The Journal of Volume 39 Number 4 December 2005 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 LARGE #2: Eduardo Inigo-Euas DIRECTOR AT LARGE #3; Michael W. Collopy DIRECTOR AT LARGE #4; Carol McIntyre DIRECTOR AT LARGE #5; John A Smallwood DIRECTOR AT LARGE #6; Daniel E. Varland Ruth Tingay NORTH AMERIGAN DIRECTOR #1: Steve Hoffman NORTH AMERICAN DIRECTOR #2: Gary Santolo NORTH AMERICAN DIRECTOR #3: Ted Swem INTERNATIONAL DIRECTOR#!: Nick Mooney INTERNATIONAI. DIRECTOR #2: PRESIDENT; Brian A. Millsap VICE-PRESIDENT; David M. Bird ^ EDITORIAL STAEE EDITOR; James C. Bednarz, Department of Biological Sciences, RO. Box 599, Arkansas State University, State University, AR 72467 U.S.A. ASSOCIATE EDITORS James R. Beltpioff Joan L. Morrison Clint W. Boal Eabrizio Sergio Cheryl R. Dykstra Ian G. Warkentin Michael I. Goldstein James W. Watson BOOK REVIEW EDITOR; Joelle Gehring, Department of Biology, Central Michigan University, Mount Pleasant, MI 48859 U.S.A. SPANISH EDITORS; Carlos Daniel Cadena, Lucio R. Malizia, Cintia Cornelius EDITORIAL ASSISTANTS: Autumn Farless, Joan Ciark 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 either electron- ic tiles or a typewritten original and three copies to the Editor. All submissions must be typewritten and dou- ble-spaced on one side of 216 X 278 mm (8% XII in.) or standard international, white, bond paper, with 25 mm (1 in.) margins. The cover page should contain a tide, 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 figure 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., 1 January 1999). Refer to a recent issue of the journal for details in format. Explicit instructions and puhlication policy are outlined in “Information for contributors,”/. Raptor Res., Vol. 39(4), and are available from the editor. Submit manuscripts to incoming Editor-in-chief Cheryl Dykstra, Raptor Environmental, 7280 Susan Springs Drive, West Chester, OH 45069-3696 U.S.A.journalofraptorresearch@juno.com COVER: Cuban Black-Hawk {Buteogallus anthracinus gundlachii). Painting by Nils Navarro. Contents Taxonomic Status and Biology of the Cuban Black-Hawk, Buteogallus ANTHRACINUS GUNDLACHII (AVES: AcCIPITRIDAE) . James W. Wiley and Orlando H. Garrido 351 I Home Range and Habitat Use of Northern Spotted Owls on the Olympic Peninsula, Washington. Eric D. Forsman, TimmothyJ. Kaminski, Jeffery C. Lewis, Kevin J. Maurice, Stan G. Sovem, Cheron Ferland, and Elizabeth M. Glenn 365 First-cycle Molts in North American Falconieormes. Peter Pyie 378 Morphometric Analysis of Large Falco Species and Their Hybrids with Implications for Conservation. Chris p. Eastham and Mike k Nichoiis 386 A Change in Foraging Success and Cooperative Hunting by a Breeding Pair of Peregrine Falcons and Their Fledglings. Dick Dekker and Robert Taylor 394 Nesting Ecology and Behavior of Broad-winged Hawks in Moist Karst Forests of Puerto Rico. Derek W. Hengstenberg and Francisco J.VUella 404 Raptor Abundance and Distribution in the Llanos Wetlands of Venezuela. Wendy J. Jensen, Mark S. Gregory, Guy A. Baldassarre, Francisco J. Vilella, and Keith L. Bildstein .... 417 A Comparison of Breeding Season Food Habits of Burrowing Owls Nesting IN Agricultural and Nonagricultural Habitat in Idaho. Coiieen e. Moulton, Ryan S. Brady, and James R. Belthoff 429 Red-tailed Hawk Dietary Overlap with Northern Goshawks on the Kaibab Plateau, Arizona. Angela E. Gatto, Teryl G. Grubb, and Carol L. Chambers 439 Bat Predation by Long-eared Owls in Mediterranean and Temperate Regions OF Southern Europe. Ana Maria Garcia, Francisco Cervera, and Alejandro Rodriguez 445 Short Communications Differential Effectiveness of Playbacks for Little Owls (Athene noctua) Surveys Before and After Sunset. Joan Navarro, Eduardo Mmguez, David Garcia, Carlos Villacorta, Francisco Botella, Jose Antonio Sanchez-Zapata, Martina Carrete, and Andres Gimenez 454 King Vultures (Sarcoramphus papa) Forage in Moriche and Cucurit Palm Stands. Marsha A. Schlee 458 Family Break Up, Departure, and Autumn Migration in Europe of a Family of Greater Spotted Eagles (Aquila clanga) as Reported by Satellite Telemetry. Bernd-U. Meyburg, Christiane Meyburg, Tadeusz Mizera, Grzegorz Maciorowski, and Jan Kowalski 462 Seasonal Patterns of Common Buzzard (Buteo buteo) Relative Abundance and Behavior in ^ POLLINO National Park, Italy. Massimo Pandolfi, Alessandro Tanfema, and Giorgia Gaibani ...... 466 New Nesting Record and Observations of Breeding Peregrine Falcons in Baja C al ifornia Sur, Mexico. Aradit Castellanos, Cerafina Arglielles, Federico Salinas-Zavala, and Alfredo Ortega-Rubio 472 Letters A Previously Undescribed Vocalization of the Northern Pygmy-Owl. Graham G. Frye 476 Book Review. Edited byjoelle Gehring 478 Information for Contributors 480 Index to Volume 39 484 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. 39 December 2005 No. 4 J. Raptor Res. 39(4) :351— 364 © 2005 The Raptor Research Foundation, Inc. TAXONOMIC STATUS AND BIOLOGY OF THE CUBAN BLACK- HAWK, BUTEOGALLUS ANTHRACINUS GUNDLACHII (AVES: ACCIPITRIDAE) James W. Wiley^ USGS Maryland Cooperative Fish and Wildlife Research Unit, University of Maryland Eastern Shore, 1120 Trigg Hall, Princess Anne, MD 21853 US. A. Orlando H. Garrido 1706 Calle 60 entre 17 y 19, Marianao 13 (Play a), La Habana, Cuba Abstract. — ^We reevaluate the taxonomic status of the Cuban population of the Common Black-Hawk {Buteogallus anthracinus) based on our examination of additional specimens, nests, eggs, and voice data. Buteogallus a. gundlachii is smaller than mainland populations of anthracinus and differs from mainland birds in plumage coloration and pattern. The common (alarm) call of gundlachii is a series of three or four notes, differing from that of mainland anthracinus, whose call consists of 9-24 notes. In the Isla de Pinos, Cuba, we observed gundlachii ediiing two species of land crabs (71.4%), centipedes (7.1%), lizards (10.7%), mammals (7.1%), and a bird (3.6%). We consider Buteogallus gundlachii Cabanis 1854 (1855), the Cuban Black-Hawk, to be a full species, endemic to Cuba, Isla de Pinos, and many of the cays of the Cuban Archipelago. Key Words: Common Black-Hawk; Buteogallus anthracinus; Cuban Black-Hawk; Buteogallus gundlachii; Buteogallus subtilis; ecology; taxonomy. ESTADO TAXONOMICO Y BIOLOGIA DE BUTEOGALLUS ANTHRACINUS GUNDLACHII (AVES: ACCIPITRIDAE) Resumen. — En este estudio re-evaluamos el estatus taxonomico de la poblacion cubana de Buteogallus anthracinus (subespecie gundlachii) con base en examenes de especimenes adicionales, nidos, huevos y datos de la voz. Los individuos de B. a. gundlachii son mas pequehos que los individuos de las poblaciones continentales de B. anthracinus, y difieren de las aves del continente en la coloracion y patron del plumaje. El llamado comun de alarma de gundlachii es una serie de tres o cuatro notas, mientras que el llamado de anthracinus en el continente consiste de entre 9 y 24 notas. En la Isla de Pinos, Cuba, observamos a gundlachii alimentandose de dos especies de cangrejos terrestres (71.4%), ciempies (7.1%), lagartijas (10.7%), mamiferos (7.1%) y un ave (3.6%). Consideramos Buteogallus gundlachii Cabanis, 1854 (1855) debe ser tratado como una especie distinta, endemica de Cuba, la Isla de Pinos y muchos de los cayos del archipielago cubano. [Traduccion del autores] The New World genus Buteogallus Lesson, 1830 includes five species, mostly restricted to tropical areas, including Great Black-Hawk {Buteogallus uru- ^ Email address: jwwiley@mail.umes.edu bitinga) of the lowlands of Mexico to northern Ar- gentina, Savanna Hawk {B. meriodionalis) inhabit- ing savannas and marshes of western Panama to northern Argentina, Rufous Crab-Hawk (B. aequi- noctialis) occurring in mangroves of northeastern 351 352 Wiley AND Garrido VoL. 39, No. 4 Venezuela to eastern Brazil (Parana), and Man- grove Black-Hawk {B. subtilis), which is restricted to the Pacific coasts and rivers of El Salvador south to northwestern Peru. The Common Black-Hawk (Buteogallus anthracinus, Deppe 1830) ranges from southwestern United States, south to extreme northern South America (coastal Venezuela to northeastern Guiana) , Colombia, to northern Peru, including Trinidad, and some of the West Indies (Bond 1950, American Ornithologists’ Union 1998). One of the West Indian populations {B. a. cancrivorus Clark 1905b) is restricted to St. Vincent, St. Lucia, Union Island (Grenadines), and Grenada (accidental and doubtful in last two is- lands; no specimens taken; Clark 1905a, 1905b, 1905c, Bond 1950, Evans 1990) in the Lesser An- tilles, whereas the only other Antillean population {B. a. gundlachii Cabanis 1854 [1855]) occurs in Cuba and its satellites. The taxonomic status of the Cuban population has been controversial, with some considering the form as a full species, Buteo- gallus gundlachii (as originally described by Cabanis [1854, actually 1855]) instead of Buteogallus anth- racinus gundlachii (American Ornithologists’ Union 1998) . Among those authorities who have consid- ered the Cuban form gundlachii conspecific with the continental species {anthracinus) are Sharpe (1874, 1899), Cory (1887, 1892), Bangs and Zap- pey (1905), Bond (1956a, 1956b), Amadon (1961), Brown and Amadon (1968), Mayr and Short (1970), Stresemann and Amadon in Mayr and Cot- trell (1979), Palmer (1988), Sibley and Monroe (1990), Ferguson-Lees and Christie (2001), Dick- inson (2003), and others. Conversely, other au- thors have considered gundlachii different from Bu- teogallus anthracinus at the species level: Cabanis (1855), Gundlach (1854, 1865-1 866a, 1865-1 866b, 1871, 1876), Ridgway (1876), Gurney (1876, 1934), Bangs (1905), Swann (1921-1922, 1930), Peters (1931), Bond (1936), Hellmayr and Conover (1949), Friedmann (1950), Monroe (1963, 1968), Wetmore (1965), and others. Some of these au- thors subsequently changed their opinions on the Cuban form’s status, later considering gundlachii conspecific with anthracinus (e.g., Gundlach 1893, Bond 1950, 1956a, 1956b). With rare exception, however, previous evaluations did not consider the important characteristics of breeding biology and voice, mainly because of the limited knowledge of the Cuban form resulting from the difficulty in reaching its breeding habitats. The lack of natural history information is not unique to gundlachii, but is also true for other forms of the genus Buteogallus, e.g., anthracinus and subtilis, which are currently recognized as different species (Aldrich and Bole 1937, Amadon 1982, Mayr and Cottrell 1979, Stiles and Skutch 1989, Sibley and Monroe 1990, Amer- ican Ornithologists’ Union 1998, Ridgely and Greenfield 2001), but with reservation by some au- thors (Stiles and Skutch 1989, American Ornithol- ogists’ Union 1998, Ridgely and Greenfield 2001). Here, we reevaluate the taxonomic status of the Cuban population of Buteogallus anthracinus gund- lachii, based on our examination of more speci- mens, nests, eggs, and behavioral data, especially vocalizations, than were considered by previous workers. All published work on the Cuban form has been based on information from the few spec- imens collected before 1960, all of which are de- posited in foreign institutions. In this study, we in- clude specimens in North American and Cuban collections, including those collected after 1960, and not evaluated previously. Our main comparison in this assessment is with anthracinus, the taxon most often linked to gund- lachii. In these comparisons, we refer to Cuban populations as gundlachii and other forms as anth- racinus. It is not the purpose of this contribution to speculate on the taxonomic status of Buteogallus subtilis (including the three subspecies), although we make some comparisons between subtilis and gundlachii. Study Area Many of the observations reported here were made during our 30 yr of field experience throughout Cuba. We made more intensive observations of nesting black- hawks from 1996 to 1998 at the Los Indios Ecological Reserve, Isla de Pinos (now Isla de la Juventud). M^or vegetational communities at Los Indios include: (1) man- grove forest formation, characterized by black mangrove (Avicennia germinans) and red mangrove {Rhizophora man- gle); (2) semi-deciduous gallery forests, with prominent Cuban royal palm (Roystonea regia), beach hibiscus {Ht- hiscus tiliaceus), and pond apple {Annona glabra); (3) the open forest (savanna) formation of an open pine {Pinus caribaea and P. tropicalis) and Cuban bottle palm ( Colpoth- rinax rmightii) , with silver saw palm (Acoelorraphe wrightn) and a sparse undergrowth; and (4) the pine-barren for- mation, with pines and palms, and an undergrowth pre- dominantly of Pachyanthus cubensis, P. poiretii, Kalmiella ag- gregata, Miconia delicatula, Polygala uncinata, Lyonia myrtilloides, and Pinguicula filifolia (Jennings 1917, Alain 1946). Black-hawk observations were made mainly in the mangrove and gallery forests. Additional intensive obser- vations in red and black mangrove habitats were made in Cienaga de Zapata in December 1999. An elevated road bed, lined with Casuarina equisetifolia and scrub veg- December 2005 Cuban Black-Hawk Status and Biology 353 elation, bisects the mangrove forest where we made our observations near Playa Larga. Methods We examined specimens of Buteogallus a. anthracinus (N = 37), B. a. gundlachii (12), B. a. cancrivorus (4), B. subtilis (25), B. aequinoctialis (3), and B. urubitinga (24) deposited in the Field Museum of Natural History (Chi- cago), Museum of Comparative Zoology (Harvard Uni- versity), American Museum of Natural History, United States National Museum of Natural History, Academy of Natural Sciences of Philadelphia, Louisiana State Univer- sity Museum of Natural History, Museo Nacional de His- toria Natural de Cuba (La Habana), and Instituto de Eco- logia y Sistematica (Cuba) (Table 1). Conventional measurements of wing chord (flattened against the rul- er), tail, tarsus, and exposed culmen were taken to the nearest 0.1 mm with calipers. Egg masses were measured to the nearest gram using spring scales. We present sum- mary descriptive statistics (mean, SD, and range) for the specimens. We plotted body measurements to assess the pattern of spatial segregation among populations and forms. The hypothesis of separation derived from the plots of body measurements was tested using discriminate function analysis (DFA; Kleinbaum and Kupper 1978). SPSS (1999) for Windows was used to run DFA. Results Morphometries and Plumage. Adult morphology. Our examinations of the two taxa of B. anthracinus (anthracinus and gundlachii) revealed differences in size and coloration. We found sexual size dimor- phism in three of the measurements taken of spec- imens of mainland anthracinus (Table 1). There- fore, size comparisons between anthracinus and gundlachii were made within sex; i.e., male anthra- cinus with male gundlachii and female anthracinus with female gundlachii. Tarsal length was not dif- ferent in either population, so for comparing anth- racinus with gundlachii tarsi we combined male and female measurements for that morphometric pa- rameter. Only measurements of wing and exposed culmen for gundlachii revealed sexual size dimor- phism (P < 0.01; Table 1), although the small sam- ple size of females (N — 5) precluded a reliable analysis. Measurements of gundlachii yielded a mean Dimorphic Index (Storer 1966) of 6.9, com- pared with a mean index of 5.6 for anthracinus (Ta- ble 1). Birds from Cuba (gundlachii) are substantially smaller than mainland (anthracinus) birds in some conventional measurements, including wing chord in both sexes and tail length in males (Table 2). Also, tarsal lengths (combined male and female measurements) were significantly different be- tween the two forms (P — 0.001). A stepwise selec- tion procedure within DFA revealed wing chord, tail length, and exposed culmen were the most im- portant of the size variables measured. Plots con- trasting these variables within sex showed anthra- cinus and gundlachii tended to occupy generally distinct regions of the morphological space (Fig. 1 ). To further examine size differences between the two populations, we used linear discriminant anal- ysis to classify specimens into two groups (“race”), mainland anthracinus and Cuban gundlachii, using lengths of wing chord, tail, culmen, and tarsus as predictors. For male anthracinus, the analysis pro- duced a true group classification proportion of 0.938 (15 of 16 correctly classified) and 0.857 (6 of 7 correctly classified) for gundlachii males, for an overall proportion correct of 0.913 (21 of 23) (Wilks’s lambda = 0.375; = 17.646; df = 4, P< 0.001). For females, the analysis produced a true group classification proportion of 0.857 (18 of 21) for anthracinus and 0.800 (4 of 5) for gundlachii individuals, for an overall proportion correct of 0.846 (22 of 26) (Wilks’s lambda = 0.4.95; = 14.781.1; df = 4, P< 0.005). The four adult female St. Vincent (B. a. canen- vorus) specimens we examined were somewhat larger in wing chord (x = 389 ± 7.63, range = 385-401; t = -4.99, P = 0.002, df = 6) than gund- lachii females, whereas we found no difference be- tween the two island forms in tail (213.3 ± 12.4; range = 200-230; t - -1.83, P > 0.05, df = 6), culmen (27.3 ± 0.8; range = 26.8-28.4; t = 0.60, P > 0.05, df = 5), or tarsus (85.5 ± 6.4; range = 81.0-94.9; t = -1.63, P> 0.05, df = 4) length. We found no differences (P > 0.05) in measurements between anthracinus and cancrivorus. In general coloration, gundlachii differs from B. anthracinus and B. subtilis in being chocolate- brown, not slate blackish or even black as in the latter two forms. However, some specimens of anth- racinus, especially of the race cancrivorus, have a tendency to be less blackish, almost dark brown. The underparts feathers of gundlachii have a light (brownish-gray) edge, more conspicuous to- ward the abdominal region and more broadly edged on the alula coverts than in anthracinus, with the edging on the terminal alula coverts becoming white bands. The margins of the flank and thigh feathers are heavily marked, forming a series of bands, although these bands tend to disappear in older birds. The shoulder feathers are boldly barred in white, contrasting with the chocolate- 354 Wiley AND Garrido VoL. 39, No. 4 Table 1. Sexual size dimorphism in four body measurements from specimens of Buteogallus anthracinus (mainland Buteogallus a. anthracinus -aind Cuban B. a. gundlachii) , B. subtilis, B. aequinoctialis, and B. urubitinga, expressed as mean, standard deviation, range, and sample size (in parentheses). Statistical analyses are between-sex comparisons (two- sample t-test; equal variances not assumed) . Species Structure Sex Male Female t df P Signif- icance® D.l.^ Buteogallus anthracinus Wing 371.69 ± 11.95 (16) 385.19 ± 11.21 (21) -3.50 31 0.001 * 3.6 341-393 360-421 Tail 195.50 ± 7.40 (16) 213.81 ± 10.61 (21) -6.18 34 0.0001 8.9 183-210 190-230 Exposed culmen 26.27 ± 0.82 (16) 27.40 ± 1.25 (20) 3.24 32 0.003 4.2 25.1-28.1 23.6-30.3 Tarsus 85.94 ± 2.65 (16) 85.42 ± 4.00 (21) 0.48 34 0.636 ns -0.6 80-90.0 80.3-92.7 Mean D.I. 4.0 Buteogallus gundlachii Wing 342.71 ± 12.16 (7) 363.00 ± 8.43 (5) -3.41 9 0.008 * 5.8 323-370 350-372 Tail 179.29 ± 9.12 (7) 191.60 ± 22.16 (5) -1.15 4 0.313 ns 6.6 167-197 182-233 Exposed culmen 25.32 ± 0.69 (7) 27.54 ± 0.61 (5) -5.84 9 0.0001 8.3 24.5-28.5 26.7-28.1 Tarsus 81.33 ± 3.57 (6) 79.67 ± 3.56 (5) 0.77 8 0.464 ns -2.1 75.4-87 79.0-87.7 Mean D.I. 4.7 Buteogallus subtilis Wing 348.0 ± 13.68 (12) 352.31 ± 13.43 (13) -0.79 22 0.436 ns 1.2 330-370 328-373 Tail 189.92 ± 14.58 (12) 191.69 ± 8.65 (13) -0.37 17 0.719 ns 0.9 168-220 180-205 Exposed culmen 25.46 ± 2.06 (12) 26.56 ± 1.56 (12) -1.47 20 0.157 ns 4.2 19.7-27.8 23.1-28.8 Tarsus 79.78 ± 3.09 (11) 79.55 ± 2.76 (13) 0.19 19 0.848 ns -0.3 73.3-84.1 75.0-84.0 Mean D.I. 1.5 Buteogallus aequinoctialis Wing 315.5 ± 0.71 (2) 322 (1) 315-316 Tail 155.0 ± 2.83 (2) 155 (1) 153-157 Exposed culmen 23.55 ± 0.92 (2) 16.8 (1) 22.9-24.2 Tarsus 74.5 ± 3.54 (2) 72.8 (1) 72-77 Buteogallus urubitinga Wing 384.94 ± 16.68 (16) 389.63 ± 18.36 (8) -0.61 12 0.555 ns 0.1 362-412 365-415 Tail 225.13 ± 13.50 (16) 234.63 ± 17.54 (8) -1.35 11 0.206 ns 4.1 190-250 210-260 Exposed culmen 29.72 ± 1.05 (16) 30.78 ± 2.20 (8) -1.84 8 0.103 ns 3.5 26.7-30.9 27.2-34.2 Tarsus 112 ± 8.25 (16) 113.16 ± 7.76 (8) -0.11 14 0.910 ns 1.0 85.9-118.9 98.8-123.0 Mean D.I. 2.2 ® Significance, * = P < 0.05, ns = not significant. D.I. = Dimorphic Index (Storer 1966). December 2005 Cuban Black-Hawk Status and Biology 355 Table 2. Mean, standard deviation, and sample size (parentheses) for wing chord, tail, culmen, and tarsus length for mainland {Buteogallus a. anthracinus) and Cuban {Buteogallus a. gundlachii) populations of the Common Black- Hawk. Statistical analyses are within-sex comparisons (two-sample <-test; equal variances not assumed) between main- land and Cuban specimens, except for tarsus, for which we found no sexual size dimorphism. Structure Sex Taxon B. A. ANTHRACINUS B. A. GUNDLACHII t df P Signieicance^* Wing M 371.69 ± 11.95 (16) 342.71 ± 12.16 (7) 5.28 11 <0.001 F 385.19 ± 11.21 (21) 363.00 ± 8.43 (5) 4.94 7 0.002 SH Tail M 195.50 ± 7.40 (16) 179.29 ± 9.12 (7) 4.14 9 0.003 :)4 F 213.81 ± 10.61 (21) 191.60 ± 22.16 (5) 2.14 4 0.099 ns Exposed M 26.27 ± 0.82 (16) 25.32 ± 0.69 (7) 2.86 13 0.013 culmen F 27.40 ± 1.25 (20) 27.54 ± 0.61 (5) -0.36 13 0.728 ns Tarsus M and F’’ 85.64 ± 3.45 (37) 80.57 ± 3.49 (11) 4.24 16 0.001 ® Significance, * = P < 0.05, ns = not significant. ^ Male and female tarsus data combined because specimens did not display sexual size dimorphism. brown ground color. Remiges are dark brown, with wing coverts edged in grayish-cinnamon, especially the secondaries. The undersides of primaries and some secondaries have an extensive white patch, which constitutes the most distinctive character of the Cuban form. In subtilis, and especially anthra- cinus, this patch is mottled with grayish-brown. The tertiaries of gundlachii are heavily mottled grayish. This mottling is similar to the coloration of the primaries and secondaries of anthracinus, which has only an inconspicuous whitish patch on the un- dersides of these feathers. On the other hand, some specimens of subtilis display more white in this region than does anthracinus, but do not ap- proach the amount shown in gundlachii. The upperparts in gundlachii are also brown, with brownish-gray or with a trace of cinnamon on the feather margins. The head and pileum are uni- formly chocolate brown. The rectrices are darker brown, almost blackish, with a broad white band of variable width (averaging 40 mm) in the middle of the tail. The tip of the tail is edged in white (as wide as 13 mm), which is a purer white than in anthracinus and subtilis. The feet and cere are yel- low, the claws are black, and the iris is dark brown. The bill is blackish at the tip, becoming more yel- lowish toward the base on maxilla and mandible. Immature morphology. Immature gundlachii individ- uals are not chocolate brown ventrally, but rather whitish, and heavily mottled with brown, having some feathers with considerable beige suffusion. Many feathers are mottled with medallion-like marks, whereas others are marked with elongated blotches, and some with streak-like dashes; these marks are seldom present in fully-feathered im- mature birds. The sides of the face and throat are whitish, speckled with hrown. The pileum, nape, and neck are heavily mottled or spotted with brown on a light (white or beige) background. Flanks and thighs also display considerable varia- tion, with younger birds showing a lighter (whitish to brownish-beige) background, whereas older birds display more mottling or barring. The thighs are distinctly barred with light and dark bands in subtilis and anthracinus, whereas gundlachii has mot- tled or very lightly barred thighs. The white patch of the underside of primaries is even more expanded and conspicuous in subadult than in adult gundlachii. Also, the subadult’s tail is distinct from that of the adult’s tail. When still not in full adult plumage, the subadult’s tail shows remnants of several (as many as nine) thin, brown- ish bands, instead of displaying a single broad white band in the middle of the tail as in the adult. Some bands are complete, whereas others are somewhat broken. In Cuban birds, these bands are straight and parallel, whereas in the other forms they are oblique (chevron-like), as well as being much wider than in gundlachii. The bands become less delimited toward the tip; compared with the adult, the white tip of the subadult’s tail is less de- marcated, more grayish than white, and becomes browner from the tip toward the base. Natural History. Habitat. Although we occasion- ally observed black-hawks within the white sand palm savanna of Los Indios, Isla de Pinos, nearly all observations were made in the coastal zone, pri- marily in mangrove forests or at the edges of that 356 Wiley AND Garrido VoL. 39, No. 4 Male Female 420 410 400 O) 390 5 380 370 360 350 23.5 24.5 25.5 26.5 27.5 28.5 29.5 30.5 23.5 24.5 25.5 26.5 27.5 28.5 29.5 30.5 Culmen Culmen 230 220 210 _ 200 <0 ^ 190 180 170 160 75 80 85 90 Tarsus Figure 1. Plots contrasting body measurements of specimens of mainland Buteogallus anthracinus anthradnus (solid dots; = 16 males, 21 females) and Cuban B. a. gundlachii (open circles; N = 7 males, 5 females). habitat. Hawks hunted in the sparsely-vegetated mangrove pannes and flooded openings, where they foraged by perching in young or dead man- groves. We also saw black-hawks foraging or roost- ing in beach and coastal habitats, frequently perch- ing in windbreaks of Casuarina equisetifolia at the edge of mangroves and dirt roads. Nidification. We examined eight nests at Los In- dios within the period of 14-27 May 1996-98. All contained eggs, except the nest examined on 27 May 1996, which had one chick. During our obser- vations at Isla de Pinos, which were well into the breeding season, we observed no aerial courtship, although individual gundlachii regularly soared si- lendy for short periods above their nesting areas. Of the eight gundlachii nests we examined at Los Indies, half were placed in black-mangroves and half in red-mangroves (Table 3) . Each of the nests Table 3. Nest and contents data for eight Buteogallus anthracinus gundlachii nests at Los Indios, Pinos, Cuba, 1996-98. December 2005 Cuban Black-Hawk Status and Biology 357 00 00 03 O) CM CM J> 03 O) pa S H z oo to 03 03 CM I I- 00 m 00 CM CO CM X UO to CO g CM S X C.O) th fti pS o 03 I-H 30 ^ ^ CM 3T3 O TfH to to ts * i ts ■r-i c ' C»i ts • esi S I s> CM CM GO CM X X 30 lO o O 03 ^ ^ CM CO to to to to CM 30 30 30 O 30 CM 00 i> ^ w CO CN| X X 00 30 to 03 30 30 30 30 03 I-H 30 03 03 I—* to 03 GO CM X X w ^ ^ to Th CO 03 CM 30 30 to 30 i Tf oo 03 Tft 03 00 r-H I t- ■S3 30 CO 00 00 30 03 00 rH U rH 03 CM 1 — i Tft Tft X X ^O CO 30 30 CT3 O CM 30 30 30 to s s •r* O 30 1> CM 'tfi X CO to 30 to ? 13 H Z W •G S U B ij ji (X Z lU a o O CO N bO be W VI m d bc be W « I f ts s be ts •«»» s s *<«* U U U G § bio V .o was in the subcanopy, shaded by foliage and was constructed completely of Avicennia and Rhizophora twigs. The nests showed a large range of sizes, probably the result of additions made in successive years. Two of the nests we monitored from 1996 through 1998 were reused by black-hawks, and in- creased in size with the addition of more nest ma- terials in subsequent years. All nests examined at Los Indios contained fresh or older lining materi- als, consisting of green leaves and sprigs of Avicen- nia and Rhizophora, and some debris. Both adults were observed bringing green lining material to nests. Nests had notably deep bowls (Table 3) and when adults were on nests incubating or brooding, they remained low in the bowl and were difficult to detect. During our inspections of nests at Los Indios, adults at three nests regularly perched plac- idly within 2 m of us while we measured eggs and chicks. Adults at a fourth nest were somewhat more aggressive, but the pair only flew low above our heads, occasionally calling, and vocalized from a nearby perch while we measured eggs. We measured 11 eggs at Los Indios (Table 3). Three eggs collected by O. H. Garrido in Cayo Candles (Archipielago de los Canarreos; deposited at Instituto de Ecologia y Sistematica) measured 55.16 X 44.1 mm, 55.8 X 42.6 mm, and 57.08 X 42.34 mm. The 14 gundlachii eggs we measured av- eraged 55.87 ± 0.69 (range = 54.7-57.08) X 42.71 ± 0.62 (range = 41.9—44.1) mm. Eggs of gundla- chii are typically short sub-elliptical to elliptical, with a finely granulated texture. Eggs have a dull grayish-white ground color, sometimes with a greenish or bluish cast early in incubation, and are marked with spots and blotches of dark or reddish- brown, particularly at the larger end. Clutch sizes at Los Indios averaged 1.57 ± 0.53 {N — 8; range = 1-2) eggs (Table 3). The egg of gundlachii is usually more colored (bluish to greenish suffu- sion) than those of anthracinus or subtilis, which are typically grayish or whitish (Bent 1937, Wetmore 1965, O. Garrido pers. obs.). Diet and foraging behavior. Cuban birds were found to feed on a variety of prey (Table 4) . No- table was the lack of fish prey, although fishes were available in tidal channels in the study area. How- ever, twice, hawks were observed wading in shallow tidal channels and making foot thrusts at probable fish prey. During our observation periods (May- June) at Los Indios, land crab populations were particularly high, and crabs were active and con- 358 Wiley AND Garrido VoL. 39, No. 4 Table 4. Prey of Buteogallus anthracinus gundlachii at Los Indios, Isla de Pinos, Cuba, 1996—1998, and Cienaga de Zapata, Cuba, 1999-2000. Prey Number (%) Observed Brought TO Nest Prey Remains Observed Captures Total (%) All Observations Invertebrates Crab Cardisoma guanhumi 4 12 2 18 (64.3) Ucides cordatus 1 1 2 (7.1) Centipede sp. 1 1 2 (7.1) Totals (invertebrates) 6 (21.4) 14 (50.0) 2 (7.1) 22 (78.6) Vertebrates Reptiles Lizards Anolis spp. 1 1 2 (7.1) Ameiva auberi 1 1 (3.6) Totals (reptiles) 2 (7.1) 1 (3.6) 3 (10.7) Birds Sora Porzana Carolina 1 1 (3.6) Totals (birds) 1 (3.6) 1 (3.6) Mammals Rattus rattus 2 2 (7.1) Totals (mammals) 2 (11) 2 (7.1) Total (vertebrates) 2 (7.1) 4 (14.3) 6 (21.4) Total (all observations) 8 (28.6) 18 (64.3) 2 (7.1) 28 spicuous in the early mornings and evenings, when most of our observations of prey delivery and cap- tures were made. In December 1999, we also ob- served gundlachii capturing several crabs {Cardiso- ma guanhumi) along the coast of Cienaga de Zapata, where the hawks hunted from a mixed mangrove- equisetifolia-codistal scrub zone. During our observations in the Los Indios man- grove habitat, gundlachii displayed passive still hunting from low (x = 1.3 ± 0.94; range = 0.2-3 m; N = 54) mangrove tree perches or from the ground. Prey captures were made in a low-angle flight, snatching the item (all observations of crabs) and continuing to a nearby perch, or the hawk landed near the crab and stalked it on foot. Once the hawk grasped the crab, it controlled the claws and legs on either side of the prey with its feet, then removed the carapace with a quick tug at the head region using its bill. We found apparent caches of uneaten, though dismembered, land crabs near (range = 5-20 m) used gundlachii nests. However, we did not observe hawks returning to the caches to feed on the stock- piled crabs. Although B. a. anthracinus has been observed (O. Garrido pers. obs.) in Mexico hunting at the edge of a meadow in a fashion similar to that of the coursing behavior of the Northern Harrier {Circus cyaneus), gundlachii was not observed for- aging aerially in an active manner. Vocal behavior. The common call of gundlachii is a series of three or, uncommonly, four notes, with emphasis on the first two elements, suggesting its Cuban common name, BA-TIS-ta (Gundlach 1893, Garrido and Schwartz 1969, Garrido and Kirkcon- nell 2000; Fig. 2A). The call has a much shorter duration and fewer elements than in other popu- lations of Buteogallus anthracinus (Table 5). The common call of mainland anthracinus consists of 9- 24 notes, with the middle to the final third of the notes accentuated (Fig. 2C-F, Table 5). Stiles and Skutch (1989) characterized the call of mainland anthracinus as ''klee klee klee KLEE KLEE klee kle kle keki ki." The comparable call of cancrivorus con&i&Xs, December 2005 Cuban Black-Hawk Status and Biology 359 Figure 2. Sonographs of common (alarm) calls of Buteogallus species. A. Buteogallus anthradnus gundlachii, showing typical three element “ba-tis-ta” phrase, Cuba (G.B. Reynard), B, Buteogallus a. cancrivorus, St. Vincent (J. Roche, courtesy British Library Sound Archive) . C, Buteogallus a. anthradnus, Costa Rica (Cornell Library of Natural Sounds 27216). D. Buteogallus a. anthradnus, Venezuela (R Schwartz). E. Buteogallus a. anthradnus, male, Arizona (courtesy J. Schnell). F. Buteogallus a. anthradnus, female, Arizona (courtesy J. Schnell). G. Buteogallus urubitinga, Venezuela (P. Schwartz) . H. Buteogallus aequinoctialis, Surinam (Paul Donahue, courtesy British Library Sound Archive) . Table 5. Characteristics {x ± SD, range in parentheses) of the common (advertisement) call of Buteogallus anthraanus anthraanus, B. a. cancnvorus, gundlachii, B. urubitinga, and B. aequinoctialis. Wiley AND Garrido VoL. 39, No. 4 360 53 Q « s C/5 H 2 o cO 50 w OI i> CM rH CM £ Ui S w h-i 1 1-H 1 50 1 00 4i 1 05 1 1 CM cO 05 50 05 M' 00 00 00 m CM CO 00 o I> GO I> CM o CO 'Ch CO iD 00 xi 1 05 Mi 4 4 1 rH 00 rfH 05 o 05 CO CM 05 oo M' o oo OO CM CM CM O ^ ' ' Hi X 50 rH OO 50 05 CM CO OO GO d 50 d CO 00 ^H 05 o M^ CO 'Cfl GO 50 00 oo CM rH +1 +1 + 1 +1 +1 + 1 +1 N CM CM 50 CO o CD 50 CO 50 i> 00 o CO iD 1-H rH G CO CO oo CO GO z w p a 05 V, CO GM 1-H rH rH CM s Ol iC 05 1-H o CO CM 50 00 rH rH CO 00 t 1 1 00 05 1 1 1 CN IM lO 50 oo o OO 05 05 oo o OO 00 i'' > CO 1-H rH 50 00 CO 0 ^H rH CM 05 05 ' ' ' ' CO d 00 o oO 00 CO cb o 05 d d cb CM 05 CM CM 05 CO rH + 1 +l +1 + 1 + 1 + 1 + 1 05 CO M^ CM CO 1-H CM 05 o CO 50 1> CO t£5 50 50 j> cO 1-H rH rH 1-H rH 1-H CM ,^-v M-. lO CM CO GO z w rH 1 f-H 1 00 ch 1 05 rH rH 1 05 CM CM 1 M^ V ^ o o 05 rH CO u o' CO 50 50 50 o d d +1 +1 + 1 +1 + 1 + 1 + 1 Z 1-H 50 CM 50 50 o CO CO CM CO CM d 1-H I> rH rH rH CM M^ CO s. 05 00 oo 50 H q oo cb od 1 rH z 0 1 00 4^ 1 1 05 1 o xO CM 1 05 c d> CM CM 1-H 50 rH d d '• — ^ ' ^ ^ ^ ^H 00 o CD rH o d rH rH rH rH d Q +1 +1 + 1 +1 +1 +1 + 1 o> oo 50 CO 00 00 o C5 CO CM I> CM i-H J> 50 CM GO CO lO ^ ^ — .. u r^ cd (2 d 0 0 u N C/D fi H 0 < u > '' 0 * Hi ‘ Hi << S s <<: S S ‘G “3 s fi H s b 53 ■53 e g 1 s s •Hi o s 3 S 53 ?3 53 2 ■ri S <3 53 53 53 e 53 =q cq cq «sq oq =q oq cl o c u B t3 a d ij-i of a large number of elements (22-37), with em- phasis on several middle elements (Fig. 2B, Table 5). Similarly, the common call of B. subtilis is sub- stantially different from that of gundlachii, consist- ing of several, rapidly repeated elements, described by Ferguson-Lees and Christie (2001) as a series of shrill whistles, indistinguishable from anthracinus. The call of Buteogallus urubitinga consists of a single note, drawn out in a high shrill keeeeeeeeh" (Fer- guson-Lees and Christie 2001; Fig. 2G), whereas that of Buteo aequinoctialis is a distinct series of whis- tle-like notes (Ferguson-Lees and Christie 2001; Fig. 2H, Table 5). Discussion As is normal among most birds of prey, female gundlachii are somewhat larger than males, with culmen and wing length significantly different be- tween genders. Sexual size dimorphism was less ev- ident in anthracinus {x Dimorphic Index = 4.0) than gundlachii, where we found a mean Dimor- phic Index of 4.7 with males significantly larger than females in wing and culmen length (Table 1). Snyder and Wiley (1976) reported a lower index (2.7) of sexual size dimorphism for B. anthracinus. Whereas measurements of selected body parts did not show complete distinction between anth- racinus and gundlachii (Table 2, Fig. 1), Cuban birds were consistently smaller or at the small end of the range for anthracinus measurements. In con- trast to our measurements. Bangs (1905) partly based his determination of separating gundlachii from anthracinus on the former being slightly larg- er than the latter, and in having a decidedly heavi- er, broader bill. As a general pattern, Schnell (1994) noted that Common Black-Hawks of conti- nental (inland) North and Central America are largest. Mainland anthracinus populations inhabit- ing mangrove habitat tend to be smaller and browner than others. The race B. subtilis rhizopho- rae, which inhabits mangrove habitat (Monroe 1963, 1968, Blake 1977), shows a dark-brown plum- age. Our observations revealed that Cuban birds, also mangrove inhabitants, are consistently brown- er with substantial differences in plumage pattern compared with mainland birds. Thus, such color differences may be a result of ecological parallel- ism, rather than of phylogenetic relationships. The species of Buteogallus are partial to wetlands, swampy woods, and seacoasts (Amadou 1982). In its mainland range, anthracinus has been charac- terized as inhabiting woodlands around coastal December 2005 Cuban Black-Hawk Status and Biology 361 swamps, ponds, and streams, and especially man- groves in the swampy woodlands adjacent to the poorly-drained inlands that are affected by tide- waters (Phillips et al. 1964, Wetmore 1965, Davis 1972, Schnell 1994). Wetmore (1965) noted that along large rivers they extend their range farther inland. Thomas (1908) reported anthracinus in stretches of sand dunes and savannas with clumps of palmettos and pines. The Cuban population shows a similar preference for lowland coastal ar- eas. Gundlach (1893) and Bangs (1905) noted gundlachii was found only in mangrove swamps and on the banks of large rivers. In broad contrast, the other West Indian population, Buteogallus anthraci- nus cancrivorus of St. Vincent, mainly keeps to the high wooded valleys, although it seldom occurs far from water (Lister 1880, Clark 1905b, Bond 1956a). Cuban populations of the black-hawk breed from January through June (Garrido and Kirkcon- nell 2000) , with egg-laying occurring in late March or April. Bangs (1905) collected a female contain- ing a soft-shelled egg and found another tending a nest on 15 April. Bond (1950) reported a nest with a newly-hatched chick on 4 April. Garrido and Schwartz (1969) and Valdes Miro (1984) com- mented gundlachii builds its nest at a considerable distance above the ground. Gundlach (1876) re- ported a nest at 8 “varas” (6.8 m), whereas Bond (1936) noted one at 6.2 m. Nests of the Cuban form are typically rough structures of twigs, lined with green leaves and, sometimes, debris (Gundlach 1893, Bond 1936, Garrido and Schwartz 1969, Valdes Miro 1984). Bond (1936), describing nests found in St. Vincent {B. a. cancrivorus) and Cuba {gundlachii), noted, “The nest, a rough mat of sticks, is placed at vari- ous elevations in trees.” All nests located by us at Los Indios in 1996-98 were in mangroves {Avicen- nia, Rhizophora). In contrast, Bond (1936) de- scribed black-hawk nests in St. Vincent as “placed on top of clumps of mistletoe and were rather small.” As Bond (1936) suggested, nests of the Cu- ban species are somewhat larger than those of birds in St. Vincent. Schnell (1994) gave the di- mensions of mainland anthracinus nests as ranging from 38 cm diameter X 20 cm deep to 1.2 m di- ameter X 0.67-1.2 m deep. Bangs (1905) and Bond (1936) also noted gundlachii re-used nests in more than one season, which we believe accounts, in part, for the larger nest size of Cuban birds. Black-hawks at Los Indios were remarkably non- aggressive toward humans at their nests and al- lowed us to approach much closer than other local raptor species tolerated, perhaps relying on their cryptic behavior to avoid detection at the nest. Others have also noted this tolerance in Cuban black-hawks (Todd 1916, Barbour 1923, Garrido and Schwartz 1969). Schnell (1994) reviewed available egg specimens for Buteogallus anthracinus, summarizing mean measurements from Bent (1937) as 57.3 X 44.9 mm {N = 60 eggs) and examples in the Western Foundation of Vertebrate Zoology as 57.30 X 45.50 mm (V = 12 clutches, 19 eggs; range = 52.61- 62.02 mm length, 42.69-47.35 mm breadth). Eggs of anthracinus we measured at the Delaware Mu- seum of Natural History averaged 57.46 (53.1- 63.2) X 45.25 (41.7-49.1) mm {N = 13 clutches, 21 eggs). Interestingly, an egg reported from St. Vincent is at the high end for the species: 61 X 47 mm (Bond 1936) and exceeds the range for gund- lachii. Eggs of gundlachii we measured at Los Indios averaged only slightly smaller than those of main- land B. anthracinus analyzed by Schnell (1994). Gundlach (1876) reported that Cuban eggs mea- sured 58 X 45 mm, whereas Bangs (1905) reported 56 X 45.5 mm. Measurements presented by Valdes Miro (1984) are obviously in error; i.e., x = 56.0 (range = 55.0-57.0) X 24.6 (23.0-26.5) mm. The mean mass (61.0 ± 1.8 g) of eggs we measured at Los Indios was somewhat lighter compared with SchnelFs estimated mean mass of 63.8 g for anth- racinus. Although we observed differences in egg color- ation and pattern among anthracinus, subtilis, and gundlachii, these characters show considerable var- iation and do not appear to be a good character for determining relationships (L. Kiff pers. comm.). Schnell (1994) noted that, in general, clutch size of Buteogallus anthracinus decreased from two eggs in the northern range to one in the southern range; several reported three-egg clutches were questionable. Clutch sizes at Los Indios fell within that range, averaging 1.57 eggs per clutch. Buteogallus anthracinus feeds mainly on inver- tebrates and lower vertebrates, with occasional small birds or mammals in the diet (Schnell 1994). Eor mainland populations, Thomas (1908) report- ed anthracinus preying on burrowing land crabs, which form almost the sole diet of the hawks in British Honduras (Belize). The St. Vincent popu- lation {B. a. cancrivorus) reportedly feeds on cray- 362 Wiley AND Garrido VoL. 39, No. 4 fish and freshwater crabs (Lister 1880). In Cuba, Gundlach (1893) reported remains of crustaceans, as well as frogs, snakes, and fishes in the stomachs of black-hawks. Barbour (1943) reported land crabs as its prey in Cuba. Garrido and Kirkconnell (2000) reported its prey as mainly crabs and birds, whereas Ramsden (C. Ramsden, Museo de His- toria Natural, Universidad de Oriente, Santiago de Cuba unpubl. data) noted the hawk fed on crabs and fishes. The hunting behavior of Buteogallus, in general, has been characterized as sluggish. Schnell {in Palmer 1988, 1994) noted B. anthracinus normally hunts from a stationary perch, often near the ground, from branches up to 15 m high, on boul- ders, other low perches, and gravel beds along streams. For Cuban hawks, Barbour (1923) de- scribed crab predation similar to our observations: “The hawk pounces on the crab, gathers the legs and claws of each side in one of its feet, and reach- ing down removes the carapace by hooking the bill under its front edge.” Kirkconnell and Garrido (1991) reported gundlachii drowning its avian prey (Common Moorhen [Gallinula chloropus]), which they suggested was unusual and perhaps related to the abundant rain that caused the raising of the water level in the swamp, rendering crabs difficult to find. We observed Cuban Black-Hawks caching crab prey near their nest, a habit that has also been re- ported for B. anthracinus in mainland sites (Thom- as 1908, Schnell 1991, 1994). As noted by Schnell (1994), descriptions of the vocal behavior of Buteogallus anthracinus have been confusing and conflicting. Schnell (1994) charac- terized the common call (= alarm call) as of a complex, un-raptor-like quality. The common call of mainland Buteogallus anthracinus is distinct from the three-note call of gundlachii, consisting of 9-24 notes (Reynard and Garrido 1988, Schnell 1994) (Figs. 2A, 2C— F, Table 5). Similarly, the common call of B. subtilis is distinct from that of gundlachii, consisting of several, rapidly-repeated elements, described by Ferguson-Lees and Christie (2001) as a series of shrill whistles, indistinguishable from that of anthracinus. The call of Buteogallus aequinoc- tialis is a series of six or seven whistle-like notes, the first three rapid, followed by slower and de- scending elements (Fig. 2H; Ferguson-Lees and Christie 2001). Finally, B. meriodionalis has a call consisting of a prolonged whistle, described as “eeeeee-eh” or ''kree-ee-ee-er” (Ferguson-Lees and Christie 2001). Conclusions We consider Buteogallus anthracinus (with its geo- graphical races, cancrivorus and anthracinus), B. urubitinga, B. aequinoctialis, and B. gundlachii as sep- arate species. This treatment of the Cuban popu- lation agrees with Wetmore (1965:234), who stated the other forms stand apart: “. . . from the bird of the island of Cuba which it appears appropriate to treat as a separate species, Buteogallus gundlachii.” Thus, the Cuban Black-Hawk Buteogallus gundlachii Cabanis, 1854 (1855), becomes a species endemic to Cuba, distributed in the main island, where it is relatively uncommon and quite localized, Isla de Pinos, and many of the keys of the Cuban Archi- pelago. Acknowledgments A grant from the American Museum of Natural History allowed Garrido to undertake studies of West Indian birds in its and other United States Museums. We grate- fully acknowledge help and support from the RARE Cen- ter for Tropical Conservation, through John Guarnaccia and Victor L. Gonzalez, enabling visits to collections in several museums. We also thank the curators of the American Museum of Natural History, the Smithsonian Institution, the Museum of Comparative Zoology, Loui- siana State University Museum of Natural History, Acad- emy of Natural Science, Delaware Museum of Natural History, Instituto de Ecologia y Sistematica, and Museo Nacional de Historia Natural de Cuba for their cooper- ation. Wiley’s research in Isla de Pinos was supported by grants from Wildlife Preservation Trust International, World Parrot Fund, International Crane Foundation, and the U. S. Department of the Interior. We are particularly grateful to Lie. Xiomara Galvez, Empresa Nacional para la Conservacion de la Flora y Fauna, for logistical support and field assistance at Los Indios. For use of tape record- ings of Buteogallus vocalizations, we thank George B. Rey- nard, Paul Schwartz, L. Irby Davis, R. Grimshaw, J Schnell, Richard Ranft, and the British Library Sound Archive. We thank Lloyd F. Kiff, Jay H. Schnell, and an anonymous reviewer for their helpful suggestions, which greatly improved the manuscript. Literature Cited Alain, H. 1946. Notas taxonomicas y ecologicas sobre la flora de Isla de Pinos. Contr. Ocas. Museo Hist. Nat. Coleg. La Salle 7:11-115. Aldrich, J.W. and B.P. Bole, Jr. 1937. The birds and mammals of the western slope of the Azuero Penin- sula, Republic of Panama. Sci. Publ. Cleveland Mus Nat. Hist. 7:1-196. Amadon, D. 1961. Remarks on the genus Buteogallus. Nov Colombianas 1:358-360. December 2005 Cuban Biack-Hawk Status and Biology 363 . 1982. A revision of the sul>Buteonine hawks (Ac- cipitridae, Aves). Am. Mus. Nov. No. 2741:1-20. American Ornithologists’ Union. 1998. Check-list of North American birds, 7th Ed. American Ornitholo- gists’ Union, Washington, DC U.S.A. Bangs, O. 1905. The Cuban Crab Hawk, Urubitinga gund- lachii (Cabanis). Auk 22:307-309. and W.R. Zappey. 1905. Birds of the Isle of Pines. Am. Nat. 39:179-215. Barbour, T. 1923. The birds of Cuba. Mem. Nuttall Or- nithol. Club 6. . 1943. Cuban ornithology. Mem. Nuttall Omithol. Club 9. Bent, A.C. 1937. Life histories of North American birds of prey. Part I. U.S. Nat. Mus. Bull. 167. Blake, E.R. 1977. Manual of Neotropical birds. Vol. I. University Chicago Press, Chicago, IL U.S.A. Bond, J. 1936. Birds of the West Indies. Acad. Nat. Sci. Philadelphia, PA U.S.A. . 1950. Check-list of birds of the West Indies. Acad. Nat. Sci. Philadelphia, PA U.S.A. . 1956a. Check-list of birds of the West Indies, 4th Ed. Acad. Nat. Sci. Philadelphia, PA U.S.A. . 1956b. First supplement to the Check-list of birds of the West Indies (1956). Acad. Nat. Sci. Philadel- phia, PA U.S.A. Brown, L. and D. Amadon. 1968. Eagles, hawks and fal- cons of the world. Vol. 2. McGraw-Hill Book Co., New York, NYU.S.A. Cabanis, J.L. 1855. Dr. J. Gundlach’s Beitrage zur Ornit- hologie Cuba’s. J. Ornithol. 2 (extra-heft 1854 [pub- lished 1855]):77-87. Clark, A.H. 1905a. The Crab Hawk {Urubitinga) in the island of St. Lucia, West Indies. Auk 22:210. . 1905b. Preliminary descriptions of three new birds from St. Vincent, West Indies. Proc. Biol. Soc. Wash. 18:61-63. . 1905c. Birds of the southern Lesser Antilles. Proc. Boston Soc. Nat. Hist. 32:203-312. Cory, C.B. 1887. The birds of the West Indies, including the Bahama Islands, the Greater and the Lesser An- tilles, excepting the islands of Tobago and Trinidad. ArA 4:37-51. . 1892. Catalogue of West Indian birds, containing a list of all species known to occur in the Bahama Islands, the Greater Antilles, the Caymans, and the Lesser Antilles, excepting the islands of Tobago and Trinidad, Privately published by the author, Boston, MA U.S.A. Davis, L.L 1972. A field guide to the birds of Mexico and Central America. University Texas Press, Austin, TX U.S.A. Dickinson, E.C. [Ed.]. 2003. The Howard and Moore complete checklist of the birds of the world, rev. and enlarged 3rd Ed. Princeton University Press, Prince- ton, NJ U.S.A. Evans, P.G.H. 1990. Birds of the eastern Caribbean. Mac- Millan Press Ltd., London, England. Ferguson-Lees, j. and D.A. Christie. 2001. Raptors of the world. Houghton Mifflin Co., New York, NY U.S.A. Friedmann, H. 1950. Birds of North and Middle America, Falconiformes. U.S. Nat. Mus. Bull. 50, pt. 11. Garrido, O.H. and a. Kirkconnell. 2000. Field guide to the birds of Cuba. Cornell University Press, Ithaca, NY U.S.A. and a. Schwartz. 1969. Anfibios, reptiles y aves de Cayo Candles. Poeyana 67:44. Gundlach, j. 1854. Beitrage zur Ornithologie Cuba’s. Nach Mittheilungen des Reisenden an Hr. Bez.-Dir. Sezekorn in Cassel; von Letzterem zusammengestellt. Mit Zustzen und Anmerkungen geordnet vom He- rausgeber. /. Ornithol. 2 (extra-heft) :77-87. . 1865-1 866a. Revista y catalogo de las aves Cu- banas. Pages 165-180 in Repertorio Fisico Natural de la Isla de Cuba. Vol. 1 (F. Poey, [Ed.]). Imprenta del Gobierno y Capitania General, La Habana. . 1865-1 866b. Revista y catalogo de las aves Cu- banas. Pages 221-224 in Repertorio Fisico Natural de la Isla de Cuba. Vol. 1 (F. Poey, [Ed.]). Imprenta del Gobierno y Capitania General, La Habana. . 1871. Neue Beitrage zur Ornithologie Cubas. Nach eigenen 30 jahrigen Beobachtungen zusam- mengestellt./. Omithol. 19:353-378. . 1876. Contribucion a la ornitologia Cubana. La Habana: “La Antilla.’’ . 1893. Ornitologia Cubana. La Habana: La Mo- derna. Gurney, J.H. 1876. Notes on a “Catalogue of the Acci- pitres in the British Museum,” by R. Bowdler Sharpe (1874). Ibis, series 3, 6:467—493. . 1934. A list of the diurnal birds of prey, with references and annotations. John Van Voorst, Lon- don, England. Hellmayr, C., and H. Conover. 1949. Catalogue of birds of the Americas. Field Mus. Nat. Hist. Zool. Ser., 13, part 1(4). Jennings, O.E. 1917. A contribution to the botany of the Isle of Pines, Cuba, based upon the specimens of plants from that island contained in the herbarium of the Carnegie Museum under date of October, 1916. Ann. Carnegie Mus. 11:19—290. Kirkconnell, A. and O. Garrido. 1991. Aberrante com- portamiento de caza del Gavilan Batista Buteogallus anthracinus gundlachii (Aves: Accipitridae) en Cuba Vol. Migrat. No. 16:27-28. Kleinbaum, D.G. and L.L. Kupper. 1978. Applied regres- sion analysis and other multivariable methods. Dux- bury Press, North Scituate, MA U.S.A. Lister, C.E. 1880. Field-notes on the birds of St. Vincent, West Indies. Ibis, ser. 4, 4:38-44. Mayr, E. and G.W. Cottrell [Eds.]. 1979. Check-list of birds of the world. Vol. 1, 2nd Ed. (rev. of the work 364 Wiley AND Garrido VoL. 39, No. 4 of Peters, J.L.). Mus. Comp. Zool. Annu. Rep., Cam- bridge, MA U.S.A. AND L.L. Short. 1970. Species taxa of North American birds. Publ. Nuttall Ornithol. Club No. 9. Monroe, B.L., Jr. 1963. Three new subspecies of birds from Honduras. Occas. Pap. Mus. Zool., Louisiana State University 136. . 1968. A distributional survey of the birds of Hon- duras. American Ornithologists’ Union Ornithol. Mo- nogr. No. 7. Washington, DC U.S.A. Palmer, R.S. [Ed.]. 1988. Handbook of North American birds. Vol. 4. Yale Universtiy Press, New Haven, CT U.S.A. Peters, J.L. 1931. Check-list of birds of the world. Vol. 1. Harvard University Press, Cambridge, MA U.S.A. Phillips, A., J. Marshall, and G. Monson. 1964. The birds of Arizona. University Arizona Press, Tucson, AZ U.S.A. Reynard, G.B. and O.H. Garrido. 1988. Bird songs in Cuba/Cantos de aves en Cuba. Cornell Lab. Ornit- hol., Ithaca, NY U.S.A. Ridgely, R.S. AND PJ- Greenfield. 2001. The birds of Ec- uador: status, distribution and taxonomy. Christopher Helm, London, England. Ridgway, R. 1876. Studies of the American Falconidae. Dept. Interior, U. S. Geolog. Geograph. Survey Terri- tories 2:91-182. Schnell, J.H. 1991. A closer look: Common Black-Hawk. Birding 23:282-285. . 1994. Common Black-Hawk, Buteogallus anthraci- nus. In Poole, A.F., and F.B. Gill [Eds.] The birds of North America, No. 122. Acad. Nat. Sci. Philadelphia, and American Ornithologists’ Union, Washington, DC U.S.A. Sharpe, R.B. 1874. Catalogue of the birds in the British Museum. Vol. I. Catalogue of the Accipitres, or diur- nal birds of prey, in the collection of the British Mu- seum. British Museum, London, England. . 1899. A hand-list of the genera and species of birds. Vol. I. British Museum, London, England. Sibley, C.G. and B.L. Monroe, Jr- 1990. Distribution and taxonomy of birds of the world. Yale University Press, New Haven, CT U.S.A. Snyder, N.F.R. and J.W. Wiley. 1976. Sexual size dimor- phism in hawks and owls in North America. American Ornithologists’ Union Ornithol. Monogr. No. 20. SPSS. 1999. SPSS for Windows 10.0.5. SPSS Inc., Chicago, IL U.S.A. Stiles, EG., and A.F. Skutch. 1989. A guide to the birds of Costa Rica. Comstock Pub. Assoc., Ithaca, NY U.S.A. Storer, R.W. 1966. Sexual dimorphism and food habits in three North American accipiters. Auk 83:423-436. Swann, H.K. 1921-1922. A synopsis of the Accipitres (di- urnal birds of prey) comprising species and subspe- cies described up to 1920, with their characters and distribution. 2*“^ edition, revised and corrected throughout. Wheldon and Wesley, London, England. . 1930. A monograph of the birds of prey (Order Accipitres). Vol. I. Part VIII. Wheldon and Wesley, London, England. Thomas, G.B. 1908. The Mexican Black Hawk. Condor 10' 116-118. Todd, W.E.C. 1916. The birds of the Isle of Pines. Incor- porating the substance of the field-notes by Gustav A Link. Ann. Carnegie Mus. 10:146-296. Valdes Miro, V. 1984. Datos de nidificacion sobre las aves que crian en Cuba. Poeyana 282:1-27. Wetmore, a. 1965. The birds of the Republic of Panama, pt. 1. Smithson. Misc. Collect. 150. Received 28 June 2004; accepted 28 July 2005 Associate Editor: Cheryl R. Dykstra J. Raptor Res. 39(4):365-377 © 2005 The Raptor Research Foundation, Inc. HOME RANGE AND HABITAT USE OF NORTHERN SPOTTED OWLS ON THE OLYMPIC PENINSULA, WASHINGTON Eric D. Forsman,i TimmothyJ. Kaminski,^ Jeffery C. Lewis,^ Kevin J. Maurice,^ and Stan G. Sovern^ USD A Forest Service, Pacific Northwest Research Station, 3200 Jefferson Way, Corvallis, OR 97331 US. A. Cheron Ferland and Elizabeth M. Glenn Department of Fisheries and Wildlife, Oregon State University, Corvallis, OR 97331 US. A. Abstract. — ^We studied movements and habitat selection of 20 adult northern Spotted Owls (Strix oc- cidentalis caurina) on the Olympic Peninsula, Washington in 1987-89. Median annual home range size of individual owls was 1147 ha based on the 75% isopleth of the Fixed Kernel (FK), 2406 ha based on the 95% FK, and 2290 ha based on the 100% Minimum Convex Polygon (MCP). Annual ranges of individual owls tracked >1 yr overlapped by a mean of 70-73%, depending on which estimator was used. Size of annual and cumulative ranges was negatively correlated with the amount of old forest within the cumulative MCP home range and within a 4.3 km radius of the center of activity. Overlap of annual ranges of owls that were paired averaged 64 ± 5% based on the MCP and 69 ± 5% based on the 95% FK. On average, ranges used during the nonbreeding season overlapped breeding season ranges by 65.0 ± 4.5%, and breeding season ranges overlapped nonbreeding season ranges by 62.6 ± 4.9%. Compositional analysis of habitat selection indicated that old forests were the most preferred cover type for foraging and roosting and that dear-cuts and non-forest cover types were rarely used. There was little evidence that owls selected riparian areas or forest edges for foraging or roosting. Our observations are consistent with the hypotheses that northern Spotted Owls use larger foraging areas in regions where northern flying squirrels {Glaucomys sabnnus)arG their primary source of food, that they prefer old forests for foraging and roosting, and that their home ranges become larger as the amount of old forest declines. The large size of annual ranges on the Olympic Peninsula may be a response to low prey biomass. Key Words: Northern Spotted Owl, Strix occidentalis caurina; home range, habitat use, radiotelemetry, Olympic Peninsular, Washington. RANGO DE HOGAR Y USO DE habitat DE strix occidentalis CAURINA EN OLYMPIC PEN- INSULA, WASHINGTON Resumen. — Estudiamos los movimientos y la seleccion de habitat de 20 individuos adultos de Strix occi- dentalis caurina en Olympic Peninsula, Washington, entre 1987 y 1989. La mediana del tamaho del area de hogar de un individuo fue de 1147 ha basada en la isolinea de 75% del kernel fijo (KE), 2406 ha basada en el 95% KF y 2290 ha basada en el 100% del poligono convexo minimo (PCM). Los rangos anuales de los individuos seguidos por menos de un ano se superpusieron en promedio entre un 70% y un 73%, dependiendo del estimador que usamos. Los tamanos de los rangos anuales y acumulativos se correlacionaron negativamente con la cantidad de bosque maduro presente dentro del PCM acu- mulativo del rango de hogar y a menos de 4.3 km del centre de actividad. La superposicion promedio de los rangos de hogar anuales de individuos que confer maban parejas fue de 64 ± 5% basado en el PCM y 69 ± 5% basado en el 95% del KF. En promedio, los rangos usados durante el periodo no reproductive se superpusieron con los rangos del periodo reproductive en 65.0 ± 4.5%, y los rangos del periodo reproductive se superpusieron con los rangos del periodo no reproductive en 62.6 ± 4.9%. ^ Email address: eforsman@fs.fed.us ^Present address: U.S. Forest Service, Bridger-Teton National Forest, P.O. Box 1888, Jackson, WY 83001 U.S.A. ^ Present address: Washington Department of Fish and Wildlife, 600 Capitol Way North, Olympia, WA 98501 U.S.A. Present address: U.S. Fish and Wildlife Service, Portland Field Office, 2600 SE 98*, Portland, OR 97266 U.S.A. ’’ Present address: Department of Fisheries and Wildlife, Oregon State University, Corvallis, OR 97331 U.S.A. 365 366 Forsman et al. VoL. 39, No. 4 Los analisis de composicion de los ambientes seleccionados indicaron que los bosques maduros fueron el dpo de cobertura preferida para alimentarse y reposar, mientras que las areas completamente taladas y no boscosas fueron usadas en muy pocas ocasiones. Encontramos muy poca evidencia de que las lechuzas seleccionan las areas riparias o los hordes de bosque para alimentarse o reposar. Nuestras observaciones son consistentes con la hipotesis de que S. o. caurina usa grandes areas de forrajeo en las regiones donde las ardillas voladoras {Glaucomys sabrinus) son su fuente principal de alimento, que prefieren bosques maduros para alimentarse y reposar y que sus areas de hogar aumentan a medida que disminuye la cantidad de bosque maduro. El gran tamano de los rangos anuales en Olympic Pen- insula podria responder a una baja biomasa de presas. [Traduccion del equipo editorial] Spotted Owls {Strix occidentalis) exhibit consid- erable variation in home range size and patterns of seasonal movements, both within and among re- gions. For example, in some parts of their range, Spotted Owls may migrate during winter, moving 16-58 km from their breeding season ranges into lowland forests (Laymon 1989, Zabel et al. 1992). In other regions, they are largely resident in the same areas throughout the year (Forsman et al. 1984, Carey et al. 1990, 1992). Home ranges and habitat selection of Spotted Owls have been studied extensively in Oregon and California, but with the exception of a study by Hamer (1988), little information is available from Washington. We examined home ranges and hab- itat selection of northern Spotted Owls on the Olympic Peninsula, Washington to determine if patterns of habitat use differed near the northern edge of the range of the owl compared to earlier studies conducted in Oregon (e.g., Forsman et al. 1984, Carey et al. 1990, 1992, Carey and Peeler 1995) and northern California (Solis and Gutier- rez 1990, Zabel et al. 1992, 1995). Study Area We conducted our study on two areas on the west side of the Olympic Peninsula, one located 3 km SE of the town of Forks, Clallam County, and the other located 10 km SE of the town of Quinault, Jefferson County (Fig. 1). Both areas were located on the Olympic National For- est, had similar climate, topography and vegetation, and will hereafter be referred to collectively as the “study area.” The study area was characterized by mountainous ter- rain covered by forests of western hemlock (Tsuga hete- rophylla) and western redcedar {Thuja plicata). Sitka spruce {Picea sitchensis) was common on mesic, low ele- vation areas, and Douglas-fir {Pseudotsuga menziesii) and Pacific silver fir (Abies amabilis) were often intermixed with western hemlock on upland sites (Henderson et al. 1986). Elevations ranged from 150-1500 m. Precipitation ranged from 280-460 cm/yr, mostly falling as rain during October-May. The area included a mosaic of serai stages, ranging from clearings in which all trees had been recently har- vested (clear-cuts) to old-growth forests in which oversto- ry trees were over 500 yr old (Henderson et al. 1986). Approximately half of the area had been clear-cut within the previous 30 yr, but harvested areas were not uniform- ly distributed within the study area. Some areas were heavily fragmented by recent clear-cuts, whereas other ar- eas had extensive blocks of mature and old-growth forest. Much of the study area was hit by hurricane-force winds in 1921 which severely damaged many stands (Pierce 1921). As a result, many stands included a mixture of 60- 80-yr-old trees that regenerated after the wind event, in- terspersed with old trees (80-500+ yr) that survived the windstorm. All types of natural (unlogged) forest typical- ly had high canopy closure (65-80%), high variation in tree size and age, and high volumes of logs and snags (Henderson et al. 1986). Regenerating stands of young trees in clear-cut areas were usually even-aged, with high canopy closure. Methods Capture and Radio-marking. We captured owls with noose poles (Forsman 1983) and marked them with back-pack transmitters (Model P2, AVM Instrument Company, Livermore, CA U.S.A.), as described by Fors- man et al. (1984). Total mass of transmitter and harness was 18-20 g, and transmitter life was 9-15 mo. We tried to obtain a minimum of 12 mo of data from each owl. We replaced transmitters on six individuals after 9-12 mo, and tracked them for nearly 2 yr. Sampling Schedule. We attempted to obtain one noc- turnal foraging location per night on each owl at least 3 nights per wk, and one diurnal roost location per owl at least 3 d per wk. Our sampling schedule was intended to reduce autocorrelation between sequential locations (Swihart and Slade 1985a, 1985b). However, Aebischer et al. (1993) and Otis and White (1999) have suggested that autocorrelation is generally irrelevant when individual animals are used as the sample unit in home-range stud- ies, so we used all of our data, including a few cases (129 of 7346 locations) when we obtained 2-3 locations on the same owl in one night. We classified all locations as foraging locations if they occurred from 0.5 hr after sun- set to 0.5 hr before sunrise. We excluded locations of incubating or brooding females from analyses of habitat selection, until females began to forage when the young were about 2 wk old. Radio Triangulation. We estimated owl locations by tri- angulating with a Telonics hand-held H-antenna and TR2 receiver (Telonics, Mesa, AZ U.S.A.). We used a hand- held compass to estimate azimuths from ^3 locations December 2005 Spotted Owl Habitat Use 367 Figure 1. Location of radiotelemetry study areas on the Olympic Peninsula, Washington, 1987-89. ) 1 I along roads (Guetterman et al. 1991). Azimuths were plotted on 1:12 000 or 1:24000 scale U.S. Geological Sur- vey orthophotos or topographic maps. We considered the position of the owl to be the geometric center of the polygon formed by the intersection of ^3 bearings (Nams and Boutin 1991). If weak signals or inconsisten- cies in the direction of bearings caused us to suspect sig- nal deflection or movement of an owl during triangula- tion, we discarded the location. We used all locations to estimate home ranges, but only locations with error poly- gons ^8 ha were used for analyses of habitat use. Telemetry Error. We estimated telemetry error with 63 blind trials in which one observer placed transmitters in trees in owl home ranges and another observer then tri- angulated on the transmitters at night. The median dis- tance between estimated and actual transmitter locations was 100 m (x = 140 ± 17 m). This estimate was similar to or less than error estimates in previous telemetry stud- ies of Spotted Owls (Carey et al. 1990, Glenn et al. 2004). Errors of this magnitude undoubtedly resulted in some locations falling in the wrong cover types, but we made the assumption that classification errors due to telemetry error were similar in all cover types, and that our overall assessment should reflect actual habitat use. Home-range Estimation. We estimated cumulative and annual ranges with the Minimum Convex Polygon (MCP) and Fixed-Kernel (FK) methods (Hayne 1949, Seaman and Powell 1996). For estimates of MCP ranges, we used 100% MCP polygons. For FK estimates, we used 95% and 75% isopleths, which we interpreted as the “home range,” and “area of concentrated use,” respec- tively. We used Program CALHOME (Kie et al. 1996) to estimate MCP ranges and Version 4.28 of Program ICER- NELHR (Seaman et al. 1998) to estimate FK ranges. Con- trary to the recommendation of Seaman and Powell (1996), we used the FK method without least-squares- cross-validation (LSCV). We did so because we believe that kernel estimates based on locations where owls stop long enough for the observer to obtain a location tend to underestimate home range areas of owls (because movements across intervening non-forest areas usually happen so quickly that they cannot be documented with a point on the map) . Thus, we feel that the LSCV option, which tends to fit the home range isopleth more tightly to the observed points, is likely to cause an even greater underestimate of home ranges. We used all locations for MCP estimates, but we only used foraging locations for FK estimates (because FK estimates that include large 368 Forsman et al. VoL. 39, No. 4 Table 1. Vegetation cover types used to map landscapes for analyses of habitat use by northern Spotted Owls on the Olympic Peninsula, Washington, 1987-89. Old Forest: Multilayered stands of western hemlock and western redcedar in which the dominant overstory trees were typically ^100 cm DBH. Pacific silver fir was often subdominant or codominant with hemlock or redcedar. Douglas-fir was codominant on a few areas. Also included mixed-age stands of mature and old forest in which both age classes were common. Many of the latter stands were the result of a hurricane force windstorm in January 1921 (Pierce 1921). Mature Forest: Conifer-dominated stands in which the overstory trees were typically 50—99 cm DBH. Young Forest: Relatively even-aged stands in which most trees were 31-60 cm DBH. Regenerated on burned areas and old dear-cuts. Mixed-young Forest: Same as Young Forest except with inclusions of mature trees, usually remnants left during previous fires or harvest. Pole-sapling: Single-layered conifer stands in which most trees were 10-30 cm DBH. Mostly young stands regenerat- ing on old dear-cuts. Hardwood/Riparian: Riparian areas dominated by red alder {Alnus rubra), bigleaf maple {Acer macrophyllum) , and variable amounts of western redcedar. Clear-cut/Non-forest: Recent dear-cuts dominated by bare soil, grasses, shrubs or small seedling conifers. Also in- cluded small areas of meadows, gravel pits, and agricultural, or residential areas. numbers of roosting locations clustered at the nest site or central place will underestimate foraging areas during the breeding season) . Estimation of Annual, Cumulative, and Seasonal Rang- es, Although we marked some owls in June or July of 1987, we did not begin regular sampling of most individ- uals until late July or August 1987, For these owls, we estimated the first annual range through the end of July 1988. If they were monitored after July 1988, we com- puted a second annual range for the second year. A few individuals were not marked until fall 1987 or summer 1988, in which case the annual range was estimated for one year only. There was only a weak positive correlation between the number of days in the tracking period and estimates of annual home-range size, regardless of which home range estimator was used (95% FK r^^ ~ 0-221, P = 0.223; MCP r^^ = 0.208, P = 0.253). Therefore, we used all annual ranges for comparisons among owls, re- gardless of the monitoring period. For six owls tracked in both years, we estimated the cumulative range from the union of the annual ranges (range A -I- range B minus the area of overlap). Estimates of home range overlap between years, seasons, pair mem- bers, or owls on adjacent territories were based on the percent of range A overlaid by range B or the percent of range B overlaid by range A. In most cases, we computed overlap of ranges based on three different frames of ref- erence (75% FK, 95% FK, and 100% MCP). For estimates of overlap of seasonal ranges, we only used the 95% FK. For seasonal analysis of home ranges, we divided each year into two phonological periods, the “breeding sea- son” (March— August) , when Spotted Owls nest and feed young, and the “nonbreeding season” (September— Feb- ruary) , when Spotted Owls are largely solitary. Estimates of seasonal ranges were limited to owls tracked sl20 d during the season of interest. Habitat Mapping and Assessment of Habitat Use. We examined second-order habitat selection (i.e., use of dif- ferent forest cover types within the home range of each owl). We developed a cover-type map of the study area that included seven cover types based on structural dif- ferences in vegetation as determined from on-the-ground examination of stands and aerial photo interpretation (Table 1). We visited virtually all stands within the study area on one or more occasions to determine the size and species composition of trees. We did not use canopy clo- sure to differentiate among cover types because nearly all forests on the study area had relatively high (>70%) can- opy closure, regardless of stand age or tree size. Cover types were mapped on 1:12 000 scale orthophotos and digitized into an ARC/INFO (ESRI Inc., Redlands, CA U.S.A.) GIS layer. For convenience, we use the term cover type, even though we recognize that our designation of cover type was based on only one component of habitat (i.e., vegetation structure). Site visits to 403 randomly se- lected grid coordinates indicated tbat map accuracy was 83%. We used compositional analysis (Aebischer et al. 1993) to evaluate relative preference of cover types for foraging and roosting. This method treats the individual as the sample unit, accounts for lack of independence among proportions, is not sensitive to serial correlation between locations, and is based on a unique set of observed and expected values for each cover type in the home range of each individual. Expected use was equal to the pro- portion of the cumulative MCP home range covered by each cover type, and the observed use was the proportion of locations in each cover type. We used Program RSW (Leban 1999) to conduct the analysis. Results of this anal- ysis included a numeric ranking of the different cover types according to their relative “preference,” as well as a table of pair-wise comparisons (Mests) indicating the degree to which preference differed between types. We used paired f-tests to determine if the distribution of foraging or roosting locations differed from random locations relative to elevation, distance to the nearest stream, or distance to the nearest open area (dear-cuts/ non-forest in Table 1). For these analyses, we computed December 2005 Spotted Owl Habitat Use 369 Figure 2. Observation periods of 20 radio-marked northern Spotted Owls observed on the Olympic Peninsula, Washington, 1987-89. Vertical lines indicate intervals used for calculation of seasonal ranges. mean expected values from a random sample of 200 lo- cations in forest areas in the 100% MCP home range of each owl. We used digital stream layers and elevation lay- ers in GIS to compute elevation and distance to the near- est stream for each owl location and each random loca- tion. Based on a preliminary analysis of our data, the Wash- ington State Forest Practices Board (1996) adopted land management guidelines in which they stipulated that land managers should maintain a minimum of 5863 acres (2372 ha) of “suitable habitat” within a 4.3-km radius around Spotted Owl site centers (known or suspected nest areas) on the Olympic Peninsula. To evaluate the amount of protection afforded by these guidelines, we examined the proportion of each cumulative owl home range that fell within a 4.3-km radius of the nest area or main roost area of each owl, and we compared median and mean areas of “suitable habitat” in cumulative owl ranges with the target in the Forest Practices Rules. All means are expressed as x ± 1 SE. Results Sample Size and Tracking Periods. We moni- tored 22 owls in 12 territories, including 10 resi- dent pairs, one territory where we marked one member of a resident pair, and one territory where we marked an adult female that did not appear to have a mate. We did not use the data from the unpaired adult female because she did not exhibit site fidelity. We also did not use data from one fe- male that died shortly after she was radio-marked. Of the 11 pairs in which one or both members were radio-marked, three nested during the study, including one pair in 1987 and two pairs in 1988. On average, we tracked individual owls for 438 ± 34 d (range = 166-711 d; Fig. 2). Total reloca- tions per owl, not counting incubation locations, averaged 366 ± 35 (range = 126-685). Of 3262 roost locations, we estimated 2360 (72%) by tri- angulation and located 902 (28%) by homing in on transmitters to locate owls visually in their roost trees. Annual Ranges. Median estimates of annual ranges of individual owls were 1147 ha (75% FK), 2406 ha (95% FK), and 2290 ha (100% MCP; Table 2) . Most mean estimates of ranges were larger than medians because means were skewed by a few in- dividuals with large ranges (Table 2). All owls with annual MCP or FK ranges >5000 ha were individ- uals that expanded their ranges substantially dur- ing fall and winter. Annual ranges were smaller for females than for males, with the exception of the 75% FK estimate (Table 2). In six cases in which we monitored owls for two yr, the sequential an- nual ranges overlapped by 70 ± 6% based on the 75% FK (range = 18-100%), 73 ± 5% based on the 95% FK (range = 27-100%), and 73 ± 4% based on the MCP (range = 38-100%). 370 Forsman et al. VoL. 39, No. 4 Table 2. Estimates of annual home-range areas of individual northern Spotted Owls on the Olympic Peninsula, Washington, 1987-89. Estimates include the 100% minimum convex polygon (MCP) and the 75% and 95% isopleths of the Fixed Kernel (FK) . Number of Days and Number of Locations IN Sample Period*^ Home Range Estimates (ha) AND Year* Days Locations Locations 75% FK 95% FK 100% MCP BPF87 410 160 191 771 1696 1779 BPF88 116 66 70 2209 4196 3295 BPM87 386 147 180 1935 4865 6122 BPM88 325 144 145 3106 8469 8351 BRF87 302 51 136 415 1189 1402 BRF88 142 39 58 804 1720 1151 BRM87 373 52 172 532 1172 1367 BRM88 123 26 37 530 1045 683 ECF87 354 72 129 7927 15 212 10 704 ECF88 276 100 128 1404 4207 6668 ECM87 370 68 159 1108 2104 1917 ECM88 65 55 62 982 1876 1294 FBF87 166 73 79 1586 3483 3086 FBM87 206 99 106 1202 2411 2230 HRM88 366 201 186 2465 6924 7954 LBF88 345 137 143 2195 4931 4950 LBM88 391 168 184 1513 3232 3288 LCF88 246 65 109 967 2235 1915 LCM87 354 45 140 678 1504 2000 LCM88 94 23 42 509 1074 894 NRF87 387 133 174 734 1786 2350 NRF88 95 58 58 2597 5795 4537 NRF88 95 58 58 2597 5795 4537 NRM87 387 165 210 1186 3084 4284 NRM88 274 153 157 5003 11558 11252 RFF88 348 208 190 980 2092 2235 RFF89 114 50 44 554 1294 975 RFM88 309 152 144 2516 7059 6704 SPF87 396 154 169 468 1115 1323 SPF88 86 48 55 1228 2402 1406 SPM87 396 154 167 568 1583 1861 SPM88 247 151 142 1072 2533 3593 SRM87 314 34 92 2817 5693 3879 Mean 1642 3736 3608 Median 1147 2406 2290 Mean 9 ^ 1656 3557 3185 Mean 1631 3893 3981 Median 967 2235 1915 Median d'* 1186 2533 3288 First two letters indicate owl name, third letter indicates sex of owl, and numbers indicate year of estimate. Total locations for NRF87 and SPF87 also included 63 and 76 incubation locations, respectively. TV = 1 6 owl years. TV = 17 owl years. December 2005 Spotted Owl Habitat Use 371 Table 3. Estimates of the cumulative home-range areas of northern Spotted Owls on the Olympic Peninsula, Wash- ington, 1987-89. Estimates include the 100% minimum convex polygon (MCP) and the 75% and 95% isopleths of the Fixed Kernal (FK) estimator. Number of Days and Number of Locations IN Sample Period Owl Code Name, Sex Days Roost Locations Forage Locations Home Range Estimates (ha) 75% FK 95% FK 100% MCP BPF 526 226 261 2216 4303 3527 BPM 711 291 325 3235 8521 8715 BRF 444 90 194 804 1746 1562 BRM 496 78 209 645 1281 1372 ECF 630 172 257 7927 15212 10916 ECM 435 134 221 1215 2166 1932 LCM 449 68 182 845 1636 2026 NRF^ 482 191 232 2580 5995 4852 NRM 661 318 367 5003 11561 11252 RFF 462 258 234 985 2164 2298 SPF^ 482 202 224 1230 2436 1831 SPM 643 305 309 1090 2643 3716 Mean^ 2315 4972 4500 Median*" 1222 2539 2912 Mean 2624 5309 4164 Mean 2006 4635 4836 Median 9'^ 1173 3384 2912 Median 1152 2404 2870 ^ Total locations for NRF and SPF also included 62 and 76 incubation locations, respectively. bN = 12. = 6. Overlap of annual ranges of nine owls of the same sex that occupied adjacent territories aver- aged 5 ± 2% for the 75% FK (range = 0-25%), 21 ± 3% for the 95% FK (range = 3-58%), and 26 ± 4% for the MCP (range = 0-58%). These estimates probably did not reflect total overlap with adjacent residents because there were adja- cent pairs that we did not have radio-marked and because tracking periods for individual owls were not always exactly the same. However, even with incomplete data on some individuals and no data on the pairs that were not radio-marked, it was clear that home ranges of neighbors overlapped considerably, particularly during winter. In one case, a male from one territory (BPM) was found on several occasions during winter, roosting in the traditional nest area of an adjacent male (HRM) . Cumulative Ranges. Median estimates of cumu- lative ranges of individual owls monitored in two sequential years were 1222 ha (75% FK), 2539 ha (95% FK), and 2912 ha (MCP; Table 3). Cumula- tive ranges of females averaged larger than cumu- lative ranges of males for all comparisons except the mean MCP (Table 3). Seasonal Ranges. Ranges of individual owls based on the 95% FK averaged 3360 ± 572 ha dur- ing the breeding season (range = 883—10 205 ha, median = 2052 ha, = 21) and 3175 ± 572 ha during the nonbreeding season (range = 611-12 352 ha, median — 2168 ha, N = 29). There was no consistent pattern of larger ranges in one season or the other. Median estimates of seasonal ranges were smaller than means because means were pos- itively skewed by a few individuals with large rang- es. Overlap of nonbreeding season ranges on breeding season ranges averaged 65 ± 4.5% (range = 8-98%, N= 36), and overlap of breeding season ranges on nonbreeding ranges averaged 63 ± 4.9% (range = 1-100%, N = 36). Overlap of breeding season ranges of two owls tracked in two different breeding seasons averaged 74 ± 8.9% (range = 58-91%). Overlap of nonbreeding rang- es of nine owls tracked in two different nonbreed- 372 Forsman et al. VoL. 39, No. 4 Table 4. Results of compositional analysis of habitat use for foraging by northern Spotted Owls on the Olympic Peninsula, Washington, 1987-89. Rank scores indicate relative preference of cover types from highest (6) to lowest (0) . Results of pairwise t-tests indicate the relative preference of cover types. A positive lvalue indicates that the row cover type ranked higher than the column cover type and a negative t-value indicates that the row cover type ranked lower than the column cover type. A significant P-value suggests that confidence in the direction of the relationship was high. Cover Type^ Old Forest Mature Forest Mixed- Young Forest Young Forest Pole- sapling Hard- wood/ Riparian Clear- cut/Non- forest Rank Old Forest t 3.127 4.459 4.637 8.427 4.443 8.103 6 P 0.006 <0.001 <0.001 <0.001 <0.001 <0.001 Mature Forest t -3.127 -1.454 0.774 2.429 0.391 4.183 4 P 0.006 0.162 0.448 0.025 0.700 0.001 Mixed-young Forest t -4.459 1.454 2.756 6.676 2.743 6.528 5 P <0.001 0.162 0.013 <0.001 0.013 <0.001 Young Forest t -4.637 -0.774 -2.756 1.561 -0.471 2.631 2 P <0.001 0.448 0.013 0.135 0.643 0.017 Pole-sapling t -8.427 -2.429 -6.676 -1.561 -2.826 2.370 1 P <0.001 0.025 <0.001 0.135 0.011 0.029 Hardwood/Riparian t -4.443 -0.391 -2.743 0.471 2.826 3.860 3 P <0.001 0.700 0.013 0.643 0.011 0.001 Clear-cut/Non-forest t -8.102 -4.183 -6.528 -2.631 -2.370 -3.860 0 P <0.001 0.001 <0.001 0.017 0.029 0.001 ing seasons averaged 59 ± 6.3% (range = 10- 100 %). During the breeding season, movements of owls were typically centered on the nest tree or, in the case of nonnesting pairs, a regularly-used roost area. Winter ranges typically included part of the breeding-season range plus areas peripheral to the breeding-season range. However, a few individuals spent little time in their breeding-season ranges during the winter season. The most dramatic ex- ample was the Elk Creek Female (ECF) . After nest- ing and producing a juvenile in 1987, she left the nest area in August and spent most of the fall and winter in an area 5-15 km away from the nest area before eventually returning to the nest area in June of 1988. The Neilton Ridge Male (NRM) also had a very large nonbreeding range in 1988-89, but in his case, the nonbreeding range overlapped most of the breeding season range. Ranges of Pairs. There were 14 cases where we monitored annual ranges of paired owls in the same year. The annual ranges of these pairs (union of annual ranges of male and female) averaged 2397 ± 558 ha for the 75% FK (median = 1570 ha), 5449 ± 1111 ha for the 95% FK (median = 4081 ha), and 5414 ± 895 ha for the MCP (median = 5032 ha). Overlap of annual ranges of paired owls averaged 70 ± 5% based on the 75% FK (range = 14-100%), 69 ± 5% based on the 95%FK (range = 14—100%), and 64 ± 5% based on the MCP (range = 14-100%). Estimates of mean over- lap of annual ranges were similar, regardless of which sex was used as the frame of reference, so we based the above averages on all possible com- binations of overlap. Cumulative ranges of five pairs that were moni- tored in both years averaged 3945 ± 1282 ha for the 75% EK (median = 4053 ha), 8278 ± 2550 ha for the 95% EK (median = 9329 ha), and 7488 ± 1951 ha for the MCP (median = 9195 ha). Overlap of cumulative 95% FK ranges of paired individuals averaged 68 ± 14% for males on females and 72 ± 12% for females on males. Habitat Selection. Use of cover types for forag- ing and roosting was nonrandom. Old Forest was the most preferred type for foraging, followed by Mixed-young Forest, Mature Forest, Hardwood/Ri- parian Forest, Young Forest, Pole-sapling, and Clear-cut/Non-forest (Table 4). Pairwise compari- sons of rank indicated that Old Forests were con- sistently preferred over all other cover types (Table 4). Although Mixed-young Forest ranked higher than Mature Forest, pairwise comparisons of rank indicated little difference between the two types December 2005 Spotted Owl Habitat Use 373 Table 5. Results of compositional analysis of habitat use for roosting by northern Spotted Owls on the Olympic Peninsula, Washington, 1987-89. Rank scores indicate relative preference of cover types from highest (6) to lowest (0) . Results of pairwise t-tests comparisons indicate the relative preference of cover types. A positive lvalue indicates that the row cover type ranked higher than the column cover type and a negative f-value indicates that the row cover type ranked lower than the column cover type. A significant P-value suggests that confidence in the direction of the relationship was high. Cover Type^* Old Forest Mature Forest Mixed- Young Forest Young Forest Pole- sapling Hard- wood/ Riparian Clear- cut/Non- forest Rank Old Forest t 2.605 3.326 4.823 8.079 4.752 16.554 6 P 0.017 0.004 <0.001 <0.001 <0.001 <0.001 Mature Forest t -2.605 0.121 1.141 5.124 1.361 10.527 5 P 0.017 0.905 0.268 <0.001 0.189 <0.001 Mixed-young Forest t -3.326 -0.121 1.163 5.195 1.711 9.010 4 P 0.004 0.905 0.259 <0.001 0.103 <0.001 Young Forest t -4.823 -1.141 -1.163 3.541 0.447 7.540 3 P <0.001 0.268 0.259 0.002 0.660 <0.001 Pole-sapling t -8.079 -5.124 -5.195 -3.541 -3.271 2.543 1 P <0.001 <0.001 <0.001 0.002 0.004 0.020 Hardwood/Riparian t -4.752 -1.361 -1.711 -0.447 3.271 6.477 2 P <0.001 0.189 0.103 0.660 0.004 <0.001 Clear-cut/Non-forest t -16.554 -10.527 -9.010 -7.540 -2.543 -6.477 0 P <0.001 <0.001 <0.001 <0.001 0.020 <0.001 (Table 4). Similarly, Young Forest ranked lower than Mature and Hardwood Forest, but pairwise comparisons indicated that these differences were weak (Table 4). Pole-sapling stands ranked lower than all other types except Clear-cuts, but the pair- wise comparisons with other types indicated that preference for Pole-sapling was not greatly differ- ent from Young Forest (Table 4). Large P-values for all pairwise comparisons of Clear-cuts relative to other cover types indicated that Clear-cuts were the least preferred cover type for foraging. In fact, out of 3822 foraging locations where cover type could be determined, only 57 (1.5%) occurred in Clear-cuts or Non-forest areas, and we suspected that some of these cases were due to telemetry or mapping error. Use of cover types for roosting indicated that Old Forests were preferred over all other cover types (Table 5) . Mature Forest ranked higher than Mixed-young, Young Forest, and Hardwood/Ripar- ian Forest, but pairwise comparisons of these types indicated that differences among them were weak (Table 5). Pole-sapling, Clear-cuts, and Non-forest areas were rarely used for roosting. Of 902 roosts located visually, none were located in Clear-cuts or Non-forest. Of 2275 roosts located by triangulation alone, and for which cover type was determined, eight were in Clear-cuts or Non-forest types; we sus- pected these were due to triangulation or mapping error. Habitat Use Relative to Forest Edges, Streams, and Elevation. On average, foraging locations and roost locations were closer to openings (233 ± 24 m, and 271.9 ± 33.0 m, respectively) than were random locations (304 ± 34 m; borage ~ — 4.10, P = 0.001, ^roost = -2.04, P = 0.055; - 20 owls). However, the number of locations within 100 m of an edge was similar between random locations and foraging locations (28.4% vs. 33.5%) and random locations and roost locations (28.4% vs. 29.9%), so we concluded that there was little evidence that owls either preferred or avoided forest edges for roosting or foraging. Mean elevations at foraging locations (315 ± 29 m) and roosting locations (322 ± 31 m) were slightly lower than elevations at random locations (354 ± 36 m; = -3.63, P = 0.002, W ^ — 3.09, P = 0.006, N — 20 owls). Mean distance to the nearest stream was similar for foraging (98 ± 14 m), roosting (112 ± 19 m), and random loca- tions (94 ± 10 m; ^forage — 0.73, P = 0.475, ~ 1.75, P = 0.097, = 20 owls). Landscape Composition and Home Range Size. Size of annual ranges was negatively correlated 374 Forsman et al. VoL. 39, No. 4 with the percent cover of older forest (cover types: Old and Mature forest) in the cumulative MCP range, regardless of whether the estimator was the 75% FK (rgi = -0.53, P = 0.002), 95% FK (r^i = -0.59, P < 0.001), or MCP = -0.67, P < 0.001). Size of annual ranges was also negatively correlated with the amount of older forest in a 4.3 km circle centered on the central place (75% FK r3i - -0.34, P = 0.058; 95% FK - -0.40, P - 0.028; MCP = -0.46, P - 0.009). Overlap of Management Circles with Home Ranges. On average, a 4.3-km radius circle cen- tered on the nest site or center of activity included 94 ± 2% of the annual 75% FK home range, 86 ± 4% of the annual 95% FK home range, and 83 ± 4% of the annual MCP range. For 12 owls tracked in both years, average overlap of the 4.3-km radius circle on the cumulative range was 99 ± 13% for the 75% FK range, 79 ± 7% for the 95% FKrange, and 76 ± 7% for the MCP range. The counter- intuitive result in which overlap of the 4.3-km cir- cle was lower on the 75% FK annual range than on the 75% FK cumulative range occurred because the estimates were based on different individuals. If we defined “suitable habitat” as the cover types that had the top three preference rankings based on compositional analysis (cover types = Old, Ma- ture, and Mixed-young Forest), then the mean amount of suitable habitat within a 4.3-km radius circle was 3105 ± 236 ha. Discussion Home Range Attributes. The large ranges ob- served in our study suggest that biomass of suitable prey for Spotted Owls is lower on the Olympic Pen- insula than in western Oregon and northwestern California, where home ranges tend to be smaller (Forsman et al. 1984, Carey et al. 1990, 1992, Zabel et al. 1995, Bingham and Noon 1997, Glenn et al. 2004) . We did not have data on total prey biomass in our study area, but Carey et al. (1992) found that flying squirrels, which are the primary prey of Spotted Owls on the Olympic Peninsula, were rel- atively uncommon on the peninsula compared to western Oregon. As in our study, Carey et al. (1990) and Glenn et al. (2004) found that home range size of north- ern Spotted Owls was inversely related to the amount of old forest in the home range. This sug- gests that Spotted Owls respond to decreasing amounts of their preferred habitat by increasing the size of their ranges to encompass more old for- est. However, Zabel et al. (1995) found no corre- lation between home-range size of Spotted Owls and the proportion of the range covered by large trees. Instead, they found that home-range size was positively correlated with the proportion of flying squirrels in the diet and negatively correlated with the proportion of woodrats (Neotoma spp.) in the diet. In our study area, the diet was dominated by flying squirrels (Forsman et al. 2001), which tend to be most abundant in old forests (Carey et al. 1992, Waters and Zabel 1995). This could explain why home ranges in our study area became larger as the amount of old forest declined. However, for a central-place forager like the Spotted Owl, the ability to increase the size of the home range and still function as a part of the resident breeding population is probably limited by energetic and so- cial constraints (Carey et al. 1992). In our study, annual home ranges of paired owls typically overlapped by 50-80%. Similar estimates were obtained in a number of other studies (Fors- man et al. 1984, Carey et al. 1990, Glenn et al. 2004). Our estimates of mean overlap of annual ranges of owls on adjacent territories were higher than values reported by Forsman et al. (1984:23; MCP overlap = 12%) and Glenn et al. (2004:41; 95% FK overlap = 14.9 ± 4.3% and 6.7 ± 2.2% on two different study areas) . Habitat Selection. Our study, and most other studies in which telemetry methods have been used to examine habitat selection by northern Spotted Owls, indicated that, given a choice, most individuals selectively used older forests for forag- ing and roosting and that younger stands generally provided lower quality habitat (e.g., Forsman et al. 1984, Call 1989, Carey et al. 1990, 1992, Solis and Gutierrez 1990, Gutierrez et al. 1995). However, there have been two radiotelemetry studies of northern Spotted Owls in landscapes dominated by young forest, where patterns of habitat selection were less clear. Glenn et al. (2004) examined hab- itat selection by Spotted Owls in young forests in northwest Oregon and did not find strong selec- tion for any cover type. In a landscape where old forest comprised less than 10% of the available cov- er, Irwin et al. (2000) found that northern Spotted Owls infrequently used stands <25 yr of age and foraged primarily in mid-age stands (25—79 yr old) or in remnant patches of old forest. However, Ir- win et al. (2000) did not conduct a landscape-level analysis of use-versus-availability with their data, so December 2005 Spotted Owl Habitat Use 375 we could not determine if use of different cover types differed from availability. California Spotted Owls (5. o. occidentalis) in the Sierra Nevada Mountains tended to forage in for- ests with ^40% canopy cover, but did not show a strong preference relative to tree age or tree size (Zabel et al. 1992). However, at two of the three study areas described by Zabel et al. (1992), the majority of foraging and roosting locations were in stands dominated by large (>53 cm DBH) trees. Of the 5-6 species of small mammals that com- prise the primary diet of Spotted Owls, several ap- pear to be most abundant in older forests. For ex- ample, there are a number of studies that suggest that red tree voles {Arborimus longicaudus) and red- backed voles {Clethrionomys californicus) are most abundant in older forests (Corn and Bury 1986, Aubry et al. 1991, Rosenberg et al. 1994). While not all studies of northern flying squirrels have found significantly higher numbers in old forests, the trend in most studies was toward higher num- bers in old forests (Carey et al. 1992, Rosenberg and Anthony 1992, Waters and Zabel 1995, Lehm- kuhl et al. (in press). Therefore, an obvious hy- pothesis is that differences in abundance of pre- ferred prey cause northern Spotted Owls to select for older forests (Forsman et al. 1984, Carey et al. 1992). Ward et al. (1998) posed a similar hypoth- esis to explain high use of forest edges by Spotted Owls in northwestern California, where the diet was dominated by dusky-footed woodrats {N. fusci- pes), which were most abundant in brushy open- ings adjacent to forests. In contrast, in areas where they feed mainly on flying squirrels. Spotted Owls either avoid non-forest edges or use them in pro- portion to availability (Zabel et al. 1995, Glenn et al. 2004, this study). Streams and Elevation. Although Glenn et al. (2004) found evidence that Spotted Owls foraged selectively in riparian vegetation, we found no ev- idence that foraging or roosting locations were closer to streams than were random locations. We concluded that there was no evidence from our data that owls were either selecting or avoiding ri- parian areas. Although Spotted Owls in our study foraged at lower elevations than expected, the mean difference between observed and expected foraging locations was only 39 m. We were not con- vinced that this relatively small difference was bi- ologically meaningful. Management Implications, Based on the results of our study, we agree with Forsman et al. (1984), Thomas et al. (1990), and Carey et al. (1992) that management for northern Spotted Owls in western Washington and Oregon should focus on the re- tention of old forests. Although Franklin et al. (2000) and Olson et al. (2004) found that north- ern Spotted Owls may have higher reproductive output in landscapes that include a mixture of old forest and edges with other forest types, those stud- ies were conducted in areas where woodrats were a primary prey, and the results may not apply to areas like the Olympic Peninsula, where flying squirrels are the primary prey. Bingham and Noon (1997, 1998) suggested that the U.S. Fish and Wildlife Service should focus on the most heavily-used portion of the home range, or “core area,” as the frame of reference for as- sessment of “take” of Spotted Owls. If this ap- proach is used on the Olympic Peninsula, then we believe it would be reasonable to use our estimates of the 75% isopleth of the FK annual range as the criteria for estimates of core areas, although other methods have been proposed (Bingham and Noon 1997). We agree with Bingham and Noon (1997) that it makes sense to use repeatable measures of home range areas as the frame of reference for assessments of “take,” but this should not be mis- construed as a recommendation to manage Spot- ted Owls based only on core areas. If the objective is to provide Spotted Owls with enough habitat to survive and reproduce on a site, then we agree with Buchanan et al. (1998) that management should be based on amounts of habitat within the entire home-range areas of radio-marked owls, not just core areas. Our estimates of the median and mean amounts of “suitable habitat” within cumulative MCP rang- es of Spotted Owls (1824 ha and 2253 ± 286 ha) are similar to or slightly lower than the manage- ment target adopted by the Washington State For- est Practices Board (1996) for management around Spotted Owl nest sites (2373 ha of suitable habitat within a 4.3-km radius). We found that a 4.3-km radius circle centered on the nest site en- compassed about 83-87% of the mean cumulative home range used by individual Spotted Owls on the peninsula. Based on these results, we see no reason to suggest changes to the 1996 Forest Prac- tices Rules (Washington State Forest Practices Board 1996). However, it remains to be seen if Spotted Owls will persist in areas where old and mature forests are gradually replaced with less-pre- ferred types that are also classified as “suitable.” 376 Forsman et al. VoL. 39, No. 4 Acknowledgments For help with data collection, we thank Duane Aubu- chon, Bruce Casler, Sue Grayson, Martha Jensen, Pete Loschl, Rich Lowell, Paul Radley, Doreen Schmidt, Margy Taylor, and Joe Zisa. Housing and administrative support was provided by the Olympic National Forest. Reviews of the manuscript by Rocky Gutierrez, Martin Raphael, Dan Varland, and Cindy Zabel were extremely helpful. This study was funded by the U.S. Department of Agriculture Forest Service, Pacific Northwest Research Station, Port- land, OR U.S.A. Literature Cited Aebischer, N.J., V. Marcstrom, R.E. Kenward, and M. Karlbom. 1993. Survival and habitat utilization: a case for compositional analysis. Pages 343-353 mJ.D. Le- breton and P.M. North [Eds.], Marked individuals in the study of bird population. Birkhauser Verlag, Basel, Switzerland. Aubry, K.B, MJ. Crites, and S.D. West. 1991. Regional patterns of small mammal abundance and community composition in Oregon and Washington. Pages 285- 294 in L.F. Ruggerio, K.B. Aubry, A.B. Carey, and M.H. Huff [Tech. Coords.], Wildlife and vegetation of unmanaged Douglas-fir forests. Gen. Tech. Rep. PNW-GTR-285. USDA, Forest Service, Pacific North- west Research Station, Portland, OR U.S.A. Bingham, B.B. and B.R. Noon. 1997. Mitigation of hab- itat “take”: application to habitat conservation plan- ning. Conserv. Biol. 11:127-139. AND . 1998. The use of core areas in com- prehensive mitigation strategies. Conserv. Biol. 12:241- 243. Buchanan, J.B., FJ. Frederickson, and D.E. Seaman. 1998. Mitigation of habitat “take” and the core area concept. Conserv. Biol. 12:238—240. Call, D.R. 1989. Home range and habitat use by Cali- fornia Spotted Owls in the central Sierra Nevada. M.S. thesis, Humboldt State University, Areata, CA U.S. A. Carey, A.B., S.P. Horton, and B.L. Biswell. 1992. North- ern Spotted Owls: influence of prey base and land- scape character. Ecol. Monogr. 62:223-250. AND K.C. Peeler. 1995. Spotted Owls: resource and space use in mosaic landscapes. J. Raptor Res. 29: 223-229. , J.A. Reid, and S.P. Horton. 1990. Spotted Owl home range and habitat use in southern Oregon Coast Ranges. /. Wildl. Manage. 54:11-17. Corn, P.S. and R.B. Bury. 1986. Habitat use and terres- trial activity by red tree voles {Arborimus longicaudus) in Oregon./. Mammalogy 67:404—406. Forsman, E.D. 1983. Methods and materials for locating and studying Spotted Owls. Gen. Tech. Rep. PNW- 162. USDA, Forest Service, Pacific Northwest Re- search Station, Portland, OR U.S.A. , E.C. Meslow, and H.M. Wight. 1984. Distribu- tion and biology of the Spotted Owl in Oregon. Wildl. Monogr. 87:1—64. , LA. Otto, S.G. Sovern, M. Taylor, D.W. Hays, H. Allen, S.L. Roberts, and D.E. Seaman. 2001. Spa- tial and temporal variation in diets of Spotted Owls in Washington./. Raptor Res. 35:141-150. Franklin, A.B., D.R. Anderson, RJ. Gutierrez, and K.P. Burnham. 2000. Climate, habitat quality, and fitness in Northern Spotted Owl populations in northwest- ern California. Ecol. Monogr. 70:539-590. Glenn, E.M., M.C. Hansen, and R.G. Anthony. 2004. Spotted Owl home range and habitat use in young forests of western Oregon. /. Wildl. Manage. 68:33-50. Guetterman, J.H., J.A. Burns, J.A. Reid, R.B. Horn, and C.C. Foster. 1991. Radiotelemetry methods for study- ing Spotted Owls in the Pacific Northwest. Gen. Tech. Rep. PNW-272. USDA, Forest Service, Pacific North- west Research Station, Portland, OR U.S.A. Gutierrez, R.J., A.B. Franklin, and W.S. LaHaye. 1995 Spotted Owl. The Birds of North America No. 179. The Birds of North America, Inc. Philadelphia, PA U.S.A. Hamer, T.E. 1988. Home range size of the Northern Barred Owl and Northern Spotted Owl in western Washington. M.S. thesis. Western Washington Univer- sity. Seattle, WA U.S.A. Havne, D.W. 1949. Calculation of size of home range. /. Mammalogy 30:1-18. Henderson, J.A., D.H. Peter, R.D. Lesher, and D.C Shaw. 1986. Forested plant associations of the Olym- pic National Forest. Ecological Technical Paper R6- ECOL-TP 001-88. USDA, Forest Service, Portland, OR U.S.A. Irwin, L.L., D.R Rock, and G.P. Miller. 2000. Stand structures used by northern Spotted Owls in managed forests./. Raptor Res. 34:175-186. Kie,J.G., J.A. Baldwin, and C.J. Evans. 1996. CALHOME- A program for estimating animal home ranges. Wildl Soc. Bull. 24:342-344. Laymon, S.A. 1989. Altitudinal migration movements of Spotted Owls in the Sierra Nevada, California. Condor 91:837-841. Leban, F. 1999. Program RSW — Resource selection for Windows, Version 1.00 Beta 8.4. University of Idaho, Moscow, ID U.S.A. Lehmkuhl, J.F., K.D. Kistler, J.S. Begley, and J. Boulan- ger. (in press) . Demography of northern flying squir- rels informs ecosystem management of western inte- rior forests. Ecol, Appl. Nams, V.O. and S. Boutin. 1991. What is wrong with er- ror polygons? /. Wildl. Manage. 55:172-176. Olson, G.S., E.M. Glenn, R.G. Anthony, E.D. Forsman, J.A. Reid, PJ. Loschl, and WJ. Ripple. 2004. Model- ing demographic performance of northern Spotted Owls relative to forest habitat in Oregon. /. Wildl Manage. 68:1039—1053. Otis, D.L., and G.C. White. 1999. Autocorrelation of lo- cation estimates and the analysis of radiotracking data.. J. Wildl. Manage. 63:1039-1044. December 2005 Spotted Owl Habitat Use 377 Pierce, F.R. 1921. Tornado destroys great forest. Pop. Mech. 35:659-662. Rosenberg, D.K. and R.G. Anthony. 1992. Characteris- tics of northern flying squirrel populations in young second- and old-growth forests of western Oregon. Can.;. Zool. 70:161-166. , K.A. Swindle, and R.G. Anthony. 1994. Habitat associations of California red-backed voles in young and old-growth forests in western Oregon. Northwest Sci. 68:266-272. Seaman, D.E., B. Griffith, and R.A. Powell. 1998. KER- NELHR: a program for estimating animal home-rang- es. WildL Soc. Bull. 26:95-100. and R.A. Powell. 1996. An evaluation of the ac- curacy of kernel density estimators for home-range analysis. Ecology 77:2075-2085. Solis, D.M., Jr. and RJ. Gutierrez. 1990. Summer hab- itat ecology of northern Spotted Owls in northwestern California. Condor 92:739-748. SwiHART, R.K. and N.A. Slade. 1985a. Influence of sam- pling interval on estimates of home-range size. J. WildL Manage. 49:1009-1025. and N.A. Slade. 1985b. Testing for independence of observations in animal movements. Ecology 66: 1176-1184. Thomas, J.W., E.D. Forsman, J.B. Lint, E.C. Meslow, B.R. Noon, and J. Verner. 1990. A conservation strat- egy for the northern Spotted Owl: report of the In- teragency Scientihc Committee to address the conser- vation of the northern Spotted Owl. USDA, Forest Service and USDI, Bureau of Land Management, Portland, OR U.S.A. Ward, J.P., Jr., RJ. Gutierrez, and B.R. Noon. 1998. Habitat selection by Northern Spotted Owls: the con- sequences of prey selection and distribution. Condor 100:79-92. Washington State Forest Practices Board. 1996. Final environmental impact statement on forest practices rule proposals for northern Spotted Owl, Marbled Murrelet, and western gray squirrel. Washington State Forest Practices Board, Olympia, WA U.S.A. Waters, J.R. and C.J. Zabel. 1995. Northern flying squir- rel densities in fir forests of northeastern California. J. WildL Manage. 59:858-866. Zabel, C.J., K. McKelvey, and J.P. Ward, Jr. 1995. Influ- ence of primary prey on home-range size and habitat- use patterns of northern Spotted Owls (Strix occiden- talis caurina). Can. J. Zool. 73:433—439. , G.N. Steger, K.S. McKelvey, G.P. Eberlein, B.R. Noon, and J. Verner. 1992. Home-range size and habitat-use patterns of California Spotted Owls in the Sierra Nevada. Pages 149-163 in]. Verner, K.S. Mc- Kelvey, B.R. Noon, RJ. Gutierrez, G.I. Gould, Jr., and T.W. Beck [Eds.], The California Spotted Owl: a tech- nical assessment of it current status. USDA, Forest Service, Pacific Southwest Research Station, Albany, CA U.S.A. Received 4 April 2003; accepted 30 August 2005 J Raptor Res. 39(4):378-385 © 2005 The Raptor Research Foundation, Inc. FIRST-CYCLE MOLTS IN NORTH AMERICAN FALCONIFORMES Peter Pyle' The Institute for Bird Populations, P.O. Box 1436, Point Reyes Station, CA 94956 U.S.A. Abstract. — I examined 1849 specimens of 20 North American Falconiform species to elucidate the occurrence and nomenclature of partial first-cycle molts. As reported in the literature, American Kestrel {Falco sparverius) and White-tailed Kite (Elanus leucurus) have relatively complete body-feather molts that occur during the first fall; in the kite, this molt can also include up to all rectrices and 2-6 secondaries, but no primaries — an unusual pattern for such partial molts in first-year birds. Evidence of partial first- cycle molts was found in 16 of 18 other species (among Pandion, Haliaeetus, Circus, Accipiter, Asturina, Buteo, Aquila, and Falco) for which such molts have not been previously elucidated. Maximum extent of body-feather replacement among individuals of these 16 species varied from 5-50%. On the other hand, most species showed evidence that this molt could be absent (11-100% of birds remaining in juvenile plumage until commencement of the complete or near-complete prebasic molt that occurs during the first summer) . I argue that these partial molts are best considered preformative molts (following Howell et al. 2003) rather than “first prebasic” molts, as defined by Humphrey and Parkes (1959). Variation in the extent and timing of preformative molts may reflect various constraints according to species- specihc breeding, migrating, and foraging strategies. The apparent lack of function for this molt suggests that ancestral Falconiformes exhibited a more extensive preformative molt, as found in related orders of birds, but that this molt has since become vestigial, at least in the larger species. Key Words: American Kestrel, Falco sparverius; White-tailed Kite, Elanus leucurus; preformative molt, hawks; falcons. MUDAS DEL PRIMER CICLO EN FALCONIFORMES NORTEAMERICANOS Resumen. — Examine 1849 especimenes de 20 especies norteamericanas de Falconiformes para estable- cer la ocurrencia y la nomenclatura de mudas parciales del primer ciclo. Como habia sido informado en la literatura, Falco sparverius y Elanus leucurus presentan mudas relativamente completas de las plumas corporales que tienen lugar durante el primer otono, aunque en E. leucurus esta muda puede tambien incluir algunas o todas las rectrices y 2-6 secundarias pero ninguna primaria. Esto constituye un patron poco usual de mudas parciales durante el primer aho de vida. Encontre evidencia de mudas parciales en el primer ciclo en 16 de las 18 especies adicionales (en los generos Pandion, Haliaeetus, Circus, Accipiter, Asturina, Buteo, Aquila y Falco) para las cuales no se habia determinado previamente la ocurrencia de mudas de este tipo. El maximo grado de reemplazo de las plumas del cuerpo en los individuos de esas 16 especies vario entre el 5% y el 50%. Por otra parte, la mayoria de las especies mostraron evidencia de que esta muda podria estar ausente, pues entre el 11% y el 100% de los individuos mantuvieron el plumaje juvenil hasta comenzar la muda prebasica completa, o casi completa, que tiene lugar durante el primer verano. Sugiero que estas mudas prebasicas deben considerase mudas preformativas (siguien- do la terminologia de Howell et al. 2003) en lugar de “primeras mudas prebasicas”, como fueron definidas por Humphrey y Parkes (1959). La variacion en la extension y en el momenta de ocurrencia de las mudas preformativas podria reflejar distintos limitantes de acuerdo a las estrategias reproductivas, de migracion y de forrajeo especificas de cada especie. La falta aparente de funcionalidad de esta muda sugiere que los Falconiformes ancestrales exhibian una muda preformativa mas extensiva, como se ha documentado en ordenes de aves relacionados, pero que esta muda se ha vuelto vestigial, al menos en las especies de mayor tamaho. [Traduccion del equipo editorial] North American Falconiformes exhibit various most species, Juvenal plumage is reportedly re- molt strategies during their first year of life. In tained until the first spring or summer (when a year old) , at which point a complete or near-com- plete prebasic molt commences, usually with the ^ Email address: ppyle@birdpop.org shedding of the innermost primaries in Accipitri- 378 December 2005 First-cycle Molts in Falconiformes 379 dae or the medial primaries in Falconidae (e.g., Miller 1941, Palmer 1988, Forsman 1999, Wheeler 2003). In American Kestrel (Falco sparverius) and certain kites, by contrast, most to all Juvenal body feathers are replaced during the first fall, well be- fore molt of the primaries commences during the following summer (Bent 1937, 1938, Parkes 1955, Palmer 1988, Miller and Smallwood 1997). For some species of Falconiformes, a limited number of body feathers are reported to be re- placed during the first fall, winter, or spring, prior to the shedding of primaries during the first spring or summer. For example, Wheeler (2003) reported that in the Red-shouldered Hawk (see Table 1 for scientific names), “the first prebasic molt begins on the breast, then is noticeable on the back and scapulars” before molt of primaries begins, and Forsman (1999) reported that in the Northern Harrier, “single body-feathers and tail-feathers may be replaced from late winter” prior to the com- plete molt the following May to October. Similar limited molts in fall, winter, or spring have been reported for Osprey (Bent 1937), Northern Gos- hawk (Bent 1937, Dement’ev and Gladkov 1951, Forsman 1999), Gray Hawk (Dickey and van Ros- sem 1938, Wheeler 2003), and several species of Falconidae (Cramp and Simmons 1980, Forsman 1999, Wheeler 2003). Most authors consider this limited body-feather replacement to be the initiation of the prebasic molt at a year of age (hereafter, “first complete molt”) rather than a separate molt; indeed, Wheel- er (2003) even interprets the fall molts in first-year American Kestrel and kites as part of the first com- plete molt, terminating with the flight feathers the following summer. This interpretation appears to disregard the additional body molt that occurs in one-year-old birds concurrent with flight-feather molt in these species (Parkes 1955, Palmer 1988). Herremans and Louette (2000), on the other hand, document that such molts in certain Old- World species of Accipiter are distinct from the first complete molt. Thus, there remains confusion about both the occurrence of partial first-cycle molts in Falconi- formes and whether or not body-feather replace- ment in the first fall, winter, or spring should be considered part of a separate partial molt or as the initiation of the first complete molt. To investigate the occurrence and extent of partial first-cycle molts in Falconiformes, I examined 1227 speci- mens of 20 North American species collected dur- ing their first year (age == 0-12 mo), prior to ini- tiation of primary molt, and 622 specimens collected in their second year (12-24 mo old) or later. Methods Specimens of Falconiformes were examined at the Cal- ifornia Academy of Sciences (CAS), San Francisco, the Museum of Vertebrate Zoology (MVZ), Berkeley, and the National Museum of Natural History (USNM), Washing- ton, DC. Specimens were collected throughout North, Central, and South America, but all specimens repre- sented species or subspecies exhibiting boreal breeding cycles. Data are presented for 20 species with A s 16 first-year individuals examined (Table 1). Specimens examined included birds collected during their first year of life, prior to the shedding of primaries during the first complete molt, and birds determined to be in their second year of life. Age of first-year birds was determined by plumage features (Palmer 1988, Wheeler 2003), the presence of indicative fault bars (Hammers- trom 1967), tapered and relatively worn outer primaries and rectrices, and absence of wear and color patterns among flight feathers indicating previous molts. Birds in their second year were aged by the presence of predefi- nitive plumage in some species or the retention of Juve- nal feathers during the first “complete” molt, particular- ly among the lesser coverts, on the rump, or within the secondaries (Wheeler 2003). Each first-year specimen was examined carefully for re- placed body feathers in patterns indicating molt. Body molt in Falconiformes typically begins on the head and throat and proceeds caudally (Palmer 1988, Wheeler 2003). Thus, feathers showing wear patterns suggesting replacement in a caudal direction were assumed to have resulted from molt rather than adventitious replacement (e.g., after accidental loss). Replaced feathers were often intermediate in color or patterning between those of first-year and second-year birds (Fig. 1), facilitating their identification. The proportion of newly-replaced body feathers, to the nearest 5%, was estimated for each spec- imen. Those showing <2.5% replacement were scored as 0%, further ensuring that birds with adventitiously re- placed feathers were not included in the sample evincing molt. All primaries, secondaries, and rectrices were also examined for evidence of symmetrical replacement in- dicating molt. Secondaries were numbered proximally from the outermost (si) to the innermost (si 3 in most species) feather. Second-year birds were examined for uniformity of feather generations, particularly in tracts for which par- tial molts were detected in first-year birds. The goal of this examination was to assess whether or not feathers replaced during the first year, prior to initiation of pri- mary molt, had been replayed for a second time during the first complete molt. Results Partial First-year Molts in White-tailed Kite and American Kestrel. I examined 22 specimens of 380 Pyle VoL. 39, No. 4 Table 1. Proportion of individuals showing evidence of molt and maximum percent of body feathers replaced, among specimens of 18 species of North American Falconiformes, during three time periods within the first year Values represent sample size of specimens examined, proportion of individuals showing evidence of molt, and max- imum proportion of body feathers replaced among individuals sampled. September-November December-February March- May Species N Proportion Molting Maximum Percent Feathers Replaced N Proportion Molting Maximum Percent Feathers Replaced N Proportion Molting Maximum Percent Feathers Replaced Osprey {Pandion haliaetus) 10 0.40 10% 3 1.00 20% 2 1.00 35% Bald Eagle {Haliaeetus leucocephalus) 8 0.12 5% 5 0.60 30% 3 1.00 40% Northern Harrier {Circus cyaneus) 35 0.00 0% 18 0.33 15% 16 0.44 50% Sharp-shinned Hawk {Accipiter striatus) 80 0.00 0% 43 0.12 10% 48 0.31 45% Cooper’s Hawk {Accipiter cooperii) 59 0.08 10% 26 0.31 10% 14 0.64 10% Northern Goshawk {Accipiter gentilis) 30 0.00 0% 6 0.00 0% 10 0.00 0% Gray Hawk {Asturina nitida) 12 0.00 0% 7 0.14 5% 6 0.33 10% Red-shouldered Hawk {Buteo lineatus) 11 0.45 10% 11 0.45 10% 2 0.00 0% Broad-winged Hawk {Buteo platypterus) 10 0.00 0% 9 0.00 0% 7 0.00 0% Swainson’s Hawk {Buteo swains oni) 7 0.57 10% 5 1.00 25% 15 0.86 40% Red-tailed Hawk {Buteo jamaicensis) 48 0.21 5% 54 0.26 10% 27 0.37 20% Ferruginous Hawk {Buteo regalis) 23 0.13 5% 10 0.50 5% 5 0.60 10% Rough-legged Hawk {Buteo lagopus) 16 0.31 5% 10 0.50 15% 4 1.00 10% Golden Eagle {Aquila chrysaetos) 6 0.17 5% 9 0.22 5% 3 1.00 5% Merlin {Falco columbarius) 51 0.04 5% 33 0.33 15% 18 0.50 50% Gyrfalcon {Falco rusticolus) 15 0.07 10% 12 0.42 30% 5 0.20 25% Peregrine Falcon {Falco peregrinus) 27 0.19 10% 13 0.46 25% 13 0.85 20% Prairie Falcon {Falco mexicanus) 24 0.29 15% 12 0.50 15% 9 0.89 20% White-tailed Kite and 183 specimens of American Kestrel collected between September and May of their first year. A partial to complete body-feather molt during the first fall was confirmed for all in- dividuals of both of these species. Proportions of replaced feathers indicated that, in both species, this molt had completed or nearly completed by December, with little or no additional molt taking place during January-May, until initiation of the first complete molt. Examination of six White- tailed Kites and 15 American Kestrels undergoing their first complete molt and 73 kites and 274 kes- trels in definitive plumage (showing uniform body plumage) confirmed that another complete body December 2005 First-cycle Molts in Falconiformes 381 A B C D Figure 1. Pigraent patterns on underpart feathers of Cooper’s Hawks. A — Typical Juvenal feather; B — Forma- tive feather replaced in September (CAS62145); C — For- mative feather replaced in January (CAS33560); D — Typ- ical definitive feather. molt occurs in these species concurrent with the flight-feather molt at a year of age. In White-tailed Kite, 13 first-year individuals col- lected between January and the initiation of the first complete molt (with shedding of the inner- most primaries) had replaced a mean 97.7% of body feathers. All 13 kites had replaced at least some lesser coverts (but no median coverts) and the central two to all 12 rectrices {x = 6.9 feath- ers) . In addition, five individuals were found that had replaced 2-6 secondaries during the partial molt, symmetrically on both wings and apparently in the same sequence as observed in complete molts, as determined from dines in wear. The two specimens with the most extensive partial first-year molts were collected in California: MVZ24493 (10 January) and CAS73280 (22 February), each with 100% of the body feathers, all 12 rectrices, and sl- s2, s5, and sll-sl3 replaced. No primaries had been replaced during this molt in any individual. In American Kestrel, 92 individuals collected be- tween their first January and the initiation of the first complete molt had replaced a mean of 90.8% of body feathers. The amount of feathers replaced ranged from 40% of the body and no wing coverts (CAS 73504 collected 26 April) to 100% of the body feathers, all lesser coverts, and up to three proximal median coverts, but no flight feathers (several specimens). First-year Molts in other North American Falco- niformes. Evidence of body-feather replacement prior to initiation of primary molt was recorded in 16 North American Falconiformes (all but North- ern Goshawk and Broad-winged Hawk; Table 1). Replaced feathers were observed primarily on the back and breast (Fig. 2). Specimens collected in fall and early winter had replaced some feathers on the crown, throat, and upper back; whereas many spring specimens had replaced larger scap- ulars and some breast feathers, but had retained most to all feathers of the crown, upper back, and throat. Newly-replaced feathers on the underparts tended to show patterns resembling Juvenal feath- ers when molted in fall and definitive feathers when molted in spring, with a clinal rate of pattern change with time (Fig. 1). The seasonal timing at which definitive characteristics in these feathers were acquired appeared to vary, occurring during the late fall and winter in some species (e.g., Sharp- shinned Hawk, Red-shouldered Hawk, and Pere- grine Falcon) and during the spring or later in oth- ers (e.g., Swainson’s and Rough-legged hawks). Extent of this molt showed substantial intraspe- cific and interspecific variation (Table 1). Among the 16 species showing these molts, the mean max- imum recorded extent for all species combined was 25.6%, varying from 5% in Golden Eagle to 50% in Northern Harrier and Merlin (Table 1). Examples of specimens showing extensive body- feather replacement included Northern Harrier MVZ144731 (collected in March with 35% replace- ment), Sharp-shinned Hawk MVZ99723 (May, 45%), Swainson’s Hawk CAS13889 (April, 40%), Red-tailed Hawk CAS27181 (April, 20%), Rough- legged Hawk MVZ173433 (December, 20%), and Peregrine Falcon CAS73587 (December, 25%; Fig. 2). On the other hand, at least some individuals (11-80%) of 12 species (or 11—100% of 14 species when Northern Goshawk and Broad-winged Hawk are included) collected in March-May showed no feather replacement (Table 1), and six specimens were recorded that had begun shedding primaries during the first complete molt, but remained in complete Juvenal body plumage. No first-year birds of these 22 species showed symmetrical replace- ment of any flight feathers prior to initiation of the first complete molt. Two replacement patterns according to season were observed among these 16 species (Table 1, Fig. 3). In Northern Harrier, Sharp-shinned, Coo- per’s, Gray, and Ferruginous hawks. Merlin, and Peregrine and Prairie falcons, molt had com- menced in few birds in fall, some birds in winter, and most birds in spring. In Osprey and Red-shoul- dered, Red-tailed, and Rough-legged hawks, molt had occurred in some birds in fall, appeared to be suspended in many birds over winter, and was re- 382 Pyle VoL. 39, No. 4 Figure 2. Dorsal and ventral aspects of a first-year Peregrine Falcon {F. p. anatum) collected 15 December in Cali- fornia (CAS73587) showing evidence of 25% body-feather replacement during a preformative molt (see Discussion). Formative feathers on the dorsum are bluish and barred, and formadve feathers on the underparts are unstreaked (breast) and barred (belly and flanks) , resembling definitive feathers in these regions. sumed or initiated in some birds in spring. The patterns for Bald and Golden eagles, Swainson’s Hawk, and Gyrfalcon appeared to be intermediate, with molt occurring throughout the first year (Ta- ble 1). Feather-replacement Patterns in Second-year Birds. Totals of 27 birds collected while undergo- ing the first complete molt and 146 second-year (12-24 mo-old) birds following completion of this molt were examined, including at least six individ- uals of all species in Table 1 , except for Gray Hawk, Red-shouldered Hawk, and Merlin, for which sec- ond-year and older individuals could not be distin- guished. For the smaller species (including all fal- cons) , there was no evidence that feathers replaced during the first fall, winter, or spring had been re- tained during the first complete molt. All 27 birds undergoing this complete molt appeared to be re- placing all body feathers (or most feathers in the case of the larger species; see below). On second- year birds that had completed body molt, all scap- ulars as well as crown, back, and underpart feathers were uniform in wear, reflecting a complete molt within a relatively short time period. For the three species mentioned above, examination of 75 birds in definitive plumage (likely including second-year birds) also showed no variation in feather wear. For five of the larger species, Osprey, Bald and Golden December 2005 First-cycle Molts in Falconiformes 383 Month Figure 3. Proportion of individuals evincing body molt, by month, in two species of North American Falconifor- mes showing differing seasonal-replacement strategies. eagles, and Red-tailed and Ferruginous hawks, var- iation in wear among the body feathers precluded confirmation that feathers replaced during the first year were replaced again during the first complete molt. In the two species of eagle, up to many Ju- venal body feathers could be retained during the first “complete” molt, so it is possible that feathers replaced during the first year may also have been retained during this molt. Discussion Traditionally, both the partial first-fall molts of North American Kestrels and kites and the com- plete molts of other 1-yr-old Falconiformes have been considered the “first prebasic molts,” accord- ing to the molt terminology of Humphrey and Parkes (1959). However, Howell and Corben (2000) suggested that the complete first prebasic molts of most Falconiformes may be homologous with the second prebasic molts of kestrels, kites, and most other birds. Accordingly, Howell et al. (2003) proposed a revised molt terminology for the first cycle, redefining the prejuvenal molt as syn- onymous with the first prebasic molt, the complete molt at the end of the first molt cycle as the second prebasic molt, and any inserted molts during the first cycle as preformative molts. Evidence of body-feather replacement during the first year was found in 18 of 20 species exam- ined and, for at least 15 species, the “first complete molt” appeared to include those feathers replaced during the first cycle. By definition (Humphrey and Parkes 1959, Howell et al. 2003), the addition- al body-feather replacement during the first year should be considered separate, inserted molts; us- ing the terminology of Howell et al. (2003), they should be considered preformative molts followed by the complete or near-complete second prebasic molt at a year of age. Under the terminology of Humphrey and Parkes (1959), the “first prebasic molt” referred to the limited body-feather molt during the first fall, winter, or spring of some in- dividuals or species and to the complete molt at a year of age in other individuals, even within species (cf. Elanus caeruleus in Marchant and Higgins 1993), and presumed homology between individ- uals and species was lost. In the remaining two species. Northern Gos- hawk and Broad-winged Hawk, evidence of prefor- mative molts may have been missed in this study due to low sample sizes, especially in spring. In- deed, body-feather replacement during the first year has been reported for Northern Goshawk (Bent 1937, Dement’ ev and Gladkov 1951, Fors- man 1999). A first-year Broad-winged Hawk (Slater Museum of Natural History No. 2367) collected 1 June reportedly had replaced some breast feathers (Clark and Anderson 1984), although this individ- ual had initiated the second prebasic molt (D. Paulson pers. comm.). It is possible that the con- straints of migration may preclude the occurrence of a preformative molt in Broad-winged Hawk; however, substantial evidence of this molt was found in the migratory Swainson’s Hawk. In gen- eral, birds that migrate to the tropics or Southern Hemisphere often display more extensive first-win- ter molts, possibly due to more abundant and sta- ble food resources and greater day-lengths with which to forage (Myers et al. 1985, Pyle 1998). Thus, preformative molts should be expected in some Broad-winged Hawks. The evidence, there- fore, suggests that preformative molts likely occur in at least some individuals of all North American Falconiformes. Including the pattern of White-tailed Kite and American Kestrel, three seasonal strategies of pre- formative molt were identified. Consideration of the ecology and life history of the species com- prising each group revealed no evident explana- tions for conditions or constraints leading to each molt pattern. There appeared to be a slight phy- logenetic component, with a majority of species among Accipiter and Falco delaying preformative molts until winter or spring (see Herremans and Louette 2000) , whereas more species among Buteo initiated preformative molts in fall. Northerly breeding and wintering species also tended to 384 Pyle VoL. 39, No. 4 show a greater amount of preformative molt in fall; perhaps first-year birds of these species can take advantage of abundant food resources in the fall, but suspend molting during winter when food be- comes scarce. Such variation might also be expect- ed within species that show a wide latitudinal breeding range. Within genera, smaller species (in- cluding Mdiite-tailed Kite and American Kestrel) generally showed higher proportions of birds molt- ing a greater amount of feathers. This correlation is expected based on the added energy required to replace larger feathers (Lindstrom et al. 1993) , and may also be part of a signaling mechanism for species more likely to undergo breeding in their first year (Kemp 1999, Herremans and Louette 2000). Finally, species inhabiting open areas ex- posed to higher amounts of UV radiation appeared to undergo more extensive preformative molts. However, there were exceptions to all of these pat- terns, and it is likely that the extent and timing of preformative molts in Falconiformes reflect various constraints according to a complex combination of species-specific, breeding, migrating, and foraging strategies. Replacement patterns by season indicate that breast feathers and scapulars can be molted later in winter or spring during preformative molts, while juvenal crown, throat, and upper back feath- ers had been retained, an unusual sequence of feather replacement for Falconiformes (Palmer 1988, Wheeler 2003). This suggests a triggering mechanism for the preformative molt, by which body molt of some feathers is bypassed, resulting in spring birds initiating molt at a later point in the sequence. Thus, sequence as well as extent may vary individually, depending on initiation date. Similarly, hormonal processes controlling feather pattern appear to develop clinally throughout the first year (and sometimes beyond), resulting in de- layed acquisition of definitive plumage that varies m timing by species. In Swainson’s Hawk and Crested Caracara ( Caracara cheriway ) , acquisition of definitive plumage is delayed until the following summer or later, resulting in identifiable second- basic plumages, intermediate in pattern between juvenal and definitive plumages (Wheeler 2003). For the two eagle species, acquisition of definitive plumage requires up to 4 or 5 yr to complete. Within many species, the preformative molt ap- pears to occur in only a proportion of individuals, with some birds retaining full Juvenal plumage un- til the second prebasic molt. Similar variation in the preformative molt was found among Eurasian Accipiters (Herremans and Louette 2000). Thus, some individuals exhibit a “Simple Basic Strategy” (lacking a preformative molt) whereas others ex- hibit a “Complex Basic Strategy ” (including a pre- formative molt, but lacking prealternate molts) ac- cording to Howell et al. (2003). However, at the species level, Falconiformes are best described as exhibiting the Complex Basic Strategy, with the preformative molt ranging in extent from absent to partial. Evolutionarily, the existence of these variable and at times ephemeral preformative molts may suggest that ancestral Falconiformes exhibited more extensive preformative molts that have be- come vestigial in most (larger) species that do not breed at age one and can commence the second prebasic molt at an earlier age (Wheeler 2003). Phylogenetic evidence suggests that ancestral Fal- coniformes branched from a common ancestor that also included Podicipediformes, Pelecanifor- mes, and Ciconiformes (Sibley and Ahlquist 1990), orders which currently display extensive prefor- mative molts (Palmer 1962, Howell et al. 2003). Alternatively, it is possible that these preformative molts in Falconiformes have become inserted over time, from an ancestral species that lacked such molts; however, the apparent lack of functionality for these molts may argue against this alternative hypothesis. A better understanding of molt and plumage homologies in Falconiformes awaits fur- ther study of both molts and phylogenetic relation- ships. Acknowledgments I thank Jack Dumbacher and Douglas J. Long (CAS), Carla Cicero (MVZ), and James Dean (USNM) for access to collections under their care and Dennis Paulson of the Slater Museum of Natural History, Puget Sound, WA, for examining a specimen of Broad-winged Hawk for me. The manuscript benefited from comments by Steve N.G. Howell, Christopher W. Thompson, William S. Clark, Jo- sef K. Schmutz, and Marc Herremans. Funding for the study of molts and plumages was provided by the Neo- tropical Migratory Bird Conservation Act (grant No 2601) administered by the U.S. Fish and Wildlife Service. This is Contribution No. 232 of The Institute for Bird Populations. Literature Cited Bent, A.C. 1937. Life histories of North American rap- tors. Part 1. U.S. Nat. Mus. Bull. 167:1-409. . 1938. Life histories of North American raptors. Part 2. U.S. Natl. Mus. Bull. 170:1-482. Clark, W.S. and C.M. Anderson. 1984. First specimen December 2005 First-cycle Molts in Falconiformes 385 record of the Broad-winged Hawk for Washington. Murrelet 65:93-94. Cramp, S. and K.E.L. Simmons. 1980. The birds of the western Palearctic. Vol. II. Oxford University Press, Oxford, England. Dement’ev, G.P. and N.A. Gladkov. 1951. Birds of the Soviet Union. Vol. I. Gosudarstvennoe Izdatel’stvo “Sovetskaya Nauka,” Moscow, Russia. Dickey, D.R. and A.J. van Rossem. 1938. The birds of El Salvador. Field Mus. Nat. Hist. Publ. Zool. Sen 23:1-609. Eorsman, D. 1999. The raptors of Europe and the Middle East. T. and A.D. Poyser, Ltd., London, England. Hammerstrom, E. 1967. On the use of fault bars in ageing birds of prey. Inland Bird-Banding News 39:35-41. Herremans, M. and M. Louette. 2000. A partial post- juvenile molt and transitional plumage in the Shikra {Acdpiter badius) and Grey Erog Hawk {Accipiter soloen- sis).J. Raptor Res. 34:249-261. Howell, S.N.G. and G. Corben. 2000. A commentary on molt and plumage terminology: implications from the Western Gull. West. Birds 31:50-56. , C. Corben, P. Pvle, and D.I. Rogers. 2003. The first basic problem: A review of molt and plumage homologies. Cowdor 105:635-653. Humphrey, P.S. and K.C. Parkes. 1959. An approach to the study of molts and plumages. Auk 76:1-31. Kemp, A.C. 1999. Plumage development and visual com- munication in the Greater Kestrel (Falco rupicoloides) near Pretoria, South Africa. Ostrich 70:220—224. Lindstrom, a, G.H. Visser, and S. Daan. 1993. The en- ergetic costs of feather synthesis is proportional to basal metabolic rate. Phys. Zool. 66:490-510. Marchant, S. and P.J. Higgins. 1993. Handbook of Aus- tralian, New Zealand, and Antarctic Birds. Vol. 2. Ox- ford University Press, Oxford, England. Miller, A.H. 1941. The significance of molt centers among secondary remiges in the Ealconiformes, Con- dor 43:113-115. Miller, K.E, and J.A. Smaixwood. 1997. Juvenal plum- age characteristics of male southeastern American Kestrels {Falco sparverius paulus) . J. Raptor Res. 31:273- 274. Myers, J.P.,J.L. Maron, and M. Sallaberry. 1985. Going to extremes: why do Sanderlings migrate to the trop- ics? Ornithol. Monogr. 36:520—535. Palmer, R.S. 1962. Handbook of North American birds. Vol. 1. Loons through flamingos. Yale University Press, New Haven, CT U.S.A. . 1988. Handbook of North American birds. Vols. 4 and 5. Diurnal raptors (Parts 1 and 2). Yale Univer- sity Press, New Haven, CT U.S.A. Parkes, K.C. 1955. Notes on the molts and plumages of the Sparrow Hawk. Wilson Bull. 67:194—199. Pyle, P. 1998. Eccentric first-year molt patterns in certain tyrannid flycatchers. West. Birds 29:29-35. Sibley, C.G. and J.E. Ahlquist. 1990. Phylogeny and clas- sification of birds. Yale University Press, New Haven, CT U.S.A. Wheeler, B.K. 2003. Raptors of western North America. Princeton University Press, Princeton, NJ U.S.A. Received 12 Eebruary 2004; accepted 6 June 2005 Associate Editor: Ian G. Warkentin J Raptor Res. 39(4) :386— 393 © 2005 The Raptor Research Foundation, Inc. MORPHOMETRIC ANALYSIS OF LARGE FALCO SPECIES AND THEIR HYBRIDS WITH IMPLICATIONS FOR CONSERVATION Chris P. Eastham^ National Avian Research Centre, The Falcon Facility, Penllynin Farm, College Road, Carmarthen, Wales SA33 5EH United Kingdom Mike K. Nicholes^ Ecology Research Group, Canterbury Christ Church University College, Canterbury, Kent, CTl IQU United Kingdom Abstract. — Morphometric examination of several large falcon species and their hybrids was conducted to ascertain whether phenotype was an accurate indicator of hybrid parentage. Six external body measure- ments were recorded from 167 Gyrfalcons {Falco rusticolus), Saker {F. cherrug). Peregrine (F. peregrinus), and New Zealand Falcons {E novaezeelandiae) and from 100 FI, F2, and backcross hybrids of these species. Principal Component Analysis separated pure species and also indicated clusters for FI peregrine X saker, gyr X peregrine, and gyr X saker hybrids. Gyr X peregrine hybrids were distinguishable from their parent species, but it was impossible to discriminate accurately between a complex (FI, F2, and backcross) of gyr X saker hybrids and between these and the parent species. Escaped or released falconry hybrids are perceived as a significant threat to the conservation of wild falcon populations. Under current legislation, gyrs and their hybrids are CITES Appendix I species, and sakers are Appendix II species. We suggest that phenotypic characteristics are not reliable for identification of such hybrids for legal purposes. Further- more, analysis of measurements also identified a “paternal effect,” whereby Fj hybrids, irrespective of gender, were phenotypically more similar to their paternal than their maternal progenitors. Keywords: Peregrine Falcon', Falco peregrinus; Gyrfalcon-, Falco rusticolus; Saker Falcon-, Falco cherrug; New Zealand Falcon', Falco novaezeelandiae; Falcon hybrids, morphometric, principal component analysis:, PCA; CITES. anAlisis morfometrico de las especies de ealco de tamano grande y de sus HIBRIDOS, E IMPLICACIONES PARA LA CONSERVACION Resumen. — Se analizo la morfologia de varias especies de halcones y de sus hfbridos para averiguar si el fenotipo es un indicador precise de la paternidad de los hfbridos. Se registraron seis medidas cor- porales para un total de 167 individuos pertenecientes a las especies Ealco rusticolus, F. cherrug, E peregrinus y E novaezeelandiae, y para un total de 100 hfbridos FI, F2 y retrocruces de estas especies. Un analisis de componentes principales separo a las especies puras e identified grupos formados por hfbridos FI E peregrinus X E. cherrug, E rusticolus X E peregrinus y E. rusticolus X E cherrug. Los hfbridos F. rusticolus X F. peregrinus se diferenciaron de las especies parentales, pero fue imposible distinguir claramente entre un complejo (FI, F2, retrocruces) de hfbridos F. rusticolus X F. cherrug, y entre este complejo y las especies parentales. Los halcones hfbridos de cetrerfa que escapan o son liberados se consideran una amenaza para la conservacidn de las poblaciones silvestres. Bajo la actual legislacidn, F. rusticolus y sus hfbridos estan registradas en el Apendice I de CITES y F. cherrug en el Apendice II. Consideramos que las caracterfsticas fenotfpicas no son confiables para la identificacidn de estos hfbridos con propdsitos legales. Ademas, el analisis morfometrico identified “efectos paternos,” en donde los hfbridos FI, in- dependientemente de su sexo, fueron fenotfpicamente mas similares a sus progenitores paternos que a los maternos. [Traduccidn del equipo editorial] ^ Present address: Klumpstugevagen 1, 61892 Kolmarden, Sweden. ^ Corresponding author’s address: LEAP, University of Greenwich at Medway, Chatham Maritime, Kent ME4 4TB, United Kingdom. Email: M.Nicholls@gre.ac.uk One of the first domestic hybrid falcons was pro- duced in 1971 from a female Saker Falcon {Falco cherrug and male Peregrine Falcon {F. peregrinus, Morris and Stevens 1971, Morris 1972). Since then, falconers and raptor breeders have produced many different hybrids from members of the Falconifor- 386 December 2005 Morphometrics oe Falcon Hybrids 387 Table 1. Identity of hybrid falcons used in the analysis. Hybrid Identity^ Samples Sizes Male Parent Fkmat.f. Parent FIs Gyr X Peregrine 6 d, 7 9 Gyr Peregrine Gyr X Saker 7 c?, 5 9 Gyr Saker Peregrine X Saker 3 c?, 13 9 Peregrine Saker Peregrine X Gyr 1 c? Peregrine Gyr Peregrine X New Zealand 1 d Peregrine New Zealand Gyr/Saker X Peregrine 1 d, 1 9 Gyr X Saker F2 hybrid Peregrine F2s Gyr X Saker 4 d, 5 9 Gyr X Saker FI hybrid Gyr X Saker FI hybrid Backcrosses — 1st generation Gyr X Gyr/Saker 3 d, 1 9 Gyr Gyr X Saker FI hybrid Saker X Gyr/Saker 3 d, 3 9 Saker Gyr X Saker FI hybrid Gyr/Saker X Saker 14 d, 17 9 Gyr X Saker FI hybrid Saker Gyr/Peregrine X Peregrine 2 d Gyr X Peregrine FI hybrid Peregrine Backcrosses — 2nd generation Gyr (3/8)VSaker 1 d Gyr X Saker FI hybrid Gyr/Saker X Saker (backcross hybrid) Gyr (5/8)VSaker 1 d, 1 9 Gyr Gyr/Saker X Saker (backcross hybrid) ® When naming hybrids, the paternal species is cited first. For example, a cross between a male gyr and female saker is a gyr X saker (or gyr/ saker) hybrid, whereas a male saker crossed with a female gyr is a saker X gyr (saker/ gyr) hybrid. These numbers represent a simple way to show the proportion of genes from the parent species, assuming that a FI hybrid inherits Vz of the genes from both the male and female parent species. For example, a gyr (%) /saker, produced by backcrossing a gyr X saker FI hybrid with a gyr/ saker backcross hybrid, has \ gyrfalcon and \ saker genes. mes (Boyd and Boyd 1975, Cade and Weaver 1976, Bunnell 1986, Weaver and Cade 1991), including intergeneric hybrids (e.g., Harris’s Hawks {Para- buteo unicinctus] X Cooper’s Hawk [Aedpiter coope- ni\ and Harris’s Hawk X Ferruginous Hawk {Buteo regalis] ; Fox and Sherrod 1999a) for falconry pur- poses. Many Fj hybrids are fully viable (Heidenreich 1997), in their turn producing Fg hybrids or back- crosses (Bj and B^ representing 1st and 2nd gen- eration backcrosses) to one or other parent spe- cies. Indeed, hybrids from within the subgenus Hierofalco, the “desert falcon” group (Heidenreich 1997), exhibit full fertility, presumably over indef- inite generations. Less closely-related pairs of spe- cies, such as gyr (F. rusticolus) and peregrine, pro- duce hybrids with reduced fertility, manifest as deformed spermatozoa, completely sterile females (Heidenreich and Kuspert 1992), or unviable em- bryos (Rosenkranz 1995). This extended viability of some falcon hybrids coupled with increasing demand over the last 10 yr for domestic falcons from North American, Eu- ropean, and Arabian falconry markets (Fox and Sherrod 1999b) has prompted conservation con- cerns. Escaped domestic hybrids may be merely a curiosity (Forseman 1999), a nuisance for bird watchers (Gantlett and Millington 1992), or a threat to the integrity of wild populations. Indeed, falcon pairs made up of an escaped hybrid and a wild, pure individual have been documented sev- eral times (e.g., Kleinstauber and Seeber 2000, Lindberg 2000) . A further conservation issue pre- sumably concerns illegal-trade in falcons, whereby protected falcon species may be “laundered” as domestic hybrids. In this study, we examine the re- lationship between falcon species and their hy- brids, particularly the accuracy of using morpho- metric characters for identification, and discuss the conservation issues concerning falcon hybrids. Methods We investigated four large falcon species, namely Per- egrine, Gyr, Saker, and New Zealand falcons {F. novaezee- landiae) and several of their hybrid types used for falcon- ry. Hybrid falcons were all bred in captivity and, therefore, their parentage was known. Six external body 388 Eastham and Nicholls VoL. 39, No. 4 Table 2. Principal Component Analysis (PCA) of six anatomical measurements from juvenile male Gyrfalcon, Saker, Peregrine, and New Zealand falcon species and hybrids of those species. Eigenvalues and eigenvectors (based on the correlation matrix). Principai. Component 1 2 3 4 Eigenvalue 3.0328 1.3691 0.7459 0.6833 Percent of variability 0.5055 0.2282 0.1243 0.1139 Cumulated Percent 0.5055 0.7336 0.8580 0.9718 Characters Eigenvectors Wing chord 0.4990 -0.2324 0.1490 -0.3852 Wing width 0.5552 -0.0321 -0.0688 -0.1697 Tail length 0.5195 0.2308 -0.2347 -0.1471 Tail step 0.0247 0.6402 0.7536 -0.1395 Tarsus length 0.3536 0.3591 -0.1598 0.7769 Digit three length 0.2170 -0.5940 0.5697 0.4221 measurements were collected from live juvenile Gyrfal- cons {N = 7 males and 6 females) , Saker {N = 34 males and 40 females). Peregrine {N = 17 males and 24 fe- males), and New Zealand falcons {N = 25 males and 14 females), and their various hybrids (Table 1). Apart from European Peregrine Falcon subspecies, the majority of which were F. peregrinus peregrinus, no other differentia- tion was made between subspecies or geographic morphs. All birds were kept at the National Avian Re- search Center’s Falcon Facility in Carmarthen, Wales, U.K. The majority of Fg and backcross 1 and 2 hybrids were between gyrs and sakers. This is because hybrids between members of the subgenus Hierofalco remain fer- tile for an indefinite number of generations, whereas hy- brids between more out-crossed falcon species, such as peregrines and sakers, have a reduced fertility. Some of the hybrids included are produced in very low numbers (e.g., Peregrine Falcon X New Zealand Falcon), and pub- lished data on these are rare. Therefore, we included them in the analysis. One of us (C. Eastham) took six measurements, name- ly wing chord length and width, tail length, tail step (the difference between the outermost tail feather [rectrix 6] and the tip of the center tail feather [rectrix 1] on the same side), tarsus length, and third digit length from each bird. Measurement protocols followed standard methods described by Baldwin et al. (1931), Fox (1977), Biggs et al. (1978), Kemp (1987), and Fox et al. (1997). Feather characters were measured to the nearest 1 mm and non-feather characters to the nearest 0.1 mm using a pair of digital calipers, a steel ruler, and tape measure. Inclusion of single individuals, for example, a male Per- egrine Falcon X New Zealand Falcon, allowed us to em- ploy Principal Component Analysis (PCA) on XLSTAT- Pro (Fahmy 1998) statistical software as a suitable method for data analysis. Male and female data were analyzed separately to eliminate background variation due to re- versed sexual size dimorphism (Brown and Amadon 1968, Cade 1982), or other sex-linked or sex-limited char- acters. We used wing chord length to distinguish males and females, as this measurement was reported by Wyllie and Newton (1994) and Eastham (2000) to be the most reliable indicator of overall body size. Results Males. Principal Component (PC) 1 (Table 2, Fig. 1) accounted for the majority (50.5%) of var- iation. Because all eigenvectors for PCI showed positive and nearly equal values, we concluded this component represents overall body size (Wiley 1981). Male gyrs and FI and F2 gyr/saker hybrids had the largest body size, whilst male peregrines and New Zealand Falcons were the smallest. PC 2 (Table 2, Fig. 1) accounted for 22.8% of the variation, as indicated by a contrast in eigen- vectors between the positively weighted tail step, tail and tarsus length, the negatively weighted digit three length, and wing chord length and width (Table 2). This component summarizes variation related to body shape. Tail step and digit three length showed the strongest positive and negative weightings, respectively. With a low negative weighting, wing width was of limited use in further analysis of PC 2. New Zealand Falcons had the rel- atively longest tail step (indicating a more rounded tail) , tail and tarsus length, and shortest digit three and wing chord, whilst peregrines, the single per- egrine X New Zealand hybrid and gyr/ saker X per- egrine hybrid had the relatively longest digit three and wing chord length and shortest tail step and tail and tarsus length. FI gyr X peregrines and sa- kers showed a wide variation in PC 2 values, with an individual saker having the highest PC 2 value. December 2005 Morphometrics of Falcon Hybrids 389 Principal Component 1 • Gyrfalcon BSaker JKPeregrine -New Zealand falcon AGyr/ Peregrine +Gyr/Saker- FI • Gyr / Saker - F2 X Peregrine / Saker A Peregrine / New Zealand ^ Gyr x Gyr/ Saker □ Gyr/ Saker x Saker OSakerx Gyr/ Saker :: Gyr (5/8) / Saker SSGyr/ Saker x Peregrine Figure 1. Principal component scores from morphometric comparison of various male falcon species and their hybrids. PC 3 and 4 (Table 2) accounted for only 12.4% and 11.3% of the residual variation, respectively. Tail step and digit three length had a high positive weighting in PC 3, and in PC 4, there was a con- trast between positively weighted tarsus and digit three length and negatively weighted wing chord length. As PC 1 and 2 together accounted for the majority (73%) of variation, we did not consider PC 3 and 4 further. Females. The PCA for juvenile female falcons showed a similar pattern of variation as that seen in juvenile males. Again PC 1 (Table 3, Fig. 2) ac- counted for the largest proportion (44.7%) of var- iation and as indicated by mostly positive and near- ly equal values represents, as with males, overall size (Wiley 1981). Gyrs and the various gyr/ saker FI, F2 and backcrosses had the largest size, whilst Peregrine and New Zealand falcons were the small- est. PC 2 (Table 3, Fig. 2) accounted for a further 21.8% of the variation, and we concluded that this, again like males, was related to shape. This was in- dicated by a contrast in eigenvectors between tbe positively weighted tail step, tail and tarsus length, the negatively weighted digit three length, and wing chord length and width (Table 3). Positively weighted tail step and negatively weighted digit three and wing chord length had the highest ei- genvectors for this PC. New Zealand Falcons had the longest tail step and the shortest digit three and wing chord length, whilst FI gyr X peregrines and peregrines had the shortest tail length and the longest digit three and wing chord length. PC 3 and PC 4 (Table 3) accounted for 18.4% and 11.4% of the variation, respectively. As for males, we did not consider these principal components further. Discussion Using PCA we found that the four falcon species, irrespective of sex, were clearly separated into groups: New Zealand Falcons with a small size, long rounded tails and tarsi, and short wings; per- egrines, also with a small size, long digit three lengths, and long narrow wings; sakers with a large 390 Eastham and Nicholls VOL. 39, No. 4 Table 3. Principal Component Analysis (PCA) of six anatomical measurements from juvenile female Gyrfalcon, Saker, Peregrine, and New Zealand falcon species and hybrids of those species. Eigenvalues and eigenvectors (based on the correlation matrix). Principal Component 1 2 3 4 Eigenvalue 2.6826 1.3071 1.1023 0.6864 Percent of variability 0.4471 0.2179 0.1837 0.1144 Cumulated percent 0.4471 0.6649 0.8487 0.9631 Characters Eigenvectors Wing chord 0.4717 -0.3853 ■0.1963 0.4118 Wing width 0.5765 -0.1767 ■0.0356 -0.0129 Tail length 0.5737 0.1670 ■0.0642 -0.1107 Tail step 0.0866 0.6865 0.1871 0.6869 Tarsus length 0.3077 0.2169 0.6899 -0.4484 Digit three length -0.1177 -0.5236 0.6672 0.3809 size, long rounded tails, and short digits; and gyrs with the largest size. Using these external body measurements, we also found it possible to identify three main hybrid groups: a complex of FI, F2 and backcross gyr/ sakers; FI gyr X peregrines and FI peregrine X sakers. Further, it was possible to separate gyr X peregrines from their parent species, but impossi- ble to separate completely the FI, F2, and back- cross gyr/ sakers hybrid-complex from pure sakers or particularly, from pure gyrs. Overall, we found that the hybrids were gener- ally of intermediate phenotype between their par- ents. However, beyond this it appears that the pa- ternal progenitor influences the phenotype to a greater extent than maternal. For example, for both males and females, the clusters representing FI gyr X peregrine hybrids, hybrids whose male parents were gyrs, were spatially closer to the gyr clusters than to the peregrine clusters (Fig. 1, 2). Thus, both male and female gyr X peregrine hy- brids have a morphology closer to that of gyrs than to peregrines. Further, that the single female per- egrine X gyr, whose sire was a peregrine, was spa- tially closer to the peregrine cluster than the gyr cluster, adds further weight to this proposed gen- erality. Similarly, the male and female FI gyr X sa- kers, both with gyrs as male parent, appear more gyr-like than saker-like in morphology. Except for male FI peregrine X sakers this “paternal effect” seems true for all species combinations. We ex- plain this by considering that two thirds of the sex linked genes in a population are carried by the homogametic sex (male in birds; Mittwoch 1977) and only one third by the heterogametic sex (Fal- coner 1967). Therefore, falcon sires (the homo- gametic sex) will contribute more sex-linked alleles to their hybrid offspring than will (heterogametic) dams. International trade in endangered species, such as some falcons, can be a profitable enterprise and, if unregulated, can threaten their conservation. Regulation of the trade in such endangered spe- cies is by international agreements, such as the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES). Accord- ing to their degree of endangerment in the wild, all species are classed in one of three CITES ap- pendices. Special conditions apply to the most en- dangered, known as Appendix I species (i.e., those threatened with extinction and whose survival could be impaired by trade) , which allows restrict- ed trade in captive and domestic bred individuals. Appendix II includes species considered less threatened. The saker is an Appendix II species, and although trade is regulated, this is less exact- ing than for Appendix I species. Despite relatively healthy world populations and for reasons which are largely political (White and Kiff 1998), gyrs and peregrines and their hybrids are included in Ap- pendix I. The results we present here show that it can be difficult to discriminate falcon species accurately from their hybrids, especially hybrids of Appendix I gyrs and Appendix II sakers. Similarly, plumage variation, especially in juvenile falcons, is difficult December 2005 Morphometrics of Falcon Hybrids 391 • Gyrfalcon ■ Saker )K Peregrine - New Zealand falcon A Gyr /Peregrine + Gyr /Saker - F1 • Gyr / Saker - F2 X Peregrine / Saker >I< Peregrine / Gyr <>Gyr X Gyr / Saker A Gyr / Peregrine x Peregrine □ Gyr / Saker x Saker O Saker x Gyr / Saker :: Gyr (5/8) / Saker — Gyr (3/8) / Saker SSGyr / Saker x Peregrine Figure 2. Principal component scores from morphometric comparison of various female falcon species and their hybrids. to assess objectively and make comparisons be- tween species and their hybrids. These observa- tions may provide fuel for two separate arguments. Ornithologists are increasingly aware of the widespread genomic compatibility and potential for hybridization amongst what appear to be very dissimilar species (Grant and Grant 1992), such that hybridization between avian species is consid- ered more common than originally thought (Gill 1998). Mayr and Short (1970) estimated that up to 10% of North American bird species regularly hy- bridize; it’s so common that hybrids are even in- cluded in birdwatchers’ field guides (Sibley 2000). The presence of natural hybrids is not believed to be a threat to the integrity of a species, even though they may challenge the biological species concept of taxonomists (Brookes 1999). Amongst free-living, wild birds of prey such hybridization is increasingly documented at the subspecific (Fefe- lov 2001), specific (Hamer et al. 1994), and even to the intergeneric levels (Corso and Gildi 1998, Yosef et al. 2001). Introgressive hybridization may therefore be a process by which species evolve, rather than some- thing that will corrupt them. Thus, if the species concept for birds is much looser than conservation law dictates, then perhaps it is the legislation and not the species concept that must be challenged. For example, the so-called “Altai falcon” (Falco al- taicus Menzbier) , whose phenotype seems to share characters with both gyrs and sakers, is believed by some to be the result of introgressive hybridization between gyrs and sakers, rendering all these as allospecies within a single superspecies (Pfander 1987, Ellis 1995a, 1995b). This being so, then to discriminate between Appendix I gyrs and Appen- dix II sakers may be irrelevant, and artificially pro- duced crosses between these two may be of no evo- lutionary threat should they escape to the wild. If, however, the Altai Falcon is merely a large, dark race of the saker (Eastham and Nicholls 2002) , and any resemblance to gyrs is merely superficial, then escaped hybrids between these species could po- tentially compromise wild populations, and the in- tegrity of gyrs and sakers must be recognized and CITES regulations enforced. A different view of the role of natural hybridiza- tion accepts that avian species are dynamic entities. 392 Eastham and Nicholls VoL. 39, No. 4 which in certain circumstances, freely exchange genes with other such entities. Despite these inter- actions, the integrity of the whole is a fragile one, and to short circuit gene flow via artificial hybrids is a danger to this integrity. CITES protocols are the response to perceived conservation status, and therefore, it should be mandatory to discriminate between species as we know them. Accurate iden- tification to assist in controlling the trade in fal- cons is paramount, and we have shown that criteria other than phenotypic characteristics (e.g., DNA markers) must be employed to identify individuals and species. This is imperative if CITES is to re- main an effective means of regulating legitimate trade and protecting species in the wild. Acknowled gments Thanks to His Highness Sheikh Khalifa bin Zayed A1 Nahyan and His Highness Sheikh Hamdan bin Zayed A1 Nahyan for sponsoring this research. Thanks also to M. A1 Bowardi, managing director of the Environmental Re- search and Wildlife Development Agency (ERWDA), Drs. Nick Fox, Director of the ERWDA Falcon Programme, Robert Kenward, and Juan Negro and Violeta Munoz for comments on this manuscript. Literature Cited Batdwin, S.R, H.C. Oberholser, and L.G. Worley. 1931. Measurements of birds. Sci, Publ. Clevel. Mus. Nat. Hist. 2:1-65. Biggs, H.C., R. Biggs, and A.C. Kemp. 1978. Measure- ments of raptors. Pages 77-82 in A.C. Kemp [Ed.], Proceedings of a Symposium on African Predatory Birds. Northern Transvaal Ornithological Society, Pre- toria, South Africa. Boyd, L.L. and N. Boyd. 1975. Hybrid falcons. Hawk Chalk 14:53-54. Brookes, M. 1999. Live and let live. New Scientist 2193: 32-36. Brown, L. and D. Amadon. 1968. Eagles, hawks and fal- cons of the world. Vol. 2. The Wellfleet Press, Secau- cus, NJ U.S.A. Bunnell, S. 1986. Hybrid falcon overview — 1985. Hawk Chalk 25:43-47. Burton, J.F. 1995. Birds and climate change. Christopher Helm Publishers, London, England. Cade, T.J. 1982. Falcons of the world. William Collins Sons and Co. Ltd., London, England. andJ.D. Weaver. 1976. Gyrfalcon X peregrine hy- brids produced by artificial insemination./. North Am. Falconers Assoc. 15:42-47. CORSO, A. AND R. Gildi. 1998. Hybrids of Black Kite and Common Buzzard in Italy in 1996. Dutch Birding 20: 226-233. Eastham, C.P. 2000. Morphological studies of taxonomy of the Saker (Falco cherrug; Gray 1833) and closely allied species. Ph.D. thesis. University of Kent at Can- terbury, Canterbury, Kent, England. and M.K. Nicholls. 2002. Morphological classi- fication of the so-called “Altai Falcon.” Pages 211-219 in R. Yosef, M.L. Miller, D. Pepler [Eds.], Raptors in the new millennium. Int. Birding and Res. Centre, Eilat, Israel. Ellis, D.H. 1995a. The Altai falcon: origin, morphology, and distribution. Pages 143-168 m Middle East Falcon Research Group [Eds.], Proceedings of the Specialist Workshop. Abu Dhabi, United Arab Emirates. . 1995b. What is Falco altaicus Menzbier? /. Raptor Res. 29.45-25. Fahmy, T. 1998. XLSTAT-PRO statistical software. Paris, France. Falconer, D.S. 1967. Introduction to quantitative genet- ics. Oliver and Boyd, London, England. Fefelov, I.V. 2001. Comparative breeding ecology and hybridization of eastern and western Marsh Harriers Circus spilonotus and C. aeruginosus in the Baikal region of eastern Siberia. Ibis 143:587-592. Forseman, R. 1999. The raptors of Europe and the Mid- dle East: a handbook for field identification. T. and A.D. Poyser, Ltd., London, England. Fox, N.C. 1977. The biology of the New Zealand Falcon (Falco novaezeelandiae; Gmelin 1788). Ph.D. disserta- tion, University of Canterbury, New Zealand. , C.P. Eastham, and HJ. MacDonald. 1997. The ERWDA handbook of falcon protocols. Environmen- tal Research and Wildlife Agency, Abu Dhabi, United Arab Emirates. and S. Sherrod. 1999a. The use of exotic and hybrid raptors in falconry. Information arising from the International Committee on Hybrids. Internation- al Wildlife Consultants Ltd., Carmarthen, Wales, Eng- land. AND S. Sherrod. 1999b. Cited in “Scale produc- tion and use of hybrids in falconry.” Falco 13, Middle East Falcon Research Group, International Wildlife Consultants Ltd., Carmarthen, Wales, England. Gantlett, S. and R. Millington. 1992. Identification of large falcons. Birding World 5:101-106. Gill, F.B. 1998. Hybridization in birds. Auk 115:281—283. Grant, P.R. and B.R. Grant. 1992. Hybridization of bird species. SAmce 256:193-197. Hamer, T.E., E.D. Forsman, A.D. Fuchs, and M.L. Wal- ters. 1994. Hybridization between Barred and Spot- ted owls. Auk 111:487-492. Heidenreich, M. 1997. Birds of prey: medicine and man- agement. Blackwell Scientific Ltd., Oxford, England. AND H. Kuspert. 1992. Metodos de la reproduc- cion en cautividad de las diferentes especies de hal- cones. El problema de los hibridos y la cooperacion entre criadores y autoridades. Proc. Congr. Nac. Cria. Aves Cetreria 1:15—17. Hoeller, T. AND P. Wegner. 2001. Der Wanderfalke ist in Gefahr! Die zweite Hybridfalken-Wildbrut in December 2005 Morphometrics of Falcon Hybrids 393 Deutschland zeigt die Brisanz der Hybridfalkenzucht. Greifvogel u. Falknerei 2000:64—66, Kemp, A.C. 1987. Taxonomy and systematics. Pages 251- 259 in B.A. Giron Pendleton, B.A. Millsap, K.W. Cline, and D.M. Bird [Eds.], Raptor management tech- niques manual. National Wildlife Federation Scientif- ic and Technical Series No. 10. Kleinstauber, G. and H.J. Seeber. 2000. Die erfolgreiche Brut eines Gerfalke-X-Wanderfalke-Hybriden {Falco rusticolus X Falco peregrinus) in freier Wildbahn-Re- port, MaBnahmen, SchluBfolgerungen. Poplationso- kol. Greifvogel- u. Eulenarten 4:323-332 Lindberg, P. 2000. Hybridfalk hackar igen. Var Fagelvdrld 8:28-29. Mayr, E, 1963. Animal species and evolution. Belknap Press of Harvard University Press, Cambridge, MA U.S.A. AND L.L. Short. 1970. Species taxa of North American birds. No. 9 Publ. Nuttal Ornithol. Club. Mittwoch, U. 1977. Genetics of sex differentiation. Ac- ademic Press, Burlington, MA U.S.A. Morris, J. 1972. Peregrine/ Saker breeding. Cap. Breed. Diurn. Birds of Prey 1:14-15. AND R. Stevens. 1971. Successful cross-breeding of Peregrine tiercel and a Saker falcon. Cap. Breed. Diurn. Birds of Prey, 1:5-7. Pennycuick, C.J. 1989. Bird flight performance: a prac- tical calculation manual. Oxford University Press, Ox- ford, England. Pfander, P.V. 1987. Once again concerning the Altay Gyr- falcon. Selevinia 2:38-42. Rosenkranz, D. 1995. Cited in M. Heidenreich. 1997. Birds of prey: medicine and management. Blackwell Scientific Ltd., Oxford, England. Sibley, C. 2000. The North American bird guide. Chris- topher Helm, London, England. Weaver, J.D. and T.J. Cade. 1991. Falcon propagation. The Peregrine Fund, Inc., Boise, ID U.S.A. White, C.M. and L.F. Kiff. 1998. Language use and mis- applied, selective “science;” their roles in swaying public opinion and policy as shown with two North American raptors. Pages 547—560 in R.D. Chancellor, B.-U. Meyburg, andJ.J. Ferrero [Eds.], Holarc tic birds of prey. ADENEX-WWGBP, Badajoz, Spain. Wiley, E.O. 1981. Phylogenetics: the theory and practice of phylogenetic systems. Wiley/Interscience, Hobo- ken, NJ U.S.A. Wyllie, I. AND I. Newton. 1994. Latitudinal variation m the body-size of Sparrowhawks {Accipiter nisus) within Britain. Ibis 136:434-440. Yosef, R., A.J. Helbig, and W.S. Clark. 2001. An intra- generic Accipiter hybrid from Eilat, Israel. Sandgrouse 23:141-144. Received 13 April 2004; accepted 24 March 2005 Former Associate Editor: Juan Jose Negro /. Raptor Res. 39(4):394-403 © 2005 The Raptor Research Foundation, Inc. A CHANGE IN FORAGING SUCCESS AND COOPERATIVE HUNTING BY A BREEDING PAIR OF PEREGRINE FALCONS AND THEIR FLEDGLINGS Dick Dekker^ 3819-112A Street NW, Edmonton, Alberta, T6J 1K4 Canada Robert Taylor P.O. Box 3105, Spruce Grove, Alberta, T7X 3A1 Canada Abstract. — The foraging habits of one pair of Peregrine Falcons (Falco peregrinus) nesting on a power plant in central Alberta were studied over seven consecutive breeding seasons (1998-2004) . We observed 386 attacks that resulted in II7 captures of prey. The success rate increased from 21.9% in the first year to 39.1% in the seventh year and averaged 30.3%. The majority of hunts (76.7%) were initiated from soaring, and the peregrines commonly used the hot air above the plant’s smoke stacks to gain height. The success rates of hunts launched from soaring versus perches were not significantly different (28.7% versus 35.6%). Tandem hunts by the pair {N = 100) were more successful than solo hunts (39.0% versus 27.3%), but the difference was not significant. The main prey species were Franklin’s Gull {Larus pipix- can) and small passerines, which made up 53.0% and 27.4% of kills, respectively. There was a significant difference in the respective capture rates of these prey types (42.5% versus 24.1%). The success rates of the male and the female peregrines were not significantly different, but there was a notable difference in the prey taxa taken by each gender. The male captured 84.4% of the passerines, but only 22.9% of the gulls. After the first year, there was a significant switch from passerines to gulls, which paralleled a significant change in gender participation in foraging. All gulls captured by the male were surrendered to the female or to the fledged Juveniles. In 23.3% of hunts, one or both parents were accompanied by one or more fledglings. The male participated in 76.7% of all parent-fledgling hunts {N = 90), the female in 10.0%, and the remainder by both parents. Aerial prey transfers from adults to flying young were feet-to-feet or through aerial drops. The hypothesis that parent peregrines make live-drops of prey to their fledged young to teach them how to hunt is confounded by observations that live-drops of just- caught prey are also made by the adult male to his mate. However, the hypothesis that peregrines assist their young in capturing prey is supported by anecdotal evidence. Keywords: Peregrine Falcon-, Falco peregrinus; tandem hunting, fledgling hunting, adult hunting cooperative hunting. CAMBIOS EN EL EXITO DE FORRAJEO Y CAZA COOPERATIVA EN UNA PAREJA DE FALCO PER- EGRINUS Y SUS VOLANTONES Resumen. — Se estudiaron los habitos de forrajeo de una pareja de Falco peregrinus que anido en una planta de energfa en el centro de Alberta a lo largo de siete epocas reproductivas consecutivas (1998 a 2004). Observamos 386 ataques, los cuales resultaron en 117 capturas de presas. La tasa de exito se incremento del 21.9% en el primer ano al 39.1% en el septimo ano, y en promedio fue del 30.3%. La mayoria de los eventos de caza (76.7%) fueron iniciados a partir de vuelos elevados; los halcones emplearon frecuentemente el aire caliente que se encontraba encima de las columnas de humo de la planta para alcanzar mayores alturas. La tasa de exito de los ataques iniciados en vuelo no fue signifi- cativamente diferente de la de los ataques iniciados desde perchas (28.7% versus 35.6%). Los eventos de caza en los que ambos miembros de la pareja participaron en tandem {N = 100) fueron mas exitosos que aquellos en los que participo un solo individuo (39.0% versus 27.3%), pero la diferencia no fue significativa. Las presas predominantes fueron la gaviota Larus pipixcan y varias aves paserinas pequenas, representando el 53.0% y el 27.4% de las capturas, respectivamente. Existio una diferencia significativa en las tasas de captura de los distintos tipos de presa (42.5% para L. pipixcan y 24.1% para las aves ^ Email address: tj_dick_dekker@hotmail.com 394 December 2005 Foraging Success of Peregrine Falcons 395 paserinas) . Las tasas de exito no fueron diferentes entre el macho y la hembra, pero existio una notable diferencia en los taxa capturados por cada individuo. El macho capture el 84.4% de las aves paserinas, pero solo el 22.9% de las gaviotas. Despues del primer ano, existio un cambio significativo de paserinas a gaviotas de forma paralela con un cambio significativo en la participacion de los miembros de la pareja en el forrajeo. Todas las gaviotas capturadas por el macho fueron entregadas a la hembra o a los volantones. En el 23.3% de los eventos de caza, uno o los dos padres estuvieron acompahados por uno o mas volantones. El macho participo en el 76.7% de todas las cacerias en las que participaron volantones {N = 90), la hembra en el 10.0% de estas, y ambos individuos en el porcentaje restante. Los adultos entregaron las presas a los juveniles en vuelo directamente de garras a garras, o dejandolas caer en el aire. La hipotesis de que los parentales dejan caer presas vivas para ensenarles como cazar a sus volantones es dificil de apoyar ya que el macho adulto tambien dejo caer presas recien capturadas en vuelo para que su pareja las tomara. Sin embargo, la hipotesis de que los peregrinos le ayudan a sus volantones a capturar presas es apoyada por informacion anecdotica. [Traduccion del equipo editorial] Cooperative or tandem hunting by mated pairs of Peregrine Falcons {Falco peregrinus), in which both partners attack the same prey simultaneously, has been reported from many locations across the species’ worldwide range (e.g., Cade 1960, Bird and Aubry 1982, Thiollay 1988, Frank 1994, Tre- leaven 1998, Jenkins 2000). The success rate of tan- dem hunts has been reported to be higher than hunts by single peregrines (Thiollay 1988), but whether or not tandem hunting by mated pairs ac- tually involves a degree of coordination between the two individuals is unclear. On their breeding range, Peregrine Falcons are also known to hunt together with their fledglings, although data are limited (Brown and Amadon 1968, Palmer 1988, White et al. 2002). The notion that juvenile pere- grines need to be taught hunting skills by their par- ents is contradicted by the fact that captive-reared peregrines, deprived of parental instruction, begin pursuing and capturing prey at about the same age as young falcons at natural nest sites (Cade and Burnham 2003). Apparently, the peregrine does not need to be taught how to chase and kill prey (Cade 1982, Sherrod 1983). However, Newton (1979) and Ratcliffe (1993) state that more critical observation is needed during the period in which fledglings achieve independence. This paper presents empirical data on the for- aging behavior of one mated pair of peregrines hunting solo or in tandem over seven consecutive breeding seasons. Additionally, we present anec- dotal observations on parent-fledgling interac- tions. Study Area and Methods The study area is in central Alberta, Canada, at 53°N. In this largely agricultural region, peregrines nested com- monly on cliffs and cutbanks along rivers and creeks until they became extirpated (Dekker 1967, Fyfe 1976). A gov- ernment program of releasing captive-reared peregrines in central Alberta began in 1975 and led to the establish- ment of breeding pairs on city buildings by 1981 (Hol- royd and Banasch 1990). By the mid 1990s, peregrines began using nest boxes on high industrial structures in rural regions. The breeding site selected for this study is a power plant on the north shore of Wabamun Lake, which is roughly 8 X 20 km in size. The nest box was built by the Alberta Falconry Association in 1993 and put in place by TransAlta Utilities (Wabamun, Alberta) on a catwalk just below the top of a 91 m smoke stack. Lower down, the flat roof of the main building functions as a landing pad for newly-fledged young. Power line pylons in the vicinity provide high perches and plucking posts. Settling ponds and bare ground adjacent to the plant are used as loafing areas by the fledglings and by the adults that bring down large prey such as gulls and ducks. The landscape around the Wabamun plant is mainly wooded with small marshes and an extensive area (5-10 km^) of excavated and reclaimed terrain to the north. The lake receives light recreational use, and some shoreline sec- tions are developed with cottages and small marinas. The nest box is in plain view from a public road that runs between the lake and the plant. Depending on weather conditions, we watched from different vantage points 0.1-2 km away, and we used 8X wide-angle bin- oculars and 20— 60X telescopes. Observation periods, last- ing 3-6 hr each, usually between mid-afternoon and sun- down, were spaced arbitrarily and increased during the period when the fledglings were on the wing. Over 7 yr, 1998—2004, the number of observation days was 196; 15 in June, 58 in July, 101 in August, 21 in September, and 1 in October. The only two days of observation in 1997 were added to the 1998 total. The median date when the first juvenile (always a male) fledged was 16 July (13-25 July). The number of young fledged was three or four per season with a mean of 3.6 (N = 11 males and 14 females). As indicated by their leg bands and plumage characteristics, the breeding pair consisted of the same individuals for the duration of the study, and we classified the female as a 2-yr old falcon in 1998 based on her brown dorsal color in 1997. The subspecific origin of these falcons is F. p. anatum and both originated from captive-reared stock nesting in the cities of Edmonton 396 Dekker and Taylor VoL. 39, No. 4 and Calgary (Gordon Court, Alberta Environment, pers. comm.). Prey species upon which peregrines are known to feed, such as waterbirds and small passerines, are common in the study area. With a mass of 220-335 g (Dunning 1984), the Franklin’s Gull {Larus pipixcan) is close to the mass limit that male peregrines can or are willing to carry over long distances. Although Franklin’s Gulls are locally scarce in early summer, large migrating flocks begin ar- riving in mid-July. Rock Pigeons (Columba livid), which are the dominant prey for peregrines in human altered environments (Ratcliffe 1993), were resident at the plant site but only in small numbers (<20). The terms hunt and attack are used interchangeably and represent one attempt at capturing prey including one or more stoops or passes at the same target of which the outcome was known. Tandem hunts were simulta- neous attacks by both adults on a flock of prey or an individual prey. Group hunts by a combination of parents and juveniles were tallied as adult hunts, and their kills were considered to have been made by the adults, al- though in a few cases it was the juvenile which actually seized (or was allowed to seize) the prey. Hunts by juve- niles alone were not tallied. Details of hunts and kills were entered into field diaries and annotated tabulations. Observer bias in comparing results between years, we believe, was not a factor as D. Dekker recorded observations over the 7-yr study period. R Taylor was associate observer the last 3 yr of the proj- ect. Data sets were compared for significance by chi- square Test of Independence with a Williams’ correction as described in Sokal and Rohlf (1981). Results Hunting Methods and Prey Species. We ob- served 386 hunts by the adult peregrines. Nearly all (99.5%) were directed at airborne prey and consisted of two primary methods: attacks launched from soaring flight (76.7%) or from a perch (23.3%; Table 1). The falcons typically be- gan a soaring sequence by circling up over the plant and using the hot air of the three stacks to gain height rapidly. Drifting downwind, the soaring falcons reached altitudes estimated to exceed 1000 m. By flapping their wings or gliding, they cruised upwind. Attacks on prey flying lower than the fal- con were made by deep stoops with wings partly or fully flexed. Some of these attacks began with a burst of wing flaps and ended in a stoop, which could either be near-vertical or oblique. If the prey evaded the initial stoop, the falcon might follow up with one or more additional stoops or passes. Still-hunting attacks were launched from high perches such as the catwalk railing near the top of the 91 m stacks. With rapid wing flaps, these fal- cons headed for targets some distance away (>100 m) . After unsuccessful or aborted attacks, the per- egrines commonly returned to the plant, either to perch or to regain altitude by soaring. Some hunt- ing sequences lasted 3-4 hr before a prey was cap- tured. On other days, we observed falcons catch three prey in less than 1 hr. The success rate of all hunts {N — 386) by the adult peregrines, either attacking prey solo or in tandem, was 30.3% (Table 1). Perch hunts were not significantly (G = 0.77, P = 0.379) more suc- cessful than soar hunts (35.5% and 28.7%, respec- tively). The success rate of the female was 37.1%, not significantly (G = 2.37, P — 0.124) different from that of the male (24.1%), but there was a notable difference in the taxa of the prey taken by the two sexes (Table 2). The male caught 84.4% of 32 small passerines, but only 22.9% of 62 gulls. Nearly all gulls seized by the male were released to his mate or a fledgling <1 sec after capture. If nei- ther the female nor any of the fledglings were nearby, the male, upon seizing a gull, brought it down steeply and left it on open ground or on the roof of the plant ( A = 5) . The female carried gulls with apparent ease over distances exceeding 1 km. The majority (84.9%) of gull kills observed at close range or found as prey remains (A = 73) were juveniles. Juvenile Franklin’s Gulls were often seized in mid air during the falcon’s first stoop. By contrast, adult gulls typically evaded a peregrine by rising steeply. Some evaded two or more additional passes and were eventually left alone. Others de- scended and plunged into water. In seven instanc- es, the female peregrine repeatedly swooped at the swimming gull. Three were retrieved from the wa- ter and carried off. Some white birds, assumed to be gulls, that evaded 20 or more diving attacks far out over the lake may have been Common Terns {Sterna hirundo). Two terns were carried to the plant, but their capture had not been observed. Ring-billed Gulls {Larus delawarensis) were some- times forced down, but we found no evidence that any were killed. Rock Pigeons were seldom attacked. On two oc- casions, the female stooped unsuccessfully at free- flying flocks of pigeons. Both adults made oppor- tunistic passes at pigeons that flushed from plant ledges below them, but the only successful pigeon hunts (A = 2) were initiated by fledglings. In at least one of these hunts, the capture was made by the adult female, who joined the attack. During the sixth year of study, we gained the im- pression that the falcons had become more suc- cessful — in particular, at capturing gulls — than dur- ing the preceding years. This impression was December 2005 Foraging Success of Peregrine Falcons 397 •D 10 1998 1999 2000 2001 2002 2003 2004 Years Figure 1. Hunting success per year of the same pair of Peregrine Falcons breeding in central Alberta, 1998— 2004. The respective number of hunts/kills for the 7 yr of study were as follows: 108/23, 47/11, 25/8, 37/12, 44/ 16, 61/22, 64/25. strengthened in year 7 and confirmed by subse- quent data analysis. The overall success rate of the adults increased from 21.3% in the first year of study to 39.1% in the seventh year of study (Fig. 1 ). Tandem Hunts. Each year from mid June on- wards, both parents often began soaring together. The male gained height quicker and always circled higher than the female. Stooping alternately, they attacked 68 prey from soaring flights (Table 1). If no prey had been sighted for some time, the pair separated or headed off into the distance, one fol- lowing the other, either in soaring or flapping flight. Tandem attacks were also launched from high perches on the plant. In tandem attacks, both falcons could take the lead, and the male flew higher than the female. While the female ap- proached the prey directly, the male typically at- tacked from above in a near vertical stoop. A high percentage (60-75%) of tandem hunts were lost from view before we could either see the target or the result. Tandem hunts of which the outcome was known (N = 100) represented 25.9% of all hunts, and their success rate was higher than in 286 solo hunts (39.0% versus 27.3%), but the dif- ference was not significant (G — 2.39, P = 0.121). The success rate of 32 tandem attacks launched from a perch was 50.0%, compared to 33.8% for tandem attacks from soaring position. Tandem hunts resulted in 39 kills (Table 1). Of these, 26 captures of prey were observed in detail: seven were seized by the female, 19 by the male. Prey caught by the male in tandem attacks were surrendered at once to the female, either by feet- to-feet transfer (N = 5) or released in the air (N = 14). Thirteen of these aerial drops were gulls, which resumed flight upon release. Nine were sub- sequently seized by the female; four evaded her passes and were let go or escaped by splashing down into water. Hunts by Groups of Adults and Fledglings. In 23.3% of hunts (N = 386), one or both parents were accompanied, and often harassed, by one or more juveniles (Table 3) . Some fledglings closely followed the adults prior to the start of a hunt, others approached hurriedly from a distance to join a hunt in progress. A significant majority of group hunts (G = 37.96, P = 0.001) were led by the adult male, which made 71.0% of the 31 kills (Table 3). Prey caught by the parents were trans- ferred, usually at once, to the approaching juve- niles, either by feet-to-feet transfers (N = 15) or through aerial drops (N = 7) . At least three small passerines that were dropped were still alive and resumed flight. They were recaptured by a juvenile or by the adult and released again. Three live- drops were gulls. In addition, four of 11 gulls caught in adult-juvenile hunts were seized in flight by the juveniles “chaperoned” by adults. After successful group hunts, one or more juve- niles fed on the kill. If the feeding site was in an open area away from the plant, the adult female typically perched on a pole nearby {N = 14). At Table 1. Hunts and kills made by a pair of Peregrine Falcons nesting on a power plant in central Alberta, 1998- 2004. The hunting success rates are presented in parentheses. Hunts/Kills from Soaring Hunts/Kills from Perch Total Hunts Adult male (24.1%) 176/44 (25.0%) 40/8 (20.0%) 216/52 Adult female (37.1%) 52/18 (34.6%) 18/8 (44.4%) 70/26 In tandem hunts (39.0%) 68/23 (33.8%) 32/16 (50.0%) 100/39 Totals (30.3%) 296/85 (28.7%) 90/32 (35.5%) 386/117 398 Dekker and Taylor VoL. 39, No. 4 the approach of people, she vocalized and “swooped” overhead. She also stooped at crows or Buteo hawks, apparently to drive them away. After the fledglings were satiated and left, the adult fe- male sometimes fed on the remains of the kill {N = 5). Discussion Hunting Success. The hunting success of adult peregrines on breeding territory is generally high- er than that of migrating or wintering peregrines (Dekker 1980, Roalkvam 1985). Jenkins (2000) compared data presented in eight publications on breeding peregrines and found an extreme varia- tion (9.3-84.1%) in hunting success, but much less variation in hunting methods. The majority (58- 75%) of attacks summarized in his review were launched from a perch. By contrast, the percent- age of perch hunts was only 23% at Wabamun. The “still-hunting strategy” of perch hunts reduces the energy cost of foraging. However, high-soaring flight is also relatively energy-efficient and widens the radius of the attack zone (Enderson and Craig 1997). Soaring flight has been reported as a com- mon hunting method in many regions, but not to such a high degree as documented in this study. The second highest use was recorded in Africa, where breeding pairs launched 30-42% of hunts from flight (Thiollay 1988, Jenkins 2000). The use of factory exhaust to facilitate soaring flight by the Wabamun peregrines was described in an earlier publication (Dekker 1999), but this phenomenon has to our knowledge not been reported else- where. It is noteworthy that the falcons in this study at- tacked nearly all of their prey in flight, often at great altitudes. This is in sharp contrast to migrat- ing or wintering peregrines, which commonly use surprise methods to attack prey on the ground or in shallow water (Dekker 1980, 2003, Cresswell 1996). However, there is also an element of sur- prise involved if a soaring peregrine stoops from a great height at prey flying far below. In solo hunts, the male frequently made long stoops that levelled out low over woods and were aimed at flocks of small passerines flying just beyond the trees. Sur- prise is probably also a factor in tandem hunts in which the male stoops from high above while the female pursues the prey at a lower altitude. A key factor in the hunting success of peregrines is the vulnerability of individual prey, which is dif- ficult to assess for the human observer. It is well- known that predation risk for land birds increases over water (Herbert and Herbert 1965). Converse- ly, water birds become vulnerable over land (Hunt et al. 1975, Dekker 1980). Mature prey on home territory should be harder to catch for a raptor than juvenile prey passing over unfamiliar terrain. In central Alberta, the migrations of juvenile prey species approximately coincide with the period when fledgling peregrines are on the wing and when the parental task of provisioning them is most demanding. In this study, the peregrines were very successful at capturing juvenile Franklin’s Gulls. For instance, on 7 d between 17 July and 9 August 2003, when thousands of gulls were passing through the area, the adult female, hunting solo, captured each of seven juvenile gulls in her first attack of the afternoon and each requiring only one stoop. While it does not seem surprising that bird-hunting falcons should become better at what they do as they become older and gain in experi- ence, data in support of that notion have, to our knowledge, not been published before. An expla- nation for the year-to-year increase in the success rate of the Wabamun pair is that these falcons be- came specialists on juvenile Franklin’s Gulls, which differ from adults by their dusky color and absence of black on the head. Juvenile targets were proba- bly pre-selected before the falcons began their at- tack. A similar pre-selection hypothesis was ad- vanced for the high success rate (73-93%) of “Red Baron,” an adult male peregrine hunting over coastal marshes in New Jersey (Cade and Burnham 2003:333). In this study, the prey component changed sig- nificantly after the first year (Tables 4 and 5). In the first year of study, gulls made up only 8.7% of kills compared to 63.8% in years 2-7 (G = 10.85, P = 0.001). The proportion of small passerines changed significantly from 60.9% in the first year to 19.1% in years 2-7 (G = 7.01, P — 0.008). Over- all, the capture success on gulls (of both age groups) was significantly higher (G = 5.35, P = 0.021) than on small passerines (42.5% versus 24.1%). Successful taking of juvenile gulls may be even higher than for adults, but an adequate sam- ple was not available for adults. As reported in re- sults, the sample of gull kills {N = 73) included 84.9% juveniles. Coincident with the observed prey switch, we re- corded a major change in foraging participation between the sexes after the first year (Table 6). The male’s solo hunts in the first year were 78.7% December 2005 Foraging Sugcess of Peregrine Falgons 399 Table 2. Prey taxa captured by the Peregrine Falcon pair hunting solo or in tandem. Franklin’s Gulls Small Passerines Smalt. Shorebirds Ducks Other or Unidentified Adult male 14 27 1 1 9 Adult female 21 2 1 2 0 Pair in tandem 27 3 5 0 4 Totals 62 32 7 3 13 of total hunts (N = 108), significantly greater (G = 8.08, P = 0.004) than the 47.1% of hunts in years 2-7 {N = 278). By contrast, the female’s share during the first year was 2-7%, and this rose significantly to 24.1% in years 2-7 (G = 23.51, P = 0.001). A possible explanation is that the female lacked skill and experience in her first year on ter- ritory. As she became older and gained expertise, she began to play a more active role in foraging. There was no significant change (G = 2.70, P = 0.1) in the proportion of tandem hunts between the first year and years 2-7 (Table 6) . Our findings that the primary prey of the male was passerines, whereas the female’s main prey was gulls (Table 2), lends support to the hypothesis that the reversed sexual size dimorphism in raptors such as seen in the peregrine allows them to ex- ploit a wider prey base and reduces competition between the sexes (Selander 1966). Cooperative Hunting. Are peregrines that attack the same prey simultaneously with their mates or fledglings actually cooperating? Or, are the individ- uals simply reacting at the same time to the stim- ulus of sighting prey? True cooperative hunting differs fundamentally from other forms of group predation, such as pseudo-cooperative hunting (El- lis et al. 1993). In true cooperative hunting, the group consists of at least two members that are a stable social unit, and their cooperation should benefit the group rather than just the individual. An example of true cooperative foraging is de- scribed by Bednarz (1988) for the Harris’s Hawk {Parabuteo unicinctus). Ellis et al. (1993) conclude that cooperative hunting is the most efficient strat- egy for capturing prey in many situations and that each form of social foraging should have evolved as an adaptive advantage enhancing the fitness of all individuals in the group. In the Aplomado Fal- con (Falco femoralis), pair hunting is more than twice as productive as solo hunting when the prey are birds (Hector 1986). Similarly, the success rate of pair hunting in peregrines may depend on the species of prey hunted (Jenkins 2000). In this study, tandem hunts by the mated pair of peregrines were more successful than solo hunts (39.0 versus 27.3%), although the difference was not statistically significant. Moreover, the individu- al success rate of the two partners in tandem hunts is actually halved. So what, if any, is the fitness value of tandem hunting for this pair of breeding pere- grines? It may lie in the fact that any action by the male that benefits his mate or their progeny should be of adaptive advantage. At Wabamun, the adult male was not seen to eat the gulls he killed, although on two occasions he fed on gull remains abandoned by fledglings. The fact that he mainly hunted such prey in the com- pany of his mate or fledglings suggests that his role was a cooperative one. This view is supported by anecdotes such as the following. By mid September 1999, after the fledglings had dispersed, the adults were still hunting together. One evening, shortly after both had soared up over the plant, the male caught a small passerine and landed on a pylon to Table 3. Hunts/kills made by one or both Peregrine Falcon parents accompanied by one or more fledglingjuveniles. 1 Juvenile 2 Juveniles 3 Juveniles Totals Adult Male 29/10 34/10 6/2 69/22 Adult female 3/2 6/4 0/0 9/6 Both adults 6/3 5/0 1/0 12/3 Totals 38/15 45/14 7/2 90/31 400 Dekker and Taylor VoL. 39, No. 4 Table 4. Prey captured by a pair of Peregrine Falcons as a percentage of total prey per year. Year Percent Gulls Percent Passerines Percent Other No. Total Prey No. Hunts 1 8.7 60.9 30.4 23 108 2 63.6 27.3 9.1 11 47 3 62.5 12.5 25.0 8 25 4 58.3 16.7 25.0 12 37 5 43.7 31.3 25.0 16 44 6 63.6 27.3 9.1 22 61 7 80.0 4.0 16.0 25 64 Totals 53.0 27.4 19.6 117 386 consume his prey. Instead of interfering, the fe- male perched on the next pylon and waited for her mate to finish his meal. Presently, both soared up again and eventually flew out high over the lake, stooping in tandem at gulls. Like many similar ob- servations, this incident suggests that the male’s role was truly a cooperative one — to assist his mate in her foraging. The question of whether attacks on the same prey by a combination of adults and fledglings can be considered cooperative seems even less clear, for it is apparent that the juvenile is primarily in- tent on kleptoparasitizing the adult. If there is co- operation, it is one-sided and benefits only the ju- veniles (i.e., this may be considered as parental investment). Nevertheless, the evolutionary value of adult/ fledgling combinations is undeniable, as a well-fed progeny enhances the future of the fam- ily genes. In support of the hypothesis that the par- ticipation of adults in joined hunts with their fledg- lings indeed represents parental care, we present a number of anecdotal observations in the appen- dix. Live-drops of Prey. Cade (1982) and Sherrod (1983) characterize the behavior of paired male peregrines in social interactions with their bigger mate as submissive and even fearful. Females take food from males in an aggressive manner, not only Table 5. Comparison of main prey taxa captured by a pair of Peregrine Falcons in study years 1 and 2-7. Prey Taxa No. Kills Percent of Totals Kills N Year 1 Gulls 2 8.7 23 Years 2-7 60 63.8 94 Year 1 Passerines 14 60.9 23 Years 2-7 18 19.1 94 during the breeding season but also on migration or at the wintering grounds. Unmated females rou- tinely force males to drop just-caught prey (Dekker 1980, 2003). Tandem hunting by unmated pere- grines is relatively common and driven by compe- tition rather than cooperation. Two or more (up to six) migrating or wintering peregrines were seen in joint pursuit of the same prey in Alberta and British Columbia. Male peregrines were forced to make live-drops of prey not only to female conspe- cifics, but also to other kleptoparasitic raptors, such as Prairie Falcons {Falco mexicanus) and Bald Eagles {Haliaeetus leucocephalus; Dekker 1995, 1998). I concur with Sherrod (1983) that the release of just-caught prey by the male peregrine at the ap- proach of an aggressive female may be triggered by the impulse to avoid close contact. The impor- tance of a timely release was demonstrated on 21 August 2003, when the Wabamun male, carrying a just-caught prey and flying 20-25 m high, was met by a screaming juvenile female in typical begging flight. Apparently, the male was just too late in re- leasing his prey, for he was seized by the juvenile. Locked by their feet, both birds tumbled about 15 Table 6. Number of hunts by the adult male and fe- male, either hunting solo or in tandem, in year 1 and years 2-7. No. Hunts Percent OF Total Year 1 Adult male 85 78.7 N = 108 hunts Adult female 3 2.7 Tandem 20 18.5 Years 2-7 Adult male 131 47.1 N = 278 hunts Adult female 67 24.1 Tandem 80 28.8 December 2005 Foraging Success of Peregrine Falcons 401 m before separating again, while the prey item fell into the bushes below and was lost. Based on a thorough review of the literature and on his own studies, Sherrod (1983) observed that the behaviors of parents and fledglings comple- ment each other. While the youngster only wants food, the adult appears to be very willing to supply that food. Some live-drops are no doubt the result of the aggressive approach of the begging falcon. However, as argued by Newton (1979), other live- drops are clearly intentional as parent peregrines will recapture the live-dropped prey and release it again if the fledgling fails to catch it the first time. Such repeat drops were seen in this and other stud- ies (Frank 1994, Treleaven 1998). Conclusion Some of the aerial drops of just-caught and still live prey from adult to fledgling (described in the appendix) lend support to the hypothesis that the parents were teaching their young by example. Al- though young falcons do not need to be taught to hunt, such extended parental care might well give them a certain survival advantage after leaving the nest site. The observation that live-drops were also made from adult to adult, both at Wabamun and at a natural nest site in northern Alberta (Dekker 1999), may seem to refute the above hypothesis. Because there can be no doubt that adult females are accomplished hunters, the male’s reason for making live-drops to his mate can have nothing to do with teaching. However, here a secondary and complementary factor comes into play. Whether hunting with his mate or a fledgling, the male plays a dual role, either in support of his mate or prog- eny. It is difficult, if not impossible, to translate all of the anecdotes detailed in the appendix into hard data in support of the teaching hypothesis. As to the alternative, but not exclusive, hypothesis that parent peregrines assist their fledglings in captur- ing their first prey, on the evidence presented here, we are convinced that the answer is affirma- tive. ACKNOWI.EDGMENTS I thank T.J. Cade for his review of an earlier draft of this paper. Referees G. Malar and I. Newton made very helpful comments on the second version. Throughout, the input ofj. Bednarz was particularly valuable. W. Nel- son provided relevant information on nest box installa- tion and the weight of Franklin’s Gulls. G. Court did the statistical tests. R. Dekker drafted the figure. Occasional co-observers were G. Court, I. Dekker, K. Kunst, and J Morrison. I thank TransAlta Utilities for access to obser- vation points out of bounds to the general public. All expenses incurred during this study were privately fund- ed by the authors except during the last year, when D Dekker received travel grants from TransAlta Utilities and the Alberta Conservation Association. Literature Cited Bednarz, J.C. 1988. Cooperative hunting in Harris’s Hawks {Parabuteo unicinctus) . Science 2^9:1525— 1527 . Bird, D.M. and Y. Aubry. 1982. Reproductive and hunt- ing behavior in Peregrine Falcons in southern Que- bec. Can. Field-Nat. 96:167-171. Brown, L. and D. Amadon. 1968. Eagles, hawks and fal- cons of the world. Country Life Books. Hamlyn House, Feltham, England. Cade, TJ. 1960. Ecology of the peregrine and Gyrfalcon populations in Alaska. Univ. Calif. Publ. Zool. 63:1561- 290. . 1982. The falcons of the world. Cornell Univer- sity Press, Ithaca, NY U.S.A. and W. Burnham. 2003. Return of the peregrine. The Peregrine Eund, Boise, ID U.S.A. Cresswell, W. 1996. Surprise as a winter hunting strategy in Sparrowhawks, peregrines, and Merlins. Ibis 138. 684-692. Dekker, D, 1967. Disappearance of the Peregrine Falcon as a breeding bird in a river valley in Alberta. Blue fay 30:175-176. . 1980. Hunting success rates, foraging habits and prey selection of Peregrine Falcons migrating through central Alberta. Can. Field-Nat. 94:371-382. . 1995. Prey capture by Peregrine Falcons winter- ing on southern Vancouver Island, British Columbia J. Raptor Res. 29:26-29. . 1998. Over-ocean flocking by Dunlins and the effect of raptor predation at Boundary Bay, British Columbia. Can. Field-Nat. 112:694—697. . 1999. Bolt from the blue. Wild peregrines on the hunt. Hancock House Publishers. Surrey, BC Canada, Blaine, WA U.S.A. . 2003. Peregrine Falcon predation on Dunlins and ducks and kleptoparasitic interference from Bald Eagles wintering at Boundary Bay, British Columbia. J. Raptor Res. 37:91-97. Dunning, J.B. 1984. Body weights of 686 species of North American birds. Western Bird Banding Assoc. Mo- nogr. No 1. Ellis, D.H., J.C. Bednarz, D.G. Smith, and S.P. Flem- ming. 1993. Social foraging classes in raptorial birds BioScience 43:14—20. Enderson, J.H. and G.R. Craig. 1997. Wide-ranging by nesting Peregrine Falcons determined by radio-telem- etry. /. Raptor Res. 31:333-338. Frank, S. 1994. City peregrines. A ten-year saga of New York city falcons. Hancock House Publishers. Surrey, BC Canada, Blaine, WA U.S.A. 402 Dekker and Taylor VoL. 39, No. 4 Fyfe, R.W. 1976. Rationale and success of the Canadian Wildlife Service peregrine breeding project. Can. Field-Nat. 90:308-319. Hector, D.P. 1986. Cooperative hunting and its relation- ship to foraging success and prey size in an avian predator. Ethology 73:247-257. Herbert, R.A. and K. Herbert. 1965. Behavior of Pere- grine Falcons in the New York City region. Auk 82: 62-94. Holroyd, G.L. and U. Banasch, 1990. The reintroduc- tion of the Peregrine Falcon into southern Canada. Can. Field-Nat. 104:203-208. Hunt, W.G., R.R. Rogers, and D.J. Slowe. 1975. Migra- tions and foraging habits of Peregrine Falcons on the Texas coast. Can. Field-Nat. 89:11—123. Jenkins, A.R. 2000. Hunting mode and success of African peregrines: does nesting habitat quality affect forag- ing efficiency? Ibis 142:235-246. Newton, I. 1979. Population ecology of raptors. Buteo Books, Vermillion, SD U.S.A. Palmer, R.S. 1988. Handbook of North American birds. Vol. 5. Diurnal raptors, part 2. Yale University Press, New Haven, CT U.S.A. Ratclifee, D. 1993. The Peregrine Falcon. Poyser Pub- lishing. London, England. Roalkvam, R. 1985. How effective are hunting pere- grines? Raptor Res. 19:27-29. Selander, R.K. 1966. Sexual dimorphism and differential niche utilization in birds. Coracior 68:113-151. Sherrod, S. 1983. Post-fledgling behavior of the Pere- grine Falcon. The Peregrine Fund, Ithaca, NY U.S.A. SoKAL, R.R. and F.J. Rohlf. 1981. Biometry: the princi- ples and practice of statistics in biological research. Freeman and Company, San Francisco, CA U.S.A. Thioliay, J.-M. 1988. Prey availability limiting an island population of Peregrine Falcons in Tunisia. Pages 701-710 m TJ. Cade, J.H. Enderson, C.G Thelander, and C.M. White [Eds.], Peregrine Falcon populations: their management and recovery. The Peregrine Fund, Boise, ID U.S.A. Treleaven, R.B. 1998. In pursuit of the peregrine. Tier- cel Publishing, Wheathampstead Herts, England. White, C.M, NJ. Clum, TJ. Cade, and W.G. Hunt. 2002. Peregrine Falcon {Falco peregrinus). In The birds of North America, No. 660. A. Poole and F. Gill [Eds.], The Birds of North America, Inc., Philadelphia, PA U.S.A. Received 18 September 2003; accepted 15 July 2005 Appendix. Observations of parental care by a pair of per- egrines toward their fledglings and nearly-fledged juve- niles. (1) On 16 August 1999, having stooped at a gull and missed his target, the adult male maneuvered under it in a peculiarly slow flying style. He appeared to drive the gull upward and prevent it from dropping into the water below. Seconds later, a juvenile female stooped from above, just missing the gull. The adult male then stooped again, just short of the target. Chased by both peregrines, the gull eventually escaped. In this case, as in others, it seemed clear that the male did not want to catch the gull himself. Instead, he attempted to set it up for the ap- proaching juvenile. (2) On two occasions, when a gull was chased by a group of one adult and one fledgling, the adult hit or touched the prey, causing it to cartwheel in the air. In the next instant, the gull was seized by the closely-follow- ing juvenile. Our impression was that the adults were not intent on taking the gull themselves. (3) On 12 August 2000, the adult male, while soaring in tandem with his mate, suddenly sprinted away (400— 500 m) to hit a shorebird that was being chased by two juveniles at an altitude estimated at 75 m. Dead or stunned, the prey fell into reeds and was lost before the juveniles could recover it. This was one of only four in- stances when these peregrines hit their prey in the air, as opposed to the common method of binding to the prey and carrying it down. The remarkable point is that the adult male did not stoop and recover the plunging shorebird himself, which would have been a normal pro- cedure if he had been hunting alone. (4) A Rock Pigeon pursued by a fledgling was taken by the adult female, who transferred it at once to the juvenile. On their own, the adults seldom pursued pi- geons. (5) Fledglings that have been on the wing for only a few days have difficulty carrying a gull. After making an aerial transfer to an unskilled juvenile, the adults will ac- company it in flight and retrieve the gull if dropped. By the same token, seemingly aware of the lack of ability in a badgering female youngster, the adult female refused to transfer a just-caught gull and instead carried it ca. 500 m to the roof of the plant, where she at once sur- rendered it to the fledgling. (6) The adult male demonstrated an awareness of the different needs of the genders. In group hunts with male offspring, he selected mostly passerines or small shore- birds (41 of 45). By contrast, he usually attacked gulls in the company of females. After capturing one or more gulls for his mate or the fledglings, he hunted smaller prey for himself. The above interactions between parents and fledglings parallel numerous anecdotes described by long-term per- egrine observers in New York and Great Britain (Frank 1994, Treleaven 1998). It is our impression that adult peregrines appear to know how to assist their progeny in gaining indepen- dence. For a start, there is a change in food exchanges when the first of the juveniles are close to fledging. Both at Wabamun and a natural nest site in central Alberta, the adult male appeared to coax a fully-feathered male youngster to fly. Instead of transferring the prey item in the usual direct manner, he cruised back and forth just December 2005 Foraging Success of Peregrine Falcons 403 out of reach, holding the prey in his lowered feet. His behavior cannot be explained as having been prompted by fear of a bigger bird for this incident involved a male juvenile. After the young male failed to fledge, the adult gave the food item to another juvenile, a fully-feathered female sitting on the catwalk. Similarly, on two other oc- casions when he arrived with a small prey, he flew right up to a screaming juvenile male that was perched on the edge of a catwalk, but instead of exchanging the prey feet-to-feet, he dropped it and recovered the falling item in a quick stoop. He repeated this teasing show three or four times. After the fully-feathered juvenile male failed to fly, the adult gave the food item directly to a less-de- veloped female still in the nest box. Another noteworthy exchange involved the female One day, after she had delivered a prey to two juvenile males, which shared the food, the male arrived, also with prey. He gave it directly to one of the same two young- sters. Instantly, the adult female interfered. She took the food item away from the young male and offered it in- stead to a juvenile female which had not been fed for several hours. Similar incidents were seen at Wabamun and at a natural nest site in central Alberta (D. Dekker unpubl. data). J Raptor Res. 39(4):404-416 © 2005 The Raptor Research Foundation, Inc. NESTING ECOLOGY AND BEHAVIOR OF BROAD-WINGED HAWKS IN MOIST KARST FORESTS OF PUERTO RICO Derek W. Hengstenberg and Francisco J. VilellaI USGS Biological Resources Division, Cooperative Research Units, MS 9691, Department of Wildlife and Fisheries, Mississippi State University, Mississippi State, MS 39762 US. A. Abstract. The Puerto Rican Broad-winged Hawk {Buteo platypterus brunnescens) is an endemic and endangered subspecies inhabiting upland montane forests of Puerto Rico. The reproductive ecology, behavior, and nesting habitat of the Broad-winged Hawk were studied in Rio Ab^o Forest, Puerto Rico, from 2001-02. We observed 158 courtship displays by Broad-winged Hawks. Also, we recorded 25 ter- ritorial interactions between resident Broad-winged Hawks and intruding Red-tailed Hawks {Buteo ja- maicensis jamaicensis) . Broad-winged Hawks displaced intruding Red-tailed Hawks from occupied terri- tories (P = 0.009). Mayfield nest survival was 0.67 across breeding seasons (0.81 in 2001, N = 6; 0.51 in 2002, N = 4), and pairs averaged 1.1 young per nest (years combined). The birds nested in mixed species timber plantations and mature secondary forest. Nests were placed in the upper reaches of large trees emerging from the canopy. Nest tree DBH, understory stem density, and distance to karst cliff wall correctly classified (77.8%) nest sites. Key Words: Broad-wing Hawk; Buteo platypterus brunnescens; endangered; nest success; prey delivery; hab- itat model; karst forest; Puerto Rico. ECOLOGIA REPRODUCTIVA DE BUTEO PiJiTYPTERUS BRUNNESCENS EN BOSQUES DE CALIZA HUMEDOS DE PUERTO RICO Resumen. Buteo platypterus brunnescens es una subespecie de rapaz endemica a los bosques montanos de Puerto Rico. Investigamos la ecologia reproductiva y el comportamiento de B. p. brunnescens en el Bosque de Rio Abajo, Puerto Rico, durante 2001 y 2002. Observamos 158 despliegues de cortejo en Rio Abajo. Observamos 25 encuentros territoriales entre B. p. brunnescens y B. jamaicensis jamaicensis. B. p. brunnescens desplazo a B. j. jamaicensis de sus territorios el 84% de las veces (P = 0.009). La supervi- vencia de los nidos en ambas temporadas fue de 0.67 (0.81 en 2001, N = 6; 0.51 en 2002, N= 4). Los nidos produjeron un promedio de 1.1 volantones por nido (afios combinados). Encontramos nidos en areas de bosque secundario maduro y plantaciones forestales. La altura del dosel, diametro del arbol, densidad del sotobosque y distancia a farallon de mogote clasificaron correctamente (77.8%) los nidos en Rio Abajo. [Traduccion de los autores] The Broad-winged Hawk {Buteo platypterus brun- nescens) is an endemic woodland raptor of upland montane forests of Puerto Rico. This subspecies is listed as endangered (Federal Register 1994) hy the Puerto Rico Department of Natural and Envi- ronmental Resources (DNER) and the U.S. Fish and Wildlife Service (USFWS). The Broad-winged Hawk in Puerto Rico is nonmigratory and exhibits a limited geographic range with known popula- tions restricted to montane forests (Delannoy 1997). Breeding in Puerto Rico begins in late De- cember, with nests placed in the upper reaches, but ^ Corresponding author email address; Mlella@cfr. msstate.edu below the high canopy (Delannoy and Tossas 2002). This insular subspecies is smaller and dark- er than its North American nominate counterpart Buteo platypterus platypterus, but larger than the Lesser Antillean subspecies (Raffaele 1989, Good- rich et al. 1996). The most recent population es- timate for the Broad-winged Hawk in forest re- serves of Puerto Rico is approximately 125 individuals (Delannoy 1997). The nesting biology of the Broad-winged Hawk in North America has been described by a number of authors (e.g., Fitch 1974, Matray 1974, Rosen- held 1984, Titus and Mosher 1987, Crocoll and Parker 1989). However, knowledge on the repro- ductive biology of the insular endemic subspecies 404 December 2005 Puerto Rican Broad-winged Hawk Ecology 405 BWHA nest sites A 2001 • 2002 Forest trails Roads Rivers Lago Dos Bocas Rio Abajo Forest N 900 0 900 Meters Figure 1. Locations of Broad-winged Hawk nest sites during the breeding season of 2001 and 2002 in Rio Abajo Forest, Puerto Rico. of Puerto Rico, and forest raptors of the West In- dies in general, are limited (Goodrich et al. 1996). Similarly, the available information on nesting be- havior and diet of Broad-winged Hawks in Puerto Rico is based on a few observations by Snyder and Kepler (1987). Additional information on the Puerto Rican Broad-winged Hawk’s reproductive ecology and nest habitat use is required to better understand current limiting factors, and provide recommen- dations for future research on habitat conservation in public and private lands. Herein, we report on the nesting ecology and behavior of the Broad- winged Hawk in a forest reserve in the moist lime- stone forest region of Puerto Rico. Specifically, we provide baseline information on courtship behav- ior and territorial defense, nest success and pro- ductivity, prey delivery rates by nesting pairs, and their habitat use. Moreover, we discuss the impli- cations our results on interactions between the Broad-winged Hawk and Red-tailed Hawk {Buteoja- maicensis jamaicensis) may have on future plans to establish a second wild population (by releasing captive-reared individuals) of the critically endan- gered Puerto Rican Parrot {Amazona vittata) in the Rio Abajo Forest (White et al. 2005). Study Area and Methods Study Area. We studied Broad-winged Hawks in the Rio Abajo Forest and surrounding private lands in Puerto Rico from 2000-02 (Fig. 1). The Rio Abajo Forest (18°20'N, 66°42'W) is managed by the Forestry Division of DNER and is in west-central Puerto Rico within the moist limestone region of the island (Ewel and Whitmore 1973). This forest reserve comprises an area of 2300 ha with elevations ranging from 200-420 m. We obtained climate data for the study period from the site closest to our study area, the Dos Bocas NOAA weather station (NOAA 2002). Annual precipitation during our study averaged 18.3 cm (range = 6.9-34.9 cm) in 2001 and 14.7 cm (range = 5.2-45.1 cm) in 2002. Mean annual temperature was 25.3°C (range = 19.9— 30.6°C) in 2001 and 25.5°C (range = 20.1-30.9°C) in 2002. The rugged limestone region (i.e., karst) of Puerto 406 Hengstenberg and Vilella VoL. 39, No. 4 Rico encompasses 27.5% of the island’s surface (Lugo et al. 2001). Topography in this region is extreme and char- acterized by karst formations of subterranean drainages, caves, dome shaped hills or “mogotes,” and deep sink- holes. Karst forest contains the largest tree species rich- ness of Puerto Rico (Lugo et al. 2001). Rio Abajo Forest IS fragmented on the eastern end by a double lane high- way and in the south-central part by a small community (Fig. 1) . About 75% of the forest is within the subtropical wet zone, the remaining quarter lies within the subtrop- ical moist zone (Ewel and Whitmore 1973). Previous studies indicated Broad-winged Hawks in Puerto Rico have a limited geographic range, and their abundance is higher in the Karst region compared to other life zones on the island (Delannoy 1997). As our primary objectives were to expand current knowledge on breeding ecology, habitat use, and movement patterns of Broad-winged Hawks (Hengstenberg and Vilella 2004), we selected the Rio Abajo Forest. Moreover, we were in- terested in comparing our results with findings of re- cently completed studies in Rio Abajo Forest (Delannoy and Tossas 2002). Vegetation of Rio Abajo is comprised of a mix of sec- ondary growth forests and timber plantations (Cardona et al. 1987). The midstory of some areas of secondary forest was characterized by abandoned shade-grown cof- fee and citrus plantations. Forest overstories were domi- nated by moca {Andira inermis) , capa prieto ( Cordia allio- dora), and guaraguao {Guarea guidonia). Approximately 6.9% of the forest (171.7 ha) are managed timber plan- tations. Some stands are actively maintained (open un- derstory), while others had a developing understory. Timber plantations, approximately 30-50 yr old, of Hon- duras mahogany {Swietenia macrophylla) , maria {Calophy- llum brasiliense) , teca (Tectona grandis), and mahoe {Hibis- cus elatus) occur in valley bottoms and along mid-slopes (Cardona et al. 1987). As part of additional research efforts, we trapped Broad-winged Hawks in Rio Abajo Forest (Hengstenberg and Vilella 2004). Between January 2001 and July 2002, we trapped eight Broad-winged Hawks in the Rio Abajo Forest. We used octagonal and quonset style bal-chatri traps (Berger and Mueller 1959, Erickson and Hoppe 1979), a modified dho-gaza trap (Hamerstrom 1963, Clark 1981) with a live Red-tailed Hawk, and a Rock Pi- geon {Columba livia) with noose harness. Each captured individual was banded with a unique color-coded leg band on the left leg and a standard U.S. Geological Survey (USGS) Bird Banding Laboratory band on the right leg. We also recorded morphometric measurements and determined gender. We collected a small amount (±5 mL) of blood from the brachia vein of the left wing of every captured individual and deter- mined gender through DNA typing. Each bird was fitted with a backpack mounted radio transmitter attached via a break-away backpack harness and a leather keel patch (Vekasy et al. 1996), Furthermore, we used visible size differences and markings to separate individual members of a broad-wing pair. Territorial Defense. We documented territorial en- counters between resident Broad-winged Hawks at Rio Abajo Forest and surrounding lands. Moreover, to ex- amine the behavioral interactions of the two sympatric Buteos, we observed territorial interactions between Broad-winged Hawks and intruding Red-tailed Hawks from limestone hilltops. We recorded the behavior of both birds (aggressive or passive) and the end result (de- terred or not deterred) . We used a binomial sign test for a single sample (Daniel 1990, Sheskin 2000) to test the hypothesis that both species of raptors displaced the oth- er randomly during aerial displays. Nest Searches and Monitoring. Broad-winged Hawk use areas were delineated through direct observations from limestone hills and hawk locations were plotted on USGS topographic quadrangles. We then extensively searched areas with radio-marked adults and document- ed aerial displays or other territorial behavior. Spot maps and historical nesting information were used to extend our searches into other potential nesting territories. All potential nests were monitored for reproductive activity beginning in February of each year. When a nest site was located, we built observation blinds 50—100 m from the nest tree at a location on a nearby cliff wall looking down on the nest with clear line- of-sight visibility. In one instance, it was not possible to build a blind with a direct view because of nest location, dense vegetation, or steep rock walls. In this instance, the nest was monitored from the ground at a similar dis- tance. Nest activities were recorded throughout the breeding season using spotting scopes, video cameras, and binoculars from the observation blind. The distance (m) between all occupied nests (nest-site spacing) for the 2001 and 2002 breeding seasons was measured on the ground with surveying tape and verified on a study area map with Geographical Information System (GIS) mea- surements. A two-sample ^-test was used to determine if the spatial distribution of nest sites varied between years (Sheskin 2000). Continuous nest observations were conducted daily throughout the breeding season. Nest checks were ran- domly conducted throughout the day to include all hours when Broad-winged Hawks were active. Based on nest ob- servations and nest checks, we estimated dates of incu- bation, hatching, and fledging. We calculated nest sur- vival (Mayfield 1975) from start of incubation to fledging (total nest survival) and determined nest attentiveness patterns. A nest was considered successful if the pair pro- duced young. We estimated Mayfield nest survival during the incubation and nestling periods using a combined total of 198 incubation exposure-days and 274 nestling exposure-days. Prey Delivery. Food provisioning by adult Broad- winged Hawks to the nest during the breeding season was determined from direct observation. To assess if the birds regularly delivered prey throughout the day, occupied nests were monitored weekly in equal proportion during four time periods: early morning (0800-1100 H), early morning to mid afternoon (1101-1400 H), mid after- noon (1401—1700 H), and late afternoon to early evening (1701-2000 H). Nests were monitored from hatching un- til the young fledged or the nest failed. We calculated the mean number of prey items delivered and the pro- portion of prey deliveries per time period. We divided the prey items into two general categories, large (e.g., birds and rats) and small (e.g., macroinvertebrates and lizards) , to determine prey provisioning patterns. December 2005 Puerto Rican Broad-winged Hawk Ecology 407 An analysis of variance in a randomized complete block design (PROC GLM; SAS Institute 1999) was used to test if Broad-winged Hawks delivered total number of prey items (response variable) per time period (block), number of large prey items per time period, and number of small prey items per time period equally throughout the day. We used multiple comparisons (Least Scientific Difference Means) of mean number of prey items (pooled, large, and small) to examine significant values (a = 0.05) and determine which time periods differed with respect to number of prey items delivered (Sheskin 2000 ). Nest Habitat Model. Areas used by broadwings for nesting at Rio Abajo are valleys with tall forest bounded by limestone ridges and cliff walls, where pairs soar along their respective ridge tops (Delannoy and Tossas 2002). Forest vegetation along cliff walls and limestone ridges are used by resident broadwings for perching, but not for nesting (Hengstenberg and Vilella 2004). Our pri- mary objective was to assess which stand-level variables best described a broadwing nest within the context of the surrounding habitat at Rio Abajo Forest. Therefore, veg- etation characteristics and structure around nest sites were measured at the end of the breeding season and following post-fledging dependency. One nest used both in 2001 and 2002, was included once in the analyses. We described a nest site as all veg- etation within a 0.04 ha plot (11.3-m radius) centered on the nest tree (Titus and Mosher 1981). We recorded hab- itat measurements on nine of ten occupied nests and nine nonuse sites using standard procedures (James 1971). We used a random numbers table to determine a distance and azimuth to travel from a particular nest site to an equivalent nonuse site. We constrained selection of random sites to forested area within a 400-m radius of the nest tree. All random sites were within the nest tree stand in valleys and side slopes. The closest tree to the plot center was chosen as the center point, and habitat variables were measured accordingly. We recorded visual obscurity of the understory using a 2 m Nudds board (Nudds 1977). The board consisted of four 0.5 m sections with 30 orange and white squares. Nudds board measurements were taken from each car- dinal direction at a distance of 10 m from center point. Percentage visual obscurity for the four cardinal direc- tions was averaged for each 0.5 m section. We recorded altitude, aspect, percentage slope, and distance to the nearest rock wall, water, and man-made opening. All woody plants over 2 m tall according to species, diameter at breast height (DBH) , height, and vertical stratification were recorded. Vertical structure was classified into three strata heights (1-understory, 2-midstory, 3-overstory). Nest heights were recorded directly with a measuring tape. Tree heights were measured either by clinometer or through visual estimation. We tested for differences in height between clinometer readings and visual estima- tions using a two-sample West (SAS Institute 1999). We used a spherical densitometer to collect four readings of canopy cover from each cardinal direction at a distance of 5 m from center point. We calculated a mean to esti- mate percentage overstory canopy cover. Because of the paucity of information on nest site char- acteristics of Broad-winged Hawks in Puerto Rico (U.S. Fish and Wildlife Service 1997), we decided to record as much information as possible. We measured all variables considered biologically relevant to woodland raptors (Ti- tus and Mosher 1981). We tested habitat variables for normality using a Kolmogorov-Smirnov Goodness-of-Fit Test (Sheskin 2000). As data were normally distributed, we then used two-sample t-tests to identify variables that differed between nest sites and random sites. Uncorre- lated significant variables were selected for the variable selection model (see Table 1 for variables considered). Logistic regression analysis and AIC modeling was used to determine which variables best discriminate a Broad- winged Hawk nest site from a random site (PROC LO- GISTIC; SAS Institute 1999). To develop broadwing-habitat nest site relationships, we utilized a two model approach (variable selection and model selection) . For both microhabitat analyses, an in- formation-theoretic approach was used for model selec- tion and inference (Burnham and Anderson 2002). We used Akaike’s information criterion (AIC,.), delta AIC (A,), and Akaike’s ranking weight (w,) to determine the best model. Because of the small number of nest sites {N = 9), we recognized the variable selection model may produce bi- ased results (Milliken and Johnson 1984). Therefore, we conducted an alternative AIC model selection approach to assess which biologically-relevant variables best de- scribed Broad-winged Hawk nest sites. This alternative analysis models the particular nest sites without use of stand-level comparisons (nest site versus random site) as in the variable selection model. For the alternative model selection, we chose 10 ex- planatory variables from a list of 27 microhabitat vari- ables (Table 1) based upon literature review and person- al field experience (Titus and Mosher 1981, Delannoy and Tossas 2002). Variables chosen were: aspect, slope, road, rock wall, DBH, canopy cover, Nudds 2, midstory number of stems, overstory number of stems, and canopy height (Table 2). Nest or center tree height was excluded because it is significantly correlated with DBH. Best mod- el selection was based on criteria stated by Burnham and Anderson (2002). Results Breeding Behavior and Territorial Defense. From January to March of 2001 and 2002, we ob- served 158 courtship display flights by known pairs. Most (68%) aerial displays occurred 0917-1107 H {x = 1012 H). Across two breeding seasons, we doc- umented courtship display behavior in 26 pairs throughout the Rio Abajo Forest and surrounding private lands. Of the eleven pairs identified in 2001, occupied nests were located for six of these. In 2002, courtship behavior was observed by 8 of the 11 pairs recorded in 2001. We observed 25 territorial interactions between Broad-winged Hawks and intruding Red-tailed Hawks. Broad-winged Hawks displaced Red-tailed Hawks 84% of the time when an intruding Red- 408 Hengstenberg and Vilella VoL. 39, No. 4 Table 1. Nest habitat characteristics (mean ± SD, range) measured within 0.04 ha of Broad-winged Hawk nest and random sites in Rio Abajo Forest, Puerto Rico, 2001 and 2002. Habitat Characteristics Nest Site (9) Random Site (9) Mean ± SD Range Mean ± SD Range Altitude (m) 235.7 ± 58.0 (150-330) 219.9 ± 62.7 (126-320) Aspect 204.9 ± 105.1 (32-340) 192.7 ± 108.8 (12-334) Slope (%) 46.0 ± 19.0 (18-88) 30.1 ± 24.3 (0-85) Distance to water (m) 194.1 ± 127.1 (37-450) 174.1 ± 120.7 (3-370) Distance to road or trail (m) 68.6 ± 35.5 (25.8-135.0) 40.8 ± 42.8 (3.5-133.0) Distance to cliff wall (m) * 41.1 ± 19.0 (13-75) 71.8 ± 30.2 (35-137) Nest or center tree height (m)* 22.2 ± 7.7 (16.0-35.1) 12.8 ± 5.2 (6.5-22.0) Nest or center tree DBH (cm)* 46.1 ± 15.6 (23.0-74.5) 25.2 ± 13.5 (6.9-42.5) Nest height (m) 16.3 ± 5.6 (10.0-25.9) — Percentage nest height 73.5 ± 6.7 (58.8-81.3) — Canopy cover (%) 85.2 ± 5.8 (72-91) 81.3 ± 6.0 (70-88) Nudds 0.5 m (%) 90.1 ± 11.5 (63.7-100.0) 75.9 ± 17.6 (53.8-97.5) Nudds 1.0 m (%) 75.8 ± 14.2 (59.6-97.5) 59.5 ± 25.4 (21.6-87.6) Nudds 1.5 m (%)* 74.4 ± 14.8 (51.3-95.9) 55.7 ± 18.0 (25.8-77.7) Nudds 2.0 m (%)* 77.1 ± 9.9 (63.7-92.6) 52.9 ± 24.5 (15.9-83.5) Midstory species richness 11.1 ± 5.6 (3-18) 12.0 ± 4.8 (7-21) Midstory # of stems 35.8 ± 16.7 (10-58) 58.6 ± 34.2 (18-124) Midstory stem DBH 1-4.9 cm 15.9 ± 12.7 (1-36) 27.4 ± 21.7 (0-65) Midstory stem DBH 5-8.9 cm 9.0 ± 6.1 (2-19) 16.0 ± 14.1 (2-50) Midstory stem DBH s 9 10.9 ± 5.3 (7-22) 16.3 ± 8.3 (4-33) Overstory species richness 3.1 ± 2.0 (1-6) 3.4 ± 1.9 (2-7) Overstory # of stems 10.2 ± 5.0 (3-17) 7.9 ± 5.1 (2-16) Overstory stem DBH ^ 25.9 cm 3.3 ± 2.9 (0-9) 3.0 ± 3.9 (0-10) Overstory stem DBH 26-49.9 cm 5.6 ± 3.2 (2-11) 4.2 ± 3.5 (0-12) Overstory stem DBH ^ 50 cm 1.3 ± 1.4 (0-5) 0.7 ± 0.9 (0-2) Basal area m^/ha 31.8 ± 13.6 (14.2-53.4) 28.2 ± 14.0 (9.0-56.7) Canopy height (m) 17.9 ± 2.8 (14.4-23.4) 16.6 ± 4.3 (13-26) * Significant Kest (P < 0.05). tailed Hawk entered an occupied territory (p^ = 21/25 = 0.84, P = 0.009, 2-tailed test). In every aerial encounter, the Red-tailed Hawk was the in- truding species. All aerial encounters involved adult birds of both species. Aerial displays involv- ing one Broad-winged Hawk and one Red-tailed Hawk occurred 72% of the time. However, Broad- winged Hawk pairs flew in unison 28% of the time to defend their territory against Red-tailed Hawks. Displays varied from “high-intensity flights” with intruders to “low-intensity flights” between pairs. During displays both birds circled together in close proximity and in the same general direction with the male broadwing flying above the female. Dur- ing low-intensity flights, adults soared upward on widespread wings and fanned tails. In high-inten- sity displays, the adults would alternate between wing flaps and soaring. When the male reached the top of the flight, he performed undulating dives or dipping flight (Wiley and Wiley 1981, Brown and Amadon 1989), consisting of a series of 2-7 shallow dives made toward the female. Most other displays ended with the birds diving straight back with cupped wings at high speeds into the canopy, known as parachuting (Wiley and Wiley 1981, Goodrich et al. 1996). Altitude gained varied among displays; generally the longer the flight, the higher the altitude ob- tained. Flights ranged from about 50 m above the tree canopy to elevations over 450 m. Courtship display flights lasted 60-900 sec (x = 334.7 ± 191.5 sec). Cartwheeling behavior or sky dancing (Good- rich et al. 1996) was only observed three times by three different pairs, in which the male was flying on top of the inverted female with their talons ex- posed. The birds proceeded to tumble with semi- locked talons at high speeds until they reached the forested canopy, where they quickly released their December 2005 Puerto Rican Broad-winged Hawk Ecology 409 Table 2. Model selection: parameters (K), relative AIC^, Delta AIC (A,), Akaike Weights (w,), Goodness-of-Fit statistic, P-value, and percent correct classification for Broad-winged Hawk nests in Rio Abajo Forest, Puerto Rico, 2001 - 02 ). Model Selction K AIC, Ai w, P-VALUE Percent DBH + Canheight + Cliffwall + Overstems 4 20.057 0.000 0.304 10.953 0.027 77.8 DBH 1 21.362 1.305 0.158 6.613 0.010 66.7 Cliffwall 1 22.309 2.252 0.099 5.283 0.022 72.2 DBH + Canheight + Cliffwall 3 22.391 2.334 0.095 9.009 0.029 72.2 Nudds 2 1 22.399 2.342 0.094 5.762 0.016 61.1 DBH + Canheight 2 22.957 2.900 0.071 7.167 0.028 66.7 Aspect + DBH -1- Cliffwall 3 23.928 3.871 0.044 8.569 0.036 72.2 DBH + Canopy cover 2 24.619 4.562 0.031 6.765 0.034 61.1 Road + Cliffwall 2 25.591 5.534 0.019 5.582 0.061 72.2 Cliffwall + DBH 2 25.591 5.534 0.019 8.493 0.014 72.2 Midstems 1 26.050 5.993 0.015 3.018 0.082 55.6 Slope 1 26.974 6.917 0.010 2.346 0.126 66.7 Road 1 27.192 7.135 0.009 2.211 0.137 55.6 Canopy cover 1 27.435 7.378 0.008 1.977 0.160 55.6 Nudds 2 + Midstems + Overstems 3 27.961 7.904 0.006 8.146 0.043 61.1 Overstems 1 28.484 8.427 0.005 1.023 0.312 59.0 Midstems + Overstems 2 28.628 8.571 0.004 3.486 0.175 50.0 Canheight 1 28.862 8.805 0.004 0.650 0.420 38.9 DBH + Canheight + Cliffwall + Aspect 4 29.070 9.013 0.003 9.047 0.060 66.7 Aspect 1 29.458 9.401 0.003 0.066 0.797 0.0 Aspect + Slope 2 30.340 10.283 0.002 2.395 0.302 72.2 talon lock, swooped up, and dispersed upward in separate directions. Territorial flights were elicited by the presence or vocalizations from intruding Red-tailed Hawks and juvenile Broad-winged Hawks in the vicinity of the residents’ territory. Flights varied in intensity and depended on the intruding species and its proximity to the nest. Generally, males were first to fly and confront the intruder (Wiley and Wiley 1981). Adults used alarm vocalizations to warn their mate of an intruding bird. During these dis- plays, adults used stuttered and whistle squeal vo- calizations (Burns 1911). Stuttered and whistle squeals vocalizations were used in high-intensity displays. When Red-tailed Hawks were detected. Broad-winged Hawks responded quickly with rapid- pursuit flights. Resident birds circled and soared to an altitude above the intruding bird and re- peatedly dived at it. The resident male continued to dive at the intruder until the intruder departed the territory. In some cases, the resident bird ex- tended its talons during its dives. In one instance, a male broadwing locked talons with an intruding Red-tailed Hawk; once the intruder left the area, the resident male silently dove or “parachuted” back to its territory. Dipping flight, or undulating display (Wiley and Wiley 1981, Brown and Amadon 1989), was a com- mon behavior used in all intense intruder inter- actions in which the resident bird was successful at chasing the Red-tailed Hawk. Territorial confron- tations between conspecific neighbors were less intense than toward Red-tailed Hawks. From the radiotelemetry study, a radio of a juvenile Broad- winged Hawk (5167) was found and all the feathers were plucked (D. Hengstenberg and F. Vilella un- publ. data). The cause of mortality was determined to be a Red-tailed Hawk, which had been observed numerous times in the same area as the juvenile broadwing. Perched intruders were generally attacked by a slower supplantation flight (Wiley and Wiley 1981) or a dive in which the intruder typically fled the area. If the intruder remained, the residents then circled above and used low angled dives until the intruder departed (Wiley and Wiley 1981). Nesting Biology. We found 10 nests during our study in Rfo Abtyo Forest. Onset of incubation was 410 Hengstenberg and Vilella VoL. 39, No. 4 Table 3. Broad-winged Hawk nests monitored during the breeding seasons 2001 and 2002, Rio Abcgo Forest, Puerto Rico. Reproductive Variables 2001 2002 Both Years Number of nests found 6 7 13 Occupied nests 6 4 10 Failed nests 1 2 3 Successful nests 5 2 7 Proportion successful nests 0.83 0.50 0.70 Mayfield nest success 0.81 0.51 0.67 Number of nestlings 7 3 10 Number fledged 6 2 8 Nestling loss 0.14 0.33 0.20 Fledglings per nest 1.2 1 1.2 from 28 February-21 March in 2001 and from 6- 16 March in 2002. Hatching occurred from 9-20 April in 2001 and from 6-17 April in 2002. In 2001, six juveniles fledged at 35-39 d between 2-25 May, and two juveniles fledged at 35-36 d from 21-22 May in 2002. Three nests failed during the study (Table 3) . Two of the failed nests were attributed to heavy rains. The third nest was depredated by a Red-tailed Hawk. For 2001 and 2002, nest survival for the incu- bation and nestling periods was 0.67 (Table 3). The probability of surviving from nest initiation through fledging was 0.81 in 2001 and 0.51 in 2002. For both years, the probability of nest surviv- al during the incubation period was 1.0. {N = 10). The number of fledglings per successful nest was 1.20 in 2001, 1.0 in 2002, and 1.14 for both years combined (Table 1). During the egg stage, females spent the majority of time incubating. However, our results differ from the available information (Raffaele 1989, U.S. Fish and Wildlife Service 1997), as we documented males engaged in all nesting duties, including in- cubation. In some instances, males were observed continuously incubating for >4 hr and overnight. Adult Broad-winged Hawks frequently brought green vegetation, especially Trichilia hirta, to the nest. During incubation, vegetation may act as a buffer between nest branches and eggs. Raptors may use fresh greenery for concealment, to reduce odors, and to avoid ectoparasites (Wimberger 1984, Sibley 2001). Nest Monitoring and Prey Delivery. We recorded 5534 min of nest observations during the incuba- tion stage and 8825 min through the nestling stage. During the incubation period, females in- cubated approximately 53% of the time and males 23%. We observed both male and female Broad- winged Hawks incubating overnight. The nest was not attended 15%, and an unknown adult incubat- ed 9% of the time. During the nestling period, the nest was not attended 69% of the time, females attended 17%, and males attended 14% of the time. As the nestlings matured, the female spent less time at the nest (Lyons and Mosher 1987). Prey deliveries away from the nest were relatively common, with either member of the pair initiating solicitation calls (Goodrich et al. 1996). The deliv- ering adult, usually the male, would vocalize back and forth until the incubating female flew off the nest to where the male was perched (<50 m from nest) to obtain the prey item. While one adult was eating, the other would fly to the nest and brood. Most prey deliveries during the nestling stage oc- curred during the early morning to mid-afternoon period. We observed 60 prey items delivered to 7 of 10 monitored nests during the brood-rearing periods of 2001 and 2002 (Table 4). Prey consisted of 35% rats {Rattus , 27% lizards, 17% birds, 12% ma- croinvertebrates, 7% unidentified prey, and 3% snakes. Pooled prey items (large and small) varied among the four time periods (F^av = 4.01, P = 0.024). More prey was delivered to nests during early morning to mid-afternoon than early even- ing. Daily prey deliveries were distributed as fol- lows: 38% early morning, 40% late morning to mid afternoon, 17% mid to late afternoon, and 5% late afternoon to early evening. However, neither the number of small prey items (F 327 = 2.83, P = 0.068), nor the number of large prey items (^ 3,27 = 2.80, P = 0.067) brought to nest sites, differed over the course of the day. The earliest prey deliv- ery was recorded at 0856 H, and the latest prey delivery was observed at 1846 H. We calculated a mean prey delivery rate of 0.38 prey (SE — 0.08) items per chick per hour (range = 0.14—0.80). Nesting Habitat. All nests were within 50 m of a rock wall. Nests sites were generally found on southwest facing slopes (x = 204°). Distance among nests averaged 838.5 m (SE = 79.98, range - 200-1455 m) in 2001 and 793.0 m (SE = 91.87, 411-1231 m) in 2002. Distance did not vary be- tween years (^0 ~ 0.362, P = 0.720) . There was no difference amongst visual estimations and clinom- eter readings of nest tree heights when compared December 2005 Puerto Rican Broad-winged Hawk Ecology 411 Table 4. Observed prey items delivered to Broad-winced Hawk nests or consumed in Rio Abajo Forest, Puerto Rico, 2001 - 02 . Common Name Taxonomic Name Observed No. OF Prey Percent Puerto Rican giant centipede / kjcuvujjt/vufu riuri '^ 4 6.7 Puerto Rican arboreal millipede Orthocricus arboreus 3 5.0 Melodius coqui Eleutherodactylus wightmanae 1 1.7 Common coqui Eleutherodactylus coqui 3 5.0 Common anole Anolis cristatellus 2 3.3 Banded anole Anolis stratulus 1 1.7 Yellow-breasted anole Anolis gundlachi 2 3.3 Small green anole Anolis evermanni 1 1.7 Orange dewlap anole Anolis krugi 1 1.7 Snake anole Anolis pulchellus 1 1.7 Green giant anole Anolis cuvieri 2 3.3 Common gecko Spaherodactylus macrolepis 2 3.3 Puerto Rican boa Epicrates inoratus 1 1.7 Ground snake Arrhyton exiguum 1 1.7 White-winged Dove Zenaida asiatica 1 1.7 Bananaquit Coereba flaveola 6 10.0 Puerto Rican Bullfinch Loxigilla portoricensis 3 5.0 Common mouse Mus musculus 12 20.0 Roof rat Rattus rattus 7 11.7 Norway rat Rattus norvegicus 2 3.3 Unidentified prey items — 4 6.7 Total prey items — 60 100 to actual tape-measured heights (P = 0.25). Nest tree height averaged 22.3 ± 7.7 m (range = 16.0- 35.1 m) and nest height averaged 16.3 ± 5.6 m (range = 10.0-25.9 m). Nest tree DBH averaged 46.1 ± 15.6 m (range = 23.0-74.5 cm). Dimen- sions for the two nests measured were: 0.79 m (long diameter) by 0.52 m (short diameter) by 0.61 m (depth), and 0.46 m (long) by 0.31 m (short) by 0.61 m (depth). Nest cup depth measured 1.27 and 2.54 cm, respectively. Within nest site vegeta- tion plots we recorded 13 species of overstory trees, but only four tree species were used as nest trees. Nests were in maria, Honduras mahogany, moca, and guaraguao trees. Of 27 microhabitat variables measured, five dif- fered between nests and random sites (Table 1). Nest sites were closer to cliff walls (t^Q = 2.578, P = 0.020), had greater tree height = —3.020, P = 0.008), larger DBH = -3.048, P = 0.008), and denser understories at 1.5 m (ti 5 = —2.409, P = 0.028) and 2.0 m = -2.742, P = 0.015) than random sites. Logistic regression produced a best nest site model containing two variables: DBH (parameter = 0.1800, SE = 0.0965, x^ = 3.4814, P = 0.062), and Nudds 2 m (parameter = 0.1297, SE = 0.0865, X^i = 2.2512, P = 0.134). This variable combina- tion correctly classified Broad-winged Hawk nests 83.3% of the time (Table 5). The best AIC model for nest sites contained DBH (parameter = 1.2853, SE = 1.5429, xS = 0.694, P = 0.405), canopy height (parameter = 5.5472, SE = 7.5492, x^i ~ 0.5399, P = 0.540), distance to rock wall (param- eter = -0.4281, SE = 0.8384, x^ = 0.2607, P = 0.610), and overstory stems (parameter = 2.6233, SE = 3.5889, x^ = 0.5343, P = 0.465). These four variables correctly classified nests 77.8% of the time (Table 2). Discussion Albeit our small sample of nests, phenology was similar across both years, with the onset of incu- bation beginning in late February and juveniles fledging by the end of May. The post-fledging de- pendency period lasted 4—8 wk after the juveniles left the nest. During the first 2-3 wk post-fledging, juveniles frequently returned to the nest to receive 412 Hengstenberg and Vilella VoL. 39, No. 4 Table 5. Variable selection of Broad-winged Hawk nest sites in Rio Abajo Forest, Puerto Rico, 2001-02. Significant parameters (K), relative AIC (AICJ, Delta AIC (AJ, Akaike Weight (w*), Goodness-of-Fit statistic (x^), P-value, and percent correct classification. Variable Selection K AIC, A, P-VALUE Percent DBH + Nudds 2 2 16.774 0.000 0.551 10.754 0.005 83.3 Cliffwall + DBH 2 19.386 2.612 0.149 8.493 0.014 72.2 Cliffwall + DBH -1- Nudds 2 3 19.630 2.856 0.132 11.553 0.009 83.3 DBH 1 21.362 4.588 0.056 6.613 0.010 66.7 Cliffwall + Nudds 2 2 21.770 4.996 0.045 8.595 0.014 83.3 Cliffwall 1 22.309 5.535 0.035 5.283 0.022 72.2 Nudds 2 1 22.399 5.625 0.033 5.762 0.016 61.1 prey deliveries from the adults and to roost for the night. Broad-winged Hawk nests in Puerto Rico aver- aged 1.1 young per nest attempt and 67% nest suc- cess. Our estimate is almost double what Delannoy and Tossas (2002) reported (0.66 fledglings/nest) for Broad-winged Hawk nests in Rio Abajo from 1994 to 1996. However, our estimates of nest suc- cess and young per nest attempt were slightly lower compared to Broad-winged Hawk studies in North America (Armstrong and Euler 1983, Crocoll 1984, Rosenfield 1984). The number of fledglings per successful nest and overall nest success was greater in 2001 than in 2002. Lower nest success in 2002 may have been attributed to rain events that occurred in April of 2002. Two nest sites were abandoned within the same week of heavy rain in April 2002. Santana and Temple (1988) reported lower success of Red- tailed Hawks nesting in the eastern Luquillo Moun- tains rainforest region during extensive rainy pe- riods. Similarly, severe rainfall was suggested as a cause of nest failures of the Puerto Rican Sharp- shinned Hawk {Accipiter striatus vennator) in the Lu- quillo Mountains (Snyder and Wiley 1976) and in forests of the central mountain range of the island (Delannoy and Cruz 1988). April and May are important months to nestling survival. During this time period, Broad-winged Hawks are brooding partially feathered nestlings. In 2001, total precipitation from April and May was 44.2 cm. Conversely, April-May precipitation in 2002 was 54.4 cm. April rains in 2002 coincided with the presence of recently-hatched chicks or young nestlings and may have caused nest aban- donment and hypothermia of young at the two failed nests. At Rio Abajo Forest, Broad-winged Hawks for- aged primarily on rats, lizards, and small birds (Ta- ble 4) . The Broad-winged Hawk is an opportunistic feeder who forages on a wide variety of prey (Rush and Doerr 1972, Reran 1978). However, our in- ability to detect a daily pattern of prey deliveries may have been a result of our small sample of nests and prey delivery observations. Prey size or type exploited at Rio Abajo Forest may be a function of seasonality (wet vs. dry), as changing weather conditions may produce differ- ences in dietary patterns (Grubb 1977, Stinson 1980). In tropical environments such as Puerto Rico, rain may limit foraging opportunities (Foster 1974). We observed Broad-winged Hawks at Rio Abajo Forest were less active during periods of rain. There was no difference in the spatial distribu- tion of nest sites between years, suggesting Broad- winged Hawks may maintain territories year round. These clusters of nests (Fig. 1) are bounded by limestone ridges and cliff walls, where pairs soar along their respective ridge tops. This may have some advantages. Pairs may be better able to detect intruding Red-tailed Hawks. Vigilance may contrib- ute to greater survival of nesting birds (Alcock 1993). Breeding Broad-winged Hawks were aggressive and successfully deterred intruding Red-tailed Hawks from their nesting territories. Delannoy and Tossas (2002) speculated similar nest-site require- ments between Broad-winged Hawks and Red- tailed Hawks could lead to aggressive encounters. However, we found no evidence of nesting by Red- tailed Hawks within the closed canopy forests of Rio Abajo. By and large, Broad-winged Hawk court- ship and territory defense behavior in Puerto Rico was similar to that of the Ridgway’s Hawk {Buteo December 2005 Puerto Rican Broau-winged Hawk Ecology 413 ndgwayi) in moist limestone forests of the Domin- ican Republic (Wiley and Wiley 1981). Territory occupancy seemed relatively stable be- tween years. The year-round residency and site fi- delity of Broad-winged Hawks in the moist lime- stone region of Puerto Rico may be indicative of long-term pair bonds (Griffin et al. 1998). Like other tropical raptors (Mader 1982, Griffin et al. 1998), the subspecies in Puerto Rico seems to ex- hibit high site fidelity (73%, N — 11), Our results from radio-marked breeding birds (D. Hengstenberg and F. Vilella unpubl. data) sug- gest Broad-winged Hawks tend to nest in the same stand or nearby from one year to the next (Reran 1978). We observed courtship behavior in 15 ad- ditional pairs, but found no evidence of nests or nest building. This may suggest that while some pairs may hold nesting territories, they do not nec- essarily build a nest or lay eggs every year, as has been documented in other raptors (Steenhof 1987). Five nests constructed in maria trees were placed atop termite nests in the main crotch of the tree. Broad-winged Hawks nesting in North America sometimes place their nest on top of old bird and squirrel nests (Goodrich et al. 1996). Distance to cliff wall, tree height and DBH, Nudds board 1.5 m, and Nudds 2.0 m differed be- tween nest and random sites. Variables DBH and Nudds 2 m best classified nest sites, suggesting that Broad-winged Hawks at Rio Abajo Forest may pre- fer large trees and dense understories (Table 5). Also, Broad-winged Hawks in North America avoid- ed smaller trees and selected large DBH trees (Ti- tus and Mosher 1987). At Rio Abajo Forest, nest sites had denser understories than random sites. Dense understories may be related to prey avail- ability for adults. These dense understories may of- fer fledglings protection from predators and great- er foraging opportunities of prey. Radio-marked adults and juveniles were frequently observed hunting in the dense understory around their nest sites. Foraging habitat studies have suggested Broad-winged Hawks select sites with high prey availability (Steblein 1991). However, further re- search is required to better understand the rela- tionships between Broad-winged Hawks in Rio Aba- jo Forest and prey populations. Model selection procedures for nests (site spe- cific) yielded a four-variable model in which DBH, canopy height, cliff wall, and overstory stems cor- rectly classified nest sites (Table 2). This suggests closeness to karst cliff walls and canopy height may be additional predictors of Broad-winged Hawk nest habitat, in addition to basal area (i.e., DBH) and understory cover. Nest tree height averaged 22.2 m, whereas can- opy heights of nest plots averaged 17.9 m and ran- dom sites averaged 16.6 m, suggesting Broad- winged Hawks select emergent trees for their nests. Our results coincide with nest site characteristics of North American broadwings (Goodrich et al. 1996). Delannoy and Tossas (2002) reported a mean nest tree height of 27.0 m and a mean can- opy height of 15.7 m. On average, nest heights in Puerto Rico were taller than nest heights reported from North America (Burns 1911, Matray 1974, Ti- tus and Mosher 1981, Armstrong and Euler 1983, Rosenfield 1984). This may reflect the relative lengths of trees in tropical versus temperate forests (Fedorov 1966). Broad-winged Hawk nest sites were located with- in 50 m of a cliff wall. In the karst region, cliff walls are very abundant. Cliff walls may offer nest sites with adequate protection from the elements (wind, rain, and sun), intruding predators, provide van- tage points, and facilitate reduction in energy re- quirements when searching for thermal updrafts. Nests sites were generally found on slopes facing southwest (x = 204°). This nest placement may help protect the nests from the prevailing easterly winds. Broad-winged Hawk nest sites in a limestone forest may be described as occurring in mature closed-canopy overstory stands sheltering a thin midstory, a dense understory, and in close prox- imity to a cliff wall. Conservation and persistence of the breeding population of Rio Abajo Forest may depend on management of the existing forest stands used by Broad-winged Hawks. Further research is required in Rio Abajo Forest to increase sample sizes of nest sites and validate the broadwing-habitat relation- ships revealed by our habitat model. Moreover, var- iance estimates of parameters from our study will provide baseline information needed to calculate sample sizes for future research. We suggest addi- tional studies to quantify broadwing habitat in oth- er localities of the karst region and to develop hab- itat models at multiple spatial scales. At Rio Abajo Forest, managers should limit dis- turbance within valleys used by broadwings during the critical nest initiation and incubation periods (i.e., February to April). Based on our preliminary results on habitat relationships, silvicultural prac- 414 Hengstenberg and Vtt.et.ia VoL. 39, No. 4 tices within Rio Abajo Forest that promote main- tenance of canopy emergent trees and dense un- derstories may improve habitat conditions for nesting pairs as well as fledglings during their de- pendence period. Moreover, Broad-winged Hawks readily used plantation tree species such as maria and Honduras mahogany for nest sites. We rec- ommend the DNER Forest Service encourage sur- rounding private landowners to engage in agrofor- estry practices using these fast-growing plantation species. Additionally, programs for private lands that promote maintenance and enhancement of forest cover (e.g., USFWS Partners for Wildlife) should be brought to the attention of the land- owners adjoining Rio Ab^o Forest. In an attempt to establish a second wild popu- lation, releases of captive-reared Puerto Rican Par- rots are scheduled for 2006 (U.S. Fish and Wildlife Service 1999). Available information suggests Puerto Rican Parrots may exceed the size of avian prey taken by Broad-winged Hawks (Snyder and Kepler 1987). At Rio Abajo, 61% of prey deliveries to nests were rodents and Anolis lizards (Table 4) . Forest songbirds (e.g., Puerto Rican Bullfinch [Lo- xigilla portoricensis] and Bananaquit [ Coereba flaveo- la] ) were the avian prey taken (Table 4) . In contrast, Red-tailed Hawks are known parrot predators (White et al. 2005) . However, our results indicated resident Broad-winged Hawks chased off intruding Red-tailed Hawks effectively in Rio Abajo Forest. Owing to the likely negative relationship between these sympatric Buteos, resident Broad- winged Hawks in Rio Abajo may indirectly provide some degree of protection to released parrots from predation by excluding intruder Red-tailed Hawks. However, research is required to examine the re- lationship between spatial overlap of parrots and broadwings and the likelihood of Red-tailed Hawk predation on released parrots. Other studies have reported some avian species select nest sites close to more aggressive species that regularly attack or mob predators (Durango 1949, Clark and Robertson 1979, Wiklund 1979, Dyrcz et al. 1981, Norrdahl et al. 1995). The Wood- pigeon {Columba palumbus) benefits from nesting in association with Eurasian (northern) Hobbies (Falco subbuteo; Bogliani et al. 1999). Nevertheless, both the Puerto Rican Parrot and Broad-winged Hawk are listed as endangered (U.S. Fish and Wildlife Service 1997, 1999). Therefore, a co-management approach will be required to en- sure habitat management activities for one species are not done at the expense of the other. We rec- ommend parrot habitat management activities (i.e., deployment of artificial cavities) should be limited to the nonbreeding season (August-De- cember) to minimize disturbance to Broad-winged Hawk nesting pairs and post-fledging dependent juveniles. Ultimately, the future of both these endangered species rests on the ability to disseminate research results to forest managers and policymakers. This information in turn will help to guide the protec- tion and conservation of the karst forest region of Puerto Rico, as further forest fragmentation will impact severely the recovery of both the broadwing and the parrot. Multiagency efforts are underway to acquire and protect a significant portion (>30 000 ha) of forest in the moist karst region of Puerto Rico (Lugo et al. 2001). Broad-winged Hawks do not limit their activities to the Rio Abajo Forest boundaries, and their fate in the surround- ing private lands may be uncertain. Therefore, DNER forest managers should work proactively with the surrounding land owners to promote land-use practices to conserve and to enhance ex- isting forest cover. Future patterns of land use around the forest boundary may indirectly and di- rectly affect the ability of the Rio Abajo Forest to function as an effective conservation unit for the Broad-winged Hawk. Acknowiedgments Funding was provided by the USFWS, Caribbean Field Office. The DNER Forestry Division provided access to Rio Abajo Forest. We are indebted to J. Casanova, DNER Management Officer at Rio Abajo Forest, and his staff for providing assistance and logistic support. The DNER Terrestrial Resources Division provided housing facilities at Rio Abajo Aviary. We are grateful to I. Llerandi, J. Rios, and A. Jordan for field assistance. Thanks to student vol- unteers from the University of Puerto Rico at Arecibo and Utuado. We are grateful for the helpful comments of C. Delannoy on the raptors of Puerto Rico. We thank J. Bednarz, K. Bildstein, L. Goodrich, K. Titus, and J. Wi- ley for suggestions that greatly improved earlier versions of the manuscript. Animal collection and handling pro- cedures were conducted under the auspices of DNER permit Ol-EPE-019, Master Station Permit 22456 of the USGS Bird Banding Laboratory, and Mississippi State University’s Institutional Animal Care and Use Commit- tee protocol No. 00-094. Literature Cited Ai,cock, j. 1993. Animal behavior: an evolutionary ap- proach. Sinauer Associates, Sunderland, MA U.S.A. Armstrong, E. and D. Euler. 1983. Habitat usage of two December 2005 Puerto Rican Broad-winged Hawk Ecology 415 woodland Buteo species in central Ontario. Can. Field Nat. 97:200-207. Berger, D.D. and H.C. Mueller. 1959. The bal-chatri: a trap for the birds of prey. Bird Banding 30:18-2&. Bogliani, G., F. Sergio, and G. Tavecchia. 1999. Wood- pigeons nesting in association with Hobby Falcons: ad- vantages and choice rules. Anim. Behav. 57:125-131. Brown, L. and D. Amadon. 1989. Eagles, hawks, and fal- cons of the world. Wellflcct Press, Sec3.nciis, U.S.A. Burnham, K.P. and D.R Anderson. 2002. Model selection and multimodel inference. Springer-Verlag, New York, NY U.S.A. Burns, F.L. 1911. A monograph of the Broad-winged Hawk {Buteo platypterus) . Wilson Bull. 23:139-320. Cardona, J.E., M. Rivera, M. Vazquez Otero, and C.R. Laboy. 1987. Availability of food resources for the Puerto Rican Parrot and Puerto Rican Plain Pigeon in the Rio Abajo Forest. Final Report, Puerto Rico Department of Natural Resources, Project W-lO-ES-1. Clark, K.L. and R.J. Robertson. 1979. Spatial and tem- poral multi-species nesting aggregation in birds as anti-parasite and anti-predator defenses. Behav. Ecol. Sociobiol. 5:359-371. Clark, W.S. 1981. A modified dho-gaza trap for use at a raptor banding station. J. Wildl. Manage. 45:1043- 1044. Crocoll, S.T. 1984. Breeding biology of Broad-winged and Red-shouldered Hawks in Western New York. M.S. thesis, State University College, Fredonia, NY U.S.A. and J.W. Parker. 1989. The breeding biology of Broad-winged and Red-shouldered Hawks in western New York. y. Raptor Res. 23:125-139. Daniel, W.W. 1990. Applied Nonparametric Statistics, 2nd Ed. PWS-KENT Publishing Company, Boston, MA U.S.A. Delannoy, C.A. 1997. Status of the Broad-winged Hawk and Sharp-shinned Hawk in Puerto Rico. Carib.J. Sci. 33:21-33. and a. Cruz. 1988. Breeding biology of the Puerto Rican Sharp-shinned Hawk {Accipiter striatus vennator). Auk 105:649—662. AND A.G. Tossas. 2002. Breeding biology and nest site characteristics of Puerto Rican Broad-winged Hawk at the Rio Abayo forest. Caribb. J. Sci. 38:20-26. Durango, S. 1949. The nesting association of birds with social insects and with birds of different species. Ibis 91:140-143. Dyrcz, a., j. Witkowski, and J. Okulewicz. 1981. Nesting by ‘timid’ waders in the vicinity of ‘bold’ ones as an antipredator adaptation. Ibis 123:542—545. Erickson, M.G. and D.M. Hoppe. 1979. An octagonal bal- chatri trap for small raptors. Raptor Res. 13:36-38. Ewel, JJ- AND J.L. Whitmore. 1973. The ecological life zones of Puerto Rico and the U.S. Virgin Islands. USDA Forest Service Research Paper ITF-18. Rio Piedras, Puerto Rico. Federal Register. 1994. Endangered and threatened wildlife and plants; determination of endangered sta- tus for the Puerto Rican Broad-winged Hawk and the Puerto Rican Sharp-shinned Hawk. 59:46710—46715. Eedorov, A.A. 1966. The structure of the tropical rain forest and speciation in the humid tropics. J. Ecol. 54' 1 - 11 . Pitch, H.S. 1974. Observations on the food and nesting of the Broad-y.'inged Hawk in northeastern Kansas. Condor 76:331—333. Poster, M.F. 1974. Rain, feeding behavior, and clutch size in tropical birds. Auk 91:722-726. Goodrich, L., J. Crocoll, and S.E. Senner. 1996. Broad- winged Hawk (Buteo platypterus ) . In A. Poole and F Gill [Eds.] . The birds of North America, No. 218. The Academy of Natural Sciences, Philadelphia, PAU.S.A., and the American Ornithologists’ Union, Washing- ton, DC U.S.A. Griffin, C.R., P.W.C. Paton, and T.S. Baskett. 1998. Breeding ecology and behavior of the Hawaiian Hawk. Condor 100:654-662. Grubb, T.C. 1977. Weather-dependent foraging in Os- preys. Auk 94:146-149. Hamerstrom, F. 1963. The use of Great Horned Owls m catching Marsh Hawks. Proc. XIII Int. Ornithol. Congr 13:866-869. Hengstenberg, D.H. and FJ. ViLELLA, 2004. Reproduc- tive biology, abundance and movement patterns of the Broad-winged Hawk Buteo platypterus brunnescens in a moist limestone forest of Puerto Rico. Final Report, Cooperative Agreement No. 14-45-009-1543-59. U.S. Geological Survey Biological Resources Division, Mis- sissippi Cooperative Fish and Wildlife Research Unit. Mississippi State, MS U.S. A. James, F.C. 1971. Ordinations of habitat relationships among breeding birds. Wilson Bull. 83:215—236. Reran, D. 1978. Nest site selection by the Broad-winged Hawk in north-central Minnesota and Wisconsin. Rap- tor Res. 12:15-20. Lugo, A.E., L. Miranda Castro, A. Vale, T. del Mar Lopez, E. Hernandez Prieto, A. Garcia, A.R. Puente Rolon, A.G. Tossas, D.A. McEarlane, T. Millar, O Ramos, and E. Helmer. 2001. Puerto Rican karst — a vital resource. US Department of Agriculture Forest Service, International Institute of Tropical Forestry, GTR-WO-65. Rio Piedras, Puerto Rico. Lyons, D.M. and J.A. Mosher. 1987. Morphological growth, behavioral development, and parental care of Broad-winged Hawks. J. Field Ornithol. 58:334—344. Mader, W.J. 1982. Ecology and breeding habits of the Savanna Hawk in the llanos of Venezuela. Condor 84: 261-271. Matray, P.F. 1974. Broad-winged Hawk nesting and ecol- ogy. Auk 91:307-324. Mayfield, H.F. 1975. Suggestion for calculating nest suc- cess. Wilson Bull. 87:456—466. Milliken, G.A. AND D.E. Johnson. 1984. Analysis of 416 Hengstenberg and Vilella VoL. 39, No. 4 messy data. Vol. 1. Designed experiments. Van Nos- trand Reinhold Company, New York, NY U.S.A. National Oceanographic and Atmospheric Adminis- tration (NOAA) . 2002. Annual Climatological Sum- mary, National Climate Data Center. Asheville, NC U.S.A. www.ncdc.noad.gov/oa/climate/station/ stationlo cator. h tml Norrdahl, K., J. Suhonen, O. Hemminki, and E. Korpim- Aki. 1995. Predator presence may benefit: kestrels pro- tect curlew nests against nest predators. Oecologia 101: 105-109. NcfDDS, T.D. 1977. Quantifying the vegetative structure of wildlife cover. Wildl. Soc. Bull. 5:113-117. Raffaele, H.A. 1989. A guide to the Birds of Puerto Rico and the Virgin Islands. Princeton University Press. Princeton, NJ U.S.A. Rosenfield, R.N. 1984. Nesting biology of Broad-winged Hawks in Wisconsin. Raptor Res. 18:6-9. Rush, D.H. and P.D. Doerr. 1972. Broad-winged Hawk nesting and food habitats. Auk 89:139-145. Santana, E.C., and S.A. Temple. 1988. Breeding biology and diet of Red-tailed Hawks in Puerto Rico. Biotropica 20:151-160. SAS Institute. 1999. SAS/STAT® user’s guide, version 8, 4th Ed., SAS Institute Inc., Cary, NC U.S.A. Sheskin, D.J. 2000. Handbook of parametric and non- parametric statistical procedures, 2nd Ed. Chapman and Hall, New York, NY U.S.A. Sibley, D.A. 2001. The Sibley guide to bird life & behav- ior. Alfred A. Knopf. New York, NY U.S.A. Snyder, N.F. and C.B. Kepler. 1987. The parrots of Lu- quillo: Natural history and conservation of the Puerto Rican Parrot. Western Foundation of Vertebrate Zo- ology, Los Angeles, CA U.S.A. AND J.W. Wiley. 1976. Sexual size dimorphism in hawks and owls in North America. Ornithol. Monogr. 20 . Steblein, RF. 1991. Influence of spatial and temporal dynamics of prey populations on patch selection by Broad-winged Hawks (Abstract)./. Raptor Res. 25:160- 161. Steenhof, K. 1987. Assessing raptor reproductive success and productivity. Pages 157-170 in B.G. Pendleton, B.A. Millsap, K.W. Cline, and D.M. Bird [Eds.], Insti- tute for Wildlife Research. Raptor Management Tech- niques Manual. Scientific and Technical series No. 10, National Wildlife Federation, Washington, DC U.S.A. Stinson, C.H. 1980. Weather-dependent foraging success and sibling aggression in Red-tailed Hawks in central Washington. Condor 82:76—80. Titus, K. 1987. Selection of nest tree species by Red- shouldered and Broad-winged hawks in two temper- ate forest regions. J. Field Ornithol. 58:274—283. AND J.A. Mosher. 1981. Nest-site habitat selected by woodland hawks in the central Appalachians. Auk 98:270-281. U.S. Fish and Wildlife Service. 1997. Puerto Rican Broad-winged Hawk and Puerto Rican Sharp-shinned Hawk Recovery Plan. U.S. Fish and Wildlife Service, Atlanta, GA U.S. A. . 1999. Technical/Agency Draft Revised Recovery Plan for the Puerto Rican Parrot {Amazona vittata). U.S. Fish Wildlife Service, Atlanta, GA U.S.A. Vekasy, M.S., J.M. Marzluff, M.N, Kochert, R.N. Leh- man, and K. Steenhof. 1996. Influence of radio trans- mitters on Prairie Falcons, f. Field Ornithol. 67:680- 690. White, T.H., J.A. Collazo, and F.J. Vilella. 2005. Sur- vival of captive-reared Puerto Rican Parrots released in the Caribbean National Forest. Condor 107:424— 432. Wiklund, C.G. 1979. Increased breeding success for Mer- lin {Falco columbarius) nesting among fieldfare (Turdus pilaris) colonies. Ibis 121:109-111. Wiley, J. A. and B.N. Wiley. 1981. Breeding season ecol- ogy and behavior of Ridgway’s Hawk (Buteo ridnuayi). CowtZor 83:132-151. Wimberger, P.H. 1984. The use of green plant material in bird nests to avoid ectoparasites. Auk 101:615-618. Received 28 June 2004; accepted 27 August 2005 J. Raptor Res. 39(4):417-428 © 2005 The Raptor Research Foundation, Inc. RAPTOR ABUNDANCE AND DISTRIBUTION IN THE LLANOS WETLANDS OF VENEZUELA Wendy J. Jensen, 1 Mark S. Gregory, and Guy A. Baldassarre State University of New York, College of Environmental Science and Forestry, 1 Forestry Drive, Syracuse, NY 13210 US. A. Francisco J. Vilella uses Biological Resources Division, Cooperative Research Units, Department of Wildlife and Fisheries, Box 9691, Mississippi State, MS 39762-9691 U.S.A. Keith L. Bildstein Acopian Center for Conservation Fearning, Hawk Mountain Sanctuary, 410 Summer Valley Road, Orwigsburg, PA 17961 U.S.A. Abstract. — The Llanos of Venezuela is a 275 OOO-km® freshwater wetland long recognized as an impor- tant habitat for waterbirds. However, litde information exists on the raptor community of the region. We conducted raptor surveys in the Southwestern and Western Llanos during 2000-02 and detected 28 species representing 19 genera. Overall, areas of the Llanos that we sampled contained 52% of all raptor species and more than 70% of the kites, buteos, and subbuteos known to inhabit Venezuela. Regional differences in the mean number per route for four of the 14 most common species, the Crested Ca- racara ( Caracara plancus ) , Black-collared Hawk {Busarellus nigricollis) , American Kestrel {Falco sparverius) , and Osprey (Pandion haliaetus), were significant {P < 0.0018) in relation to the wet or dry seasons. Of the 14 less common species, six were detected in only one season (wet or dry). The Southwestern and Western regions of the Llanos support a rich raptor community composed primarily of nonmigratory wetland-dependent and upland-terrestrial species. Key Words: Neotropics-, Venezuela, Llanos, savanna, wetlands', roadside surveys. DISTRIBUCION YABUNDANCIA DE RAPACES EN HUMEDALES DE LOS LLANOS DE VENZUELA Resumen. — Los llanos de Venezuela constituyen un humedal de agua dulce de 275 000 km^ que ha sido tradicionalmente reconocido como un ambiente importante para las aves acuaticas. Sin embargo, existe poca informacion sobre la comunidad de rapaces de la region. Realizamos censos de aves rapaces en el sudoeste y el oeste de los llanos entre 2000 y 2002 y detectamos 28 especies que representaron 19 generos. En total, las areas de los llanos que censamos contuvieron el 52% de todas las especies de rapaces y mas del 70% de los elanios, buteos y subbuteos que habitan en Venezuela. Las diferencias regionales en el numero medio por ruta para cuatro de las 14 especies mas comunes, Caracara plancus, Busarellus nigricollis, Falco sparverius y Pandion haliaetus, fueron significativas (P< 0.0018) con relacion a las estaciones humeda y seca. De las 14 especies menos comunes, seis fueron detectadas en una sola estacion (humeda o seca). Las regiones del sudoeste y del oeste de los llanos albergan una rica comu- nidad de aves rapaces compuesta primariamente por aves no migratorias que dependen de humedales y de especies terrestres de lugares elevados. [Traduccion del equipo editorial] South America comprises 12% of the world’s land surface, yet supports 28% of all raptors (Bie- rregaard 1998). Most South American raptors do not appear threatened globally, but more infor- mation is needed to confirm current assessments ^ Email address: wjjensen@syr.edu and appropriately address threats (Bierregaard 1998, Bildstein et al. 1998). Community level rap- tor research in South America has been primarily focused in forest habitats (e.g., Thiollay 1984, Thiollay 1989, Alverez et al. 1996, Manosa and Pe- drocchi 1997, Manosa et al. 2003); thus, little is known about the raptor populations within the ex- 417 418 Jensen et al. VoL. 39, No. 4 tensive savanna and grassland regions of the con- tinent. Raptor surveys in nine South American countries during 1979 detected the greatest num- ber of species in the savannas, mixed riparian for- ests, pastures, and open areas of interior Venezuela (Ellis et al. 1990), indicating that these areas are important habitats for South American raptors. Furthermore, Neotropical raptors of open land and savanna habitats are currently threatened by habitat loss, including wetland depletion and land- scape homogenization (Alvarez-Lopez and Kattan 1995). In Venezuela, the savannas of the interior are part of an extensive (275 000 km^) wetland com- plex called the Llanos. The Llanos cover approxi- mately 31% of Venezuela (Mittermeier et al. 2003) and are located in the latitudinal region character- ized by the greatest avian endemism in the North- ern Hemisphere (Bibby et al. 1992). Between 32- 36 nonmigratory and North American migratory raptor species use some or all of the Llanos (Fer- guson-Lees and Christie 2001, Hilty 2003). Al- though the natural history, biology, and habitat as- sociations of some of these species have been studied locally (Mader 1981, 1982, Beissinger et al. 1988, Balgooyen 1989, Kirk and Currall 1994), community based, landscape-level surveys are lack- ing. Our objective was to document and compare the species richness, relative abundance, and distribu- tion of nonmigratory and migratory raptors in the savannas of the Southwestern and Western regions of the Venezuelan Llanos. We also compared these population parameters between the distinct wet and dry seasons that characterize the Llanos. Study Area Venezuela supports 1381 species of birds (Hilty 2003) and is considered a globally important region of biodi- versity in part due to its rich avifauna (Mittermeier and Mittermeier 1997, Myers et al. 2000), The Venezuelan Llanos is located between ca. 6-9°N and 63— 71 TV and is bordered by the Coastal Cordillera to the north, the Ori- noco Delta and Guiana Shield to the east and southeast, Colombia to the south and southwest, and by the Andes Mountains to the northwest. Annual rainfall in the Venezuelan Llanos ranges from 90-180 cm (Silva and Moreno 1993), with most rain fall- ing and widespread flooding occurring from April through November (Cole 1986). In contrast, the late No- vember-late April dry season typically is rain free (Troth 1979). The Venezuelan Llanos is divided into three general areas: western, central, and eastern (Huber and Alcaron 1988). Covering 90 000 km‘^, the western area comprises 35% of the freshwater wetland in Venezuela and spans north and west of the Orinoco River from ca. 69-7l°W and 6-9°N (Bulla et al. 1990). This area is relatively flat, with elevation ranging from sea level at the Orinoco Riv- er to 155 m near the foothills of the Andes. The western area is further divided into two distinct regions known as the Southwestern and Western Llanos (Fig. 1; Huber and Alcaron 1988). The Southwestern Llanos includes the vast open-sa- vanna-wetland habitats that extend from the Meta and Orinoco rivers northwest to the agricultural-savanna-for- est mosaic habitats of the Western Llanos. The South- western Llanos is characterized by poor soils, savanna habitats with small patches of trees, gallery forest, and extensive wet season flooding that renders the region largely unsuitable for agriculture (Huber and Alcaron 1988). The Western Llanos encompasses the alluvial plains bordered by the foothills of the Andes and extends south- east to the open savannas of the Southwestern Llanos. This region is characterized by fertile soils and partial flooding that support agricultural production and native forests dominated by tree species similar to those of the Amazon basin (Huber and Alcaron 1988, Silva and Mo- reno 1993). Methods Sampling Design. We surveyed raptors along the sparse network of roads accessible during both the wet and dry seasons in the Southwestern and Western Llanos (Fig. 1) Despite limitations inherent in roadside surveys (Millsap and LeFranc 1988, Bunn et al. 1995), such counts can be used to survey relative raptor abundance, community composition, and habitat associations across large land- scapes (Woffinden and Murphy 1977, Thiollay 1978, Ellis et al. 1990, Sorley and Andersen 1994, Seavy and Apo- daca 2002). However, roadside surveys in open habitats may be inadequate for detecting small and uncommon raptor species unless surveys incorporate frequent and regular stops (Whitacre and Turley 1990). Therefore, we used the North American Breeding Bird Survey model (Droege 1990) to establish stationary surveying points along road routes. We placed 50 survey routes along roads where the over- all distribution of survey routes was dictated by accessi- bility of roads during the wet season (Fig. 1). Each route was 22.5 km long with 16 sample points spaced 1.5 km apart (Jensen 2003: Appendix A) . Each sample point was surveyed by a 2-person surveying team for 3 min to the right and left side of the road for a total of 6 min/ point. All raptors were tallied within a 500-m radius as measured by rangefinder binoculars. Except for the King Vulture {Sarcoramphus papa), we did not tally Cathartid vultures. Scientific and common names of birds follow Ferguson- Lees and Christie (2001). Raptors were surveyed during rainless periods of the day, primarily from 0700-1200 H, and never later than 1400 H. Surveys were conducted over ca. 6-wk periods twice each year between August 2000 and March 2002 The 6-wk periods coincided with the end of the wet (Au- gust-October) and dry seasons (January-March). We surveyed 27 routes in the wet season of 2000, 39 in the December 2005 The Raptor Community of the Llanos 419 North Central Llanos Central Western Llanos Llanos Southwestern Llanos Legend c j Regions of the Venezuelan Llanos n The Colonijian L i^nos crwTis 10°N 8"N 6"N 70" W 68" W 66"Vy Figure 1. Study area map of the raptor survey during wet seasons 2000 and 2001 and dry seasons 2001 and 2002 in the Southwestern and Western Llanos of Venezuela. Route clusters are areas where groups of survey routes con- taining sample points are located. 420 Jensen et al. VoL. 39, No. 4 dry season of 2001, 50 in the wet season of 2001, and 50 in the dry season of 2002 . Analysis. We calculated the total number of individuals detected on all surveys, as well as species totals, percent composition, and frequency of detection. We classified percent composition into four species abundance classes: Very common (10-26% of all individuals detected), Com- mon (3-5%), Uncommon (1-2%), and Rare (<1%). We used Estimates Version 6 . Obi software (Colwell 2000) to generate a species accumulation curve to evaluate the probability that our 2-yr 50-route roadside survey design was adequate for documenting all detectable raptor spe- cies. We then calculated the mean number of individuals per species per route (mean number per route) by year, season, and region to investigate annual changes in abun- dance between the 2000 and 2001 wet seasons and the 2001 and 2002 dry seasons for each region. For this com- parison we used the 27 routes surveyed during the first survey visit (wet season 2000 ) and resurveyed for the du- ration of the survey (2001 and 2002): Southwestern Lla- nos (15 routes) and Western Llanos (12 routes). Rare abundance class species were detected in numbers too low to meet the underlying assumptions for a /-test and were omitted from all subsequent statistical tests. For the 14 most common species, we used paired /-tests to eval- uate four hypothesis (Hoj: mean number per route Southwestern Llanos wet season 2000 = mean number per route Southwestern Llanos wet season 2001; Hog: mean number per route Southwestern Llanos dry season 2001 = mean number per route Southwestern Llanos dry season 2002; Hog: mean number per route Western Lla- nos wet season 2000 = mean number per route Western Llanos wet season 2001; H 04 : mean number per route Western Llanos dry season 2001 = mean number per route Western Llanos dry season 2002). Differences were considered significant at L* < 0.10 because we were in- terested in large-scale broad patterns. However, we used the Bonferroni Method to control for inflated experi- mentwise type I error rate resulting from simultaneous multiple comparisons (Beal and Khamis 1991). For all analysis, Bonferroni corrected significance for 56 com- parisons was determined at {P < 0.0018). We then combined data from both years to investigate seasonal and regional differences in species numbers, species diversity, community composition, and mean number per route for each species. During the 2-yr study, 50 routes were surveyed (Southwestern Llanos 28 routes, Western Llanos 22 routes) . We surveyed 27 routes during both survey years (wet season 2000 and dry season 2001 ) and an additional 23 routes during the second survey year (wet season 2001 and dry season 2002). To stan- dardize the number of individuals detected on the 27 routes surveyed over 2 -yr, we averaged the number of individuals per survey point across survey years for each species. We then combined this 27-route average with the data from the 23 routes surveyed only during the second survey year to yield 50 total routes. Comhining the data using this method preserved the majority of data collect- ed, accounted for variation in numbers of individuals de- tected for each species on routes replicated over survey years, and preserved data from routes surveyed only dur- ing the second year. We calculated species numbers and Simpson’s Inverse Diversity Index (D = 1/X pi^) for the wet and dry seasons within each region (Hayek and Buzas 1996). We also cal- culated Jaccard’s Coefficient of Community Similarity to estimate the percent overlap between communities in both seasons and regions (Magurran 1988). Again, we used Estimates version 6.0b 1 software (Colwell 2000) to generate species accumulation curves for each regional and seasonal dataset to evaluate the probability that the number of routes surveyed in each region for each sea- son were adequate for documenting all detectable raptor species. To investigate seasonal and regional differences in numbers for each species, we first calculated the mean numbers per route for each region and season. To com- pare seasonal di ff erences in mean numbers per route for the 14 most abundant species, we used paired /-tests to evaluate two hypothesis (Hop mean number per route Southwestern Llanos wet seasons = mean number per route Southwestern Llanos dry seasons; Hog: mean num- ber per route Western Llanos wet seasons = mean num- ber per route Western Llanos dry seasons). For the same 14 species, we also used 2-sample /-tests to evaluate re- gional differences in mean numbers per route (Hop mean number per route Southwestern Llanos wet sea- sons = mean number per route Western Llanos wet sea- sons; Hog: mean number per route Southwestern Llanos dry seasons = mean number per route Western Llanos dry seasons). The 14 rare abundance class species are presented only as present or absent for each region and season. Results General Patterns. We counted 5735 raptors rep- resenting 28 species and 19 genera (Table 1). The four most abundant species, Crested Caracara ( Ca- racara plancus). Yellow-headed Caracara (Milvago chimachima) , Savanna Hawk {Buteogallus meridiona- lis), and Roadside Hawk (Buteo magnirostris) , com- prised 72% of all individuals and were seen on 98- 100% of all routes. Four additional species were classified as common and together comprised 15% of all individuals. These were the Black-collared Hawk {Busarellus nigricollis) , White-tailed Hawk {Buteo albicaudatus) , Snail Kite {Rostrhamus sociabi- lis) , and American Kestrel {Falco sparverius) . Six ad- ditional species were uncommon and represented 10% of all detections. They were the Aplomado Falcon {Falco femoralis) , Great Black Hawk {Buteo- gallus urubitinga), White-tailed Kite {Elanus leucu- rus) , Crane Hawk ( Geranospiza caerulescens) , Laugh- ing Falcon {Herpetotheres cachinnans), and Osprey {Pandion haliaetus). Fourteen additional species comprised the remaining 3%. Species classified as common were detected on 52-84% of all routes, whereas uncommon species were detected on 44- 78% routes. The 14 rare species were detected on December 2005 The Raptor Community of the Lianos 421 Table 1. Species detected during raptor surveys in the Southwestern and Western Llanos of Venezuela during wet seasons in 2000 and 2001 and dry seasons in 2001 and 2002. Species are listed in order of relative abundance based on percent composition. English common names and taxonomy follow Ferguson-Lees and Christie (2001). Relative Abundance Class"' Species Status'^ Total Number Percent Composition Frequency Occurrence (%)<= Very common Creasted Caracara ( Caracara plancus) R 1473 26 100 Yellow-headed Caracara {Milvago chimachima) R 1061 19 100 Savanna Hawk {Buteogallus meridionalis) R 1005 18 98 Roadside Hawk {Buteo magnirostris) R 604 11 98 Common Black-collared Hawk {Busarellus nigticollis) R 269 5 84 Snail Kite {Rostrhamus sodabilis) R 226 4 68 American Kestrel {Falco sparverius) R/NAM 183 3 52 White-tailed Hawk {Buteo albicaudatus) R 155 3 72 Aplomado Falcon {Falco femor alls) R 113 2 78 Uncommon Great Black Hawk {Buteogallus urubitinga) R 110 2 58 White-tailed Kite {Elanus leucurus) R 137 2 70 Crane Hawk {Geranospiza caerulescens) R 74 1 56 Laughing Falcon {Herpetotheres cachinnans) R 70 1 44 Osprey {Pandion haliaetus) NAM 66 1 48 Rare King Vulture ( Sarcoramphus papa) R 41 0.7 24 Harris’s Hawk {Parabuteo unidnctus) R 36 0.5 28 Grey-lined Hawk {Buteo nitidus) R 27 0.5 28 Zone-tailed Hawk {Buteo albonotatus) R 19 0.3 32 Slender-billed Kite {Rostrhamus hamatus) R 19 0.3 16 Bat Falcon {Falco rufigularis) R 16 0.3 24 Plumbeous Kite {Ictinia plumbea) R 9 0.2 14 Hook-billed Kite ( Chondrohierax undnatus) R 5 0.08 10 Long-winged Harrier {Circus buffoni) R 5 0.09 6 Common Black Hawk {Buteogallus anthradnus) R 5 0.09 10 Peregrine Falcon {Falco peregrinus) NAM 5 0.09 8 Short-tailed Hawk {Buteo brachyurus) R 5 0.09 6 Grey-headed Kite {Leptodon cayanensis) R 3 0.05 4 Pearl Kite ( Gampsonyx swainsonii) R 3 0.05 6 ® Relative abundance class: Very common = 10-26% of all individuals detected; Common = 3-5%, Uncommon = 1-2%, Rare = < 1 %. ^ Status: R = nonmigratory population, NAM = North American migratory population, R/NAM = nonmigratory and North American migratory populations. Frequency of occurrence: the percent of routes on which a species was detected. <32% of all routes. There were three species of North American migrants: Osprey, American Kes- trel, and Peregrine Falcon {Falco peregrinus). The Osprey and Peregrine Falcon combined comprised 1.2% of all individuals seen. In contrast, the Amer- ican Kestrel comprised 3%, having both nonmigra- tory and migratory populations (Hilty 2003). At least one of these three species was seen at least once on 82% of routes. The species accumulation curve indicated the number of routes surveyed over the two survey years was adequate for documenting all species de- tectable by roadside point count surveys in the study area (Fig. 2) . Specifically, all 28 species were detected with 34 routes (68% of all routes sur- veyed) . Yearly Comparisons. Among the 14 most com- mon species there were no regional or seasonal differences (P > 0.0018) in mean number per route between survey years (Table 2). Regional and Seasonal Patterns. The greatest number of species, 25, was detected in the South- western Llanos during the wet season and in the Western Llanos during the dry season (Fig. 3) . Al- though species numbers were equal between re- gions, diversity was higher in the Western Llanos 422 Jensen et al. VoL. 39, No. 4 Nutnbar of Routes Surveyed Figure 2. Species accumulation graph of raptor species detected on 50 survey routes in the Southwestern and Western Llanos of Venezuela during the wet seasons in 2000 and 2001 and dry seasons 2001 and 2002. during both seasons. Species accumulation curves indicated the number of routes surveyed during each season was adequate to detect the majority of species for both regions (Fig. 4) . The raptor community (Jaccard’s Coefficient of Community Similarity) in the Southwestern versus Western Llanos differed by 12%, both in the wet and dry seasons. For seasons combined, 25 species were detected in each region, of which 22 species were shared between regions. The seasonal chang- es in community composition within regions (26%) were greater than the regional differences within the wet and dry season (12%). Four of the 14 most common species exhibited regional differences {P< 0.0018) in mean number per route in relation to the wet or dry seasons (Ta- ble 3). The Crested Caracara (P < 0.001) and Os- prey (P < 0.001) were more numerous in the Southwestern Llanos than the Western Llanos dur- ing the wet season. The Black-collared Hawk (P = 0.001) was more numerous in the dry season in the Southwestern Llanos than the Western Llanos, whereas the American Kestrel (P = 0.001) was more numerous in the dry season in the Western Llanos than the Southwestern Llanos. Of the 14 rare species, six were detected only during one season or region (Table 4). The Pere- grine Falcon and Grey-headed Kite were seen only in the dry season, and the Pearl Kite and Slender- billed Kite were seen only in the wet season. The Common Black Hawk was detected only in the Southwestern Llanos in the wet season, and the Plumbeous Kite was detected only in the Western Llanos in the dry season. Discussion General Findings. The savannas of the South- western and Western Llanos of Venezuela are par- ticularly rich in raptors, supporting 52% (32 of 61) of all regularly occurring migrant and resident spe- cies found in Venezuela (Hilty 2003) . Indeed, dur- ing our 2-yr study of the Venezuelan Llanos, these regions included 55% of all hawk species (10 of 18), 70% (7 of 10) of kite species, 67% (4 of 6) of vulture species, and 50% (3 of 6) of the regularly occurring North American migratory species (Hilty 2003). However, although the Llanos sup- ported species assemblages equivalent to all other Venezuelan life zones for most raptors, we did not detect any of the eight eagle species that occur in Venezuela. Except for the three common Cathartid species we did not count, we detected 24 of 27 resident species expected to occur in the Llanos (Hilty 2003) . We did not detect three uncommon forest- dwelling raptors thought to occur in the region: the Collared Forest-Falcon (Micrastur semitorqua- tus). Bicolored Hawk {Accipiter bicolor), and Ornate Hawk Eagle (Spizaetus ornatus). We likely failed to see these species because of the inherent difficulty of detecting forest-dwelling species from roadside surveys (Millsap and LeEranc 1988). Of the three North American migrant species we detected, two occur year-round in the Llanos. The American Kestrel occurs year-round in the Llanos because there are nonmigratory and migratory populations (Hilty 2003) . Although the Osprey population is wholly migratory, the Osprey occurs year-round in the Llanos because first-year birds reaching the Llanos remain for at least 18 mo (Martell et al. 2001, Hilty 2003). North American migratory species not seen dur- ing surveys were the Northern Harrier {Circus cya- neus), Broad-winged Hawk {Buteo platypterus) , Swainson’s Hawk {B. swainsoni), and Merlin {Falco columbarius) . However, one Merlin was detected in the Llanos, but not on survey routes. Historical sightings of the Swainson’s Hawk and Northern Harrier in the Llanos are considered accidental (Hilty 2003), and satellite-tracking of Swainson’s Hawks confirms that their migrations to and from Argentina occur along the central and eastern slopes of the Andes (Euller et al. 1998) . The Broad- winged Hawk is thought to winter in portions of the Llanos west and north of our study area (Hilty 2003). December 2005 The Raptor Community of the Llanos J= U > u in U O •i-j a, rt be 'S "d o "d fl u -4-1 d o ^ o s ^ a gvj V a .a i/i C O tn V V ^ C^-d xi Oi o X X X i> X X 05 CM X q q q X q on rH X rH CM CM i-H iH rH d d d rH d d d d d d d O o +1 +1 + 1 + 1 +1 + l + l + 1 + 1 +1 +1 +1 +1 + 1 X CM CM o TjH o 05 rH CM X X 05 05 q q O' q X X X 05 X X CM X on II »rj d J> CO d d CO d d d rH d d d CM Th X X X X o- O' X CM i> i> X CM Tb X CM rH c/: Q rH o f— s rH +1 rH +1 CO +1 d + 1 d + 1 o d + l d + 1 d + 1 d + 1 d +1 d +1 d +1 d +1 u CM CM l-H CM CM X X rH X X i> X X § q 05 05 q T q q X q CM X hJ irj X d CO d CM d d rH d d d k-] rH W H O) O) on X X X 05 X X O' X X X q m X CM X q X CM rH X o CM rH d lH d d d d 1 — 1 d d d d d d o o +1 + 1 +1 + 1 + 1 + l + 1 + 1 + 1 + 1 + 1 + 1 + 1 o X CM CO on X X CM o X X X O' o X X X on q q X q q q X o X II CM d rH CO d rH CM d d d rH d d CO CM X o iH O' X X rH X X H X I— H X X tH iH O' iH CM o CM o d rH d d d d d d d d d d o o + 1 + 1 + 1 +1 +1 +l +1 +1 +1 o + 1 +1 + 1 o CM UO X o X CM X o X X CM 1> in q X iH iH q CM X q o d d H d d rH d d 00 CM X o X O' rH X X o X X X CM i> o> I — 1 X X X iH CM X rH rH fH CM CM o l-H d rH d d d d d d d d d d d +1 + 1 + 1 + 1 +l + l +1 + 1 +1 + 1 +1 + 1 +1 +1 ,,»— V o X CM o X X 05 CM o iH X 05 X CM o o X 05 X CM q rH Tb O' tH II i> d CO CM d d d d d d rH o X j> X X 05 o J> X o CM CM CO 0 q q rH iq X CM rH CM rH q H rH o rH P rH O (*»S CM + 1 d +1 rH + 1 d + 1 d +l d +l d +1 d + 1 d +1 d +1 o +1 d +1 d +1 d + 1 GNT X CM O) T— 1 -O X CM O' rH 05 X o o »-) CM CM q q q q rt< X q q q q |Z CM in d CM CM d d d d d d d d d s rH M r . ce X o -t^ X X O' 05 -O' 05 TfH 05 X 05 X on 05 X X q o X CM iH iH CM * Xi CM lH d d d rH d d d d d d d d H o + 1 + 1 +1 +1 + l + l + 1 +1 +1 +1 +1 +1 +1 +1 P 0 cr5 X q in O' q q CM H CM O X CM rH d d d CM d d d d d d o o ,»^ + 1 + 1 + 1 +1 + l + l o +1 + 1 o + 1 + 1 + 1 + 1 r> i> t'» X O o X CO X O' CM q q rH O' CO X o X 1-H in CM CM rH Tb H d d d d d tH IZ) M M U bj Pl, V5 U P ^ ^ S i 3 « P u d o o u U d o a a o u d o a a o o d p5 CM Cl O O u C P m cl o e a o o X (D M o u -4-1 u X d X •g X c VD X CM X I 2 O o d II A § 5^ e d a ^ o y u X a 1> c" . . w ^ o ^ C Ij 1) u c rfl X d d X cfl OJ > d o 'C d Oh a 0 u d 2 ^ 423 424 Jensen et al. VoL. 39, No. 4 Figure 3. Raptor species numbers and diversity in the Southwestern and Western Llanos of Venezuela for com- bined years during the wet season (2000 and 2001) and the dry season (2001 and 2002). Number of Routes Surveyed Figure 4. Species accumulation graphs for raptors de- tected in the Southwestern and Western Llanos of Ven- ezuela for combined years during the wet seasons (2000 and 2001) and the dry seasons (2001 and 2002). Overall, the Southwestern and Western regions of the Llanos lacked the eagle diversity character- istic of African savannas (Thiollay 1978). Nonethe- less, these Llanos regions supported numbers of raptor genera (21) similar to those found in the seasonally-flooded savannas and agricultural-forest mosaic habitats of Kidepo Valley National Park of Uganda (22; Thiollay 1978). Consequently, migra- tory species in the Southwestern and Western Lla- nos only comprise a small portion of the raptor community (7%), whereas migratory species ac- count for 33% of the raptor community in Kidepo Valley National Park of Uganda. Table 3. Relative abundance of species and mean number of individuals (± SE) detected per route (mean/route) during raptor surveys for combined years in the wet (2000 and 2001) and the dry season (2001 and 2002) in the Southwestern and Western Llanos of Venezuela. Relative Abundance Class* Southwestern (N = 28) Western {N = 22) Species Status*’ Wet'^ DrV Wet^ DrV' Very common Crested Caracara R 10.21 + 1.37^ 13.80 H- 1.88 2.98 ± 0.88^ 6.00 + 1.62 Yellow-headed Caracara R 6.09 0.70 6.70 + 0.74 5.21 ± 0.97 8.05 H- 1.25 Savanna Hawk R 4.45 - 1 - 0.63 6.34 1.06 3.16 ± 0.76 12.70 3.92 Roadside Hawk R 3.25 0.62 3.59 0.76 5.18 ± 0.81 5.41 0.80 Common Black-collared Hawk R 2.36 0.53 2.73 0.55“ 0.52 ± 0.21 0.66 0.20“ Snail Kite R 3.00 0.94 0.93 0.35 0.73 ± 0.22 0.23 0.09 American Kestrel R/NAM 0.25 + 0.13 0.23 + 0.10“ 2.21 ± 0.76 2.37 0.57“ White-tailed Hawk R 1.41 + 0.24 0.82 + 0.18 0.48 ± 0.24 0.68 0.17 Uncommon Aplomado Falcon R 0.61 + 0.17 0.84 + 0.20 0.57 ± 0.13 0.60 0.16 Great Black Hawk R 0.84 + 0.30 1.21 + 0.37 0.59 ± 0.19 0.43 + 0.18 White-tailed Kite R 0.32 + 0.09 0.39 0.12 1.48 ± 0.40 1.50 0.32 Crane Hawk R 0.30 0.10 0.61 0.14 0.27 ± 0.11 0.55 0.19 Laughing Falcon R 0.23 -h 0.12 0.18 0.13 0.93 ± 0.27 1.09 -4- 0.38 Osprey NAM 0.93 0.2U 0.38 0.14 O'* 0.28 + 0.11 “Relative abundance class: Very common = 10-26% of zill individuals detected; Common = 3-5%, Uncommon = 1-2%. *^ Status: R = resident species, NAM = North American migratory species, R/NAM = resident and North American migratory population. ^ Means denoted by “d” differed (P < 0.0018) between regions for the wet season. Means denoted by “D” differed {P < 0.0018) between regions for the dry season. All other comparisons were not significant (P > 0.0018). December 2005 The Raptor Community of the Li^nos 425 Table 4. Rare species detected during raptor surveys for combined years in the wet (2000 and 2001) and the dry season (2001 and 2002) in the Southwestern and Western Llanos of Venezuela. Relative Southwestern {N = 28) Western {N = 22) Abundance Class'^ Species Status'’ Wet Dry Wet Dry Rare King Vulture R X X X Harris’s Hawk R X X X X Gray-lined Hawk R X X X X Zone-tailed Hawk R X X X X Slender-billed Kite R X X Bat Falcon R X X X X Plumbeous Kite R X Hook-billed Kite R X X X X Long-winged Harrier R X X X Common Black Hawk R X Peregrine Falcon NAM X X Short-tailed Hawk R X X X Grey-headed Kite R X X Pearl Kite R X X “Relative abundance class: Rare = <1% of all individuals detected. ’’ Status: R = nonmigratory populations, NAM = North American migratory population. Regional and Seasonal Patterns. The Western Llanos was characterized by higher levels of raptor diversity. This region underwent a period of exten- sive deforestation prior to 1825, followed by forest regeneration through 1950, and another period of deforestation by 1975 (Veillon 1976). Forest ex- ploitation cycles, coupled with agricultural activity and year-round water sources, have resulted in a dynamic mosaic of forest, savanna, agricultural, pasture, and early successional habitats that likely account for the high raptor diversity in this region. Overall, raptor communities in the Southwest- ern and Western Llanos were similar. However, varying vegetation cover types, large-scale flooding, and the availability of year round water sources during the dry season almost certainly influence the raptor community. For example, Balgooyen (1989) reported the American Kestrel preferred the forest-agriculture mosaic habitats in the Llanos. Our data indicated this pattern was pronounced in the dry season, when American Kestrel numbers increased in the Western Llanos, likely due to the arrival of wintering North American migrants. Fur- thermore, several examples of seasonal influences are apparent in the Southwestern Llanos, where flooding of savanna and gallery forest is extensive in the wet season, but also where wetland com- plexes persist throughout the year. The higher numbers of the Black-collared Hawk in this region during the dry season were likely explained by the presence of year-round wetland complexes. Simi- larly, the higher numbers of Osprey in the South- western Llanos versus Western Llanos during the wet season indicated that Osprey used both regions during the dry season, but first-year birds spent the wet season in the Southwestern Llanos, where flooding was extensive. Therefore, the year-round availability of wetland complexes and the extensive inundation of savanna in the wet season likely ex- plains why these aquatic-dependent species are more abundant in this region. Raptor Distribution. Our results on the relative abundance and distribution of the 14 most com- mon species in our study were consistent with pre- vious findings. However, 4 of 14 less common spe- cies were not expected to occur in the Llanos, or there was little information on their distribution and seasonal occurrence: Common Black Hawk (5 individuals). Pearl Fate (6), Plumbeous Kite (9), and Short-tailed Hawk (5). The Pearl Kite was re- ported as scarce or absent in the Llanos (Hilty 2003), but was detected in greater numbers than many species considered uncommon residents. Al- though the distribution of the Short-tailed Hawk was previously unknown in the Llanos (Hilty 2003), we detected this species in low numbers through- out the study area. Our observation that the Plum- beous Kite was absent during the wet season sug- 426 Jensen et al. VoL. 39, No. 4 gests the Llanos population was similar to those of Central America, Mexico, and Trinidad. All of these raptors migrate southward in August and September and return north to breed in February and March (Ferguson-Lees and Christie 2001). Conservation Implications. The rich raptor com- munity of the Southwestern and Western Llanos is comprised of wetland-dependent and upland-ter- restrial species, both nonmigratory and migratory, of which several appear seasonally nomadic and may depend on both regions as they move in and out in response to the wet and dry seasons. This pattern suggests that, at least in part, the diverse raptor community in the region owes its origins to a combination of a large landscape with a substan- tial seasonal influx of water and the forest-agricul- tural mosaic that creates a temporally and spatially diverse mix of habitats. The Harris’s Hawk {Parabuteo unicinctus ) , Savan- na Hawk, and White-tailed Hawk are now absent from the Cauca Valley in Columbia as a result of landscape homogenization and wetland depletion (Alvarez-Lopez and Kattan 1995). We commonly detected the Savanna Hawk and White-tailed Hawk, and to a lesser extent the Harris’s Hawk, throughout the Southwestern and Western Llanos, which suggests large-scale landscape homogeniza- tion and wetland degradation are not yet occurring in these regions. In contrast, because hawk-eagles may be less tolerant of human alterations on the landscape than many other raptors (Burnham et al. 1994), the Ornate Hawk-Eagle may have been affected by historical deforestation and subsequent lack of suitable habitats in these regions. However, with the exception of the Ornate Hawk-Eagle, the raptor community in the Llanos may represent a community model for Neotropical savanna-forest- agricultural regions. Importantly, additional studies on the raptor community in the Llanos are required to better understand abundance patterns and seasonal fluc- tuations of raptors. Eurthermore, research is need- ed to evaluate local areas where deforestation, in- tensive agriculture, and man-made impoundments are currently expanding, which may be useful for assessing the long-term stability of the current rap- tor community. Finally, comparisons between the Llanos and other Neotropical wetland-savanna complexes, such as the Brazilian Pantanal, will help determine the scope and representation of our findings. Acknowledgments Duck’s Unlimited and the U.S. Department of Agri- culture Forest Service provided the major financial sup- port for this project, along with support from the Hawk Mountain Sanctuary. Special thanks to our field assistants Alexis Araujo, Mireya Barrera, David Peraza, Mariana Es- cobar, Maria Doris Escovar, Andres DeGraf, Sara Seijas, and Graciela Barrera and to the Universidad Nacional Experimental de Los Llanos Occidentales Ezequiel Za- mora (UNELLEZ) in Guanare for their collaboration with the project and the generous use of their facilities. Thanks to Don Taphorn, Jose Angel Anez of AsoMuseo, Luis Altuve, Jose Gregorio Quintero, Gilberto Rios, An- tonio Utrera, Andres Seijas, Franklin Rojas-Suarez of Pro Vita, Alvaro Velasco from Profauna in Caracas, Jose Gregorio Garcia Tenia and Yuri Cedeno from Profauna in San Fernando, Richard Schargel and Heberto Pacheco from BioCentro, Clemencia Rodner and Robin Restall from the Audubon Society in Caracas, and Luis Gonzalez Morales from the Universidad Central de Venezuela. Fi- nally we gratefully thank the following landowners in the Llanos for access to their properties and use of facilities, the Maldonado family and staff members Alexis Aguirre, Jose Ayarzaguena, and Sara Candela of Hato El Frio; Ing. Jesus Pacheco and Enrique Loreto who authorized visits to Hato El Cedral, where staff members Edgar Chiapan- na and Ing. Tulio Aquilera provided assistance; the Bran- ger family at Hato Pinero, where Edgar Useche autho- rized visits; staff at Hato Fernando Corrales; and Don Porfirio Martinez from Mantecal. This is Hawk Mountain Sanctuary contribution to conservation science number 119. Literature Cited Alverez, E., D.H. Ellis, D.G. Smith, AND C.T. Larue 1996. Diurnal raptors in the fragmented forest of Si- erra Imataca, Venezuela. Pages 263-273 in D. Bird, D. Varland, andj. Negro [Eds.], Raptors in human land- scapes. Academic Press, London, England. A1.VAREZ-L0PEZ, H. AND G.H. Kattan. 1995. Notes on the conservation status of resident diurnal raptors of the Middle Cauca Valley, Colombia. Bird Cons. Inti. 5:341- 348. Balgooyen, T.G. 1989. Natural history of the American Kestrel in Venezuela. J. Raptor Res. 23:85-93. Beai., K.G. AND H.J. Khamis. 1991. A problem in statistical analysis: simultaneous inference. Condor 93:1023— 1025. Beissinger, S.R., B.T. Thomas, and S.D. Strahi.. 1988. Vocalizations, food habits, and nesting biology of the Slender-billed Kite with comparisons to the Snail Kite Wilson Bull. 100:604—616. Bibby, C.J., N.J. Collar, M.J. Crosby, M.F. Heath, C.H. Imboden, T.H. Johnson, AJ. Long, A.J. Sattersfield, AND S.J. Thirgood. 1992. Putting biodiversity on the map: priority areas for global conservation. ICBP, Cambridge, England. Bierregaard, R.O., Jr. 1998. Conservation status of birds December 2005 The Raptor Community of the Llanos 427 of prey in the South American tropics. J. Raptor Res. 32:19-27. Bildstein, K.L., W. Schelsky, J. Zalles, and S. Ellis. 1998. Conservation status of tropical raptors./. Raptor Res. 32:3-18. Bulla, L., J. Pacheco, and G. Morales. 1990. Seasonally flooded Neotropical savanna closed by dikes. Pages 177-210 in A.I. Breymeyer [Ed.], Ecosystems of the world: managed grasslands. Regional Studies 17A. El- sevier Science, Amsterdam, Netherlands. Bunn, A.G., W. Klein, and K.L. Bildstein. 1995. Time- of-day effects on the numbers and behavior of non- breeding raptors seen on roadside surveys in eastern Pennsylvania./. Field Ornithol. 66:544—552. Burnham, W.A., D.F. Whitacre, and J.P. Jenny. 1994. The Maya Project: use of raptors as tools for conservation and ecological monitoring of biological diversity. Pag- es 257-264 in B.-U. Meyburg and R.D. Chancellor [Eds.], Raptor conservation today. WWGBP/The Pica Press, Tonbridge, Kent, England. Cole, M.M. 1986. The savannas: biogeography and geo- botany. The savanna woodlands, savanna grasslands and low tree and shrub savannas of northern tropical America. Academic Press, London, England. Colwell, R.K. 2000. Estimates: statistical estimation of species richness and shared species from samples, ver- sion 6. Obi. University of Connecticut, Storrs, CT, U.S.A. Droege, S. 1990. The North American Breeding Bird Survey. Pages 1-4 mJ.R. Sauer and S. Droege [Eds.], Survey designs and statistical methods for the esti- mation of avian population trends. U.S. Fish and Wildlife Service, Biological Report 90. Ellis, D.H., R.L. Glinski, and D.G. Smith. 1990. Raptor road surveys in South America. /. Raptor Res. 24:98- 106. Ferguson-Lees, j. and D.A. Christie. 2001. Raptors of the world. Houghton Mifflin Company, New York, NY U.S.A. Fuller, M.R., S.W. 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. Hayek, L.C. and M.A. Buzas. 1996. Surveying natural populations. Columbia University Press, New York, NY U.S.A. Hilty, S.L. 2003. Birds of Venezuela, 2nd Ed. Princeton University Press. Princeton, NJ U.S.A. Huber, O. and C. Alcaron. 1988. Mapa de vegetacion de Venezuela. Minestro del ambiente y de los Recur- sos Naturales Renovables. DGHA, Division de Vege- tacion, Caracas, Venezuela. Jensen, W.J. 2003. The abundance and distribution of fal- coniformes in the central and western llanos of Ven- ezuela. M.S. thesis. State Univ. of New York, College of Environmental Science and Forestry, Syracuse, NY U.S.A. Kirk, D.A. and J.P. Currall. 1994. Habitat associations of migrant and resident vultures in central Venezuela /. Avian Biol. 25:327—337. Mader, W.J. 1981. Notes on nesting raptors in the Llanos Condor 83:48-51. . 1982. Ecology and breeding habits of the Savan- na Hawk in the Llanos of Venezuela. Condor 84:261- 271. Magurran, A.E. 1988. Ecological diversity and its mea- surement. Princeton University Press, Princeton, NJ U.S.A. Manosa, S., E. Mateos, and V. Pedrocchi. 2003. Abun- dance of soaring raptors in the Brazilian Atlantic rain- forest. /. Raptor Res. 37:19-30. and V. Pedrocchi. 1997. A raptor survey in the Brazilian Atlantic rainforest./. Raptor Res. 31:203-207. Martell, M.S., C.J. Henny, RE. Nye, and MJ. Solensky 2001. Fall migration routes, timing, and wintering sites of North American Ospreys as determined by sat- ellite telemetry. Condor 103:715—724. Millsap, B.A. and M.N. LeFranc, Jr. 1988. Road transect counts for raptors, how reliable are they? /. Raptor Res 22:8-16. Mittermeier, R.A and C.G. Mittermeier. 1997. Venezue- la. Pages 448 — 467 in Megadiversity-Earth’s biologi- cally wealthiest nations. Conservation International CEMEX de SV, The University of Chicago Press, Chi- cago, IL U.S.A. , , P.R. Gil, J. Pilgrim, G. Fonesca, T. Brooks, and W.R. Konstant. 2003. The llanos. Pages 265—270 in Wilderness: earth’s last wild places. Con- servation International CEMEX de SV, The University of Chicago Press, Chicago, IL U.S.A. Myers, N., R.A. Mittermeier, C.G Mittermeier, G.A.B. DE Fonesca, and J. Kent. 2000. Biodiversity hotspots for conservation priorities. Nature 403:853-858. Seavy, N.E. and C.K. Apodaca. 2002. Raptor abundance and habitat use in a highly-disturbed-forest landscape in Western Uganda. / Raptor Res. 36:51-57. Silva, J.F and A. Moreno. 1993. Land use in Venezuela Pages 239—257 mM.D. Young and O.T. Solbrig [Eds ], The world’s savannas; economic driving forces, eco- logical constraints, and policy options for sustainable land use. UNESCO, Paris, France. SoRLEY, C.S. AND D.E. Andersen. 1994. Raptor abun- dance in south-central Kenya in relation to land use patterns. Afr. J. Ecol. 32:30-38. Thiollay, J.-M. 1978. Population structure and seasonal fluctuations of the falconiformes in Uganda National Parks. East Afr. Wildl. J. 16:145-151. . 1984. Raptor community structure of a primary rainforest in French Guiana and effect of human hunting pressure. Raptor Res. 18:117-122. . 1989. Area requirements for the conservation of 428 Jensen et al. VoL. 39, No. 4 rain forest raptors and game birds in French Guiana. Conserv. Biol. 2:128—137. Troth, R.G. 1979. Vegetational types on a ranch in the central llanos of Venezuela. Pages 17-30 mJ.F. Eisen- berg [Ed.], Vertebrate ecology in the neotropics. Smithsonian Institute Press, Washington, DC U.S.A. Veillon, J.P. 1976. Las Deforestaciones en la Region de los Llanos Occidentales de Venezuela, desde 1950 has- ta 1975. Pages 67-112 mL.S. Hamilton, J. Steyermark, J.P. Veillon, and E. Mondolfi, [Eds.], Conservacion de los Bosques Humedos de Venezuela. Sierra Club y Consejo del Bienestar Rural, Caracas, Venezuela. Whitacre, D.F. and C.W. Turley. 1990. Further compar- isons of tropical rainforest census techniques. Pages 71-92 mW.A. Burnham, D.F. Whitacre, and J.P. Jenny [Eds.], Maya Project; use of raptors as environmental indices for design and management of protected ar- eas and for building local capacity for conservation in Latin America. The Peregrine Fund, Boise, ID U.S.A. WOFFINDEN, N.D. AND J.R. MuRPHY. 1977. A roadside rap- tor census in the eastern Great Basin, 1973-1974. Rap- tor Res. 11:62-66. Received 27 May 2004; accepted 15 August 2005 J. Raptor Res. 39(4):429-438 © 2005 The Raptor Research Foundation, Inc. A COMPARISON OF BREEDING SEASON FOOD HABITS OF BURROWING OWLS NESTING IN AGRICULTURAL AND NONAGRICULTURAL HABITAT IN IDAHO Colleen E. Moulton,^ Ryan S. Brady, and James R. Belthoff Department of Biology and Raptor Research Center, Boise State University, Boise, ID 83725 U.S.A. Abstract. Through analysis of regurgitated pellets and prey remains collected at nests between 2001- 02, we characterized diet composition of western Burrowing Owls {Athene cunicularia hypugaea) in the Snake River Birds of Prey National Conservation Area (NCA) of southwestern Idaho. We hypothesized that diet differs between owls nesting in agricultural and nonagricultural habitat, because at least one important prey species, montane voles (Microtus montanus), occurs predominately in the former. From 859 pellets, we identified 7402 prey items representing 23 species, and identified 403 prey remains of 19 species. Invertebrates dominated the diet in numbers of prey, whereas rodents contributed the great- est biomass. Montane voles, which were not present in pellets in nonagricultural areas, represented the greatest percent biomass of pellets in agricultural areas. Invertebrates (predominately Gryllidae) also were more abundant in diets of owls nesting in agricultural habitat. Pellets of owls nesting in agricultural areas had greater species richness, whereas pellets from nonagricultural areas had greater species even- ness and broader food-niche breadths. Finally, we estimated food-niche breadth of Burrowing Owls in the NCA to be broader than previously reported. Key Words: Burrowing Owl; Athene cunicularia; agriculture; food habits; food-niche; Idaho. UNA COMPARACION DE LOS hAbITOS ALIMENTICIOS DE INDIVIDUOS NIDIFICANTES DE ATHENE CUNICULARIA EN AMBIENTES AGRICOLAS Y NO AGRICOLAS EN IDAHO Resumen. — A traves del analisis de egagropilas y de restos de presas recolectados en nidos en 2001 y 2002, caracterizamos la composicion de la dieta de Athene cunicularia hypugaea en el Area Nacional de Conservacion de Aves de Presa Snake River, sudoeste de Idaho. Nos planteamos la hipotesis de que la dieta difiere entre las lechuzas que nidifican en ambientes agricolas y no agricolas, debido a que al menos una de las especies de presa importantes, Microtus montarius, se encuentra predominantemente en las areas agricolas. De un total de 859 egagropilas, identificamos 7402 items de presas correspon- dientes a 23 especies, e identificamos 403 restos de presas provenientes de 19 especies. Los invertebrados dominaron la dieta en terminos del numero de presas, mientras que los roedores representaron la mayor biomasa. Microtus montanus no estuvo presente en las egagropilas de las areas no agricolas y represento el mayor porcentaje de biomasa en las egagropilas de las areas agricolas. Los invertebrados (predominantemente Gryllidae) tambien fueron abundantes en las dietas de las lechuzas que nidifica- ron en los ambientes agricolas. Las egagropilas de las lechuzas que nidificaron en las areas agricolas presentaron mayor riqueza de especies, mientras que las provenientes de las areas no agricolas presen- taron mayor equidad y nichos alimenticios mas amplios. Finalmente, estimamos que el nicho alimenticio de A. c. hypugaea en el area silvestre de conservacion estudiada es mas amplio de lo que habia sido informado previamente. [Traduccion del equipo editorial] Agricultural practices historically have provided many different types of wildlife habitat, including shelterbelts, hedgerows and fencerows, cultivated fields, and fields in rotation. Although some spe- cies nest, seek cover, and forage in these habitats, 1 Present address: Wildlife Bureau, Idaho Department of Fish and Game, P.O. Box 25, Boise, ID 83707 U.S.A.; email address: cmoulton@idfg.idaho.gov many wildlife populations have declined signifi- cantly in areas of agricultural conversion (Carlson 1985, Murphy 2003). In fact, there is mounting ev- idence that converting natural landscapes into ag- ricultural use can affect a wide array of wildlife populations through erosion, exposure to herbi- cides and pesticides, and destruction of nesting and cover habitat (Carlson 1985, Jahn and Schenck 1991, Gervais et al. 2000). These effects may be 429 430 Moulton et al. VoL. 39, No. 4 amplified by the shift from small-scale farming practices to large-scale monoculture farming seen throughout the United States and Canada (Peter- john 2003). Western Burrowing Owls {Athene cunicularia hypugaea) are listed as Endangered in Canada and several western U.S. states, and their populations are declining in many areas (e.g., James and Espie 1997, Clayton and Schmutz 1999, Klute et al. 2003). These owls suffer deleterious effects from agricultural practices (James and Fox 1987, Haug et al. 1993, Bellocq 1997, Gervais et al. 2000) and, in Canada, often avoid agricultural fields (Haug and Oliphant 1990, Clayton and Schmutz 1999). However, throughout some portions of their west- ern U.S. range. Burrowing Owls associate with ag- riculture (Rich 1986, DeSante et al. 2004, Moulton et al., in press), and they are the only raptor spe- cies that shows significant affinity for agriculture in southern Idaho (Leptich 1994). Rich (1986) sug- gested that proximity to montane voles {Microtus montanus) in farmlands could explain some of this habitat selection. Moulton et al. (in press) con- firmed that owls did not nest in agricultural areas because of decreased nest predation or increased availability of nesting sites but noted that prey con- sumption was greater in agricultural areas. If Burrowing Owl nesting distributions can be affected by prey, as Rich (1986) and Moulton et al. (in press) hypothesize, then diet composition may differ for owls occupying agricultural and nonag- ricultural areas. Thus, the objective of our study was to examine breeding season food habits of Burrowing Owls in the Snake River Birds of Prey National Conservation Area (NCA), where Burrow- ing Owls inhabit both agricultural and nonagricul- tural areas. Specifically, we tested the hypotheses that (1) diets of owls in agricultural areas contain more montane voles than those in nonagricultural habitats and (2) because of influences of agricul- tural practices, diet diversity and food-niche breadths differ. We predicted that Burrowing Owls nesting in agricultural habitats would have greater prey diversity and broader food-niche breadths than owls nesting in nonagricultural habitats. Fi- nally, we compared our food-niche breadth esti- mates with those of a previous study (Marti et al. 1993) on raptor food habits in the NCA. Methods We studied Burrowing Owls nesting within and near the NCA in southwestern Idaho during 2001-02. This area was once representative of a typical shrub-steppe community dominated by large expanses of big sage- brush {Artemesia tridentata xvyomingensis; Hironaka et al. 1983) and other shrubs, and scattered perennial bunch- grasses. However, disturbances, such as range fires, mili- tary training, grazing, and off-road vehicle use, have helped convert much of the area to exotic annual grass- lands dominated by cheatgrass {Bromus tectorum), tumble mustard {Sisymbrium altissimum), and other non-native species (Hironaka et al. 1983). Surrounding areas also contained scattered residential homes, paved and dirt roads, a military training area, and public lands managed by the Bureau of Land Management. Cattle and sheep graze much of the area, especially during winter. Irrigat- ed agricultural fields (primarily alfalfa, sugar beets, and mint) constituted <5% of the NCA and were located pri- marily along its margins (USDI 1996). For the purpose of this study, we considered Burrowing Owl nests that were within 1 km of an irrigated agricultural field, to be in “agricultural” habitat (hereafter agricultural nests). Agricultural nests were located in the natural vegetation surrounding agriculture fields rather than in the irrigat- ed portions where crops grew. However, adult owls fre- quently hunted within these fields and perched on fence posts adjacent to them (Moulton et al. in press). “Non- agricultural” habitat was the term we used to categorize nests that were greater than 3 km from irrigated fields (hereafter nonagricultural nests). Because this distance exceeded the typical foraging range of Burrowing Owls (Haug and Oliphant 1990, Rosenberg and Haley 2004), we are almost certain that owls from nonagricultural nests were not collecting prey from or near irrigated fields. Nonagricultural areas were generally disturbed shrublands and grasslands much like that in the agricul- tural areas, but there were no crops or irrigation nearby Diet Composition. Regurgitated pellets are reliable in- dicators of the diet of Burrowing Owls (Marti 1974), al- though amphibians and reptiles can be underrepresent- ed in pellets (Thomsen 1971, Haug 1985). Similarly, prey remains alone do not provide reliable information re- garding overall diet composition, as many prey items con- sumed by Burrowing Owls are too small to cache (such as small insects) . But, remains provide better information than pellets concerning amphibians and reptiles in the diet. Therefore, to determine diet composition, we both documented prey remains at nests and collected and an- alyzed regurgitated pellets. Pellet Collection and Analysis. We collected regurgitat- ed pellets from tunnel entrances, perches, and nearby mounds within 20 m of nest burrows every 3-10 d from hatching through 25 d post-hatch (May-June) . For nests at which we collected more than 20 pellets (29 of 51 nests; 22 agricultural, 7 nonagricultural), we analyzed a random sample of 20 pellets per nest. For all other nests, we analyzed all collected pellets (11.2 — 1.0 [SE] pellets per nest, range = 4-19). We analyzed and quantified remains of each pellet us- ing standard procedures (Marti 1987) and by comparing prey species to a museum collection at Boise State Uni- versity. Skulls, jaws, dentition patterns, head capsules, pronota, elytra, legs, scales and other distinguishing body parts helped identify prey. Prey Remains. Owls in this study nested in artificial burrows deployed for other studies (Smith and Belthoff 2001, Belthoff and Smith 2003, Brady 2004, Moulton et al. in press), which provided access to nest chambers. December 2005 Burrowing Owl Food Habits in Idaho 431 where we found most prey remains. We could therefore document cached and uneaten prey remains at all oc- cupied nests and adjacent satellite burrows (non-nest burrows used by owls for roosting, cover, and caching prey) . We quantified prey remains each time we excavat- ed an artificial burrow (2-5 visits per nest) between hatching and 25 d post-hatch. Biomass Estimation. We determined biomass of rep- resentative mammalian, avian, and amphibian prey using Smith and Murphy (1973) and Steenhof (1983). Biomass of invertebrate prey species was determined using our own estimates obtained from captured live specimens (Moulton 2003) and values reported in Smith and Mur- phy (1973) and Olenick (1990). Statistical Analysis. Because we obtained prey remains from only two nonagricultural nests, we did not include data from prey remains in diversity calculations described below or in statistical comparisons; instead, pellet data provided all information used for these calculations and comparisons. We determined each prey type as a percent of total prey items per nest (percent number) and per- cent biomass per nest. We determined food-niche breadth for agricultural and nonagricultural nests by calculating the reciprocal of Simpson’s index (Simpson 1949). We calculated dietary evenness using the Alatalo (1981) modification of Hill’s (1973) index: F = - l)/(Wi - 1). To determine if differences in diets existed between owls nesting in each habitat, we compared percent num- ber and percent biomass of each prey taxa (vertebrates) , class (invertebrates), or order (invertebrates) per nest us- ing Wilcoxon’s ranked sums tests (Zar 1999). If there were differences between habitats in taxa/ order of prey, we then compared species (vertebrates) or families (in- vertebrates) of that taxa/order. Because we made multi- ple comparisons of prey categories, we adjusted alpha lev- els using sequential Bonferroni corrections (Rice 1989). To determine if diet diversity differed between agri- cultural and nonagricultural nests, we compared food- niche breadth (Simpson’s index), species richness (num- ber of species in the diet), and dietary evenness (Alatalo’s index) using Wilcoxon’s ranked sums tests. Statistical analyses were performed using JMPIN V.5 (SAS Institute, Inc., Cary, NC), and evaluated at an alpha level of 0.05 unless otherwise noted. Throughout, we present means with their standard errors. Results Pellet Remains. We analyzed 602 regurgitated pellets from 34 agricultural nests and 257 pellets from 19 nonagricultural nests. From these, we identified 7402 prey items representing 23 differ- ent prey species. Overall pellet composition. Invertebrates were the most frequent prey in pellets, representing 93% of prey items; however, they represented only 23% of biomass (Table 1). Conversely, vertebrates (ro- dents, birds, and herpefofauna) comprised 7% of prey items, but 77% of biomass. Coleopterans (beetles) and Orthopterans (crick- ets, grasshoppers) were the most common inver- tebrates in pellets, constituting 47% and 32% of total prey, respectively (Table 1 ) . Of Coleopterans, ground beetles (Carabidae) and darkling beetles (Tenebrionidae) were most common (33% and 22% of Coleopteran prey items, respectively). Or- thopteran prey remains were predominately Grylli- dae (crickets), which constituted 73% of Orthop- teran prey items. Rodents were the most common vertebrates in pellets and represented 97% of vertebrates detect- ed and 73% of overall prey biomass (Table 1). Pocket mice {Perognathus parvus) and deer mice (Peromyscus maniculatus) were the most abundant rodents (37% and 25%, respectively), but montane voles represented the greatest biomass (18%). Habitat variation. Invertebrates were the most fre- quent prey in pellets for both agricultural and non- agricultural nests, representing 95% and 90% of total prey items, respectively (Table 2). Vertebrate prey (mostly rodents) represented the greatest per- cent biomass in both agricultural (76%) and non- agricultural (79%) nests. Coleopterans were the most common inverte- brates in both habitats (Table 2). However, Arach- nids contributed the greatest biomass (52%) of invertebrates in nonagricultural nests, and Orthop- terans contributed the greatest biomass (52%) of invertebrates in agricultural nests. Of rodent spe- cies found in pellets, deer mice and pocket mice were most common in agricultural and nonagri- cultural nests, respectively. Pocket mice also con- tributed the greatest biomass of rodents at nonag- ricultural nests, but montane voles contributed the greatest biomass of rodents at agricultural nests. Only owls at agricultural nests preyed on montane voles (Table 2). Agricultural and nonagricultural nests did not differ in percent biomass of vertebrates or inver- tebrates (Table 3) . However, agricultural nests had a greater percent number of invertebrates, and nonagricultural nests had a greater percent num- ber of vertebrates. Pellets from agricultural nests had greater percent number and percent biomass of montane voles (Table 4). Nonagricultural nests had greater percent number and biomass of pock- et mice (Table 4) . Among invertebrates. Arachnids and Orthopterans differed between habitats (Table 3). Solpugida (windscorpions) and Acrididae oc- curred in greater percent number and biomass in pellets of nonagricultural nests, while Gryllidae oc- curred in greater number and biomass in pellets at agricultural nests (Table 5). For all nests combined, food-niche breadth was 432 Moulton et al. VoL. 39, No. 4 Table 1. Mean (±SE) percent number and percent biomass per nest of prey items detected in pellets collected at 53 Burrowing Owl nests in southwestern Idaho, 2001-02. Prey Category Percent Number Percent Biomass Mammals 6.7 “h 0.7 72.9 — b 2.5 Spermophilus mollis 0.2 0.1 10.2 -1- 3.4 Thomomys townsendii 0.2 0.1 12.4 3.5 Perognathus parvus 2.5 + 0.6 12.5 + 2.5 Dipodomys ordii 0.6 -h 0.2 9.4 + 2.4 Reithrodontomus megalotis 0.1 — K 0.1 0.3 + 0.1 Peromyscus maniculatus 1.7 “b 0.4 11.1 -b 2.5 Mus musculus 0.2 0.1 1.2 “b 0.5 Microtus montanus 0.9 0.2 13.2 -b 2.8 Rodent — unidentified^ 0.4 “h 0.1 2.7 -b 0.7 Birds — unidentified^ 0.2 H- 0.1 2.2 0.8 Reptiles and Amphibians'^ <0.1 -h 0.1 1.9 -b 1.2 Arachnids 13.8 -h 1.9 6.3 -b 1.0 Scorpionida 5.8 -h 1.0 3.5 -b 0.9 Solpugida 8.0 H- 1.5 2.8 -b 0.5 Orthopterans 31.6 -b 3.2 9.6 -b 1.6 Acrididae 2.9 -b 0.6 0.7 fr 0.1 Gryllidae 23.2 -b 3.5 7.7 -b 1.7 Unknown Orthoptera 5.5 -b 0.9 1.2 -b 0.2 Dermapterans (Forficulidae) 0.4 -b 0.2 0.2 H- 0.1 Homopterans (Cicadidae) 0.1 -b 0.1 0.1 0.1 Coleopterans 47.0 -b 2.6 6.8 0.7 Carabidae 15.7 2.2 1.9 -b 0.3 Scarabidae 7.5 1.3 1.1 H- 0.2 Silphidae 8.0 1.4 1.1 H- 0.2 Tenebrionidae 10.6 -b 1.9 2.3 -H 0.4 Coleoptera — unidentified 5.3 + 1.2 0.5 + 0.1 Total vertebrates 6.9 H- 0.7 77.0 2.1 Total invertebrates 93.1 + 0.7 23.0 -b 2.1 ^ Mouse species: likely P. parvus, R. megalotis, P. maniculatus, or M. musculus. ^ Likely Eremophila alpestris or Sturnella neglecta. Includes Bufo woodhousei, Phrynosoma platyrhinos, and unknown snake species. 4.22 ± 0.22 {N = 53). Nonagricultural {N = 19) nests had greater species evenness than agricultur- al {N = 34) nests (0.76 ± 0.03 versus 0.60 ± 0.02; Z = 3.89, P < 0.001) and broader food-niche breadth (5.21 ± 0.33 versus 3.67 ± 0.25; Z = 3.24, P — 0.001). However, agricultural nests had higher species richness (11.82 ± 0.40 versus 9.79 ± 0.54; Z = -2.69, P = 0.007). Prey Remains. We recorded cached and other uneaten prey remains at 43 nests {N = 41 agricul- tural, N— 2 nonagricultural) and documented 403 prey items representing 19 species (Table 6). Be- cause we had so few nonagricultural nests, we made no comparisons between habitats and pooled data from all nests for descriptions of prey remains. Although common in pellets, invertebrate prey remains were uncommon in nest burrows {N = 50 individual invertebrate prey items) . The majority of prey remains in both percent number (87.6%) and percent biomass (99.7%) were vertebrates, most of which were rodents. Of rodent species, montane voles were most common by number (36%), and pocket gophers represented the greatest biomass (50%). Although rare in pellets, we occasionally found herpetofauna (N — 38) and birds {N = 18) cached in burrows. Woodhouse’s toads {Bufo woodhousei) were the most common (92%) herpetofauna in nest burrows. All toads were in nests adjacent to agricultural fields. Burrowing Owl nestlings were the most common (50%) cached avian prey item December 2005 Burrowing Owl Food Habits in Idaho 433 Table 2. Mean (±SE) percent number and percent biomass per nest of prey items detected in pellets of Burrowing Owls nesting in agricultural (N = 34) and nonagricultural (N = 19) habitats of southwestern Idaho, 2001-02. Agricultural Nonagricultural Prey Percent No. Percent Biomass Percent No. Percent Biomass Mammals 4.9 0.8 70.7 -h 3.1 10.1 ± 1.1 76.9 ± 4.1 Spermophilus mollis 0.1 -h 0.1 5.3 + 4.1 0.5 ± 0.2 0.2 ± 0.1 Thomomys townsendii 0.4 -H 0.1 1.9 4.1 — Perognathus parvus 0.7 0.6 4.9 ± 2.6 5.6 ± 0.8 26.0 ± 3.4 Dipodomys ordii 0.4 ± 0.2 4.9 2.8 1.2 ± 0.3 17.5 ± 3.8 Peromyscus maniculatus 1.4 + 0.5 10.5 3.1 2.1 ± 0.7 12.3 ± 4.2 Mus musculus 0.2 0.1 1.9 + 0.6 — — Microtus montanus 1.4 -h 0.3 20.6 + 3.1 — — Birds — unidentified^ 0.1 H- 0.1 2.3 + 1.0 0.3 ± 0.1 2.0 ± 1.3 Reptiles and Amphibians*’ 0.1 -+- 0.1 2.8 + 1.5 0.1 ± 0.1 0.3 ± 2.0 Arachnida 6.7 1.7 3.7 + 1.1 26.5 ± 2.3 10.9 ± 1.4 Scorpionida 3.3 -h 1.1 2.0 1.0 10.3 ± 1.5 6.0 ± 1.4 Solpugida 3.4 H- 1.6 1.7 ± 0.6 16.2 ± 2.1 4.9 ± 0.8 Orthoptera 40.0 3.5 12.7 ± 1.9 16.7 ± 4.7 4.0 ± 2.6 Acrididae 1.8 -t- 0.8 0.5 H- 0.2 4.9 ± 1.0 1.1 ± 0.2 Gryllidae 34.8 ± 3.5 11.4 -h 1.9 2.3 ± 4.7 1.0 ± 2.5 Coleoptera 47.6 ± 3.2 7.5 0.8 46.1 ± 4.3 5.7 ± 0.1 Carabidae 21.4 -t- 2.4 2.7 0.4 5.5 ± 3.2 0.7 ± 0.5 Scarabidae 5.6 -h 1.5 1.0 0.3 10.9 ± 2.0 1.3 ± 0.4 Silphidae 5.1 1.7 0.9 + 0.3 13.2 ± 2.3 ' 1.4 ± 0.4 Tenebrionidae 11.1 -h 2.3 2.5 H- 0.6 9.6 ± 3.1 1.9 ± 0.7 Total vertebrates 5.0 + 0.8 75.8 + 2.6 10.3 ± 1.1 79.1 ± 3.5 Total invertebrates 95.0 + 0.8 24.2 2.6 89.7 ± 1.1 20.9 ± 3.5 “ Likely Eremophila alpestris or Sturnella neglecta. Includes Bufo woodhousei, Phrynosoma platyrhinos, and unknown snake species. we found. These Burrowing Owl nestlings all were individuals from nests other than the nest in which we found them. Whether they wandered into the nest on their own and subsequently starved or were killed or were taken directly from their nest is un- known. We suspect that adults tending nearby nests preyed upon these nestlings because they frequent- ly were too young to have wandered into nests oth- er than their own. Discussion The NGA supports one of the highest densities of breeding raptors in the world (Marti et al. 1993), and many previous studies have examined food habits of nesting raptors there (e.g., Marks and Marks 1981, Marks and Doremus 1988, Marti 1988, Steenhof and Kochert 1988). However, die- tary habits and trophic relationships of Burrowing Owls remain the least well-understood of raptors breeding in the NCA (Marti pers. comm.). Thus, our study filled an important knowledge gap in raptor ecology within the NCA. Our study found: (1) no one species dominated the vertebrate com- ponent of Burrowing Owl diets, unlike owls in oth- er regions; (2) diets differed by habitat, most no- tably that montane voles and crickets were important prey for agricultural nests, but they were not part of the diet for nonagricultural nests; and (3) the food-niche breadth of Burrowing Owls in the NCA is broader than previously estimated. Burrowing Owl Diet in the NCA. Burrowing Owls are considered opportunistic predators (Gleason and Craig 1979, Green et al. 1993, Haug et al. 1993), and the wide variety of prey owls con- sumed in our study area is consistent with this no- tion. Similar to studies in Colorado (Marti 1974), Saskatchewan (Haug 1985), and the Idaho Nation- al Engineering Laboratory (INEEL) in Idaho (Gleason and Craig 1979), invertebrates represent- ed approximately 90-95% of prey items in regur- gitated pellets, but they constituted only 20-30% of biomass of prey. In contrast, Olenick (1990), in 434 Moulton et al. VoL. 39, No. 4 Table 3. Mean (±SE) percent number and percent biomass per nest of vertebrate (taxa) and invertebrate (class/ order) prey detected in pellets of Burrowing Owls nesting in agricultural {N = 34) and nonagricultural {N = 19) habitats of southwestern Idaho, 2001-02. Habitat Prey Category Agricultural Nonagricultural Z® P-value Percent Number Mammal 4.9 0.8 10.1 -h 1.1 3.01 0.003* Bird 0.1 0.1 0.3 -h 0.1 0.22 0.823 Reptile and Amphibian 0.1 -h 0.1 0.1 0.1 -0.38 0.701 Arachnid 6.7 ± 1.7 26.5 2.3 4.98 <0.001* Orthopteran 40.0 3.5 16.7 4.7 -3.61 <0.001* Goleopteran 47.6 ± 3.2 46.1 4.3 -0.29 0.774 Total vertebrates 5.0 0.8 10.3 + 1.1 3.05 0.002* Total invertebrates 95.0 + 0.8 89.7 1.1 -3.05 0.002* Percent Biomass Mammal 70.7 + 3.1 76.9 + 4.1 1.03 0.303 Bird 2.3 ± 1.0 2.0 1.3 0.02 0.988 Reptile and Amphibian 2.8 1.5 0.3 2.0 -0.51 0.613 Arachnid 3.7 1.1 10.9 1.4 4.12 <0.001* Orthopteran 12.7 1.9 4.0 -1- 2.6 -3.24 0.001* Goleopteran 7.5 ± 0.8 5.7 0.1 -0.99 0.321 Total vertebrates 75.8 -1- 2.6 79.1 -K 3.5 0.96 0.340 Total invertebrates 24.2 ± 2.6 20.9 3.5 -0.96 0.340 ® Data were compared using Wilcoxon’s ranked sums tests. * Significant based on sequential Bonferroni corrections adjusted from an original alpha level of 0.05 for a total of 16 comparisons Table 4. Mean (±SE) percent number and percent biomass per nest of rodent species detected in pellets of Bur- rowing Owls nesting in agricultural {N = 34) and nonagricultural {N = 19) habitats of southwestern Idaho, 2001-02. Habitat Prey Species Agricultural Nonagricultural P-value Percent Number Spermophilus mollis 0.1 0.1 0.5 0.2 1.83 0.067 Thomomys townsendii 0.4 0.1 0.0 -h 0.1 -2.72 0.007 Perognathus parvus 0.7 -+- 0.6 5.6 0.8 4.23 <0.001* Dipodomys ordii 0.4 -+- 0.2 1.2 -h 0.3 1.67 0.095 Peromyscus maniculatus 1.4 ± 0.5 2.1 -1- 0.7 -1.43 0.153 Mus musculus 0.2 0.1 0.0 0.1 -2.25 0.025 Microtus montanus 1.4 0.3 0.0 0.4 -4.32 <0.001* Percent Biomass Spermophilus mollis 5.3 ± 4.1 0.2 -h 0.1 1.86 0.063 Thomomys townsendii 1.9 H- 4.1 0.0 -1- 5.4 -2.72 0.007 Perognathus parvus 4.9 2.6 26.0 3.4 4.00 <0.001* Dipodomys ordii 4.9 2.8 17.5 3.8 1.79 0.073 Peromyscus maniculatus 10.5 3.1 12.3 4.2 -1.05 0.294 Mus musculus 1.9 0.6 0.0 ± 0.8 -2.25 0.025 Microtus montanus 20.6 3.1 0.0 4.2 -4.32 <0.001* Data were compared using Wilcoxon’s ranked sums tests. * Significant based on sequential Bonferroni corrections adjusted from an original alpha level of 0,05 for a total of 14 comparisons December 2005 Burrowing Owl Food Habits in Idaho 435 Table 5. Mean (±SE) percent number and percent biomass per nest of Arachnid orders and Orthopteran families detected in pellets of Burrowing Owls nesting in agricultural {N = 34) and nonagricultural {N = 19) habitats of southwestern Idaho, 2001-02. Prey Order/Family Habitat Agricultural Non agricultural P-VALUE Percent Number Arachnida Scorpiones 3.3 ± 1.1 10.3 ± 1.5 2.22 0.026 Solpugida 3.4 ± 1.6 16.2 ± 2.1 4.04 <0.001* Orthoptera Acrididae 1.8 ± 0.8 4.9 ± 1.0 2.81 0.005* Gryllidae 34.8 ± 3.5 2.3 ± 4.7 -5.43 <0.001* Percent Biomass Arachnida Scorpiones 2.0 ± 1.0 6.0 ± 1.4 1.72 0.086 Solpugida 1.7 ± 0.6 4.9 ± 0.8 3.68 <0.001* Orthoptera Acrididae 0.5 ± 0.2 1.1 ± 0.2 2.38 0.017* Gryllidae 11.4 ± 1.9 1.0 ± 2.5 -5.26 <0.001* Data were compared using Wilcoxon’s ranked sums tests. * Significant based on sequential Bonferroni corrections adjusted from an original alpha level of 0.05 for a total of four comparisons each for Arachnida and Orthoptera. southeastern Idaho, reported that invertebrates represent only 60% of the number of prey items and less than 3% of the biomass, and owls in the Imperial Valley, California, feed almost exclusively on invertebrates (York et al. 2002). Although invertebrates generally constitute a large percentage of prey Burrowing Owls consume, the orders and families that are most common in the diet vary among regions. For example, Cole- opterans were the most abundant invertebrate spe- cies in our study, as well as in Colorado (Marti 1974), Washington (Green et al. 1993), and Oregon (Green et al. 1993), whereas Jerusalem crickets (Stenopelmatus spp.) were the most impor- tant invertebrate prey species, in terms of biomass, for Burrowing Owls in Oregon (Green et al. 1993), California (Thomsen 1971), and southeastern Ida- ho (Gleason and Craig 1979). Vertebrates accounted for most of the biomass in our study, but no one vertebrate species domi- nated the diet. Percent biomass of montane voles (17%), pocket mice (16%), pocket gophers (16%), and deer mice (14%) were similar. In contrast, Mi- crotus sp. were the predominant vertebrate prey item in Montana (Holt et al. 2001) and represent- ed 80% of biomass in owl diets in southeastern Ida- ho (Olenick 1990), and pocket mice dominated rodent prey in Oregon (97%; Green 1983). This lack of a dominant vertebrate prey may indicate a diverse prey base in our study area (Moulton et al. in press). Agricultural versus Nonagricultural Nests. Com- parisons of pellet remains from Burrowing Owl nests in agricultural and nonagricultural areas re- vealed different prey composition, species rich- ness, species evenness, and food-niche breadth. Al- though both habitats had similar biomass of vertebrates, nonagricultural areas had greater numbers of rodent prey. In contrast, owls nesting adjacent to agricultural fields in southeastern Ida- ho had a higher proportion of rodents in their diet than those nesting in more natural areas (Gleason 1978). Agricultural nests had a higher proportion of invertebrates than nonagricultural nests, which resulted from the high numbers of crickets present in pellets from agricultural nests. Crickets were rare in pellets of owls nesting in nonagricultural habitats. Moulton et al. (in press) reported greater prey consumption by Burrowing Owls nesting near agricultural fields in the NCA; this difference pri- marily resulted from greater invertebrate prey in agricultural habitats. While some have suggested 436 Moulton et ai.. VoL. 39, No. 4 Table 6. Percent number, percent biomass, and total number of cached and other uneaten prey remains docu- mented at 43 Burrowing Owl nests in southwestern Idaho, 2001-02. Prey Category Percent No. Percent Biomass Total No. Mammals 73.70 87.67 297 ^Ivilagus nuttallii 0.25 1.03 1 Thomomys townsendii 10.91 44.73 44 Dipodomys ordii 11.41 13.01 46 Perognathus parvus 2.48 0.77 10 Mus musculus 2.98 1.17 12 Mouse species® 18.86 5.86 76 Microtus montanus 26.80 21.10 108 Birds 4.47 8.09 18 Eremophila alpestris 0.25 0.09 1 Sturnus vulgaris 0.74 1.22 3 Sturnella neglecta 0.50 0.41 2 Passerine sp,^ 0.25 0.15 1 Athene cuniculana — juv. 2.23 4.16 9 A. cunicularia — adult 0.25 1.03 1 Raptor sp.'^ 0.25 1.03 1 Amphibians 8.68 3.60 35 Bufo woodhousei 8.68 3.60 35 Reptiles 0.74 0.29 3 Pituophis catenifer 0.74 0.29 3 Scolopendromorpha 0.50 0.00 2 Arachnids 10.92 0.34 44 Scorpiones 10.67 0.33 43 Solpugida 0.25 0.01 1 Orthopterans 0.50 0.01 2 Acrididae 0.25 0.00 1 Gryllidae 0.25 0.00 1 Total vertebrates 87.59 99.65 353 Total invertebrates 12.41 0.35 50 Total 403 Likely P. parvus, R. megalotis, P. maniculatus, or M. musculus. Likely Eremophila alpestris or Sturnella neglecta. Small juvenile hawk or Prairie Falcon {Falco mexicanus) . that Burrowing Owls associate with irrigated agri- culture because of the high abundance of montane voles (Gleason 1978, Rich 1986), presence of high numbers of invertebrate prey in the diet of owls in agricultural habitat may indicate an overlooked im- portance of invertebrate prey to breeding Burrow- ing Owls in these areas. Agricultural nests also had greater species rich- ness than nonagricultural nests. Common rodent species in agricultural habitats, such as montane voles, were not in pellets of nonagricultural nests and likely were not available in that habitat type. However, nonagricultural nests had greater species evenness than agricultural nests. This greater spe- cies evenness likely contributed to our finding that diets of owls nesting in nonagricultural areas had a broader food-niche (i.e., greater diversity), as Simpson’s diversity index can be greatly influenced by species evenness. Narrower food-niche breadths of Burrowing Owls nesting near agricultural fields may indicate a more specialized diet. As MacArthur and Pianka (1966) proposed, one expects a species to special- ize when prey availability is high (i.e., a productive environment), and thus search time is low. A spe- cies will generalize in unproductive environments December 2005 Burrowing Owl Food Habits in Idaho 437 where search times are high. Therefore, if owls in agricultural areas exhibit more specialized diets relative to owls in nonagricultural areas, we pro- pose that owls nesting in agricultural areas are ex- periencing greater prey availability. This is consis- tent with suggestions by previous researchers (Gleason 1978, Rich 1986, Moulton et al. in press) that Burrowing Owls associate with agriculture be- cause of increased prey. However, further research is needed to determine if the narrower food-niche breadth of owls in agricultural areas results from greater prey availability, where owls can be selec- tive, or lower prey diversity. Food-niche Breadth of Burrowing Owls in the NCA. Prior to our study. Burrowing Owls in the NCA were thought to have a very narrow food- niche breadth compared to other raptor species breeding there. Marti et al. (1993) estimated food- niche breadth of Burrowing Owls to be only 2.43, which was the narrowest food-niche breadth of all 12 raptor species studied. In contrast, food-niche breadth of Burrowing Owls in our study was 4.22 ± 0.22, which ranks Burrowing Owls seventh in terms of food-niche breadth (first being the broad- est). This disparity may be explained in part by smaller sample sizes in Marti et al. (1993) com- bined with different levels of identification; that is, the 1993 study identified invertebrate prey to or- der, whereas we identified invertebrates to family when possible. Because this difference in prey level identification would only affect the food-niche breadth estimates of a species whose diet has a large invertebrate component, only Burrowing Owl estimates likely would be affected. Compared to other raptors breeding within the NCA, our study estimated food-niche breadths of Burrowing Owls to be similar to Golden Eagles {Aquila chrysaetos; 4.07) and Long-eared Owls {Asio otus; 4.79; Marti et al. 1993). However, Burrowing Owl diet composition is more similar to American Kestrels {Falco sparverius), which also frequently prey on invertebrates (Marti et al. 1993). In fact, Burrowing Owls and American Kestrels are the only two raptor species in the NCA for which in- vertebrate prey comprises >1% of the diet (in terms of biomass: 23% and 5%, respectively). Acknowitdgments We thank T. Booms, K. Donohue, C. Eaton, J. Egbert, G. Fairhurst, K. Hasselblad, S. Lysne, R. Medeck, J. Rausch, J. Soules, M. Troxler, and J. Urbanic for assis- tance with fieldwork. Financial and logistical support was provided through an ERA STAR Fellowship to C. Moul- ton, grants from the Bureau of Land Management and Idaho Department of Fish and Game, and the Depart- ment of Biology and Raptor Research Center at Boise State University. J. Clark, J. Doremus, and J. Sullivan fa- cilitated our work in the Snake River Birds of Prey Na- tional Conservation Area, and M. Fuller, Director of the Raptor Research Center, was helpful in numerous ways Finally, we thank J. Bednarz, C.S. Houston, J. Munger, T. Rich, and I. Robertson for helpful comments on previous versions of our manuscript. Literature Cited Alatalo, R.V. 1981. Problems in the measurement of evenness in ecology. Oikos 37:199-204. Bellocq, M.A. 1997. Ecology of the Burrowing Owl in agrosystems of central Argentina. Pages 52—57 m J.L. Lincer [Ed.], The Burrowing Owl, its biology and management including the proceedings of the first international Burrowing Owl symposium. Raptor Re- search Report 9. Belthoff, J.R. AND B.W. Smith. 2003. Patterns of artificial burrow occupancy and reuse by Burrowing Owls in Idaho. Wildl. Soc. Bull. 31:138-144. Brady, R.S. 2004. Nest-lining behavior, nest microclimate, and nest defense behavior of Burrowing Owls. M.S. thesis, Boise State University, Boise, ID U.S.A. Carlson, C.A. 1985. Wildlife and agriculture; can they coexist? J. Soil Water Conserv. 40:263-266. Clayton, K.M. and J.K. Schmutz. 1999. Is the decline of Burrowing Owls (Speotyto cunicularia) in prairie Can- ada linked to changes in Great Plains ecosystems? Bird Conserv. Int. 9:163-185. DeSante, D.F., E.D. Ruhlen, and D.K. Rosenberg. 2004 Density and abundance of Burrowing Owls in the ag- ricultural matrix of the Imperial Valley, California Stud. Avian Biol. 27:116-119. Gervais, J.A., D.K. Rosenberg, D.M. Fry, L. Trulio, and K.K. Sturm. 2000. Burrowing Owls and agricultural pesticides: evaluation of residues and risks for three populations in California, U.S.A. Environ. Toxicol Chem. 19:337-343. Gleason, R.L. 1978. Aspects of the breeding biology of Burrowing Owls in southeastern Idaho. M.S. thesis. University of Idaho, Moscow, ID U.S.A. AND T.H. Craig. 1979. Food habits of Burrowing Owls in southeastern Idaho. Great Basin Nat. 39:273— 276. Green, G.A. 1983. Ecology of breeding Burrowing Owls in the Columbia Basin, Oregon. M.S. thesis, Oregon State University, Corvallis, OR U.S.A. , R.E. Fitzner, R.G. Anthony, and L.E. Rogers. 1993. Comparative diets of Burrowing Owls in Oregon and Washington. Northwest Sci. 67:88-93. Haug, E.A. 1985. Observations on the breeding ecology of Burrowing Owls in Saskatchewan. M.S. thesis. Uni- versity of Saskatchewan, Saskatoon, Canada. , B.A. Millsap, and M.S. Martell. 1993. Burrow- ing Owl {Speotyto cunicularia). In A. Poole and F. Gill [Eds.], The birds of North America, No. 61. The Academy of Natural Sciences, Philadelphia, PA and 438 Moulton et al. VoL. 39, No. 4 The American Ornithologists’ Union, Washington, DC U.S.A. AND L.W. Oliphant. 1990. Movements, activity patterns, and habitat use of Burrowing Owls in Sas- katchewan./. Wildl. Manage. 54:27-35. Hill, M.O. 1973. Diversity and evenness: a unifying no- tation and its consequences. Ecology 54:427-432. Hironaka, M., M.A. Fosberg, and A.H. Winward. 1983. Sagebrush-grass habitat types of southern Idaho. Bul- letin Number 35. Forest, Wildlife and Range Experi- ment Station, University of Idaho, Moscow, ID U.S.A. Holt, D.W., W.D. Norton, and W.C. Atkinson 2001. Breeding season food habits of Burrowing Owls in south-central Montana. Intermt. J. Sci. 7:63—69. Jahn, L.R. and E.W. Schenck. 1991. What sustainable ag- riculture means for fish and wildlife. / Soil Water Con- serv. 46:251-254. James, P.C. and R,M. Espie. 1997. Current status of the Burrowing Owl in North America: an agency survey. Pages 3-5 mJ.L. Lincer [Ed.]. The Burrowing Owl, its Biology and Management including the proceed- ings of the first international Burrowing Owl sympo- sium. Raptor Research Report 9. and G.A. Fox. 1987. Effects of some insecticides on productivity of Burrowing Owls. Blue Jay 45:65-71. Klute, D.S., L.W. Ayers, M.T. Green, W.H. Howe, S.L. Jones, J.A. Shaffer' S.R. Sheffield, and T.S. Zimmer- man. 2003. A status assessment and conservation plan for the western Burrowing Owl in the United States. U.S. Department of Interior, Fish and Wildlife Ser- vice, Biological Technical Publication FWS/BTP- R6001-2003, Washington, DC U.S.A. Leptich, D. 1994. Agricultural development and its influ- ence on raptors in southern Idaho. Northwest Sci. 68: 167-171. MacArthur, R.H. and E.R. Pianka. 1966. On optimal use of a patchy environment. Am. Nat. 100:603-609. Marks, J.S. and J.H. Doremus. 1988. Breeding-season diet of Northern Saw-whet Owls in southwestern Ida- ho. Wilson Bull. 100:690-694. AND V.A. Marks. 1981. Comparative food habits of the Screech Owl and Long-eared Owl in south- western Idaho. Murrelet 62:80-82. Marti, C.D. 1974. Feeding ecology of four sympatric owl s . Condor 7 6 : 45—6 1 . . 1987. Raptor food habits studies. Pages 67-80 in B.A. Pendleton, B.A. Millsap, K.W. Kline, and D.M. Bird [Eds.], Raptor management techniques manual. National Wildlife Federation Scientific and Technical Series No. 10, Washington, DC U.S.A. . 1988. A long-term study of food-niche dynamics in the Common Barn Owl: comparisons within and between populations. Can. J. Zool. 66:1803—1812. , K. Steenhof, M.N. Kochert, and J.S. Marks. 1993. Community trophic structure: the roles of diet. body size, and activity time in vertebrate predators. Oikos 67:6—18. Moulton, C.E. 2003. Ecology of Burrowing Owls in southwestern Idaho: association with agriculture, food habits, and territorial behavior. M.S. thesis, Boise State University, Boise, ID U.S. A. , R.S. Brady, and J.R. Belthoff. In press. Associ- ation between wildlife and agriculture: underlying mechanisms and implications in Burrowing Owls. / Wildl. Manage. Murphy, M.T. 2003. Avian population trends within the evolving agricultural landscape of eastern and central United States. Auk 120:20-34. Olenick, B.E. 1990. Breeding biology of Burrowing Owls using artificial nest burrows in southeastern Idaho. M.S. thesis, Idaho State University, Pocatello, ID U.S.A. Peterjohn, B.G. 2003. Agricultural landscapes: can they support healthy bird populations as well as farm prod- ucts? Auk 120:14-19. Rice, W.R. 1989. Analyzing tables of statistical tests. Evo- lution 43:223-225. Rich, T. 1986. Habitat and nest-site selection of Burrow- ing Owls in the sagebrush steppe of Idaho. /. Wildl. Manage. 50:548-555. Rosenberg, D.K and K.L. Hai.ey. 2004. Ecology of Bur- rowing Owls in the agroecosystem of the Imperial Val- ley, California. Stud. Avian Biol. 27:120-135. Simpson, E.H. 1949. Measurement of diversity. Nature 163:688. Smith, B.W. and J.R. Belthoff. 2001. Effects of nest di- mensions on use of artificial burrow systems by Bur- rowing Owls./. Wildl. Manage. 65:318-326. Smith, D.G. and J.R. Murphy. 1973. Breeding ecology of raptors in the eastern Great Basin of Utah. Brigham Young Univ. Sci. Bull. Biol. Ser. 18:69-76. Steenhof, K. 1983. Prey weights for computing percent biomass in raptor diets. Raptor Res. 17:15—27. AND M.N. Kochert. 1988. Dietary responses of three raptor species to changing prey densities in a natural environment./ Anim. Ecol. 57:37-48. Thomsen, L. 1971. Behavior and ecology of Burrowing Owls on the Oakland Municipal Airport. Condor 73: 177-192. U.S. Department of the Interior. 1996. Effects of mil- itary training and fire in the Snake River Birds of Prey National Conservation Area. BLM/IDARNG Research Project Final Report. U.S. Geological Survey, Biol Res. Div., Snake River Field Station, Boise, ID U.S.A. York, M.M., D.K. Rosenberg, and K.K. Sturm. 2002. Diet and food-niche breadth of Burrowing Owls {Athene cu- nicularia) in the Imperial Valley, California. West. N Am. Nat. 62:280-287. Zar, J.H. 1999. Biostatistical analysis, 4th Ed. Prentice Hall, Upper Saddle River, NJ U.S.A. Received 1 December 2003; accepted 3 July 2005 J Raptor Res. 39(4) :439-444 © 2005 The Raptor Research Foundation, Inc. RED-TAILED HAWK DIETARY OVERLAP WITH NORTHERN GOSHAWKS ON THE KAIBAB PLATEAU, ARIZONA Angela E. Gatto^ Bureau of Land Management, Lake Havasu, 2610 Sweetwater Avenue, Lake Havasu City, AZ 86406 U.S.A. Teryl G. Grubb Rocky Mountain Research Station, 2500 South Pine Knoll Drive, Flagstaff, AZ 86001 U.S.A. Carol L. Chambers School of Forestry, College of Ecosystem Science and Management, Northern Arizona University, P.O. Box 15018, Flagstaff, AZ 86001 U.S.A. Abstract. — ^We determined food habits of Red-tailed Hawks {Buteo jamaicensis) for comparison with published information for Northern Goshawks {Accipiter gentilis) to evaluate potential competition on the Kaibab Plateau, Arizona. We collected prey remains and pellets from 42 Red-tailed Hawk nests at the end of the nesting season between August-October 1998-2001, and opportunistically from below nest trees during site visits, May-July 2000-01. We identified 478 prey items, including 17 mammal, 7 bird, and 2 reptile species. Prey species frequency did not vary among years (P = 0.3), across habitat types (P = 0.8), or by collection technique (P = 0.4). Annual food niche breadth for Red-tailed Hawks averaged 0.57. Published mean niche breadth for Northern Goshawks was 0.32, supporting that Red- tailed Hawks were feeding generalists, while Northern Goshawks were more specialized. However, 48% of Red-tailed Hawk diet on the Kaibab Plateau consisted of species comprising a major portion of the documented diet of Northern Goshawks, including Nuttall’s cottontail {Sylvilagus nuttallii), golden-man- tled ground squirrel {Spermophilus lateralis lateralis), rock squirrel (5. variegates grammurus) , and Northern Elicker ( Colaptes auratus) . Because raptor communities with high dietary overlap and lack of prey par- titioning show food-limited nesting success, greater agnonistic behavior, and territoriality, Red-tailed Hawks could be negatively affecting Northern Goshawks on the Kaibab Plateau. Key Words: Red-tailed Hawk, Buteo jamaicensis; Northern Goshawk, Accipiter gentilis; competition', diet, food habits; food niche breadth; foraging. SOBRELAPAMIENTO DE LA DIETA DE BUTEO fAMAICENSlSY ACCIPITER GPAP/L/SEN LAMESETA DE KAIBAB, ARIZONA Resumen. — Determinamos la dieta de Buteo jamaicensis en la planicie de Kaibab, Arizona y la compara- mos con informacion publicada sobre Accipiter gentilis con la finalidad de evaluar la posible existencia de competencia entre estas dos especies. Recolectamos restos de presas y egagropilas de 42 nidos de B. jamaicensis al final de la temporada reproductiva entre agosto y octubre de 1998-2001, y de forma oportunista debajo de arboles de anidacion durante visitas a la zona de estudio realizadas entre mayo y julio de 2000 y 2001. Identificamos 478 tipos de presas, induyendo 17 especies de mamiferos, 7 de aves y 2 de reptiles. La frecuencia de las especies de presa no vario entre anos (P = 0.3), tipos de habitat (P = 0.8) y tecnicas de colecta (P = 0.4). La amplitud anual del nicho alimentario de B. jamaicensis promedio 0.57. Registros publicados indican que la amplitud del nicho alimentario de A. gentilis es 0.32, lo cual sugiere que B. jamaicensis es una especie generalista, mientras que A. gentilis es mas especializada. Sin embargo, el 48% de la dieta de B. jamaicensis en la planicie de Kaibab consiste de especies que forman la mayor parte de la dieta documentada para A. gentilis, entre las que se encuentran Sylvilagus nuttallii, Spermophilus lateralis lateralis, S. variegates grammurus y Colaptes auratus. Debido a que las comunidades de rapaces que presentan un alto grado de superposicion en su dieta y en las que no hay reparticion de presas muestran un bajo exito de anidacion debido a falta de alimento, ^ Email address: angela_gatto@blm.gov 439 440 Gatto et ai.. VoL. 39, No. 4 un mayor comportamiento antagonico y mayor territorialidad, B. jamaicensis podria estar afectando negativamente a A. gentilis en la planicie de Kaibab. [Traduccion del equipo editorial] Fire suppression, timber harvesting, and live- stock grazing over the past 100 yr have caused changes in southwestern forests, including those on the Kaibab Plateau in northwestern Arizona (e.g.. Weaver 1951, Cooper 1960, Covington and Moore 1994). Decreasing the quality and quantity of climax forest can favor habitat generalists (Carey 1984) such as Red-tailed Hawks {Buteo jamaicensis) , which may out compete species dependent on late successional forests, such as Northern Goshawks {Accipiter gentilis; Andersen et al. 2003). Creation of 259-ha buffer zones on the plateau by the Kaibab National Forest, although maintaining the quality of old-growth for nesting sites, does not ensure suf- ficient quantity or quality of old-growth forest be- yond these protected areas (Reynolds et al. 1992). Red-tailed Hawks often nest in abandoned North- ern Goshawk nests after timber harvesting (Crock- er-Bedford 1990). Beyond old-growth-managed buffer zones, open areas may favor the hunting style, and thus, increase foraging habitat of Red- tailed Hawks more than Northern Goshawks. Ex- ploring the potential relationship between these two species facilitates a better understanding of how upper level avian predators have adjusted to human-caused alterations in southwestern forest ecosystems. Further, this understanding could aid resource managers in maintaining a viable North- ern Goshawk population on the Kaibab Plateau. Competition occurs when use or defense of a resource by one individual reduces the availability of that resource to other individuals (Gill 1990). Interspecific competition occurs when individuals of coexisting species require a resource that is in limited supply relative to their needs such that sur- vival or reproduction of at least one species is de- creased (Ricklefs and Schluter 1993). On the Kai- bab Plateau, Red-tailed Hawks and Northern Goshawks, based on their proximity and similar habitat use, could be competing for nest sites, for- aging areas, or prey, despite potential partitioning resulting from morphological and behavioral dif- ferences between species (Ballam 1984, Squires and Reynolds 1997). In this study, we focused on determining the diet of Red-tailed Hawks for com- parison with published information on Northern Goshawks to evaluate potential overlap. We hypoth- esized that Red-tailed Hawks and Northern Gos- hawks, because of their close proximity and use of similar nesting and foraging habitat, could also be using the same prey species, creating the potential for competition. Study Area The Kaibab Plateau, on the North Rim of the Grand Canyon, is located in northwestern Arizona within the Kaibab National Forest. The plateau encompasses 2980 km^ above 1830 m elevation. Our study area was confined to 1732 km^ above the 2075 m contour to be consistent with concurrent long-term Northern Goshawk research on the plateau (Reynolds et al. 1994). Goshawk and Red- tailed Hawk nest sites are interspersed throughout this study area. Vegetation on the plateau consists of ponde- rosa pine {Finns ponderosa) forest between 2075-2500 m; mixed-conifer forest (ponderosa pine, Douglas-fir [Pseu- dotsuga menziesii], white fir [Abies concolor], blue spruce [Picea pungens] , and quaking aspen [Populus tremuloides ^ ) between 2500-2650 m; and Engelmann spruce-subalpine fir {Picea engelmannii-Abies lasiocarpa) forest between 2650-2800 m (Rasmussen 1941, White and Vankat 1993) Methods Prey Remains and Pellet Identification. We collected prey remains and regurgitated pellets from all known oc- cupied Red-tailed Hawk nests {N = 21, 24, 32, 11 for 1998-2001, respectively), from below and in the nest, at the end of the nesting season, August-October 1998- 2001, and opportunistically during the nesting season, May-Jnly 2000-01. Collections from each nest were sep- arated by >3 d, and end-of-season samples were collected >30 d after previous collections. We treated each collec- tion of pellets and prey remains from an occupied nest as one sample for that nest site. We assumed prey species identified in each sample had been consumed since the last sample and represented new prey. To identify prey, we separated samples into bones/frag- ments, feathers, and hair. Bones and feathers were iden- tified to the lowest taxon possible, through comparison with S. Bayard’s reference prey collection stored at the Rocky Mountain Research Station, Fort Collins, CO. We identified hairs using keys (Williams 1938, Stains 1958, Moore et al. 1974) and by comparing hairs directly with samples from the Northern Arizona University, Depart- ment of Biological Sciences’ collection and the private collection of H.E. Graham (Flagstaff, AZ U.S.A.). The characteristics we considered included color banding, shape, presence of a hair shield, and configuration of a medulla if present (Moore et al. 1974). Determining diet via indirect means requires cautious interpretation because of inherent biases associated with each method (Lewis et al. 2004); however, identification of a prey species in pellets and prey remains is an abso- lute indication of presence. Therefore, we recorded one occurrence whenever a species was found in a sample. December 2005 Red-tailed Hawk Diet 441 We then pooled samples across years for each nest and summed the number of times a prey species occurred. This provided a conservative estimate of the relative im- portance of prey species consumed by Red-tailed Hawks at each nest. We also calculated the total number and percent of nests at which each prey species occurred. Prey Species Dissimilarity. Dyer (1978) developed a lin- ear statistical analysis for comparing species dissimilarity that can be used in conjunction with any species dissim- ilarity index and is designed for data sets that involve both multiple species and multiple environmental vari- ables. Total species dissimilarity is divided into compo- nents with one component being assigned to each envi- ronmental variable or interaction of environmental variables: = Po + + W' + - + where D^yis the (dis) similarity between observations zand j, 5,yh) is a known function of i and j which corresponds to the environmental variable or interaction, Pj is an unknown parameter which represents the contribution of the environmental variable or interaction to the total dissimilarity, and cij is an error term with an ex- pected value of 0. We used this linear model with the Jaccard dissimilarity index (Krebs 1998) to estimate whether the species iden- tified in prey remains and pellets of Red-tailed Hawks varied among years (1998-2001), vegetation types (pon- derosa pine only, mixed conifer with pine dominant, or mixed conifer only), or collection techniques (end of season samples from nest or opportunistic samples col- lected from below nest trees during nesting season). Jac- card’s index is specifically designed for presence-absence data, and because, by definition, rare species are typically absent, rare species have little influence on the value of the index (Krebs 1998). The Jaccard dissimilarity index (D^J) is calculated by: Djy = a/ {a + b + c) where a is the number of binary characteristics present m sample i and sample j, b is the number present only in i, and c is the number present only in j. We estimated Dyer’s model using the Jaccard index. We used permu- tation methods (Edington 1995) to estimate a signifi- cance level (P-value) for each environmental variable. Statistical tests were significant if P < 0.05. Niche Breadth. Niche breadth and niche overlap are widely applied to analysis of foraging and community ecology to estimate competition (Greene and Jaksic 1983). Niche breadths were calculated according to Lev- ins’ (1968) equation: P = for i = \ to n Zj Pi^ where pi is the proportion of Red-tailed Hawk nests with the taxon present. The value of p varies from 1 to n, where n is the number of taxa. If prey taxa occur equally among all nests, then p = ra. Niche-breadth values were standardized and converted to a fraction ranging from 0 to 1 by the equation: Pstandard = (P “ ^) / {n ~ 1). To calculate niche breadth, we created a prey species list based on pooled prey species present across all oc- cupied Red-tailed Hawk nests by year. This yielded a sep- arate food niche breadth measure for each of the four study years. We also calculated a 95% confidence interval for Red-tailed Hawk niche breadth on the Kaibab Plateau for comparison with calculated niche breadths for North- ern Goshawks from other studies. Resuits Pellet and Prey Remain Analysis. Considering each visit’s collection of pellets and prey remains as one discreet sample, we obtained 140 prey sam- ples {N — 80 end of nesting season, 1998-2001; N = 60 opportunistic during nesting season, 2000- 01) , consisting of 1-10 collections from 42 nests {x = 3.3). We identified 478 prey items, including at least 17 mammal, seven bird, and two reptile spe- cies (Table 1). The number of species at any one nest site ranged from 1-6. For all 4 yr combined, mammals represented 72% of Red-tailed Hawk diet and birds represented 27% (Table 1). Only two reptile species were identified; they did not contribute greatly to the overall diet (<1%). Six species accounted for 67% of prey by frequency of occurrence: Nuttall’s cottontail {Sylvilagus nuttallu, 17.6%), Kaibab squirrel (Sciurus aberti kaibabensis, 7.7%), rock squirrel {Spermophilus variegatus gram- murus, 10.0%), golden-mantled ground squirrel {S. lateralis lateralis, 10.3%), Northern Flicker {Colaptes auratus, 10.7%), and Steller’sJay {Cyanocitta stellen, 10.3%; Table 1). Rare occurrences of porcupine (Erethizon dorsatum) , coyote ( Canis latrans) , and mule deer {Odocoileus hemionus) comprised <2% of overall Red-tailed Hawk diet. Dissimilarity Measures and Niche Breadth. Prey species frequency identified from pellet and prey remains did not vary among years {P = 0.3), across habitat types {P = 0.8), or by collection technique (P = 0.4). Dietary niche breadth for Red-tailed Hawks pooled across nests for each year was 0.58 in 1998, 0.52 in 1999, 0.51 in 2000, and 0.65 in 2001. The mean and 95% confidence interval for Red-tailed Hawks niche breadth for all four years combined was 0.57 ± 0.11, Calculated food niche breadths for Northern Goshawks for Arizona and several other western states did not fall within our confidence interval for Red-tailed Hawks (Table 2) . Dietary overlap between the Red-tailed Hawk during this study and a Northern Goshawk diet from BoaPs (1993) earlier study in the same area was 55% (number of common species/sum of spe- cies recorded for both raptors). 442 Gatto et al. VoL. 39, No. 4 Table 1. Frequency of prey species in 140 samples (478 identified prey items) of pellets and prey remains collected from 42 Red-tailed Hawk nests, May-October 1998-2001, on the Kaibab Plateau, Arizona, U.S.A. Prey Species Frequency (N = 478) Percent Frequency Nest Occurrence {N = 42) Percent Nest Occurrence Nuttall’s cottontail {Sylvilagus nuttallii) 84 17.6 36 85.7 Golden-mantled ground squirrel {Spermophilus lateralis) 49 10.3 26 61.9 Rock squirrel {Spermophilus variegatus grammurus) 48 10.0 29 69.0 Kaibab squirrel {Sciurus aberti kaibabensis) 37 7.7 21 50.0 Chipmunk {Eutamias sp.) 27 5.6 20 47.6 Northern pocket gopher {Thomomys talpoides) 22 4.6 21 50.0 Long-tailed vole {Microtus longicaudus) 17 3.6 17 40.5 Red squirrel ( Tamiasciurus hudsonicus) 15 3.1 13 31.0 Mouse {Peromyscus sp.) 13 2.7 9 21.4 Shrew {Sorex sp.) 11 2.3 10 23.8 Black-tailed jackrabbit {Lepus californicus) 8 1.7 8 19.0 Long-tailed weasel {Mustela frenata arizonenst) 5 1.0 5 11.9 Mule deer (Odocoileus hemionus) 5 1.0 5 11.9 Plains pocket mouse {Perognathus flavescens) 3 0.6 2 4.8 Porcupine {Erethizon dorsatum) 2 0.4 2 4.8 Ringtail {Bassariscus astutus) 1 0.2 1 2.4 Coyote {Canis latrans) 1 0.2 1 2.4 Northern Flicker {Colaptes auratus) 51 10.7 28 66.7 Steller’s Jay ( Cyanocitta stelleri) 49 10.3 26 61.9 Clark’s Nutcracker {Nucifraga Columbiana) 13 2.7 10 23.8 Unknown bird 8 1.7 7 16.7 Common Raven {Corvus corax) 3 0.6 3 7.1 Western Bluebird (Sialia mexicana) 1 0.2 1 2.4 Hairy Woodpecker {Picoides villosus) 1 0.2 1 2.4 Common Nighthawk ( Chordeiles minor) 1 0.2 1 2.4 Unknown snake 3 0.6 2 4.8 Mountain short horned lizard^ {Phrynosoma douglassi) 0 0.0 0 0.0 ^ Mountain short horned lizard was positively identified as a prey species during observations, but was not found in the prey collec- tions. December 2005 Red-tailed Hawk Diet 443 Table 2. Food niche breadth of nesting Northern Goshawks calculated from prey remains collected in Arizona and other western states. Location Number of Nests Food Niche Breadth Source Arizona 20 0.29 Boal and Mannan (1994) California 114 0.41 Bloom et al. (1986) New Mexico 8 0.36 Kennedy (1991) Oregon 4 0.38 Reynolds and Meslow (1984) Discussion Our results on the Kaibab Plateau were consis- tent with Red-tailed Hawks being feeding general- ists and preying primarily upon rabbits {Sylvilagus spp.), black-tailed jackrabbits {Lepus californicus) , and ground squirrels {Spermophilus spp; Preston and Beane 1993); however, Northern Flickers and Steller’s Jays were also frequent in this diet analysis. Previous observation and prey remains analyses for Northern Goshawks on the plateau indicated gos- hawks also preyed mostly upon rabbits and hares, tree and ground squirrels. Northern Flickers, and Steller’s Jays (Boal and Mannan 1994, Kaufmann et al. 1994, and Reynolds et al. 1994). Bosakowski and Smith (1992) found that in eastern forests, competition between accipiters and buteos is usu- ally minimized by a difference in prey selection, with buteos typically having a higher proportion of mammals in their diet, and accipiters more avian prey. Red-tailed Hawks frequently exhibit switching behavior, which is the capability to utilize which- ever species is most abundant at the time (Steen- hof and Kochert 1988). Predators that are gener- alists often have weak and variable prey preferences and will exhibit switching behavior, while specialists with strong or consistent prefer- ences do not (Murdoch 1969). The wider niche breadth of Red-tailed Hawks on the Kaibab Plateau indicated weaker preferences compared to the nar- rower niche breadth of Northern Goshawks. Red-tailed Hawks and Northern Goshawks fre- quently occupy similar nesting habitat on the Kai- bab Plateau. Both species tend to nest in larger trees (x DBH = 68.3-72.5 cm) and mid-slope in drainages (x slope position = 0.36-0.37; LaSorte et al. 2004) . Because they also nest in close proximity (<2000 m apart, USDA Forest Service, North Kai- bab Ranger District unpubl. data), we suggest that they may also be utilizing the same foraging habi- tat. Unfortunately, pellets and prey remains do not provide insight into any spatial or temporal parti- tioning of prey by these two species. However, our research clearly shows Red-tailed Hawks are prey- ing upon many of the same species utilized by Northern Goshawks. Thus, we believe the potential for competition exists. We further hypothesize that the more generalist nature of Red-tailed Hawk diet and nesting habitat, in combination with the de- terioration of late-successional forest habitat and concurrent creation of openings on the Kaibab Plateau, may tend to exacerbate any potential con- flict. We suggest that additional research should be implemented to examine this potential competi- tion, better quantify numbers of prey in both spe- cies’ diets, and determine potential effects on such competition on Northern Goshawk management alternatives. Acknowledgments We thank R. Reynolds for first conceiving of this pro- ject; he, S. Bayard de Volo, S. Salafsky, and his summer crews on the Kaibab Plateau provided field and logistical support. A. Hales, R. Lopez, and T. Rohmer assisted with data collection. R. Baida and P. Beier offered helpful comments on study design, as well as the original man- uscript. R. King provided invaluable statistical assistance. We also thank D. Andersen and S. Lewis for thorough and constructive reviews of the manuscript. This research was part of the senior author’s Master of Science project at Northern Arizona University, funded by U.S. Depart- ment of Agriculture Forest Service, Rocky Mountain Re- search Station. Literature Cited Andersen, D.E., S. DeStefano, M.I. Goldstein, K. Titus, C. Crocker-Bedford, JJ- Keane, R.G. Anthony, and R.N. Rosenfield. 2003. The status of Northern Gos- hawks in the western United States. Wildlife Society Technical Review 04-1 . The Wildlife Society, Bethesda, MD U.S.A. Bai.lam, J.M. 1984. The use of soaring by the Red-tailed Hawk. Auk 101:519-524. Bloom, P.H., G.R. Stewart, and BJ. Walton. 1986. The status of the Northern Goshawk in California, 1981— 444 Gatto et al. VoL. 39, No. 4 1983. California Department of Fish and Game, Wild- life Management Branch Administration Report 85-1, Boal, C.W. 1993. Northern Goshawk diets in ponderosa pine forests in northern Arizona. M.S. thesis. Univer- sity of Arizona, Tucson, AZ U.S.A. AND R.W. Mannan. 1994. Northern Goshawk diets in ponderosa pine forests on the Kaibab Plateau. Stud. Avian Biol. 16:97-102. Bosakowski, T. and D.G. Smith. 1992. Comparative diets of sympatric nesting raptors in the eastern deciduous forest biome. Can.]. Zool. 70:984—999. Carey, A.B. 1984. A critical look at the issue of species- habitat dependency. Pages 28-33 in Challenges for wildlife and fish-the old growth ecosystem in managed forests. Proc. 1983 Technical Session of Wildlife and Fish Ecology. Working Group of Society of American Forests, Washington, DC U.S.A. Cooper, C.F. 1960. Changes in vegetation, structure, and growth of southwestern ponderosa pine forests since white settlement. Ecol. Monogr. 30:129—164. Covington, W.W. and M.M. Moore. 1994. Southwestern ponderosa forest structure: changes since Euro-Amer- ican settlement. J. For. 92:39-47. Crocker-Bedford, C.D. 1990. Goshawk reproduction and forest management. Wildl. Soc. Bull. 18:262-269. Dyer, D.P. 1978. An analysis of dissimilarity using multi- ple environmental variables. Ecology 59:117-125. Edington, E.S. 1995. Randomization tests, 3rd Ed. Mar- cel Dekker, New York, NY U.S.A. Gill, F.B. 1990. Ornithology. W.H. Freeman & Co., New York, NYU.S.A. Greene, H.W. and F.M. Jaksic. 1983. Food-niche relation- ships among sympatric predators: effects of level of prey identification. 40:151-154. Kaufmann, M.R., R.T. Graham, D.A. Boyce, W.H. Moir, L. Perry, R.T. Reynolds, R.L. Bassett, P. Mehlhop, C.B. Edminster, W.M. Block, and P.S. Corn. 1994. An ecological basis for ecosystem management. USDA Forest Service Rocky Mountain Forest 8c Range Ex- periment Station, Ft. Collins, CO U.S.A. Kennedy, P.L. 1991. Reproductive strategies of Northern Goshawks and Cooper’s Hawks in north-central New Mexico. Ph.D. dissertation, Utah State University, Lo- gan, UT U.S.A. Krebs, C.J. 1998. Ecological methodology, 2nd Ed. Ben- jamin/Cummings, Menlo Park, CA U.S.A. LaSorte, F.A., R. William Mannan, R.T. Reynolds, T.G. Grubb. 2004. Habitat association of sympatric Red- tailed Hawks and Northern Goshawks on the Kaibab Plateau./ Wildl. Manage. 68:307-317. Levins, R. 1968. Evolution in changing environments: some theoretical explorations. Princeton University Press, Princeton, NJ U.S.A. Lewis, S.B., M.R. Fuller, and K. Titus. 2004. A compar- ison of 3 methods for assessing raptor diet during the breeding season. Wildl. Soc. Bull. 32:373-385. Moore, T.D., L.E. Spence, C.E. Dugnolle, and W.G. Hepworth. 1974. Identification of the dorsal guard hairs of some mammals of Wyoming. Wyoming Game and Fish Department, Cheyenne, WY U.S.A. Murdoch, W.W. 1969. Switching in general predators, experiments on predator specificity and stability of prey population. Ecol. Monogr. 39:335-354. Preston, C.R. and R.D. Beane. 1993. Red-tailed Hawk The Birds of North America. The Academy of Natural Sciences, Philadelphia, PA U.S.A. and The American Ornithologists’ Union, Washington, DC U.S.A. Rasmussen, D.I. 1941. Biotic communities of Kaibab Pla- teau, Arizona. Ecol. Monogr. 11:231-274. Reynolds, R.T., R.T. Graham, M.H. Reiser, R.L. Bassett, P.L. Kennedy, D.A. Boyce, G. Goodwin, R. Smith, and E.L. Fisher. 1992. Management recommendations for the Northern Goshawk in the southwestern United States. USDA Forest Service Rocky Mountain Forest & Range Experiment Station, Ft. Collins, CO U.S.A. , S.M. Joy, and D.G. Leslie. 1994. Nest productiv- ity, fidelity, and spacing of Northern Goshawks in Ar- izona. Stud. Avian Biol. 16:103-113. AND E.C. Meslow. 1984. Partitioning of food and niche characteristics of coexisting Accipters during breeding. Auk 101:761-779. Ricklefs, R.E. AND D. Schluter. 1993. Ecological com- munities: historical and geographical perspectives. University of Chicago Press, Chicago, IL U.S.A. Squires, J.R. and R.T. Reynolds. 1997. Northern Gos- hawk. The Birds of North America. The Academy of Natural Sciences, Philadelphia, PA and The American Ornithologists’ Union, Washington, DC U.S.A. Stains, H.J. 1958. Field key to guard hair of middle west- ern furbearers. J. Wildl. Manage. 22:95-97. Steenhof, K. and M.N. Kochert. 1988. Dietary respons- es of three raptor species to changing prey densities in a natural environment./. Anim. Ecol. 57:37-48. Weaver, H. 1951. Fire as an ecological factor in the southwestern ponderosa pine forests./. For. 49:93-98. White, M.A. and J.T. Vankat. 1993. Middle and high el- evation coniferous forest communities of the North Rim region of the Grand Canyon National Park, Ari- zona, U.S.A. Vegetatio 109:161-174. Williams, C.S. 1938. Aids to the identification of mouse and shrew hairs with general comments on hair struc- ture and hair determination./. Wildl. Manage. 2:239- 269. Received 15 March 2004; accepted 26 March 2005 Associate Editor: Clint W. Boal J Raptor Res. 39(4) :445-453 © 2005 The Raptor Research Foundation, Inc. BAT PREDATION BY LONG-EARED OWLS IN MEDITERRANEAN AND TEMPERATE REGIONS OF SOUTHERN EUROPE Ana Maria Garcia^ IMEDEA (CSIC-UJB), Miquel Marques 21, E-07190 Esporles, Mallorca, Spain Francisco Cervera C.R.F. Granja de El Saler, Conselleria de Territori i Habitatge, Avda. dels Pinars 106, E-46012, Valencia, Spain Alejandro Rodriguez Department of Applied Biology, Estadon Bioldgica de Donana, CSIC, Avda. Maria Luisa s/n, E-41013, Sevilla, Spain Abstract. — ^We described spatial and temporal variation in bat consumption by Long-eared Owls {Asio otus) at a coastal site of eastern Spain and examined the importance of bats in the diet of this raptor in nine temperate and 21 Mediterranean localities of southern Europe. In our study site in Spain, bats accounted for 2% of prey items, which is the largest percentage so far reported for the species. The vast majority of bats were Pipistrellus spp. Bat predation occurred in all seasons, but was significantly higher in spring and summer. The temporal pattern of bat predation was unrelated to temporal variation in the consumption of rodents, the dominant prey in the diet. Although a consistent increase in bat intake only in years of rodent scarcity predicts an aggregation of occurrences over time, bat occurrence during 31 successive seasons was not different from a random sequence. Pellets containing bat remains originated mainly from one communal roosting site. Bat remains appeared in pellets from five of 16 nests, accounting for 17% of prey items on average. In southern Europe, bats occurred in 38% of diets in the Mediterranean region, while they were absent in diets from adjacent temperate localities. Our results suggest that Long-eared Owls prey on bats rarely and opportunistically in Mediterranean sites, but also that bat aggregations could be a locally important food source for some individual owls during certain periods. Key Words: Long-eared Owl, Asio otus; Chiroptera; diet, Mediterranean basin; trophic plasticity. PREDACION de MURCIELAGOS POR el BUHO CHICO EN REGIONES TEMPLADAS YMEDITER- rAneas del sur de EUROPA Resumen. — Describimos la variacion espacial y temporal en el consumo de murcielagos por parte del buho Chico Asio otus en una localidad costera del este de Espana, y examinamos la importancia de los quiropteros en la dieta de esta rapaz en 30 local! dades del sur de Europa, 9 de clima templado y 21 de clima mediterraneo. En nuestra area de estudio, los quiropteros constituyeron el 2% de las presas ingeridas, cifra que representa el mayor consumo conocido para la especie. Casi todos los murcielagos consumidos fueron a Pipistrellus spp. Su predacion se produjo en todas las estaciones, pero fue signifi- cativamente mas alta en primavera y verano. El patron temporal de predacion de murcielagos no estuvo relacionado con la variacion temporal en el consumo de roedores, la presa dominante en la dieta. El incremento en el consumo de murcielagos solo en anos en los que los roedores son escasos predice una agregacion temporal de las apariciones. Sin embargo, la presencia de murcielagos en la dieta a lo largo de 31 estaciones sucesivas no difirio de una secuencia aleatoria. La mayor parte de las egagropilas que contuvieron murcielagos procedieron del dormidero comunal. Encontramos restos de quiropteros en cinco de los 16 nidos muestreados, donde constituyeron en promedio el 17% de las presas. En Europa meridional, los murcielagos aparecieron en el 38% de las dietas de la region mediterranea, pero en ninguna de las dietas de la region templada adyacente. Nuestros resultados indican que A. otus consume murcielagos con baja frecuenda de forma oportupista en la region mediterranea, pero tambien sugieren que las agrtipaciohes de quiropteros pueden ser una fuente de alimento localmente importante para algunos individuos durante periodos concretos. [Traduccion del autor] ^ Email address: panamel@ono.com 445 446 Garcia et al. VoL. 39, No. 4 Throughout the boreal and temperate regions of Europe, Long-eared Owls {Asio otus) prey almost exclusively upon microtine rodents (Herrera and Hiraldo 1976, Lundberg 1979, Marks et al. 1999). As a result, owl numbers decrease locally with de- clining vole (Microtus spp.) populations and, at larger spatial scales, owls may use nomadic or ir- ruptive movements to track peaks in vole abun- dance (Lundberg 1979, Korpimaki 1985, Hanski et al. 1991). Such a numerical response is a trait of specialist predators that may be explained in part by low availability of alternative prey (Weber et al. 2002) and by high predictability of vole population peaks in such ecosystems (Korpimaki 1985). How- ever, Long-eared Owls seem to depend less on ro- dents at lower latitudes (Bertolino et al. 2001), es- pecially in Mediterranean regions (Garcia and Cervera 2001) which feature lower environmental predictability, resulting in strong seasonal and an- nual fluctuations in the abundance of rodents and other prey (Blondel and Aronson 1999). Indeed diet diversification with decreasing latitude may be a general pattern in nocturnal raptors (Herrera and Hiraldo 1976, Mikkola 1983, Korpimaki and Marti 1995) and other predators (Revilla and Pa- lomares 2002, Clavero et al. 2003) , suggesting that specialization may not always reflect species-specif- ic constraints in physiology or morphology, but be- havioral flexibility (Futuyma and Moreno 1988, Martin et al. 1995). Bats have regularly been reported, albeit in small amounts, as prey of a variety of diurnal and noc- turnal raptors (e.g., Baker 1962, Ruprecht 1979, Barclay et al. 1982). Long-eared Owls are no ex- ception. For example, in the British Isles, this spe- cies was the second most important bat predator among raptors (Speakman 1991). In this paper, we describe the pattern of bat consumption by Long- eared Owls during an 8-yr period in a Mediterra- nean site with good habitat for bats in terms of roost (buildings) and food availability (insects in rice fields and other flooded areas) . We also review European studies of Long-eared Owl diet in the Mediterranean basin and the adjacent temperate zone to examine the geographical pattern of pre- dation on bats. We predicted that occurrence of bats in the Mediterranean sites would be higher than in temperate sites of similar latitude because (1) rodent abundance undergoes pronounced sea- sonal and annual fluctuations, and owls must search for alternative prey and (2) the season of bat activity is longer, and bat average abundance higher, in warmer Mediterranean environments (Avery 1985, Altringham 1996). Methods We studied food habits of Long-eared Owls in Devesa de I’Albufera, one Mediterranean coastal site near Valen- cia city, Spain (39°21'N, 0°19'W). The owl habitat is a mosaic of pine forest (Pinus halepensis) with dense un- derstory and open areas, mostly dunes and mesic inter- dune depressions (Costa et al. 1982). This forested land- scape is highly disturbed (many buildings and regular recreational activities) and surrounded by a large ex- panse of rice fields. From November 1995 to June 2003, we collected owl pellets from beneath roost and nest sites on a monthly basis. We identified prey remains and, for each pellet, determined occurrence and minimum num- ber of individuals of prey species. Using these data, we analyzed spatio-temporal fluctuations in bat predation. For analysis of seasonal variation in bat consumption, sea- sons were defined as winter ( January-March) , spring (April-June) , summer ( July-September) , and fall (Oc- tober-December) . We carried out the biogeographic comparison of bat predation using data from 30 diet studies from southern Europe (Table 1). Each study area was assigned to the Mediterranean or the temperate climate region accord- ing to Emberger et al. (1963; Fig. 1). We excluded north- ern temperate localities to avoid diets almost completely dominated by voles. For diets containing bats, we used Spearman correlation analysis to test the hypothesis that proportion of bats decreased with increasing latitude and altitude. Results and Discussion Bat Consumption in Devesa de I’Albufera, Coast- al Spain. We collected 2012 pellets that contained 6210 prey items. Pellets containing bat remains originated mainly from a communal roosting site (60%) and also near 16 nest sites, where we re- corded successful owl reproduction. Bats account- ed for 2% of prey items (Table 1), which is the largest percentage thus far reported for Long- eared Owls (Marti 1976, Mikkola 1983; Table 1). Of 126 bats, 124 were pipistrelle bats {Pipistrellus spp; Table 2) . Bat remains occurred in all seasons, but predation on bats was significantly higher dur- ing the peak of bat activity and abundance in spring and summer (G = 47.3, df = 3, P < 0.001; Fig. 2) . In our study area, the first flights of young pipistrelle bats take place between mid-July and mid-August (D. Almenar and M. Monsalve pers. comm.), and during the initial 2 wk their flight skills are less than those of adults (Blanco 1998). Thus, the combination of the annual peak in abun- dance associated with the emergence of young bats and their relatively higher vulnerability, associated with their reduced flight capability, may help to December 2005 Bat Predation by Long-eared Owls 447 Table 1. Long-eared Owl diet composition in 30 localities of southern Europe. Each locality is assigned to a climatic region either Mediterranean or temperate. The percentage of bats, rodents, and other prey are calculated on the total number of prey individuals. Numbers assigned to each study area are the same as in Fig. 1. Study Country Climatic Region^* No. OF Prey Percent Bats Percent Rodents Percent Other Prey Source 1 Spain M 874 0.00 86.61 13.26 Alegre et al. 1989 2 Spain M 232 0.00 96.10 3.90 Delibes et al. 1984 3 Spain M 6929 0.04 90.70 9.20 Araujo et al. 1974 4 Spain M 3726 0.00 92.60 7.30 San Segundo 1988 5 Spain M 3185 0.03 78.50 21.50 Veiga 1980 6 Spain M 255 0.00 96.50 2.80 Lopez-Gordo et al. 1977 7 Spain M 804 0.00 72.60 27.40 Amat and Soriguer 1981 8 Spain M 6210 2.03 52.77 45.20 This study 9 Spain M 6249 0.08 89.00 10.70 Corral et al. 1979 10 France M 368 0.00 58.40 39.13 Kayser and Sadoul 1996 11 Italy M 494 0.00 90.08 9.72 Gerdol and Perco 1977 12 Italy M 121 0.00 95.87 4.13 Gerdol and Perco 1977 13 Italy M 103 0.00 54.37 45.63 Gerdol et al. 1982 14 Italy M 1157 0.00 93.52 6.49 Casini and Magnani 1988 15 Italy M 181 0.00 97.24 2.76 Capizzi et al. 1998 16 Italy M 338 0.30 98.20 1.50 Plini 1986 17 Italy M 1787 0.00 81.60 18.50 Guidoni et al. 1999 18 Italy M 201 0.00 95.10 5.00 Capizzi and Luiselli 1996 19 Italy M — 0.11 70.70 28.40 Sublimi and Scalera 1991 20 Italy M 234 <1.40 93.60 5.00 Sara 1990 21 Greece M 961 0.30 87.90 11.70 Alivizatos and Goutner 1999 22 Italy T 1787 0.00 81.60 18.40 Bertolino et al. 2001 23 Italy T 1836 0.00 85.52 14.44 Galeotti and Can ova 1994 24 Italy T 519 0.00 83.63 16.37 Mezzavilla 1993 25 Italy T 655 0.00 93.44 6.56 Malavasi 1995 26 Italy T 593 0.00 90.58 9.42 Aloise and Scaravelli 1995 27 Italy T 98 0.00 79.59 20.40 Riga and Capizzi 1999 28 Slovenia T 10991 0.00 95.48 4.52 Tome 2003a 29 Romania T 1268 0.00 88.18 11.82 Marariu et al. 1991 30 Switzerland T 4639 0.00 99.23 0.77 Roulin 1996 M = Mediterranean and T = temperate region. explain the high occurrence of bats in summer samples. The relative importance of bats in the diet varied by year (G = 79.3, df = 6, P < 0.001; Fig. 2) . In our study area and other Mediterranean en- vironments, rodent populations typically show an- nual minima during summer, especially during warm years (Blondel and Aronson 1999) . Although owl consumption of rodents roughly follows avail- ability (Fig. 3) , bat predation was unrelated to the proportion of rodents in the summer diet (r^ = 0.048, N = 8, P = 0.911; Fig. 3). This suggests that bats are not specifically sought as an alternative prey. The hypothesis that bats are taken only in years of marked rodent scarcity also predicts a clumped occurrence of bats in the diet over the 8 study yr. However, bat occurrence during 31 suc- cessive seasons did not differ from a random se- quence (runs test, Zar 1984; Z = 0.726, P = 0.233; Fig. 3). Bat remains occurred in 42 pellets (2.1%). In 27% of these, bats were the only prey item, with 2-8 individuals per pellet. In 62% of pellets, bats occurred together with other prey, but accounted for ^50% of prey items in each sample. The dis- tribution of the number of individual bats per pel- let was aggregated (did not fit a Poisson distribu- tion with X = 0.055, = 42.62, df = 1, P< 0.001; only pellets with bats, X = 2.643, = 13.23, df — 3, P = 0.004). These results suggested that bats were clumped when captured. Pipistrelle bats are 448 Garcia et al. VoL. 39, No. 4 Figure 1. Localities where Long-eared Owl diet was studied in southern Europe. The broken line separates the temperate (squares) and the Mediterranean localities (circles) following Emberger et al. (1963). Black circles indicate sites where bat predation has been recorded. Numbers correspond to study numbers in Table 1. Table 2. Bat species in the diet of Long-eared Owls in the Mediterranean region of Europe. Seasons when predation occurred and the range of body mass (g) are also shown. Body masses are from Palomo and Gisbert (2002). Study number and location are as in Table 1 and Fig. 1. Species Body Mass (g) Season Source Study Greater horseshoe bat 14.6-31.6 Winter Alivizatos and Goutner 1999 21 (Rhinolophus ferrumequinum) All seasons This study 8 Greater mouse-eared bat 21.0-35.0 Winter-Spring Corral et al. 1979 9 (Myotis myotis) Fall Sublimi and Scalera 1991 19 Lesser mouse-eared bat 18.0-29.5 Spring— Summer Veiga 1980 5 {Myotis blythii) Whiskered bat {Myotis mystacinus) 4.0-8.0 All seasons Plini 1986 16 Myotis spp. Winter Alivizatos and Goutner 1999 21 Pipistrellus spp. 3.5-10.0 All seasons This study 8 Serotine bat {Eptesicus serotinus) 14.0-33.0 Winter-Spring Corral et al. 1979 9 Brown long-eared bat 6.8-12.0 Spring-Fall Araujo et al. 1974 3 {Plecotus auritus) Schreibers’ bat {Miniopterus 10.1-20.8 All seasons This study 8 schreibersii) European free-tailed bat 22.0-54.0 Winter Alivizatos and Goutner 1999 21 {Tadarida teniotis) December 2005 Bat Predation by Long-eared Owls 449 (0 E >. E OL (0 (0 (0 4 -* CO OQ o A E Year I 1 winter i— i spring summer fall Figure 2 . Temporal variation in the number of individuals taken by Long-eared Owls (bars) and in the percent of bats in terms of the total number of prey items (line) in Devesa de I’Albufera between 1996 and 2003. very abundant in the study area and roost in col- onies, often in buildings. Long-eared Owls could capture them at emergence as pipistrelle bats leave the roosts in large groups, but return as single in- dividuals or in small groups and much more spaced over time. Although pipistrelle bats do not gather while foraging over large rice fields, they can form large aggregations when feeding along 18 n FWSS FWSS FWSS FWSSFWSS FWSS FWSS FWS 1995 1996 1997 1998 1999 2000 2001 2002 2003 Figure 3. Seasonal variation in the percent (on the total number of prey items) of bats taken by Long-eared Owls in Devesa de I’Albufera between 1995 and 2003. Below the bars and for each year, shading intensity of cells indicates the ranked percentage of rodents in the diet, from the highest (dark) to the lowest importance (light) in the seasonal diet. 450 Garcia et al. VoL. 39, No. 4 drainage channels or near street lamps (Blake et al. 1994), and the owls could hunt them there as well. Predation by nocturnal raptors on predictable accumulations of bats has been previously docu- mented (Barclay et al. 1982, Fenton et al. 1994, Hoetker and Gobalet 1999). The spatial distribution of bat remains during the owl breeding season was not random. Bat re- mains appeared in pellets from only five (A^ = 205 prey items) of 16 nests sampled. The mean per- centage of bats per positive nest sample was 17.6 ± 13.0% (SD) of prey items. Only in one owl nest did bats account for <11% of prey, and the maxi- mum observed was 37%. These figures make less plausible the idea of an opportunistic capture of bats as a result of accidental encounters (Ruprecht 1979) . Our results were consistent with the hypoth- esis of individual differences in ability to catch bats or with individual knowledge of the location of pi- pistrelle colonies, which may be more profitable to exploit than solitary bat species if the emergence of large numbers of bats increases hunting success (Fenton et al. 1994). Even small-sized pipistrelle bats (body mass = 7.5 g) could be a profitable prey for Long-eared Owls if available in large quantities. The biomass of 8 pipistrelle bats (60 g; the maxi- mum number of bats found per pellet) may satisfy two thirds of the daily energy needs (93.3 g for a 280 g owl; Wijnandts 1984), perhaps with little en- ergy expenditure during foraging. Biogeographic Pattern. We considered 21 diets for the Mediterranean region (34 410 prey items) and nine diets in the adjacent temperate region (22 386 prey items; Fig. 1). In the Mediterranean region, 38% of diets included bats as prey, whereas bats did not occur in any diet for the temperate region (Table 1). These differences in bat occur- rence were significant (G = 6.885, df — 1, P = 0.009). Even excluding our study in coastal Spain, where we found an unusually high quantity of bat remains, the mean proportion of bats in the Med- iterranean diets was significantly higher than in the diets of adjacent temperate sites (Mann-Whitney IT- test, Z = 1.98, P — 0.048). In the Mediterranean region, the overall importance of bats in the diet of Long-eared Owls (0.43% of prey-items) was at least twice as high as in other geographical areas. But the large number of bats in the diet of owls in our study area was very influential in this compar- ison. In fact, omitting our results from eastern Spain, bats only represent 0.06% of prey items found in the combined diets from the Mediterra- nean region, which is similar to figures found else- where. In the diets from North America (23 888 prey items) and temperate Europe, plus Iraq, re- viewed by Marti (1976), bats did not occur. In later reviews, Mikkola (1983) and Speakman (1991) found that bats accounted for <0.20% of prey items in Europe (67 805 prey items) and 0.05% in the British Isles (12 870 prey items). In the studies we reviewed, bat predation was re- stricted to latitudes 37— 43°N and altitudes 0—1400 m. Differences between localities in bat occurrence in the diet could not be attributed to a decline in bat species richness northwards, as species richness is almost constant at latitudes 35-50°N in Europe, which encompass all localities in Figure 1 (Perez- Barberia 1991, Mitchell-Jones et al. 1999). Howev- er, bat abundance increases with decreasing lati- tude and altitude (Perez-Barberia 1991, Kunz and Fenton 2003). Moreover, following the temporal pattern of insect availability, bats in Mediterranean environments show an extended activity season (Avery 1985, Altringham 1996, Blondel and Aron- son 1999). Indeed, bat predation occurs in all sea- sons (Table 2). If an extended period of activity and a higher abundance were indicators of in- creased availability of bats for owls, we would ex- pect increasing bats in the diet with decreasing lat- itude and altitude. We found no such correlations (latitude, = —0.05, P= 0.91; altitude, = —0.45, P = 0.26), but these analyses were based on small sample sizes {N = 8 diets containing bats) . Long-eared Owls preyed on nine of 29 bat spe- cies present in southern Europe (Mitchell-Jones et al. 1999; Table 2). Speakman (1991) suggests that large bat species would be more profitable prey than small ones, and therefore selected by raptors. However, Long-eared Owls consumed a variety of bat species, very different in body size, and there was no bias toward large species (Table 2) . All the bat species that owls consumed, except Tadarida te- niotis, forage low in open areas (Altringham 1996), just as Long-eared Owls do (Mikkola 1983, Tome 2003b) while hunting terrestrial prey on the wing (Marks et al. 1999). Excluding our results in coastal Spain, in Mediterranean environments, mean bat intake per diet, standardized as bats per 1000 prey items, was 3.4 individuals, suggesting that preda- tion is in most cases opportunistic (Ruprecht 1979). Comparable results have been obtained for Barn Owls (Tyto alba; Perez-Barberia 1991), which are regarded as opportunistic predators of bats. We conclude that bat aggregations could be a December 2005 Bat Predation by Long-eared Owls 451 locally important food source for some individual owls during certain periods, as exemplified by the population of Devesa de TAlbufera. More gener- ally, this evidence supports the view that Long- eared Owls may show substantial trophic plasticity, in contrast to their widespread recognition as a ro- dent specialist. In other words, their trophic re- sponse may be context-dependent rather than im- posed by morphological or behavioral constraints that typically affect all populations across the range of true specialists. At the geographical scale, bat abundance does not seem to reflect bat availability for Long-eared Owls, maybe because hunting strategies for pre- ferred prey such as rodents are not compatible with a regular exploitation of flying bats. Accord- ingly, bats occur in a number of diets across the Mediterranean region, but their contribution re- mains largely irrelevant. Acknowledgments We thank D. Almenar, M.A. Monsalve, and J. Quetglas for providing helpful information on bat ecology and R. Barclay, A, Gastello, D. Tome, and H. Ulmschneider for valuable comments that improved the manuscript. A.M. Garcia was supported by a grant from the Conselleria de Territori i Habitatge, Generalitat Valenciana. A. Rodri- guez benefited from a research contract by the Conseje- ria de Educacion, Junta de Andalucia. Literature Cited Alegre, J., A. Hernandez, and AJ. SAnchez. 1989. Datos sobre la dieta invernal del buho chico {Asio otus) en la provincia de Leon. Donana Acta Vert. 16:305-309. Alivizatos, H. and V. Goutner. 1999. Winter diet of the Barn Owl ( Tyto alba) and Long-eared Owl {Asia otus) in northeastern Greece: a comparison. J. Raptor Res. 33:160-163. Aloise, G. and D. Scaravelli. 1995. Ecologia alimentare del gufo comune, Asio otus, in un roost del basso Man- tovano. Avocetta 19:110. Altringham, J.D. 1996. Bats. Biology and behavior. Ox- ford University Press, Oxford, England. Amat, J.A. and R.C. Soriguer. 1981. Analyse comparative des regimes alimentaires de I’effraire, Tyto alba, et du moyen-duc, Asio otus, dans I’ouest de I’Espagne. Alau- 49:112-120. Araujo, J., J.M. Rey, A. Landin, and A. Moreno. 1974. Contribucion al estudio del Buho Chico {Asio otus) en Espaha. Ardeola 19:397—428. Avery, M.I. 1985. Winter activity of pipistrelle bats. J. Anim. Ecol. 54:721-738. Baker, J.K. 1962. The manner and efficiency of raptor depredation on bats. Condor 64:500-503. Barclay, R.M.R., C.E. Thomson, and FJ.S. Phelan. 1982. Screech Owl, Otus asio, attempting to capture little brown bats, Myotis lucifugus, at a colony. Can. Field-Nat 96:205-206. Bertolino, S., E. Ghiberti, and A. Perrone. 2001. Feed- ing ecology of the Long-eared Owl {Asio otus) m northern Italy: is it a dietary specialist? Can. J. Zool 79:2192-2198. Blake, D., A.M. Hutson, P.A. Racey, J. Rydell, and JR Spearman. 1994. Use of lamplit roads by foraging bats in southern England. /. Zool. 234:453-462. Blanco, J.C. 1998. Mamfferos de Espana I. Insectivoros, Quiropteros, Primates y Carnivoros de la peninsula Iberica, Baleares y Canarias. Planeta, Barcelona, Spain. Blondel, J. and J. Aronson. 1999. Biology and wildlife of the Mediterranean region. Oxford University Press, Oxford, England. Capizzi, D., L. Caroli, and P. Varuzza. 1998. Feeding habits of sympatric Long-eared Owl, Asio otus. Tawny Owl, Strix aluco, and Barn Owl, Tyto alba, in a medi- terranean coastal woodland. Acta Ornithol. 33:85-92 and L. Luiselli. 1996. Feeding relationships and competitive interactions between phylogenetically un- related predators (owls and snakes). Acta Oecol. 17: 265-284. Casini, L. and a. Magnani. 1988. Alimentazione inver- nale di gufo comune, Asio otus, in un area agricola dell’Emilia orientale. Avocetta 12:101—106. Clavero, M., j. Prenda, and M. Delibes. 2003. Trophic diversity of the otter {Lutra lutra L.) in temperate and mediterranean freshwater habitats./. Biogeogr. 30:761- 769. Corral, J.F., J. A. Cortes, and J.M. Gil. 1979. Contribu- cion al estudio de la alimentacion de Asio otus en el sur de Espana. Donana Acta Vert. 6:179-190. Costa, M., J.B. Peris, and R. Figuerola. 1982. La vege- tacion de la Devesa de FAlbufera. Monografies 01. Delegacio del Medi Ambient i Espais Oberts, Ajunta- ment de Valencia, Valencia, Spain. Delibes, M., P. Brunet-Lecomte, and M. Manez. 1984. Datos sobre la alimentacion de la lechuza comun {Tyto alba), el buho chico {Asio otus) y el mochuelo {Athene noctua) en una misma localidad de Castilla la Vieja. Ardeola 30:57-63. Emberger, L., H. Gaussen, M. Kassas, and A. Philippis 1963. Carte bioclimatique de la zone mediterraneen- ne (Etude ecologique de la zone mediterrraneenne). UNESCO-FAO, Paris, France. Fenton, M.B., I.L. Rautenbach, S.E. Smith, G.M. Swa- nepoel, J. Grosell, and j. van Jaarsveld. 1994. Rap- tors and bats: threats and opportunities. Anim. Behav. 48:9-16. Futuyma, D.J. AND G. Moreno. 1988. The evolution of ecological specialization. Annu. Rev. Ecol. Syst. 19:207- 233. Galeotti, P. AND L. Canova. 1994. Winter diet of Long- eared Owl {Asio otus) in the Po plain (northern Italy) . J. Raptor Res. 28:265-268. 452 Garcia et al. VoL. 39, No. 4 Garcia, A.M. and F. Cervera. 2001. Notas sobre la varia- cion estacional y geografica de la dieta del Bubo Chi- co Asia otus. Ardeola 48:75-80. Gerdol, R., E. Mantovani, and F. Perco. 1982. Prelimi- nary comparative study on dietary habits of three Stri- giform species in the Carso Triestino. Riv. Ital. Ornitol. 52:55-60. and F. Perco. 1977. Ecological observations on the Long-eared Owl Asio otus (L) in north-eastern It- aly. Boll. Soc. Adriat. Sci. 41:37-59. Guidoni, R., D. Capizzi, L. Caroli, and L. Luiselli. 1999. Feeding habits of sympatric owls in an agricultural and forested landscape of central Italy. Folia ZooL 48: 199-202. Hanski, L, L. Hansson, and H. Henttonen. 1991. Spe- cialist predators, generalist predators, and the micro- tine rodent cycle. J. Anim. Ecol. 60:353-367. Herrera, C.M. and F. Hiraldo. 1976. Food-niche and trophic relationships among European owls. Ornis Scand. 7:29-41. Hoetker, G.M. and K.W. Gobalet. 1999. Predation on Mexican free-tailed bats by Burrowing Owls in Cali- fornia. Raptor Res. 33:333—335. Kayser, Y. and N. Sadoul. 1996. Cas original de preda- tion exercee sur des colonies de larides par le hibou moyen-duc {Asio otus) dans les salins d’Aigues-Mortes (Camargue, France). Nos 43:485-496. Korpimaki, E. 1985. Rapid tracking of micro tine popu- lations by their avian predators: possible evidence for stabilizing predation. Oikos 45:281-284. and C.D. Marti. 1995. Geographical trends in tro- phic characteristics of mammal-eating and bird-eating raptors in Europe and North America. Auk 112:1004- 1023. Kunz, T.H. and B. Fenton. 2003. Bat ecology. Chicago University Press, Chicago, IL U.S.A. Lopez Gordo, J.L., E. Lazaro, and A. Fernandez Jorge. 1977. Comparacion de las dietas de Strix aluco, Asio otus y Tyto alba en un mismo biotopo de la provincia de Madrid. Ardeola 23:189-221. Lundberg, a. 1979. Residency, migration and a compro- mise: adaptations to nest site scarcity and food spe- cialization in three Fennoscandian owl species. Oeco- logia 41:273-281. Malavasi, D. 1995. Preliminary data on the diet of the Long-eared Owl {Asio otus) in an intensively cultivated agricultural land of the Po plain near Mantova. Suppl. Ric. Biol. Selvaggina 22:255-256. Marariu, D., I. Andreescu, and V. Nesterov. 1991. Les composants de la nourriture d’hiver d' Asio otus otus (L., 1758) du nord-est de Bucarest (Roumanie). Trav. Mus. Hist. Nat. “Gngore Antipa” 51 Alb— 420. Marks, J.S., RJ. Cannings, and H. Mikkola. 1999. Family Strigidae. Pages 76—151 in], del Hoyo, A. Elliott, and J. Sargatal [Eds.] , Handbook of the birds of the world. Vol. 5. Lynx Editions, Barcelona, Spain. Marti, C.D. 1976. A review of prey selection by the Long- eared Owl. Condor 78:331-336. Martin, R., A. Rodriguez, and M. Delibes. 1995. Local feeding specialization by badgers {Meles meles) in a mediterranean environment. Oecologia 101:45-50. Mezzavilla, E. 1993. Study on the winter diet of the Long-eared Owl, Asio otus, in the Treviso province. Lav. Soc. Ven. Sc. Nat. 18:173-182. Mikkola, H. 1983. Owls of Europe. T. & A.D. Poyser, Cal- ton, U.K. Mitchell-Jones, A.J., G. Amori, W. Bogdanowicz, B Krystufek, P.J.H. Reijnders, F. Spitzenberger, M. Stubbe, J.B.M. Thissen, V. Vohralik, and J. Zima. 1999. The atlas of European mammals. T. and A.D. Poyser, London, England. Pai.omo, J.L. AND J. Gisbert. 2002. Atlas de los mamiferos terrestres de Espaha. Direccion General de Conser- vacion de la Naturaleza-SECEM-SECEMU, Madrid, Spain. Perez-BarberIa, F.J. 1991. Influencia de la variacion lati- tudinal en la contribucion de los murcielagos (Chi- roptera) a la dieta de la lechuza comun {Tyto alba). Ardeola 38:61—68. Plini, P. 1986. Primi dati sull’alimentazione del gufo co- mune, Asio otus, nel Lazio. Avocetta 10:41-43. Reviiia, E. and E. Palomares. 2002. Does local feeding specialization exist in Eurasian badgers? Can. J. Zool. 80:83-93. Riga, F. and D. Capizzi. 1999. Dietary habits of the Long- eared Owl, Asio otus, in the Italian peninsula. Acta Or- nithol. 34:45-51. Roulin, a. 1996. La forme fouisseuse du campagnol te- rrestre {Arvicola terrestris) : une proie dominante chez le hibou moyen-duc {Asio otus). Nos Oiseaux 43:289- 294. Ruprecht, A.L. 1979. Bats (Chiroptera) as constituents of the food of Barn Owls {Tyto alba) in Poland. Ibis 121:489-494. San Segundo, C. 1988. Notas sobre la alimentacion del Buho Chico {Asio otus) en Avila. Ardeola 35:150-155. SarA, M. 1990. Aspetti della nicchia ecologica degli stri- giformi in Sicilia. Naturalista Sicil. 14:109-122. Speakman, J.R. 1991. The impact of predation by birds on bat populations in the British Isles. Mammal Rev. 21:123-142. SuBLiMi, S. AND L. Scalera. 1991. Dati sulla predazione di gufo comune {Asio otus) svernanti in una dolina pugliese. Suppl. Ric. Biol. Selvaggina 17:119-122. Tome, D. 2003a. Functional response of the Long-eared Owl {Asio otus) to changing prey numbers: a 20-year study. Ornis Fenn. 80:63—70. . 2003b. Nest site selection and predation driven despotic distribution of breeding Long-eared Owls Asio otus. J. Avian Biol. 34:150-154, Veiga, J.P. 1980. Alimentacion y relaciones troficas entre la Lechuza Comun {Tyto alba) y el Buho Chico {Asio December 2005 Bat Predation by Long-eared Owls 453 otus) en la sierra de Guadarrama (Espana). Ardeola'2,b: 113-142. Weber, J.M., S. Aubry, N. Ferrari, C. Fischer, N. Lachat Feller, J.S. Meia, and S. Meyer. 2002. Population changes of different predators during a water vole cy- cle in a central European mountainous habitat. Eco- graphy 25:95-101. WijNANDTS, H. 1984. Ecological energetics of the Long- eared Owl (Asia otus). Ardea 72:1-92- Zar, J.H. 1984. Biostatistical analysis, 2nd Ed. Prentice Hall, Englewood Cliffs, NJ U.S.A. Received 12 August 2004; accepted 3 July 2005 Associate Editor: James R. Belthoff Short Communications J. Raptor Res. 39 (4) :454-457 © 2005 The Raptor Research Foundation, Inc. Differential Effectiveness of Playbacks for Little Owls {Athene noctua) Surveys Before and After Sunset Joan Navarro^ Departament de Biologia Animal (Vertebrats) , Facultat de Biologia, Universitat de Barcelona, Avda, Diagonal 645, 08028 Barcelona, Spain Eduardo Minguez, David Garcia, Carlos Villacorta, Francisco Botella, Jose Antonio Sanchez-Zapata, Martina Carrete, and Andres Gimenez Area de Ecologia, Departamento de Biologia Aplicada, Universidad Miguel Hernandez, Avda de la Universidad s/n, 03202, Elche, Spain Key Words: Little Owl, Athene noctua; survey method', population surveys', playback effectiveness. Most nocturnal owls respond to broadcast of conspe- cific recordings, and this technique may be used to study their behavior (Galeotti and Pavan 1993, Galeotti et al. 1997), to map territories (Finck 1990, Lane et al. 2001), to identify individuals (Galeotti and Sacchi 2001, Delport et al. 2002) , or to study population trends (Exo and Hen- nes 1978, Martinez and Zuberogoitia 2004). Several fac- tors affecting the effectiveness of playback techniques or spontaneous vocalizations have been identified, includ- ing response distance (Proudfoot et al. 2002), season (Zuberogoitia and Campos 1998), weather (Lengane and Slater 2002) , gender, and social status (Appleby and Red- path 1997). However, only a few studies have investigated systematically how these factors influence playback meth- odology (McGarigal and Fraser 1985, Redpath 1994, Cen- tili 2001). When comparing the accuracy of sampling us- ing spontaneous owl vocalizations or conspecific playbacks, several authors demonstrated that sampling error increases when using spontaneous calls (McGarigal and Fraser 1985, Haug and Didiuk 1993, Redpath 1994). However, for Eurasian Eagle-Owl {Bubo bubo), passive au- ditory surveys provide better results than surveys based on broadcast calls (Penteriani and Pichera 1991, Marti- nez and Zuberogoitia 2002). The Little Owl {Athene noctua) is a territorial species widely distributed in Palearctic regions. This small raptor inhabits a wide variety of semi-open areas, from steppes and stony semideserts to farmlands and open woodlands, and villages and urban areas (Cramp 1985). Little Owls prey on insects, small mammals, and birds, and hunt both during diurnal and nocturnal hours (Negro et al. ^ Email address: joannavarro@ub.edu 1990) . Researchers have surveyed Little Owls by listening to spontaneous vocalizations or by playing typical calls to provoke the territorial vocalization after sunset (Finck 1990, Exo 1992, Mastrorilli 1997, Zuberogoitia and Cam- pos 1998, Verwaerde et al. 1999, Pirovano and Galeotti 1999) or before sunset (Martinez and Zuberogoitia 2004) . Territorial defense is performed mostly by males, which are more vocal than females (Mikkola 1983, Finck 1990, Zuberogoitia and Campos 1998). Here, we examine the effectiveness of the playback method to detect Little Owls before or after sunset. Spe- cifically, we tested whether broadcast before or after sun- set would elicit the greater response frequency and whether duration of playback affected the Little Owl re- sponse rate. We also offer some suggestions to improve the survey methodology. Study Area and Methods We conducted the study in Clot de Galvany Council Natural Park (southeastern Spain, Province of Alicante). The study area (ca. 650 ha.) was characterized by a mo- saic of shrubs, saline grasslands, and mixed forest, inter- spersed with extensive abandoned arboreal cultures such as almond and olive trees. The area exhibits a semiarid Mediterranean climate (Sancho and Lopez 2002). Between 19 April-17 May 2002 (the courtship and ter- ritorial defense period; Mikkola 1983, Finck 1990), we surveyed twice (before and after sunset) for Little Owls at 14 permanent stations during five sessions (10 d). Each day, survey stations were sampled with three differ- ent “survey” experiments, seven before and seven after sunset. The first survey was made 2 hr before sunset (1900-2100 H). The second survey was done 2 hr after sunset (2130-2330 H). These survey experiments were’ (1) Spontaneous Calls: the observer listened for 2 min, registering the response rate (number of different Little Owls heard); (2) First Playback: after the spontaneous calls trial, a territorial intrusion was simulated by broad- casting territorial calls of Little Owl for 2 min using a 454 December 2005 Short Communications 455 cassette-player. Then, the observer listened for 1 min, re- cording the response; (3) Second Playback: after the first playback (1 min later), a new territorial intrusion was simulated by broadcasting territorial calls for 2 min by using a cassette-player. The observer listened for 1 min, recording the total response rate (response rate of first and second playback) . The sequence of stations surveyed was reversed every other visit. To reduce the number of Litde Owls that could be potentially counted twice, the distances between survey stations were at least 500 m (Finck 1990, Exo 1992, Centili 2001). To avoid differenc- es associated with possible bias in sound direction, we always used the same cassette-player (power: 4 watts, Sony WN-FX 195, Barcelona, Spain), placed on the ground with the loudspeakers directed upwards. To minimize the potential of lower response rates (i.e., owls becoming less responsive because they habituated to our broadcasts), we used territorial male calls from five different individ- ual owls (Roche 1996, SEO 1998, Llimona et al. 2002). We did not conduct experiments on windy or rainy days. Because data were not normally distributed, we used nonparametric tests for statistical analysis (Zar 1996). To avoid pseudoreplication, we used the mean of response rate for each survey station. Statistical analyses were car- ried out using the SPSS statistical package (SPSS for Win- dows 1999). Two-tailed P-values were used throughout and statistical significance was set at P < 0.05. Results Before sunset, spontaneous Little Owl calls were heard at only two stations (Table 1). In contrast, Little Owls were detected after sunset by spontaneous calls at nine stations (Table 1). As expected, more Little Owls sang spontaneously after sunset than before (Kruskal-Wallis test, = 10.39, df = 1, P < 0.001). Numbers of Little Owls detected by passive auditory surveys were signifi- cantly lower than those detected by playback surveys, both before and after sunset (Fig. 1). These differences were shown in both the comparison with the first play- back (before sunset: = 22.07, df = 1, P< 0.001; after sunset: x^ = 19.84, df = 1, P < 0.001), and the second playback (before sunset: x^ = 28.17, df = 1, P < 0.001; after sunset: x^ = 25.73, df = 1, P < 0.001). Playback surveys detected more individuals after sunset than be- fore (first playback before vs. after sunset: x^ = 6.83, df = 1, P < 0.001; second playback before vs. after sunset: X^ = 4.51, df = 1, P = 0.03), but there were no differ- ences between the two playback experiments within each period (first playback vs. second playback before sunset: X^ = 0.55, df = 1, P = 0.46, first playback vs. second playback after sunset: x^ = 0.03, df = 1, P = 0.87; Fig. 1 ). Discussion Our results strongly suggest that nocturnal broadcast surveys were the most effective method for surveying Lit- tle Owls, both for detecting presence and counting in- dividuals or territories (Zuberogoitia and Campos 1998, Verwaerde et al. 1999, Centili 2001, van Nieuwenhuyse et 2.0 1 Before Sunset After Sunset Figure 1. Call rates (calling means for Little Owls per survey station) detected without using playback (sponta- neous calls) , using first and second playback both before and after sunset. al. 2002). These results are in accordance with similar studies with other owls: Barred Owl {Strix varia; Mc- Garigal and Fraser 1985), Tawny Owl {Strix aluco; Red- path 1994), Burrowing Owl {Athene cunicularia; Haug and Didiuk 1993), Ferruginous Pygmy-Owl {Glauddium brasi- lianum; Proudfoot and Beasom 1996), and Long-eared Owl {Asio otus; Martinez et al. 2002). Interestingly, our study indicated that Little Owls re- sponded at a similar rate in the first and the second play- back (before and after sunset). The fact that our first playback experiment, broadcasting for 2 min, elicited a similar number of owls as when both playbacks were in- cluded, suggested that 2 min of continuous playback was sufficient for detecting Little Owls, which was a shorter period than that used in other studies of this species (5 min, Zuberogoitia and Campos 1998; 4 min, Verwaerde et al. 1999; 3 min, Centili 2001). Detection rate of Little Owls after sunset was less vari- able and higher than before sunset (Table 1, Fig. 1). Thus, before sunset, surveys may underestimate the num- ber of breeding Little Owls in an area. However, the use of playback before sunset would be useful in finding ter- ritories and nests because individuals might be observed when calling at perches (Martinez and Zuherogoitia 2004, pers. obs.). Estimating the breeding density of owl species is an important part of population studies, and comparisons are widely used to assess the abundance of a species across years and geographical cireas. Biased estimates of breeding pair density are misleading and prevent com- parisons between studies. Thus, increased efficiency of survey methods and knowledge of potential error is nec- essary. Our results suggest broadcasting 2 min of conspe- Table 1. Mean call rates (number of different owls calling at each survey station /number of trials ±SD) of Little Owls at Clot de Galvany Park, Alicante Province, Spain. Data were obtained during five visits/ survey station; protocol included recording spontaneous calls followed by playback of conspecific calls. 456 Short Communications VoL. 39, No. 4 q 00 03 m in m o o i-H t-H in CM q 00 00 Til 00 in m q q q CJ5 j Cm I-H d d d d d d t-H I-H r-H d 1-H d d Q + 1 + 1 +1 +1 +1 + 1 +1 +1 +1 + 1 +1 + 1 + 1 +1 +1 o o o o o o o o O o o o o o esn 0 0^1 cy CD 00 CM q CD (M cy q q q q q u w c6 d d GO l-H iM d ^H f-H d 1-H d 1-H o Oh C/5 4 (U .kJ cd OJ (5 OJ 43 4h o a; 4 0 Oh (U ■O (U 43 H 3 0 H 4/ o w Q z: o H w § H H ^ Q ^ H e z O w 0, w Q ^ ^ 2 Lh T [jj M L_J MH l-H Pi t4 n ^ 2 O i ^ „ - 2 o p 52 Pi u S D P 0 4 o O c« cc o bD o cj 43 1 43 2 3 o u xi 4 3 cn 43 H I— 1 VJ (d (d CD > cd JP 2 ^ T 3 cd O ■a ^ p kH (L) cn lU O XS CD ^ O Oh 00 U 5:1 0 o cb O D ‘y hh — ^ 3 -5 pS tn O • ^ ct u bo bo V V CM SO flj o 43 4h ° 'v B “ a o O a o 43 cd 43 o 0 ■O 03 (U Oh 4S t) U CCI 43 ^ . c/5 t 3 cd Th CD Cd X CO g 00 C5/ CD 00 CD 4J 3 V O 4 V h/5 rb' V u 1 'H T3 'bc 3 V > -b' u 2 -HH Oh V qj 44 • ^ o 44 OJ C/J u 3 43 C/5 C7i d OJ H H 4 o CM a o Oh ■4 Oh 05 c /5 45 cd 43 05 C /5 0 CM 05 4 Ph a 44 GO 05 a 44 J> ID IM I> r- GO TfH o o o i-H o o 4 4 cS Cd Cd 4 4 Oh Oh O. 0/ 0) 0/ c /5 c /5 c /5 CD O CD CM CM CM ^ P P 01 ^ OJ GO C/5 ^ C/5 O 00 00 (M CD 05 2 3 a ttJ t4 CD CM CD 05 be ^ .a ^ 4 3 Oh C pCG 4 464 Short Communications VoL. 39, No. 4 -52°S -48°S -44°S Figure 1. The autumn migration of the three Greater Spotted Eagles in Europe determined by satellite telemetry in 1996; dates of arrival at selected points en route are indicated. December 2005 Short Communications 465 Table 2. The outward migration of the juvenile Greater Spotted Eagle (see Fig. 1) 1996 was determined by satellite telemetry. Beginning and End of Each Stage"* Length of the Different Stages IN km Duration OF Each Stage (days) Mean Length OF Daily Flight Distances Countries Transversed ca 21/22 Sept 26 Sept: 0044 H 257 km ? ? Poland 26 Sept: 0044 H 30 Sept: 1134 H Roosting 4.5 — Poland 30 Sept: 1134 H 4 Oct: 1534 H 242 km 4.5 54 km/day Poland 4 Oct: 1720 H 9 Oct: 0005 H 76 km 4 19 km/day Poland and Slovakia 9 Oct: 0547 H 13 Oct: 0600 H 329 km 4 82 km/day Slovakia and Hungary 13 Oct: 0600 H 17 Oct: 1327 H 202 km 4.5 45 km/day Hungary and Croatia 17 Oct: 1738 H 22 Oct: 0109 H 361 km 4 90 km/day Bosnia - Herzegovina 22 Oct: 0109 H 26 Oct: 0619 H 220 km 4 55 km/day Montenegro and Albania Each migration stage was 4 d, based on the duty cycle of the satellite transmitter. Location data provided by Argos Service, Inc (Toulouse, France). Dementiev and Gladkov (1951) and Ivanov et al. (1951) believed that Greater Spotted Eagle families de- parted together on migration. We know of no study in the literature on this species, or any other raptor, in which the dates of departure on migration, the break up of the family, and their combined or separate migrations have been investigated by satellite telemetry. The family of Greater Spotted Eagles studied here clearly broke up when leaving the breeding territory. We also have studied a pair of the closely-related Lesser Spotted Eagle using this method over several years. The members of this pair migrated separately in 1997—98 and 1998-99 and overwintered ca. 1000 km apart both years in southern Africa (Meyburg et al. 2006). However, in this case, the offspring were not tracked. Registro de la Ruptura Familia, Partida y Migracion DE Otono de Una Familia de Aquilas Moteadas {Aquila clanga) en Europa Usando Telemetria Satelital Resumen. — ^Ambos adultos y el polluelo de una famila de Aquila clanga fueron estudiados mediante telemetria sa- tehtal en el noreste de Polonia para determinar la fecha de inicio de su migracion, la disolucion de la familia y sus patrones de migracion combinados e independientes. La familia se disolvio al abandonar el territorio de cria. La hembra fue la primera en partir, el inmaduro lo hizo alrededor dos o tres dias mas tarde y el macho partio una semana despues de la hembra. Los adultos se diri- gieron directo al Bosforo. El inmaduro recorrio 1687 km hasta Albania, donde aparentemente murio a fines de octubre. [Traduccion del autor] Acknowledgments The authors wish to thank the Deutsche Forschungs- gemeinschaft (DFG) in Bonn, Germany, for its unstinting financial support of the Satellite Telemetry Greater Spot- ted Eagle Project, the Polish Environment Ministry in Warsaw, and the administration of the Biebrza National Park for kindly allowing us to study and, in particular, to trap adult birds. We are also grateful to the Poznan Zoo for providing a live eagle-owl to use as a decoy in trapping the eagles, to Joachim and Hinrich Matthes, as well as Mike McGrady, for help in the field, and to Robin Chan- cellor for linguistic help. Prof. Kai Graszynski kindly made comments on the first draft of the manuscript, as well as three anonymous referees. Literature Cited Allison, J.B. 1997. Mapit. Version 2.0. Allison Software, Apollo, PA U.S.A. Alonso, J.C., L.M. Gonzalez, B. Heredia, andJ.L. Gon- zalez. 1987. Parental care and the transition to in- dependence of Spanish Imperial Eagles {Aquila helia- cd) in Donana National Park, southwest Spain. Ibis 129:212-224. Bahat, O. 1992. Post-fledging movements of Golden Ea- gles {Aquila chrysaetos homeyeri) in the Negev Desert, Israel, as determined by radio-telemetry. Pages 612- 466 Short Communications VoL. 39, No. 4 621 in LG. Priede and S.M. Swift [Eds.], Wildlife te- lemetry: remote monitoring and tracking of animals. Ellis Horwood Ltd., New York, NYU.S.A. Bloom, P,H. 1987. Capturing and handling raptors. Pag- es 99-123 in B.A.G, Pendleton, B.A. Millsap, K.W. Cline, and D.M. Bird [Eds.], Raptor management techniques manual. National Wildlife Federation, Washington, DC U.S.A. Brown, L. and D. Amadon. 1968. Eagles, hawks and Fal- cons of the world. Vol. 1. Country Life Books, Felt- ham, England. Bustamante, J. 1995. The duration of the post-fledging dependence period of Ospreys {Pandion haliaetus) at Loch Garten, Scotland. Bird Study 42:31-36. Clark, W.S. 1981. A modified dho-gaza trap for use at a raptor banding station. J. Wildl. Manage. 45:1043- 1044. Dementiev, G.P. and N.A. Gladkov. 1951. Birds of the Soviet Union. Moscow, Russia. (In Russian). Hamerstrom, F. 1963. The use of Great Horned Owls in catching Marsh Hawks. Proc. Int. Ornithol. Congr. 13: 866-869. Ivanov, A.I., E.V. Kozlova E.V., L.A. Portenko, and A.Y. Tugarinov. 1951. Birds of Soviet Union. Vol. 1. Iz- datelstvo Akademi Nauk SSSR, Moscow, Russia. (In Russian). Meyburg, B.-U., X. Eichaker, C. Meyburg, and P. Pai- LLAT. 1995a. Migrations of an adult Spotted Eagle tracked by satellite. Brit. Birds 88:357-361. , C. Meyburg, and J. Matthes. 2006. Annual cycle, timing and speed of migration of a pair of Lesser Spotted Eagles {Aquila pomarina) tracked by satellite. J. Ornithol. In press. Mizera, T, G. Maciorowski, and B.-U. Meyburg. 2001. {Aquila clanga (Pallas, 1811) Greater Spotted Eagle] Pages 145-148 in Z. Glowadnski [Ed.], Polish red data book on animals. Vertebrates. Panstwowe Wydaw- nictwo Rolnicze I Lesne, Warszawa, Poland. (In Polish with English summary). Morvan, R. and F. Dobchies. 1990. Dependance de jeu- nes Aigles de Bonelli {Hieraaetus fasciatus) apres I’en- vol: variations individuelles. Alauda 58:150-162. Rafanomezantsoa, S.A. 2000. Behavior and range move- ments during the post-fledging dependence period of the Madagascar Fish-Eagle {Haliaeetus vociferoides) . Pages 113-119 inR.D. Chancellor and B.-U. Meyburg [Eds.], Raptors at risk. Hancock House and WWGBP, Berlin, Germany. Real, J., S. Manosa, and J. Godina. 1998. Post-nestling dependence period in the Bonelli’s Eagle {Hieraaetus fasciatus). Ornis Fennica 75:129-137. Received 24 November 2003; accepted 5 September 2005 J Raptor Res. 39(4):466-47l © 2005 The Raptor Research Foundation, Inc. Seasonal Patterns of Common Buzzard {Buteo buteo) Reiatwe Abundance and Behavior in PoLLiNO National Park, Italy Massimo Pandolfi, Alessandro Tanferna, and Giorgia GaibanP Istituto di Zoologia, Universitd di Urbino, via Oddi 21, 61029 Urbino, Italy Key Words: Common Buzzard', Buteo buteo; relative abun- dance] roadside surveys. Nest-site selection and habitat use have been described m the Common Buzzard {Buteo buteo) by several authors (e.g., Penteriani and Faivre 1997, Kruger 2002, Lohmus 2003, Bustamante and Seoane 2004, Sergio et al. 2005), but few studies have documented annual variations in the abundance and habitat associations of this species (Meu- nier et al. 2000). We conducted monthly roadside surveys of Common Buzzards in a mountainous area of southern Italy. Al- ^ Present address and corresponding author: Museo di Storia Naturale, Dipartimento di Biologia Evolutiva e Funzionale, Universita di Parma, Via Farini 90, 43100 Parma, Italy; e-mail address: gaibani@biol.unipr.it though roadside surveys have well-known limitations (e.g., Andersen et al. 1985, Fuller and Mosher 1987, Mill- sap and LeFranc 1988, Vinuela 1997), they remain a use- ful technique for monitoring local abundance and distri- bution of raptors (Fuller and Mosher 1987, Ellis et al. 1990). Because roadside surveys are easy to conduct, they can be carried out at frequent intervals. Here, we present results from monthly roadside surveys of Common Buz- zards. Using these data, we examine habitat associations, describe seasonal patterns of Common Buzzard behavior and abundance and, in particular, discuss the effective- ness of roadside surveys to monitor changes in abun- dance. Methods The Common Buzzard (hereafter buzzard) surveys were conducted from October 2000-September 2001 in Polhno December 2005 Short Communications 467 transects I — I Pollino National Park % § I % I 0,0 4,0 8,0 12,0 16,0 20,0 km Figure 1. Locations of seven routes (thick dark lines) used for roadside surveys in Pollino National Park, southern Italy, in 2000-01 (thin lines indicate the boundaries). National Park (39°58'N, 16°08'E), a 1821 km^ area located in the southern Italian Apennines (Fig. 1). The elevation ranges from 170-2266 m. The land uses include farmlands and oak woods {Quercus ilex, Q. pubescens, Q. cenris) in the northern section of the park and grassland and beech woods {Fagus sylvatica) in the southern portion. Over the study period, the mean monthly temperature was 14.5°C, with a mean of 28.8°C during July— September and 9.7°C during October-February. The annual rainfall in the 12 mo of the survey was 841 mm. We surveyed buzzards along seven paved roads (Fig. 1) selected randomly with restrictions (Caughley and Sin- clair 1994). Specifically, we rejected routes adjoining those roads previously chosen. Each road was surveyed once each mo, during the third or fourth wk of the mo and only on calm and clear days. We did not sample on days with snow, rain, fog, or strong winds. Each month, the routes were surveyed over 4 d by means of two cars, each one with a driver and two observers. All surveys were conducted in the morning (0900-1200 H), typically the best time to count raptors (Robbins 1981). We drove at a speed of 40—45 km/hr and stopped the car to con- firm each sighting. For each buzzard detected, we re- corded if it was flying or perched and if it was alone or with other buzzards. Also, we discounted any buzzard that may have represented a re-sighted bird. We calculated the relative abundance as the number of buzzards seen per 100 km sampled. For abundance computations we excluded the stretches of roads lined by trees within tunnels, forests, or villages. Therefore, al- 468 Short Communications VoL. 39, No. 4 Table 1. Number of Common Buzzards recorded along seven routes in Pollino National Park (southern Italy, 2000-01). For computation of relative abundance, we used the survey length obtained discounting the stretch- es of roads lined by trees or passing through tunnels, forests, or villages. Routes Length ( km) OF Routes Survey Length ( km) OF Routes Monthly Mean ±SD OF Relative Abundance 1 57.8 40.4 11.2 ± 1.8 2 18.3 18.3 10.2 ± 2.0 3 44.4 28.9 6.7 ± 2.0 4 47.3 34.0 12.7 ± 2.7 5 38.5 38.3 7.7 ± 1.6 6 43.8 24.2 20.1 ± 5.7 7 65.1 53.6 9.3 ± 1.2 though the total road length was 315.2 km, we consid- ered for analysis only 237.6 km (survey km in Table 1). For the analysis of buzzard habitat associations, we re- ported the sightings as presence/absence in a 1 X 1 km UTM grid. We created a 1 km buffer on both sides of each route and we considered only buzzards observed inside this buffer. We analyzed the habitat associations within this buffer using the Corine Land Cover 1:100 000 digital map (Legend level 3, Ministero dell’Ambiente e del Terri to rio — Ente Parco), identifying 10 land cover types which we then pooled into four general vegetation cover types: (1) arable land (cultivated areas regularly plowed and generally under a rotation system), (2) het- erogeneous agricultural areas (areas principally occupied by agriculture, interspersed with natural areas), (3) for- ests, and (4) shrub or herbaceous vegetation (Table 2). In each 1X1 km grid cells or portions of a cell included inside the buffer, we calculated the surface of each cover type by means of a Geographical Information System (GIS) analysis (Geomedia Professional 2002). Buzzard habitat associations were analyzed in four periods: Feb- ruary-April (courtship), May-July (incubation and nest- ing), August-September (post-fledging), and October- January (winter). Although observers recorded the number of individual buzzards sighted, we only consid- ered their presence/absence in each grid cell. We used the Friedman repeated measures analysis of variance (F,.) to detect any difference in the relative abun- dance among months and among periods. Using the same test, we evaluated if the number of flying or perched buzzards varied among months or periods. We used the Kruskal-Wallis test to detect any difference in the relative abundance among routes and the Mann- Whitney U test to ascertain whether flying buzzards were observed more frequently than perched buzzards. We used the arcsin-transformation to convert the proportion of sightings composed of buzzard groups. The Kolmo- gorov-Smirnov one-sample test (Z) was used to examine if the distribution of relative frequencies was uniform. Finally, we used a stepwise logistic regression to deter- o o o O, 3 iso "u 3 ~a u 0 3 V u Cu T3 Sh 01 N M 3 PQ 3 O s 6 o U t3 3 33 Ki 33 3 OJ o V .3 Cu • rH 33 73 3 O ♦ rH 'o (h tJ3 3 ’u O "a X CD 00 CD d d d 3 o 73 Q ^ I ^ § 0 ^ S 1 CM (L) 3 ^ I cd OJ d •ui 3 u bo cs o ^ bo S ’S 5 Si Shrub or herbaceous vegetation 0.83 1.14 0.53 1.60 0.20 3.79 0.80 0.84 ^ P = statistical significance of the Wald statistic, a chi-square distribution used to ascertain if a variable is a significant predictor of the outcome (presence or absence of buzzards; Field 2000) . Exp (B) = indicator of the change in odds (probability of an event occurring divided by the probability of that event not occurring) resulting from a unit change in the predictor (Field 2000). December 2005 Short Communications 469 Figure 2. Monthly mean ±SD of the relative abundance of Common Buzzards recorded along seven routes in Pollino National Park, southern Italy (2000-01). The dark columns show the relative abundance calculated consid- ering both grouped and single individuals; the white columns are the relative abundance calculated based on single individuals only. mine whether the probability of detecting buzzards var- ied among the four cover types in each of the periods. Means are presented ±SD. The nonparametric tests were from Siegel and Castellan (1988), and the logistic re- gression analysis followed Field (2000). All of the statis- tical tests were performed with SPSS 10.0 (SPSS 2000). Results During the study period, we recorded 328 buzzard sightings. The mean relative abundance per roadside sur- vey was 11.1 ± 4.4 {N = 84) buzzards/ 100 km (Table 1). Buzzard abundance varied among months {F^ — 23.1; df = 11; P < 0.05; Fig. 2) and, marginally, among periods (Fr = 7.6; df = S, P = 0.054). In particular, a post-hoc multiple comparisons test showed that the abundance was greater during the courtship period than in the other three periods and in the incubation-nesting period than in the winter (P < 0.05 for all comparisons). However, the relative abundance estimates showed a high variation for all periods (17.2 ± 11.9 for courtship, 10.2 ± 5.0 for incubation and nesting, 7.2 ± 4.3 for post-fledging, and 9.0 ± 2.4 for winter). We found no difference (x^ = 10.3; df = 6; P = 0.11) in abundance among routes, although the survey routes crossed different land cover types. We detected more buzzards flying (87.2%) than perched (12.8%; U = 1.5; N = 24; P < 0.0001). The number of perched buzzards did not vary among months (p. = 15.6; df = 11; P > 0.05), whereas the number of flying buzzards did (P^ = 21.0; df = 11; P< 0,05) (Fig. 3), In addition, the number of flying buzzards varied among periods (p. = 10.8; df = 3; P< 0.05); during courtship, flying buzzards were more numerous than in all other periods (Multiple Compari- son test P < 0.05). Most buzzards detected (67.4%, N = 221) were alone, while 23.2% {N = 76) were paired and 9.4% {N = 31) were in larger groups, with a mean group size of 3.9 ± 0.6 individuals. The proportion of sightings composed of a group of buzzards showed a uniform distribution throughout the year, ranging from 0.4 in July to 0.0 in January (Z = 1.5; N= 12; P> 0.05). No land cover variable entered the stepwise logistic re- gression discriminating between grid cells with or without buzzards, in any of the four periods of the year (Table 2 ). Discussion The relative abundance of the study population was relatively stable throughout the year, apart from a peak during the courtship period (February-April) . A possible explanation is that during the courtship period, many young buzzards return to their natal area. A high ten- dency for philopatric movements among dispersers has been recorded both in Common Buzzards (Walls and Kenward 1998) and other raptors (e.g., Ferrer 1993, Newton et al. 1994, Carter 2001). A second explanation for the observed increase in relative abundance may be linked more to buzzard behavior and detectability than to variations in actual population density. During court- ship, buzzards participate in more aerial displays than in other periods, and are thus more detectable. This expla- nation is supported by the higher proportion of flying buzzards recorded during the courtship period, particu- larly in March, when all the detected individuals were flying (Fig. 3). The decrease in relative abundance during the incu- bation and nesting periods is probably related to the fact 470 Short Communications VoL. 39, No. 4 60 ■o 50 i_ CO N N m 40 T3 > I 30 O o 20 CO E 3 10 B Flying □ Perched IxJ Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Figure 3. Total number of flying and perched Common Buzzards observed per month during vehicle surveys con- ducted at Pollino National Park, southern Italy in 2000-01. that half of the breeding population were tending nests or were close to their nests most of the time. However, contrary to expectations, during the post-fledging peri- od, when flying young join the adult population, the rel- ative abundance was not higher than that estimated dur- ing the other periods. This could be caused by the relatively inconspicuous behavior of fledged buzzards that often are perched in the immediate neighborhood of the nest for several weeks after fledging (Tubbs 1974, Tyack et al. 1998). Moreover, because most natal dispers- al usually occurs in the autumn (Picozzi and Weir 1976, Walls and Kenward 1998), it is possible that our survey methods underestimated fledged young during the post- fledging period. In our study, buzzards did not show a preference for any land cover type during any periods of the year. This is supported by lack of variation among the seven road- side surveys, although these routes crossed different hab- itats. This lack of association with cover type may be due to the buzzard’s high plasticity and varied diet (Cramp and Simmons 1979, Sergio et al. 2002). However, San- chez-Zapata and Calvo (1999), Kruger (2002), and Sergio et al. (2005) found that buzzard breeding sites were linked to some landscape characteristics, particularly for- est cover. There are three possible explanations for this discrepancy in results. First, the previous studies exam- ined the relationship between habitat characteristics and nesting sites, while we examined the association of indi- vidual locations with land cover type. Second, we record- ed buzzards when in flight or perched, whether they were hunting or involved in other activities. Finally, the dis- agreement could be related to the coarse scale of our landscape analysis. However, Sanchez-Zapata and Calvo (1999) found a linkage between buzzard nest sites and some landscape characteristics using a coarse-scale (1. 200 000) land use map. In summary, our results showed that the relative abun- dance of a raptor population recorded by road counts may be sensitive to temporal behavioral changes, which ultimately affect detectability, thus biasing a potential as- .sessment of seasonal variations in numbers. For this rea- son, roadside surveys would be a coarse method to mon- itor intra-annual population changes, or to compare the relative abundance recorded in different years, unless the data are collected in the same time period or season. On the other hand, we suggest further investigation into whether roadside surveys can be a useful tool for long- term monitoring of a population when comparing the same months in different years. Patrones Estacionaits en ia Abundancia Relativa y el COMPARTAMIENTO DE BUIEO BUTEO EN EL PARQUE NACION- AL POLl.INO, ItAUA Resumen. — Se estudio la abundancia relativa de una po- blacion de Buteo buteo a lo largo de 12 meses en el Parque Nacional Pollino (sur de Italia) para comparar datos en- tre meses y para determinar las asociaciones de habitat Realizamos censos mensuales en carreteras a lo largo de siete rutas (total = 315 km). La media anual de aves detectadas fue de 11.1 ± 4.4 individuos/100 km, aunque este valor vario significativamente entre meses. Esta varia- December 2005 Short Communications 471 cion probablemente reflejo la actividad de vuelo de B. buteo y no las fluctuaciones en el niimero de individuos durante el ano, ya que la mayoria de los registros tuvie- ron lugar durante el periodo de cortejo. Con base en nuestros resultados, sugerimos que los censos realizados para esta especie a lo largo de carreteras son mas efec- tivos durante el periodo reproductivo que en otras epo- cas, cuando los individuos realizan vuelos elevados con menor frecuencia. Finalmente, la presencia y la distri- bucion de B. buteo dentro del parque no se asociaron con el tipo de cobertura del suelo. [Traduccion del equipo editorial] Acknowledgments We are indebted to Pollino National Park, which pro- vided support for this project. We are grateful to P. Perna for GIS analysis and to A. Aradis, B. Carelli, E. Giardi- nazzo, N. Pantone, L. Paternostro, R. Rotondaro, P. Sto- rino, and S. Urso for their help in the field data record- ing. We thank N.E. Seavy, F. Sergio, and G. Boano for their comments on an early draft of the paper. Literature Cited Andersen, D.E., O.J. Rongstad, and W.R. Mitton. 1985. Line transect analysis of raptor abundance along roads. Wildl. Soc. Bull. 13:533-539. Bustamante, J. and J. Seoane. 2004. Predicting the dis- tribution of four species of raptors (Aves: Acdpitridae) in southern Spain: statistical models work better than existing maps./. Biogeogr. 31:295-306. Carter, I. 2001. The Red Kite. Arlequin Press, Chelms- ford, Essex, England. Caughley, G. and A.R.E. Sinclair. 1994. Wildlife ecology and management. Blackwell Science, Cambridge, England. Cramp, S. and K.E.L. Simmons [Eds.]. 1980. Handbook of the birds of Europe, the Middle East, and North Africa. The birds of the western palearctic. Vol. II. Hawks to Bustards. Oxford University Press, Oxford, England. Ellis, D.H., R.L. Glinski, and D.G. Smith. 1990. Raptor road survey in South America. J. Raptor Res. 24:98- 106. Ferrer, M. 1993. Juvenile dispersal behavior and natal philopatry of a long lived raptor, the Spanish Imperial Eagle. Ibis 134:128-133. Field, A. 2000. Discovering statistics using SPSS for Win- dows. SAGE Publications, London, England. Fuller, M.R. and J.A. Mosher. 1987. Raptor survey tech- niques. Pages 37—65 in B.A. Pendleton, B.A. Millsap, K.W. Cline and D.M. Bird [Eds.], Raptor manage- ment techniques manual. Nat. Wildl. Fed., Washing- ton, DC U.S.A. Geomedia Professional. 2002. Geomedia Professional 5.0 for Windows. Intergraph Corporation, Huntsville, AL U.S.A. Kruger, O. 2002. Analysis of nest occupancy and nest reproduction in two sympatric raptors: Common Buz- zard {Buteo buteo) and Goshawk {Accipiter gentilis) . Eco- graphy 25:523-532. Lohmus, a. 2003. Are certain habitats better every year? A review and a case study on birds of prey. Ecography 26:545-552. Meunier, F.D., C. Vereyden, and P. Jouventin. 2000. Use of roadside by diurnal raptors in agricultural land- scapes. Biol. Conserv. 92:291-298. Millsap, B.A. and M.N. LeFranc, Jr. 1988. Road transect counts for raptors: how reliable are they? /. Raptor Res. 22:8-16. Newton, L, RE. Davies, and D. Mo.ss. 1994. Philopatry and population growth of Red Kites, {Milvus milvus), in Wales. Proc. R. Soc. Load. B 257:317-323. Penteriani, V. and B. Fatvre. 1997. Breeding density and landscape-level habitat selection of Common Buzzard {Buteo buteo) in a mountain area (Abruzzo Appenni- nes, Italy). / Raptor Res. 31:208-212- PiCOZZI, N. AND D.N. Weir. 1976. Dispersal and causes of death in buzzards. Br. Birds 69:193-220. Robbins, C.S. 1981. Effect of time of day on bird activity. Pages 275-286 in C.J. Ralph and J.M. Scott [Eds.], Estimating numbers of terrestrial birds. Stud. Avian Biol. 6. SAnchez-Zapata, J.A. and J.F. Calvo. 1999. Raptor distri- bution in relation to landscape composition in semi- arid Mediterranean habitats. / Appl. Ecol. 36:254—262 Sergio, R, A. Boto, C. Scandolara, and G. Bogliani. 2002. Density, nest-sites, diet, and productivity of Common Buzzards {Buteo buteo) in the Italian pre- Alps./. Raptor Res. 36: 24-32. , C. Scandolara, L. Marchesi, P. Pedrini, and V. Penteriani. 2005. Effect of agro-forestry and land- scape changes on Common Buzzards {Buteo buteo) m the Alps: implications for conservation. Anim. Conserv. 7:17-25. Siegel, S. and N J. Castellan, Jr. 1988. Nonparametric Statistics for the Behavioral Sciences. McGraw-Hill, Inc., New York, NYU.S.A. (Italian translation.) SPSS Inc. 2000. SPSS 10.0 for Windows: base, profession- al statistics and advanced statistics. SPSS Inc., Chicago, IL U.S.A. Tubbs, C.R. 1974. The Buzzard. David 8c Charles, Lon- don, England. Tyack, A.J., S.S. Walls, and R.E. Kenward. 1998. Behav- ior in the post-nestling dependence period of radio- tagged Common Buzzards {Buteo buteo). Ibis 140:58— 63. ViNUELA, J. 1997. Road transects as a large-scale method for raptors: the case of the Red Kite {Milvus milvus) in Spain. Bird Study 44:155—165. Walls, S.S. and R.E. Kenward. 1998. Movements of ra- dio-tagged buzzards {Buteo buteo) in early life. Ibis 140’ 561-568. Received 7 October 2004; accepted 5 September 2005 Associate Editor: Fabrizio Sergio 472 Short Communications VoL. 39, No. 4 J Raptor Res. 39(4):472-475 © 2005 The Raptor Research Foundation, Inc. New Nesting Record and Observations of Breeding Peregrine Falcons IN Baja California Sur, Mexico Aradit Castellanos, 1 Cerafina Arguelles, Federico Salinas-Zavala, and Alfredo Ortega-Rubio Centro de Investigaciones Biologicas del Noroeste, S.C. Apdo. Postal No. 128 CP 23000, La Paz, Baja California Sur, Mexico Keywords: Peregrine Falcon-, Falco peregrinus; nesting re- cord-, Baja California-, Mexico. The B^a California peninsula has been an area where resident Peregrine Falcons {Falco peregrinus) were com- mon in the past (Bancroft 1927, Banks 1969). Historical records of their presence in the region were published by Bryant in 1889 (see Grinnell 1928). Detailed accounts of nesting territories for the peninsula and Gulf of Cali- fornia islands were made by Banks (1969), Anderson (1976), and Porter et al. (1988). According to Banks (1969), prior to 1967, 54 known peregrine locations ac- counted for approximately 66 nest sites in this region. Porter et al. (1988) identified 67 eyries in this area in 1976-84. As in other parts of the world from the late 1960s to the early 1980s, the peninsular Peregrine Falcon population declined, likely due to the impact of organ- ochlorine pesticides (Kiff 1988). However, published data show that recovery began by the late 1980s (Porter et al. 1988, Castellanos et al. 1997). Historical nesting territories were mainly located on sea cliffs of the western side of the state of Baja Califor- nia, from Tijuana to Santa Catarina, and on islands along both coasts of the peninsula (Banks 1969, Porter et al. 1988, Castellanos et al. 1997, Ruiz-Campos and Contrer- as-Balderas 2000). A small number of inland territories were also known (Banks 1969, Castellanos et al. 1997). However, no nesting pair has ever been reported on Ba- hia Magdalena region, which is located on the west coast of the peninsula (Fig. 1). Here, we provide the first re- port of Peregrine Falcons nesting in this area. Our find- ing extends the breeding range of this species to an area of the Baja California peninsula lacking suitable natural nesting sites. We also report on the reproductive output of this pair. Methods We observed the nest from vantage points 100-200 m from the metal tower, where the peregrine nest was lo- cated. We monitored activities around the nesting area, and with the help of binoculars and a spotting scope, recorded the behavior and attendance of adults at the nest. During each observation period, we registered the ^ Email address; arcas04@cibnor.mx contents in the nest, the birds’ activities, time, the num- ber of interactions between birds, and disturbance events that occurred. Results and Discussion On 25 January 2004, we found a pair of Peregrine Fal- cons at 25°05'2T'N and 112°04'41"W in Santa Elenita, a remote, abandoned industrial port 7 km north of Adolfo Lopez Mateos in Bahia Magdalena, Baja California Sur, Mexico. Both birds displayed activities (mutual roosting, cooperative hunting excursions, courtship flights) sug- gesting they were a territorial pair. The same day we sighted another single adult Peregrine Falcon 7.5 km south of Santa Elenita. After our finding, we visited Santa Elenita between February to mid-October to monitor the pair’s reproductive status (20 hr 25 min of observation in 9 d between 1000-1600 H). The port is a concrete platform about 30 m wide by 120 m long located in a mangrove estuary. Above this platform there are four metal towers and a central crane 40 m tall. The crane has a metal horizontal arm about 30 m long oriented northeast to southwest (Fig. 2). The peregrine pair nest- ed on an old Osprey {Pandion haliaetus) nest at the ex- treme southwestern end of the arm (Fig. 2). On the op- posite side was another Osprey nest. A small fishing camp near the platform was operating daily during the study. On 28 February, we saw another single bird (male Per- egrine Falcon, by relative size) approaching the nest. Both males “fought,” and after the interaction and the intruder left, the nesting male copulated with the female. On 18 March, we found the female incubating eggs. From 8-28 May, three nestlings were observed in the nest, and they began to fly in late June. On 22 July, we observed the parents and one of the fledglings. The adults were still on the territory on 1-4 and 11-15 Oc- tober. The tower was shared with an Osprey pair during the entire study period. However, both pairs showed toler- ance despite their proximity. Daily and incidental human disturbance at the hase of the towers was relatively in- tense (22 events or 1.08 events/hr, including small boats, people, cars, and motocross traffic); however, the pere- grine nest was not deserted and breeding was successful. Peregrine response to disturbance occurred more fre- quently when other birds approached to within about 20 December 2005 Short Communications 473 Figure 1. Current and historical distribution of nesting territories of Peregrine Falcon in Baja California Sur, Mexico 474 Short Communications VoL. 39, No. 4 Figure 2. View of Santa Elenita mangrove estuary in Bahia Magdalena, B. C. S., Mexico, and Peregrine Falcon (a) and Osprey (b) nesting sites. m of the nest. We observed 16 interactions (0.79 events/ hr), including eight with Magnificent Frigatebirds {Fre- gata magnificens ) , four with Ospreys, three with Common Ravens {Corvus corax), and one with a gull {Larus sp.). Peregrines made territorial attacks or cacking-calls in all interactions. Breeding dates and the number of nestlings and fledg- lings reared by this p£iir were similar to those reported for other areas on the western coast of the peninsula (Porter et al. 1988, Castellanos et al. 1997). Previous re- ports (Castellanos et al. 1997) also show the number of nestlings and fledglings produced by pairs on the western side of the peninsula are greater than those along the Gulf of California. Reasons for this higher productivity are unknown. However, it may be an indication of a healthier environment; the west coast of the peninsula is a relatively low organochlorine pesticide-polluted area in North America (total DDT concentration levels on Os- prey eggs between 5-311 ppm lipid basis; Spitzer et al. 1977). The southwest coastline of the peninsula is quite dif- ferent from the northwestern coast. The terrain is rela- tively flat and covered by sparse desert shrubs. The lack of suitable natural nesting sites such as sea cliffs, ledges on vegetated slopes, and high trees precludes establish- ment of breeding pairs in spite of a variety and abun- dance of shore birds and waterfowl. The Santa Elenita towers provided an opportunity for this nest-site-limited species to breed. Future conservation of Peregrine Falcon south of the U.S.A. should be focused on protection of the natural landscape (Temple 1988). This strategy is on course in Mexico. In 1972, the islands in the Gulf of California were protected as wildlife refuges. In 1988, the entire central west coast, including the small islands and Lagun- as Ojo de Liebre and San Ignacio, was declared a bio- sphere reserve. These refuges preserve prime Peregrine Falcon breeding range with low human impact. Nuevo Registro de Anidacion y Observaciones de Hal- coNES Peregrines Reproductores en Baja Caufornia S uR, Mexico Resumen. — Encontramos una pareja reproductiva y dos adultos no reproductivos de Falco peregrinus en Bahia Magdalena, en la costa suroeste de Baja California Sur, Mexico. El nido estaba localizado en una torre metalica en un puerto abandonado. La anidacion fue exitosa y tres volantones abandonaron el nido. Nuestro hallazgo amplia el rango de anidacion conocido en la peninsula. [Traduccion de los autores] December 2005 Short Communk:ations 475 Acknowledgments We thank A. Cozar, A. Ortiz, E. Rivera, and P. Tamez for field assistance. Financial support was provided by Centro de Investigaciones Biologicas del Noroeste, S. C. and Secretaria del Medio Ambientes y de los Recursos Naturales-Consejo Nacional de Ciencia y Tecnologia proj- ect 2002-GO 1-0844. We are grateful to Jim Watson, Tom Cade, and one anonymous reviewer for their comments on an earlier draft of this paper. Thanks to Dr. E. Glazier for editing the English-language text. Literature Cited Anderson, D.W. 1976. The Gulf of California, Mexico. Pages 270-271 in R.W. Fyfe, S.A. Temple, and T.J. Cade [Eds.], The 1975 North American Peregrine Falcon survey. Can. Field-Nat. 99:228-273. Bancroft, G. 1927. Notes on the breeding coastal and insular birds of central Lower California. Condor 29: 228-273. Banks, R.C. 1969. The Peregrine Falcon in B^ya Califor- nia and the Gulf of California. Pages 81—91 in JJ- Hickey [Ed.], Peregrine Falcon Populations. Univer- sity Wisconsin Press, Madison, W1 U.S.A. Castellanos, A., F. Jaramillo, F. Salinas, A. Ortega-Ru- BIO, AND C. Arguelles. 1997. Peregrine Falcon recov- ery along the west central coast of the Baja California peninsula, Mexico./, Raptor Res. 31:1-6. Grinnell, J. 1928. A distributional summation of the or- nithology of Lower California. Univ. Calif. Publ. Zool. 32:1-300. Kief, L.F. 1988. Changes in the status of the peregrine in North America: an overview. Pages 123-139 in T Cade, J. Fnderson, C. Thelander and C. White [Eds.], Peregrine Falcon populations: their management and recovery. The Peregrine Fund, Inc., Boise, ID U.S.A. Porter, R.D., M. A. Jenkins, M.N. Kirven, D.W. Anderson AND J.O. Keith. 1988. Status and reproductive perfor- mance of marine peregrines in Baja California and the Gulf of California, Mexico. Pages 105-114 in T. Cade, J. Fnderson, C. Thelander and C. White [Eds.], Peregrine Falcon populations: their management and recovery. The Peregrine Fund, Inc., Boise, ID U.S.A. Ruiz-Campos, G. AND AJ. Contreras-Balderas. 2000. New northern nesting record of the Peregrine Falcon in Baja California, Mexico./. Raptor Res. 34:151. Spitzer, P.R., R.W. Risebrough, J.W. Grier, and C.R. Sin- DElAR, Jr. 1977. Eggshell thickness-pollutant relation- ships among North American Ospreys. Pages 13—19 mJ.C. Ogden [Ed.], Trans. North American Osprey research conference. U.S. Nat. Park Serv. Proc. Ser. 2, Washington, DC U.S.A. Temple, S.A. 1988. Future goals and needs for the man- agement and conservation of the Peregrine Falcon. Pages 843-848 in T. Cade, J. Fnderson, C. Thelander and C. White [Eds.], Peregrine Falcon populations: their management and recovery. The Peregrine Fund, Inc., Boise, ID U.S.A. Received 7 December 2004; accepted 5 September 2005 Associate Editor: James W. Watson Letters J. Raptor Res. 39(4):476-477 © 2005 The Raptor Research Foundation, Inc. A Previously Undescribed Vocalization of the Northern Pygmy-Owl Vocalizations of the Northern Pygmy-Owl {Glaucidium gnoma) were summarized by Holt and Peterson (2000, The Birds of North America, No. 494, Philadelphia, PA U.S.A.). The known repertoire of adult vocalizations consists of: the “toot song,” which functions as the primary song for both sexes; the “trill call,” which often accompanies the “toot song” and for which a function is not yet known; and the “chitter call,” which accompanies prey deliveries and certain interactions between breeding adults. Also, a “chatter call” used during copulation was described by Righter (1995, Colo. Field Ornithol. J. 29:21—23). Here, I describe a previously undocumented vocalization of the Northern Pygmy-Owl. As part of an ongoing study of Northern Pygmy-Owls in northern Montana, I have radio-tracked 11 (between one and five annually) owls since 2001. Owls were tracked before, during, and after the nesting season. Periods of behavioral observation were intermittently conducted in durations ranging from ca. 15 min to 4 hr. On three occa- sions, I observed a “weet” vocalization, which somewhat resembled a single note of the “toot song,” but was slightly prolonged with a “screeching” quality and a slight upward bend in pitch. In abruptness and duration, this call superficially resembled the sudden, high-pitched alarm calls of some Spermophilus ground squirrels, but was deeper in pitch and slightly more hollow sounding. The most similar avian call that I am familiar with is the ksew call of the Northern Saw-whet Owl {Aegolius acadicus; Cannings 1993, The Birds of North America, No. 42, Washington DC, U.S.A.). However, that call is lower in pitch, descending, and usually repeated in a brief series, while the weet call I describe here was a single ascending note. Proudfoot and Johnson (2000:7, The Birds of North America, No. 498, Philadelphia, PA U.S.A.) describe an alarm call of the congeneric Ferruginous Pygmy-Owl {Glaucidium brasilianum) as “short and sharp with upward inflection, pee weet, repeated at irregular intervals.” Although I am unfamiliar with the Ferruginous Pygmy-Owl alarm call, this describes accurately the Northern Pygmy-Owl vocalization I observed, except that the former consists of two syllables and the latter only one. On 21 July 2003, while observing a Northern Pygmy-Owl family group, I first observed an adult male give the ''iveet" call. A few seconds later an adult Northern Goshawk {Accipiter gentilis) flew within 50 m of the family group and disappeared. On 1 July 2004 at 0936 H, while observing a family group with recently fledged young, I observed both the adult male and adult female give the weet call. A Northern Goshawk appeared and perched briefly in the same stand as the family group a few seconds after the calls. In that instance, the male gave the call first and was followed by the female. The female then repeated the call one more time just before the goshawk entered the stand The female’s call was slightly higher pitched to a degree approximately equivalent to the difference in pitch of the “toot song” between the .sexes (Holt and Peterson 2000, pers. obs.). At 1003 H, during the same observation period on 1 July 2004, a Northern Goshawk flew through an opening adjacent to the stand in which the family group (same stand and same family group as the previous observation) was located. Both adults gave weet calls, and both repeated the calls after 5—10 sec. In that instance, several of the seven fledglings had been calling actively before the adults gave the weet calls. The fledglings’ calls immediately ceased after the weet calls were given, but resumed less than 30 sec later, at which time the goshawk was no longer visible to me. 1 am not certain whether any of the Northern Goshawks in these instances were aware of the owls. Twice the goshawks simply flew past, and the one time a goshawk perched nearby, it quickly left the perch when it apparently became aware of my presence. The contexts of these observations suggest that the function of the weet call is an alarm call. During a Northern Pygmy-Owl study in Washington, A. Giese (unpubl. data) observed what was likely the call described here, and likewise suspected its function as an alarm call. The call was given several times during a 1.5 hr period by an adult female in the presence of fledged young. The young generally ceased vocalizing after the call was given. Unlike the calls I observed, however, that female vocalized in bouts of 3-4 calls at a time. Interestingly, I have observed instances in which Northern Pygmy-Owls might have been expected to give alarm calls, but did not give the weet call. I observed a domestic dog in close proximity (<5 m) to a group of fledgling and adult owls perched low on branches on two occasions in 2004 and heard no calls. When banding young owls or climbing nest trees to check nests, I have observed adults give short versions of the toot song, trill calls, and display agitated behavior (e.g., rapid tail twitching, perching close to and staring at the human intruder). However, I have not heard the weet call I describe here in those situations. On 7 May 2001, I observed a Northern Goshawk pass 476 December 2005 Letters 477 within 5 m of a solitary nonbreeding Northern Pygmy-Owl perched with prey. The owl watched the goshawk intently, but gave no call. If the weet call was indeed an alarm call, these observations suggested that it may be associated specifically with avian predators, as the mammalian (dog and human) observations I described did not elicit the call. Additionally, the observation of the solitary owl failing to give a weet call suggests that it may be used only when fledglings or mated owls are present. Alarm calls are well documented for many strigids (e.g.. Ferruginous Pygmy-Owl, Proudfoot and Johnson 2000; Great Gray Owl, Strix nebulosa, Bull and Duncan 1993, The Birds of North America, No. 41, Philadelphia, PA U.S.A.; Long-eared Owl Asio otus, Marks et al. 1994, The Birds of North America, No. 133, Phila- delphia, PA U.S.A.) and are often given in response to threats to nests and fledged young. However, further study is needed to better understand the causes and contexts of this and other Northern Pygmy-Owl vocalizations. I am grateful to the many individuals who have volunteered their time to conduct fieldwork during this study. Private contributions have helped make this ongoing study possible. The Conservation Research Foundation, Mar- mot’s Edge Conservation, the Owl Research Institute, and the Rocky Mountain Ranger District of the Lewis and Clark National Forest have generously provided assistance with various parts of this project as well. A. Giese and G Proudfoot provided criticism and field observations that improved the manuscript. — Graham G. Frye (e-mail address: ggfrye@rmf-inh,org) Rocky Mountain Front Institute of Natural History, P.O. Box 186, Ghoteau, MT 59422 U.SA. Received 12 August 2004; accepted 18 June 2005 Associate Editor: Ian G. Warkentin BOOK REVIEW J. Raptor Res. 39(4) :478-479 © 2005 The Raptor Research Foundation, Inc. Hawks and Owls of Eastern North America. By Donald S. Heintzelman. 2004. Rutgers University Press, Piscataway, NJ U.S.A. viii + 203 pp., 5 color plates, numerous black and white photos. ISBN 0-8135-3350-3. Cloth, $29.95.— Donald Heintzel- man ’s original purpose with this volume was to pre- pare a second edition of his earlier work, Hawks and Owls of North America (1979, Universe Books, New York, NYU.S.A.). However, this book restricts coverage to the raptors of eastern North America, which is loosely defined as birds east of the Missis- sippi River, except for parts of Minnesota and On- tario. The volume begins with six chapters dealing with general topics: Introduction to Raptor Ecolo- gy, Hawk Migrations, Owl Migrations and Inva- sions, Raptor Conservation, Citizen Scientists, and Recreational Raptor Watching; and follows with eight chapters of raptor species accounts mostly in taxonomic order, beginning with Ospreys {Pandion halieatus) and ending with Northern Saw-whet Owls {Aegolius acadicus). For the most part, the book is nicely illustrated with a selection of outstanding black and white photos. I found the introductory chapter, “An Introduc- tion to Raptor Ecology,” to be well out of date. In the preface, Heintzelman acknowledges that this chapter includes substantial portions of the text from his 1979 contribution. Upon reading this chapter, it is clear that many of the ideas expressed reflect antiquated ecological opinions and specu- lations of the 1960s and early 1970s. Statements such as. Northern Goshawks {Accipiter gentilis) are beneficial to Ruffed Grouse (Bonasa umbellus) pop- ulations, that raptors control numbers of prolific rodents, and raptors maintain a delicate and effec- tive ecological balance between predator and prey, without presentation of supporting data or sources are made freely throughout this short chapter. The subsequent introductory chapters, although brief, provided easier reading and were based on more-current information. The chapter on Raptor Conservation reported a series of interesting and relatively-recent anecdotes. However, when discuss- ing habitat degradation and loss, Heintzelman em- phasizes the “ambitious, long-term restoration” ef- fort of the Lehigh Gap Restoration Project, which has a goal of restoring 750 acres of woodland on the Kittatinny Ridge in Pennsylvania. From a rap- tor perspective, restoring ca. 3 km^ of deciduous woodland habitat is extremely trivial and will likely have little impact on the population of any raptor. I am familiar with several other governmental and private (e.g., the Nature Conservancy) land acqui- sition/habitat restoration programs that involve many thousands of hectares that are much more likely to have substantial population impacts on several species of raptors. I would have liked to have seen Heintzelman discuss some of these ma- jor habitat restoration efforts, perhaps in addition to smaller isolated projects that he has personally spear-headed. Some minor distractions for me were the provi- sion of selective contact information. Although most major raptor conservation organizations were mentioned in the volume, contact information was only provided for a select few. Perhaps, in this day and age in which an interested person can quickly obtain contact information by googling the name of an organization, this is not necessary. But, why provide detailed contact information (postal ad- dress, e-mail address, phone numbers) for a few selected organizations, and no information for oth- ers? Probably more bothersome for a professional or- nithologist using this volume is the style of not cit- ing references in the text. Although most chapters are reasonably-well researched and supported with references, albeit selectively, all references are in- cluded in the back of the volume listed alphabeti- cally by chapter. Thus, when you encounter a state- ment in a given chapter, identifying the responsible source is exceedingly difficult. Also, the supporting sources represent a very mixed bag in which some very current and important studies are mentioned and cited, while most of the refer- ences are from state journals and generally repre- sent novel anecdotes. Although I found most of the material presented in the species accounts to be accurate, I ran across several reported “facts” that in my opinion. 478 December 2005 Book Review 479 amounted to unsubstantiated and rather far-reach- ing speculations. Some examples include a state- ment that evidence supports that mated pairs of Rough-legged Hawks {Buteo lagopus) perch togeth- er on their wintering grounds; statements or im- plications that several species of raptors covered in the volume relatively commonly exhibit coopera- tive hunting; that Gyrfalcons {Falco rusticolus) are faster than Peregrine Falcons {F. peregrinus); and that the food habits of several small raptors include large birds and mammals, such as waterfowl (An- atidae), grouse (Tetraoninae) , raccoons {Procyon lo- tor), woodchucks (Marntota monax), and hares (Le- pus spp.). The latter is true, although most of these relatively large preys are taken rarely and most like- ly involve very-young animals or carrion. This clar- ification is generally not included and the way ac- counts are worded, the text implies that such large prey are just as commonly taken as small rodents. The layout of the book, with citations lumped by chapter, makes tracking down the specific sources for these far-reaching speculations and implica- tions nearly impossible. As a scientist, I found this aggravating. Besides my mild complaints indicated above, the species accounts were concise and informative. Within each account, the known longevity record for each species is reported, which I found to rep- resent an interesting anecdote. Each account is il- lustrated with one or more high-quality black and white photos of the featured species. For the most part, the volume is well-edited and I found only a few typographical errors scattered about the text. Generally, I feel this volume would make an ex- cellent primer for an amateur or beginning stu- dent interested in North American birds of prey. The presentation of the basic natural history and promotion of recreational hawk watching to the beginning student of birds is clearly the intended target of this publication. I would recommend tbis volume to a high school or underclass university student that expresses an interest in raptors. Also, this book would make an excellent resource for local public libraries throughout eastern North America. — James C. Bednarz, Department of Bio- logical Sciences, Arkansas State University, P.O. Box 599, Jonesboro, AR 72467 U.S.A. J Raptor Res. 39(4);480-483 © 2005 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 correct any printer errors and returned by the fast- est mail within two days of receipt TO THE EDI- TOR. 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 reprint order will accompany page proofs to en- able authors to buy reprints. Costs of reprints are the author’s responsibility and payment for re- prints ordered 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 hy 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. 480 December 2005 Information for Contributors 481 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 2*78 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 near- letter-quality printer. DOUBLE SPACE THROUGH- OUT including title, text, tables, figure legends, and literature cited. D 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: journalofraptorresearch® juno.com □ 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 North 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. n 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. □ 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. □ Use “continental” dating (e.g., lOJuly 1993, 1-3 June, 1 1 May to 1 1 June) . □ Use 24-hour clock (e.g., 0800 H, 1345-1400 H) D 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. G 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 authorG^ones (1993) or (Jones 1993) b. Two authons-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., (LG. Birds 1993, /. Raptcyr Res. 27:45-50) . D 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. □ 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 correspondence 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 482 Information for Contributors VoL. 39, No. 4 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. n 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. LiTEaiATURE Cited n 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. n Verify all entries against original sources includ- ing diacritical marks and spelling in languages other than English. Capitalize all nouns in Ger- man. □ Cite references in alphabetical order by hrst 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. □ On-line sources should be cited if necessary and listed in the Literature Cited. Include au- thor (s), year, appropriate “title” of web site, publisher or sponsor of web site, web site ad- dress, and date last accessed in parentheses. Example: Sauer, J.R., J.E. Hines, and J. Fallon. 2005. The North American breeding bird survey re- sults and analysis, 1966-2004. Version 2005.2 USGS Patuxent Wildlife Research Center, Laurel, MD U.S.A. http://www.mbr-pwrc. usgs.gov/bbs/bbs.html (last accessed 31 De- cember 2005). □ “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). O 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 headings should be typed in large and small capital letters. D Use same size of type as in text. □ Indicate footnotes by lowercase superscript let- ters. □ Do not use vertical lines. VII. Figure Legends □ Print all figure legends on one page, double spaced. □ Number using Arabic numbers consecutively m the same order the figures appear in the text (i.e., Figure L, Figure 2., etc.). VIII. Preparation of Illustrations (Illustrations are referred to as figures and include drawings, graphs, and black and white half-tones December 2005 Information for Contributors 483 [photographs] . CONSULT THE EDITOR IN AD- VANCE ABOUT COLOR.) D 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. n 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 X 11") or standard international (210 X 297 mm). □ All graphics and images should be scanned at a minimum resolution of 300 pixels per inch (ppi). Line art should be scanned at 1200 ppi. Low resolution figures or graphics are not ac- ceptable. □ 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. □ Use the same style of lettering and presentation for all figures. Capitalize each word of axes titles except prepositions and articles. IX. What to Send □ Cover letter. □ Copy of this checklist completed. □ 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) . □ Electronic submissions are encouraged. See sec- tion I. □ Submit to: Cheryl R. Dykstra, Editor Raptor Environmental 7280 Susan Springs Drive West Chester, OH 45069-3696 U.S.A. More information: Telephone: (513) 779-1744 E-mail: journalofraptorresearch@juno.com J Raptor Res. 39(4):484-491 © 2005 The Raptor Research Foundation, Inc. Index to Volume 39 Prepared By Autumn A. Farless 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, 80-83 relative, 466-471 Accipiter cooperii, 109 gentilis, 210-221, 222-228, 229-236, 237-246, 247-252, 264-273, 274-285, 286-295, 296-302, 303-309, 310-323, 32^334, 335-341, 342-350, 439-444 atricapillus, 192-209 laingi, 253-263 Activity centers, 253-263 Adaptive kernel, 253-263 Aegolius acadicus brooksi, 134-141 Agostini, Nicolantonio, Are earlier estimates of Accipitri- formes crossing the channel of Sicily (Central Med- iterranean) during spring migration accurate?, 184— 186 Agricultural areas, 55-60 Agriculture, 429-438 Alarm call, 475-476 Alencar Carvalho, Carlos Eduardo, see Mendes de Car- valho Filho, Eduardo Pio Allen, Deborah J., see DeCandido, Robert Anchor-bolt ladder, 109 Andersen, David E., Stephen DeStefano, Michael I. Gold- stein, Kimberly Titus, Cole Crocker-Bedford, John J. Keane, Robert G. Anthony, Robert N. Rosenfield, Technical review of the status of Northern Goshawks in the western United States, 192-209 Andersen, David E., see Boal, Clint W. Andersen, David E,, see Smithers, Brett L. Andrade, Analia, see Sauthier, Daniel E. Udrizar Anthony, Robert G., see Andersen, David E. Anthropogenic disturbance, 97-101 Antolin, Michael R, see Bayard de Volo, Shelley Aquila clanga, 462-466 Argentina, 65-69 Argiielles, Cerafina, see Castellanos, Aradit Arizona, 274-285 Arkansas, 74-79 Asio otus, 445-453 Athene cuniculania, 429—438 noctua, 156-159, 454—457 Attalea maripa B Baja California, 472-475 Baldassarre, Guy A., see Jensen, Wendy J. Bayard de Volo, Shelley, Richard T. Reynolds, J. Rick To- pinka, Bernie May, and Michael F. Antolin, Popula- tion genetics and genotyping for mark-recapture studies of Northern Goshawks {Accipiter gentilis) on the Kaibab Plateau, Arizona, 286—295 Bechard, Marc J., see Fairhurst, Graham D. Bednarz, J.G., sec Radley, Paul M. Bednarz, J.G., A Review of Hawks and Owls of Eastern North America, by Donald S. Heintzelman, 2004, 478-479 Beier, Paul, see Boyce, Douglas A., Jr. Begall, Sabine, The relationship of foraging habitat to the diet of Barn Owls ( Tyto alba) from central Ghile, 97-101 Belthoff, James R., see Moulton, Colleen E. Bennett, Jason R. and P.H. Bloom, Home range and hab- itat use by Great Horned Owls {Bubo virginianus) m southern California, 119-126 Bildstein, Keith L., see Jensen, Wendy J. Bird, David M., see Laing, Dawn K. Bird, David M., see Chubbs, Tony E. Black-Hawk, Common, 351-364 Cuban, 351-364 Bloom, Peter H., see Bennett, Jason R. Bloszyk, Jerzy, see Gwiazdowicz, Dariusz J. Boal, Clint W., Preface: Proceedings of the international symposium on the ecology and management of Northern Goshawks, 189 Boal, Clint W., Productivity and mortality of Northern Goshawks in Minnesota, 222-228 Boal, Clint W., see Smithers, Brett L. Boavista, 80-83 Bonin Islands, 173-179 Bootstrapping, 253-263, 274-285 Botella, Francisco, see Navarro, Joan Boyce, Douglas A., Jr., Patricia L. Kennedy, Paul Beier, Michael F. Ingraldi, Susie R. MaeVean, Melissa S. Sid- ers, John R. Squires, and Brian Woodbridge, When are goshawks not there? Is a single visit enough to infer absence at occupied nest areas?, 296-302 Brady, Ryan S., see Moulton, Colleen E. Brazil, southeastern, 89-92 Breeding, 222-228, 229-236 chronology, 74-79 cooperative, 92-94 range, 70-74 484 December 2005 Index to Volume 39 485 British Columbia, 1-10, 335-341 Bubo magellanicus, 163-166 virginianus, 111-118, 119-126 Bustamante, Javier, see Rodriguez, Carlos Buteo buteo, 466-474 jamaicensis, 108, 439-444 platypterus brunnescens, 404—416 toyoshimai, 173-179 Buteogallus anthracinus, 351-364 gundlachii, 351—364 subtilis, 351-364 Buzzard, Common, 173-179, 466-474 C California, southern, 119-126 Calvo, Jose F., see Martinez, Jose E. Camera, remote, 303-309 Camiha, Alvaro, see Garrido, Jose Rafael Canary Islands, 186-187 Caprimulgus macrurus, 106-107 Capture-recapture, 286-295 Caracara, Yellow-headed, 94—97 Caribbean, 94—97 Carrete, Martina, see Navarro, Joan Casado, Eva and Miguel Ferrer, Analysis of reservoir se- lection by wintering Ospreys {Pandion haliaetus hal- iaetus) in Andalusia, Spain: a potential tool for re- introduction, 168-173 Castellanos, Aradit, Cerafina Arguelles, Federico Salinas- Zavala, Alfredo Ortega-Rubio, New nesting record and observations of breeding Peregrine Falcons in Baja California Sur, Mexico, 472-475 Cebus olivaceus, 458—461 Cervera, Francisco, see Garcia, Ana Maria Chambers, Carol L., see Gatto, Angela E. Chile, 55-60, 97-101 Chiroptera, 445-453 Chubbs, Tony E., Matthew J. Solensky, Dawn K. Laing, David M. Bird, and Geoff Goodyear, Using a porta- ble anchor-bolt ladder to access rock-nesting Osprey, 103-106 see Laing, Dawn K. Circus spilonotus, 106-107 CITES, 386-393 Colorado, 166—168 Community, 156-159 Competition, 156-159, 439-444 Connecticut, 342-350 Corales Stappung, Ema Soraya, see Rojas, Ricardo A. Fi- gueroa Crete, 179-183 Crocker-Bedford, Cole, see Andersen, David E. Crozier, Michelle L., The effect of broadcasting Great Horned Owl vocalizations on Spotted Owl vocal re- sponsiveness, 111-118 D DeCandido, Robert and Deborah J. Allen, First nesting of Cooper’s Hawks {Accipiter cooperii) in New York City since 1955, 109 Dekker, Dick and Robert Taylor, A change in foraging success and cooperative hunting by a breeding pair of Peregrine Falcons and their fledglings, 394-403 de la Rocha, J.L. Paz, see Garrido, Jose Rafael DeLong, John R, Timothy D. Meehan, and Ruth B. Smith, Investigating fall movements of hatch-year Flammulated Owls {Otus flammeolus) in central New Mexico using stable hydrogen isotopes, 19-25 Departure, 462-466 Desimone, Steven M. and Stephen DeStefano, Temporal patterns of Northern Goshawk nest area occupancy and habitat: a retrospective analysis, 310-323 DeStefano, Stephen, see Andersen, David E. DeStefano, Stephen, see Rogers, Andi S. DeStefano, Stephen, see Desimone, Steven M. DeStefano, Stephen, A review of the status and distribu- tion of Northern Goshawks in New England, 342- 350 Detectability, 274-285 Detection rates, 296-302 Diet, 55-60, 65-69, 80-83, 97-101, 163-166, 173-179, 179-183, 237-246, 264-273, 303-309, 439-444, 445- 453 Digital image analysis, 127-133 Dispersal, 11-18 natal, 253-263 Distance sampling, 237-246 Distribution, 342-350 Doyle, Donald D., see McClaren, Erica L. Doyle, Frank I., see Mahon, Todd E Eastham, Chris P. and Mike K. Nicholls. Morphometric analysis of large Falco species and their hybrids with implications for conservation, 386-393 Eagle, Bald, 1-10, 11-18 Booted, 92-94, 159-163 Greater Spotted, 462-466 Ecology, 351-364 Egea, Maria, see Garrido, Jose Rafael Eggs, 89-92 Egyptian Vulture, 186-187 Elanoides forficatus, 94—97 Elanus leucurus, 378—385 Elliott, John E., see Elliott, Kyle Hamish Elliott, Kyle Hamish, Christopher E. Gill, and John E Elliott, The influence of tide and weather on provi- sioning rates of chick-rearing Bald Eagles in Vancou- ver Island, British Columbia, 1-10 Endangered, 404—416 Enderson, James H., Changes in site occupancy and nest- 486 Index to Volume 39 VoL. 39, No. 4 ing performance of Peregrine Falcons in Colorado, 1963-2004, 166-168 Europe, eastern, 36-54 Experiment, cross-over, 111-118 F Eairhurst, Graham D. and Marc J. Bechard, Relationships between winter and spring weather and Northern Goshawk (Accipiter gentilis) reproduction in northern Nevada, 229-236 Falco cherrug, 386—393 femoralis, 55-60 naumanni, 12V-133 nexvtoni, 149-155 novaezeelandiae, 386-393 peregrinus, 166-168, 386—393, 394^403, 472—475 rusticolus, 386-393 sparverius, 84—88, 94—97, 378—385 tinnunculus alexandri, 80—83 Falcon, Aplomado 55-60 New Zealand, 386-393 Peregrine, 166-168, 386-393, 394-403, 472-475 Saker, 386-393 Falcons, 378-385 Family break up, 462-466 Feather, 84-88 Feeding ecology, 97-101 Ferland, Cheron, see Forsman, Eric D. Ferrer, Miguel, see Casado, Eva Fidelity, site, 134—141 Fish production, 168-173 Fitness, 210-221 Fitzsimons, James A., Attempted predation on a Large- tailed Nightjar (Caprimulgus macrurus) by an Eastern Marsh-Harrier ( Circus spilonotus) in coastal Vietnam, 106-107 Fledging-dependency period, 253-263 success, 26-35 Food habits, 429-438, 439-444 niche, 429-438 breadth, 439-444 partitioning, 159-163 Foraging, 439-444 association, 458-461 Forest, 159-163 karst, 404—416 management, 296-302, 324—334 Forsman, Eric D., TimmothyJ. Kaminski, Jeffery C. Lewis, Kevin J. Maurice, Stan G. Sovern, Cheron Ferland, and Elizabeth M, Glenn, Home range and habitat use of Northern Spotted Owls on the Olympic Pen- insula, Washington, 365-377 Fortabat, Sofia Heinonen, see Pardihas, Ulyses FJ. French, John B., Jr., see Quinn, Michael J.,Jn Frye, Graham G., A previously undescribed vocalization of the Northern Pygmy-Owl, 476-477 G Gaibani, Giorgia, see Pandolfi, Massimo Gangoso, Laura and Cesanjavier Palacios, Ground nest- ing by Egyptian Vultures {Neophron perenopterus) in the Canary Islands, 186-187 Garcia, Ana Maria, Francisco Cervera, and Alejandro Rodriguez, Bat predation by Long-eared Owls m Mediterranean and temperate regions of southern Europe, 445-453 Garcia, David, see Navarro, Joan Garrido, Jose Rafael, Alvaro Camina, Mariangela Guinda, Maria Egea, Nourdine Mouati, Alfonso Godino, and J.L. Paz de la Rocha, Absence of the Eurasian Grif- fon {Gyps fulvus) in northern Morocco, 70-74 Garrido, Orlando H., see Wiley, James W. Gatto, Angela E., Teryl G. Grubb, and Carol L. Cham- bers, Red-tailed Hawk dietary overlap with Northern Goshawks on the Kaibab Plateau, Arizona, 439-444 Gender determination, 127-133 Genetic structure, 192-209 variability, 142-148 Geographical-catchment area, 19-25 Gill, Christopher E., see Elliott, Kyle Hamish Gimenez, Andres, see Navarro, Joan Glaucidium gnoma, 476-477 Glenn, Elizabeth M., see Forsman, Eric D. Godino, Alfonso, see Garrido, Jose Rafael Goldstein, Michael L, see Andersen, David E. Gonzalez, Carlos, see Martinez, Jose E. Gonzalez-Bravo, Betzabeth, see Meraz, Juan Goodyear, Geoff, see Chubbs, Tony E. Goshawk, Northern, 192-209, 210-221, 222-228, 229- 236, 237-246, 247-252, 253-263, 264-273, 274-285, 286-295, 296-302, 303-309, 310-323, 324-334, 335- 341, 342-350, 439-444 Gregory, Mark S., see Jensen, Wendy J. Griffon, Eurasian, 70-74, 179-183 Ground nesting, 186-187 Grubb, Teryl G., see Gatto, Angela E. Guinda, Mariangela, see Garrido, Jose Rafael Gutierrez, R.J., see Crozier, Michelle L. Gwiazdowicz, Dariusz J., Jerzy Bloszyk, Tadeusz Mizera, and Piotr Tryjanowski, Mesostigmatic mites (Acari Mesostigmata) in White-tailed Sea Eagle nests {Hal- iaeetus albicilla) , 60—65 (dyps fulvus, 70-74, 179-183 Gyrfalcon, 386—393 H Habitat, 310-323 modek 404—416 relations, 192-"209 use, 119-126, 365-377 Habits, food, 149^155 Haliaeetus dlMcilla, 60—65 leucOcephalus, 1—10, 11-18 December 2005 Index to Volume 39 487 Harrier, Eastern Marsh, 106-107 Hatching success, 26-35 Hawk, Broad-winged, 404-416 Cooper’s, 109 Red-tailed, 108, 439-444 Hawks, 378-385 Hegyi, Zoltan, see Sasvari, Lajos Hengstenberg, Derek W. and Francisco Vilella, Nesting ecology and behavior of Broad-winged Hawks in moist karst forests of Puerto Rico, 404-416 Heterogeneity, individual, 210-221 Hieraaetus pennatus, 92-94, 159-163 Hoffmann, Gyula, see Matics, Robert Holschuh, Carmen I. and Ken A. Otter, Using vocal in- dividuality to monitor Queen Charlotte Saw-whet Owls (Aegolius acadicus brooksi), 134—141 Home range, 119-126, 365-377 Horvath, Gyozo, see Matics, Robert Honey-buzzard, European, 184—186 Hunting, adult, 394-403 cooperative, 394—403 fledgling, 394-403 tandem, 394—403 Hybrids, falcon, 386-393 I IberVNeembucu wetlands, 65-69 Ictinia mississippiensis, 108 Idaho, 429-438 Immature movements, 253—263 Individual identification, 286—295 Ingraldi, Michael F., A skewed sex ratio in Northern Gos- hawks: is it a sign of a stressed population?, 247-252 Ingraldi, Michael F., see Boyce, Douglas A., Jr. Ingraldi, Michael F., see Rogers, Andi S. Introduced animals, 173—179 Introgression, 142-148 Iverson, W.F., 102-103 J Juvenile, 11-18 Jensen, Wendy J., Mark S. Gregory, Guy A. Baldassarre, Francisco Vilella, and Keith L. Bildstein, Raptor abundance and distribution on the Llanos wedands of Venezuela, 417-428 Joy, Suzanne M., see Reynolds, Richard T. K Kaibab Plateau, 210-221 Kaminski, TimmothyJ., see Forsman, Eric D. Kato, Yuka and Tadashi Suzuki, Introduced animals in the diets of the Ogasawara Buzzard, an endemic in- sular raptor in the Pacific Ocean, 173-179 Keane, John J., see Andersen, David E. Kennedy, Patricia L., see Boal, Clint W. Kennedy, Patricia L., see McClaren, Erica L. Kennedy, Patricia L., see Boyce, Douglas A., Jr. Kestrel, Alexander’s, 80-83 American, 84-88, 94-97, 378-385 Common, 80-83 Lesser, 127-133 Madagascar, 149-155 Kite, Black, 184-186 Gray-headed, 89-92 Mississippi, 108 Swallow-tailed, 94-97 White-tailed, 378-385 Klein, Akos, see Matics, Robert Kowalski, Jan, see Meyburg, Bernd-U. L Labrador, 11-18 Laing, Dawn K., David M. Bird, and Tony E. Chubbs, First complete migration cycles for juvenile Bald Eagles {Haliaeetus leucocephalus) from Labrador, 11-18 Laing, Dawn K., see Chubbs, Tony E. Landscape change, 310-323 Leptodon cayanensis, 89-92 Lewis, Jeffery C., see Forsman, Eric D. Livezey, Kent B., Iverson (2004) on Spotted Owls and Barred Owls: comments on methods and conclu- sion, 102-103 Llanos, 417-428 M Maciorowski, Grzegorz, see Meyburg, Bernd-U. MacVean, Susie R., see Boyce, Douglas A., Jr. Mahon, Todd and Frank I. Doyle, Effects of timber har- vesting near nest sites on the reproductive success of Northern Goshawks {Accipiter gentilis) , 335-341 Maine, 342-350 Management, adaptive, 335-341 Martinez, Jose Antonio, see Zuberogoitia, Inigo Martinez, Jose E., Carlos Gonzalez, and Jose F. Calvo, Co- operative nesting by a trio of Booted Eagles {Hieraae- tus pennatus) , 92-94 Martinez, Jose Enrique, see Zuberogoitia, Inigo Martinez, Jose E. and Jose F. Calvo, Prey partitioning be- tween mates in breeding Booted Eagles {Hieraaetus pennatus), 159-163 Martinez-Cruz, Begona, see Rodriguez, Carlos Massachusetts, 342-350 Matics, Robert, Sandor Varga, Balazs Opper, Akos Klein, Gydzo Horvath, Alexandre Roulin, Peter Putnoky, and Gyula Hoffmann, Partitioning of genetic (RAPD) variability among sexes and populations of the Barn Owl {Tyto alba) in Europe, 142-148 Maurice, Kevin J., see Forsman, Eric D. Mauritia Jlexuosa, 458-461 May, Bernie, see Bayard de Volo, Shelley McClaren, Erica L., Patricia L. Kennedy, and Donald D Doyle, Northern Goshawk {Accipiter gentilis laingi) 488 Index to Volume 39 VoL. 39, No. 4 post-fledging areas on Vancouver Island, British Co- lumbia, 253—263 McNabb, F.M. Anne, see Quinn, Michael J., Jr. Mediterranean basin, 445-453 central, 184—186 Meehan, Timothy D., see DeLong, John P. Mendes de Carvalho, Gustavo Diniz, see Mendes de Car- valho Filho, Eduardo Pio Mendes de Carvalho Filho, Eduardo Pio, Gustavo Diniz Mendes de Carvalho, and Carlos Eduardo Alencar Carvalho, Observations of nesting Gray-headed Kites {Leptodon cayanensis) in southeastern Brazil, 89-92 Meraz, Juan and Betzabeth Gonzalez-Bravo, First summer records of Ospreys {Pandion haliaetus) along the coast of Oaxaca, Mexico, 18V-188 Mesostigmata, 60-65 Mexico, 187-188, 472-475 Meyburg, Bernd-U., Christiane Meyburg, Tadeusz Mizera, Grzegorz Maciorowski, and Jan Kowalski, Family break up, departure, and autumn migration in Eu- rope of a family of Greater Spotted Eagles {Aquila clanga) as reported by satellite telemetry, 462-466 Meyburg, Christiane, see Meyburg, Bernd-U. Migration, 11-18, 94—97, 462—466 counts, 184—186 patterns, 19—25 spring, 184-186 Miller, Ikarl E., Red-tailed Hawk depredates Mississippi Kite nestling at dawn, 108 Milvago chimachima, 94—97 Milvus migrans, 184-186 Minguez, Eduardo, see Navarro, Joan Minnesota, 222-228, 264-273 Mites, 60-65 Mizera, Tadeusz, see Gwiazdowicz, Dariusz J. Mizera, Tadeusz, see Meyburg, Bernd-U. Molecular sexing, 286-295 Molt, 84-88 preformative, 378-385 Monitoring, 274-285 raptor, 324-334 Monkeys, wedge-capped capuchin, 458-461 Morocco, 70-74 Morphometric, 386-393 Mortality, 222-228 Mouati, Nourdine, see Garrido, Jose Rafael Moulton, Colleen E., Ryan S. Brady, and James R. Belt- hoff, A comparison of breeding season food habits of Burrowing Owls nesting in agricultural and non- agricultural habitat in Idaho, 429-438 N Navarro, Joan, Eduardo Minguez, David Garcia, Carlos Villacorta, Francisco Botella, Jose Antonio Sanchez- Zapata, Martina Carrete, and Andres Gimenez, Dif- ferential effectiveness of playbacks for Little Owls {Athene noctua) surveys before and after sunset, 454— 457 Negro, Juan Jose, see Rodriguez, Carlos Neophron percnopterus, 1 86-1 87 Neotropics, 417-428 Nest alternate, 274-285 area, 296-302, 335-341 historical, 310-323 biology 60-65, 89-92 depredation, 108 structure, 89-92 success, 404-416 Nesting, 109 biology, 89-92, 149-155 record, 472-475 Nesting structures, artificial, 74-79 Nestling, 89-92, 127-133, 247-252 Nests, 149-155 Nevada, 229-236 New England, 342-350 New Hampshire, 342-350 New York City, 109 Nicholls, Mike K., see Eastham, Chris P. Nightjar, Large-tailed, 106-107 Nijman, Vincent, Tineke G. Prins, andJ.H. (Hans) Reu- ter, Timing and abundance of migrant raptors on Bonaire, Netherlands Antilles, 94-97 Noon, Barry R., see Salafsky, Susan R. O Oaxaca, 187-188 Occupancy, 296-302 nest site, 324-334 rate, 166-168 Ogasawara Islands, 173-179 Olympic Peninsula, 365-377 Ontiveros, Diego, Abundance and diet of Alexander’s Kestrel {Falco tinnunculus alexandri) on Boavista Is- land (Archipelago of Cape Verde), 80-83 Opper, Balazs, see Matics, Robert Oregon, 310-323 Ortega-Rubio, Alfredo, see Castellanos, Aradit Osprey, 103-106, 168-173, 187-188 Otter, Ken A., see Holschuh, Carmen I. Ottinger, Mary Ann, see Quinn, Michael J., Jr. Otus Jlammeolus, 19—25 scops, 156-159 Owl, Barn, 65-69, 74-79, 97-101, 142-148, 156-159, 163-166 Barred, 102-103 Burrowing, 429-438 California Spotted, 111-118 Flammulated, 19-25 Great Horned, 111-118, 119-126 Little, 156-159, 454-457 Long-eared, 445-453 December 2005 Index to Volume 39 489 Magellanic Horned, 163-166 Northern Saw-whet, 134-141 Northern Spotted, 184—186, 365-377 Rufous-legged, 163-166 Scops, 156-159 Tawny, 26-35, 156-159 P Palacios, Cesar-Javier, see Gangoso, Laura Palms, Attalea maripa, 458-461 Pandion haliaetus, 103-106, 168-173, 187-188 Pandolfi, Massimo, Alessandro Tanferna, and Giorgia Gaibani, Seasonal patterns of Common Buzzard (Bu- teo buteo) relative abundance and behavior in Pollino National Park, Italy, 466-471 Paraguay, 65—69 Pardihas, Ulyses F.J., see Sauthier, Daniel E. Udrizar Pardihas, Ulyses F.J., Pablo Teta, and Sofia Heinonen For- tabat. Vertebrate prey of the Barn Owl ( Tyto alba) in subtropical wetlands of northeastern Argentina and eastern Paraguay, 65-69 Parental age, 26-35 condition, 26-35 Patla, Susan M., Monitoring results of Northern Goshawk nesting areas in the greater Yellowstone ecosystem: is decline in occupancy related to habitat change?, 324-334 PGA, 386-393 Pellet analysis, 179-183 Pernis apivorus, 184—186 Playback effectiveness, 454-457 Plumage, 84-88 Poland, 60-65 Polygamy, 92-94 Polygyny, 92-94 Population change, 166-168 status, 36-54 trend, 192-209, 229-236 Predation, 156-159 attempted, 106—107 Predator-prey dynamics, 237-246 Predictive model, 168-173 Prey biomass, 65-69 delivery, 404—416 rate, 303-309 density, 237-246 diversity, 264-273 provisioning, 159-163 Principal component analysis, 386-393 Pnns, Tineke G., see Nijman, Vincent Probability of identity, 286-295 Productivity, 1-10, 74-79, 149-155, 166-168, 222-228, 237-246 Puerto Rico, 404—416 Putnoky, Peter, see Matics, Robert Pygmy-Owl, Northern, 476-477 Pyle, Peter, First-cycle molts in North American Falconi- formes, 378-385 Q Quinn, Micheal Jr., John B. French, Jr., F.M. Anne McNabb, and Mary Ann Ottinger, The role of thy- roxine on the production of plumage in the Amer- ican Kestrel {Falco sparverius) , 84—88 R Rabearivony, Jeanneney, see Rene de Roland, Lily-Arison Radiotelemetry, 365-377 Radley, Paul M. and James C. Bednarz, Artificial nest structure use and reproductive success of Barn Owls in northeastern Arkansas, 74-79 RAPD, 142-148 Razafimanjato, Gilbert, see Rene de Roland, Lily-Arison Recruitment, 210-221 Reintroduction, 168-173 Rene de Roland, Lily-Arison, Jeanneney Rabearivony, Harilalaina Robenarimangason, Gilbert Razafiman- jato, and Russell Thorstrom, Breeding biology and food habits of the Madagascar Kestrel {Falco newtom) in northeastern Madagascar, 149-155 Reproduction, 274-285 Reproductive success, 74-79, 102-103, 166-168, 210-221, 335-341 Reservoirs, 168-173 Response, heterospecific, 111-118 Reuter, J.H. (Hans), see Nijman, Vincent Reversed size dimorphism (RSD), 159-163 Review, 192-209 Reynolds, Richard T., In raemoriam: Suzanne Merideth Joy, 190-191 Reynolds, Richard T., see Wiens, J. David Reynolds, Richard T., see Salafsky, Susan R. Reynolds, Richard T, J. David Wiens, Suzanne M. Joy, and Susan R. Salafsky, Sampling considerations for demographic and habitat studies of Northern Gos- hawks, 274—285 Reynolds, Richard T, see Bayard de Volo, Shelley RGB values, 127-133 Rhode Island, 342-350 Robenarimangason, Harilalaina, see Rene de Roland, Lily-Arison Rodents, sigmodontine, 163-166 Rodriguez, Alejandro, see Garcia, Ana Maria Rodriguez, Carlos, Javier Bustamante, Begona Martinez- Cruz, and Juan Jose Negro, Evaluation of methods for gender determination of Lesser Kestrel nestlings, 127-133 Rogers, Andi S., Stephen DeStefano, and Michael F. In- graldi. Quantifying Northern Goshawk diets using remote cameras and observations from blinds, 303- 309 Rojas, Ricardo A. Figueroa and Ema Soraya Corales Stap- pung, Seasonal diet of the Aplomado Falcon {Falco 490 Index to Volume 39 VoL. 39, No. 4 femaralis) in an agricultural area of Araucania, south- ern Chile, 55—60 Rosenfield, Robert N., see Andersen, David E. Roulin, Alexandre, see Matics, Robert S Salafsky, Susan R., Richard T. Reynolds, and Barry R. Noon, Patterns of temporal variation on goshawk re- production and prey resources, 237-246 Salafsky, Susan R., see Reynolds, Richard T. Sahnas-Zavala, Federico, see Castellanos, Aradit Sampling, 274-285 repeated, 296-302 Sanchez-Zapata, Jose Antonio, see Navarro, Joan Sarcoramphus papa, 458-461 Sasvari, Lajos and Zoltan Hegyi, Effects of breeding ex- perience on nest-site choice and the reproductive performance of Tawny Owls {Strix aluco), 26-35 Satellite telemetry, 11-18, 462-466 Sauthier, Daniel E. Udrizar, Analia Andrade, and Ulyses FJ. Pardihas, Predation of small mammals by Rufous- legged Owl, Barn Owl, and Magellanic Horned Owl in Argentinian Patagonia Forests, 163-166 Savanna, 417-428 Schlee, Marsha A., King Vultures {Sarcoramphus papa) for- age in moriche and cucurit palm stands, 458-461 Sea Eagle, White-tailed, 60-65 Seamons, Mark E., see Crozier, Michelle L. Sex allocation, 247-252 ratio, 247-252 Sicily, Channel of, 184—186 Siders, Melissa S., see Boyce, Douglas A., Jr. Smith, Ruth B., see DeLong, John P. Smithers, Brett L., Clint W. Boal, and David E. Andersen, Northern Goshawk diet in Minnesota: an analysis us- ing video recording systems, 264—273 Solensky, Matthew J., see Chubbs, Tony E. Severn, Stan G., see Forsman, Eric D. Spain, southern, 168-173 Squires, John R., see Boyce, Douglas A., Jr. Squirrel, red, 264-273 Stable-hydrogen isotopes, 19-25 Status, 192-209, 342-350 Stopover, 11-18 Stnx aluco, 26-35, 156-159 occidentalis caurina, 365-377 occidentalis, 102-103, 111-118 rufipes, 163-166 varia, 102—103 Subspecies, 142-148 Insular 173—179 Success, reproductive, 166—168 Survey, auditory, 111-118 method, 454—457 population, 454—457 roadside, 417-428, 466-471 techniques, 324—334 Suzuki, Tadashi, see Kato, Yuka T Tagging, genetic, 286-295 Tamiasciurus hudsonicus, 264—273 Tanferna, Alessandro, see Pandolfi, Massimo Taxonomy, 351-364 Taylor, Robert, see Dekker, Dick Techniques, 103-106 Territoriality, 111-118 Territory occupancy, 274—285 Teta, Pablo, see Pardihas, Ulyses FJ. Thorstrom, Russell, see Rene de Roland, Lily-Arison Threats, 36-54 Thyroxine, 84—88 Tides, 1-10 Timber harvesting, 335-341 Titus, Kimberly, see Andersen, David E. Topinka, J. Rick, see Bayard de Volo, Shelley Trio, 92-94 Trophic plasticity, 445-453 Tryjanowski, Piotr, see Gwiazdowicz, Dariusz J. Turan, Levent, The status of diurnal birds of prey in Tur- key, 36—54 Turkey, 36-54 Tyto alba, 65-69, 97-101, 142-148, 156-159, 163-166 pratincola, 74-79 U United States, northeastern, 342-350 western, 192-209 V Varga, Sandor, see Matics, Robert Venezuela, 417-428 Vermont, 342-350 Video surveillance, 303—309 Vietnam, 106—107 Vilella, Francisco, see Jensen, Wendy J. Vilella, Francisco, see Hengstenberg, Derek W. Villacorta, Carlos, see Navarro, Joan Vocal individuality, 134—141 Vocalization, 476—477 Vulture, King, 458-461 W Washington, 365-377 Weather, 1-10, 229-236 condition, 26-35 Wetlands, 417-428 Wiens, J. David and Richard T. Reynolds, Is fledging suc- cess a reliable index of fitness in Northern Gos- hawks?, 210-221 Wiens, J. David, see Reynolds, Richard T. Wiley, James W. and Orlando H. Garrido, Taxonomic sta- December 2005 Index to Volume 39 491 tus and biology of the Cuban Black-Hawk, Buteogallus anthracinus gundlachii (Aves: Accipitridae) , 351—364 Wintering, 70-74 Woodbridge, Brian, see Boyce, Douglas A., Jr. X Xirouchakis, Stavros M., The diet of Eurasian Griffons {Gyps fulvus) in Crete, 179-183 Z Zabala, Jabi, see Zuberogoitia, Inigo Zuberogoitia, Inigo, Jose Antonio Martinez, Jabi Zabala, and Jose Enrique Martinez, Interspecific aggression and nest-site competition in a European owl com- munity, 156-159 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 Fabrizio Sergio Ian G. Warrentin James W. Watson BOOK REVIEW EDITOR Jeffrey S. Marks CONTENTS FOR VOLUME 39, 2005 Number 1 The Influence of Tide and Weather on Provisioning Rates of Chick-rearing Bald Eagles in Vancouver Island, British Columbia. Kyle Hamish Elliott, Christopher E. Gill, and John E. Elliott 1 First Complete Migration Cycles for Juvenile Bald Eagles {Haliaeetus LeUCOCEPHALUS) from Labrador. Dawn K. Laing, David M. Bird, and Tony E. Chubbs ........ 11 Investigating Fall Movements of Hatch-year Flammulated Owls ( Otus FLAMMEOLUS) IN CENTRAL NeW MEXICO USING STABLE HYDROGEN ISOTOPES. John P. DeLong, Timothy D. Meehan, and Ruth B. Smith 19 Effects of Breeding Experience on Nest-site Choice and the Reproductive Performance of Tawny Owls (Strix aluco) . Lajos Sasvari and Zoitan Hegyi 26 The Status of Diurnal Birds of Prey in Turkey. Levent Turan 36 Short Communications Seasonal Diet of the Aplomado Falcon (Falco femoralis) in an Agricultural Area of Araucania, Southern Chile. Ricardo A. Figueroa Rojas and Ema Soraya Corales Stappung 55 Mesostigmatic Mites (Acari: Mesostigmata) in White-tailed Sea Eagle Nests {Haliaeetus albicilla) . Dariusz J. Gwiazdowicz, Jerzy Bloszyk, Tadeusz Mizera, and Piotr Tryjanowski 60 Vertebrate Prey of the Barn Owl ( Tyto alba) in Subtropical Wetlands of Northeastern Argentina and Eastern Paraguay. Ulyses FJ. Pardinas, Pablo Teta, and Sofia Heinonen Fortabat 65 Absence of the Eurasian Griffon {Gyps fulvus) in Northern Morocco. Jose Rafael Garrido, Alvaro Camiha, Mariangela Guinda, Marfa Egea, Nourdine Mouati, Alfonso Godino, and J.L. Paz de la Rocha 70 Artificial Nest Structure Use and Reproductive Success of Barn Owls in Northeastern Arkansas. Paul M. Radley and James C. Bednarz 74 Abundance and Diet of Alexander’s Kestrel {Falco tinnunculus alexandri) on Boavista Island (Archipelago of Cape Verde) . Diego Ontiveros 80 The Role of Thyroxine on the Production of Plumage in the American Kestrel {Falco sparverius) . Michael J. Quinn, Jr., John B. French, Jr., F.M. Anne McNabb, and Mary Ann Ottinger 84 Observations of Nesting Gray-headed Kites {Leptodon cayanensis) in Southeastern Brazil. Eduardo Pio Mendes de Carvalho Filho, Gustavo Diniz Mendes de Carvalho, and Carlos Eduardo Aencar Carvalho 89 Cooperative Nesting by a Trio of Booted Eagles {Hieraaetus pennatus). Jose E. Martmez, Carlos Gonzalez, and Jose F. Calvo 92 Timing and Abundance of Migrant Raptors on Bonaire, Netherlands Antilles. Vincent Nijman, Tineke G. Prins, andJ.H. (Hans) Reuter 94 The Relationship of Foraging Habitat to the Diet of Barn Owls ( Tyto alba) from Central Chile. Sabine Begall 97 Letters Iverson (2004) on Spotted Owls and Barred Owls: Comments on Methods and Conclusion. Kent B. Livezey 102 Using a Portable, Anchor-bolt Ladder to Access Rock-nesting Osprey. Tony E. Ghubbs, Matthew J. Solensky, Dawn K. Laing, David M. Bird, and Geoff Goodyear 103 Attempted Predation on a Large-tailed Nightjar ( Caprimulgus macrurus) by an Eastern Marsh-Harrier (Circus spilonotus) in Coastal Vietnam. James A. Fitzsimons 106 Red-tailed Hawk Depredates Mississippi Kite Nestling at Dawn. Karl E. Miller 108 First Nesting of Cooper’s Hawks (Accipiter cooperu) in New York City Since 1955 . Robert DeCandido 1 09 Manuscript Referees 110 Number 2 The Effect of Broadcasting Great Horned Owl Vocalizations on Spotted Owl Vocal Responsiveness. Michelle L. Crozier, Mark E. Seamans, and R.J. Gutierrez Ill Home Range and Habitat Use by Great Horned Owls {Bubo virginianus) in Southern California. Jason R. Bennett and Peter H. Bloom 119 Evaluation of Methods for Gender Determination of Lesser Kestrel Nestlings. Carlos Rodriguez, Javier Bustamante, Begona Martmez-Cruz, and Juan Jose Negro 127 Using Vogal Individuality to Monitor Queen Charlotte Saw-whet Owls (AeGOLIUS ACADICUS BROOKS !) . Carmen I. Holschuh and Ken A. Otter 134 Partitioning of Genetic (RAPD) Variability among Sexes and Populations OF THE Barn Owl ( TyTO alba) in Europe. Robert Matics, Sandor Varga, Balazs Opper, Akos Klein, Gyozo Horvath, Alexandre Roulin, Peter Putnoky, and Gyula Hoffmann 142 Breeding Biology and Food Habits of the Madagascar Kestrel {Falco newtoni) IN Northeastern Madagascar. Lily-Arison Rene de Roland, Jeanneney Rabearivony, Harilalaina Robenarimangason, Gilbert Razafimanjato, and Russell Thorstrom 1 49 Short Communications Interspecific Aggression and Nest-site Competition in a European Owl Community. Inigo Zuberogoitia, Jose Antonio Martinez, Jabi Zabala, and Jose Enrique Martinez 156 Prey Partitioning between Mates in Breeding Booted Eagles (Hieraaetus pennatus) . Jose E. Martinez and Jose F. Calvo 159 Predation of Small Mammals by Rufous-legged Owl, Barn Owl, and Magellanic Horned Owl IN Argentinean Patagonia Forests. Daniel E. Udrizar Sauthier, Analia Andrade, and Ulyses F. J. Pardinas 163 Changes in Site Occupancy and Nesting Performance of Peregrine Falcons in Colorado, 1963-2004. James H. Enderson 166 Analysis of Reservoir Selection by Wintering Ospreys (Pandion haliaetus hauaetus) in Andalusia, Spain: A Potential Tool for Reintroduction. Eva Casado and Miguel Ferrer 168 Introduced Animals in the Diets of the Ogasawara Buzzard, an Endemic Insular Raptor in the Pacific Ocean. Yuka Kato and Tadashi Suzuki 173 The Diet of Eurasian Griffons {Gyps fulvus) in Crete. Stavros M. Xirouchakis 179 Letters Are Earlier Estimates of Accipitriformes Crossing the Channel of Sicily (Central Mediterranean) During Spring Migration Accurate? Nicolantonio Agostini 184 Ground Nesting by Egyptian Vultures {Neophron percnoptekus) in the Canary Islands. Laura Gangoso and Cesar-Javier Palacios 186 First Summer Records of Ospreys {Pandion haliaetus) Along the Coast of Oaxaca, Mexico. Juan Meraz and Betzabeth Gonzalez-Bravo 187 Number 3 Introduction Preface: Proceedings of the International Symposium on the Ecology and Management of Northern Goshawks. Clint W. Boal 189 In Memoriam: Suzanne Merideth Joy. Richard T. Reynolds 190 Status Technical Review of the Status of Northern Goshawks in the Western United States. David E. Andersen, Stephen DeStefano, Michael I. Goldstein, Kimberly Titus, Cole Crocker-Bedford, John J. Keane, Robert G. Anthony, and Robert N. Rosenfield 192 Biology Is Fledging Success a Reliable Index of Fitness in Northern Goshawks? J. David Wiens and Richard T Reynolds 210 Productivity and Mortai.ity of Northern Goshawks in Minnesota. Clint W. Boal, David E. Andersen, and Patricia L. Kennedy 222 Relationships Between Winter and Spring Weather and Norihern Goshawk {AcapiTER gentilis) Reproduction in Northern Nevada. Graham D. Fairhurst and Marc J. Bechard 229 Patterns of Temporal Variation in Goshawk Reproduction and Prey Resources. Susan R. Salafsky, Richard T. Reynolds, and Barry R. Noon 237 A Skewf.d Sex Ratio in Northern Goshawks; Is It a Sign of a Stressed Population? Michael E Ingraldi 247 Northern Goshawk {Accjpiter gentilis laingi) Post-fledging Areas on Vancouver Isiand, British Columbia. Erica L. McClaren, Patricia L. Kennedy, and Donald D. Doyle 253 Northern Goshawk Diet in Minnesota: An Anaiasis Using Video Recording Systems. Brett L. Smithers, Clint W. Boal, and David A. Andersen 264 Techniques Sampling Considerations for Demographic and Habitat Studies of Northern Goshawks. Richard T. Reynolds, J. David Wiens, Suzanne M. Joy, and Susan R. Salafsky 274 Population Genetics and Genotyping for Mark-Recapture Studies of Northern Goshawks {Acciitper gentilis) on the Kaibab Plateau, Arizona. Shelley Bayard de Volo, Richard T. Reynolds, J, Rick Topinka, Bernie May, and Michael F. Antolin 286 When Are Goshawks Not There? Is a Single Visit Enough to Infer Absence at Occupied Nest Areas? Douglas A. Boyce, Jr., Patricia L. Kennedy, Paul Beier, Michael F. Ingraldi, Susie R. MacVean, Melissa S. Siders, John R. Squires, and Brian Woodbridge 296 Quantifying Northern Goshawk Diets Using Remote Cameras and Observations from Blinds. Andi S. Rogers, Stephen DeStefano, and Micheal F. Ingraldi 303 Conservation Temporal Pati'erns of Northern Goshawk Nest Area Occupancy and Habitat: A Retrospective Anai.ysis. Steven Desimone and Stephen DeStefano 310 Monitoring Results of Northern Goshawk Nesting Areas in the Greater Yellowstone Ecosystem: Is Decline in Occupancy Reiated to Habitat Change? Susan M. Patla 324 Effects of Timber Harvesting Near Nest Sites on the Reproductive Success of Northern Goshawks {Accipiter gentius). Todd Mahon and Frank I. Doyle 335 A Review of the Status and Distribution of Northern Goshawks in New England. Stephen DeStefano 342 Number 4 Taxonomic Status and Biology of the Cuban Black-Hawk, Buteogallus ANTHRACINUS GUNDLACHII (AVES: AcCIPITRIDAE) . James W. Wiley and Orlando H. Garrido.... 351 Home Range and Habitat Use of Northern Spotted Owls on the Olympic Peninsula, Washington. Eric D. Forsman, TimmothyJ. Kaminski, Jeffery C. Lewis, Kevin J. Maurice, Stan G. Sovern, Cheron Ferland, and Elizabeth M. Glenn 365 First-cycle Molts in North American Falconiformes. Peter Pyle 378 Morphometric Analysis of Large Falco Species and Their Hybrids with Implications for Conservation. Chris p. Eastham and Mike k. Nichoiis 386 A Change in Foraging Success and Cooperative Hunting by a Breeding Pair OF Peregrine Falcons and Their Fledglings. Dick Dekker and Robert Taylor 394 Nesting Ecology and Behavior of Broad-winged Hawks in Moist Karst Forests of Puerto Rico. Derek W. Hengstenberg and Francisco J. Vilella 404 Raptor Abundance and Distribution in the Llanos Wetlands of Venezuela. Wendy J. Jensen, Mark S. Gregory, Guy A. Baldassarre, Francisco J. Vilella, and Keith L. Bildstein .... 417 A Comparison of Breeding Season Food Habits of Burrowing Owls Nesting IN Agricultural and Nonagricultural Habitat in Idaho. Colleen e. Moulton, Ryan S. Brady, and James R. Belthoff 429 Red-tailed Hawk Dietary Overlap with Northern Goshawks on the Kaibab Plateau, Arizona. Angela E. Gatto, Teryl G. Grubb, and Carol L. Chambers 439 Bat Predation by Long-eared Owls in Mediterranean and Temperate Regions OF Southern Europe. Ana Marfa Garcfa, Francisco Cervera, and Alejandro Rodriguez 445 Short Communications Differential Effectiveness of Playbacks for Little Owls {Athene noctua) Surveys before and AFTER Sunset. Joan Navarro, Eduardo Mfnguez, David Garcfa, Carlos Villacorta, Francisco Botella, JosJ Antonio S1000 pocket sized telemetry receiver will impress you. and the prne will convince you. Othtr ftatum Imludt: • factory toned to any 4MH/ wide segment in the 148 174MHz Band • Very high sensitivity of •148dBm to ISOdBm • Illuminated display and keypad for use in low light or darkness • User selectable Kan rates from 1-30 seconds in 1 second steps • Rechargeable battenes operate the receiver for 12 hours and can be replaced with standard AA Alkaline batteries in the field. Both 12vdc and llOvac chargers are included. • b.r (IS.bcm) high, 2.6" (6.6cm) wide. 1.5" (3.8cm) deep. • 3 year warranty • 1 day delivery $ 695.00 Please specify desired 4MHz wide segment in the 148-174HHZ band Visit our website for complete specifications, operating manual and information on the R- 1000 or call our toll free number to order your receiver now. Try th€ New R IOOO and YduH Impressed! COMMUNICATIONS SPECIALISTS. INC. ■ • .''ri ('i-- ! A.I-' ! ; 1 A OCj • 1 '^.^0 r, •.! .S .S A Buteq Books Toll Free: 800-722-2460 phone: 434-263-8671 fax: 434-263-4842 Specializing in Ornithology Buteo Books is the largest retailer of Ornithology books in North America, with over 2,000 in-print titles, and hundreds of out-of-print titles available. Rffutyt P^m NEW, IN-PRINT ORNITHOLOGY The most popular authors, cutting edge studies, and unusual subjects. We stock scientific texts, hard-to-find foreign books, and entertaining reading; plus audio, video, and software. RARE AND OUT-OF-PRINT ORNITHOLOGY Buteo Books has a wide selection of used, out-of-print, and scarce titles available, including classics on birds of prey and falconry. Our stock changes weekly, so call to check availability. BIRDS OF NORTH AMERICA SERIES This series consists of individual species accounts for each of the species which breed in the United States and Canada, including diurnal and nocturnal raptors. These illustrated reviews provide comprehensive summaries of the current knowledge of each species, with range maps and extensive lists of references. As the exclusive distributor of print copies in this series, Buteo Books is pleased to offer these accounts for $7.50 each. All 716 profiles are listed in taxonomic order on our website. ■a Buteo Books; 3130 Laurel Road; Shipman, VA 22971; USA Visit our website for more information; www.buteobooks.com r Essential Guides "One of the best bird field guides ever published." — Oriental Bird Club Bulletin Birds of Southeast Asia Craig Robson • More than 140 full-color plates • All 1 ,270 species covered in detail • Up-to-date text covers identification, voice, habitat, behavior, and range of all the region's species and distinctive subspecies • Complete coverage of some fifteen Southeast Asian countries and regions 304 pages. 142 color plates. 5 3/4 x 8 1/4. Princeton Field Guides Series Paper $29.95 0-691-12435-3 Not for sale in the Commonwealth (except Canada) and the European Union "Jerry Liguori has created a book that can easily be toted in the field, and is an absolute must-have for any raptor enthusiast!"— Brian K. Wheeler, author of Raptors of Eastern/Wesfern North America Hawks from Every Angle How to Identify Raptors In Flight Jerry Liguori Foreword by David Allen Sibley • The essential new approach to identifying hawks in flight • Innovative, accurate, and field-tested • Compares and contrasts species easily confused with one another • Provides the pictures (and words) needed for identifica- tion in the field • Covers the 1 9 species common to North America • User-friendly format for quick use 339 color photos. 32 b/w photos. 7 1/2x9 1/2. Paper $19.95 0-691-11825-6 Cloth $55.00 0-691-11824-8 Cclchratint^ lOO Years of Excellence PRINCETON .1 , ’ ■ Kf jd cxccrpls or-inc ‘Umversitv ‘.Press V 2006 ANNUAL MEETING The Raptor Research Foundation, Inc. 2006 annual meeting will be held in conjunction with the Fourth North American Ornithological Conference on 3-7 October 2006 in Veracruz, Mexico. For more information about the meeting see http:/ /www.naoc2006.org/ Persons interested in predatory birds are invited to join The Raptor Research Foundation, Inc. (see: http:/ /biology.boisestate.edu/ raptor/). Send requests for information concerning membership, subscriptions, special publications, or change of address to OSNA, 5400 Bosque Blvd., Suite 680, Waco TX 76710, U.S.A. The, Journal of Raptor Research (ISSN 0892-1016) is published quarterly and available to individuals for $40.00 per year and to libraries and institutions for $65.00 per year from The Raptor Research Foundation, Inc., 14377 1 17th Street South, Hastings, Minnesota 55033, U.S.A. (Add $3 for destinations outside of the continental United States.) Periodicals postage paid at Hastings, Minnesota, and additional mailing offices. POSTMASTER: Send address changes to The Journal of Raptor Research, OSNA, P.O. Box 1897, Lawrence, KS 66044-8897, U.S.A. Printed by Allen Press, Inc., Lawrence, Kansas, U.S.A. Copyright 2005 by The Raptor Research Foundation, Inc. Printed in U.S.A. @ This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). Raptor Research Foundation, Inc. Grants and Awards For details and additional information visit: http:/ /biology.boisestate.edu/raptor/grants%20and%20awards.htm Awards for Recognition of Significant Contributions The Tom Cade Award is a non-monetary award that recognizes an individual who has made significant advances in the area of captive propagation and reintroduction of raptors. The Fran and Frederick Hamerstrom Award is a non-monetary award that recognizes an individual who has contributed significantly to the under- standing of raptor ecology and natural history. Submit nominations for either award to: Dr. Clint Boal, Texas Cooperative Fish and Wildlife Research Unit, BRD/USGS, Texas Tech University, 15th Street & Boston, Ag Science Bldg., Room 218, Lubbock TX 79409-2120 U.S.A.; phone: 806-742-2851; e-mail: cboal@ttu.edu Awards for Student Recognition and Travel Assistance The James R. Koplin Travel Award is given to a student who is the senior author and presenter of a paper or poster to be presented at the RRF meeting for which travel funds are requested. Application deadline: due date for meeting abstract. Contact: Dr. Patricia A. Hall, 5937 E. Abbey Rd., Flagstaff, AZ 86004; phone: 520-526-6222 U.S.A.; e-mail: pah@spruce.for.nau.edu The William C. Anderson Memorial Award is given to both the best student oral and poster presentation at the annual RRF meeting. The paper cannot be part of an organized symposium to be considered. Application deadline; due date for meeting abstract, no special application is needed. Contact: Rick Gerhardt, Sage Science, 319 SE Woodside Ct., Madras, OR 97741 U.S.A; phone: 541-475-4330; email; rgerhardt@madras.net Grants Application deadline for all grants is February 15 of each year; selections will be made by April 15. The Dean Amadon Grant for up to $1000 is designed to assist persons working in the area of systematics (tax- onomy) and distribution of raptors. The Stephen R. TuUy Memorial Grant for up to $500 is given to sup- port research and conservation of raptors, especially to students and amateurs with limited access to alter- native funding. Agency proposals are not accepted. Contact for both grants; Dr. Carole Griffiths, 251 Mar ding Ave., Tarrytown, NY 10591 U.S.A.; phone: 914-631-2911; e-mail; cgriff@liu.edu The Leslie Brown Memorial Grant for up to $1400 is given to support research and/or the dissemination of information on African raptors. Contact: Dr. Jeffrey L. Lincer, 9251 Golondrina Drive, La Mesa, CA 91941, U.S.A.; e-mail: JeflLincer@tns.net