Cinnyris of artificial nectar (Nectarinia) talatala

Cinnyris of artificial nectar  (Nectarinia) talatala
The feeding response of white-bellied sunbirds (Cinnyris
(Nectarinia) talatala) to sugar concentration and viscosity
of artificial nectar
Eckart Stolle
By
Carolina D. C. Leseigneur
Supervised by Prof. S. W. Nicolson
Submitted in partial fulfilment of the requirements for the degree MSc Zoology
in the Faculty of Natural and Agricultural Science
University of Pretoria
Pretoria
February 2008
© University of Pretoria
MSc Candidate
Miss Carolina D. C Leseigneur
Department of Zoology & Entomology
University of Pretoria
Pretoria
0002
South Africa
[email protected]
[email protected]
Student no.: 20077582
Carolina Leseigneur
One must ask children and birds how cherries and strawberries taste.
- J. W. Von Goethe
Supervisor
Prof. S. W. Nicolson
Department of Zoology & Entomology
University of Pretoria
Pretoria
0002
South Africa
[email protected]
2
Acknowledgements
I would firstly like to thank my supervisor, Prof. Sue W. Nicolson, without her guidance
and support this thesis would certainly have taken me a lot longer to put together. I have
enjoyed learning from you and your sincerity and encouragement was much appreciated.
Next I would like to thank all my colleagues and friends, who gave me some valuable
comments in the preparation of this thesis, and who were a wonderful support group
throughout this time. A special word of thanks goes to Luke Verburgt, for all his assistance
and statistical help in the work of the third chapter of this thesis. Building circuit boards,
soldering cables and programming computers late into the night with a beer was both
enjoyable and productive. My fellow “sunbirders”, Craig Symes, Cromwell Purchase and
Angela Köhler, and also Dr. Hannelie Human and Dr. Christian Pirk, your inputs were very
helpful, and the general conversations to take our minds off work were very fun. To my
special friends in and outside the department, Anne Dolinschek, Rosie van Zyl, Lia
Rotherham, Katarina Medger, René Wolmerans, Michelle Fourie, Tracy Shaw, Eckart
Stolle, Mike Ellis, Cornel Du Toit and Bernard Coetzee, thanks for being there when I just
needed to get away from it all. Your kindness, love, laughter and general enthusiasm has
been great to share in. Special thanks to Eckart Stolle for the cover photo of the whitebellied sunbird.
Most importantly, I would like to thank my family (Mom, Dad, Ivan and our feathered
and four-footed family members). Your love, support and helping me stay sane during these
years has been fantastic. Never saying no to anything I asked of you and encouraging me
throughout is truly appreciated.
No bird soars too high if he soars with his own wings.
- William Blake
3
General Abstract
Plant nectar is a simple food and is easily digested by many different species of
pollinators. Many compounds make up the composition of floral nectars, but the most
abundant are sugars, generally dominated by sucrose and the hexoses, glucose and fructose.
Nectar sugars have been measured for many plant species visited by hummingbirds,
sunbirds and other passerines, revealing a range of concentrations. The nectars of passerinepollinated flowers are generally dilute compared to those of bee-pollinated flowers. The
question why bird nectars are so dilute has been addressed in many studies. Many
hypotheses have been proposed, among them the relationship between viscosity and
drinking by birds. The viscosity of sugar solutions increases exponentially with increasing
concentration, and capillarity is inversely proportional to viscosity. Nectarivorous birds
imbibe nectar by capillarity, and high sugar concentrations could impose constraints on their
feeding efficiency. Feeding in nectarivorous birds, especially hummingbirds, has been
mostly devoted to assessing sugar type preferences. However, concentration preferences
have received less attention, and the effect of viscosity on feeding has not been examined
separately from sugar concentration for any bird species. Do nectarivorous birds show a
preference for specific concentrations at a broad and a fine scale of difference, given a
specific sugar type? Does viscosity impose a feeding limitation on nectarivorous birds?
Does it affect their feeding behaviour? Sunbirds and other nectar-feeding birds can choose
amongst various flowering plant species at any one time. Their feeding responses may have
important consequences on pollination ecology.
In this study, concentration preferences of white-bellied sunbirds were examined using
paired solutions of either sucrose or equicaloric 1:1 mixtures of glucose and fructose, at a
4
both a broad and a fine scale of difference between pairs over the concentration range of
0.25 to 2.5 M. I hypothesized that sunbirds would prefer concentrations of 1 M and higher
on sucrose solutions, while preferring concentrations less than 1 M on hexose solutions. On
both sugar types at the broad scale, the higher concentration was significantly preferred up
to 1 M, suggesting a preference for 1 M sugar solutions. At a finer scale, white-bellied
sunbirds were able to discriminate 0.03 and 0.05 M (1 and 2% w/w) concentration
differences between sucrose and hexose solutions respectively. This discrimination is
similar to that reported at low concentrations for other passerine nectar-feeders, and at
higher concentrations for hummingbirds.
To determine if high viscosity nectars limit the sugar intake of avian nectar consumers,
white-bellied sunbirds were exposed to three different test series of sucrose solutions:
control series (CS, pure sucrose 0.25 – 2.5 M), constant viscosity series (CVS, 0.25 – 0.7 M
with increased viscosity equivalent to that of 1 M sucrose) and constant concentration series
(CCS, 1 M with increased viscosities equivalent to that of 1.5, 2 and 2.5 M sucrose).
Viscosities were artificially altered with Tylose ®. The sunbirds had reduced intake rates
and gained less energy on more viscous sucrose solutions. Also, sunbirds did not alter their
feeding behaviour (feeding frequency, feeding duration, total feeding duration and feeding
interval) in any significant way when feeding on more viscous sucrose solutions. This lack
of change in feeding behaviour led to lower sugar intake rates and sugar consumption.
These results suggest that sunbirds suffer a preingestional limitation when consuming
nectars with viscosities higher than those due to sugar concentration alone.
5
TABLE OF CONTENTS
Acknowledgements ................................................................................................................ 3
General Abstract ................................................................................................................... 4
Disclaimer............................................................................................................................... 7
Chapter 1................................................................................................................................ 8
General Introduction ..................................................................................................................... 8
The white-bellied sunbird........................................................................................................................... 8
The nectars of bird-pollinated flowers ....................................................................................................... 9
Sugar type preferences of bird pollinators............................................................................................... 10
Concentration preferences of bird pollinators ......................................................................................... 13
Discrimination abilities of nectarivorous birds ....................................................................................... 15
Viscosity of sugar solutions and effects on preingestional processing..................................................... 16
Aim of study.............................................................................................................................................. 19
Chapter 2.............................................................................................................................. 29
Nectar concentration preferences of the white-bellied sunbird, Cinnyris talatala
(Nectariniidae).............................................................................................................................. 29
ABSTRACT............................................................................................................................................ 30
Introduction............................................................................................................................................ 31
Materials & Methods ............................................................................................................................. 35
Results ..................................................................................................................................................... 39
Discussion................................................................................................................................................ 48
Chapter 3.............................................................................................................................. 59
The viscosity of artificial nectar: effect on the feeding behaviour of white-bellied sunbirds,
Cinnyris talatala (Nectariniidae) ................................................................................................. 59
ABSTRACT............................................................................................................................................ 60
Introduction............................................................................................................................................ 61
Materials & Methods ............................................................................................................................. 66
Results ..................................................................................................................................................... 71
Discussion................................................................................................................................................ 84
Chapter 4.............................................................................................................................. 96
Conclusion .................................................................................................................................... 96
Appendix ............................................................................................................................ 102
Further experiments on fine scale concentration preferences of white-bellied sunbirds .... 102
Summary............................................................................................................................................... 102
Materials & Methods ........................................................................................................................... 103
Results ................................................................................................................................................... 104
General References............................................................................................................ 109
6
Disclaimer
This thesis consists of a series of chapters that have been prepared as stand-alone
manuscripts, for subsequent submission for publication purposes. Consequently, repetitions
may occur between chapters.
7
Chapter 1
General Introduction
Carolina Leseigneur
The white-bellied sunbird
Sunbirds are specialized nectarivorous passerines and the Old World counterparts of
hummingbirds (family Trochilidae). Sunbirds are important pollinators as they feed
primarily on floral nectar, occurring throughout Africa and Asia (Skead 1967; Cheke et al.
2001). They are all typified by their small size (ranging between 6 and 22 g in body mass),
long, thin curved bills, and highly energetic lifestyles. The sunbirds’ primary diet is floral
nectar from a variety of plant species, and many sunbirds species often form groups with
other sunbirds and passerines at rich nectar sources (Skead 1967). They also consume pollen
and glean foliage for small arthropods, especially during the breeding season (van Tets &
Nicolson 2000; Hockey et al. 2005).
8
The white-bellied sunbird Cinnyris (Nectarinia) talatala, (family Nectariniidae) is one
of the most common sunbirds in the greater Johannesburg and Pretoria area of South Africa
(Skead 1967). It occurs over a wide range in north-eastern sub-Saharan Africa, and also to
the west in northern Botswana and Namibia (Hockey et al. 2005). Throughout its range it
has a preference for semi-arid savannas and woodlands (Hockey et al. 2005), especially dry
Acacia thickets (Skead 1967).
Mature males of the species have a characteristic white breast and contrasting iridescent
green and blue on the head, neck, mantle, chin, throat, upper breast and lesser and median
wing coverts (Hockey et al. 2005). The flight feathers and tail tip are a dull black. Females
are duller with the upper parts a grey-olive brown, and the tail a darker brown. The young
resemble the female. The white-bellied sunbird occurs solitarily at times, but mostly in pairs
and occasionally in groups (Cheke et al. 2001; Hockey et al. 2005).
White-bellied sunbirds habituate well to captivity, and learn to feed from an artificial
feeder within three hours of capture (personal observation). This nectarivorous passerine has
proven to be a highly suitable model animal to examine relationships between floral nectars
and energy management, and the physiological adaptations of the birds in response to a
simple, watery diet (for example, Fleming et al. 2004a; Köhler et al. 2006, Napier et al.
2008).
The nectars of bird-pollinated flowers
Plant nectar is a simple food consumed by a great variety of floral visitors and
pollinators. Floral nectars contain many solutes, but the most abundant are sugars, generally
dominated by sucrose and the hexoses glucose and fructose (Baker & Baker 1982; Baker &
9
Baker 1983; Nicolson & Thornburg 2007). Nectar sugars have been measured for many
plant species visited by hummingbirds, sunbirds and other passerines, revealing a range of
sucrose, hexose and even xylose concentrations (Pyke & Waser 1981; van Wyk & Nicolson
1995; Nicolson 2002, Nicolson & Fleming 2003a). There are however marked differences
in the properties of nectar from specialized (specialized nectarivores) and generalized
(occasional nectarivores) bird pollination systems (Johnson & Nicolson 2008). Flowers
pollinated by specialized nectarivores, like hummingbirds and sunbirds, tend to be
characterized by small volumes, high sugar concentrations and high sucrose content. In
contrast, nectar from plants pollinated by generalist bird pollinators has large volumes,
dilute sugar concentrations and low sucrose content (Johnson & Nicolson 2008). In general
however, bird-pollinated floral nectars are more dilute than the nectars of bee-pollinated
flowers (Pyke & Waser 1981; Nicolson & Fleming 2003a).
Sugar type preferences of bird pollinators
Sugar type preferences in birds have been studied for some species of Japanese quail
(Harriman & Milner 1969), the Red Lory (Downs 1997) and several hummingbird and
sunbird species (see below, reviewed by Lotz & Schondube 2006). Stiles (1976) conducted
several experiments using Anna’s hummingbird (Calypte anna) and three other species of
hummingbirds (Selasphorus rufus, Archilochus alexandri, Thalurania furcata). Using pairwise tests, he found that sucrose (S) at 30% w/v was preferred over an equicaloric mixture
of the hexoses glucose (G) and fructose (F), which was preferred over glucose, while
fructose was rejected by all birds (preference order from most to least preferred: S > G:F >
G > F). However, given no other choice, hummingbirds will consume fructose over the
course of a day in equal amounts (ml h-1) to sucrose (Stiles 1976). The hierarchy of sugar
preference was the same for all hummingbird species, and fructose was rejected by nearly
10
all birds. Martínez del Rio (1990) offered 17.1% w/v solutions of sucrose, a 1:1 glucose and
fructose mixture, and glucose and fructose alone, in pairs to three species of Mexican
hummingbirds (Amazilia rutila, Cynanthus latirostris, Chlorostilbon canivetii). The birds
showed the same hierarchy of preference as those studied by Stiles (1976), even though all
birds digested all the sugars with very high and similar efficiencies. Further echoing these
results, Hainsworth and Wolf (1976) found for five species of hummingbirds that sucrose
alone was preferred when paired with glucose, and fructose alone was never preferred.
However, none of these preferences were strong, possibly due to the fact that sugars were
made up on a % w/w basis. Most articles and experiments use % w/w (g solute per 100 g
solution) concentrations due to the wide use of refractometers. Refractometers are hand held
devices which measure the refractive index of a solution placed onto the prism, which varies
according to the nature of the solute, concentration and temperature (Corbet 2003). This can
easily be converted to % w/v (g solute per 100 ml solution), by multiplying by the density of
the sugar solution at the set concentration (Corbet 2003). This is based on the assumption
that the solutions are equicaloric when measured as % w/w. However, because the
molecular mass of sucrose (342.3 g mol-1) is less than that of glucose and fructose combined
(180.2 g mol-1 each), hexose solutions will have only 95% of the energy of a sucrose
solution if mixed on a % w/w basis (Fleming et al. 2004b). Additionally, if one is examining
whether nectarivorous birds select one sugar type over another depending on factors other
than the energy value of solutions, then solutions should be presented on an equicaloric
basis (Brown et al. 2008) Another cause of weak preferences could be short-term exposure
to solutions, as feeders in Hainsworth and Wolf’s (1976) experiments were alternated in
position every half hour. Similarly, frugivorous and nectarivorous bats consistently prefer
sucrose over equicaloric solutions of glucose and fructose, and solutions of the single
hexoses in pairwise tests, when concentrations are maintained at values equivalent to the
11
nectar of bat-pollinated flowers (Herrera 1999). However, some bat species have no sugar
type preferences when concentrations of paired test solutions have the same concentrations
(Rodríguez-Peña et al. 2007). Interestingly, and in contrast to other studies on
hummingbirds, broadtailed hummingbirds (Selasphorus platycercus) do not show any
preference for sucrose or hexoses at 21ºC Ta over the range of 0.25 to 1 mol l-1 (molar,
hereafter M; Fleming et al. 2004b).
However, for other avian nectarivores, preferences have been found to be different. For
three species of nectarivorous tanagers (Tachyphonus cristatus, Dacnis cayana and
Chlorophanes spiza), pairwise tests offering sucrose and glucose at 20% concentration
yielded no preferences (Schaefer et al. 2003). Cape white-eyes (Zosterops pallidus), which
are generalist passerine nectarivores/frugivores, prefer sucrose to both fructose and glucose
though they do not discriminate significantly between fructose, glucose and equicaloric
hexose mixture (Franke et al. 1998). Since xylose (X) has been found in some floral nectars
of Proteaceae (van Wyk & Nicolson 1995), it has also been tested for some southern
African bird species. Cape white-eyes avoid xylose, yielding a preference ranking of S > F
= G = G:F > X (Franke et al. 1998). In the southern double-collared sunbird (Cinnyris
chalybeus) the order of preference for 20% sugars presented in pairwise tests was S = G:F =
F > G > X (Lotz & Nicolson 1996). The fact that fructose was preferred over glucose is
surprising in view of hummingbird preferences. Cape sugarbirds (Promerops cafer) were
also found to significantly prefer fructose over glucose at 20% in pairwise tests (Jackson et
al. 1998).
12
Concentration preferences of bird pollinators
The limited research on nectar-sugar concentration preferences in birds shows that
preferences tend to be for concentrations higher than those of natural nectars. Stiles (1976)
found that when Anna’s hummingbirds were presented with choices between 15, 30 and
45% w/v or between 30, 45 and 60% concentrations in three-way preference tests, repeated
for all three sugars (sucrose, glucose and fructose), the birds preferred the highest
concentrations available for both sucrose and glucose, but favoured the lower concentrations
for fructose. Tamm and Gass (1986) found similar results for rufous hummingbirds
(Selasphorus rufus) using four-way feeder tests with sucrose solutions. The birds preferred
more concentrated sucrose solutions when offered concentrations less than 45%, but
preferred more dilute solutions when offered concentrations higher than 55%.
Hummingbirds were found to maximize their gain (amount of energy per unit time) of
nectar by selecting the more concentrated sucrose solutions between concentration pairs
(Hainsworth & Wolf 1976). Interestingly, flower visiting bats (Glossophaga soricina
antillarum) also prefer the more concentrated of two solutions when offered honey diluted
with water to concentrations less than or equal to 50% w/w sucrose equivalents, SE (Roces
et al. 1993). Concentration of sucrose differs from that of the hexoses glucose and fructose
due to their different molecular masses: 30% w/w sucrose is 1 M, while 30% w/w of either
glucose or fructose is 1.85 M (Weast 1980, see Appendix). Hence, concentrations are often
made to be equivalent to that of sucrose. However, when concentrations are equal to or
higher than 60% SE, bats prefer the more dilute solution, and also consume more free water
which has a dilution effect. In another study, Saussure’s long-nosed bat (Leptonycteris
curasoae) was found to prefer concentrated over dilute artificial nectar in paired tests
(sucrose- or hexose-dominated) regardless of sugar type at various concentrations, whereas
Glossophaga soricina showed no preference except for preferring the more concentrated
13
(27% w/v) hexose-dominated solution over a more dilute (18% w/v) sucrose-dominated
solution (Rodríguez-Peña et al. 2007). In general, hummingbird and bat preferences seem to
be for sugar concentrations much higher than those found in natural nectars. In the southern
double-collared sunbird (Cinnyris chalybeus), preliminary results show that sucrose and
fructose at 20 and 30% w/w were equally accepted and preferred over 10% solutions; but
for glucose, 10 and 20% were equally accepted and preferred over 30% solutions (Lotz &
Nicolson 1996).
In more recent studies, sugar type preferences have been found to be concentrationdependent. In white-bellied sunbirds a strong preference for hexose sugars at 0.1 M SE was
found at 21ºC, while they were indifferent to hexose or sucrose solutions of 0.25 M or
higher, although they significantly preferred sucrose solutions at 0.75 M (Fleming et al.
2004b). In the same study, broadtailed hummingbirds showed only a slight (non-significant)
preference for hexoses on 0.25 M at 10ºC (as hummingbirds use torpor, two temperature
extremes were used), but no preference for sucrose. Hummingbirds also showed no
preferences for sucrose or hexoses at 21ºC (Fleming et al. 2004b). The findings on the
white-bellied sunbird however are similar to data for the cinnamon-bellied flowerpiercer
(Diglossa baritula) and the magnificent hummingbird (Eugenes fulgens) (Schondube &
Martínez del Rio 2003). Sugar type preferences in the latter species were concentrationdependent, and preference for hexoses was found at lower concentrations (146 mM SE) in
contrast to a preference for sucrose at higher concentrations (1168 mM SE). At intermediate
concentration (584 mM SE) the birds were indifferent to hexose or sucrose solutions. The
switch from preferring sucrose at high concentrations to hexose preference at low
concentrations could be common among specialized nectarivorous birds (see Lotz &
Schondube 2006). This interesting as hexose dominated nectars tend to be more dilute than
14
sucrose dominated nectars (Nicolson 2002). But it is important to note that nectarivorous
birds, particularly sunbirds, adjust the volume of nectar consumed according to sugar
concentration (Downs 1997; Lotz & Nicolson 1999), and have been found to show perfect
compensatory feeding on sucrose concentrations between 8.5 and 65% w/w (0.25 and 2.5
M) (Nicolson & Fleming 2003b).
Discrimination abilities of nectarivorous birds
All the studies mentioned above reported feeding preferences over broad scales of
concentration. There is very limited data on preferences at a finer scale. Lloyd (1989) found
that the greater double-collared sunbird (Cinnyris afer) and the malachite sunbird
(Nectarinia famosa) selected the more concentrated of two sucrose solutions (0.25 to 1 M
solutions, pairs differing in concentration by either 0.05, 0.1 and 0.2 M) at low
concentrations, though at higher concentrations this discrimination decreased. For three
species of tanagers (Family Dacnidinae), the difference between sugar solutions was
reduced to 1% in pairwise tests (either glucose or sucrose), and the birds preferred the 6%
over the 5% solution, though at 12 and 13% they did not show a preference (Schaefer et al.
2003). Thus discrimination varies with sugar concentration. Rufous hummingbirds
(Selasphorus rufus) were tested for sucrose concentration preferences in four-way field
trials (Blem et al. 2000). Concentrations ranged from 10 to 70% w/v, and for each test the
four concentrations offered together differed by 10% increments. There were significant
differences between all concentrations from 10 to 50%, but above 50% hummingbirds never
showed a preference. In a separate experiment, concentration differences were reduced by 4
or 6%, and when a significant preference was obtained, the difference between
concentrations was further reduced until a non-significant result was found. When
concentrations approximated values typical of those found in hummingbird-pollinated
15
flowers (approximately 20% on average), the birds could distinguish solutions differing by
as little as 1% (Blem et al. 2000).
Viscosity of sugar solutions and effects on preingestional processing
The question why bird nectars, particularly those of passerine-pollinated flowers, are so
dilute in sugar concentration has been addressed in several studies (for example, Nicolson
2002, Johnson & Nicolson 2008). Many hypotheses have been proposed (for a review see
Nicolson 2002), among them the relationship between viscosity and drinking by birds which
was first proposed by Baker (1975). Baker hypothesized that low concentration-low
viscosity nectars were necessary for efficient extraction of nectar from flowers. Dynamic
viscosity (measured in mPa.s) can be defined as the internal friction of a fluid which resists
forces that cause change in its form (Weast 1980; Bourne 1982; Vogel 1994). The viscosity
of sugar solutions increases exponentially with increasing concentration, and capillarity is
inversely proportional to viscosity. Capillarity is defined as a physical phenomenon caused
by surface tension that results in the surface of a liquid rising or falling in contact with a
solid (Bourne 1982; Vogel 1994). Nectarivorous birds imbibe nectar by capillarity along
their tongues, either down bifurcated “tubes” as in sunbirds and hummingbirds, or by
collection on brush-tipped tongues as in honeyeaters (Schlamowitz et al. 1967; Skead 1976;
Kingsolver & Daniel 1983; Paton & Collins 1989; Cheke et al. 2001; Downs 2004), and
high sugar concentrations with high viscosities could impose constraints on the feeding
efficiency of birds. The effect of viscosity on feeding has not been explicitly examined
separately from sugar concentration for any bird species (but see Stromberg & Johnsen
1990).
16
The concentration of nectar sugars, as well as ambient temperature and other nectar
solutes affect the viscosity of nectar (Heyneman 1983; Bourne 1982; Mathlouthi &
Génotelle 1995). Models based on the feeding of hummingbirds propose that for large
volumes that require several licks, both fluid loading and unloading of the tongue are
required and higher concentrations maximize extraction efficiency in terms of energy
content. Fluid loading occurs through capillarity up the grooves of the tongues of
nectarivorous birds. The unloading phase requires muscular work of the tongue and related
muscles (see Schlamowitz et al. 1976; Skead 1976; Downs 2004). Small volumes of nectar
can be loaded on a single lick, and because of capillarity it requires no unloading phase
(Heyneman 1983; Kingsolver & Daniel 1983; Gass & Roberts 1992). The viscosity of
sucrose solutions increases exponentially as a function of concentration while the energy
content increases linearly, and volumetric flow rate decreases with concentration (Weast
1980; Gass & Roberts 1992; Josens & Farina 2001). It is expected then that maximum
energy intake rates are a compromise between energy density and viscosity. This implies
that the most efficient energy intake by nectarivorous birds could fall at intermediate levels
of sugar concentration (Baker 1975; Pyke & Waser 1981).
Studies on insects reveal that viscosity does impact feeding behaviour. In honeybees
(Apis mellifera), trophallactic food transfer increases with sucrose concentration up to 30%,
but is impeded at higher concentrations by the effect of viscosity, either because it is not
energetically worthwhile for the donor or because of a direct physical limitation (Tezze &
Farina 1999). In a nectarivorous ant (Camponotus mus), it has been found that foraging
behaviour is affected by body size in conjunction with viscosity (large versus small workers
varied in feeding time on different viscosity of solutions; Medan & Josens 2005). In the
hovering hawk moth (Macroglossum stellatarum), intake rate was analyzed by separating
17
the effects of concentration and viscosity (measured as kinematic viscosity – dynamic
viscosity divided by the density of the solution) using Tylose ®, an inert polysaccharide
(Josens & Farina 2001). Presented with different solutions at single feeders following a
detailed experiment of varying concentrations and viscosities, it was found that both
viscosity and concentration of sucrose in a solution influence the intake rate. For
concentrations lower than 50% w/w with viscosities equivalent to 50%, hawk moths had
higher ingestion rates. Solutions of equal viscosity should be ingested at the same rate if it
was the only physical limitation, but this was not the case for hawk moths.
The effect of viscosity on the feeding variables of nectar-feeding birds has only been
looked at to date by Stromberg & Johnsen (1990). They examined the independent effects of
artificial sweeteners (including aspartame), sucrose solutions and viscosity (equal to that of
sucrose solutions) in field experiments on the feeding preferences of black-chinned
hummingbirds (Archilochus alexanderi). Artificial sweetener solutions were increased to
sucrose equivalent viscosities with sodium carboxymethylcellulose (CMC), an anionic
substance. Artificial sweeteners alone were first found to be ignored by the hummingbirds,
and increasing their viscosity made no difference to their acceptance. A 20% sucrose
solution was significantly preferred over a CMC solution of equal viscosity to 20% sucrose,
and the authors concluded that sweet stimuli rather than the mechanical effect of high
viscosity is required to induce feeding in black-chinned hummingbirds. However, these
results are questionable for two important reasons. Firstly, CMC is anionic and ions can
alter viscosity of fluids due to the charge they carry which may alter chemical bonds or form
new bonds (Bourne 1982; Vogel 1994. Secondly, many artificial sweeteners, like aspartame,
are comprised primarily of amino acids. Amino acids can also alter the viscosity of
solutions, depending on concentration or molecular mass, or even the apparent specific
18
volumes of the amino acids, which directly affect taste perception (Bourne 1982; Birch &
Kemp 1989; Nicolson 2007). It has also been found that both hummingbirds and sunbirds
avoid sugar solutions with high concentrations of amino acids, and are indifferent to them at
lower concentrations (Hainsworth & Wolf 1976; Leseigneur et al. 2007).
Aim of study
In this study I will address the effects of concentration and viscosity of artificial nectars
on the feeding response of white-bellied sunbirds (Cinnyris talatala).
Chapter 1 – Concentration preferences
Given a specific sugar type, will nectarivorous birds show a preference for specific
concentrations, at both a broad and a fine scale of difference? White-bellied sunbirds, a
specialized nectarivorous passerine, will be exposed to concentration pairs of sucrose and
pairs of equicaloric glucose and fructose mixtures of different concentrations. Acuteness of
discrimination at a fine scale of difference between concentrations will be assessed
similarly. Since sugar type preferences have been shown to be concentration-dependent in
some bird species (Schondube & Martínez del Rio 2003; Fleming et al. 2004b; Fleming et
al. 2008), it is expected that there will be a difference in concentration preferences between
sucrose and hexose sugars – sunbirds should prefer higher concentrations on sucrose
solutions, but lower concentrations on hexose solutions. It is further expected that sugar
(energy) intake will be constant over the concentration range as white-bellied sunbirds
employ compensatory feeding (Nicolson & Fleming 2003b). At a finer scale of difference in
concentration, I expect that sunbirds, like hummingbirds and some other passerines (Levey
1987; Lloyd 1989; Blem et al. 2000; Schaefer et al. 2003), will be able to discriminate
between small differences in sugar concentration.
19
Chapter 2 – The effect of viscosity on feeding behaviour
Baker (1975) suggested that dilute (low sugar concentration) nectar of bird-pollinated
flowers was necessary to increase the efficiency of nectar extraction from flowers by birds,
as higher concentrations with higher viscosities may impede extraction efficiency. Does
viscosity impose a feeding limitation on nectarivorous birds, and if so, how does it affect
their feeding behaviour? I will determine dynamic viscosity values of artificial nectars
containing pure sugars (sucrose and hexose mixture). The effect of viscosity will then be
examined (independently of concentration) using a polysaccharide (Tylose ®), to determine
differences in sugar (energy) consumption rates and feeding behaviour. The sunbirds will be
exposed to three different treatment series as per Josens & Farina (2001), with varied
concentrations and artificially increased viscosities. As white-bellied sunbirds use
compensatory feeding (Nicolson & Fleming 2003b) and assuming that higher viscosity
limits the amount of food ingested per unit volume, it is expected that sunbirds will
consume solutions of low concentration with high viscosity at higher than normal rates.
Concurrently, they should increase their feeding frequency, feeding duration per feeding
event, total feeding duration and decrease the feeding interval between feeding events so as
to compensate for increased viscosities behaviourally. However, if viscosity is the only
physical limitation, solutions of equal viscosity should be ingested at the same rates.
References
Baker HG (1975) Sugar concentrations in nectars from hummingbird flowers. Biotropica
7:37-41
20
Baker HG, Baker I (1982) Chemical constituents of nectar in relation to pollination
mechanisms and phylogeny. In: Nitecki HM (ed) Biochemical aspects of evolutionary
biology. University of Chicago Press
Baker HG, Baker I (1983) Floral nectar sugar constituents in relation to pollinator type. In:
Jones CE, Little RJ (eds) Handbook of experimental pollination biology. Van Nostrand
Reinhold, New York
Birch GG, Kemp SE (1989) Apparent specific volumes and tastes of amino acids. Chem
Senses 14:249-258
Blem CR, Blem LB, Felix J, van Gelder J (2000) Rufous hummingbird sucrose preference:
precision of selection varies with concentration. Condor 102:235-238
Bourne MC (1982) Food science and technology: Concept and measurement. Academic
Press Inc., New York, USA
Brown M, Downs CT, Jonson SD (2008) Sugar preferences of nectar feeding birds: a
comparison of experimental techniques. J Avian Biol in press
Cheke CA, Mann CF, Allen R (2001) Sunbirds: A guide to the sunbirds, flowerpeckers,
spiderhunters and sugarbirds of the world. Christopher Helm, London, UK
Corbet SA (2003) Nectar sugar content: estimating standing crop and secretion rate in the
field. Apidologie 34:1-10
21
Downs CT (1997) Sugar preference and apparent sugar assimilation in the Red Lory. Aust J
Zool 45:613-619
Downs CT (2004) Some preliminary results of studies on the bill and tongue morphology of
Gurney’s sugarbird and some southern African sunbirds. Ostrich 75:169-175
Fleming PA, Gray DA, Nicolson SW (2004a) Circadian rhythm of water balance and
aldosterone excretion in the whitebellied sunbird Nectarinia talatala. J Comp Physiol B
174:341-346
Fleming PA, Hartman Bakken B, Lotz CN, Nicolson SW (2004b) Concentration and
temperature effects on sugar intake preferences in a sunbird and a hummingbird. Funct Ecol
18:223–232
Franke E, Jackson S, Nicolson S (1998) Nectar sugar preferences and absorption in a
generalist African frugivore, the Cape White-eye Zosterops pallidus. Ibis 140:501-506
Gass CL, Roberts WM (1992) The problem of temporal scale in optimization: three
contrasting views of hummingbird visits to flowers. Am Nat 140:829-853
Hainsworth FR, Wolf LL (1976) Nectar characteristics and food selection by
hummingbirds. Oecologia 25:101–113
Harriman AE, Milner JS (1969) Preference for sucrose solutions by Japanese quail
(Coturnix coturnix japonica) in two-bottle drinking tests. Am Midl Nat 81:575-578
22
Herrera LG (1999) Preferences for different sugars in neotropical nectarivorous and
frugivorous bats. J Mammal 80:683-688
Heyneman AJ (1983) Optimal sugar concentrations of floral nectars – dependence on sugar
intake efficiency and foraging costs. Oecologia 60:198-213
Hockey PAR, Dean WRJ, Ryan PG (eds) (2005) Roberts: Birds of Southern Africa, VIIth
edition. The Trustees of the John Voelcker Bird Book Fund, Cape Town, South Africa
Jackson S, Nicolson SW, Lotz CN (1998) Sugar preferences and “side bias” in Cape
Sugarbirds and Lesser Double-Collared Sunbirds. Auk 115:156–165
Johnson SD, Nicolson SW (2008) Evolutionary associations between nectar properties and
specificity in bird pollination systems. Biol Lett 4:49-52
Josens RB, Farina WM (2001) Nectar feeding by the hovering hawk moth Macroglossum
stellatarum: intake rate as a function of viscosity and concentration of sucrose solutions. J
Comp Physiol A 187:661-665
Kingsolver JG, Daniels TL (1983) Mechanical determinants of nectar feeding strategy in
hummingbirds: energetics, tongue morphology, and licking behaviour. Oecologia 60:214226
23
Köhler A, Verburgt L, Nicolson SW (2006) Short-term energy regulation of whitebellied
sunbirds (Nectarinia talatala): effects of food concentration on feeding frequency and
duration. J Exp Biol 209:2880-2887
Leseigneur CDC, Verburgt L, Nicolson SW (2007) Whitebellied sunbirds (Nectarinia
talatala, Nectariniidae) do not prefer artificial nectar containing amino acids. J Comp
Physiol B 177:679-685
Lloyd P (1989) Sucrose concentration preferences of two southern African sunbirds. Ostrich
60:134-135
Lotz CN, Nicolson SW (1996) Sugar preferences of a nectarivorous passerine bird, the
Lesser Double-collared Sunbird (Nectarinia chalybea). Funct Ecol 10:360-365
Lotz CN, Nicolson SW (1999) Energy and water balance in the lesser double-collared
sunbird (Nectarinia chalybea) feeding on different nectar concentrations. J Comp Physiol B
169:200-206
Lotz CN, Schondube JE (2006) Sugar preferences in nectar- and fruit-eating birds:
behavioural patterns and physiological causes. Biotropica 38:3-15
Martínez del Rio C (1990) Sugar preferences in hummingbirds: The influence of subtle
chemical differences on food choice. Condor 92:1022-1030
24
Mathlouthi M, Génotelle J (1995) Rheological properties of sucrose solutions and
suspensions. In: Mathlouthi M, Reiser P (eds) Sucrose properties and applications. Blackie
Academic & Professional
Medan V, Josens RB (2005) Nectar foraging behaviour is affected by ant body size in
Camponotus mus. J Insect Physiol 51:853-860
Napier KR, Purchase C, McWhorter TJ, Nicolson SW, PA Fleming (2008) The sweet life:
diet sugar concentration influences paracellular glucose absorption. Biol Lett (doi:
10.1098/rsbl.2008.0253)
Nicolson SW (2002) Pollination by passerine birds: why are the nectars so dilute? Comp
Biochem Physiol B 131:645-652
Nicolson SW, Fleming PA (2003a) Nectar as food for birds: the physiological consequences
of drinking dilute sugar solutions. Plant Syst Evol 238:139-153
Nicolson SW, Fleming PA (2003b) Energy balance in the whitebellied sunbird Nectarinia
talatala: constraints on compensatory feeding, and consumption of supplementary water.
Funct Ecol 17:3-9
Nicolson SW (2007) Amino acids concentrations in the nectars of southern African birdpollinated flowers, especially Aloe and Erythrina. J Chem Ecol 33:1707-1720
25
Nicolson SW, Thornburg RW (2007) Nectar chemistry. In: Nicolson SW, Nepi M, Pacini E
(eds) Nectaries and nectar. Springer, Dordrecht
Paton DC, Collins BG (1989) Bills and tongues of nectar-feeding birds: a review of
morphology, function and performance, with intercontinental comparisons. Aust Ecol
14:473-506
Pyke GH, Waser NM (1981) The production of dilute nectars by hummingbird and
honeyeater flowers. Biotropica 13:260-270
Roces F, York W, von Helversen O (1993) Nectar concentration preference and water
balance in a flower visiting bat, Glossophaga soricina antillarum. In: Barthlott W et al.
(eds) Animal-plant interactions in tropical environments. Bonn: Museum Koenig
Rodríguez-Peña N, Stoner KE, Schondube JE, Ayala-Berdón J, Flores-Ortiz CM Martínez
del Rio C (2007) Effects of sugar composition and concentration on food selection by
Saussure’s long-nosed bat (Leptonycteris curasoae) and the long-tongued bat (Glossophaga
soricina). J Mammal 88:1466-1474
Schaefer HM, Schmidt V, Bairlein F (2003) Discrimination abilities for nutrients: which
difference matters for choosy birds and why? Anim Behav 65:531-541
Schlamowitz R, Hainsworth FR, Wolf LL (1976) On the tongues of sunbirds. Condor
78:104-107
26
Schondube JE, Martínez del Rio C (2003) Concentration-dependent preferences in nectarfeeding birds: mechanisms and consequences. Funct Ecol 17:445-453
Skead CJ (1967) Sunbirds of southern Africa. Cape & Transvaal Printers Ltd. Cape Town,
South Africa
Stiles FG (1976) Taste preferences, colour preferences, and flower choice in hummingbirds.
Condor 78:10-26
Stromberg MR, Johnsen PB (1990) Hummingbird sweetness preferences: taste or viscosity?
Condor 92:606-612
Tamm S, Gass CL (1986) Energy intake rates and nectar concentration preferences by
hummingbirds. Oecologia 70:20-23
Tezze AA, Farina WM (1999) Trophallaxis in the honeybee, Apis mellifera: the interaction
between viscosity and sucrose concentration of the transferred solution. Anim Behav
57:1319-1326
van Tets IG, Nicolson SW (2000) Pollen and nitrogen requirements of the lesser doublecollared sunbird. Auk 117:826-830
van Wyk BE, Nicolson SW (1995) Xylose is a major nectar sugar in Protea and Faurea. S
Afr J Sci 91:151-153
27
Vogel S (1994) Life in moving fluids: the physical biology of flow. 2nd edn. Princeton
University Press
Weast RC (1980) CRC handbook of chemistry and physics, 60th edn. CRC Press, Inc.
Florida USA
28
Chapter 2
Nectar concentration preferences of the white-bellied sunbird,
Cinnyris talatala (Nectariniidae)
Carolina Leseigneur
29
ABSTRACT
The responses of nectar-feeding birds to nectar sugar concentrations are much less studied
than those to sugar types. Given a specific sugar type, do nectarivorous birds show a
preference for specific concentrations at both a broad and a fine scale of difference?
Concentration preferences of white-bellied sunbirds, Cinnyris talatala, were examined using
paired solutions of either sucrose or equicaloric 1:1 mixtures of glucose and fructose.
Preferences were first examined at a broad scale over the concentration range 0.25 to 2.5 M
(0.25 or 0.5 M differences). On both sugar types, the higher concentration was significantly
preferred up to 1 M, suggesting a preference for 1 M sugar solutions. Sunbirds ingested
more sugar on the hexose mixtures than on sucrose on all concentrations. Discrimination
between concentrations was then examined at a finer scale of 0.03 or 0.05 M differences
(molar equivalents of 1 and 2% w/w). White-bellied sunbirds were able to discriminate 0.03
and 0.05 M concentration differences between sucrose and hexose solutions respectively, at
both low and high concentrations. This discrimination is similar to that reported at low
concentrations for other passerine nectar-feeders, and at higher concentrations for
hummingbirds.
30
Introduction
Nectar-feeding birds have access to a wide range of nectar concentrations (Nicolson &
Fleming 2003a; Nicolson & Thornburg 2007) and this has consequences for satisfying
energy requirements and water balance. Nectar solutes are varied, but the most abundant are
sucrose and the hexoses glucose and fructose (Baker & Baker 1982; Baker & Baker 1983;
Nicolson & Thornburg 2007). There are marked differences in the properties of nectar from
specialized and generalized bird pollination systems. Flowers pollinated by specialized
nectarivores, like hummingbirds and sunbirds, tend to contain small volumes of nectar (1030 μl), high sugar concentrations (15-25% w/w or 0.45-0.8 M sucrose equivalents, SE) and
high sucrose content (40-60% total sugar). In contrast, nectar from plants pollinated by
generalist bird pollinators has large volumes (40-100 μl), dilute sugar concentrations (8-12%
or 0.23-0.37 M SE) and low sucrose content (0-5%) (Johnson & Nicolson 2008).
Because there are differences in the sucrose content of nectar, sugar type preferences
have been intensively investigated in various nectar-feeding birds (for a review, see Lotz &
Schondube 2006). In more recent studies sugar type preferences have been found to be
concentration-dependent. In white-bellied sunbirds a strong preference for hexose sugars at
0.1 M (sucrose equivalent) was found, while the birds were indifferent to hexose or sucrose
solutions of 0.25 M or higher, although they preferred sucrose solutions at 0.75 M (Fleming
et al. 2004b). This finding is similar to studies on the cinnamon-bellied flowerpiercer
(Diglossa baritula) and the magnificent hummingbird (Eugenes fulgens) (Schondube &
Martínez del Rio 2003). A similar change of sugar preference with concentration occurs in
two honeyeaters, the New Holland honeyeater (Phylidonyris novaehollandiae) and the red
wattlebird
(Anthochaera
carunculata),
and
the
rainbow
lorikeet
(Trichoglossus
31
haematonotus) (Fleming et al. 2008). However, the preference for hexoses was only
significant at very dilute concentrations (less than 0.25 M) for the two honeyeaters, whereas
rainbow lorikeets demonstrated hexose preferences up to 0.75 M. Sugar type preferences in
all these species are concentration-dependent, and the switch from preferring sucrose at high
concentrations to hexose at low concentrations may be common among specialized
nectarivorous birds (see Lotz & Schondube 2006).
The preferences of nectar-feeding birds for different food concentrations have received
less attention. Most nectar of plants pollinated by hummingbirds, sunbirds and other
passerines are considered dilute, averaging 8 to 25% w/w (0.23 – 0.8 M SE) in sugar
concentration (Baker 1975; Pyke & Waser 1981; Nicolson 2002; Johnson & Nicolson
2008). However, in choice tests with sucrose solutions it has been consistently found that
birds prefer the highest available concentrations. Presented with choices between 15, 30 and
45% or between 30, 45 and 60% w/v in three-way preference tests (30% w/v ≈ 0.88 M SE),
Anna’s hummingbirds (Calypte anna) showed a preference for the highest concentrations
available (Stiles 1976). In four-way preference tests, Tamm & Gass (1986) found that
rufous hummingbirds (Selasphorus rufus) preferred the highest solutions up to 45% w/w.
Similarly, in four-way field preference tests, rufous hummingbirds consistently chose the
highest concentrations available up to 50% w/v (Blem et al. 2000). In a passerine
nectarivore, the bananaquit (Coereba flaveola), the results were similar: given a choice of
paired feeders, they chose the highest sucrose concentrations up to 0.74 M; Mata & Bosque
2004).
Hexoses have been used less frequently in concentration preference tests, but
hummingbirds also chose the highest concentrations of glucose, while the lowest
32
concentrations of fructose solutions were favoured (Stiles 1976; Martínez del Rio 1990).
Interestingly, similar results have been obtained for flower-visiting bats. Given paired
feeders with diluted honey at various concentrations, Glossophaga soricina antillarum bats
preferred the highest concentrations up to 50% w/w SE (approximately 1.8 M SE; Roces et
al. 1993). Honey contains predominantly glucose and fructose, and most contain very little
sucrose (Johannsmeier 2001). In another study, Saussure’s long-nosed bat (Leptonycteris
curasoae) was found to prefer concentrated over dilute artificial nectar in paired tests
(sucrose- or hexose-dominated solutions) regardless of sugar type at various concentrations,
whereas Glossophaga soricina showed no preference except for preferring the more
concentrated (27% w/v) hexose-dominated solution over a more dilute (18% w/v) sucrosedominated solution (Rodríguez-Peña et al. 2007). Preliminary results for the southern
double-collared sunbird (Cinnyris chalybeus) showed that sucrose and fructose at 20 and
30% w/w (0.65 and 1 M SE) were equally accepted and preferred over 10% (0.33 M SE)
solutions. But for glucose, 10 and 20% were equally accepted and preferred over 30%
solutions (Lotz & Nicolson 1996).
Hummingbirds have been found to maximize their extraction efficiency of nectar
(Hainsworth & Wolf 1976) as well as energy intake rates per foraging bout (leaving perch to
visit a feeder and return to perch; Roberts 1996) by selecting the more concentrated sucrose
solutions between pairs. In order to obtain the same amount of energy at various
concentrations, choosing dilute nectars would involve more time spent foraging. Choosing
more concentrated nectars would allow the birds more free time for other activities that may
increase their fitness (Roberts 1996). This could help explain such choices in many nectarfeeding birds.
33
If concentration preferences can be established for a species, how fine are these
preferences? It is essential to quantify discrimination abilities of species at a finer scale to
further understand food choices and the mechanisms involved. For two African sunbird
species in laboratory pairwise tests, Lloyd (1989) found that the birds selected the more
concentrated of two sucrose solutions differing by as little as 0.05 M (2% w/w), up to 0.5 M.
Levey (1987) found in pairwise laboratory trials that three tanager species (Family
Dacnidinae) consistently chose the higher concentration of fruit pulps differing by only 2%
in sucrose concentration (8, 10 and 12% or 0.23, 0.33 and 0.37 M SE). In a more recent
study, three species of tanagers offered solid diet pairs (5 vs. 6%) of sucrose or glucose
consistently chose the higher concentration (Schaefer et al. 2003). Blem et al. (2000)
conducted four-way preference tests on rufous hummingbirds in the field. Concentration
differences were reduced by 4 or 6% w/v, and when a significant preference was obtained,
the difference between concentrations was further reduced until a non-significant result was
found. When concentrations approximated values typical of those found in hummingbirdpollinated flowers (20% w/v) the birds could distinguish sucrose solutions differing by as
little as 1%. These studies suggest that discrimination ability varies with concentration in
nectarivorous birds. From the limited data available, passerines appear to have more acute
discrimination abilities at low concentrations, whereas hummingbirds seem to discriminate
best at higher concentrations.
Do nectarivorous birds have a preference for specific concentrations within a sugar
type? If so, how acute is their discrimination ability between concentrations? In this study I
assessed the concentration preferences of a specialized nectarivorous passerine, the whitebellied sunbird (Cinnyris talatala) using both sucrose and equicaloric hexose solutions.
Sugar solutions were prepared using molar (mol l-1 or M) concentrations. Many sugar
34
preference trials are based on sugars mixed on a % w/w (g of solute per 100 g of solution)
basis, on the assumption that the solutions are equicaloric. However, because the molecular
mass of sucrose (342.3 g mol-1) is less than that of glucose and fructose combined (180.2 g
mol-1 each), hexose solutions will have only 95% of the energy of a sucrose solution if
mixed on a % w/w basis (Fleming et al. 2004b). The birds were exposed to concentration
pairs of sucrose and equicaloric hexose (glucose and fructose) mixtures to determine their
broad preferences. Fine scale concentration preferences were similarly assessed for both
sugar types. Since sugar type preferences have been shown to be concentration dependent
for some bird species (Schondube & Martínez del Rio 2003; Fleming et al. 2004b, Fleming
et al. 2008), it is expected that there will be a difference in concentration preferences
between sucrose and hexose solutions – sunbirds should prefer high concentrations for
sucrose solutions while preferring low concentrations for hexose solutions. It is further
expected that sugar (energy) intake will be constant over the concentration range as whitebellied sunbirds employ compensatory feeding (Nicolson & Fleming 2003b). At a finer
scale of difference between concentrations, I expect that sunbirds, like hummingbirds and
other passerines (Levey 1987; Lloyd 1989, Blem et al. 2000; Schaefer et al. 2003), will be
able to discriminate between small differences in sugar concentration.
Materials & Methods
Birds and Maintenance
White-bellied sunbirds were captured by mist-netting at Jan Cilliers Park in Pretoria,
South Africa. Seven males and three females (8.91 ± 0.25 g SE), all with full adult plumage,
were captured during April 2006. Birds were maintained in an outdoor aviary (8 x 2 x 5 m),
at the University of Pretoria’s Experimental Farm. They were fed a maintenance diet of 0.63
35
M sucrose mixed with a nutritional supplement for protein, vitamins and minerals (Ensure®,
Abbott Laboratories, Johannesburg, South Africa). This diet and supplementary water were
available to the birds ad libitum in inverted stoppered syringes (feeders), and water baths
were also provided.
During the laboratory trials, the birds were maintained individually in cages of 40 x 42 x
28 cm in a climate-controlled (CC) room at 20 ± 2 ºC and 45% RH on a 12:12 h light: dark
cycle (lights on at 07.00 h) as in previous studies on white-bellied sunbirds (for example,
Nicolson & Fleming 2003b; Fleming et al. 2004a and b). Temperature and humidity were
maintained by automated Siemens RLU 220 and 210 temperature and humidity regulators
and an indoor air conditioner (Johnson Controls cr 722 CAREL). Dawn and dusk were
simulated with 0.5 h of dimmed light at the beginning and the end of the photoperiod. The
same maintenance diet was provided. Birds were allowed to acclimate to laboratory
conditions for one week prior to the experiments. Birds were released at the end of the
study.
Experiments
Solutions of sucrose and equicaloric 1:1 mixtures of glucose and fructose (hereafter
referred to as hexose) ranging from 0.25 to 2.5 M were tested separately. The difference
within each concentration pair was 0.5 M, except for the lowest concentration pair which
differed by 0.25 M. Below 0.25 M the birds cannot maintain their energy balance at 20ºC
(Nicolson & Fleming 2003b).
Sunbirds (n = 10) were simultaneously tested over two days per pairwise preference test.
Feeders were weighed at the start and end of the 6 h experimental period (07.00 to 13.00 h)
36
with a digital balance (± 0.01g, Mettler Toledo PB-602S, Microsep Ltd., Johannesburg) in
order to determine the amount of food ingested. White-bellied sunbirds in captivity have
been reported to generally show the most constant food intake rate during the first half of
the day (Fleming et al. 2004a), and have longer feeding durations and lower feeding
frequency in the early morning (Köhler et al. 2006), hence the choice of experimental
period. Experimental diets were randomized and feeders switched at 10.00 h (after 3 h) to
compensate for possible side biases (Jackson et al. 1998). To further compensate for
possible side biases, the initial feeder position was also reversed at the start of the second
test day. The maintenance diet was provided at the end of the experimental period each day
so that the birds could replace any mass lost as a result of feeding on sugar-only solutions
(Nicolson et al. 2005). Spilt food was collected beneath feeders in containers with liquid
paraffin, to prevent evaporation, and corrections were made to the consumption values (this
averaged 0.089 ± 0.014 SE g per 6 h experiment, n = 10).
Based on the results of the broad scale concentration preferences (see Results),
discrimination ability was examined at a finer scale following the same experimental
procedures. The difference between pairs around concentrations of 0.25, 0.50 and 1 M was
reduced to 0.03 M for sucrose and 0.05 M for equicaloric glucose and fructose mixtures
(Table 1).
Table 1: Concentration pairs differing by 0.03 M for sucrose and 0.05 M for hexose mixtures
used in discrimination tests (representing differences of 1 and 2 % w/w respectively).
Sucrose
(M)
Glucose: Fructose
(M)
0.27 vs. 0.3
0.25 vs. 0.3
0.47 vs. 0.5
0.45 vs. 0.5
0.97 vs. 1
0.95 vs. 1
37
Data were collected as mass (g) of each solution consumed (mean ± SE). This was then
converted to volumetric consumption (ml) based on the density of sugar solutions at each
concentration (from Weast 1980), and to sugar (energy) consumption (in grams). Hexoses
and sucrose have the same energy value per unit mass, 16.48 x 103 kJ/g (Nicolson &
Fleming 2003b). Sugar consumption data were analyzed using the STATISTICA 7.1
(StatSoft Inc. ©, Tulsa, OK, USA, 2006) software package. Total consumption is defined as
the sum of the sugar consumed for both feeders of paired diets for each sugar type per test
day. The preference index is defined as the sugar consumed at the higher concentration
divided by the total consumption.
Statistical Analysis
For initial analysis of both broad and fine scale data, an overall repeated-measures
ANOVA (RM-ANOVA) was performed for all of the consumption data to determine where
differences occurred. The overall RM-ANOVA comprised three within-effects, nested as
follows: “DIET(CONC(TESTDAY))” where TESTDAY (the difference between the two days
of each preference test) was nested within CONC (the difference between the two
concentrations), which was nested within DIET (the difference between the five (broad
scale) or three (fine scale) different concentration pairs). Following the overall RMANOVA, consumption values were tested separately for each concentration pair by separate
RM-ANOVA. The structure of these RM-ANOVAs therefore comprised two within-effects
nested as follows: “CONC(TESTDAY)”. Differences between sugar types were compared by
separate
RM-ANOVA
comprising
two
within-effects
nested
as
follows:
“SUGAR(TESTDAY)” where SUGAR is the difference between sucrose and the hexose
mixture. Post-hoc analyses for each RM-ANOVA were conducted using Tukey’s Honest
38
Significant Difference (HSD) test for equal sample sizes. Data are presented as means ± 1
SE.
For broad scale concentration preferences, preference indices were further analysed to
compare concentration preferences for each diet to a value of 0.5 (no preference) by
Wilcoxon’s test for non-parametric data (Sokal & Rohlf 1995).
Results
Broad Scale Concentration Preferences
The overall RM-ANOVA revealed that there were significant differences between
concentration pairs (diets) and the concentrations of each concentration pair for the hexose
mixtures (Table 2). There was a highly significant interaction between diet and
concentration for both sugar types. The interaction between concentration and test day was
significant for sucrose, which suggests that birds varied their feeding preferences between
the two test days. Post-hoc analysis (Tukey’s HSD) revealed that these significant effects
were due to the mean total consumption of each concentration pair increasing as
concentrations increased, and to variation in consumption between concentration pairs and
test days.
Based on the above differences, data were then separated according to concentration
pairs, and consumption patterns for both sugar types were analysed by separate RMANOVA. These results are summarized for both sugar types in Table 3 and the values for
sugar consumption are illustrated in Figure 1.
39
Table 2: Results of the overall RM-ANOVA for sugar consumption (g) of sucrose and the
equicaloric hexose mixture (glucose and fructose) by Cinnyris talatala. The RM-ANOVA
comprised three within-effects structured as follows: “DIET(CONC(TESTDAY))”. Significant
effects after Tukey’s HSD are indicated by asterisks (*P<0.05, ***P<0.001); n = 10 in all cases.
Sucrose
Glucose: Fructose
Effect
d.f.
F
P
F
P
DIET
4
0.924
0.461
7.688
0.000***
Error (DIET)
36
CONC
1
3.697
0.086
6.327
0.033*
Error (CONC)
9
TESTDAY
1
0.125
0.732
0.586
0.463
Error (TESTDAY)
9
DIET*CONC
4
14.435
0.000***
7.158
0.000***
Error (DIET*CONC)
36
DIET*TESTDAY
4
0.483
0.748
1.821
0.146
Error (DIET*TESTDAY)
36
CONC*TESTDAY
1
7.827
0.021*
0.327
0.581
Error (CONC*TESTDAY)
9
DIET*CONC*TESTDAY
4
0.613
0.656
0.633
0.642
Error (DIET*CONC*TESTDAY)
36
Table 3: Difference in sugar consumption (g) between concentrations (CONC), days
(TESTDAY), and their interaction (CONC x TESTDAY) for sucrose and the hexose mixture (1:1
equicaloric glucose and fructose) for Cinnyris talatala. Values are F-values from RM-ANOVA
(within-effects: CONC(TESTDAY)), with significant effects after Tukey’s HSD indicated by
asterisks (*P<0.05, **P<0.01, ***P<0.001); n = 10 and d.f. = 1 in all cases.
Sucrose
Glucose: Fructose
Concentration
Pairs (M)
CONC
TESTDAY
CONC x
TESTDAY
CONC
TESTDAY
CONC x
TESTDAY
0.25 vs. 0.5
73.21***
0.67
4.98
36.79***
1.86
0.79
0.5 vs. 1
10.44*
0.54
2.38
13.5**
3.52
1.25
1 vs. 1.5
1.846
0.069
0.008
0.185
0.211
0.487
1.5 vs. 2
5.187*
0.011
0.558
0.134
0.038
0.032
2 vs. 2.5
1.859
0.878
0.718
2.302
2.998
0.183
40
a
Low Concentration
High Concentration
1.6
***
1.2
*
*
0.8
0.4
Sugar Consumption (g ± SE)
0.0
0.25 vs. 0.5
0.5 vs. 1
1 vs. 1.5
1.5 vs. 2
2 vs. 2.5
1.5 vs. 2
2 vs. 2.5
Concentration Pairs (M)
b
1.6
***
**
1.2
0.8
0.4
0.0
0.25 vs. 0.5
0.5 vs. 1
1 vs. 1.5
Concentration (M)
Figure 1: Sugar consumption (averaged over 2 days) by Cinnyris talatala of sucrose (a) and
hexose (equicaloric 1:1 glucose and fructose) (b) solutions during paired preference tests (n =
10). Asterisks (*) indicate significant differences in consumption between solutions of a pair (*
P<0.05; ** P<0.01; *** P<0.001).
41
There is a significant difference between concentrations for the two lowest concentration
pairs in both sugar types, because more sugar was consumed at the higher concentration,
and also for 1.5 vs. 2 M sucrose, where more sugar was consumed at the lower
concentration. Post-hoc analysis (Tukey’s HSD) showed that neither consumption between
the different test days (TESTDAY, P>0.05 for all cases), nor the interaction between
concentrations and test days (CONC x TESTDAY, P>0.05 for all cases) was significant.
Preference indices are shown in Figure 2. Wilcoxon’s test was used to test the median of
the preference index per concentration pair against 0.5 (indicating no preference). On both
sucrose and hexose the sunbirds showed a strong preference for the higher concentration in
the lowest concentration pairs (Z9 = 2.80, median = 0.925, P<0.01, 0.25 vs. 0.5 M sucrose,
and Z9 = 2.29, median = 0.708, P<0.05, 0.5 vs. 1 M sucrose; Z9 = 2.80, median = 9.59,
P<0.01, 0.25 vs. 0.5 M hexose, and Z9 = 2.59, median = 0.765, P<0.01, 0.5 vs. 1 M hexose).
This result supports the findings of the RM-ANOVA. However, the preference for 1.5 vs. 2
M sucrose was not significant (Z9 = 1.78, median = 0.359, P=0.07). This is due to the
sensitivity of the Wilcoxon test, which is more powerful than the RM-ANOVA (Sokal &
Rohlf 1995).
An additional RM-ANOVA and post-hoc analysis was performed to compare total
consumption between sugars (see Table 4). There was a significant difference between
sugar types for 1 vs. 1.5 M (F1,9 = 9.045, P<0.05), 1.5 vs. 2 M (F1,9 = 10.11, P<0.05) and 2
vs. 2.5 M (F1,9 = 25.72, P<0.001).
42
Mean Concentration Preference (preference index ± SE)
1.00
Sucrose
Glucose:Fructose
High
0.50
Low
0.00
0.25 vs 0.5
0.5 vs 1
1 vs 1.5
1.5 vs 2
2 vs 2.5
S
G:F
Concentration Pairs (M)
Figure 2: Mean concentration preference (preference index ± SE, n = 10, averaged over 2 days
for each bird) for solutions of sucrose and hexose (equicaloric 1:1 glucose and fructose) by
Cinnyris talatala. The preference index is defined as the grams of sugar consumed at the higher
concentration divided by the total grams of sugar consumed. Bold lines indicate significant
differences in consumption between concentration pairs (sucrose (S) and hexose (G:F);
Wilcoxon’s test, P<0.05).
43
Table 4: Mean sugar (g ± SE) intake on all paired diets for Cinnyris talatala (defined as the sum
of sugar consumed for both feeders of paired diets, n = 10). See Figure 1. Significant differences
between sugar types from RM-ANOVA (within-effects: SUGAR(TESTDAY)), after Tukey’s
HSD, are indicated by asterisks (*P<0.05, **P<0.01) as for Table 3; n = 10 and d.f. = 1 in all
cases.
Sugar (g ± SE)
Concentration Pairs (M)
Sucrose
Glucose:Fructose
0.25 vs. 0.5
1.42 ± 0.08
1.55 ± 0.08
0.5 vs. 1
1.49 ± 0.10
1.79 ± 0.14
1 vs. 1.5
1.49 ± 0.10
*
1.77 ± 0.15
1.5 vs. 2
1.56 ± 0.10
*
1.93 ± 0.13
2 vs. 2.5
1.53 ± 0.13
**
2.10 ± 0.18
Fine Scale Concentration Preferences
The analysis of this experiment followed the same procedure as for the broad scale
concentration preferences. The overall RM-ANOVA for sugar consumption of both sugar
types revealed that there were significant differences between concentration pairs for both
sucrose and hexose (Table 5). There was also a significant difference between test days for
hexoses. Some interactions between effects were also significant for both sugar types. After
post-hoc analysis (Tukey’s HSD), these significant effects were due to the mean
consumption of each concentration pair increasing as concentrations increased.
44
Table 5: Results of the overall RM-ANOVA for sugar consumption (g) of sucrose and the
equicaloric hexose mixture (glucose and fructose) by Cinnyris talatala. The RM-ANOVA
comprised three within-effects structured as follows: “DIET(CONC(TESTDAY))”. Significant
effects after Tukey’s HSD are indicated by asterisks (*P<0.05, **P<0.01, ***P<0.001); n = 10
in all cases.
Sucrose
Hexose
Effect
d.f.
F
P
F
P
DIET
2
4.134
0.033*
60.626
0.000***
Error (DIET)
18
CONC
1
57.462
0.000***
25.636
0.000***
Error (CONC)
9
TESTDAY
1
0.215
0.654
17.474
0.002**
Error (TESTDAY)
9
DIET*CONC
2
31.853
0.000***
8.712
0.002**
Error (DIET*CONC)
18
DIET*TESTDAY
2
5.954
0.010*
7.902
0.003**
Error (DIET*TESTDAY)
18
CONC*TESTDAY
1
0.007
0.935
4.718
0.058
Error (CONC*TESTDAY)
9
DIET*CONC*TESTDAY
2
0.519
0.604
2.068
0.155
Error (DIET*CONC*TESTDAY)
18
Based on the above mentioned differences, data was separated as for broad scale
preferences for further analysis. These results are summarized in Table 6 and shown in
Figure 3. Concentrations were significantly different for all experimental diets of sucrose
and hexose except for the mid-range diets of both sugars. The difference in concentration
for 0.45 vs. 0.5 M hexose was almost significant, but this is because three birds showed a
slight preference for the lower concentration; differences between the concentrations for all
other individual birds were small.
45
Table 6: Differences in sugar consumption (g) between concentrations (CONC), test days
(TESTDAY), and their interaction (CONC x TESTDAY) for sucrose and the hexose mixture (1:1
equicaloric glucose and fructose) for mature white-bellied sunbirds, Cinnyris talatala. Values
are F-values from RM-ANOVA (within effects: CONC(TESTDAY)), with significant effects
after Tukey’s HSD indicated by asterisks (*P<0.05, **P<0.01, ***P<0.001); n = 10 and d.f. = 1
in all cases.
Sucrose
Concentration
Pairs (M)
0.27 vs. 0.3
Glucose: Fructose
CONC
TESTDAY
6.104*
0.006
CONC x
TESTDAY
0.414
CONC
TESTDAY
-
-
CONC x
TESTDAY
-
0.47 vs. 0.5
0.786
1.142
0.114
-
-
-
0.97 vs. 1
7.61*
31.64***
0.39
-
-
-
0.25 vs. 0.3
-
-
-
175.7***
1.6
1.5
0.45 vs. 0.5
-
-
-
3.064
0.007
0.016
0.95 vs. 1
-
-
-
30.36***
10.89**
1.5
These results reveal that sunbirds could discriminate differences in sucrose solutions of
0.03 M (representing 1% w/w) and differences in equicaloric hexose solutions of 0.05 M
(representing 2% w/w). Post-hoc analysis (Tukey’s HSD) showed that consumption
differences between the two test days were significant for 0.97 vs. 1 M sucrose and 0.95 vs.
1 M hexose. This may be attributable to individual variation and possible side biases in this
experiment. From the raw data, it is evident that these effects were significant due to the fact
that birds (in each case three different birds) consumed different combined amounts
between days (individual variation) or consumed more from one feeder position on both
days, attributable to side bias (Jackson et al. 1998).
46
Low Concentration
a
High Concentration
1.6
1.2
*
*
0.8
Sugar Consumption (g ± SE)
0.4
0.0
0.27 vs. 0.3
0.47 vs. 0.5
0.97 vs. 1
b
1.6
***
***
1.2
0.8
0.4
0.0
0.25 vs. 0.3
0.45 vs. 0.5
0.95 vs. 1
Concentrations(M)
Concentration
(M)
Figure 3: Sugar consumption (g, averaged over 2 days) by Cinnyris talatala of sucrose (a) and
hexose (equicaloric 1:1 glucose and fructose) (b) solutions during paired preference tests (n =
10). Asterisks (*) indicate significant differences in consumption between solutions of a pair
(*P<0.05; ***P<0.001).
47
Discussion
White-bellied sunbirds show concentration preferences within a sugar type at both broad
and fine scales. Contrary to expectations, sunbirds showed similar preferences in
concentration over broad scale differences for both sucrose and equicaloric glucose: fructose
solutions. The general preference found was for concentrations around 1 M for both sugar
types, and this confirms an earlier finding that white-bellied sunbirds can effectively dilute
higher concentrations of artificial nectar to approximately 1.05 M by consuming
supplementary water (Nicolson & Fleming 2003b). This concentration (1 M) is higher than
that of most natural nectars for sunbirds (Nicolson 2002; Johnson & Nicolson 2008), and the
result concurs with previous laboratory findings of other birds preferring concentrations
higher than those common for natural nectars (Stiles 1976; Tamm & Gass 1986; Blem et al.
2000; Martínez del Rio et al. 2001). The reasons for this are unclear. It may be that birds can
more easily maintain energy and water balance when consuming higher concentrations than
those commonly found in floral nectars, through decreased foraging time and reduced water
loading. Based on the results of the present study, and the dilution of concentrated nectars
by consumption of supplementary water (Nicolson & Fleming 2003b), it appears that
sunbirds may better maintain energy and water requirements at a concentration of 1 M
sugar. Bypassing of water absorption in the kidneys occurs with dilute nectars in Palestine
sunbirds, Nectarinia osea (McWhorter et al. 2003). This ability may allow other species of
sunbirds to cope with very dilute nectars, more so than hummingbirds (Nicolson & Fleming
2003b). This water shunting requires rapid absorption of sugars and other solutes from the
ingested nectar (Nicolson 2006). Activities such as foraging and physiological processes
incur energetic costs which may be reduced with nectar of 1 M sugar concentration. On 1 M
sucrose the sunbirds have the lowest feeding frequency, take the longest breaks between
48
feeding events and have the lowest total feeding duration from 9.00 to 14.00 (Chapter 2,
Figures 6, 8 and 9). This implies that nectar of 1 M sucrose, and likely the same for
equicaloric hexose, the birds can meet their energy and water demands while having more
spare time to devote to other activities.
Since sunbirds show perfect compensatory feeding in the range of 0.25 to 2.5 M sucrose
(Nicolson & Fleming 2003b), their energy intake was expected to be fairly constant across
all concentrations tested. White-bellied sunbirds have been previously found to average 2.77
± 0.42 g of daily (13 h light) energy intake over the same range of concentrations (average
body mass 9.27 ± 0.34 g; Nicolson & Fleming 2003b). This is approximately 0.21 g h-1.
Over the same concentration range in this study, the sunbirds (average body mass 8.91 ±
0.25 g) had high energy consumption gaining an average of 1.49 ± 0.10 g of sucrose over 6
h, or 0.25 g h-1. The mean body mass of the two groups of sunbirds was very similar, and
therefore sugar (energy) intakes are comparable. White-bellied sunbirds have a higher
energy intake (Nicolson et al. 2005) and total feeding duration (Köhler et al. 2006) in the
early morning under laboratory conditions. Since the 6 h experimental period was from the
early morning until early afternoon (7.00 to 13.00 h), it appears that the birds can easily gain
the bulk of their energy requirements during the early hours of the day.
Sunbirds consumed greater volumes of the hexose mixture than sucrose solutions at the
highest concentrations (1.5 to 2.5 M), and therefore gained more grams of sugar (energy)
from hexoses. This is difficult to explain, as both sugar types are assimilated with high
efficiency (Lotz & Nicolson 1996; Jackson et al. 1998; McWhorter & Martínez del Rio
2000) and sucrose and equicaloric hexose solutions are equally accepted by sunbirds (Lotz
& Nicolson 1996; Fleming & Nicolson 2004b). Given the high assimilation efficiency,
49
sunbirds likely have intestinal sucrase activity, though this has not been measured to date
(Nicolson & Fleming 2003a). The hydrolysis of sucrose to glucose and fructose may
however still be a rate limiting step for sunbirds, leading to a preference for hexoses as they
can be actively and passively absorbed. Passive uptake of glucose has been shown for
several bird species. For example, northern bobwhite quail show almost complete passive
absorption of L-glucose (92%, Levey & Cipollini 1996). The yellow-rumped warbler
(Dendroica coronata) shows high D-glucose (90%) and L-glucose (91%) passive uptake
efficiency (Afik et al. 1997). House sparrows (Passer domesticus) have been found to have
>70% passive glucose absorption (Chang & Karasov 2004). In nectarivorous rainbow
lorikeets, passive glucose absorption has also been found to be high (80%, Karasov & Cork
1994). Rufous and Anna’s hummingbirds have also been found to have a high passive
permeability to glucose, and rely partially on passive uptake to meet energy demands
(McWhorter et al. 2005). It is possible that passive absorption in sunbirds may be more
heavily relied on to meet energy demands than active intestinal uptake of sugars (see
Martínez del Rio and Karasov 1990; Levey & Martínez del Rio 2001), and in a recent study
it has been shown that white-bellied sunbirds have extensive paracellular (passive) glucose
absorption (Napier et al. 2008). Aside from possible physiological reasons, preingestional
limitations caused by preingestional food processing (interference such as corolla length and
other possible mechanical influences or constraints (Kingsolver & Daniel 1983)) may be
affecting intake rates. Viscosity increases dramatically at the high end of the concentration
range used in this study. From 1.5 to 2.5 M, the viscosity of sucrose solutions is
approximately double that of equicaloric glucose: fructose solutions (Chapter 2, Figure 2),
and this may impose limitations on volumetric ingestion rates (see McWhorter 2005), which
in turn would influence energetic intake. These physiological and preingestional factors may
help explain why the sunbirds ingested more sugar on hexose solutions.
50
At a finer scale of preference, I hypothesized that sunbirds, like hummingbirds and some
other passerines (Levey 1987; Lloyd 1989; Blem et al. 2000; Schaefer et al. 2003) should be
able to discriminate between small differences in sugar concentrations. Sunbirds were found
to discriminate between diets differing by as little as 0.03 M (1% w/w) for sucrose and 0.05
M (2% w/w) for equicaloric hexose mixtures, at both low and high nectar concentrations.
However, they did not show this ability at mid-range nectar concentrations. Lloyd’s (1989)
results for the fine (2%) discrimination between sucrose solutions at low concentrations (8,
10 and 12%) for two African sunbirds (Cinnyris afer and Nectarinia famosa, the southern
double-collared and malachite sunbirds) were based on a small sample size (n = 3). The
results shown here confirm that sunbirds do possess fine discrimination abilities for sucrose
at low concentrations. Sunbirds are specialized nectarivorous passerines, and my results
indicate that they can discriminate between diets as well as both a hummingbird and other
passerines. White-bellied sunbirds respond to high concentrations like hummingbirds, which
can discriminate small differences (as little as 1% w/v) around 20% w/v solutions in choice
tests in the field (Blem et al. 2000). At low concentrations, white-bellied sunbirds respond
like other sunbirds and tanagers, which can discriminate small differences (2%) around 0.25
to 1 M on liquid diets (sunbirds, Lloyd 1989), and 1% differences around 6 and 8% w/w on
solid diets (banana mash with agar and sucrose mix; Levey 1987; mixed ingredients;
Schaefer et al. 2003).
Since sugars and concentrations in nectar vary naturally over wide ranges (see Pyke &
Waser 1981; Johnson & Nicolson 2008), the ability to discriminate among foods plays a
vital role in foraging decisions. It must be said that the results shown here for fine scale
concentration preference are to be interpreted with caution, as there was much individual
variation and side bias. Some birds could discriminate “better” between diets (their selection
51
of one diet over the other was almost exclusive) than others. Other birds simply liked
feeding from one feeder regardless of the diet, and this may be attributable to side bias
(Jackson et al. 1998) which could have developed due to time spent in captivity (see
Appendix). Overall, the results indicate that sunbirds do have the ability to discriminate
between concentrations within a sugar type. This ability, at both a broad and a fine scale,
may facilitate their ability to employ foraging strategies. By selecting concentrations that
will allow them to gain sufficient energy and water, and allow them more free time to
allocate to other behaviours like territory defence and mate choice, they may improve their
fitness.
Acknowledgements Experiments were approved by the Animal Use and Care Committee
of the University of Pretoria, and were funded by the South African National Research
Foundation (NRF). Craig Symes is thanked for mist-netting sunbirds and Jan Cilliers Park
for permission to mist-net birds under permit from the Gauteng Department of Nature
Conservation.
References
Afik D, McWilliams SR, Karasov WH (1997) A test for passive absorption of glucose in
yellow-rumped warblers and its ecological implications. Physiol Zool 70:370-377
Baker HG (1975) Sugar concentrations in nectars from hummingbird flowers. Biotropica
7:37-41
52
Baker HG, Baker I (1982) Chemical constituents of nectar in relation to pollination
mechanisms and phylogeny. In: Nitecki HM (ed) Biochemical aspects of evolutionary
biology. University of Chicago Press
Baker HG, Baker I (1983) Floral nectar sugar constituents in relation to pollinator type. In:
Jones CE, Little RJ (eds) Handbook of experimental pollination biology. Van Nostrand
Reinhold, New York
Blem CR, Blem LB, Felix J, van Gelder J (2000) Rufous hummingbird sucrose preference:
precision of selection varies with concentration. Condor 102:235-238
Chang M, Karasov WH (2004) How the house sparrow Passer domesticus absorbs glucose.
J Exp Biol 207:3109-3121
Fleming PA, Gray DA, Nicolson SW (2004a) Circadian rhythm of water balance and
aldosterone excretion in the white-bellied sunbird Nectarinia talatala. J Comp Physiol B
174:341–346
Fleming PA, Hartman Bakken B, Lotz CN, Nicolson SW (2004b) Concentration and
temperature effects on sugar intake preferences in a sunbird and a hummingbird. Funct Ecol
18:223–232
Fleming PA, Xie S, Napier K, McWhorter TJ, Nicolson SW (2008) Nectar concentration
affects sugar preferences in two Australian honeyeaters and a lorikeet. Funct Ecol
(doi:10.1111/j.1365-2435.2008.01401.x)
53
Hainsworth FR, Wolf LL (1976) Nectar characteristics and food selection by
hummingbirds. Oecologia 25:101–113
Jackson S, Nicolson SW, Lotz CN (1998) Sugar preferences and “side bias” in Cape
Sugarbirds and Lesser Double-Collared Sunbirds. Auk 115:156–165
Johannsmeier MF (ed) (2001) Beekeeping in South Africa. Agricultural Research Council,
pp 152
Johnson SD, Nicolson SW (2008) Evolutionary associations between nectar properties and
specificity in bird pollination systems. Biol Lett 4:49-52
Karasov WH, Cork SJ (1994) Glucose absorption by a nectarivorous bird: the passive
pathway is paramount. Am J Physiol 267:G18-G26
Kingsolver JG, Daniels TL (1983) Mechanical determinants of nectar feeding strategy in
hummingbirds: energetics, tongue morphology, and licking behaviour. Oecologia 60:214226
Köhler A, Verburgt L, Nicolson SW (2006) Short-term energy regulation of whitebellied
sunbirds (Nectarinia talatala): effects of food concentration on feeding frequency and
duration. J Exp Biol 209:2880-2887
Levey DJ (1987) Sugar-tasting ability and fruit selection in tropical fruit-eating birds. Auk
104:173-179
54
Levey DJ, Cipollini ML (1996) Is most glucose absorbed passively in northern bobwhite?
Comp Biochem Physiol 113A:225-231
Levey DJ, Martínez del Rio C (2001) It takes guts (and more) to eat fruit: lessons from
avian nutritional ecology. Auk 118:819-831
Lloyd P (1989) Sucrose concentration preferences of two southern African sunbirds. Ostrich
60:134-135
Lotz CN, Nicolson SW (1996) Sugar preferences of a nectarivorous passerine bird, the
Lesser Double-collared Sunbird (Nectarinia chalybea). Funct Ecol 10:360-365
Lotz CN, Schondube JE (2006) Sugar preferences in nectar- and fruit-eating birds:
behavioural patterns and physiological causes. Biotropica 38:3-15
Martínez del Rio C (1990) Sugar preferences in hummingbirds: The influence of subtle
chemical differences on food choice. Condor 92:1022-1030
Martínez del Rio C, Karasov WH (1990) Digestion strategies in nectar- and fruit-eating
birds and the sugar composition of plant rewards. Am Nat 136:618-637
Martínez 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. Am Zool 41:902-915
55
Mata A, Bosque C (2004) Sugar preferences, absorption efficiency and water influx in a
Neotropical nectarivorous passerine, the bananaquit (Coereba flaveola). Comp Biochem
Physiol A 139:395-404
McWhorter TJ, Martínez del Rio C (2000) Does gut function limit hummingbird food
intake? Physiol Biochem Zool 73:313-324
McWhorter TJ, Martínez del Rio C, Pinshow B (2003) Modulation of ingested water
absorption by Palestine sunbirds: evidence for adaptive regulation. J Exp Biol 206:659-666
McWhorter TJ (2005) Carbohydrate hydrolysis and absorption: lessons from modelling
digestive function. In: Starck JM and Wang T (eds) Physiological and ecological
adaptations to feeding in vertebrates. Science Publishers, Enfield, New Hampshire.
McWhorter TJ, Hartmann Bakken B, Karasov WH, Martínez del Rio (2005) Hummingbirds
rely on both paracellular and carrier-mediated intestinal glucose absorption to fuel high
metabolism. Biol Lett 2:131-134
Napier KR, Purchase C, McWhorter TJ, Nicolson SW, PA Fleming (2008) The sweet life:
diet sugar concentration influences paracellular glucose absorption. Biol Lett (doi:
10.1098/rsbl.2008.0253)
Nicolson SW (1998) The importance of osmosis in nectar secretion and its consumption by
insects. Am Zool 38:418-425
56
Nicolson SW (2002) Pollination by passerine birds: why are the nectars so dilute? Comp
Biochem Physiol B 131:645-652
Nicolson SW, Fleming PA (2003a) Nectar as food for birds: the physiological consequences
of drinking dilute sugar solutions. Plant Syst Evol 238:139-153
Nicolson SW, Fleming PA (2003b) Energy balance in the whitebellied sunbird Nectarinia
talatala: constraints on compensatory feeding, and consumption of supplementary water.
Funct Ecol 17:3-9
Nicolson SW, Hoffmann D, Fleming PA (2005) Short-term energy regulation in nectarfeeding birds: the response of white-bellied sunbirds (Nectarinia talatala) to a midday fast.
Funct Ecol 19:988-994
Nicolson SW (2006) Water management in nectar-feeding birds. Am J Physiol Rgul Integr
Comp Physiol 291:R828-R829
Nicolson SW, Thornburg RW (2007) Nectar chemistry. In: Nicolson SW, Nepi M, Pacini E
(eds) Nectaries and nectar. Springer, Dordrecht
Pyke GH, Waser NM (1981) The production of dilute nectars by hummingbird and
honeyeater flowers. Biotropica 13:260-270
Roberts WM (1996) Hummingbirds’ nectar concentration preferences at low volume: the
importance of time scale. Anim Behav 52:361-370
57
Roces F, York W, von Helversen O (1993) Nectar concentration preference and water
balance in a flower visiting bat, Glossophaga soricina antillarum. In: Barthlott W et al.
(eds) Animal-plant interactions in tropical environments. Bonn: Museum Koenig
Rodríguez-Peña N, Stoner KE, Schondube JE, Ayala-Berdón J, Flores-Ortiz CM Martínez
del Rio C (2007) Effects of sugar composition and concentration on food selection by
Saussure’s long-nosed bat (Leptonycteris curasoae) and the long-tongued bat (Glossophaga
soricina). J Mammal 88:1466-1474
Schaefer HM, Schmidt V, Bairlein F (2003) Discrimination abilities for nutrients: which
difference matters for choosy birds and why? Anim Behav 65:531-541
Schondube JE, Martínez del Rio C (2003) Concentration-dependent preferences in nectarfeeding birds: mechanisms and consequences. Funct Ecol 17:445-453
Sokal RR, Rohlf FJ (1995) Biometry. WH Freeman & Co., NY
Stiles FG (1976) Taste preferences, colour preferences, and flower choice in hummingbirds.
Condor 78:10-26
Tamm S, Gass CL (1986) Energy intake rates and nectar concentration preferences by
hummingbirds. Oecologia 70:20-23
Weast RC (1980) CRC handbook of chemistry and physics, 60th edn. CRC Press, Inc.
Florida USA
58
Chapter 3
The viscosity of artificial nectar: effect on the feeding behaviour
of white-bellied sunbirds, Cinnyris talatala (Nectariniidae)
Carolina Leseigneur
59
ABSTRACT
Nectars of plants pollinated by birds, and by passerines in particular, have been widely
reported to be dilute in terms of their sugar concentration. It has been proposed that the low
viscosity of dilute nectars makes drinking easier for birds. Does the viscosity of nectar
represent a preingestional limitation to nectarivorous birds? How does it affect their feeding
behaviour? The present study represents the first analysis of the effect of viscosity on the
feeding response of a bird, separate from the effects of sugar concentration. A specialized
nectarivorous passerine, the white-bellied sunbird (Cinnyris talatala), was exposed to three
different test series of sucrose solutions (control series, pure sucrose from 0.25 to 2.5 M CS, constant viscosity series, 0.25 to 0.7 M – CVS, and constant concentration series, 1 M CCS). Viscosities were artificially altered with Tylose ®. On dilute sucrose concentrations
with increased viscosity (CVS - all equivalent to the viscosity of a pure 1 M solution), the
sunbirds reduced intake rates and gained less energy compared to pure solutions of the same
concentrations and pure 1 M (CS). On sucrose solutions of 1 M but with increased viscosity
(CCS - equivalent to 1.5, 2 and 2.5 M solutions respectively), the birds consumed solutions
at similar rates to those on pure 1.5, 2 and 2.5 M solutions (CS), but gained less energy as
these intake rates are lower than the intake rate on pure 1 M sucrose. These reduced intake
rates when feeding on more viscous artificial nectars occurred because sunbirds did not alter
their feeding behaviour (feeding frequency, feeding duration, total feeding duration and
feeding interval) in any significant way. These results suggest that sunbirds suffer a preingestional limitation when consuming nectars with viscosities higher than those due to
sugar concentration alone, leading to an energy deficit. On dilute solutions sunbirds appear
to regulate their feeding behaviour according to viscosity, while on the mid-range (1 M)
solutions they regulate feeding behaviour according to concentration.
60
Introduction
The question of why bird nectars, particularly those of passerine-pollinated flowers, are
so dilute has been addressed in many studies (Johnson & Nicolson 2008). Of the many
hypotheses proposed (for a review, see Nicolson 2002; Johnson & Nicolson 2008), the
importance of low viscosity for drinking by birds was first proposed by Baker (1975).
Nectarivorous birds imbibe nectar by capillarity, which is inversely proportional to viscosity
(Heyneman 1983). Capillarity is defined as a physical phenomenon caused by surface
tension that results in the surface of a liquid rising or falling in contact with a solid (Bourne
1982; Vogel 1994). Thus high sugar concentrations could impose constraints on the feeding
behaviour and efficiency of these birds. Feeding in nectarivorous birds, especially
hummingbirds, has been extensively studied, but the effect of viscosity on feeding
behaviour has not been explicitly examined separately from sugar concentration for any bird
species (but see Stromberg & Johnsen 1990).
Dynamic viscosity can be defined as the internal friction of a fluid which resists forces
that cause change in its form (Weast 1980; Bourne 1982; Vogel 1994) and is measured in
mPa.s. Sugar solutions are Newtonian fluids (Bourne 1982). A Newtonian fluid is a fluid
that continues to display fluid properties regardless of the speed at which it is deformed, its
stress versus rate of strain is linear. The viscosity of a Newtonian fluid depends only on
temperature, pressure and its chemical composition (Vogel 1994). The viscosity of sucrose
solutions increases exponentially as a function of concentration while the energy content
increases linearly, and volumetric flow rate (capillarity) decreases with concentration
(Weast 1980; Heyneman 1983; Gass & Roberts 1992; Josens & Farina 2001; Nicolson &
Thornburg 2007). Based on these variables, it is expected that maximum energy intake rates
61
are a compromise between energy density and viscosity. This implies that the most efficient
energy intake by nectarivorous birds (and other pollinators) could fall at intermediate levels
of sugar concentration (Baker 1975; Pyke & Waser 1981). Based on empirical
measurements of sucrose intake rates, this has been shown to be true for many different
pollinators across various taxa, including bird species (Nicolson 2007b).
Ambient temperature and nectar solutes affect the viscosity of nectar (Heyneman 1983;
Bourne 1982; Mathlouthi & Génotelle 1995). Salts are mostly present in nectars as charged
ions such as sodium (Na+), potassium (K+) and chloride (Cl-) (Hiebert and Calder 1983).
Ions can affect the viscosities of solutions by forming new chemical bonds or altering
existing bonds (Bourne 1982; Vogel 1994). Amino acids can also alter the viscosity of
solutions, the effect depending on concentration and the molecular mass of the amino acids
(Bourne 1982; Nicolson 2007a). Low molecular weight amino acids (both D- and Lenantiomers) are always sweet to humans, but spatial barriers (their effective size in water
or solutions) according to apparent specific volumes of amino acids are reported to account
for the exclusion of larger L-amino acids from sweet receptors, hence they are perceived as
bitter (Birch & Kemp 1989).
Sunbirds and hummingbirds imbibe nectar by licking it from flowers, and it flows via
capillarity along grooves on their long tongues (Skead 1967; Schlamowitz et al. 1976;
Kingsolver & Daniel 1983). Heyneman (1983) modelled feeding in hummingbirds and
proposed that for large nectar volumes, a 22 to 26% w/w sucrose concentration would be
optimal as feeding costs are high beyond such concentrations, assuming a steady-state
continuous nectar flow. In contrast, Kingsolver and Daniel (1983) proposed that nectar flow
is not a steady-state if it is induced by capillarity; they state that flow will be discontinuous,
62
and they predicted that optimal concentrations should be 40 to 45% for high volume nectar
pools that require several licks, and 20 to 25% for small volumes that require single licks for
hummingbirds. However, temporal scale must also be considered. Gass and Roberts (1992)
modelled hummingbird visits to flowers based on three temporal scales: the time it takes to
complete tongue loading, a licking cycle (inserting the bill, inserting and withdrawing the
tongue, and withdrawing the bill) and entire visits to flowers. Small volumes of nectar can
be loaded onto the tongue in a single lick via capillarity, and because of fluid flow due to
capillarity they require no unloading phase. The unloading phase is when the tongue is
actively retracted by the bird by pulling the tongue back and up against the upper bill with
the hyoid muscles, causing a negative pressure which sucks the nectar from the tongue
grooves into the back tongue cavity for swallowing (Skead 1967). However for larger
volumes that require several licks, fluid loading and unloading are required and higher
concentrations maximize extraction efficiency in terms of energy content (Gass & Roberts
1992). This provides support for Kingsolver and Daniel’s (1983) model.
Studies on certain insects provide some insight into the effects of viscosity on feeding
behaviour. In the following studies, the effects of viscosity and concentration were
separated by adding Tylose H 10000 YP2 ®, an inert polysaccharide, to sugar solutions. In
honeybees, trophallactic food transfer by donor bees increases with sucrose concentration
over a range of concentrations (10-30%). However, trophallaxis is impeded at higher
concentrations by the effect of viscosity, either because it is not energetically worthwhile for
the donor or because of a direct physical limitation (Tezze & Farina 1999). In a
nectarivorous ant (Camponotus mus), foraging behaviour is affected by body size in
conjunction with viscosity (Medan & Josens 2005). When given sucrose solutions with
artificially increased viscosity, only large workers increased feeding time. When given
63
solutions of constant viscosity with different concentration, large ants were indifferent in
their feeding time, but smaller workers increased their feeding time on the more
concentrated solution. Variations observed between different sized workers in their response
to different characteristics of their food source may be a foraging specialization based on
worker size. Josens and Farina (2001) analysed intake rates in response to artificially
increased viscosity of sucrose solutions in the hovering hawk moth (Macroglossum
stellatarum). Hawk moths were presented with different solutions at single feeders, and
ingestion volumes and feeding times were recorded. The first series consisted of 5 to 60%
sucrose solutions (base series, BS). The second comprised a series of 10 to 50% sucrose
solutions mixed with Tylose ® to keep the viscosity constant at a value corresponding to
that of a 50% sucrose solution (constant viscosity series, CVS). The third series kept
concentration constant at 30% sucrose, but viscosity was increased by adding Tylose ® to
correspond with sucrose solutions up to 60% (constant concentration series, CCS). It was
found that both viscosity and concentration of sucrose in a solution influence the nectar
intake rate.
The only study to date on nectarivorous birds is that of Stromberg & Johnsen (1990)
who examined the independent effects of artificial sweeteners, sucrose solutions and
viscosity (equal to that of sucrose solutions, altered artificially with sodium
carboxymethylcellulose (CMC)) in field experiments on the feeding preferences of blackchinned hummingbirds (Archilochus alexanderi). The authors concluded that a sweet
stimulus, rather than the mechanical effect of high viscosity, is required to induce feeding in
black-chinned hummingbirds. However, these results are tentative as CMC is anionic, and
ions can alter viscosity (Bourne 1982; Vogel 1994). In addition, many artificial sweeteners,
like aspartame, are comprised primarily of amino acids. Aside from the chemistry of amino
64
acids in solutions, it has been found that amino acids in artificial nectar affect the
preferences of both hummingbirds and sunbirds (Hainsworth & Wolf 1976; Leseigneur et
al. 2007).
Considering Baker’s (1975) hypothesis, does viscosity impose a feeding limitation on
nectarivorous birds, and if so, how does it affect their feeding behaviour? Preingestional
limitations caused by preingestional food processing (interference caused by factors such as
corolla length and other mechanical influences or constraints (Kingsolver & Daniel 1983))
may be affecting intake rates. In this study I assessed the effects of viscosity separately from
those of concentration on a specialized nectarivorous passerine, the white-bellied sunbird
(Cinnyris talatala). Sucrose solutions were prepared using molar concentrations, ranging
from 0.25 to 2.5 M: below 0.25 M the birds cannot maintain their energy balance at 20ºC
(Nicolson & Fleming 2003b). The birds were exposed to three different treatment series as
per Josens & Farina (2001) – control series (CS, pure sucrose solutions), constant viscosity
series (CVS, dilute solutions with increased viscosity equivalent to that of 1 M sucrose), and
constant concentration series (CCS, 1 M solutions with variably increased viscosity). CVS
and CCS sucrose solutions were altered with Tylose ® to artificially increase viscosity. To
maintain energy consumption at required levels, it was expected that sunbirds would
consume solutions of low concentration with high viscosities at higher rates by increasing
their feeding frequency, feeding duration per feeding event, total feeding duration and by
decreasing the feeding interval between feeding events. This is based on the fact that whitebellied sunbirds employ compensatory feeding over the range of sucrose concentrations
used in this study (Nicolson & Fleming 2003b), and the assumption that increased viscosity
will limit the amount of food ingested per unit volume. Alternatively, if viscosity is the only
physical limitation, solutions of equal viscosity should be ingested at the same rates.
65
Materials & Methods
Birds and Maintenance
White-bellied sunbirds were captured by mist-netting at Jan Cilliers Park in Pretoria,
South Africa. Six males and five females (8.41 ± 0.43 g), all with full adult plumage, were
captured during April 2007. Birds were maintained in an outdoor aviary (8 x 2 x 5 m), at the
University of Pretoria’s Experimental Farm. They were fed a maintenance diet of 0.6 M
sucrose (20% w/w) mixed with a nutritional supplement (Ensure®, Abbott Laboratories,
Johannesburg, South Africa). This diet and supplementary water were available to the birds
ad libitum in inverted stoppered syringes (feeders), and water baths were also provided.
During the laboratory trials, the birds were maintained individually in cages of 40 x 42 x
28 cm in a climate-controlled room at 20 ± 2 ºC and 45% RH on a 12:12 h light: dark cycle
(lights on at 07.00 h) as in previous studies on white-bellied sunbirds (for example,
Nicolson & Fleming 2003b; Fleming et al. 2004a and b). Temperature and humidity were
maintained by automated Siemens RLU 222 and 210 temperature and humidity regulators
and an indoor air conditioner (Johnson Controls cr 722 CAREL). The birds were allowed to
acclimatize to laboratory conditions for one week prior to experiments. Dawn and dusk
were simulated with 0.5 h of dimmed light at the beginning and the end of the photoperiod.
The same maintenance diet was provided during the acclimatization period. Birds were
released at the end of the experiments.
66
Experiments
Viscosity measurements
The dynamic viscosities (mPa.s) of sugar solutions were determined with an Anton Paar
(GmbH) Rheometer (Model Physica MCR 301, Cup – C-PTD200, Spindle – CC27,
Advanced Laboratory Solutions, Johannesburg), with the RHEOPLUS/32 v.3.00 software
package at 20°C and a shear rate 60.8 s-1 (automatically maintained by the system). The cup
and spindle were chosen based on the nature (Newtonian fluid) and density (low density
solutions) of the sugar solutions. The rheometer was calibrated with pure water as the
standard (1.002 mPa.s at 20°C). Concentrations measured were sucrose and equicaloric 1:1
glucose and fructose mixtures at 0.25, 0.5, 0.7, 1, 1.5, 2, and 2.5 M. Additionally, a 0.7 M
sucrose solution mixed with asparagine, serine and glutamine at 10 mM each, and 0.7 M
sucrose solutions each containing either 10 or 20 mM NaCl were also measured.
The dynamic viscosity of sucrose solutions with added Tylose H 10 000 YP2 ® (SE
Tylose GmbH & Co. KG, Wiesbaden, Germany - hereafter Tylose ®) was also measured.
Tylose ® is a hydroxyethylcellulose ether (a polysaccharide, molecular weight 10 000 Da)
used as a binder or thickener in various substances, including industrial paints and cement,
make-up materials and food substances. It is non-ionic and does not alter the nutritional
value or density of foods. Because it is a polysaccharide, solutions containing it will still be
Newtonian fluids (Bourne 1982; Vogel 1994). Tylose ® amounts added to sucrose solutions
were estimated by extrapolating from Josens & Farina’s (2001) data of kinematic viscosity
of sugar solutions, using a regression equation. Sucrose solutions containing Tylose ® were
then measured with the rheometer. This was necessary to determine whether the added
67
amounts of Tylose ® increased viscosities of sucrose solutions to the desired levels. These
values are tabulated in Table 1.
Table 1: Quantities of Tylose ® (following Josens & Farina 2001) added to sucrose solutions to
alter viscosities for CVS and CCS series. Dynamic viscosities were measured with the
rheometer at 20°C, 60.8 s-1 shear rate.
Series
Sucrose
Concentration
Desired
Viscosity
Tylose ® added
Dynamic Viscosity
(mPa.s)
Pure
with Tylose
Sucrose
®
M
Label
M
mPa.s
% w/w
CVS
0.25
0.25+T1
1
3.23
0.160
1.37
3.21
CVS
0.5
0.5+T1
1
3.23
0.108
1.79
3.23
CVS
0.7
0.7+T1
1
3.23
0.076
2.27
3.25
CCS
1
1+T1.5
1.5
8.08
0.100
3.23
8.08
CCS
1
1+T2
2
70.3
0.185
3.23
70.7
CCS
1
1+T2.5
2.5
162
0.230
3.23
161
Feeding experiments
Each bird was tested with the three different test series to differentiate between the
effects of viscosity and concentration on feeding behaviour. Sunbirds were exposed to the
test series in individual cages fitted with infra-red photo-detection LED systems on the
feeders for automated recordings of feeding frequency, feeding duration, total feeding
duration and feeding intervals (Figure 1). Sunbirds were trained to feed from these feeders
during the acclimation period prior to the experiments. The birds took approximately 3 h to
learn to feed from the novel food source after being initially fed by hand, except for one
male that took 3 days to learn.
68
Feeder and
detection
setup
Bird cage
Computer
Feeding
perch
Feeder
Hole
Automated photo-detection LED system
Infrared light
source
Photo-detector
Figure 1: Schematic representation of the experimental setup and infra-red photo-detection LED
system for automated recordings. Feeding frequency, feeding duration, feeding interval and total
feeding duration by white-bellied sunbirds (Cinnyris talatala) were simultaneously recorded
using custom software (L Verburgt) during test series.
The control series (CS) consisted of pure sucrose solutions of 0.25, 0.5, 0.7, 1, 1.5, 2 and
2.5 M. Feeding responses were determined by presenting single solutions in random order to
all sunbirds (n = 11). This was conducted from 9.00 to 14.00 each day as white-bellied
sunbirds in captivity have longer feeding durations (in s h-1) and lower feeding frequencies
in the early morning (Köhler et al. 2006), and show the most constant food intake rate
during the first half of the day (Fleming et al. 2004). The maintenance diet was provided at
the end of the experimental period each day so that the birds could replace any mass lost as
69
a result of feeding on sugar-only solutions (Nicolson et al. 2005). The birds were given a
rest day between test series and were also provided with the maintenance diet and water.
In the second series (constant viscosity series, CVS), the birds underwent the same
procedure but were presented with dilute sucrose solutions with the viscosity of each
solution held equivalent to that of a 1 M sucrose solution (Table 1). In the third series
(constant concentration series, CCS), the birds again followed the same procedure but were
presented with 1 M sucrose solutions with viscosities altered to be equivalent to that of 1.5,
2 and 2.5 M sucrose solutions respectively (Table1). A 1 M control solution was included in
both the CVS and the CCS.
The feeder mass was measured at the start and end of the experimental period to
determine food intake and thereby sugar (energy) consumption (± 0.01g, Mettler Toledo
PB-602S, Microsep Ltd., Johannesburg). Containers with liquid paraffin were placed
beneath feeders to collect any potential drips. These were also weighed together with the
feeders, and corrections to the consumption data were made where necessary. For all test
series, food consumption in grams was converted to millilitres based on the density of sugar
solutions (Weast 1980), and grams of sugar (energy) consumed per solution was calculated
(g ± SE). Hexoses and sucrose have the same energy value per unit gram, 16.48 x 103 kJ/g
(Nicolson & Fleming 2003b). Feeding frequency (FF), feeding duration (FD), feeding
interval (FI) and total feeding duration (Tot FD) was automatically recorded for all birds
simultaneously on all solutions (resolution = 1 ms). The sugar (energy) intake rate (hereafter
intake rate) was then calculated and averaged (μmol s-1 ± SE) by dividing sugar (energy)
consumption (converted to μmol sugar consumed) by the total feeding duration (Tot FD)
recorded.
70
Statistical analysis
After normality tests, all data were log-transformed for analyses. For sugar (energy)
consumption (g), intake rates (μmol s-1) and feeding behaviour data (FF, FD, FI and Tot
FD), separate overall repeated-measures ANOVA (RM-ANOVA) were performed for CS,
CS vs. CVS and CS vs. CCS. The RM-ANOVAs comparing the test series comprised two
within-effects, nested as follows: “SERIES(CONC)” where CONC (the different
concentrations) was nested within SERIES (the test series). Post-hoc analyses for each RMANOVA were conducted using Tukey’s Honest Significant Difference (HSD) test for equal
sample sizes, followed by sequential Bonferroni corrections for multiple comparisons (Rice
1989). Repeatability ± SE was also calculated for the 1 M solutions of all three test series.
Results
Viscosity measurements
Sugar solutions from 0.25 to 2.5 M showed the expected exponential increase in
dynamic viscosity with increasing concentration (Figure 2). Viscosity is similar for the
sucrose and equicaloric hexose solutions up to 1 M. For higher concentrations, the viscosity
of sucrose solutions is approximately twice that of equicaloric hexose solutions. The slight
difference in viscosity at 1 M sucrose (3.23 mPa.s) and hexose (3.63 mPa.s) was unexpected
and can be attributed to the natural range in viscosity of the fluid when measured (L
Moitsheki pers. comm.).
71
170
160
150
140
130
120
Viscosity (mPa.s)
110
100
90
80
70
60
50
40
30
20
10
0
0.25
0.5
0.7
1
1.5
2
2.5
Sucrose
1.37
1.79
2.27
3.23
8.08
70.3
162
Glucose:Fructose
1.42
1.77
2.07
3.63
3.91
34.5
71.4
Concentration (M)
Figure 2: Dynamic viscosity (data table, mPa.s) of sucrose and equicaloric hexose (glucose and
fructose, 1:1) solutions at 0.25, 0.5, 0.7, 1, 1.5, 2 and 2.5 M. All solutions were measured at
20°C and 60.8 s-1 shear rate with an Anton Paar (GmbH) rheometer.
For 0.7 M sucrose solutions containing salt (NaCl at 10 and 20 mM) or an amino acid
mixture (asparagine, serine and glutamine, each added at 10 mM), the difference in
viscosity was small compared to a pure 0.7 M sucrose solution (Figure 3). Viscosity was
increased most by the inclusion of 20 mM NaCl.
72
0.7 M S
NaCl 10 mM
NaCl 20 mM
3 Amino Acids ea. 10mM
2.93
3
2.45
Viscosity (mPa.s)
2.27
2.37
2
1
0
0.7 M
0.7 M+NaCl
0.7 M+NaCl
Solutions
10 mM
20 mM
0.7 M+ASN,
GLN, SER
10 mM each
Figure 3: Dynamic viscosity (mPa.s) of 0.7 M sucrose solutions: pure sucrose; with added salt
(NaCl) at 10 and 20 mM respectively; with a combination of three amino acids – asparagine
(ASN), glutamine (GLN) and serine (SER) - each added at 10 mM. All solutions were measured
at 20°C and 60.8 s-1 shear rate with an Anton Paar (GmbH) rheometer.
Energy consumption
There were no differences in energy consumption between the CS concentrations except
for 0.25 M which was significantly different to all the other concentrations (not shown, but
see Figure 4; F6,60 = 11.4, P<0.01). The general trend in consumption across concentrations
from 0.5 to 2.5 M is attributable to compensatory feeding (Nicolson & Fleming 2003b).
Energy consumption on CS solutions was significantly different to the CVS solutions
(F3,30 = 13.233, P<0.01, Figure 4). The sunbirds ingested less sugar (gained less energy) on
all CVS solutions with artificially increased viscosities than on the pure CS solutions of the
same concentrations. For example, on the 0.25 M CS solution, the birds ingested 1.16 ± 0.06
g sugar over the experimental period (5 hours), while on the 0.25+T1 CCS solution the birds
73
ingested on average 0.47 ± 0.04 g sugar. Energy consumption of all CVS solutions was also
significantly less than the 1 M CVS control solution.
CVS
Sugar Consumption
Average
Consumption (g +/- SE)
1.5
CS
CCS
***
***
*
***
***
‡
1.0
†
***
‡
‡
†
†
0.5
0.0
0.25+T1 │ 0.25
0.5+T1 │ 0.5
0.7+T1 │ 0.7
1
1.5 │ 1+T1.5
2 │ 1+T2
2.5 │ 1+T2.5
Concentration
(M)
Concentration
(M)
Figure 4: Average sugar consumption (g ± SE) over 5 hours by Cinnyris talatala (n = 11) for
three different test series: pure sucrose solutions (CS, 0.25 – 2.5 M), dilute sucrose solutions
with artificially increased viscosity (CVS, concentration+T1; viscosity increased with Tylose ®
to that of 1 M sucrose), and 1 M sucrose solutions with increasing viscosity (CCS, 1+T1.5, 2 or
2.5 - viscosities equivalent to the stated molar concentrations respectively). After RM-ANOVA
and post-hoc analyses, significant differences between test series are indicated by asterisks
(*P<0.05, ***P<0.001). All CVS (except the CVS 1 M control) are significantly different to 1
M CS (†P<0.01), and all CCS are significantly different to the CCS 1 M control (‡P<0.001).
At higher concentration and viscosities, energy consumption on all of the CCS solutions
was significantly less than on the 1 M CCS control solution (F3,30 = 22.525, P<0.001, Figure
4). Birds ingested less sugar (energy) on the CCS solutions. Sugar consumption of the CCS
solutions was also significantly less than that of the 1.5, 2 and 2.5 M CS solutions
respectively (P<0.01). Energy consumption on the 1 M solutions of all three test series did
not differ (F2,20 = 0.560, P=0.579) and was highly repeatable (0.66 ± 0.15 SE).
74
Energy intake rates
Significant differences were found between the intake rates of the different CS
concentrations (not shown but see Figure 5, F
6,60
= 12.599, P<0.001). The intake rates of
the 0.25 and 2.5 M CS solutions were significantly different to the intake rates of 0.5 to 1.5
M and of 0.5 to 2 M CS solutions respectively (P<0.001 and P<0.05 in all cases
respectively). From 0.5 to 2 M CS, the intake rates differed, increasing with concentration
up to 1 M then decreasing with concentration up to 2.5 M.
Sugar Intake Rate
Rate(μmol s-1 +/- SE)
Average Consumption
CVS
CS
CCS
16
* **
14
* **
12
10
††
8
*
6
†
†
‡
†
‡
‡
†
‡
‡
‡
**
4
2
0
0.25+T1 │ 0.25
0.5+T1 │ 0.5
0.7+T1 │ 0.7
1.5 │ 1+T1.5
1
2 │ 1+T2
2.5 │ 1+T2.5
Concentraion (M)
Figure 5: Average sugar intake rates (μmol s-1 ± SE) over 5 hours by Cinnyris talatala (n = 11)
for three different test series (for description of series, see Figure 4). The intake rates were
significantly different between CS and CVS (**P<0.01), and all CVS intake rates were
significantly different to the 1 M control intake rate (†P<0.01). The intake rates of CCS
solutions were only significantly different to the 1 M CCS control rate (‡P<0.01).
Between test series there were marked differences in intake rates at concentrations
below 1 M. All the CVS solutions were consumed at significantly lower intake rates than
the CS solutions of the same concentrations (F
3,30
= 4.824, P<0.01 in all cases), and were
also consumed at significantly lower rates than the 1 M CVS control solution (P<0.01).
75
Intake rates of the CVS solutions followed an increasing trend similar to that of the CS
solutions, i.e. intake rate increased with increasing concentration up to 1 M.
At higher concentrations and viscosities, the intake rates of all the CCS solutions were
significantly lower than the 1 M CCS control solution (F
3,30
= 21.025, P<0.01, Figure 5).
Interestingly, the intake rates of 1+T2 and 1+T2.5 CCS solutions were not different to pure
2 and 2.5 M CS solutions respectively. There was no significant difference between the 1 M
solutions of the CS, CCS control and the CVS control (F2,20 = 1.167, P=0.332), which again
indicates that feeding by the birds remained regular between test series. However, intake
rate was not highly repeatable (0.29 ± 0.21 SE), and this may be attributed to individual
variation between birds.
Feeding behaviour
The results for the automatically recorded data revealed very few differences in feeding
behaviour for the different test series (Table 2, Figures 6 - 9). The results are discussed
separately for each feeding behaviour measured.
Feeding frequency (FF)
There were differences between the feeding frequencies of the CS solutions (not shown,
but see Figure 6): the feeding frequency on 0.25 M was significantly higher than on 1 and
1.5 M, the 2 M feeding frequency was higher than on 1 M, and the 2.5 M feeding frequency
was higher than on 0.5, 0.7, 1 and 1.5 M (F 6,60 = 17.159, P<0.01 in all cases). The general
trend for the CS solutions was a decrease in feeding frequency with increasing concentration
up to 1 M, followed by an increase with increasing concentration up to 2.5 M. The high
76
feeding frequency measured on pure 0.25 M CS is attributable to the need to ingest large
volumes.
CVS
CS
CCS
**
FF (no. feeds over Tot FD +/- SE)
1200
1000
800
**
600
400
200
0
0.25+T1 │ 0.25
0.5+T1 │ 0.5
0.7+T1 │ 0.7
1
1.5 │ 1+T1.5
2 │ 1+T2
2.5 │ 1+T2.5
Concentration (M)
Figure 6: Feeding frequency (FF) on sucrose solutions by Cinnyris talatala (n = 11) over a 5
hour experimental period on three different test series (for description of series see Figure 4).
Significant differences after RM-ANOVA and post-hoc analyses are indicated by asterisks
(**P<0.01).
Feeding frequency was highest for the two highest concentrations, 2 and 2.5 M CS. This
higher feeding frequency however does not necessarily mean that the birds ate more at these
concentrations, as is demonstrated by the sugar (energy) consumed and sugar intake rate
(Figures 4 and 5).
77
Table 2: Comparison of feeding behaviours of Cinnyris talatala (n = 11) feeding on various
diets of pure sucrose solutions (CS) and sucrose solutions with added Tylose ® (T) (CVS and
CCS – viscosities altered to be equivalent to that of the concentration shown, “Tconc.”). FF =
feeding frequency, FD = feeding duration per feeding event, FI = feeding interval per feeding
event, Tot FD = total feeding duration. Significant differences after RM-ANOVA and post-hoc
analyses are indicated by asterisks (*P<0.05, **P<0.01, ns = not significant).
Series &
Concentration
(M)
Mean FD (s)
FF
CS 0.25
557.4 ± 104.2
CVS 0.25+T1
465.0 ± 119.7
CVS 1 control
308.8 ± 51.9
CS 0.5
382.4 ± 61.6
CVS 0.5+T1
348.4 ± 64.2
CVS 1 control
308.8 ± 51.9
CS 0.7
364.9 ± 78.3
CVS 0.7+T1
353.4 ± 66.0
CVS 1 control
308.8 ± 51.9
CVS 1 control
308.8 ± 51.9
CS 1
304.2 ± 46.9
CCS 1 control
242.6 ± 44.0
CS 1.5
334.3 ± 63.2
CCS 1+T1.5
309.6 ± 63.1
CCS 1 control
242.6 ± 44.0
CS 2
535.7 ± 97.5
CCS 1+T2
270.5 ± 45.4
CCS 1 control
242.6 ± 44.0
ns
1.05 ± 0.20
ns
1.19 ± 0.36
ns
ns
1.21 ± 0.51
1.07 ± 0.21
ns
1.21 ± 0.51
1.08 ± 0.24
ns
ns
ns
1.21 ± 0.51
1.21 ± 0.51
ns
ns
ns
1.55 ± 0.91
1.28 ± 0.46
ns
ns
**
1.55 ± 0.91
1.29 ± 0.38
ns
ns
1.55 ± 0.91
ns
77.1 ± 50.4
66.0 ± 35.4
77.1 ± 50.4
77.1 ± 50.4
93.9 ± 54.4
68.6 ± 42.3
ns
ns
93.9 ± 54.4
41.3 ± 35.2
ns
ns
93.9 ± 54.4
288.7 ± 42.6
319.0 ± 20.2
ns
ns
288.7 ± 42.6
291.0 ± 18.6
ns
ns
321.1 ± 21.3
ns
ns
288.7 ± 42.6
288.7 ± 42.6
ns
ns
261.7 ± 15.3
ns
ns
249.8 ± 15.0
321.7 ± 41.0
ns
ns
317.3 ± 21.2
ns
**
249.8 ± 15.0
483.2 ± 51.6
ns
ns
324.7 ± 24.8
84.7 ± 51.6
ns
N
s
298.7 ± 25.5
80.1 ± 48.1
1.55 ± 0.49
ns
59.3 ± 34.7
ns
69.9 ± 39.7
1.44 ± 0.35
ns
77.1 ± 50.4
456.9 ± 28.2
427.8 ± 48.8
65.4 ± 41.5
1.09 ± 0.24
ns
ns
71.3 ± 42.1
1.18 ± 0.34
ns
41.5 ± 25.6
Tot FD (s)
68.2 ± 39.5
1.16 ± 0.31
ns
Mean FI (s)
ns
249.8 ± 15.0
78
ns
Table 2 continued.
Series &
Concentration
(M)
FF
CS 2.5
894.1 ± 0.18
CCS 1+T2.5
288.4 ± 0.49
CCS 1 control
242.6 ± 44.0
Mean FD (s)
**
1.02 ± 0.36
**
1.98 ± 0.78
ns
1.55 ± 0.91
Mean FI (s)
27.1 ± 32.8
Tot FD (s)
**
82.8 ± 52.1
ns
93.9 ± 54.4
587.9 ± 11.8
ns
357.5 ± 38.1
ns
249.8 ± 15.0
When comparing test series, I expected to see significant differences between the
consumption of pure (CS) and increased viscosity (CVS) solutions. There were in fact few
differences (Table 2, Figure 6). Feeding frequency remained similar to the 1 M CVS control
solution across all concentrations of the CVS solutions. The small decrease in feeding
frequency for the CVS solutions compared to the CS solutions of the same concentrations
suggests a weak negative effect of viscosity. However, the similarity between the feeding
frequencies of the CVS solutions suggests that birds were treating all solutions as being a
pure 1 M sucrose concentration. For comparisons between the CS and CCS solutions I had
the same expectations as for CS versus CVS solutions. Feeding frequency was significantly
lower for the 1+T2 and 1+T2.5 CCS solutions compared to that for pure 2 and 2.5 M CS
solutions respectively, and this is contrary to expectations. Feeding frequency was similar
for both 1+T2 and 1+T2.5 CCS solutions to the 1 M CCS control solution. The feeding
frequency of the 1+T1.5 CCS solution was also similar to that of the 1 M CCS control
solution. The fact that feeding frequency was similar for all the CCS solutions implies that
birds were responding to CCS solutions as if they were pure 1 M sucrose solutions,
regardless of viscosity. For the feeding frequencies on 1 M CS, CVS and CCS solutions,
repeatability was very high (0.87 ± 0.06 SE). But there were significant differences between
the series (F2,20 = 7.788, P<0.01). Post-hoc analysis showed that this was due to the feeding
frequency of the 1 M CCS solution being significantly lower to those on both the 1 M CS
and CVS solutions (P<0.05 both cases).
79
ns
Feeding duration (FD)
There were no significant differences between any of the CS concentrations for mean
feeding duration per feeding event (not shown, but see Figure 7). Feeding duration was
similar for all concentrations between 0.25 and 2.5 M CS.
CVS
CS
CCS
3.0
**
Mean FD (s +/- SE)
2.5
2.0
1.5
1.0
0.5
0.0
0.25+T1 │ 0.25
0.5+T1 │ 0.5
0.7+T1 │ 0.7
1
1.5 │ 1+T1.5
2 │ 1+T2
2.5 │ 1+T2.5
Concentration (M)
Figure 7: Mean feeding duration (FD) per feeding event on sucrose solutions by Cinnyris
talatala (n = 11), over a 5 hour experimental period on three different test series (for description
of series see Figure 4). Significant differences after RM-ANOVA and post-hoc analyses are
indicated by asterisks (**P<0.01).
Between test series (Table 2, Figure 7), the birds had higher feeding durations on the
CVS solutions compared to the CS solutions of the same concentrations. All the CCS
solutions also showed higher mean feeding durations compared to the CS solutions, but this
was significant only for the 2.5 M CS solution compared to the 1+T2.5 CCS solution. All
the CCS solutions were similar to the pure 1 M CCS control solution, suggesting a response
to concentration alone. For the birds’ feeding durations on 1 M CS, CVS and CCS solutions,
repeatability was also very high (0.77 ± 0.11 SE), though again there were significant
differences across the series for 1 M solutions (F2,20 = 7.495, P<0.01). Post-hoc analysis
80
again showed that feeding duration on 1 M CCS was significantly higher than on both 1 M
CS and CVS solutions (P<0.05 both cases).
Feeding interval (FI)
The feeding interval is inversely related to feeding frequency. For the CS solutions,
there was an increase in mean feeding interval up to 1 M, and thereafter a decrease with
increasing concentration, opposite to the pattern of feeding frequency. Significant
differences in mean feeding intervals were the same as for feeding frequency (F
6,60
=
17.173, P<0.01 in all cases, not shown but see Figure 8).
CVS
CS
CCS
120
**
**
Mean FI (s +/- SE)
100
80
60
40
20
0
0.25+T1 │ 0.25
0.5+T1 │ 0.5
0.7+T1 │ 0.7
1
1.5 │ 1+T1.5
2 │ 1+T2
2.5 │ 1+T2.5
Concentration (M)
Figure 8: Mean feeding interval (FI) between feeding events of sucrose solutions by Cinnyris
talatala (n = 11) over a 5 hour experimental period on three different test series (for description
of series see Figure 4). Significant differences after RM-ANOVA and post-hoc analyses are
indicated by asterisks (**P<0.01).
When feeding on 0.25 M CS, birds had a lower (i.e. shorter) mean feeding interval than
most other CS solutions. This is because the birds have to feed more often on this dilute
solution to meet energy demands. Fewer feeding intervals were also recorded for 2 and 2.5
81
M CS solutions. This can be linked to the birds’ high feeding frequencies at these high
concentrations, implying that the birds had to feed more often, and further suggests that they
may not necessarily have been getting food on every feeding attempt, as a result of high
viscosity.
Comparisons between test series revealed small differences in mean feeding intervals
(Table 2, Figure 8). There were no significant differences in mean feeding interval between
the CS and CVS solutions, though feeding intervals of the CVS solutions were longer,
especially for the 0.25+T1 CVS solution. The increase in feeding intervals of the CVS
solutions overall may be explained by the response of the birds to CVS solutions as if they
were pure 1 M sucrose solutions. Mean feeding interval of the CVS solutions were also all
similar to the 1 M CVS control solution. Mean feeding interval for the CCS solutions was
longer than for the CS solutions. This was significant for 1+T2 and 1+T2.5 CCS solutions
compared to the 2 and 2.5 M CS solutions respectively (F
3,30
= 20.730, P<0.01 in both
cases). Feeding intervals of the CCS solutions were very similar, indicating that birds may
have responded to the CCS solutions as pure 1 M sucrose solutions (i.e. a weak response to
viscosity). Mean feeding intervals of 1 M CS, CVS and CCS solutions had very low
repeatability (0.22 ± 0.21 SE). The feeding intervals on 1 M CS, CVS and CCS solutions
were significantly different (F=2,20 = 6.956, P<0.01). Post-hoc tests showed this was due to
the feeding interval on 1 M CCS being significantly higher than on both 1 M CS and CVS
(P<0.05 both cases).
Total feeding duration (Tot FD)
The total feeding duration was significantly higher for 2 and 2.5 M CS solutions against
the 1 M CS solution (F 6,60 = 6.947, P<0.01 in both cases, not shown but see Figure 9). The
82
total feeding duration of 0.25 M CS was higher (not significantly) than other CS solutions
up to 1.5 M CS, which can also be attributed to compensatory feeding. The differences seen
in the most dilute (0.25 M) and the two most concentrated (2 and 2.5 M) CS solutions
suggest that the birds ‘struggle’ to meet energy demands at the two extremes of the
concentration range.
CVS
CS
CCS
800
700
Total FD (s +/- SE)
600
500
400
300
200
100
0
0.25+T1 │ 0.25
0.5+T1 │ 0.5
0.7+T1 │ 0.7
1
1.5 │ 1+T1.5
2 │ 1+T2
2.5 │ 1+T2.5
Concentration (M)
Figure 9: Total feeding duration (Tot FD) of sucrose solutions by Cinnyris talatala (n = 11)
over a 5 hour experimental period on three different test series (for description of series see
Figure 4). There were no significant differences after RM-ANOVA and post-hoc analyses.
Between test series, no significant differences were found in total feeding duration
(Table 2, Figure 9). Total feeding duration of the CVS solutions was very similar to the CS
solutions of the same concentrations, and also largely comparable to the 1 M CVS control
solution. The 0.25+T1 CVS was slightly higher in total feeding duration than all other CVS
solutions, which compares well to the total feeding duration of the pure 0.25 M CS.
There is a small increase in total feeding duration across all the CCS solutions compared
to the pure 1 M CCS control solution, but all CCS solutions were broadly similar to each
83
other. However, the CCS total feeding durations were still lower than those of pure 2 and
2.5 M CS solutions. Total feeding duration between the 1 M solutions of the CS, CVS and
CCS were not different (F2,20 = 0.67, P=0.52). For the total feeding durations of 1 M CS,
CVS and CCS solutions, repeatability was low (0.43 ± 0.19 SE), and this contributed to the
low repeatability in energy intake rates.
Discussion
There are consequences for nectarivorous birds feeding on dilute nectar which have led
to several different adaptations (Nicolson & Fleming 2003a; Nicolson 2007b). However, the
effect of the viscosity of nectar on the feeding behaviour of nectarivorous birds has received
little attention. Baker (1975) first suggested that the high viscosity of concentrated nectars
limited the energy intake rates of avian nectar consumers and may explain the occurrence of
dilute nectars. The results of the present study show that for dilute and mid-range sucrose
concentrations with artificially increased viscosities, sunbirds have lower energy intake rates
which lead to decreased energy consumption. This indicates that viscosity is determining the
feeding behaviour of the sunbird and suggests a preingestional limitation that may lead to
energy deficits in these birds.
It was expected that the sunbirds would alter their feeding behaviour in response to
artificial nectars with increased viscosities in a compensatory manner. However when
feeding on more viscous artificial nectars sunbirds did not alter their feeding behaviour in
any significant way to compensate for the decrease in energy intake rates and energy
consumption. The high feeding frequency and total feeding duration observed on 0.25 M
CS, and the similarity in mean feeding duration across all the concentrations of the CS are
84
comparable to results obtained by Köhler et al. (2006). They looked at short-term feeding
patterns in white-bellied sunbirds using 10, 20 and 30% w/w sucrose solutions (0.3, 0.7 and
1 M respectively). Significantly more feeding events and a longer total feeding duration
were recorded on 10% sucrose, and there were no differences in feeding duration between
the concentrations. The only behaviour that showed a weak response to viscosity was mean
feeding duration per feeding event, which increased slightly on all concentrations with
artificially increased viscosities. This is similar to the response seen in large workers of
nectarivorous ants, which increased feeding duration on 30% w/w sucrose solutions with a
viscosity equivalent to that of a 60% solution (Medan & Josens 2005). This lack of change
in feeding behaviour with higher viscosities led to the lower energy intake rates and energy
consumptions observed. In terms of feeding behaviour, sunbirds broadly responded to the
CVS and CCS solutions as if they were pure 1 M sucrose solutions. For CVS solutions, this
implies that birds responded to viscosity as a feeding stimulus, while for CCS solutions, it
implies that birds responded to concentration as a feeding stimulus.
The overall trend in energy intake rates, with a peak at 1 M sucrose, is loosely
comparable to the data of Josens & Farina (2001): intake rates in hovering hawk moths
increased with increasing concentration to 20%, and decreased beyond this concentration as
viscosity increased. Low intake rates were recorded for the sunbirds on the most dilute (0.25
M) and two most concentrated (2 and 2.5 M) concentrations. According to Baker’s (1975)
hypothesis, energy intake rate is limited by the low energy content at low concentrations –
this is evident by the fact that sunbirds had very high feeding frequencies, total feeding
duration and short feeding intervals between feeding events on 0.25 M sucrose. At high
sugar concentrations, energy intake rate is limited by the mechanical effect of high viscosity
– this is also evident from the high feeding frequency, total feeding duration and short
85
feeding intervals recorded for sunbirds on 2 and 2.5 M sucrose. The birds “struggle” to feed
at the two extremes of the concentration range, at the dilute end because of energy content,
at the high concentration end because of higher viscosity. The mechanical effect of viscosity
on tongue loading may explain this phenomenon. The tongues of sunbirds ranging from 6 to
10 g in body mass are long bifurcated “tubes” (Skead 1967; Cheke et al. 2001),
approximately 0.2 mm in diameter and 0.46 μl in volume (Schlamowitz et al. 1976; Downs
2004). Nectar is loaded on the tongue by capillarity (Schlamowitz et al. 1976; Skead 1976;
Downs 2004) and high sugar concentrations with high viscosities could impose constraints
on the feeding efficiency of birds. If sunbirds were not fully loading their tongues on 2 and
2.5 M sucrose because the solutions were too viscous for efficient capillarity, it would
explain the need for higher feeding frequencies, total feeding durations and shorter feeding
intervals between feeding events observed.
When viscosity was kept constant (CVS), hovering hawk moths had higher intake rates
for lower concentrations than they did for pure 50% sucrose solution (Josens & Farina
2001). This is in contrast to sunbirds, which decreased intake rates when viscosity was kept
constant (CVS) at lower sugar concentrations. However, intake rates did not differ between
CCS and pure solutions of the same viscosities in the hovering hawk moth (Josens & Farina
2001). The sunbirds also showed no difference in intake rates between CCS and CS (pure)
solutions of the same viscosity, which indicates that viscosity is the only physical limitation
to feeding behaviour if solutions of different viscosity are presented in an equal manner
(such as similar flower morphology).
The results lead to another question - why do white-bellied sunbirds not respond in a
compensatory manner to a viscosity higher than that caused by sugar concentration in
86
nectar? One possibility is that sugar concentration alone may be the driving force behind the
feeding behaviour of nectarivorous birds, regardless of viscosity. The results for the CCS
solutions imply that the birds were responding to concentration as the feeding stimulus. This
provides some support for the conclusion of Stromberg & Johnsen (1990) regarding blackchinned hummingbirds, though there are reservations about this work due to the use of
CMC (sodium carboxymethylcellulose) and amino acid containing artificial sweeteners. The
viscosity of sucrose solutions (10, 20, 30 and 40% w/v) and solutions containing artificial
sweeteners (Equal ®, pure aspartame and saccharin, mixed to mimic the sweetness of 20%
sucrose) were measured with a capillary viscometer at the ambient temperatures of the study
site. Artificial sweetener solutions were increased to sucrose equivalent viscosities with
CMC. These solutions were supplied to the birds together with solutions containing sucroseonly and CMC-only (sucrose equivalent viscosities) in three-way tests. Artificial sweeteners
alone were first found to be ignored by the hummingbirds, and increasing their viscosity
made no difference to their acceptance. 20% sucrose was significantly preferred over a
CMC solution of equal viscosity to 20% sucrose. Thus the authors concluded that sweet
stimuli rather than viscosity were driving feeding responses in hummingbirds.
Alternatively, or concurrently, it may be that viscosities of natural nectars never increase
beyond the viscosities caused by sugar concentration alone, and may therefore be negligible
for sunbirds. As the 0.7 M sucrose solutions with added salt or amino acids show, other
solutes common in floral nectars may not greatly affect the viscosity. 0.7 M is a typical
concentration for floral nectars of plants pollinated by passerines (Nicolson 2002, Johnson
& Nicolson 2008). Amino acids typically occur at low concentrations in floral nectars
(Nicolson 2007a, Nicolson & Thornburg 2007). Salt concentrations in the nectar of 19
species of hummingbird-pollinated flowers averaged 24.7 mM for potassium, 3.4 mM for
87
sodium and 9.9 mM for chloride (Hiebert & Calder 1983). In some southern African plant
species, potassium can occur from 4.2 to 17 mM and sodium from 3.3 to 18 mM in nectar
(Nicolson & Thornburg 2007). The impact of other solutes and fallen pollen on nectar
viscosity is not known, and small quantities of certain solutes may have a large effect on
viscosity. High molecular weight polymers cause a jelly-like consistency in the nectar of
vertebrate-pollinated flowers (Johnson et al. 2001; Sazima et al. 2001).
Sunbirds were clearly limited in sugar intake rates and sugar (energy) consumption by
viscosity. But why did the sunbirds’ feeding behaviour on CVS solutions resemble that on
pure 1 M sucrose solutions? In food science, it is generally understood that increasing
viscosity through the addition of a thickener results in a decrease in the perceived sensitivity
of volatile and non-volatile components in humans, especially reducing sweetness
perception (Hollowood et al. 2002). At low concentrations (<0.5g/100g) of hydroxyl propyl
methylcellulose (HPMC), the perception of sweetness (sucrose solutions) was unaffected,
but at higher concentrations (>0.6g/100g) perceived sweetness decreased. However, if this
was the case for sunbirds, they should have shown a response to the perceived decrease in
sweetness as a “lower concentration”, and should have fed at higher frequencies and longer
feeding durations, due to the employment of compensatory feeding. This was not the case.
In Chapter 1 I reported that concentration preferences within a sugar type peak at 1 M
for both sucrose and equicaloric hexoses. Data on the feeding behaviour presented here
support this finding. White-bellied sunbirds had the lowest feeding frequency and total
feeding duration, as well as the highest feeding interval between feeding events when
offered 1 M sucrose. This gives the birds more “free time” to devote to other behaviours,
such as territorial defence and courtship, and would imply an advantage to selecting nectar
88
of 1 M sugar concentration. But this does not really explain the results on the dilute-high
viscosity solutions (CVS), unless the birds base their feeding preferences of dilute (less than
1 M sugar) concentrations on viscosity. This is possible, as the viscosities of sucrose
solutions from 0.25 to 1 M are very similar. Most natural nectars occur in this range and
concentrations above 1 M are generally not common (Pyke & Waser 1981; Nicolson 2002;
Nicolson & Fleming 2003a; Johnson & Nicolson 2008). In a natural context, the comparison
of sucrose and equicaloric hexose mixtures is not important for viscosity – the effect of
viscosity will occur regardless of sugar type. However, effects may be somewhat different
between sugar types: nectars have complex compositions, and sugars are never pure, though
some nectars are dominated by sucrose or by hexoses (Nicolson & Thornburg 2007;
Johnson & Nicolson 2008). The viscosity of equicaloric hexoses is approximately half that
of sucrose from 1.5 to 2.5 M, and suggest an advantage to feeding on hexose-dominated
nectars at high concentrations.
Despite these possible explanations for the lack of response in feeding behaviour by
sunbirds, it is clear that viscosities of artificial nectar higher than those caused by sugar
concentration alone do impede energy intake rates, and therefore energy consumption. The
results imply that this limitation is mechanical as the birds did not alter their behaviour to
compensate for higher viscosities. In other words, the sunbirds are experiencing reduced
sugar intake rates potentially leading to energy deficits because the higher viscosity of their
food may be reducing the amount of nectar consumed per lick of the tongue. This is
possible, because the birds appear to regulate their feeding behaviour according to viscosity
on dilute solutions (less than 1 M), while on the mid-range solution (1 M) they appear to
regulate their feeding behaviour according to concentration.
89
Acknowledgements Experiments were approved by the Animal Use and Care Committee
of the University of Pretoria, and were funded by the South African National Research
Foundation (NRF). Hennie Johnson is thanked for his assistance in the building the housing
for the infra-red photo-detection LED system. The Centre for Applied Materials at the
University of Pretoria and Mr. L Moitsheki are thanked for assistance and use of the Aanton
Paar (GmbH) rheometer. Craig Symes is thanked for mist-netting sunbirds with permission
from Jan Cilliers Park and the Gauteng Department of Nature Conservation. Thanks also go
to SE Tylose GmbH & Co. KG in Wiesbaden, Germany, for providing me with a free
sample of Tylose ®.
References
Baker HG (1975) Sugar concentrations in nectars from hummingbird flowers. Biotropica
7:37-41
Birch GG, Kemp SE (1989) Apparent specific volumes and tastes of amino acids. Chem
Senses 14:249-258
Bourne MC (1982) Food science and technology: concept and measurement. Academic
Press Inc., New York, USA
Cheke CA, Mann CF, Allen R (2001) Sunbirds: A guide to the sunbirds, flowerpeckers,
spiderhunters and sugarbirds of the world. Christopher Helm, London, UK
90
Downs CT (2004) Some preliminary results of studies on the bill and tongue morphology of
Gurney’s sugarbird and some southern African sunbirds. Ostrich 75:169-175
Fleming PA, Gray DA, Nicolson SW (2004) Circadian rhythm of water balance and
aldosterone excretion in the white-bellied sunbird Nectarinia talatala. J Comp Physiol B
174:341–346
Gass CL, Roberts WM (1992) The problem of temporal scale in optimization: three
contrasting views of hummingbird visits to flowers. Am Nat 140:829-853
Hainsworth FR, Wolf LL (1976) Nectar characteristics and food selection by
hummingbirds. Oecologia 25:101–113
Heyneman AJ (1983) Optimal sugar concentrations of floral nectars – dependence on sugar
intake efficiency and foraging costs. Oecologia 60:198-213
Hiebert SM, Calder WA (1983) Sodium, potassium, and chloride in floral nectars: energyfree contributions to refractive index and salt balance. Ecology 64:399-402
Hollowood TA, Linforth RST, Taylor AJ (2002) The effect of viscosity on the perception of
flavour. Chem Senses 27:583-591
Johnson SD, Pauw A, Midgley J (2001) Rodent pollination in the African lily Masonia
depressa (Hyacinthaceae). Am J Bot 88:1768-1773
91
Johnson SD, Nicolson SW (2008) Evolutionary associations between nectar properties and
specificity in bird pollination systems. Biol Lett 4:49-52
Josens RB, Farina WM (2001) Nectar feeding by the hovering hawk moth Macroglossum
stellatarum: intake rate as a function of viscosity and concentration of sucrose solutions. J
Comp Physiol A 187:661-665
Kingsolver JG, Daniel TL (1983) Mechanical determinants of nectar feeding strategy in
hummingbirds: energetics, tongue morphology, and licking behaviour. Oecologia 60:214226
Köhler A, Verburgt L, Nicolson SW (2006) Short-term energy regulation of whitebellied
sunbirds (Nectarinia talatala): effects of food concentration on feeding frequency and
duration. J Exp Biol 209:2880-2887
Leseigneur CDC, Verburgt L, Nicolson SW (2007) Whitebellied sunbirds (Nectarinia
talatala, Nectariniidae) do not prefer artificial nectar containing amino acids. J Comp
Physiol B 177:679-685
Lotz CN, Martínez del Rio C (2004) The ability of rufous hummingbirds Selasphorus rufus
to dilute and concentrate urine. J Avian Biol 35:54-62
Mathlouthi M, Génotelle J (1995) Rheological properties of sucrose solutions and
suspensions. In: Mathlouthi M, Reiser P (eds) Sucrose properties and applications. Blackie
Academic & Professional
92
Medan V, Josens RB (2005) Nectar foraging behaviour is affected by ant body size in
Camponotus mus. J Insect Physiol 51:853-860
Nicolson SW (2002) Pollination by passerine birds: why are the nectars so dilute? Comp
Biochem Physiol B 131:645-652
Nicolson SW, Fleming PA (2003a) Nectar as food for birds: the physiological consequences
of drinking dilute sugar solutions. Plant Syst Evol 238:139-153
Nicolson SW, Fleming PA (2003b) Energy balance in the whitebellied sunbird Nectarinia
talatala: constraints on compensatory feeding, and consumption of supplementary water.
Funct Ecol 17:3-9
Nicolson SW, Hoffmann D, Fleming PA (2005) Short-term energy regulation in nectarfeeding birds: the response of white-bellied sunbirds (Nectarinia talatala) to midday fast.
Funct Ecol 19:988-994
Nicolson SW (2007a) Amino acids concentrations in the nectars of southern African birdpollinated flowers, especially Aloe and Erythrina. J Chem Ecol 33:1707-1720
Nicolson SW (2007b) Nectar consumers. In: Nicolson SW, Nepi M, Pacini E (eds)
Nectaries and nectar. Springer, Dordrecht
Nicolson SW, Thornburg RW (2007) Nectar chemistry. In: Nicolson SW, Nepi M, Pacini E
(eds) Nectaries and nectar. Springer, Dordrecht
93
Pyke GH, Waser NM (1981) The production of dilute nectars by hummingbird and
honeyeater flowers. Biotropica 13:260-270
Rice WR (1989) Analyzing tables of statistical tests. Evolution 43:223-225
Sazima M, Vogel S, do Prado AL, de Oliveira DM, Franz G, Sazima I (2001) The sweet
jelly of Combretum lanceolatum flowers (Combretaceae): a cornucopia resource for bird
pollinators in the Pantanal, western Brazil. Plant Sys Evol 227:195-208
Schlamowitz R, Hainsworth FR, Wolf LL (1976) On the tongues of sunbirds. Condor
78:104-107
Skead CJ (1967) Sunbirds of southern Africa. Cape & Transvaal Printers Ltd. Cape Town,
South Africa
Stromberg MR, Johnsen PB (1990) Hummingbird sweetness preferences: taste or viscosity?
Condor 92:606-612
Tezze AA, Farina WM (1999) Trophallaxis in the honeybee, Apis mellifera: the interaction
between viscosity and sucrose concentration of the transferred solution. Anim Behav
57:1319-1326
Vogel S (1994) Life in moving fluids: the physical biology of flow. 2nd edn. Princeton
University Press
94
Weast RC (1980) CRC handbook of chemistry and physics, 60th edn. CRC Press, Inc.
Florida USA
95
Chapter 4
Conclusion
Sunbirds are specialized nectarivorous passerines and important pollinators of many
plant species (Skead 1967; Cheke et al. 2001). They rely heavily on nectar as their primary
food source and have highly energetic lifestyles. The white-bellied sunbird Cinnyris
talatala, in particular, habituates well to captivity and has been the subject of much research
concerning floral nectars and energy management, and the physiological adaptations of the
birds to a simple, watery diet (for example, Nicolson & Fleming 2003; Fleming et al. 2004;
Köhler et al. 2006). For this reason, the white-bellied sunbird was chosen for the present
study to address specific aspects of their feeding preferences and behaviour.
Though research into feeding behaviour, physiology and plant-pollinator relationships of
nectarivorous birds is extensive (for some reviews see Martinéz del Rio et al. 2001;
Nicolson 2002; Lotz & Schondube 2006; Johnson & Nicolson 2008), there is a lack of
research with regards to the concentration preferences of nectarivorous birds within a sugar
type, and nothing clear is known about the effects of viscosity on feeding behaviour. The
present study included the analysis of concentration preferences by the white-bellied sunbird
within sugar types, and the first analysis of the effect of viscosity separate from sugar
concentration on the feeding behaviour of a nectarivorous bird.
96
The results from the concentration preference experiments revealed that sunbirds do
prefer certain concentrations over others within a sugar type. The peak preference occurred
at 1 M for both sucrose and equicaloric hexose solutions. This is significant in terms of the
birds’ physiology. White-bellied sunbirds effectively dilute higher concentrations to near 1
M by consuming supplementary water (Nicolson & Fleming 2003). This indicates that a 1
M solution not only meets energy demands, but also meets water requirements perhaps
without the necessity of shunting excess water past the kidneys (McWhorter et al 2003;
Nicolson & Fleming 2003; Nicolson 2006). Palestine sunbirds (Necatrinia osea) can bypass
the kidneys by not absorbing up to two thirds of ingested water when feeding on dilute diets,
passing the water to be expelled (McWhorter et al. 2003). It has also been found that
fractional water reabsorption in the kidney is sensitive to water status (McWhorter et al.
2004). The results are also of ecological significance. Birds that can discriminate between
concentrations and choose those that maximize their free time for other activities, such as
territory defence and mate choice, may more effectively increase their fitness. Fine scale
discrimination between concentrations was also observed for sunbirds, as in hummingbirds
and other nectarivorous passerines (Levey 1987; Lloyd 1989, Blem et al. 2000; Schaefer et
al. 2003). However, this is stated with caution, as there was high individual variation and
side bias in this experiment and the fact that it was not repeatable (see Appendix) casts some
doubt on the certainty of the result.
The present study also revealed that viscosity appears to be of great importance as a
limiting factor in energy intake rates and energy consumption of white-bellied sunbirds.
Given the concentration preferences, and the expected mechanical effect of viscosity on
feeding by affecting tongue loading, sunbirds were expected to alter their feeding behaviour
as viscosity of sugar solutions changed. However, the results showed that sunbirds do not
97
alter their feeding behaviour to compensate for higher viscosity, but they did respond in an
unexpected way. They responded in their feeding behaviour to dilute solutions whose
viscosity was equivalent to a pure 1 M sucrose as if they were pure 1 M solutions, and they
treated 1 M sucrose solutions with variedly increased viscosities as pure 1 M solutions too.
In other words, viscosity appears to be the primary factor in driving foraging choices on
dilute solutions, while concentration becomes the primary factor from mid-range
concentrations. This is a highly significant result as it agrees with the findings of
concentration preference, and the results provide support for Baker’s (1975) hypothesis: the
viscosity of nectar increases exponentially with sugar concentration, and low concentrations
are necessary for efficient extraction of nectar from flowers.
Of great significance in the present study was the fact that sunbirds consumed greater
volumes of equicaloric hexose than sucrose solutions at the highest concentrations (1.5 – 2.5
M). This suggests an advantage to feeding on hexose dominated nectars for white-bellied
sunbirds, as the viscosity of hexose solutions at these high concentrations is approximately
half that of sucrose solutions. It is important to note that these concentrations (1.5 - 2.5 M)
are not common in nature (Pyke & Waser 1981; Nicolson 2002; Nicolson & Fleming 2003a;
Johnson & Nicolson 2008). In a natural context, the comparison of sucrose and equicaloric
hexose mixtures is not important – the effect of viscosity will occur regardless of sugar type.
However, effects may be somewhat different between sugar types due to the difference in
viscosity. Nectars have complex compositions and sugars are never pure, though some are
dominated by either sucrose or by hexoses (Nicolson & Thornburg 2007; Johnson &
Nicolson 2008).
98
In conclusion, concentration drives foraging in the white-bellied sunbird, but viscosity
impedes normal intake rates and determines choices of dilute solutions. It is however
important to state that the viscosity of artificial nectars comprising solutes typical of natural
nectars (at typical concentrations) do not increase viscosity as much as was done in this
study. But other solutes may more adversely affect viscosity, especially trace amounts of
certain carbohydrates. For example, high molecular weight polymers cause a jelly-like
consistency in the nectar of an African lily and a species of Combretum, which are
vertebrate-pollinated flowers (Johnson et al. 2001; Sazima et al. 2001).
References
Baker HG (1975) Sugar concentrations in nectars from hummingbird flowers. Biotropica
7:37-41
Blem CR, Blem LB, Felix J, van Gelder J (2000) Rufous hummingbird sucrose preference:
precision of selection varies with concentration. Condor 102:235-238
Cheke CA, Mann CF, Allen R (2001) Sunbirds: A guide to the sunbirds, flowerpeckers,
spiderhunters and sugarbirds of the world. Christopher Helm, London, UK
Fleming PA, Hartman Bakken B, Lotz CN, Nicolson SW (2004) Concentration and
temperature effects on sugar intake preferences in a sunbird and a hummingbird. Funct Ecol
18:223–232
99
Johnson SD, Pauw A, Midgley J (2001) Rodent pollination in the African lily Masonia
depressa (Hyacinthaceae). Am J Bot 88:1768-1773
Johnson SD, Nicolson SW (2008) Evolutionary associations between nectar properties and
specificity in bird pollination systems. Biol Lett 4:49-52
Köhler A, Verburgt L, Nicolson SW (2006) Short-term energy regulation of whitebellied
sunbirds (Nectarinia talatala): effects of food concentration on feeding frequency and
duration. J Exp Biol 209:2880-2887
Levey DJ (1987) Sugar-tasting ability and fruit selection in tropical fruit-eating birds. Auk
104:173-179
Lloyd P (1989) Sucrose concentration preferences of two southern African sunbirds. Ostrich
60:134-135
Lotz CN, Schondube JE (2006) Sugar preferences in nectar- and fruit-eating birds:
behavioural patterns and physiological causes. Biotropica 38:3-15
Martínez 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. Am Zool 41:902-915
McWhorter TJ, Martínez del Rio C, Pinshow B (2003) Modulation of ingested water
absorption by Palestine sunbirds: evidence for adaptive regulation. J Exp Biol 206:659-666
100
McWhorter TJ, Martínez del Rio C, Pinshow B, Roxburgh L (2004) Renal function in
Palestine sunbirds: elimination of excess water does not constrain energy intake rate. J Exp
Biol 207:3391-3398
Nicolson SW (2002) Pollination by passerine birds: why are the nectars so dilute? Comp
Biochem Physiol B 131:645-652
Nicolson SW, Fleming PA (2003) Nectar as food for birds: the physiological consequences
of drinking dilute sugar solutions. Plant Syst Evol 238:139-153
Nicolson SW (2006) Water management in nectar-feeding birds. Am J Physiol Rgul Integr
Comp Physiol 291:R828-R829
Sazima M, Vogel S, do Prado AL, de Oliveira DM, Franz G, Sazima I (2001) The sweet
jelly of Combretum lanceolatum flowers (Combretaceae): a cornucopia resource for bird
pollinators in the Pantanal, western Brazil. Plant Sys Evol 227:195-208
Schaefer HM, Schmidt V, Bairlein F (2003) Discrimination abilities for nutrients: which
difference matters for choosy birds and why? Anim Behav 65:531-541
Skead CJ (1967) Sunbirds of southern Africa. Cape & Transvaal Printers Ltd. Cape Town,
South Africa
101
Appendix
Further experiments on fine scale concentration preferences of
white-bellied sunbirds
Summary
In Chapter 1, fine scale concentration preferences of white-bellied sunbirds were
examined. The results demonstrated that the birds could distinguish between concentrations
differing by as little as 0.03 M (1% w/w) for sucrose and 0.05 M (2% w/w) for equicaloric
hexose solutions. These resembled some previous findings in hummingbirds and other
nectarivorous passerines (see Chapter 1 for results and discussion). However, since there
was much individual variation between birds and side bias in the first fine scale
concentration preference experiment, and a discrepancy between sucrose and hexose
differences (1 and 2 % w/w differences), I opted to repeat the experiment. The experiment
was repeated twice: once more on the same set of birds (after one year in captivity – the
original experiment (Chapter 1) was after six months in captivity) and again on a new set of
relatively young birds (after one month in captivity). Both groups of sunbirds however
showed no fine scale preferences except for the second group which showed a preference
for the highest concentration available – but this result is tentative as there was high
individual variation between these birds.
102
Repetition of this experiment on fine scale concentration preferences has produced
results that conflict with those in the original experiment. This can not be explained by the
time in captivity, since similar data were obtained with the original group of birds (now
older) and the new group of sunbirds. It may be that there is a factor of experience missing
for the new group of sunbirds as several were relatively young, though their ages are not
defined and this is speculative. The reasons for these conflicting results are unclear and
warrant further investigation.
Materials & Methods
Birds and maintenance
The same mature birds used for experiments in Chapter 1 were used for the first fine
scale concentration preference experiment reported here (set 1, captured in July 2006), (n =
8, 8.51 ± 0.18 g). The second set of sunbirds for the second experiment reported here (set 2),
three males and four females (7.7 ± 0.69 g), were captured in April 2007 at Jan Cilliers
Park. Maintenance and housing of the birds followed the procedure described in Chapter 1.
Fine scale concentration preference experiments were repeated as described in Chapter 1
for both sets of sunbirds. The differences between concentrations in pairwise tests in the
new experiments were 0.03 M (1% w/w) for both sucrose and equicaloric hexose solutions
(0.27 vs. 0.3 M, 0.47 vs. 0.5 M and 0.97 vs. 1 M). Data were arranged and analysed as
described in Chapter 1: data were collected as mass (g) of each solution consumed (mean ±
SE), then converted to volumetric consumption (ml) based on the density of sugar solutions
at each concentration, and to sugar consumption (g). Sugar consumption data were analysed
using the STATISTICA 7.1 (StatSoft Inc. ©, Tulsa, OK, USA, 2006) software package.
103
For initial analysis, an overall repeated-measures ANOVA (RM-ANOVA) was
performed for all of the consumption data to determine where differences occurred. The
overall
RM-ANOVA
comprised
three
within-effects,
nested
as
follows:
“DIET(CONC(TESTDAY))” where TESTDAY (the difference between the two days of each
preference test) was nested within CONC (the difference between the two concentrations),
which was nested within DIET (the difference between the three different concentration
pairs). Following the overall RM-ANOVA, consumption values were tested separately for
each concentration pair by separate RM-ANOVA. The structure of these RM-ANOVA
therefore comprised two within-effects nested as follows: “CONC(TESTDAY)”. Post-hoc
analyses for each RM-ANOVA were conducted using Tukey’s Honest Significant
Difference (HSD) test for equal sample sizes.
Results
Fine scale concentration preferences – set 1
The first set of sunbirds showed no preferences between concentrations differing by 1%
w/w (Figure 1). The overall RM-ANOVA for sugar consumption for both sugar types show
that there were also significant differences between diets for sucrose and hexose mixtures
(sucrose F2,14 = 112.8, P<0.001, hexose F2,14 = 27.226, P<0.001), but there were no
significant differences between concentrations for either sugar type (sucrose F1,7 = 1.409,
P=0.273, hexose F1,7 = 0.063, P=0.809). Post-hoc analysis (Tukey’s HSD) revealed that
these significant effects were, as for broad scale preferences (Chapter 1), due to the mean
consumption of each concentration pair increasing as concentrations increased, and some
birds also showed side biases. Also, consumption varied more between test days than during
the first fine scale preference experiment (Chapter 1).
104
Data were separated by diet for further analysis. No significant differences were found
between any concentration pairs for either sugar type (Table 1 and Figure 1). The difference
between test days for sucrose diets was due to individual variation and side biases.
Table 1: Sugar consumption (g) differences between concentrations (CONC), days (TESTDAY),
and their interaction (CONCxTESTDAY) for sucrose and the hexose mixture (1:1 equicaloric
glucose and fructose) for white-bellied sunbirds, Cinnyris talatala (set 1). Values are F-values
from RM-ANOVA, with significant effects after Tukey’s HSD indicated by asterisks (*P<0.05);
n = 8 and d.f. = 1 in all cases.
Sucrose
Glucose: Fructose
Concentration
Pairs (M)
CONC
TESTDAY
CONC x
TESTDAY
CONC
TESTDAY
CONC x
TESTDAY
0.27 vs. 0.3
0.3072
10.1396*
6.8416
0.0867
2.8729
0.0572
0.57 vs. 0.6
0.9650
6.2815*
4.4634
0.0115
0.6129
0.0005
0.97 vs. 1
3.0637
0.225
0.2831
0.154
0.882
0.002
Fine scale concentration preferences – set 2
The same experiment was conducted on a second set of sunbirds. Sunbirds failed to
show preferences between concentrations differing by 0.03 M (1% w/w) except at the
highest concentrations for sucrose (Figure 2). The overall RM-ANOVA for sugar
consumption showed significant differences between experimental diets for sucrose and
hexose mixtures (sucrose F2,14 = 46.391, P<0.001, hexose F2,14 = 52.601, P<0.001), and
between concentrations for sucrose (F1,7 = 6.714, P<0.05). Post-hoc analysis (Tukey’s HSD)
revealed that significant effects were due to the mean consumption of each concentration
pair increasing as concentrations increased.
105
Low Concentration
a
High Concentration
1.6
1.2
0.8
Sugar Consumption (g ± SE)
0.4
0.0
0.27 vs. 0.3
0.57 vs. 0.6
0.97 vs. 1
0.27 vs. 0.3
0.57 vs. 0.6
0.97 vs. 1
b
1.6
1.2
0.8
0.4
0.0
Concentration Pairs (M)
Figure 1:
Sugar consumption (g) by Cinnyris talatala (set 1) of sucrose (a) and hexose
(equicaloric 1:1 glucose and fructose) (b) solutions from paired preference tests (n = 8). No
significant differences in consumption between pairs were found after Tukey’s HSD (RMANOVA Table 1).
106
Data were again separated by diet for further analysis. There was a significant difference
in concentration between 0.97 and 1 M sucrose, indicating a preference by young birds for
0.97 M (Table 2 and Figure 2). However this result is uncertain. On day one of the
preference test all birds were indifferent to either concentration, and on day two all but two
of the birds consumed the 0.97 M sucrose solution almost exclusively over the 1 M solution.
This caused a significant difference between test days (Table 2), and this was possibly due
to experience gained from the first test day.
Table 2: Sugar consumption (g) differences between concentrations (CONC), between days
(TESTDAY), and their interaction (CONCxTESTDAY) for sucrose and the hexose mixture (1:1
equicaloric glucose and fructose) for white-bellied sunbirds, Cinnyris talatala (set 2). Values are
F-values from RM-ANOVA, with significant effects after Tukey’s HSD indicated by asterisks
(*P<0.05); n = 8 and d.f. = 1 for all effects.
Sucrose
Glucose: Fructose
Concentration
Pairs (M)
CONC
TESTDAY
CONC x
TESTDAY
CONC
TESTDAY
CONC x
TESTDAY
0.27 vs. 0.3
3.9682
0.1998
1.0921
0.647
2.354
0.394
0.57 vs. 0.6
0.5543
0.0005
0.9102
0.095
0.196
0.514
0.97 vs. 1
8.3522*
6.0473*
5.1215
1.031
0.691
0.276
107
Low Concentration
a
High Concentration
1.6
*
1.2
0.8
Sugar Consumption (g ± SE)
0.4
0.0
0.27 vs 0.3
0.57 vs 0.6
0.97 vs 1
b
1.6
1.2
0.8
0.4
0.0
0.27 vs. 0.3
0.57 vs. 0.6
0.97 vs. 1
Concentration Pairs (M)
Figure 2:
Sugar consumption (g) by Cinnyris talatala (set 2) of sucrose (a) and hexose
(equicaloric 1:1 glucose and fructose) (b) solutions from paired preference tests (n = 8).
Asterisks (*) indicate significant differences in consumption between pairs after Tukey’s HSD
(RM-ANOVA Table 2; CONC: *P < 0.05).
108
General References
Afik D, McWilliams SR, Karasov WH (1997) A test for passive absorption of glucose in
yellow-rumped warblers and its ecological implications. Physiol Zool 70:370-377
Baker HG (1975) Sugar concentrations in nectars from hummingbird flowers. Biotropica
7:37-41
Baker HG, Baker I (1982) Chemical constituents of nectar in relation to pollination
mechanisms and phylogeny. In: Nitecki HM (ed) Biochemical aspects of evolutionary
biology. University of Chicago Press
Baker HG, Baker I (1983) Floral nectar sugar constituents in relation to pollinator type. In:
Jones CE, Little RJ (eds) Handbook of experimental pollination biology. Van Nostrand
Reinhold, New York
Birch GG, Kemp SE (1989) Apparent specific volumes and tastes of amino acids. Chem
Senses 14:249-258
Blem CR, Blem LB, Felix J, van Gelder J (2000) Rufous hummingbird sucrose preference:
precision of selection varies with concentration. Condor 102:235-238
Bourne MC (1982) Food science and technology: Concept and measurement. Academic
Press Inc., New York, USA
109
Brown M, Downs CT, Jonson SD (2008) Sugar preferences of nectar feeding birds: a
comparison of experimental techniques. J Avian Biol in press
Chang M, Karasov WH (2004) How the house sparrow Passer domesticus absorbs glucose.
J Exp Biol 207:3109-3121
Cheke CA, Mann CF, Allen R (2001) Sunbirds: A guide to the sunbirds, flowerpeckers,
spiderhunters and sugarbirds of the world. Christopher Helm, London, UK
Corbet SA (2003) Nectar sugar content: estimating standing crop and secretion rate in the
field. Apidologie 34:1-10
Downs CT (1997) Sugar preference and apparent sugar assimilation in the Red Lory. Aust J
Zool 45:613-619
Downs CT (2004) Some preliminary results of studies on the bill and tongue morphology of
Gurney’s sugarbird and some southern African sunbirds. Ostrich 75:169-175
Fleming PA, Gray DA, Nicolson SW (2004a) Circadian rhythm of water balance and
aldosterone excretion in the whitebellied sunbird Nectarinia talatala. J Comp Physiol B
174:341-346
Fleming PA, Bakken Hartman B, Lotz CN, Nicolson SW (2004b) Concentration and
temperature effects on sugar intake preferences in a sunbird and a hummingbird. Funct Ecol
18:223–232
110
Fleming PA, Xie S, Napier K, McWhorter TJ, Nicolson SW (2008) Nectar concentration
affects sugar preferences in two Australian honeyeaters and a lorikeet. Funct Ecol
(doi:10.1111/j.1365-2435.2008.01401.x)
Franke E, Jackson S, Nicolson S (1998) Nectar sugar preferences and absorption in a
generalist African frugivore, the Cape White-eye Zosterops pallidus. Ibis 140:501-506
Gass CL, Roberts WM (1992) The problem of temporal scale in optimization: three
contrasting views of hummingbird visits to flowers. Am Nat 140:829-853
Hainsworth FR, Wolf LL (1976) Nectar characteristics and food selection by
hummingbirds. Oecologia 25:101–113
Harriman AE, Milner JS (1969) Preference for sucrose solutions by Japanese quail
(Coturnix coturnix japonica) in two-bottle drinking tests. Am Midl Nat 81:575-578
Herrera LG (1999) Preferences for different sugars in neotropical nectarivorous and
frugivorous bats. J Mammal 80:683-688
Heyneman AJ (1983) Optimal sugar concentrations of floral nectars – dependence on sugar
intake efficiency and foraging costs. Oecologia 60:198-213
Hiebert SM, Calder WA (1983) Sodium, potassium, and chloride in floral nectars: energyfree contributions to refractive index and salt balance. Ecology 64:399-402
111
Hockey PAR, Dean WRJ, Ryan PG (eds) (2005) Roberts: Birds of Southern Africa, VIIth
edition. The Trustees of the John Voelcker Bird Book Fund, Cape Town, South Africa
Hollowood TA, Linforth RST, Taylor AJ (2002) The effect of viscosity on the perception of
flavour. Chem Senses 27:583-591
Jackson S, Nicolson SW, Lotz CN (1998) Sugar preferences and “side bias” in Cape
Sugarbirds and Lesser Double-Collared Sunbirds. Auk 115:156–165
Johannsmeier MF (ed) (2001) Beekeeping in South Africa. Agricultural Research Council,
pp 152
Johnson SD, Pauw A, Midgley J (2001) Rodent pollination in the African lily Masonia
depressa (Hyacinthaceae). Am J Bot 88:1768-1773
Johnson SD, Nicolson SW (2008) Evolutionary associations between nectar properties and
specificity in bird pollination systems. Biol Lett 4:49-52
Josens RB, Farina WM (2001) Nectar feeding by the hovering hawk moth Macroglossum
stellatarum: intake rate as a function of viscosity and concentration of sucrose solutions. J
Comp Physiol A 187:661-665
Karasov WH, Cork SJ (1994) Glucose absorption by a nectarivorous bird: the passive
pathway is paramount. Am J Physiol 267:G18-G26
112
Kingsolver JG, Daniels TL (1983) Mechanical determinants of nectar feeding strategy in
hummingbirds: energetics, tongue morphology, and licking behaviour. Oecologia 60:214226
Köhler A, Verburgt L, Nicolson SW (2006) Short-term energy regulation of whitebellied
sunbirds (Nectarinia talatala): effects of food concentration on feeding frequency and
duration. J Exp Biol 209:2880-2887
Leseigneur CDC, Verburgt L, Nicolson SW (2007) Whitebellied sunbirds (Nectarinia
talatala, Nectariniidae) do not prefer artificial nectar containing amino acids. J Comp
Physiol B 177:679-685
Levey DJ (1987) Sugar-tasting ability and fruit selection in tropical fruit-eating birds. Auk
104:173-179
Levey DJ, Cipollini ML (1996) Is most glucose absorbed passively in northern bobwhite?
Comp Biochem Physiol 113A:225-231
Levey DJ, Martínez del Rio C (2001) It takes guts (and more) to eat fruit: lessons from
avian nutritional ecology. Auk 118:819-831
Lloyd P (1989) Sucrose concentration preferences of two southern African sunbirds. Ostrich
60:134-135
113
Lotz CN, Nicolson SW (1996) Sugar preferences of a nectarivorous passerine bird, the
Lesser Double-collared Sunbird (Nectarinia chalybea). Funct Ecol 10:360-365
Lotz CN, Nicolson SW (1999) Energy and water balance in the lesser double-collared
sunbird (Nectarinia chalybea) feeding on different nectar concentrations. J Comp Physiol B
169:200-206
Lotz CN, Martínez del Rio C (2004) The ability of rufous hummingbirds Selasphorus rufus
to dilute and concentrate urine. J Avian Biol 35:54-62
Lotz CN, Schondube JE (2006) Sugar preferences in nectar- and fruit-eating birds:
behavioural patterns and physiological causes. Biotropica 38:3-15
Martínez del Rio C (1990) Sugar preferences in hummingbirds: The influence of subtle
chemical differences on food choice. Condor 92:1022-1030
Martínez del Rio C, Karasov WH (1990) Digestion strategies in nectar- and fruit-eating
birds and the sugar composition of plant rewards. Am Nat 136:618-637
Martínez 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. Am Zool 41:902-915
114
Mata A, Bosque C (2004) Sugar preferences, absorption efficiency and water influx in a
Neotropical nectarivorous passerine, the bananaquit (Coereba flaveola). Comp Biochem
Physiol A 139:395-404
Mathlouthi M, Génotelle J (1995) Rheological properties of sucrose solutions and
suspensions. In: Mathlouthi M, Reiser P (eds) Sucrose properties and applications. Blackie
Academic & Professional
McWhorter TJ, Martínez del Rio C (2000) Does gut function limit hummingbird food
intake? Physiol Biochem Zool 73:313-324
McWhorter TJ, Martínez del Rio C, Pinshow B (2003) Modulation of ingested water
absorption by Palestine sunbirds: evidence for adaptive regulation. J Exp Biol 206:659-666
McWhorter TJ, Martínez del Rio C, Pinshow B, Roxburgh L (2004) Renal function in
Palestine sunbirds: elimination of excess water does not constrain energy intake rate. J Exp
Biol 207:3391-3398
McWhorter TJ (2005) Carbohydrate hydrolysis and absorption: lessons from modelling
digestive function. In: Starck JM and Wang T (eds) Physiological and ecological
adaptations to feeding in vertebrates. Science Publishers, Enfield, New Hampshire.
McWhorter TJ, Hartmann Bakken B, Karasov WH, Martínez del Rio (2005) Hummingbirds
rely on both paracellular and carrier-mediated intestinal glucose absorption to fuel high
metabolism. Biol Lett 2:131-134
115
Medan V, Josens RB (2005) Nectar foraging behaviour is affected by ant body size in
Camponotus mus. J Insect Physiol 51:853-860
Napier KR, Purchase C, McWhorter TJ, Nicolson SW, PA Fleming (2008) The sweet life:
diet sugar concentration influences paracellular glucose absorption. Biol Lett (doi:
10.1098/rsbl.2008.0253)
Nicolson SW (1998) The importance of osmosis in nectar secretion and its consumption by
insects. Am Zool 38:418-425
Nicolson SW (2002) Pollination by passerine birds: why are the nectars so dilute? Comp
Biochem Physiol B 131:645-652
Nicolson SW, Fleming PA (2003a) Nectar as food for birds: the physiological consequences
of drinking dilute sugar solutions. Plant Syst Evol 238:139-153
Nicolson SW, Fleming PA (2003b) Energy balance in the whitebellied sunbird Nectarinia
talatala: constraints on compensatory feeding, and consumption of supplementary water.
Funct Ecol 17:3-9
Nicolson SW, Hoffmann D, Fleming PA (2005) Short-term energy regulation in nectarfeeding birds: the response of white-bellied sunbirds (Nectarinia talatala) to a midday fast.
Funct Ecol 19:988-994
116
Nicolson SW (2006) Water management in nectar-feeding birds. Am J Physiol Rgul Integr
Comp Physiol 291:R828-R829
Nicolson SW (2007a) Amino acids concentrations in the nectars of southern African birdpollinated flowers, especially Aloe and Erythrina. J Chem Ecol 33:1707-1720
Nicolson SW (2007b) Nectar consumers. In: Nicolson SW, Nepi M, Pacini E (eds)
Nectaries and nectar. Springer, Dordrecht
Nicolson SW, Thornburg RW (2007) Nectar chemistry. In: Nicolson SW, Nepi M, Pacini E
(eds) Nectaries and nectar. Springer, Dordrecht
Paton DC, Collins BG (1989) Bills and tongues of nectar-feeding birds: a review of
morphology, function and performance, with intercontinental comparisons. Aust Ecol
14:473-506
Pyke GH, Waser NM (1981) The production of dilute nectars by hummingbird and
honeyeater flowers. Biotropica 13:260-270
Rice WR (1989) Analyzing tables of statistical tests. Evolution 43:223-225
Roberts WM (1996) Hummingbirds’ nectar concentration preferences at low volume: the
importance of time scale. Anim Behav 52:361-370
117
Roces F, York W, von Helversen O (1993) Nectar concentration preference and water
balance in a flower visiting bat, Glossophaga soricina antillarum. In: Barthlott W et al.
(eds) Animal-plant interactions in tropical environments. Bonn: Museum Koenig
Rodríguez-Peña N, Stoner KE, Schondube JE, Ayala-Berdón J, Flores-Ortiz CM Martínez
del Rio C (2007) Effects of sugar composition and concentration on food selection by
Saussure’s long-nosed bat (Leptonycteris curasoae) and the long-tongued bat (Glossophaga
soricina). J Mammal 88:1466-1474
Sazima M, Vogel S, do Prado AL, de Oliveira DM, Franz G, Sazima I (2001) The sweet
jelly of Combretum lanceolatum flowers (Combretaceae): a cornucopia resource for bird
pollinators in the Pantanal, western Brazil. Plant Sys Evol 227:195-208
Schaefer HM, Schmidt V, Bairlein F (2003) Discrimination abilities for nutrients: which
difference matters for choosy birds and why? Anim Behav 65:531-541
Schlamowitz R, Hainsworth FR, Wolf LL (1976) On the tongues of sunbirds. Condor
78:104-107
Schondube JE, Martínez del Rio C (2003) Concentration-dependent preferences in nectarfeeding birds: mechanisms and consequences. Funct Ecol 17:445-453
Skead CJ (1967) Sunbirds of southern Africa. Cape & Transvaal Printers Ltd. Cape Town,
South Africa
118
Sokal RR, Rohlf FJ (1995) Biometry. WH Freeman & Co., NY
Stiles FG (1976) Taste preferences, colour preferences, and flower choice in hummingbirds.
Condor 78:10-26
Stromberg MR, Johnsen PB (1990) Hummingbird sweetness preferences: taste or viscosity?
Condor 92:606-612
Tamm S, Gass CL (1986) Energy intake rates and nectar concentration preferences by
hummingbirds. Oecologia 70:20-23
Tezze AA, Farina WM (1999) Trophallaxis in the honeybee, Apis mellifera: the interaction
between viscosity and sucrose concentration of the transferred solution. Anim Behav
57:1319-1326
van Tets IG, Nicolson SW (2000) Pollen and nitrogen requirements of the lesser doublecollared sunbird. Auk 117:826-830
van Wyk BE, Nicolson SW (1995) Xylose is a major nectar sugar in Protea and Faurea. S
Afr J Sci 91:151-153
Vogel S (1994) Life in moving fluids: the physical biology of flow. 2nd edn. Princeton
University Press
119
Weast RC (1980) CRC handbook of chemistry and physics, 60th edn. CRC Press, Inc.
Florida USA
120
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement