Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
j RaptorRes.30(4):219-223 ¸ 1996 The Raptor ResearchFoundation, Inc. FACTORS INFLUENCING NIGEL THE SIZE OF IN RAPTORS W. BARTON AND DAVID SOME INTERNAL ORGANS C. HOUSTON AppliedOrnithology Unit, DivisionofEvolutionaryand Environmental Biology,GrahamKerr Building, Universityof Glasgow, Glasgow,G12 8Q• Scotland ABSTP•CT.--Thesize of the small intestine, stomach,kidney, liver and heart were compared among raptor speciesand consideredin relation to hunting strategyand body size. Speciesrelying on rapid accelerationand maneuverability to capturepreyin flight,suchassparrowhawk (Accipiter nisus),goshawk (A. gentilis),and peregrine (Falcoperegrinus), had the smallestdigestivetracts for their size. Species dependingmore on soaringflight which do not need fastaccelerationto captureprey,suchascommon buzzard (Buteobuteo),red kite (Milvus milvus), and European kestrel (Falco tinnunculus),had heavy digestiveorgans.In the Strigiformes,the same relationshipwasfound, and specieswhich hunt by active flight, such as barn owl (Tyt0 alba), and 1ong-earedowl (Asiootus),had significantlylighter digestive tracts than the tawny owl (Strix aluco),a specieswhich mainly hunts from a perch dropping onto its prey from above.Body condition waspositivelycorrelatedwith organ weights,including the heart, but to a lesserextent with linear measuresof size such as intestine length. KEYWOe.DS: internalorgans;predatorybirds;condition. Factoresque infiuencian el tamafio de algunos6rganosinternos en rapaces RESUMEN.--EItamafio del intestinodelgado,est6mago,riffones,higado y coraz6n,rue comparadoentre especiesde rapacesque difieren en estrategiade caza y tamafio corporal. Especiesbasadasen r•pida aceleraci6ny maniobrabilidadpara capturarpresasen vuelo, talescomoAccipiter nisus,A. gentilis y Falco peregrinus ten/an el tracto digestivom•s pequefio para su tamafio corporal. Especiesque dependen m•s bien de un vuelo de planeo y que no necesitande una r•pida aceleraci6n,talescomo Buteobuteo, Milvus milvusy Falcotinnunculus,tenlan 6rganosdigestivosm:ispesados.En Strigiformesse encontr6 la misma relaci6n, especiesque cazanpor vuelo activo,tales como Tytoalba,Asiootus,ten/an un tracto digestivo significativamentem•s liviano que Atrix aluco,una especieque caza esperandoa la presa desde una percha.La condici6ncorporalestuvocorrelacionadapositivamentecon el pesode los 6rganos,incluyendo el coraz6n. [Traducci6n de Ivan Lazo] We previouslyreported that birds of prey show considerablevariation in the length of their small intestines (Barton and Houston 1994) and suggestedthat these differencescould be due to adaptationsfor different stylesof predation. We hypothesized that raptors which capture prey by pouncing on it from abovewould have the longest gut lengths for efficient digestion while species which need to acceleraterapidly to capture active prey, such as predatorstaking small prey in flight, would have very light organ systemsto enable them to acceleratemore rapidly. Reductionin the length of the small intestine probablyresultsin less efficient absorption of food but this could still be selected for if this disadvantagewere more than compensatedfor by a higher rate of prey capture. We were able to demonstrate that there was an as- 219 sociation between the length of the small intestine and the predatory strategyshownby a species,and that after correcting for differences in body size, those species that specialized in the capture of small birds in flight had smallintestinallengthsup to 50% shorterthan specieswith a lessactivepredatory style. The small intestine is responsiblefor nutrient absorption and is therefore the organ most likely to effect absorptionefficiency.We have shown that specieswith short small intestines digest their food less efficiently, and as a consequence select only prey items of high energy content (Barton and Houston 1993a, 1993b). This paper considerswhether there may be associations between predatory strategy and the weight of other organ systems.We compared the massof the gizzard, proventriculus,small intestine, 220 N. BARTON AND D. HOUSTON VOL. 30, NO. 4 Table 1. Mean wet weights (_SD) of internal organs of various raptors expressedas a percentage of total body weight. Means are basedon a combinationof the sizesof internal organsof both sexesand samplesizesare given in parentheses. SMALL SPECIES Kestrel Merlin Sparrowhawk Peregrine Buzzard Red Kite Goshawk Long-earedowl Barn owl Short-Pared owl Tawnyowl LWER HEART 3.24 -+ 0.46 (14) 1.13 -+ 0.14 (12) 3.24 -+ 0.84 (3) 2.20 +- 0.31 (3) 2.63 -+ 0.53 (86) 1.12 +_0.14 (75) 2.76 +_0.43 (3) 1.81 _+0.05 (3) 2.15 +_0.47 (5) 0.86 +_0.15 (31) -1.13 _+ 0.26 (7) 1.65 _+0.36 (50) 0.86 -+ 0.09 (44) 2.10 (2) 0.98 -+ 0.28 (6) 2.83 (40) 0.87 (38) 3.17 (2) 1.22 (2) 2.55 _ 0.19 (3) 0.71 -+ 0.10 (3) KIDNEY 1.14_+ 0.22 1.1o 0.81 ___ 0.12 0.86_+0.2 0.92 +_0.05 --0.84 0.78 0.98 -- STOMACH (10) (1) (59) (3) (3) 2.62 +- 0.34 1.56 ___ 0.32 0.95 -+ 0.13 1.24 1.48 _+0.41 (12) (3) (38) (1) (4) 0.75 0.56 0.63 0.53 0.99 -- M_ASS _+0.42 (18) (1) _+0.28 (22) +_0.05 (3) _+2.62 (7) -- -(2) (36) (1) BODY INTESTINE 1.93 1.94 (1) (12) 2.48 1.91 (1) (2) 0.46 0.75 0.42 1.64 1.26 _+0.38 180 190 220 760 850 1040 (1) (2) (2) (1) (3) 1110 270 310 350 460 significant loss in tissue weight (Barton and Houston 1992); we therefore discardedany birdswhich had obviouslylain for some time before being frozen. In order to test our hypothesis,we first needed to categorize the predatory styleof each raptor species.For the liver, and kidney between eight speciesof Falconiformes and four Strigiformes.We also examined heart weight in thesespeciesto seeif this is heavier in specieswith an energetic form of prey capture. Finally, we considered the correlation between body condition and the size of these organs. Falconiformes we did this based on literature review of the proportion of avianprey in each speciesdiet. Species taking more than 75% avian prey were categorizedas attackers and included peregrine falcons (Falcoperegnnus), sparrowhawks(Accipiternisus),and goshawks(A. gentilis),while speciestaking predominantly small mammals and carrion were categorizedas searchersand in- METHODS Most of the birds used in this studyhad been found dead and sent to the Institute of TerrestrialEcologyasa part of their program to monitor pesticideresidues.In addition, a sampleof goshawks wasexaminedwhich had been shot under licensefor game preservationand imported under license (CITES 038371A). Birds were stored frozen at -20øC for up to 6 wk. It is known that freezing has little effect on gut morphology, but prolonged periods betweendeath and freezing can lead to cluded buzzard (Buteo buteo), red kite (Milvus milvus), and kestrel (Falco tinnunculus)(Brown1978). Attackers were specieswhich capture prey by high-speedaerial chase,requiring great accelerationand agilitywith an averagehunting success of 13% (TemPles1985). Searchers, however,were speciescapturingmammalianprey mainly Table 2. Correlation betweenorgan size (g) and index of body condition of variousspeciesof raptors.Samplesizes given in parentheses. SPECIES SEX Sparrowhawk M r = 0.71, P= 0.0018 (16) F r = 0.53, P= 0.0018 (32) r = 0.63, P= 0.02 (13) Peregrine M r = 0.96, P = 0.03 (4) r = 0.96, P = 0.01 (5) Kestrel M F M F M M F r = 0.55, P = -r = 0.93, P = -r = 0.69, P = r = 0.33, P= r = 0.44, P = r = 0.34, P= r= r= r= r= r= r= r= Buzzard Red Kite Goshawk Barn Owl TawnyOwl GIZZARD -- HEART r = 0.89, P= 0.02 (6) 0.09 (10) 0.0001 (10) 0.31 0.12 0.07 0.19 (4) (23) (18) (16) -0.80, P = 0.87, P= 0.77, P= 0.95, P = 0.58, P= 0.79, P = 0.78, P= 0.03 (7) 0.0001 (16) 0.0003 (17) 0.004 (6) 0.002 (25) 0.0001 (19) 0.02 (8) r = 0.72, P = 0.28 (4) DECEMBER 1996 INTERNAL ORGAN SIZE IN RAPTORS by dropping on it from above and not depending so heavilyon high-speedpowered flight, with higher rates of prey capture (Temeles 1985). For the owl speciesexamined, the main difference in hunting strategybetween specieswasthat barn owls( Tytoalba)and long-earedowls (Asiootus)have an active flight mode with prey mainly locatedwhile in flight, while tawnyowls (Styixaluco)are more passivepredators, locating prey from perch sites (Cramp and Simmons 1980). 221 and sternum diagonal length (from base of sternum to distal point of coracoid) as the best predictors.We used the product of thesetwo measuresas our body sizefactor and used this factor as the covariatein ANCOVA analysis to correct for body size differences.We log-transformed data on both axes. The Bonferroni method was used for pairwise comparison of adjusted treatment means (Day and Quinn 1989). Since our study dealt with evolutionary adaptation of There was some variation in the condition of the birds digestiveorgan sizewith relation to hunting strategy,we availablefor this study.We used dry weightsto record considered including a phylogenetic component in our organ massbecausewe did not know the extent to which analysis.However,sincethe speciessamplewassmalland carcasseshad become dehydrated after death. Wet we had specieswith contrastinghunting strategiesin the weightswere used only from birds known to have died samegenus,we decided there would be no advantageto accidentallyand whose carcasses were collected shortly adding a phylogeneticcomponent. after death. To considerthe contribution that each organ makesto The proventriculus,gizzard, small intestine, liver, kid- total body mass,we used a smaller sample of fresh carney, and heart were dissectedfrom 583 carcasses. Stom- cassesto obtain wet weightsof liver, kidney,proventncuachswere separatedinto the proventriculus,whosefunc- lus, gizzard, small intestine, and heart. The weight of uon is the production and releaseof gastricsecretions, each organ wascalculatedas a percentageof total body and the gizzard, which provides mechanical digestion mass. We dissectedthe pectoralismusclefrom each carcass and preliminary proteolysis(Duke 1986). The length and w•dth of the proventriculuswere measuredwhen laid flat After weighing,it wasdried at 70øCto constantweight. to estimate internal surface area. All other organswere Larger specieshad 10 g samplesof muscle tissuedried, cleaned of mesenteryand fat bodies:the heart had the from which total muscle dry weight was estimated from chambersopened and any blood clots removed.All tis- the wet weight of the sample and total muscle. Subsamsueswere then oven dried at 70øCto constantweight. pies were shown to be representativeof the whole musThe combined weight of proventriculus and gizzard is cle. A condition index which accountedfor body sizedifhere referred to as stomach, and the combined mass of ferenceswasthen calculatedby regressingthe dry weight proventriculus,gizzard, and small intestine as gut. Fal- of the two pectoral musclesagainstthe skeletalbody s•ze coniformeswere analyzedseparatelyby sex, becauseof measureand savingthe residuals.The variablesbeing extheir sexual dimorphism, and sufficientdata were avail- amined were standardizedin the same way,again sawng able to analyzemale and female sparrowhawks, peregrine the residuals. Correlations between the two sets of residfalcons, kestrels, common buzzards,red kites, and gos- uals were then examined. hawksseparately.We useda skeletalbody sizevariableto correct for differencesin body sizebetweenspecies(Bar- RESULTS ton and Houston 1994). We used Principal Components Analysis to identify which of the six body measureswere best predictors of body size, and identified the length of the sternum keel (from base of the sternum to anterior edge of the keel) Male and female sparrowhawksand goshawks had the lightestgizzardsand stomachsfor birds of their size, while red kites and common small Table 2. SMALL SMALL INTESTINE LENGTH r = 0.51, P = 0.009 (25) r = 0.06, P = 0.72 (39) r = 0.14, P = 0.79 (6) r = 0.60 P= r = 0.15 P = r= 0.60. P = r = 0.09 P-r= 0.63. P = r = 0.09 P = r= 0.35. P = 0.12 0.63 0.01 0.70 0.18 0.65 0.13 intestines and common buzzards and red kites the heaviest.When total mass of the gut was considered,the guts of the goshawk,sparrowhawk, and peregrine were the lightest for their body size Extended. INTESTINE buzzards had the heaviest. Sparrowhawkshad the lightest (8) (13) (17) (22) (6) (25) (20) DRY WEIGHT r = 0.45, P = 0.06 (18) r = 0.30 P: 0.07 (36) r= r= r= r= r= r= r= r= 0.95 0.89 0.79 0.69 0.34 0.99 0.04 0.69 P-- 0.01 (5) P = 0.001 (9) P= 0.0005 (15) P= 0.0017 (18) P-P: P= P= 0.12 (22) 0.0001 (6) 0.83 (25) 0.001 (19) r = 0.28 P = 0.17 (24) r = 0.37 P= 0.14 (17) r = 0.99 P = 0.004 (4) r = 0.95 P = 0.05 (4) and those of the kestrel and common buzzard the heaviest. All comparisons were significant (P < 0.05) with no significant heterogeneity in regression slopes. In considering heart weight, there were significant differences among the speciesexamined, but these were not consistent. Among males, kestrels had the largest heart relative to body massbut, in females, goshawksand sparrowhawks were heavier. Among owls, the tawny owl had a significantly heavierproventriculus,smallintestine,and total digestivetract than expected for a bird of this size. There were no significant differences in heart weight between owls. 222 N. BARTON AND D. HOUSTON We were not able to obtain dry weights for liver and kidney tissuebut determined their wet weights asa percentageof total bodyweight (Table 1). This wasa lesssatisfactorymethod of indicating relative organ size becausebirds differed in their fat content, gut content, and overall muscle condition making body weight a poor predictor of actual body size (Barton and Houston 1994). To minimize this problem, we only included data from birds which were freshly killed from accidental causesand which seemedto be in good condition. Sample sizeswere small but there is no indication VOL. 30, NO. 4 waysbe the best measure of gut tissuebecausethe alimentary tract can stretch, so that differencesin length may not necessarilyreflect a corresponding difference in tissue mass. Here we have shown that small intestine tissue weight also varies between species,and that those specieswith an attacking mode of predation have significantlylessintestine tissuethan those that do not depend so heavilyon rapid accelerationfor prey capture.Apart from the small intestine, the stomachsizeis alsoimportant, for this not only involves the mass of the organ itself, but also determines the weight of food that thatliver,kidney,or heartmasswasassociated with a speciescan consume.We found that a species predatory strategy.Liver and kidney made up ap- such as the sparrowhawk which faces selection proximately the same proportion of body mass pressuresto minimize bodyweight doesshowsmall acrossall species.The largestrelativeheart sizewas stomach mass. This may accentuate the imporfound in the peregrine and merlin (Falcocolumbar- tance of diet quality between raptors.We showed zus)both of which havehigh power output require- earlier (Barton and Houston 1993) that species ments. However, the goshawk and buzzard, two with short intestine lengths are lessefficient at dispecieswith directlycontrastinghunting strategies, gestion and can only maintain body weight when had the same relative heart size. fed on prey speciesof high calorificvalue. If they Stomachsizewasvariable. The kestrel had a very alsohave significantlysmaller stomachs,and so can large stomach for a bird of its size. Compared to consumea smaller amount of prey, the difference the sparrowhawk,the kestrel had a stomachthree in organ size emphasizesthe need to take prey of times larger. The relative size of the kestrel gut high energy content to compensatefor the lower more closely resembled that of the Strigiformes overall intake. In Strigiformes,proventriculusand suggestingit might be a dietary adaptationfor spe- small intestine weight (and also length, see Barton and Houston 1994) was greatest in the tawny owl cies feeding largely on small mammals. We used weight of the pectoral muscle as an in- which is the specieswith the most passivehunting dex of body condition becausethis varies consid- mode compared to the barn owl or long-earedowl erably and is the largest muscle in the body of a (Barton and Houston 1994). bird. Individualswhich we categorizedsubjectively While there appearsto be clear associations beas having died of starvationhad on averageonly tween predatory strategyand the total weight of 54% of lean dry pectoralmuscleweightcompared the digestivetract, this is not the casefor the liver, to that found in individuals which had been killed kidney, and heart. The size of these organsseems by collisions.For all speciesthere wasa significant to be strictly determined by metabolic body size, positive correlation between condition and the and presumablycannot be reduced in masswithweight of the small intestine (Table 2). The length out seriouslyimpairing tissuefunction and the fitof the small intestinewasalsopositivelycorrelated ness of the bird. with condition in male sparrowhawks,male buzThe variations that are found with body condizards and tawny owls. Gizzard weight was also tion were perhaps to be expected.When animals found to be significantlycorrelated with body con- enter periodsof food deprivation,they depend on dition for sparrowhawks,male peregrines, and endogenousreservesof fat and protein to mainmale common buzzards.Heart weight was highly tain their metabolic requirements. The extent to correlated with condition in every speciesexcept which tissuefrom the alimentary tract is depleted, the tawny owl. however,is little known. Our finding that species we examined showedvery significantlossesof tisDISCUSSION sue when in poor condition demonstratesthat gut In an earlier analysiswe showedthat the length tissueis widely used as a reserve. Presumablyany of the smallintestinein raptorsvaried betweenspe- reduction in gut tissuehas a detrimental effect on cies and seemedto be stronglyassociatedwith the digestiveefficiency,and so birds do not experience predatory strategy.However, length might not al- this tissuelossunless they have no alternative. The DECEMBER 1996 INTERNAL ORGAN SIZE IN RAPTORS finding that heart muscle is also significantlylost in birds in poor condition also demonstratesthat when birds reach poor body condition even the most vital organs are affected. ACKNOWLEDGMENTS We are extremely grateful to Ian Wyllie and Ian Newton at the Institute of Terrestrial Ecology and to Steve Petty at the Forestry Commissionfor providing most of the birds used for this study.The project was funded by a Natural Environment Research Council CASE Student- ship with the Nature ConservancyCouncil to N.W.H.B. LITERATURE CITED B^RTON, N.W.H. ^ND D.C. HOUSTON. 1992. Post-mortem changesin avian grossintestinal morphology. Can.J. Zool. 70:1849-1851. 1993a. The influence of gut morphologyon digestiontime in raptors. Comp.Biochem. Physiol.A. 105: 571-578. 1993b. A comparison of digestiveefficiency in birds of prey. Ibis 135:363-371. 223 1994. Morphological adaptation of the digestive tract in relation to feeding ecologyin raptors.jr. Zool. Lond. 232:133-150. BROWN,L.H. 1978. British birds of prey. Collins, London, UK. CP,AMP, S. AND K.E.L. SIMMONS. 1980. Handbook of the birds of Europe, the Middle East and North Al•ica. The birds of the western Palearctic. Vol. 2. Oxford University Press,Oxford, UK. DAY,R.W. AND G.P. QUINN. 1989. Comparisons of treatments after an analysisof variance in ecology. Ecol. Monog• 59:433-463. DuRE,G.E. 1986. Alimentary canal:secretionand digestion, special digestivefunctions and absorption. In P.D. Sturkie, Avian Physiology.Springer-Verlag, New York, NY U.S.A. TEMELES, E.J. 1985. Sexual size dimorphism of bird-eating hawks: the effect of prey vulnerability. Am. Nat. 125:485-499. Received4 October 1995; accepted 26 August 1996