-article/doi/10.1093/icb/icab111/6288459 by Missouri State University user on 12 August 2021
Stationary interference and traveling exploitation
strategies may also influence the benefits for the
plant, beyond simple differences in conspecific pollen
transfer. For example, stationary interference is likely
to reduce pollen dispersal of the defended plants,
assuming that displacing other pollinators reduces
floral visitation rate, and could also increase hetero-
specific pollen transfer when interferers visit multiple
plant species in the same area (Ashman and Arceo-
G
omez 2013). In addition, stationary interference is
likely to reduce outcrossing, as pollen will be mostly
transferred among the defended plants (Torres-
Vanegas et al. 2019). Traveling exploitation on the
other hand involves the pollinator visiting flowers in
a discrete sequence across the landscape, thus shap-
ing the direction of pollen movement. As a result,
one plant in the visitation sequence may always serve
as a pollen donor, while another in the sequence
may always serve as a pollen recipient, even if the
individual plants are widely separated (e.g., Ohashi
and Thomson 2009, but see Torres-Vanegas et al.
2019). Traveling exploitation and stationary interfer-
ence lie at the extremes of a continuum between
exploitative and interference competition (Rico-
Guevara et al. 2019) that can select for a variety of
behavioral and morphological traits that can directly
impact pollination. In an assemblage of spatiotem-
porally coexisting hummingbirds and hummingbird-
visited plants (a pollinator interaction network, see
Approaches to studying hummingbird–plant (bill–
corolla) trait matching section), the foraging links
between particular plants and birds, as well as their
strength (e.g., the proportion of visits for both par-
ties), depend on pollinator behavior. Similarly, plant
and hummingbird abundance and phenology can
also affect the links between them and thus the
strength of selection (V
azquez et al. 2009). At the
community level, additional evolutionary processes
need also to be considered to link mechanisms and
patterns at different ecological scales. For example,
character displacement on hummingbirds feeding on
coexisting plants can partially explain their astonish-
ing diversity in bill shape and size (Maglianesi et al.
2014). Similarly, intersexual floral resource partition-
ing is thought to at least in part explain sexual di-
morphism in bill morphology (Temeles et al. 2000,
2010). Furthermore, character displacement likely
drives the evolution of diverse floral morphologies
and pollen deposition and collection strategies
among plants (e.g., different lengths of anthers aim-
ing at different body surfaces, lever mechanisms, ex-
plosive pollen release, modified petals, etc. (Aluri
and Reddi 1995;Rengifo et al. 2006;Temeles and
Rankin 2011;LoPresti et al. 2020)).
Approaches to studying hummingbird–
plant (bill–corolla) trait matching
Inferences about reciprocal specialization between
hummingbird bills and the flowers they pollinate
are frequently drawn from research on humming-
bird–plant interaction networks (examples in
Supplementary Table S1). These studies establish
which hummingbird and plant species interact and
can generate hypotheses for putative benefits of bill–
corolla trait matching (e.g., Maglianesi et al. 2014).
While plant–pollinator networks suggest that func-
tional trait matching maintains links among species,
the determinants of both pollination and nectarivory
at the level of the floral visits are poorly understood.
Moreover, evaluating the inferred benefits or draw-
backs of particular interactions for plants and polli-
nators requires a detailed look at what happens
during a floral visit. Therefore, to help guide future
research, we provide a comparison of different net-
work building methods (Supplementary Table S1)
used to characterize benefits to hummingbird and
plant and thus assess the mechanisms enabling pu-
tative bill–corolla coevolution. We focus in particular
on the value of video, because while some methods
effectively characterize which plants and humming-
birds are interacting, as well as the strength of these
interactions (e.g., number of visits, their frequency,
and the amount of pollen found on the pollinator),
only video recordings enable precise characterization
of hummingbird behavior on flowers and the inter-
action between floral reproductive organs and the
hummingbird. Recent video footage of plant–polli-
nator interactions has unveiled pollinator fidelity to
their resources even in times of low abundance and a
relatively high occurrence of nectar-robbing
(Weinstein and Graham 2017), as well as highlighted
the influence of spatial distributions in trait-
matching and resulting morphotypes (Sonne et al.
2019).
In addition to the importance of quantifying pol-
lination network interactions, it is necessary to un-
derstand how they are modulated through a
common currency: net energy gain. The main pro-
posed benefit of bill–corolla matching from the pol-
linator’s perspective is an increase in net energy gain
(Box 1), which is a consequence of (1) nectar accu-
mulation in specialized flowers (if competitors can-
not access or are morphologically discouraged from
accessing the reward, Behavioral components of
hummingbird–plant trait matching section), (2) re-
duced access time to the flower entrance and nectary
(Physical components of hummingbird–plant trait
matching and pollination section), and (3) increased
Bene“fit” in bill–corolla coevolution 9
Downloaded from https://academic.oup.com/icb/advance-article/doi/10.1093/icb/icab111/6288459 by Missouri State University user on 12 August 2021
nectar intake efficiency. Measuring nectar replenish-
ment (without affecting floral visitation) and feeding
performance in wild flowers was practically impossi-
ble until recent technological advances, thus, nectar
drinking has been mostly studied under artificial
conditions (e.g., Roberts 1995;Collins 2008;Rico-
Guevara et al. 2019). In particular, advances in vid-
eography capabilities and affordability (see discus-
sion about costs in Rico-Guevara and Mickley
2017) have been a game changer. Video recordings
are comprehensive in capturing both successful and
unsuccessful attempts to extract nectar from focal
flowers and enabling characterization of alternative
behaviors such as nectar robbing. Similarly, in com-
bination with field data (e.g., nectar concentration
and flower dimensions) they can provide the means
to quantify performance variables involved in evalu-
ating how bill–corolla match shapes net energy gain
(e.g., Fig. 2). We present an overview of the video
techniques that would be applicable to the study of
plant–pollinator and other ecological interactions,
and we finish with an example of one out of the
many possible combinations of those techniques.
To capture hummingbird fast behaviors (e.g., lick-
ing rates up to 17 Hz, Ewald and Williams 1982), we
need high-speed video; to resolve movement and
contact with the floral sex organs or nectar inside
small flowers and tiny nectaries (e.g., the exclusion
of insect pollinators or the movement of the bill
inside the corolla), we need macro-photography;
and to mitigate the influence of the camera on hum-
mingbird behavior, we need tele-photo capabilities
(Supplementary Table S2). Solving these challenges
is possible with specialized and field-friendly cameras
(e.g., Rico-Guevara and Mickley 2017). We present a
combination of backlit-filming, long duration, and
high-speed recording techniques that allows us to
measure rates of nectar depletion and replenishment
in wild flowers. This is key, because to the best of
our knowledge, until this point researchers have not
had a way to measure nectar extraction, which
requires knowing both the uptake rate and the pre-
existing volume of nectar in the flower. Backlit film-
ing allows estimation of the nectar volume without
physically manipulating the flower (e.g., extracting
nectar manually) and thus does not affect floral vis-
itation or damage floral tissues (Fig. 3;
Supplementary Table S2). We performed volumetric
estimations (Supplementary Table S3) of the nectar
through the visualization of the liquid inside the
flower (Supplementary Video S1) and floral dissec-
tions to calculate the internal dimensions of the nec-
tar chamber (Supplementary Fig. S1). Measuring
nectar extraction in unmanipulated wild flowers
takes us a step closer to directly testing the bill–co-
rolla matching benefits from the hummingbird’s
perspective.
Concluding remarks and implications
for the study of other coevolutionary
systems
Half a century of research on floral foraging by hum-
mingbirds has provided us with a wealth of knowl-
edge about hummingbird nectar preferences
(Hainsworth and Wolf 1976;Stiles 1976;Calder
1979;Tamm and Gass 1986;Mart
ınez del Rio
1990;Stromberg and Johnsen 1990;Roberts 1996;
Blem et al. 1997,2000;Fleming et al. 2004;
Chalcoff et al. 2008), nectar extraction efficiency
(Houston and Krakauer 1993;Roberts 1995;Collins
2008), the role of cognition (Healy and Hurly 2013;
Gonz
alez-G
omez and Araya-Salas 2019), and optimi-
zation of foraging and energetics (DeBenedictis et al.
1978;Hixon and Carpenter 1988;Gass and Roberts
1992;Shankar et al. 2019). Despite the breadth of
this literature, we still lack a good understanding of
the mechanisms underlying bill–corolla matching
and their role in maintaining hummingbird–plant
interactions as well as driving coevolution.
Therefore, we endeavored to elucidate these gaps in
our knowledge via a discussion of the evolution and
physical and behavioral components of bill–corolla
matching. To help move this field forward, we
have highlighted key questions and methods that
we hope will facilitate and expand characterization
of functional trait matching for both plant and pol-
linator (Supplementary Tables S1 and S2).
In particular, we emphasize taking a mechanistic
perspective when considering the drivers of adapta-
tions in hummingbird–trochiliphilous plant interac-
tions. Specialized videography is uniquely suited to
characterize the physical interactions between floral
reproductive structures and pollinator surfaces, mak-
ing it possible to quantify the benefits of trait match-
ing for the plant (Physical components of
hummingbird–plant trait matching and pollination
section; Supplementary Table S2). Similarly, video
of hummingbirds drinking from flowers makes it
possible to quantify key parameters needed to deter-
mine costs and benefits for the birds (Fig. 2), such as
energy intake efficiency (Fig. 3), which is one of the
drivers of foraging decisions (Behavioral components
of hummingbird–plant trait matching section) and
thus reciprocal adaptation. Similarly, a combination
of methods (Supplementary Table S1) can be used to
characterize the interaction networks resulting from
those foraging decisions. In addition, with
10 A. Rico-Guevara et al.
Downloaded from https://academic.oup.com/icb/advance-article/doi/10.1093/icb/icab111/6288459 by Missouri State University user on 12 August 2021
videography we can document visits by different
hummingbirds to focal flowers and hummingbirds
preventing competitors from accessing those flowers,
thus revealing the interplay between exploitative and
interference competition. In other words, videogra-
phy permits us to contrast null expectations of
plant–hummingbird interactions based exclusively
on bill–corolla matching, with the actual interactions
occurring in natural communities and thus the con-
ditions influencing bill–corolla coevolution. Finally,
by combining videography and quantification of
nectar extraction performance with the wealth of
knowledge about hummingbird aerodynamics and
energetics in the context of floral foraging (Sargent
et al. 2021, it becomes possible to quantify net en-
ergy gain, the key benefit of trait matching for the
hummingbird. Altogether, these quantitative and
mechanistic approaches make hummingbird–plant
interactions a model system for studying the benefits
of functional trait matching for both plant and pol-
linator and, more generally, the drivers of
coevolution.
Our proposed framework unites theoretical
expectations and empirical observations to improve
our understanding of the mechanisms constraining
functional trait matching and its evolution. While
the methods and framework presented here concern
the biomechanics and energetics of hummingbird–
flower interactions, they hopefully serve as a tem-
plate for quantifying costs and benefits in other
plant–pollinator systems and ecological interactions.
For example, some of the questions asked about
hummingbird–plant interactions, like bill–corolla
length-matching, are equivalent to questions asked
in bumble bee–plant interactions (tongue-corolla
length-matching e.g., Miller-Struttmann et al.
2015). Similarly, with slight modifications—such as
infrared lights and recording capabilities—equivalent
measurements could be collected for a variety of di-
urnal (e.g., other anthophilous birds, butterflies,
bees) and nocturnal (e.g., bats and moths) nectari-
vores, that sometimes even feed on the same flowers
(e.g., Fig. 1A). Additionally, while behavioral plastic-
ity in flower feeding mechanisms potentially strongly
affects functional trait matching, we have barely
scratched the surface of this subject in a variety of
pollination systems, including anthophilous birds,
rodents, bats, and butterflies (but see work on moths
e.g., Goyret and Raguso 2006;Goyret and Kelber
2011; and work on bees e.g., Russell et al. 2017,
2018;Wei et al. 2020). Finally, biomechanics per-
spectives are often incompletely developed in other
coevolutionary systems, such as the snake–newt–te-
trodotoxin system (e.g., feeding biomechanics that
potentially influences the costs of prey consumption)
and the anemone–anemonefish system (e.g., anemo-
nefish biomechanics that potentially benefits the
host). All in all, the study of physical mechanisms
of behavior is an important avenue for reintegrating
biological sciences and will be an active field of re-
search for years to come.
Acknowledgments
We thank Patrick Green for organizing the Physical
Mechanisms of Behavior Symposium. We thank
many field assistants and researchers for their sup-
port and discussions in the development of filming
techniques. We are grateful to Patrick Green and
Melissa Morado for their extensive support in man-
uscript preparation and review, and to Derrick
Groom and Alyssa Sargent for discussions. We thank
two anonymous reviewers for their excellent
suggestions.
Funding
This work was supported by the Walt Halperin
Endowed Professorship and the Washington
Research Foundation as Distinguished Investigator
(to A.R-G.), and by The Company of Biologists
and the Society of Integrative and Comparative
Biology.
Conflicts of interest
The authors declare no conflicts of interest.
Extraction Rate (µl/s)
Nectar In Flower (µl)
Time (s)
Fig. 3 Nectar depletion (in blue, left axis, and circles) and ex-
traction rate (in magenta, right axis, and triangles) for a single
visit of a Speckled hummingbird (Adelomyia melanogenys)toa
flower of Palicourea sp. (Supplementary Video S1). Measurements
of nectar pool depletion were performed after every lick
(Supplementary Table S3). Our methods allow for the assess-
ment of feeding performance, at the flower visit level, quantifying
variables that have not been possible to measure to date in the
wild (e.g., maximum bill insertion, access times, licking rates,
liquid collection rates, etc.; Fig. 2).
Bene“fit” in bill–corolla coevolution 11
Downloaded from https://academic.oup.com/icb/advance-article/doi/10.1093/icb/icab111/6288459 by Missouri State University user on 12 August 2021
Animal ethics statement
Activities are covered by UW IACUC protocol 4498-
03.
Data availability statement
All the data presented and used in graphs are in-
cluded in the supplement.
Supplementary data
Supplementary Data available at ICB online.
References
Abrahamczyk S, Renner SS. 2015. The temporal build-up of
hummingbird/plant mutualisms in North America and
temperate South America. BMC Evol Biol 15:104.
Abrahamczyk S, Souto-Vilar
os D, Renner SS. 2014. Escape
from extreme specialization: passionflowers, bats and the
sword-billed hummingbird. Proc R Soc B Biol Sci
281:20140888.
Aluri JSR, Reddi C. 1995. Explosive pollen release and polli-
nation in flowering plants. Proc Indian Natl Sci Acad
B61:323–32.
Arcos L, Lara C, Castillo-Guevara C, Cuautle M, Ornelas J.
2018. ‘Pro-bird’ floral traits discourage bumblebee visits to
Penstemon gentianoides (Plantaginaceae), a mixed-
pollinated herb. Naturwissenschaften 106:1.
Arditti J, Elliott J, Kitching IJ, Wasserthal LT. 2012. ‘Good
Heavens what insect can suck it’—Charles Darwin,
Angraecum sesquipedale and Xanthopan morganii praedicta.
Bot J Linn Soc 169:403–32.
Ashman T-L, Arceo-Gomez G. 2013. Toward a predictive
understanding of the fitness costs of heterospecific pollen
receipt and its importance in co-flowering communities.
Am J Bot 100:1061–70.
Betts MG, Hadley AS, Kress WJ. 2015. Pollinator recognition
by a keystone tropical plant. Proc Natl Acad Sci U S A
112:3433–8.
Blem CR, Blem LB, Cosgrove CC. 1997. Field studies of ru-
fous hummingbird sucrose preference: does source height
affect test results? (Estudios de Campo Sobre las
Preferencias de Sucrosa por Parte de Selasphorus rufus:
afecta los Resultados la Altura de la Fuente de
Alimentos?). J Field Ornithol 68:245–52.
Blem CR, Blem LB, Felix J, Gelder JV. 2000. Rufous hum-
mingbird sucrose preference: precision of selection varies
with concentration. The Condor 102:235–8.
Brown JH, Kodric-Brown A. 1979. Convergence, competition,
and mimicry in a temperate community of hummingbird-
pollinated flowers. Ecology 60:1022–35.
Buatois A, Lihoreau M. 2016. Evidence of trapline foraging in
honeybees. J Exp Biol 219:2426–9.
Calder WA. 1979. On the temperature-dependency of optimal
nectar concentrations for birds. J Theor Biol 78:185–96.
Castellanos MC, Wilson P, Thomson JD. 2004. ‘Anti-bee’ and
‘pro-bird’ changes during the evolution of hummingbird
pollination in Penstemon flowers. J Evol Biol 17:876–85.
Castellanos MC, Wilson P, Thomson JD. 2003. Pollen transfer
by hummingbirds and bumblebees, and the divergence of
pollination modes In Penstemon. Evolution 57:2742–52.
Chalcoff VR, Aizen MA, Galetto L. 2008. Sugar preferences of
the green-backed firecrown hummingbird (Sephanoides
Sephaniodes): a field experiment. Auk 125:60–6.
Clarke D, Whitney H, Sutton G, Robert D. 2013. Detection
and learning of floral electric fields by bumblebees. Science
340:66–9.
Collins BG. 2008. Nectar intake and foraging efficiency:
responses of honeyeaters and hummingbirds to variations
in floral environments. Auk 125:574–87.
Cooney CR, Bright JA, Capp EJR, Chira AM, Hughes EC,
Moody CJA, Nouri LO, Varley ZK, Thomas GH. 2017.
Mega-evolutionary dynamics of the adaptive radiation of
birds. Nature 542:344–7.
Cotton PA. 1998. Coevolution in an Amazonian
hummingbird-plant community. Ibis 140:639–46.
Cronk Q, Ojeda I. 2008. Bird-pollinated flowers in an evolu-
tionary and molecular context. J Exp Bot 59:715–27.
Dalsgaard B, Gonzalez A, Olesen J, Ollerton J, Timmermann
A, Andersen L, Tossas A. 2009. Plant–hummingbird inter-
actions in the West Indies: floral specialisation gradients
associated with environment and hummingbird size.
Oecologia 159:757–66.
Dalsgaard B, Gonz
alez AMM, Olesen JM, Timmermann A,
Andersen LH, Ollerton J. 2008. Pollination networks and
functional specialization: a test using lesser antillean plant-
hummingbird assemblages. Oikos 117:789–93.
Dalsgaard B, Maruyama P, Sonne J, Hansen K, Zanata T,
Stefan A, Alarc
on R, Araujo A, Ara
ujo F, Buzato S, et al.
2021. The influence of biogeographical and evolutionary
histories on morphological trait-matching and resource
specialization in mutualistic hummingbird-plant networks.
Funct Ecol 35:1120–33.
Darwin CR. 1862. Letter 3411-Darwin, C. R., to Hooker, J. D.
1862 Jan 25.
DeBenedictis PA, Gill FB, Hainsworth FR, Pyke GH, Wolf LL.
1978. Optimal meal size in hummingbirds. Am Nat
112:301–16.
Dellinger AS, Chartier M, Fern
andez-Fern
andez D, Penneys
DS, Alvear M, Almeda F, Michelangeli FA, Staedler Y,
Armbruster WS, Scho¨ nenberger J. 2019. Beyond buzz-pol-
lination—departures from an adaptive plateau lead to new
pollination syndromes. New Phytol 221:1136–49.
Ewald PW, Williams WA. 1982. Function of the bill and
tongue in nectar uptake by hummingbirds. Auk 99:573–6.
Feinsinger P, Colwell RK. 1978. Community organization
among neotropical nectar-feeding birds. Am Zool
18:779–95.
Fleming PA, Bakken BH, Lotz CN, Nicolson SW. 2004.
Concentration and temperature effects on sugar intake
and preferences in a sunbird and a hummingbird. Funct
Ecol 18:223–32.
Fleming TH, Muchhala N. 2008. Nectar-feeding bird and bat
niches in two worlds: pantropical comparisons of verte-
brate pollination systems. J Biogeogr 35:764–80.
Fonseca L. d, Rech AR, Bergamo PJ, Gonc¸alves-Esteves V,
Sazima M. 2016. Heterospecific pollen deposition among
plants sharing hummingbird pollinators in the Brazilian
Atlantic Forest. Rodrigu
esia 67:335–45.
12 A. Rico-Guevara et al.
Downloaded from https://academic.oup.com/icb/advance-article/doi/10.1093/icb/icab111/6288459 by Missouri State University user on 12 August 2021
Gass CL, Roberts WM. 1992. The problem of temporal scale
in optimization: three contrasting views of hummingbird
visits to flowers. Am Nat 140:829–53.
Gegear RJ, Burns R, Swoboda-Bhattarai KA. 2017.
“Hummingbird” floral traits interact synergistically to discour-
age visitation by bumble bee foragers. Ecology 98:489–99.
Gonz
alez-G
omez PL, Araya-Salas M. 2019. Perspectives on
the study of field hummingbird cognition in the neotrop-
ics. In: Reboreda JC, Fiorini VD, Tuero DT, editors.
Behavioral ecology of neotropical birds. Cham: Springer
International Publishing. p. 199–212.
Goyret J, Raguso RA. 2006. The role of mechanosensory in-
put in flower handling efficiency and learning by Manduca
sexta. J Experi Biol 209:1585–93.
Goyret J, Kelber A. 2011. How does a diurnal hawkmoth find
nectar? Differences in sensory control with a nocturnal rel-
ative. Behav Ecol 22:976–84.
Grant V, Temeles EJ. 1992. Foraging ability of rufous hum-
mingbirds on hummingbird flowers and hawkmoth flow-
ers. Proc Natl Acad Sci U S A 89:9400–4.
Hainsworth FR. 1973. On the tongue of a hummingbird: its
role in the rate and energetics of feeding. Comp Biochem
Physiol A Physiol 46:65–78.
Hainsworth FR, Wolf LL. 1976. Nectar characteristics and
food selection by hummingbirds. Oecologia 25:101–13.
Healy SD, Hurly TA. 2013. What hummingbirds can tell us
about cognition in the wild. Comp Cogn Behav Rev 8:13–28.
Heinrich B. 1975. Energetics of pollination. Annu Rev Ecol
Syst 6:139–70.
Hernandez-Conrique D, Ornelas JF, Garc
ıa-Franco JG, Vargas
F. 2007. Nectar production of Calliandra longipedicellata
(Fabaceae: mimosoideae), an endemic mexican shrub
with multiple potential pollinators. Biotropica 39:459–67.
Hixon MA, Carpenter FL. 1988. Distinguishing energy max-
imizers from time minimizers: a comparative study of two
hummingbird species. Am Zool 28:913–25.
Houston AI, Krakauer DC. 1993. Hummingbirds as net rate
maximisers. Oecologia 94:135–8.
Johnson SD, Anderson B. 2010. Coevolution between food-
rewarding flowers and their pollinators. Evol Educ
Outreach 3:32–9.
Johnson SD, Mor
e M, Amorim FW, Haber WA, Frankie GW,
Stanley DA, Cocucci AA, Raguso RA. 2017. The long and
the short of it: a global analysis of hawkmoth pollination
niches and interaction networks. Funct Ecol 31:101–15.
Kamath A, Wesner AB. 2020. Animal territoriality, property
and access: a collaborative exchange between animal behav-
iour and the social sciences. Anim Behav 164:233–9.
Lagomarsino LP. 2015. The systematics and evolution of neo-
tropical bellflowers (Campanulaceae: Lobelioideae)
[Doctoral dissertation], Harvard University, Graduate
School of Arts & Sciences.
Lara C, Ornelas J. 2001. Preferential nectar robbing of flowers
with long corollas: experimental studies of two hummingbird
species visiting three plant species. Oecologia 128:263–73.
Lindberg A, Olesen J. 2001. The fragility of extreme special-
ization: passiflora mixta and its pollinating hummingbird
Ensifera ensifera. J Trop Ecol 17:323–9.
LoPresti EF, Goidell J, Mola JM, Page ML, Specht CD, Stuligross
C, Weber MG, Williams NM, Karban R. 2020. A lever action
hypothesis for pendulous hummingbird flowers: experimental
evidence from a columbine. Ann Bot 125:59–65.
Lunau K, Ren Z-X, Fan X-Q, Trunschke J, Pyke GH, Wang H.
2020. Nectar mimicry: a new phenomenon. Sci Rep 10:7039.
Mackin CR, Pe~
na JF, Blanco MA, Balfour NJ, Castellanos
MC. 2021. Rapid evolution of a floral trait following ac-
quisition of novel pollinators. J Ecol Early View
109:2234–46.
Maglianesi MA, Blu¨ thgen N, Bo¨ hning-Gaese K, Schleuning
M. 2014. Morphological traits determine specialization
and resource use in plant–hummingbird networks in the
neotropics. Ecology 95:3325–34.
Mart
ınez del Rio C. 1990. Dietary, phylogenetic, and ecolog-
ical correlates of intestinal sucrase and maltase activity in
birds. Physiol Zool 63:987–1011.
Mart
ınez del Rio C, Schondube JE, McWhorter TJ, Herrera
LG. 2001. Intake responses in nectar feeding birds: digestive
and metabolic causes, osmoregulatory consequences, and
coevolutionary effects 1. Am Zool 41:902–15.
Mayr G. 2004. Old world fossil record of modern-type hum-
mingbirds. Science 304:861–4.
Medeiros LP, Garcia G, Thompson JN, Guimar~
aes PR. 2018.
The geographic mosaic of coevolution in mutualistic net-
works. Proc Natl Acad Sci U S A 115:12017–22.
Miller-Struttmann NE, Geib JC, Franklin JD, Kevan PG,
Holdo RM, Ebert-May D, Lynn AM, Kettenbach JA,
Hedrick E, Galen C. 2015. Functional mismatch in a bum-
ble bee pollination mutualism under climate change.
Science 349:1541–4.
Montgomerie RD. 1984. Nectar extraction by hummingbirds:
response to different floral characters. Oecologia 63:229–36.
Morales CL, Traveset A. 2008. Interspecific Pollen Transfer:
Magnitude, Prevalence and Consequences for Plant Fitness.
CRC Crit Rev Plant Sci 27:221–38.
Muchhala N. 2006. Nectar bat stows huge tongue in its rib
cage. Nature 444:701–2.
Naval
on G, Nebreda SM, Bright JA, Fabbri M, Benson RBJ,
Bhullar B-A, Marug
an-Lob
on J, Rayfield EJ. 2021.
Craniofacial development illuminates the evolution of
nightbirds (Strisores). Proc R Soc B Biol Sci
288:rspb.2021.0181, 20210181.
Ohashi K, Thomson J. 2009. Trapline foraging by pollinators:
its ontogeny, economics and possible consequences for
plants. Ann Bot 103:1365–78.
Pauw A. 2019. A bird’s-eye view of pollination: biotic inter-
actions as drivers of adaptation and community change.
Annu Rev Ecol Evol Syst 50:477–502.
Pauw A, Cocucci AA, S
ersic AN. 2020. The least effective
pollinator principle: specialized morphology despite gener-
alized ecology. Plant Biol 22:924–31.
Proctor M, Yeo P, Lack A. 1996. The natural history of pol-
lination. London: Harper Collins New Naturalist
Pyke G. 2019. Optimal foraging and plant–pollinator co-evo-
lution. Encycl Anim Behav 209–16.
Rengifo C, Cornejo L, Akirov I. 2006. One size fits all: corolla
compression in Aphelandra runcinata (Acanthaceae), an
adaptation to short-billed hummingbirds. J Trop Ecol
22:613–9.
Rico-Guevara A. 2008. Morfolog
ıa Y Forrajeo Para Buscar
Artropodos Por Colibr
ıes Altoandinos. Ornitolog
ıa
Colomb 7:43–58.
Bene“fit” in bill–corolla coevolution 13
Downloaded from https://academic.oup.com/icb/advance-article/doi/10.1093/icb/icab111/6288459 by Missouri State University user on 12 August 2021
Rico-Guevara A. 2014. Morphology and function of the
drinking apparatus in hummingbirds [Doctoral disserta-
tion]. University of Connecticut, The Graduate School.
Rico-Guevara A. 2017. Relating form to function in the hum-
mingbird feeding apparatus. PeerJ 5:e3449.
Rico-Guevara A, Araya-Salas M. 2015. Bills as daggers? A test
for sexually dimorphic weapons in a lekking hummingbird.
Behav Ecol 26:21–9.
Rico-Guevara A, Fan T-H, Rubega MA. 2015. Hummingbird
tongues are elastic micropumps. Proc R Soc B Biol Sci
282:20151014.
Rico-Guevara A, Hurme KJ. 2019. Intrasexually selected
weapons. Biol Rev 94:60–101.
Rico-Guevara A, Mickley J. 2017. Bring your own camera to
the trap: an inexpensive, versatile, and portable triggering
system tested on wild hummingbirds. Ecol Evol 7:4592–8.
Rico-Guevara A, Rubega MA. 2011. The hummingbird
tongue is a fluid trap, not a capillary tube. Proc Natl
Acad Sci U S A 108:9356–60.
Rico-Guevara A, Rubega MA. 2017. Functional morphology
of hummingbird bill tips: their function as tongue wring-
ers. Zoology 123:1–10.
Rico-Guevara A, Rubega MA Hurme KJ, Dudley R. 2019.
Shifting paradigms in the mechanics of nectar extraction
and hummingbird bill morphology. Integr Org Biol
1:oby006.
Roberts MW. 1996. Hummingbirds’ nectar concentration
preferences at low volume: the importance of time scale.
Anim Behav 52:361–70.
Roberts WM. 1995. Hummingbird licking behavior and the
energetics of nectar feeding. Auk 112:456–63.
Rocca MA, Sazima M. 2010. Beyond hummingbird-flowers:
the other side of ornithophily in the neotropics. Oecologia
Aust 14:67–99.
Rodr
ıguez-Giron
es MA, Llandres AL. 2008. Resource compe-
tition triggers the co-evolution of long tongues and deep
corolla tubes. PLoS ONE 3:e2992.
Rojas-Nossa SV, S
anchez JM, Navarro L. 2021. Nectar rob-
bing and plant reproduction: an interplay of positive and
negative effects. Oikos 130:601–8.
Russell AL, Mauerman K, Golden R, Papaj D. 2018. Linking
components of complex signals to morphological part: the
role of anther and corolla in the complex floral display.
Anim Behav 135:223–36.
Russell AL, Buchmann SL, Papaj DR. 2017. How a generalist
bee achieves high efficiency of pollen collection on diverse
floral resources. Behav Ecol published online (doi:10.1093/
beheco/arx058).
Sargent AJ, Groom DJE, Rico-Guevara A. 2021. Locomotion
and energetics of divergent foraging strategies in humming-
birds: A review. Integr Comp Biol (doi: 10.1093/icb/
icab124).
Sargent RD, Vamosi JC. 2008. The influence of canopy position,
pollinator syndrome, and region on evolutionary transitions in
pollinator guild size. Int J Plant Sci 169:39–47.
Serrano-Serrano ML, Rolland J, Clark JL, Salamin N, Perret
M. 2017. Hummingbird pollination and the diversification
of angiosperms: an old and successful association in
Gesneriaceae. Proc R Soc B Biol Sci 284:20162816.
Shankar A, Graham CH, Canepa JR, Wethington SM, Powers
DR. 2019. Hummingbirds budget energy flexibly in re-
sponse to changing resources. Funct Ecol 33:1904–16.
Simmons BI, Vizentin-Bugoni J, Maruyama PK, Cotton PA,
Mar
ın-G
omez OH, Lara C, Rosero-Lasprilla L, Maglianesi
MA, Ortiz-Pulido R, Rocca MA, et al. 2019. Abundance
drives broad patterns of generalisation in plant–humming-
bird pollination networks. Oikos 128:1287–95.
Smith CE, Stevens JT, Temeles EJ, Ewald PW, Hebert RJ,
Bonkovsky RL. 1996. Effect of floral orifice width and
shape on hummingbird-flower interactions. Oecologia
106:482–92.
Sonne J, Zanata TB, Gonz
alez AMM, Torres NLC, Fjeldsa˚J,
Colwell RK, Tinoco BA, Rahbek C, Dalsgaard B. 2019. The
distributions of morphologically specialized hummingbirds
coincide with floral trait matching across an Andean eleva-
tional gradient. Biotropica 51:205–18.
Soteras F, Mor
e M, Iba~
nez AC, Iglesias M. d R, Cocucci AA.
2018. Range overlap between the sword-billed humming-
bird and its guild of long-flowered species: an approach to
the study of a coevolutionary mosaic. PLoS ONE
13:e0209742.
Stein BA. 1992. Sicklebill hummingbirds, ants, and flowers.
Bioscience 42:27–33.
Stiles FG. 1975. Ecology, flowering phenology, and humming-
bird pollination of some Costa Rican Heliconia species.
Ecology 56:285–301.
Stiles FG. 1976. Taste preferences, color preferences, and
flower choice in hummingbirds. Condor 78:10–26.
Stiles FG. 1978. Ecological and evolutionary implications of
bird pollination. Am Zool 18:715–27.
Stiles FG. 1981. Geographical aspects of bird-flower coevolution,
with particular reference to central America. Ann Mo Bot
Gard 68:323.
Stiles FG. 2008. Ecomorphology and phylogeny of humming-
birds: divergence and convergence in adaptations to high
elevations. Ornitologia Neotropical 19:511–9.
Stromberg MR, Johnsen PB. 1990. Hummingbird sweetness
preferences: taste or viscosity? Condor 92:606.
Tamm S, Gass CL. 1986. Energy intake rates and nectar con-
centration preferences by hummingbirds. Oecologia 70:20–3.
Tello-Ramos MC, Hurly TA, Healy SD. 2015. Traplining in
hummingbirds: flying short-distance sequences among sev-
eral locations. Behav Ecol 26:812–9.
Temeles E, Kress W. 2010. Mate choice and mate competition
in a tropical hummingbird at a floral resource. Proc Biol
Sci 277:1607–13.
Temeles EJ. 1996. A new dimension to hummingbird-flower
relationships. Oecologia 105:517–23.
Temeles EJ, Koulouris CR, Sander SE, Kress WJ. 2009. Effect
of flower shape and size on foraging performance and
trade-offs in a tropical hummingbird. Ecology 90:1147–61.
Temeles EJ, Linhart YB Masonjones M. Masonjones HD.
2002. The role of flower width in hummingbird bill
length–flower length relationships. Biotropica 13: 68–80.
Temeles EJ, Miller JS, Rifkin JL. 2010. Evolution of sexual
dimorphism in bill size and shape of hermit hummingbirds
(Phaethornithinae): a role for ecological causation. Philos
Trans R Soc B Biol Sci 365:1053–63.
14 A. Rico-Guevara et al.
Downloaded from https://academic.oup.com/icb/advance-article/doi/10.1093/icb/icab111/6288459 by Missouri State University user on 12 August 2021
Temeles EJ, Pan IL, Brennan JL, Horwitt JN. 2000. Evidence
for ecological causation of sexual dimorphism in a hum-
mingbird. Science 289:441–3.
Temeles EJ, Rankin AG. 2011. Effect of the lower lip of
Monarda didyma on pollen removal by hummingbirds.
Can J Bot 78:1164–8.
Temeles EJ, Roberts WM. 1993. Effect of sexual dimorphism
in bill length on foraging behavior: an experimental anal-
ysis of hummingbirds. Oecologia 94:87–94.
Thomson JD, Wilson P. 2008. Explaining evolutionary shifts
between bee and hummingbird pollination: convergence,
divergence, and directionality. Int J Plant Sci 169:23–38.
Torres-Vanegas F, Hadley AS, Kormann UG, Jones FA, Betts
MG, Wagner HH. 2019. The landscape genetic signature of
pollination by trapliners: evidence from the tropical herb,
Heliconia tortuosa. Front Genet 10:1206
Tripp EA, Manos PS. 2008. Is floral specialization an evolu-
tionary dead-end? Pollination system transitions in Ruellia
(acanthaceae). Evolution 62:1712–37.
V
azquez DP, Chacoff NP, Cagnolo L. 2009. Evaluating mul-
tiple determinants of the structure of plant–animal mutu-
alistic networks. Ecology 90:2039–46.
Waser NM, CaraDonna PJ, Price MV. 2018. Atypical flowers can
be as profitable as typical hummingbird flowers. Am Nat
192:644–53.
Waser NM, Fugate ML. 1986. Pollen precedence and stigma clo-
sure: a mechanism of competition for pollination between
Delphinium nelsonii and Ipomopsis aggregata. Oecologia 70:573–7.
Wei J, Huo Z, Gorb SN, Rico-Guevara A, Wu Z, Wu J. 2020.
Sucking or lapping: facultative feeding mechanisms in hon-
eybees (Apis mellifera). Biol Lett 16:20200449.
Weinstein BG, Graham CH. 2017. Persistent bill and co-
rolla matching despite shifting temporal resources in
tropical hummingbird-plant interactions. Ecol Lett
20:326–35.
Wessinger CA, Rausher MD, Hileman LC. 2019.
Adaptation to hummingbird pollination is associated
with reduced diversification in Penstemon. Evol Lett
3:521–33.
Wilson P, Castellanos MC, Wolfe AD, Thomson JD. 2006.
Shifts between bee and bird pollination in Penstemons. In:
Waser NM, Ollerton J, editors. Plant-pollinator interac-
tions: from specialization to generalization. Chicago, CA:
The University of Chicago Press. p. 47–68.
Wolf LL, Hainsworth FR, Stiles FG. 1972. Energetics of for-
aging: rate and efficiency of nectar extraction by humming-
birds. Science 176:1351–2.
Wolf LL, Hainsworth FR, Gill FB. 1975. Foraging efficiencies
and time budgets in nectar-feeding birds. Ecology
56:117–28.
Bene“fit” in bill–corolla coevolution 15
Downloaded from https://academic.oup.com/icb/advance-article/doi/10.1093/icb/icab111/6288459 by Missouri State University user on 12 August 2021
