ume 38 Nuinbef 4 Decdfhber 20 Published by j ■ -i T*1 vF- ■ ^mimfr-L. ■ r -J r” 4i J The Raptor Research Foundation, Inc. THE RAPTOR RESEARCH FOUNDATION, INC. (Founded 1966) http: //biology.boisestate.edu/ raptor/ OFFICERS SECRETARY: Judith Henckel TREASURER: Jim Fitzpatrick BOARD OF DIRECTORS INTERNATIONAL DIRECTOR #3; Steve Redpath DIRECTOR AT lARGE #1; Jemima ParryJones DIRECTOR AT I ARGE #2: Eduardo Inigo-Eeias DIRECTOR AT LARGE #3: Michael W. Coli.opy DIRECTOR AT LARGE #4: Carol McIntyre DIRECTOR AT LARGE #5: John A. Smallwood DIRECTOR AT lARGE #6: Daniel E. Variand Ruth Tingay NORTH AMERICAN DIRECTOR #1: Jeff Smith NORTH AMERICAN DIRECTOR #2: Gary Santolo NORTH AMERICAN DIRECTOR #3: Ted Swem INTERNATIONAL DIRECTOR #1 : Beatriz Arroyo INTERNATIONAL DIRECTOR #2: PRESIDENT: Brian A. Millsap VICE-PRESIDENT: David M. Bird EDITORIAL STAFF EDITOR: James C. Bednarz, Department of Biological Sciences, P.O. Box 599, Arkansas State University, State University, AR 72467 U.S.A. ASSISTANT EDITOR: Jennifer L. Norris ASSOCIATE EDITORS James R. Belthoef Juan Jose Negro Clint W. Boat Marco Restani Cheryl, R. Dykstra Eabrizio Sergio Michael I. Goldstein Ian G. Warkentin Joan L. Morrison James W. Watson BOOK REVIEW EDITOR: Jeffrey S. Marks, Montana Cooperative Research Unit, University of Montana, Missoula, MT 59812 U.S.A. SPANISH EDITORS: Carlos Daniel Cadena, Luc jo R. Malizia, Cintia Cornelius EDITORIAL ASSISTANT: Joan Clark The Journal of Raptor Research is distributed quarterly to all current members. Original manuscripts dealing with the biology and conservation of diurnal and nocturnal birds of prey are welcomed from throughout the world, but must be written in English. Submissions can be in the form of research articles, short communications, letters to the editor, and book reviews. Contributors should submit a typewritten original and three copies to the Editor. All submissions must be typewritten and double-spaced on one side of 216 X 278 mm (814 X 11 in.) or standard international, white, bond paper, with 25 mm (1 in.) mar- gins. The cover page should contain a title, the author’s full name(s) and address (es). Name and address should be centered on the cover page. If the current address is different, indicate this via a footnote. A short version of the title, not exceeding 35 characters, should be provided for a running head. An abstract of about 250 words should accompany all research articles on a separate page. Tables, one to a page, should be double-spaced throughout and be assigned consecutive Arabic numer- als. Collect all 6gure legends on a separate page. Each illustration should be centered on a single page and be no smaller than final size and no larger than twice final size. The name of the author (s) and figure number, assigned consecutively using Arabic numerals, should be pencilled on the back of each figure. Names for birds should follow the A.O.U. Checklist of North American Birds (7th ed., 1998) or another authoritative source for other regions. Subspecific identification should be cited only when pertinent to the material presented. Metric units should be used for all measurements. Use the 24-hour clock (e.g., 0830 H and 2030 H) and “continental” dating (e.g., I January 1999). Refer to a recent issue of the journal for details in format. Explicit instructions and publication policy are outlined in “Information for contributors,”/. Raptor Res., Vol. 38(4), and are available from the editor. Submit manuscripts to J. Bednarz at the address listed above. COVER: Nesting Red-shouldered Hawk (Buteo lineatus). Painting by Nicole Perretta; for more informa- tion and images, visit www.hawkandjaguar.com Contents Responses of Nesting Bald Eagles to Experimental Pedestrian Activity. James W. Watson 295 Dispersal and Mortality of Red-shouldered Hawks Banded in Ohio. Cheryi r. Dykstra, Jeffrey L. Hays, Melinda M. Simon, John B. Holt, Jr., G. Ronald Austing, and F. Bernard Daniel 304 Decreases in a Population of Red-shouldered Hawks Nesting in Central Maryland. Elwood M. (Woody) Martin 312 Seasonal Variation in Sex Ratio of Nestling Eleonora’s Falcons. Dietrich Ristow and Michael Wink 320 Tawny Fish-Owl Predation at Fish Farms in Taiwan. Yuan-Hsun Sun, Hsin ju Wu, and YmgWang 326 Geographic Variation in Morphology of Four Species of Migratory Raptors. Elise Vernon Pearlstine and Daniel B. Thompson 334 Home-range Size of the Javan Hawk-Eagle (Spizaetus bartelsi) Estimated from Direct Observations and Radiotelemetry, jan Ove Gjershaug, Nils R0v, Torgeir Nygard, Dewi M. Prawiradilaga, M. Yayat Afianto, Hapsoro, and Adam Supriatna 343 Population Status and Reproductive Performance of Eurasian Griffons ( Gyps FULVUS) IN Eastern Spain. Pascual Lopez-Lopez, Clara Garcfa-Ripolles, and Jose Verdejo 350 Short Communications Gender Determination in the Swainson’s Hawk (Buteo swainsoni) Using Molecular Procedures and Discriminant Function Analysis. Jose Heman Sarasoia and Juan Jose Negro 357 Productivity and Fledgling Sex Ratio in a Cinereous Vulture {Aegypius MONACHUS) Population in Spain. Auxiliadora Villegas, Juan Manuel Sanchez-Guzman, Emilio Costillo, Casimiro Corbacho, and Ricardo Moran 361 Nest Provisioning of the Oriental Honey-buzzard {Pernis ptilorhynchus) in Northern Taiwan. Kuang-Ying Huang, Yao-Sung Lin, and Lucia Liu Severinghaus 367 Winter Diet of the Greater Spotted Eagle {Aquila clanga) in the Amvrakikos Wetlands, Greece. Haralambos Alivizatos, Dimitris Papandropoulos, and Stamatis Zogaris 371 Gender Determination of Eurasian Eagle-Owls {Bubo bubo) by Morphology. Maria del mar Delgado and Vincenzo Penteriani 375 Letters A Possible Case of Double Brooding of Eagle-Owls {Bubo bubo) in Spain. Joaquin Ortego 378 Insect Hawking Observed in the Long-eared Owl {Asio otus). Darren J.H. Sleep and Rowan D.H. Barrett 379 Osprey Scavenges Common Murre Carcass in Coastal Washington. Joseph B. Buchanan 380 How Long is Too Long? A Case of Fostering Nestling Bonelli’s Eagles {Hieraaetus fasciatus) . Satish A. Pande, Amit P. Pawashe, Banda Pednekar, Anil Mahabal, and Reuven Yosef 381 Information For Contributors 383 Index to Volume 38 387 The Raptor Research Foundation, Inc. gratefully acknowledges funds and logistical support provided by Arkansas State University to assist in the publication of the journal. THE JOURNAL OF RAPTOR RESEARCH A QUARTERLY PUBLICATION OF THE RAPTOR RESEARCH FOUNDATION, INC. VoL. 38 December 2004 No. 4 J Raptor Res. 38 (4) ;295-303 © 2004 The Raptor Research Foundation, Inc. RESPONSES OE NESTING BALD EAGLES TO EXPERIMENTAL PEDESTRIAN ACTIVITY James W. Watson ^ Wildlife Program, Washington Department of Fish and Wildlife, 600 Capitol Way North, Olympia, WA 98501 U.S.A. Abstract. — In 1993 and 1994, I tested effects of pedestrian activity on Bald Eagles {Haliaeetus leucoce- phalus) at nests on 21 territories in western Washington. A pedestrian walked around each nest 10 min/ hr for 6 hr during 65 trials. Eagles averaged 8.0 ± 1.8 (SE) responses/trial (i.e., per hr of pedestrian activity, N = 524 combined responses), and 10.7 ± 2.0 min of response time/trial {N = 681 min combined response time). Responses accounted for 23 ± 3% by frequency, and 3 ± 0.3% by time, of all perch behaviors on treatment days. I observed no damage or injury to eggs or young from the encounters, although nest flushes had a mean of 3.1 ± 0.9 responses/ trial. Treatments resulted in a two-fold increase in mean egg exposure time/trial (x = 7.8 ±1.6 min/exposure) compared to controls (x = 3.3 ± 0.8 min/exposure). Precipitation reduced the time adults left eggs and young exposed from a mean of 7.9 ± 0.9 min/hr to 5.1 ± 1.0 min/hr on control days. Nest height, nest screening, pedestrian distance, and phenology affected eagle responses; responses were substantially reduced at nests that were >40 m high and highly-screened, when pedestrian distance to nests increased from 60-120 m, and during incubation compared to the brood period. To reduce risks from increased exposure of eggs and young, I recommend that pedestrian activities be restricted near Bald Eagle nests during incubation and the first 3 wk of brooding. Eor Bald Eagle nests in forested, non-pristine areas of residential devel- opment, pedestrian activity less than 120 m from nests can be restricted as a function of nest height and screening to minimize disturbance. Key Words: Bald Eagle', Haliaeetus leucocephalus; behavior, disturbance response, human activity; Washington. RESPUESTA DE HALIAEETUS LEUCOCEPHALUS NIDIFICANTES ANTE ACTIVIDAD PEATONAL EXPERIMENTAL Resumen. — En 1993 y 1994, evalue los efectos de la actividad de peatones sobre aguilas Haliaeetus leu- cocephalus en nidos ubicados en 21 territorios en el oeste de Washington. Un peaton camino alrededor de cada nido 10 min/hr por 6 hr durante 65 ensayos. Las aguilas presentaron en promedio (±SE) 8.0 ± 1.8 respuestas por ensayo(i.e., por hora de actividad peatonal, N = 524 respuestas combinadas), y 10.7 ± 2.0 min como tiempo de respuesta/ensayo {N = 681 min combinando el tiempo de respuesta). Del total de comportamientos de percha en los dias de estudio, las respuestas correspondieron al 23 ± 3% en terminos de frecuencia y al 3 ± 0.3% en terminos de tiempo. No observe danos o lesiones en los pichones o huevos como consecuencia de los encuentros, aunque en promedio las aves fueron espantadas del nido en 3.1 ± 0.9 ocasiones por ensayo. Los tratamientos condujeron a un incremento del doble en el tiempo promedio de exposicion de los huevos (x = 7.8 ±1.6 min/exposicion) en comparacion con controles (x = 3.3 ± 0.8 min/exposicion). La precipitacion redujo el tiempo por el cual los adultos dejaron expuestos los huevos y pichones de un promedio de 7.9 ± 0.9 min/hr a 5.1 ± 1.0 min/hr en dias de control. La altura del nido, su nivel de proteccion, la distancia al peaton y la fenologia afectaron la respuesta de las aguilas; las respuestas fueron sustancialmente reducidas en nidos que estaban ubicados a mas de 40 m de altura y altamente protegidos, cuando la distancia al peaton se ^ E-mail address; Watson@valleyint.com 295 296 Watson VoL. 38, No. 4 incremento de 60-120 m, y durante la incubacion, en comparacion con el periodo de erapollamiento. Para reducir los riesgos del incremento en la exposicion de los huevos y pichones, recomiendo que las actividades peatonales scan restringidas cerca de los nidos de H. leucocephalus durante la incubacion y las tres primeras semanas de cria de los pichones. Para areas forestales no pristinas con desarrollo residencial, la actividad de los peatones a menos de 120 m de los nidos puede restringirse en funcion de la altura y proteccion de los nidos para minimizar el disturbio. [Traduccion del equipo editorial] As Bald Eagle {Haliaeetus leucocephalus) popula- tions recover, increasing numbers of eagles are nesting in landscapes subject to human activities and habitat alteration. Post-recovery persistence of some populations, such as in Florida and the Ches- apeake Bay region, may be directly tied to the abil- ity of eagles to reproduce successfully in urban and residential environments (Buehler et al. 1991). Hu- man impacts are common in Washington, where m the past 25 yr increasing eagle populations and residential development in Puget Sound led to the development of >1150 eagle management plans with private landowners (Stinson et al. 2001). Man- agement plans attempt to minimize effects of hu- man activities primarily by imposing restrictions near nests during the nesting period, maintaining nest screening, and maximizing distances between human activities and nesting eagles. The effective- ness of management planning depends on our ability to understand how Bald Eagles respond to human activities near residential developments. Behavioral changes of Bald Eagles that might precede nest failure often are not studied because they are difficult and time-consuming to assess (Anthony and Isaacs 1989). Human activities can change behavior of nesting eagles through audi- tory and visual disturbance (Grier 1969, Fraser et al. 1985, Grubb and King 1991, Steidl and Anthony 1996). Such activities may cause reductions in in- cubation time, brooding time, feeding time, and other adult behaviors that affect attendance of eggs or young, and potentially survival of young. These changes in adult behavior may be possible indicators of impending nest failure of raptors (Holthuijzen et al. 1989). Experimental studies of human activities on Bald Eagles provide an opportunity to understand how eagles respond to different types of disturbance and variable disturbance parameters (e.g., dis- tance) in controlled situations. Fraser et al. (1985) were hrst to evaluate experimentally flush distanc- es of nesting Bald Eagles in response to a pedes- trian approaching the nest directly, and made rec- ommendations for buffer zones based on those distances. McGarigal et al. (1991) experimentally tested responses of foraging eagles to stationary boating activity and demonstrated that Bald Eagles were affected by passive human activities. More re- cently, in interior Alaska, Steidl and Anthony (2000) found experimental recreational activity re- sulted in signihcant behavioral changes to Bald Ea- gles nesting, potentially leading to reduced nest- ling survival. To better understand the impact of human ac- tivities on behavior and reproduction of Bald Ea- gles in rural areas undergoing residential devel- opment, I conducted an experiment in western Washington in 1993-94. I used a walking pedestri- an to simulate the most common human activity in the rural environment, and recorded the fre- quency and duration of Bald Eagle responses to the pedestrian during different nest stages, at dif- ferent nest heights, with different degrees of screening cover, and at varying distances from nests. Analysis of these factors provided guidelines for reducing the effects of human activity in areas of residential development. Study Area The Puget Sound region in northwest Washington is characterized by diverse saltwater, brackish, and fresh- water ecosystems. Bald Eagles nest in coniferous stands dominated by Douglas-fir (Pseudotsuga menziesii) along marine shorelines, and in riparian stands dominated by black cottonwood {Populus trichocarpa) along lakes and rivers. Nests are typically placed in dominant or codom- inant trees in forest stands. High annual precipitation (e.g., 100-150 cm) results in closed forest canopies and dense understories, except where stands have been re- cently logged, or trees have been limbed. Thus, visibility of human activities from eagle nests varies. Land devel- opment in recent years has resulted in increased residen- tial activities near nesting eagles (D. Stinson, J. Watson, and K. McAllister unpubl. data). The statewide Bald Ea- gle population has increased exponentially in the past 25 yr and growing numbers of eagles are nesting along Pu- get Sound (Watson et al. 2002), where humans common- ly approach within 400 m of Bald Eagle nests (unpubl. data) . Methods Eagle responses to human activity were assessed by comparing eagle behaviors when a pedestrian was pres- December 2004 Bald Eagle Response to Pedestrians 297 em (treatments) and absent (controls). Six-hour treat- ments and controls were applied to each eagle nest on consecutive days beginning at dawn. The order of each trial was randomized (treatment-control vs. control- treatment). During each treatment, a person walked a circular path around the nest tree once/hr (i.e., six times/ treatment) to simulate naturally occurring pedes- trian activity near the eagle nest, as opposed to direct nest approaches which are atypical (pers. obs.). An ob- server recorded the following information: duration (sec) of the pedestrian activity; frequency, duration, and type of eagle disturbance responses (i.e., none, flush and reperch, perch, vocalize, return flight to nest, fly in and respond, surveillance flight, flush and soar, flush and re- spond, redirected aggression toward conspecific); and perch locations and flight paths of eagles on 1:12000 ae- rial photos. Photos were used for measuring the distance of the eagle to the pedestrian at the point of response. On both control and treatment days the observer record- ed time and duration (sec) of all behaviors for both adult eagles <200 m from the nest tree (i.e., nest building, incubating, brooding, feeding young, resting), and at dis- tances >200 m where visibility allowed. Behavior fre- quencies and duration were summed for both adults on each territory because response was sometimes cued by the behavior of the other adult. Wind speed (low = calm or slight breeze; high = brisk or gusty winds) , cloud cover (<50% vs. >50%), and precipitation (yes/no) were re- corded. 1 conducted trials (i.e., one control-treatment on con- secutive days) at a 60-m radius from each nest once dur- ing incubation and once during the first 3 wk of brood- ing (i.e., 42 trials). Trials conducted at the same nest were separated by >5 wk to minimize possible effects of repeated treatments (e.g., habituation) on responses. To assess the effect of pedestrian distance, 1 repeated the experiment at the seven nests with partial screening dur- ing incubation and brooding at 30 and 120 m (i.e., 28 trials). This resulted in a total of two trials/nest at each of the three distances. 1 standardized the total time and distance traversed by pedestrians among treatments by walking the circumference of the 120-m distance once, the circumference of the 60-m distance twice, and cir- cumference of the 30-m distance four times. Duration of pedestrian activity was ca. 10 min long (x duration = 9.0 ± 2.5 min [SD]) for a total of 60 min of activity/ 6-hr treatment, with variability due to differences in ground cover and topography. 1 described screening cover at nests by scoring nest visibility as high (0), partial (50), and low (100) at 30 m, 60 m, 120 m, and 200 m along four transects at each cardinal direction from the nest. Screening was assessed prior to leafout in January and after leafout in May, and categorized by the mean score as: <33 = little or no screening; 34—66 = partial screening; or ^67 = high screening. Seven nests were identified for each screening category. Height of each nest was measured with a cli- nometer and categorized (20-29 m, 30-39 m, 40-49 m, >49 m) for analysis. Ne.sts on 21 Bald Eagle territories were studied; 12 ad- jacent to Puget Sound, six along rivers, and three at lakes. These were randomly selected from 323 nests with one to five homes <400 m away, which was typical of human activity levels near eagle nests in Puget Sound. I did not select nests that were at remote locations isolated from human activities because my intent was to provide rec- ommendations for Bald Eagles with moderate prior ex- posure to human activity. Nests were chosen in settings with varied vegetative screening that still afforded an ob- server, stationed >400 m away, good visibility of the area <200 m from the nest tree. Data Analysis. 1 summarized the frequency and dura- tion of eagle disturbance responses and perch behaviors during controls and treatments by computing the grand mean and standard error from means computed at each nest. Paired /5-tests were used to evaluate the significance of changes in responses and behaviors between controls and treatments. 1 used multivariate analysis of variance to test effects of nest height, nest screening, distance to pedestrian, and nest phenology on standardized frequencies and dura- tions of three eagle disturbance responses (total respons- es, nest flushes only, and combined nest responses in- cluding flushes, perch alert, and vocalization) and six eagle behaviors (perching <200 m from nest, nest build- ing, incubation, brooding, feeding young, resting). Ef- fects of treatments on eagle responses and behaviors were computed as the arithmetic difference between con- trols and treatments. I tested full models that included interactive effects. If no interactions were identified (PS: 0.10), effects found to be significant were tested in re- duced models, and factor level effects were identified with the Bonferroni’s method (Miller 1981). I used Rests to compare mean exposure time (sec) for eggs and young on control days, on treatment days while the pedestrian was present, and on treatment days after the pedestrian activity. Also, Rests were used to compare exposure times (sec) for eggs and young on control days among categories of precipitation, wind, and cloud cover 1 did not assess the effects of weather on egg and young exposure during pedestrian activity because I intention- ally avoided conducting trials on days with inclement weather. Repeated measures analysis of variance was used to assess if eagles habituated to treatments among the six exposures to the pedestrian on treatment days. Where appropriate, variables were tested to verify the assumption of normality. No transformations were nec- essary for variables deemed significant in final models Significance level for all tests was a = 0.05. Results Behavioral Responses. I conducted 65 of 70 scheduled trials during the 2-yr study (nest failures eliminated five brooding trials). Disturbance re- sponses accounted for 23 ± 3% (SE) by frequency and 3 ± 0.3% by time, of all behaviors on treat- ment days. Eagles had a mean of 8.0 ±1.8 distur- bance responses/ nest for each trial, or each hr of pedestrian activity (N — 524 combined responses) . The typical response sequence was a flush from the nest, followed by reperching on the nest, and even- tual resumption of pretreatment perch behavior. Eagles flushed and perched a mean of 3.1 times/ 298 Watson VoL. 38, No. 4 Table 1. Bald Eagle responses to experimental pedestrian activity in northwestern WA, 1993—94. Response rates (number of responses/nest, N = 21) are standardized per trial (1 hr of pedestrian activity). Response FREQUENCV/hr Duration (sec/hr) Mean SE Mean SE Flush and reperch 3.1 0.9 123.2 39.1 Perch or vocalize 2.7 0.9 271.4 67.8 Return flight to nest 1.1 0.3 74.2 29.5 Fly in and respond 0.2 0.1 25.2 12.1 Surveillance flight 0.2 0.1 64.1 56.3 Flush and soar 0.1 0.1 37.4 23.8 Flush and respond 0.2 0.1 8.3 3.8 Redirected aggression 0.2 0.1 33.0 22.8 nest (±0.9) during each trial (Table 1). Eagles av- eraged 10.7 (±2.0) min of response time/nest for each hr of pedestrian activity {N =681 min com- bined response time), and typically perched or vo- calized half of that time (Table 1 ) . Eagles were initially perched on the nest a mean of 48 ± 1% of time when pedestrian activity was initiated during trials. For eagles on nests, the mean encounter distance for flush responses was 76 ± 9 m/nest, and 72 ± 9 m/nest for combined responses. For eagles perched anywhere on the ter- ritory, the mean encounter distance for flush re- sponses was 72 ± 10 m/nest, and 71 ± 10 m/nest for all responses. Pedestrian activity caused changes in eagle be- havior, manifested as increased frequency of com- bined perch behaviors <200 m from nests (Table 2). Accordingly, frequency of pedestrian-induced responses increased, while frequency of nest build- ing decreased. Pedestrian activity had less of an ef- fect on duration of combined behaviors (Table 2). Although incubation time per trial was not re- duced by pedestrian activity, duration of individual egg exposures/ trial was longer {t = 3.39, P = 0.009, N = 21) during treatments (A = 7.8 ± 1.6 min/exposure; range = 0.3-59.0) compared to controls (x — 3.3 ± 0.8 min/exposure; range = 0.1-110.2), and was also longer {t = 9.25, P < 0.0001) than egg exposures immediately after treatments (A = 1.9 ± 0.5 min/exposure; range = 0.1—27.1). Mean dnration of posttreatment egg ex- posure was partially dependent on mean duration of treatment exposure; eagles reduced posttreat- ment egg exposures by 30 sec for every 5 min in- crease in egg exposure during treatments (linear regression, r = 0.29, P = 0.018). During incuba- tion, eagles responded a mean of 23 ± 5% of the time the pedestrian was present (14 ± 3 sec/min of experimental human activity) . Total brood time decreased due to pedestrian activity (Table 2), but time that young were exposed/ trial during the brood period was the same {P — 0.101) for treat- ments {x = 33.1 ± 9.4 min/exposure; range = 1.3- 316.7), controls (x = 35.4 ± 14.1 min/exposure; range = 0.1-360.0), and immediately after treat- ments {P = 0.432; X — 29.5 ±11.4 min/exposure; range = 0.1-319.4). For nests with young {N= 19), eagles responded a mean of 43 ± 10% of the time the pedestrian was present (26 ± 6 sec/min). I did not observe direct effects to eggs due to treatments (e.g., eggs rejected or broken by flush- ing adults). Hatching success was unaffected at a nest where eggs were exposed for 59 min. Direct effects from weather were mixed; total time eggs and young were exposed was higher {t — 2.05, P = 0.045) in the absence of precipitation (7.9 ± 0.9 min exposure/hr) than when there was some level of precipitation (5.1 ± 1.0 min exposure/hr). There was no effect of wind (P = 0.325) or cloud cover (P = 0.370) on exposure time. Nest success averaged 1.14 young/ occupied territory during the experiment, and 1.11 young/occupied territory in 1994, the year following the experiment. All nine territories treated in 1993 were reoccupied in 1994. The presence of other raptors increased eagle responsiveness. Seventy-eight encounters with con- specihcs and four encounters with Red-tailed Hawks {Buteo jamaicensis) were recorded during tri- als. Sixty-six encounters (85%) with other eagles involved a non-incubating or non-brooding adult chasing an intruder. Eight encounters (10%) were of an incubating or brooding bird vocalizing or standing above the eggs or young in response to December 2004 Bai,d Eac;le Response to Pedestrians 299 Table 2. Effects of pedestrian activity on the frequency and duration of Bald Eagle behaviors during 65 experimental trials. Each trial consisted of a 6-hr control and a 6-hr treatment in which pedestrian activity was conducted <200 m away for 10 min/hr. Effects were examined with paired t-tests. CONTROI. Treatment Percenf Behavior Grand x SE Grand x SE Change t P Frequency All perch behaviors Nest building/ main- 25.8 1.4 30.8 2.1 19.4 2.36 0.029 tenance 9.9 0.9 7.2 0.8 -27.3 3.11 0.006 Incubate 9.4 0.7 8.6 0.7 -8.5 1.43 0.169 Brood 6.5 0.7 6.2 0.6 -6.2 0.62 0.544 Feed young 3.9 0.4 3.7 0.5 -5.1 0.41 0.698 Rest 4.6 0.7 5.2 0.7 13.0 1.09 0.289 Response^ 0.01 0.01 8.0 1.8 >100 4.46 0.0003 Duration All perch behaviors Nest building/ main- 456.6*^ 17.8 439.3 16.8 -3,9 0.99 0.333 tenance 33.9 6.7 28.7 6.7 -15.3 0.71 0.484 Incubate 324.8 9.7 318.9 7.0 -1.8 1.19 0.247 Brood 223.2 21.0 178.7 22.4 -19.9 2.25 0.038 Feed young 26.8 3.3 33.0 5.6 23.1 0.95 0.355 Rest 108.9 20.0 103.0 17.4 -5.4 0.13 0.902 Response^ <0.001 <0.001 10.7 2.0 >100 5.31 <0.0001 Pooled disturbance responses included flushes, standing in alert posture before and after flushing, and flight (evasion, surveillance, or aggression) before eagles resumed pre-disturbance activity. ^ Behavior of both adults on each territory was pooled so duration of all perch behaviors exceeded 6 hr. Figure 1. Effect of nest height on responses of Bald Ea- gles to experimental pedestrian activity at 21 nests in western Washington. Response rates (number of respons- es or min of response/hr of activity) were compared with the Bonferroni procedure. Responses included flushes, alert posture, and flight before eagles resumed previous activity. Means with different shading are statistically dif- ferent (error bars = SE). an intruding eagle. The remaining four encoun- ters (5%) involved incubating or brooding adults engaged in physical aggression with an intruder on the nest, or leaving the eggs or young unattended while pursuing an intruding adult. Two encounters on the nest occurred during treatments. Eagles did not habituate during six pedestrian exposures on a treatment day based on frequency {P = 0.192) or time (P = 0.663) of responses. Effects of Nest Height, Screening, Distance, and Nesting Stage. No interactions were identified be- tween the four factors on nine eagle behaviors (P > 0.116), but individual factor effects were signif- icant. Nest height mitigated the effects of pedes- trians near nests; four eagle responses were re- duced at nests >29-m high (Fig. 1). Combined responses of eagles on the nest were reduced to 40-high, and eagles did not flush from nests >50-m high. Nest screening also ameliorated effects of pedestrian ac- tivity for three responses (Fig. 2). There was a mean decrease of three responses/hr between low and partially-screened nests, and a further reduc- 300 Watson VoL. 38, No. 4 F = 3.74, P = 0.030 Low Partial High 1 1 Figure 2. Effect of nest screening on responses of Bald Eagles to experimental pedestrian activity at 21 nests in western Washington. Response rates (number of responses or min of response/hr of activity) were compared with the Bonferroni procedure. Responses included flushes, alert posture, and flight before eagles resumed previous activity. Means with different shading are statistically different (error bars = SE). tion of five responses/hr when nests were highly- screened. Highly-screened nests also significandy reduced eagle response time and nest flushes com- pared to other screening classes. Pedestrian dis- tance to nests only affected eagle response time (^ 8,56 ~ 4.01, P = 0.024). Eagle response time at 30 m (x = 19.9 ± 3.7 min/hr of activity) was not significandy different from 60 m (x = 18.4 ± 2.8 min/hr of activity), but both were significandy greater than at 120 m (x = 11.9 ± 4.4 min/hr of activity). Nest stage only affected total responses (7^8,56 = 7.37, P = 0.009) . Eagles responded almost twice as often during the brood period (x - 6.b ± 1.4 responses/hr of activity) compared to incuba- tion (x = 3.3 ± 0.7 responses/hr of activity). At given levels of screening and nest height the predicted eagle response frequency was reduced by a mean of 1.6/hr when pedestrian distance in- creased from 60-120 m (Table 3). At nests 20-29 m in height, response frequencies were reduced up to 18% at specific screening levels when dis- tance increased from 60-120 m. At nests 30-39 m in height, response frequencies were reduced up to 33% at specific screening levels when distance increased to 120 m. At nests >40 m in height, re- sponses were reduced from 50-100% as pedestrian distance increased from 60-120 m. Discussion Successful management of Bald Eagle habitat de- pends, in part, on identifying factors that affect ea- gle responses to human activity, and understanding how to manipulate those factors to reduce their impacts. This study illustrates that in non-pristine, forested environments, nest height, and vegetative screening are important factors for mitigating ea- gle responses to brief exposures of a single pedes- trian. Height of nests in which eagle responses were reduced to insignificant levels (i.e., >40 m) was considerably less than average response dis- tance to pedestrians (i.e., 72 m). This suggests that nest height affords eagles a more substantial buffer against disturbance than horizontal distance. Nest height was a significant predictor of call rates, dive rates, and minimum approach distances that Red- tailed Hawks exhibited toward an observer (An- dersen 1990). Flushes from nests are potentially the most detrimental eagle response, and this re- Table 3. Predicted reduction in mean frequency (95% Confidence Interval) of Bald Eagle responses per hr of pedestrian activity when distance to the nest is increased from 60-120 m (iV = 21 nests). Nest Screening Nest Low Partial High Height (m) 60 m 120 m 60 m 120 m 60 m 120 m 20-29 11.5 (2.1) 9.9 (2.8) 10.2 (1.8) 8.6 (2.7) 8.9 (3.1) 7.3 (3.6) 30-39 7.4 (2.1) 5.8 (2.8) 6.1 (1.1) 4.5 (2.2) 4.8 (2.2) 3.2 (3.0) >40 3.3 (2.8) 1.7 (3.3) 2.0 (1.6) 0.4 (2.4) 0.7 (2.0) 0.0 (0.9) December 2004 Bald Eagle Response to Pedestrians 301 sponse was ameliorated only by nest height and vegetative screening. Screening had marked effects on the frequency of flush and flight responses for eagles both on and away from the nest. Earlier re- search found nesting eagles responded to human activity when the activity first became visible (Grubb et al. 1992, Steidl and Anthony 1996). Par- tial screening reduced response time by 32% com- pared to open nests, and in some situations ap- peared to allow just enough visibility of the pedestrian to alert the eagle, but prevent it from monitoring pedestrian activity, resulting in a flush. Seventy-two percent of surveillance flights, when eagles circled the nest site during and after treat- ments, involved situations with the nest partially screened. Manipulation of encounter distance produced less dramatic changes in responses. Mean response distance (71 m) and flush response distance (72 m) of eagles I studied were similar to breeding ea- gles at 33 nests in western Washington that flushed in response to humans in nonexperimental set- tings at a mean distance of 86 m, and exhibited alert responses at 143 m (J. Watson unpubl. data). Similarly, camping <100 m from Bald Eagle nests affected eagle behavior, compared to >500 m away (Steidl and Anthony 2000). The distances that breeding eagles in Washington responded to pe- destrians were less than has been documented in other studies in the United States (e.g., 185 m, Grubb et al. 1992; 275 m, Grubb and King 1991; 57-991 m, Fraser et al. 1985), possibly reflecting a higher degree of habituation to human activities by eagles in western Washington. Studies which evaluated six characteristics of Bald Eagle respons- es in Arizona and Michigan, found distance to be the most important characteristic of pedestrian dis- turbance, followed by duration or sound of activity, and then visibility (Grubb and King 1991, Grubb et al. 1992). Nest stage affected eagle tolerance to pedestrian disturbance with brooding eagles responding twice as long, on average, than incubating eagles. For both periods most response time was spent passive- ly perching following flushing, prior to resuming the previous activity. During incubation, eagles usu- ally sat more tightly on nests, resulting in fewer responses while the pedestrian was present. After treatments, incubation resumed more quickly than did brooding. Bald Eagles exhibited the same dif- ferences in nesting responses to close helicopter approaches in northwest Washington (Watson 1993). Control comparisons for exposure of eggs (3 min) and young (57 min) showed eagles exhib- ited a natural decrease in tenacity to nests through- out nesting, which amplified the time they spent off nests in the brood period. Eaglets begin to ther- moregulate at ca. 15 d (Bortolotti 1984) and re- quire less brooding thereafter. Coincident increas- es in daytime temperatures and decreasing precipitation throughout nesting may have pro- gressively reduced the need for parental atten- dance. Wetter and colder days in early spring cor- responded to increased time on eggs and small young after exposures. Such compensation is im- portant during incubation to reheat cooled eggs of nesting raptors, especially during frequent expo- sures between 30-60 min that lower egg tempera- tures to <35°C (Fox 1995). I found mean egg ex- posure times decreased 3 min/hr in the absence of pedestrian activity, with an additional reduction in exposure time of 3 min/hr on days with precip- itation. Captive Bald Eagles reduced the length of time they left eggs exposed from ca. 2 min/hr to 0.6 min/hr at ambient temperatures <7.2°C and wind velocities >16.2 km/hr (Gerrard et al. 1979). Thus, Bald Eagles exhibit flexibility in their incu- bation strategies depending on human disturbance and weather conditions. This flexibility probably explains the unexpectedly long exposure times for eggs even on control days (e.g., maximum 110 min). Among nesting pairs of Bald Eagles there is a wide range of tolerance to pedestrians that may result from certain pairs being more habituated to higher existing human-activity levels on their ter- ritories (McGarigal et al. 1991). Research on rap- tors and American Crows {Corvus brachyrhynchos) indicated that increased human interactions from urbanization influenced bird behavior in such a way that they have become more tolerant of hu- man intrusion near nests (Newton 1979, Knight et al. 1987, Grubb et al. 1992). However, I did not find that eagles habituated to experimental activity during a treatment day. Steidl and v\nthony (2000) found short-term (e.g., daily) habituation of eagles, but no long-term (e.g., weekly) habituation to hu- man activity levels. Different tolerance limits of in- dividuals may, in part, result from past experience and nestling imprinting (Newton 1979, Harmata 1984). This study illustrated the potential for detrimen- tal behavioral responses of Bald Eagles during brief exposures to human activity. Because intensity of 302 Watson VoL. 38, No. 4 pedestrian activity during the study (10.0 min/hr for 6 hr) was in the range observed in typical cir- cumstances in western Washington (J. Watson un- publ. data) , I believe the results are representative of existing conditions. The need to reduce human activities that elicit eagle responses is based on the assumption that such efforts will increase survival and productivity (e.g., Fraser 1981, Grier and Fyfe 1987, Anthony and Isaacs 1989, Steidl and Anthony 2000). I did not detect damage or injury to eggs or young that occurred when adults were flushed from nests (Grier and Fyfe 1987, Yates and Mc- Clelland 1989); although nearly half of the eagles were on nests during pedestrian activity and flush- es from nests constituted half of the responses. I observed no predation of eggs or young (Yates 1989), although exposure during two treatment episodes provided the opportunity for conspecifics to attack young, and for adults to incidentally harm them during violent agonistic encounters. A two- fold increase in mean egg exposure time increased the likelihood that eggs would be affected by cool- ing, overheating, or loss of moisture (Gerrard and Bortolotti 1988). A mean reduction in incubation time by 14 min/hr was related to nest failure at 40 Bald Eagle nests in western Washington ( J. Watson unpubl. data). Management Recommendations Effects of pedestrian activity <120 m from nests on Bald Eagles with some previous exposure to hu- man activities in non-pristine, rural environments can be reduced by regulating the distance of dis- turbance as a function of nest height and screen- ing vegetation on a site specific basis (Table 3). Guidelines to promote reduced disturbance in- clude maintaining and enhancing vegetative and topographic features that totally screen Bald Eagle nests from pedestrian activities. A high degree of screening, as opposed to partial screening, is par- ticularly critical when pedestrian activity must be allowed <60 m from nests. Where partial screening exists, or is proposed by removal of vegetation on undeveloped property (e.g., creating “view win- dows” or limbing), maintaining some degree of screening vegetation is preferred to completely re- moving the cover, but will provide eagles substan- tially less protection from disturbance compared to dense cover. Planting trees or vegetation near ex- posed nests will not reduce responses of eagles in the short term unless they effectively conceal hu- man activities from the nest, but should be en- couraged to provide future nesting habitat. Management plans should not reduce pedestri- an restrictions based solely on nest tree height, be- cause eagles will select new trees and build nests at different heights over time. On eagle territories with limited management options, maintaining an adequate number of trees >40 m tall may reduce the long-term impacts of reduced screening and closer human activities. Timing conditions in Bald Eagle management plans should restrict pedestrian activities during in- cubation and the first 3 wk of brooding when adults exhibit similar flush rates and subject eggs and young to the possibility of being crushed or ejected from nests (e.g., second wk of Eebruary through the fourth wk in May for western Wash- ington; J. Watson unpubl. data) . Time away from eggs or small young would be more critical during inclement weather, in harsher climates, and at higher elevation sites. Restrictions during later brooding (4-5 wk) through fledging (e.g., 12 wk), a period not specifically addressed in this research, should address activities that may inhibit delivery of prey to young, potentially affecting their survival (Bortolotti 1989, Anthony et al. 1994, Steidl and Anthony 2000). Outreach programs explaining Bald Eagle man- agement should inform the public that untrained observers may greatly underestimate their distur- bance of nesting Bald Eagles. Incubating and brooding eagles will respond roughly 30% of the time pedestrians are present <120 m from nests, and nearly 70% of the eagle response time will be spent perching passively, exposing eggs and young to the elements and predators. Because Bald Ea- gles nesting in pristine environments may exhibit greater sensitivity to pedestrians than those in this study, similar experiments should be conducted in those areas to evaluate distance and cover relation- ships before management conditions are imple- mented. Acknowledgments This research was funded by the Washington Depart- ment of Fish and Wildlife, Wildlife Program. I thank R. Steidl, F. Isaacs, K. Jensen, R. Yates, G. Montopoli, M. Restani, and J. Bednarz for constructive comments that improved this paper, and B. Cunningham and L. Aamot for providing primary field assistance throughout the study. K. McGarigal, R. Steidl, J. Skaski, J. Pierce, and C. Rice contributed statistical advice and fruitful discussions on study design. December 2004 Bald Eagle Response to Pedestrians 303 Literature Cited Andersen, D.E. 1990. Nest-defense behavior of Red-tailed Hawks. Condor 92:991-997 . Anthony, R.G. and F.B. Isaacs. 1989. Characteristics of Bald Eagle nest sites in Oregon. J. Wildl. Manag. 53: 148-159. , RJ. Steidl, and K. McGarigal. 1994. Recreation and Bald Eagles in the Pacific northwest. Pages 223- 242 in R.L. Knight and K.J. Gutzwiller [Eds.] , Wildlife and recreationists: coexistence through management and research. Island Press, Washington, DC U.S.A. Bortolotti, G.R. 1984. Physical development of nestling Bald Eagles with emphasis on the timing of growth events. Wilson Bull. 96:524—542. . 1989. Factors influencing the growth of Bald Ea- gles in north central Saskatchewan. Can. J. Zool. 67: 606-611. Buehler, D.A., T.J. Mersmann, J.D. Fraser, and J.K.D. Seeger. 1991. Effects of human activity on Bald Eagle distribution on the northern Chesapeake Bay. /. Wildl. Manag. 55:282-290. Fox, N. 1995. Understanding the bird of prey. Hancock House Publishers, Blaine, WA U.S.A. Fraser, J.D. 1981. The breeding biology and status of the Bald Eagle on the Chippewa National Forest. Ph.D. dissertation, Univ. of Minnesota, St. Paul, MN U.S.A. , L.D. Frenzel, and J.E. Mathisen. 1985. The im- pact of human activities on breeding Bald Eagles in north-central Minnesota./. Wildl. Manag. 49:585-592. Gerrard, J.M. and G.R. Bortolotti. 1988. The Bald Ea- gle, haunts and habits of a wilderness monarch. Smithsonian Institution Press, Washington, DC U.S.A. Gerrard, R, S.N. Wiemeyer, and J.M. Gerrard. 1979. Some observations on the behavior of captive Bald Eagles before and during incubation. Raptor Res. 13: 57-64. Grier, J.W. 1969. Bald Eagle behavior and productivity responses to climbing nests./. Wildl. Manag. 41:438— 443. and R.W. Fyfe. 1987. Preventing research and management disturbance. Pages 173-182 in B.A. Gi- ron Pendleton, B.A. Millsap, K.W. Gline, and D.M. Bird [Eds.], Raptor management techniques manual. Nat. Wildl. Fed., Washington, DC U.S.A. Grubb, T.G. and R.M. King. 1991. Assessing human dis- turbance of breeding Bald Eagles with classification tree models./. Wildl. Manag. 55:500—511. , W.W. Bowerman, J.P. Giesy, and G.A. Dawson. 1992. Responses of breeding Bald Eagles, Haliaeetu^ leucocephalus, to human activities in northcentral Min- nesota. Can. Field Nat. 106:443-453. Harmata, A.R. 1984. Bald Eagles of the San Luis Valley, Golorado: their winter ecology and spring migration Ph.D. dissertation, Montana State Univ., Bozeman, MT U.S.A. Holthuijzen, A.M.A., W.G. Eastland, A.R. Ansell, M N Kochert, R.D. Williams, and L.S. Young. 1989. Ef- fects of blasting on behavior and productivity of nest- ing Prairie Falcons. Wildl. Soc. Bull. 18:270-281. Knight, R.L., D.J. Grout, and S.A. Temple. 1987. Nest- defense behavior of the American Crow in urban and rural areas. Conrfor 89:175-177. McGarigal, K., R.G. Anthony, and F.B. Isaacs. 1991. In- teractions of humans and Bald Eagles on the Colum- bia River estuary. Wildl. Monogr. 115. Miller, R.G., Jr. 1981. Simultaneous statistical inference Springer-Verlag, New York, NY U.S.A. Newton, I. 1979. Population ecology of raptors. Buteo Books, Vermillion, SD U.S.A. Steidl, R.J. and R.G. Anthony. 1996. Responses of Bald Eagles to human activity during the summer in inte- rior Alaska. Ecol. Appl. 6:482-491. and . 2000. Experimental effects of human activity on breeding Bald Eagles. Ecol. Appl. 10:258- 268. Stinson, D.W., J.W. Watson, and K.R. Mc at, lister. 2001. Washington state status report for the Bald Eagle Washington Department of Fish and Wildlife, Olym- pia, WA U.S.A. Watson, J.W. 1993. Responses of nesting Bald Eagles to helicopter surveys. Wildl. Soc. Bull. 21:171-178. , D. Stinson, K.R. McAllister, and T.E. Owens 2002. Population status of Bald Eagles breeding m Washington at the end of the 20th century. /. Raptor Res. 36:161-169. Yates, R.E. 1989. Bald Eagle nesting ecology and habitat use: Lake McDonald, Glacier National Park, Montana M.S. thesis, Univ. Montana, Missoula, MT U.S.A. AND B.R. McClelland. 1989. Unusual leg injury in a nestling Bald Eagle./. Raptor Res. 23:14—16. Received 21 October 2003; accepted 23 July 2004 Associate Editor: Marco Restani J Raptor Res. 38(4);304-311 © 2004 The Raptor Research Foundation, Inc. DISPERSAL AND MORTALITY OF RED-SHOULDERED HAWKS BANDED IN OHIO Cheryl R. Dykstra^ U.S. Environmental Protection Agency, National Exposure Research Laboratory, Cincinnati, OH 45268 U.S.A. Jeffrey L. Hays RAPTOR, Inc., 1586 Covered Bridge Road, Cincinnati, OH 45251 U.S.A. Melinda M. Simon 9016 Winthrop, Cincinnati, OH 45249 U.S.A. John B. Holt, Jr. 858 Johnson Street, North Andover, MA 01845 U.S.A. G. Ronald Austing P.O. Box 428, Dillshoro, IN 47018 U.S.A. F. Bernard Daniel US. Environmental Protection Agency, National Exposure Research Laboratory, Cincinnati, OH 45268 U.S.A. Abstract. — We banded nestling Red-shouldered Hawks {Buteo lineatus) in southwestern Ohio and north- ern Kentucky (SW OHIO, hereafter) to examine movements and determine causes of mortality in this suburban population. For comparison, we examined band recovery records for nestling Red-shouldered Hawks banded in rural northern Ohio. Of 899 nestlings banded in SW OHIO from 1955-2002, 43 (4.8%) were encountered (dead or alive) some time after fledging. Mean distance from natal nest at time of encounter was 38.5 ± 13.6 km and was not correlated with hawk age (P> 0.58). Distance from natal nest did not differ for hawks of three age classes or between those encountered in the breeding and nonbreeding seasons {P > 0.13). Cumulative exponential distribution (CED) analysis of distance from natal nest at time of encounter indicated that 50% of SW OHIO Red-shouldered Hawks were found <15 kra from their natal nest, 75% were <29 km away, and 95% were <62 km away. Mean age of hawks recovered dead was 1.9 ± 0.4 yr {N =31). CED analysis of age at recovery indicated that 50% of Red-shouldered Hawks were dead by age 1.2 yr, 75% by 2-4 yr, and 95% by 5.2 yr. SW OHIO hawks did not differ from hawks banded in northern Ohio in either distance from natal nest or age at recovery. EIey Words; Red-shouldered Hawk, Buteo lineatus; natal dispersal, survival, banding, urban. DISPERSION Y MORTALIDAD DE BUTEO LINEATUS ANlllADOS EN OHIO Resumen. — ^Anillamos pichones de Buteo lineatus en el sudoeste de Ohio y norte de Kentucky (SO de Ohio, en adelante) para examinar los movimientos y determinar las causas de mortalidad en esta pobla- cion suburbana. De modo comparativo, examinamos datos de anillos recuperados de pichones de Buteo lineatus anillados en areas rurales del norte de Ohio. De 899 pichones anillados en el SO de Ohio entre 1955 y 2002, 43 (4.8%) fueron encontrados (muertos o vivos) algun tiempo despues de abandonar el nido. La distancia al nido natal en el momento del encuentro fue de 38.5 ± 13.6 km y no estuvo correlacionada con la edad del ave {P > 0.58). La distancia al nido natal no difirio entre aguilas de tres clases de edad ni entre aquellas encontradas en la estacion reproductiva y no reproductiva {P > 0.13). Los analisis de la distribucion exponencial acumulativa (DEA) de la distancia desde el nido natal en el momento del encuentro indicaron que el 50% de las aguilas del SO de Ohio fueron encontradas ^ Present address: 7280 Susan Springs Dr., West Chester, OH 45069 U.S.A.; E-mail address: cheryldykstra@juno.com 304 December 2004 Red-shoui.dered Hawk Dispersal 305 a <15 km del nido natal, 75% a <29 km de distancia y 95% a <62 km de distancia. La edad media de las aves recobradas muertas fue de 1.9 ± 0.4 anos {N = 31). Los analisis de DEA de la edad en el momento del encuentro indicaron que el 50% de los individuos ya habian muerto a una edad de 1.2 anos, 75% a una edad de 2.4 anos y 95% a una edad de 5.2 anos. Las aguilas del SO de OHIO no se diferenciaron de las anilladas en el norte de Ohio en la distancia desde el nido natal ni en la edad al momento de recuperacion del anillo. [Traduccion del equipo editorial] The breeding habitat of the Red-shouldered Hawk {Buteo lineatus) in some areas of North Amer- ica has been described as remote (Johnson 1989, Bosakowski et al. 1992, Bosakowski and Smith 1997). However, the species also inhabits suburban areas, at least in California (Bloom and McCrary 1996, Rottenborn 2000) and in southwestern Ohio near Cincinnati (Dykstra et al. 2000, 2001a, 2003). In southwestern Ohio, Red-shouldered Hawks built nests a mean of 75 m from human residences (Dyk- stra et al. 2000) , and their 90-ha home ranges con- tained a mean of 169 residences each (Dykstra et al. 2001b). Two pairs of southwestern Ohio Red- shouldered Hawks have nested on rooftops and one pair on a gas grill on the deck of a residence (Hays 2000, Dykstra et al. 2001b). Suburban-nesting raptors may experience differ- ent threats than those encountered by their rural- nesting conspecifics (Love and Bird 2000). They may endure repeated human disturbance (Preston and Beane 1996) and risk of collision with vehicles and buildings (e.g.. Peregrine Falcons [Falco pergri- nus]; Sweeney et al. 1997). They also may face a higher risk of other kinds of human-induced mor- tality, such as electrocution on power lines (e.g., Harris’s Hawks [Parabuteo unidnctus ] ; Dawson and Mannan 1995) and persecution. Although suburban areas can provide significant breeding habitat for species facing habitat loss in more traditional natural and rural settings, subur- ban areas also are subject to rapid development that may displace even the most tolerant raptor species. The continual conversion of forested areas to lawn and other nonnative vegetation may re- duce prey populations to levels inadequate to sus- tain raptor populations, even if suitable nest sites are available. As urbanization proceeds and habitat deteriorates, fledglings able to disperse long dis- tances from their natal nest may have the best op- portunity to locate suitable breeding habitat. Fledglings of species with short natal dispersals may be less likely to find adequate breeding sites, which eventually may result in regional population decline. We banded nestling Red-shouldered Hawks in southwestern Ohio to examine natal dispersal and determine causes of mortality in this suburban population. For comparison, we obtained band-re- covery records from the Bird Banding Lab for nest- lings banded in rural areas of northern Ohio. We anticipated that suburban hawks from southwest- ern Ohio and northern Kentucky have longer na- tal-dispersal distances than rural hawks because the fragmented-habitat mosaic of the suburbs might result in suitable nesting habitat interspersed among highly-developed, unsuitable space. Study Area We banded nestling Red-shouldered Hawks in south- western Ohio and northern Kentucky (SW OHIO, here- after), in Hamilton, Clermont, and southwestern Warren Counties, OH, and northern Boone and Kenton Coun- ties, KY, <27 km south of the Ohio-Kentucky border. Most nests at which nestlings were banded occurred in a wide band of suburban development and semirural areas surrounding the city of Cincinnati. SW OHIO is a hilly, unglaciated area in the Interior Plateau ecoregion (Omernik 1987). The hills are dissect- ed by many small streams located in ravines and by two large rivers, the Great Miami and the Little Miami. Native forests are dominated by second-growth oak-hickory {Quercus spp.-Carya spp.) and beech-maple (Fagus gran- difolia-Acer saccharum) associations, with lowland-riparian forests characterized by sycamores {Platanus occidentahs) and beech. Suburban areas in SW OHIO varied from densely populated (residential lots ca. 20 X 35 m) to sparsely populated (>2.5-ha residential lots, as well as un- developed private land). Most residences and other buildings were surrounded by lawns and other nonnative vegetation, but residences tended to be located on level ground, with steep slopes and riparian areas left in native vegetation. Public land within the study area consisted primarily of native vegetation, with some developed areas for sports and other recreational uses. Methods Banding. Red-shouldered Hawk nestlings in SW OHIO were banded with U.S. Fish and Wildlife Service (USFWS)/ U.S. Geological Survey (USGS) bands be- tween 1955-59, 1963-77, and 1996-2002. Most nestlings banded between May 1998 and June 2002 were also band- ed with colored-plastic bands (Haggie Engraving, Crump- ton, MD U.S.A.) inscribed with individual alpha-numeric codes large enough to be observed from the ground with 306 D^tcstra et al. VoL. 38, No. 4 binoculars or a spotting scope. Nestlings were banded at ca age 2-5 wk. Band Recoveries and Encoimters. We defined a band recovery as a report of a hawk that had died, and a band encounter as any report of a banded hawk, dead or alive. Most reports were further investigated by contacting the individual who had reported the band. We also encountered color-marked hawks in the course of other fieldwork. Color bands were read with a spotting scope or binoculars, or the marked bird was captured using a bal-chatri trap baited with a mouse (Bloom 1987). Other banded birds {N= 2) encountered by birders were reported directly to the bander. Four banded birds that were injured were brought to RAPTOR, Inc., a local re- habilitation organization. Additional bands recovered in nests or on the ground under nests were not included in this study. We determined causes of death for recovered birds from uses Bird Banding Laboratory (BBL) records (“How obtained” codes) or by carcass examination. We determined gender of dead hawks, when possible, by ex- amination {N = 4) , and gender of live hawks by behavior (i e., copulation observed) or the presence/absence of a brood patch (N = 5). For comparison, we obtained from the BBL reports of band recoveries and encounters for birds banded at var- ious rural locations in northern Ohio (>40°N latitude) by seven banders. The birds represented were banded from 1940-72. Data Analyses. Banding locations in SW OHIO were defined by street addresses in most cases (N = 37) and plotted on USGS 7.5' topographic maps. Encounter lo- cations in SW OHIO were defined by street addresses in most cases {N = 27) or by the nearest town as indicated on the BBL “Report to the bander” data card {N = 9). Specific location data were lacking for some older band- ing locations {N = 6), encounters {N = 7), and for all banding and encounter locations in northern Ohio, so we designated these locations as the center of the 10- mmute block indicated in BBL records. To estimate age of hawks at the time of the encounter, we assumed that all nestlings hatched on 23 April, the mean hatch date for SW OHIO 1997-99. We classified encounters into three categories based on age at the time of encounter; <298 d, 298-663 d, and >663 d. Birds <298 d were those encountered before 15 February in the year following the year in which they were banded, and thus, were not breeding birds. We selected 15 Feb- ruary as a cut-off date because by that date most birds in SW OHIO had begun breeding activities, such as terri- tory occupancy and nest-building (Dykstra et al. 2000, 2001a). Birds 298-663 d were those encountered be- tween 15 February of the year following that in which they were banded and the subsequent 15 February, and thus, were in immature plumage and possibly breeding. Birds >663 d were those in mature plumage and were probably breeding birds. We also classified encounters according to the season in which they occurred. We con- sidered 15 February— 31 July to be the breeding season (Dykstra et al. 2001b), and 1 August-14 February the nonbreeding season. Although Red-shouldered Hawks normally begin breeding in their third spring at 2 yr, they may breed first as immature-plumaged yearlings in their second spring (Wiley 1975, Crocoll 1994). Natal dispersal has been de- fined as the movement from birthplace to the site of the first breeding attempt (Greenwood 1980, Greenwood and Harvey 1982). However, because adult Red-shoul- dered Hawks in SW OHIO apparently are year-round res- idents (Dykstra et al. 2001a, 2001b), any bird encoun- tered at age >663 d or older was likely located at or very near its breeding site, regardless of the season. For sim- plicity, we assumed this location to be the first breeding site. Thus our measure of natal dispersal included all birds >663 d encountered in any season, as well as year- ling birds known to be breeders. Results are shown as mean ± 1 standard error of mean. Because of skewed distributions, the distance from natal nest and age at encounter data were log-transformed be- fore statistical analysis. We used Rests, analysis of variance (ANOVA), and analysis of covariance (ANCOVA) to test for differences in distance from natal nest, and linear regression to examine the relationship between distance from natal nest and age at encounter. We examined age at recovery using cumulative expo- nential distribution (CED) following Harmata et al. (2001) and Harmata (2002). Age in years (x) of each recovered hawk from SW OHIO was listed in order from youngest to oldest and the cumulative proportion of re- coveries calculated for each. For the purposes of calcu- lation, the proportion of 1.0 was expressed as 0.9999999. We fitted cumulative proportion of recoveries by age to the CED function y = 1 — c**. The predicted proportion of recoveries (y) generated from this function were used in the transformed function hx = In (1 ~ y). The ln(l — y) was calculated and regressed with the observed x (with zero intercept) to determine the coefficient ((3) of age variable x. We also examined distance from natal nest at time of encounter using CED analysis following Harmata et al. (2001). Resui.ts Band Encounter Rate and Circumstances of En- counters. Of 899 nestling Red-shouldered Hawks banded in SW OHIO from 1955-2002, 43 (4.8%) were encountered some time after fledging and be- fore January 2004. Of 28 SW OHIO hawks that were recovered, 14 (50%) were simply “found dead,” five (18%) were hit by vehicles, four (14%) were electrocuted, two (7%) were found injured and later died, and one each (4%) were caught in a trap, shot, and found as a band with bone or skeleton only. Three other bands were returned; we assumed these hawks were dead and thus in- cluded them with the recoveries. Twelve hawks were encountered alive: nine (75%) of these were color-marked birds that we sighted or trapped in 1999-January 2004, one (8%) was hand-caught during a storm and released, one was trapped by a bander, and one was obtained without informa- tion. December 2004 Red-shouldered Hawk Dispersal 307 Table 1. Movements of Red-shouldered Hawks banded as nestlings in southwestern Ohio and northern Ohio, and encountered after fledging. Age at Distance from Natal Nest (km) Banding Region Encounter N Mean ± 1 SE Median Range Southwestern Ohio <298 d 14 68.5 ± 36.5 13.7 3.9-500.0 298-663 d 11 33.6 ± 23.4 8.5 0.0-266.1 >663 d 18 18.2 ± 5.4 11.8 0.8-103.2 Total 43 38.5 ± 13.6 12.3 0.0-500.0 Northern Ohio'^ <298 d 11 38.4 ± 13.6 18.5 0.0-145.3 298-663 d 2 40.4 ± 17.2 40.4 23.2-57.6 >663 d 10 92.6 ± 58.7 28.0 0.0-612.9 Total 23 62.1 ± 26.2 27.9 0.0-612.9 “ From uses Bird Banding Laboratory records, banded 1940-72. Figure 1. Map of long-distance (>100 km) dispersal of Red-shouldered Hawks banded as nestlings in southwest- ern OH, 1955-2002. Lines join natal sites and encounter locations. All birds shown were recovered dead and four of the five birds shown were < 663-d old at recovery. Distance from Natal Nest. For SW OHIO birds, mean distance from natal nest at time of encounter was 38.5 ± 13.6 km (Table 1). Most birds moved <30 km, but five birds were recovered 103-500 km from their natal nest (Fig. 1). CED analysis of dis- tance from natal nest at time of encounter indi- cated that 50% of SW OHIO Red-shouldered Hawks were found <15 km from their natal nest, 75% were <29 km away, and 95% were <62 km away {F? = 0.98, (3 = -0.048, P < 0.001, N= 43; Fig. 2). Natal dispersal averaged 18.2 ± 4.9 km, N = '2,0 (males 9.3 ± 2.6 km, W = 4; females 16.5 ± 4.7 km, N = 5, 11 sex undetermined). One bird was recovered 103 km from its natal nest; the other 19 were encountered <30 km from their natal nests (x = 13.7 ± 2.0 km). Distance from Natal Nest (km) Figure 2. Cumulative exponential distribution of dis- tance from natal area for Red-shouldered Hawks banded as nestlings in southwestern OH and northern KY, 1955- 2002 . 308 Dykstra et ajl. VoL. 38, No. 4 Figure 3. Cumulative exponential distribution of age at recovery for Red-shouldered Hawks banded as nestlings in southwestern OH and northern KY, 1955-2002. Distance from the natal nest differed between birds recovered dead and those encountered alive {t = 2.455, df = 41, P — 0.018), so the two cate- gories were tested both separately and combined for analyses below; the combined results are pre- sented when there were no differences. Distance from natal nest was not correlated with age at en- counter {P > 0.58), and did not differ among age classes (P > 0.18). Birds encountered during the breeding season were neither nearer nor farther from their natal nests than birds encountered dur- ing the nonbreeding season (P > 0.13, N = 24 for breeding season, N = \9 for nonbreeding season). When compared to BBL records for birds band- ed as nestlings in rural northern Ohio, those band- ed in SW OHIO moved slightly shorter distances (Table 1), bnt there were no differences in dis- tance from natal nest for all age classes combined or for birds <298 d or those >663 d (P > 0.250; Table 1). Among rural-northern Ohio birds alone, the distance from the natal nest did not differ be- tween birds encountered in the breeding and non- breeding seasons (P > 0.583); however, when ad- justed for age class, there was a tendency for birds to be farther from the natal nest during the non- breeding season than during the breeding season (ANCOVA, age class as covariate, P = 3.719, P = 0.068, IP = 0.17, A - 23). Age at Recovery. Mean age at recovery for Red- shouldered Hawks banded in SW OHIO was 1.9 ± 0.4 yr (N = 31 hawks). CED analysis indicated that 50% of Red-shouldered Hawks were dead by age 1.2 yr, 75% by age 2.4 yr, and 95% by age 5.2 yr {IP = 0.96, P < 0.001, 3 = -0.577; Fig. 3). SW OHIO birds did not differ from northern Ohio birds in age at recovery (t = 0.038, df = 51, P = 0.97). For northern Ohio birds alone, the mean age at recovery was 2.0 ± 0.4 yr {N = 22). CED analysis indicated that 50% of northern Ohio Red-shouldered Hawks were dead by age 1.1 yr, 75% by age 2.3 yr, and 95% by age 4.9 yr {IP — 0.93, P < 0.001, 3 = -0.607, N = 22). Discussion Dispersal from the Natal Nest. Mean dispersal distance was 38.5 km, with 50% of the hawks found <15 km from their natal nest. Natal-dispersal dis- tance, the distance from birthplace to a breeding site, was 18.2 ± 4.9 km. Similarly, Red-shouldered Hawks in other parts of their breeding range also have short natal-dispersal distances. In Wisconsin, 1 1 banded nestlings that were recaptured as breed- ing birds had dispersed a mean of 17 km from their natal site (Jacobs and Jacobs 2002). Jacobs and Jacobs (2002) also determined from BBL data that >54% of 99 eastern Red-shouldered Hawks recovered in the breeding season were <30 km from their natal site. In our study, birds encountered while alive, pri- marily color-banded birds we sighted or captured, had significantly shorter dispersal distances than those recovered dead. The inclusion of birds en- countered alive may have caused the mean dis- persal distance to be underestimated because we did not search for color-marked birds outside the study area. Also, birds were more likely to be en- countered within the heavily-populated suburban region surrounding Cincinnati than in rural re- gions outside the study area. Underestimation of dispersal distance is not uncommon in dispersal studies because long-distance dispersers are less likely to be detected than short-distance dispersers (Koenig et al. 1996). Nonetheless, within the study area, we believe that local-dispersal distances were likely correct because, despite ca. equal effort in banding and trapping throughout the study area, we found only two birds that moved as far as the distance from the west side of the study area to the east side, a span of 30-50 km. We anticipated that young birds <663 d might be encountered farther from their natal nest than those of breeding age (>663-d old), because young raptors of some species tend to move far from the natal nest after they gain independence and before they begin breeding (Walls and Ken- ward 1998, Forero et al. 2002, Byholm et al. 2003). December 2004 Red-shouldered Hawk Dispersal 309 Although we did not find significant differences in encounter distance for birds of different ages, we did note that four of the five birds found at long distances from their natal nest were <2-yr old, sug- gesting that young Red-shouldered Hawks might also wander. Although not significant statistically, mean dis- persal distance for SW OHIO was slightly smaller than for rural northern Ohio hawks (Table 1). This difference might reflect true differences be- tween suburban and rural birds, or it may have resulted from an ecological difference between the two populations. Our age-adjusted analysis indicat- ed that northern Ohio birds encountered in the nonbreeding season tended to be farther from their natal nest than those encountered in the breeding season, suggesting that some northern birds migrate. A more comparable rural popula- tion that is likely nonmigratory is located in south- eastern Ohio; in this population, we banded 217 nestlings from 1997-2002, but recovered only two bands (0.9%, C. Dykstra andj. Hays unpubl. data). Mortality. Most mortality for Red-shouldered Hawks in SW OHIO occurred within the first 14 mo of life, as it did for northern Ohio birds. Henny (1972) examined band-recovery data for Red- shouldered Hawks in six regions of North America and determined mortality rate for the first year of life to be 0.58. High first-year mortality is typical of raptors (Newton 1979). The oldest Red-shouldered Hawk recovered in this study was over 10 yr 3 mo old, but Jacobs and Jacobs (2002) report several hawks that were at least 10—14 yr old and one 17 yr old. The oldest wild Red-shouldered Hawk re- corded was 19 yr 11 mo (Clapp et al. 1982). Most SW OHIO hawks died of unknown causes, but of those for whom cause of death was known, 38% were killed by motor vehicles and an addi- tional 31% by electrocution on power lines or elec- tric fences. Although sample sizes are small, these data suggest that interactions with humans and hu- man-made structures may be an important agent of mortality for urban/ suburban raptors. Similarly, for urban Harris’s Hawks in Tucson, at least 72% of mortality in which cause could be determined was due to electrocution (Dawson and Mannan 1995). Among midwestern Peregrine Falcons, a primarily urban population, 81% of injured falcons admitted to the Raptor Center at University of Min- nesota had sustained injuries from collisions with vehicles, buildings or utility lines (Sweeney et al. 1997) . Mortality of urban adult Lesser Kestrels {Fal- co naumanni) in southern Spain was ascribed to collision with vehicles (13% of mortalities with known cause), persecution by humans (25%), elec- trocution (8%), and entanglement in safety nets erected for building restoration work (21%), for a total of 67% of mortality due to interaction with humans (Telia et al. 1996). In contrast, in a species with a typically rural distribution. Red-tailed Hawks {Buteo jamaicensis) , only 32% of mortalities with known causes were due to collisions with vehicles, electrocution, and gunshot wounds, while the ma- jority were due to poisonings by agricultural pesti- cides (19%), emaciation (25%), and disease (16%; Franson et al. 1996). Dispersal and Urbanization. The short natal-dis- persal distances for Red-shouldered Hawks, com- bined with the increasing urbanization of the Cin- cinnati area and its suburbs, may make it increasingly difficult for young Red-shouldered Hawks fledged in SW OHIO to find suitable breed- ing habitat. Currently, this suburban population does not appear to be compromised in any way. Compared to more rural populations in Ohio and elsewhere, the SW OHIO birds reproduced well at a fairly high nest density, found suitable nest sites (Dykstra et al. 2000) , and maintained home ranges that were typical in size for Red-shouldered Hawks (Dykstra et al. 2001b), although they were less for- ested than those measured elsewhere (Howell and Chapman 1997). However, anecdotal evidence sug- gests that hawks may be losing nesting habitat as urbanization proceeds: in a sample of 22 nesting territories, where hawks were banded in 1963—77, only 10 of them still contained nesting hawks by 1997-98 (Dykstra et al. 2000). Red-shouldered Hawks of SW OHIO may be able to maintain their population if they are able to fur- ther adapt to humans and suburban landscapes. Red-shouldered Hawks in southern California may be even more adjusted to urban conditions than those in Ohio; nesting urban birds there tolerated large crowds attending athletic events as well as people camping directly underneath their nest trees (Bloom and McCrary 1996). A few individual hawks in SW OHIO may be similarly tolerant, as evidenced by the two nests located on rooftops and one located on a suburban deck (Hays 2000, Dyk- stra et al. 2001b) . In summary, the suburban Red- shouldered Hawk population of SW OHIO is ap- parently well-adapted to humans, although it remains uncertain whether these suburban birds 310 Dykstra et ai VoL. 38, No. 4 will be able to maintain their numbers in the face of further urbanization and suburban sprawl. Acknowledgments We would like to thank K. Klimkiewicz and staff at the Bird Banding Lab, as well as V. Koppelberger and the other hawk banders for providing banding and recovery data. We especially thank the many landowners in the Cincinnati area who allowed access to their private prop- erty. This research was supported in part by RAPTOR Inc., M. andj. Wilz of Hamilton County, B. and M. Lind- ner of Hamilton County, and by an appointment to the Postgraduate Research Participation Program at the Na- tional Exposure Research Laboratory administered by the Oak Ridge Institute for Science and Education through an interagency agreement between the U.S, De- partment of Energy and the U.S. Environmental Protec- tion Agency. We thank P. Bloom, B. Chapman, M. Res- tani, and J. Zelenak for valuable comments on an earlier draft. Literature Cited Bloom, P.H. 1987, Capturing and handling raptors. Pag- es 99-123 mB.A. Giron Pendleton, B.A. Millsap, K.W. Cline, and D.A. Bird [Eds.], Raptor management techniques manual. Nat. Wildl. Fed., Washington, DC U.S.A. AND M.D. McCrary. 1996. The urban buteo: Red- shouldered Hawks in southern California. Pages 31- 39 in D.M. Bird, D.E. Varland, and J.J. Negro [EdS.], Raptors in human landscapes: adaptations to built and cultivated environments. Academic Press Limit- ed, London, U.K. Bosakowski, T, D.G. Smith, and R. Speiser. 1992. Status, nesting density, and macrohabitat selection of Red- shouldered Hawks in northern New Jersey. Wilson Bull. 104:434-446. AND . 1997. Distribution and species richness of a forest raptor community in relation to urbaniza- tion./. Raptor Res. 31:26-33. Byholm, R, P. Saurola, H. Linden, and M. Wikman. 2003. Causes of dispersal in Northern Goshawks (Ac- cipiter gentilis) in Finland. Auk 120:706-716. Clapp, R.B., M.K. Klimkiewicz, and J.H. Kennard. 1982. Longevity records of North American birds: Gaviidae through Alcidae./ Field Ornithol. 53:81-124. Crocoll, S.T. 1994. Red-shouldered Hawk {Buteo linea- tus). In A. Poole and F. Gill [Eds.], The birds of North America, No. 107. The Birds of North America, Inc., Philadelphia, PA U.S.A. Dawson, J.W. and R.W. Mannan. 1995. Electrocution as a mortality factor in an urban population of Harris’s Hawks./. Raptor Res. 29:55. Dykstra, C.R., J.L. Hays, F.B. Daniel, and M.M. Simon. 2000. Nest-site selection and productivity of suburban Red-shouldered Hawks in southern Ohio. Condor 102’. 401-408. , F.B. Daniel, J.L. Hays, and M.M. Simon. 2001a. Correlation of Red-shouldered Hawk abundance and macrohabitat characteristics in southern Ohio. Condor 103:652-656. , J.L. Hays, F.B. Daniel, and M.M. Simon. 2001b. Home range and habitat use of suburban Red-shoul- dered Hawks in southwestern Ohio. Wilson Bull. 113’ 308-316. , , M.M. Simon, and F.B. Daniel. 2003. Be- havior and prey of nesting Red-shouldered Hawks in southwestern Ohio./. Raptor Res. 37:177-187. Forero, M.G., J.A. Donazar, and F. Hiraldo. 2002. Caus- es and fitness consequences of natal dispersal in a population of Black Kites. Ecology 83:858-872. Franson, J.C., NJ. Thomas, M.R. Smith, A.H. Robbins, S. Newman, and P.C. McCartin. 1996. A retrospective study of postmortem findings in Red-tailed Hawks. / Raptor Res. 30:7-14. Greenwood, P.J. 1980. Mating systems, philopatry, and dispersal in birds and mammals. Anim. Behav. 28- 1140-1162. AND P.H. Harvey. 1982. The natal and breeding dispersal of birds. Ann. Rev. Ecol. Syst. 13:1-21. Harmata, A.R., M. Restani, G. Montopoli, J.R. Zelenak, J.T. Ensign, and P.J. Harmata. 2001. Movements and mortality of Ferruginous Hawks banded in Montana / Field Ornithol. 72:389—398. . 2002. Encounters of Golden Eagles banded in the Rocky Mountain west. / Field Ornithol 73:23-32. FIays, J.L. 2000. Red-shouldered Hawks nesting on hu- man-made structures in southwest Ohio. Pages 469- 471 in R.D. Chancellor and B.-U. Meyburg [Eds.], Raptors at risk; proceedings of the world conference on birds of prey and owls. World Working Group on Birds of Prey and Owls, Berlin, Germany. Henny, C.J. 1972. Analysis of population dynamics of se- lected avian species with special reference to changes during the modern pesticide era. U.S. Bureau of Sport Fisheries and Wildlife, Washington, DC U.S.A. Howell, D.L. and B.R. Chapman. 1997. Home range and habitat use of Red-shouldered Hawks in Georgia. Wil- son Bull 109:131-144. Jacobs, J.P. and E.A. Jacobs. 2002. Conservation assess- ment for Red-shouldered Hawk national forests of north central states, http://www.fs.fed.us/r9/wildlife/ tes/ ca-overview/birds. Johnson, G. 1989. Status and breeding ecology of the Red-shouldered Hawk in north-central New York. M.S. thesis. State Univ. New York, Syracuse, NYU.S.A. Koenig, W.D., D.V. Vuren, and P.H. Hooge. 1996. De- tectability, philopatry, and the distribution of dispersal distances in vertebrates. Trends Ecol. Evol. 11:514—517. Love, O.P. and D.M. Bird. 2000. Raptors in urban land- scapes; a review and future concerns. Pages 425-434 in R.D. Chancellor and B.-U. Meyburg [Eds.], Raptors at risk: proceedings of the world conference on birds of prey and owls. World Working Group on Birds of Prey and Owls, Berlin, Germany. December 2004 Red-shouldered Hawk Dispersal 311 Newton, I. 1979. Population ecology of raptors. Buteo Books, Vermillion, SD U.S.A. Omernik, J.M. 1987. Ecoregions of the conterminous United States. Ann. Assoc. Am. Geograph. 77:118-125. Preston, C.R. and R.D. Beane. 1996. Occurrence and distribution of diurnal raptors in relation to human activity and other factors at Rocky Mountain Arsenal, Colorado. Pages 365-374 in D. Bird, D. Varland, and J. Negro [Eds.], Raptors in human landscapes: adap- tations to built and cultivated environments. Raptor Research Foundation and Academic Press, London, U.K. Rottenborn, S.C. 2000. Nest-site selection and repro- ductive success of urban Red-shouldered Hawks in central California./. Raptor Res. 34:18-25. Sweeney, S.J., P.T. Redig, and H.B. Tordoff. 1997. Mor- bidity, survival, and productivity of rehabilitated Per- egrine Falcons in the upper midwestern United States. / Raptor Res. 31:347-352. Tella, J.L., F. Hiraldo, J.A. Donazar-Sancho, and JJ Negro. 1996. Costs and benefits of urban nesting in the Lesser Kestrel. Pages 53-60 in D. Bird, D. Varland, and J. Negro [Eds.], Raptors in human landscapes adaptations to built and cultivated environments. Rap- tor Research Foundation and Academic Press, Lon- don, U.K. Walls, S.S. and R.E. Kenward. 1998. Movements of ra- dio-tagged Buzzards Buteo huteo in early life. Ibis 140‘ 561-568. Wiley, J.W. 1975. The nesting and reproductive success of Red-tailed Hawks and Red-shouldered Hawks m Orange County, California, 1973. Condor 77:133-139 Received 19 September 2003; Accepted 13 July 2004 Associate Editor: Marco Restani J. Raptor Res. 38(4):312-319 © 2004 The Raptor Research Foundation, Inc. DECREASES IN A POPULATION OF RED-SHOULDERED HAWKS NESTING IN CENTRAL MARYLAND Elwood M. (Woody) Martin^ 10815 Loblolly Pine Drive, Laurel, MD 20708 U.S.A. Abstract. — I report the results of a 32-yr (1971-2002) nesting study of the Red-shouldered Hawk {Buteo lineatus) in central Maryland that adds 31 yr of observations to an earlier long-term study. Regression analysis indicated that from 1975-2002 the number of nesting pairs in the study area decreased by at least 78%. An estimate of the population change based on the number of successful nests (fledging at least one young) indicated a decrease of about 88%. The number of young fledged/ successful nest decreased slightly. Modest downward trends in the numbers of Red-shouldered Hawks observed during the local Christmas Bird Counts since 1972 provide further evidence of a population in decline. These long-term trends in this nesting population’s size and nesting success were contrary to patterns expected as the density of hawks decreased. Human activities resulting, both directly and indirectly, in habitat changes detrimental to this species were likely the principal reasons for these local decreases; similar to declines observed in other Red-shouldered Hawk populations. Key Words: Red-shouldered Hawk, Buteo lineatus; central Maryland', breeding success; population decline. DISMINUCIONES EN UNA POBIACION NIDIFICANTE DE BUTEO LINEATUS EN EL CENTRO DE MARYIAND Resumen. — En este trabajo documento los resultados de un estudio de 32 ahos de duracion (1971- 2002)sobre la nidificacion de Buteo lineatus en el centre de Maryland, ahadiendo 31 anos de observa- ciones a un estudio previo de largo plazo. Analisis de regresion indicaron que el numero de parejas nidiheantes en el area de estudio disminuyo por lo menos en un 78% entre 1975 y 2002. Otro estimado del Cambio poblacional basado en el numero de nidos exitosos (con al menos un pichon emplumado) indico una disminucion de alrededor del 88%. El numero de pichones emplumados por nido exitoso disminuyo ligeramente. Las tendencias moderadas de disminucion en el numero de B. lineatus obser- vados durante los conteos navidenos desde 1972 proveen evidencia adicional de que la poblacion esta en disminucion. Estas tendencias de largo plazo en el tamaho de esta poblacion nidificante y en su exito de nidificacion fueron contrarias a los patrones esperados con la disminucion de la densidad poblacional. Las actividades humanas que llevan directa o indirectamente a cambios nocivos en el habitat para esta especie fueron probablemente las razones principals que explican estas disminuciones locales, de forma similar a las disminuciones observadas en otras poblaciones de B. lineatus. [Traduccion del equipo editorial] In the spring of 1947, U.S. Fish and Wildlife Ser- vice biologist Robert Stewart (1949) led a nesting study of the Red-shouldered Hawk {Buteo lineatus) in the Patuxent River watershed in parts of Prince George’s and Anne Arundel counties, MD cover- ing the coastal plain from Laurel and Fort Meade to tidewater. That study, which included records dating back to 1943, reported a variety of infor- mation on the Red-shouldered Hawk (hereafter RSHA) including habitat requirements, population densities, reproductive performance, and food habits. From 1960 through 1967 Fred Schmid, an- ^ E-mail address: woody_martin@fws.gov other biologist at the Patuxent Wildlife Research Center (PWRC), located nests and banded nest- lings in the heart of Stewart’s study area. His ob- servations are included in Henny et al. (1973). In late 1970, a group of biologists led by Charles Hen- ny designed a follow-up study on a portion of Stew- art’s original area to determine this species’ status, a study centered on but not limited to the PWRC. I, with many volunteers, have continued to moni- tor this RSHA population to provide a continuous 32-yr record using the same methods to extend ear- lier observations. Because many birds and other species are known to decline in abundance as their habitat patches decrease in size and quality, such 312 December 2004 Declining Red-shouldered Hawks in Maryland 313 long-term monitoring is especially useful in this area of growing urbanization as the PWRC area be- comes an increasingly isolated large patch of for- est. Henny et al. (1973) noted that the PWRC re- mained an atypical “island of remaining habitat,” where the RSHA population still seemed to be do- ing well at a time when many other populations of this species were declining. Bednarz and Dinsmore (1981) cited 14 references indicating that the RSHA population had declined in Iowa and else- where and was listed as rare or endangered in five states, probably largely due, directly or indirectly, to habitat change. These authors stated that “Tim- ber harvesting (selective or clear-cutting), dam construction, and channelization all have major detrimental impacts on natural bottomland com- munities. Clearly, the Red-shouldered Hawk will continue to decline as river systems and lowland habitats continue to be modified and developed” citing several additional references to support this prediction. Titus et al. (1989) reported that five northeastern states had listed the RSHA as either “threatened,” of “special concern,” or a candidate for listing. Data reported here supplement the earlier pop- ulation data for central Maryland and enable me to test the hypothesis that the PWRC is still an is- land of prime RSHA habitat. I will thus examine the size and breeding success of the RSHA popu- lation on and around this area to learn if changes have occurred therein since the early 1970s and suggest possible causes for any changes. Study Area and Methods The study area, as described by Henny et al. (1973), included a mostly mature woodland in the floodplain and adjacent upland along the Patuxent River from Lau- rel (39°06'07"N, 76°50'22"W) downstream to Bowie State University (39°02'04"N, 76°45'06"W). Since 1972, I have continued to cover this same area plus a 23% extension downstream to the old Bowie Race Track (39°00'36"N, 76°44'11"W), a total of about 1077 ha. Henny et al. (1973) also provided a detailed description of their meth- ods and a thorough analysis of the population data avail- able for the PWRC segment for the 29-yr period 1943- 71. Even more detailed information on the area’s physical characteristics together with annotated lists of much of the flora (Hotchkiss and Stewart 1947) and fau- na of the PWRC were published in one booklet (Anon- ymous 1979). Following the methods of Henny et al. (1973), the study area was searched on foot from early March through May each year for any evidence of RSHA nesting ranging from courtship and nest building through fledging. Total nests found showing evidence of use by nesting RSHA (e.g., presence of an adult, fresh leafy material or down on the nest, droppings indicative of young in the nest, and young seen in the nest) pro- vided an index of population size each season. Any such nest from which at least one young most likely fledged (usually indicated by the number of young banded at >2.5 wk old, but including a few nests observed to have fledged unbanded young) is further defined as a success- ful nest. Generally nest trees were not climbed until time of banding, but nests that appeared unsuccessful were checked sooner, and several nests were visited at periodic intervals from the time of hatching to fledging. At these nests wing chord measurements were taken on several young every few days to provide information on growth rate and corresponding age in days. By taking the same measurements of the young in other nests when I banded them, I was usually able to estimate hatching dates within a few days. In 1981, a fairly typical year in terms of study effort, about 110 hr were spent searching the area and an ad- ditional 50 hr returning to check the status of nests found, band the young, and record details on nests and young. I was able to get out mostly after work and on weekends, and thus, covered only a small portion of the study area each period in the field. As with any field work, some years had more days with adverse weather or trees leafed out earlier, while in other years, conditions for conducting this type of work were better. Thus, cov- erage of the area varied from year to year. However, I believe that over such a long period of data collection, sampling effort fluctuated around an average which was not biased relative to the changes I observed in this RSHA population. On the other hand, I suggest that my ability to identify and include unsuccessful nests and find a greater proportion of the total population likely im- proved with experience. Each year, I found or was told about RSHA nests m central Maryland outside my study area. Also, in 1975 and 1976, I spent quite a bit of time searching for nests in several nearby areas. Data for nests located outside the study area are labeled as “other nests” and included in only those analyses which would not be affected by the tendency for this subsample to include more successful nests. Collection and contaminant analysis of a small sam- ple of eggs (Henny et al. 1973) that failed to hatch con- tinued into the raid 1970s. I used the linear-regression program provided by Lotus 123, Release 5, (IBM Software Group, Cambridge, MA U.S.A.) primarily to assess changes over time; levels of apparent statistical significance (Snedecor and Cochran 1980) were included to emphasize patterns and the rel- ative magnitude of suspected changes. I also examined local Christmas Bird Count (CBC) data using linear re- gression. I estimated percent change over the entire pe- riod by dividing the difference between the expected val- ues provided by linear regression for the first and last years by the value for the first year. I used 2-tailed statis- tical tests for this analysis. Results Despite increasing search efforts over the years, the number of RSHA nests found in my study area has decreased substantially each 10-12-yr period (Table 1). Also, nest success has declined from the Table 1. Summary of Red-shouldered Hawk population status and nesting performance and changes therein in central MD, 1971-2002. Other nests are 314 Martin VoL. 38 , No. 4 T 3 h c OJ o O o T 3 Ed o T) :3 CO OJ a; n D j-i tJ aj a cd tu t:5 V Cm O ^04 04 00 CO^ CDOdCDOCOcOS^OiDGM ccj aj M rS tu Cd Cd toj u e -d Cd ccj I O M U Dd rS u Sk ni c/l u cd M -S cd td CO O rt u 3 - CO O Z) OJ C kH 00 04 Cd d CM 50 CD 00 CD 1T5 00 CM lO 00 !-k d OO 00 OO 00 CM m OO 04 OO 04 OO d CO 04 CM 04 CM CM CM CM 04 04 CM CM CM O 00 1> ^ o CD CD 1— i 00 05 m 05 d CD OO <35 J> ■d !— 1 CT) O in 00 d OO CD <35 d in 00 d 1-H CM <35 CM r- 1 CD O 05 04 m t-h 05 CM 05 q ■rH 05 . CD d CD D D D CnO 4 rH 0000 cOCr) 00 J> CDOO^ 05 onooooGOJ>cDtsoo oo^ t-H cm T— ' kk 105 rL. n. (T-. 0-. 0-. rk. r»-k 00 00 04 CM f-H tH D + O + + <— 1 + + s> + + 00 + d m d d D D 00 kH D 00 CM OO 00 d o D 1—1 04 GO CM CM CM D X c/5 K! U U ^ 05 03 « c« cd o3 U ■ CO ■d cd o3 -I o ^ c3 kJ CM CM 0 O O M o o o w o d! 00 1 05 1 CM 1 O CM 1 .2 kk kk kk U kk 'm 1> 00 35 J> u <35 <35 (35 CM (35 IP kH kk kk cO kk a; > 'co T3 C rt 0.500 RSHA/party-hr from BCBC (1972-2002) = 0.16 - 0.002x -23.1% 0.030 Probability that slope (estimated change) is zero. Estimates based on all nests found may be biased because the proportion of unsuccessful nests not found was not uniform through- out the study period. first decade of study (69.6%) relative to the most recent period of monitoring (47.9%; Table 1). Sta- tistically significant decreases are indicated both for nests found and for nesting success (Table 2). The number of nests found in the study area de- creased by 69.3% since 1971 and by 78.3% since 1975. The number of successful nests (fledging young) has decreased by 88.2% since 1972 (Table 2). Because a smaller area was studied in 1971 and my skill at finding nests, especially unsuccessful nests, likely improved for the first several years, I consider the 1972-2002 estimate ( — 88.2%) based only on successful nests, the best measure of the population decrease. For the same reason, the per- centage of nests successful and the number of young fledged/nest found (Table 2) were likely overestimates, especially early in the study. On the other hand, young fledged per successful nest, which should also be relatively unbiased, has shown little long-term decrease (Tables 1 and 2). Similar re- sults were evident among the nests found outside my study area (Table 1). Mean hatching date (28 April) may have changed slightly (to 26 April; Ta- ble 1) since about the mid 1980s, but this pattern was not statistically significant (Table 2). Discussion Based on the difficulties of finding unsuccessful nests, as noted above (also see Johnson and Shaffer 1990), I suggest my data pertaining to successful nests are more reliable than that on total nests found, and I emphasize it here. With this subsam- ple, there was a slightly sharper downward trend indicated in the numbers of successful nests found compared to all nests found (Table 2). Because successful RSHA nests tend to be more visible over a longer period than unsuccessful nests, they have a higher probability of being discovered by an ob- server than unsuccessful nests, especially those that fail early in the season. With experience, 1 feel I have become better at finding and identifying such unsuccessful nests and included an increasing pro- portion of them in my sample as the study pro- gressed. However, nesting RSFIAs often continue to occupy a territory again the following year, espe- cially after a successful nesting season, tending to nest near the old nest site or even reuse the same nest. This makes nests in territories with a history of success easier to find year after year, and the likelihood of finding a higher portion of the suc- cessful nests thus increases as years go by. Overall, however, I believe the improvement in my ability to find unsuccessful nests had the greatest influ- ence on my results. This more complete sample of unsuccessful nests at least partly explains the sharp- er decrease in young/nest found compared to young/ successful nest (Table 2) . However, finding more of the unsuccessful nests (and successful nests as well) should have caused total nests found in my sample to increase if the population was ac- tually stable or increasing. Thus, as my population 316 Martin VoL. 38, No. 4 figures must be more complete now than in the early years of the study, the population decrease may be even steeper than my data indicate. Also, although more people become aware of this study as years go by, fewer reports of nesting hawks now come to me from other observers; another indi- cation that the RSHA nesting population has de- creased in central Maryland. Henny et al. (1973) looked at population density and nesting success on the PWRC from 1960-71 by comparing four years with up to six RSHA nests found and five years with more than six nests lo- cated. He found that young fledged/nest was 31% lower when there were more nests, which seemed to support a density-dependent response. However, young fledged/ successful nest was only 12% lower. From 1971-2002, I had 15 years (most before 1984), in which nine or more nests were found. These sites had a mean of 2.19 young fledged/ successful nest and 1.12 young fledged/ nest found. I had 17 years, mostly since 1983, with eight or fewer nests located, from which 2.12 young fledged/ successful nest ( — 3%), and 1.27 young fledged/nest found ( + 15%). If the data back to 1960 are included, there were 25 years with fewer than nine nests found, which produced 2.15 young fledged/successful nest ( — 2%), and 1.32 young fledged/nest found ( + 18%). These data do not support the occurrence of density-dependent re- sponses in reproductive success. However, the question of whether or not the unsuccessful nests were included in the proper proportion, especially in the early years, remains. Thus, I consider the figures based on successful nests to be the most reliable. During the early years of this research, I assumed I was seeing normal annual fluctuations in the size and reproductive success of a relatively stable breeding population. I considered differences from year to year to be due to variations in weather conditions and other natural variables. Henny et al. (1973) concluded that rainfall had no influence on the number of young fledged/ nest found. How- ever, their figures show the difference in young/ successful nest to be somewhat larger and success higher when there were 1-2 d with at least 19 mm of rain during the nesting period compared to no ram. Their study did not include any years with a longer period of heavy rain. At least short-term fluctuations in the numbers of nesting pairs and their success may be related to larger weather changes. For example, in 1995, both young A 0 I — ^ ^ ^ ^ ^ — ■ — > — ' — ' — ' — ' — ' — ' — ^ ^ — > « ■ ' — ^ ^ ^ ^ ^ 1970 1975 1980 1985 1990 1995 2000 Year Obser\)ed - Expected Figure 1. Trends shown in selected measurements of a Red-shouldered Hawk population in a central MD study area: (A) number of nests found from 1975-2002 (r^ = 0.88, b = —0.98), (B) number of successful nests found from 1972-2002 (+ = 0.67, b = —0.67), and (C) number of young fledged/successful nest from 1971-2002 = 0.06, b = -0.009). fledged/successful nest (Fig. 1) and young fledged/nest found were the highest recorded in this study, an exception to the general long-term trend. This unusually successful nesting season was preceded by milder and drier than normal winter and early spring weather and followed several years of markedly colder, wetter weather with lower nest- ing success (Fig. 1). Generally, local weather con- ditions appear to me to have continued to fluctu- ate normally while the nesting RSHA population has declined to new lows (Table 1), a decrease which seems to have been largely independent of local weather conditions. However, changes in weather cannot be entirely ruled out. After a re- view of over 2000 published papers, the IPCC (Gi- tay et al. 2002) published an overview of the effects of climate changes, especially global warming, on biodiversity around the world. Among their find- ings: “There has been a discernible impact of re- gional climate change, particularly increases in temperature, on biological systems in the 20th cen- tury.” They go on to say “Such systems include, for December 2004 Declining Red-shouldered Hawks in Maryland 317 example . . . species distributions, and population sizes.” They point out that such climate changes may impact different species in a community in different ways. For example, by putting the timing of breeding out of synchrony with the times when food is available to prepare adults for breeding or later to feed their young. Development-related activities, particularly con- struction of buildings, roads, and power lines around the edges of and in the study area accom- panying a growing human population, have caused a gradual deterioration in and even destruction of at least some RSHA bottomland habitat in the PWRC area as it has in many other areas (e.g., Bed- narz and Dinsmore 1981, Bryant 1986). Motor-bike trails now run through the bottomland in several areas. Near Laurel in an area where “paintball wars” were conducted, I found two paintballs rest- ing in an unsuccessful RSHA nest. Also, an ex- panding beaver ( Castor canadensis) population has resulted in flooding of numerous lowland sites and the cutting and drowning of many trees and other vegetation. In addition, early in this study some logging occurred in and near the bottomland in the downstream portion of the study area. 1 have not attempted to measure the habitat loss, but there does not yet appear to me to be any shortage of nest sites or food for hawks in the study area, though the size and quality of many RSHA terri- tories has likely been changed for the worse. The changes seen in the RSHA population provide the best indication of this decline in habitat. While some authors (e.g., Howell and Chapman 1997) suggest that openings such as those made by log- ging and beaver in woodlands benefit RSHAs, oth- ers (e.g., Bednarz and Dinsmore 1982, Moorman and Chapman 1996) found that the Red-tailed Hawk (Buteo jamaicensis) tended to replace RSHAs when a floodplain was opened up and fragmented. Bryant (1986), aware of widespread concerns for the status of the RSHA, studied a local population in Ontario and concluded that selective logging there had created habitat more attractive to the Red-tailed Hawk (hereafter RTHA), which forced out the RSHAs previously nesting there. I have ob- served increased nesting by, competition with, and predation on RSHAs in my study area by Great Horned Owls {Bubo virginianus) and RTHAs which now find this area better suited to their habitat re- quirements. The impact of Great Horned Owls was especially evident in 2000, when the remains of at least four adult RSHAs were found, three in or near their nests and a fourth in a Great Horned Owl nest. Signs of such losses have become an an- nual event, and I consider this as evidence that the decrease in this RSHA population was the result, both directly and indirectly, of the habitat changes observed. While acknowledging comparison with recruit- ment standards (Henny 1972) that “maybe slightly biased high,” Henny et al. (1973) concluded ten- tatively that the observed recruitment rate of 1.95 fledgings/breeding pair/yr with 77% of the nest- ing pairs successful on the PWRC appeared to be adequate for maintaining the population. Howev- er, for most of my study, both recruitment rate and percentage of pairs nesting successfully (Table 1) have been well below the means reported by Henny et al. (1973). Thus, I have concluded that my study area and nearby areas contained a RSHA popula- tion that was not stable between 1971-2002, but was in fact decreasing significantly. Henny et al. (1973) also concluded that “Therefore, it is doubt- ful that the relatively low pesticide levels in the eggs had a detrimental effect on the reproductive performance of the population.” Eggs collected in this study later in the 1970s gave results similar to those shown in Henny et al. (1973) and continue to support that conclusion. The Bowie Christmas Bird Count (BCBC) , spon- sored by the local chapter of the National Audu- bon Society, provides additional evidence that this RSHA population has decreased (Table 2). This count, in which I have been a regular participant, began in 1972 and includes nearly all my study area and a much larger nearby area (a Christmas count circle includes almost 45770 ha), encom- passing many of my other nest sites. Regression analysis indicated that the BCBC RSHA count/par- ty-hr has decreased by 23.1% since 1972 (Table 2). Many, if not all, of the RSHAs in my study area are year-round residents as indicated by a radiote- lemetry study of local adult RSHAs in the late 1980s by M. Fuller and his assistants (Senchak 1991) and by band-recovery information. Thus, both my breeding population data and the BCBC winter population data, the latter including local young-of-the-year, likely relate to the same resident population, and both provided evidence of a pop- ulation decrease. In another analysis of CBC data, McKay et al. (2002) found a substantial decrease in the RSHA population in the 1960s and no sign of recovery to former levels in a portion of the Mississippi River valley in Iowa and Illinois. An ear- 318 Martin VoL. 38, No. 4 Her study (Brown 1971) of CBC data from 1950- 69, showed the RSHA population to have “de- creased markedly” in a group of northern and eastern states, with a decrease of about 50% indi- cated for Maryland. As discussed above, Henny et al. (1973) suggest- ed that the nesting success of this RSHA popula- tion was density-dependent with pairs nesting clos- er to each other producing fewer young than pairs nesting farther apart. Thus, other factors being equal, a less dense population should have higher nesting success. However, in my study, both popu- lation density and nesting success showed decreas- es. In contrast, Rottenborn (2000) found both an increasing RSHA population and high nesting suc- cess in a study in California, crediting this to the unusual habitat, particularly Eucalyptus spp. and Washingtonia spp., introduced species, which in- creased nest site availability. Few, if any, other stud- ies provide long-term data on both population lev- els and nesting success. It seems clear from my study that other factors may not be equal, that nest- ing success can be affected by more than just pop- ulation density. Evidence of compensatory inter- action between RSHA nesting success and nesting density, at least over three decades in central Mary- land, appears to be lacking. On a broader scale, Henny (1972) compared mean numbers of young banded/successful nest, an estimator of young fledged/ successful nest, during the periods 1900- 45 (2.50 birds) and 1946-68 (2.33 birds) in a mid- Atlantic region centered on Maryland. The de- crease of about 7% between these periods was not statistically significant. However, he found even larger decreases in the three other regions of the nation for which similar data were available sug- gesting that widespread decreases in nesting suc- cess had occurred between the early 1900s and 1968. During this period, the continental RSHA population was also believed to have undergone a widespread decline (e.g., Henny 1972, Bednarz and Dinsmore 1981). Thus, these apparent de- creases in both population size and nesting success were not confined to a few scattered study areas. Again, support for the operation of density depen- dence and compensatory interaction appears to be lacking, at least under conditions of declining hab- itat quality and quantity. RSHA populations in many other areas seem to have been decreasing at least through the 1960s. The relatively stable population apparent here un- til at least the early 1970s was viewed by Henny et al. (1973) as a local phenomenon — a population in an “island of remaining habitat.” An analysis by Bednarz et al. (1990) suggested that the numbers of RSHA counted at Hawk Mountain, PA declined significantly between 1946-86, consistent with de- creases in all five northeastern Breeding Bird Sur- vey (BBS) strata. I do not expect the PWRC RSHA population to continue to decline in a straight line as assumed by linear regressions. Rather, I propose this population may be beginning to stabilize at a lower level as suggested by recent patterns in my data (Fig. 1). A number of authors including Bednarz and Dinsmore (1981) and Bryant (1986) have present- ed evidence and argued that widespread habitat changes in recent years could be expected to pro- duce relatively large-scale decreases in the RSHA population. This appears to have happened in my study area and in at least some other areas. How- ever, this evidence of relatively widespread RSHA population decreases through at least the 1990s seems to be contradicted by data from the North American BBS that indicated for the period 1966- 99 that the RSHA population increased in the U.S. by a mean of 2.5%/yr. This change was statistically significant, but one to be viewed with caution for a variety of sampling and biological reasons (Par- dieck and Sauer 2001). Analysis for the 1966-2002 period (Sauer et al. 2003) gives a similar result at the U.S. level and indicates a 4.8% increase for Maryland (almost reaching statistical significance) for the 1972-2002 period. Could deterioration of RSHA habitats leading to lower populations have forced the remaining birds to range over larger, more open territories to survive, making them more conspicuous along BBS routes? Because of low detection rates, the BBS is likely a less perfect technique for measuring most raptor populations than it is for measuring most other bird popula- tions. Or, as I speculate above, perhaps the RSHA population is beginning to stabilize at a relatively low level or even to recover in some areas. More work is needed to resolve these contrary indications from different data sets. Continuation of RSHA population monitoring here together with broader studies here and elsewhere should shed further light on causes and patterns of RSHA population change. Particular attention should be given to habitat modification and destruction, but also to some perhaps less obvious threats. While the RSHA was apparently among the species rela- tively unaffected by DDT during or after the period December 2004 Declining Red-shouldered Hawks in Maryland 319 of its use, 1946-72 in most areas (Bednarz et al. 1990), other environmental contaminants may be a factor. Also, the threat of West Nile virus (Flavi- viris sp.) to the RSHA population needs to be ex- amined. Acknowledgments I will not attempt to list the many people who assisted with various phases of this study from locating nests through banding nestlings as it would take much space and some would undoubtedly be overlooked; however, their help was essential and was much appreciated. I am especially indebted to C.S. Houston who kindly read sev- eral of my early annual updates on this study, provided much editorial help, and urged me to publish my data, and to G. Allen, J.C. Bednarz, C.R. Dykstra, C.S. Robbins, JR- Sauer, and several anonymous reviewers who also provided many helpful editorial suggestions and com- ments on various versions of this paper. Literature Cited Anonymous. 1979. Vegetation and vertebrates of the Pa- tuxent Wildlife Research Center: outline of ecology and annotated lists. Patuxent Wildlife Research Cen- ter, USDI Fish and Wildlife Service, Laurel, MD U.S.A. Bednarz, J.C. andJ.J. Dinsmore. 1981. Status, habitat use, and management of Red-shouldered Hawks in Iowa. /. Wildl. Manag. 45:236-241. and . 1982. Nest sites and habitat of Red- shouldered and Red-tailed Hawks in Iowa. Wilson Bull. 94:31-45. , D. Klem, Jr., LJ. Goodrich, and S.E. Senner. 1990. Migration counts of raptors at Hawk Mountain, Pennsylvania, as indicators of population trends, 1934-86. Auk 107:96-109. Brown, W.H. 1971. Winter population trends in the Red- shouldered Hawk. Am. Birds 25:813-817. Bryant, A.A. 1986. Influence of selective logging on Red- shouldered Hawks, Buteo lineatus, in Waterloo region, Ontario, 1953-78. Can. Field-Nat. 100:520-525. Gitay, H., a. Suarez, R.T. Watson, and D.J. Dokken (Eds.). 2002. Climate change and biodiversity. IPCC Technical Paper, http://www.ipcc.ch/pub/techrep. htm. Henny, C.J. 1972. An analysis of the population dynamics of selected avian species with special reference to changes during the modern pesticide era. USDI Fish and Wildlife Service, Wildl. Res. Rept. 1, Washington DC U.S.A. , F.C. Schmid, E.M. Martin, and L.L. Hood. 1973. Territorial behavior, pesticides, and the population ecology of Red-shouldered Hawks in central Mary- land, 1943-1971. Ecology 54:545-554. Hotchkiss, N. and R.E. Stewart. 1947. Vegetation of the Patuxent Research Refuge, Maryland. Am. Midi. Nat. 38:1-75. Howell, D.L. and B.R. Chapman. 1997. Home range and habitat use of Red-shouldered Hawks in Georgia. Wil- son Bull. 109:131-144. Johnson, D.H. and T.L. Shaffer. 1990. Estimating nest success: when Mayfield wins. Auk 107:595—600. McKay, K.J., J.W. Stravers, P.C. Peterson, C.J. Kohrt, J.S. Lundh, and G.V. Swenson. 2002. Long-term trends of raptors on Christmas bird counts in the mid- west. Am. Birds 5:15-21. Moorman, C.E. and B.R. Chapman. 1996. Nest-site selec- tion of Red-shouldered and Red-tailed hawks in a managed forest. Wilson Bull. 108:357-368. Pardieck, K.L. AND J.R. Sauer. 2001. The 1995-99 sum- mary of the North American Breeding Bird Survey. Bird Popul. 5:30-48. Rottenborn, S.C. 2000. Nest-site selection and repro- ductive success of urban Red-shouldered Hawks in central California, y. Raptor Res. 34:18-25. Sauer, J.R., J.E. Hines, andJ. Fallon. 2003. The North American breeding bird survey, results and analysis 1966-2002. Version 2003.1, USGS Patuxent Wildife Research Center, Laurel, MD U.S.A. Senchak, S.S. 1991. Home ranges and habitat selection of Red-shouldered Hawks in central Maryland: eval- uating telemetry triangulation errors. M.S. thesis, Vir- ginia Polytechnic Institute and State Univ., Blacks- burg, VA U.S.A. Snedecor, G.W. and W.G. Cochran. 1980. Statistical methods. Iowa State College Press, Ames, lA U.S.A Stewart, R.E. 1949. Ecology of a nesting Red-shouldered Hawk population. Wilson Bull. 61:26-35. Titus, K., M.R. Fulier, D.F. Stauffer, and J.R. Sauer 1989. Buteos. Pages 53-64 in B. Giron Pendleton [Ed.], Proceedings of the northeast raptor manage- ment symposium and workshop. Natl. Wildl. Fed., Washington, DC U.S.A. Received 10 December 2002; accepted 7 June 2004 J Raptor Res. 38(4):320-325 © 2004 The Raptor Research Foundation, Inc. SEASONAL VARIATION IN SEX RATIO OF NESTLING ELEONORA’S FALCONS^ Dietrich Ristow^ Pappelstrafie 35, D-855 79 Neubiberg, Germany Michael Wink Universitdt Heidelberg, Institut fur Pharmazie und Molekulare Biotechnologie, Im Neuenheimer Feld 364, D-69120 Heidelberg, Germany Abstract. — In a breeding colony of Eleonora’s Falcon {Falco eleonorae) on an island offshore of Crete, we determined the gender of 95% of the chicks by molecular (PCR) methods; 1028 samples were collected between 1997-2001. Hatching occurred between 11 August and 12 September with a variation of up to 5 d between years. The overall sex ratio was biased toward male fledglings (52.1%) . The percent of males was positively related to the time of hatching. Falcons that hatched during the first 10 d of the hatching period had a higher daughter to son ratio. We propose that such bias may be adaptive because males with a higher fitness start to breed earlier and invest preferentially in female offspring than in lighter males. This hypothesis could explain some of the discrepancies in earlier sex-ratio studies on other raptors. Keywords: Eleonora’s Falcon; Falco eleonorae; molecular sexing, sex allocation; sex ratio; male fitness. VARIACION ESTACIONAL EN EL COCIENTE DE SEXOS EN PICHONES DE FALCO ELEONORAE Resumen. — En una colonia reproductiva de halcones Falco eleonorae en una isla cercana a la costa de Greta, determinamos el genero del 95% de los pichones por medio de metodos moleculares (PCR) empleando 1028 muestras colectadas entre 1997-2001. La eclosion se produjo entre el 11 de agosto y el 12 de septiembre con una variacion de hasta 5 dias entre los ahos. El cociente de sexos global estuvo sesgado hacia volantones machos (52.1%). El porcent^e de los machos estuvo positivamente relacio- nado con el momento de la eclosion. Los halcones que eclosionaron durante los primeros 10 dias del periodo de eclosion tuvieron un cociente mayor de hijas a hijos. Proponemos que este sesgo podria ser adaptativo ya que los machos con una adecuacion mayor comienzan a reproducirse mas temprano e invierten preferentemente en crias hembras que en machos mas livianos. Esta hipotesis podria explicar algunas de las discrepancias observadas en estudios previos del cociente de sexos en otras rapaces. [Traduccion del equipo editorial] The reason why primary and secondary sex ra- tios in raptors and other birds can deviate from parity has been widely debated (Bennet and Owens 2002, Hardy 2002, Komdeur and Pen 2002). Be- cause it is often difficult to monitor a large number of nests within a single season, some studies suffer from small sample size. Several studies are biased toward species with a pronounced gender size di- morphism because this gender difference was for- merly used to sex fledglings. As a consequence, de- ^ This contribution represents part 28 of a series on Eleo- nora’s Falcon. ^ E-mail address: dietrich.ristow@t-online.de; Wink® uni-hd.de viations from parity tended to be discussed in relation to body size, the main explanations focus- ing on the advantage for female nestlings in food competition with male nestlings or on the advan- tage for males because of their smaller size and lower food requirements. Several obstacles which limited the selection of study species and timing of such studies have been overcome with the intro- duction of molecular techniques for gender deter- mination (Ellegren and Sheldon 1997). We selected Eleonora’s Falcon {Falco eleonorae) for a sex-ratio study because this species breeds in colonies with broods of 1-3 nestlings, so that a suf- ficiently large sample size could be obtained. In spite of its colonial habits, this falcon is monoga- 320 Df.cember 2004 Sex Ratio in Eleonora’s Falcons 321 mous (Swatschek et al. 1993). Furthermore, egg laying starts at the end of July during the stable weather conditions of the Mediterranean summer so that breeding performance rarely differs be- tween years (Walter 1979, Wink et al. 1985, Wink and Ristow 2000) . This falcon is insectivorous dur- ing the courtship period when it feeds upon un- predictable food patches far away from the breed- ing cliff, but is mainly an avian predator during the brood-rearing period in September, when it feeds upon autumn-migrant passerine captured over the sea. Typically, half of the falcon population is 7 yr and older, and males begin to breed at 3 yr of age (Ristow et al. 1989). Methods Field Work. Eleonora’s Falcon is a species of conser- vation concern and included in Annex 1 of the European Union’s Wild Bird Directive. Thus, to avoid disturbances caused by measuring eggs or trapping adults, we restrict- ed our sampling to the period when the young are older than 10 d. Under natural conditions, egg losses are fairly high and nestling mortality comparatively low. We studied a colony of about 150 falcon pairs on an island off Crete (<1 km^ in size) between 1997-2001. Nests were visited once as a rule in mid-September. In ca. 20% of situations young were too small for measure- ment at the first visit, and these nests were revisited a second time about two weeks later. Thus, more than 95% of the fledglings of the colony were banded annually, their wing chord measured, and blood samples of ca. 50 fxl each were taken and stored in DNA buffer (EDTA buffer) in a vial. For both genders, wing chord was converted into hatching date by means of the growth curve formula WC + 52.1 A = r for 45 < WC < 113 and 9.71 WC - 8.3 A = for 113 < WC < 242 6.16 m which WC is the wing chord in mm and A is the age of a nestling in days. The accuracy of this formula is ±1 d (Wink et al. 1991). In some nests with three young, the third bird may experience extended periods of lower growth rate than its older siblings. In such cases the ap- plication of the growth formula would give an age differ- ence in excess of 5 d between second and third nestling, a value which would exceed the maximum difference be- tween egg laying dates (Wink et al. 1985); differences between hatching dates should be smaller or equal to this value in our population (Wink et al. 1991). In such cases (ca. 2% of cases) simply a 5-d age difference was assumed for the runt young. A calculated age difference of up to 7 d was accepted in nests with two young if an unhatched egg explained the gap (ca. 1% of cases). DNA Isolation. Blood samples were preserved in an EDTA buffer (0.1 M Tris, pH 7.4, 10% EDTA, 1% NaF, 0.1% thymol; Wink 1998) and stored at — 20°C until pro- cessing. Total DNA was extracted from the blood samples by an overnight incubation at 37°C in ly.sis buffer (10 mM Tris [pH 7.5], 25 mM EDTA, 75 mM NaCl, 1% SDS) including 1 mg of Proteinase K (Merck, Darmstadt, Ger- many), followed by a standard phenol-chloroform pro- tein extraction. DNA was precipitated from the super- natant with 0.8 volume of cold isopropanol, centrifuged, washed, dried, and resuspended in TE buffer (10 mM Tris-Cl, pH 7.5; 1 mM EDTA). Molecular Sexing. Molecular sexing was modified (Becker and Wink 2002) according to the methods out- lined in Kahn et al. (1998), which are based on the de- tection of the CHD gene on avian sex chromosomes. In most species, males produce one DNA band and females two, presumably reflecting differing intron sizes of the W versus Z chromosomes (Kahn et al. 1998). Polymerase chain reaction (PCR) used were 1237L: GAG AAA CTG TGC AAA ACA G and 1272H: TCC AGA ATA TCT TCT GCT CC. PCR conditions: the PCR mix consisted of 60 ng (2 pi) total DNA in 25 pi total volume, 0.12 pi 1272H Primer (97.45 pmol/pl), 0.103 pi 1237L Primer (83 1 pmol/pl), 1 pi nucleotide-mix (100 pM of GTP, CTP, TTP, and ATP), 2.5 pi lOX buffer with 15 mM MgCl, 0.15 pi Taq-Polymerase (0.6 Units; Pharmacia Biotech, Frei- burg, Germany), and 0.1 pi a-dATP(l pCi). PCR pro- gram: 2 min at 94°C, 31 cycles with 30 sec at 94°C, 1 mm at 56°C, 2 min at 72°C, and finally 10 min at 72°C. After 32 cycles the reaction temperature was main- tained at 72°C for 4 min and then lowered to 4°C for further storage. PCR products were separated electro- phoretically on a denaturing Sequagel matrix at 65 W for 1.5 hr (length 40 cm). After drying, the gel was exposed to an X-ray film (Hyperfilm-MP, Amersham, Freiburg, Germany), for 1-2 d, and developed (X-ray developer and fixer, Kodak, New Haven, CT U.S.A.). Results The gender of 1028 young falcons from 556 nests was determined by molecular sexing (Table 1 ) . As the number of infertile eggs and premature deaths amount to about 10% and <2%, respective- ly (Ristow and Wink 1985, Ristow et al. 1989, Wink et al. 1991), our data mostly reflect the primary sex ratio. There was a tendency to a higher per- centage of sons as compared to daughters when all broods were considered (52.1% compared to the expected frequency of 50%; chi-square test, P < 0 . 1 ). The mean hatching date varied by 5 d between years (Table 2). If the data are corrected for year- to-year variation by setting the first hatching date as day 1, a positive correlation (r = 0.685; N — 24; P< 0.001) was detected between the sex-ratio and the date of hatching. Daughters were more abun- dant during the first 10 d of the hatching period, whereas sons dominated in the middle and final period (Fig. 1). Table 1. Seasonal variation of sex ratios in Eleonora’s Falcons. Numbers of sons (M) and daughters (F) in relation to hatching date and year (Crete, Greece, 1997-2001). Ristow and Wink VoL. 38, No. 4 322 H W FLh [JH 0 H o o (M O o o CM C7) O) cn o i>oiTioO’^oeDijooocnrMi-HO'^^j>ix:ic£5ir>ooooi>ooo o ccoidooo^ocioooo-GcMcaoco^co^doocd^'^idodoooooinoo t£>o^i>cooco«2oouDooT-Hoooo'^oOrHCDCMJ>coooor)^coir)Tt^cO'^ ^1— (G^oOGOiT)^C'Oi>mi>i>i>ir)ir:m'^oocMooi— it-h^ 1 1 CM m CD 00 o m CM 00 CM CM CO CD 1> CMO^CMCOm00 001> 1 CM CM y^ I 1 1 CM 1-H 00 CM CO CM GO GO GO CM CM CM I— H 1— I 1 1 1 iTjHiomcoooin^GM o CD C) CTi 00 1-H CTi 1> O ooooouoi>gogocmcmco I II 1 — 1 i-H CM CM Tfl CO GO CO 00 CM 00 CM CM CM T— 1 !-H I cMio^oo^i-HC^O'fo^icoino CM |r-( I I |cMiO<.OiC5'i5CM-^iO001>COir3ir2CM0O^(MCMrt-^ CMGOcocDcD'tiecT^ooooioai^m-^GM'^ so SOtDiO)C01>00^0-^l:^SOOotCM^CM CO 00 c?^ J> Oi CO CMCM CMiT3rHCMTfc^OrHini>l>i0i00^1>00^CM i— ( I I I I looi-H I |lOM^^cClcor'Oomo^DlOO^;o«Ocooot-^ COoh^O^CMJ>00Mt^0^^Oa)COl>00tO00CMCM^CMC0 T^GM00CMTt^iOl>a3CD00OO'^xi0x0)l>GM00GM | jtMCO | CNCOcO^COiO^Ot-'CO'^^O'.Or'GOGMCM CM CM CO^^^'^tDOOtDt^COOOCOCO^lOi^OGO^^^COr-H w p he a. 1) ^CMcootxncoi>aoo^Oi-HCMoo-^mcOJ>ooCT)O^^GMcoTtiincoi>oocriO CM CM CM CM CM CM CM CM CM CM CO CO r-l CM Sons 124 117 93 102 100 536 Daughters 110 118 91 86 87 492 Total 234 235 184 188 187 1028 December 2004 Sex Ratio in Eleonora’s Falcons 323 Table 2. Yearly variation of hatching dates of sons and daughters, and the percentage of males in Eleonora’s Falcon broods (Crete, Greece, 1997—2001). The median hatch dates are used to characterize the center of the asymmetrical yearly distributions. Year Median FOR Sons Median for Daughters Difference Percentage OF Males 1997 23.5 22.3 1.2 53.0 1998 25.7 24.3 1.4 49.8 1999 28.5 27.0 1.5 50.5 2000 26.0 25.6 0.4 54.3 2001 22.8 21.7 1.1 53.5 Total 24.4 23.5 0.9 52.1 Discussion This study was carried out during a period of population decline caused by poisoning of falcons on mainland Crete resulting with instant death of many adults (Ristow 2001, Anonymous 2002); thus, the number of nestlings sampled decreased from 234-187 between 1997-2001 (Table 1). We assume that this effect was not responsible for the ob- served gender bias, which was present within each year (Table 2) . Nor did we find evidence of a gen- Figure 1. Seasonal variation of sex ratio in Eleonora’s Falcons {N = 1028 fledglings, Crete, Greece, 1997-2001). The Y-axis indicates mean daily percentage of males and the X-axis standardized days of hatching. The following standardization procedure was adopted: 13 August was set as day 1 for 1997 as the first year; the distributions of the following years in Table 1 were shifted by 2, 5, 3, and 0 d, respectively toward earlier dates (Table 2). Then, all broods earlier than day 1 were pooled with those of day 1. Similarly, 5 September was set as day 24 in 1997 and all later broods were pooled with those of day 24. Re- gression equation: Y = 0.94x + 40.33; = 0.469, P < 0 001 . der bias in embryo mortality, although our data from unhatched eggs are few (3 males, 3 females). We interpret the early bias toward daughters in the first third of the hatching period to reflect to some degree the fitness of parents and of males in particular. From previous studies on fitness in Eleo- nora’s Falcon, Wink et al. (1985) established that large and heavy males have larger clutches and that mass between partners was uncorrelated. Mass of males (and likely fitness) increases with age, three- egg clutches tend to be started earlier in the breed- ing period than smaller clutches, and the first egg laid is the heaviest within a clutch (Wink et al. 1982a, 1985, 1991). The sum of these details was that experienced and successful pairs tended to start clutches early, and these produced more daughters. Vice versa, light males apparently pro- duce more sons. In the case of Eleonora’s Falcon, our results may be interpreted that the heaviest (fittest) males can afford to invest into the rarer gender, which needs a larger food supply (daugh- ters were 15-20% heavier than sons; Wink et al. 1982b, 1991). This interpretation is in line with the observation that 1 1 pairs had three daughters each as compared to only seven pairs with three sons each. If the above interpretation is valid, then the age of the adult males should be taken into account when comparing sex ratios among species. For ex- ample, no skewed sex allocation was found in Per- egrine Falcons {Falco peregrinus; Burnham et al. 2003) in North America. The reevaluation of a study with German peregrines (data from Fig. 62 and Table 31, Rockenbauch 2002) also did not re- veal a statistically significant trend in sex ratios. These data sets were obtained in populations that were recovering after severe declines in the 1960s and 1970s and which consisted of a large percent- 324 Ristow and Wink VoL. 38, No. 4 age of young pairs. This might explain why in Aus- tralia, where such a decline had not taken place, a female-biased sex ratio was reported (Olsen and Cockburn 1991). In Eurasian Kestrels {Falco tinnunculus) the pro- portion of sons increased with later laying in years of low and moderate food supply, whereas in years of good food supply the sex ratio was son biased throughout (Korpimaki et al. 2000) . In this Finnish study of typically 30 falcon pairs, the population turn-over was >75% per yr, so that young breeders dominated. Also, in years of good food supply kes- trels of inferior fitness could reproduce without chick loss. We suggest that this pattern is similar to what we observed for Eleonora’s Falcons, in that less-fit males seemed to breed later in the season and produced male-biased broods. In the American Kestrel {Falco sparverius), no trend or the opposite seasonal trend in the sex ra- tio of fledglings was reported (Anderson et al. 1997, Smallwood and Smallwood, 1998, Griggio et al. 2002). A simple explanation for this deviation from the other falcon species was not evident to us, but the extended laying season of 4 mo and the fact that males of this small falcon breed at 1 yr of age may be of relevance. Also in Lesser Kestrels {Falco naumanni) a secondary sex-ratio bias toward daughters as the breeding season progressed has been reported (Telia et al. 1996). Information about parents’ age distribution would have helped to integrate these differing results into our sug- gested broader concept for falcons. After having discussed the available data on fal- con species that agree with or do not agree with our interpretation, it is worthwhile to examine data for other raptor species. Age of parents had been considered in the sex allocation of the Eurasian Sparrowhawk {Accipiter nisuy, Risch and Brinkhof 2002), and these results were consistent with our interpretation of the Eleonora’s Falcon data. How- ever, neither parents’ age nor fitness can explain the mechanism of how birds skew their offsprings’ sex ratio. This becomes obvious when non-raptors are considered. For example, Cory’s Shearwater ( Calonectris diomedea) nesting on the same study is- land showed the opposite seasonal trend in sex ra- tio as compared to the Eleonora’s Falcon (D. Ris- tow and M. Wink unpubl. data). Acknowledgments L. Witte assisted in the field, Mrs. H. Sauer-Giirth, M. Bissinger, A. Ring, S. Wolf, F. Coban, and I. Obreiter in the laboratory. Field work was carried out under permit of the Greek Ministry of Agriculture, Athens, Greece Dimitris Bakaloudis, David Ellis, and Clayton White com- mented on an earlier draft of the manuscript. Literature Cited Anderson, D.J., J. Reeve, and D.M. Bird. 1997. Sexually dimorphic eggs, nestling growth, and sibling compe- tition in American Kestrels Falco sparverius. Fund. Ecol 11:331-335. Anonymous. 2002. RRF resolution: inadvertent poisoning of Eleonora’s Falcon. Wingspan 11:4-5. Becker, P. and M. Wink. 2002. Geschlechtsabhangige GroBenunterschiede von Flugglingen der FluB- .seeschwalbe {Sterna hirundo).]. Ornithol. 143:51-56. Bennet, P.M. and J.P.F. Owens. 2002. Evolutionary ecol- ogy of birds. Oxford Univ. Press, Oxford, U.K. Burnham, W., C. Sandfort, and J.R. Beltoff. 2003. Per- egrine Falcon eggs: egg size, hatchling sex, and clutch sex ratios. Condor 105:327-335. Ellegren, H. and B.C. Shei.don. 1997. New tools for sex identification and the study of sex allocation in birds. Trends Ecol. Evol. 4:143-147. Griggo, M., F. Hamerstrom, R.N. Rosenfield, and G Tavecchia. 2002. Seasonal variation in sex ratio of fledgling American Kestrels: a long term study. Wilson Bull. 114:474-478. Hardy, C.W. 2002. Sex ratios: concepts and research methods. Cambridge Univ. Press, Cambridge, U.K. Kahn, N.W., J. John, and T.W. Quinn. 1998. Chromo- some-specific intron size differences in the avain CHD gene provide an efficient method for sex identifica- tion in birds. Auk 115:1074-1078. Komdeur, J. AND 1. Pen. 2002. Adaptive sex allocation m birds: the complexities of linking theory and practice. Philos. Trans. R. Soc. Lond. Ser. B. Biol. Sci. 357:373- 380. Korpimaki, E., C.A. May, D.T. Parkin, J.H. Welton, and J. Wiehn. 2000. Environmental- and parental-condi- tion related variation in sex ratio of kestrel broods. J Avian Biol. 31:128-134. Olsen, P. and A. Cockburn. 1991. Female-biased sex al- location in Peregrine Falcons and other raptors. Be- hav. Ecol. Sociobiol. 28:417-423. Risch, M. and M.W.G. Brinkhof. 2002. Sex ratios of spar- rowhawk {Accipiter nisus) broods: the importance of age in males. Ornis Fenn. 79:49-59. Ristow, D. 2001. Poison is causing the sudden popula- tion decline in Eleonora’s Falcon. Int. Hawkwatch. 3' 10-17. and M. Wink. 1985. Breeding success and conser- vation management of Eleonora’s Falcon. JCBP Tech- nical Publ. No. 5:147-152. , W. Schariau, and M. Wink. 1989. Population structure and mortality of Eleonora’s Falcon Falco eleo- norae. Pages 321-326 in B.-U. Meyburg and R.D Chancellor [Eds.], Raptors of the modern world. December 2004 Sex Ratio in Eleonora’s Falcons 325 World Working Group on Birds of Prey and Owls, Ber- lin, Germany. Rockenbauch, D. 2002. Der Wanderfalke in Deutschland und umliegenden Gebieten. C. Holzinger, Ludwigs- burg, Germany. Smallwood, P.D. and J.A. Smallwood. 1998. Seasonal shifts in sex ratios of fledgling American Kestrels {Fal- co sparverius paulus): the early bird hypothesis. Evol. Ecol. 12:839-853. SwATSCHEK, I., D. Ristow, W. Scharlau, C. Wink, and M. Wink. 1993. Populationsgenetik und Vaterschaftsana- lyse beim Eleonorenfalken {Ealco eleonorae) . J. Ornithol. 134:137-143. Telia, J.L., J.A. Donazar, JJ- Negro, and F. Hiraldo. 1996. Seasonal and interannual variation in the sex- ratio of Lesser Kestrel Ealco naumanni broods. Ibis 138: 342-345. Walter, H. 1979. Eleonora’s Falcon: adaptations to prey and habitat in a social raptor. Univ. Chicago Press, Chicago, IL U.S.A. Wink, M. 1998. Application of DNA-markers to study the ecology and evolution in raptors. Pages 49-71 in R.D. Chancellor, B.-U. Meyburg, and J.J. Ferrero [Eds.], Holarctic birds of prey. World Working Group on Birds of Prey and Owls, Berlin, Germany. , H. Biebach, F. Feldmann, W. Scharlau, I. Swat- scHEK, C. Wink, and D. Ristow. 1991. Contribution to the breeding biology of Eleonora’s Falcon {Ealco eleonorae). Pages 59-72 in M.K. Nicholls and R. Clark [Eds.], Proceedings of hawk and owl trust conference- biology and conservation of small falcons. Hawk and Owl Trust, London, U.K. and D. Ristow. 2000. Biology and molecular ge- netics of Eleonora’s Falcon Ealco eleonorae, a colonial raptor of Mediterranean islands. Pages 653-668 m R.D. Chancellor and B.-U. Meyburg [Eds.], Raptors at risk. Hancock House, Blaine, WA U.S.A. , , and C. Wink. 1985. Biology of Eleonora’s Falcon {Ealco eleonorae): 7. variability of clutch size, egg dimensions, and egg coloring. Raptor Res. 19:8-14. , C. Wink, and D. Ristow. 1982a. Biologic des Eleonorenfalken: 10. EinfluB der Horstlage auf den Bruterfolg. y. Ornithol. 123:401-408. , , and . 1982b. Biologie des Eleo- norenfalken: 11. Biometrie des Sexualdimorphismus adulter und fltigger Falken. Vdgelwelt 103:225-229. Received 23 March 2004; accepted 13 September 2004 Associate Editor: Fabrizio Sergio J Raptor Res. 38(4):326-333 © 2004 The Raptor Research Foundation, Inc. TAWNY FISH-OWL PREDATION AT FISH FARMS IN TAIWAN Yuan-Hsun Sun^ Institute of Wildlife Conservation, National Pingtung University of Science and Technology, Pingtung, Taiwan 912 Hsin-Ju Wu and Ying Wang Department of Biology, National Taiwan Normal University, Taipei, Taiwan 117 Abstract. — We examined the conflict between cold-water fish farmers and endangered Tawny Fish-Owls {Ketupa flavipes) in Taiwan. From 1994-2000, we surveyed 144 fish farms to assess the level of fish predation by Tawny Fish-Owls and to document farmers’ responses to owl predation. From July 1994- May 1996, studies were conducted at five farms on Nanshih Stream in northern Taiwan and Tachia Stream in central Taiwan to determine the size of fish taken by the owls and the factors affecting predation rates. Owl predation was reported at 25 (17.4%) of the fish farms. Most farmers claimed that owl predation was most frequent during winter, then spring, fall, and summer. At 16 of these farms, owls were trapped with steel leg-hold traps or mist nets, and 10 owls were found drowned or floating in the fish ponds of eight farms. At the five intensively-studied fish farms, the owls took 8-131 (0.04— 0.66%) of ca. 20 000 fish available each year. As the water level in streams increased, owls visited fish farms more often than expected. Owls foraged more frequently on clear nights and caught 101-400 g of fish more often than expected. Key Worus: Tawny Fish-Owl, Ketupa flavipes; fish predation] mortality] fish farms] Taiwan. DEPREDACION DE PECES EN GRANJAS POR PARTE DE KETUPA FLAVIPES EN TAIWAN Resumen. — En este estudio examinamos el conflicto entre los cultivadores de peces de agua fria y la especie de buho araenazada Ketupa flavipes en Taiwan. Entre 1994 y 2000, estudiamos 144 granjas de peces para establecer el nivel de depredacion de peces por parte de K. flavipes y para documentar la respuesta de los cultivadores ante la depredacion por parte de estas aves. Entre julio de 1994 y mayo de 1996, se realizaron estudios en cinco granjas en el arroyo Nanshih en el norte de Taiwan y el arroyo Tachia en el centra del pais para determinar el tamano de los peces consumidos por K. flavipes y los factores determinantes de las tasas de depredacion. Se reporto depredacion por parte de esta especie en 25 (17.4%) cultivos de peces. La mayoria de los cultivadores dijeron que la frecuencia de depreda- cion era maxima durante el invierno y seguidamente menor en la primavera, el otoho y el verano. En 16 de estas granjas se capturaron buhos con trampas de acero o redes de niebla y 10 individuos fueron encontrados ahogados o flotando en los lagos de ocho cultivos. En las cinco granjas estndiadas inten- sivamente, los buhos capturaron entre 8 y 131 (0.04—0.66%) de los aproximadamente 20 000 peces disponibles anualmente. A medida que el nivel del agua en los arroyos se incremento, las aves visitaron los cultivos de peces mas frecuentemente que lo esperado. Los buhos forrajearon mas frecuentemente en noches claras y capturaron 101-400 g de peces con mayor frecuencia que lo esperado. [Traduccion del equipo editorial] Fish-owls are often regarded as nocturnal coun- terparts of the diurnal Osprey {Pandion haliaetus), fish eagles {Ichthyophaga spp.), and sea eagles (Hal- laeetus spp.). Fish-owls include four species of Ke- tupa in Asia and three species of Scotopelia in Africa (Fogden 1973). The Tawny Fish-Owl {K. flavipes), the only fish-owl found in Taiwan, occurs from the Himalayan foothills of Kashmir and Garhwal, east ' E-mail address: ysun@mail.npust.edu.tw to the mountains of northern Laos, Vietnam, and south China, and north almost to the Yellow River (Voous 1988). In Taiwan, the Tawny Fish-Owl is rare, primarily due to the degradation of riparian habitat and illegal hunting (Severinghaus 1987). This species is listed as endangered under the 1989 “Wildlife Conservation Law” (Council of Agricul- ture 1989). In Taiwan, cold-water fish farming in low-eleva- tion mountain streams began in 1960-65 (Tzeng 326 December 2004 Fish Predation by the Tawny Fish-Owl 327 1988). Fish-owls prey upon farmed fish, including rainbow trout {Oncorhynchus mykiss) and ayu {Ple- coglossus altivelis; Wang et al. 1994). Although pro- tected, the owls have been illegally trapped or killed by farmers. While some farmers claim that owl predation causes major losses, to date no data have been presented to substantiate these claims. Understanding the extent and cost of fish pre- dation by owls is necessary to prioritize conserva- tion activities and to implement effective manage- ment of this rare bird. In this study, we investigated Tawny Fish-Owl predation on farmed fish and the interactions between fish-owls and fish farmers. Study Area and Methods We conducted this work in Taiwan. With an annual precipitation of 1000-6700 mm and an annual mean temperature of 22-24°C (Taiwan Forest Bureau 1995), about one half of the island is dominated by luxuriantly- forested mountains. Taiwan has ca. 129 streams, ranging from 10-200 km in length. From the Taiwan Department of Fisheries database, we acquired information on 220 registered cold-water fish farms. We excluded farms in deforested suburban areas, where fish-owls do not reside (Sun 1996). The remaining 144 farms were located in the mountains in potential fish-owl habitat. From 1994-2000, the farmers of these farms were interviewed by questionnaire (with an owl pic- ture) and by telephone. Questions included in the ques- tionnaire simply asked whether farmers had seen this owl depredate fish or found fish scales and remains on walk- ways at their farms. Of the surveyed farms, 108 (75%) were located in central and northern Taiwan, where the climate is cooler and more suitable for cold-water fish farming, and 36 were located in southern Taiwan. We visited each farm that reported owl predation or found evidence of predation and asked farmers to rank the in- tensity of owl predation by season (spring: March-May, summer: June-August, fall: September-November, win- ter: December-February). Farmers provided a predation- intensity score from 1-4, which corresponded to owl pre- dation that was very rare, rare, common, or very common, respectively. We also documented measures taken by farmers to protect fish from owl predation and the fate of owls trapped or taken by farmers. From 1994-96, we closely examined Tawny Fish-Owl predation at four fish farms (two in each of two fish-owl territories) on Nanshih Stream (Sun et al. 2000) and at one fish farm in a fish-owl territory on Tachia Stream. Nanshih Stream ranges from 250-550 m above sea level, and is in northern Taiwan, 30 km south of Taipei. Three streams, Hawun, Chakung, and Talolan, join Nanshih Stream near Fusan, an aboriginal village. On the east and south banks of the stream, the vegetation consists mostly of subtropical rainforest dominated by Ficus and Laura- ceae (Taiwan Forest Bureau 1995). Makino bamboo {Phyllostachys makinoi) and Cryptomeria ( Cryptomeria ja- ponica) plantations, farmland, and human habitations oc- cupy much of the west and south banks. Tachia Stream IS located in central Taiwan. It runs through warm-tem- perate, montane forests of Lauraceae trees (Taiwan Foi- est Bureau 1995), Cyclobalanopsis sp., alder {Alnus formo- sana), Taiwan red pine {Pinus taiivanensis) and Taiwan short-leaf pine {P. morrisonicola) . We documented hsh predation by Tawny Fish-Owls at the Malin Fish Farm, 1000 m in elevation. Each fish farm hatched 30 000-100 000 trout and ayu each year, mainly during January-March. Fish were kept in circular or rectangular fish ponds; each circular pond has an outlet in the center, unlike the rectangular one with an outlet located at the other side of the inlet. Trout grow to marketable size (>500 g) in 12-14 mo and ayu (80-120 g) take 5-7 mo. After harvest, 100-900 trout re- mained at each fish farm where they continued to grow (to 1500-2000 g) through the following year. These fish were sold or consumed at a later date. Unlike trout, ayu perished after breeding. Hence, farmed fish usually were available to owls throughout the year. At these hve farms, we documented hsh stocks, prey remains, the size (g), and species of hsh taken by owls and the dates that owls caught hsh. The size of hsh taken by owls from ponds with hsh of only one age class was easily estimated. For hsh taken from ponds with hsh of more than one age class, we estimated prey mass from sizes of the gills and scales in the remains. We were able to record the time of some predation events by opportunistic observations and radio-tracking. Owls were seen grabbing hsh out of the water by their talons, holding them in talons on the bank, and sometimes plucking the gills and bladders out before swallowing pieces of meat. We captured two Tawny Fish-Owls in each two territo- ries by trapping them at night on tree branches or on pond banks at hsh farms with foot-snare traps. The owls were then radio-tagged prior to release. Radio transmit- ters (MD-205; Telonics Inc., Mesa, AZ U.S.A.) weighed 70-80 g (< 3.5% of the owl’s body mass) and had a life- span of ca. 2 yr. Transmitters were attached dorsally with a backpack harness of wire (1.5 mm in diameter) wrapped inside a tubular teflon ribbon. Owls were locat- ed by homing a directional hand-held H-antenna with a TR-2 receiver (Telonics Inc., Mesa, AZ U.S.A.) and by triangulation, taking at least two bearings for each loca- tion. We also examined whether owls more frequently preyed on hsh in different stream flow conditions or dur- ing periods of different rainfall levels than expected. For each season and owl territory, the expected values were determined based on the proportion of the number of nights in different rainfall or stream flow categories dur- ing our observation period. The observed values were based on proportion of nights that we observed owl pre- dation on farm hsh during the different rainfall and stream-flow categories. We obtained rainfall (mm) data from the Taiwan Central Weather Bureau and water flow (m^s"^) data from the local hydrographic station of the Taiwan Power Company, which was <1 km from hsh farms. Rainfall was categorized into two levels: 0-10 and >10 mm/d. Stream flow was classihed as: low (<10 m^s“9> moderate (11-20 m'^s^^) or high (>20 m%^^). Likewise, we examined whether owls preyed on hsh m certain size classes more often than expected. For each owl territory, the availability of hsh in each size class was estimated as the product of hsh quantity (XlO"*^) and the 328 Sun et al. VoL. 38, No. 4 number of months the fish stayed in the pond. The pro- portion of fish in each size class taken by owls served as a measure of resource use. Chi-square analysis (Conover 1980) was used to deter- mine whether stream flow or rainfall was related to owl predation and to assess owl selection of fish by size. For significant relationships, analyses of selection (Bonfer- roni’s Z test) described by Neu et al. (1974) were tested in terms of the use (observed) versus availability (ex- pected) data. Differences in owl predation intensity among seasons were tested with the Friedman test (Con- over 1980). Data were managed and analyzed with the Statistical Analysis System (SAS Institute 1989). Results A total of 25 (17.4%) of the 144 fish farms re- ported owl predation. Two farms raised ayu, 18 had trout, and five farms raised both species. These farms were all from central and northern Taiwan, where most farms were located. Owl pre- dation rates varied among seasons (Friedman, = 17.0, P = 0.007). Farmers claimed that owl pre- dation was most frequent during winter (x preda- tion intensity score = 3.5, N — 25), than spring (x = 2.4, N — 25), fall (x = 2.2, N = 25), and summer (x = 1.9, A = 25). However, at the two ayu farms (Nanao and Tungao) owl predation was highest during the summer (predation reported as very common, rank = 4). After discovering fish remains on banks, farmers would set steel leg-hold traps or mist nets to catch the predator. Usually, it was only after the owl was captured that the farmers became aware of this rare owl species. Of the 25 fish farms reporting predation, Tawny Fish-Owls were caught at 16 fish farms from 1970-present. One owl was caught at 10 farms, nine farms caught two owls, and three farms caught three or more owls. Alarmingly, two farmers who had been in business for over 10 yr had caught more than five owls each. Farmers re- moved 28-37 owls that were caught in steel leg- hold traps and three in mist nets. In addition, we recorded 10 incidents in which fishing owls were unable to get out of a pond at eight farms. Five owls drowned and five were alive and floating when they were found by farmers in the morning. Most accidents {N = 8) involved owls preying on large trout (600-900 g); an equal num- ber of such incidents occurred in circular and rect- angular fish ponds. At the three Tawny Fish-Owl territories moni- tored, the seasonal pattern of owl predation was somewhat mixed (Fig. 1). Predation intensity gen- erally increased in October or November and was Hsiapen (July 94-May 96) g JASONDJFMAMJJASONDJFMAM o ^ Fusan (October 94-May 96) 15 T JASONDJ FMAMJ JASONDJ FMAM Malin (March 95-May 96) JASONDJ FMAMJJASONDJ FMAM h“ 1994 1 1995 b -1996 H Month Figure 1. Tawny Fish-Owl foraging trip frequency (num- ber of nights/mo) at cold water fish farms in three owl territories along Nanshih and Tachia streams, Taiwan, 1994—96. Data collected at the Hsiapen and Fusan farms were based on radiotelemetry, and date collected at Mal- in Farm based on fish remains found on the banks of fish ponds. highest during the winter. At the Malin Farm, owl predation was highest in April, and it also ap- peared to increase during November 1995 and February 1996. We recorded 206 hunting events, including sightings and fish remains, at the five farms. Except for one early morning hunt, hunting only oc- curred at night. Of the 53 hunts for which the time was known, 28 (52.8%) occurred before midnight. Individual owls visited 1-3 times each night, spend- ing 6-22 min, with a mean of 12.1 min (SD = 7.0, N = 7 nights), hunting for trout near the water surface. Of 15 foraging attempts observed, five were successful (33.3%). Two trout were eaten im- mediately on the bank, and the other three were taken into a nearby forest. Sometimes, we found fish scales and remains at foraging perches near the farms. We observed a pair of owls fishing at the Fusan Fish Farm on nine nights. This farm was lit all night by lights. Usually, the owls perched on a nearby snag, immediately adjacent to the fish farm, where they watched for 1-53 min (x = 9.8 ± 16.4 min) before flying to the farm. December 2004 Fish Predation by the Tawny Fish-Owi. 329 EZa Expected -•—Observed Hsiapen, N^\12 ^ 100 80 ^ Li_ 60 'B 40 I 20 z 0 <100 101-200 201-300 301-400 401-600 601-1000 . Expected —•— Observed Fusan, Af=80 Malin, A^=16 ESa Expected —•—Observed Fish size(g) (D Figure 2. The observed number of fish taken by Tawny Fish-Owls and the availability of fish based estimated fish- months in mass classes (g) at fish farms in Taiwan, July 1994-May 1996. An asterisk p'') indicates P< 0.05, Bonferrom Z test. We found the fresh remains of 60 trout on the banks of fish farms. All trout remains included bloodstains and scales, most also contained gills (77.7%), and some contained the stomach and/or jaws (16.7%). However, only bloodstains and scales were found at fresh ayu remains {N — 37) . Tawny Fish-Owls took fish ranging from 80-1000 g. They chose fish of specific sizes at Hsiapen (x^ = 84.2, df = 5, P — 0.001), Fusan (x^ = 46.5, df = b, P — 0.001) and Malin farms (Fisher exact test, P — 0.004; Fig. 2) . Owls preyed on medium-sized fish (101-400 g) more frequently than expected, and on small fish (<100g) and large fish (>600 g) less often than expected (Bonferroni Ztest, P< 0.05). At Malin Farm, owls foraged on slightly larger fish (300-600 g). In summer, fall, and winter. Tawny Fish-Owls fished on nights with no or light rainfall (Table 1; Bonferroni Z test, P > 0.05). In spring, owls for- aged slightly more frequently than expected dur- ing heavy rain at Fusan and Hsiapen, but this result was not significant (P > 0.05; Table 1). Owls preyed at fish farms more frequently than expected when stream flow was medium and high (Bonferroni Z test, P < 0.05). However, the rela- tionship between water flow and owl predation was not significant (P > 0.05; Table 2). Small sample size may have been a factor. In all seasons and ar- 530 Sun et al. VoL. 38, No. 4 Table 1. Tawny Fish-Owl nighttime foraging trips to cold water hsh farms in three owl territories in Taiwan, in relation to rainfall and season, July 1994-May 1996. Rainfall (mm) Territory Season 0-10 >10 Nights Fusan Summer Expected 82 10 92 Observed 21* 0* 21 Fall Expected 56 5 61 Observed 16 1 17 Spring Expected 84 8 92 Observed 13 2 15 Hsiapen Summer Expected 54 6 60 Observed 16* 0* 16 Fall Expected 140 12 152 Observed 45* 1* 46 Spring Expected 84 8 92 Observed 27 5 32 Winter Expected 160 18 178 Observed 53 5 58 Malin 7/95-6/96 Expected 311 23 334 Observed 33* 0* 33 * Indicates observed value was significantly different than the ex :pected value; P < 0.05, Bonferroni Z test. eas, owls went to fish farms as often, or less often. The stock of fish of the most-often-taken sizes than expected when water flow was low (Table 2). From July 1994— June 1996, Tawny Fish-Owls took a total of 288 fish, including 260 trout (90.3%) and 28 ayu (9.7%) from five fish farms in three fish- owl territories (Table 3). At each of the five fish farms, the owls were known to take 8-131 (0.04- 0.66%) of ca. 20 000 fish available during the year. The estimated annual cost of the fish taken from each farm ranged from $18-$316 US. In 1994—95, owls killed the greatest number of fish at Loshan- chun and Hsinshen farms. In 1995-96, owl preda- tion at these two farms decreased. Discussion Tawny Fish-Owls took fish from less than 20% of the cold-water fish farms in Taiwan. We postulate those fish farms at which owls are not a problem do not lie within owl territories because original riparian forests have heen eradicated (Sun 1996). Our data suggested that Tawny Fish-Owls visited fish farms most frequently during the winter. This pattern may have occurred because owls required greater amounts of energy during the cold winter and because fish of suitable sizes were available. For instance, Sun and Wang (1997) reported that the daytime foraging activities of the predominate- ly nocturnal owls were higher in the winter than in other seasons, based on radiotelemetry data. (101-400 g in the Fusan and Hsiapen territories, and 301-600 g at Malin) was greatest in winter, when trout were 8-12 mo old. The availability of these medium-sized classes may have encouraged owl predation. Neither weather, stream flow, nor fish behavior seemed to explain the higher owl predation in the winter. In the study areas, rainfall was lowest during the winter, especially in central Taiwan (Central Weather Bureau 1995). Although rainbow trout that dwelled in deep (>20 m), cold water during the summer, resided in shallower wa- ter (<10 m in depth) in winter (Fast 1993), we suggest that seasonal changes in the depth at which fish live do not explain seasonal changes in the incidence of owl predation at fish farms. First, in cold water fish ponds, the water temperature changes very little over the year. Second, the water was less than 1.5 m deep in fish ponds. Finally, sick trout were more common during the hot summer, when water temperatures exceeded 24°C. Sick trout usually swam just beneath the surface, mak- ing them easy targets for owls. Tawny Fish-Owl breeding activity may affect their use of Nanshih Stream fish farms during the spring. In spring 1995, a pair of owls nested in vir- gin riparian forest along Chakung Stream (Fusan territory), ca. 800 m from two fish farms. From December 2004 Fish Predation by the Tawny Fish-Owl 331 Table 2. Tawny Fish-Owl hunting trips to cold water fish farms in three owl territories in Taiwan, in relation to water flow and season, July 1994— May 1996. Water Flow Number Territory Season Low Medium High OF Nights Fusan Summer Expected 52 2Tl 3 92 Observed 2 18* 1 21 Fall Expected 28 15 18 61 Observed 0* 5 12* 17 Winter Expected 32 43 15 90 Observed 2^' 7 2 11 Spring Expected 45 42 5 92 Observed 8 5 2 15 Hsiapen Summer Expected 22 26 16 60 Observed 4 8 4 16 Fall Expected 53 56 43 152 Observed 1* 41* 4* 46 Spring Expected 45 42 5 92 Observed 12 16 4 32 Winter Expected 70 93 15 178 Observed 13* 35* 10 58 Malin July 95-June 96 Expected 183 119 32 334 Observed 18 11 5 34 * Indicate.s observed value was significantly different than ex :pected value; P < 0.005, Bonferroni Z test. early February to mid-May, the owls stopped visit- remains were mostly located within 500 m of the ing the farms. We speculate that the distance be- nest of a pair of breeding owls. In addition, for tween the nest and the farms may have been too males, nest defense may be more crucial than ac- great, especially for the male, who delivered food cess to a readily available food source. After mid- to the female during incubation and to the young May, as more food was needed to feed the young. during the brood-rearing period (Sun et al. 1997). parent owls may be stimulated to take additional At Sakatang Stream, pellets, droppings, and prey risks and forage at fish farms. In the remaining two Table 3. Tawny Fish-Owl predation on farmed, cold-water fish and the estimated cost to fish farms in three owl territories in Taiwan, July 1994-96. Number of Fish Cost Date Territory Fish Farm Trout Avu Total (US) July 94-June 95 Fusan Hsinshen 55 0 55 208.6 Fusan 14 0 14 34.9 Total 69 0 69 243.5 Hsiapen Loshanchun 122 9 131 316.3 Hsiapen 13 8 21 58.3 Total 135 17 152 364.6 July 95-June 96 Fusan Hsinshen 21 0 21 79.6 Fusan 3 5 8 18.0 Total 24 5 29 97.6 Hsiapen Loshanchun 20 2 22 98.5 Hsiapen a — — — July 95-June 96 Malin Malin 16 0 16 60.4 Data not recorded. 332 Sun et al. VoL. 38, No. 4 territories and other areas surveyed, variation in the predation rates during the spring and summer may have resulted from the distance between nest- ing sites and fish farms. In Taiwan, most fish farms were built near developed riparian zones in low- land areas. Tawny Fish-Owls are unlikely to nest near these farms or forage at them during the breeding season. Mist netting was not as effective as steel leg-hold traps in capturing Tawny Fish-Owls that preyed on farm fish. This was because mist nets were erected to capture smaller predatory birds such as the Black-crowned Night Heron {Nycticorax nycticorax) , a common nuisance in Taiwan. Therefore, the owls often can escape after initial entanglement. Tawny Fish-Owls can become trapped in pond water and drown. Poole (1989) proposed that drowned Os- preys were not pulled into the water by the large fish they seized. He argued that Ospreys, with the ability to catch prey weighing up to 1500 g, could readily remove their talons from prey if they were too heavy. No fish were found in the talons of in- jured or drowned Tawny Fish-Owls. However, the Tawny Fish-Owls we found in ponds containing large trout (>600 g). Photographs taken with an automatic camera placed at a fish pond revealed that fish-owls catch fish by plunging into the water, as do Osprey (Poole 1989). Blakiston’s Fish-Owls (K. blakistoni) were also seen catching fish by plunging into the water (Yamamoto 1988). We as- sume that catching large fish takes more energy and increases the chance of a struggle in the water. Thus, the chance of injury and death may increase with fish mass, especially for slow-flying birds, such as most owls (Norberg 1987). Two owls preying on small fish also were trapped in the water. In these cases, we believe the strong current in the circular pond, which has an outlet in the center that gen- erates a vortex, was probably responsible for these accidents. We also suggest that owls that fish near the outlet could be sucked into the vortex. During this study, the fishing success of Tawny Fish-Owls at fish farms was 33.3%, somewhat lower than that of the Blakiston’s Fish-Owl (45-50%) in a stream (Yamamoto 1988). Usually, Tawny Fish- Owls spent <1 hr fishing at a fish farm. The owls could quickly catch all the fish they needed be- cause prey was abundant and they eat only 114- 228 g of prey per day (Sun 1996). Owls tended to avoid foraging at fish farms when it rained hard at night. Heavy rain also might re- duce or stop owl foraging in streams by making it difficult for owls to detect prey. After moderate rains made the water in Nanshih Stream turbid. Tawny Fish-Owl hunting of farmed trout increased. This likely occurred because fish-pond water re- mained clear. Most of the small creeks that provide water for the fish ponds drain heavily-vegetated slopes. Water in these smaller streams remained clear during and after moderate rain. Consequent- ly, owls foraged at the fish farms when ponds were clear. However, heavy rains made the water of small creeks and fish ponds turbid, probably decreasing owl predation. Most farm fish caught by Tawny Fish-Owls weighed 101-400 g, or 4.1-16.5% of the owls’ body mass (2200-2650 g). The prey/ predator body mass ratio for Osprey, which catches 150-300 g fish, was also 8.3-16.7% (Poole 1989). We saw a number of large farm trout (600-1000 g) with scratches on both flanks, suggesting an owl had tried and failed to capture the fish. After owls were captured and released at fish farms, they seemed to reduce their hunting at these facilities. In 1994-95, owls killed fewer fish at the Fusan and Hsiapen farms, compared to the Hsinshen and Loshanchun farms. This probably occurred because we first trapped, marked, and re- leased two owls each at the Fusan and Hsiapen farms in the fall of 1993. In 1995-96, we captured, marked, and released the two owls that used the Hsinshen Farm within the Fusan territory, and one owl that used the Loshanchun Farm, of the Hsia- pen territory, again causing a decrease in preda- tion at the farms again. Specifically, the number of fish eaten by owls decreased by 83.2% and 61.8% at the Hsinshen and Loshanchun farms, respec- tively. Lower owl depredation at Malin probably result- ed from other factors, such as an increase in nat- ural prey abundance and effective use of dogs as an aversion measure. The Fusan Farm owner even used the presence of Tawny Fish-Owls to attract birders and photographers, who paid boarding fees. Based on our data, fish farmers were relieved to find out losses to owls were relatively minimal. ACKNOW1.EDGMENTS The following people assisted in the collection of field data: P. Chiang, C. Fang, T. Fu, L. Hsiao, T. Hsu, H. Lee, L. Liao, Y. Liao, H. Mai, P. Mark, Y. Sun, T. Tin, T. Wang, H. Wu, S. Wu, and Y. Wu. Without their great assistance, the study would not have been completed. Y. Cheng, C. Hsu, W. Huang, C. Lee, W. Lee, T. Liao, and M. Shih made this research possible by allowing us access to their December 2004 Fish Predation by the Tawny Fish-Owl 333 fish farms. We thank R. Wallis, D. Johnson, and an anon- ymous referee who provided suggestions on an earlier draft of this manuscript. This research was supported by the Council of Agriculture, Taiwan. Literature Cited Central Weather Bureau. 1995. Astronomical almanac 1995. The Central Weather Bureau, Taipei, Taiwan. Conover, W.J. 1980. Practical nonparametric statistics. John Wiley 8c Sons, New York, NY U.S.A. Council of Agriculture. 1989. The compilation of wild- life conservation laws. Council of Agriculture, Taipei, Taiwan. Fast, A.W. 1993. Distributions of rainbow trout, large- mouth bass, and thread-fin shad in Lake Casitas, Cal- ifornia, with artificial aeration. Calif. Fish Game 79:13- 27. Fogden, M. 1973. Fish-owls, eagle owls, and the Snowy Owl. Pages 53-85 in J.A. Burton [Ed.], Owls of the world: their evolution, structure, and ecology. A&W Visual Library, New York, NY U.S.A. Neu, C.W., C.R. Byers, and J.M. Peek. 1974. A technique for analysis of utilization-availability data. J. Wildl. Manag. 38:541-545. Norberg, R.A. 1987. Evolution, structure and ecology of northern forest owls. Pages 9-43 in R.W. Nero, R.J. Clark, R.J. Knapton, and R.H. Hamre [Eds.], Biology and conservation of northern forest owls. USDA For- est Service, Gen. Tech. Rpt. RM-142, Corvallis, OR U.S.A. Poole, A.F. 1989. Osprey: a natural and unnatural his- tory. Cambridge Univ. Press, Cambridge, U.K. SAS Institute. 1989. SAS user guide: statistics. SAS In- stitute, Cary, NC U.S.A. Severinghaus, L.L. 1987. The Tawny Fish-Owl. Pages 354—355 in A.W. Diamond, L.L. Severinghaus, and C. Chen [Eds.], Save the birds. Pro Nature, Frankfurt, Germany. Sun, Y. 1996. The ecology and conservation of Tawny Fish-Owl in Taiwan. Ph.D. dissertation, Texas A&M Univ., College Station, TX U.S.A. and Y. Wang. 1997. Activity pattern of Tawny Fish- Owl. Wilson Bull 109:377-381. , , AND K.A. Arnold. 1997. Notes on a nest of the Tawny Fish-Owl at Sakatang Stream, Taiwan J Raptor Res. 31:387—389. , , AND C. Lee. 2000. Habitat selection by Tawny Fish-Owls {Ketupa flavipes) in Taiwan./. Raptor Res, 34:102-107. Taiwan Forest Bureau. 1995. The third forest resource and land use inventory in Taiwan. Taiwan Forest Bu- reau, Taipei, Taiwan. Tzeng, C. 1988. The freshwater fishes of Taiwan. Dept, of Education, Taipei, Taiwan. Voous, K.H. 1988. Owls of the northern hemisphere MIT Press, Cambridge, MA U.S.A. Wang, Y, Y. Sun, and L. Liu. 1994. The distribution, ac- tivity pattern, food selection, habitat use, breeding ter- ritory of the Tawny Fish-Owl. The Ecological Research Report of the Council of Agriculture, Taipei, Taiwan. Yamamoto, S. 1988. The hunting techniques of Blakis- ton’s Fish-Owl {Ketupa blakistoni) in Hokkaido. Nemuro Mun. Mus. 7:15-28. Received 27 August 2003; accepted 8 September 2004 J. Raptor Res. 38(4):334-342 © 2004 The Raptor Research Foundation, Inc, GEOGRAPHIC VARIATION IN MORPHOLOGY OF FOUR SPECIES OF MIGRATORY RAPTORS Elise Vernon Pearlstine^ University of Florida, JFAS, 3205 College Avenue, Davie, FL 33314 U.S.A. Daniel B. Thompson University of Nevada, Las Vegas, 4505 Maryland Parkway, Las Vegas, NV 89154 U.S.A. Abstract. — ^We studied geographic variation in morphology of four species of migratory raptor to test large-scale hypotheses of adaptive divergence in quantitative characters among migratory flyways. The Sharp-shinned Hawk {Accipiter striatus) , Cooper’s Hawk (Accipiter cooperii) , Red-tailed Hawk {Buteojamai- censis), and American Kestrel {Falco sparverius) are co-distributed throughout North America. We ex- amined patterns of morphological variation among raptors migrating along two western flyways, the Goshute Mountains of Nevada and Manzano Mountains of New Mexico, and one eastern flyway, Cape May Point in New Jersey. Although they were lower in mass, raptors from western flyways had significantly longer wings, longer tails or both, compared to conspecifics from an eastern flyway. It is significant that parallel variation in flight morphology occurs across four taxa that differ widely in taxonomic affinity, flight habits, size, and shape. Key Words: Sharp-shinned Hawk, Accipiter striatus; Cooper's Haxvk, Accipiter cooperii; Red-tailed Hawk, Buteo jamaicensis; American Kestrel, Falco sparverius; ecomorphology, migration', principal components analysis. VARIACION GEOGRAFICA EN LA MORFOLOGIA DE CUATRO ESPECIES DE RAPACES MIGRATORIAS Resumen.— Estudiamos la variacion geografica en la morfologfa de cuatro especies de rapaces migra- torias para evaluar hipotesis a gran escala sobre la divergencia adaptativa en caracteres cuantitativos entre rutas migratorias. Accipiter striatus, Accipiter cooperii, Buteo jamaicensis, y Falco sparverius encuentran co-distribuidos en America del Norte. Examinamos los patrones de variacion morfologica entre las rapaces encontradas migrando a lo largo de dos rutas migratorias del oeste, las montahas Goshute de Nevada y las montahas Manzano de New Mexico, y una mta del este, Cape May Point en New Jersey. Aunque presentaron menor masa corporal, las rapaces de las rutas del oeste tuvieron alas significati- vamente mas largas, colas significativamente mas largas, o ambas, comparadas con aves coespecfficas de la ruta del este. Es significativo que exista variacion paralela en la morfologia del vuelo en cuatro taxa que difieren enormemente en afinidad taxonomica, habitos de vuelo, tamaho y forma. [Traduccion del equipo editorial] Species with wide distributions are often ex- posed to a variety of environmental conditions that may result in a number of populations having unique morphologies that reflect local conditions. Evolutionary responses to natural selection associ- ated with environmental variation may consist of population-level genetic divergence or phenotypic plasticity (James 1983, 1991, Via and Lande 1985, ^ Corresponding author: research completed as Elise Ver- non Schmidt at University of Nevada, Las Vegas, 4505 Maryland Parkway, Las Vegas, NV 89154 U.S.A; present e-mail: epearls@ufl.edu Bull 1987). Studies of geographic variation in avian populations have documented differences in body size, wing length, leg and foot shape, and bill size and shape that have been correlated with variation in climate, habitat, or ecology (e.g., James 1970, 1991, Wattel 1973, Leisler et al. 1989, Whaley and White 1994, Fitzpatrick and Dunk 1999). In this study, we examine geographic variation of morphology within four species of migratory raptors to determine the degree to which different species exhibit concordant patterns. If there is geo- graphic variation in morphology, it might be attri- buted to environmental conditions in breeding or wintering habitat. James (1991) found that birds 334 December 2004 Migratory Raptor Morphology 335 from cool, dry climates tend to be large while those from warm, humid climates tend to be small as predicted by Bergmann’s rule. Alternatively, mi- gration patterns and habits may influence mor- phology and be exhibited in wing and tail charac- teristics relative to body size. Sharp-shinned Hawks {Accipiter striatus). Coo- per’s Hawks {A. cooperii), Red-tailed Hawks {Buteo jamaicensis) , and American Kestrels (Falco sparver- ius) are all found throughout North America with many populations exhibiting seasonal north-south migration. Individuals were sampled from two mi- gratory routes in western North America and one on the east coast. While little can be said about the breeding grounds for birds captured on migration, some hypotheses can be developed regarding mi- gratory patterns and some general statements about potential adaptations to wintering and breeding habitat can be made. The Goshute Mountains of Nevada and the Man- zano Mountains of New Mexico are monitoring points along major raptor flyways in the west (Hoff- man et al. 2002), and Cape May Point in New Jer- sey is situated on a major eastern flyway (Clark 1985). The Goshute and Manzano mountain fly- ways are both situated along mountainous-ridge systems. Migrants through Cape May Point build up along the Atlantic coastline and funnel into the southern New Jersey peninsula to cross the Dela- ware Bay. Based on available band returns, breed- ing grounds are thought to be north of the western flyways (Smith et al. 1990, Hoffman et al. 2002), and north and east of the eastern flyway (Clark 1985, W. Clark pers. coram.). Goshute and Man- zano migrants travel each fall to wintering grounds in central and western Mexico, a distance that may be as much as twice that of eastern migrants, which tend to remain in the southeastern United States (Clark 1985, Smith et al. 1990, W. Clark pers. comm.). The wintering grounds of raptors provide important habitat for a substantial portion of the year. Consistent variation in morphology across taxa is supportive of the idea that ecological or environ- mental factors may drive such patterns. The single flight-related activity that all these species have in common (western vs. eastern populations) is their migratory pattern. With flights through the Man- zanos and Goshutes occurring in high-altitude-arid mountains and continuing for longer distances, greater flight surfaces may decrease wing loading, and hence may increase flight efficiency in these habitats. A habitat-related variable on both breed- ing and wintering grounds that might influence body size would be climate. However, we were un- able to make specific predictions with regards cli- mate and body size without additional information as to the origin and destination of the birds. Methods Morphological Variables. Between September and No- vember 1991-94 we collected mensural information from raptors trapped while in migration on each of three fly- ways. We measured mass, length of central retrix (Tail), wing chord (Wing) , tarsus length (Tarsus Length) , tarsus width at the narrowest point (Tarsus Width), culmen length (Culmen; bill from cere to tip) and hallux length (Hallux; length of hind claw) using a balance, dial cali- pers, metric ruler, and wing-chord ruler. In the absence of wing and tail surface area measurements, wing and tail length were taken to be indicators of flight-surface area. Flight-surface area is a function of length and width, and wing area increases proportionally as a square of wing span in wings of similar shape (Tucker and Parrott 1970, Pennycuick 1975). Any birds with a noticeable crop were removed from analyses involving body mass. Sharp- shinned Hawks, Cooper’s Hawks, and American Kestrels were grouped according to gender and age because of sexual dimorphism and differential growth of feathers Red-tailed Hawks were grouped only by age. Results are reported for hatch-year birds only; samples of adult birds were too small for statistical analyses. We log-transformed all measurements to conform to a multivariate-normal distribution and compared character by character using StatGraphics (Manguistics, Inc., Rock- ville, MD U.S.A.) multiple analysis of variance. We also performed multivariate analyses to remove possible allo- metric relationships and to investigate morphological shape variables. Principal component scores were deter- mined for each species and gender, after which scores were assigned to individuals. For the principal compo- nents analysis, mass was eliminated from analyses of size and shape because it was highly variable and may have simply reflected the physical condition of migrants. We used analysis of variance to test for significant differences between flyways for each component score. Correlation analysis of the first three principal components and the morphological measurements was also performed. Results Univariate Analyses. With the exception of mar- ginal pattern in the female Sharp-shinned Hawk, all four species of raptor and both genders were significantly heavier in the east than in the west (Tables 1-4). Sharp-shinned Hawks averaged 5% heavier in the east than in the west, American Kes- trels were 7% (males) and 9% (females) heavier in Gape May than in the west. Red-tailed Hawks were about 17% heavier in the east than in the west, and Cooper’s Hawks showed the greatest difference at 23% (males) and 29% (females) heavier in the east 336 Pearlstine and Thompson VoL. 38, No. 4 Table 1. Results of multiple analysis of variance for significant differences in mean morphological characters of male and female hatch-year Sharp-shinned Hawks in three flyways. Significant differences are represented by unique letters for flyways. Standard deviations are in parentheses. Mass was given in grams and linear measurements in millimeters. Values for males are in the first line, females in the second. Character Mai.es Females Goshutes N= 100 N= 81 Manzanos V= 21 N= 9 Cape May N= 24 N= 37 P-value Mass 98.3 (6.3)" 98.4 (7.1)" 102.9 (7.1)” 0.009 162.9 (12.2)" 170.1 (15.5)"*’ 170.5 (15.6)” 0.016 Tail 138.4 (3.0)" 139.3 (3.5)" 131.5 (4.9)” <0.001 162.7 (5.0)" 160.0 (6.3)" 154.1 (5.4)” <0.001 Wing 170.7 (2.9)" 172.3 (2.7)*’ 164.8 (4.2)^ <0.001 202.2 (4.1)" 202.4 (4.5)" 195.6 (4.7)” <0.001 Tarsus length 50.4 (1.2)" 49.6 (TO)” 48.7 (1.8)" <0.001 56.3 (1.4)" 54.8 (1.9)” 55.2 (1.6)” <0.001 Tarsus width 3.4 (0.2)" 3.6 (0.1)” 3.5 (0.2)” 0.001 4.3 (0.3)" 4.4 (0.3)"” 4.5 (0.3)” 0.003 Hallux 11.4 (0.4)" 11.4 (0.4)" 11.1 (0.5)” 0.001 14.3 (0.5)" 14.3 (0.6)" 14.3 (0.6)" 0.944 Gulmen 9.8 (0.3)" 9.7 (0.4)" 10.0 (0.4)” 0.006 11.9 (0.4)" 11.9 (0.3)" 12.2 (0.5)” <0.001 than in the west. Three species had significantly through the Goshute Mountains to have longer longer mean wing and tail length in the west than and/or thinner tarsi than other migrants but the in the east, with the exception of the Cooper’s pattern was not significant for all species. Culmen Hawk. length was longest in Cape May for Sharp-shinned Leg length and width also varied between fly- Hawks, female Cooper’s Hawks and Red-tailed ways. There was a tendency for raptors migrating Hawks. Table 2. Results of analysis of variance for significant difference in morphological characters between flyways for male and female hatch-year Cooper’s Hawks. Significant differences are represented by unique letters for flyways. Standard deviations are in parentheses. Mass was measured in grams and linear measurements in millimeters. Values for males are in the first line, females in the second. Character Goshutes Manzanos Cape May Males N= 37 V= 36 N= 37 Females N= 28 N= 39 N=2l P-value Mass 253.7 (15.5)" 266.6 (26.0)" 339.2 (38.7)” <0.001 378.5 (25.9)" 400.5 (35.5)” 530.3 (41.0)^ <0.001 Tail 188.8 (6.4)" 194.3 (5.1)” 191.0 (5.6)" <0.001 214.1 (8.1)" 217.2 (6.6)" 216.9 (7.8)" 0.254 Wing 220.1 (4.8)" 226.0 (5.0)” 228.0 (4.9)^ <0.001 251.5 (4.8)" 255.2 (5.9)” 260.1 (6.7)'= <0.001 Tarsus length 62.2 (1.9)" 61.7 (1.6)" 65.0 (1.8)” <0.001 68.6 (2.1)" 67.3 (2.3)” 73.2 (1.8)<= <0.001 Tarsus width 5.4 (0.3)" 5.7 (0.3)” 6.0 (0.5)" <0.001 6.6 (0.4)" 7.1 (0.4)” 7.4 (0.5)" <0.001 Hallux 18.8 (0.6)" 18.8 (0.6)" 19.7 (0.6)” <0.001 22.3 (0.8)" 22.2 (0.6)" 23.8 (1.7)” <0.001 Culmen 14.7 (0.5)" 14.7 (1.1)" 14.9 (2.0)" 0.715 17.2 (1.2)" 17.0 (1.5)" 19.2 (0.9)” <0.001 December 2004 Migratory Raptor Morphology 337 Table 3. Results of analysis of variance for significant differences in morphological characters between flyways for hatch-year Red-tailed Hawks. Significant differences are represented by unique letters for flyways. Standard deviations are in parentheses. Mass was measured in grams and linear measurements in millimeters. Character Goshutes N= 152 Manzanos N= 62 Gape May N= 12 P-value Mass 933.4 (150.5)" 950.2 (124.3)" 1134.4 (143.6)b <0.001 Tail 233.7 (11.3)" 235.8 (11.0)" 222.1 (9.5)i> <0.001 Wing 397.4 (17.2)" 403.6 (17.2)^^ 387.3 (14.8)" 0.004 Tarsus length 87.7 (4.2)" 88.5 (4.2)" 85.8 (9.0)" 0.145 Tarsus width 10.8 (0.9)" 11.0 (0.8)" 12.8 (0.7)b <0.001 Hallux 28.8 (2.0)" 29.2 (1.6)" 32.6 (2.2)*^ <0.001 Culmen 25.1 (1.4)" 25.1 (1.4)" 28.1 (1.4)b <0.001 Multivariate Analyses. The first principal com- ponent (PCI) was interpreted as an overall size component in the Cooper’s Hawk and Red-tailed Hawk (Fig. 1). This component explained 37% and 45% of the variation in male and female Coo- per’s Hawks, respectively, and 59% of the variation in Red-tailed Hawks. Weightings (Manly 1994) of all the characters were, generally, equally high. Val- ues of PCI were highest for wing and tail, with val- ues for other characters being lower, in the Sharp- shinned Hawk and in the American Kestrel (Fig. 1). This component explained 32% and 33% of the variation in male and female Sharp-shinned Hawks, and 27% and 29% of the variation in male and female American Kestrels. Individuals of all four species were significantly Length Width ■ Male □ Female Sharp-shinned Hawk Cooper’s Hawk 0.7 - 0.1 Tall Wing Tarsus Tarsus Hallur Culmen Ler^h Width Figure 1. Principal component weightings (PCI) for six morphological variables (Tail = tail length, Wing = wing length, tarsus length, tarsus width, Hallux = hallux length, and Culmen = culmen length) are presented as bar graphs. Genders could not be distinguished in Red-tailed Hawks. 338 Pearlstine and Thompson VoL. 38, No. 4 Table 4. Results of analysis of variance for significant differences in morphological characters between flyways for male and female hatch-year American Kestrels. Significant differences are represented by unique letters for flyways. Standard deviations are in parentheses. Mass was measured in grams, linear measurements are in millimeters. Values for males are in the first line, females in the second. Character Mates Females Goshutes N= 205 N= 158 Manzanos A= 21 V= 13 Cape May N= 14 N= 13 P-VALUE Mass 100.9 (7.9)" 102.5 (7.6)" 109.6 (10.2)'" 0.001 107.6 (9.1)" 114.1 (9.7)b 121.7 (7.5)" <0.001 Tail 121.9 (4.9)" 121.1 (6.2)" 116.4 (4.5)1^ 0.0001 126.6 (5.3)" 124.5 (7.3)">^ 121.8 (3.3)*^ 0.005 Wing 188.7 (5.6)" 191.6 (5.3)*’ 180.1 (5.5)" <0.001 197.1 (6.1)" 196.7 (8.5)" 191.2 (5.7)*^ 0.005 Tarsus length 36.4 (2.1)" 35.6 (1.6)b 35.5 (2.5)"^’ 0.050 36.3 (1.9)" 35.6 (2.0)" 35.4 (1.7)" 0.152 Tarsus width 4.1 (0.3)" 4.1 (0.4)" 3.8 (0.2)'" 0.011 4.2 (0.4)" 4.4 (0.4)b 4.3 (0.4)"!" 0.083 Hallux 9.4 (0.5)" 9.8 (0.7)" 9.5 (0.5)" 0.704 9.8 (0.7)" 9.7 (0.5)" 9.8 (0.4)" 0.954 Culmen 11.6 (0.6)" 11.6 (0.7)" 11.8 (0.8)" 0.254 11.9 (0.7)" 12.3 (0.6)" 12.2 (0.5)" 0.083 Male Female Goshutes Manzanos Cape May Qoshutes Manzanos Cape May Sharp-shinned Hawk Cooper’s Hawk Male Female Goshutes Manzanos Cape May Red-tailed Hawk American Kestrel -Male ■ Female Figure 2. Principal component 1 scores (y-axis) were assigned to individuals and significant differences between flyways were determined by analysis of variance. Significantly different means {P < 0.05) for flyways (Goshute Moun- tains, Manzano Mountains, and Cape May Point) are indicated by unique letters. Genders could not be distinguished in Red-tailed Hawks. December 2004 Migratory Raptor Morphology 339 1 - 0.4 Tail V\fing Tarsus Tarsus Cutmen Hallux Ler^h Width Tail Wirrg Tarsus Tarsus hfalliK Ciimen Length Width Sharp-shinned Hawk Cooper's Hawk o.e 0.8 Tail Wing Tarsus Tarsus Haliux Cuimen Length Width O.e 0.6 0.4 0.2 0 - 0.2 - 0.4 - 0.6 - 0.8 A 1 - Tail Wing Tarsus Tarsus Hallux Cuimen Length Width ■ Male □ Female Red-tailed Hawk American Kestrel Figure 3. Principal component weightings (PC2) for six morphological variables (Tail = tail length, Wing = wing length, tarsus length, tarsus width, Hallux = hallux length, and Cuimen = cuimen length) are presented as bar graphs. Genders could not be distinguished in Red-tailed Hawks. different for mean PCI scores between eastern and western flyways except for female American Kes- trels, which were significantly different only be- tween Goshute Mountains and Cape May Point (Fig. 2). The general pattern indicated large over- all size in eastern Cooper’s Hawk and Red-tailed Hawk migrants, and long mean wings and tail in western Sharp-shinned Hawks and American Kes- trels. PC2 was interpreted to be a shape component in Cooper’s and Red-tailed hawks (Fig. 3) . Tail and wing loaded opposite to other characters. Cooper’s Hawks migrating through the Manzano Mountains were significantly different for mean PC2 scores from those migrating through the other flyways and had longer tails than Goshute and Cape May migrants (Table 2, Fig. 4). This component ex- plained 19% and 20% of the variation in male and female Cooper’s Hawks, and 13% of the variation in Red-tailed Hawks. Red-tailed Hawks migrating through Cape May Point were significantly differ- ent from the other flyways for this component and had shorter wings and tail than western migrants (Table 3, Fig. 3). On the other hand, PC2 consisted of variable weightings of cuimen, hallux, and leg characters in Sharp-shinned Hawks and American Kestrels (Fig. 3). PC2 explained 19% and 20% of the variation in male and female Sharp-shinned Hawks, and 18% of the variation in both sexes of the American Kestrel. Although there were signif- icant differences between some flyways (Fig. 4), variation for this component was difficult to inter- pret in the smaller raptors. Discussion Four species exhibited considerable variation in size, taxonomy, predatory habits, and flight styles, yet we found a common pattern of morphological variation. The four species of raptors exhibited geographic variation in morphology among migra- tory flyway pathways. We were unable to correlate variation in body size with climate due to uncer- tainty of specific geographic location of either breeding or migratory habitat. The morphological 340 Pearlstine and Thompson VoL. 38, No. 4 Goshutes Manzanos Cape May Goshutes Manzanos Cape May Sharp-shinned Hawk Cooper’s Hawk Red-tailed Hawk American Kestrel Figure 4. Principal component 2 scores (y-axis) were assigned to individuals and significant differences between flyways were determined by analysis of variance. Significantly different means (P< 0.05) for flyways (Goshute Moun- tains, Manzano Mountains, and Cape May Point) are indicated by unique letters. Genders could not be distinguished in Red-tailed Hawks. patterns observed could result from either genetic differentiation for the traits or from phenotypic plasticity (James 1983). Of the four species in this study, only the Red-tailed Hawk consists of a num- ber of subspecies or races in North America in the area of the study (Preston and Beane 1993). In a separate study using mitochondrial DNA (mtDNA) from the same individuals, only the Red-tailed Hawk exhibited genetic differences between east- ern and western flyways indicating population ge- netic structure (Pearlstine 2004). Morphological Variation and Migration Distance. The Sharp-shinned Hawk, Red-tailed Hawk, and American Kestrel exhibited variation in body mass, wing chord, and tail length consistently among the migratory flyways, whereas the Cooper’s Hawk did not. Although wing area was not measured, we used wing length as an approximate indication of relative wing area within a species. Increased wing length in western raptors may result in greater wing area, thus reducing wing loading, and per- haps the cost of migratory flight. In contrast to the raptors in this study, migratory populations of pas- serines do not have longer wings but they are more pointed than nonmigratory populations (Mulvihill and Chandler 1991, Senar et al. 1994, Monkkonen 1995). There is a difference in flight strategy, how- ever, as small birds migrate primarily through flap- ping flight, whereas raptors tend to utilize gliding and soaring strategies to minimize flight costs (Ker- linger 1989, Alerstam 1990). The tail is also a vital component of flight sur- face in birds, especially when lift is important (Thomas and Balmford 1995). Raptors can reduce wing loading by from 19.7% (Red-tailed Hawks) to 29.3% (Sharp-shinned Hawks) by simply spreading the tail (Kerlinger 1989). Despite the smaller body mass of western migrants, wings and tail were sig- nificantly longer in Sharp-shinned and Red-tailed hawks relative to Cape May migrants. Wings and tail were significantly longer in male American Kes- trels from the west than from the east and wings were significantly longer in western female Amer- ican Kestrels than eastern. Parallel variation be- December 2004 Migratory Raptor Morphology 341 tween the three species from three flyways suggests possible natural selection relative to migration. Morphological Variation and Climate. Variation in body mass among the three flyways revealed consistently larger body size in the east than in the west. Mean mass for all species was greater in the east, and PCI in the large raptors supports the ob- servation of relatively large eastern individuals for Cooper’s and Red-tailed hawks. Other studies of geographic variation correlated with climate have provided support for Bergmann’s rule, though not for the same geographic areas as this study (James 1970, Aldrich 1984, Murphy 1985, Aldrich and James 1991). A comprehensive review of avian morphology by Zink and Remsen (1986) however, revealed that only 42% of studies provide unam- biguous evidence of correlation between body size and climate. One study of Red-tailed Hawks also failed to support variation in body size according to Bergmann’s rule (Fitzpatrick and Dunk 1999). Additionally, predictions from Bergmann’s rule have not been supported for migratory species. The variation in body size revealed by this study was consistent with results of Zink and Remsen’s (1986) analysis. Bergmann’s rule may hold true however, for body size comparisons made with re- spect to wintering habitat (e.g., Johnston and Fleischer 1981, Wiedenfeld 1991). Clearly, raptors must be studied on both wintering and breeding grounds to further understand this pattern. Acknowledgments The author thanks Peter C. Frederick and JefR. Jaeger for valuable comments on the manuscript. This study could not have been completed without the assistance of Charles L. Douglas, Brett R. Riddle, Donald H. Baepler, Clayton M. White, and James E. Deacon. Personnel from Hawkwatch International and Cape May Point were an integral part of this study. Special thanks go to Stephen Hoffman, William C. Clark, Chris Schultz, Phil Magasich, and Paul A. Napier. Funding was provided by a doctoral dissertation improvement grant (DEB 9321656); the Ste- phen R. Tully Memorial Grant from the Raptor Research Foundation, Inc.; the Arizona-Nevada Academy of Sci- ences; Hawkwatch International; the University of Neva- da, Las Vegas (UNLV) Graduate College; the Department of Biological Sciences at UNLV; the Marjorie Barrick Mu- seum and fellowship at UNLV; and the Women in Sci- ence Award at UNLV. Additional support for this re- search was received from the Florida Agricultural Experiment Station, and approved for publication as Journal series R-10274. Literature Cited Aldrich, J.W. 1984. Geographical variation in size and proportions of Song Sparrows (Melospiza melodia). Or- nithol. Monogr. 35. AND F.C. James 1991. Ecogeographic variation in the American Robin {Turdus migratorius) . Auk 108* 230-249. Alerstam, T. 1990. Bird migration. Cambridge Univ. Press, Cambridge, U.K. Bull, JJ- 1987. Evolution of phenotypic variance. Evol. 41:303-315. Clark, W.S. 1985. The migrating Sharp-shinned Hawk at Cape May Point: banding and recovery results. Pages 137-148 in M. Harwood [Ed.], Proceedings of hawk migration conference IV. Hawk Migration Association of North America, Washington Depot, CT U.S.A. Eitzpatrick, B.M. and J.R. Dunk. 1999. Ecogeographic variation in morphology of Red-tailed Hawks in west- ern North America./. Raptor Res. 33:305-312. Hoffman, S.W., J.R Smith, and T.D. Meehan. 2002. Breeding grounds, winter ranges, and migratory routes of raptors in the mountain west. /. Raptor Res 36:97-110. James, F.C. 1970. Geographic size variation in birds and its relationship to climate. Ecology 51:365—390. . 1983. Environmental component of morpholog- ical differentiation in birds. Science 221:184-186. . 1991. Complementary descriptive and experi- mental studies of clinal variation in birds. Am. Zool. 31:694-706. Johnston, R.F. and R.C. Fleischer. 1981. Overwinter mortality and sexual size dimorphism in the House Sparrow. Auk 98:503—511. Kerlinger, P. 1989. Flight strategies of migrating hawks. Univ. Chicago Press, Chicago, IL U.S.A. Leisler, B., H.-W. Ley, and H. Winkler. 1989. Habitat, behavior, and morphology of warblers: an integrated analysis. Ornis Scand. 20:181-186. Manly, B.F. 1994. Multivariate statistical methods. Chap- man and Hall, New York, NY U.S.A. Monkkonen, M. 1995. Do migrant birds have more pointed wings?: a comparative study. Evol. Ecol. 9:520- 528. Mulvihill, R.S. and C.R. Chandler. 1991. A comparison of wing shape between migratory and sedentary Dark- eyed Juncos (Junco hyemalis) . Condor 93:172-175. Murphy, E.C. 1985. Bergmann’s rule, seasonality, and geographic variation in body size of House Sparrows Evol. 39:1327-1334. Pearlstine, E.V. 2004. Variation in mitochondrial DNA of four species of migratory raptors. /. Raptor Res. 38' 250-255. Pennycuick, C. 1975. Animal flight. The Institute of Bi- ology’s Studies in Biology, London, U.K. Preston, C.R. and R.D. Beane. 1993. Red-tailed Hawk {Buteo jamaicensis) . In A. Poole and F. Gill [Eds.], The birds of North America, No. 52. The Birds of North America, Philadelphia, PA U.S.A. Senar, J.C., J. Lleonart, and N.B. Metcalfe. 1994. Wing- shape variation between resident and transient win- tering siskins Carduelis spinus. J. Avian Biol. 25:50-54. 342 Pearlstine and Thompson VoL. 38, No. 4 Smith, J.P., S.W. Hoffman, andJ.A. Gessaman. 1990. Re- gional size differences among fall-migrant accipiters in North America./. Field Ornithol. 61:192-200. Thomas, A.L.R. and A. Bai.mford. 1995. How natural se- lection shapes birds’ tails. Am. Nat. 146:848-868. Tucker, V.A. and G.C. Parrott. 1970. Aerodynamics of gliding flight in a falcon and other birds. J. Exp. Biol. 52:345-367. Via, S. and R. Lande. 1985. Genotype-environment inter- action and the evolution of phenotypic plasticity. Evol. 39:505-522. Wattel, J. 1973. Geographical differentiation in the ge- nus Accipiter. Nuttall Ornithological Club, Cambridge, MA U.S.A. Whaley, W.H. and C.M. White. 1994. Trends in geo- graphic variation of Cooper’s Hawk and Northern Goshawk in North America: a multivariate analysis. Proceedings of the Western Foundation of Vertebrate Zoology, Camarillo, CA U.S.A. Wiedenfeld, D.A. 1991. Geographical morphology of male Yellow Warblers. Co«(for 93:712-723. Zink, R.M. and J.V. Remsen, Jr. 1986. Evolutionary pro- cesses and patterns of geographic variation in birds Curr. Ornithol. 4:1—69. Received 30 July 2003; accepted 7 August 2004 J. Raptor Res. 38(4);343— 349 © 2004 The Raptor Research Foundation, Inc. HOME-RANGE SIZE OF THE JAVAN HAWK-EAGLE {SPIZAETUS BARTELSI) ESTIMATED FROM DIRECT OBSERVATIONS AND RADIOTELEMETRY Jan Ove Ojershaug^, Nils Rgv, and Torgeir NygArd Norwegian Institute for Nature Research, Tungasletta 2, N-7485 Trondheim, Norway Dewi M. Prawiradiiaga Research Centre for Biology-LIPI, Fauna Cf Flora International-IP, P O. Box 230, Bogor 1 6002, Indonesia M. Yayat Afianto and Hapsoro Telapak Indonesia, f Sempur Kaler No. 16, Bogor 16154, Indonesia Adam Supriatna KPB CIBA, Kotak Pos 66, Sindanglaya, Cianjur 43253, Indonesia Abstract. — The mean home-range size of the Javan Hawk-Eagle {Spizaetus bartelsi) on Java was estimated to be ca. 400 ha based on three different methods. The distance between nests of neighboring pairs was ca. 3 km in Gede-Pangrango and 2 km in the Salak Mountains. In the Halimun Mountains, the mean distance between territories was 1.8 km. Radiotracking of one adult male indicated a home-range size of 300 ha in the nonbreeding season. This finding suggested that earlier population estimates probably were too low, as they were based on home-range estimates of 2000-5000 ha per pair depending on habitat quality. The species should still be considered endangered, as it is threatened both from habitat loss and illegal hunting. Key Words: favan Hawk-Eagle, Spizaetus bartelsi; home range, population status', radio-tracking, Indonesia. TAMANO DEL ArEA DE HOGAR DE SPIZAETUS BARTELSI ESTIMADO A PARTIR DE OBSERVA- CIONES DIRECTAS Y RADIOTELEMETRIA Resumen. — -El tamano medio del area de hogar de Spizaetus bartelsi en Java ha sido estimado en apro- ximadamente 400 ha considerando tres metodos distintos. La distancia entre los limites de las areas de hogar de parejas vecinas fue de aproximadamente 3 km en Gede-Pangrango y 2 km en las montahas de Salak. En las montahas de Halimun, la distancia media entre territorios fue de 1.8 km. El seguimiento con radiotransmisores de un macho adulto indico un area de hogar de 300 ha en la estacion no reproductiva. Esto sugirio que las estimaciones poblacionales anteriores fueron probablemente muy bajas, ya que estuvieron basadas en estimaciones de areas de hogar de 2000-5000 ha por pareja, de- pendiendo en la calidad del habitat. La especie debe aun ser considerada en peligro, ya que esta amenazada tanto por la perdida de habitat como por la caceria ilegal. [Traduccion del equipo editorial] The Javan Hawk-Eagle (Spizaetus bartelsi) is en- demic to the rainforests of Java, Indonesia, where less than 10% of the original natural forests remain (Whitten et al. 1996). Small population size, severe habitat loss, forest fragmentation, and illegal hunt- ing have all contributed to the “endangered” sta- tus of this species on the world list of threatened ' E-mail address: jan.o.gjershaug@nina.no birds (BirdLife International 2000, BirdLife Inter- national 2001). The population size has been esti- mated differently by various authors; “Not more than 60 breeding pairs” (Meyburg et al. 1989), “67-81 pairs” (van Balen and Meyburg 1994), “81-108 pairs” (Sozer and Nijman 1995), and “137-188 pairs” (van Balen 1999, van Balen et al 2000) . These estimates were based on data on the size of two home ranges. Thiollay and Meyburg (1988) estimated the home-range size to be 2000— 343 344 Gjershaug et al. VoL. 38, No. 4 3000 ha, but used 1700-4500 ha when they calcu- lated their population estimates. Meyburg et al. (1989) suggested that suboptimal habitat may sup- port home ranges as large as 12 000 ha. The home- range size of a breeding male studied in west Java was estimated at a minimum of 1200 ha (Sozer and Nijman 1995), and of another adult in central Java at ca. 3600 ha (van Balen 1999). To obtain accurate knowledge of population size, it is necessary to obtain reliable data on den- sity and area of suitable habitat. Density estimates have usually been based on the size of a few home ranges mapped by sight observations (Meyburg et al. 1989, Sozer and Nijman 1995, van Balen and Meyburg 1994, van Balen et al. 2000) under the assumptions that the eagle pairs occupy contiguous breeding territories and that entire forest areas are used by breeding pairs. We studied the home range of the Javan Hawk-Eagle in western Java by three different methods: (1) direct observation of territorial behavior of breeding pairs, (2) the dis- tances between nests, and (3) radio-tracking of one adult male. Distance between neighbor pairs was defined as the distance between the centroids of their terri- tories. When nests were known, we used the dis- tance between occupied nests of the same year. In this paper, we assume that home range is the same as territory, as the entire home range seems to be defended by Javan Hawk-Eagles during the breed- ing season (pers. obs.). Study Area and Methods Halimun. This study area is close to Ciptarasa village on the slopes of the Halimun Mountains in west Java, and is close to and partly inside Halimun National Park (Fig. 1). Most observations were made from open cultivated areas along the forest borders. The rainforests were of lowland or lower-montane type at ca. 1000 m above sea level (masl). The national park is comprised of ca. 20% lowland forest (Whitten et al. 1996). Halimun National Park (established 1992) covers an area of 40000 ha (Whitten et al. 1996) and is one of the largest forested areas in Java. Around it are large forested areas in ad- ministrative management as production or protected for- ests. Relatively large areas of primary rainforest still exist outside the borders of the park. However, because of need for cultivated land, the surrounding forests are gradually transformed into gardens and rice fields. An- nual rainfall is between 4000-6000 mm. During large parts of the year, the highest mountains are covered by mist and fog, which its Indonesian name indicates. Mt. Salak. This mountain is a volcano 2211 masl, well vegetated to the top (Fig. 1). The area of forest contain- ing Javan Hawk-Eagle has been estimated to 10 000 ha (van Balen et al. 1999). The forests have administrative Figure 1. The study areas: 1 = Halimun, 2 = Mt. Salak, 3 = Gede-Pangrango. status as production forest. Part of the forest, particularly on the lower slopes of the mountain, has been trans- formed into tree plantations or secondary forests. The lowermost part of the forest, bordering cultivated areas, is used by local villagers for collecting forest products There are large areas of primary forest present on the mountain, mostly at higher elevations. Mt. Salak has a very high annual rainfall and is an important water-catch- ment area. Gede-Pangrango. This national park includes the vol- canoes Mt. Gede (2958 m) and Mt. Pangrango (3019 m) It contains 15 196 ha of rainforest (Fig. 1) and includes some of the oldest, protected forests in Indonesia. The annual rainfall is 3000-7000 mm (RePPProT 1990). There is little seasonal variation with only a slight de- crease in rainfall from May-August. However, El Nino events (such as in 1997) may result in extended and more pronounced dry seasons, lasting until November. The park consists mainly of montane forest and includes the botanical gardens of Cibodas. The nearby Telaga Warna Nature Reserve surrounds a small lake; the re- serve proper covers an area of 350 ha. The total area of this study area is ca. 20 000 ha (van Balen et al. 2000) . Field Observations. Behavior of territorial pairs. When weather conditions are favorable, eagles may soar above the forest within their home ranges, and perform terri- torial displays. Therefore, the birds can be observed from places with a good view of the surrounding terrain. To identify the home ranges of the different pairs, we used December 2004 Home-range of Javan Hawk-Eagle 345 Figure 2. The distribution of Javan Hawk-Eagle home ranges at the southeastern border of Halimun. The home-range sizes were estimated using the distance be- tween centroids of neighboring territories as the diame- ter of a hypothetical-circular home range. The dashed line denotes the boundary of the national park, the dark gray area represents forest, and light gray area represents open areas. Triangles and numbers = high points and elevation in m. one of the following criteria: either simultaneous obser- vation of neighboring pairs, individual recognition based on variation in plumage color, or molt pattern. To obtain an estimate of die density of territorial eagles in the area, four observers performed observations along a 10 km stretch of forest edge in the Halimun study area over 9 d during two time periods, 30 August-2 September and 26—30 September 1997. The total observation time was 67 hr. The map of the study area (Fig. 2) is based on maps in Whitten et al. (1996) and from Biodiversity Con- servation Project in Indonesia (1997). The border be- tween home ranges was mapped based on observations of territorial display and other flight activity, and the dis- tances between the centroids of the five neighboring ter- ritories were measured from the map (Fig. 2) . Distances between nests. In the study areas in Mt. Salak and Gede-Pangrango, we were given the locations of six nests by local inhabitants. Therefore, we were able to measure the distances between nests of neighbor pairs in these areas. For this study, we included three nests which were identified in 1997 in Salak. These nests were plotted on available maps (Bakosurtanal 1997) and the distances between them were measured on the maps. The same method was used on three occupied neighbor nests in Gede-Pangrango in 1998, which were plotted on maps from Bakosurtanal (1990a, 1990b). Radiotelemetry. An adult Javan Hawk-Eagle male was caught by use of a snare on the nest in Mt. Salak on 19 October 1997, and equipped with a VHF transmitter weighing 25 g (BioTrack Inc., Dorset, U.K.). The trans- mitter was equipped with a mercury activity switch. It was attached as a backpack by a harness made of Teflon (Bally Ribbon Mills, Bally, PA U.S.A.) ribbon using a Y- type attachment (Buehler et al. 1995). This male was ra- dio-tracked intensively from five fixed receiver stations from 31 March-15 May 1998 with a four-element yagi hand-held antenna. The coordinates of these receiver sta- tions (locations not differentially corrected) were deter- mined with a GPS receiver (GPS 45, Garmin Internation- al Inc., Olathe, KS U.S.A.). During observations, data were collected every 30-min during a 5-10 min interval. Locations were obtained by simultaneous triangulations from two different stations. The number of locations estimated from each station var- ied from 2-24/d. The positions were often confirmed by visual observations of the eagle flying or perching. The accuracy of the location estimates is probably reasonable due to the short distances (typically <500 m) from the receiver sites to the eagle and the good overview of the area from overlooks. Two types of behaviors were inter- preted from radio signals: long pulse as sitting, short or variable pulse as flying or eating. We collected 126 locations for this Javan Hawk-Eagle during our fieldwork. Home range was calculated with the minimum convex polygon and fixed kernel methods (Worton 1989) using the software program Arc\dew and the Animal Movements extension (Hooge and Eichen- laub 2000). Results Home Range. Direct observation of territorial behav- ior. In Halimun, we found six pairs of Javan Hawk- Eagle along a 10-km distance of forest edge (Fig. 2) . The mean estimated distance between cen- troids of five neighboring territories was 1.8 km, which gives a hypothetical-circular home-range size of 254 ha (Table 1). One home range in Gede- Pangrango was mapped based on sight observa- tions of eagles. This range was 530 ha, of which 220 ha was forest and the rest was tea plantation. Distances between nests. In Gede-Pangrango, we found that the distances between three occupied nests from the same year were ca. 3 km, which gives a hypothetical circular home-range size of 710 ha. The distances between three nests occupied the same year in the Salak area were ca. 2 km, resulting in a hypothetical circular home-range size of 314 ha (Table 1). Radiotelemetry. The home range of the radio- equipped Javan Hawk-Eagle male from 31 March- 15 May 1998 was confined within an area of 1.5 X 2.2 km. It was located at 6°40'-6°41'S and 106°44'- 106°00'E, and between 620-1550 m altitude (Fig. 3) . The home-range size according to the 95%- probability contour of the fixed-kernel method was 289 ha. A minimum-convex polygon around all fix- es gave a home range of 310 ha. Hence, we esti- mated the home range of this eagle at ca. 300 ha (Table 1). The types of habitat in the home range consisted 346 Gjershaug et al. VoL. 38, No. 4 Table 1. Home-range sizes estimated by different methods: direct observations of territorial behavior of breeding pairs, distances between nests, and radio-tracking of an adult male. Method Distance Between Activity Centers ( km) (A) Estimated Size of Home Range ( ha) Study Area Distance between home-range centroids 1.8 (4) 254 Halimun 1997 Mapped by sight observations (1) 530 Gede-Pangrango 1998 Distance between nests 3 (2) 710 Gede-Pangrango 1998 Distance between nests 2 (2) 314 Salak 1997 Mapped by radiotelemetry (1) 300 Salak 1997 of undisturbed primary forest, production forest {Pinus sp.) and disturbed natural forest. Frequent observations of soaring and displaying Crested Ser- pent Eagle {Spilornis cheela). Black Eagle {Ictinaetus malayensis), and Changeable Hawk-Eagle (Spizaetus N 0 0.5 1 1.6 2 km Figure 3. Home range of a VHF radio-equipped adult male Javan Hawk-Eagle in the nonbreeding season. The 95, 75, 50, and 25% probability contours using a fixed- kernel method as shown. The black circles show 126 ra- diotelemetry locations. A least-square cross validation for the smoothing factor (H) was performed, and its value was set to 200. drrhatus limnaeetus) indicated that the home range of the Javan Hawk-Eagle partly overlapped the home ranges of these eagles. In Cede-Pangrango, no observations of Javan Hawk-Eagles were made at altitudes over 2000 mask The six nests were situated between 1200- 1400 mask Time Budget and Behavior. The time budget data from 75 hr of observations between 0600- 1800 H of the radio-tracked male in Salak showed that this Javan Hawk-Eagle spent 42.8% of its time flying or feeding, and 57.2% perching. Discussion Home-range Size. When using a hand-held an- tenna in the field, it is difficult to obtain better than 5 degrees accuracy on the signal bearing. This would result in a 10 m error at a tracking distance of 100 m (Kenward 2001). The outermost fixes were up to 1 km away from the observer, thus in- volving a potential error of ca. 100 m in the loca- tion accuracy. Although we did not test the accu- racy of locations, we suggest that some of our location errors may have been compensatory (i.e., one location could have had an error of several hundred meters to the east, while another location could have had an error similar distance to the west. Therefore, even though the accuracy of our telemetry data were limited, we feel that our results provided a reasonable approximation of the home- range size used by one Javan Hawk-Eagle. Our estimates of home-range sizes of 230-710 ha, suggesting a median value of ca. 400 ha (Table 1 ) are considerably lower than those given by other authors (Meyburg et al. 1989, Sozer and Nijman 1995, Thiollay and Meyburg 1988), which ranged from 1200-12 000 ha. Madrid et al. (1991) found that the home-range size of the Ornate Hawk-Eagle December 2004 Home-range oe Javan Hawk-Eagle 347 {Spizaetus ornatus) in Guatemala was 800 ha for males and 1300-2100 ha for females. In the Japa- nese Mountain Hawk-Eagle {Spizaetus nipalensis or- lentalis), Yamazaki (2000) found that home ranges were normally >2000 ha, (neighbor-nest distances 1.5-5. 6 km, X = 4 km). The mean density of breed- ing pairs of this species has been calculated at one pair per 2500-2800 ha uniformly throughout Ja- pan (Yamazaki 2000). A similar home-range size (1270-3230 ha) has been recorded in the Philip- pine Hawk-Eagle {Spizaetus philippensis; Preleuthner and Gamauf 1998). However, relatively small home ranges (ca. 650 ha) have been documented in Crowned Hawk-Eagles {Stephanoaetus coronatus; Shultz 2002), and the nearest-neighbor distance between nests averaged 1.8 km. In the huge Harpy Eagle {Harpia harpyja), occupied nests have been recorded as close as 3-5 km apart in South and Central America (del Hoyo et al. 1994). Our results from a single radio-tracked male probably represent the home-range size of a pair. We suggest that the male typically uses the com- bined home ranges of the male and the female exploited during the breeding season. Our telem- etry data were collected about 5 mo after the young left the nest. This use area probably repre- sents the breeding home range, as the juveniles stay with their parents for a year or longer (Nijman et al. 2000) . Also, we repeatedly saw territorial in- teractions between neighboring pairs, suggesting that adjacent home ranges were defended. In the Halimun area, we observed Changeable Hawk-Eagles, Black Eagles, and Crested Serpent Eagles commonly in the home ranges of Javan Hawk-Eagles indicating home-range overlap among these species (R0v et al. 2000). Habitat Use. Thiollay and Meyburg (1988) sug- gested that the Javan Hawk-Eagle was dependent on primary rainforest, although those authors also mention that the species was seen in three degrad- ed forest areas around Bogor. They also stated that hawk-eagles were seen flying over a plantation be- tween two patches of forest as well as perched near a road in secondary forest at Mem Betiri in east Java. Nijman and van Balen (2003) found that the prime habitat for adult Javan Hawk-Eagles were ev- ergreen forest and to a lesser degree secondary for- est, and that immatures and juveniles had a greater preference for open woodland (forests with large clearings, small forest fragments, and young tree plantations) than adults. Our observations suggest that the species uses both primary and secondary forests for hunting and nesting. We have observed Javan Hawk-Eagles hunting over cultivated areas both in Halimun and Gede-Pangrango with home ranges that included such areas. We did not observe this for our radio- tagged bird in Salak as this home range did not include cultivated sites. When the home ranges are bordering open areas, the eagles may include some of these areas in their home ranges (Nijman and van Balen 2003, Nijman 2004, pers. obs.). However, our observations may be biased because they were made along forest borders. We have observed that the eagles can obtain food opportunistically from outside its primary habitat. This is based on prey remains of Barred Buttonquail ( Turnix suscitator ) , an open-habitat species, and shows that some prey must be taken outside the forest. We have also ob- tained information from local people that domes- tic chickens have been taken in a village close to a nest of Javan Hawk-Eagle (Prawiradilaga et al. 2000 ). Population Status. The number of breeding pairs of this species has been estimated by dividing the area of presumed suitable habitat by assumed home-range size. The Javan Hawk-Eagle normally does not nest above ca. 1400 masl (Nijman et al. 2000, pers. obs.); therefore, we excluded these high altitude areas to estimate population size. In areas such as Gede-Pangrango, the eagles some- times soared to high altitudes, but they were never seen to be encountered by other conspecifics at these higher elevations. If we use 3 km as a mean home-range diameter for the Gede-Pangrango area, this habitat would provide space for about 20 pairs, which is 2-3 times more than previous esti- mate of 6-10 pairs (van Balen et al. 2001). Van Balen et al. (2000) estimated the size of available Javan Hawk-Eagle habitat on all of Java at ca. 5480 km^ in 22 forest areas (also see van Balen et al. 2001). Based on this finding and a density estimate of one pair per 2000—5000 ha, they esti- mated the total population of Javan Hawk-Eagles to be between 137 and 200 pairs (van Balen 1999, Nijman et al. 2000). Our data indicated that this estimate probably was too conservative. Extrapola- tion of our Gede-Pangrango densities to the entire forest habitat would place the population size be- tween 270-600 (median = 435) pairs. However, we must admit that the accuracy of our home-range estimates are limited and our island-wide popula- tion estimate is based on a number of assumptions. Therefore, we recommend more studies to be car- 348 Gjershaug et al. VoL. 38, No. 4 ried out in different forest habitats in other parts of Java. Nevertheless, the species should still be re- garded as endangered; it is threatened by both habitat loss and illegal hunting. Because of this, we urge implementation of the Species Recovery Plan (Sozer et al. 1998) proposed by the Javan Hawk- Eagle Working Group to ensure the future conser- vation of the species. Acknowledgments We are grateful to the following institutions and per- sons: The Indonesian Institute for Sciences provided re- search permits. We had a fruitful cooperation with The Ministry of State for Environment. BirdLife International Indonesia Programme provided valuable advice and help during all parts of the study. We thank Harry, Mulyadi, Bob, Arbi, Dwi (Telapak members), Usep, Eddy, Mulyadi, Bapa Dili, Siti Nuraeni (KPB CIBA members) , local peo- ple at Mt. Salak, (especially Pak Ocen), Pak Otang, Pak Onyon, and especially Pak Kokon at Halimun, for their invaluable help during fieldwork. Literature Cited Bakosurtanal. 1990a. Petarupabumi cugeunang lembar 1209-213, 1: 25000. Bakosurtanal, Cibinong-Bogor, Indonesia. . 1990b. Peta rupaburai cisarua lembar 1209-142, 1: 25000. Bakosurtanal, Cibinong-Bogor, Indonesia. . 1997. Peta rupabumi G. salak lembar 1209-141, 1: 25000. Bakosurtanal, Cibinong-Bogor, Indonesia. Biodiversity Conservation Project in Indonesia. 1997. Research and conservation of biodiversity in Indone- sia. LIPI, PHPA and JICA, Bogor, Indonesia. BirdLife International. 2000. Threatened birds of the world. Lynx Edicions and BirdLife International, Bar- celona, Spain. . 2001. Threatened birds of Asia: the BirdLife In- ternational red data book. BirdLife International, Cambridge, U.K. Buehler, D.A., J.D. Fraser, M.R. Fuller, L.S. McAllis- ter, and J.K.D. Seegar. 1995. Captive and field-tested radio transmitter attachment for Bald Eagles. J. Field Ornithol. 66:173-180. del Hoyo, J., a. Elliott, and J. Sargatal. 1994. Hand- book of the birds of the world. Lynx Edicions, Bar- celona, Spain. Hooge, P.N. and B. Eichenlaub. 2000. Animal move- ment extension to ArcView: version 2.0. Alaska Sci- ence Center, Biological Science Office, U.S. Geologi- cal Survey, Anchorage, AK U.S.A. Kenward, R.E. 2001. A manual for wildlife radio tagging. Academic Press, San Diego, GA U.S.A. Madrid, J.A., H.D. Madrid, S.H. Funes,J. Lopez, R. Bot- ZOC, AND A. Ramos. 1991. Reproductive biology and behavior of the Ornate Hawk-Eagle {Spizaetus ornatus) in Tikal National Park. Pages 93-113 mD.F. Whitacre, W.A. Burnham, and J.P. Jenny [Eds.], Progress report of the Maya project: use of raptors and other fauna as environmental indicators for design and manage- ment of protected areas and for building local capac- ity for conservation in Latin America. The Peregrine Fund, Boise, ID U.S.A. Meyburg, B.-U., S. van Balen, J.-M. Thiollay, and R.D. Chanceli.or. 1989. Observations on the endangered Java Hawk-Eagle Spizaetus bartelsi. Pages 279-299 in B.-U. Meyburg and R.D. Chancellor [Eds.], Raptors in the modern world. World Working Group on Birds of Prey and Owls, Berlin, Germany. Nijman, V. 2004. Habitat segregation in two congeneric hawk-eagles {Spizaetus bartelsi and S. cirrhatus) in Java, Indonesia./. Trap. Ecol. 20:105-111. , S. van Balen, and R. Sozer. 2000. Breeding bi- ology of Javan Hawk-Eagle Spizaetus bartelsi in west Java, Indonesia. Emu 100:125-132. AND . 2003. Wandering stars: age-related habitat use and dispersal of Javan Hawk-Eagles {Spt- zaetus bartelsi). J. Ornithol. 144:451-458. Prawiradilaga, D.M., N. R0v, J.O. Gjershaug, Hapsoro, AND A. SUPRIATNA. 2000. Feeding ecology of Javan Hawk-Eagle Spizaetus bartelsi during the nestling peri- od. Pages 559-563 in R.D. Chancellor and B.-U. Mey- burg [Eds.], Raptors at risk. World Working Group on Birds of Prey and Owls, Berlin, Germany. Preleuthner, M. and a. Gamauf. 1998. A possible new subspecies of the Philippine Hawk-Eagle {Spizaetus philippensis) and its future prospects./. Raptor Res. 32' 126-135. RePPProT. 1990. The land resources of Indonesia: a na- tional overview: regional physical planning pro- gramme for transmigration. Directorate General of Settlement Preparation, Ministry of Transmigration and Land resources, London Natural Resources Insti- tute, and Overseas Development Administration, Ja- karta, Indonesia. R0V, N.,J.O. Gjershaug, and Hapsoro. 2000. Abundance of territorial rainforest eagles in the Halimun Moun- tain, west Java. Pages 111-116 in H. Ichinose, T. In- oue, and T. Yamazaki [Eds.], Proceedings of the first symposium on raptors of south-east Asia. Asian Raptor Research and Conservation, Biwa, Japan. Shultz, S. 2002. Population density, breeding chronolo- gy, and diet of Crowned Eagles Stephanoaetus coronatus in Tai National Park, Ivory Coast. Ibis 144:135-138. Sozer, R. and V. Nijman. 1995. Behavioral ecology, dis- tribution, and conservation of the Javan Hawk-Eagle Spizaetus Stresemann, 1924. Versl. Tech. Gegevens 62:1-122. , , I. Setiawan, S. van Balen, D.M. Prawir- adilaga, AND J. SUBIJANTO. 1998. Javan Hawk-Eagle re- covery plan. BirdLife International, Bogor, Indonesia. Thiollay, J.-M. and B.-U. Meyburg. 1988. Forest frag- mentation and conservation of raptors: survey on the island of Java. Biol. Conserv. 44:229—250. van Balen, S. 1999. Birds on fragmented islands: persis- December 2004 Home-range of Javan Hawk-Eagle 349 tence in the forests of Java and Bali. Ph.D. disserta- tion, Wageningen Univ., Wageningen, Netherlands. AND B.-U. Meyburg. 1994. The Java Hawk-Eagle Spizaetus bartelsi: results of recent research on distri- bution, status and ecology. Pages 89-92 in B.-U. Mey- burg and R.D. Chancellor [Eds.], Raptor conserva- tion today. World Working Group on Birds of Prey and Owls, Berlin, Germany. , V. Nijman, and H.T. Prins. 2000. The Javan Hawk-Eagle: misconceptions about rareness and threat. Biol. Conserv. 96:297-304. , , AND R. SozER. 1999. Distribution and con- servation of the Javan Hawk-Eagle Spizaetus bartelsi. Bird Conserv. Int. 9:333-349. , , AND . 2001. Conservation of the endemic Javan Hawk-Eagle Spizaetus bartelsi Strese- mann, 1924: density, age-structure and population numbers. Contrib. Zool. 70:161-173. Whitten, T, R.E. Soeriaatmadja, and S.A. Afiff. 1996 The ecology of Java and Bali: the ecology of Indonesia series, Vol. 2. Periplus Editions, Singapore, Singapore. WORTON, B.J. 1989. Kernel methods for estimating the utilization distribution in home range studies. Ecology 70:164-168. Yamazaki, T 2000. Ecological research and its relation- ship to the conservation programme of the Golden Eagle and the Japanese Mountain Hawk-Eagle. Pages 415—422 in R.D. Chancellor and B.-U. Meyburg [Eds.], Raptors at risk. World Working Group on Birds of Prey and Owls, Berlin, Germany. Received 15 July 2002; accepted 10 July 2004 J Raptor Res. 38(4):350-356 © 2004 The Raptor Research Foundation, Inc. POPULATION STATUS AND REPRODUCTIVE PERFORMANCE OF EURASIAN GRIFFONS {GYPS FULVUS) IN EASTERN SPAIN Pascual L6PEZ-L6PEZ^ Clara Garcia-Ripolles, and Jose Verdejo Departamento de Ecologia, Universidad de Valencia, Dr. Moliner 50, 46100 Burjassot, Spain Abstract. — The Eurasian Griffon ( Gyps fulvus) has experienced a measurable increase in population numbers in the Iberian Peninsula and, particularly, in the Castellon province during the last two de- cades. In Castellon, we have located 18 breeding colonies on a 6670 km^ area in 2002. These included two new nesting colonies, expanding the known distribution in the province by 64 km^. The breeding success was 0.83 chicks/laying pair, productivity was 0.66 chicks per detected pair, and the percentage of pairs initiating breeding activities was 79.5%. Two variables were included in the logistic model that best explained the probability of raising a chick successfully (type of nest and its interaction with ori- entation) . The probability of nest success increased with nests located inside caves and with nests located on open ledges that were oriented to the south. There was a significant relationship between the dis- tance to the nearest-neighbor nest with the nest success (0 or 1) in the logistic-regression analysis. We suggest that conservation of native fauna is necessary to maintain the griffon population in the province. The reduction and closure of vulture restaurants in existing range may have also stimulated southward dispersal of griffons searching for new trophic resources. Key Words: Eurasian Griffon', Gyps fulvus; breeding success; coloniality; productivity; vulture restaurants', Spain. ESTATUS POBLACIONAL Y PARAMETROS REPRODUCTORES DEL BUITRE LEONADO GYPS FULVUS EN EL ESTE DE ESPANA Resumen. — Durante las ultimas dos decadas, el buitre leonado Gyps fulvus ha experimentado un consid- erable aumento de su tamano poblacional en la Peninsula Iberica, especialmente en la provincia de Castellon. En esta provincia hemos localizado un total de 18 colonias de reproductivas en un area de 6670 km^ en 2002. Dos nuevas colonias de cria no se habian citado previamente, y suponen una ex- pansion de 64 km^ de la distribucion conocida de la especie en la provincia. El exito reproductivo fue de 0.83 pollos/huevo, la productividad de 0.66 pollos/pareja y el 79.5% de las parejas iniciaron activi- dades de reproduccion. El modelo logistico que mejor explico la probabilidad de criar un polio con exito incluyo dos variables significativas: el tipo de nido y su interaccion con la orientacion. Los nidos emplazados en cuevas, asi como los ubicados en repisas abiertas orientadas hacia el sur, mostraron mayor probabilidad de exito. Se encontro una relacion significativa en cl analisis de regresion logistica entre la distancia al nido vecino mas proximo y el exito de cada nido (0 6 1). Sugerimos que la conservacion de la fauna silvestre es necesaria para mantener las poblaciones de G. fulvus en Castellon. La reduccion y cierre de muladares podria haber estimulado la dispersion de los buitres hacia el sur en busca de nuevos recursos troficos. [Traduccion de los autores] The large growth in the Eurasian Griffon {Gyps fulvus) population in the Iberian Peninsula has been well documented since the 1970s (Errando et al. 1981, Donazar 1987, Arroyo et al. 1990, Don- azar and Fernandez 1990). The recent national censuses showed increases in breeding pairs of 135% between 1979-89 and 130% from 1989-99 (Del Moral and Marti 2001). In the Castellon prov- ince, this growth has been constant and greater ^ E-mail address: Pascual.lopez@uv.es (>150%) than the national mean (Del Moral and Marti 2001). In spite of this growth, expansion of griffon distribution both on a national and a local level has not occurred (Arroyo et al. 1990, Donazar 1993, Del Moral and Marti 2001). Conspecific at- traction to nest places (Sarrazin et al. 1995) seems to be among the causative factors for this pattern. In this paper, we present results related to the population status, reproductive performance (breeding success and productivity) , and range ex- pansion for the Eurasian Griffon in the Castellon 350 December 2004 Status of Eurasian Griffons in Eastern Spain 351 Figure 1. Left: Iberian Peninsula. The shaded area shows the study area, Castellon province. Right: study area general map with Universal Transverse Mercator 10 X 10 km squares. Shaded area squares show frequent Eurasian Griffon presence. Circles indicate the two new breeding colonies. province. Age of breeding pairs, type of nest, nest orientations, and distance to nearest conspecific nest were recorded and statistically related to nest success. We also provide an analysis of the physical variables recorded at griffon colonies and their re- lationship with reproductive performance. Study Area The study area comprises the Castellon province (Fig. 1; located in the east of the Iberian Peninsula), including 6670 km2; 40°47'N, 39“42'S, 0“5UW, 0°32'E; 0-1814 m above sea level (masl). The area is geomorphologically characterized as the confluence of two mountain ranges: the Iberian system, oriented northwest-southeast on the one hand, and the east-northeast-oriented structures of the Catalanides, parallel to the coastline. This results in a subtabular and a folded-peak line with calcareous ma- terial, mostly sedimentary, which supports many cliffs and walls suitable for the nesting of the griffon. Climatologi- cally, it belongs to the Mediterranean area, with an an- nual mean temperature varying between I7°C in the coast area and 8°C in the inner highlands (where the entire griffon population is located). The annual mean precipitation varies from 400-900 mm, with maximum values during the fall and minimum values in the sum- mer, characterized by the great interannual irregularity of the Mediterranean weather (Quereda et al. 1999). Bio- climatologically, the study area supports an assortment of vegetation types and ecosystems (Rivas-Martinez 1987). This heterogeneity also manifests itself locally, alternating cultivation zones both irrigated and nonirrigated with forest patches dominated by pines {Pinus spp.) and, to a lesser extent, oaks {Quercus spp.) and Juniperus spp. The livestock industry in the study area is mainly re- stricted to intensive feed-lot farms, in which animals are confined and not available to vultures (CAPA 1999). Pas- ture grazing does occur in the northern regions, the only area where a griffon population has existed for the last 30 yr (Arroyo et al. 1990, Urios et al. 1991, Del Moral and Marti 2001). The livestock is mainly porcine (92.44%), followed in importance by ovine (3.96%), and bovine (2.35%). The availability and use of wild ungu- lates as a trophic resource for griffons is unknown, Also in the study area, a small number of “vulture restau- rants” are scattered in the central part of the province which are available to griffons. “Vulture restaurants” are traditional places close to villages where shepherds and farmers drop carcasses and serve as supplementary feed- ing sources for carrion-eating birds. Methods We monitored the reproductive success of 112 breed- ing pairs from December 2001-June 2002. All areas where breeding by griffons was known were observed, as were cliffs larger than 40 m in height with suitable ledges that could be colonized by nesting vultures (Donazar 1993). For the latter, we monitored 85% of the potential nesting cliffs, thus may have missed an isolated-reproduc- tive pair. The coastal area, which has high human density and is less suitable for nesting vultures, was surveyed less intensively; 60% of the potential nesting cliffs surveyed. In the interior areas, we monitored 98% of the potential nesting cliffs. Observations were made with a 20-60 X telescope dur- ing clear days and >300 m from nesting cliffs to avoid disturbance to vultures (Fernandez et al, 1996, Olea et al. 1999, Gil-Sanchez 2000). At least three visits, in some cases five, were made to every reproductive colony. A pre- liminary search was made between 20 December and 10 January, in which nuptial flights and copulations were observed (Donazar 1993). The first visit was made be- tween 16 February and 23 February, in which a sketch of the cliff that hosts the colonies was made, noting the lo- cations of pairs and nests. The second visit was made be- tween 21 March and 3 April to confirm the presence/ 352 Lopez-Lopez et al. VoL. 38, No. 4 Table 1. Physical variables recorded at Griffon Vulture colonies {N = 16 colonies). Name Description Cliff height Cliff height (m); 1:10 000 digital map. Distance to top of cliff Distance (ra) from the highest nest to the top of cliff; digitally treated im- ages based on known-length segments previously measured on the field. Recurring measures were only counted once (Olea et al, 1999). Distance to base of cliff Distance (m) from the lowest nest to the base of cliff; digitally treated im- ages. Distance from nearest colony Distance (m) from nearest colony; 1:50 000 Spanish Army Cartographic Service. MAST Mean elevation (m) above sea level; 1:50 000 Spanish Army Cartographic Service. Distance to nearest nesV Distance (m) to nearest neighbor nest in the same colony; digitally treated images. ^ N = 82 nests were measured; we could not obtain an adequate digital picture to measure distances accurately at colony No. 13. absence of the previously-detected pairs, the existence of new nests, and the newly-hatched chicks. Our third visit took place between 19 April-11 May to monitor the de- velopment of previously-detected chicks and the pres- ence of new hatchings. Finally, a fourth visit was made in the period between 21 May and 25 June, when 11 of the 18 known colonies were visited. During this last visit, breeding success was recorded, as was the presence of late broods. Those cliffs where the species was not de- tected during the first two visits were not visited subse- quently (Martinez et al. 1997, Del Moral and Marti 2001) . A pair was considered as a laying pair if it was building a nest, incubating, and the griffons were taking turns in the nest, or if typical pair behavior, such as close contact with mutual preening, was observed (Donazar and Fer- nandez 1990, Blanco and Martinez 1996, Olea et al. 1999). A cliff was considered as a colony if it was occu- pied by at least two pairs and was at least 1000 m away from its closest neighbor, according to the methodology used in the species’ Spanish National Census (Del Moral and Marti 2001). These criteria were modified for two of the colonies because of the more rugged orogra- phy of the terrain and its different abiotic characteristics; at these sites we reduced the nearest-neighbor criterion to 600 m. Information gathered for each nest included Its orientation and type of location (open ledge, shel- tered ledge, or cave). Orientation was measured on the Valencian Cartographic Institute computer-cartographic database to a 1:10 000 scale. For each pair, age was re- corded according to plumage (subadult or adult) . Those individuals with brown ruff and non-nacreous bill were considered subadults (Donazar 1993, Blanco and Marti- nez 1996). The following reproductive laying-pair com- binations were found: adult-adult, adult-subadult, and subadult-subadult pairs. Selected physical parameters of cliffs were also derived from the cartographic database (Table 1). The following reproductive parameters were calculated for each breeding colony and for the entire study area: productivity = fledged chicks/detected pairs; breeding success = fledged chicks/laying pairs (Del Moral and Marti 2001). A chick was considered as fledged if, given its development level during the last visit, it was older than 70 d (Del Moral and Marti 2001). Based on the locations of nesting colonies, we calculated a minimum convex polygon to estimate the area in the province oc- cupied by breeding Griffon Vultures (Olea et al. 1999) Statistical Analysis. Descriptive statistics were calculated for the most variables: productivity, breeding success, cliff height, distance to base of cliff, distance to top of cliff, distance from nearest colony, and mask We used linear regressions (Sokal and Rohlf 1981, Draper and Smith 1998) with productivity and breeding success of each col- ony as dependant variables to examine the influence of measured variables on reproductive performance. Final- ly, in order to assess the relationships between a depen- dant (response) factor (each nest’s success: 0 or 1) against the independent (explanatory) categorical pre- dictors: orientation, type of pair, type of nest, and dis- tance to nearest nest, a logistic regression analysis was performed (Sokal and Rohlf 1981, Fernandez et al. 1996, Everitt and Dunn 2001) . To employ this method, we used the link function in the generalized-linear models pro- cedure of SPSS (1998), where the predicted variable was logistically transformed (Agresti 1990, Everitt and Dunn 2001). We assumed a binomial distribution of errors and each nest was considered as one case. Standard-stepwise backward procedure was used, including all variables and then removing not significant variables by Wald's method (Johnson 1998, McNally 2000) . If the Wald statistic was significant then the parameter was included in the mod- el. We selected the last significant model that included the fewest variables. All calculations were made using the SPSS vll.5 (SPSS Inc. 1998), Results We located 16 colonies and two isolated nests in the study area, with 89 laying pairs and 112 de- tected pairs. Productivity was 0.66 chicks per de- tected pair {N = 112) and breeding success was 0.83 chicks per laying pair {N — 89). The mean number of visits per colony was 3.22 (SD = 0.88, December 2004 Status oe Eurasian Grieeons in Eastern Spain 353 Table 2. Generalized linear models for nest success (0 or 1) of Eurasian Griffon, using binomial error and logistic links. Significant model (x^ = 25.33, df = 14, P = 0.03, IP = 0.25) involved type of nest and its interaction with orientation. 6 Standard Error Wald P All nests 7.37 0.03 Sheltered ledge -2.37 1.02 5.43 0.02 Open ledge -1.96 0.88 4.93 0.03 Nest by orientation 7.95 0.79 Sheltered ledge toward north 21.20 17 974.84 0.00 0.99 Sheltered ledge toward northeast 21.20 17 974.84 0.00 0.99 Sheltered ledge toward east 1.61 1.37 1.39 0.24 Sheltered ledge toward southeast 21.20 16 408.71 0.00 0.99 Sheltered ledge toward south 21.20 12118.64 0.00 0.99 Sheltered ledge toward southwest 21.20 23 205.42 0.00 0.99 Sheltered ledge toward west 21.20 16 408.71 0.00 0.99 Open ledge toward northeast 20.80 15191.52 0.00 0.99 Open ledge toward east 20.80 20 096.49 0.00 0.99 Open ledge toward south 2.13 0.88 5.86 0.01 Open ledge toward southwest 1.32 0.81 2.66 0.10 Open ledge toward west 0.51 1.06 0.23 0.63 Constant 2.37 0.60 15.37 <0.01 Residual deviance 105.57 N = 58) . The percentage of pairs initiating repro- duction was '79.46%. The number of pairs per col- ony ranged from 2-18 (x = 6.88, SD = 5.45, N = 16). All the colonies were located on cliffs over 40 m high (x = 131.67 m, SD = 74.06, range = 40-320 m), between 645 and 1150 masl (x = 850.56, SD = 139.80). The mean distance from nearest colony was 5523.72 m (SD - 7743.66, range = 325-24 000 m). The mean distance from the highest nest to the top of cliff 32.45 m (SD = 31.11, range = 10- 120 m), and the distance from the lowest nest to the base of the cliff was 43.45 m (SD = 24.91, range - 10-80 m). The location of the nests was as follows: open ledge, 55 nests; sheltered ledge, 26 nests; cave, 19 nests {N = 100). The nest orientations were mostly south (33.33%, N = 33) and southwest (25.25%, N — 25). Breeding pairs were comprised of: adult- adult (24.11%, N— 27), at least one adult (28.57%, N = 32), adult-subadult (8.93%, N = 10), at least one subadult (4.46%, N = 5), subadult-subadult (1.79%, iV = 2), and unknown age (32.14%, N = 36) . The mean distance to the nearest nest inside the colony was 72.93 m (SD = 99.54, range = 4— 498 m). Our results indicated that the nesting distribu- tion of griffons in Castellon province has expand- ed in recent years. The species now occupies two new nesting colonies located 41 and 64 km south- west from the previously-reported distribution in 1999 (Del Moral and Marti 2001), which implies an increase of 64 km^ (Fig- 1)- No statistical relationship was found between the cliff variables (Table 1) and productivity or breed- ing success for each colony. There was a significant relationship between the distance to nearest-neigh- bor nest and nest success in the logistic regression analysis {P = 0.014) . Using nest-site characteristics, two significant var- iables were involved in the model that best ex- plained the probability of breeding successfully: type of nest and its interaction with orientation (Table 2). The probability of nest success de- creased with nests located on open ledges and on sheltered ledges, and increases in success with nests inside caves. Also those nests located on open ledges and oriented to the south had a greater probability of producing a chick. Discussion The number of nesting colonies in the Castellon province has increased from 14 in 1999 (Del Moral and Marti 2001) to 18 in 2002. However, the num- 354 Lopez-Lopez et al. VoL. 38, No. 4 her of isolated pairs decreased from five to two. Productivity (0.66 chicks/pair) and breeding suc- cess (0.83 chicks/laying pair) were lower and high- er, respectively, than that observed in 1999 (pro- ductivity = 0.80, breeding success = 0.81). These small differences could be due to a more thorough monitoring of the population in 2002, with a mean of 3.6 visits per colony compared to 2.6 visits made during the last census. More complete monitoring may have allowed for the detection of a greater number of nest failures (Martinez et al. 1997). We could not make direct comparisons of repro- ductive parameters among regions because of dif- ferent methodologies (Donazar et al. 1988, Arroyo et al. 1990, Leconte and Som 1996, Martinez et al. 1997, Olea et al. 1999). However, the breeding suc- cess found in the Castellon province (0.83 chicks/ laying pair) was similar to that estimated for the Cantabrian Mountains (0.84; Arroyo et al. 1990) and for the Navarre pre-Pyrenees (0.86; Donazar et al. 1988). Also, the breeding success estimated during this study was higher than the breeding suc- cess found in the Spanish Pyrenees (0.77; Arroyo et al. 1990), French Pyrenees (0.76; Leconte and Som 1996), and the Hoces del Duraton (0.63-0.68; Palacin et al. 1993). The productivity in the study area (0.66 chicks/detected pair) was close to that reported for other Spanish territories (e.g., Extre- madura — 0.67, Murcia = 0.64, Euskadi = 0.67) and was also near the national mean (0.69; Del Moral and Marti 2001). The expansion of the species’ distribution to- ward the southwest of the previously known range m the province (Del Moral and Marti 2001) was documented during this study. The two new colo- nies were probably established recently, and we found the southernmost one in 2002. This area was surveyed exhaustively for the Peregrine Falcons (Falco peregrinus) over the past 20 yr (Verdejo 1991, 1994, Gil-Delgado et al. 1995) and no griffon col- onies were detected previously. These two new col- onies, along with the previously-documented colo- nies, lie geographically in the western and northern portions of the province (Fig. 1). This distribution corresponds with lower human-density zones where the agrarian and livestock-related ac- tivity is still present (CAPA 1999, Garcia-Rippoles et al. 2004) . We suggest that this southward expan- sion may aid in the species’ recovery and compen- sate for those locations where the species was ex- tirpated, such as the northern portion of Valencia province (Urios et al. 1991). It is notable that 91.66% of the reproductive individuals in the two new colonies are adults {N =11 pairs). Only one breeding pair formed by subadult-plumage individ- uals was recorded and they failed in their nesting attempt. This proportion matches that recorded in a northern region of Spain in a 1997 (Olea et al. 1999), where 92% were adults {N = 38). The percentage of adults composing breeding pairs (61.61%, N — 69) in the study area was lower than that of other Spanish peninsular territories such as the West Pyrenees (75%), the Castilla Mes- eta (72.2%), Extremadura (72.7%), and Cadiz (73.1%; Blanco and Transverso 1996). We also not- ed the low proportion of two-subadult breeding pairs (1.79%) was only comparable to Castilla Mes- eta (3.3%; Blanco and Transverso 1996). Our results from the logistic-regression analysis showed that the probability of raising a chick was related to the type of nest and its orientation. Age of breeding pairs was not related to the probability of nest success. To a lesser extent, the logistic mod- el indicated a positive relationship between south- oriented nests located on open ledge and the probability of success. In our study area, this may have resulted in a higher probability of success be- cause these sites were less exposed to the Mediter- ranean winter and thus, inclement weather. For the same reason, as was suggested by Elosegui (1989), nests located in caves probably had a higher prob- ability of supporting a successful nest. Neverthe- less, Donazar (1993) argued that there was gener- ally no preference for a specific nest orientation by griffons, and that patterns observed were likely dic- tated by mountain range orientation. The Eurasian Griffon is mostly a colonial nesting bird (Cramp and Simmons 1980, Donazar 1993). This likely is related to the spatial distribution of the species’ food resources (Horn 1968, Donazar 1993). The unpredictable nature of the availability of carrion has been proposed as the main factor favoring this behavior (Donazar 1993). In our study, this was supported by the negative relation- ship observed between the distance to the nearest- neighbor nest in the same colony and its success. We suggest that the carrion availability in our study area was unpredictable, and that the number of permanent functioning “vulture restaurants’’ were few. This unpredictability of food resources may ex- plain why these colonies exhibiting greater nest ag- gregation have the highest reproductive perfor- mance in our province. We suggest that the colonies might act as effective information centers December 2004 Status of Eurasian Griffons in Eastern Spain 355 (Ward and Zahavi 1973). In addition, the colonial habit could also reduce the predatory risk by Com- mon Ravens (Corvus corax; Elosegui 1989). Eactors like the increasing of livestock-trophic re- sources, the recovery of wild fauna, the creation of vulture restaurants, and the lessening of direct and indirect prosecution of the species by humans in recent years (Donazar 1993), probably has fostered the species’ range expansion. Moreover, due to “mad-cow” disease in Europe in 1999 (Garcia-Rip- poles et al. 2004), the National Spanish Govern- ment prohibited the ad hoc dumping of carcasses in an attempt to prevent any possible effects on scavengers. New national laws concerning the es- tablishment of “vulture restaurants,” and succes- sive cases of livestock diseases, might have elimi- nated the access to previously available food resources to carrion-eating birds. This reduction in food availability may have helped to stimulate southward range expansion by Eurasian Griffons in the Castellon province. Acknowledgments We thank Francisco Garcia and Jose Miguel Aguilar, who helped us to find colonies and gave valuable sug- gestions on the original manuscript. Without their highly regarded help, the writing of this paper would not have been possible. Ramon Prades and some members of the Forest Guard helped locate some colonies. J.C. Bednarz, C. Griffith, R. Yosef, and two anonymous referees re- viewed the manuscript and provided valuable sugges- tions. P. Montes and G. Ayala provided statistical assis- tance. Finally, Javier Garcia gave us logistic support and showed a stoic patience during the never-ending field days. Manuel Viciano helped the authors translate this paper. Our many thanks to all of these contributors. Literature Cited Agresti, a. 1990. Categorical data analysis. Wiley Press, New York, NY U.S.A. Arroyo, B., E. Ferreiro, and V. Garza [Eds.]. 1990. II Censo Nacional de buitre leonado (Gyps fulvus): pob- lacion, distribucion, demografia y conservacion. Na- tional Institute for Conservation Nature, Madrid, Spain. Bianco, G. and F. Martinez. 1996. Sex difference in breeding age of Griffon Vulture (Gyps fulvus). Auk 113:247-248. and J.M. Transverso. 1996. Pair bond and age distribution of breeding Griffon Vultures Gyps fulvus in relation to reproductive status and geographic area in Spain. Ibis 139:180-183. CAPA. 1999. Agrarian production data, 1999. Generalitat Valenciana, http://www.capa.gva.es Cramp, S. and K.E.L. Simmons. 1980. Handbook of the birds of the western palearctic. Vol. 2. Oxford Univ Press, Oxford, U.K. Del Moral, J.C. and R. Marti [Eds.]. 2001. El buitre leonado en la Peninsula Iberica. BirdLife Internation- al, Madrid, Spain. Donazar, J. A. 1987. Apparent increase in vulture (Gyps fulvus) population in Spain./! Raptor Res. 21:75-77. , J. Elosegui, and A. Senosiain. 1988. Exito re- productor del buitre leonado (Gyps fulvus) en Navar- ra. Donana Acta Vertebr. 15:187-192. AND C. Fernandez. 1990. Population trends of the Griffon Vulture Gyps fulvus in northern Spain be- tween 1969 and 1989 in relation to conservation mea- sures. Biol. Gonserv. 53:83—91. . 1993. Los buitres ibericos biologia y conserva- cion. Pages 25-37 m J.M. Reyero [Ed.], Los buitres ibericos biologia y conservacion. Madrid, Spain. Draper, N.R. and H. Smith. 1998. Applied regression analysis, 3’’'^ Ed. Wiley Interscience Press, Quebec, Canada. Elosegui, I. 1989. Vautour fauve (Gyps fulvus) , gypaete barbu (Gypaetus barbatus), percnoptere d'egypte (Neo- phron percnopterus): synthese bibliographique et re- cherches. Acta Biol. Mont. 3:1-278. Errando, E., J.A. Gil-Delgado, and B. Miguel. 1981 Primer censo de buitreras (1979), informe sobre ah- cante, Castellon y Valencia. Ardeola 26-27:165-312. Everitt, B.S. and G. Dunn. 2001. Applied multivariate data analysis. Oxford Univ. Press Inc., New York, NY U.S.A. Fernandez, C., P. Azkona, and J.A. Donazar. 1996. Den- sity-dependent effects on productivity in the Griffon Vulture Gyps fulvus-. the role of interference and hab- itat heterogeneity. Ibis 140:64-69. Garcia-Rippoles, C., P. Lopez-Lopez, and F. Garcia- Lopez. 2004. Management and monitoring of a vul- ture restaurant in Castellon province, Spain. Vulture News 50:5-14. Gil-Delgado, J.A., J. Verdejo, and E. Barba. 1995. Nest- ling diet and fledgling production of Eurasian Kes- trels (Falco tinnunculus) in eastern Spain./. Raptor Res 29:240-244. GlI -SAnchez, J.M. 2000. Efectos de la altitud y de la dis- ponibilidad de presas en la fecha de puesta del aguila azor-perdicera (Hieraaetus fasciatus) en la provincia de Granada. Ardeola 47:1-8. Horn, H.S. 1968. The adaptive significance of colonial nesting in the Brewer's Blackbird (Euphagus cyanoceph- alus). Ecology 49:682—694. Johnson, D.E. 1998. Applied multivariate methods for data analysis. Brooks Cole Publishing Co., Florence, KS U.S.A. Leconte, M. and J. Som. 1996. La repoduction du Vau- tour fauve Gyps fulvus dans les pyrenees occidentales. historique d’une restauration d’effectifs et parametres reproducteurs. Alauda 64:135-137. McNally, R. 2000. Regression and model building in 356 Lopez-Lopez et al. VoL. 38, No. 4 conservation biology, biogeography, and ecology: the distribution between and reconciliation of “predic- tive” and “explanatory” models. Biodiver. Conserv. 105: 395-403. Martinez, R, R.F. Rodriguez, and G. Blanco. 1997. Ef- fects of monitoring frequency on estimates of abun- dance, age distribution, and productivity of colonial Griffon Vultures. J. Field Ornithol. 68: 392-399. Olea, RR, J. Garcia, and J. FalagAn. 1999. Expansion del buitre leonado Gyps fulvus: tamano de la pobla- cion y parametros reproductores en un area de re- ciente colonizacion. Ardeola 46:81-88. Palagin, G., j. Marchamalo, and M.A. Carro. 1993. Pob- lacion y tasas reproductoras del buitre leonado {Gyps fulvus) en la cuenca del no Duraton. Alytes 6:99-103. Quereda, j., E. Monton, and J. Escrig. 1999. El clima de la provincia de Castellon. Pages 51-60 in M.J. Gi- meno [Ed.] , La provincia de Castellon, Diputacion de Castellon, Castellon, Spain. Rivas-Martinez, S. 1987. Memoria del mapa de series de vegetacion de Espaha. National Institute for Conser- vation Nature, Madrid, Spain. Sarrazin, E, C. Bargnolini, J.L. Pinna, and E. Danchin. 1995. Breeding biology during establishment of a re- introduced Griffon Vulture Gyps fulvus population. Ibis 138:315-325. SOKAL, R.R. AND FJ- Rohlf. 1981. Biometry. W.H. Free- man and Company, New York, NY U.S.A. SPSS, Inc. 1998. SPSS for Windows version 10.1. SPSS Inc., Chicago, IL U.S.A. Urios, V, J.V. Escobar, R. Pardo, andJ.A. Gomez. 1991. Atlas de las aves nidificantes de la Comunidad Val- enciana. Generalitat Valenciana, Valencia, Spain. Verdejo, J. 1991. Las aves de presa diurnas y nidificantes en el Alto Palancia. Ph.D. dissertation, Univ. Valencia, Valencia, Spain. . 1994. Datos sobre la reproduccion y alimenta- cion del azor {Accipiter gentilis) en un area mediter- ranea. Ardeola 41:37-43. Ward, R and A. Zahavi. 1973. The importance of certain assemblages of birds as “information centers” for food finding. Ibis 115:517-534. Received 30 January 2003; accepted 17 August 2004 Short Communications J Raptor Res. 38(4);357-361 © 2004 The Raptor Research Foundation, Inc. Gender Determination in the Swainson’s Hawk (Buteo swainsoni) Using Molecular Procedures and Discriminant Function Analysis Jose HernAn Sarasola^ and Juan Jose Negro Department of Applied Biology, Estacion Biolo^ca de Donana (CSIC), Avda. de Maria Luisa s/n, Pabelldn del Peru, 41013 Sevilla, Spain Key Words: Swainson ’s Hawk, Buteo swainsoni; gender de- termination', molecular sexing, morphometric measures', winter- ing grounds. Gender identification of individuals is important in many studies of wild animals. However, easy determina- tion of sex is difficult for monomorphic species including most birds of prey. Many raptor species show littie plum- age dimorphism, and although females are generally larger than males, overlap in morphometric measure- ments and body mass make gender determination diffi- cult even when birds are captured and handled. The Swainson’s Hawk {Buteo swainsoni) breeds in North America and migrates to southern South America during the boreal winter that involves a trip of ca. 10 000 km each way (second longest migration distance among raptors; England et al. 1997, Fuller et al. 1998). Male and female Swainson’s Hawks are similar in plumage when adults (Wheeler and Clark 1995, England et al. 1997), and as in most of the genus Buteo, plumage polymor- phism occurs both in immature and adult birds. Al- though accurate methods for gender determination us- ing morphometric data have been developed for several raptor species (Bortollotti 1984a, 1984b, Garcelon et al. 1985, Edwards and Kochert 1986, Ferrer and De Le Court 1992, Balbontm et al. 2001, Palma et al. 2001), no reliable criteria for gender determination of Swainson’s Hawks using external characteristics have been de- scribed. During the last decade, the development of laboratory techniques involving molecular procedures has provided reliable methods for the accurate gender determination of the majority of avian species (Ellegren and Sheldon 1997). PCR-based methods targeting CHDl-Z and CHDl- W genes are purported to be of universal application to birds, with the exception of ratite species (Ellegren 1996, Fridolfsson and Ellegren 1999). The aim of this study was to develop an accurate method for gender determination of Swainson’s Hawks using molecular procedures and ^ E-mail address: sarasola@ebd.csic.es morphometric criteria. Our goal was to obtain a general model, derived from discriminant analysis, to determine gender of immature and adult Swainson’s Hawks. Methods We captured and sampled free-living Swainson’s Hawks during two wintering seasons (austral summers) in three study areas located in central Argentina. We captured 34 hawks in the vicinity of a roost site near Las Varillas, Cor- doba province (31°58'S, 62°50'W), from 19-26 January 2003. One hawk was captured in northern La Pampa province (35°14'S, 63°57'W) on 21 November 2002, and 34 hawks from 7-10 December 2003 at the same site. The sample was completed with 35 hawks captured near Santa Rosa (36°33'S, 64°07'W), La Pampa province, from 21- 29 January 2004. The habitat where trapping was con- ducted consisted of agricultural fields of continuous crops, with soybeans as the principal crop. Planted pas- tures and natural fields comprised the remaining habitat Hawks were captured in open fields near the roost us- ing bal-chatri traps (Berger and Mueller 1959) in early morning and during the afternoon. Traps were set m front of fence posts usually used by hawks for perching, both when they left roosts in the morning and during late afternoon before returning. Captured hawks were aged as juveniles or adults based on plumage character- istics (Wheeler and Clark 1995), with immature birds grouped with juveniles using the same criteria employed by Goldstein et al. (1999). Hawks were banded and weighed with a 1500 g Pesola scale (Pesola AG, Baar, Swit- zerland) to the nearest 2 g. Six morphometric measure- ments were taken from adults and juveniles. We mea- sured the length of wing chord (WING) and tail (TAIL) using a plastic rule to the nearest 1 mm, and length of the exposed culmen (CULMEN), tarsus (TARSUS), and hallux claw (FIALLUX) using a caliper to the nearest 0.05 mm. We also measured the forearm length (FORE- ARM), or the length from the front of the folded wrist to the proximal extremity of the ulna, also using a caliper (Ferrer and De Le Court 1992, Balbontm et al. 2001; Fig 1). For a few birds only some of the body measurements were recorded (Tables 1, 2). Approximately 2 ml of blood were taken from each bird from the brachial vein. The blood was placed in tubes with 96% ethanol that were kept in coolers until analysis in the laboratory. The cellular fraction of the 357 358 Short Communications VoL. 38, No. 4 Figure 1, Measurement of forearm length in Swainson’s Hawk. blood sample was used to determine gender for all hawks. For this analysis, we used primers 2550F (5'- GTTACTGATTCGTCTACGAGA-3') and 2718R (5'-ATT- GAAATGATCCAGTGCTTG-3') to amplify the W chro- mosome gene following Fridolfsson and Ellegren (1999; Fig. 2). We performed multivariate analysis of variance (MAN- OVA) followed by univariate analysis of variance (ANO- VA) to check for differences in morphometric measures among age and sex classes (Zar 1996). Sexes were dis- criminated with a stepwise discriminant analysis. We used a cross-validation (also called Jacknife) procedure to as- sess the predictive power of the discriminant functions, in which each individual was classified using a function derived from the total sample less the individual being classified (Manly 1986). Cohen’s Kappa statistic was also calculated and significance tests were performed for each of the resulting discriminant functions. This statistic es- timates the correct classification rate adjusted by chance, considering also the effect of unequal group sizes in the probability of correct classification (Titus and Mosher 1984). Resuits The total sample analyzed included 66 males (32 ju- veniles and 34 adults) and 38 females (17 juveniles and 21 adults) . Juvenile and adult Swainson’s Hawks differed significantly in overall size (MANOVA, F = 3.21, df = 7, 93, P < 0.01). An ANOVA for each morphometric vari- able indicated that the difference was primarily due to body mass, culmen, and hallux length differences be- tween adult and immature hawks, while there were no significant differences in the remaining measurements (Table 1). MANOVA test also showed differences be- tween males and females {F = 35.8, df = 7, 93, P < 0.001). Univariate analysis of variances showed males be- ing significantly smaller than females for all measures (Table 2). The ranges for the six variables were overlap- ping between gender groups in all cases. Due to age-related differences in some of the morpho- metric measures, and in order to obtain a general meth- od to individualize male and female hawks independently of age, we excluded body mass, culmen, and hallux length from the discriminant analysis and considered only those morphometric measures that did not differ between age groups. Discriminant function analysis using single measurements showed that most of the variables were good predictors of gender (Table 3), but every var- iable considered separately failed to classify 100% of the individuals in the sample correctly. Forearm was the best predictor variable considering the percentage of cases correctly classified and the value of Cohen’s Kappa, with a resulting standardized linear function equal to = 0.49 FOREARM — 69.85. The function assigned all but 11 individuals to the correct sex after cross-validation (four males and seven females, overall success 89.4%), where values of D > 0 identified females and values of Table 1. Morphometric measurements of juvenile (includes immatures or second year hawks) and adult Swainson’s Hawks and analysis of variance (ANOVA) results for age class differences. All measurements, except mass (g) are m mm. Juveniles Adults ANOVA X SD Range N X SD Range N F P Wing 390.0 16.0 350.0-420.0 49 395.0 16.0 370.0-430.0 55 3.21 0.07 Tail 201.0 12.0 180.0-230.0 49 202.0 11.0 180.0-220.0 55 0.17 0.68 Culmen 22.5 1.3 20.3-25.6 48 23.3 1.2 20.5-26.1 54 8.55 <0.01 Tarsus 70.8 4.5 60.3-79.4 49 70.5 3.8 64.2-80.1 54 0.08 0.77 Hallux 23.9 1.3 21.4-26.7 49 24.5 1.2 22.3-28.1 55 5.63 <0.05 Forearm 136.9 6.2 124.0-149.0 49 138.4 7.1 127.0-157.0 55 1.47 0.22 Mass 759.7 116.7 540.0-1100.0 48 824.8 110.9 580.0-1110.0 55 8.40 <0.01 December 2004 Short Communications 359 Table 2. Morphometric measurements of Swainson’s Hawks in wintering grounds and analysis of variance (ANOVA) test results for gender differences. All measurements except mass (g) are in mm. Females Males ANOVA X SD Range N X SD Range N F P Wing 409.0 11.0 390.0-430.0 38 383.0 11.0 350.0-420.0 66 127.74 <0.001 Tail 211.0 9.0 180.0-240.0 38 195.0 8.0 180.0-230.0 66 88.98 <0.001 Culmen 24.1 1.1 21.0-26.1 37 22.3 1.0 20.3-25.6 65 69.53 <0.001 Tarsus 72.6 4.0 64.6-80.1 37 69.5 3.8 60.3-78.4 66 15.68 <0.001 Hallux 25.3 1.1 22.6-28.1 38 23.7 1.0 21.4-26.4 66 55.97 <0.001 Forearm 144.2 4.8 134.5-157.0 38 134.0 4.3 124.0-144.8 66 122.68 <0.001 Mass 895.1 118.3 590.0-1110.0 38 735.6 66.8 540.0-880.0 65 76.64 <0.001 D < 0 identified males. The dividing point between gen- ders for forearm length obtained by solving for 0 was 140.2 mm, with values over this point representing fe- males and values under it representing males. Forearm, tail length, and wing chord length were retained in the stepwise discriminant analysis. The resulting linear func- tion (£>2 = 0.36 FOREARM + 1.36 TAIL + 1.04 WING — 120.95) increased our predictive power against Dj (Ta- ble 3) and a lower value of Wilk’s lambda indicated that females and males were better separated with this linear combination of variables than using only forearm. Discussion Our data indicated that there were significant differ- ences between male and female Swainson’s Hawks. How- ever, there was considerable overlap in the ranges of the morphometric measurements, suggesting that the use of relative size as the only criterion to determine gender in this raptor would result in errors. Bill depth, toe-pad length, and body mass are morpho- metric measurements frequently used in determining gender of size-dimorphic birds of prey (Bortollotti 1984a, 1984b, Garcelon et al. 1985, Edwards and Kochert 1986) Forearm length is a body feature traditionally not record- ed for wild raptors, but its use has become more common as it is a measure easily obtained from museum skins. Furthermore, it has been shown to be a low variance mea- sure with high repeatability among observers (Ferrer and De Le Court 1992). These features make forearm a re- liable morphometric measure that should be considered as a general and standard measure in morphometric gen- der determination of birds of prey. Ideally, a technique for gender determination would be applicable to individuals of all ages and under differ- ent conditions. Our model based on forearm as a single explanatory variable can be used even on birds showing incomplete molt or evidence of loss of corporal mass due to fasting. The alternative model {D^) can be also applied in cases in which loss or gain of body mass is suspected to occur (see Smith et al. 1986, Goldstein et al. 1999) The second function, which classified the highest per- centage of cases after the cross-validation and chance cor- rection tests, produced a better separation of groups (lower value of Wilk’s lambda) , but needed more vari- Figure 2. Gender determination using PCR methods. A multiple amplification with 2550F amplify a 420 bp fragment of W chromosome in females and 2550F + 2718R that amplifies 600 bp fragments in both sexes. Males and females are indicated as M and F, respectively. 360 Short Communications VoL. 38, No. 4 Table 3. Accuracy of sexing Swainson’s Hawks obtained from discriminant analysis using single measurements or combinations of morphometric variables, and assessed by cross-validation procedure and Cohen’s Kappa calculation Single variables are ordered from higher to lower values of Cohen’s Kappa. Variable Cases Correctly Separated Cohen’s Kappa P WiLK’S Lambda N Pergent Femai.es Percent Mai.es Percent Overall Forearm 0.454 (104) 81.6 93.9 89.4 0.77 <0.01 Wing 0.444 (104) 81.6 90.9 87.5 0.73 <0.01 Tail 0.534 (104) 81.6 90.9 87.5 0.73 <0.01 Tarsus 0.866 (103) 70.0 87.5 82.4 0.28 >0.05 Forearm -f Tail + Wing 0.332 (104) 86.8 97.0 93.3 0.85 <0.01 ables to be applicable. Its use would be better when com- plete data sets of the variables are available and given that measurements are not biased by external factors (e.g., molting). Furthermore, Dj discriminant function allows for gender determination of dead birds when natural de- composition and the effect of scavengers after days of exposure in the field makes classification of sex by stan- dard forensic methods impossible. Insecticide poisoning of hawks in their wintering ground during 1995-96 and 1996-97 austral summers (Woodbridge et al. 1995, Goldstein et al. 1996) has been documented in this species, with ca. 20 000 birds poi- soned in 1996-97, Gender determination of Swainson’s Hawks would provide a valuable tool for a complete as- sessment of these mortality incidents, including the gen- der of affected birds. Resumen. — Buteo swainsoni es un ave de presa poco di- morfica, y aunque las hembras suelen ser mas grandes que los machos, la determinacion del sexo en esta es- pecie puede ser dificil, aun cuando las aves son captu- radas y manipuladas. En este articulo presentamos un metodo de sexado para B. swainsoni basado en tecnicas moleculares y analisis discriminantes. Los datos emplea- dos corresponden a medidas morfometricas de 104 in- dividuos silvestres capturados en el area de invernada de la especie durante los veranos australes 2002-03 y 2003- 04 Encontramos diferencias significativas entre machos y hembras en todas las medidas morfometricas conside- radas, mientras que los juveniles se diferenciaron de los adultos solo en su masa corporal, la longitud del culmen y la longitud del halux. Usando solo la longitud del an- tebrazo como variable de prediccion, nuestra funcion dis- criminante cla.sific6 correctamente el 89.4% de los ma- chos y el 93.9% de las hembras. Una segunda funcion que incluia la longitud del antebrazo, de la cola y de la cuerda alar mejoro la separacion de los grupos y tambien el porcentaje de individuos correctamente clasificados (97.0% y 93.3% de los machos y hembras, respectiva- mente). El uso de medidas relacionadas con el tamano estructural de las aves como la longitud del antebrazo, de la cola y del ala hacen de este un metodo seguro y de amplia aplicacion, aun para aves pertenecientes a distin- tas clases de edad. [Traduccion de los autores] Acknowled gments We wish to thank N. Witheman, R.A. Sosa, V. Salvador, and M. Santillan for help in trapping hawks during field- work. R Bazaga did the lab work and J. Balbontin helped with data analysis. We are grateful for comments by J Balbontin, M. Bechard, and an anonymous reviewer that greatly improved this manuscript. We also thank person- nel from Ea. La Independencia and Ea. I.a Armenia II, the Agencia Cordoba Ambiente of Cordoba province, and the Direccion de Recursos Naturales of La Pampa province for providing permission to carry out trapping in the study areas. We thank A. Lanusse, S. Salva, M, Al- liaga, and J. Montoya for their hospitality and logistic sup- port during fieldwork in La Pampa province. J.H. Sara- sola had a fellowship from the Consejo Nacional de Investigaciones Cientificas y Tecnicas of Argentina. This study was supported by a grant from the Wildlife Con- servation Society (U.S.A.). Literature Cited BaibontIn, J., M. Ferrer, and E. Casado. 2001. Sex de- termination in Booted Eagles {Hieraaetus pennatus) us- ing molecular procedures and discriminant function analysis./. Raptor Res. 35:20-23. Berger, D.D. and H.C. Muelter. 1959. The bal-chatri: a trap for the birds of prey. Bird-Banding 30:18-2&. Bortototti, G.R. 1984a. Age and sex size variation in Golden Eagles. / Field Ornithol. .55:54—56. . 1984b. Criteria for determining age and sex of nestling Bald Eagles. / Field Ornithol. 55:467-481. Edwards, T.C. and M.N. Kociiert. 1986. Use of body weight and length of footpad as predictors of sex in Golden Eagles. /. Field Ornithol. 57:317—319. Eiiegren, H. 1996. Eirst gene on the avian W chromo- some (CHD) provides a tag for universal sexing of non-ratite birds. Proc. R. Soc. Land. B. 263:1635-1641. AND B.C. Sheldon. 1997. New tools for sex iden- December 2004 Short Communications 361 tification and the study of sex allocation in birds. Trends Ecol. Evol. 12:255-259. Engiand, A.S., M.J. Beciiard, and C.S. Houston. 199V. Swainson’s Hawk {Buteo swainsoni). In A. Poole and F. Gill [Eds.], The birds of North America, No. 265. The American Ornithologists’ Union, Washington, DC U.S.A. Ferrer, M. and C. De Le Court. 1992. Sex identification in the Spanish Imperial Eagle. /. Field Ornithol. 62:359- 364. Fridoitsson, A.K. and H. Ellegren. 1999. A simple and universal method for molecular sexing of non-ratite birds./. Avian Biol. 30:116-121. Fuller, M.R., W.S. Seegar, and L.S. Schueck. 1998. Routes and travel rates of migrating Peregrine Fal- cons Falco peregrinus and Swainson’s Hawk Buteo swain- soni in the Western Hemisphere./. Avian Biol. 29:433- 440. Garcelon, D.K., M.S. Martell, P.T. Redig, and L.C. Buoen. 1985. Morphometric, karyotypic, and laparo- scopic techniques for determining sex in Bald Eagles. /. Wildl. Manage. 49:595-599. Goldstein, M.I., B. Woodbridge, M.E. Zaccagnini, S.B. Canavelli, and a. Ianusse. 1996. An assessment of mortality of Swainson’s Hawks on wintering grounds in Argentina./. Raptor Res. 30:106-107. , P.H. Bloom, J.H. Sarasola, and T.E. Lachlr. 1999. Post-migration weight gain of Swainson’s Hawks in Argentina. Wilson Bull. 111:428-432. Manly, B.E.J. 1986. Multivariate statistical methods a primer. Chapman and Hall, Eondon, U.K. Palma, L., S. Mira, P. Cardia, P. Be)A, T. Guillemaud, N. Ferrand, M.L. Cancela, and L. Cancela da Fonseca 2001. Sexing Bonelli’s Eagle nestlings: morphometries versus molecular techniques. / Raptor Res. 35:187- 193. Smith, N.G., D.L. Goldstein, and G.A. Bartholomew. 1986. Is long-distance migration possible using only stored fat? Auk 103:607-611. Titus, K. and J.A. Mosher. 1984. Chance-corrected clas- sification for use in discriminant analysis: ecological applications. Am. Midi. Nat. 111:1-7. Wheeler, B.K. and W.S. Clark. 1995. A photographic guide to North American raptors. Academic Press, San Diego, CA U.S.A. Woodbridge, B., K.K. Finley, and S.T. Seager. 1995. An investigation of the Swainson’s Hawk in Argentina / Raptor Res. 29:202-204. Zar, J.H, 1996. Biostatistical analysis, 3rd Ed. Prentice Hall, Princeton, NJ U.S.A. Received 12 August 2003; accepted 11 July 2004 /. Raptor Res. 38(4) :361— 366 © 2004 The Raptor Research Foundation, Inc. Productivity AND Fledgling Sex Ratio in a Cinereous Vulture {Afgypius monachus) Population in Spain Auxiliadora Villegas, 1 Juan Manuel SAnchez-Guzman, Emilio Costillo, CAsimiro Corbacho, and Ricardo Moran Grupo de Jnvestigacion en Conservacion, Area de Zoologia, Universidad de Extremadura, Avda. Elvas s/n, 06071 Badajoz, Spain Key Words: Cinereous Vulture, Aegypius monachus; breed- ing conditions; brood size, fledgling-sex ratio; sexual dimor- phism; Spain. Upon initial review, the mechanism of chromosomal gender determination in birds and mammals seems to be a factor limiting the parents’ ability to modify the sex ratio of their progeny (Charnov 1982). However, sex al- location theory (Fisher 1930, Charnov 1982) predicts that the sex ratio can deviate from the expected 1:1, par- ticularly when the costs of rearing the two genders are ^ E-mail address: villegas@unex.es different. The optimal sex allocation for individuals can be predicted from three basic non-mutually exclusive hy- potheses (reviewed by Frank 1990). (1) Fisher (1930) proposed that parental expenditure in the population should be equal for all sons and daughters, which would result in a population sex ratio biased toward the gender that costs less to produce. (2) Trivers and Willard (1973) hypothesized that if the reproductive return differs be- tween genders depending on parental condition at the time of breeding, natural selection would favor faculta- tive adjustments of offspring sex ratios to obtain the max- imum fitness from a breeding attempt. (3) Charnov (1982) generalized Trivers and Willard’s hypothesis to cover any socio-environmental variable that might pre- 362 Short Communications VoL. 38, No. 4 dictably affect the fitness of sons and daughters unequal- ly Biases in the fledgling sex ratio may reflect an adaptive manipulation by the parents, by mechanisms that are not clearly understood (Krackow 1995). Otherwise, they may arise from nonadaptive mechanisms such as gender dif- ferences in chick survival rate between hatching and fledging (Clutton-Brock et al. 1985, Arroyo 2002) due to the parents’ inability to rear the more costly sex. In raptors, as in other groups of birds, some reproduc- tive parameters may reflect the environmental or paren- tal conditions during the breeding season (Dawson and Bortolotti 2000). For instance, a decrease in food intake due to the parents’ low foraging efficiency or reduction in food availability may be reflected in such reproductive variables as clutch size (Corbacho et al. 1997) or produc- tivity (Corbacho and Sanchez 2000, Dawson and Borto- lotti 2000). Therefore, interannual variations in popula- tion productivity may be good indicators of any environmental stresses affecting the birds. This is espe- cially true in Mediterranean environments, where the harsh and highly-variable conditions of summer are a ma- jor constraint on reproduction (Corbacho and Sanchez 2000, Costillo et al. 2002a). Recent empirical studies on raptors have provided ev- idence for the existence of significant biases in the sex ratio of the progeny depending on the environmental and social conditions experienced by the parents during the breeding season (e.g., Dzus et al. 1996, Appleby et al 1997, Post et al. 1999, Korpimaki et al. 2000, Byholm et al. 2002, Hipkiss et al. 2002). Most of these studies have focused on species with marked sexual dimorphism in size. In such cases, one can predict that a bias in the sex ratio, whether adaptive or not, will arise as a result of the difference in energy requirements or in vulnerability to adverse conditions between male and female chicks (Torres and Drummond 1997, Arroyo 2002) . There have been fewer studies of variations in the nestling sex ratio m species that are monomorphic or have only slight sex- ual dimorphism (Gooch et al. 1997, Sheldon et al. 1998, South and Wright 2002). The main limitation has been the difficulty of assigning gender to the chicks of these species. However, the advances in molecular techniques m the last decade allow the relatively easy and accurate gender determination of a wide survey of bird species (Ellegren 1996). The Cinereous Vulture {Aegypius monachus) is a large raptor widely distributed throughout the western Pale- arctic. Except for in the Iberian Peninsula, its popula- tions have been greatly reduced (Cramp and Simmons 1980, del Hoyo et al. 1994). This .species is of conserva- tion concern, being classified as near-threatened world- wide (Collar et al. 1994) and as vulnerable in Europe and Spain (Blanco and Gonzalez 1992, Tucker and Heath 1994). Knowledge of its basic ecology, including aspects related to reproduction, is therefore of great importance lor the conservation of the species. Here, we report the patterns of variation in productiv- ity and fledgling sex ratio in a Cinereous Vulture popu- lation in Spain during 3 yr. We analyze whether the inter- year variations in breeding success had any relationship to fledgling sex ratio or on the fledglings’ nutritional condition. Also we examined whether there were differ- ences in nutritional condition between the genders that may favor manipulation of the sex ratio in this species under certain conditions. There are two fundamental aspects in which this spe- cies differs from other raptors related to studies of sex ratio. Firstly, its sexual dimorphism is only slight, females are larger than males (Donazar 1993). Secondly, clutches are always composed of a single egg (Cramp and Sim- mons 1980). Study Area and Methods The study was carried out in 1998-2000, in the Sierra de San Pedro (Extremadura, southwestern Spain). This area and the region of Extremadura in general, support the most numerous populations of Cinereous Vultures in the Palearctic and those that have experienced the great- est growth in Spain in recent decades (Gonzalez 1990, Tucker and Heath 1994, Sanchez 1998, Costillo et al. 2002b). At the beginning of each breeding season (February- March), the entire area was surveyed to determine whether pairs were present. Then nesting attempts were monitored periodically during the breeding season to de- termine which pairs laid a clutch (Breeding Pairs), and how many reared a chick to fledging (Costillo et al 2002a) . As a measure of reproductive success we used the productivity, estimated as the number of fledglings pro- duced per monitored pair. The population’s annual pro- ductivity was used as an indicator of the breeding con- ditions that the individuals experienced in the study area A variable number of chicks (11-26) were captured in the nests and their blood sampled each year at 45-60 d of age. Assuming that the same pair reused the same nest in .successive years (Cramp and Simmons 1980), we se- lected 12 nests to monitor during the three breeding sea- sons. In five nests, we were able to sample a chick each year. In the other seven, either the pair did not lay in one of the years, or the chick did not .survive long enough to be sampled. In total, blood from 57 chicks was sampled during the study period, 29 of them belonged to one of the selected family groups. The blood sample was drawn from the brachial vein using 2 ml syringes and disposable needles. A small part (50 pi) was collected in a capillary tube and transferred to a tube with ethanol. For the remaining sample, pla.sma was obtained by centrifuging at 8000 g for 10 min, and stored at — 20°C. We determined the sex of the chicks u.sing polymerase-chain reaction (PCR) amplifications of the CHD genes (Ellegren 1996). Blood was boiled in NaOFI 100 mM for 10 min at 100°C before being added to the PCR reaction. PCR protocols were modified from Fridolfsson and Ellegren (1999). We u.sed the primer set 2552F-2781R, scoring one band in males and two bands in females in a 2% agarose-gel stained with ethydium bro- mide. December 2004 Short Communications 363 □ Breeding Pairs Success B Productivity Figure 1. Reproductive performance of the Cinereous Vulture population at Sierra de San Pedro (Extremadura, Spain) in 1998-2000. Breeding pairs = percent of mon- itored pairs that laid a clutch; Success = percent of re- productive pairs (pairs that laid a clutch) that produced a young; Productivity = Number of fledglings produced per monitored pair. The alkaline-phosphatase (ALP) concentration was measured in the plasma sample using a multiparametric autoanalyzer (Falcor 300, Menarini Diagnostics, Barce- lona, Spain) with the reagents recommended by Mena- gent (Menarini Diagnostics) . This variable was used as an index of chicks’ condition at the moment of sampling, because it has been found to be positively associated with physical condition in the chicks of this species (Villegas et al. 2002 ). The annual fledgling sex ratio is expressed as the pro- portion of males. The G-test was used to analyze devia- tions from the proportion 1 : 1 , inter-yearly differences in this parameter, and in the productivity. Inter-year and gender variations in the chicks’ physical condition were analyzed using a two-way analysis of variance in which sex and year were used as the principal factors and the al- kaline-phosphatase concentration as the dependent var- iable (Zar 1999). Results The percentage of Cinereous Vulture pairs initiating a clutch was similar in the 3 yr of study (G = 0.52, df = 2, P = 0.77). Productivity, however, declined from 0.81-0.53 during the study period (Fig. 1). There were statistically significant differences among years (G = 28.9, df = 2, P < 0.01), with the lowest value in 2000 (1998 vs. 2000: G^dj = 27.6, df = I, P< 0.01; 1999 vs. 2000: G^dj = 5.77, df = 1, P< 0.05). Of the 57 chicks monitored, 32 were males and 25 fe- males, so that the overall fledgling sex ratio for the 3 yr was 56.1%. This was not significantly different from 50% (Table 1). By years, there was an increase in the propor- tion of males in 2000 (0.7), although the difference from the 1:1 proportion was not statistically significant in any year (Table 1). Likewise, there were no inter-year differ- ences in the proportion of males (G = 1.057; df = 2, P = 0.78). The results were the same for the analysis of the 29 chicks of the selected family groups (Table 2). The fledglings’ physical condition, as measured by the ALP concentration (Table 3), differed among years (p 2 , 5 i = 4.17, P = 0.02), but not between sexes (Pj 51 = 1.01, P = 0.31), and the interaction of these factors was not significant (P 2.51 ~ 0.73, P = 0.49). In 2000, the chicks of both genders exhibited condition indices that were lower than in 1998 and 1999 (Table 3). Discussion The inter-year differences found in the productivity re- flect differences in the breeding conditions experienced by the individuals in the three seasons. At the time of breeding, the environmental, parental quality, or both were poorer in 2000, than in 1998 or 1999. The decline (Fig. 1) was mainly due to egg loss or the death of hatch- lings, as the percentage of pairs that initiated a clutch was similar in each of the 3 yr. This decrease in produc- tivity was accompanied by a significant decline in nutri- tional condition of the chicks at the time of sampling (Table 3). Costillo et al. (2002a) suggested adverse weath- er conditions as the cause of this decline in the area in the year 2000. Rain and cold at the time of hatching significantly affected the parents’ capacity to provide food, thus could have led to a decrease in the hatchlings’ physical condition, and even to their death from starva- tion (Donazar et al. 1988, Corbacho and Sanchez 2000, Dawson and Bortolotti 2000). Various studies on raptors have found evidence for the parents being able to adjust the sex ratio of their prog- eny, or for the existence of a gender bias in early nestling mortality, according to each year’s parental or ecological conditions (e.g., Olsen and Cockburn 1991, Wiebe and Bortolotti 1992, Dzus et al. 1996, Korpimaki et al. 2000, Arroyo 2002, Van den Burg et al. 2002). These studies Table 1. Sex ratio variation at the fledging stage in Cinereous Vultures by year in Spain. Sex ratios are expressed as percentage of males. \Lar Males Females Sex Ratio frTEST P-value 1998 6 5 54.5 0.00 1.00 1999 13 13 50.0 0.00 1.00 2000 13 7 65.0 1.26 0.26 Pooled 32 25 56.1 0.63 0.43 364 Short Communications VoL. 38, No. 4 Table 2. Sex-ratio variation at the fledging stage in Cinereous Vultures by year, within selected groups in Spain. Sex ratios are expressed as percentage of males. Year Males Females Sex Ratio Gtest P-value 1998 3 4 42.8 0.00 1.00 1999 7 5 58.3 0.08 0.77 2000 7 3 70.0 0.91 0.34 Pooled 17 12 58.6 0.55 0.46 used model species with marked sexual dimorphism and with brood sizes greater than one, in which gender-de- pendent food demand in combination with hatching asynchrony cause the cost-beneht ratio of rearing male and female chicks to differ (Dijkstra et al, 1998). Most raptors exhibit reversed sexual size dimorphism, with the females being larger (Newton 1979) . Growth de- mands, and hence the nutritional requirements, of fe- male chicks are therefore greater, so that females are more costly for the parents to rear and they are more vulnerable to adverse environmental conditions (Richner 1991). Differences between species in the magnitude of their sexual dimorphism are often related to the agility of the prey, with greater dimorphism corresponding to greater agility (Newton 1979). In vultures, which feed on carrion, although the females are larger, the difference IS slight, and there is a high degree of overlap in the sizes of the two sexes (Cramp and Simmons 1980, Donazar 1 993) . Studies of bird species that show little dimorphism have not shown any gender-environment interaction in chick performance (e.g., Sheldon et al. 1998, South and Wright 2002). Differences in nutritional condition between males and females should most likely be observed when the environmental conditions were seriously limiting — a sit- uation that did not arise in this area during the study period, as deduced from the high percentage of pairs that initiated a clutch and successfully reared a fledgling (Costillo et al. 2002a). Nonetheless, this does not imply that the costs associated with rearing the two genders are equal. In the case that the growth rate or the energy requirements of one sex are greater than those of the other, rearing a chick of the more costly gender in years with suboptimal conditions, as year 2000 was, would in- volve greater effort for the parents. The result could be a greater decline in their physical condition over the course of the breeding season, which could compromise chick’s survival and even the parents’ capacity to breed in the following year (Donazar 1993, Korpimaki et al 2000, Genovart 2002) . In such case, female Cinereous Vultures might obtain htness benefits by producing a chick of the cheaper sex when the breeding conditions are not optimal (Meyers 1978), with the advantage that, by laying only a single egg, they can allow themselves to reabsorb a zygote of the unsuitable gender without in- curring the costs of hatching asynchrony or of a signih- cant lengthening of time on the clutch (Emlen 1997). In summary, our results suggest a shift toward males of this Cinereous Vulture population’s fledgling sex ratio in years with suboptimal breeding conditions (Table 1), al- though probably because of the small sample size, this bias was not statistically significant. This bias could reflect a greater production of males at hatching or a greater early mortality incidence of females. Resumen. — Estudios empiricos recientes en rapaces han proporcionado evidencias de la cxistencia de sesgos sig- nificativos en la proporcion de sexos de volantones en funcion de las condiciones ambientales experimentadas por los padres durante la estacion reproductiva. La ma- yoria de estos estudios se han centrado en especies con un marcado dimorfismo sexual y con tamanos de nidada mayores que uno. En este trabajo estudiamos los patro- nes de variacion en la proporcion de sexos de volantones en una poblacion de la Peninsula Iberica de Aegypius mon- achus, un ave rapaz con un dimorfismo sexual poco acu- Table 3. Alkaline phosphatase concentration (Ul/L) in male and female Cinereous Vultures during 3 yr in Spain. Year Males Females POOLEIV 7* Mean ± SD (V) Mean ± SD (A) Mean ± SD (V) 1998 1574.5 ± 232.9 (6) 1376.4 ± 186.4 (5) 1484.5 ± 227.4 (11) A ns 1999 1437.1 ± 262.5 (13) 1343,0 ± 341.0 (13) 1390.0 ± 302.0 (26) A ns 2000 1168.9 ± 284.9 (13) 1219.6 ± 249.9 0) 1186.6 ± 267.6 (20) B ns The different letters in this column indicate that the differences between the year 2000 and the years 1998 and 1999 were significant Probability that concentrations were different between males and females. December 2004 Short Communications 365 sado y con un tamano de nidada igual a uno, durante el periodo 1998-2000. Analizamos si las variaciones anuales en el exito reproductivo tienen alguna influencia en la proporcion de sexos y en la condicion nutricional de los polios, medida por la concentracion de fosfatasas alcali- nas en plasma, y si esta muestra diferencias entre polios de distinto sexo. Tanto la productividad de la poblacion como la condicion fisica de los polios fue menor en el ano 2000 que en los dos anos previos. I.a proporcion de sexos no se aparto de forma significativa de 1:1 ni mostro diferencias interanuales signiflcativas, aunque la propor- cion de machos incremento a 0.70 en el ano de menor productividad. La condicion nutricional no mostro dife- rencias signiflcativas entre polios de distinto sexo. Estu- dios mas detallados son necesarios para determinar los mecanismos que actuan en la desviacion de la propor- cion de sexos en esta especie. [Traduccion de los autores] Acknowledgments We are grateful to all the people who helped with the field and laboratory work. Comments of JJ. Negro and an anonymous referee greatly improved a first version of the manuscript. Financial support for the study and for E. Costillo was provided by the project LIFE-NAT “Ges- tion de Zonas de Especial Proteccion par alas Aves en Extremadura: Aguila Perdicera y Buitre Negro” of the Government of Extremadura (Spain) . A. Villegas was sup- ported by the pre-doctoral grant FIC99B096 from the Government of Extremadura. Literature Cued Appleby, B.M., SJ. Petty, J.K. Blakey, P. Rainey, and D.W. Macdonald. 1997. Does variation of sex ratio en- hance reproductive success of offspring in Tawny Owls {Strix aluco)? Proc. R. Soc. Land, i? 264:1 1 1 1—1 1 16. Arroyo, B. 2002. Sex-biased nestling mortality in the Montagu’s Harrier Circus pygargus. J Avian Biol. 33: 455-460. Blanco, J.C. andJ.L. Gonzalez. 1992. Libro Rojo de los vertebrados de Espana. Institute para la Conservacion de la Naturaleza, Madrid, Spain. Byiiolm, R, E. Ranta, V. Kaitata, H. Linden, P. Saurola, and M. Wikman. 2002. Resource availability and gos- hawk offspring sex ratio variation: a large-scale eco- logical phenomenon./. Anim. Ecol. 71:994—1001. Charnov, E.L. 1982. The theory of sex allocation. Prince- ton Univ. Press, Princeton, NJ U.S.A. Clutton-Brock, T.H., S.D. Albon, and F.E. Guinness. 1985. Parental investment and sex differences in ju- venile mortality in birds and mammals. Nature 313: 131-133. Gollar, N.J., M.S. Crosby, and A.J. Statterseield. 1994. Bird to watch 2: the world list of threatened birds. BirdLife Conservation Series, BirdLife International, Cambridge, U.K. CoocH, E.G., D.B. Lank, R.J. Robertson, and F. Cooke. 1997. Effects of parental age and environmental change on offspring sex-ratio in a precocial bird. J Anim. Ecol. 66:189-202. CoRBACHO, C., J.M. Sanchez-Guzman, and a. Sanchez. 1997. Breeding biology of Montagu’s Harrier Circus pygargus in agricultural environments of southwest Spain; comparison with other population in the west- ern palearctic. Bird Study 44:166-175. AND J.M. SAnchez. 2000. Clutch size and egg size in the breeding strategy of Montagu's Harriers in a Mediterranean area. Bird Study 47:24.5-248. CosTiLi.o, E., A. Villegas, G. Lagoa, and P. Corbacho 2002a, Biologfa y parametros reproductores de las poblacion es de buitre negro [Aegypius monachus) de Extremadura. Congreso Gestion de ZEPAs en Extre- madura: Aguila Perdiciera y Buitre Negro, Caceres, Spain. , R. Moran, G. Lagoa, P. Corbacho, and A. Vil- legas. 2002b. Nucleos reproductores y evolucion de las poblaciones de buitre negro {Aegypius monachus) de Extremadura. Congreso Gestion de ZEPAs en Ex- tremadura: Aguila Perdiciera y Buitre Negro, Gaceres, Spain. Cramp, S. and K,E. Simmons, 1980. Handbook of the birds of the western palearctic. Vol. 2. Oxford Univ Press, Oxford, U.K. Dawson, R.D. and G.R. Bortolotti. 2000. Reproductive success of American Kestrels: the role of prey abun- dance and weather. Condor 102:814-822. DEI. Hoyo, J., A. Elliott, and J. Sargatal. 1994. Hand- book of the birds of the world. Vol. 2. Lynx Edicions, Barcelona, Spain. Dijkstra, C., S. Daan, and I. Pen. 1998. Fledgling sex ratios in relation to brood size in size-dimorphic al- tricial birds. Behav. Ecol. 9:287-296. Donazar, J.A. 1993. Los buitres ibericos: biologia y con- servacion. J.M. Reyero [Ed.], Madrid, Spain. , J. Elosegui, and a. Senosiain. 1988. Exito re- productor del buitre leonado {Gyps fulvus) en navar- ra. Acta Vertebr. 15:187-192. Dzus, E.H., G.R. Bortolotti, and J.M. Gerrard. 1996 Does sex-biased hatching order in Bald Eagles vary with food resources? Bioscience 3:252-258. Ellegren, H. 1996. First gene on the avian W chromo- some (CHD) provides a tag for universal sexing of non-ratite birds. Proc. R. Soc. Lond. B 263:1635-1641. Emlen, S.T. 1997. When mothers prefer daughters over sons. Trends Ecol. Evol. 12:291-292. Fisher, R.A. 1930. The genetical theory of natural selec- tion. Oxford Univ. Press, Oxford, U.K. Frank, S.A. 1990. Sex allocation theory for birds and mammals. Ann. Rev. Ecol. Syst. 21:13—55. Fridolfsson, A.K. and H. Ellegren. 1999. A simple and universal method for molecular sexing of non-ratite birds./. Avian Biol. 30:116—121. Genovart, M. 2002. Molecular genetics applied to the study and conservation of seabirds. Ph.D. dissertation. Univ. Barcelona, Barcelona, Spain. 366 Short Communications VoL. 38, No. 4 Gonzaitz, L.M. 1990. Situacion de las poblaciones de Aguila imperial y buitre negro en Espafia. Quercus 58: 16-22. Hipkiss, T., B. Hornfeldt, U. Eki.und, and S. Berlin. 2002. Year-dependent sex-biased mortality in supple- mentary-fed Tengmalm’s Owl nestlings. J. Anim. Ecol. 71:693-699. Korpimakj, E., C.A. May, D. Parkin, J. Wetton, and J. WiEHN. 2000. Environmental- and parental condition- related variation in sex ratio of kestrel broods./. Avian Biol 31:128-134. Krackow, S. 1995. Potential mechanisms for sex ratio ad- justment in mammals and birds. Biol Rev. 70:225-241 . Myers, J.H. 1978. Sex-ratio adjustment under food stress — maximization of quality or numbers of off- spring. Am.. Nat. 112:381-388. Newton, 1. 1979. Populaiton ecology of raptors. Buteo Books, Vermillion, SD U.S.A. Olsen, P.D. and A. Cockburn. 1991. Female-biased sex allocation in Peregrine Falcons and other raptors. Be- hav. Ecol SociohioL 28:417-423. Post, E., M.C. Forchhammer, N.C. Stenseth, and R. Lancjvatn. 1999. Extrinsic modification of vertebrate sex ratios by climatic variation. Am. Nat. 154:194-204. Richner, H. 1991. The growth dynamics of sexually di- morphic birds and Fisher’s theory: does sex-specific growth contribute to balanced sex ratios? Fund. Ecol 5:19-28. Sanchez, J.J. 1998. The recovery of the Cinereous Vul- ture, {Aegypius monachus), in Spain. Pages 89-99 m E. Tewes, JJ- Sanchez, J.J. Heredia, and M. B. van Lex- mond [Eds.], International symposium on the Cine- reous Vulture in south eastern Europe and adjacent regions. Cinereous Vulture Conservation Foundation, Palma de Mallorca, Spain. Sheldon, B.C., J. Merila, G. Lind(;ren, and H. Elle- gren. 1998. Gender and environmental sensitivity in nestling collared flycatchers. Ecology 79:1939-1948. South, J.M. and T.F. Wright. 2002. Nestling sex ratios in the Yellow-naped Amazon: no evidence for adaptive modification. Condor 104:437-440. Torres, R. and H. Drummond. 1997. Female-biased mor- tality in nestlings of a bird with size dimorphism. J Anim. Ecol. 66:859-865. Trivers, R.L. and D.E. Willard. 1973. Natural selection of parental ability to vary the sex ratio of offspring Science 179:90-92. Tucker, G.M. and M.F Heath. 1994. Birds in Europe- their conservation status. Birdlife International, Cam- bridge, U.K. Van den Burg, A.B., J. Van Diermen, J.D.M. Muskens, S. Van Rijn, and R. Zollinger. 2002. Sex ratio compar- isons between nestlings and dead embryos of the spar- rowhawk Accipiter nisus. Ibis 144:E39-E44. Villegas, A., J.M. Sanchez, E. Costillo, and C. Corba- cho. 2002. Blood chemistry and haematocrit of the Cinereous Vulture {Aegypius monachus) . Comp. Biochem. Physiol A. 132:489-497. WlEBE, K.L. AND C.R. Bortolotti. 1992. Facultative sex ratio manipulation in American Kestrels. Behav. Ecol Sociobiol. 30:379— .386. Zar, J.H. 1999. Biostatistical analysis, 4th ed. Prentice Hall, Simon and Schuster, Englewood Cliffs, NJ U.S.A. Received 31 December 2003; accepted 11 September 2004 Associate Editor: Juan Jose Negro December 2004 Short Communications 367 J Raptor Res. 38(4):367-37l © 2004 The Raptor Research Foundation, Inc. Nest Provisioning of the Oriental Honey-buzzard (Pernis ptilorhyncus) in Northern Taiwan Kuang-Ying Huang Yangmingshan National Park Headquarters, Yangmingshan, Taipei, 112, Taiwan, R.O.C. and Institute of Ecology and Evolutionary Biology, National Taiwan University, Taipei, 106, Taiwan, R.O.C. Yao-Sung Lin Institute of Ecology and Evolutionary Biology, National Taiwan University, Taipei, 1 06, Taiwan, R. O. C. Lucia Liu Severinghaus^ Institute of Zoology, Academia Sinica, Taipei, 115, Taiwan, R.O.C. Key Words: Oriental Honey-buzzard', Pernis ptilorhyncus; nest provisioning, paper wasps', Polistes; Parapolybia; Taiwan. Raptors in the genus Pernis have a unique diet consist- ing primarily of the larvae, pupae, and adults of social Hymenoptera (Clark 1994, Orta 1994, Bijlsma 1999, Ga- mauf 1999). The European Honey-buzzard {P. apivorus) has been studied in some detail (Kostrzewa 1987a, 1987b, 1987c, Steiner 1992, Saporetti et al. 1994, Selas 1997, Van Manen 2000), but the biology of the Oriental Honey- buzzard {P. ptilorhyncus) has not been studied in depth (Clark 1994, Orta 1994, Morioka et al. 1995). Based on our monthly raptor surveys in northern Tai- wan’s Yangmingshan National Park (25°10'N, 121°33'E) between 1993 and 2003, the Oriental Honey-buzzard was a common spring and autumn migrant, a rare winter vis- itor, and an uncommon summer resident. We found one Oriental Honey-buzzard nest in 1994 and one in 1999. This paper reports the composition, seasonal variation, and relative importance of the prey species provisioned to the nest in 1999 and compares the male and female contributions to the nestling. Methods We constructed a hide eye-level with the nest on a steep slope about 30 m away. We monitored the nest pro- visioning of the parents and recorded the time and the gender each time an adult returned to the nest. The male could be distinguished based on an orange-brown iris and broader dark tail bands, while the female has a yellow iris and narrower dark tail bands (Morioka et al. 1995). We simultaneously videotaped the nest-provision- ing events. From these videotapes, we identified prey by comparing these with reference specimens in terms of structure, shape, color, and size of the wasp nests, the color and placement of the stem of the wasp nests, and ^ Corresponding author’s e-mail address: zobbowl@gate. sinica. edu.tw the color of the seal on cells containing pupae. Obser- vations were made from 12 July— 19 August, the day be- fore the nestling fledged. Because it was impossible to determine the exact sizes of the wasp nests from the videotapes due to filming an- gles, we collected empty wasp nests discarded under the honey-buzzard nest for species identification and count- ed the number of cells per wasp nest. Based on the sizes of these wasp nests, we categorized all the wasp nests m the videotapes into small, medium, or large classes (Table 1 ). To understand the relationship between wasp activity and prey capture rate, we identified the paper wasps ac- tive in July and August. We searched for wasp nests along three trails in the study area through wooded areas with small cabins, each 1 km in length. We scanned the veg- etation from canopy to the shrub layer and checked buildings carefully. Each transect line was searched at least twice to ensure the discovery of all wasp nests. Eor each wasp nest found, we measured height above ground, its length and width, recorded the number of cells, and described its shape. We collected one actively- used nest for each species found in order to obtain the proportion of cells occupied by eggs, different larval in- stars and pupae, and measured the mass of larvae at each in star. Biomass per wasp nest was calculated with the follow- ing equation: Biomass = the number of cells/nest X pro- portion of cells occupied X the mean biomass/larva. The number of cells found on each wasp nest was related to size (Table 1). We used the wasp nests we collected as standards to determine cell occupancy and larval mass Biomass estimates of reptiles and frogs were based on the mean mass of >10 individuals per species caught in the study area. Results Food Composition and Seasonality. We observed the honey-buzzard nest for 26 d (204.7 hr) and recorded 123 food items delivered by parents. Among these, 78.9% were complete wasp nests {N = 97), 16.3% were Swin- hoe’s brown frogs {Rana swinhoana', N = 20), and 4.9% 368 Short Communications VoL. 38, No. 4 Table 1. Wasp nest-size classes and their cell numbers. Species Size Rank Cell Number Polistis tenebriocosus Small <39 Medium 40-69 Large >70 Polistis rothneyi Small <119 Medium 120-219 Large >220 Polistis gigas Small 10-19 Medium 20-29 Large >30 Polistis sp. Medium 780 Parapolybia varia Small <179 Medium 180-349 Large >350 Unidentified wasp comb Small 60 {Po. tenebriocosus or Medium 80 Po. rothneyi) Large 100 were yellow-mouthed japalura lizards {Japalura polygonata xanthostoma; N = 6). All 97 wasp nests belonged to paper wasps (Polistinae, Vespidae), but two belonged to an un- known species and 24 could not be distinguished be- tween Po. tmebriocosus and Po. rothneyi because either the nestling blocked the camera or the speed of delivery was too rapid. Among prey items that could he identified, Po. tenebriocosus nests (N = 36) constituted 29.3%, Po. rothneyi 13% {N - 16), Po. gigas 9.8% {N = 12), Parapolybia varia 5.7% {N = 7), and Polistis sp. 1.6% {N = 2). The fre- quency distribution of wasp nests brought back to the nestling was not different from those found along our transect lines (nests recorded: 29, Po. tenebriocosus =21, Po. rothneyi = 3, Po. gigas = 2, and Pa. varia = 3, Polistis sp. = 0, = 5.657, df = 4, P > 0.05). The body sizes of these five species of paper wasps var- ied from the world’s largest paper wasp {Po. gigas-, 4.5 cm for adult males) to a new species, Polistis sp, (1,3 cm). The colony sizes of these wasps also varied from 12 cells in a Po. gigas nest to more than 400 cells in a Pa. varia nest. Based on biomass estimates from the nests we col- lected, we found that the pupae and the fourth and fifth mstars of Po. tenebriocosus, Po. rothneyi, Po. gigas, and Pa. varia constituted 93.4%, 95.4%, 95.7%, and 87.6% of the total biomass of a nest, and occupied 44%, 45%, 56%, and 39.6% of the cells, respectively. We did not find any nests of the new Polistis sp. to obtain larva/pupa mass measurements. Because nests of this species were brought back only twice during the entire study period, we decided not to include this species in our biomass calculations. Considering the cumulative biomass of all prey types delivered to the honey-buzzard nest, frogs and lizards to- gether made up 16.9%, while Polistinae wasps made up Table 2. Relative biomass of different prey types brought to the Oriental Honey-buzzard nest in northern Taiwan. Species Total Biomass (g) Percent Polistis tenebriocosus 1003.00 31.59 Po. rothneyi 689.34 21.71 Po. ^gas 244.50 7.70 Parapolybia varia 90.84 2.86 Unidentified nests=* 611.16 19.25 Frogs 500.00 15.75 Lizards 36.00 1.13 Total 3174.84 100.00 ® Either Po. tenebriocosus or Po. rothneyi. We calculated the biomass by using the mean mass and mean cell occupancy of the two species. 83.1% (Table 2). Frogs were delivered in July and early August, and lizards were delivered in July only. By mid- August, honey-buzzards provisioned only paper wasps (Fig. 1). The patterns of seasonal variation were similar by prey frequency and by biomass. The biomass delivered to the nest per hour of observation fluctuated between 5-40 g throughout the nestling period except on the third d before fledging (Fig. 2) . We did not monitor the nest on 18 August and so were unable to determine if this increased rate of prey delivery took place two con- secutive days before fledging. Paper Wasp Nest Height Distribution. The 29 paper wasps nests we found along trails hung from vegetation {N = 10) or eaves {N = 19) from 0.5—11.5 m above the ground. More than 86% were between 0.5-4 m high, the remaining four nests were above 6 m (3 Po. tenebriocosus and 1 Po. rothneyi). Sexual Differences in Contribution. The male made By Frequency By Biomass Figure 1. Diet of Oriental Honey-buzzard nestling based on the frequency of prey deliveries to the nest. Paper wasps in diet included: Polistis tenebriocosus, Po. rothneyi, Po gigas, and Parapolybia varia. December 2004 Short Communications 369 Biomass Delivered / hr hO 0 ) 00 o 3 0 0 0 0 0 10 day average: 18.45 g/hr 20.36 gftir 13.45 g/hr 22.99 g/hi .-till ill J llll -III liill ll L A u 12 14 16 18 20 22 24 26 28 30 1 3 5 7 9 11 13 15 17 19 July August Figure 2- Biomass delivered to an Oriental Honey-buz- zard nest in northern Taiwan per hour of observation. The nestling fledged on 20 August. The mean biomass delivered per hour for each 10-d period is given at the top of the histogram. 49.6% of the visits and the female 29.3%; the sex was uncertain in the remaining visits. Among the biomass de- livered by adults of known sex, the male contributed 56% and the female 44% in July and in early August. After 9 August, female contribution ceased completely. The dif- ference in biomass contribution between sexes across time periods was significant (x^ = 305.31, P < 0.0001). Even if all the unknown feedings were made by the fe- male, the male still would have contributed more biomass than the female during the final 10 d before fledging. Discussion Prey Choice. Oriental Honey-buzzards captured prey that were on the ground, such as frogs in the shrub layer, or in the sub-canopy paper wasp nests. The pair we stud- ied in Yangmingshan primarily brought Polistes nests as food for their young. Among known wasp prey, the fre- quency of different species of wasp nests they brought to the nestling was proportional to those available in the environment. However, they ignored the nests of the small paper wasp, Po. takasagonus, even though these nests were as common as the Po. tenehriocosus nests along our transect lines. A colony of Po. takasagonus was con- structing a nest just below the honey-buzzard nest, but the honey-buzzards never showed interest in it. The bio- mass of the largest Po. takasagonus nest we found weighed less than 9 g, less than the smallest nest brought back by parents (9.91 g). There may be a minimum-size thre.sh- old for a wasp nest to be energetically profitable for hon- ey-buzzards. Another potential prey are Vespa hornets (Vespidae, Vespinae), which are large, conspicuous, and usually number by the hundreds per colony. The nests of some species are built on high branches and can be as large as 60 cm in diameter (M.C. Kuo pers. comm.). Hornets and wild bees (Apidae) were active within the home range of the honey-buzzard pair, but the nests of these families were never delivered to the nestling. The honey-buzzards’ food choice was most likely influ- enced by the accessibility of prey and its energy content in relation to the handling cost. Hornets are aggressive and usually defend their combs vigorously (Evans and Eberhard 1970, Kuo and Yeh 1987). Ranking the aggres- siveness of wasps into five classes (1 being most aggres- sive), Po. tenehriocosus and Po. gigas would in be class 5, Po. rothneyi and Pa. varia in class 3, while the Vespa species would be in classes 1-3 (Kuo and Yeh 1987). Although the stomach contents of European Honey-buzzards con- tained small numbers of hornets (Loskutova 1985), and adult Oriental Honey-buzzards probably consume hor- nets, preying on hornet nests no doubt demands a high energy cost. Specifically, honey-buzzards would need to evade tbe attack of a large number of hornets and would have a high risk of injury. Wild bees are common in Taiwan, but most nest in rock crevices and tree cavities (Kuo and Yeh 1987). Pol- istinae nests in Taiwan always hang by a stem, which makes them easy for honey-buzzards to remove in flight When paper wasps are abundant, there should be little incentive for a honey-buzzard to open logs or remove rocks to access a bee nest. Seasonal Variation in Prey Composition. Paper wasps in Taiwan generally establish colonies in late April or May, and colony expansion usually occurs in June and July, followed by a mating period in July or August before they enter the dormant wintering stage, when breeding and foraging activities cease (Kuo and Yeh 1987, Lu et al. 1992). Based on this chronology, wasp nests would be small in the early part of the breeding season (July) . Dur- ing this period, the Oriental Honey-buzzards delivered wasp nests to the nestling and supplemented the nest- ling’s diet with frogs and lizards. By mid-August, although frogs and lizards were still common, wasp colonies were large and the honey-buzzards delivered only paper wasp nests. European Honey-buzzards also foraged almost en- tirely on wasp nests when wasps were available (Martin 1992, Ferguson-Lee and Christie 2001). European Honey-buzzards find wasp nests by watching the insects return from foraging trips (Grigor’yev et al 1977, Martin 1992, Ferguson-Lee and Christie 2001). We do not know if Oriental Honey-buzzards could determine whether a wasp was departing from or returning to its nest. However, by following a flying wasp, honey-buzzards could either find the wasp foraging or come upon a nest Timing of Breeding. If the two nests we found are typ- ical for this species, the Oriental Honey-buzzard has a much later breeding season than the other species of raptors in Taiwan. The hatching of eggs in July appears to be timed with the increase of paper wasp colonies m our study area. The delayed onset of its breeding season could result from the unavailability of wasp nests earlier in the spring. Wasp abundance has been known to influ- ence the breeding success of European Honey-buzzards (Kostrzewa 1987c, Saporetti et al. 1994, Steiner 2000). Sexual Differences in Provisioning. Erom midjuly to early August, the female Oriental Honey-buzzard contrib- 370 Short Communications VoL. 38, No. 4 uted more than 30% of the biomass delivered to the nest- ling, and the male at least 40%. After 9 August, the fe- male’s contribution ceased completely even though she remained in the vicinity of the nest. Using a modeling analysis, Brodin et al. (2003) showed that whether a fe- male raptor assisted with hunting for the young late in the nestling period depended on the hunting success of the male, environmental conditions, and the energy de- mands of the young. Because the pair of honey-buzzards we monitored had only one nestling, perhaps the male alone could provide sufficient prey for the young (re- ported clutch size for Oriental Honey-buzzard is two eggs; Sung et al. 1991). Our study advanced what was known about Oriental Honey-buzzard. We found that adults primarily brought paper wasps as prey for their nestling, and the frequency of each species of wasp nest brought back reflected prey availability. Low energetic profitability most likely made some species of wasp nests undesirable. In addition, our results from one nest showed that the female delivered food less often and contributed less biomass to the nest- ling than the male. Resumen. — Se conoce muy poco sobre la ecologia y el comportamiento de Pernis ptilorhyncus. Con base en cen- sos realizados entre 1993 y 2003, se determine que esta especie es un visitante poco comun durante el verano en Yangmingshan, Taipei, Taiwan. Encontramos un nido en 1994 y otro en 1999. A traves de observaciones directas y de grabaciones de video realizadas en 1999, registramos las presas llevadas al nido durante 26 dias, entre el 12 de julio y el 19 de agosto {N = 204.7 hr, N = 123 presas). Nuestros resultados revelaron que el 78.9% (83.7% en terminos de biomasa) de los items alimenticios fueron nidos de avispas, el 16.2% (15,3% en biomasa) fueron ranas y el 4.9% (1.1% en bioraasa) fueron lagartijas. Cin- co especies de avispas que construyen nidos de papel fue- ron identificadas: Polistes tenebriocosus, Po. rothneyi, Po. gi- gas, Parapolybia variay Po. sp. Las ranas y lagartijas fueron registradas solo entre el 12 de julio y el 10 de agosto. La dieta de la cria de P. ptilorhyncus reflejo la fauna disponi- ble de avispas y el ciclo anual de estas en el area de es- tudio. Los machos y hembras no contribuyeron igual- mente a alimentar a su cria. Las hembras dejaron de llevar alimento diez dias antes de que el pichon aban- donara el nido. [Traduccion del equipo editorial] ACKNOWI.EDGMF.NTS We wish to thank Chung Kun-yen, Hsieh Shu-Shin, Huang Hui-Tzu, Huang Yung-Fen, Lee Tien-Fu, Liang Chieh-Te, Tseng Wen-Shiu, Jack Yu, and Raptor Research Group of Taiwan for help in field data collection; Dr.J.T. Chao for helping to identify paper wasps; Prof. M.C. Kuo for information on wasp natural history; Drs. Ian Newton, Nigel Collar, R. Yosef, and two anonymous reviewers for their valuable suggestions that greatly improved an ear- lier draft of this paper; and Mr. Wayne Hsu for English editing. Literature Cited Biji ^SMA, R.G. 1999. Do honey buzzards Pernis apivorus produce pellets? Limosa 72:99-103. Brodin, A., K.I. Joensson, and N. Hoi.mgren. 2003. Op- timal energy allocation and behaviour in female rap- torial birds during the nestling period. Ecoscience 10 140-150. Clark, W.S. 1994. Barred Honey-buzzard. Page 112 mj del Hoyo, A. Elliott, andj. Sargatal [Eds.], Handbook of the birds of the world. Vol. 2. Lynx Edicions, Bar- celona, Spain. Evans, H.E. and M.J. Eberhard. 1970. The wasps. Univ. Michigan Press, Ann Arbor, MI U.S.A. Eerguson-Lee, J. and D. Christie. 2001. Raptors of the world. Houghton Mifflin Company, New York, NY U. S.A. Gamauf, a. 1999. Is the European Honey-buzzard {Perms apivorus) a feeding specialist? The influence of social Hymenoptera on habitat selection and home range size. Egretta 42:57-85. Grigor’yev, N.D., V.A. Popov, and YK. Popov. 1977. Otryad sokoloobraznye. Pages 76-116 in Y.A. Popov [Ed.], Ptitsy volzhsko-kamskogo kraya. Nauka Press, Moscow, Russia. Kuo, M.S. and W.H. Yeh. 1987. Ecological and natural history research for Taiwan Vespidae (III): Vespa, Pol- istes, Parapolybia, Ropalidia. Ann. Res. Rep. of Chai-Yi Agr.l Coll. 16:77-104. Kostrzewa, a. 1987a. Territorial behavior competition and nest nse of three tree-breeding raptors accipitres J. Ornithol. 128:495-496. . 1987b. Quantitative analyses of nest habitat sep- aration in the Common Buzzard Buteo buteo, goshawk Accipiter gentiles, and honey buzzard Pernis apivorus. J Ornithol. 128:209-230. . 1987c. The effect of weather on density and re- production success in honey buzzard Pernis apivorus Vogelwarte 34:33—46. 1996. A comparative study of nest-site occupancy and breeding performance as indicators for nesting- habitat quality in three Enropean raptor species. Ethol. Ecol. Evol. 8:1-18. Loskutova, N.M. 1985. Khishchnye ptitsy i sovy Bashkir- skogo zapovednika. Pages 45-58 in V.M. Galushin and V. G. Krever [Eds.], Khishchnye ptiLsy i sovy v zapov- ednikakh RSFSR. CNIL Glavokhoty Press, Moscow, Russia. Lu, S.S., J.T. Chao, and L.L. Lee. 1992. Colony cycle of Polistes jadwigae Dalla Torre (Hymenoptera: Vespidae) in northern Taiwan. Chin. J. Entomol. 12:171-181. Martin, B.P. 1992. Birds of prey of the British Isles. David &: Charles, Newton Abhot, Devon, U.K. Morioka, T, T Kanoucti, T Kawata, and N. Yamagata December 2004 Short Communications 371 1995. Birds of prey in Japan. Okumura Printing, To- kyo, Japan. Orta, J. 1994. Western Honey-buzzard and Crested Hon- ey-buzzard. Pages 111-112 mj. del Hoyo, A. Elliott, and J. Sargatal [Eds.], Handbook of the birds of the world. Vol. 2. Lynx Edicions, Barcelona, Spain. Saporetti, E, W. Guenzani, and P. Pavan. 1994. Density, habitat, and breeding success of diurnal raptors in the Prealps of NW Italy. Riv. Ital. Ornitol. 63:145—173. Selas, V. 1997. Nest-site selection by four sympatric forest raptors in southern Norway./. Raptor Res. 31:16-25. Steiner, H. 1992. The diurnal raptor community of a farmland area in upper Austria. Egretta 35:96—110. . 2000. Forest fragmentation, competition, and cli- matic dependence in the honey buzzard {Pernis api- vorus ) . /. Ornithol. 141: 68—7 6 . Sung, YJ., WJ. Lo, M. Wei, and X.L. Chang. 1991. Pre- liminary study on the breeding biology of the Orien- tal Honey-buzzard {Pernis ptilorhyncus) . Science Print- ing, Beijing, China. Van Manen, W. 2000. Reproductive strategy of honey buzzards Pernis apivorus in the northern Netherlands. Limosa 73:81—86. Received 5 January 2004; accepted 3 September 2004 J Raptor Res. 38(4):371-374 © 2004 The Raptor Research Foundation, Inc. Winter Diet of the Greater Spotted Eagle {Aquila clanga) in the Amvrakikos Wetlands, Greece Haralambos Alivizatos^ 4 Zaliki Str., GR-115 24 Athens, Greece Dimitris Papandropoulos 21 Zaimi Str., GR-26 500 Rion, Greece Stamatis National Center for Marine Research, Institute of Inland Key Words: Greater Spotted Eagle, Aquila clanga; diet, win- ter, wetland] Amvrakikos', Greece. The Greater Spotted Eagle {Aquila clanga) is a globally- threatened raptor, which breeds from Eastern Europe to the Pacific coast of the northern Far East and winters locally in southern Europe, Asia, the Middle East, and sub-Saharan Africa as far south as Uganda and Kenya (Tucker and Heath 1994). The main reasons for its de- cline are habitat destruction and degradation (in both forests, where the species nests, and wetlands, where it forages), disturbance during the breeding season, illegal shooting (mainly in migration), and to a lesser extent, nest robbing (Tucker and Heath 1994, Meyburg et al. 2001). The primary proposed conservation measures in- clude the establishment of protected areas in the breed- ing habitat of the species, preservation of a mosaic of breeding-feeding habitat, protection of wetlands, avoid- ance of disturbance during the breeding season within 300 m of nests, as well as improvement of legislation. ^ E-mail address: xaraaliv@otenet.gr ZOGARIS Waters, P.O. Box 712, 19 013 Anavyssos Attiki, Greece international cooperation, monitoring, and research (Meyburg et al. 2001). The biology of the Greater Spot- ted Eagle has not been examined to a great extent; ad- ditionally, its diet has been examined almost exclusively during the breeding season (e.g., Priklonsky 1960, Gal- ushin 1962, Pankin 1972, Ivanovsky 1996). Few studies of any kind have been conducted in the winter quarters of this species (Moltoni 1943, Francois 1992, Qingxia 1996) Although an important population of Greater Spotted Eagles winters in the extensive wetlands of northern and central Greece (45-85 individuals), up to now it has re- ceived very little attention (Hallmann 1989, Handrinos and Akriotis 1997). Here, we present the results of an examination of Greater Spotted Eagles at a key wintering site, the Amvrakikos wetlands, western Greece, where up to 12 individuals winter each year. Study Area and Methods Our study area was in the Amvrakikos wetlands, west- ern Greece (38°59'-39°lTN, 20°44'-21°07'E). Amvraki- kos is a Ramsar Wetland, a Special Protection Area, and proposed National Park. It is one of the largest wetlands in Greece covering 220 km^ including river deltas, coastal 372 Short Communications VoL. 38, No. 4 lagoons, extensive saltmarshes, reedbeds, relic riparian woods, and grazed grasslands. Large areas bordering the salt marshes have been drained and support intensive ag- riculture and livestock farming. The roost site of the Greater Spotted Eagles observed was located in a small clump of very tall Eucalyptus trees on the banks of the Arachthos River, <2 km away from a major village. This roost is ca. 2 km from the estuarine wetlands of the river delta and is in a region of intensive agriculture, domi- nated by citrus plantations. The scattered trees along the bank of this region of the river include white poplar {Po- pulus alba), willows (Salix sp), ashes (Fraxinus sp.), and common alder (Alnus glutinosa) . Amid the native riparian trees are a few planted eucalyptus {Eucalyptus sp.), some reaching a considerable height (ca. 40 m). In the winter of 2001—02, eagles roosted exclusively on the Eucalyptus sp , but they did not seem to use these trees in the winter of 2002-03, although they did gather to roost in the tree stands in the vicinity. In other parts of Amvrakikos, eagles roost on oaks {Quercus spp.) in the adjacent limestone hills (100-300 ra elevation) 1-5 km away from the wet- land habitats. We located four such roosts in the Am- vrakikos, but pellets where collected only from the one at the Arachthos Delta because the others were in rela- tively inaccessible areas (steep, forested hills), or difficult to pin-point with precision (woodlots in marshes) . The diet of the .species was studied by the analysis of pellets that were collected in January and March 2002 from under the regular roost, in a small Eucalyptus clump, which was used by six to seven Greater Spotted Eagles. Up to two Common Buzzards {Buteo buteo) were also seen to use the trees near the Eucalyptus trees, but not near the eagles. Buzzard pellets were easily distinguished from those of the eagles by a combination of both size and form (eagle pellets were ca. 70-120 X 30—50 mm and had a generally loose form, while those of buzzards were ca 50-75 X 20—35 mm and were quite compact). Be- cause of these differences and in combination with their location, we feel that it is unlikely that pellets of the two species were confused. Prey remains were identified with the help of reference books (Brown et al. 1987, Chinery 1993, Macdonald and Barrett 1993). The number of prey Items was determined by counting the number of skull and major limb bones represented, when these were present; unless these indicated otherwise, we assumed the presence of one individual of each prey type per pel- let REsui;rs We collected 57 pellets from the Eucalyptus roost, used by 6-7 individuals. The size of the pellets was ca. 70-120 X 30-50 mm. In the study area, the Greater Spotted Ea- gle preyed primarily on birds, but it also took mammals, reptiles, amphibians, hsh, and insects (Table 1 ) . By num- bers, among 95 prey items, birds comprised 84.2% of the diet, followed by insects, mammals, reptiles, fish and am- phibians (Table 1). By mass, birds were even more im- portant (94.6%), while other prey were of comparatively minor importance. The most important birds, in terms of biomass, were ducks, particularly Eurasian Wigeons {Anas penelope) and Gommon Teals {Anas crecca) , as well as Common Moorhens {Gallinula chloropus), Common Coots {Fulica atra), Little Egrets {Egretta garzetta), and small gulls {Larus spp). We also found lead shot in 3.2% of the pellets. Discussion Although the sample was comparatively small, some in- teresting conclusions may be drawn. The proportion of birds found in the diet was higher in the Amvrakikos wetlands than in any other study of the Greater Spotted Eagle. Priklonsky (1960), in the mouth of the Belaya Riv- er, Russia, based on pellet analysis (78 pellets and 274 prey items) reported the diet of the species, consisting of 69.0% mammals, 23.0% reptiles, 7.7% birds, and 0.3% fish (only vertebrates included). In addition, insects oc- curred in 29.4% of the pellets and carrion (primarily moose [Alces alces^) in at least 20.5% of the pellets. Study of the prey remains in the nests of the eagles also re- vealed a predominance of mammals and birds. Most mammals were small rodents, while most reptiles were snakes. Priklonsky (1960) summarized the results of sev- eral studies as follows: mammals 66.0-88.6%, birds 46.2- 97.0%, insects 5.5—34.0%, and reptiles 0.0-12.3% (occur- rence in pellets). Galushin (1962) in the valley of the Oka River during the period 1954—57, found the follow- ing yearly variation in diet: mammals 42.6-68.0%, birds 32.0—69.0%, fish 0.0— 1.8%, and reptiles and amphibians each 0.0-0. 6% (based on prey numbers). In Byelorussia, Ivanovsky (1996) found that 50 prey items (from re- mains) were 53.8% mammals, 21.1% amphibians, 17.4% birds, and 7.7% reptiles. Qingxia (1996) studied the diet of the eagles wintering in the Lishan Nature Reserve (China) and found the diet consisted of about equal numbers of mammals and birds {N = 68 prey items) . The mammals consumed were mostly rodents, as well as some rabbits, collectively making up 52-1% of the diet by bio- mass. A variety of bird species were also eaten (47.9% of diet by biomass), including several gamebirds (Phasiani- dae) . Often, the study of pellets can result in biased esti- mates of diet composition; that is, this methodology can lead to the underestimate of certain prey types (particu- larly amphibians) and the overestimate in others. Vlachos and Papageorgiou (1996) found that a large proportion of the prey delivered in the nests of Lesser Spotted Eagles {Aquila pomarina) , as recorded by direct observation, con- sisted of frogs, but these were only infrequently repre- sented in the pellet analysis. Also many, or all, of the insects found in pellets in this and other studies, may have been ingested secondarily (i.e., consumed by the eagle prey). Nearly all the prey species identified inhabited wetland habitats, particularly the salt marshes and coastal lagoons (Table 1 ) . Surprisingly, few species of prey from the ad- jacent agricultural lands were taken (e.g., the small ro- dents and beetles). The high proportion of water birds in our study was Df.cember 2004 Short Communications 373 Table 1. Diet of the Greater Spotted Eagle {Aquila danga) in the Amvrakikos Wetland {N = 57 pellets). Common Name Scientific Name Number Percent Number Perctnt Biomass Insects Insecta 5 5.3 <0.1 Ground beetles Carabidae 5 5.3 <0.1 Fish Pisces 2 2.1 3.1 Mullets Mugilidae 2 2.1 3.1 Amphibians Amphibia 1 1.1 0.1 Frogs Rana sp. 1 1.1 0.1 Reptiles Reptilia 3 3.2 0.5 Snakes Colubridae 3 3.2 0.5 Birds Aves 80 84.2 94.6 Little Egret Egretta garzetta 3 3.2 4.7 Eurasian Wigeon Anas penelope 11 11.6 27.4 Common Teal Anas crecca 17 17.9 15.9 Unidentified ducks Anas spp. 5 5.3 7.8 Spotted Crake Porzana porzana 1 1.1 0.3 Water Rail Rallus aquaticus 2 2.1 0.7 Common Moorhen Gallinula chloropus 16 16.8 14.9 Common Coot Fulica atra 4 4.2 8.7 Unidentified rails Rallidae 1 1.1 0.3 Northern I^apwing Vanellus vanellus 1 1.1 0.6 Common Redshank Tringa totanus 1 1.1 0.4 Common Snipe Gallinago gallinago 1 1.1 0.3 Unidentified scolopacids Scolopacidae 1 1.1 0.3 Yellow-legged Gull Larus cachinnans 1 1.1 3.7 Unidentified small gulls Earns spp. 5 5.3 4.7 Hooded Crow Corvus corone 1 1.1 1.6 Reed Bunting Emheriza schoenidus 3 3.2 0.3 Unidentified passerines Passeriformes 4 4.2 0.4 Unidentified birds Aves 4 4.2 1.9 Mammals Mammalia 4 4.2 1.7 Eastern Hedgehog Erinaceus concolor 1 1.1 1.6 Unidentified voles Microtus sp. 1 1.1 0.1 Unidentified mice Mus spp. 2 2.1 0.1 Total — 95 — — probably due to their high availability in the area, com- pared with other potential prey during the study period. The winter of 2001-02 was relatively “severe” with peri- ods of frosts, thus reptiles and amphibians would hardly have been available to the eagles (in fact, the presence of reptiles and amphibians, even in low numbers, was surprising) . This may be true, to a lesser extent, for the smaller mammals as well. On the other hand, large num- bers of waterfowl winter in the Amvrakikos wetlands, in- cluding up to 100 000 ducks and coots in certain years. The eagles mostly hunt prey from perches (often on low tamarisk, Tamarix spp.), or from embankments, or pa- trolling by gliding low over extensive wetland areas. where many water birds are present. These areas, as well as providing relatively abundant prey, also provide cover suitable for the eagles to surprise their prey. We made relatively few observations of the eagles over large open lagoons or the surrounding sandy spits, in which the highest numbers of waterfowl winter. As well as capturing healthy prey, the eagles probably also took injured or dead birds, as suggested by the pres- ence of lead shot in two pellets. The presence of lead shot in the pellets indicated that the Greater Spotted Ea- gle fed on birds shot, but not retrieved by hunters (see also Hallmann 1989). Lead shot is considered to be a serious threat for many wetland raptors, such as the West- 374 Short Communications VoL. 38, No. 4 ern Marsh-Harrier ( Circus aeruginosus) in France (Pain et al 1997). Therefore, the potential effect of lead shot on the Greater Spotted Eagle should be investigated. Stands of trees, which provide perches for look-outs and roosting, are important for the hunting eagles and should be protected. Riparian woodlots and large trees are often cut by local inhabitants, and thus roost sites (actual or potential) may be threatened by this activity. Poisoning is also a potential threat. Because several un- authorized refuge dumps exist in the Amvrakikos Wet- lands, the threat of poisoning through scavenging at dumps is present. Also, eagles could feed on poisoned carrion laid out illegally by local farmers for deterring foxes and other canines. Therefore, extent of scavenging and the susceptibility of the Greater Spotted Eagle to such poisoning should be investigated. Resumen. — Se examino la dieta de aguilas Aquila clanga en los humedales de Amvrakikos en el oeste de Grecia por medio de analisis de egagropilas. Con base en el nu- mero de presas, la dieta estuvo compuesta principal- men te por aves (84.2%) y por cantidades menores de msectos (5.3%), mamiferos (4.2%), reptiles (3.2%), pe- ces (2.1%) y anfibios (1.1%). Las aves fueron suprema- mente importantes en terminos de biomasa (94.6%); las presas principales fueron patos del genero Anas, y los ralidos Gallinula chloropus y Fulica atra. La gran mayoria de las presas correspondio a especies que se encuentran casi exclusivamente en humedales. [Traduccion del equipo editorial] Acknowi.edgments We are grateful to OIKOS-Nature Management, Ltd., who supported this study, and in particular to T. Arapis, G. Rigas, and V. Hatzirvassanis. We are also grateful to the wardens and other field workers of the LIFE-Nature project, and in particular to K. Floudas and Y Rousso- poulos. We also thank A. Dimitropoulos and I. Schogolev for the translation of the Russian articles and Dr. V. Gout- ner for his valuable comments on the original draft. We also appreciate B.-U. Meyburg, D. Ontiveros, and an anonymous reviewer for their comments on an earlier draft of this manuscript. Liierature Cited Brown, R., J. Ferguson, M. Lawrenct:, and D. Lees. 1987. Tracks and signs of the birds of Britain and Eu- rope: an identification guide. Christopher Helm, Lon- don, U.K. Chinery, M. 1993. Collins field guide: insects of Britain and northern Europe. Harper-Collins Publishers, London, U.K. Francois, J. 1992. Observations sur la presence hivernal de Faigle criard {Aquila clanga) en Moselle. Ciconia 16’ 117-125. Galushin, V.M. 1962. The Greater Spotted Eagle in the valley of the Oka River and its influence in the num- bers of some birds. Uch. Zap. Mask. Pedagog. Inst. Len- ina. 186:115-151. Hallmann, B. 1989. Status and distribution of Aquila in Greece. Biol. Gallo-Hell. 15:171-176. HANdrinos, G. and T. Akriotis. 1997. The birds of Greece. Christopher Helm, London, U.K. Ivanovsky, V. 1996. Notes on the breeding biology of .spotted eagles Aquila clanga and A. pomarina in Bye- lorussia. Pages 297-299 in B.-U. Meyburg and R.D. Chancellor [Eds.], Eagle studies. World Working Group on Birds of Prey and Owls, Berlin, Germany. Macdonald, D. and P. Barrett. 1993. Collins field guide: mammals of Britain and Europe. Harper Collins Pub- li.shers, London, U.K. Meyburg, B.-U., L. Haraszthy, M. Strazds, and N Schaffer. 2001. European species action plan for Greater Spotted Eagle {Aquila clanga). Pages 1-16 in N. Schaffer and U. Gallo-Orsi [Eds.], European Union action plans for eight priority bird species. Eu- ropean Commission, Brussels, Belgium. Moltoni, D.E. 1943. L’alimentazione delF aquila ana- traia {Aquila clanga). Riv. Ital. Ornitol. 13:97-100. Pain, D.J., C. Bavaux, and G. Burneleau. 1997. Seasonal blood concentration of lead in Marsh Harrier Circus aeruginosus from Charente Maritime (France): rela- tionship with the hunting season. Biol. Conserv. 81:1- 7. Pankin, N.S. 1972. On the feeding of the Greater Spotted Eagle {Aquila clanga) pallas in the Bureya River Valley (Amur region). Pages 331-333 in Zoological prob- lems of Siberia. NAUKA, Novosibirisk, Russia. Priklonsky, S.G. 1960. On the food of the Greater Spot- ted Eagle at the mouth of the Belaya River. Ornitol. 3‘ 174-179. Qingxia, Z. 1996. Winter ecology of Aquila clanga in Lis- han Nature Reserve. Sichuan J. Zool. 15:170-172. Tucker, G.M. and M.F. Heath. 1994. Birds in Europe, their conservation status. BirdLife Conservation Se- ries No. 3. BirdLife International, Cambridge, U.K. ViACHOS, C.G. and N.K. Papageorgiou. 1996. Breeding biology and feeding of the Lesser Spotted Eagle Aq- uila pomarina in Dadia Forest, north-eastern Greece Pages 337-347 in B.-U. Meyburg and R.D. Chancellor [Eds.], Eagle studies. World Working Group on Birds of Prey and Owls, Berlin, Germany. Received 10 December 2003; accepted 23 September 2004 December 2004 Short Communications 375 /. Raptor Res. 38(4) :375-377 © 2004 The Raptor Research Foundation, Inc. Gender Determination oe Eurasian Eagle-Owls {Bubo bubo) by Morphology Marla del mar Delgado^ and Vincenzo Penteriani Department of Applied Biology, Estacion Biologica de Donana (EBD), C.S.I.C., Avda. De Maria Luisa s/n. Pabellon del Peru, Apdo. 1056, 41013 Sevilla, Spain Key Words: Eurasian Eagle-Owl; Bubo bubo; discriminant function analysis; forearm; sexing, gender determination. Gender determination is an important prerequisite to studies on many aspects of avian biology such as foraging ecology (e.g., Anderson and Norberg 1981), evolutionary ecology (e.g., Glutton-Brock 1986), survivorship (e.g., Newton et al. 1983), and conservation genetics (e.g., Griffith and Tiwari 1995). Many avian species show no sexual dimorphism in plumage, but the gender of indi- viduals may be determined by body measurements. Most raptors are dimorphic in size, which allows for the de- velopment of gender determination methods based on morphometric data. Nonetheless, this method has been applied to a relatively small number of species (e.g., Bor- tolotti 1984a, 1984b, Garcelon et al. 1985, Edwards and Kochert 1987, Ferrer and De Le Court 1992, Balbontin et al. 2001). The Eurasian Eagle-Owl {Bubo bubo) is a sexually mono- morphic species and, although females are bigger than males (i.e., reversed sexual dimorphism) gender deter- mination in the field is only possible through detection of gender-specific calls (Penteriani 1996). Due to its con- servation concerns, the high density of this species in several Mediterranean areas of its breeding range (e.g., Penteriani et al. 2002, Delgado et al. 2003, Penteriani et al. 2004), its eclectism in habitat preferences (e.g., Pen- teriani et al. 2001, Marches! et al. 2002, Martinez et al. 2003), its complex social communication (e.g., Penteri- ani 2002, Penteriani 2003), and its impact on bird com- munities (e.g., Sergio et al. 2003), this species has been the subject of increasing research in the last few years. In this context, determination of gender for this species represents a useful tool in future studies examining in- tersexual and intrasexual patterns. Our objective was to provide an inexpensive and practical tool to determine the gender of eagle-owls in the field using a minimum number of morphometric measurements. Methods We measured 13 morphological characteristics of 50 skins of Eurasian Eagle-Owls {N = 22 males and N = 2S females) from the collections of the Estacion Biologica de Donana (Andalusia, Spain) and the Natural Science ^ E-mail address: mmdelgado@ebd.csic.es Museum of Madrid. All eagle-owls analyzed came from Spain and gender was previously determined by internal examination of reproductive organs. To avoid the con- founding effect of age, we only used skins of adult indi- viduals when morphometric differences seem to be most- ly related to gender rather than age. Length of claws, tarsus, bill including cere, exposed culmen without cere, and bill depth were taken using a caliper (±0.1 mm) (Bortolotti 1984a, 1984c). The four claws of the left foot were measured from the hallux claw (toe number one) to the outer claw (toe number four). Length of wing chord, tail, ear tufts, and forearm (the length from the front of the folded wrist to the proximal extremity of the ulna) were measured with a metal ruler to the nearest mm (Bortolotti 1984a, 1984c). To mini- mize measurement errors, each specimen was measured three times. For analyses, we used the mean values of these three measurements. To determine which morphometric variables were the best predictors for gender determination, we conducted a two-step analysis. First, a Mest was conducted for the 13 variables to identily the descriptors for which the be- tween gender variance was higher. Secondly, we used a discriminant function analysis (DFA) to obtain the func- tion best discriminating between males and females. Chi- square analysis was employed to test the significance of the gender clas.sification established by the DFA proce- dure (Sokal and Rohlf 1995). DFA has been widely used for gender determination in bird species with monomor- phic plumage (e.g., Scolaro et al. 1983, Maran and Myers 1984, Manners and Patton 1985, Malacalaza and Hall 1988). A DFA produced a linear combination of several morphometric variables that best discriminated samples of individuals of known gender. This function was then used to predict the sex of unknown birds (Sokal and Rohlf 1981, Norusis 1988). Because large discriminant functions can be cumbersome (McCloskey and Thomp- son 2000), we established a level of significance of P < 0.0001 as a threshold to select the significant Ltest vari- ables that were used in the DFA. Results and Discussion The i-test revealed that females were significantly larger than males in all the variables measured except tail, wing chord, and ear tufts (Table 1). Second claw, forearm, length of exposed culmen without cere, and bill depth were the most dimorphic variables (P< 0.0001). The DFA produced the following discriminant equation: D = —28.740 + 0.204(second claw) + 0.714(forearm) + 0.158 (culmen without cere) + 0.1 13 (bill depth). 376 Short Communications VoL. 38, No. 4 Table 1. Morphometric of study skins of males and female Eurasian Eagle-Owls {Bubo bubo) from Spain. Claws are numbered according to toe numbers (hallux = 1, outer claw = 4). X Females {N SD = 28) Range Males {N X = 22) SD Range t df P Claw of toe 1 34.62 3.56 26.74-40.10 30.60 3.33 21.66-33.98 -3.512 37 0.0010 Claw of toe 2 34.89 2.36 27.72-38.50 31.34 2.19 27.87-38.84 -5.252 44 0.0001 Claw of toe 3 30.53 3.12 25.32-36.84 28.35 2.42 24.99-33.67 -2.436 39 0.0201 Claw of toe 4 29.70 3.39 20.09-33.62 26.61 1.30 24.22-28.86 -3.806 38 0.0001 Tarsus (L) 102.5 6.58 83.98-112.00 93.8 4.5 80-104.00 -4.543 47 0.0001 Tail 258.76 14.73 229.67-293.67 250.36 15.70 232.33-293.00 -1.919 47 0.0613 Wing^ 44.13 3.07 32.26-47.46 43.19 1.87 40.67-48.90 -1.260 47 0.2140 Forearm 20.04 0.84 18-21.93 18.83 0.72 17.06-19.76 -5.288 46 0.0001 BCER*^ 48.56 2.87 42.60-54.35 44.87 3.15 38.90-52.03 -4.280 47 0.0001 BCUL'^ 32.89 1.82 27.25-35.17 30.10 1.61 27.18-34.90 -5.523 46 0.0001 Bill depth 28.47 3.19 20.25-33.97 24.81 3.61 12.22-30.09 -3.750 47 0.0001 Ear tuft (left) 72.96 6.70 46-84 72.23 4.53 63.33-81.33 -0.421 44 0.6762 Ear tuft (right) 72.16 10.51 42.33-86.67 74.75 3.82 65-79.67 1.096 45 0.2790 ® Wing chord. Bill including cere. Exposed culmen without cere. A correct classification was obtained for 90.5% of males and 90.9% of females. Hence, overall 90.7% of cases were classified correctly. This classification was significantly better than random (chi-square = 41.360, P = 0.0001). There was a clear separation between males and females along the first discriminant axis (Fig. 1). Variables used in this study were easy to measure in the field and have been shown to be good predictors of gen- der in several other bird species (e.g., Calvo and Bolton 1997, Renner and Davis 1999, Leader 2000). Also, in comparison with other proposed morphometric criteria for gender determination (e.g., wing and body mass), the descriptors we used were not influenced by molting, con- dition of specimens, or of the feathers. 12 n - 3.6 - 2.8 -2 - 1.2 Scor«s Figure 1. Discriminant Function Analysis (DFA) scores of male {N = 22) and female {N = 28) Eurasian Eagle- Owl study skins. The four variables used in classifying genders were: second claw, forearm, length of exposed culmen without cere, and bill depth. The length of the forearm has been used successfully for gender determination in two other raptor species, Spanish Imperial Eagles {Aquila adalbertv, Ferrer and De Le Court 1992) and Bonelli’s Eagles {Hieraaetus fasciatus; Balbontin et al. 2001). In a similar study, Martinez et al. (2002) also considered this parameter to be the best pre- dictor of gender for Eurasian Eagle-Owls. Additionally, our study revealed a small overlap between males and females. Finally, the forearm variable has two additional advantages: it is easy to measure, and repeated measure- ments taken by both the same and different observers show little variation (Ferrer and De Le Court 1992). Gender determination by DFA is applicable to adults year round, when most alternative methods are limited by season (e.g., during the breeding season) or expensive (e.g., karyotyping). However, the application of our DFA model may be limited because of the pronounced geo- graphical variation of body size exhibited by eagle-owls (Penteriani 1996). This factor needs to be taken into ac- count when applying our DFA model to other popula- tions. However, our approach could be used to derive similar DFA models for other Eurasian Eagle-Owl popu- lations. Resumen. — Bubo bubo es un ave rapaz nocturna grande que presenta dimorfismo sexual de tamano revertido. A traves del analisis de 13 parametros morfologicos colec- tados de 50 especimenes de museo {N = 22 machos y N = 28 hembras), asignamos correctamente el genero a 90.7% de los individuos por medio de analisis de funcion discriminante. Las variables morfologicas usadas para predecir el genero incluyeron la profundidad del pico, December 2004 Short Communications 377 longitud de la segunda garra, longitud del antebrazo y longitud de la parte expuesta del culmen sin cera. [Traduccion del equipo editorial] Acknowledgments We thank J. Barreiro of the National Museum of Nat- ural Sciences (Madrid), P. Cabot and C. Valle of the Es- tacion Biologica de Dohana (Seville) for their precious help during the analyses of Eagle Owl specimens. G. Bor- tolotti, M. Di Vittorio, A. Harmata, H. Mikkola, F. Sergio, and an anonymous referee provided helpful comments on the first draft of the manuscript. Literature Cited Anderson, M. and R.A. Norberg. 1981. Evolution of re- versed sexual size dimorphism and role partitioning among predatory birds, with a size scaling of flight performance. Biol. J. Linn. Soc. 15:105-30. Balbontin, J., M. Ferrer, and E. Casado. 2001. Sex de- termination in Booted Eagle (Hieraaetus pennatus) us- ing molecular procedures and discriminant function analysis./. Raptor Res. 35:20-23. Bortolotti, G.R. 1984a. Criteria for determining age and sex of nestling Bald Eagles. /. Field Ornithol. 55: 467-481. . 1984b. Age and sex variation in Golden Eagles. J. Field Ornithol. 55:54-66. . 1984c. Sexual size dimorphism and age-related size variation in Bald Eagles./. Wildl. Manag. 48:72-81. Calvo, B. and M. Bolton. 1997. Sexing Shags Phalacro- corax aristotelis from external measurements using dis- criminant analysis. Ringing Migr. 18:50-56. Clutton-Brock, T.H. 1986. Sex ratio variation in birds. Ibis 128:317-329. Delgado, M.M., C. Maggio, A. Basanta, C. Escot, and V. Penteriani. 2003. Preliminary data on an eagle owl Bubo bubo population in S-W Spain. World Working Group on Birds of Prey and Owls, Berlin, Germany. Edwards, T.C. and M.N. Kochert. 1987. Use of body weight and length of foodpad as predictors of sex in Golden Eagles./. Field Ornithol. 58:144—147. Ferrer, M. and C. De Le Court. 1992. Sex determina- tion in the Spanish Imperial Eagle. /. Field Ornithol. 62:359-364. Garcelon, D.K., M.S. Martell, P.T. Redig, and L.C. Buoen. 1985. Morphometric, karyotypic, and laparo- scopic techniques for determining sex in Bald Eagles. /. Wildl. Manag. 49:595-599. Griffith, R. and B. Tiwari. 1995. Sex of the last wild Spix's Macaw. Nature 375:454. Hanners, L.A. and S.R. Patton. 1985. Sexing Laughing Gulls using external measurements and discriminant analysis./. Field Ornithol. 56:158-164. Leader, N. 2000. Predicting the sex of Blackstarts {Cer- comela melanura) by discriminant analysis. /. Zool. 46: 149-154. Malacalaza, V.E. and M.A. Hall. 1988. Sexing adult King Cormorants {Phalacrocorax albiventer) by discrim- inant analysis. Colon. Waterbirds . Maran, J.L. and J.P. Myers. 1984. A discrimination and evaluation of two techniques for sexing wintering Sanderlings. /. Field Ornithol. 55:336-342. Marchesi, L., F. Sergio, and P. Pedrini. 2002. Costs and benefits of breeding in human-altered landscapes for the eagle owl Bubo bubo. Ibis 144:164-177. Martinez, J.A., I. Zuberogoitia, and R. Alonso. 2002 Rapaces nocturnas: guia para la determinacion de la edad y el sexo en Estrigiformes Ibericas. Monticola Editions, Madrid, Spain. , D. Serrano, and I. Zuberogoitia. 2003. Predic- tive models of habitat preferences for the Eurasian Eagle-Owl Bubo bubo: a multiscale approach. Ecography 26:21-28. McCloskey, J.T. and J.E. Thompson. 2000. Aging and sexing common snipe using discriminant analysis. /. Wildl. Manag. 64:960—969. Newton, I., M. Marquiss, and P. Rothery. 1983. Age structure and survival in a sparrowhawk population. /. Anim. Ecol. 52:.591-602. Norusis, M.J. 1988. Advanced statistics. SPSS/PC -H. SPSS Inc., Chicago, IL U.S.A. Penteriani, V. 1996. The eagle owl. Calderini Ed., Bolo- gna, Italy. , M. Gallardo, and P. Roche. 2001. Landscape structure and food supply affect eagle owl {Bubo bubo) density and breeding performance: a case of intra- population heterogeneity. /. Zool. Lond. 257:365-372. . 2002. Variation in the function of eagle owl vocal behavior: territorial defence and intra-pair commu- nication. Ethol. Ecol. Evol. 14:275-281. , M. Gai.lardo, P. Roche, and H. Cazassus. 2002 Effects of landscape spatial structure and composition on the settlement of the eagle owl Bubo bubo in a Med- iterranean habitat. Ardea 89:331-340. . 2003. Breeding density affects the honesty of bird vocal displays as possible indicators of male/ territory quality. Ibis 145:E127-E135. , M.M. Delgado, C. Maggio, A. Aradis, and F. Sergio. 2004. Development of chicks and pre-dispers- al behaviour of young in the eagle owl. Ibis 3:150. Renner, M. and L.S. Davis. 1999. Sexing Little Penguins Eudyptula minor from Cook Strait, New Zealand using discriminant function analysis. Emu 99:74—79. ScoLARO, J.A., M.A. May, and I.M. Ximenez. 1983. The Magellanic Penguin {Spheniscus magellanicus) , sexing adults by discriminant analysis of morphometric char- acters. Auk 100:221-224. Sergio, R, L. Marchesi, and P. Pedrini. 2003. Spatial re- fugia and the coexistence of a diurnal raptor with its intraguild owl predator, /. Anim. Ecol. 72:232-245. SOKAL, R.R. AND rj- Rohlf. 1995. Biometry, the princi- ples and practice of statistics in biological research, 3rd Ed. W.H. Freeman & Co., New York, NY U.S.A Received 13 November 2003; accepted 3 June 2004 Associate Editor: Fabrizio Sergio Letters J Raptor Res. 38(4):378-379 © 2004 The Raptor Research Foundation, Inc. A Possible Case of Double Brooding of Eagle-Owls {Bubo bubo) in Spain Multiple breeding in the same reproductive season is a common life-history tactic by which individuals can increase their genetic representation in future generations (Roff 2002, Life history evolution. Sinauer, Sunderland, U.K.) However, this strategy implies costs in terms of survival, future fledgling production, or both, although such costs of reproduction depend closely on environmental conditions (Verhulst 1998, Fund. Ecol. 12:132-140). Perrins (1970, Ibis 112:242-255) hypothesized that females should start laying as soon as they are physiologically capable, and that mterindividual differences in the timing of breeding could be caused by differential acquisition of food (food supply hypothesis) . Early laying pairs are expected to be more able to carry out second clutches than pairs that lay later, because females are in a better physiological condition and they have time enough for additional breeding attempts in the same reproductive season (e.g., Morrison 1998, Auk 115:979-987; Marks and Perkins 1999, Wilson Bull 11. 273-276). Laying second and even third clutches has been reported as usual in a wide variety of bird species, mainly passerines (e.g., Friesen et al. 2000, Wilson Bull 112:505-509; Mahony et al. 2001, Wilson Bull 113:441-444), while it IS considered as exceptional in others (e.g., Miller 2003, Wilson Bull 115:94—95). In raptors, there are reports of double clutches (Newton 1979, Population ecology of raptors. T. & A.D. Poyser, London, U.K.), although most of them correspond to small-sized species with short reproductive periods and in favorable areas or years of high food availability (e.g., Korpimaki 1988a,/. Anim. Ecol 57:1027-1039; Marks and Perkins 1999). For large species with long reproductive periods, even replacement clutches are rarely reported (Newton 1979, Bull and Henjum 1990, Ecology of the Great Gray Owl. USDA For. Serv. Gen. Tech. Rept. PNW-GTR-265, Portland, OR U.S.A.; Cabeza and de la Cruz 2001, Ardeola 48:233-236; Margalida and Bertran 2002,/. Raptor Res. 36:154—155; Martinez and Blanco 2002, Ardeola 49 297-299) . Owls of the family Strigidae typically raise no more than one brood per year, but some records of double brooding have been reported (Kellomaki et al. 1977, OrnisFenn. 54:124—135; Millsap and Bera 1990, Wilson Bull 102. 313-317; Forsman et al. 1997, Cowrfor 97:1078-1080; Marks and Perkins 1999). Replacement clutches are known for Eurasian Eagle-Owls {Bubo bubo) (e.g., Blondel and Badan 1976, Nos Oiseaux 33:189-219), but only one possible double clutch has been reported in southeastern Spain (Martinez et al. 2003, Ardeola 50:77-79) . Collaborators and 1 have monitored a Eurasian Eagle-Owl population since 1999 in the province of Toledo, central Spain (39°47'N, 4°04'W). The study area extends over 2400 km^ with meso-mediterranean climate, with mean tem- peratures of 26°C and 5°C in July and January, respectively, and 300-400 mm of rainfall concentrated in spring and autumn. To date, we have located 100 pairs of eagle-owls, but we estimated that at least twice this number could be breeding in the study area. Mean nearest neighbor distance (hereafter NND; x = 1.4, SD =1.7 km, N = 100) is the lowest, whereas clutch size {x =■ 3.67, SD = 0.53, N = 36) and mean number of fledglings per successful pair {x = 2 72, SD = 0.78, N = 50; J. Ortego unpubl. data) is the highest reported to date (Marches! et al. 2002, Ibis 144:164— 177). These population traits are likely related to the fact that rabbit {Oryctolagus cuniculus) density in the study area IS one of the largest reported for the Iberian Peninsula (Blanco 1998, Mamiferos de Espaha. Geoplaneta, Barcelona, Spain) . On 13 April 2002, I found three fledged chicks, which flew away when we approached them, around the nest of an eagle-owl pair. Approximately 50 m away from this nest, and in the same cliff, I found an adult bird incubating two eggs. In spite of the high density of eagle-owls in the study area, it seems unlikely that the second clutch belonged to a different pair. Eagle-owls are closely linked to ravines in the study area, which provide both nest sites with low human disturbance and high rabbit availability (Ortego and Diaz in press, Seleccion del habitat de nidificadon del buho real [Bubo bubo hispanus] en la provincia de Toledo. In Actas de las XVI jornadas ornitologicas espaholas. Sociedad Espanola de Ornitologia, Madrid, Spain). The minimum NND recorded in the study area was 150 m, and the minimum mean for pairs living in the same ravine was 389 m {N =14 pairs). The NND for the pairs settled in the ravine where the reported nest was located was 895 m {N = 11). This relatively low density makes the settlement of two pairs in the same cliff unlikely. According to chick development, I estimated the laying date for the three fledglings around 5 January. Laying date of this pair in the following breeding season was estimated around 19 December, the earliest for a sample of 31 pairs {x = 28 January, SD = 17.5). The early laying date of this pair could have facilitated a second clutch (Morrison 1998; Marks and Perkins 1999; however, see Martinez et al. 2003). In addition, in the previous breeding season I found 17 378 December 2004 Letters 379 young rabbits in this nest when the last egg was still hatching. Storage of rabbits in the nest before hatching is common in our study area (J. Ortego unpubl. data), and I have never noticed such a large number of rabbits in a sample of 36 nests. These data suggest that the pair involved could be living in a high quality territory that yields relatively large numbers of available prey of high-energetic value, consequently minimizing the costs of a multiple brooding (Verhulst 1998, Fund. Ecol. 12:132-140). Martinez et al. (2003) offered two alternative explanations that could explain the apparent double-brooding ob- servations in southwestern Spain. Death of the female could have allowed the male to pair with another female physiologically ready to start the reproduction, or the male could have been polygynous (Bull and Henjum 1990), as has been observed in other raptors responding to a superabundant food supply (Korpimaki 1988b, Oecologia 77. 278-285; Marti 1992, Condor 92:261-265) . The latter explanation, polygyny, would be an usual breeding behavior in the eagle-owl (Dalbeck et al. 1998, Vdgelwelt 119:331-344). Neither the pair reported by Martinez et al. (2003), nor the pair reported here were marked, so it was not possible to conclude if a lone pair was involved, or if a replacement, or if two females were involved in these cases of double-brooding. Nevertheless, all proposed explanations are likely related to the effects of high prey availability on the reproductive behavior of eagle-owls, which can reduce repro- ductive costs and lead to multiple breeding attempts. Such conditions in Spain seem to be infrequent, especially after the recent population crash of rabbits (Villafuerte et al. 1995, Mammalia 59:651-659; Martinez and Calvo 2001, J Raptor Res. 35:259-262; Martinez and Zuberogoitia 2001,/. Ornithol. 142:204-211). However, intensive research m high-prey situations, such as reported here may provide further examples of double brooding that could be more common than previously thought (Marks and Perkins 1999; Mahony et al. 2001). Consejeria de Agricultura y Medio Ambiente de Castilla-La Mancha provided the permits for monitoring eagle-owl nests. I wish to thank Mario Diaz for idiomatic and editorial advice and Jose Arcadio Calvo for helping during field work. — Joaquin Ortego (e-mail address: joaquinortegolO@latinmail.com), Departamento de Ciencias Ambientales, Facultad de Ciencias del Medio Ambiente, Universidad de Castilla-La Mancha, Avda, Carlos III s/n, 45071 Toledo, Spain. Received 4 February 2004; accepted 9 August 2004 Associate Editor: Juan Jose Negro J Raptor Res. 38(4):379-380 © 2004 The Raptor Research Foundation, Inc. Insect Hawking Observed in the Long-eared Owl {Asio otus) The Long-eared Owl {Asio otus) has been described as a specialist on a relatively narrow range of species of small mammals (Errington 1932, Condor 34:176-186; Craighead and Craighead 1979, Hawks, owls, and wildlife. Stackpole Co., Harrisburg, PA U.S.A; Marks and Marks 1981, Murrelet 62:80-82), and highly dependent on Microtus spp in many parts of North America and Europe (Marks 1984, Can. J. Zool. 62:1528—1533; Marks and Marti 1984, Ornis Scand. 15:135-143). Asio otus has also been found to shift dietary preference seasonally among different Microtus spp. in southern Sweden and among other small mammals in central Slovenia (Nilsson 1981, Ornis Scand. 12:216-223, Tome 2003, Ornis Fenn. 80:63-70). Invertebrates are a minor component of this species’ diet (0.5-0. 2% by number, <0.1% by mass; Marti 1974, Condor 76:45-61; Marti 1976, Condor 78:331-336; Tome 1994, J. Raptor Res. 28:253-258; Alivizatos and Goutner 1999,/. Raptor Res. 33:160-163) as are larger prey, such as juvenile (100-150 g) lagomorphs (0.75% by number, 2.5% by biomass, Marks 1984). Foraging behavior among Long-eared Owls is less understood than diet. The long-pointed wings and relatively low- wing loading of Long-eared Owls suggests the ability to hunt aerially, which has been observed in the form of quartering the ground for prey. Such adaptations are similar to Caprimulgids such as the Common Nighthawk (Chordeiles minor) which “hawk” prey aerially (catching prey on the wing; Poulin et al. 1996, Common Nighthawk {Chordeiles minor). In K. Poole and F. Gill [Eds.], The birds of North America, No. 213. The Birds of North America, Inc., Philadelphia, PA U.S.A.). In the Long-eared Owl, hawking behavior has never been documented (Marks et al 1994, Long-eared Owl {Asio otus). In A. Poole and F. Gill [Eds.], The birds of North America, No. 133. The Birds of North America, Inc., Philadelphia, PA U.S.A.) . There are very few published observations of Long-eared Owl foraging 380 Letters VoL. 38, No. 4 behavior, as this species is strictly nocturnal and difficult to observe. Glue and Hammond (1974, Br. Birds 67:361- 369) report Long-eared Owls “hovering” seconds before making a kill of a small mammal, but not otherwise. During nocturnal owl and bat surveys, we were frequently able to observe the behavior of several owl species. Here, we report observations of a hovering/hawking approach to aerial feeding by a Long-eared Owl, not previously reported in this species. Observations took place in the boreal forest of northern Ontario, Canada, south of the municipality of Ear Falls. The topography of the area is highly variable, with many lakes, and is dominated by stands of black spruce {Picea mariana) and to a lesser extent jack pine {Firms banksiana ) . On 25 June 2001, between 2220-2240 H, we first observed a Long-eared Owl perched on an aspen tree (Populus tremuloides) on the roadside. We confirmed the owl’s identification with a flashlight and a pair of binoculars. We were able to approach the bird three times to within 10 m as it perched on various trees. As we tried to find the bird a fourth time, it flew out from the side of the road and began to hover, slowly sweeping back and forth across the road ca 2 m off the ground within 5 m of our vehicle. In the headlights, we were able to observe the owl as it “hawked” moths over a large water puddle in the middle of the road. The moths were large enough to be clearly visible (ca 5. 7-6. 3 cm wingspan), and were later confirmed to be moths of the genus Actius or Smerinthus (Ross 1873, The butterflies and moths of Canada. Rowsell and Hutchison, Toronto, Canada), which had previously been observed in the area. While we watched, the owl captured at least three moths, which were apparently consumed whole. The owl then flew back into the woods in the direction from which it came, and was not seen again that night. Comments on this observation from M.C. Drever and T.D. Nudds were greatly appreciated. We wish to thank the Sustainable Forest Management Network of Centres of Excellence and The University of Guelph for funding, and for cooperation from Weyerhaeuser Inc., all of whom contributed to our presence in the field during the summer of 2001. — Darren J.H. Sleep (e-mail address: dsleep@uoguelph.ca) and Rowan D.H. Barrett, Department of Organ- ismal Biology, Ecology and Evolution, University of Guelph, Guelph, Ontario NIG 2W1, Canada. Received 27 December 2003; accepted 19 July 2004 Associate Editor: Ian G. Warkentin J Raptor Res. 38(4);380— 381 © 2004 The Raptor Research Foundation, Inc. Osprey Scavenges Common Murre Carcass in Coastal Washington Ospreys (Pandion haliaetus) feed almost exclusively on fish (Poole et al. 2002, In A. Poole and F. Gill [Eds.], The birds of North America, No. 683. The Birds of North America, Inc., Philadelphia, PA U.S.A.). They rarely capture non-fish items or scavenge non-fish carcasses. Poole et al. (2002) provided no records of Ospreys scavenging bird carcasses. On 9 September 2002, I observed an Osprey in immature plumage scavenging a Common Murre {Uria aalge) carcass on northern Grayland Beach, Grays Harbor County, WA. The carcass was one of >15 on the beach during my visit. Grayland Beach is a relatively flat, sandy beach situated between the mouths of Grays Harbor and Willapa Bay on Washington’s outer coast. At 1304 H, I saw an Osprey on the beach; it faced south and used its bill to twice tear at the flesh of a carcass that I later identified as a Common Murre. The Osprey then turned, apparently saw me (ca. 100 m away), and flew south and out of view. I approached the carcass, which lay on its back, and noted the pectoralis muscles were exposed and had been partially removed. I did not see the Osprey again, but at 1314 H saw another Osprey fly over heading south above the beach. It is possible that the Osprey I observed was merely investigating an unusual item, a behavior that has been noted in post-fledging juveniles (L. Gilson pers. comm.), and that scavenging was not its initial intent. However, it seems reasonable that most scavenging is preceded by investigation, particularly in juveniles. Consequently, regardless of the original intent, the outcome was that the Osprey extracted flesh from the carcass of a dead bird. Although Ospreys rarely capture or consume non-fish prey, Wiley and Lohrer (1973, Wilson Bull. 85:468-470) identified a number of factors to explain the occasional use of non-fish food. Among these factors were: (1) the presence of easily-captured prey and (2) an abundant alternate food source. The coastal beaches of Washington often have abundant dead birds (e.g.. Northern Fulmar [Fulmarus glacialis], scoters [Melanitta spp.], gulls [Larus December 2004 Letters 381 spp.], and Common Murres) that wash ashore and are deposited at or above the high-tide line. These carcasses are occasionally scavenged by Peregrine Falcons {Falco peregrinus; Buchanan 1991, Northwest. Nat. 72:28-29), Bald Eagles {Haliaeetus leucocephalus) , and Northern Harriers {Circus cyaneus; ]. Buchanan unpubl. data). The presence of nu- merous carcasses on the beach during my visit represented an easily accessible and abundant source of food, two of the conditions proposed to explain use of non-fish food by Ospreys (Wiley and Lohrer 1973). Ospreys that use coastal habitats, especially during migration, have access to an easily obtained food source in some areas. I suggest that scavenging, although apparently rare, may be more likely in this coastal habitat than in other areas. I thank Tracy Fleming for providing literature citations. Lauren Gilson, Jim Belthoff, and an anonymous reviewer provided comments that improved the manuscript. — Joseph B. Buchanan (e-mail address: huchajbb@dfw.wa.gov), Washington Department of Fish and Wildlife, 600 Capitol Way North, Olympia, WA 98501 U.S.A. Received 24 February 2004; accepted 2 September 2004 Associate Editor; James R. Belthoff J. Raptor Res. 38(4) :381-382 © 2004 The Raptor Research Foundation, Inc. How Long is Too Long? A Case of Fostering Nestling Bonelli’s Eagles {Hieraaetus fasciatus) After monitoring eight nests of Bonelli’s Eagle {Hieraaetus fasciatus) for more than 11 yr in the state of Maharashtra, India, we recorded two incidents wherein eaglets were either found to have fallen out of nests due to human distur- bance or removed by local children. In the two incidents, we replaced the previously-removed eaglets into the nest immediately upon discovery and then verified continued parental care. To help ensure the continued survival of raptors in the wild, a wide range of techniques have been developed and applied to maximize the survival of the brood (see Cade et al. 1988, Peregrine Falcon populations: their management and recovery. The Peregrine Fund, Boise, ID U.S.A.) including “add-on” techniques (an abandoned nestling is introduced into a wild brood of similar age); guarding of nests during the breeding season; relocation of nests away from sources of mortality; presentation of alternative, artificial nest platforms or sites; translocation; hacking in natural or artificial nests; and inter- or intra-specific fostering by parents with young of the same age (Allen 1982, Pages 5- 19 in T.N. Ingram [Ed.], Proceedings of the Bald Eagle conference on Bald Eagle restoration. USDI Fish and Wildlife Service, Rochester, NYU.S.A.). All of these techniques are very costly, do not always ensure success, and many are employed in human-modified environments. Here, we present a method that has not been previously employed in a natural environment with Bonelli’s Eagles, though similar experiments have been conducted for the Spanish Im- perial Eagle {Aquila adalberti; Gonzalez et al. 1986,/. Raptor Res. 20:77-78; Ferrer 1993,/. Ornithol. 134:335-337). Our experiment was conducted unintentionally on 14 March 2003 when we discovered that local children had removed an eaglet from a Bonelli’s Eagle nest in Jejuri, Pune district (18°3TN, 73°55'E), India. The eaglet was 40- 42 d old and it was returned to the nest. We noted that there were no green branches on the nest, which was unusual because in previous seasons parents were observed to layer the nest with new leaves and branches almost on a daily basis. Moreover, the parents continuously evicted the eaglet over the next 3 d by pushing it out of the nest with their wings. Therefore, we decided to foster the eaglet into the nest of another pair. This foster pair nested ca. 250 km away at Pawangad, Kolhapur district (16°42'N, 74°16'E), and on 5 March had two chicks in the nest. Upon arrival on 18 March, we found the nest empty and the eagle family soaring over the nest tree. Nevertheless, we decided to place the eaglet into the vacant nest. Based on previous observations, we knew that eaglets and parents roosted at the nest for at least 2 wk after the young had fledged and that the young at this stage were still dependent on their parents for food (pers. obs.). After almost 2 hr the family was observed to land on the branches adjacent to the nest to roost for the night. Neither the young nor the parents displayed any signs of aggression toward the foster eaglet. Initially, the family ignored the eaglet but in the evening one of the adults dropped an un-plucked chicken by the eaglet and after observing the inability of the foster eaglet to feed upon the chicken, one of the fledglings, which were ca. 60 d of age, then plucked the chicken and consumed a small portion of it. The foster eaglet observed the actions of the fledgling and imitated its movements of plucking the chicken and swallowing, and when the fledgling 382 Letters VoL. 38, No. 4 left the nest, the translocated eaglet immediately grasped the partially-consumed chicken and proceeded to dismem- ber and consume it. For the next week (20-26 March), the parents and the fledged young alternated in bringing prey to the foster nestling. The eaglet was not observed to have any further prey handling difficulties and between feedings stood at the nest and flapped its wings vigorously. The foster chick eventually fledged on 26 March 2003. We conclude from this episode that at least some parent Bonelli’s Eagles will accept a foster young in an empty nest, but only if egg or chick loss happened a few hours earlier. We have done such translocations successfully in the past, but within 24 hr of the nest becoming empty (Pande 2003, Newsletter for Birdwatchers 43:31-33). Similarly, a 24- d-old chick of Changeable Hawk-Eagle {Spizaetus cirrhatus) has been fostered successfully in a nest of a similar species, that had been empty for just over 24 hr in India (Naoroji 1984,/. Bombay Nat. Hist. Soc. 82:278-308). We assume that the finding of translocating an eaglet into a foster family during the post-fledging dependence period may also be applicable to other raptor species, provided that siblicide is not common in the host species. This technique seems mainly suitable as an emergency action to save individual birds. We thank Mr. J.C. Daniel, and the other authorities of the Bombay Natural History Society in Mumbai, and Mr. Prakash Thosare, and the staff of the Forest Department in Pune for their cooperation and permission for the above Drs. T.J. Cade, M. Ferrer, Javier Balbontin, Fabrizio Sergio, and an anonymous referee greatly improved an earlier version of this manuscript. — Satish A. Pande, Amit P. Pawashe, Banda Pednekar, ELA Foundation, C-9 Bhosale Park, Sahakarnagar 2, Pxme 411009, India; An il Mahabal, Zoological Survey of India, WRS, Pune, India; Reuven Yosef (corresponding author’s e-mail address: ryosef@eilatcity.co.il). International Birding and Research Centre in Eilat, P.O. Box 774, Eilat 88000, Israel. Received 6 April 2004; accepted 10 August 2004 Associate Editor: Fabrizio Sergio J. Raptor Res. 38(4):383— 386 © 2004 The Raptor Research Foundation, Inc. Journal of Raptor Research INFORMATION FOR CONTRIBUTORS The Journal of Raptor Research (JRR) publishes original research reports and review articles about the biology of diurnal and nocturnal birds of prey. All submissions must be in English, but contribu- tions from anywhere in the world are welcome. Manuscripts are considered with the understand- ing that they have not been published, submitted or accepted for publication elsewhere. Manuscripts are subjected to peer review for evaluation of their significance and soundness, and edited to improve communication between authors and readers. De- cisions of the editor are final. Material is published as feature articles, short communications (usually not longer than four printed pages) , and letters (see recent issue of the JRR for examples) . Submissions that adhere closely to the JRR’s format greatly enhance the efficiency and cost of the editorial and publishing processes. Author’s efforts in this regard are deeply appreci- ated by the editorial staff. When submitting scholarly papers, send the orig- inal and three copies, a completed checklist (see below), and a cover letter that includes: (1) a state- ment that the data in the manuscript have not been published or accepted for publication in the same form, and have not been submitted simulta- neously elsewhere, (2) the name and address of the corresponding author (in multiauthored pa- pers) including any temporary addresses where that author will be during the review process (also the phone number and, if possible, a FAX number and e-mail address of the corresponding author), and (3) if applicable, any special instructions. Au- thors may also suggest potential reviewers. If the manuscript submitted was produced on a word processor, also send a diskette (3 1/2") or CD containing a single file that is identical with the printed copy. The electronic copy should be sup- plied as an IBM-compatible text file (Word or WordPerfect) . Optional electronic submissions are encouraged (see General Instructions below) . Manuscripts are accepted upon the condition that the revision must be returned to the editor within 60 days. Manuscripts held longer will lose their priority and may be treated as new submis- sions. The editor should be notified if extenuating circumstances prevent a timely return of the man- uscript. Authors will receive proofs of their articles prior to publication. Proofs must be read carefully to cor- rect any printer errors and returned by the fastest mail within two days of receipt TO THE EDITOR. Changes in typeset text are expensive and authors making changes, not due to printer error, will be billed for the costs ($3.50 US per change). A re- print order will accompany page proofs to enable authors to buy reprints. Costs of reprints are the author’s responsibility and payment for reprints or- dered must accompany the order form. Both must be sent TO THE EDITOR. Publication is expensive and member dues do not cover the entire cost of producing the JRR. Hence, the Raptor Research Foundation, Inc. ex- pects that authors defray the high costs of publi- cation through payment of page costs (currently $115.00 U.S. per page). Authors who are not as- sociated with a research institution or simply do not have access to such grants may request the page charges be waived. Such a request can only be approved if the author is a member of the Foun- dation and the article is short. Payments of amounts less than the full page charges will be ac- cepted. Authors of long manuscripts are expected to pay publishing costs. It is unlikely that articles longer than 10 printed pages or 18 typewritten pages including tables and illustrations can be pub- lished without full payment. Authors employed by government agencies, universities, or firms that will meet reprint and page charges may forward a statement to the editor indicating intent to pay. Upon receipt of such a statement, reprints will be mailed to the author and the agency will be billed with the understanding that payment will be made within 30 days. All checks should be made payable to the Raptor Research Foundation, Inc. All per- sonal payments toward publication costs are tax- deductible in the United States. 383 384 Information for Contributors VoL. 38, No. 4 Journal of Raptor Research CHECKLIST FOR PREPARATION OF MANUSCRIPTS {check items and submit with manuscript) I. General Instructions (Consult recent issues for additional guidance on format) I I Type manuscripts on one side of either 216 X 278 mm (8.5 X 11") or standard international size (210 X 297 mm) good quality paper (do not use erasable or lightweight paper) . Word-processor-generated manuscripts must be done with a letter-quality or neai- letter-quality printer. DOUBLE SPACE THROUGH- OUT including title, text, tables, figure legends, and literature cited. □ Electronic submission. Submit e-mail message with two attached file copies of manuscript: (1) pdf file (must include figures) and (2) word processing file (Word Perfect or MS Word files accepted). Follow same format guidelines as for standard mail submis- sion. Letter of transmittal should be included in the e-mail message. E-mail to: jrr@astate.edu □ Give the scientific name at the first mention of a spe- cies, both in the abstract and in the article. Scientific names of birds should follow the usage of the AOU Check-list of Nm th American Birds (7th. Ed. 1998 and subsequent supplements in the Auk) or an authorita- tive source corresponding to other geographic re- gions. Do not give subspecific identification unless it is pertinent. Capitalize first letter of words in com- plete common names for birds. Use lower case for all other common names. □ Use American spelling and Webster’s Tenth New Colle- giate Dictionary (1996, Merriam-Webster, Inc.) as a spelling authority. □ Leave at least a 25 mm (1") margin on all sides. Avoid hyphens or dashes at ends of lines; do not divide a word at the end of a line. □ Use a nonproportional font of at least elite size (4.7 characters/cm = 12 characters/inch) or 12 point, preferably Courier. DO NOT USE RIGHT JUSTIFI- CATION-LEAVE RIGHT MARGIN RAGGED. □ Use italic type for addresses, scientific names, journal names, and third level headings. D Type last name(s) of author (s) and page number in upper right-hand corner of page 2 and all following pages. □ Cite each figure and table in the text. Do not repeat material in two forms (i.e., in text and table, or table and figure). Organize text, as far as possible, so that tables and figures are cited in numerical order. □ Use “Figure” only to start a sentence; otherwise “Fig.” if singular, “Figs.” if plural (e.g., Fig. 1; Figs. 2, 3; Figs. 4—6). □ Use metric units throughout. □ Use these abbreviations without spelling out: hr, min, sec, yr, mo, wk, d, km, cm, mm; designate temperature as 32“C. n Use “continental” dating (e.g., 10 July 1993, 1-3 June, 1 1 May to 1 1 June) . □ Use 24-hour clock (e.g., 0800 H, 1345-1400 H) □ Write out numbers one to nine unless a measurement (e.g., four birds, 3 km, 40 sites, 6 yr). Use 1000 and 10 000; 0.15 instead of .15; % instead of percent. □ Each reference cited in text must be listed in the Lit- erature Cited section, and vice versa. Double check the accuracy of all entries — THE EDITORIAL STAFF CANNOT DO THIS FOR YOU. □ Literature citations in the text are as follows: a. One author-Jones (1993) or (Jones 1993) b. Two authors— Smith and Jones (1991) or (Smith and Jones 1991) c. Three or more authors-Hernandez et al. (1990) or (Hernandez et al. 1990) d. Manuscripts accepted for publication but not yet published-Howard (in press) or (Howard in press) e. Unpublished materials-K. Jacobson (unpubl. data); (K. Jacobson pers. comm.); or K. Jacobson (pers. comm.)— do not place in the Literature Cited section. f. When citing several references within parentheses, separate with commas and put in chronological or- der, oldest first) . g. For manuscripts submitted as letters, place cita- tions in text in abbreviated form, e.g., (I.C. Birds 1993,/. Raptor Res. 27:45-50). □ Assemble manuscripts for regular articles in this or- der: (1) title page, (2) abstract page, (3) text, (4) ta- bles, (5) figure legends, (6) figures. DO NOT STA- PLE. n Avoid any unnecessary or special formatting. II. Title Page □ Place full title 6-8 lines below top of page in all capital letters. Below title, center author’s name(s) in all capital letters and address (es) followed by a running title (short title) not to exceed 30 characters. If the author(s) is/are currently at another location from where the work was done, use superscript number(s) fol- lowing author (s) name(s) to indicate current address in footnote at bottom of the page. In multiauthored papers, indicate the author re- sponsible for corre.spondence and requests for reprints. Give phone number and, if possible, FAX number and e-mail address of the corre- sponding author. III. Abstract/summary □ For regular articles, include an abstract of about 250 words in one paragraph that is completely December 2004 Information for Contributors 385 without reference to the text. Be concise, in- clude the paper’s purpose, but emphasize the results. Statements like “results will be dis- cussed” are not appropriate. The abstract will also be published in Spanish. Authors fluent in both languages are encouraged to include both versions, otherwise the JRR will provide the Spanish translation. □ Include five to seven key words for indexing af- ter the abstract. □ Short communications will be printed with a Spanish summary only. Authors must provide an English summary to be translated into Span- ish unless they are fluent in Spanish. □ Avoid citing references in the abstract. If they must be cited, include journal name, volume, pages, and year, all in parentheses. IV. Text □ Follow instructions in section I, □ Main headings are all capital letters and flush with left margin. □ Typical main headings for regular articles are; METHODS, RESULTS, and DISCUSSION. An introduction begins the text but does not have a heading. □ Put second-level headings in bold. Use normal indentation and capitalize first letter of each word in the second-level headline except prep- ositions and articles. □ Put third-level headings in italics. Capitalize first letter of first word only. □ Short communications and letters may or may not have headings within the text depending upon the need. V. Literature Cited □ Type references in large and small capital let- ters, including all authors’ names (e.g., Ne- meth, N.M. AND J.L. Morrison. 2002. Natal dis- persal of the Crested Caracara . . . etc. . . .). □ Do not put space between initials. D Verify all entries against original sources includ- ing diacritical marks and spelling in languages other than English. Capitalize all nouns in Ger- man. n Cite references in alphabetical order by first au- thor’s surname. References by a single author precede multiauthored works by the same se- nior author regardless of date. □ List works by the same author (s) chronologi- cally, beginning with the oldest. □ Use six hyphens when the author is the same as in the preceding citation. □ “In press” citations must have been accepted for publication and must include date, volume number, and the name of the journal or pub- lisher. □ Initials of second, third, and . . . authors pre- cede their surname. □ Abbreviate journal names according to the Se- rial Sources for the BIOSIS Data Base (published annually by the BioSciences Information Ser- vice). □ Do not list personal communications and un- published reports. VI. Tables (Tables are expensive to print-keep them to a min- imum and put each on a separate page— try to de- sign them to fit a single column.) □ Double space throughout. Assign each table an Arabic number followed by a period. □ Table titles must be complete sentences. □ Use same size of type as in text. □ Indicate footnotes by lowercase superscript let- ters. □ Do not use vertical lines. VTI. Figure Legends □ Print all figure legends on one page, double spaced. □ Number using Arabic numbers consecutively in the same order the figures appear in the text (i.e., Eigure L, Figure 2., etc.). VIII. Preparation of Illustrations (Illustrations are referred to as figures and include drawings, graphs, and black and white half-tones [photographs]. CONSULT THE EDITOR IN AD- VANCE ABOUT COLOR.) □ Use professional standards in preparing figures; their reproduction in the JRR is virtually iden- tical to what is submitted. Consult issues of JRR for examples and Steps Toward Better Scientific Il- lustrations (Allen Press, P.O. Box 368, Lawrence, KS 66044) for more information. □ Plan figures to fit proportions in the JRR, pref- erably for a single column-printed size is 72 mm for single column width, 148 mm for full page width or 195 mm for lengthwise figures. Figures should be submitted no smaller than the final size nor larger than twice the final size (on paper no larger than 216 X 278 mm (8.5 386 Information for Contributors VoL. 38, No. 4 X 11") or standard international (210 X 297 mm). □ Submit drawings as original art (undiluted In- dia ink on heavy-weight, smooth-finish drafting paper) or as photomechanical transfers (PMTs) . Submit graphs as mechanical drawings or as high-resolution laser prints. Typed or handwritten text or symbols are not acceptable. Add text and symbols with press-on symbols and letters or mechanical lettering. Review copies of figures can be photographic copies but must ap- proach the quality of the original. □ Figure text must be a plain (sans serif) typeface (e.g., Helvetica), not compressed, and large enough so that it will be as large as the text type (8-10 point) when in print. □ Photographs must be sharp, high-contrast, glossy prints approximately the size that they will appear in print. If several photographs are to be included in one figure, group them butt- ed together with no space between. D Use the same style of lettering and presentation for all figures. IX. What to Send □ Cover letter. □ Copy of this checklist completed. n Original and three copies of manuscript and il- lustrations. □ Diskette containing a text file of the manuscript text, tables, and figures (file in MS Word or WordPerfect) . □ Submit to: James C. Bednarz, Editor Journal of Raptor Research Department of Biological Sciences P.O. Box 599, 117 Caraway Road Arkansas State University State University, AR 72467 U.S.A. More information? Telephone: (870) 972-3082 FAX: (870) 972-2638 E-mail: jrr@astate.edu J. Raptor Res. 38(4):387-393 © 2004 The Raptor Research Foundation, Inc. Index to Volume 38 By Dana Ripper and Jennifer L. Norris The index includes references to general topics, common names, keywords, and authors. Reference is also made to book reviews and letters. Taxa other than raptors are included where referenced by authors. A Abundance, 91-95 Accipiter cooperii, 26-34, 118-124, 163-168, 250-255, 334— 342 gentilis, 243-249 nisus, 243—249 Striatus, 62-68, 69-77, 163-168, 250-255, 334-342 Aegyptus monachus, 361-366 Afianto, M. Yayat, see Gjershaug, Jan Ove Agostini, Nicolantonio, Charles Coleiro, and Michele Panuccio, Analysis of the autumn migration of ju- venile honey-buzzards {Pernis apivorus) across the central Mediterranean, 283-286 Agriculture, 152-157 ALAD, 118-124 Alaska, 81-85, 158-160, 275-277 interior, 9-18 Alcalde, Leandro and Sergio D. Rosset, Observation of the Chimango Caracara {Milvago chimango) feeding on common lesser toads {Bufo fernandezae) , 190-191 Aleutian archipelago, 81—85 Alivizatos, Haralambos, Dimitris Papandropoulos, and Stamatis Zogaris, Winter diet of the Greater Spotted Eagle {Aquila clanga) in the Amvrakikos Wetlands, Greece, 371—374 Alvarado, Sergio, see Pavez, Eduardo F. Amvrakikos, 371—374 Amputation, 85-88 Anthony, Robert G., see Forsman, Eric D. Anthony, Robert G., see Ricca, Mark A. Aquila chrysaetos, 192, 195-207, 270-275 clanga, 371—374 Arborimus longicaudus, 214-230 Arctic, 158-160 Argentina, 1-8, 152-157 Arizona, 26-34, 238-242 Arnello, Alejandra, see Pavez, Eduardo F. Asio otus, 379-380 Asturina nitida, 238-242 Athene cunicularia hypugaea, 91-95 noctua, 35-46 Austing, G. Ronald, see Dykstra, Cheryl R. Avian energetics, 163-168 B Barrett, Rowan D.H., see Sleep, Darren J.H. Bass, Orin L., see Mealey, Brian Bakaloudis, Dimitris, see Vlachos, Christos Banding, 304—311 Behavior, 295-303 Bibles, Brent D. and R. William Mannan, Productivity and nest-site characteristics of Gray Hawks in south- ern Arizona, 238-242 Bloise, Carolina, see Tome, Ricardo Bossart, Gregory D., see Mealey, Brian Brandes, David and Daniel W. Orabalski, Modeling rap- tor migration pathways using a fluid-flow analogy, 195-207 Brazil, southeastern, 191 Breeding, 186-189 biology, 1-8, 263-269 success, 161-163, 312-319, 350-356 Breeding Bird Atlas, 91-95 Survey, 91-95 Breeding conditions, 361-366 Broerman, Fred J., see Harwood, Christopher M. Brood size, 361-366 Brown, Jessi L., William R. Heinrich, J. Peter Jenny, and Brian D. Mutch, Development of hunting behavior in hacked Aplomado Falcons, 148-152 Bubo bubo, 375-377, 378-379 scandiacus, 275-277 Buchanan, Joseph B., Osprey scavenges Common Murre carcass in coastal Washington, 380-381 Buchanan, Joseph B., Tracy L. Fleming, and Larry L. Ir- win, A comparison of Barred and Spotted owl nest- site characteristics in the eastern Cascade Moun- tains, Washington, 231-237 Bufo fernandezae, 190-191 Butastur indicus, 263—269 Buteo albigula, 1—8, 186-189 buteo, 243-249 jamaicensis, 19-25, 181—186, 334—342 lagopus, 181-186 lineatus, 290, 304-311, 312-319 swainsoni, 357—361 Buzzard, Common, 243-249 Grey-faced, 263-269 Buzzard-Eagle, Black-chested, 101-102 Byrd, Mitchell A., see Watts, Bryan D. 387 388 Index to Volume 38 VoL. 38, No. 4 c Cage-flight exercise, 125-132 Cangarato, Rogerio, see Palma, Luis Cape Verde Islands, 141-147 Caracara, Chimango, 152-157, 190-191 Crested, 152-157 Mountain, 290-292 Caracara plancus, 152-157 Cathartes aura, 288—289 Creance-flight exercise, 125-132 Chandler, Richard B., Allan M. Strong, and Carlin C. Kaufman, Elevated lead levels in urban House Spar- rows; a threat to Sharp-shinned Hawks and Merlins?, 62-68 Chatzinikos, Evangelos, see Vlachos, Christos Chen canagica, 81—85 Chesapeake Bay, 47-54 Chihuahua, Mexico, 107-117 Desert, 107-117 Chile, central, 186-189 Christmas Bird Count, 91-95 Circus cyaneus, 133-140, 243-249 macrosceles, 256-262 pygargus, 133-140 Clarke, Michele L., see Tingay, Ruth E. Cloud cover, 208-213 Clutch size, 263-269 Coleiro, Charles, see Agostini, Nicolantonio Coloniality, 350-356 Communal roosts, 278-282 Competition, 88-91, 178-181 Contaminants, 69-77 Cooperative hunting, 148-152 Coragyps atratus, 288-289 Corbacho, Casimiro, see Villegas, Auxiliadora Corona, Rodrigo Sierra, see Lopez Gonzalez, Carlos A. Costillo, Emilio, see Villegas, Auxiliadora D Daniel, F. Bernard, see Dykstra, Cheryl R. de Frutos, Angel, see Olea, Pedro P. De Lucca, Eduardo R. see Saggese, Miguel D. de Roland, Lily-Arison Rene, Jeanneney Rabearivony, Ig- nace Randriamanga, and Russell Thorstrom, Nesting biology and diet of the Madagascar Harrier {Circus macrosceles) in Ambohitantely Special Reserve, Mad- agascar, 256-262 Deciduous forests, 186-189 del mar Delgado, Maria and Vincenzo Peiateriani, Gen- der determination of Eurasian Eagle-owls {Bubo bubo) by morphology, 375-377 DeLong, John P, and Stephen W. Hoffman, Fat stores of migrant Sharp-shinned and Cooper’s hawks in New Mexico, 163—168 Deng, Wen-Hong, Wei Gao, and Jiang Zhao, Breeding biology of the Grey-faced Buzzard {Butastur indicus) in northeastern China, 263-269 Development, 263-269 Discriminant function analysis, 375-377 Diet, 69-77, 178-181, 214-230, 256-262, 371-374 Dietary shift, 174—177 Dispersal, 26-34 Distribution, 107-117, 181-186 Disturbance response, 295—303 Dominance hierarchy, 85-88 Dominguez, Jesus, see Tapia, Luis Double brooding, 378-379 Dykstra, Cheryl R., Jeffery L. Hays, Melinda M. Simon, John B. Holt, Jr., G. Ronald Austing, and F. Bernard Daniel, Dispersal and mortality of Red-shouldered Hawks banded in Ohio, 304-311 E Eakle, Wade L., Carl Millier, Pierre Mineau, and Janos Vilagosi, An example of cooperative hunting by Sa- ker Falcons in Hungary, 292-293 Eagle, Bald, 81-85, 96-100, 295-303 Bonelli’s, 381-382 Booted, 168-174 Golden, 192, 195-207, 270-275 Greater Spotted, 371—374 Eagle-Owl, Eurasian, 375-377, 378-379 Ecomorphology, 334-342 Elanus leucurus, 178—181 Ellis, David H. and Marc Kery, Variable retention times for retrices at different loci in a Golden Eagle, 270- 275 Engelmann, Mathias, see Greene, Dana M. Estes, Wendy A., see Mannan, R. William Everglades, 96-100 Exercise cage-flight, 125-132 creance-flight, 1 25-1 32 F Falco cherrug, 292-293 columbarius, 62-68 eleonorae, 320-325 femoralis, 107-117, 148-152 naumanni, 161-163, 278-282 peregrinus, 9-18, 158-160 sparverius, 250-255, 334-342 subbuteo, 287-288 Falcon, Aplomado, 107-117, 148-152 Eleonora’s, 292-293, 320-325 Peregrine, 9-18, 158-160 Saker, 292-293 Sooty, 292-293 False eyes, 287 Fat scores, 163-168 stores, 163-168 Ferguson, Howard L., Relative abundance and diversity December 2004 Index to Volume 38 389 of winter raptors in Spokane County, eastern Wash- ington, 181-186 Ferreira, Joao, see Palma, Luis Finland, 243-249 Fire perturbations, 174—177 Fish-Eagle, Madagascar, 85-88 Fish farms, 326-333 predation, 326—333 Fitness, male, 320—325 muscle, 125-132 Fish-Owl, Tawny, 326-333 Fledgling sex ratio, 361—366 Fledglings, 161-163, 263-269 Fleming, Tracy L., see Buchanan, Joseph B. Florida, 55-61, 96-100 Bay, 96-100 Food habits, 81-85 Forearm, 375-377 Foraging behavior, 148-152 theory, 9-18 Forsman, Eric D., Robert G. Anthony, E. Charles Meslow, and Cynthia J. Zabel, Diets and foraging behavior of northern Spotted Owls in Oregon, 214—230 Eostering nestlings, 381-382 Fundy National Park, 69-77 G Gabella, Juan P., see Pavez, Eduardo F. Gao, Wei, see Deng, Wen-Hong Garcia Dios, Ignacio S., Spanish ringing and recovery re- cords of Booted Eagle {Hieraaetus pennatus) , 168—174 Garcia-Ripolles, Clara, see Lopez-Lopez, Pascual Gelain, Mariano, see Trejo, Ana Gender determination, 357-361, 375-377 Genetic variation, 250-255 Geranoaetus melanoleucus, 101-102 Giving-up time, 19-25 Gjershaug, Jan Ove, Nils R0v, Torgeir Nygard, Dewi M. Prawiradilaga, M. Yayat Afianto, Hapsoro, and Adam Supriatna, Home range size of the Javan Hawk-Eagle {Spizaetus bartelsi) estimated from direct observations and radiotelemetry, 343-349 Glaucomys sabnnus, 214—230 Goose, Emperor, 81-85 Goldstein, Michael I. and Toby J. Hibbitts, Summer road- side raptor surveys in the western pampas of Argen- tina, 152—157 Gonzalez, Benito A., see Pavez, Eduardo F. Gonzalez, Christian, see Pavez, Eduardo E. Goshawk, Northern, 243-249 Grande, Juan Manuel, see Negro, Juan Jose Greene, Dana M., Mathias Engelmann, and Todd R. Steck, An assessment of cage flight as an exercise method for raptors, 125-132 Granzinolli, Marco Antonio Monteiro, see Mottajunior, Jose Carlos Graves, Gary R., Commensal foraging by a Red-sboul- dered Hawk {Buteo lineatus) with Wild Turkeys {Me- leagris gallopavo) , 290 Greece, 371-374 Griffon, Eurasian, 350-356 Gross, Howard P., see McBride, Tobias J. Grouse, 243-249 Guevara, Eduardo Ponce, see Lopez Gonzalez, Carlos A. Guyer, Craig, see Ress, Sara Gyps fulvus, 350—356 H Habitat, 158-160, 231-237 selection, 26-34, 35-46 use, 133-140 Hacking, 148-152 Haliaeetus leucocephalus, 81—85, 96—100, 295—303 vociferoides, 85-88 Hapsoro, see Gjershaug, Jan Ove Harrier, Hen, 133-140 Madagascar, 256-262 Montagu’s, 133-140 Northern, 243-249 Harwood, Christopher M., Brian J. McCaffery, Ered J Broerman, and Paul A. Liedberg, A local concentra- tion of Snowy Owls in the Yukon-Kuskokwim Delta in Summer 2000, 275-277 Hawk, Cooper’s, 26-34, 118-124, 163-168, 250-255, 334- 342 Gray, 238-242 Red-shouldered, 290, 304-311, 312-319 Red-tailed, 19-25, 181-186, 334-342 Rough-legged, 181-186 Sharp-shinned, 62-68, 69-77, 163-168, 250-255, 334— 342 Swainson’s, 357-361 White-tailed, 191 White-throated, 1-8, 186-189 Hawk-Eagle, Javan, 343-349 Ornate, 193-194 Hays, Jeffery L., see Dykstra, Cheryl R. Health assessment, 96-100 Heinrich, William R., see Brown, Jessi L. Hematology, 96-100 Hibbitts, TobyJ., see Goldstein, Michael I, Hieraaetus fasciatus, 381-382 pennatus, 168-174 Hobby, Eurasian, 287-288 Hoffman, Stephen W., A review of raptors of eastern North America, by Brian K. Wheeler, 2003, 103-104 Hoffman, Stephen W., see DeLong, John P. Holt, Jr., John B., see Dykstra, Cheryl R. Home range, 238-242, 343-349 size, 26-34 Honey-buzzard, European, 283-286 Oriental, 367-371 390 Index to Volume 38 VoL. 38, No. 4 Hooper, Michael J., see McBride, Tobias J. Huang, Kuang-Mng, Yao-Sung Lin, and Lucia Liu Sever- inghaus, Nest provisioning of the Oriental Honey- buzzard {Pernis ptilorhynchus) in northern Taiwan, 367-371 Huhtala, Kauko, see Reif, Vitali Human activity, 295-303 Hungary, 292-293 Hunting, 19-25 behavior, 148-152 cooperative, 292-293 I Indonesia, 343-349 Injury, 77-81 Insect hawking, 379-380 Iverson, Wayne R, Reproductive success of Spotted Owls sympatric with Barred Owls in western Washington, 88-91 Irwin, Larry L, see Buchanan, Joseph B. J Jenny, J. Peter, see Brown, Jessi L. K Kalavah, Loukman, see Tingay, Ruth E. Kaufman, Carlin C., see Chandler, Richard B. Kery, Marc, see Ellis, David H. Kestrel, American, 250-255, 334-342 Lesser, I61-I63, 278-282 Ketupa flavipes, 326-333 Kingfisher, Ringed, 191 Kirk, David Anthony, see Woodley, Stephen J. Kite, White-tailed, 178-181 Korpiraaki, Erkki, see Tome, Ricardo L Lead, 62-68, 118-124 Lehman, Chad P. and Dan J. Thompson, Golden Eagle {Aquila chrysaetos) predation attempts on Merriam’s Turkeys (Meleagris gallopavo merriami) in the south- ern Black Hills, South Dakota, 192 Leyhe, Jennifer E. and Gary Ritchison, Perch sites and hunting behavior of Red-tailed Hawks {Buteo jamai- censis), 19-25 Leveau, Carlos M., see Leveau, Lucas M. Leveau, Lucas M., Carlos M. Leveau, and Ulyses FJ. Par- dihas, Trophic relationships between White-tailed Kites {Elanus leucurus) and Barn Owls {Tyto alba) in southern Buenos Aires Province, Argentina, 178-181 Liedberg, Paul A., see Harwood, Christopher M. Lin, Yao-Sung, see Huang, Kuang-Wng I,ongevity, 85-88, 168-174 Lopez Gonzalez, Carlos A., Eduardo Ponce Guevara, Kar- la Pelz Serrano, Hugo Luna Soria, and Rodrigo Si- erra Corona, A record of the Ornate Hawk-Eagle {Spizaetus ornatus) in Nayarit, Mexico, 193-194 Lopez-Lopez, Pascual, Clara Garcia-Ripolles, and Jose Verdejo, Population status and reproductive perfor- mance of Eurasian Griffons ( Gyps fulvus) in eastern Spain, 350-356 Lunar phase, 208-213 M Madagascar, 256-262 Mahabal, Anil, see Pande, Satish A. Male fitness, 320-325 Mannan, R. William, Wendy A. Estes, and William J. Mat- ter, Movements and survival of fledgling Cooper’s Hawks in an urban environment, 26-34 Mannan, R. William, see Bibles, Brent D. Martell, Mark S., Michael A. McMillian, Mathew J. Solen- sky, and Brian K. Mealey, Partial migration and win- tering use of Florida by Ospreys, 55-61 Martin, Elwood (Woody) M., Decreases in a population of Red-shouldered Hawks nesting in central Mary- land, 312-319 Maryland, central, 312-319 Matter, William J., see Mannan, R. William McBride, To- bias J., Jeff P. Smith, Howard P. Gross, and Michael J. Hooper, Blood-lead and ALAD-activity levels of Cooper’s Hawks {Accipiter cooperii) migrating through the southern Rocky Mountains, 118-124 McCaffery, Brian J., see Harwood, Christopher M. McIntyre, Nancy E., Historical and current status of breeding and wintering western Burrowing Owls {Athene cunicularia hypugaea) in Texas, 91-95 McMillian, Michael A., see Martell, Mark S. Mealey, Brian, Greta M. Parks, Carlos M. Pages, Brian A. Millsap, Orin L. Bass, and Gregory D. Bossart, Se- rum chemistry values for nestling Bald Eagles {Hal- iaeetus leucocephalus) in Florida Bay, Everglades Na- tional Park, 96—100 Mealey, Brian K., see Martell, Mark S. Mediterranean region, 35-46 Meleagris gallopavo, 290 Merlin, 62-68 Meslow, E. Charles, see Forsman, Eric D. Mesquite, 152-157 Mexico, Chihuahua, 107-117 Nayarit, 193-194 Meyer, Kenneth D., see Woodley, Stephen J. Migration, 55-61, 62-68, 118-124, 163-168, 195-207, 283-286, 334-342 Millier, Carl, see Eakle, Wade L. Millsap, Brian A., see Mealey, Brian Milvago chimango, 152-157, 190-191 Mineau, Pierre, see Eakle, Wade L. Mitochondrial DNA, 250-255 Modeling, 133-140, 195-207 Molecular sexing, 320-325, 357-361 December 2004 Index to Voi.ume 38 391 Molt, 270-275 Montoya, Angel B., see Young, Kendal E. Moran, Ricardo, see Villegas, Auxiliadora Morphometric measures, 357-361 Mortality, 168—174, 326—333 factor, 77—81 Motta-Junior, Jose Carlos and Marco Antonio Monteiro Granzinolli, Consumption of a Ringed Kingfisher {Megaceryle torquata) by a White-tailed Hawk {Buteo alhicaudatus) in southeastern Brazil, 191 Mutch, Brian D., see Brown, Jessi L. Murre, Common, 380—381 N Natal dispersal, 304-311 Negro, Juan Jose, Juan Manuel Grande, and Jose Hernan Sarasola, Do Eurasian Hobbies {Falco subbuteo) have “false eyes” on the nape?, 287-288 Negro, Juan Jose, see Sarasola, Jose Hernan Nest characteristics, 238-242 provisioning, 367-371 sites, 231-237 substrate, 47-54 Nesting, 161-163 behavior, 256-262 Nestling diet, 9-18 provisioning rates, 9-18 Nestlings, 263-269 Nests, 1-8, 256-262 New Brunswick, 69-77 Nordmeyer, Dana L., see Palmer, Angela G. Northern flying squirrel, 214-230 Nygard, Torgeir, see Gjershaug, Jan Ove O Ojeda, Valeria, see Trejo, Ana Olea, Pedro R, Ruben Vera, Angel de Erutos, and Hugo Robles, Premigratory communal roosts of the Lesser Kestrel in the boreal summer, 278-282 Ombalski, Daniel W., see Brandes, David Orellana, Sergio Alvarado, see Rojas, Ricardo Figueroa Orientation, 283-286 Osprey, 47-54, 55-61, 141-147, 380-381 Ortego, Joaquin, A possible case of double-brooding of Eagle-Owls {Bubo bubo) in Spain, 378-379 Owl, Barn, 174-177, 17-181 Barred, 88-91, 231-237 Little, 35-46 Long-eared, 379-380 Northern Spotted, 88-91, 214-230, 231-237 Snowy, 275-277 Western Burrowing, 91-95 P Palma, Luis, Joao Ferreira, Rogerio Cangarato, and Pe- dro Vaz Pinto, Current status of the Osprey in the Cape Verde Islands, 141-147 Palmer, Angela G., Dana L. Nordmeyer, and Daniel D. Roby, Nestling provisioning rates of Peregrine Fal- cons in interior Alaska, 9-18 Pande, Satish A., Amit P. Pawashe, Banda Pednekar, Anil Mahabal, and Reuven Yosef, How long is too long? A case of fostering nestling Bonelli’s Eagles {Hieraae- tus fasdatus), 381-382 Pandion haliaetus, 47-54, 55—61, 141-147, 380-381 Panuccio, Michele, see Agostini, Nicolantonio Papandropoulos, Dimitris, see Alivizatos, Haralambos Pages, Carlos M,, see Mealey, Brian Paper wasps, 367-371 Parapolybia, 367—371 Pardinas, Ulyses FJ., see Leveau, Lucas M. Parks, Greta M., see Mealey, Brian Passer domesticus, 62-68 Patagonia, 174-177 northwestern, 1-8 Pathways, 195-207 Pavez, Eduardo R, Christian Gonzalez, Benito A. Gonza- lez, Cristian Saucedo, Sergio Alvarado, Juan P. Ga- bella, and Alejandra Arnello, Nesting of the White- throated Hawk {Buteo albigula) in deciduous forests of central Chile, 186—189 Pawashe, Amit R, see Pande, Satish A. Pb, 62-68 Pearce, Peter A., see Woodley, Stephen J. Pearlstine, Elise V, Variation in mitochondrial DNA of four species of migratory raptors, 250-255 Pearlstine, Elise Vernon and Daniel B. Thompson, Geo- graphic variation in morphology of four species of migratory raptors, 334-342 Pednekar, Banda, see Pande, Satish A. Pentriani, Vincenzo, see Del Mar Delgado, Maria Perch site, 19-25 Pernis apivorus, 283-286 ptilorhynchus, 367-371 Phalcoboenus megalopterus, 290-292 Philopatry, 168-174 Plumage, 270-275 Polistes, 367—371 Population declines, 69-77, 312-319 increase, 47-54 status, 343—349 trend, 91-95 Prawiradilaga, Dewi M., see Gjershaug, Jan Ove Predation, 243-249 risk, 35-46 Premigratory aggregations, 278-282 Prey selection, 214-230 Principal components analysis, 334-342 Productivity, 238-242, 350-356 R Rabearivony, Jeanneney, see de Roland, Lily-Arison Rene 392 Index to Volume 38 VoL. 38, No. 4 Randriamanga, Ignace, see de Roland, Lily-Arison Rene Radio-tracking, 343-349 Raptor rehabilitation, 77-81, 125-132 Red tree vole, 214-230 Relative abundance, 181—186 Release, 77—81 Reproductive success, 35-46, 263-269 Ress, Sara and Craig Guyer, A retrospective study of mor- tality and rehabilitation of raptors in the southeast- ern United States, 77—81 RFLP, 250-255 Ricca, Mark A., Robert G. Anthony, and Jeffrey C. Wil- liams, Bald Eagles consume Emporer Geese during late-winter in the Aleutian archipelago, 81-85 Rief, Vitali, Risto Tornberg, and Kauko Huh tala. Juvenile grouse in the diet of some raptors, 243-249 Ristow, Dietrich and Michael Wink, Seasonal variation in sex ratio of nestling Eleonora’s Falcons, 320-325 Ritchie, Robert J., Ann M. Wildman, and Clayton M. White, Peregrine Falcons nesting on lake bluffs on the Arctic coastal plain of northern Alaska, 158-160 Ritchison, Gary, see Leyhe, Jennifer E. Roadside survey, 181-186 Robles, Hugo, see Olea, Pedro P. Roby, Daniel D., see Palmer, Angela G. Roen, Keely T. and Richard H. Yahner, Vulture winter roost abandonment and reestablishment, 288-289 Rodriguez, Luis, see Tapia, Luis Rojas, Ricardo Figueroa, Sergio Alvarado Orellana, and E. Soraya Corales Stappung, Notes on a range ex- pansion and summer diet of the Mountain Caracara in the Andes of south-central Chile, 290-292 Rosset, Sergio D., see Alcalde, Leandro R0V, Nils, see Gjershaug, Jan Ove S Saggese, Miguel D. and Eduardo R. De Lucca, Live mam- mal prey (Zaedyus pichiy) in a nest of the Black-chest- ed Buzzard-Eagle {Geranoaetus melanoleucus) , 101— 102 Sahores, Mercedes and Ana Trejo, Diet shift of Barn Owls {Tyto alba) after natural fires in Patagonia, Argenti- na, 174-177 Sanchez-Guzman, Juan Manuel, see Villegas, Auxiliadora Sarasola, Jose Hernan and Juan Jose Negro, Gender de- termination in the Swainson’s Hawk {Buteo swain- soni) using molecular procedures and discriminant function analysis, 357-361 Sarasola, Jose Hernan, see Negro, Juan Jose Satellite telemetry, 55-61 Saucedo, Cristian, see Pavez, Eduardo F. Scavenging, 380-381 Schmutz, Josef K., A review of a hawk in the sun: adven- tures studying hawks, by Leon R. Powers, 2003, 294 Seavy, Nathaniel E., Environmental correlates of African Wood-Owl calling activity in Kibale National Park, Uganda, 208-213 Serum chemistry, 96-100 Serrano, Karla Pelz, see Lopez Gonzalez, Carlos A. Severinghaus, Lucia Liu, see Huang, Kuang-Mng Sex allocation, 320-325 ratio, 320-325 Sexing, 375-377 Sexual dimorphism, 361-366 Simon, Melinda M., see Dykstra, Cheryl R. Sleep, Darren J.H. and Rowan D.H. Barrett, Insect hawk- ing observed in the Long-eared Owl {Asio otus) , 379- 380 Smith, Jeff R, see McBride, Tobias J. Solensky, Mathew W., see Martell, Mark S. Soria, Hugo Luna, see Lopez Gonzalez, Carlos A. Spain, 133-140, 350-356, 361-366, 378-379 Sparrow, House, 62-68 Sparrowhawk, Eurasian, 243-249 Spizaetus bartelsi, 343-349 ornatus, 193-194 Stappung, E. Soraya Corales, see Rojas, Ricardo Figueroa Status, 141-147 Steck, Todd R., see Greene, Dana M. Strong, Allan M., see Chandler, Richard B. Strix occidentalis caurina, 88—91, 214—230, 231—237 varia varia, 88-91, 231-237 woodfordii, 208-213 Sun, Yuan-Hsun, Hsin-Ju Wu, and Ying Wang, Tawny Fish-Owl predation at fish farms in Taiwan, 326-333 Supriatna, Adam, see Gjershaug, Jan Ove Survey, 141-147, 152-157, 158-160, 181-186, 208-213 Sympatry, 181-186 Sympson, Lorenzo, see Trejo, Ana Survival, 304-311 T Taiwan, 326-333, 367-371 Tapia, Luis, Jesus Dominguez, and Luis Rodriguez, Mod- eling habitat use and distribution of Hen Harriers (Circus cyaneus) and Montagu’s Harrier (Circus pygar- gus) in a mountainous area in Galicia, northwestern Spain, 133-140 Terrain, 195-207 Terrazas, Alberto Lafon, see Young, Kendal E. Texas, 91-95, 148-152 Thompson, Bruce C., see Young, Kendal E. Thompson, Dan J., see Lehman, Chad P. Thompson, Daniel B., see Pearlstine, Elise Vernon Thorstrom, Russell, see de Roland, Lily-Arison Rene Thorstrom, Russell, see Tingay, Ruth E. Tingay, Ruth E., Michele L. Clarke, Richard T. Watson, Russell Thorstrom, and Loukman Kalavah, Survival and behavior of a one-footed Madagascar Fish-Eagle in the wild, 85-88 Tome, Ricardo, Carolina Bloise, and Erkki Korpimaki, Nest-site selection and nesting success of Little Owls (Athene noctua) in Mediterranean woodland and open habitats, 35-46 December 2004 Index to Volume 38 393 Tornberg, Risto, see Reif, Vitali Toxicology, 118-124 Trejo, Ana, Valeria Ojeda, Lorenzo Sympson, and Mari- ano Gelain. Breeding biology and nest characteris- tics of the White-throated Hawk {Buteo albigula) in northwestern Argentine Patagonia, 1-8 Trejo, Ana, see Sahores, Mercedes Trophic overlap, 178-181 Tucson, Arizona, 26-34 Tyto alba, 174-177, 178-181 U Updraft, 195-207 Uganda, 208-213 Urban, 304-311 ecosystem, 62—68 environments, 26-34 Una aalge, 380-381 V Valdez, Raul, see Young, Kendal E. Vaz Pinto, Pedro, see Palma Luis Vera, Ruben, see Olea, Pedro P. Verdejo, Jose, see Lopez-Lopez, Pascual Vilagosi, Janos, see Eakle, Wade L. Villegas, Auxiliadora, Juan Manuel Sanchez-Guzman, Em- ilio Costillo, Casimiro Corbacho, and Ricardo Mor- an, Productivity and fledgling sex ratio in a Cinere- ous Vulture (Aegypius monachus) population in Spain, 361-366 Vlachos, Christos, Dimitris Bakaloudis, and Evangelos Chatzinikos, Unusual nesting of the Lesser Kestrel (Falco naumanni) in Thessaly, Greece, 161—163 Vocalizations, 208-213 Vulture, Black, 288-289 Cinereous, 361—366 Turkey, 288-289 restaurants, 350-356 W Wang, Ymg, see Sun, Yuan-Hsun Washington, 88-91, 231-237, 295-303 Water-crossing, 283-286 Watson, James W., Responses of nesting Bald Eagles to experimental pedestrian activity, 295-303 Watson, Richard T., see Tingay, Ruth E. Watts, Bryan D., Mitchell A. Byrd, and Marian U. Watts, Status and distribution of breeding Ospreys in the Chesapeake Bay: 1995-1996, 47-54 Watts, Marian U., see Watts, Bryan D. Wetland, 371-374 White, Clayton M., see Ritchie, Robert J. Wildman, Ann M., see Ritchie, Robert J. Williams, Jeffrey C., see Ricca, Mark A. Wink, Michael, see Ristow, Dietrich Winter, 371-376 distribution, 181—186 Wintering area, 55-61 grounds, 357-361 Wood-Owl, African, 208-213 Woodley, Stephen J., Kenneth D. Meyer, David Anthony Kirk, and Peter A. Pearce, Contaminant levels, egg- shell thinning, and productivity in Sharp-shinned Hawks in Fundy National Park, New Brunswick, 69- 77 Wu, Hsin-Ju, see Sun, Yuan-Hsun Y Yahner, Richard H., see Roen, Keely T. Yosef, Reuven, see Pande, Satish A. Young, Kendal E., Bruce C. Thompson, Alberto Lafon Terrazas, Angel B. Montoya, and Raul Valdez, Aplo- mado Falcon abundance and distribution in the northern Chihuahuan Desert of Mexico, 107-117 Yukon Delta National Wildlife Refuge, 275-277 Yukon-Kuskokwim Delta, 275-277 Z Zabel, Cynthia J., see Forsman, Eric D. Zaedyus pichiy, 101-102 Zhao, Jiang, see Deng, Wen-Hong Zogaris, Stamatis, see Alivizatos, Haralambos THE JOURNAL OF RAPTOR RESEARCH A QUARTERLY PUBLICATION OF THE RAPTOR RESEARCH FOUNDATION, INC. (Founded 1966) EDITOR IN CHIEF James C. Bednarz ASSOCIATE EDITORS James R. Belthoff Clint W. Boal Cheryl R, Dykstra Michael I. Goldstein Joan L. Morrison Juan Jose Negro Marco Restani Fabrizio Sergio Ian G. Warrentin James W. Watson BOOK REVIEW EDITOR Jeffrey S. Marks CONTENTS FOR VOLUME 38, 2004 Number 1 Breeding Biology and Nest Charagteristics of the White-throated Hawk {Buteo albigula) in Northwestern Argentine Patagonia. Ana Trejo, Valeria Ojeda, Lorenzo Sympson, and Mariano Gelain 1 Nestling Provisioning Rates of Peregrine Falcons in Interior Alaska. Angela G. Palmer, Dana L. Nordmeyer, and Daniel D. Roby 9 Perch Sites and Hunting Behavior of Red-tailed Hawks (Buteo jamaicensis) . Jennifer E. Leyhe and Gary Ritchison 19 Movements and Survival of Fledgling Cooper’s Hawks in an Urban Environment. R. William Mannan, Wendy A. Estes, and William J. Matter 26 Nest-site Selection and Nesting Sugcess of Little Owls (Athene noctua) in Mediterranean Woodland and Open Habitats. Ricardo Tome, Carolina Bioise, and Erkki Korpimaki 35 Status and Distribution of Breeding Ospreys in the Chesapeake Bay: 1995 — 96. Bryan D. Watts, Mitchell A. Byrd, and Marian U. Watts 47 Partial Migration and Wintering Use of Florida by Ospreys. Mark s. Marteii, Michael A. McMillian, Mathew J. Solensky, and Brian K. Mealey 55 Elevated Lead Levels in Urban House Sparrows: A Threat to Sharp-shinned Hawks and Merlins? Richard B. Chandler, Allan M. Strong, and Carlin C. Kaufman 62 Short Communications Contaminant Levels, Eggshell Thinning, and Productivity in Sharp-shinned Hawks in Fundy National Park, New Brunswick. Stephen j. Woodley, Kenneth d. Meyer, David Anthony Kirk, and Peter A. Pearce 69 A Retrospective Study of Mortality and Rehabilitation of Raptors in the Southeastern United States. Sara Ress and Craig Guyer 77 Bald Eagles Consume Emperor Geese during Late-Winter in the Aleutian ArGHIPELAGO. Mark A. Ricca, Robert G. Anthony, and Jeffrey C. Williams 81 Survival and Behavior of a One-footed Madagascar Fish-Eagle in the Wild. Ruth E. Tingay, Michele L. Clarke, Richard T. Watson, Russell Thorstrom, and Loukman Kalavah 85 Reproductive Success of Spotted Owls Sympatric with Barred Owls in Western Washington. Wayne F. Iverson 88 Historical and Current Status of Breeding and Wintering Western Burrowing Owls (Athene CUNICULARIA HYPUGAEA) IN Texas. Nancy E. McIntyre 91 Serum Chemistry Values for Nestling Bald Eagles (Haliaeetus leucocephalus) in Florida Bay, Everglades National Park. Brian Mealey, Greta M. Parks, Carlos M. Pages, Brian A. Millsap, Orin L, Bass, and Gregory D. Bossart 96 Letter Live Mammal Prey {Zaedyus pichiy) in a Nest of the Black-chested Buzzard-Eagle {Geranoaetus MELANOLEUCUS) . Miguel D. Saggese and Eduardo R. De Lucca 101 Book Review. Edited by Jeffrey S. Marks 103 Manuscript Referees 105 Number 2 Aplomado Falcon Abundance and Distribution in the Northern Chihuahuan Desert of Mexico. Kendal E. Young, Bmce C. Thompson, Alberto Lafon Terrazas, Angel B. Montoya, and Raul Valdez 107 Blood-Lead and ALAD Activuy Levels of Cooper’s Hawks {Accipiter cooperii) Migrating Through the Southern Rocky Mountains. Tobias j. McBride, Jeff p. Smith, Howard P. Gross, and Michael J. Hooper 118 An Assessment of Cage Flight as an Exercise Method for Raptors. Dana m. Greene, Mathias Engelmann, and Todd R. Steck 125 Modeling Habitat Use and Distribution of Hen Harriers {Circus cyaneus) and Montagu’s Harrier {Circus pygargus) in a Mountainous Area in Galicia, Northwestern Spain. Luis Tapia, Jesus Dominguez, and Luis Rodriguez 133 Current Status of the Osprey in the Cape Verde Islands. Luis Palma, joao Ferreira, Rogerio Cangarato, and Pedro Vaz Pinto 141 Short Communications Development of Hunting Behavior in Hacked Aplomado Falcons, jessi l. Brown, William R. Heinrich, J. Peter Jenny, and Brian D. Mutch 148 Summer Roadside Raptor Surveys in the Western Pampas of Argentina. Michael i. Goldstein and Toby J. Hibbitts 152 Peregrine Falcons Nesting on Lake Bluffs on the Arctic Coastal Plain of Northern Alaska. Robert j. Ritchie, Ann M. Wildman, and Clayton M. White 158 Unusual Nesting of the Lesser Kestrel {Falco naumanni) in Thessaly, Greece. Christos Vlachos, Dimitris Bakaloudis, and Evangelos Chatzinikos 161 Fat Stores of Migrant Sharp-shinned and Cooper’s Hawks in New Mexico. John r DeLong and Stephen W. Hoffman 1 63 Spanish Ringing and Recovery Records of Booted Eagle {Hieraatus pennatus). Ignacio S. Garcia Dios 168 Diet Shift of Barn Owls ( Tyto alba) after Natural Fires in Patagonia, Argentina. Mercedes Sahores and Ana Trejo.... 174 Trophic Relationships Between White-tailed Kites {Elanus ieucurus) and Barn Owls ( Tyto alba) in Southern Buenos Aires Province, Argentina. Lucas m. Leveau, Carlos M. Leveau, and Ulyses E.J. Pardihas 178 Relative Abundance and Diversity of Winter Raptors in Spokane County, Eastern Washington. Howard L. Ferguson 181 Nesting of the White-throated Hawk {Bitteo albigula) in Deciduous Forests of Central Chile. Eduardo F. Pavez, Christian Gonzalez, Benito A. Gonzalez, Cristian Saucedo, Sergio Alvarado, Juan P. Gabella, and Alejandra Arnello 186 Letters Observation of the Chimango Caracara (Milvago chimango) Feeding on Common Lesser Toads {Bufo FERNANDE7AE) . Leandro Alcalde and Sergio D. Rosset 190 Consumption of a Ringed Kingeisher {Megaceryle torquata) by a White-tailed Hawk {Buteo albicaudatus) in Southeastern Brazil. Jose Carlos Motta-Junior and Marco Antonio Monteiro Granzinolli 191 Golden Eagle {Aquila chrysaetos) Predation Attempts on Merriam’s Turkeys {Meleagris gallpavo MERRlAMl) IN THE SOUTHERN Black Hills, South DAKOTA. Chad P. Lehman and Dan J. Thompson 192 A Record of the Ornate Hawk-Eagle (Spizaetus ornatus) in Nayarit, Mexico. Carlos A. Lopez Gonzalez, Eduardo Ponce Guevara, Karla Pelz Serrano, Hugo Luna Soria, and Rodrigo Sierra Corona 193 Number 3 Modeling Raptor Migration Pathways Using a Fluid-flow Analogy. David Brandes and Daniel W, Ombalski 195 Environmental Correlates of African Wood-Owl Calling Activiiy in Kibale National Park, Uganda. Nathaniel e. Seavy 208 Diets and Foraging Behavior of Northern Spotted Owls in Oregon, ehc d. Forsman, Robert G. Anthony, E. Charles Meslow, and Cynthia J. Zabel 214 A Comparison of Barred and Spotted Owl Nest-site Characteristics in the Eastern Cascade Mountains, Washington. Joseph B. Buchanan, Tracy L. Fleming, and Larry L. Irwin 231 Productivity and Nest-site Characteristics of Gray Hawks in Southern Arizona. Brent D. Bibles and R. William Mannan 238 Juvenile Grouse in the Diet of Some Raptors. Vitaii Reif, Risto Tomberg, and Kauko Huhtala 243 Variation in Mitochondrial DNA of Four Species of Migratory Raptors. Eiise v. Pearlstine 250 Nesting Biology and Diet of the Madagascar Harrier ( Circus macrosceles) in AmBOHITANTELY Special Reserve, Madagascar. Lily-Arison Rene de Roland, Jeanneney Rabearivony, Ignace Randriamanga, and Russell Thorstrom 256 Breeding Biology of the Grey-faced Buzzard {BiriASTUR indicus) in Northeastern China. Wen-Hong Deng, Wei Gao, and Jiang Zhao 263 Short Communications Variable Retention Times for Rectrices at Different Loci in a Golden Eagle. David H. Ellis and Marc Kery 270 A Local Concentration of Snowy Owls on the Yukon-Kuskokwim Delta in Summer 2000. Christopher M. Harwood, Brian J. McCaffery, Fred J. Broerman, and Paul A. Liedberg 275 Premigratory Communal Roosts of the Lesser Kestrel in the Boreal Summer. Pedro P. Olea, Ruben Vera, Angel de Frutos, and Hugo Robles 278 Analysis of the Autumn Migration of Juvenile Honey-buzzards {Pernis apivorus) Across the Central Mediterranean. Nicolantonio Agostini, Charles Coleiro, and Michele Panuccio 283 Letters Do Eurasian Hobbies {Falco subbuteo) Havt'. “Fai.se Eyes” on the Nape? Juan Jose Negro, Juan Manuel Grande, and Jose Hernan Sarasola 287 Vulture Winter Roost Abandonmeni and Reestablishment. Keely T. Roen and Richard H. Yahner 288 Commensal Foraging by a Red-shoulderi-;u Hawk {Buteo lineatus) wrra Wild Turkeys (AlEiJiAGius GALLOPAVO). Gary R. Graves 290 Notes on a Range Expansion and Summer Diet of the Mountain Caracara in the Andes of South- central Chile. Ricardo Figueroa Rojas, Sergio Alvarado Orellana, and E. Soraya Corales Stappung 290 An Example of Cooperative Hunting by Saker Fai cons in Hungary. Wade L. Eakle, Carl Millier, Pierre Mineau, and Janos Vilagosi 292 Book Review. Edited by Jeffrey S. Marks 294 Number 4 Responses of Nesting Bald Eagles to Experimental Pedestrian Activity. James W. Watson 295 Dispersal and Mortality of Red-shouldered Hawks Banded in Ohio. Cheryl R. Dykstra, Jeffrey L. Hays, Melinda M. Simon, John B. Holt, Jr., G. Ronald Austing, and F. Bernard Daniel 304 Decreases in a Population of Red-shouldered Hawks Nesting in Central Maryland. Elwood M. (Woody) Martin 312 Seasonal Variation in Sex Ratio of Nestling Eleonora’s Ealcons. Dietrich Ristow and Michael Wink 320 Tawny Fish-Owl Predation at Fish Farms in Taiwan. Yuan-Hsun Sun, Hsinju Wu, and Ymg Wang 326 Geographic Variation in Morphology of Four Species of Migratory Raptors. Elise Vernon Pearlstine and Daniel B. Thompson 334 Home-range Size of the Javan Hawk-Eagle (Spizaetus bartelsi) Estimated from Direct Observations and Radiotelemetry, jan Ove Gjershaug, Nils R0v, Torgeir Nygard, Dewi M. Prawiradilaga, M. Yayat Afianto, Hapsoro, and Adam Supriatna 343 Population Status and Reproductive Performance of Eurasian Griffons ( Gyps FULVUS) IN Eastern Spain. Pascual L6pez-L6pez, Clara Garcfa-Ripolles, and JoseVerdejo 350 Short Communications Gender Determination in the Swainson’s Hawk {Buteo swainsoni) Using Molecular Procedures and Discriminant Function Analysis, jose Heman Sarasoia and Juan Jose Negro 357 Productivity and Fledgling Sex Ratio in a Cinereous Vulture (Aegypius MONACHUS) Population in Spain. Auxiliadora Villegas, Juan Manuel Sanchez-Guzman, Emilio Costillo, Casimiro Corbacho, and Ricardo Moran 361 Nest Provisioning of the Oriental Honey-buzzard {Pernis ptilorhynchus) in Northern Taiwan. Kuang-Ying Huang, Yao-Sung Lin, and Lucia Liu Severinghaus 367 Winter Diet of the Greater Spotted Eagle {Aquila clanga) in the Amvrakikos Wetlands, Greece. Haralambos Alivizatos, Dimitris Papandropoulos, and Stamatis Zogaris 37l Gender Determination of Eurasian Eagle-Owls (Bubo bubo) by Morphology. Maria del mar Delgado and Vincenzo Penteriani 375 Letters A Possible Case of Double Brooding of Eagle-Owls {Bubo bubo) in Spain. Joaquin Ortego 378 Insect Hawking Observed in the Long-eared Owl {Asio otus). Darren J.H. Sleep and Rowan D.H. Barrett 379 Osprey Scavenges Common Murre Carcass in Coastal Washington. Joseph B. Buchanan 380 How Long is Too Long? A Case of Fostering Nestling Bonelli’s Eagles {Hieraaetus fasciatus) . Satish A. Pande, Amit P. Pawashe, Banda Pednekar, Anil Mahabal, and Reuven Yosef 381 Information for Contributors 383 Index to Volume 38 387 orders:8oo-722-246o Phone:434-263-4842 fax:434-263-4842 Buteo Books is the largest retailer of Ornithology books in North America^ with over 2,000 in-print titles in stock. A FEW BOOKS ON BIRDS OF PREY AND FALCONRY Handbook of the Birds of the World. The first seven volumes of this projected sixteen-volume work are available for $185 each. Volume 2 includes diurnal raptors, Volume 3 includes owls. Volume 8, the first to cover Passerines, due Summer 2003. A Photographic Guide to North American Raptors. Wheeler & Clark. $24.95 (New Eastern and Western editions by Wheeler are due September 2003.) Ecology and Conservation of Owls. Ian Newton. $95 Birds of Prey: Health and Disease. Third Edition. Cooper. $96.95 North American Falconry and Hunting Hawks. Beebe & Webster. $55 Buteo Books has a wide selection of used and out-of-print titles available. Our stock changes daily, so call to check availability of that scarce title you're seeking. In 1992 the American Ornithologists' Union, in partnership with the Academy of Natural Sciences of Philadelphia, undertook the publication of species accounts for each of the more than 700 species which breed in the United States and Canada. These illustrated reviews provide comprehensive summaries of the current knowledge of each species, with range maps and an extensive list of references. Buteo Books is pleased to offer these individual species accounts for $7.50 each. All 716 profiles are listed in taxonomic order on our website (conforming with the Seventh Edition of the A.O.U. Check-list of North American Birds). Singles may be ordered by contacting Buteo Books. Shipping and handling is $4 for the first profile and $1 for each additional profile to a maximum of $10 per order. Visit our website for more information: www.buteobooks.com RARE AND OUPOF-PRINT ORNITHOLOGY BIRDS OF NORTH AMERICA SERIES allen@buteobooks . com 3130 Laurel Road; Shipman, VA 22971; USA A Telemetry Receiver Designed with The Researcher in Mind What you've been waiting for! fin illy, ,1 liKjf'.ly '.nn'.it i v*‘ v f'f'.'.’nil !»v'»*ivi.‘t ih.it <^oiq}n Io‘/% th.iM H ounce*., is c I'mpij.tri V ti'.ei prn:jr,i-:un )!).*• .ind ..iM.iDif .m f.ito'. uvc: .ill frequoncics. To: e