The Auk 109(4):722-730, 1992
HOW
DO
FRUGIVORES
TRANSIT
AND
PROCESS
GLUCOSE
FRUIT?
GASTROINTESTINAL
ABSORPTION
IN CEDAR
WAXWINGS (BOMBYCILLA CEDRORUM)
DOUGLASJ. LEVEY• AND GARY E. DUKE2
1Department
of Zoology,Universityof Florida,Gainesville,
Florida32611,USA; and
2Department
of VeterinaryBiology,1988FitchAvenue,Universityof Minnesota,
St. Paul, Minnesota 55108, USA
ABSTRACT.--We
used image-intensificationradiology to examine gut function in six Cedar
Waxwings (Bombycilla
cedrorum)fed artificial fruits that containedbarium-labeledpulp and
seeds.Despite the absenceof an anatomicalcrop, waxwings were able to storefruits orad to
the gizzardin a distensibleportion of the esophagus.
By doing so,they were able to increase
meal sizesabovewhat their gizzardscould containat any one time. Thus,they couldpartially
overcomethe digestivebottleneckthat resultsfrom a bulky diet. Gizzardcontractionincreased
in frequencywith gizzard fullness.Seedsand pulp were separatedin the gizzard; mostpulp
passed into the intestine before the seeds. Once in the intestine, however, seeds moved
especiallyquickly (g = 0.14cm/s) into the rectum.Bothpulp and seedsresidedfor a relatively
long time in the rectum, where antiperistalsis was clearly visible. Unlike other birds that
havebeenstudied,the functionsof antiperistalsisin waxwingsprobablydo not includececal
filling, sincececaof waxwingsare minute. We hypothesizethat the function of antiperistalsis
in waxwings (and other fruit-eating birds) is to mix digestain the intraluminal zone adjacent
to the rectal mucosa,thereby increasingnutrient absorption.This hypothesisis consistent
with the observationthat fruit-eating birds typically have low digestiveefficiencies.In particular, absorptionof nutrients in the rectum may be important, given that digestatherein
may still contain significant amounts of nutrients. Indeed, active uptake of D-glucose,a
primary constituentof fruit pulp, was as high in the rectumas elsewherein the gut. We also
found that waxwings typically defecatedonly the distal 50%of rectal contents.Presumably,
digestain the proximal portion was higher in nutrients and so was retained longer for more
thorough absorption. This pattern contraststo most other vertebrates in which the rectum
emptiescompletelyand sugaruptake in the rectum is negligible. We suggestthat functional
adaptationsto frugivory (i.e. how fruit is processed)are likely more important than structural
adaptations.Received
9 May 1991,accepted
15 December
1991.
in fruit-eating birds. Most studieshave simply
treated birds as "black boxes"--patternsof ina nutritious,easilydigestiblecovering.Presum- gestionand defecationare compared,then tranably, fruit-eating birds possess
gut adaptations sit and digestive processesare inferred (Johnthat allow them to efficiently processthis mix sonet al. 1985,Bairlein1987,Worthington1989,
of high and low digestibilitycomponents.
Many Levey and Grajal 1991).For example,retention
studieshave attemptedto find suchadaptations, time is one of the most important parameters
but there is little agreement among them. For of food processing(Van Soest1982,Penry and
example,someresearchers
reportthat fruit-eat- Jumars 1987). The statistic that is usually reing birdshave especiallyshortguts(e.g.Wals- ported is the mean time from ingestion to defberg 1975, Pullainen et al. 1981), others that ecation. Yet, what is needed for accurate intertheyhavelongguts(e.g.A1-Dabbagh
et al. 1987, pretation is the time a meal spends in each
Jordano 1987), and yet others that they have segmentof the gut (Penry and Jumars1987,
guts of normal length (Herrera 1984).Appar- Martinez del Rio and Karasov1990).This type
ently, there are few (if any) generalmorpho- of information can yield new insights.In the
logicaladaptationsof avian guts to frugivory. case of hummingbirds, for example, if cropThe major adaptationsmay be functional,re- emptying time determines foraging bout frelating to how fruit is processedrather than to quency, these birds would likely be energy
gut structure(Herrera1984,KarasovandLevey maximizersrather than foraging-timeminimiz-
FRUITS
AREa unique foodsourcethat combine
a largeproportionof well-protectedseedswith
ers (Diamond et al. 1986, Karasov et al. 1986).
1990).
Little is known about functioning of the gut
722
Image-intensificationradiology (IIR) allows
October1992]
Gastrointestinal
Transit
in CedarWaxwings
continuousviewing of barium-labeledmealsas
they move through the digestive tract. Although this techniquehas been used to study
gastricmotility in severalordersof birds (Dziuk
and Duke 1972, Rhoades and Duke 1977, Duke
et al. 1989), it has only been applied once to a
passerine(Duke 1989a)and has not been used
on a frugivore.Here we useIIR to examinegut
processingof fruits by Cedar Waxwings(Bombycillacedrorum),one of the most heavily frugivorous speciesin North America (Martin et
al. 1951).Our goalwas to relategut function to
patternsof fruit digestion.More specifically,
we
wanted to determine whether pulp and seeds
are treated differently in the gut, quantify retention timesin the gizzardand intestine,measure gizzard contractionfrequency,and determine the presence, timing, and location of
intestinalantiperistalsis.
We alsousedan in vitro
technique to measureintestinal and rectal uptakesof D-glucose,a primaryconstituentof fruit
pulp.
METHODS
Five CedarWaxwingswere capturedwith mist nets
in Alachua County, Florida and held for approximately I0 months on a synthetic fruit-based diet
(Denslow et al. 1987) before this study began. All
regainedtheir initial body massand maintainedgood
health. Birdswere housedindividually in cages(approximately 50 x 50 x 50 cm) at a constanttemperature (21øC)and photoperiod(12 h dark, 12 h light).
Image-intensification
radiology.--We observed food
passagevia IIR on a Technomed1250 fiuoroscopy
machine.During trials, eachbird was held in a small
cardboard box (30 x 20 x 15 cm) with a translucent
plastictop to allow illumination.The bird usuallysat
on the box's single perch, thus giving a consistent
lateral view when the box wasplacedwith its perch
parallel to the x-ray beam.
To trackfoodthroughthe gastrointestinal
tract,we
usedartificialfruits (6-mmdiameter)composedof 2%
agar, 15%glucose,10%barium sulfate,"seeds"(4-mm
plasticbeads),redfoodcoloring,andwater.The"pulp"
and "seeds"of theseartificialfruitsarepassedthrough
the gastrointestinaltract of captivewaxwingsat rates
similar to thosefound in wild birds eating natural
fruits (Levey and Grajal 1991).Barium sulfate,which
is radio-opaque,wasalsopackedinto the bore of each
beadand held in placewith a drop of superglue.This
made the seedseasily visible in the gut.
Each day, within 30 min after "dawn" in the holding room, a waxwing was fed three to four fruits
without barium sulfate or seeds, enough to fill its
foregutand initiate digestiveprocesses.
It was then
placed in the box and taken to the radiology area,
723
where approximately 15 min later it was force-feda
barium-labeled
two fruits with
meal of one fruit without
seeds. The seedless fruit
a seed and
had much
morebarium-containing
pulp than the seededfruits
and served to coat the gut with barium for easier
viewing. We followedthe mealthroughthe gut until
most barium had been excreted or until the equipment was needed for clinical cases. At least three trials
were completedfor each individual. We timed gizzard contractionsand rectal antiperistalsis,estimated
percentof pulp in the gizzard,intestine,and rectum
at 1-min intervals, recordedpassagetimes of seeds
throughthe gizzardand intestine,and estimatedpercent of rectum emptied by each defecation.
Mean retention time of pulp in the gizzard was
calculatedby multiplying the proportionof the meal
leaving the gizzard during each l-min interval by
elapsedtime and then summing theseproductsover
all intervals (see Castle 1956, Warner 1981).
Becauseour birds were not feeding continually,
their barium-labeled meal probably resided in the
small intestine and rectum longer than if another
bolusof food had been immediatelybehind it. Consequently,we presentresidencetimes of pulp in the
rectum only up to 25 min postingestion,which is
approximately the retention time of the artificial fruit
pulp in freely-feedingwaxwings(Leveyand Grajal
1991).Even though pulp and seedsoften residedin
the intestine for longer than 25 min, we feel that
reporting such times would be misleading.Instead,
we focuson processing
time in the gizzard.Thismeasurementwasprobablylessaffectedby the short-term
fast,becausethe rest of the gut was full at the time,
andgizzardemptyingappearedto bea necessary
precursorfor ingestinganother meal (D. Levey and G.
Duke pers.observ.;alsoseeWorthington 1989).
Exceptwhere noted we report statistics(average
and standard deviations) calculated from mean values
of each individual (i.e. n = 5 in most cases).
Glucose
uptake.--Wemeasureduptakeof D-glucose
acrossthe brush-border membrane (not transmural
flux) of four waxwings.Methodologyis detailedin
Karasovand Diamond (1983).In brief, the digestive
tractfrom gizzardto vent wasremovedfrom an anesthetizedbird and evertedin ice-coldRingersolution
over
a 4-mm
diameter
rod.
The
1-cm sleeves
were
then mounted individually on 4- or 5-mm diameter
steel rods and, after a 5-min preincubationin 37øC
Ringer,were verticallysuspendedseveralmillimeters
above a stir bar (1,200 rpm) for 30 s in a 50 mM
D-glucosesolutioncontainingtracer concentrations
of 3H D-glucoseand •4CL-glucose.(Justificationof
concentration
andincubationtime isprovidedin Karasovand Levey 1990.)Tissueswere then blotted on
tissuepaper, removed from the rod, weighed, solubilized, and countedfor disintegrationsper minute.
In calculationsof uptakerate,we correctedfor D-glucoseabsorbedpassivelyand in adherentmucosalfluid
by usinguptakeof L-glucose,a stereoisomer
of D-glu-
724
LEVE¾
^NDDYKE
[Auk,Vol. 109
16
•' 14
ß
[]
• 12
•
10
ß
õ
•
•
...
Thick
muscle
'•-•,•/•
•C ]•
ß
8
ß
• 2 ß
m ßß
ß•
o
0
ß
,- ......
, - , . , ........
010
20
30
40
50
60
70
80
90
100
Percent
of Meal
in Gizzard
2,Cizza•d
co•t•act•o•
•e•ue•cy
as
o•Fi•,
pe•ce•t
o[ath•ee-•uJt
meal
J••Jzza•d
(r•u•ctio•
•: 0.•3,
• < 0,05; data •om all bi•ds combined),
cosethat is •ot actively t•a•spo•ted (see•a•aso• a•d
Diamond Z983).Uptake •a•ues were
cm o• i•test•al length, We were able to mou•t a•d
measureuptakesJ• o•e to two sleeves•om
testJ•e,distal i•testi•e, a•d •ectum (postceca),
RESULTS
Generalmorphology.--Theentire gastrointestinal tract weighed approximately 11%of total
body mass.The esophaguswas distensible,but
lacked a crop. A small but obvious proventriculus led into the gizzard (muscularstomach),
which was comprisedof thin and thick muscle
pairs, as describedin other birds (Fig. 1; Duke
1989b).The duodenumwasapproximatelytwice
the diameter of the proximal ileum. Rudimentary cecawere found upon closeexamination.
Image-intensification
radiology.--Despitethe
absenceof an anatomicalcrop,fruits were often
storedin the distensibleregion of the esophagus before entering the gizzard. In 54% of the
trials, all seedsmoved directly into the gizzard
within 3 min of feeding. In the remaining trials,
fruits were retained in the esophagusfor 27 +
9 min. Typically, two or three fruits moved into
the gizzard at approximatelythe sametime, after it had mostlyemptied the previousmeal and
o
1
2
I
i
I
at the end of the thick-muscle
contraction.
Fruits
always passedrapidly through the proventriccm
ulus. Almost immediately upon entering the
gizzard, pulp and seeds were separated and
Fig. 1. Gastrointestinaltracts
of long-term-captive thoroughly mixed by the alternating contrac-
waxwing (left) and wild waxwing (right). Note larger tions of the thin and thick muscles. Contraction
gizzard and longer intestinesin wild individual, and frequency was initially high, averaging 7.9 +
the largeduodenumin both. Drawn from specimens. 2.3 contractions/min, and contraction rates were
October
1992]
Gastrointestinal
Transit
inCedar
Waxwings
725
14
12
•-
8
O'
6
U_
4
2
5
1•
1•
20
25
0
o
•o •o .o o
•o •o •o •o oo
Percent of Rectum Emptied
Fig. 4. Frequency distribution of defecationsthat
emptied specified percent of materia! from rectum.
Data from
5
10
15
20
combined.
intestine;n = 25). Becausethe seedsnearly filled
the lumen of the intestine, any pulp in the intestine was pushedahead of them.
Despite the large size of the duodenum, we
rarely saw pulp there becauseit moved so rapidly. Also, it was often hard to distinguish the
duodenumfrom the adjacentgizzard. Duodenal
o
o
all birds
25
1oo-
E
80-
refluxes (Duke 1989b) were rare; we saw two in
n"'
30 h of observation.
.__.
60-
Peristalsis
was evident
in
the ileum and, although pulp generally moved
through it at a steady pace,we frequently ob-
40-
servedextremelyrapid orador aboradflow over
1- to 2-cm segmentsof intestine. Pulp rapidly
20'
accumulated
tained
0
0
Time (rain)
Fig. 3. Percent of single meal remaining in (A)
gizzard, (B) intestine, and (C) rectum as function of
time since ingestion. Bars represent SD. Note relatively small percentin intestine at any one time, and
rapid accumulationin rectum.
in the rectum, which
30 to 40% of the meal within
often con10 min after
feeding (Fig. 3). Once the rectum was more than
approximately 30% full, antiperistaltic waves
becameobvious,startingat the cloacaand usually traveling the full length of rectum to the
ileocecalrectaljunction. Contractionfrequency
was 18 + 5 per min and was not significantly
correlated with rectal fullness (r2 = 0.12, P >
0.5). Defecationswere precededby a gradual
packingof pulp in the cloacaand distal rectum.
positively correlatedwith gizzard fullness(Fig.
Nevertheless, defecation volume was often small
2).
of pulp in the gizzard was 7.7 + 2.2 min (Fig.
3). In contrast, seedsremained in the gizzard
comparedto the amount of pulp in the rectum.
Usually only 50% of the rectum was emptied,
and often much less(Fig. 4). We never observed
the rectum empty entirely in one defecation.
Following defecationthe bird would usually
for œ = 27
become
Gizzard contents emptied at the end of the
thin-muscle
contraction.
+
12 min.
Mean
Once
retention
seeds
exited
time
the
more
active
than
it had been
for the
gizzard, however, they moved extremely rapidly through the intestineat approximately0.14
previous5 to 10 min. Often a fruit left the "crop"
or a seed left the gizzard within 1 min after a
cm/s (or 155 + 40 s to transverse the entire
defecation,presumablyfreeing the foregut for
726
LEVE¾
ANDDUKE
[Auk,Vol. 109
TA13Lœ
1. Gutdimensions
andmasses
(+SD) of waxwingsheldin captivity(n = 2) for 10monthsor collected
from field (window-killed birds; n = 9).
Intestine
Gizzard
Length
(cm)
Mass
(g)
Diameter
(cm)
Mass
(g)
Captive birds
14.5 _+0.5
0.32 _+0.09
0.85 _+0.05
0.14 + 0.02
Wild birds
18.6 _+ 2.1
0.27 _+ 0.09
1.32 + 0.10
0.43 + 0.14
another meal. Our sample sizes,however, were
too small
to determine
whether
these actions
were significantlyassociated
with defecation.
Becausethe waxwingsusedin this studyprocessedourartificialfruitsmoreslowlythansimilarly treated waxwings that had not been in
captivity as long (see Levey and Grajal 1991),
we becameconcernedthat the pattern of gut
function describedabovemight not be typical
of wild individuals.We capturedoneadditional
waxwing, which adjustedto captivity within
five days. We then ran an identical seriesof
trials on it. Mean retention times of pulp and
phology associatedwith the 10-month period
of captivity,we sacrificedtwo of our long-term
captivesand measuredgizzard diameter, intestine length, and dry massof both organs.We
then comparedthese measurementsto those
collectedfrom eight window-killed waxwings
(Table 1). Intestine length of the captive waxwings was approximately 22% shorter than in
the window-killed individuals(Mann-Whitney
U = 17.5,P = 0.042),although intestinedry mass
of the two groupswas not significantlydifferent. Gizzard diameter and dry masswere significantly smaller in the captive individuals
seeds in the gizzard indeed were shorter for (Mann-Whitney U = 18, P < 0.05 and U = 16,
this individual than for the others (4.4 vs. 7.7 P < 0.05, respectively).
min for pulp and 6.8 vs. 27 for seeds),although
Glucose
uptake.--Activeuptake rate of D-glunot statisticallydifferent (valuesfor new bird cose in the rectum was relatively high comfell within 95% but outside 85% confidence inpared to rates in the midintestine and distal
tervals of other five birds). We stressthat the intestine;uptakeratesdid not vary significantly
overall pattern of processingwas essentially from midgut to rectum (F2,n= 0.5; P = 0.39;
identical betweenthe two groups.The only no- repeated-measures
ANOVA). We doubtthat this
ticeabledifferencewasthat in the recentlycap- result is an artifact of low sample size (n = 4
tured individual, seedsleft the gizzard ahead birds)becauseno trendsin uptakerateswithin
of much of the pulp.
or among birds were apparent (Fig. 5).
To examine possible changes in gut morDISCUSSION
200'
150'
T
100'
0
Midgut
Hindgut
Rectum
Position
Fig. 5. Carrier-mediatedD-glucoseuptake at 50
mM in everted sleevesof waxwing gut (n = 4 birds).
Storing
fruitsandprocess-rate
limitation.--Many
fruit-eatingvertebratesappearprocess-rate
limited (i.e. their rate of ingestionis limited by the
rateat which their gutscanprocessthe previous
meal; Sorensen 1984, Tedman and Hall 1985,
Worthington1989,Leveyand Grajal1991).One
way to alleviatethisconstraintis to increasegut
size (Sibly 1981). However, volant frugivores
generally do not have large gut volumes,presumably becauseof the associatedincreasein
the cost of flight that such large guts would
entail. A surprisingobservationin the present
studywasthat, althoughwaxwingshave short,
simple guts with no anatomicallydistinguishable crop, they nonethelesshave functional
crops.By storing ingested fruits, their distensible esophagiallow them to consumemore
October1992]
Gastrointestinal
Transit
in CedarWaxwings
fruits per feedingbout than their gizzardsand
intestinescan processat any one time. In short,
waxwings appearto offsetthe problem of process-ratelimitation by storing fruits orad to the
gizzard. Phainopeplas(Phainopepla
nitens),an-
727
tion of its gizzard capacity.We also observed
that gizzardcontractionfrequencywas higher
in thisbird. In bothchickens(Gallus
gallus;Roche
and Decerprit 1977) and turkeys (Meleagrisgallopavo;
Dukeet al. 1975,Duke 1986)the motility
other highly frugivorousspecies,apparently of the gizzardappearsto be importantin regstore fruit in a similar way (Walsberg 1975). ulating motility of the smallintestine,cropand
More generally, these results demonstrate the
esophagus.If such regulation occursin waxdangerof the simpleand often dogmaticnotion wings,then their higher gastriccontractionfrethat function dependson structure;waxwings quency would also result in higher intestinal
do not require a specialstorageorgan to store contractionfrequenciesand a more rapid pasfruits.
Note that the mean retention time of pulp
from small meals(i.e. mealsin which fruits pass
directlyto the gizzard)is approximately25 min
(Leveyand Grajal 1991).This agreesquite closely with the averagetime fruits from largermeals
are stored in the "crop" (g = 27 min). Thus, it
appearsthe distensibleesophagusallows waxwingsto ingestapproximatelytwo mealsof fruit
in a single foraging trip.
Gut rnorphology.--Waxwings
have a shortgut
with a relativelywide lumen,a typeof gut found
in other frugivores(Berndtand Meise 1959,in
Cvitanic 1967). However, the structure of our
waxwings' gastrointestinaltract differed from
that reportedfor waxwingsby Walsberg(1975).
In particular, the duodenum of our birds (both
captive and window-killed individuals) was
distinctly larger than the rest of the intestine,
whereasWalsberg'sspecimensshowedno such
contrast.This differencemay be due to regional
and seasonal variation in diet. Intraspecific
changesin gut structureare often correlated
with dietary shifts,which are usually seasonal
(Ziswiler and Farner 1972,Miller 1975,Savory
and Gentle 1976, AI-Dabbagh et al. 1987). Indeed, the gut morphology of wild individuals
and our long-term captiveswas conspicuously
different (Table 1). The smaller guts of the captive birds may have been due to their unusual
diet, which was relatively low in bulk and wellbalancednutritionally. Birdson poorer-quality
dietsare known to increasegut volume (Savory
and Gentle 1976, AI-Joborae 1980).
The differencein gut structurebetween the
long-term captivesand the newly capturedindividual may explain their differencesin fruit
sage rate of contents.
The longer retentiontimesof our five longterm captivescomparedto the newly captured
individual clearly illustrate that rate of gut passageneed not be correlatedwith gut length. In
particular, the common assumptionthat short
guts promote rapid passage(e.g. Ziswiler and
Farner 1972) is not supported. An individual
waxwingwith a long gut processed
fruits much
more quickly than did five individuals with
shorterguts.
The lack of food processingin the proventriculus and passageof food into the gizzard at
the end of the thick-musclecontractionapparently are typical of other bird species(Dziuk
and Duke 1972, Rhoades and Duke 1977). Gizzard-contraction frequencies, however, were
much higher in waxwingsthan in theseother
species (Duke et al. 1975, Kostuch and Duke
1975, Roche and Decerprit 1977).
Rectalfunctionin waxwings.--Rectalantiperi-
staltic contractionfrequency is approximately
the samein waxwingsand domesticturkeys,14
to 18/min (Lai and Duke 1978). Previously, the
function of rectal antiperistalsiswas assumed
to be for urinary refluxand cecalfilling. In waxwings,the latter functionclearly doesnot occur,
becausethe cecaeare extremelysmall and presumably nonfunctional.We suggestan alternative explanation for antiperistalsisin waxwings--enhancementof nutrient absorption.
The intensive mixing of rectal contents
throughantiperistalsis
suggests
that the rectum
is active in nutrient absorption. Karasov and
Levey (1990)reportedthat frugivorousbirds in
general and waxwings in particular appear
atypical in their high ratesof glucoseabsorp-
processing.The latter bird passedboth the seeds tion in the distal third of the small intestine.
and pulp much more rapidly than did the long- Here, we extendedthis result by documenting
term captives.We speculatethat the larger gizzard of the newly captured individual could
processthe barium-labeledmeal fasterbecause
the meal occupieda relatively smaller propor-
high ratesof glucoseabsorptionin the rectum
of waxwings. In another frugivorousspecies,
the AmericanRobin (Turdusrnigratorius),
glucoseuptake alsooccursin the rectum(D. Levey
728
LEvU•ANDDUNCE
and W. Karasovunpubl. data). In addition, enzymaticactivity suggestsnutrient uptakein the
rectumof other frugivores(Ogunbiyi and Okon
1976). With the exception of hindgut fermenters,relatively high ratesof nutrient absorption
and enzymatic activity in the rectum of vertebratesis unusual.Typically, water and electrolytes are the principal substancesabsorbedrec-
[Auk, Vol. 109
seedsmay appear to passmore quickly than
pulp if the pulp that follows the last seed is
mixed with unlabeled pulp from the next meal,
which is likely given the high ratesof antiperistalsisand incompleteemptyingof the rectum.
This mixing would generatea long tail on the
plot of marker excretionversustime and, there-
by, inflate meanretentiontime of pulp relative
tally (Ruckebuschet al. 1991). In frugivores, to seeds, since seeds cannot be mixed in a simrectal absorption of sugar can be explained by
the low digestiveefficiencyof pulp, which has
such a short residence time in the intestine
that
it is poorly assimilated(Levey and Karasov1989,
Karasovand Levey 1990,Karasov1990).Thus,
pulp in the rectum may still contain relatively
ilar fashion. The different conclusionsof Levey
and Grajal (1991) and this study provide a good
example of the danger of examining digestive
processesonly on the basis of ingestion and
excretion.
high levels of sugars,which are thought to induce nutrient transport activity (Diamond and
Karasov 1987, Karasov and Diamond 1987; but
seeLevey and Karasov1992).
Nutrient absorptionin the rectum could also
help explain another distinctive feature of gut
processingin CedarWaxwings.Unlike turkeys
and chickens(Duke 1986),which void virtually
ACKNOWLEDGMENTS
We thank Norm Ackermanand the Collegeof Veterinary Medicine at the University of Florida for use
of the radiologyequipment.The radiologystaffhelped
with setup and Kris Brugger helped with data collection.
We thank
Bill Karasov
for use of his labora-
defecation,CedarWaxwingstypically void only
the distal 50%.In most other species,relatively
little absorptiontakesplacein the colon (except
for water and polyvalent ions), presumablybe-
tory to measurenutrient uptakes, and Bruce Darken,
Danny Afik, and C. Martinez del Rio for assistance
with the measurements.Daryl Harrison prepared the
illustrations. Lee Gass and an anonymous reviewer
made valuable suggestionsfor improving the manuscript. The project was sponsoredby the University
of Florida, Department of Zoology and Division of
cause the contents
SponsoredResearch,the University of Minnesota
the entire
contents
of their
rectum
have little
with
nutritional
each
value
to the bird. In frugivores, however, the pulp in
the colon still contains a relatively high proportion of nutrients. We suggestthat only the
lower portion of the rectum is voided so that
nutrient absorption can continue on the pulp
in the upper portion, which is probably more
nutrient dense.A longer retention time of pulp
in the
rectum
also would
aid
in water
reab-
sorption,an importantbenefitto waxwings,who
are sometimesin negative water balance (Studier et al. 1988).
GraduateSchool,and NSF grant BSR9020911to D.J.L.
LITERATURE CITED
AL-DABBAGH, K. Y., J. H. JIAD, AND I. N. WAHEED.
1987. The influenceof diet on the intestinelength
of the White-cheeked
Bulbul.
Ornis
Scand.
18:
150-152.
AI•-JoI3ORAE,
F.F.
1980. The influence of diet on the
gut morphology of the starling (Sturnusvulgaris
L. 1758). Ph.D. dissertation, Univ. Oxford, Oxford.
Separation
ofpulpandseeds.--Leveyand Grajal BAIRLEIN,F. 1987. Nutritional requirements for
(1991) reported that waxwings internally sepmaintenanceof body weight and fat deposition
in the long-distancemigratory Garden Warbler,
arate pulp and seeds,and defecate the seeds
Sylviaborin(Boddaert).Comp. Blochem.Physiol.
beforepulp. We found that, althoughwaxwings
A Comp. Physiol. 86A:337-347.
separatedthe two componentsof fruit, they did
not defecatethe seedsfirst. Most pulp seemed CASTLE,E.J. 1956. The rate of passageof foodstuffs
to arrive in the rectum before seeds and then,
through the alimentary tract of the goat. Br. J.
Nutr.
10:15-23.
typically, both were defecated together. We CVlTANIC,A. 1967. The relationships between insuggestthe especiallyshort retention times of
testineand bodylength and the nutrition of some
seedsreportedby Levey and Grajalwere likely
bird species.Larus (Engl. Transl.) 21:181-190.
a statisticalartifact causedby comparinga pulp DENSLOW,J. S., D. J. LEVEY,T. C. MOERMOND, AND B.
marker that generatesa continuousdistribution
C. WENTWORTH.
1987. A syntheticdiet for fruiteating birds. Wilson Bull. 99:131-134.
(phenol red) with a discretemarker (the seeds).
Even if pulp and seedspassat the same rate, DIAMOND, J. M., W. H. KARASOV,D. PHAN, AND F. L.
October
1992]
Gastrointestinal
Transit
inCedar
Waxwings
CARPENTER. 1986. Hummingbird digestive
physiology,a determinant of foraging bout frequency. Nature (Lond.) 320:62-63.
DIAMOND,J. M., ANDW. •-[. KARASOv.1987. Adaptive
regulationof intestinalnutrient transporters.Proc.
Natl.
Acad.
Sci. USA
84:2242-2245.
729
LAI, H. C., AND G. E. DUKE. 1978. Colonic motility
in domesticturkeys.Am. J. Dig. Dis. 23:673-681.
LEvEY,D. J., AND A. GRAJAL.1991. Evolutionary im-
plicationsof fruit processingand intake limitation in Cedar Waxwings.Am. Nat. 137:171-189.
LEVEY,D. J., AND W. H. KARASOV.1989. Digestive
responsesof temperatebirds switchedto fruit or
DUKE,G.E. 1986. Alimentary canal: Anatomy, reginsect diets. Auk 106:675-686.
ulation of feeding and motility. Pages269-288 in
Avian physiology (P. D. Sturkie, Ed.). Springer- LEVEY,D. J., AND W. •-[. KARASOV.1992. Digestive
Verlag, New York.
modulationin a seasonalfrugivore,the American
DUKE,G.E. 1989a. Avian cecaland colonic motility
Robin (Turdusmigratorius).Am. J. Physiol. 262:
G711-G718.
and the relationship of colonicantiperistalsisto
diet, habitatand cecalanatomyin severalspecies. MARTIN, A. C., I-I. S. ZIM, AND A. L. NELSON. 1951.
Proc.First Int. Sympos.on Avian Cecal Function.
Americanwildlife and plants:A guideto wildlife
food habits. Dover Publications, New York.
J. Exp. Zool. Suppl. 3:38-47.
DUKE,G. E. 1989b. Avian gastrointestinalmotor MARTINEZ DEL RIO, C., AND W. I'-I. KARASOV. 1990.
Digestion strategies in nectar- and fruit-eating
function. Pages1283-1300in Handbook of physbirdsand the sugarcompositionof plant rewards.
iology-The gastrointestinalsystem,vol. 1(2) (J.
T. Wood, Ed.). Oxford Univ. Press, New York.
DUKE, G. E., T. E. KOSTUCH,AND O. A. EVANSON. 1975.
Gastro-duodenal
electricalactivityin turkeys.Am.
J. Dig. Dis. 20:1047-1058.
DUKE,G. E., A. R. PLACE,AND B. JONES. 1989. Gastric
emptyingand gastrointestinalmotility in Leach's
Storm-Petrel chicks (Oceanodroma
leuchorhoa).
Auk
106:80-85.
DZIUK,•-[. E., ANDG. E. DUKE. 1972. Cineradiographic studiesof gastricmotility in the Turkey. Am.
J. Physiol. 222:159-166.
HERRERA,
C. M. 1984. Adaptation to frugivory of
Mediterranean avian seeddispersers.Ecology65:
609-617.
JOHNSON,R. A., M. F. WILLSON, J. N. THOMPSON,AND
R. I. BERTIN.
1985.
Nutritional
values
of wild
fruit and consumptionby migrant frugivorous
birds.Ecology66:819-827.
JORDANO,
P. 1987. Frugivory, external morphology
and digestive system in Mediterranean sylviid
warblers Sylviaspp. Ibis 129:175-189.
KARASOv,
W. iuI. 1990. Digestion in birds: Chemical
and physiological determinants and ecological
implications. Stud. Avian Biol. 13:391-415.
KARASOV,
W. iUI.,ANDJ. M. DIAMOND. 1983. Asimple
method for measuringintestinal soluteuptake in
vitro. J. Comp. Physiol. 152:105-116.
KARASOv,
W. H., ANDJ.M. DIAMOND. 1987. Adaption
of intestinal nutrient transport.Pages1489-1496
in Physiology of the gastrointestinal tract (L. R.
Johnson, Ed.). Raven Press, New York.
KARASOv,W. iuI.,AND D. J. LEvEY. 1990. Digestive
systemtrade-offsand adaptationsof frugivorous
passerinebirds. Physiol. Zool. 63:1248-1270.
KARAsov, W. H., D. PHAN, J. M. DIAMOND, AND F. L.
CARPENTER.1986. Food passageand intestinal
nutrient absorptionin hummingbirds.Auk 103:
453 -464.
KOSTUCH, T. E., AND G. E. DUKE. 1975. Gastric mo-
tility in Great-horned Owls. Comp. Biochem.
Physiol.A Comp. Physiol. 51A:201-205.
Am. Nat.
136:618-656.
MILLER,M.R. 1975. Gut morphologyof Mallards in
relation to diet quality. J.Wildl. Manage.39:168173.
OGUNBIYI, O. A., AND E. E. OKON. 1976. Studies on
the digestiveenzymesof the African fruit bat
Eidolon
helvum.Comp.Biochem.Physiol.A Comp.
Physiol. 55A:359-361.
PENRY,D. L., AND P. g. JUMARS.1987. Modeling
animal guts as chemicalreactors.Am. Nat. 129:
69-96.
PULLAINEN, E., P. I-IELLE, AND P. TUNKKARI. 1981.
Adaptive radiation of the digestivesystem,heart
and wings of Turduspilaris,Bombycilla
garrulus,
Sturnusvulgaris,Pyrrhulapyrrhula,Pinicolaenucleator,and Loxiapityopsittacus.
Ornis Fenn. 58:2128.
RHOADES,D. D., AND G. E. DUKE. 1977. Cineradio-
graphicstudiesof gastricmotility in Great-horned
Owls (Bubovirginianus).
Condor 79:328-334.
ROCHE, M., AND J. DECERPRIT. 1977. Contrfiles hormonal
et nerveux
de la motricit6
du tractus di-
gestif de la poule. Ann. Rech. Vet. 8:25-40.
RUCKEBUSCH,Y., L. PHANEUF, AND R. DUNLOP. 1991.
Physiology of small and large animals. B.C.
Decker, Philadelphia, Pennsylvania.
SAVORY,
C. J., AND M. J. GENTLE.1976. Changesin
food intake and gut size in JapaneseQuail in
responseto manipulation of dietary fiber content.
Br. Poult.
Sci. 17:571-580.
SIBLY,R.M. 1981. Strategiesof digestionand defecation. Pages 109-139 in Physiological ecology:
An evolutionary approach to resourceuse (C. R.
Townsend and P. Calow, Eds.). Sinauer, Sunderland, Massachusetts.
$ORENSEN,
A.E. 1984. Nutrition, energyand passage
time: Experiments with fruit preference in European Blackbirds(Turdusmerula).J. Anita. Ecol.
53:545-557.
$TUDIER,E. H., E. J. $ZUCH, T. M. TOMPKINS,AND V.
W. COPE. 1988. Nutritional budgetsin free fly-
730
LEVEY
ANDDUKE
[Auk, Vol. 109
ing birds: Cedar Waxwings (Bombycilla
cedrorum) WARNER,A. C. I. 1981. Rate of passageof digesta
feeding on Washington hawthorn fruit (Crataegus
through the gut of mammalsand birds.Nutrition
Abstr. Rev. 51(B):789-820.
phaenopyrum).
Comp. Biochem.Physiol.A Comp.
WORTHINGTON,A. H. 1989. Adaptations for avian
Physiol. 89A:471-474.
TEDMAN, R. A., AND L. $. HALL. 1985. The morfrugivory: Assimilationefficiencyand gut transit
phology of the gastrointestinaltract and food
time of Manacusvitellinusand Pipramentalis.
Oecotransit times in the fruit batsPteropusalectoand
1ogia80:381-389.
V., ANDD. $. FARNER.1972. Digestion and
P. poliocephalus
(Megachiroptera).Aust. J. Zool. ZISWILER,
33:625-640.
the digestivesystem.Pages343-430 in Avian biVAN $OEST,P.J. 1982. Nutritional ecology of the
ology,vol. 2 (D. S. Farner and J. R. King, Eds.).
Academic Press, New York.
ruminant. O & B Books,Corvallis, Oregon.
WALSBERG,
G. E. 1975. Digestive adaptations of
Phainopepla
nitensassociatedwith the eating of
mistletoe
berries.
Condor
77:169-174.