Blood Chemistry, Cytology, and Body Condition in Adult Northern Goshawks (Accipiter Gentilis)

j. RaptorRes. 37(4):299-306 ½32003 The Raptor ResearchFoundation, Inc.

BLOOD CHEMISTRY, CYTOLOGY, AND BODY CONDITION IN ADULT NORTHERN GOSHAWKS (ACCIPITER GENTILIS)

L•UPdA. HANAUSKA-BROWN1 AND ALFRED M. DUFTY,JR. RaptorResearch Centg Department of Biology,Boise State University, 1910 UniversityDrive, Boise, 1D 83725 U.S.A.

GARYJ.ROLOFF TimberlandResources, Boise Cascade Corporation, 1564 TomlinsonRoad, Mason, MI 48854 U.S.A.

ABSTRACT.---Abird's physiologicalstate contributes to its reproductivesuccess and survival,yet few baseline physiologicaldata havebeen publishedfor wild raptors.Mean levelsof protein, cholesterol,calcium, uric acid, aspartateaminotransferase, lactate dehydrogenase,and creatine kinasewere measuredin 29 Northern Goshawks(Accipiter gentilis) during 1998-99. None of these substanceswere significantlydif- ferent betweenmales (N = 8) and females(N = 2). Levelsof all substanceswere highlyvariable among individual birds and unrelated to the body condition index (mass/wing chord X tail length X culmen length). Total white blood cell count estimatesand differential white blood cell countswere not significantly different between the sexes.Of the blood cell measures,only the percent of heterophils and lymphocytes,and the ratio of heterophilsto lymphocytes(H/L) differed betweenbirds in good body condition and birds in relativelypoor body condition. The H/L ratio has recentlybeen proposedas a reliable measureof stress.Thus, variable H/L ratios between groups of wild birds may indicate differences in stress levels and overall health.

KEYWORDS: NorthernGoshawk; Accipiter gentilis; blood chemistry; H/L ratio;body condition.

QU_iMICASANGUiNEA, CITOLOGIA, Y CONDICION CORPORAL ENADULTOS DEL AZOR NOR- TENO ( A CCIPITER GENTILlS) RESUMEN.--E1estado fisio16gico de una ave contribuyea su 6xito reproductivoy supervivencia,hasta el momento muy pocosdatos fisio16gicos de 1ineabase han sido publicadospara rapacessilvestres. Los nivelesmedios de proteinas,colesterol, •cido firico, aminotransferasaaspartato, lactato deshidrogenasa, y creatina quinasase midieton en 29 azoresnortefios (Accipitergentilis) durante 1998-99. Ninguna de estassubstancias rue significativamentediferentes entre machos (N = 8) y hembras (N = 2). Los niveles de todaslas substanciasrueton altamentevariables entre avesindividuales y no estuvieronrelacionadas con el indice de condici6n del cuerpo (masa/cuerda alar H longitud de la cola H longitud del culmen). E1 conteo de leucocitostotales estimados y el conteo diferencial de leucocitos,no rueton significativa- mente diferentesentre sexos.Entre las medidasde las cdulas sanguineas,finicamente el porcentajede heterofilosy linfocitos,y la raz6n de heterofilos a linfocitos (H/L) difiri6 entre avescon buena condi- ci6n corporal y avesen condici6n corporal relativamentepobre. La raz6n H/L ha sido propuestare- cientemente como una medida confiable de estr6s.Asi, las razonesH/L variablesentre grupos de aves silvestrespueden indicar diferenciasen los nivelesde estrfsyen la saludglobal. [Traducci6n de C6sar M•rquez]

Northern Goshawks(Accipiter gentilis) are found cial concern" by the Idaho Department of Fish throughout forestedportions of the intermountain and Game (Beals and Harris 1996). Basic knowl- west (Squiresand Reynolds1997) where they are edge of adult animal health, including blood- classified as a "sensitive species" by the United chemistry information, is needed for the man- States Forest Service (Beals and Harris 1996, agement of vulnerable species (Ferrer and Squiresand Reynolds1997) and a "speciesof spe- Dobado-Berrios1998), such as the goshawk.How- ever,past research on free-ranginggoshawk life his- 1Present address: Idaho Dept. of Fish& Game, 1515Lin- tory or management does not include an assess- coln Rd., Idaho Falls, ID 83401 U.S.A.; e-mail address: ment of adult health from blood chemistry or [email protected] cytology (blood cell) measures. These measures

299 300 H•,:•USK•-BROWN ET AL. VOL. 37, NO. 4

provide information regarding nutritional and im- MATERIALS AND METHODS munological status,and help us understand eco- This study was conducted from 1998-99 in the west- logical and behavioral problems (Ferrer 1990, central mountainsof Idaho. The studyarea encompassed Stein et al. 1998). Two objectiveswere addressed parts of the Boise, Payette, Snake, and Salmon River drainages where elevations range from 460-$050 m in this study: (1) determine baseline means and (Steele et al. 1981). Public and private lands were sur- ranges of blood chemistry and cytologylevels in veyedfor goshawksin Washington,Adams, Idaho, Valley, adult breeding goshawks,and (2) determine the and Boise counties. The study area is a mix of lowland relationshipsamong blood chemistry,blood cytol- meadowsand rangeland and higher elevationmixed co- niferous forests.A more complete description of the veg- ogy, and body condition. etation and climatic conditions can be found in Hanaus- Blood chemistryis an indirect method of assess- ka-Brown (2000). ing health asblood actsas a meansto mobilize and We trapped adult goshawkswith a live Great Horned transport nutrients, metabolic products, immune Owl (Bubovirginianus) lure to elicit aggressiveresponses from nesting goshawksand capture them in a dho-gaza cells, and hormones (Brown 1996). Metabolic sub- net (Detrich and Woodbridge 1994). Upon capturing stancespresent in blood plasma reflect avian nu- one adult from a nest, we reset the dho-gaza net in an tritional health and overall condition (Snyder and attempt to catch the other adult. Terry 1986, Mauro 1987, Ferrer 1993). We collected up to 500 •1 of whole blood after puncturing the brachial vein with a sterile lancet. Plasmawas Much like plasma metabolites,plasma enzymes sent to IDEXX VeterinaryServices Laboratory, Inc. (Port- alsoreflect avian health. Enzymessuch as aspartate land, OR) which measuredprotein, cholesterol,calcium, aminotransferase (AST), lactate dehydrogenase uric acid, and three blood enzymes:AST, CK, and LDH. Additional substances were measured in fewer than 10 (LDH), and creatine kinase (CK) help regulate individuals:albumin, globulin, phosphorous,potassium, metabolicreactions and respondto changesin the sodium, amylase,and alkaline phosphatase.Descriptive body (Mauro 1987). Interpretationof plasmaen- statisticsonly are presentedfor these variablesas statisti- zyme levelsis complicatedby natural variation un- cal analyseswere deemed inappropriate. lesslevels are significantlyhigher or lower than ref- We evaluated adjusted Sidak P-values(Zar 1996) from multiple pairwisecomparisons (MULTTEST, SASRelease erence samples(Ivins et al. 1986, Campbell 1995). 6.12, 1996) to determine significant differencesbetween Differential white blood cell counts provide in- samplescollected from females and males, with N-> 8 formation on the activitylevel of a bird's immune samplesfor each sex (Zar 1996). Adjusted SidakP-values systemand stressresponse (Dhabhar et al. 1996). are slightlyless conservative than the Bonferroni testand control for comparison-wiseerror (Zar 1996). We per- Heterophils and lymphocytesare the predominant formed Spearman correlation analysesof metabolicsub- white blood cells found in raptors (Lavin et al. stancesand enzymeswhen N -> 8 individualsfor each 1992, Stein et al. 1998). The ratio of heterophils sex. Assumptionsof normality and heterogeneitywere to lymphocytes(H/L) has proven to be a reliable tested prior to statisticalcomparisons, and correlations and nonparametric testswere performed when assump- measureof stressin poultry and wild birds and a tions were violated. more accurate measure of stress than individual A blood smear was made for each bird and sent to cell numbers or corticosterone levels (Gross and IDEXX Veterinary ServicesLaboratory, Inc. The labora- Siegel 1983, Maxwell 1993, Vleck et al. 2000). tory estimated total white blood cell numbers and per- centagesof cell typesincluding: heterophils,eosinophils, Finally, morphological characteristicsmeasured monocytes,and lymphocytes.We used multiple pairwise to provide a body conditionestimate can be indic- comparisonsand Sidak P-valuesagain to evaluatemean ative of individual health (Brown 1996). Body mass differencesbetween the sexes.We also performed Spear- •s relative to structural size (i.e., wing, tail, culmen man correlation analysesfor all cytologicalmeasures. length), which is defined as the nutrient-reserve For the calculation of body condition, we measured mass,unfiattened wing chord length, tail length, and cul- independent size of an individual (Piersma and men length. Wing chord length wasmeasured to the tip Davidson 1991). Birds in good condition should of the longestprimary, tail length to the tip of the longest have a relativelyhigh ratio of massto structuralsize rectrix, and culmen length the distance between bill tip and anterior end of the cere. The mass:structural size (Brown 1996). We hypothesizedthat goshawksas- ratio wasdetermined by dividing mass(g) by the product sessedto be in good body condition by this index of wing chord length (mm), tail length (mm), and cul- would exhibit relativelyhigh levelsof blood chem- men length (mm). istry measures(e.g., protein, calcium) indicativeof Preliminary resultsfrom our data demonstratedblood metaboliteand enzymemeasures (N = 7) were signifi- adequate nutrition and low levelsof cytologymea- candy correlated, as were white blood cell measures(N sures (e.g., H/L ratio) indicative of an activated = 6). Thus, we predicted that a suite of blood chemistry immune system,high stress,or disease. and cytologymeasures, rather than any individual mea- DECEMBER2003 GOSH^WKBLOOD AND BODY ME^SU•S 301

Table 1. Metabolitesand metabolicenzymes measured in Northern Goshawksin west-centralIdaho during the breedingseasons of 1998and 1999 along with ranges reported for 24 captiveadult female and male goshawks (Benyon et al. 1996).

BENYON ET AL. FEMALES MALES (1996) SUBSTANCE N MEAN (SE) RANGE N MEAN (SE) RANGE RANGE Totalprotein (g/dL) 21 1.8 (0.2) 0.9-3.4 8 2.3 (0.2) 1.3-3.2 2.6-4.2 Cholesterol(mg/dL) 21 155.1(14.8) 33.0-286.0 8 201.0 125.0-282.0 154.7-4447 Uric acid (mg/dL) 21 12.6(2.3) 1.8-40.2 8 20.9 (6.6) 9.4-63.0 8.6-14.4 Calcium(mg/dL) 21 4.7 (0.5) 1.8-9.5 8 5.9 (0.7) 3.3-8.0 8.6-10.9 Phosphorous(mg/dL) 4 3.7 (0.7) 2.4-5.0 4 2.2 (0.5) 1.2-3.7 NA Potassium(mEq/L) 5 3.0 (0.3) 2.5-4.0 4 2.5 (0.3) 1.9-3.1 NA Sodium(mEq/L) 5 158.6(7.4) 136.0-181.0 4 159.8(2.3) 155.0-165.0 NA A/G ratioa 4 0.7 (0.1) 0.5-1.0 4 0.7 (0.03) 0.6-0.7 NA Globulin(g/dL) 4 1.6 (0.2) 1.2-2.2 4 1.6 (0.1) 1.4-1.8 NA Albumin(g/dL) 4 1.1 (0.1) 0.9-1.2 4 1.0 (0.1) 0.9-1.2 NA AST (IU/L) b 21 286.2(26.2) 108.0-561.0 8 375.5 199.0-556.0 176.0-409.0 LDH (IU/L) c 21 347.6 (47.9) 136.0-840.0 8 616.4 322.0-1040.0 120.0-9060 CK (IU/L) a 21 127.0(12.3) 56.0-285.0 8 160.6 56.0-412.0 218.0-7750 Amylase(IU/L) 4 900.5(44.1) 790.0-1006.0 4 864.0 567.0-1195.0 NA Alk. phos.e (IU/L) 4 20.5(4.7) 9.0-31.0 4 20.5(4.8) 15.0-35.0 NA • A/G = albumin/globulin ratio. b AST = aminotransferase. c LDH = 1 acetate dehydrogenase. d GK = creatine kinase. ½Alk. phos.= alkaline phosphatase. sure,would bestrepresent goshawk physiological health. for all birds) and the individual physiologicalvariables We hypothesizedthat goshawkswith higherbody condi- (StatSoft 1998). Thus, the magnitude and sign of factor tion indiceswould exhibit a more healthyphysiological loadings can be used to determine the relative impor- profilecompared to birdsin relativelypoor bodycondi- tance of individualvariables in explainingthe two-sample tion. To testthis hypothesis we reducedthe physiological comparison results. data from each bird to an individual factor score using principlecomponents analysis (StatSoft 1998). The factor RESULTS score was a multi-dimensionalphysiological representa- tion of goshawkhealth. In thisanalysis, we usedthe factor Metabolite, enzyme, and white blood cell con- scoresassociated with the first eigenvector.Subsequently, centrations were measured in 21 adult female and factor scoresfor the 10 goshawksin the best and the 8 adult male goshawks(Tables 1-2). There were worstbody condition (as indicatedby bodycondition in- no between-sex differences in metabolite or en- dex) were evaluatedusing a nonparametrictwo-sample comparisontest (Mann-WhitneyU:test; Bailey 1995). We zyme levels (P > 0.50). Data on metabolic sub- used the body condition index as the measureof overall stancesand enzymes from 24 captive adult gos- conditionas it wasindependent of the other physiologi- hawks (Benyon et al. 1996) are provided for cal measures (Hanauska-Brown 2000) and has been used to representhealth in other studies(Rising and Somers comparisonwith our data (Tables 1-2). There 1989, Brown 1996). We combined sexes in the analysis were no between-sex differences in total white because there were no differences in blood chemistry, blood cell estimate, individual cell type, or H/L cytology,or body condition means.Two-sample compar- ratio for goshawksmeasured in this study (P > isonswere performedon factorscores derived from the 0.45); however, we note that the differences be- entire suite of physiologyvariables (N = 13), the combined metabolicand enzymevariables (N = 7), and the tween female and male monocyte numbers ap- suite of blood cell variables (N = 6). For significanttwo- proachedsignificance (t = -3.10, P = 0.07). The samplecomparisons, factor loadingsfrom the principle correlationanalysis of all blood chemistryvariables componentsanalysis were usedto help determinewhich showed that all metabolites and enzymeswere cor- variables had greatest influence on the factor scores (StarSoft1998). Factorloadings can be interpretedas the related (P-< 0.05) with the exception of the en- correlations between factors (i.e., the set of factor scores zyme CK. Correlationanalysis of the blood cell 302 HANAUSKA-BROWNET AL. VOL. 37, NO. 4

Table 2. White blood cell total estimates(X109/1) and white blood cell types(%) measuredin Northern Goshawks in west-centralIdaho along with rangesreported for 43 captiveadult female and male goshawks(Benyon et al. 1996).

BENYON ET AL. FEMALES MAL•S (1996)

WBC N MEAN (SE) RANGE N ME&N (SE) RANGE RANGE Total estimate 21 9.8 (1.0) 4.0-20.0 8 8.7 (1.2) 4.0-13.0 4.0-11.0 Monocytes 21 2.3 (0.4) 0.0-6.0 8 5.3 (1.3) 0.0-12.0 1.0 Lymphocytes 21 38.2 (2.5) 16.0-58.0 8 46.8 (5.0) 26.0-66.0 14.0-19.0 Eosinophils 21 19.6 (1.4) 6.0-33.0 8 13.6 (2.3) 4.0-23.0 7.0 Heterophils 21 40.0 (2.4) 19.0-58.0 8 34.3 (5.0) 12.0-54.0 36.0-65.0 H/L ratioa 21 1.27 (0.17) 0.37-3.31 8 0.89 (0.23) 0.18-2.00 NA a Heterophils to lymphocytes. counts and white blood cell estimates demonstrat- bers, and low H/L ratios for goshawksin good con- ed significant patterns (P --<0.05) among several dition (Hanauska-Brown2000). There was a signif- variables.The mean body condition ratio was 4.39 icant negative correlation between H/L and body -+ 0.08 (range = 3.81-4.96) for females and 4.90 condition (N = 30, fl = -0.48, P < 0.01). __-0.03 (range = 3.86-6.16) for males (Table 3). DISCUSSION There was no significant difference between female and male body condition ratios (Wilcoxon Concernsover the population statusof goshawks signed rank, z -- 1.51, P = 0.13). throughout the intermountain west region (Ken- The two-samplecomparison test using all physi- nedy 1997) promptedus to seeka meansof index- ological variables (N = 13) for the 10 goshawks ing the health of adult breeding goshawks.We with the highest and lowestbody condition scores based this index on morphologicalmeasurements, detected no difference (N = 20, U = 27, P = 0.25). blood chemistry,and cytologicalcharacteristics. We Similarly,the comparisonusing the combinedme- focused on these variables because other research- t•,l•,o •', enzymevariables (N = 7) showedno ers have questionedthe assessmentof population difference between goshawks of different body viability using occurrence,density, or annual pro- condition (N = 20, U = 36, P = 0.29). Two-sample ductivity estimates(e.g., Van Horne 1983, Franklin comparison tests on factor scores derived from et al. 2000). blood cell variables (N = 6), however, showed a Our data representbaseline metabolite, enzyme, difference betweengood and poor body condition and blood cytologylevels for a sampleof free-rang- goshawks(N = 20, U = 3, P < 0.01). Factor loading goshawksin central Idaho. The only other ings for each variable indicated that the H/L ratio baselinedata availablefor goshawksare from "nor- (negative loading) was the variable primarily re- mal, healthy" male and female goshawksof various sponsiblefor the observed difference. These re- ages, housed in rehabilitation centers (Benyon et sults are consistentwith paired comparisonsper- al. 1996). Captivebirds are providedwith a contin- formed on individual variables that demonstrated uous food source,but alsoexperience a multitude high lymphocyte numbers, low heterophil num- of stressors(e.g., confinement and handling) not

Table 3. Morphologymeasures from Northern Goshawksin west-centralIdaho.

FEMALES MALES

MORPHOLOGY N MEAN (SE) RANGE N MEAN (SE) RANGE Mass(g) 21 1006.8 (14.8) 870-1134 8 766.9 (15.7) 710-824 Wing (mm) 21 353.8 (1.7) 341-372 8 325.4 (3.8) 310-337 Tail (mm) 21 261.7 (1.2) 249-272 8 225.1 (2.4) 212-232 Culmen (mm) 21 24.8 (0.3) 23.2-29.9 8 21.1 (0.3) 19.7-25.1 DECEMBER 2003 GOSHAWK BLOOD ^•4D BODY MEASURES 303 present in the wild (Stein et al. 1998). Captive other factors (Gee et al. 1981, Hoffman et al. 1985, birds receive a steady,high protein diet (Gee et al. Stein et al. 1998). For example, Stein et al. (1998) 1981, Garcia-Rodriguezet al. 1987) that provides documentedmore variabilityin wild populationsof an adequatesupply of amino acidsfor the synthesis American Kestrels (Falcosparverius) and Red-tailed of bloodproteins (Ferrer et al. 1987).Indeed, total Hawks (Buteojamaicensis) than in captive birds. protein was higher for the captive goshawksthan Mean levels of AST and LDH measured in this for the birds sampled in this study.Wild goshawks studywere within the wide range of enzymelevels may experienceperiods of fasting(Newton 1979) in captivegoshawks (Benyon et al. 1996). However, and may be forced to specializeon certain prey lower levels of CK were measured in wild versus species(Newton 1979, Younk 1996) leading to a captive birds (Table 1). Average or low CK levels mineral or dietary deficiency (Snyder and Terry can indicate enhanced muscle strength and endur- 1986). Birds in the wild engage in activities (e.g., ance (Apple and McGue 1983, Knuth and Chaplin courtship,territoriality, nestling care) that reduce 1994). For example, flight training in captiveRed- the time availablefor foraging,which alsomay lead tailed Hawks improved the structural integrity of to dietary deficiencies,particularly in poor-quality musclesand loweredplasma CK levels(Knuth and habitat. Chaplin 1994). The low levels of CK observed in Calcium levelswere lower in wild than in captive this study may reflect increased muscle strength goshawks(Benyon et al. 1996), which also can be and flight endurance (Apple and McGue 1983, attributed to dietary differences.Captive birds re- Knuth and Chaplin 1994) in the wild birds. ceiving chicken carcasses, rabbits, and other Individual variation due to proximate factors 'whole' food items (Gee et al. 1981) obtain calcium may play a major role in variableblood chemistry from ingestingbones and other body parts. Con- measures,but many other influencescan affect the versely,captive birds fed only red meat (including measurement of health in wild birds. Garcia-Ro- muscle, heart, and liver) were found to be defi- driguez et al. (1987) demonstrateddiurnal and cir- cient in calcium (Graham and Halliwell 1986). Di- cadian rhythms of calcium, uric acid, and choles- etary calcium in the wild comesfrom a varied diet terol levels in raptors. We could not control the including whole prey items (Graham and Halliwell time of day of sample collection because of the 1986, Squiresand Reynolds1997). However,wild logistical constraints of trapping breeding gos- birds may not be able to capture a steadydiet of hawks. Time of sample collection in this study suitableprey, and low levelsof calciummay reflect ranged from 0700-1900 H. Variation in the levels fasting periods (Halliwell 1981). of protein, uric acid, calcium, and cholesterol in The mean uric acid level for females was within individualgoshawks in this study,therefore, may be the range reported fo•: captive goshawks,but the attributed to recent feeding, fastingperiods, or cir- mean for males was above the reported range cadian and diurnal rhythms. (Benyon et al. 1996) by more than 6 mg/dL. Uric Geographiclocation alsomay affectblood chem- acid, a nitrogenouswaste product, is typicallyele- istry measures. Soil types and environmental con- vated in birds with high protein diets (Bell and tamination affect physiology,as raptors incorpo- Sturkie 1965, Gee et al. 1981). High uric acid levels rate locally-obtained nutrients and toxins from in free-ranging birds can also be attributed to de- their prey items (Hoffman et al. 1985, Ferrer and creased body condition or food stress(Handrich Dobado-Berrios1998). Differencesin total protein, et al. 1993, Balbontin and Ferrer 2002, Casado et uric acid, cholesterol, and calcium were found be- al. 2002). Males are under high energy demands tween two populationsof Adalbert's Eagle (Aquzla during the breeding season as they hunt for the adalberti)foraging in different regions of Spain adult female and young, as well as for themselves. (Ferrer and Dobado-Berrios 1998). Qualitative dif- Mean cholesterollevels from sampledbirds were ferencesin each population'sdiet were suggested within the range of cholesterolvalues reported for as the cause of the physiologicaldifferences (Fer- captivegoshawks (Benyon et al. 1996). rer and Dobado-Berrios1998). Similarly,differenc- The three enzymeswe measuredare quite vari- es in prey abundance,prey availability,and habitat able in captive goshawks(Benyon et al. 1996). in central Idaho may be reflected in goshawk Such individual differences are common in avian blood chemistrymeasures. For example, goshawks speciesdue to natural variation in habitat, genetics, in the western region of the study area likely con- body condition, environmental influences, and sumed more ground squirrels (Spermophilusspp.) 304 HAN^USK•-BP.OWN ET AL. VOL. 37, NO. 4

than goshawksin the northern regions, which ratio highlights the potential sensitivityand utility probably consumedmore birds (pers. observ.). of this technique. Goshawk samplesfrom this study showed in- Our resultsalso illustrate the difficulty in evalu- creased numbers of lymphocytes,monocytes, and ating the physiologicalstate of a bird at any given eosinophilscompared to the ranges reported for time. Metabolite and enzyme levels appear to be captive goshawks(Benyon et al. 1996). Increased influenced by too many external and individual levels of particular leukocytesor white blood cell factors to serve as accurate barometers of individ- total estimatescan indicate physiologicalstress or ual health. Despite the limitations inherent in decreased immunocompetence (Smith and Bush physiologicalstudies of wild raptors such as ours, 1978, Kontecka et al. 1999). A shift from normal the H/L ratio and body condition index hold valuesin any direction can indicate disease(Camp- promise as indicatorsof adult goshawkhealth. Fur- bell and Dein 1984). Blood loss or tissuedamage ther study of free-living raptor physiologyis war- at the collection site can also affect white blood cell ranted to assess the value of the H/L ratio as a counts. Our blood collection method may have measure of individual health. overestimatedcirculating white blood cellsbecause ACKNOWLEDGMENTS hemorrhage induced by puncturing the wing vein produces tissuedamage that triggers clotting and We thank the following organizationsfor funding and supporting this project: Boise CascadeCorporation, Boi- attracts white blood cells. Capillary blood in hu- se State University, and the Idaho Department of Fish mans (e.g., blood collected into capillary tubes af- and Game. We thank the biologistsof the Boise and Pa- ter finger pricks) has higher white blood cell yette National Forestsfor their cooperation and time. Fi- counts than blood collected intravenously(Daee et nally, we thank field techniciansTrent Brown and Greg al. 1988, Kayiran et al. 2003). We minimized this Burak, and graduate student Lynda Leppert. potential change in white blood cellsby collecting LITERATURE CITED blood immediately upon puncturing the vein. The H/L ratio was significantly different be- APPLE, F.S. AN•) M.K. McGuE. 1983. Serum enzyme changesduring marathon training. Am.J. Clin. Pathol. tween birds in relatively good, compared to birds 79:716-719. in relativelypoor, condition. Such differencessug- BMLE¾,N.TJ. 1995. StatisticalMethods in Biology,3rd Ed. gest decreased immunocompetence or higher Cambridge Univ. Press,Cambridge, U.K. rates of stressin birds of poor condition. The H/ B^LBONTiN,J. ANt)M. FEP•R. 2002. Plasmachemistry ref- L ratio has been shown to be a less variable indi- erence valuesin free-livingBonelli's Eagle (Hieraaetus cator of stress than differential counts and more fasciatus)nestlings. J. RaptorRes. 36:231-235. reliable than corticosterone levels in reflecting BV•LS,J.W. AND C.E. H^RP•S. 1996. Northern Goshawk chronic stress (Vleck 2002). nest area monitoring in the Twin FallsRanger District One of the main objectivesof this work was to of the SawtoothNational Forest.Unpubl. Report, Ida- explore the relationships between a noninvasive ho Fish and Game, Boise, ID U.S.A. body condition index and the blood chemistryand BELL,DJ. AN•) P.D. STumca•.1965. Chemical constituents of blood. Pages 32-84 in P.D. Sturkie [ED.], Avian cytologyof wild breeding goshawks.We expected Physiology,2rid Ed. ComstockPubl. Assoc.,Ithaca, NY lessvariability in the blood parameterswe observed U.S.A. and strongerrelationships among the variousphys- BENYON, P.H., N.A. FORBES,AND N.H. H^RCOURT-BROWN. iological measurements and body condition than I996. Manual of raptors, pigeons,and waterfowl. Brit- we detected. Our small sample size undoubtedly ish Small Animal Vet. Assoc. Ltd., Iowa State Univ influenced variability in our measures, and some Press, Ames, IA U.S.A. of the assumptionsunderlying the mass/length BROWN,M.E. 1996. Assessingbody conditions in birds technique we used to assessbody condition have Pages 67-135 in V. Nolan, Jr. and E.D. Ketterson recently been questioned (Green 2001). Further- [E•)s.], Current Ornithology, Vol. 13. Plenum Press, New York, NY U.S.A. more, considering that all birds sampled were (-]AMPBELL,W.W. 1995. Avian hematology and cytology breeding, our "poor" condition birds may not Iowa State Press, Ames, IA U.S.A. have been in poor condition relative to the overall -- ANDFJ. DEIN. 1984. Avian hematology:the basics. wild population. Despite these concerns,our find- Vet. Clin. North Am. Small Anim. Pract. 14:223-248. ing that the relative differencesin body condition C^s^t)o, E., J. B;•BONTiN, AN•) M. F•RRER.2002. Plasma among breeding birds were reflected in the H/L chemistry in Booted Eagle (Hieraaetuspennatus) dur- DECEMBER 2003 GOSHAWK BLOOD AND BODY MEASURES 305

ing breeding season.Comp. Biochem. Physiol. 131:233- growth, and adult health in relation to a central Idaho 241. habitat potential model. M.S. thesis,Boise State Um- DAEE, L.N., S. HALVORSEN, P.M. MATHISEN, AND K. MI- versity,Boise, ID U.S.A. RONSK•.1988. A comparisonbetween haematological HANDRICH, Y., L. NICOLAS, AND Y. LE MEHO. 1993. Winter parameters in 'capillary' and venous blood from starvationin captive common Barn Owls: physiologi- healthyadults. Scand. J. Clin. Invest.48:723-726. cal state and reversible limits. Auk 110:458-469. DETRICH,P.J. AND B. WOODBRIDGE.1994. Territory fidel- HOFFMAN,DJ., J.C. FRANSON,O.H. P^TrEE, C.M. BUNCK, ity, mate fidelity, and movementsof color-marked ANDH.C. MURRAY.1985. Biochemicaland hematolog- Northern Goshawks(Accipiter gentilis) in the southern ical effectsof lead ingestionin nestlingAmerican Kes- Cascades of California. Stud. Avian Biol. 16:130-132. trels (Falcosparverius). Comp. Biochem. Physiol. 80:431- DHABHAR, F.S., A.H. MILLER, B.S. MCEWEN, AND R.L. SPEN- 439. CER.1996. Stress-inducedchanges in blood leukocyte IVINS, G.K., G.D. WEDDLE, AND W.H. HALLIWELL. 1986 distribution: role of adrenal steroid hormones. Im- Hematology and serum chemistry in birds of prey. munology157:1638-1644. Pages 280-290 in M.E. Fowler [ED.], Zoo and Wild FERRER,M. 1990. Hematological studies in birds. Condor Animal Medicine. W.B. Saunders Co., Philadelphia, 92:1085-1086. PA U.S.A. ß 1993. Blood chemistry studiesin birds: some ap- KAYIV,AN, S.M., N. OZBEK, M. TUP,AN, AND B. GURAKAN plications to ecological problemsßTrends Comp. Bio- 2003. Significant differencesbetween capillary and ve- chem.Physiol. 1:1031-1044. nous complete blood counts in the neonatal period. --AND P. DOBADO-BERRIOS.1998. Factors affecting Clin. Lab. Haematol. 25:9-16. plasma chemistryvalues of the SpanishImperial Ea- KENNEDY,P.L. 1997.The Northern Goshawk(Accipitergen- gle, Aquila adalberti.Comp. Biochem. Physiol. 120:209- tilis):is there evidenceof a populationdecline?J. Rap- 217. tor Res. 31:95--106. , T. G^RCIA-RODRIGUEZ,J.G. CARRILLO,AND J. GAS- KNUTH, S.T. AND S.B. CHAPLIN. 1994. The effect of exer- TROVIEJO.1987. Hematocrit and blood chemistryval- cise on plasma activitiesof lactate dehydrogenaseand ues in captive raptors (Gypsfulvus,Buteo buteo, Milvus creatine kinase in Red-tailed Hawks.J. RaptorRes. 28: migrans,Aquila heliaca).Comp. Blochem. Physiol. 87: 27-33. 1126-1127. KONTECr•A,H., J.M. KSI•ZRIEWICZ,^NI) L. NOGOWSRI. FP,ANKLIN, A.B., D.R. ANDERSON,RJ. GUTIERREZ,AND K.P. 1999. Effects of different stressorson laying rate and BURNHAM.2000. Climate, habitat quality, and fitness selected blood indices in reproductive ducks. Atom in northern Spotted Owl populationsin northwestern Feed Sci. Technol. 8:63-72. California.Ecol. Monogz 70:539-590. LaviN, S., R. CUENCA,I. 1VL•co, R. VELARDE,AND L. VIN^ GARCIA-RODRIGUEZ,T., M. FERRER,F. RECIA,AND J. GAS- 1992. Hematology and blood chemistry of the Marsh TROVIEJO.1987. Circadian rhythms of determined Harrier (Circus aeruginosus).Comp. Biochem. Physzol blood chemistry values in buzzards and eagle-owls. 103:493-495. Comp.Biochem. Physiol. 88:663-669. M^URO,L. 1987. Hematologyand blood chemistry.Pages GEE,G.F., J.W. CARPENTER,AND G.L. HENSLER.1981. Spe- 269-275 in B.A. Giron Pendleton, B.A. Millsap, K.W. cies differences in hematological values of captive Cline, and D.M. Bird lEDs.I, Raptor Management cranes,geese, raptors, and quail.J. Wildl.Manage. 45: Techniques Manual National Wildlife Federation Sci- 463-483. ence and Technical Series, No. 10. Port City Press, GRAHAM, D.L. AND W.H. HALLIWELL. 1986. Malnutrition Inc., Baltimore, MD U.S.A. in birdsof prey.Pages 379-385 in M.E. Fowler [ED.], MAXWELL,M.H. 1993. Avian blood leukocyte responses Zoo and Wild Animal Medicine. W.B. Saunders Go., to stress.World's Poult. Sci.J. 49:34-43. Philadelphia, PA UßS.A. NEWTON,I. 1979. Population Ecology of Raptors. T. & GREEN,A. 2001. Mass/length residuals:measures of body A.D. Poyser,London, U.K. condition or generatorsof spuriousresults? Ecology 82: PIERSMA,T. AND N.C. DAVIDSON. 1991. Confusions of mass 1473-1483. and size. Auk 108:441-444. GROSS,W.B. AND H.S. SIEGEL. 1983. Evaluation of the het- RISING,J.D. AND K.M. SOMERS.1989. The measurement erophil/lymphocyte ratio as a measure of stressin of overallbody size in birds. Auk 106:666-674. chickens. Avian Dis. 27:972-979. SAS P•LEASE6.12. 1996. SAS User's Guide. SAS System HALLIWELL,W.H. 1981. Serum chemistry profiles in the Inst. Inc., Carey,NC U.S.A. health and diseaseof birds of prey. Pages111-112 in SMITH,E.E. ^ND M. BUSH.1978. Haematologicparame- J.E. Cooper and A.G. Greenwood [EDs.], Recent Ad- ters on various speciesof strigiformes and falconifor- vancesin the Study of Raptor Diseases.Chiron Publ., mes. Wildl. Dis. 14:447-450. West Yorkshire, U.K. SNYDER,R.L. ANDJ. TERRY.1986. Avian nutrition. Pages HANAUSKA-BROWN,L.A. 2000. Evaluation of Northern 190-200 in M.E. Fowler [ED.], Zoo and Wild Animal Goshawk (Accipiter gentilis) productivity, nestling Medicine. W.B. SaundersCo., Philadelphia, PA U.S A 306 HANAUSKA-BRO•2qET AL. VOL. 37, NO. 4

SQUIRES,J.R. AND R.T REYNOLDS.1997. Northern Gos- VLECK,C.M. 2002. Comparisonof corticosteroneand het- hawk (Accipitergentilis) in A. Poole and E Gill lEDs.I, erophil to lymphocyteratios as indicatorsof stressin The Birds of North America, No. 298. Academy of free-living birds. Pages 401-411 in A. Dawson and Natural Sciences,Philadelphia, PA, and American Or- C.M. Chaturvedi [EDs.], Avian Endocrinology.Narosa nithologists'Union, Washington,DC U.S.A. PublishingHouse, New Delhi, India. STATSOFT, INC. 1998. STATISTIGA for Windows, Com- VLECK, C.M., N. VERTALINO, D. VLECK, AND T.L. BUCHER. puter program manual. StatSoft, Inc., Tulsa, OK 2000. Stress,corticosterone, and heterophil to lym- U.S.A. phocyte ratio in free-living Adelie Penguins. Condor STEELE,R., R.D. PHISTF•R,R.A. R•R, ANDJ.A. I•TTAMS. 102:392-400. 1981. Forest Habitat Types of Central Idaho. USDA YOUNK,J.V. 1996. Breeding ecologyof the Northern Gos- Forest Service,Gem Tech. Rep. INT-114, Ogden, UT U.S.A. hawk in relation to surfacegold mining in naturally fragmented aspen forests of northern Nevada. M.S. STEIN,R.W., J.T. YAMAMOTO,D.M. FRY,AND B.W. WILSON. thesis,Boise State University,Boise, ID U.S.A. 1998. Comparativehematology and plasmabiochem- istry of Red-tailed Hawks and American Kesu-elwin- ZAR,J.H. 1996. BiostatisticalAnalysis, 3rd Ed. Prentice tering in California.J. RaptorRes. 32:163-169. Hall, Upper Saddle Rivm; NJ U.S.A. VAN HORNE,B. 1983. Density as a misleadingindicator of habitat quality.J. WildkManage. 47:893-901. Received9 July 2002; accepted28 July 2003