The Journal of Raptor Research Volume 34 Number 4 December 2000 Published by The Raptor Research Foundation, Inc. THE RAPTOR RESEARCH EOUNDATION, INC. (Eounded 1966) OFFICERS PRESIDENT: Michael N. Kochert VICE-PRESIDENT: Keith L. Bildstein SECRETARY: Patricia A. Hall TREASURER: Jim Fitzpatrick BOARD OF DIRECTORS NORTH AMERICAN DIRECTOR #1: Philip Detrich INTERNATIONAL DIRECTOR #3: INTERNATIONAL DIRECTOR#!: Eduardo Inigo-Elias NORTH AMERICAN DIRECTOR #2: Petra Bohall Wood NORTH AMERICAN DIRECTOR #3: Robert Lehman Beatriz Arroyo DIRECTOR AT LARGE #1 : Jemima ParryJones DIRECTOR AT LARGE #2: Robert Kenward DIRECTOR AT LARGE #3: Michael W. Coelopy DIRECTOR AT LARGE #4; Miguel Ferrer DIRECTOR AT LARGE #5: John A. Smallwood DIRECTOR AT LARGE #6: Brian A. Milsap INTERNATIONAL DIRECTOR #2: Reuven Yosef ^ >£< >!< %l0 sirf ^ ^ ^ ^ ^ ^ "T* 'T’ k^Jw EDITORIAL STAFF EDITOR: MarcJ. Bechard, Department of Biology, Boise State University, Boise, ID 83725 U.S.A. BOOK REVIEW EDITOR: Jeffrey S. Marks, Montana Cooperative Research Unit, University of Montana, Missoula, MT 59812 U.S.A. SPANISH EDITOR: Cesar Marquez Reyes, Institute Humboldt, Colombia, AA. 094766, Bogota 8, Colombia EDITORIAL ASSISTANTS: Joan Ciark, Keleigh Hague-Bechard, Elise Vernon Schmidt 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, letters to the editor, thesis abstracts and book reviews. Contributors should submit a typewritten original and three copies to the Editor. All submissions must be typewritten and double-spaced on one side of 216 X 278 mm (8% X 11 in.) or standard international, white, bond paper, with 25 mm (1 in.) margins. The cover page should contain a title, the author’s full name(s) and address (es). Name and address should be centered on the cover page. If the current address is different, indicate this via a footnote. A short version of the title, not exceeding 35 characters, should be provided for a running head. An abstract of about 250 words should accompany all research articles on a separate page. Tables, one to a page, should be double-spaced throughout and be assigned consecutive Arabic numer- als. Collect all 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 2Uhour clock (e.g., 0830 H and 2030 H) and “continental” dating (e.g., 1 January 1990). Refer to a recent issue of the journal for details in format. Explicit instructions and publication policy are outlined in “Information for contributors,” J. Raptor Res., Vol. 34(4), and are available from the editor. COVER: Adult male Shikra {Accipiter badius) . Painting by Lorenzo Starnini. ASSOCIATE EDITORS Juan Jose Negro Charles J. Henny Ian G. Warkentin Marco Restani Allen M. Fish Fabian Jaksic Daniel E. Varland Cole Crocker-Bedford James C. Bednarz Contents A Partial PostJuvenile Molt and Transitional Plumage in the Shikra {Accipiter BADIUS) AND GrEY FrOG HAWK (AcaPUER SOLOENSIS) . Marc Herremans and Michel Louette .. 249 Turnover and Dispersal of Prairie Falcons in Southwestern Idaho. Robert n. Lehman, Karen Steenhof, Leslie B. Carpenter, and Michael N„Kochert 262 Roost Sites of Radio-Marked Mexican Spotted Owls in Arizona and New Mexico: Sources of Variability and Descriptive Characteristics. Joseph l. Ganey, William m. Block, and Rudy M. King 270 Barred Owl and Spotted Owl Populations and Habitat in the Central Cascade Range of Washington. Dale R. Herter and Lorin L. Hicks 279 Food Habits of Bald Eagles Wintering in Northern Arizona. Teryi G. Gmbb and Roy G. Lopez 287 Nest Features and Nest-Tree Characteristics of Short-Toed Eagles {Circaetus GALUCUS) IN THE DADIA-LeFKIMI-SOUFU FOREST, NORTHEASTERN GREECE. Dimitris E. Bakaloudis, Christos G. Vlachos, and Graham J. Holloway 293 Are Northern Saw-Whet Owls Nomadic? Jeffrey s. Marks and John H. Doremus 299 Relationship Between Raptors and Rabbits in the Diet of Eagle Owls in Southwestern Europe: Competition Removal or Food Stress? David Serrano 305 An Evaluation of Methyl Anthranilate, Aminoacetophenone, and Unfamiliar Coloration as Feeding Repellents to American Kestrels. Michael K Nkhoiis, Oliver P. Love, and David M. Bird 311 Short Communications Responsiveness of Nesting Eurasian Kest re ls Falco tinnunculus to Call Playbacks. Luca Salvati, Alberto Manganaro, and Simone Fattorini 319 The Breeding Success of Tawny Owls (Strixaluco) in a Mediterranean Area: A Long- Term Study in Urban Rome. Lamberto Ranazzi, Alberto Manganaro, and Luca Salvati 322 Nocturnal AcnviTY of Lesser Kestrels Under ARnriciAL Lighting Condthons in Seville, Spain. Juan Jose Negro, Javier Bustamante, Ciro Melguizo, Jose Luis Ruis, and Juan Manuel Grande 327 Nest-Site Characteristics of Crested Caracaras in La Pampa, Argentina. Michael I. Goldstein 330 Diet of the Barn Owl (Tyto alba tuidara) in Northwestern Argentine Patagonia. Maria S. Pillado and Ana Trejo 334 Letters 339 Book Review. Edited by Jeffreys. Marks.............. 342 The Raptor Research Foundation, Inc. gratefully acknowledges a grant and logistical support provided by Boise State University to cissist in the publication of the journal. THE JOURNAL OF RAPTOR RESEARCH A QUARTERLY PUBLICATION OF THE RAPTOR RESEARCH FOUNDATION, INC. VoL. 34 December 2000 No. 4 J. Raptor Res. 34(4) ;249-261 © 2000 The Raptor Research Foundation, Inc. A PARTIAL POST-JUVENILE MOLT AND TRANSITIONAL PLUMAGE IN THE SHIKRA (ACCIPITER RADIUS) AND GREY FROG HAWK {ACCIPITER SOLOENSIS) Marc Herremans and Michel Louette Royal Museum for Central Africa, Department Zoology, Leuvensesteenweg 13, B-3080 Tervuren, Belgium Abstract. — Molt has been poorly studied in the Accipitridae. Examination of museum specimens showed that there are three age-related plumages in the Shikra (Accipiter badius) and Grey Frog Hawk (A. soloensis) similar to the pattern known in the Levant Sparrowhawk {A. brevipes) . The juvenile plumage with its distinctively-spotted underside is replaced by a transitional post-juvenile plumage during a partial contour molt between 4-10 mo of age. More feathers on the ventral side than on the dorsal side are replaced during this first contour molt, which is arrested at various stages of incomplete feather replace- ment. Usually, a significant part of the ventral pattern changes from spotted to barred, whereby the barring is on average more prominent than in adults. The early development of a transitional post- juvenile plumage might be related to early sex signaling. The adult plumage replaces the transitional post-juvenile plumage during a complete molt at about one year of age. In the subspecies A. b. poliopsis of the Shikra, which has almost no sexual dimorphism in the adult plumage, the transitional plumage is uncommon and very poorly developed. Key Words: Shikra; Accipiter badius; Grey Frog Hawk; Accipiter soloensis; Levant Sparrowhawk, Accipiter brevipes; contour molt, transitional postfuvenile plumage. Muda parcial post juvenil y de transicion de plumaje en Accipiter badius y Accipiter soloensis Resumen. — La muda ha sido poco estudiada en las Accipitridae. El examen de especimenes de museo demostro que hay tres plumajes relacionadas con la edad en Accipiter badius y en A. soloensis similar al patron conocido en A. brevipes. El plumaje juvenil con su distintivo salpicado por debajo es remplazado por un plumaje de transicion post juvenil durante una muda parcial entre los 4—10 meses de edad. Mas plumas del costado ventral que en el dorsal son remplazadas durante esta muda, la cual se detiene en varias etapas del reemplazo incompleto de plumas. Usualmente una parte insignificante del patron ventral cambia de salpicado a barrado, en donde el barrado es en promedio mas prominente que en los adultos. El desarrollo temprano de un plumaje post juvenil de transicion puede estar relacionado con sehales sexuales tempranas. El plumaje adulto reemplaza al plumaje post-juvenil de transicion dur- ante una muda completa al aho de edad. En la subespecie A. b. poliopsis la cual tiene un dimorfismo sexual en el plumaje adulto, el plumaje transicional es poco comun y pobremente desarrollado. [Traduccion de Cesar Marquez] The molt of flight feathers has been studied in some species of Accipiter (e.g., A. gentilis, A. nisus, A. cooperii, A. striatus, A. melanoleucus, A. badius; Stresemann and Stresemann 1966, Hartley 1976, Fischer 1980, Newton and Marquiss 1982, Schmitt et al. 1982, Henny et al. 1985), but body molt is less well-documented. The larger goshawks gener- ally undergo a complete molt taking several months during the second year of life (Hartley 1976, Fischer 1980). In the Eurasian Sparrowhawk (A. nisus) , the best-studied species, adults molt dur- ing the breeding season in summer. Juveniles also undergo a complete molt that lasts several months during late summer and can continue into the sec- 249 250 Herremans and Louette VoL. 34, No. 4 Figure 1. Ventral aspect of the plumages of the Levant Sparrowhawk {Accipiter brevipes). From left to right, juvenile (BMNH 1965. M. 1087), October, Israel; trcmsitional male (BMNH 1934.1.1.1221), June, Iran (note bold stripes remain from the juvenile plumage and tail feathers juvenile) ; adult male (BMNH 1956.57.13), May, Caucasus; adult female (BMNH 1934.1.1.1220), May, Iran. Photograph courtesy of the BMNH. ond year of life (Stresemann and Stresemann 1960, 1966, Newton and Marquiss 1982). During this first molt, all the juvenile plumage is replaced by the adult plumage, except for the odd feather which allows one to identify second-year birds during the next year (Newton and Marquiss 1982). The mi- gratory Sharp-shinned (A. striatus) and Cooper’s (A. cooperii) Hawks similarly molt direcdy from the juvenile plumage to the adult plumage with a com- plete molt during the first summer after hatching (Mueller et al. 1979, 1981). Like the European Sparrowhawk, adult Levant Sparrowhawks (A. brevipes) start a complete molt during the breeding season in summer and com- plete it in autumn, generally before migration (Cramp and Simmons 1980, Forsman 1999). Juve- niles, however, undergo a partial contour molt on the wintering grounds in Africa when only about six months old, and return to the breeding grounds in a transitional post^uvenile plumage. Adult plumage is acquired during a complete molt in summer at about one year of age; some birds carry over some juvenile feathers for the next molt (Cramp and Simmons 1980, Clark and Yosef 1998, Forsman 1999). The transitional plumage, which on the underside has a striking mixture of boldly streaked juvenile feathers and barred adult type feathers (Fig. 1; Clark and Yosef 1998, Forsman 1999) is kept only for about half a year, from winter until summer. December 2000 Juvenile Molt in Small Accipiters 251 Molt information for the Shikra {A. badius) is scanty and contradictory. The Asian race A. b. cen- chroides is said to follow the general pattern of the Eurasian Sparrowhawk (Cramp and Simmons 1980). Thiollay (1975) mentioned that A. b. sphen- urus in the Ivory Coast only adopts adult plumage in the course of the second year. According to Friedmann (1930), there is an immature plumage with variable underside pattern, and Zimmerman et al. (1996) described an immature plumage re- sulting from a first molt. Verheyen (1953) rejected the existence of an immature plumage between ju- venile and adult in A. b. polyzonoides in the Congo. Similarly, Schmitt et al. (1982) did not find any indication of an intermediate plumage in this race in South Africa, where the name Little Banded Goshawk is commonly used, despite the fact that they documented a partial post-juvenile contour molt. The Shikra and Grey Frog Hawk (also called Chi- nese Goshawk; A. soloensis) are excellent species for the study the sequence of plumages on museum skins, because they have contrastingly different pat- terning on the underside between the juvenile and adult; boldly spotted and striped in the juvenile and finely barred (or mainly plain in the Grey Frog Hawk) in the adult. Tail feathers are boldly banded in juveniles, and particularly the inner and outer pair have reduced markings in the adult. Plumage classification is further facilitated by discrete breed- ing seasons, and discrete breeding and nonbreed- ing ranges in the Grey Frog Hawk. The migratory Grey Frog Hawk and the Asian races of the Shikra breed in spring (Ali and Ripley 1983), while in Af- rica the Shikra breeds late in the dry and early in the wet season, although this means in different months of the year at opposite sides of the equator (Elgood et al. 1973, Smeenk and Smeenk-Enserink 1977, Brown et al. 1982, Allan 1997). The extent of movements differs in the Shikra. The northcen- tral African race A. b. sphenurus is migratory in West Africa, where it moves north after breeding to molt (Elgood et al. 1973). It seems to be more resident in the eastern part of its range (Brown et al. 1982, del Hoyo et al. 1994). The southern Af- rican race A. b. polyzonoides does not undertake a regular migration, but is highly nomadic, particu- larly in the dry season (Allan 1997). Of the four Asian races, only the westernmost A. b. cenchroides is migratory (Blanford 1895, King et al. 1978, Ali and Ripley 1983, del Hoyo et al. 1994). During work on plumages and ecology of some African Accipiters (Louette 2000, Herremans et al. 2001), we became aware of the existence in several of the smaller species of a distinct transitional plumage, kept for a short period between the typ- ical juvenile and adult plumage. Herein, we de- scribe this transitional post-juvenile plumage in the Shikra and Grey Frog Hawk with reference to the similar and better-documented pattern in the Le- vant Sparrowhawk. We report on aspects of the molt sequence relevant to the development of the transitional plumage, and on its possible function. Methods We examined the plumages of the Grey Frog Hawk at the Natural History Museum (BMNH), Tring, and stud- ied the two African and four Asian subspecies of the Shik- ra in the Royal Museum for Central Africa (RMCA) , Ter- vuren, and BMNH collections: 148 A. b. polyzonoides (southern Africa), 119 A. b. sphenurus (northcentral Af- rica), 150 A. b. dussumieri (India, Bangladesh), 115 A b. poliopsis (northeastern India to Thailand and Vietnam), 43 A. b. cenchroides (Azerbaijan to northwestern India), and 32 A. b. badius (southwestern India, Sri Lanka) We noted the state of the plumage (juvenile/ adult) separate- ly for the contour feathers on uppersides, undersides, and rectrices and checked for active molt of primaries and tail feathers. Contour feather renewal was estimated in percent (mostly in steps of 10%) for the dorsal and ventral side separately. Sample sizes differed because some specimens were undated while, in others, the state of preparation precluded the assessment of contour molt extent, or limited extent in transitional birds precluded the comparison of ventral barring with that of adults Results Similar to the pattern that develops in the Le- vant Sparrowhawk, juveniles of the migratory Grey Frog Hawk undergo a partial body molt during their first winter in southeastern Asia and Wallacea. They replace a varying amount of contour feathers and return to the breeding grounds in China in a transitional, post-juvenile plumage with a mixture of juvenile and adult-type feathers, most conspic- uous on the underside because of the differences in pattern: bold barring versus almost plain rufous- buff (Fig. 2) . In general, the post-juvenile contour molt appears to advance in parallel on both ventral and dorsal sides. The average individual difference between the extent of renewal of ventral and dor- sal feathers was insignificant (0.25 ± 3% (±SE, range = — 20%-20%; N = 12). We are not certain of timing of the complete molt in adults but it may terminate on the wintering grounds as evidenced by an undated adult from Jilolo Island (Moluccas) and two adults taken on Java that were growing outer primaries. However, an adult collected on 12 252 Herremans and Louette VoL. 34, No. 4 U Figure 2. Ventral aspect of the plumages of the Grey Frog Hawk {Accipiter soloensis). From left to right, juvenile (BMNH 73.5.12.1593), no date, Batchian; trzmsitional male (BMNH 1934.6.20.1), April, China (note underside predominantly boldly blotched and tail feathers juvenile); transitional male (BMNH 1903.7.3.94), May, China (note that fewer juvenile bars remain on lower underside and thigh feathers); adult female (BMNH 1905.12.24.955), May, China (note faintly barred on lower underside); adult male (BMNH 1914.5.1.69), March, China (plain rufous and white underside with marginal indication of barring). Photograph courtesy of the BMNH. October in Thailand had old outer primaries but the others were newly-grown feathers, suggesting that most of the molt occurs during summer and autumn on the breeding grounds (immatures do normally return in transitional plumage to the breeding grounds, and it is unlikely that this bird had remained in its winter quarters). Adults may start molting on the breeding grounds, suspend molt during migration, and complete it in winter. The fact that some specimens showed a contrast between worn inner and new outer series of pri- maries seemed to confirm the existence of molt suspension. In the southern African race of the Shikra (A. b. polyzonoides ) , the typical juvenile plumage (boldly marked below and brown with rusty edges above) was found unmolted from November-July (Table 1). Recently-fledged juveniles had been collected in November and January. From March onwards, some juveniles had molted body feathers. From May to October, body molt advancement showed great individual variation, but was never completed (Fig. 3). Because most birds examined had no growing feathers when collected, molt was appar- ently arrested before completion. Replacement of the juvenile plumage started on the ventral side with the upper throat. Molt on the upperside most- ly started in the neck and the upper parts of the mande, or on the head. In 44 of 47 transitional birds, replacement was more advanced on the ven- tral than dorsal side, while three birds had molted to a similar extent ventrally and dorsally. None of the birds that had replaced part of their juvenile plumage had already started to molt primaries or Table 1. Monthly distribution (number of birds in collections studied) of plumage types and adult primary molt in subspecies of the Shikra {Accipiter badius) . December 2000 Juvenile Molt in Small Accipiters 253 w Q 0 u w Q > 0 H U O Cm W ITl O M-l P :z p ca w pin (O CO CM 00 O O t?0 00 o O OO 00 o O 00 o O I-H LO O t-h 1T3 to O O 00 1 -H o O I— I 00 00 OO O O Of ^ CM CC to CM O Of cj! ^ O Of Of CM fO Of fiO O! O OO ’'Ch Of lOi O Of OO CM O O OO CO o 00 cjfOr^-^ oioootoo "OS * MM o s o N o O o o OO OO o O to Of 00 lo to o o o o C30 I-H to O Tft Of 00 o ^ 1-H O O o o o o tO 1-H Of ^H O O OO 1-H m OO o o o o o ^ O O O 1-H CM O O 1-H 00000f> OOOOtO OOOi-H 1-HOOOO CM CM Th o o ^ Of O to ^ O CM OO O 1-H O O OO ICO O O hh + O O OO (1J OO O O O O 1-H <0 to 1— H 1-H 1 -H o o o Of O 1 -H o O O 1 -H o o o o o O ^ 00 o O O Of o Of (Of lO o o ICO O OO OO o Of OO X o O lO o o Of + r3 00 CO O Of o o o o o 0 0^0 O OO o ito O OO o OO 1^ O m OO o O hh oO O OO o o ^ o „ o ^ S o o 2 C/!} f— < 2 ^ - H < < O' HH HH HH 2 2 2 *>* 2 w g d-j ' HM £ w s G G i4-> 'o 2 iG 0 0 3 s 2 G 0 0 3 • S'* OJ G 0 0 3 £ nS? V G 0 M 0 3 S ‘.B G B G G c5^ ;g G G • pN W G G 2 2 -i-t 3 -f-> 2 "ts 2 tU 'c/3 HH 2 -4-> 2 0 ■a- 1 *c/3 •4—1 2 •4-) 2 2 *:/3 HH 2 ■4-) 2 t3 < § T3 < -ci § 5 T3 < -C> § G3 < T3 < -cj 1 t; < < 2 2 2 Of OO to s ■ MM <3 -S3 -si O O Of o O HH O o HH ”0 ctS £S 0 . .2 o s QC w H < < G QJ M -o a rG V s 3- 3 o ;a *w G (0 Si o s o G n -g biD ‘-D .s oJ s G G u o 254 Herremans and Louette VoL. 34, No. 4 >» T3 O cn polyzonoides MAR APR MAY JUN JUL AUG SEP OCT 2 8 12 9 5 1 6 2 sphenurus 100 - 90 - 80 70 - ■ ventral □ dorsal SEP OCT NOV DEC JAN FEB MAR APR MAY 1 2 4 8 8 1 2 1 dussumieri 80 T 70 poliopsis 80 n 70 60 50 40 - OCT NOV DEC JAN FEB MAR APR MAY JUN JUL AUG SEP 1 3 6 14 4 3 3 1 1 1 1 Figure 3. Progress of the post^uvenile body-molt on the dorsal and ventral side in subspecies of the Shikra {Accipiter badius ) . Monthly values for medians of ventral shown in dark bars, dorsal molt stage in white bars, and ranges in thin lines. Numbers under the figure indicate monthly sample sizes. secondaries. One bird with well-advanced contour molt in June had new inner and outer tail feathers on both sides, but none growing. Most juveniles, therefore, acquire a transitional second plumage by a partial molt of the body feathers only, retain- ing most remiges and rectrices. The extent of this partial contour molt is highly variable individually (Fig. 3). A single second-year bird (BMNH 1910.7.1.108) was collected in November molting directly from a much-worn juvenile plumage to a fiilly adult plumage, apparently without having de- veloped transitionsil plumage. Despite wide individ- ual variation, the barring of the new feathers on the ventral side of the transitional post-juvenile plumage tended to be broader and more rusty in color than in adults (Table 2; Fig. 4) . Birds changed from transitional plumz^e into full, adult plumage through a complete molt when about 1 year old, almost synchronous with the molt of adults. Adults undergoing a complete molt from one definitive plumage to the next were collected from December (early stages) to May (latest stages) . Recently-fledged young of A. b. sphenurus were dated February-^eptember. Juveniles with transi- tional plumage appeared from September onwards (Table 1) . The contour molt never completed, with December 2000 Juvenile Molt in Small Accipiters 255 Table 2. Intensity of ventral barring of the transitional, post-juvenile plumage compared to adult plumage in sub- species of the Shikra {Accipiter hadius) . PJ> Max Ad^ Mean Ad 20% renewed feathers ventrally, and three had 70-90% renewed. The limited data suggested that, among the Asian races, A. b. badius may develop the most extensive transitional plumage, while molt extent in the migratory A. b. cenchroides ap- 256 Herremans and Louette VoL. 34, No. 4 Figure 4. Ventral aspect of the plumages of the Shikra {Accipiter badius polyzonoides ) . From left to right, juvenile male (BMNH 1911.12.23.430), January, Zambia; transitional female (BMNH 1950,50.125), April, Namibia (note limited replacement and new feathers more prominently barred than in adult); transitional female (BMNH 80.1.30.3), no date, Zambia (note only some barring from juvenile plumage remaining on flanks and thighs, juvenile tail, and replaced feathers on underside more prominently barred than in adult); adult female (BMNH 1932.5.10.598), March, Tanzania; adult male (BMNH 94.6.16.170), no date, Zambia. Photograph courtesy of the BMNH. pears more similar to A. b. dussumieri. There were too few specimens to compare the post-juvenile barring with that of adults. Adults had been col- lected from nests with eggs in April for A. b. cen- chroides, and molt in adults also followed breeding in this race (Table 1). From the timing of appear- ance of juveniles in the population and molt in adults (Table 1), it appeared that the same molt pattern also applied to the nominate race A. b. badr ius. Discussion In A. brevipes, A. soloensis, and A. badius, juvenile birds in nestling plumage have undersides with large rufous-brown spots and stripes, and broad barring on the flanks; the upperside has rusty brown edges to a generally dark brown plumage and all tail feathers are heavily banded. A transi- tional plumage occurs in the second half of the first year of life during which time remiges, most of the rectrices, and most of the larger upper wing coverts are retained from the juvenile plumage, but between 4-10 mo of age some of the body feathers are replaced. New feathers on the upper- side resemble the adult type, while those on the underside become barred with a tendency for wid- er, bolder barring than in adults. The extent of the partial contour molt is highly variable between in- dividuals and taxa, and some juveniles do not de- velop transitional plumage at all, particularly those of the race A. b. poliopsis of the Shikra. In it, the transitional plumage is replaced by adult plumage during a complete molt when one year old, more or less synchronous with the complete molt of December 2000 Juvenile Molt in Small Accipiters 257 Figure 5. Ventral aspect of the plumages of the Shikra {Accipiter badius sphenurus) . From left to right, juvenile male (RMCA 109477), July, Ethiopia; transitional male (RMCA 102.759), May, Uganda (note some spots from juvenile plumage left, tail juvenile and heavily banded, and more distinctly barred underside than adult); adult male (RMCA 95921), August, D.R. Congo (note less distinctly barred than female and unmarked outer tail feathers); transitional female (RMCA 102.736), June, Kenya (note some spots left from juvenile plumage, tail juvenile and heavily banded, and more prominently barred than adult); adult female (RMCA 103.515), November, D.R. Congo (note more dis- tinctly barred than adult male, but less than post-juvenile female, unmarked outer tail feathers). adults. Adults are uniformly bluish-grey dorsally (dark slate in A. b. soloensis) and have barring on the underside (almost plain in soloensis); females are generally more prominently barred than males. At least some tail feathers of adults have re- duced banding. Adult plumage is replaced by a complete molt following breeding. We lack information on whether birds may or may not breed when 1 year old and in transitional plumage. In migratory species such as the Levant Sparrowhawk and Grey Frog Hawk, transitional birds migrate to the breeding grounds. Brown et al. (1982) indicated that the Shikra might breed at one year of age, while Thiollay (1975) suggested that some birds do not breed at age one. Zimmer- man et al. (1996) felt that the race A. b. sphenurus might breed in transitional plumage but no cases of it were observed in A. b. polyzonoides by Tarboton ( 2000 ). Schmitt et al. (1982) found that half of the 20 juveniles caught in South Africa were undergoing solely a contour molt in April and May, most likely the partial contour molt of the post-juvenile plum- age we report here. The fact that a bird they clas- 258 Herremans and Louette VoL. 34, No. 4 Figure 6. Ventral aspect of the plumages of the Shikra (Accipiter badius dussumieri) . From left to right, juvenile female (BMNH 85.8.19.487), no date, India; transitional male (BMNH 85.8.19.481), October, India (note some spots from Juvenile plumage left, tail juvenile and heavily banded, and more distinctly barred underside than adult); adult male (BMNH 1949.WHI. 1.223), March, India (note underside less distinctly barred than female and unmarked outer tail feathers); transitional female (BMNH 85.8.19.508), March, Nepal (note many blotches left from the juvenile plumage, tail juvenile and heavily banded, and ventrally more prominently barred than adult); transitional female (BMNH 1949.25.87), November, Rawalpindi, Pakistan (note a few blotches left from the juvenile plumage and more promi- nently barred than adult); adult female (BMNH 86.3.25.79), December, India (note more distinctly barred than adult male but less than transitional female, and reduced markings on outer tail feathers). Photograph courtesy of the BMNH. sified as a juvenile in May was recaptured a year later in adult plumage should not be seen as proof for the absence of 2 ui intermediate plumage. In fact, their observations fit exactly the plumage se- quence we present here. A bird in recognizable juvenile plumage in May, whether or not the re- placement of body feathers has started, is expected either to be in the transitional plumage or to have molted recendy from the transitional to the adult plumage by next May. Schmitt et al. (1982) did not recognize the barred transitional plumage as dif- ferent from the adult and, therefore, identified the bird as an adult when recapturing it a year later. The individual they caught may have showed com- plete, adult plumage adding further evidence that the transitional plumage is retained for only about half a year and is replaced by the adult plumage in synchrony with the molt of breeding adults. We found several specimens for the different taxa molting from transitional plumage to adult plum- age with a complete molt at about the same time as adults undergo the complete post-breeding molt. This meant that the transitional plumage is only worn for about half a year. Some transitional birds may molt slightly ahead of adults with non- breeding birds molting earlier. December 2000 Juvenile Molt in Small Accipiters 259 If the transitional plumage is only retained for about half a year, are the feathers molted during the first year replaced again in the subsequent molt to adult plumage, or is all replacement part of the same molt cycle, with part of the contour molt shifted half a year forwards? Is the transitional plumage, therefore, a distinct plumage, or is it merely part of a protracted molt process? A pattern of a protracted molt with advanced contour molt would be similar to that found in several long-dis- tance migrants (e.g., waders in the genus Calidris, Cramp and Simmons 1983) and in some swallows {Hirundo spp., Delichon spp., Riparia spp.) and war- blers {Acrocephalus spp., Locustella spp., Hippolais spp., Sylvia spp., Jenni and Winkler 1994), which start a contour molt when still on the breeding grounds during the northern summer, but post- pone the replacement of most remiges and rectri- ces of the same molt cycle until arrival on the win- tering grounds some months later. In accipiters, if feathers of the transitional plumage are not re- placed after half a year, there should be birds in the population with new, adult-type rectrices and two generations of contour feathers with under- sides showing a mixture of worn, prominently barred feathers from the transitional plumage, and fresh, less barred, adult feathers. We have not found such birds in collections. Furthermore, be- cause markings on the undersides of transitional birds were on average more prominent, we con- cluded that the molt at age one included all feath- ers. We found no evidence in the Accipiterliteiature of a possible split molt, whereby replacement of body feathers significantly precedes that of flight feathers during the same molt cycle. In the Eur- asian Sparrowhawk, replacement of the primaries spans the entire molt period and no significant body molt occurs outside the period of primary replacement (Newton and Marquiss 1982). How- ever, it has a rather variable plumage with poorly- differentiated adult and juvenile patterns (Nilsson 1992, Engstrom and Edelstam 1995). Although it is the best studied Accipiter species, it may thus be an unfortunate choice as the standard for the ge- nus illustrating plumage sequences. The transitional plumage in the three accipiters we studied clearly showed a separate, intermediate feather generation between juvenile and adult plumages (Humphrey and Parkes 1959). It results from its own, albeit partial and individually-variable contour molt, giving a plumage type which is dif- ferent from that of adults. Recently, a second-year plumage was described from the South American Gray-bellied Hawk {A. poliogaster) , a species with a unique juvenile plum- age (Whittaker and Oren 1999). We found a sec- ond-year plumage with broader ventral barring than in adults in the Mayotte subspecies of the Frances’ Sparrowhawk {Accipiter francesiae brutus, Herremans et al. 2001). A transitional plumage, af- ter a partial contour molt, is also more frequent m the Black-shouldered Kite {Elanus caeruleus) than previously known (Herremans 2000). The occur- rence of early, partial postjuvenile molts in small raptors seems to be associated with a distinctive ju- venile plumage with the underside pattern differ- ing from that of adults. There are regions in the world where several accipiters, with essentially the same juvenile plumage co-occur (e.g.. King et al. 1978, Zimmerman et al. 1996), and molt into a more adult-like plumage may be essential for prop- er species recognition before they can enter the breeding population. However, interspecific pres- sures may not be the main force behind the early molt, because the phenomenon also exists where no confusing juveniles of other species co-occur (e.g., on Mayotte Island; Herremans et al. 2001). At this point we are uncertain how it functions but the underside plumage is probably important in social communication in accipiters. Under this functional hypothesis, it is important that the adult-like ventral plumage be acquired before the start of the next breeding season. In the well- marked race of the Shikra (A. b. poliopsis) , the sex- ual differentiation in ventral colors and patterning of adults is minimal and less than in the other rac- es and the transitional, post-juvenile plumage is poorly developed. Possibly, the development of the transitional plumage functions as an early indica- tion of the individual’s sex and, because of the poor sexual plumage dimorphism of adults, there is no functional need for young A. b. poliopsis to change plumage at an early stage. Individual vari- ation in the development of the post-juvenile molt might be dependent on condition and age, and the extent could also vary between years. Molt var- iation could, therefore, offer juveniles opportuni- ties to advertise their sex, age, and individual qual- ity. Age and quality have been demonstrated to have important impact on breeding performance in the Eurasian Sparrowhawk (Risch 1998), and it is likely that early advertisement of sex and quality for young birds entering their first breeding season is ultimately beneficial for their reproductive per- 260 Herremans and Louette VoL. 34, No. 4 formance. Under this hypothesis, aspects such as differences in territorial establishment, mating suc- cess, or recruitment into the breeding population at age one, may prove to be related to the extent of the post^juvenile molt. Kemp (1999) demonstrated an early partial con- tour molt in the Greater Kestrel {Falco rupicoloides) which coincided with changes in territorial behav- iors of adults. Small investments in a partial, post- juvenile molt, paralleled by changes in soft parts (e.g., eye color) resulted in important effects on communication and signaling. As in most accipi- ters, eye color changes with age in the Shikra from pale bluish-yellow in the fresh juvenile to yellow or orange in the transitional plumage and bright red in adults, with some variation according to sex and race. As in the Greater Kestrel, such changes in eye color could contribute to age and sexual commu- nication in the Shikra. Acknowledgments The staff of the Natural History Museum at Tring pro- vided all necessary facilities for the study of skins, and prepared the photographs of their specimens. The paper benefitted from the comments of the referees Alan Kemp and W.S. Clark. J.-M. Vandyck made the photographs of the RMCA specimens and prepared the prints. Literature Cited Ali, S. and S.D. Ripley. 1983. Handbook of the birds of India and Pakistan, Compact Ed. Oxford Univ. Press. Oxford, U.K. Altan, D.G. 1997. Little Banded Go.shawk. Pages 226— 227 mJ.A. Harrison, D.G. Allan, L.G. Underhill, M. Herremans, A.J. Tree, V. Parker, and C.J. Brown [Eds.], The atlas of southern African birds. Vol. I. BirdLife South Africa, Johannesburg, South Africa. Blanford, W.T. 1895. The fauna of British India, includ- ing Ceylon and Burma. Vol. Ill; birds. Taylor and Francis. London, U.K. Brown, L.H., E.K. Urban, and K. Newman. 1982. The birds of Africa. Vol. I. Academic Press. London, U.K. Clark, W.S. and R. Yosef. 1998. In-hand identification guide to Palearctic raptors. IBCE Tech. Publ. 7. Inter- national Birdwatching Centre, Eilat, Israel. Cramp, S. and K.E.L. Simmons. 1980. The birds of the Western Palearctic. Vol. II. Oxford Univ. Press. Ox- ford, U.K. and . 1983. The birds of the western Pa- learctic. Vol. III. Oxford Univ. Press. Oxford, U.K. del Hoyo, J., a. Elliott, and J. Sargatal [Eds.]. 1994. Handbook of the birds of the world. Vol. 2. Lynx Ed- icions, Barcelona, Spain. Elgood, J.H., C.H. Fry, and R.J. Dowsett. 1973. African migrants in Nigeria. Ibis 115:1-45. Engstrom, H. and C. Edelstam. 1995. Om utseendet hos sparvhokar — doda och levande. Anser 54:195-202. Fischer, W. 1980. Die Habichte. Neue Brehm-Bucherei, Wittenberg/Lutherstadt, Halle-Saale, Germany. Forsman, D. 1999. The raptors of Europe and the Middle East. T. and A.D. Poyser, London, U.K. Friedmann, H. 1930. Birds collected by the Childs Frick expedition to Ethiopia and Kenya colony. Part 1. Non- passerines. Smithsonian Institution, Washington, DC U.S.A. Hartley, R. 1976. Some notes on the plumages of Black Sparrowhawks. Bokmakierie 28:61-63. Henny, C.J., R.A. Olson, and T.L. Fleming. 1985. Breed- ing chronology, molt, and measurements of accipiter hawks in Northeastern Oregon. J. Field Ornithol. 56; 97-212. Herremans, M. 2000. Serial descendant primary molt (Staffelmauser) in the Black-shouldered Kite Elanus caeruleus. Ringing Migr. 20:15-18. , M. Louette, and J. Stevens. 2001. Biology of the Frances’s Sparrowhawk Accipiter francesiae on the Com- oro Islands. Ostrich 72:in press. Humphrey, P.S. and K.C. Parkes. 1959. An approach to the study of molt and plumages. Auk 76:1-31. Jenni, L. and R. Winkler. 1994. Molt and aging of Eu- ropean passerines. Academic Press, London, U.K. Kemp, A.C. 1999. Plumage development and visual com- munication in the Greater Kestrel Falco rupicoloides near Pretoria, South Africa. Ostrich 70:220-224. King, B., M. Woodcock, and E.C. Dickinson. 1978. A field guide to the birds of south-east Asia. Collins, London, U.K, Louette, M. 2000. Evolutionary exposition from plum- age pattern in African accipiter. Ostrich 71:45-50. Mueller, H.C., D.D. Berger, and G. At.i.f. z. 1979. Age and sex differentiation in size of Sharp-shinned Hawks. Bird-Banding 50:34—44. , , AND . 1981. Age, sex, and seasonal differences in size of Cooper’s Hawks./. Field Ornithol 52:112-126. Newton, I. and M. Marquiss. 1982. Moult in the Spar- rowhawk. Ardea 70:163-172. Nilsson, L. 1992. Hur ser vuxna sparvhokar egentligen ut? 31:279-281. Risch, M. 1998. Der Einfluss individueller Qualitat auf die Reproduktion des Sperbers Accipiter nisus. Disser- tation abstract./. Ornithol. 139:112-113. Schmitt, M.B., S. Baur, and F. von Maltitz. 1982. Men- sural data, moult, and abundance of the Little Band- ed Goshawk in the Transvaal. Ostrich 53:74-78. Smeenk, C. and N. Smeenk-Enserink. 1977. Observations on the Shikra Accipiter badius in Nigeria. Ardea 65:148- 164. Stresemann, E. and V. Stresemann. 1966. Die Mauser der Vogel./. Ornithol. 107 (Sonderheft):3-445. Stresemann, V. and E. Stresemann. 1960. Die Handsch- December 2000 Juvenile Molt in Small Accipiters 261 wingenmauser der Tagraubvogel.J. Ornithol. 101:373- 403. Tarboton, W. 2000. Little Banded Goshawk: family mat- ters. Africa Birds &' Birding 4:46—53. Thiollay, J.-M. 1975. Les rapaces d’une zone de contact savane-foret en Cote-d’Ivoire: densite, dynamique et structure du peuplement. Alauda 43:387-416. Verheyen, R. 1953. Exploration du Parc National de I’Upemba. Fascicule 19: Oiseaux. Institut des pares nationaux du Congo Beige, Bruxelles, Belgium. Whittaker, A. and D.C. Oren. 1999. Important ornitho- logical records from the Rio Jurua, western Amazonia, including twelve additions to the Brazilian avifauna Bull. Br Ornithol. Club 119:235-260. Zimmerman, D.A., D.A. Turner, and DJ. Pearson. 1996 Birds of Kenya and northern Tanzania. Christopher Helm, London, U.K. Received 1 April 1999; accepted 27 July 2000 J. Raptor Res. 34(4):262-269 © 2000 The Raptor Research Foundation, Inc. TURNOVER AND DISPERSAL OF PRAIRIE FALCONS IN SOUTHWESTERN IDAHO Robert N. Lehman^ and Karen Steenhoe USGS Forest and Rangeland Ecosystem Science Center, Snake River Field Station, 970 Lusk Street, Boise, ID 83706 U.S.A. Leslie B. Carpenter Boise State University, Raptor Research Center, Boise ID 83725 U.S.A. Michael N. Kochert uses Forest and Rangeland Ecosystem Science Center, Snake River Field Station, 970 Lusk Street, Boise, ID 83706 U.S.A. Abstract. — We studied Prairie Falcon (Falco mexicanus) breeding dispersal, natal dispersal, and turn- over at nesting areas in the Snake River Birds of Prey National Conservation Area (NCA) from 1971- 95. Of 61 nesting areas where falcons identified one year were known to be present or absent the following year, 57% had a different falcon. This turnover rate was 2-3 times higher than that reported elsewhere for large falcons, and may have been related to high nesting densities in the NCA. Turnover at nesting areas was independent of nesting success in the previous year, but was significantly higher for females nesting on large cliffs. Mean distance between natal and breeding locations for 26 falcons banded as nestlings and later encountered as nesting adults was 8.9 km. Natal dispersal distances were similar for males and females, but more than twice as many males marked as nestlings were later encountered nesting in the NCA. Fourteen adult falcons found on different nesting areas in successive years moved an average of 1.5 km between nesting areas; males dispersed significantly farther than females. Natal and breeding dispersal distances in the NCA were lower than those reported for Prairie Falcons in other study areas. Only four falcons banded as nestlings were found outside NCA bound- aries during the breeding period, and only one of these birds was known to be occupying a nesting area. We encountered no falcons banded outside the NCA occupying nesting areas in the NCA during this study. Key Words: Prairie Falcon-, Falco mexicanus; banding and marking, breeding dispersal, natal dispersal, nest- site fidelity, population turnover. Renovacion y dispersibn de Falco mexicanus en el suroeste de Idaho Resljmen. — Estudiamos la dispersion reproductiva, la dispersion natal y la renovacion en areas de ani- dacibn en el Area Nacional de Conservacibn del Snake River (ANC), desde 1971-95. De las 61 areas de anidacibn que fueron identificadas en un aho y en las cuales los halcones estuvieron presentes o ausentes en el aho siguiente, 57% tenian un halcon diferente. Esta tasa de renovacibn fue 2—3 veces mas alta que la reportada en otros sitios para halcones grandes, y pudo haber estado relacinada con las altas densidades en ANC. La renovacibn en las areas de anidacibn fue independiente del exito de anidacibn en el aho anterior, pero fue significativamente mayor para las hembras anidando en grandes riscos. La distancia media entre los sitios de reproduccibn y natalidad para 26 halcones anillados como pichones y posteriormente encontrados como adultos en anidacibn fue de 8.9 km. Las distancias de la dispersibn natal fueron similares para machos y hembras; pero fueron mas del doble en los machos marcados como pichones que fueron encontrados anidando en la ANC. Catorce halcones adultos en- contrados iCn distintas areas de anidacibn en ahos sucesivos se desplazaron un promedio de 1.5 km entre areas de anidacibn; los machos se dispersaron significativamente mas lejos que las hembras. Las distancias de la dispersibn natal y reproductiva en la ANC fueron mas bajas que las reportadas para estos halcones en otras areas de estudio. Solo cuatro halcones anillados como pichones fueron encon- 1 Present address: Dept, of Fishery and Wildlife Biology, Colorado State University, Fort Collins, CO 80523 U.S.A. 262 December 2000 Prairie Faix:on Turnover 263 trados por fuera de los limites de la ANC durante el periodo reproductivo y tan solo una de estas aves ocupo el area de anidacion. No encontranios ningun halcon anillado por fuera de la ANC ocupando algun area de anidacion durante este estudio. [Traduccion de Cesar Marquez] Many raptor nesting areas are occupied year af- ter year, but usually it is not known whether the same individuals occupy the same nesting areas in successive years (Newton 1979). Fidelity to nesting areas varies among species and populations. Small species with short life spans and those that occupy unpredictable environments are less likely to be sedentary than large, long-lived species occupying stable habitats (e.g., James et al. 1989, Jenkins and Jackman 1993, Rosenfield and Bielefeldt 1996). Within populations, older birds, successful breed- ers, and birds occupying high quality nesting areas often are more faithful to their nesting areas in successive years (Newton 1986, Village 1990, Fore- ro et al. 1999). In most bird species, females move farther from the natal site to breed, change nesting locations more often, and move greater distances between nesting areas used in different years than males (Greenwood 1980, Greenwood and Harvey 1982). Few data are available on Prairie Falcon {Falco mexicanus) turnover and dispersal. Steenhof et al. (1984) reported natal dispersal data for southwest Idaho for the 1970s and early 1980s, and Runde (1987) reported turnover and dispersal data for study areas in Wyoming, Colorado, and Alberta. Since the early 1980s, more than 1000 Prairie Fal- cons have been marked in the Snake River Birds of Prey National Conservation Area (NCA) in southwest Idaho, providing an opportunity for a more thorough analysis of turnover and dispersal than was possible in 1984. We were particularly in- terested in whether patterns of dispersal were re- lated to gender, environmental factors, or repro- ductive success, and whether patterns in the NCA were similar to those in other study areas. We hy- pothesized that female falcons would exhibit high- er turnover rates and disperse farther than males, and that turnover rates would be highest at nesting areas that failed to produce young in the previous year. We also expected turnover rates to be high at nesting areas with low long-term occupancy rates and at those on large cliffs, where potential nest sites were abundant. Study Area The NCA comprises 196 225 ha of canyonlands and shrubsteppe de.sert adjoining the Snake River in south- we.stern Idaho (USDI 1995). Our study focused on a 130- km stretch of the Snake River Canyon extending from Walters Ferry on the west to Hammett on the east. This area supports the densest known nesting concentration of Prairie Falcons (USDI 1979). During years when full counts of nesting pairs were obtained (1976-78, 1990- 94), numbers ranged from 160 to 206 (Steenhof et al 1999). In parts of the NCA where falcon densities are highest, canyon walls reach 125 m in height and often stretch uninterrupted for many kilometers. In these ar- eas, falcons typically nest <200 m and occasionally <50 m from other pairs (Steenhof 1998). In other parts of the NCA, nesting pairs may be up to 5 km apart. Prairie Falcons leave the NCA soon after young fledge and mi- grate to widely separated post-nesting and wintering ar- eas, primarily east and south of the NCA (Steenhof et al. 1984). Methods Marking. From 1970-94, 2060 Prairie Falcons were banded in the NCA primarily during two periods of intensive research (USDI 1979, 1996), and as part of annual monitoring efforts (Table 1). All falcons were banded with U.S. Fish and Wildlife Service (FWS) leg bands, and 1189 birds also received colored leg bands, patagial markers (Kochert et al. 1983), and/or radio- transmitters (USDI 1979, Vekasy et al. 1996, Marzluff et al. 1997), depending on research objectives (Table 1) Adults were trapped between March and June each year, and nestlings were banded in May and June, just before fledging. We applied several types of colored leg bands, particularly during the late 1980s and 1990s. Plastic bands were used in 1977 and 1986, and anodized alu- minum bands were used in 1987 and from 1990-94. From 1986-94, all color bands were inscribed with a unique alpha-numeric code. Data Collection. We defined a historical nesting area as any area of cliff where a Prairie Faleon pair was found in one or more years but where no more than one pair nested in the same year (Newton and Marquiss 1982). From 1973-95, we mapped 317 historical nesting areas based on records of 3170 nesting attempts. Delineated areas included scrapes, perches, and defended areas We defined an encounter as a determination of a bird’s FWS band number or alpha-numeric code by any means (Har- mata et al. 1999). During most years, data collection was incidental to other research and monitoring efforts (see Steenhof et al. 1999 for methods). However, the intensive field efforts conducted in the late 1970s and early 1990s provided more opportunities to encounter marked birds than in other years. In 1995, efforts to locate and identify marked birds were more systematic. That year, we tried to identify falcons at all nesting areas where falcons had been radiotagged in 1994, by trapping adults or reading band numbers from a distance. If 1994 occupants were not present in their nesting areas in 1995, we searched for them in the two nearest nesting areas (typically the 264 Lehman et al. VoL. 34, No. 4 Table 1. Number of Prairie Falcons marked in the NCA, by year and age class, 1970-94. Columns give the number banded with FWS leg bands. Additional markers (num- ber and type) are given in parentheses (P = patagial tag, R = radiotransmitter, C = colored leg band). Year Nestlings Banded Adults Banded Totals 1970 16 0 16 1971 no 0 110 1972 142 0 142 1973 0 0 0 1974 41 0 41 1975 79 (8P, 7R) 3 (3R) 82 1976 151 (107P, 9R) 8 (2P, 6R) 159 1977 118 (lOP, 32C) 11 (5C, 4R) 129 1978 104 (74P) 14 118 1979 75 (57P) 0 75 1980 79 0 79 1981 7 0 7 1982 4 0 4 1983 19 0 19 1984 68 0 68 1985 0 0 0 1986 25 (25C) 0 25 1987 104 (91C) 0 104 1988 0 1 1 1989 0 0 0 1990 80 20 (20C, 18R) 100 1991 154 (148C) 29 (28C, 28R) 183 1992 213 (195C, 79R) 34 (33C, 31R) 247 1993 118 (116C, 73R) 49 (42C, 38R) 167 1994 151 (147C) 33 (33C, 32R) 184 Total 1858 202 2060 adjacent upstream and downstream sites). During most years, we used binoculars and 15-60X spotting scopes to observe falcons. From 1991-95, we also used a 160X Questar telescope to read alpha-numeric codes on col- ored leg bands. Turnover. We estimated turnover as the proportion of nesting areas occupied in successive years where marked adults identified the first year were not present the fol- lowing year. We considered that turnover occurred if a different individual was trapped during the second year; if a bird’s color band inscription did not match that of the previous occupant; if a bird’s band color or place- ment differed from that of the previous occupant; if the new occupant was unmarked; or if the previous occupant was encountered in another nesting area or was found dead. We used nesting areas more than once in the anal- ysis if the same occupant was identified in more than two consecutive years or if both occupants were marked in the same year. Nesting areas that were vacant the year after birds were marked were not used in the analysis. To assess factors that might influence turnover, we clas- sified nesting areas according to nesting success in the year before turnover was assessed, long-term occupancy. and cliff height. We considered a nesting area successful if >1 young reached 30 d of age (Steenhof 1987). We based occupancy rates on the proportion of years that pairs were present at nesting areas in our sample during 8 yr when full surveys of the NCA were conducted (1976- 78, 1990-94) (Steenhof et al. 1999). We classified nesting areas that had pairs present ^65% of years (N = 10) as low occupancy sites. Those with pairs present >65% of years {N = 36) were classified as high occupancy sites. We computed cliff-height categories at nesting scrapes from studies that interpreted aerial photographs using standard parallax methods (Bentley and Hardyman un- publ. data) . We considered cliffs ^30.6 m to be small and those >30.6 m to be large. These categories reflected the fact that most cliffs in the NCA are under 30.6 m in height, but higher cliffs often reach 125 m. Dispersal. We calculated natal and breeding dispersal distances from Universal Transverse Mercator coordi- nates assigned to banding and encounter locations. To assess natal dispersal, we recorded all cases where falcons marked as nestlings were encountered later as breeding adults. To assess breeding dispersal, we recorded all cases where breeding adults were encountered in different nesting areas in subsequent years, including birds band- ed as nestlings if they moved to different nesting areas after we found them breeding. We measured distances between nesting scrapes if known, or between centers of nesting areas if locations of nesting scrapes were un- known. We also counted the number of historical nesting areas between nesting areas used by the same falcon. Data Analysis. We ran all statistical tests using SAS soft- ware (SAS Institute Inc. 1990). Because our investigation was exploratory in nature and our sample sizes were small, we opted to increase power and reduce the risk of Type 11 errors by considering P-values ^0.10 as signifi- cant. We used contingency table analysis (G-tests) to as- sess gender differences in turnover and in the tendency of falcons to breed near their natal areas. We also used G-tests to relate turnover to nesting success in the pre- vious year, cliff height, and long-term occupancy. Because dispersal data were not normally distributed, we used the Wilcoxon rank sums test, a nonparametric alternative to the ^-test, to assess differences in natal and breeding dis- persal distances. We identified gender of nestling and adult falcons using foot pad length at the time they were banded (Marzluff et al. 1991) and copulatory behavior observed after release. Birds with foot pads <86 mm were considered to be males; those with foot pads >86 mm were considered to be females. Results Encounters with Marked Birds. We recorded 76 encounters with 63 marked falcons (34 males and 29 females) at 46 nesting areas during breeding seasons from 1976-95. Sixty-five encounters oc- curred between 1990—95. Twenty-six (41%) of 63 individuals encountered were marked as nestlings, and 37 (59%) were marked as breeding adults. We recorded more than one encounter with 12 birds: 11 falcons were recorded at nesting areas in two December 2000 Prairie Fai.con Turnover 265 Table 2. Status and turnover at Prairie Falcon nesting areas in the NCA one year after they were occupied by marked adults, 1975-94. Number of Nesting Areas Same Bird Different Bird Unknown Occupant Vacant Turnover Males 11 11 51 24 50% Females 15 24 55 28 61% Both sexes 26 35 106" 52 57% ^ Includes 4 nesting areas for which occupancy was not confirmed the following year. different years after they were banded, and one bird was seen in three different years. Turnover. We evaluated 219 nesting areas where marked adults were known to be present in at least one nesting season from 1970 to 1994. Of these, 102 were occupied the following year by falcons we did not check for identity, 52 were vacant, and oc- cupancy was unconfirmed for 4 (Table 2) . This left 61 cases for turnover assessments. Of these, 26 had the same bird and 35 had a different bird the fol- lowing year. Thus, turnover was 57% for both sexes combined. Turnover was similar for males (50%) and females (61%) (G^ = 0.76, P = 0.38). Of 35 nesting areas where turnover of marked birds occurred, 18 were occupied by new individ- uals with known band numbers, and 17 were oc- cupied by unidentified birds. In the latter 17 cases, we knew turnover occurred because eight previous occupants were on different nesting areas, six new occupants were unbanded, two new occupants wore different colored bands than those of the pre- vious occupant, and one previous occupant was found dead soon after its first encounter. We were able to account for only one missing bird from the 18 nesting areas where new birds were identified. This bird was found dead near its former nesting area early in the second breeding season. Sample sizes were inadequate to assess annual turnover for most years. However, in 1995 we con- firmed if marked birds had returned to their for- mer nesting areas in 19 cases. Of these, seven nest- ing areas (37%) were occupied by the same individual, and 12 nesting areas (63%) were oc- cupied by a different bird. This turnover rate (63%) did not differ from that of all other years (55%) (Gi = 0.38, P= 0.54). Turnover of marked falcons was independent of nesting success in the previous year and long-term occupancy of the nesting area. New birds appeared at 4 of 10 nesting areas with low long-term occu- pancy rates (40%), compared to 31 of 51 nesting areas with high occupancy rates (61%) (Gj = 1.46, P = 0.30). Turnover occurred at 19 of 32 successful nesting areas (59%), compared to 10 of 18 unsuc- cessful nesting areas (56%) (G^ = 0.07, P = 0.79). When we considered relationships between turn- over and nesting success in the previous year and turnover and long-term occupancy by sex, we found no differences for males or females {Gs ^ 2.50, Ps > 0.11). However, cliff height was related to female but not male turnover. Only 3 of 14 fe- males returned to nesting areas on large cliffs, compared to 12 of 25 females on small cliffs (Gj - 2.80, P - 0.09). Site Fidelity. At least 26 falcons returned to their former nesting areas the year after they were marked or last seen (Table 2). Of these, two fe- males returned to the same nesting area for a third consecutive year. Both birds occupied their nesting areas from 1977—79. The 26 falcons also included a male that returned to the same nesting area for two consecutive years and one nonconsecutive year. This bird, banded as a nestling in 1990, was found at a nesting area >14.2 km downstream from its natal area in 1991, 1992, and 1994. It likely occupied this same nesting area in 1993, but iden- tification was inconclusive. In 1995, it was replaced by a marked, 1 -yr-old male. Five other individuals (3 males and 2 females), not included in the sam- ple of 26 falcons described above, were found in the same nesting areas in two nonconsecutive years. Natal Dispersal. Of 1858 Prairie Falcons banded as nestlings (Table 1), 26 (1.4%) were encountered during subsequent breeding seasons. These 26 fal- cons were 1-5-yr old when encountered on nesting areas (Table 3) . More than twice as many males {N = 18) were encountered as females {N = 8) (G^ = 2.01, P = 0.08). Distances between natal areas and breeding sites 266 Lehman et al. VoL. 34, No. 4 Table 3. Natal dispersal distances (km) of Prairie Falcons marked as nestlings between 1972-94 that returned to occupy nesting areas in the NCA as breeders. Males® Females'’ Both Sexes Age in Years No. Birds Mean ± SD No. Birds Mean ± SD No. Birds Mean ± SD Range 1 3 12.3 ± 8.1 0 3 12.3 ± 8.1 3.4-19.3 2 6 5.6 ± 4.4 1 5.8 7 5.6 ± 4.0 1.1-13.0 3 4 19.0 ± 14.9 2 8.2 ± 2.1 6 15.4 ± 12.8 6.3-35.6 4 4 6.7 ± 5.0 5 5.6 ± 2.9 9 6.1 ± 3.7 1.7-14.1 5 1 8.3 0 1 8.3 Total 18 10.1 ± 9.3 8 6.2 ± 2.6 26 8.9 ± 8.0 1.1-35.6 “Range (all males): 1.1-35.6. ‘’Range (all females): 1. 7-9.8. were similar for males (x — 10.1 km; median = 6.35 km; range = 1.1—35.6 km) and females (x — 6.2 km; median = 5.9 km; range = 1.7-9. 8 km) (S = 98; F = 0.59) (Table 3) . Numbers of historical nesting areas between natal and breeding sites for males (x = 24.4; range — 2-95) and females (x = 19.6; range = 5-44) also were similar (S = 101; F — 0.72). In three cases where we had dispersal data on closely related individuals, distances and direc- tion moved were similar (Table 4) . Four falcons banded as nestlings in the NCA were encountered as yearlings outside the NCA during the breeding season (Steenhof et al. 1984). These birds were encountered in northern Idaho, western Montana (>300 km from the natal areas in both cases), eastern Oregon (101 km from the natal area), and southern Idaho (41-116 km from the natal area). Only the bird in southern Idaho was known to be occupying a nesting area at the time of the encounter; however, it was not identi- fied to individual, so we could not determine the exact distance it dispersed. Breeding Dispersal. We recorded 20 encounters with marked adults in different nesting areas one or more years after their last known breeding lo- cation. None of these encounters occurred outside the NCA. Falcons found one year later (N = 14) moved an average of 1.5 km (Table 5). Males dis- persed significantly farther (x - 3.3 km; median — 2.75; range = 1.5— 6.2 km) than females (x = 0.7 km; median = 0.5; range = 0.1-1. 9 km) (S — 49, F = 0.009). When we included six falcons (a fe- male and 5 males) encountered two to three years after their last known nesting location, mean dis- tance between nesting areas increased to 2.0 km (Table 5). With movements in nonconsecutive years included, dispersal distances of males were still greater than females (S = 117, F = 0.09). Of 20 individuals encountered in new nesting areas one or more years after they were last iden- tified, 10 (2 males and 8 females) moved to adja- cent nesting areas. The remaining 10 individuals (5 males and 5 females) crossed 1-28 nesting areas (x — 7.5) during dispersal movements. Males (x — Table 4. Natal dispersal of related individuals in the NCA. All birds shown were banded as nestlings. Relation- ship Year Banded Year Encountered Direction Moved Distance Moved (km) Brother 1987 1990 Southeast 27.5 Sister 1987 1991 Southwest 1.7 Brother 1990 1991 Northwest 14.2 Brother 1990 1992 Northwest 13.0 Father 1987 1990 Southeast 6.5 Son 1990 1992 Southeast 4.8 Son 1990 1993 Southeast 6.3 December 2000 Prairie Faccon Turnover 267 Table 5. Breeding dispersal distances (km) for Prairie Falcons that moved from a previous year’s nesting area. Sample size is in parentheses. Sex Mean Dispersal Distance ± SD Dispersal Range Consecutive Male 3.3 ± 2.0 (4) 1. 5-6.2 years Female 0.7 ± 0.6 (10) 0.1-1.9 Both 1.5 ± 1.6 (14) 0.1-6.2 All encounters Male 3.4 ± 4.5 (9) 0.2-14.4 Female 0.8 ± 0.7 (11) 0.1-1.9 Both 2.0 ± 3.2 (20) 0.1-14.4 11.0; range = 1-28) and females (x = 4.4; range = 1-7) crossed similar numbers of nesting areas during these movements (5 = 31.5, P = 0.46). Of 26 falcons that returned to the same nesting areas in consecutive years, five used the same nest- ing scrape as the previous year. Of 21 falcons found in the same nesting areas hut in different nesting scrapes, 17 moved <0.5 km. The other four falcons moved 0.5-0. 8 km. When these falcons were in- cluded in estimates of breeding dispersal, mean dispersal distance for encounters in consecutive years dropped to 0.8 km: 1.2 km for males {N = 13), and 0.5 km for females {N = 22). Discussion Turnover. Turnover for Prairie Falcons in the NCA was 2-3 times higher than rates reported in other studies of large falcons (Mearns and Newton 1984, Runde 1987, Court et al. 1989). High turn- over rates in the NCA may be due partly to high falcon densities. Mean density of nesting pairs throughout the NCA is 0.7 pairs per linear km, and densities reach 4.3 pairs per km in areas where cliffs exceed 100 m in height (Steenhof 1998). Most nesting areas where we studied turnover were in the deepest parts of the canyon; thus, densities in our study area were more than 4 times those reported elsewhere. The highest mean densities re- corded for Prairie Falcons in other study areas were 0.6 pairs per km in southwestern Wyoming (Runde 1987), and 0.3 pairs per km on the Kevin Rim, Montana (Harmata et al. 1991). Density may affect turnover by influencing the availability of po- tential mates. Where densities are low, returning to a previous breeding area may be the most effec- tive means of finding a mate. Where densities are high, abundance of potential mates may reduce a falcon’s need to return to a specific former nesting area. High turnover in the NCA also may be related to the abundance of nest sites in the area, and the fact that nesting cliffs tend to be continuous. Each year, many historical nesting areas in the NCA are unoccupied. No more than 206 Prairie Falcon pairs have been recorded nesting in the NCA in any given year (Steenhof et al. 1999), yet 317 his- torical nesting areas have heen identified. Where nest sites are abundant and distances between al- ternate sites are low, frequent moves to different nesting locations might be expected. Other factors may affect turnover in large fal- cons, including gender, previous nesting success, long-term occupancy, and nest site or mate quality. The fact that females tended to move more often than males is consistent with studies that suggest turnover in large falcons is slightly higher for fe- males (Runde 1987, Enderson and Craig 1988, Court et al. 1989). In this respect, our data are consistent with overall trends in birds (Greenwood 1980, Greenwood and Harvey 1982). However, our data are inconsistent with other studies in two re- spects. In Eurasian Sparrowhawks (Accipiter nisus) and Eurasian Kestrels {Falco tinnunculus ) , turnover was lower for birds that nested successfully the pre- vious year, and for those from nesting areas with high long-term occupancy rates (Newton 1986, Vil- lage 1990). In the NCA, birds from previously suc- cessful nesting areas and nesting areas with high long-term occupancy rates showed no greater ten- dency to return than birds from unsuccessful nest- ing areas and nesting areas with low long-term oc- cupancy; but females nesting on large cliffs were more likely to move than those on small cliffs. Our results again may be related to the abundance of ledges, cavities, and potential mates in the NCA. Ealcons from some previously successful nesting ar- eas may have been displaced by other individuals seeking higher quality sites or better mates, or some falcons may have moved to better nesting ar- eas. The abundance of potential partners and plac- es to nest made it likely that individuals displaced from former nesting areas would find other breed- ing opportunities. Dispersal. Both natal and breeding dispersal dis- tances in the NCA were shorter than previously re- ported for Prairie Falcons (Runde 1987). As in most studies of dispersal, the natal and breeding dispersal distances we recorded may be biased downward because we did not search for marked 268 Lehman et al. VoL. 34, No. 4 birds outside the NCA. Furthermore, our searches for marked birds focused on previously used and adjacent nesting areas. Thus, we were more likely to find birds within a few kilometers of their natal or previous breeding sites. This bias may partly ex- plain why both our natal and breeding dispersal distances were shorter than for Alberta, Colorado, and Wyoming (Runde 1987). As with turnover, population density and habitat features also could explain shorter dispersal dis- tances in the NCA. Falcons do not need to travel far to find nesting sites or mates in the NCA be- cause stretches of cliff tend to be continuous and potential mates are abundant. In other parts of western North America, nesting cliffs are widely scattered and nesting pairs are clumped, typically with <10 pairs on individual buttes or escarpments (Runde pers. comm., Harmata et al. 1991). Birds that do not return to their natal or former nesting areas have fewer alternative sites available for nest- ing and fewer mate choices. Those that leave the local area to find breeding opportunities must trav- el greater distances. In contrast to Runde’s (1987) findings and the predictions of Greenwood (1980) and Greenwood and Harvey (1982), natal dispersal distances for fe- males in the NCA were not longer than males. However, if females dispersed outside the NCA, we would not have found them and recorded the dis- tances. Although our encounter rate likely under- estimates the total number of returning birds, it is relatively unbiased with respect to gender because we banded similar numbers of males (923) and fe- males (935) as nestlings and we trapped similar numbers of adult males (90) and females (100) as breeders. Thus, our findings suggest that males were more likely than females to return to breed in the NCA and are consistent with predictions of female-biased dispersal in raptors. The fact that we encountered few birds marked as nestlings in later years is partly an artifact of our failure to check for birds at all nesting areas, and should not be construed as evidence for frequent emigration. Low encounter rates also may be re- lated to high post-fledging mortality in the NCA (McFadzen and Marzluff 1996). Short natal and breeding dispersal distances, and the fact that only one falcon banded inside the NCA is known to have nested outside NCA boundaries, may indicate that very little dispersal from or into the NCA oc- curs. However, because many falcons inside the NCA in any given year were unmarked, and be- cause we conducted no searches for marked fal- cons outside the NCA, conclusions regarding the extent of falcon immigration into and dispersal from the NCA must remain tentative. Acknowledgments This paper is a contribution of the U.S. Geological Sur- vey, Forest and Rangeland Ecosystem Science Center, Snake River Field Station (formerly the Raptor Research and Technical Assistance Center). The U.S. Bureau of Land Management (BLM) provided funds from 1972-94 as part of its Snake River Birds of Prey Research Project and the cooperative BLM/Idaho Army National Guard (IDARNG) Project. IDARNG funding was provided pri- marily through W.S. Seegar, U.S. Army Chemical Re- search and Engineering Center. The National Biological Service (NBS) and Idaho Power Company provided ad- ditional funding and support. A.R. Bammann, J.H. Doremus, T.C. Dunstan, D.L. Evans, J.S. Marks, M.Q. Moritsch, S. Sawby, and G. Sitter played a key role in collecting data in the early years of the study. We thank the Idaho Cooperative Research Unit, M.G. Hornocker, and V.T. Ogden for the 1970-72 banding data, and A.M.A. Holthuijzen for banding falcons in 1987. This pa- per would not have been possible without the assistance of more than 30 individuals who encountered previously marked falcons. We thank Greenfalk Consultants person- nel, especially J.C. Bednarz, J.M. Marzluff, J.O. McKinley, L.S. Schueck, R. Townsend, and M.S. Vekasy, who were responsible for trapping and radiotagging Prairie Falcons and for providing observations of marked birds from 1990-94. We also thank all the BLM, NBS, and Boise State University field technicians who reported observa- tions and assisted in banding young. Special thanks go to J.O. McKinley and R.R. Olendorff, Jr. for 1995 data col- lection. J.C. Bednarz, J.M. Marzluff, D.E. Runde, J.H. En- derson, and two unidentified reviewers made valuable comments on earlier drafts. Finally, we thank Deana Par- rish for data automation. Literature Cited Court, G.S., D.M. Bradley, C.C. Gates, and D.A. Boag. 1989. Turnover and recruitment in a tundra popula- tion of Peregrine Falcons Falco peregrinus. Ibis 131:487- 496. Ender.son, J.H. AND G.R. Craig. 1988. Population turn- over in Colorado peregrines. Pages 685-688 in T.J. Cade, J.H. Enderson, C.G. Thelander, and C.M. White [Eds.], Peregrine Falcon populations: their manage- ment and recovery. The Peregrine Fund, Boise, ID U.S. A. Forero, M.G., J.A. Donazar, J. Bias, and F. Hiraldo. 1999. Causes and consequences of territory change and breeding dispersal distance in the Black Kite. Ecol- ogy 80:1298-1310. Greenwood, PJ- 1980. Mating systems, philopatry, and dispersal in birds and mammals. Anim. Behav. 28: 1140-1162. AND P.H. Harvey. 1982. The natal and breeding dispersal of birds. Ann. Rev. Ecol. Syst. 13:1-21. December 2000 Prairie Falcon Turnover 269 Harmata, a., M. Restani, B. Hatcmer, A. House, and S. Grantham. 1991. Impacts of oil and gas develop- ments on raptors associated with Kevin Rim, Mon- tana. Unpubl. rep. U.S. Dep. Inter., Bur. Land Man- age., Great Falls, MT U.S.A. , G.J. Montopoli, B. Oakleaf, RJ. Harmata, and M. Restani. 1999. Movements and survival of Bald Ea- gles banded in the Greater Yellowstone Ecosystem. J. Wildl. Manage. 63:781-793. James, P.C., I.G. Warkentin, and L.W. Oliphant. 1989. Turnover and dispersal in urban Merlins Falco coluni- barius. Ibis 131:426-447. Jenkins, J.M. and R.E. Jackman. 1993. Mate and nest site fidelity in a resident population of Bald Eagles. Condor 95:1053-1056. Kochert, M.N., K. Steenhof, and M.Q. Moritsch. 1983. Evaluation of patagial tag markers for raptors and ra- vens. Wildl. Soc. Bull. 11:271-281. Marzluff, J.M., C. Goody, T.L. Maechtle, M. McFadzen, T.C. Miles, L.S. Schueck, and M. Vekasy. 1991. Influ- ence of military training on the behavior of raptors in the Snake River Birds of Prey Area, 1991. Pages 37-95 in K. Steenhof [Ed.], Snake River Birds of Prey Area 1991 Annual Report. U.S. Dep. Inter., Bureau of Land Manage., Boise, ID U.S.A. , B.A. Kimsey, L.S. Schueck, M.E. McFadzen, M.S. Vekasy, and J.C. Bednarz. 1997. The influence of hab- itat, prey abundance, sex, and breeding success on the ranging behavior of Prairie Falcons. Condor 99: 567-584. McFadzen, M.E. and J.M. Marzluff. 1996. Mortality of Prairie Ealcons during the fledging-dependence pe- riod. Condor 98:791-800. Mearns, R. and I. Newton. 1984. Turnover and dispersal in a Peregrine falco peregrinus population. Ibis 126: 347-355. Newton, I. 1979. Population ecology of raptors. T. & A.D. Poyser, Ltd., Cal ton, U.K. . 1986. The sparrowhawk. T. & A.D. Poyser, Ltd., Gallon, U.K. AND M. Marquiss. 1982. Fidelity to breeding area and mate in sparrowhawks Accipiter nisus.J. Anim. Ecol. 51:327-341. Rosenfield, R.N. AND J. Bielefeldt. 1996. Lifetime nest- ing area fidelity in male Cooper’s Hawks in Wisconsin Condor 98:165-167. Runde, D.E. 1987. Population dynamics, habitat use and movement patterns of the Prairie Falcon {Falco mexi- canus). Ph.D. dissertation, Univ. Wyoming, Laramie, WY U.S.A. SAS Statistical Institute Inc. 1990. SAS/STAT user’s guide. Vol. 2. Version 6, 4th Ed. SAS Statistical Insti- tute, Cary, NC U.S.A. Steenhof, K. 1987. Assessing raptor reproductive success and productivity. Pages 157-170 in B.A. Giron Pen- dleton, B.A. Millsap, K.W. Cline, and D.M. Bird [Eds.], Raptor management techniques manual. Natl. Wild. Fed., Washington, DC USA. . 1998. Prairie Falcon {Falco mexicanus) . In A. Poole and F. Gill [Eds.], The birds of North America, No 346. Academy of Natural Sciences, Philadelphia, PA and the American Ornithologists’ Union, Washing- ton, DC U.S.A. , M.N. Kochert, and M.Q. Moritsch. 1984 Dis- persal and migration of southwestern Idaho raptors. J. Field Ornithol. 55:357-368. , M.N. Kochert, L.B. Carpenter, and R.N. Leh- man. 1999. Long-term Prairie Falcon population changes in relation to prey abundance, weather, land uses, and habitat conditions. Condor 101:28-41. U.S. Department of the Interior. 1979. Snake River Birds of Prey special research report to the Secretary of the Interior. Bur. Land Manage., Boise, ID U.S A . 1995. Snake River Birds of Prey National Con- servation Area management plan. Bur. Land Manage , Boise, ID U.S.A. . 1996. Effects of military training and fire in the Snake River Birds of Prey National Conservation Area BLM/IDARNG Research Project Final Report. U S Geol. Surv., Biol. Res. Div., Snake River Field Station, Boise, ID 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. J. Field Ornithol. 67:680- 690. Village, A. 1990. The kestrel. T. & A.D. Poyser, Ltd., Lon- don, U.K. Received 3 March 2000; accepted 28 July 2000 /. Raptor Res. 34(4):270-278 © 2000 The Raptor Research Foundation, Inc. ROOST SITES OF RADIO-MARKED MEXICAN SPOTTED OWLS IN ARIZONA AND NEW MEXICO: SOURCES OF VARIABILITY AND DESCRIPTIVE CHARACTERISTICS Joseph L. Ganey and William M. Block USDA Forest Service, Rocky Mountain Research Station, 2500 S. Pine Knoll, Flagstaff, AZ 86001 U.S.A. RudyM. King USDA Forest Service, Rocky Mountain Research Station, Ft. Collins, CO 80526 U.S.A. Abstract. — ^To increase understanding of roosting habitat of Mexican Spotted Owls (Strix occidentalis lucida) and factors that influence use of roosting habitat, we sampled habitat characteristics at 1790 sites used for roosting by 28 radio-marked Mexican Spotted Owls in three study areas in Arizona and New Mexico. We explored potential patterns of variation in roost-site characteristics by estimating similarity among all possible pairs of roost sites and summarizing patterns in these similarity estimates using a linear model. Factors in the model included owl identity and season. We conducted these analyses within study areas, because habitat characteristics differed among study areas. We used a repeated-measures model which assumed that similarity estimates computed between roost sites of the same owl or pairs of owls were correlated. This model significantly improved model goodness-of-ht over a null model assuming no such correlation structure. Similarity estimates were relatively high (0.744-0.775) in all three study areas, suggesting consistent patterns of selection among owls within areas. Owl and season effects were relatively small but detectable in all study areas, with the relative magnitude of these effects differing among areas. The seasonal effect was greatest in the area dominated by pine-oak forest and relatively slight in two areas where owls roosted primarily in mixed-conifer forest. Relative to areas where owls roosted in mixed-conifer forest, roosts in pine-oak forest occurred on moderate slopes, on southwest to northwest aspects, and were less concentrated on lower portions of slopes. We suspected that much of this difference reflected differences in stand-development processes in different forest types. This suggested that land managers should incorporate knowledge of such patterns in different forest types and topographic locations in planning decisions involving management of Spotted Owl habitat. Key Words; Mexican Spotted Owl, Strix occidentalis lucida; Arizona; New Mexico; radiotelemetry; repeated measures; roost sites; sources of variation. Sitios de perchas de Strix occidentalis lucida marcados con radio-transmisores en Arizona y Nuevo Mexico: fuentes de variabilidad y caracteristicas descriptivas Resumen. — Para aumentar el conocimiento de los habitats de perchas de Strix occidentalis lucida y los factores que influyen en su uso, muestreamos las caracteristicas del habitat en 1790 sitios utilizados como perchas por 28 buhos con radio transmisores en tres areas de estudio en Arizona y Nuevo Mexico. Exploramos los patrones de variacion dentro de las caracteristicas mediante la estimacion de la similar- idad entre todos los posibles pares de sitios de perchas y resumimos los patrones dentro de estos estimativos de similaridad utilizando un modelo linear. Los factores en el modelo incluyeron la identidad de los buhos y la epoca. Condujimos estos analisis dentro de las areas de estudio debido a que las caracteristicas de habitat dihrieron entre las areas de estudio. Utilizamos un modelo de repeticion de medidas el cual asumio que las estimaciones de similaridad estimadas computados entre los sitios de perchas de las mismas parejas de buhos estaban correlacionadas. Este modelo mejoro significativamente al modelo de bondad de ajuste sobre el modelo nulo, asumiendo la correlacion de estructura. Los estimativos de similaridad fueron relativamente altos (0.744-0.775) en las tres areas de estudio, sugi- riendo patrones consistentes de seleccion entre buhos y dentro de las areas. Los efectos de buhos y estacion fueron relativamente pequenos pero detectables en todas las areas de estudio, con una mag- nitud relativa de estos efectos diferidos entre areas. El efecto de la estacionalidad fue mayor en el area dominada por los bosques de roble y relativamente pequeho en las dos areas en donde los buhos se percharon principalmente en bosques de coniferas mixtas. Con relacion a las areas en donde los buhos 270 December 2000 Roost Sites of Mexican Spotted Owls 271 se percho en bosques mixtos de coniferas, las perchas en los bosques de robles y pinos ocurrieron en vertientes moderadas en el suroeste y noroeste, estas fueron menos concentradas en la porcion baja de las vertientes. Sospechamos que buena parte de esta diferencia es producto del proceso de desarrollo de arboles en distintos tipos de bosques. Esto sugiere que los planificadores deben incdrporar el con- cimiento de estos patrones en diferentes tipos de bosques y situaciones topograficas en las desiciones de planificacion que involucran el manejo de habitat de los buhos. [Traduccion de Cesar Marquez] The Mexican Spotted Owl {Strix occidentalis luci- da) occurs throughout the southwestern United States and northern Mexico in forested mountains and canyonlands (Gutierrez et al. 1995, Ward et al. 1995). It is frequently associated with late-succes- sional forests (Ganey and Dick 1995, Gutierrez et al. 1995) and was listed as threatened in 1993 be- cause of concerns over loss of forested habitat to timber harvest (Cully and Austin 1993). Previous studies (reviewed in Ganey and Dick 1995) suggest that Mexican Spotted Owls are highly selective in terms of roosting and nesting habitat but forage in a wider array of habitats. Consequently, a recovery plan prepared for the Mexican Spotted Owl (Block et al. 1995) explicitly assumed that availability of roosting/nesting habitat was a key factor limiting the distribution of this owl. Thus, understanding factors underlying use of roosting habitat by Mex- ican Spotted Owls may be critical to managing hab- itat for this owl. Several studies have examined roosting habitat used by Mexican Spotted Owls. Rinkevich and Gu- tierrez (1996) and Willey (1998) described roost- ing habitat in the canyon country of southern Utah. Owls in this region were not closely associ- ated with forests and typically roosted on cliffs near the bottoms of narrow rocky canyons with complex architecture. Johnson (1997) also observed owls as- sociated with steep canyons and roosting on cliffs in Colorado, but most of the roosts he located were in trees. Farther south, owls in Arizona and New Mexico were more closely associated with forests and typically roosted in trees (Ganey and Baida 1989, 1994, Fletcher and Hollis 1994, Zwank et al. 1994, Seamans and Gutierrez 1995, Hodgson 1996, Stacey and Hodgson 1999). Roost trees were typi- cally located in well-shaded areas, often low on can- yon slopes or in canyon bottoms, in relatively cool areas. Similar results have been reported for both Northern (S. o. caurina, Thomas et al. 1990) and California (S. o. occidentalis, Gutierrez et al. 1992) Spotted Owls. This may be at least partially due to an aversion to high daytime temperatures during the breeding season (Forsman 1976, Barrows 1981, Ganey et al. 1993, but see Verner et al. 1992). Several factors limit our understanding of forest roosting habitat of Mexican Spotted Owls. With the exception of Zwank et al. (1994) , most information is from the breeding season and does not address potential variation in habitat use between seasons. Most studies have either presented little quantita- tive information on roost sites (Ganey and Baida 1989) , were based on small numbers of owls in lim- ited areas (Ganey and Baida 1994, Zwank et al. 1994, Hodgson 1996, Stacey and Hodgson 1999), or lumped sites from widely-disparate geographic areas or forest types when summarizing roost-site characteristics (Fletcher and Hollis 1994). All of these factors limit our understanding regarding the extent and sources of variability in habitat use by roosting owls. In conjunction with studies of home-range size and habitat-use patterns of radio-marked Mexican Spotted Owls in Arizona and New Mexico, we sam- pled habitat characteristics throughout the year at 1790 roost sites. Our objectives were to explore pat- terns of variation (owls, areas, and seasons) in roost-site characteristics and describe those roost sites by study area and season. In doing so, we hoped to increase understanding of roost-site char- acteristics in general and of the extent and sources of variability in roost-site characteristics. Study Areas We radio-marked Mexican Spotted Owls in three study areas. The Bar-M Canyon study area was located within the Bar-M and Woods Canyon watersheds, Coconino Na- tional Forest, approximately 40 km south of Flagstaff, Ar- izona. The other study areas were selected to represent different habitat situations within the Sacramento Moun- tains of southcentral New Mexico. The first area (mesic study area) was located along the Rio Penasco drainage, approximately 12 km southeast of Cloudcroft, New Mex- ico. The second study area (xeric study area) was located in and around the Sixteen Springs drainage, approxi- mately 18 km northeast of Cloudcroft and approximately 30 km from the mesic study area. Elevation in the Bar-M Canyon study area ranged from 1850-2440 m. Topography was relatively gentie with roll- ing terrain broken by scattered volcanic buttes and small 272 Ganey et al. VoL. 34, No. 4 canyons. Most of the study area consisted of ponderosa pine (Pinus ponderosa) forest with scattered meadows or parks. Gambel oak {Quercus gambelii) was a common as- sociate in forested areas. Alligatorbark juniper (Juniperus deppeana) was present in many stands, particularly on warmer, drier sites. Small pockets of quaking aspen {Po- pulus tremuloides) also occurred throughout the study area and small numbers of narrowleaf cottonwood {P angus- tifolia) and box elder {Acer negundo) occurred in some canyons. Topography in the Sacramento Mountains was domi- nated by moderate to steep montane canyons. Elevation m the mesic study area ranged from approximately 2400- 2800 m. Many canyon bottoms consisted of meadows, whereas forests dominated canyon slopes and ridgetops. The predominant forest type was a relatively mesic mixed-conifer forest dominated by Douglas-fir {Pseudot- suga menziesii) and/or white fir {Abies concolor). South- western white pine {P strobiformis) was prominent in most stands and ponderosa pine and quaking aspen were fre- quently present. Elevation in the xeric study area ranged from approximately 2000-2500 m. This study area con- tained a complex mosaic of mesic and xeric forest types. Mixed-conifer forest was restricted to cool microsites such as drainage bottoms and north-facing slopes. Most south-facing slopes and ridgetops were dominated by woodlands of pinyon pine {P. edulis) and alligatorbark juniper, sometimes intermixed with ponderosa pine. Other slopes were dominated by ponderosa pine forest, sometimes with a prominent component of Gambel oak. Methods Field Sampling. We sampled habitat characteristics at 1790 diurnal roost sites used by 28 radio-marked owls (12 females and 16 males). All radio-marked owls were ^1- yr-old. Roost sites were located by homing in on the radio signal until the owl was observed. If the observer moved slowly, it was often possible to locate the owl and sample habitat characteristics without causing the owl to move. When it appeared that the owl might move, sampling of some variables was omitted to minimize disturbance to the owls. This resulted in missing data, as did human errors (e.g., forgetting to bring sampling equipment). These missing data limited the types of analyses we could conduct, but appeared to be randomly distributed and unrelated to factors in analyses. Further details on cap- ture, radio-marking, and tracking of owls are given in Ganey et al. (1999). Habitat sampling was essentially plotless, but focused on the roost “microsite,” including the roost tree and its immediate surroundings. The sampling scale represent- ed a tradeoff between our desire to sample characteristics at the actual site used by the owl (rather than simply in a forest stand or general area used by the owl) and our desire to minimize disturbance to roosting owls. Because it was usually possible to sample the microsite quickly, we suspected that sampling at this scale minimized distur- bance to radio-marked owls relative to sampling larger plots. Methods for sampling habitat characteristics largely fol- lowed Solis (1983). We estimated percent slope using a clinometer. Two samples were taken per site, one up and one down-slope, then averaged for an overall estimate. We estimated aspect of the meyor slope axis using a com- pass. To estimate percent canopy cover around the roost tree, we used a spherical densiometer to sample canopy cover at a point 5 m from the roost tree in each cardinal direction, then averaged these samples for an overall es- timate. Although we use the term canopy cover here, we recognize that the densiometer actually indexes both ver- tical and horizontal cover, and thus provides a composite measure of both types of cover. For roost trees sampled, we recorded tree species and measured diameter at breast height (dbh) to the nearest cm using a dbh tape. Roost tree and owl perch heights were estimated to the nearest m using a clinometer. We estimated overslory height as the average of the heights of the three overstory trees nearest to the roost tree (sampled with a clinome- ter). We computed an index of relative roosting height as (owl roost height/ roost tree height) X 100. We also recorded information on forest cover type, roost tree species, and slope position. Cover type assign- ment was based on a visual assessment of the dominant and co-dominant tree species present. Mixed-conifer for- ests were dominated by Douglas-fir and/or white fir. Pine- oak forests were dominated by ponderosa pine with Gam- bel oak co-dominant; pine forests without a prominent oak component were classified as ponderosa pine forest Forests that did not fit one of the above descriptions were classified as “other.” Slope position was based on a combination of visual assessment in the field and use of topographic maps Three categories were recognized: upper third of slopes and ridgetops, middle third of slopes, and lower third of slopes and canyon bottoms. Data Analysis. Potential sources of variation in roost- site characteristics included individuals, sexes, study are- as, and seasons. Because of problems with missing data and diverse variable scales and types, we could not use standard multivariate techniques to partition the variance among these potential sources. Consequently, we ex- plored patterns of variation within study areas by esti- mating similarity among all possible pairs of roost sites within a study area and summarizing patterns in these similarity estimates using a linear model developed by Dyer (1978). Analyses were conducted within study areas because habitats randomly available varied, sometimes greatly, among study areas. We used Gower’s (1971) coefficient (5,y) to estimate similarity. This coefficient measures similarity on a scale ranging from 0 (where all characteristics differ between samples) to 1 (where all characteristics are identical be- tween samples). The coefficient handles both quantita- tive and categorical variables, deals conservatively with missing data, and is not sensitive to differences in the scale at which variables were measured (Gower 1971) Similarity between roost sites i and j over k variables was estimated as: where s^f, measures similarity between roost sites i and j over variable k, and represents the possibility of com- paring variable k between roost sites i and / (8,y^ = 0 when data are missing for either or both roost sites, 1 other- wise). Where 8,y^ = 0, we set = 0 (Gower 1971). Ten habitat variables were included in the similarity December 2000 Roost Sites of Mexican Spotted Owls 273 estimates. Quantitative variables included percent slope, roost tree dbh, roost tree height, owl perch height, over- story height, canopy cover, and relative owl height. Cat- egorical variables included cover type, position on slope, and roost tree species. For categorical variables, we set = 1 if roost sites i and j agreed for variable k, 0 otherwise (Gower 1971). For continuous variables with values f X ^ ) . . , x„, of variable k over n roost sites, we set = 1 — [(x, — x^)/i?^], where R/^ is the range of variable k in the sample. We computed Sy using a Fortran program. We then used a regression model (Dyer 1978) to estimate the ef- fect of two factors (owl and season) on similarity esti- mates for all possible pairs of roost sites: Sy= ^0+ where Sy is the similarity estimate for roost sites i and j, and dummy variable = 0 if roost sites i and j were from the same owl and 1 if roost sites i and j were from different owls. Similarly, = 0 if the two roost sites were from the same season and 1 if not. Sex and territory were not included as factors because they were confound- ed with owl and because both pair members were radio- marked for only 3 of 1 1 pairs of owls represented in the Sacramento Mountains. We recognized two seasons, breeding (1 March-30 August) and nonbreeding (1 Sep- tember-28 February). Because we sampled multiple roost sites for individual owls, and because a given roost site was included in mul- tiple similarity estimates, there was potentially a high de- gree of correlation among these estimates (Dyer 1978). To account for this correlation among similarity coeffi- cients estimated between two observations on the same owl, or between two observations on the same pair of owls, we used a repeated-measures model (Morrison 1976, Littell et al. 1996) to estimate regression coeffi- cients. This model estimated a separate within-subject variance and correlation for the same owl or same pair of owls for each season. Degrees of freedom for test sta- tisdcs on regression coefficients were calculated based on the number of individual owls per study area, rather than on the number of roost sites or pairwise comparisons. This is a conservative approach, similar to a Greenhouse- Geisser maximum reduction in degrees of freedom (Mor- rison 1976:214), designed to address nonindependence of within-owl samples. We used the likelihood ratio test statistic comparing the model with the correlation struc- ture to the null model without correlation structure to assess the improvement in model fit due to incorporating the correlation structure (Littell et al. 1996). Computa- tions were done using SAS PROG MIXED (v 6.12; SAS Institute Inc. 1997), We were interested in data on aspect of roost sites be- cause some studies have suggested that roost sites are concentrated on north- or east-facing slopes (Fletcher and Hollis 1994, Seamans and Gutierrez 1995) and pre- vious evidence suggested that owls may select cool mi- crosites (Barrows 1981, Ganey et al. 1993), which may occur mainly on certain aspects. We did not include data on aspect at roost sites in the above analysis, however, because we were not certain how use of circular data would affect similarity estimates. Instead, we analyzed data on roost-site aspect separately, using Oriana for Win- dows (version 1.01, Kovach Computing Services, Pen- traeth, Anglesey, Wales, U.K.). For each individual owl, we estimated the mean slope aspect {a, hereafter re- ferred to as mean azimuth) and the angular deviation (5) around the mean azimuth by season. We tested the hy- pothesis that roost sites of individuals within season were not significantly concentrated around the mean azimuth, using Rayleigh’s z statistic (Zar 1974). Where this hypoth- esis was rejected, we tested the hypothesis that mean az- imuths of individuals did not differ between seasons us- ing the Watson-Williams test (Zar 1974). This test was conducted separately for each owl. For each study area and season, we estimated an over- all a and 5 for that study area, using mean azimuths of individual owls as input. We tested the hypothesis that mean azimuths of individuals were not significantly con- centrated around the mean azimuth for the study area, using Rayleigh’s z statistic. We tested the hypothesis that mean azimuths did not differ between seasons within study area, using the Watson-Williams test. Results and Discussion The repeated-measures model, which assumed that pairs of roost sites compared between the same owl or pair of owls within a season were var- iably correlated, significantly improved model goodness-of-fit over a null model assuming no cor- relation {P < 0.0001) . After accounting for the cor- relation structure inherent in the data, similarity between roost sites was relatively high in all study areas, ranging from 0.744 in the Sacramento Mountains xeric area to 0.775 in the Bar-M Canyon area. The effects of including a different owl or season in comparisons were slight but detectable in all three areas (Table 1). Because owls often return to the same stand or general area to roost, especially during the breed- ing season, similarity estimates could be biased high. Arguing against this explanation, however, is the fact that owl and season effects were relatively slight. That is, comparing roost sites between dif- ferent owls (which use different portions of a study area) or different seasons decreased similarity only slightly. This suggested that similarity estimates were not biased high by repeated use of the same area by individuals. Rather, it suggested that within a study area, roosts sites varied little among owls or between seasons. The relative magnitude of the effects of owl and season differed among study areas, however. The owl effect was an order of magnitude greater in the mesic area than the season effect. In contrast, this pattern was reversed in the Bar-M Canyon area, and neither effect was pronounced in the xe- ric area (Table 1). Because at least one study area 274 Ganey et al. VoL. 34, No. 4 Table 1. Regression coefficients for repeated-measures models relating Gower’s similarity coefficient between roost sites of radio-marked Mexican Spotted Owls^ to owl identity and season. Separate models were estimated for each of three study areas in Arizona and New Mexico. Effect Bar-M Canyon, Arizona Sacramento Mountains, New Mexico Mesic Study Area Xeric Study Area P SE P P SE P P SE P Intercept 0.775 0.0013 <0.001 0.746 0.0005 <0.001 0.744 0.0008 <0.001 Owl -0.012 0.0014 <0.001 -0.011 0.0005 <0.001 -0.007 0.0008 <0.001 Season -0.041 0.0008 <0.001 -0.001 0.0003 0.014 -0.004 0.0006 <0.001 ® Number of owls represented by study area = 13 (Bar-M Canyon), 8 (Sacramento Mountains mesic area), and 7 (Sacramento Mountains xeric area). Number of roost sites sampled = 418 (Bar-M Canyon), 831 (Sacramento Mountains mesic area), and 541 (Sacramento Mountains xeric area). Number of pairwise comparisons = 87,153 (Bar-M Canyon), 344,865 (Sacramento Mountains mesic area), and 146,070 (Sacramento Mountains xeric area). showed a relatively strong seasonal effect on simi- larity estimates between roost sites and availability of habitat characteristics varied among study areas, we stratified descriptive statistics for roost-site char- acteristics by study area and season (Tables 2, 3). Examination of roost-site characteristics provided some possible explanations for the observed dif- ferences among areas in similarity estimates. For example, several variables (canopy cover, roost tree species, and slope position) showed more seasonal variation in the Bar-M Canyon area than the other study areas, perhaps explaining the greater season- al effect observed there. Relative to the breeding season, owls in this area roosted less frequently in Gambel oak during the nonbreeding season, and roosted more often in the middle third of slopes (Table 3). They also used roost sites with markedly lower canopy cover than those used during the breeding season (Table 2) . We suspected that the reduced use of deciduous Gambel oak could be explained by the fact that it loses most of its foliage during most of the nonbreeding season. Thus, it would provide neither hiding nor thermal cover for roosting owls for much of this season. The shedding of oak leaves may also explain the lower canopy cover observed at nonbreeding-season roosts in this study area. Most of these roost sites were in pine-oak forest (Table 3). Canopy cover should have been uniformly lower in this forest type to the extent that oak foliage no longer con- Table 2. Descriptive characteristics of roost sites of radio-marked Mexican Spotted Owls® in three study areas in Arizona and New Mexico during the breeding and nonbreeding seasons. Shown are means and standard deviations m parentheses. Sacramento Mountains, New Mexico Bar-M Canyon, Arizona Mesic Study Area Xeric Study Area Variable Breeding Nonbreeding Breeding Nonbreeding Breeding Nonbreeding Slope (%) 18.9 (13.4) 15.9 (10.8) 35.5 (17.6) 32.6 (16.4) 37.2 (16.0) 29.5 (15.9) Canopy cover (%) 74.0 (17.0) 59.4 (17.5) 76.0 (13.0) 79.7 (11.8) 69.9 (14.0) 70.3 (20.0) Roost tree dbh (cm) 32.3 (14.2) 31.1 (11.6) 40.0 (17.1) 42.7 (19.2) 28.5 (13.0) 32.3 (14.6) Roost tree height (m) 15.2 (7.1) 15.5 (5.5) 20.3 (8.8) 20.9 (7.5) 15.1 (5.8) 16.1 (5.8) Overstory height (m) 22.3 (5.5) 21.0 (5.7) 29.0 (5.2) 27.3 (5.5) 22.3 (5.2) 20.9 (5.6) Owl perch height (m) 9.5 (5.2) 10.0 (4.2) 8.2 (4.2) 8.9 (4.0) 6.6 (2.6) 6.9 (3.1) Relative owl height (%)’’ 64.0 (19.5) 65.8 (18.1) 44.5 (19.6) 45.5 (19.6) 46.0 (16.4) 44.1 (16.7) ® Number of owls represented by study area = 13 (Bar-M Canyon), 8 (Sacramento Mountains mesic area), and 7 (Sacramento Mountains xeric area). Number of roost sites sampled for breeding and nonbreeding seasons = 148 and 270 (Bar-M Canyon), 467 and 364 (Sacramento Mountains mesic area), and 287 and 254 (Sacramento Mountains xeric area). Sample sizes varied for individual variables due to missing data. ‘’Relative owl height = (owl height/ roost tree height) X 100. December 2000 Roost Sites of Mexican Spotted Owls 275 Table 3. Summary statistics (% of sites) for categorical variables at roost sites of radio-marked Mexican Spotted Owls on three study areas in Arizona and New Mexico. Sample sizes (in parentheses) differed by variable, and refer to number of roosts for which variable was recorded. Bar-M Canyon, Arizona Sacramento Mountains, New Mexico Mesic Study Area Xeric Study Area Breeding Nonbreeding Breeding Nonbreeding Breeding Nonbreeding Cover type {N = 146) {N= 262) II {N = 364) II ro 00 {N = 254) Mixed-conifer 97.4 96.4 90.7 82.4 Ponderosa pine 0.7 3.1 2.7 6.1 Pine-oak 99.3 96.9 Other 2.6 3.6 6.6 11.5 Slope position (N= 131) {N = 268) II 7 d) from single owls of either sex within a 0.2-km radius area, were designated as site centers for that species. If an ac- tual nest tree was located, this location then became the site center. The techniques we used to designate site cen- ters for both owl species were essentially identical to pro- cedures used to determine regulatory Spotted Owl site centers by state and federal agencies. Each site center is considered likely to represent a territorial individual or pair (U.S. Fish and Wildlife Service 1992). Previously- known Spotted Owl site centers which were not occupied during our three survey years were not included in the sample. Although we did not follow-up on night respons- es to determine nest sites for Barred Owls, designation of site centers was usually apparent based on clusters of responses and consistency of response locations in mul- tiple years. We provided six opportunities for territorial owls to respond to our calls and often over three respons- es were used to determine a site center. In addition, the mountainous terrain helped delimit responses, which were often located in distinct valleys and separated from a nearby site center by an obvious ridge (thereby out of hearing range of the other pair) . Simultaneous or near- simultaneous calls from adjacent pairs or singles of the same sex on a given night also helped delimit one site from the next for each species. The actual center was placed on the earliest record of a pair (or nest for Spot- ted Owls) during a season, and likewise the earliest re- cord of a single if no pair was ever found. Early season responses were assumed to be closer to a potential nest site than late season responses, although pair responses always took priority over single responses. We overlaid isopleths of annual precipitation on study area maps to compare the effects of the east-west mois- ture gradient across the Cascades on owl distribution. We also plotted site centers on habitat maps digitized from 1:64 000 aerial photography. We used mapping that was originally prepared for Spotted Owl management plan- ning based on Washington DNR habitat definitions in use at the time, and separated all habitats into three types: (1) Old Forest Habitat which was dominated by coniferous trees typically over 100 yr old, >60% canopy cover, one to multiple canopy layers, and at least 40% cover of Douglas-fir; (2) Young Forest Habitat which was dominated by trees typically <100 yr old (but of sufficient height and spacing to allow movement of owls during foraging), >60% canopy cover, and typically a single can- opy layer; (3) Non-habitat which was made up of forested habitats with overstory trees <10 m in height, stands with <60% canopy cover and/or <40% cover of Douglas-fir, deciduous stands or mixed stands with >25% deciduous overstory, and all forests >1525 m elevation. Shrub, herb and grass-dominated habitats, bare rock and talus slopes, farmland, and water were also included in the Non-hab- itat category. To compare habitats near site centers of both species, we drew concentric circles of 0.8- and 1.6-km radius around each site center. Circular areas around Spotted Owl sites have been used in similar investigations of hab- itat patterns (Lehmkuhl and Raphael 1993, Meyer et al 1998, Swindle et al. 1999). Habitat comparisons in our study were restricted to mature and young coniferous for- est habitats because of the demonstrated importance of mature forests to Spotted Owls (Thomas et al. 1990) and the suggestion that Barred Owls could use stands of youn- ger forest (Hamer 1988). We followed recent habitat studies of Spotted Owls (Meyer et al. 1998, Swindle et al. 1999) in selecting the two circular areas for determining the nest-site locations of Spotted Owls. Radii of ^0.8 km have been shown to have significant differences in comparisons of habitat around nest sites and random forest sites (Meyer et al. 1998, Swindle et al. 1999), and differences in the amount of old forest may occur up to 1.6 km (Swindle et al. 1999). We stratified owl sites in our area into three sub- units based primarily on precipitation criteria: west of the Cascade Range crest, east of the crest to the 150 cm pre- cipitation isopleth, and east of the 150 cm isopleth. Our comparisons of average amounts of each habitat type within the tested radii were achieved using multiple anal- ysis of variance (MANOVA), following tests for normality and use of the Wilks’ lambda (likelihood ratio criterion) to test for significant interaction between variables (SYS- TAT version 8.0). We randomly sampled circular areas around site centers, and used mutually exclusive (site centers of both species tested at 0.8-km radius were not used for tests at 1.6-km radius), nonoverlapping areas for both radii. Results Population Size and Distribution. Spotted Owl surveys, when combined over three breeding sea- December 2000 Barred and Spotted Owl Populations in Central Washington 283 sons, covered 1280 km^. Portions of the surveyed area above 1525 m, or extensive areas classified as nonhabitat were not surveyed. We may have missed some Barred Owls by not surveying in forested habitats containing >75% deciduous trees. Large stands of mixed and deciduous forests comprised <2% of the study area and occurred only in the floodplain of the Green and Yakima Rivers. Like- wise, stands of <60% canopy closure were uncom- mon and small in size, and often occurred adjacent to surveyed stands, therefore receiving limited sur- vey coverage (Fig. 1 ) . A total of 62 Spotted Owl site centers and 53 Barred Owl site centers were identified. Spotted Owls were well-distributed across the area (0.047, 0.043, and 0.053/km^ from west to east by subunit; Fig. 1). Barred Owls were most abundant west of the Cascade crest (0.063/km^), with similar densi- ties (0.063/km^) extending east of the crest only within the 150 cm/yr isopleth for annual precipi- tation. East of this line. Barred Owl densities dropped to O.OlO/km^. To the west of our survey area within the Cascade Range, only Barred Owls were located during similar surveys from 1991-93 (L. Young pers. comm.). To the east of our study area, several additional Spotted Owl sites and a few Barred Owl sites have been located across north- ern Kittitas County almost to the forest/sagebrush steppe interface (S. Sovern and M. Taylor pers. comm.). Also, we have found Barred Owls breed- ing at sites both lower and higher in elevation than known Spotted Owl nest locations. Barred Owls have completely overlapped the known geographic and altitudinal distribution of Spotted Owls in cen- tral Washington. Habitat Analyses. We found no significant differ- ences in the mean amount of all habitat types be- tween Spotted and Barred Owls within the 1.6 km radius analysis area (Wilks’ X = 0.946, P — 0.475) around site centers. Within the 0.8 km radius sur- rounding Spotted Owl and Barred Owl sites, how- ever, significant differences in mean habitat amounts were detected (Wilks’ X = 0.725, P — 0.003). Spotted Owl sites contained more old for- est close to the site center than Barred Owl sites. Within the three geographic regions we tested, MAN OVA results indicated that mean amount of habitat differed significantly within the 0.8 km (Wilks’ X = 0.594, P — 0.001) radius. There was consistendy more old forest surrounding Spotted Owl sites than Barred Owl sites in all subunits (Ta- ble 1 ) . Barred Owl sites also contained more young forest than Spotted Owl sites in the far west and far east subunits. In the dry zone of the eastern Cascades east of the 150-cm isopleth, 8 of 12 Barred Owl site cen- ters were found in moister forest situations, such as those along major river or stream drainages or near lakes or wooded swamps or at higher eleva- tions where the true amount of precipitation may actually have exceeded 150 cm/yr. On both slopes of the Cascade Range, several Barred Owl sites oc- curred in deciduous and mixed forest stands found exclusively in major river valleys. Forest stands dominated by deciduous trees are not considered important Spotted Owl habitat in Washington (WSFPB 1996). East of the 150-cm isopleth, Spot- ted Owl sites were typically located in coniferous forests on the sides of slopes and were not found in the habitats described above for Barred Owls. West of the 150-cm isopleth and above m^or river valleys, however. Spotted Owl sites occurred in very similar situations to those of Barred Owls. We did not find Spotted Owl nests in high-elevation, true fir-dominated forests. Our own observations of Barred Owls, plus those of Wright and Hayward (1998), suggest that this species is also more com- mon in lower elevation mixed conifer forests than in high elevation spruce-fir forests. Discussion Population Size and Distribution. The full im- pact of the Barred Owl range expansion into the Pacific Northwest on resident Spotted Owls prob- ably has yet to be fully realized. We detected almost as many Barred Owls as Spotted Owls, and in some portions of the Washington Cascades, Barred Owls have become more numerous than Spotted Owls. We could have missed some territories of both owl species, particularly Barred Owls; however, we re- ceived consistent responses from both species at night and during the day, even though we only broadcast Spotted Owl calls. Responses obtained during the day were typically at closer range than at night. Daytime surveys were designed with closer spacing of calling stations and transects to account for this tendency. Even so, we could have missed some owls, particularly Barred Owls, because of in- dividual variation in response levels to calls of a congener. All Spotted Owl sites known in the survey area were monitored for occupancy and reproduction annually from 1991-98. Of the 62 known sites, 22 were unoccupied at least temporarily by both 284 Herter and Hicks VoL. 34, No. 4 Table 1 . Comparison of mean hectares of habitat present within selected radii around Barred Owl and Spotted Owl site centers across three geographic regions in the Central Cascade range of Washington. Radius’’ West=* East^* East 150^ (km) Mean 95% Cl N Mean 95% Cl N Mean 95% Cl N 0.8 km Old/Mature Forest Barred Owl 57 43 10 81 27 11 55 47 4 Spotted Owl 83 32 8 106 42 4 98 30 13 Young Forest Barred Owl 72 34 10 42 29 11 41 41 4 Spotted Owl 51 36 8 5 10 4 40 20 13 Non-habitaU Barred Owl 73 33 10 79 30 11 106 83 4 Spotted Owl 68 35 8 87 33 4 64 20 13 1 6 km Old/Mature Forest Barred Owl 173 153 8 430 112 9 304 125 8 Spotted Owl 182 136 5 334 85 9 323 97 15 Young Forest Barred Owl 354 178 8 91 98 9 163 94 8 Spotted Owl 420 133 5 117 71 9 178 55 15 Non-habitat Barred Owl 403 161 8 442 144 9 545 200 8 Spotted Owl 405 42 5 456 156 9 489 103 15 ^ West = west of the Cascade Range crest. East = east of the Cascade Range crest but west of the 150 cm/yr rainfall isopleth, East 150 = east of the 150 cm/yr rainfall isopleth. •’Area within 0.8 km radius = 201 ha; 1.6 km radius = 804 ha. Non-habitat included non-forest, deciduous-dominated forests, and high-elevation forests. members of the original pairs. Of these 22 sites, half remained unoccupied through 1998. Of the remaining 11 sites, six were reoccupied by differ- ent Spotted Owl pairs or single individuals, while Barred Owls were present at or near five site cen- ters. In most cases. Barred Owls were already pre- sent in the vicinity (^0.8 km) prior to the disap- pearance of the Spotted Owl pairs. In one instance, a newly established pair of Spotted Owls nested within 1 km (and hearing distance) of an estab- lished Barred Owl site. Surveys over additional years are needed to determine whether Spotted Owls regularly reoccupy sites in close proximity to Barred Owl territories. Habitat Analyses. In portions of the western Washington Cascades west of our study area where less old forest remained, Barred Owls have occu- pied second-growth Douglas-fir/western hemlock stands with remnant large trees and snags which provide nest cavities. Spotted owls have been known to occur in landscapes where young forests predominate (Forsman et al. 1988, Irwin et al. 1991) , but they persist at low densities and typically nest in a patch of old forest. In our study area, where relatively large stands (>200 ha) of old for- est habitat remained, surrounded by a mosaic of managed and unmanaged fire-regenerated habitat, both Spotted Owls and Barred Owls occupied nest- ing territories and produced young. Our data sug- gested that Barred Owls persisted in areas with less old forest than Spotted Owls. Within conservation areas designed for Spotted Owl habitat protection, management options that consolidate and protect preferred habitat for Spot- ted Owls in well-spaced, large blocks (>100 ha) may help them compete with Barred Owls in Cas- cade Range forests. Recent studies by Meyer et al. (1998) and Swindle et al. (1999) have also noted a preference for an unfragmented patch of old for- est around Spotted Owl nest sites. This does not mean that Barred Owls cannot successfully occupy areas of extensive cover of old forest. Observations by Wright and Hayward (1998) and our own ob- servations in neighboring wilderness areas and na- December 2000 Barred and Spotted Owl Populations in Central Washington 285 tional parks indicated that territorial Barred Owls can occur in wilderness valleys with extensive cover of old forest. Some competition for resources likely takes place where the two species are sympatric because of significant overlap in habitat use, prey species, and nest-site preferences. Spotted Owls and Barred Owls were previously sympatric in only one other area in North America, at the southern limit of the ranges of both species in the southern Sierra Ma- dre Occidental of Mexico (Enriquez-Rocha et al. 1993, Howell and Webb 1995). In Mexico, there are two different subspecies and the duration of the sympatry has been longer. In our study area, the northern subspecies of both owls appear to co- exist in very similar habitats in the wet, western Cascades, but they may be exhibiting greater hab- itat separation in the eastern Cascades. In these dryer forests, the predominance of Spotted Owls in conifer forests at mid-slope (Buchanan et al. 1995), and Barred Owls in forested wetlands, mixed riparian stands, and high elevation moist co- niferous forests, mirrored the habitat use of the species over the majority of their respective ranges. Spotted Owls, outside the coastal Pacific North- west, are primarily found in relatively-dry, western mountains, while Barred Owls occur in more mesic habitats in eastern mixed or deciduous forests and boreal forests. Barred Owls were already well-established on our study area by the time we began our surveys. We found no mixed-species (Barred Owl/Spotted Owl) pairs or hybrid owls, but hybrids have been reported from Washington and other parts of the Northern Spotted Owl range (Hamer et al. 1994). Widespread hybridization in the central Washing- ton Cascades did not appear to be continuing. As shown in other species (Short 1969, Rohwer 1972), it is likely that once Barred Owls established self- sustaining local populations, individuals of the in- vading species no longer had trouble finding con- specific mates, minimizing the incidence of mixed-species pairing. Although this study suggests only minor differ- ences in the amount of old and mature forest hab- itat surrounding Spotted and Barred Owl site cen- ters based on the broad serai stages used in our analyses, perhaps more detailed habitat use studies would indicate more partitioning. The extent that habitat or niche separation will keep the two spe- cies from competing directly for resources should be considered speculative. However, direct com- petition in some habitats appears likely and may negatively affect Spotted Owl population recovery. Acknowledgments We are indebted to the many field biologists who spent long nights surveying for owls and hiked many miles m steep terrain to provide complete survey coverage; prom- inent among them were: L. Melampy, A. Stabins, M. Mac- Donald, J. Bottelli, C. Smith, M. Rabanal, K. Jorgensen, D. Malkin, H. Smith, S. Sagor, M. Lanphere, B. Shepard, C. Holloway, A. Raedeke, G. Riddick, M. Richey, and C. Eakins. R. Early deftly carried out the GIS analysis and B. Marx, T. Hitzroth, and M. Baumgartner produced the habitat maps. C. Olson, H. Stabins, and T. Hillman as- sisted with data analyses. We also thank S. Sovern and M Taylor of the U.S. Forest Service, Pacific Northwest Re- search Laboratory, for providing additional field survey coverage at owl sites and review of the manuscript. Drafts of the manuscript were also improved by the comments of G. Hayward, H. Stabins, and two anonymous reviewers. Literature Cited Allen, H.L., K.R. Dixon, and K.L. Knutson. 1989. Co- operative administrative study to monitor Spotted Owl management areas in national forests in Washington Unpubl. Rep., Washington Depart. Wild!., Olympia WA, U.S.A. Buchanan, J.B., L.L. Irwin, and E.L. McCutchen. 1995. Within-stand nest site selection by Spotted Owls in the eastern Washington Cascades. J. Wildl. Manage. 59: 301-310. Dark, S.J., RJ. Gutierrez, and G.I. Gould, Jr. 1998. The Barred Owl (Strix varia) invasion in California. Auk 115:50-56. Devereaux, J.G. and J.A. Mosher. 1984. Breeding ecol- ogy of the Barred Owl in the central Appalachians Raptor Res. 18:39-58. Dunbar, D.L., B.P. Booth, E.D. Forsman, A.E. Hether- iNGTON, and D.J. Wilson. 1991. Status of the Spotted Owl, Strix ocddentalis, and Barred Owl, Strix varia, in southwestern British Columbia. Can. Field-Nat. 105 464-468. Enriquez-Rocha, R, J.L. Rangel-Salazar, and D.W Holt. 1993. Presence and distribution of Mexican owls: a review./. Raptor Res. 27:154—160. Forsman, E.D. 1988. A survey of Spotted Owls in young forests in the northern Coast Range of Oregon. Mur- relet 69:65-68. Gutierrez, R.J., A.B. Franklin, and W.S. LaHaye. 1995 Spotted Owl {Strix ocddentalis) . Pages 1-28 m A. Poole and F. Gill [Eds.], The birds of North America, No. 179. The Academy of Natural Sciences, Philadelphia, PA and American Ornithologists’ Union, Washington, DC 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, W. Wash. Univ., Bellingham, WA U.S.A. 286 Herter and Hicks VoL. 34, No. 4 , E.D. Forsman, A.D. Fuchs, and M.L. Walters. 1994. Hybridization between Barred and Spotted Owls. Auk 111:487-492. Haney, J.C. 1997. Spatial incidence of Barred Owl (Strix varia) reproduction in old-growth forest of the Ap- palachian Plateau./. Raptor Res. 31:241—252. Howell, S.N.G. and S.B. Webb. 1995. The birds of Mex- ico and northern Central America. Oxford Univ. Press, New York, NY U.S.A. Irwin, L., XL. Fleming, S.M. Speich, and J.B. Buchanan. 1991. Spotted Owl presence in managed forests of southwestern Washington. Tech. Bull. No, 601, Na- tional Council of the Paper Industry for Air and Stream Improvement, Inc., Corvallis, OR U.S.A. Lehmkuhl, J.F. and M.G. Raphael. 1993. Habitat pattern around Northern Spotted Owl locations on the Olym- pic Peninsula, Washington. / Wildl. Manage. 57:302- 315. Mazur, K.M., PC. James, M.J. Fitzsimmons, G. Langen, AND R.H.M. Espie. 1997. Habitat associations of the Barred Owl in the boreal forest of Saskatchewan, Can- ada. /. Raptor Res. 31:253-259. Meyer, J.S., L.L. Irwin, and M.S. Boyce. 1998. Influence of habitat abundance and fragmentation on Northern Spotted Owls in western Oregon. Wildl. Monogr. 139: 1-51. Rohwer, S.A. 1972. A multivariate assessment of inter- breeding between the meadowlarks, Sturnella. Syst. Zool. 21:313-338. Sharp, D.U. 1989. Range extension of the Barred Owl in western Washington and first breeding record on the Olympic Peninsula./. Raptor Res. 23:179-180. Short, L.L. 1969. Taxonomic aspects of avian hybridiza- tion. Auk 86:84—105. Swindle, K.A., W. J. Ripple, E.C. Meslow, and D. Schaf- er. 1999. Old-forest distribution around Spotted Owl nests in the central Cascade Mountains, Oregon. / Wildl. Manage. 63:1212-1221. Taylor, A.L., Jr. and E.D. Forsman. 1976. Recent range extensions of the Barred Owl in western North Amer- ica, including the first records for Oregon. Condor 78: 560-561. 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. U.S. Fish Wildl. Serv. Pordand, OR U.S.A. U.S. Fish and Wildlife Service. 1991. Protocol for sur- veying proposed management activities that may im- pact Northern Spotted Owls. U.S. Fish Wildl. Serv., Portland, OR U.S.A. . 1992. Protocol for surveying proposed manage- ment activities that may impact Northern Spotted Owls (Revised) . U.S. Fish Wildl. Serv., Portland, OR U.S.A. Washington State Forest Practiges Board (WSFPB). 1996. Final environmental impact statement on forest practices rule proposals for: Northern Spotted Owl, Marbled Murrelet, Western Gray Squirrel. Wash. Dept. Nat. Resources, Olympia, WA U.S.A. Wright, A.L. and G.D. Hayward. 1998. Barred Owl range expansion into the central Idaho wilderness./. Raptor Res. 32:77-81. Received 28 January 1999; Accepted 1 August 2000 / Raptor Res. 34(4):287-292 © 2000 The Raptor Research Foundation, Inc. FOOD HABITS OF BALD EAGLES WINTERING IN NORTHERN ARIZONA Teryl G. Grubb and Roy G. Lopez USDA Forest Service, Rocky Mountain Research Station, 2500 S. Pine Knoll Drive, Flagstaff, AZ 86001-6381 U.S.A. Abstract. — ^We used pellets collected from roosts to supplement incidental foraging observations to iden- tify prey species of Bald Eagles {Haliaeetus leucocephalus) and to evaluate spatial and temporal trends in their food habits while wintering in northern Arizona between 1994—96. We analyzed 1057 pellets collected from 14 roosts, and identified five mammal and 13 bird species. American Coot {Fulica americana, N = 447) and elk/ deer (Cervus elaphus/ Odocoileus hemionus, N = 412) were the most common prey remains we identified and they varied annually and inversely with each other (11-58% for coots and 21-78% for elk/ deer). Diving ducks (92%) were more heavily represented in pellets with identifiable bird prey {N = 701) than dabblers (1%), although Christmas Bird Counts indicated 64% divers and 36% dabblers in the study area (A = 18 202; = 46.3, df = 1 , P < 0.01). Almost all pellets consisted mostly of mammal or bird remains (N = 366 and 689, respectively). The overall ratio of mammal to bird pellets was 59:41, with relative class frequencies varying between years = 118.29, df = 2, P < 0.01). At roosts <3 km from water (A = 752), 90% of the pellets contained birds; whereas, at roosts >3 km from water (A = 303), 96% of the pellets contained mammals (x^ = 698.54, df = 1, P < 0.01). In three successive winters of varying weather conditions, wintering eagles foraged primarily on mammals, fish, and waterfowl, respec- tively; but only mammals and waterfowl were accurately represented in pellets. Key Words: Bald Eagle', Haliaeetus leucocephalus; pellets', diet', food habits', winter roosts', winter habitat. Habitos alimenticios de Haliaeetus leucocephalus en Arizona Resumen. — Utilizamos las egragopilas recolectadas en sitios de perchas para complementar las obser- vaciones de forrajeo e identificar las especies de presas de Haliaeetus leucocephalus como tambien para evaluar las tendencias espaciales, temporales y sus habitos alimenticios durante su estadia de invierno en Arizona entre 1994-96. Analizamos 1057 egragopilas recolectadas en 14 sitios de perchas, identifi- camos 5 especies de mamiferos y 13 de aves. Fulica americana (A — 447) y Cervus elaphus /Odocoileus hemionus (A = 412) fueron los restos de presas mas comunes identificados. Estos variaron anualmente e inversamente entre ellos (11-58% para americana y 21-78% para Cervus elaphus /Odocoileus hem- ionus). Diving ducks (92%) fueron mas representados en las egagropilas como aves presa (A = 701) que los Dabblers (1%), aunque los Conteos de Navidad indicaron una representatividad de 64% para Olivers y 36% para Dabblers en el area de estudio (A = 18 202; x^ = 46.3, df = 1, P < 0.01). Casi todas las egagropilas fueron restos de mamiferos o aves (A = 366 y 689 respectivamente) . La proporcion total de egagropilas de mamiferos y aves fue de 59:41, con frecuencias relativas de clase entre anos (x^ = 118.29, df = 2, P < 0.01). En los sitios de perchas <3 km del agua (A = 752), el 90% de las egagropilas contenian aves, mientras que los sitios de perchas a > 3km del agua (A = 303) , el 96% de las egagrdpilas contenian mamiferos (x^ = 698.54, df = 1, P < 0.01). En tres inviernos subsecuentes con variaciones climaticas, las aguilas forrajearon principalmente mamiferos, peces, y aves acuaticas respectivamente, pero solo los mamiferos las aves fueron representadas con certeza en las egagropilas. [Traduccion de Cesar Marquez] Food habits of nesting Bald Eagles {Haliaeetus leucocephalus) in Arizona are well-documented (Haywood and Ohmart 1986, Hunt et al. 1992, Grubb 1995), but information on the diet of winter migrants is limited (Grubb and Coffey 1982, Grubb and Kennedy 1982, Brown 1993). As part of a long-term study of wintering Bald Eagles in northern Arizona (Grubb et al. 1989, Grubb et al. 1994, Grubb 1996), we collected pellets from be- neath roost trees during three winters from 1994- 96. Typically, pellets effectively supplement direct observations and prey remains in determining lo- cal diets (Grubb and Kennedy 1982, Stalmaster and Plettner 1992, Isaacs et al. 1993). Fish and 287 288 Grubb and Lopez VoL. 34, No. 4 large mammals, for example, can be underrepre- sented in pellets, and small mammals overrepre- sented. However, the overall proportion of birds and mammals in pellet analyses tends to reflect ac- tual diet (Mersmann et al. 1992). In northern Ar- izona, mid-winter foraging on fish is rare (Grubb and Lopez 1997), and prey remains of waterfowl and large ungulate carrion are often difficult to find. In addition, the relatively-small, ephemeral population of wintering eagles limits foraging ob- servations (Grubb and Kennedy 1982, Grubb et al. 1989). Therefore, to supplement limited, inciden- tal foraging observations, we relied primarily on pellets to identify prey species and to evaluate spa- tial and temporal trends in the food habits of Bald Eagles wintering in northern Arizona. Study Area and Methods Our study area was the Coconino National Forest, sur- rounding the town of Flagstaff, Arizona, in Coconino and Yavapai counties. Habitat is dominated by ponderosa pine {Pinus ponderosa) forest transitioning into forests of pinyon pine-juniper {P. edulis-Juniperus spp.) at lower el- evations (Brown 1982). Elevation ranges between 1524- 2439 m in semimountainous terrain. The only perma- nent water bodies in the vicinity of our study roosts were several small lakes (<1000 ha). Winter weather in north- ern Arizona varies within and between years with occa- sional heavy snows (<0.6 m) and cold temperatures (lows exceeding — 18°C), interspersed with dry periods of mild temperatures (highs to 10°C) and general loss of snow cover. We collected pellets during three winters: 1993-94, 1994—95, and 1995-96, referred to hereafter as the win- ters of 1994, 1995, and 1996, respectively. We used daily minimum low temperatures (°C) measured at the Flag- staff airport (National Oceanographic and Atmospheric Administration, on-line data) to contrast winter weather conditions between study years (October through March, 1994—96). We measured the distance from each roost to the nearest permanent water on U.S. Geological Survey, 7 5-min quadrangle maps. We collected pellets from beneath roost trees in 14 previously-identified night roosts (Grubb et al. 1989, Dar- gan 1991). Bald Eagle use of winter roosts was highly variable, ranging from one eagle in a single tree to >40 eagles in >25 trees. Pellets were collected by walking around roost trees clockwise then counter-clockwise, 2- and 5-m away from the base of the tree, or farther if necessary to accommodate tree lean. We removed pellets from each roost prior to the next field season. We were unable to collect pellets from every 1994 roost during the following two winters because of weather and scheduling limitations. Because 42% of all the pellets collected, and 68% of the 1996 pellets resulted from an exceptional concentration of >40 eagles foraging on waterfowl and using a single roost (hereafter referred to as Roost 8) in 1996, we evaluated annual variation and calculated over- all class composition with and without data from this roost. We dissected pellets in the laboratory to determine their contents. Identification of prey items was made to the lowest taxonomic level possible by comparison to avi- an study skins and mammalian hair samples. Visual esti- mates of percent volume were made for each taxon. Pel- lets comprised of >50% mammal or >50% bird remains by volume were classified by predominant class to facili- tate comparisons between years, roosts, and habitat Adorjan and Kolenosky (1969) and Moore et al. (1974) facilitated hair identification, but we could not differen- tiate elk (Cervus elaphus) and mule deer {Odocoileus hem- ionus) hair. However, based on our field observations of available large ungulate carrion, we estimated elk com- prised >85% of the elk/deer hair sample. Similarly, Red- head {Aythya americana) and Canvasback (A. valisineria) feathers in pellets could not be differentiated. Because Redheads greatly outnumbered Canvasbacks on local lakes (Morrall and Coons 1996; National Audubon Soci- ety Christmas Bird Counts 1994—96, Laboratory of Orni- thology, Cornell University, on-line data; pers. obs.), we estimated they comprised >80% of this feather sample. We used SPSS 7.5 for Windows (SPSS 1997) to calcu- late frequencies and descriptive statistics, and chi-square tests for evaluating variation in frequencies of mammal and bird pellets among years and between lake and up- land roosts. Sample sizes and percentages are not always additive because some pellets contained more than one class or species. We used the frequency of occurrence (i.e., the number or percent of pellets containing a class or species) as the measure of relative abundance. Results We collected 1057 Bald Eagle pellets (Table 1). Of the 885 pellets with distinguishable prey spe- cies, 823 (93%) contained only a single species, but 61 (7%) had two species and one (<1%) had three species. We identified five mammal and 13 water- fowl species (Table 2) . American Coot {Fulica amer- icana, N = 447) and elk/deer {N = 412) were the most common prey items. Relative frequencies of these two species varied annually and inversely with each other (11-58% for coots and 21-78% for elk/ deer) . Diving ducks were more heavily represented (92%, including American Coots) than dabblers (1%) in eagle pellets with identifiable avian prey {N = 701). However, National Audubon Society Christmas Bird Counts for northern Arizona be- tween 1994—96 (Laboratory of Ornithology, Cor- nell University, on-line data), indicated much less difference in the numbers of divers (64%) and dabblers (36%) during our study {N = 18 202; = 46.3, df — 1, P < 0.01). Nine of 12 locally-com- mon species of divers (75%) were identified in our pellet analysis, and four of seven dabblers (57%). Based on the most prevalent prey species in pel- lets, we identified 366 mammal, 689 bird, and 2 December 2000 Winter Bald Eagle Food Habits 289 Table 1. Number of pellets collected each year and the relative frequency (% total number) of mammalian and avian prey by roost and by year, for 1057 pellets collected beneath 14 winter Bald Eagle roosts in northern Arizona, 1994-96. Roost Distance TO Water No. Pellets (% Mammal/ % Bird) (Trees) (km)^ 1994 1995 1996 Total 1 (4) 0.05 2^ (50/0) 1 (0/100) 1 (0/100) 4b (25/50) 2 (19) 0.25 30 (13/87) 0 6 (0/100) 36 (11/89) 3 (14) 0.4 51 (69/31) 15 (40/60) 29 (41/59) 95 (56/44) 4 (9) 0.5 22 (0/100) 0 0 22 (0/100) 5 (13) 0.85 14 (7/93) 5 (0/100) 23 (9/91) 42 (7/93) 6 (6) 1.7 2 (0/100) — — 2 (0/100) 7 (18) 1.9 0 0 98 (2/98) 98 (2/98) 8 (27) 2.2 5 (60/40) — 449 (2/98) 454 (3/97) 9 (1) 3.5 1 (100/0) — — 1 (100/0) 10 (27) 5.0 38 (100/0) 19 (100/0) 4 (100/0) 61 (100/0) 11 (5) 6.5 12 (100/0) — 0 12 (100/0) 12 (3) 7.6 1 (100/0) — — 1 (100/0) 13 (2) 13.0 12 (100/0) — — 12 (100/0) 14 (31) 18.2 110 (91/9) 61 (97/3) 46b (96/2) 217 (94/6) Totals; 14^ (179) 8L/6U Totals without Roost 8*^ 300 (69/30) 101 (83/17) 656 (11/89) 1057 (35/65) 13^ (152) 7L/6U 295 (69/30) 101 (83/17) 207 (31/69) 603 (59/41) ® Roosts <3.0 km (x = 1.0, SD = 0.8) from permanent lakes were classified as lake (L), and roosts >3.0 km (x — 9.0, SD = 5 1) were classified as upland (U). One pellet was entirely fish remains. Total number of roosts. ^ Since 42% of all the pellets collected, and 68% of the 1996 sample, came from an unusual concentration of >40 eagles foraging on waterfowl and using Roost 8 in 1996, both typical annual variation and overall class composition were more accurately represented without Roost 8 data. fish pellets. The overall ratio of mammal to bird pellets, excluding Roost 8, was 59:41, with relative class frequencies varying between years (x^ = 118.3, df = 2, P < 0.01). The overall ratio with Roost 8 included was 35:65. Class frequencies were generally consistent at individual roosts from year to year; only Roosts 3 and 8 varied among years (Table 1). However, class frequencies varied be- tween roosts and appeared related to roost dis- tance from permanent water. At roosts <3 km from water (classified as lake roosts), 90% of the pellets contained mostly bird remains {N — 752), whereas at roosts >3 km from water (classified as upland roosts), 96% of the pellets contained mam- malian remains {N = 303, x^ = 698.5, df = 1, P < 0.01). Conversely, 98% of all bird pellets occurred in lake roosts and 79% of all mammal pellets oc- curred in upland roosts. Weather conditions varied among the three years of our study. Temperatures generally de- clined from October through mid-December with repeated, brief cold cycles throughout the winter of 1994. During the winter of 1995, generally cold temperatures in November and early December with shorter warming periods than the previous year led to a freeze-over of local lakes by late De- cember. January also had nearly two weeks of un- seasonable cold before temperatures began to in- crease through March. The winter of 1996 was generally mild and characterized by only three, 10- 14 d cold cycles between mid-December and late February. In the relatively-typical winter of 1994, pellets confirmed that Bald Eagles fed primarily on large mammal carrion (69%, Table 1). In 1995, water- fowl numbers were again minimal after freeze-over and, although the number of pellets was down, de- pendence on mammalian carrion was evident 290 Grubb and Lopez VoL. 34, No. 4 Table 2. Relative class and species abundance in 1057 pellets collected beneath 14 Bald Eagle winter roosts in northern Arizona, 1994-96. No. Pellets (% Annual Total"') Class/ Species 1994 {N = 300) 1995 {N= 101) 1996 {N = 656) Total {N= 1057) Mammal 217 (72) 86 (85) 146 (22) 449 (42) Elk/mule deer'’ {Cervus da- 198 (66) 78 (78) 136 (21) 412 (39) phus/Odocoileus hemionus) Cottontail rabbit {Sylvilagus 27 (9) 4 (4) 6 (1) 37 (4) spp.) Black-tailed jackrabbit {Lepus 3 (1) 2 (2) 0 5 (<1) californicus) Coyote {Canis latrans) 2 (1) 1 (1) 0 3 (<1) Unknown mammal 27 (9) 11 (11) 21 (3) 59 (6) Bird 96 (32) 17 (17) 588 (90) 701 (66) American Coot {Fulica ameri- 59 (20) 11 (11) 377 (58) 447 (42) cana) Ruddy Duck {Oxyura jamaicen- 11 (4) 0 93 (14) 104 (10) sis) Ring-necked Duck {Aythya col- 7 (2) 0 26 (4) 33 (3) laris) Redhead/ Canvasback'’ {Aythya 6 (2) 1 (1) 23 (4) 30 (3) americana/Aythya valisinena) Northern Shoveler {Anas cly- 1 (<1) 0 1 (<1) 2 (<1) peala) Mallard {Anas platyrhynchos) 4 (1) 0 0 4 (<1) Northern Pintail {Anas acuta) 0 0 1 (<1) 1 (<1) Green-winged Teal {Anas crec- 0 0 1 (<1) 1 (<1) ca) Lesser Scaup {Aythya affinis) 0 0 1 (<1) 1 (<1) Western Grebe {Aechmophorus 11 (4) 5 (5) 6 (1) 22 (2) occidentalis) Pied-billed Grebe {Podilymbus 1 (<1) 0 4 (<1) 5 (<1) podiceps) Eared Grebe {Podiceps nigticol- 0 0 3 (<1) 3 (<1) Us) Unknown bird 10 (3) 4 (4) 275 (42) 289 (27) Unknown fish 1 (<1) 0 2 (<1) 3 (<1) Other materiab 6 (2) 3 (3) 27 (4) 36 (3) ■> Numbers and percentages of pellets are not additive because some pellets contained >1 class or species. '' Species were grouped because hair or feathers present in pellets were not distinguishable. ‘ Vegetation, seeds, soil, sand, small stones; monofilament line was also found in one pellet. (83%). However, the harsh weather conditions caused an extensive die-off of feral fathead min- nows {Pimephales promelas) which became the pri- mary food we observed Bald Eagles eating that win- ter (Grubb and Lopez 1997). During 1996, waterfowl remained locally abundant all winter, with large flocks of several hundred birds concen- trated in small openings in lake ice during cold periods. Very few road- or winter-killed elk were observed and pellets confirmed our observations that Bald Eagles fed primarily on birds (89%). Disc:ussion The 59:41 frequency ratio of mammal to bird pellets, excluding Roost 8, was consistent with our combined local field experience over the past 24 yr. Bald Eagles wintering in northern Arizona typ- ically depend on elk carrion as their primary food. December 2000 Winter Bald Eagle Food Habits 291 beginning in late fall when visceral piles left during hunting season are abundant. Waterfowl provide an opportunistic, alternative food source when available. In similar habitat around Nav^o Lake in northern New Mexico, pellets indicated Bald Eagle use of deer and elk carrion varied inversely with small mammal and waterfowl consumption, de- pending on weather and prey availability (Grubb 1984). Mild weather permits waterfowl to remain on northern Arizona lakes, whereas harsher winter conditions force them to leave and large ungulates to become more vulnerable to road- and winter- kill. Therefore, there is a weather-driven, relatively- stable food base for visiting, winter eagles even though prey classes vary (Grubb and Kennedy 1982) . The 35:65 overall ratio of mammal to bird pellets we obtained by including Roost 8 was more representative of a mild winter with abundant-wa- terfowl, such as in 1996. Differential foraging by wintering Bald Eagles on diving and not dabbling waterfowl may be a func- tion of winter icing conditions and differing re- sponse behaviors to hunting eagles. Since dabbling ducks can launch into flight quickly, they are vul- nerable for only a brief period. Divers, on the oth- er hand, require a stretch of open water to get airborne, increasing their exposure to eagle pre- dation. Diving to escape can also leave them vul- nerable to foraging Bald Eagles. We have observed eagles circling overhead, singly (Brattstrom 1989) and in cooperative groups (Sherrod et al. 1976), to repeatedly drive their prey back underwater un- til exhausted. Icing of lake surfaces exacerbates diving duck vulnerability by concentrating large numbers of waterfowl in small areas, precluding flight by reducing take-off space, and limiting the diving area for underwater maneuvering. In the three successive winters of our study, dif- ferent weather conditions resulted in Bald Eagles foraging primarily on mammals, fish, and water- fowl, respectively. However, only the 1994 depen- dence on mammals and 1996 dependence on wa- terfowl were accurately represented by pellets. Nonetheless, our results suggest that pellets can provide effective assessments of long-term trends in mammal and bird use, and an indication of the relative abundance of prey within each class (Mers- mann et al. 1992, Stalmaster and Plettner 1992). Pellet analyses need not be avoided in winter diet assessment (Stalmaster and Plettner 1992), espe- cially under limiting circumstances such as those that we encountered. As demonstrated by our win- ter 1995 results, pellet analyses should be substan- tiated with observations as much possible, and the general absence of fish representation should be taken into account. In addition, the variation we recorded in wintering Bald Eagle diet, both an- nually and among roost locations, mandates a large pellet sample well-distributed over time and space. Acknowi.edgments We thank J. Yazzie for field a.ssistance and S. Masek Lopez for initial pellet analysis. We also appreciate use of the skin collections at the Museum of Northern Arizona, Northern Arizona University, University of Arizona, and Museum of Southwest Biology at the University of New Mexico, and the assistance of D. Hill, T. Huels, and R. Baida who facilitated access. R. Smith of the Arizona De- partment of Game and Fish also provided helpful infor- mation. F. Isaacs, R. Lehman, and R. McClelland con- structively reviewed the original manuscript. Literature Cited Adorjan, A.S. AND G.B. Kolenosky. 1969, A manual for the identification of hairs of selected Ontario mam- mals. Res. Rep. No. 90. Ontario Dept, of Lands and Forests, Res. Bur., Toronto, Ontario, Canada. Brattstrom, B.H. 1989. Predation of Bald Eagles {Hal- iaeetus leucocephalus) on American Coots {Fulica amer- icana).J. Raptor Res. 23:16-17. Brown, B.T. 1993. Winter foraging ecology of Bald Eagles in Arizona. Cowdor 95:132-138. Brown, D.E. [Ed.]. 1982. Biotic communities of the American Southwest-United States and Mexico. Desert Plants 4:1—342. Dargan, C.M. 1991. Roost-site characteristics of Bald Ea- gles wintering in north-central Arizona. M.S. thesis, Northern Arizona Univ., Flagstaff, AZ U.S.A. Grubb, T.G. 1984. Winter activity of Bald Eagles {Hahaee- tus leucocephalus) at Navajo Lake, New Mexico. South- west. Nat. 29:335—341. . 1995. Food habits of Bald Eagles breeding in the Arizona desert. Wilson Bull. 107:258-274. . 1996. Wintering Bald Eagle sightings on the Co- conino National Forest, 1975-1996. USDA For. Serv., Rocky Mtn. For. and Rang. Exp. Sta. Rep., Dec. 1996. Flagstaff, AZ U.S.A. and M.A. Coffey. 1982. Evidence of Bald Eagles feeding on freshwater mussels. Wilson Bull. 94:84—85. and C.E. Kennedy. 1982. Bald Eagle winter habi- tat on southwestern national forests. USDA For. Serv., Rocky Mtn. For. and Rang. Exp. Sta. Res. Pap., RM- 237. Fort Collins, CO U.S.A. AND R.G. Lopez. 1997. Ice fishing by wintering Bald Eagles in Arizona. Wilson Bull. 109:546—548. , W.W. Bowerman, AND P.H. Howey. 1994. Track- ing local and seasonal movements of wintering Bald Eagles Haliaeetus leucocephalus from Arizona and Mich- 292 Grubb and Lopez VoL. 34, No. 4 igan with satellite telemetry. Pages 347-358 in B.-U. Meyburg and R.D. Chancellor [Eds.], Raptor conser- vation today. IV World Conference on Birds of Prey and Owls. Pica Press, London, U.K. , S.J. Nagiller, W.L. Eakle, and G.A. Goodwin. 1989. Winter roosting patterns of Bald Eagles {Hal- laeetus leucocephalus) in north-central Arizona. South- west. Nat. 34:453-459. Haywood, D.D. and R.D. Ohmart. 1986. Utilization of benethic-feeding fish by inland breeding Bald Eagles. Condor 88:35-42. Hunt, W.G., J.M. Jenkins, R.E. Jackman, C.G. Theiander, AND A,T. Gerstell, 1992. Foraging ecology of Bald Eagles on a regulated river./. Raptor Res. 26:243-256. Isaacs, F.B., R. Goggans, R.G. Anthony, and T. Bryan. 1993. Habits of Bald Eagles wintering along the Crooked River, Oregon, Northwest Sci. 67:55-62. Mersmann. T.J., D A. Buehler, J.D. Fraser, and J.K.D. Seegar. 1992. Assessing bias in studies of Bald Eagle food habits. J. Wildl. Manage. 56:73-78. Moore, T.D., L.E. Spence, and C.E. Dugnolle. 1974. Identification of the dorsal guard hairs of some mam- mals of Wyoming. Wyoming Dept, of Game and Fish, Laramie, WY U.S.A. Morrall, E. and J. Coons. 1996. Checklist of the birds — Mormon Lake, Arizona, and nearby areas (Lakes Mary and Ashurst, Anderson Mesa) . Northern Arizo- na Audubon Society, Flagstaff, AZ U.S.A. Sherrod, S.K., C.M. White, and F.S.L. Williamson. 1976. Biology of the Bald Eagle on Amchitka Island, Alaska. Living Bird 15:143-182. SPSS, Inc. 1997. SPSS 7.5 for windows: base professional statistics, advanced statistics. SPSS, Chicago, IL U.S.A. Stalmaster, M.V. and R.G. Plettner. 1992. Diets and foraging effectiveness of Bald Eagles during extreme winter weather in Nebraska./. Wildl. Manage. 56:355- 367. Received 3 March 2000; accepted 4 August 2000 J. Raptor Res. 34(4):293-298 © 2000 The Raptor Research Foundation, Inc. NEST FEATURES AND NEST-TREE CHARACTERISTICS OF SHORT-TOED EAGLES ( CIRCAETUS GALLICUS) IN THE DADIA-LEFKIMI-SOUFLI FOREST, NORTHEASTERN GREECE Dimitris E. Bakaloudis^ School of Animal and Microbial Sciences, University of Reading, Whiteknights, P. O. Box 228, Reading, RG6 6AJ, U K Christos G. Vlachos Department of Forestry and Natural Environment, Aristotle University of Thessaloniki, P.O. Box 241, Thessaloniki 54006, Greece Graham J. Holloway School of Animal and Microbial Sciences, University of Reading, Whiteknights, P O. Box 228, Reading, RG6 6AJ, U K Abstract. — Data on nest features and nest-tree characteristics of 29 nest trees of Short-toed Eagles (Circaetus gallicus) were compared with the same number of paired, randomly-selected trees in the Dadia- Lefkimi-Soufli forest complex, northeastern Greece. Short-toed Eagles usually nested in Calabrian pine {Pinus brutia, 83%) trees that were either dominant (87%) or intermediate (13%) in the canopy. Most nests were in the largest trees in terms of height (x = 13.8 ± 0.4 m, ±SE) and diameter at breast height (x = 49.7 ±1.6 cm) in stands. Nests were located in the lower or middle third of the canopy at a mean height of 8.6 ± 0.41 m and on horizontal branches at a mean distance of 133 cm ± 12.4 cm from trunks. A tendency for building nests on the south-facing side of canopies of nest trees was detected (mean angle = 178°, angular deviation s = 58°). Short-toed Eagles selected nest trees that provided them with easy access while also providing protection from predators and inclement weather. Key Words; Short-toed Eagle, Circaetus gallicus; nest features', nest-tree characteristics', Greece. Caracteristicas del nido y de los arboles con nido de Circaetus gallicus en el bosque Dadia-Lefkimi-Soufli en el noreste de Grecia Resumen. — Comparamos los datos sobre las caracteristicas de 29 nidos de Circaetus gallicus con el mismo numero de arboles seleccionados al azar en el complejo de bosques de Dadia-Lefkimi-Soufli en el noreste de Grecia. Circaetus gallicus anida usualmente en arboles de Pinus brutia 83% los cuales fueron dominantes (87%) o intermedio (13%) en el dosel. La mayoria de los nidos se encontraron en los arboles mas grandes en terminos de altura (x = 13.8 ± 0.4 m, ±SE) en los rodales. Los nidos fueron localizados en la parte baja y el tercio medio del dosel a una altura media de 8.6 ± 0.41 m en ramas horizontales a una distancia media de 133 ± 12.4 cm del tronco. Se detecto la tendencia de construir los nidos en el costado sur del dosel (mean angle = 178°, angular deviation s = 58°). Circaetus gallicus selecciono arboles que le proporcionaron un acceso facil, como tambien proteccion de los depredadores y del inclemente clima. [Traduccion de Cesar Marquez] The Short-toed Eagle ( Circaetus gallicus) is a tree- nesting accipitrid (Cramp and Simmons 1980), nesting in a variety of forest types, such as open coniferous forests in France (Thiollay 1968), dense ^ Present address: Aristotle University of Thessaloniki, Department of Forestry and Natural Environment, Lab- oratory of Wildlife and Freshwater Fisheries, P.O. Box 241, Thessaloniki 54006, Greece. evergreen oak and mixed deciduous woodland in central Italy (Petretti 1988), mixed conifer-decid- uous forests in north-western Italy (Bocca 1989), and dry pine forests with mosses {Sphagnum spp.) in the ground layer in Belarus (Ivanovsky 1992). Despite considerable interest in the ecology of Short-toed Eagles in Mediterranean countries, few studies have been conducted to describe the struc- ture of nest trees favored by the species (Petretti 293 294 Bakaloudis et al. VoL. 34, No. 4 1988, Vlachos and Papageorgiou 1994). No previ- ous studies have been attempted to describe and compare actual nest trees with available trees. A detailed analysis of Short-toed Eagle nest trees in the Dadia-Lefkimi-Soufli forest complex was car- ried out to determine the most important char- acteristics determining the choice of nest trees. The aims of this study were to describe nest struc- ture and to evaluate nest-tree characteristics of Short-toed Eagles by comparing actual nest trees with randomly-selected trees. Study Area The study was conducted in the Dadia-Lefkimi-Soufli (D-L-S) forest complex, in the central part of Evros Pre- fecture, northeastern Greece (40°59'-41°15'N, 26°19'- 26°36'E). The region is on the eastern edge of the Ro- dopi mountain chain in western Thrace. Elevations range from 20-700 m and steep-sided valleys crisscross the area. The climate is submediterranean and mean monthly temperatures in the area range from 25°C in July to 4°C in January. Mean annual precipitation is 664 mm. North- erly winds predominate during the year following the north-south orientation of the Evros valley. The structure and composition of the vegetation in the D-L-S forest complex are the result of a combination of climate, soils, and intensive past hnman influence (Dabs 1973). The study area is covered by a mosaic of different habitat types, such as agricultural lands, grasslands, shrublands, rocky areas, pine forests, oak forests, degra- dated oak forests, and mixed pine-oak forests. The main overstory tree species are pines, including Calabrian pine {Pinus brutia) and black pine {P. nigra) . The understory IS mixed with pines, oaks (Quercus conferta, Q. pubescens, Q. sessilifl(yra, and Q. cerris), and various shrub species (Phyllirea media, Arbutus andrachne, Erica arborea, Juniperus oxycedrus, Carpinus orientalis, Ostrya carpinifolia, and Fraxi- nus omus). Approximately 19.5% (7250 ha) of the 37 156 ha study area consists of two core areas that were estab- lished as protected areas for birds of prey in 1980. The study area supports a remarkable diversity of wild- life including Black Vultures (Aegypius monachus) , Griffon Vultures ( Gyps fulvus) , Lesser Spotted Eagles {Aquila po- marina), Imperial Eagles {Aquila heliaca). Booted Eagles (Hieraaetus pennatus), wolves (Canis , jackals {Canis aureus), wild cats {Felis sylvestris), brown hares {Fepus eu- ropaeus), wild boars {Sus scrofa), large whip snakes {Col- uber jugularis), gra.ss snakes {Natrix natrix), dice snakes {Natrix tesselata), nose-horned vipers {Vipera ammodytes), and green lizards {Lacerta viridis) (Bakaloudis et al. 1998). Methods As many occupied Short-toed Eagle territories as pos- sible were located in the stndy area during the 1996-97 field seasons using (a) historical descriptions of tradition- al nesting sites, (b) territorial behaviors of breeding pairs noted from high vantage points, and (c) extensive ex- ploratory surveys on foot (Fuller and Mosher 1987). A total of 29 nests were located in 22 Short-toed Eagle ter- ritories including occupied and old nests. Data on nest- tree characteristics were collected during August and September of 1996-97 after fledging. We recorded the following information to describe each nest tree: tree species, crown class (dominant, intermediate, or sup- pressed), trunk shape (straight, slightly crooked, crook- ed, forked, pitchforked, or without top), and canopy shape (condensed, dense, slack, or light) . The condition of nest trees was described as good, medium (evidence of hre on bark), or bad (both evidence of hre and epi- phytic growth). Nest-tree branches were measured and classified according to their density (1 — >20 branches on the trunk, II — 10-20 branches on the trunk, or III — <10 branches on the trunk) and their size as (thick — >50% of branches with a diameter >12 cm, medium — >50% of branches with a diameter 8-12 cm, or thin — >50% of branches with a diameter <8 cm). Diameter at breast height (dbh) was measured using a dbh tape and the age was determined using an increment core by counting growth rings. Height of nest trees, height of nests above ground, and height to living canopy were measured with a Blumme-Leiss altimeter (accuracy ±0.25 m). Canopy height of nest trees was estimated by subtracting the height of the bottom of the canopy to the ground from the height of the tree. Each nest was assigned to the low- er, middle, or upper third of the canopy. In order to compare nest-tree characteristics within the same forest stand, the same number (29) of nonnest trees were randomly selected from neighboring areas. Each random tree was situated from 70-400 m from nest trees. Three steps were followed to establish each ran- dom tree. First, the area centered on the nest tree was divided into four quadrants (1 = northeast, 2 = south- east, 3 = southwest, and 4 = northwest) and one of these was randomly selected. Secondly, two randomly-selected numbers between 0-400 were selected to calculate the distance of the random point along the north-south axis and the east-west axis. The intersection of lines extending from these points identified the location of the center of the random tree. Finally, the closest dominant tree to this centered point that was similar in dbh with the nest tree was selected and defined as the random nest tree (Titus and Mosher 1981). When random points identified points in nonforested areas such as grasslands, shrub- lands, and/or cultivated areas, or in an area with only young trees, they were rejected and the above procedure was reinitiated. The same measurements were made on the randomly-selected trees as nest trees, apart from var- iables concerning nest characteristics. Nest features were described quantitatively in terms of the distance to the trunk of the tree in cm, the diameter of branches supporting nests against trunks in cm, max- imum and minimum diameter axis of nests in cm, depth of nest cups in cm, and the height of the nest in cm. Nest orientation was determined as the mean flying di- rection to and from the nest, and nest orientation in re- lation to trunk was determined as the angle between a line joining the nest with the trunk and magnetic north. Nest orientation and nest orientation in relation to the tree trirnk were measured with a compass, recording the angle in degrees from magnetic north. All variables were tested for heterogeneity of variances using Bartlett’s test (Zar 1996) and for normality using December 2000 Nests of Short-toed Eagles in Greece 295 Table 1. Characteristics of 29 Short-toed Eagle nests in the Dadia-Lefkimi-Soufli forest complex, northeastern Greece. Variable Mean ± SE Range CV Height of nest above ground (m) 8.67 ± 0.41 5.5-12.25 20.47 Diameter of branch supporting the nest (cm) 12.46 ± 0.91 7.5-26 31.71 Distance from trunk (cm) 133.6 ± 12.4 45-231 38.72 Diameter of nest (cm) max. 61.75 ± 1.42 8-74 10.26 Diameter of nest (cm) min. 48.44 ± 1.06 40-59 9.52 Depth of nest-cup (cm) 7.17 ± 0.43 5-13 26.63 Height of nest (cm) 17.13 ± 0.93 12-26 23.74 Orientation of nest in relation to trunk (°) 178 Orientation of nest (°) 171 Anderson-Darling test. Variables that did not meet the assumptions of homoscedasticity and normality were log- transformed prior to parametric analysis. Normally-dis- tributed variables were analyzed using paired-sample t- tests, but those not meeting normality assumptions after transformation were analyzed using the nonparametric equivalent Wilcoxon matched-pair test. Nominal vari- ables were compared using chi-square analysis. We used Kolmogorov-Smirnov tests to test for uniformity in nest location in nest trees. Variables expressed as percentages were arcsine transformed to standardize variance. Circu- lar variables were analyzed using Rayleigh’s z test for cir- cular uniformity (Batschelet 1981, Zar 1996). All statisti- cal analyses were performed using the Minitab statistical software (version 12) and differences were considered significant with ot = 0.05. Results Three types of Short-toed. Eagle nests were re- corded in the study area according to their position within the canopy of nest trees. Type I nests were located in the lower third of the canopy on large, horizontal forked branches and away from trunks. Type II nests were similar, but were located in the middle third of the canopy, and Type III nests were located on the top of relatively-flat canopies near to trunks and open from above. Nests were not distrib- uted uniformly across the three nest types (Kolmo- gorov-Smirnov test; — 1, P < 0.05) and were more often located in the lower or middle third than the upper third of the canopy. Short-toed Eagles had a tendency to build nests on the south-facing sides of the canopy (mean an- gle = 178°, angular deviation s = 58°, measure of concentration r — 0.49; Table 1), which was signif- icantly different from a uniform distribution (Ray- leigh’s test: z = 6.84, P < 0.001; Fig. la). The po- sition of each nest offered incoming eagles a particular direction of approach to the nest. The mean orientation of nests was also south (mean angle = l7l°, s = 53°, r — 0.57) and the distribu- tion of orientation deviated significantly from ran- dom (Rayleigh’s test: z = 9.34, P< 0.001; Fig. lb). Nests generally were constructed using dead pine twigs, with or without needles, and oak twigs mea- suring 5-15 cm long and 1-3 cm in diameter. Nest cups were lined with green pine needles and green oak leaves. Materials were added to nests by adult eagles during the breeding season until young ea- gles fledged from nests. Nests measured on aver- age 61.7 ± 1.4 X 48.4 ± 1.1 cm {N = 29). Short- toed Eagles tend to build new nests each breeding season. In a sample of 35 nesting attempts in the study area during 1996—97, seven pairs repaired and reused the same nest for two consecutive years and three pairs used the same nest for more than two years (unpubl. data) . Short-toed Eagles nested exclusively in Calabrian (83%) and black (17%) pines. All nest trees were alive and fell into the largest diameter size classes. The structure of nest trees was similar to random nest trees (Table 2). We could not detect a differ- ence between nest tree and randomly-selected trees in terms of their dbh, height, canopy height, or age. Nest trees were either dominant (87%) or intermediate (13%) in the canopy; random trees were all dominant (2X2 contingency test, — 2.071, df — 1, P = 0.150). Nest trees had either slightly crooked (90%) or straight (10%) trunks, which was not significantly different from random trees (93% slightly crooked, 7% straight) (2X2 contingency test, — 0.25, df = 1, P = 0.61). Ad- ditionally, there were no differences between nest trees and randomly-selected trees in canopy shape (2X3 contingency test, x^ = 3.39, df = 2, P = 0.18) or the condition of the trunk (2X3 contin- gency test, x^ = 0.9, df = 2, P — 0.63). However, 296 Bakaloudis et al. VoL. 34, No. 4 (b) N Figure 1. (a) Orientation of Short-toed Eagle nests in relation to nest-tree trunk and (b) orientations of flight path of Short-toed Eagles to and from nests in the Dadia- Lefkimi-Soufli forest complex (N = 29). the majority of nest trees had many (42%) and some (55%) branches, as opposed to random trees which had only some (65%) and few (28%) branches (2X3 contingency test, “ 12.84, df = 2, P = 0.002). Branches were also thicker in nest trees than in random trees (2X3 contingency test, = 9.68, df = 2, P = 0.008). Discussion Short-toed Eagles were found to use mostly Cal- abrian (83%) and black (17%) pines for nesting Table 2. Characteristics of 29 Short-toed Eagle nest trees and 29 randomly-selected mature trees in the Dad- ia-Lefkimi-Soufli forest complex, northeastern Greece {x ± SE). P-value indicates statistical significance of differ- ence between the pairs of means. Variables Nest Tree Random Tree P-Value dbh (cm) 49.7 ± 1.6 47.9 ± 0.6 0.66^ Height (m) 13.8 ± 0.4 14.2 ± 0.4 0.38 Canopy height (m) 5.8 ± 0.3 6.0 ± 0.2 0.38 Age (years) 87.5 ± 3.1 85.3 ± 3.0 0.51 “ Value based on paired sample Wilcoxon test. in our study area. Calabrian pine trees generally have only a few, thick branches and an oval-shaped canopy. Black pines have many more and thinner branches with a relatively flat canopy. These differ- ences may explain the Short-toed Eagle’s prefer- ence for Calabrian pines for nesting. In the same study area, similar findings were also demonstrated by Vlachos and Papageorgiou (1994) who found that 80% and 20% of nests were built on Calabrian and black pines, respectively. In Belarus, Short- toed Eagles nest exclusively in pines (Ivanovsky 1992). In northwestern Italy, they nest in Larix de- cidua and Pinus silvestris (Bocca 1989) and in cen- tral Italy they nest in evergreen oak and deciduous trees (Petretti 1988). All nest trees were dominant with mean height and mean height of canopy of 13.8 and 5.8 m, re- spectively. Nest trees were also mature ranging from 72-135-yr old and they belonged to the high- est diameter size classes with a mean dbh of 49.7 cm. Short-toed Eagles showed a preference for building their nests in trees containing thicker branches and >10 branches per trunk. Trees with this structure probably provide greater shelter from predators and inclement weather while, at the same time, provide support for nests (Solonen 1982). Such trees are the result of a lack of com- petition during early successional stages (Dafis 1990) or from varying intensities of competition experienced over time (Begon et al. 1996). Short-toed Eagles preferred to build their nests on the south-facing side of the canopy of nest trees. We suggest several reasons they do this. First, the strong relationship between nest position in rela- tion to trunk and slope orientation offers a partic- ular direction for adults to access nests (Newton 1979, Sieg and Becker 1990) and provides a favor- able setting for fledglings when they first fly from December 2000 Nests of Short-toed Eagles in Greece 297 nests. Petretti (1988) also noted that 42.8% of Short-toed Eagle nests in Italy were situated on lat- eral branches overlooking steep slopes. Secondly, a preferred orientation for nests has also been ob- served in several raptors and may be related to breeding performance (Vinuela and Sunyer 1992) by providing a favorable environment both for the incubating female and nesdings. The main mete- orological factors that might influence nest orien- tation and reproductive success are temperature early in the breeding season, direct solar radiation during hotter days, and avoidance of other inclem- ent conditions. In D-L-S forest complex, Short-toed Eagles probably gain warmth in the beginning of breeding season when temperatures are still low by situating nests to the south sides of nest trees. Sim- ilarly, Ivanovsky (1992) in Belarus, found that nests were directed towards the south or southeast when they were situated below the top of trees. Mosher and White (1976) in Alaska and Poole and Brom- ley (1988) in the central Canadian Arctic noted the tendency for Golden Eagles {Aquila chrysaetos) to place their nests in a southeasterly direction and Buchanan et al. (1993) have recorded a mean southeasterly direction for Spotted Owl {Strix occi- dentalis) nests in the Cascade Mountains in Wash- ington for Spotted Owl nests. Additionally, Tjern- berg (1983) in Sweden, has mentioned the preference of Golden Eagles to breed on cliffs fac- ing south or southwest. Eighty-six percent of the nests were located with- in the foliage of nest trees on large horizontal branches at a mean distance of 133 cm from trunks. Shadowed by a roof of branches from above, these nests probably provide both conceal- ment from other avian predators (Newton 1979) and direct insulation from the sun. Short-toed Ea- gles build their nests in foliage to protect incubat- ing females, eggs, and nestlings from aerial avian predators (e.g., Eagle Owls {Bubo bubo^. Common Ravens [ Corvus corax^ , and Hooded Carrion Crows [ Corvus corone cornix] ) , which were common in the study area. In addition, branches above nests pro- vide cover from direct solar radiation minimizing thermal stress in newly-hatched nestlings, especial- ly during the hottest days when, in some cases, the temperature rises to 40-43° C. In our study area, the predominant winds are northerly and the mtyority of the storms arrive from the north (Flokas 1990). Overall, the place- ment of nests by birds opposite to prevailing winds and storms may be critical to avoid the most in- clement weather (Colias and Colias 1984) . In the Donana National Park, Spain, where westerly winds prevail. Black Kites {Milvus migrans) build their nests on the leeward sides of tree crowns maximiz- ing the sheltering effect of trees (Vinuela and Sun- yer 1992). At Sagehen Creek in California, where inclement weather is mainly from the south, Amer- ican Kestrels {Falco sparverius) situate their nests to avoid this cold direction (Balgooyen 1976, 1990, Raphael 1985). Similarly, Olsen and Olsen (1989) in Canberra, Australia, noted that only 10.3% of Peregrine Falcon {Falco peregrinus) nests faced southwest where most inclement weather originat- ed. In our study, nest features of Short-toed Eagles agreed with those reported in previous studies. Newton (1979), Petretti (1988), and Vlachos and Papageorgiou (1994) noted the tendency for Short-toed Eagles to build small nests for their body size, and to build a new nest in a different place each year. Petretti (1988) reported a total of 39 nesting attempts, but the same nest was used in two consecutive years twice and for three consec- utive years only once. Ivanovsky (1992) reported that each pair had 1-9 alternate nests, and the dis- tance between them varied from 300—1500 m. Ea- gles in our study differed in that they tended to maintain the same nests for longer periods of time, or to use another nest in close proximity to the first. This may have been due to nest site compe- tition with other raptors. The D-L-S forest complex supports a diverse concentration of raptors which possibly creates strong interspecific competition for nesting sites (Vlachos 1989). Another expla- nation is that Short-toed Eagles are possibly at car- rying capacity in the D-L-S forest complex (Hall- man 1979) and so remain, year after year, at the same nest sites because no other sites are available. We observed Short-toed Eagles frequently car- rying green twigs to their nests, a behavior that has also been noted by Petretti (1988). Newton (1979) reported this habit for other raptors, and many ex- planations including nest sanitation and the main- tenance of optimum humidity have been suggested for this behavior. The most widely accepted expla- nation is that raptors bring green vegetation to their nests to advertise territory occupancy (New- ton 1979). A further explanation is that the contin- ual addition of nesting material increases the size of the nest to accommodate the increasing size and activity of the nestlings, particularly when they be- gin to exercise their wings (Newton 1979). Perhaps 298 Bakaloudis et al. VoL. 34, No. 4 this is more important for Short-toed Eagles be- cause they build very small nests in comparison to their size. Acknowledgments We are grateful to K. Bakaloudis, S. Alatzias, P. Gou- diakas, M. Tsakaldimi, and K. Poirazidis for field assis- tance, and to the Ministry of Agriculture, General Sec- retariat of Forests and Natural Environment, for permission to carry out this study. G.R. Bortolotti and an anonymous reviewer kindly made many valuable recom- mendations for which we are most grateful. This study was funded by the Technical Institute of Athens (‘S. Chlo- rou’ legacy). Literature Cited Bakaloudis, D.E., C.G. Vlachos, and G.J. Holloway. 1998. Habitat use by Short-toed Eagles Circaetus galli- cus and their reptilian prey during the breeding sea- son in Dadia Forest (north-eastern Greece)./. Appl. Ecol. 35:821-828. Balgooyen, T.G. 1976. Behavior and ecology of the American Kestrel {Falco sparverius) in the Sierra Ne- vada of California. Univ. Calif. Publ. Zool. 103:1-87. . 1990. Orientation of American Kestrel nest cavi- ties: revisited. / Raptor Res. 24:27-28. Batschelet, E. 1981. Circular statistics in biology. Aca- demic Press, London, U.K, Begon, M,, J.L. Harper, and C.R. Townsend. 1996. Ecol- ogy: individuals, populations and communities, 3rd Ed. Blackwell Scientific Publications, Oxford, U.K. Bocca, M. 1989. Status del Biancone {Circaetus galHcus), dell’Aquila reale (Aquila chrysaetos)-e del Pellegrino {Falco peregrinus) in Valle d ’Aosta. Boll. Mus. Reg. Sci. Nat. Torino 7:165-183. Buchanan, J.B., L.L. Irwin, and E.L. McCutchen. 1993. Characteristics of Spotted Owl nest trees in the We- natchee National Forest./. Raptor Res. 27:1-7. COLIAS, N.E. AND E.C. CoUAS. 1984. Nest building and bird behavior. Princeton Univ. Press, Princeton, NJ U.S.A. Cramp, S. and K.E.L. Simmons. 1980. Handbook of the birds of Europe, the Middle East and North Africa. II hawks to bustards. Oxford Univ. Press, London, U.K. Dafis, S. 1973. Classification of forest vegetation in Greece. Sci. Bull. Agric. Fore.st. Sch., Thessaloniki 15: 74-91. . 1990. Applied silviculture. Giachoudis-Giapoulis Press, Thessaloniki, Greece. (In Greek). Floras, A. 1990. Climatology and meteorology. Univ. Stu- dio Press, Thessaloniki, Greece. (In Greek). Fuller, M.R. and J.A. Mosher. 1987. Raptor survey tech- niques. Pages 37-65 in B.A. Giron Pendleton, B.A. Millsap, K.W. Cline and D.M. Bird [Eds.], Raptor management techiques manual. Natl. Wildl. Fed., Sci. Tech. Sen No. 10. Washington, DC U.S.A. Hallman, B.C.G. 1979. Guidelines for the conservation of birds of prey in Evros. lUCN/WWF Rep. 1684, lUCN/WWF, London, U.K. Ivanovsky, V.V. 1992. Short-toed Eagle in Belarus: cur- rent status and ecology. Pages 69-77 in E.N. Kuro- chkin [Ed.], Modern ornithology 1991. Nauka Pub- lishers, Moscow, Russia. Mosher, J.A. and C.M. White. 1976. Directional expo- sure of Golden Eagle nests. Can. Field-Nat. 90:356-359. Newton, I. 1979. Population ecology of raptors. T. & A.D. Poyser, London, U.K. Olsen, P.D. and J. Olsen. 1989. Breeding of the Pere- grine Falcon Falco peregrinus: III. Weather, nest quality and breeding success. Fmu 89:6-14. Petretti, F. 1988. Notes on the behavior and ecology of the Short-toed Eagle in Italy. Gerfaut 78:261-286. Poole, K.G. and R.G. Bromley. 1988. Interrelationships within a raptorguild in the Central Canadian Arctic. Can. J. Zool. 66:2275-2282. Raphael, M.G. 1985. Orientation of American Kestrel nest cavities and nest trees. Condor 87:457-458. SiEG, C.H. and D.M. Becker. 1990. Nest-habitat selected by Merlins in southeastern Montana. Condor 92:688- 694. SOLONEN, T. 1982. Nest sites of the Common Buzzard {Buteo buteo) in Finland. Ornis Fenn. 59:191-192. Thiollay, J.M. 1968. Essai sur les rapaces du Midi de la France: distribution, ecologie. Circaete Jean-le-Blank {Circaetus gallicus). Alauda 36:179-189. Titus, K. andJ.A. Mosher. 1981. Nest-site habitat select- ed by woodland hawks in the central Appalachians. Auk 98:270-281. Tjernberg, M. 1983. Habitat and nest site features of Golden Eagle Aquila chrysaetos (L.), in Sweden. Swed. Wildl. Res. 12:131-163. ViNUELA, J. AND C. SuNYER. 1992. Nest Orientation and hatching success of Black Kites Milvus migrans in Spain. Ibis 134:340-345. Vlachos, C.G. 1989. The ecology of the Lesser-spotted Eagle {Aquila pomarina) in Dadia forest, Thrace, Greece. Ph.D. dissertation, Aristotle Univ., Thessalon- iki, Greece. (In Greek). AND N.K. Papageorgiou. 1994. Diet, breeding succe.ss, and nest-site selection of the Short-toed Eagle {Circaetus gallicus) in north-eastern Greece./. Raptor Res. 28:39-42. Zar, J.H. 1996. Biostatistical analysis, 3rd Ed. Prentice Hall Inc., London, U.K. Received 30 October 1999; accepted 31 July 2000 J. Raptor Res. 34(4) :299-304 © 2000 The Raptor Research Foundation, Inc. ARE NORTHERN SAW-WHET OWLS NOMADIC? Jeffrey S. Marks Montana Cooperative Wildlife Research Unit, University of Montana, Missoula, MT 59812 US. A. John H. Doremus us. Bureau of Land Management, Boise District, 5948 Development Avenue, Boise, ID 83705 U.S.A. Abstract. — The first known nesting of a Northern Saw-whet Owl {Aegolius acadicus) in the Snake River Birds of Prey National Conservation Area occurred in a nest box in 1986, 4 yr after nest boxes were constructed in the study area. Occupancy of nest boxes by Northern Saw-whet Owls varied substantially over the next 13 yr (0-8 nests per yr). The number of mice counted on nocturnal surveys fluctuated widely during this same period, and the annual number of Northern Saw-whet Owl nests in the boxes was positively correlated with an index of mouse abundance. Only one of the 52 breeding adults that we banded was recaptured in a subsequent year, and none of the 139 nestlings produced in the boxes was reencountered. A male that we banded at a nest in April 1990 was found dead in British Columbia in January 1993, more than 900 km NNW of our study area. Data from the Bird Banding Laboratory were insufficient to evaluate breeding-site fidelity because few researchers have banded adult Northern Saw-whet Owls at nests. Northern Saw-whet Owls seem to exhibit some of the characteristics associated with nomadism in birds (e.g., high fecundity and low survival), but they differ from typical nomadic species because they do not specialize on cyclic prey. We suggest that Northern Saw-whet Owls are nomadic in some parts of their range, settling to breed in areas of high food abundance that they encounter during the nonbreeding season. Key Words: Northern Saw-whet Owl, Aegolius acadicus; nomadism] breeding biology, Idaho. Es Aegolius acadicus una especie nomada? Resumen. — El primer registro de un Aegolius acadicus en anidacion en el area de conservacion del Snake River ocurrio en una caja de anidacion en 1986, 4 anos despues que las cajas fiieron construidas en el area de estudio. La ocupacion de las cajas por los buhos vario substancialmente en los proximos 13 anos. (0-8 nidos por ano). El numero de ratones contabilizados de noche fluctuo ampliamente durante este mismo perfodo y el numero anual de nidos de buhos en cajas fue positivamente correlacionado con el indice de abundancia de ratones. Solo uno de los 52 adultos en reproduccion anillado, fue recapturado y ninguno de los 139 pichones producidos en las cajas de anidacion fue reencontrado. Un macho que fue anillado en el nido en abril de 1990 fue encontrado muerto en Columbia Britanica en enero de 1993, a mas de 900 km NNW, del area de estudio. Los datos del laboratorio de anillacion fueron insuficientes para evaluar la fidelidad al territorio de reproduccion debido a que pocos investi- gadores han anillado adultos de Aegolius acadicus en los nidos. Aegolius acadicus parece mostrar algunas de las caracteristicas asociadas con el nomadismo en las aves (e.g., alta fecundidad y baja sobrevivencia) , pero difieren de las especies nomadas tipicas debido a que no se especializan en una presa ciclica. Sugerimos que los buhos son nomadas en parte de su rango, estableciendose para reproducirse en areas de alta abundancia de comida que encuentran durante la estacion no reproductiva. [Traduccion de Cesar Marquez] The raptors nesting in the Snake River Birds of Prey National Conservation Area (NGA) in south- western Idaho have heen under intensive study since the mid-1970s (USDI 1979). Through 1985, six species of owls were known to nest in the area; Barn Owl {Tyto alba), Western Screech-Owl {Otus kennicottii) , Great Horned Owl {Bubo virginianus) , Burrowing Owl {Athene cunicularia) , Long-eared Owl {Asia otus) , and Short-eared Owl {A. flammeus) . The first three species are permanent residents in the NCA, the Burrowing Owl is a typical migrant, and the two species of Asio are year-round residents in some years and migrants in others (J.S. Marks andJ.H. Doremus pers. ohs.). 299 300 Marks and Doremus VoL. 34, No. 4 The first known nest of a Northern Saw-whet Owl {Aegolius acadicus) in the NCA occurred in a nest box in 1986, 4 yr after boxes were constructed in the study area. In marked contrast to the six species of owls that occur regularly in the NCA, the presence of nesting Northern Saw-whet Owls var- ied substantially over the next 13 yr (0-8 nests per yr). This variation in numbers, combined with a nearly complete lack of recaptures of adults, led us to speculate that Northern Saw-whet Owls are no- madic in the NCA. Many species of birds exhibit strong breeding- site fidelity, remaining in or returning to the same breeding places year after year (Andersson 1980) . Site fidelity may be adaptive because it allows in- dividuals to learn the best places to feed, nest, and avoid predators, which in turn may enhance the ability of territory holders to attract mates (Hinde 1956, Greenwood and Harvey 1976, Greenwood 1980). Nomadism (i.e., lack of breeding-site fidel- ity) in birds is much less common than site fidelity and tends to be restricted to species that feed on cyclic prey or for which environmental variation (e.g., periodic rains) results in large fluctuations in the availability of suitable breeding habitat (An- dersson 1980, 1981). The classic examples of nomadism in birds come from boreal seedeaters such as finches and cross- bills that move large distances in response to changing availability of beechmast and conifer seeds (Newton 1972). In owls, nomadism is best known in vole specialists such as Boreal Owls {Ae- golius funereus), Long-eared Owls, and Short-eared Owls (Wallin and Andersson 1981, Village 1987, Korpimaki and Norrdahl 1991), although docu- mentation of the same individuals breeding in vast- ly different locations is rare. Several female Boreal Owls have been captured at nests more than 500 km apart in different years (Wallin and Andersson 1981, Korpimaki et al. 1987), but the extent of no- madism in this species varies widely among popu- lations, and individuals may remain on the same home ranges for two or more years in succession (Korpimaki et al. 1987, Hayward et al. 1993). In short, compared with “classic” nomadism exhibit- ed by boreal seedeaters, the owl species noted above seem to be “periodically” nomadic. The Northern Saw-whet Owl is common in for- ested habitats across the northern United States and southern Canada. Despite its abundance, rel- atively little is known about its breeding biology (Cannings 1993). In this paper, we present data to suggest that Northern Saw-whet Owls are nomadic in southwestern Idaho. If we are correct, then our study birds would constitute the only known ex- ample of nomadism in a strigid that does not spe- cialize on cyclic prey. Study Area and Methods We studied breeding Northern Saw-whet Owls in the NCA in southwestern Idaho from 1986-99. The NCA is a shrubsteppe desert dominated by big sagebrush {Arte- misia tridentata) . Compared with typical breeding habitat for Northern Saw-whet Owls (i.e., coniferous forest), trees are scarce and are confined to riparian areas and human settlements. All of the owls that we studied bred in nest boxes placed in willows {Salix spp.) , Russian olives {Elaeagnus angustifolia) , and black locusts {Robinia pseu- doacacia). We set out boxes in pairs (1-40 m apart) be- ginning in 1981. Since 1986, when Northern Saw-whet Owls first nested in one of our boxes, two or more boxes have been available at 25-47 sites each year along 95 lin- ear km of the Snake River and its tributaries. To assess the availability of small mammals, we con- ducted nocturnal surveys in the NCA each spring from 1984—94 (see Marks et al. 1989). With the aid of a spot- light, observers in a slowly moving vehicle counted all “mice” seen along 547-709 km of secondary roads. Based on these surveys, we calculated an index of mouse numbers by dividing the total number of mice seen by the total length of roads surveyed each year. No surveys were conducted after 1994. We captured breeding female owls in nest boxes dur- ing the brood-rearing period and caught males at night in mist nets set in front of the boxes. We determined the sex of adults by the presence (females) or absence (males) of an incubation patch. Adults were weighed, measured, and banded at first capture, and nestlings were banded about a week before they left the nest. We used partial correlation analysis to assess the rela- tionship between the mouse index (log transformed) and the number of owl nests in the boxes and to see whether the number of Northern Saw-whet Owl nests was correlated with the number of Western Screech-Owl nests. All tests were one-tailed because we predicted that numbers of nesting owls of both species would be posi- tively correlated with the mouse index. Similarly, we pre- dicted that the number of nesting Northern Saw-whet Owls would be negatively correlated with the number of nesting Western Screech-Owls in the boxes because West- ern Screech-Owls are permanent residents in the NCA, and their body mass is two to three times that of North- ern Saw-whet Owls (J.S. Marks and J.H. Doremus unpubl. data) . Results The number of Northern Saw-whet Owl nests in our boxes varied considerably among years, rang- ing from zero to eight {x = 2.7 ± 3.09 [±SD]) (Table 1 ) . In contrast. Western Screech-Owls nest- ed in the boxes every year (range 4-13), and the number of nests per year was larger {x — 8.8 ± December 2000 Nomadism in Northern Saw-whet Owls 301 Table 1. Number of Northern Saw-whet Owl and West- ern Screech-Owl nests in boxes in the Snake River Birds of Prey National Conservation Area, 1984-99. Pairs of nest boxes were available at 25-47 sites each year. The number of adult Northern Saw-whet Owls captured at nests each year is in parentheses. Year Northern Saw-whet Owl Western Screech- Owl 1984 0 7 1985 0 5 1986 1 5 1987 7 (11) 8 1988 0 4 1989 0 6 1990 8 (11) 9 1991 6 (10) 11 1992 7 (5) 13 1993 1 13 1994 0 8 1995 5 (6) 13 1996 3 (4) 11 1997 0 8 1998 0 9 1999 5 (5) 10 2.93) and less variable than that of Northern Saw- whet Owls (Table 1). Rather than the negative re- lationship that we expected, the number of North- ern Saw-whet Owl nests in the boxes was positively correlated with the number of Western Screech- Owl nests in the boxes each year (partial r = 0.58, N = 16, P — 0.038; Fig. 1). Indeed, in several cases Northern Saw-whet Owls nested in boxes within oc- cupied Western Screech-Owl territories. Thus, the presence of Western Screech-Owls appeared to have no negative effect on yearly fluctuations in the number of Northern Saw-whet Owl nests in the boxes. The number of mice counted during nocturnal surveys also varied substantially from year to year (Table 2) . The years with the highest mouse num- bers tended to coincide with the highest numbers of nesting Northern Saw-whet Owls in the boxes (Tables 1, 2), and the number of Northern Saw- whet Owl nests was positively correlated with the mouse index during the years that we had data on small mammals (partial r = 0.69, N = 11, P = 0.013; Fig. 2a). In contrast, no significant correla- tion existed between the mouse index and the No. of screech-owl nests Figure 1 . Relationship between the number of North- ern Saw-whet Owl nests in boxes and the number of West- ern Screech-Owl nests in boxes. Snake River Birds of Prey National Conservation Area, 1984-99. Only 15 points are shown because two were identical. number of Western Screech-Owl nests in the boxes (partial r = -0.28, N= 11, P = 0.21; Fig. 2b). We caught 52 adult Northern Saw-whet Owls (29 females, 23 males) at the 43 nests that we moni- tored (Table 1). Only one returned to breed in a subsequent year, a female that nested in boxes 360 m apart in 1990 and 1991. A male that bred suc- cessfully in the study area in 1990 was found freshly Table 2. Results of nocturnal spotlight surveys for small mammals in the Snake River Birds of Prey National Con- servation Area, 1984-94 (no surveys were conducted af- ter 1994). Year Mice^* Survey Length (km) MiCE/km 1984 2 547 0.004 1985 30 547 0.055 1986 4 547 0.007 1987 293 612 0.479 1988 20 612 0.033 1989 43 709 0.061 1990 146 700 0.208 1991 24 604 0.040 1992 47 603 0.078 1993 15 660 0.023 1 1 21 659 0.032 ® Total number counted on surveys each year. “Mice” includes Perognathus parvus, Onychomys leucogaster, Peromyscus maniculatus, Reithrodontomys megalotis, and Microtus montanus. 302 Marks and Doremus VoL. 34, No. 4 14 -| • • (b) (A CO 0) c 12 - •

200 prey items per study (Table 1). Most of these studies used the analysis of pellets from nestlings and adults to determine the diet which seems to accurately reflect overall owl diet. Nu- merical and biomass contribution were reported since frequency is important when looking at the number of competitors removed and biomass reflects the energetic yield of each taxonomic group. Percent biomass of each taxonomic group in the diet were calculated following Hiraldo et al. (1975a), Real et al. (1985), and Perrins (1987). A value of 500 g was assigned to each rabbit, as Eagle Owls actively select young and subadult rabbits (Donazar and Ceballos 1989). Falconiformes are active diurnally, whereas Strigifor- mes are mainly nocturnal, and frequently share foraging habitats with Eagle Owls. Moreover, nocturnal raptors can be potentially detected by Eagle Owls through their vocalizations. Thus, for statistical analysis, Falconiformes and Strigiformes were considered as two separate groups Results Falconiformes and Strigiformes comprised 97 (0.55%) and 223 (1.27%) out of the 17 557 prey items identified. Biomass frequencies were 0.7 and 1.0%, respectively. This included the depredation of 10 species of Falconiformes and six of Strigifor- mes (Table 2), although at least four other species of Falconiformes (Egyptian Vulture [Neophron perc- nopterus], Bonelli’s Eagle [Hieraaetus fasciatus], Booted Eagle [Hieraaetus pennatus^, and Red Kite [Milvus milvus'\) have also been reported to be prey of Eagle Owls in Mediterranean ecosystems of Europe (Perez-Chiscano 1974, Real and Manosa 1990, Telia and Manosa 1993). European Kestrels {Falco tinnunculus) were the most frequently taken Falconiform, while Little {Athene noctua) and Barn {Tyto alba) Owls were the most commonly taken Strigiforms (Table 2). Frequency of occurrence of diurnal raptors and rabbits was negatively related (r, = —0.61, N — 19, P — 0.006), but this trend was not significant for nocturnal raptors (r, = —0.39, N = 19, P = 0.103). The contribution of raptor and rabbit biomass to the owl diet was negatively related when analyzed separately (Falconiformes: r, = —0.65, P = 0.002; Strigiformes: r^ = —0.55, N = 19, P — 0.014). These results could have been due to the high biomass December 2000 Raptors in Diet of Mediterranean Eagle Owls 307 Table 1. Numerical (N) and Biomass (B) frequencies of Falconiformes (Falc), Strigiformes (Stri), and European rabbits (Rabb) in 19 populations of Mediterranean Eagle Owls. Sample sizes in each population is given (N). Locality Falc Stri Rabb N Reference N B N B N B Bardenas 0 0 3.7 2.0 64.1 85.1 245 P Navarra E 0.4 0.7 3.3 2.6 22.7 46.1 958 P Navarra W 0.3 1.6 1.6 2.5 10.2 26.9 1355 1^ Ebro N 0.4 1.6 0.9 2.9 5.2 31.3 2141 2a Ebro S 0.8 1.2 1.9 1.3 33.5 65.5 1529 2a Murcia 1.4 0.8 1.5 0.6 53.6 66.0 1398 3^ Toledo 1 0.1 0.3 0 0 79 77.5 829 qb Villuercas 0.3 0.9 0 0 42.4 57.4 361 4b Malaga 0 0 0 0 61.3 77.6 256 5*’ S. Morena 0.2 0.2 0.2 0.1 67.9 76.1 1590 5^ Extremadura 0.2 0.6 0.5 0.4 41.2 53.2 417 5^^ Toledo 2 0 0 0.7 0.3 77.1 83.6 266 5’= Salamanca 0.7 1.2 0.1 0.1 25.8 43.3 732 5b Massif Central 3.2 4.2 6 4.7 15.3 19.0 216 6b Tarn 1 0.6 1.5 1.0 22.8 25.4 595 7b Herault 0.5 1.0 1.4 1.2 26.5 39.2 623 7b Provence 1 0.6 0.4 1.4 1.4 19 31.2 2923 8b Valles 0.3 0.4 0.1 0.3 22.1 34.5 724 9b Provence 2 0.2 0.2 1.5 0.9 35.8 45.9 399 10b I Don^ar 1989; 2: Serrano 1998; 3: Martinez et al. 1992; 4: Perez-Mellado 1980; 5: Hiraldo et al. 1975b; 6: Choussy 1971; 7: Cugnasse 1983; 8: Orsini, 1985; 9: Real et al. 1985; 10: Blondel and Badan 1976. “ Pellet remains from adults. Pellet remains from adults and nestlings. Table 2. Number of raptors (N) and numerical (%N) and biomass (%B) frequencies of each species of raptor taken by Eagle Owls in 19 Mediterranean populations. Species competing with Eagle Owls for rabbits are shown (*). Species N %N %B Falco tinnunculus 64 20.0 13.2 Falco peregtinus 1 0.3 1.0 Accipiter nisus 4 1.3 0.7 Accipiter gentilis (*) 2 0.6 2.4 Circus aeruginosus (*) 1 0.3 0.6 Circus pygargus 4 1.3 1.2 Buteo buteo (*) 10 3.1 9.8 Milvus migrans n 5 1.6 5.1 Pernis apivorus 1 0.3 0.9 Circaetus gallicus 1 0.3 2.1 Unidentified Falconiforms 4 1.3 4.1 Tyto alba 76 23.7 23.4 Strix aluco 28 8.7 12.9 Athene noctua 88 27.5 15.4 Asia otus 23 7.2 6.5 Otus scops 8 2.5 0.7 Total 320 100 100 308 Serrano VoL. 34, No. 4 % biomass contribution of lagomorphs % biomass contribution of lagomorphs Figure 1. Percent biomass of lagomorphs in relation to percent biomass of Falconiformes (A) and Strigiformes (B) in the diet of 19 populations of Eagle Owls {Bubo bubo) in western Palearctic. The symbol * shows the population of Massif Central. of raptors in the study from the Massif Central (Fig. 1). Removing this sample from the analysis reduced the strength of the relationships, but they were still significant (Falconiformes: = —0.59, P — 0.009; Strigiformes: r, = -0.47, N = 19, P = 0.049). Only four species of Falconiformes in these studies may compete with Eagle Owls for rabbits (Table 2; see del Hoyo et al. 1994). After removing the rest from the analysis, the relationship was not maintained when raptor and rabbit contribution December 2000 Raptors in Diet of Mediterranean Eagle Owls 309 was analyzed (numerical frequencies: = —0.17, N = 19, P — 0.484; biomass frequencies: r, — -0.32, A^= 19, P = 0.181). Discussion My results suggest a close relationship between rabbits, the main prey, and raptors in the diet of Eagle Owls. This agreed with Telia and Mahosa (1993) who found that Eagle Owls seemed to take a larger proportion of raptors when rabbits were scarce. The weak relationship between rabbits and raptors competing for food with Eagle Owls sug- gested a food-searching rather than a competitor- removal process. Moreover, the fact that the spe- cies most frequently taken seldom compete for trophic resources with Eagle Owls supported this hypothesis. Rabbits are social mammals inhabiting flat or gently undulated Mediterranean scrub habitats of the western Palearctic. In areas of high density of rabbits. Eagle Owls seem to concentrate their hunt- ing effort in distinct perches around burrows. In contrast, other types of habitat are exploited in ar- eas with low rabbit densities (Serrano 1998). Eagle Owls discriminate prey by size rather than by a tax- onomic criterion, and are capable of killing most Mediterranean raptors. Thus, when rabbits are scarce, Eagle Owls probably search for alternative prey more frequently in habitats used by other rap- tors for nesting or roosting (e.g., cliffs, woodlands, and river groves). Thus, an increase of raptors in the diet is related to diet diversification as a con- sequence of low rabbit abundance (see Hiraldo et al. 1976, Donazar et al. 1989, Serrano 1998 for changes in trophic diversity according to rabbit availability) . Rabbits constitute one of the staple food sources for the top predator assemblage of Mediterranean ecosystems in western Europe (Delibes and Hiral- do 1981). Thus, the decline of European rabbits in recent decades as a consequence of human-in- duced epizootic diseases (first myxomatosis, see Delibes and Hiraldo 1981, and then viral haemor- rhagic disease, Villafuerte et al. 1995) has been sug- gested to affect some threatened species of raptors (e.g., Eernandez 1993, Garza and Arroyo 1996, Gonzalez 1996, Villafuerte et al. 1998, but see On- tiveros and Pleguezuelos 2000) . Regardless of the low incidence of raptors in the overall diet of Eagle Owls, my results indicated that low rahbit popula- tions could be influencing the structure of Medi- terranean raptor communities and the conserva- tion of endangered sympatric raptors. In this sense, preliminary results of an ongoing research carried out in eight study areas in the Italian pre-Alps have highlighted a significant effect of Eagle Owl abun- dance on the pattern of nest dispersion, territory occupation, and productivity of some species in the diurnal raptor assemblage (Sergio et al. 1999a, 1999b). However, to determine the abundance and distribution of other raptors in relation to Eagle Owl diet and rabbit abundance, additional re- search is needed to assess the impact of this top predator on raptor communities. Acknowledgments J.A. Donazar, J.A. Martinez, J.L. Telia, F. Sergio, S.J. Petty, and J. Calzada made useful comments on a previ- ous version of the manuscript. Literature Cited Blondel, J. AND D.O. Badan. 1976. La biologie du Hibou Grand Due en Provence. Nos Oiseaux 33:189—219. Cugnasse, J.M. 1983. Contribution a I’etude de Hibou Grand Due {Bubo bubo) dans le sud Massif Central. Nos Oiseaux 37:117-128. Choussy, F. 1971. Etude d’une population de Grand Due {Bubo bubo) dans le Massif Central. Nos Oiseaux 31 :S7- 56. DEL Hoyo, J., A. Elliott, and J. Sargatal. 1994. Hand- book of the birds of the world. Vol. 2. New World vultures to guineafowl. Lynx Edicions, Barcelona, Spain. Delibes, M. and E. Hiraldo. 1981. The rabbit as prey in the Iberian Mediterranean ecosystem. Pages 614-622 in K. Myers and C.D. Meinnes [Eds.], Proceedings of the world lagomorph conference. Univ. Guelph, Guelph, Ontario, Canada. Donazar, J.A. 1989. Variaciones geograficas y estacion- ales en la alimentacion del Buho Real {Bubo bubo) en Navarra. Ardeola 36:25—39. , F. Hiraldo, M. Delibes, and R.R. Estrella. 1 989. Comparative food habits of the Eagle Owl Bubo bubo and the Great Horned Owl Bubo virginianus in six Pa- learctic and Nearctic biomes. Ornis Scand. 20:298-306. AND O. Ceballos. 1989. Selective predation by Ea- gle Owls Bubo bubo on rabbits Oryctolagus cuni cuius: age and sex preferences. Ornis Scand. 20:117-122. Fernandez, C. 1993. Effect of the viral haemorrhagic pneumonia of the wild rabbit on the diet and breed- ing success of the Golden Eagle Aquila chrysdetos (L ). Rev. Ecol. Terre Vie 48:323-329. Garza, V. and B. Arroyo. 1996. Situacion del Aguila per- dicera {Hieraaetus fasciatus) en Espaha. Pages 219-229 in]. Muntaner and J. Mayol [Eds.], Biology and con- servation of Mediterranean raptors, 1994. Monografia No. 4. SEO-BirdLife, Madrid, Spain. Gonzalez, L.M. 1996. Tendencias poblacionales y estatus 310 Serrano VoL. 34, No. 4 de conservacion del Aguila Imperial Iberica {Aquila adalberti) en Espana durante los ultimos 20 aiios. Pag- es 61-75 mj. Muntaner and J. Mayol [Eds.], Biology and conservation of Mediterranean raptors, 1994. Monografia No. 4. SEO-BirdLife, Madrid, Spain. Hakkarainen, H. and E. Korpimaki. 1996. Competitive and predatory interactions among raptors; an obser- vational and experimental study. Ecology 77:1134— 1142. Hiraldo, R, F. FernAndez, and F. Amores. 1975a. Diet of the Montagu’s Harrier (Circus pygargus) in south- western Spain. Donana Acta Vertebrata 2:25-55. , J. Andrada and F.F. Parreno. 1975b. Diet of the Eagle Owl (Bubo bubo) in Mediterranean Spain. Don- ana Acta Vertebrata 2:161-177. , F.F. Parreno, V. Andrada, and F Amores. 1976. Variations in the food habits of the European Eagle OyA(Bubo bubo). Donana Acta Vertebrata 3:137-156. Jaksic, FM. and C.D. Marti. 1984. Comparative food habits of Bubo owls in Mediterranean-type ecosystems. Condor 86:288-296. Litvaitis, J.A. and R, Villafuerte. 1996. Intraguild pre- dation, mesopredator release, and prey stability. Con- serv. Biol. 10:676-677. Martinez, J.E., M.A. Sanchez, D. Carmona, J.A. Sanchez, A. Ortuno, and R. Martinez. 1992. The ecology and conservation of the Eagle Owl Bubo bubo in Murcia, south-east Spain. Pages 84—88 in C.A, Galbraith, I.R. Taylor, and S. Percival [Eds.], The ecology and con- servation of European owls. U.K. Nature Conserva- tion No. 5. Joint Nature Conservation Committee, Pe- terborough, U.K. Mikkola, H. 1976. Owls killing and killed by other owls and raptors in Europe. Br. Birds 69:144—154. . 1983. Owls of Europe. Poyser, Berkhamsted, U.K. Ontiveros, D. and J.M. Pleguezuelos. 2000. Influence of prey densities in the distribution and breeding suc- cess of Bonelli’s Eagle (Hieraaetus fasdatus) \ manage- ment implications. Biol. Conserv. 93:19-25. Orsini, P. 1985. Le regime alimentaire du Hibou Grand- due Bubo bubo en Provence. Alauda 53:11-28. Palomares, F, P. Gaona, P. Ferreras, and M. Delibes. 1995. Positive effects on game species of top predators by controlling smaller predator populations: an ex- ample with lynx, mongooses, and rabbits. Conserv. Biol. 9:295-305. AND TM. Caro. 1999. Interspecific killing among mammalian carnivores. Am. Nat. 153:492—508. Penteriani, V. 1996. The Eagle Owl. Calderini Edagri- cole, Bologne, Italy. (In Italian) Perez-Chiscano, J.L. 1974. Sumario informe sobre la al- imentacion de rapaces en el nor-este de la provincia de Badajoz. Ardeola 19:331-336. Perez-Mellado, V. 1980. Alimentacion del Buho Real (Bubo buboV.) en Espana Central. Ardeola 25:93-112 Perrins, C. 1987. Aves de Espana y Europa. Omega, Bar- celona, Spain. PoLis, G.A. and R.D. Holt. 1992. Intraguild predation: the dynamics of complex trophic interactions. Tree 7: 151-154. Real, J., A. Galobart, and J. Fernandez. 1985. Estudi preliminar d’una poblacio de Due (Bubo bubo) al Val- les i Bages. Medi Natural del Vallesr.l'7b-\%1 . AND S. Manosa. 1990. Eagle Owl (Bubo bubo) pre- dation on juvenile Bonelli’s Eagles (Hieraaetus fasaa- tus).J. Raptor Res. 24:69—71. Rohner, C. and FI. Doyle. 1992. Food-stressed Great Horned Owl kills adult goshawk: exceptional obser- vation or community process? J. Raptor Res. 26:261- 263. Rudolph, S.G. 1978. Predation ecology of coexisting Great Horned and Barn Owls. Wilson Bull. 90:134- 137. Serrano, D. 1998. Diferencias interhabitat en la alimen- tacion del Buho Real (Bubo bubo) en el valle medio del Ebro (NE Espana) : efecto de la disponibilidad de conejo (Oryctolagus cuniculus). Ardeola 45:47-53. Sergio, F, L. Marchesi, P. Pedrini, and F Rizzolli 1999a. II Gufo reale Bubo bubo come potenziale fattore limitante per tre specie di rapace diurni. Avocetta 23: 113. , L. Marchesi and P. Pedrini. 1999b. Black Kite (Milvus migrans) density and productivity in relation to predation pressure by the Eagle Owl (Bubo bubo) Buteo (Suppl.):30. Tella, J.L. AND S. Manosa. 1993. Eagle Owl predation on Egyptian Vulture and Northern Goshawk: possible effect of a decrease in European rabbit availability./. Raptor Res. 27:111-112. Villafuerte, R., C. Calvete, J.C. Blanco, and J. Lu- CIENTES. 1995. Incidence of viral hemorrhagic disease in wild rabbit populations in Spain. Mammalia 59:651- 659. , J. ViNUELA, AND J.C. Blanco. 1998. Extensive predator persecution caused by population crash in a game species: the case of Red Kites and rabbits in Spain. Biol. Conserv. 84:181-188. Received 3 March 2000; accepted 28 July 2000 J Raptor Res. 34(4) ;31 1-31 8 © 2000 The Raptor Research Foundation, Inc. AN EVALUATION OF METHYL ANTHRANILATE, AMINOACETOPHENONE, AND UNFAMILIAR COLORATION AS FEEDING REPELLENTS TO AMERICAN KESTRELS Michael K. Nicholls Ecology Research Group, Canterbury Christ Church University College, Canterbury, Kent, CTl IQU, UK Oliver P. Love and David M. Bird Avian Science and Conservation Centre, Macdonald Campus of McGill University, 21,111 Lakeshore Road, Ste-Anne-de-Bellevue, Quebec, H9X 3V9, Canada Abstract. — A comparison of methyl anthranilate and 4-aminoacetophenone as feeding repellents to a captive colony of American Kestrels (Falco sparverius) was made to determine whether aversive condi- tioning to these chemicals is possible in a bird of prey species. Our results suggested that, while these chemicals seemed to cause some food rejection by kestrels, they did not deter them from eating treated dead, day-old cockerels. A second study using a combination of chemical additives linked to food dyed an unfamiliar color revealed that color, and not the chemicals, was a more aversive agent. This suggested that manipulation of a kestrel’s visual perception of a prey item alone had potentially more success than conditioning it to avoid a chemical additive. These results may prove useful in practical applications such as protecting game bird young at wild release sites or domestic homing pigeons associated with a particular home loft. These measures may in turn help to protect birds of prey from persecution as competitors for prey of human economic importance. Key Words: American Kestrel, Falco sparverius; conditioned taste aversion; CTA; food choice, appetite suppres- sant, visual perception; aposmatic coloration. Metyl antranilato, aminoacetofen y la coloracion inusual como repelentes alimenticios do: Falco sparverius Resumen. — Una comparacion de metil antranilato y 4 aminoacetofen como repelentes alimenticios de una colonia en cautiverio de Falco sparverius fue utilizada para determinar si un acondicionamiento de aversion a estos quimicos es posible en una especie de ave rapaz. Nuestros resultados sugieren que mientras estos quimicos pudieron haber causado algun tipo de rechazo por los cernicalos, esto no los detuvo de alimentarse de polios muertos de un dia de nacidos. Un segundo estudio utilizando una combinacion de aditivos quimicos ligados a una comida tehida de un color inususal, revelo que el color y no los quimicos obraron mas como agente de aversion. Esto sugirio que la manipulacion de la per- cepcion visual de una presa tuvo potencialmenmte mas exito que el acondicionamiento para evitar los aditivos quimicos. Estos resultados pueden ser utiles en la aplicacion practica como en la proteccion de juveniles de aves de caza en los sitios de liberacion o de palomas mensajeras asociadas a ciertos sitios. Estas medidas pueden a la vez ayudar a proteger a las aves rapaces de la persecucion como competidoras de presas de importancia economica. [Traduccion de Cesar Marquez] Although charismatic and often of high conser- vation priority, birds of prey are regarded as pests when taking prey of human economic interest, such as when Peregrine Falcons {Falco peregrinus) take domestic pigeons {Columba livia) (Ratcliffe 1993), Hen Harriers (Circus cyaneus) and other rap- tors kill Red Grouse (Lagopus lagopus) (Redpath and Thirgood 1997), and Northern Goshawks (Ac- cipiter gentilis) kill Ring-necked Pheasants (Phasi- anus colchicus) (Kenward 1977). Such conflicts of interest have resulted in the illegal killing of birds of prey (Cadbury 1992, Etheridge et al. 1997). Musgrove (1997) has suggested the use of aversive conditioning to chemical deterrents as an accept- able (in the sense of Liss 1997) way of reducing Peregrine Falcon predation on pigeons and his pi- lot studies have shown that methyl anthranilate mixed with food causes vomiting in several falcon species, and so was presumably potentially aversive. Limitations of methyl anthranilate application in 311 312 Nicholls et al. VoL. 34, No. 4 field situations are associated with its volatility and degradation in sunlight (Askham 1992). Isomers of aminoacetophenone appear to be up to 10 times more repellent to European Starlings {Sturnus vul- garis) than does methyl anthranilate (Mason et al. 1991), while the importance of intramolecular hy- drogen bonds in the different isomers suggests greater repellency (Clark and Shah 1991) and low- er vapor pressure (i.e., lower volatility). Thus 4- aminoacetophenone probably combines lower vol- atility and higher repellency when compared with methyl anthranilate (M. Baldwin pers. comm.). The use of chemical repellents to instigate chemical aversion conditioning has been used with varying success in the control of many vertebrate pests (Mason 1997) including numerous avian spe- cies (Belant et al. 1997, Mason and Clark 1997, Clark 1998), Other workers (Reynolds and Nico- laus 1994, Reynolds 1999) have concentrated on the predation deterrent value of conditioned taste aversion (CTA) and report varying degrees of suc- cess in field application. To date, however, largescale replicated trials of aversive conditioning on a bird of prey species re- main untried. Accordingly, this paper reports on assessment of methyl anthranilate and 4-aminoac- etophenone as feeding repellents to American Kes- trels, a useful raptor model (Bird 1982) and also whether aversive conditioning to these chemicals is possible in this species. Methods Thirty-three adult male, captive-bred American Kes- trels at the Avian Science 8c Conservation Centre (ASCC) of McGill University were housed individually in open- fronted, wooden cages (60 X 40 X 48 cm) during April 1998 in an ambient temperature room on a 14 hr/ 10 hr light/ dark regime. A rope perch was attached diagonally across each cage and floors were lined with waxed paper to facilitate daily cleaning. Before the experiment began all birds were examined, weighed, randomly ascribed to cages, and then left to condition for 3 d. Caged kestrels were each fed two day- old cockerel chicks per d at 0900 H each morning and uneaten food was removed at 1600 H. Cockerel chicks are the staple diet used throughout the McGill colony and kestrels fed ad lib normally eat 1-1.5 cockerels/d. After the conditioning period, the kestrels were fasted for 1 d before each experiment was begun. Mason et al. (1991) report that isomers of aminoace- tophenone are at least an order of magnitude more re- pellent to European Starlings than is methyl anthranilate. Therefore, 1% (m/m) 4-aminoacetophenone and 10% (m/ m) methyl anthranilate in 85% ethanol were chosen for investigation. Test food was prepared daily by spraying cockerel chicks with one of these solutions until all the perinatal down was saturated, or with 85% ethanol only in the case of controls. The ethanol was then allowed to evaporate for 1 hr before the chicks were packaged and stored under refrigeration. Treated chicks and controls were visually indistinguishable to human experimenters. However, cockerels were discretely labelled by amputa- tion of distal toes and half of the lower mandible. This label was randomly alternated between test and control cockerels for each d of the food choice experiments. Two variables were scored in the feeding trials. One of these was “first choice” (i.e., the cockerel which a kestrel moved directly to and took hold of from perching). In the test situation this is not necessarily the food item which the kestrel eventually consumed but was consid- ered analogous to a wild kestrel perch-hunting, the spe- cies’ most frequent hunting technique (Bildstein and Collopy 1987, Varland and Klaas 1991). The second var- iable scored was the amount of food eaten by each kestrel between 0900-1600 H each day. It is difficult to measure quantitatively the amounts of cockerels consumed by caged kestrels. After thawing, wa- ter evaporates from partly consumed cockerels, which usually also become contaminated with kestrel feces. Therefore, taking fresh weights of intact cockerels and leftovers is not a reliable way of calculating food con- sumed. However, cockerels are remarkably constant in size; mean fresh mass in this study was 41.04 ± 0.61 g, (CV = 0.87%, N — 30). This remains so for the propor- tions of their body parts. The head and neck is 0.17 of a whole cockerel, eviscerated torso 0.40, yolk sac 0.16, oth- er viscera 0.09, each pectoral limb 0.02, thighs 0.06, and feet 0.01 each. Food consumption could, therefore, be accurately assessed as proportions of “day-old cockerel units.” When feeding, kestrels sometimes first plucked some of the perinatal down from a cockerel and then began eating the head. It is possible that in this way they avoided ingesting chemical additives. Experiment 1. Kestrels were divided randomly into three groups of 11 birds each. No significant differences between groups was found for the body mass of the kes- trels {x = 113.71 ± 0.46 g, N = 33, CV = 6.02%). The first group of 11 was used to test the reaction to food treated with methyl anthranilate, a second the reaction to food treated with 4-aminoacetophenone, and the final control group assessing voluntary food intake. A two-choice experimental procedure (Mason et. al. 1989) was followed for the first two groups. Kestrels were given a choice of one treated (treated with either repel- lent) and one untreated day-old cockerel, each day for 4 d. Each day at 0900 H, the two cockerels were placed in each cage, and within approximately 15 min after food introduction the first choice selection by the kestrels was recorded. At 1600 H, food remains were removed and assessed to determine the amount of food consumed in “day-old cockerel units.” Kestrels in the control group were each fed daily with two control cockerels and food consumption similarly measured. On day 5, all birds were reweighed, given two untreated cockerels each and the opportunity to bathe. Total food consumed was mea- sured for each bird at the usual time. Experiment 2. In nature, predators may learn to avoid unpalatable prey animals which are aposematically (warn- ingly) colored (Mathews 1977, Turner 1977), some even December 2000 Feeding Repellents to American Kestrels 313 Table 1. Numbers of kestrels in methyl anthranilate treated group {N = 11) and 4-aminoacetophenone treated group {N — 11) chosing treated day-old cockerels for days 1-4 (n = 11), P-values refer to significance of a test with df = 1. Treatment Day 1 Day 2 Day 3 Day 4 Number P Number P Number P Number P Methyl anthranilate group With methyl anthranilate 4 1 2 4 Untreated 7 0,104 10 0.007 9 0.035 7 0.336 4-aminoacetophenone group With 4-aminoacetophenone 6 2 6 4 Untreated 5 0.763 9 0.035 5 0.763 7 0.336 possessing an innate ability to avoid certain colors (Lind- strom et al. 1999). The second experiment tested wheth- er aversive conditioning in kestrels is facilitated by linking an unfamiliar color to a potentially aversive chemical. In this experiment, 10% 4-aminoacetophenone, which was 10 times the concentration used in Experiment 1, was used. This higher concentration was chosen in order to give the maximum likelihood of achieving a conditioned response to the chemical additive. Day-old cockerels are usually pale yellow and this color was masked by adding green or blue food dyes to the ethanol mixture sprayed onto them. Kestrels from Experiment 1 were rested and well fed for 1 week in large flight cages. Twenty-two kestrels from the first experiment and 1 1 additional kestrels were then weighed and reintroduced to test cages and ascribed to three random groups as in Experiment 1. Cross-sampling assured that kestrels exposed to a particular chemical in the first experiment were not in the second. After aid fasting, they were offered four day-old cockerels each day for 3 d according to one of the following three regimes: (1) Control group — two cockerels dyed with green (green control) and two dyed with blue (blue control) food coloring; (2) Green + 4-aminoacetophenone group — two cockerels dyed green then treated with 10% 4-aminoacetophenone (green -I- 4-aminoacetophenone) and two dyed blue (untreated blue); (3) Blue + 4-ami- noacetophenone group — two cockerels dyed green (un- treated green group) and two dyed blue then treated with 10% 4-aminoacetophenone (blue + 4-aminoaceto- phenone group). First choice and total food consumed was measured each day as in Experiment 1 and kestrels were re-weighed after 4 d. A further experiment showed that kestrels did not dis- criminate between undyed cockerels and those dyed with yellow food color. Results Experiment 1. Significantly more (x^ test, df = 1) kestrels chose untreated day-old cockerels on day 2 for both the methyl anthranilate (P = 0.007) and 4-aminoacetophenone (P = 0.03) groups and also on day 3 for methyl anthranilate (P — 0.03; Table 1). On day 1 and 4, treated and untreated cockerels were chosen at random and there was no significant differences in the number of kestrels choosing the two types of food. Although it seemed that kestrels ate more un- treated than treated food, treated food is also eat- en, apparendy peaking on day 3 for both treatment groups (Fig. 1). There was no evidence of vomiting caused by ingestion of treated food. What is clear, however, was that total food consumed by kestrels in the treatment groups tracked closely that vol- untarily consumed by the control group. Had the chemical additives deterred kestrels from eating treated food, then it might have been expected that total food consumption by treatment kestrels would have been less than for controls. Analysis of variance showed this to be the case (P = 0.0 for treatment effect, repeated measures analysis of var- iance with df = 2,115) with rank order of food consumed in the sequence: control group > meth- yl anthranilate treatment group > 4~aminoaceto- phenone treatment group. Presumably, the lower rates of food consumption of test groups was caused by kestrels avoiding treat- ed food and so limiting their food intake. Paired t- tests confirmed this to be so (Table 2), however, no significant difference (AN OVA, P = 0.901, df = 2,29) in body mass between the three groups could be detected at the end of the trial on day 5. It seemed, therefore, that although kestrels in the treatment groups limit their food intake, they did not completely avoid treated food (Fig. 1) nor compromise their body reserves. Further compar- ison of post-test (day 5) intake of untreated day- old cockerels showed no difference between treat- ment groups and controls (ANOVA, P = 0.199; df = 2,28; 2 missing data items) . These results togeth- er suggested that although the two test chemicals 314 Nicholls et al. VoL. 34, No. 4 Figure 1. Consumption data for American Kestrels (N = 11) in (a) the methyl anthranilate treated group and (b) the 4-aminoacetophenone treated group compared with voluntary food intake of control groups {N = 11). were at least avoided, 4-aminoacetophenone more so than methyl anthranilate, they were not truly aversive. Aversive chemicals would cause kestrels to avoid a particular food type even at the expense of compromising basal energy requirements and fur- ther cause them to avoid eating that food type even after chemical treatment had ceased. Experiment 2. Only green- and no blue-colored day-old cockerels were consumed during the ex- periment. All but two kestrels chose green cock- erels as their first choice in all groups; the remain- ing two birds chose not to eat at all. The green cockerels eaten by kestrels in the control and blue + 4-aminoacetophenone groups were all untreat- ed; effectively, therefore, this second group acted as a further control. Moreover, kestrels in the December 2000 Feeding EIepellents to American Kestrels 315 Table 2. Summary of a series of paired Wests (each with df = 10) comparing consumpdon of methyl anthranilate treated vs. untreated and 4-aminoacetophenone treated vs. untreated food by American Kestrels. AT = 1 1 kestrels in each treatment group and, where significant differences were found, the mean consumption of untreated food was greater than that of the treated food. Day 1 Day 2 Day 3 Day 4 Treatment t P t P t P t P Methyl anthranilate treated vs. untreated food 33.39 0.0 7.1 0.0 3.78 0.0036 1.04 0.32* 4-aminoacetophenone treated vs. untreated food 3.92 0.003 3.27 0.009 0.65 0.53* 2.99 0.014 * Not significant. green + 4-aminoacetophenone group preferred to eat 4-aminoacetophenohe treated green cockerels rather than untreated blue cockerels (Fig. 2). Al- though there was a trend (control group > blue + 4-aminoacetophenone group > green + 4-ami- noacetophenone group) in the total amount of food consumed, analysis of variance showed this as not significant {P = 0.35, df = 2,29). In all cases, however, food intake was very low, averaging less than 0.5 day-old cockerels per kestrel per d, and, as mentioned, some kestrels refused to eat during the entire experimental period. Although most ■ Control Grp, □ Blue + AAP Grp. B Green AAP Grp, 1 3 3 Di{r «f Expouneid Figure 2. Comparison of total amounts of green-dyed day-old cockerels consumed by American Kestrels in Experi- ment 2. Treatment 1 — control group or kestrels offered untreated green and blue day-old cockerels; treatment 2 — blue -I- 4^aminoacetophenone group or kestrels offered untreated green and blue day-old cockerels and day-old cockerels treated with 4-aminoacetophenone; and treatment 3 — green + 4-aminoacetophenone group or kestrels offered untreated blue and green day-old cockerels and day-old cockerels treated with 4-aminoacetophenone. In no case were blue-dyed day-old cockerels eaten. 316 Nicholls et al. VoL. 34, No. 4 kestrels lost weight during the trial, no significant differences in the amount of weight loss could be found by analysis of variance {P — 0.54, df = 2,29). Because food intake was low throughout and to avoid fatality due to starvation, the experiment was terminated after 3 d. Discussion The use of deterrent feeding chemicals may be loosely divided into those which cause food to be unpalatable, can be detected by the predator ei- ther directly or by associated visual cues and so em- ulate aposematic protection, and those tasteless substances which cause feelings of sickness and so evoke a conditioned taste aversion response to a particular food (Clark 1997, Reynolds 1999). Kestrels in our experiments seemed to be able to discriminate and avoid day-old cockerels treated topically with 4-aminoacetophenone and methyl anthranilate when they were novel. However, in the absence of alternate adequate food, kestrels ate cockerels treated with these chemicals, presumably to maintain their caloric needs. There was, there- fore, no evidence to suggest that kestrels were pre- pared to starve and so compromise body condi- tion. As a contrast, McKay et al. (1999) showed that lasting aversion to dead trout could be conditioned in cormorants {Phalacro corax carbo) fed previously on trout treated with carbochal. Although Mus- grove (1997) showed that methyl anthranilate mixed into chopped meat caused vomiting in large falcons, we found no evidence of such violent re- action in kestrels where methyl anthranilate and 4- aminoacetophenone was applied topically to food. Both chemicals seemed, therefore, to be unpalat- able rather than truly aversive to kestrels and so a conditioned taste aversion response did not seem possible. Interestingly, the kestrels’ aversion to unfamiliar color, particularly blue, was stronger than that to the chemical additives. The test kestrels are famil- iar solely to food of one type, yellow day-old cock- erels. Blue-dyed cockerels were such a deterrent to some kestrels that, rather than eat them, they pre- ferred food treated with 4-aminoacetophenone. Further, although they would eat green-dyed cock- erels, their food intake was low. It seemed, there- fore, that unfamiliar color, and not the chemicals, was a more aversive agent. It may be said that these domestic kestrels feed- ing on dead cockerels are not a proper test of wild circumstances. However, the findings in our study appear to be compatible with previous studies on captive kestrels and analogous to studies on wild kestrels and the innate avoidance of certain colors by other predators (Lindstrom et al. 1999). In lab- oratory studies, Mueller (1987) showed that while American Kestrels developed long-term preferenc- es for particular types of prey, they would still sam- ple novel prey if it was still within the limits of what occurred in nature. He further inferred from the literature on laboratory and field studies that such specihc search images are also formed by free- ranging kestrels and other birds of prey. A more serious consideration is whether preda- tory birds conditioned to avoid dead prey, would transfer that avoidance behavior to live alterna- tives. Whether kestrels conditioned to avoid dead cockerels will transfer this behavior to live prey needs further study. It may, however, be feasible to condition free-living raptors to avoid a potential prey by treating live prey with chemical deterrents. It is reasonable to assume that in a field application with ample alternative prey available, the applica- tion of methyl anthranilate and 4-aminoacetophen- one, or indeed some other proven agent, to a group of potential prey animals may condition a response in the predator causing it to avoid the treated group and hunt elsewhere. However, as our results suggest, if prey abundance is locally limiting and no alternative available, then treatment of prey with methyl anthranilate or 4-aminoaceto- phenone, at least at the concentrations tested nor higher concentrations, is not likely to deter pre- dation. Methyl anthranilate is an oily liquid at nor- mal temperatures and 4-aminoacetophenone a crystalline solid. Both chemicals were tested at what appeared to be maximum possible levels (10% m/m), but consumption of treated food by kestrels still occurred. Higher levels applied to, say, game birds or pigeons in the field may cause im- pairment of feather maintenance, increased time spent preening, and possibly increased vulnerabil- ity to factors such as cold or wet weather. Research into the repercussions of the chemicals applied to the plumage of the prey animals they are meant to protect would have to be undertaken. Perhaps, an alternate solution would be to feed potential prey items chemicals which would render their flesh un- palatable to their avian predators, emulating more closely naturally unpalatable and therefore pro- tected animals. In addition, it may even be more efficient to use visual cues such as sufficiently novel colors, rather than chemicals, to at least initially December 2000 Feeding Repellents to American Kestrels 317 deter raptors from taking prey of human economic importance. However, the degree to which a color remains novel to a particular predator in wild cir- cumstances is a complicated question (see Allen and Clarke 1968). Given that the use of dead baits poses problems of transference of avoidance to live prey, this raises an important question concerning in what situa- tions, if any, such management practices would best be aimed. The chemical protection of wild adult prey would not be feasible due to the diffi- culty and cost of catching them and then applying a sufficient amount of the deterrent chemicals. A more practical application of chemical deterrents may be their use in the protection of game bird young at wild release sites or domestic homing pi- geons associated with a particular home loft. Pro- tection from a single, attentive raptor, in these cir- cumstances, could possibly be insured by the combination of an aversive chemical and associat- ed aposmatic visual cues. Acknowledgments Thanks go to all those people who gave help, advice and support especially A. Musgrove, I. Ritchie, C. Semen- luk, J. Dione, and M. Baldwin. This research was sup- ported in part by a research bursary from Canterbury Christ Church University College. Literature Cited Allen, J.A. and B. Clarke. 1968. Evidence for apostatic selection by wild passerines. Nature 220:501-502. Askham, L.R. 1992. Efficacy of methyl anthranilate as a bird repellent on cherries, blueberries and grapes. Proc. Vertebr. Pest Conf. 15:137-141. Belant, J.L., T.W. Seamans, R.A. Dolbeer, and P.P. Wo- RONECKI. 1997. Evaluation of methyl anthranilate as a woodpecker repellent. Int. J. Pest Manag. 43:59-62. Bildstein, K.L. AND M.W. COLLOPY. 1987. Hunting be- havior of Eurasian {Falco tinnunculus) and American kestrels (Falco sparverius): a review. In D.M. Bird and R. Bowman [Eds.], The ancestral kestrel. Raptor Re- search Report 6, Raptor Research Eoundation, Inc., Macdonald Raptor Research Centre of McGill Uni- versity, Canada. Bird, D.M. 1982. The American Kestrel as a laboratory animal. Aatwrc 299:300-301. Cadbury, J. 1992. This illegal killing must stop: a review of bird persecution and poison abuse. R.S.P.B. Conserv. Rev. 6:28-35. Clark, L.C. 1997. A review of the bird repellent effects of 117 carbocyclic compounds. Pages 343-352 mJ.R. Mason [Ed.], Repellents in wildlife management. USDA, National Wildlife Research Center, Eort Col- lins, CO U.S.A. . 1998. Physiological, ecological and evolutionary bases for the avoidance of chemical irritants in birds. Curr. Ornithol. 14:1-37. AND P.S. Shah. 1991. Non-lethal bird repellents in search of a general model relating repellency to chemical structure./. Wildl. Manage. 55:538—545. Etheridge B., R.W. Summers, and R.E. Green. 1997. The effects of illegal killing and destruction of nests by humans on the population dynamics of the Hen Har- rier Circus cyaneus in Scotland./. Appl. Ecol. 34:1081- 1105. Kenward, R.E. 1977. Problems of predation on released pheasants {Phasianus colchinus) by Goshawks (Acapiter gentilis) in central Sweden. Viltrevy 10:79-112. Lindstrom L., R.V. Alatalo, and J. Mappes. 1999. Reac- tions of hand-reared and wild-caught predators to- ward warningly colored, gregarious and conspicuous prey. Behav. Ecol. 10:317-322. Liss, C.A. 1997. The public is attracted by the use of re- pellents. Pages 429-433 in J.R. Mason [Ed.] , Repel- lents in wildlife management. USDA, National Wild- life Research Center, Fort Collins, CO U.S.A. Mason, J.R. [Ed.]. 1997. Repellents in wildlife manage- ment. USDA, National Wildlife Research Center, Fort Collins, CO U.S.A. , M.L. Avery, J.F. Glahn, D.L. Otis, R. Matteson, and C.O. Nelms. 1989. Evaluation of methyl anthra- nilate and starch-plated methyl anthranilate as bird repellent feed additives./. Wildl. Manage. 55:182-187. , L.C. Clark, and P.S. Shah. 1991, Ortho-ami- noacetophenone repellency to birds: similarities to methyl anthranilate./. Wildl. Manage. 55:334—340. and L. Clark. 1997. Avian repellents: options, modes of action, and economic considerations. Pages 371-391 m J.R. Mason [Ed.], Repellents in wildlife management. USDA, National Wildlife Research Cen- ter, Fort Collins, CO U.S.A. Mathews, E.G. 1977. Signal-based frequency-dependent defence strategies and the evolution of mimicry. Am Nat. 111:213-222. McKay, H., R. Furness, I. Russell, D. Parrott, M. Reh- fisch, G. Watola, j. Packer, E. Gill, and P. Robert- son. 1999. The assessment of the effectiveness of management measures to control damage by fish-eat- ing birds to inland fisheries in England and Wales Report to the Ministry of Agriculture, Fisheries and Food (MAFF Project VCO107). Central Science Lab- oratory, York, U.K. Mueller, H.C. 1987. Prey selection by kestrels: a review. In D.M. Bird and R. Bowman [Eds.], The ancestral kestrel. Raptor Research Report 6, Raptor Research Foundation, Inc., Macdonald Raptor Research Centre of McGill University, Canada. Musgrove, A.J. 1997. Peregrines and pigeons: investiga- tions into raptor human conflict. Ph.D. dissertation, Univ. Bristol, Bristol, U.K. 318 Nicholls et al. VoL. 34, No. 4 Ratcliffe, D.A. 1993. The Peregrine Falcon, 2nd Ed. Poyser, London, U.K. Redpath, S.M. and S.J. Thirgood. 1997. Birds of prey and Red Grouse. Her Majesty’s Stationary Office, London, U.K. Reynolds, J.C. 1999. The potential for exploiting condi- tioned taste aversion (CTA) in wildlife management. In D.P. Cowan and C.J. Feare [Eds.], Advances in ver- tebrate pest management. Filander Verlag, Furth, Germany. AND L. Nicolaus. 1994. Learning to hate game- birds! Game Conserv. Rev. 25:97-99. Turner, J.R.G. 1977. Butterfly mimicry: the genetical evolution of an adaptation. Evol. Biol. 10:163-206. Varland, D.E. and E.E, Klaas. 1991. Development of the foraging behaviour in the American Kestrel. J. Raptor Res. 25:9-17. Received 30 October 1999; accepted 16 July 2000 Short Communications J. Raptor Res. 34(4) :319-321 © 2000 The Raptor Research Foundation, Inc. Responsiveness of Nesting Eurasian Kestrels Falco tinnunculus to Call Playbacks Luca Salvati Piazza F. Morosini 12, 1-00136 Rome, Italy Alberto Manganaro Via di Donna Olimpia 152, 1-00152 Rome, Italy Simone Fattorini Dipartimento di Biologia Animale e delVlIomo (Zoologia), Universitd di Roma “La Sapienza, ” Viale delVUniversita 32, 1-00185 Rome, Italy Key Words: Eurasian Kestrel, Falco tinnunculus; playback, nesting period; survey technique. Many birds of prey, both diurnal and nocturnal, re- spond to playback of their own calls (Fuller and Mosher 1981, Morrell et al. 1991, Cerasoli and Penteriani 1992, Redpath 1994). This responsiveness has lead to the de- velopment of acoustic methods for detecting and count- ing raptors, particularly in woodland habitats (e.g., Mosh- er et al. 1990, Cerasoli and Penteriani 1992). By contrast, diurnal raptors found in open habitats are seldom sur- veyed using taped broadcasts because they are easily de- tected by observers. Nevertheless, difficulties in locating these raptors can arise, especially when there is a mosaic of habitat types across the landscape or when the archi- tectural complexity of old buildings and ruins in urban areas supplies additional nest sites. The aim of this study was to assess the response of breeding pairs of Eurasian Kestrels {Falco tinnunculus) , which can be widespread in European urban areas (Village 1990), to the playback of their calls in order to evaluate the efficiency of this meth- od for surveying for nests of this species. Methods The study was carried out on the Eurasian Kestrel pop- ulation that nests in scaffolding holes of Roman ruins and old buildings in the center of Rome, Italy. This popula- tion has been studied since 1995 and has shown a very high breeding density (Piattella et al. 1999). During April-June 1998, 36 taped broadcast sessions were per- formed at 20 known occupied nests. Nine sessions were performed in April, 16 in May, and 11 in June. To avoid habituation to playback and disturbance during the breeding season, breeding pairs were not tested more than twice (e.g., Redpath 1994) during the entire study period. Playbacks were performed at each occupied nest in early morning (0700-1000 H, N = 13 playback sessions). late morning (1000-1300 H, N = 9), early afternoon (1300-1600 H, N = 7), and late afternoon (1600-1900 H, N = 7) . Signal calls, a series of high-pitched trills es- pecially uttered by females denoting the presence of breeding pairs near nests (Village 1990), were used for eliciting responses of kestrels. Each playback session was conducted giving five taped calls lasting approximately 1 min at 2-min intervals. Taped broadcasts were performed using a portable stereo with 6 W amplifiers. Taped calls were stopped once a bird responded (latency time). Broadcast points were located in the streets around build- ings used by breeding pairs of kestrels, at a minimum distance ranging 30-40 m. During each playback session, each of the following were recorded: the date and time of stimulation, minimum daily temperature and wind speed (values obtained from the meteorological station of “Ufficio Centrale di Ecologia Agraria” in the city cen- ter), latency (time from start of broadcast to first re- sponse) , sex of the responding individual, and response type classified as (1) appearance of a kestrel at the nest entrance, (2) advertising or alarm calls, (Village 1990), (3) flights around the nest site (taking into account only individuals flying away from nest entrances or from perches close to nest holes), (4) copulation near nests, (5) no response, and (6) behavior of young (only for nests completely visible by the observer) . Nearest-neigh- bor distance (nnd) was also calculated including all breeding pairs, even those not tested by playbacks. Breeding pairs were located in the study area (about 10 km^) by visiting known nest sites and checking other suitable nest sites with standard census methods (Village 1990) . Breeding success was determined by checking nest sites at least twice and counting all visible fledglings. Nests where complete counts of fledglings could not be made were not included in the analysis. Percentages of successful playbacks were calculated with reference to the number of playbacks performed in each month and over the study period, respectively. In calculating percentages of successful playbacks, we re- garded as successful each playback that caused any type 319 320 Short Communications VoL. 34, No. 4 (n=13) (n = 9| (n = 7) (n-7) Figure 1. Daily response pattern of breeding Eurasian Kestrels to taped broadcasts. of response by males, females or both together. Also, we calculated separately the percentages of males and fe- males that responded to playbacks in each month for the total playbacks performed in each month. Finally, a “re- sponse index” indicating the intensity of responses was calculated for each pair by summing of all response types of the male and female (giving the value 1 to all positive types of responses) and dividing it by the total number of playbacks. Values of response time and response index recorded for each pair were compared to date, time of stimulation, temperature, wind, nearest-neighbor dis- tance, and breeding success using the Spearman rank correlation tests. Differences between female and male latency were tested by using Mann-Whitney [/-test ( C-val- ue corrected to a z score) . All tests were two-tailed and a = 0.05. Table 1 . Responses of breeding Eurasian Kestrels to call playbacks in Rome, Italy. Total responses, appearances, vocalizations, and flights are expressed for each sex as number of observed events and percentages on both sex- es Percentage of copulations is calculated for the total number of taped broadcast stimulations. Response time, as well as response index, are expressed as x ± SD. Females Males N Response time(s) 17 ± 28 65 ± 108 36 Total responses 18 (58.1%) 13 (41.9%) 31 Appearances 16 (72.7%) 6 (27.3%) 22 Vocalizations 9 (50.0%) 9 (50.0%) 18 Flights 3 (23.1%) 10 (76.9%) 13 Copulations 4 (11.1%) 36 Response index 2.2 ± 1.5 16 Results A total of 26 of 36 playbacks resulted in responses by kestrels. Eight (88.9%) of 9 playbacks performed in April, 10 (62.5%) of 16 playbacks performed in May, and 8 (72.7%) of 11 playbacks performed in June resulted in responses. Four playbacks (44.4%, N = 9) in April, 5 (31.3%, N= 16) in May, and 4 (36.4%, N= 11) in June resulted in responses by male kestrels. Seven playbacks (77.8%, N = 9) in April, 7 (43.8%, N = 16) in May, and 4 (36.4%, N = 11) in June resulted in responses by fe- male kestrels. In some cases, males and females of the same pair responded together to the stimulation (Fig. 1, Table 1). Thus, the same playback could have produced a double response. Likewise, multiple types of behavioral reactions were sometimes elicited by a single stimulation All the individuals responded within 5 min from the start of playbacks and the difference between male and female latency was not significant (z = —1.06, P = 0.29, N= 31). During incubation and brooding, females appeared at nest entrances and called regularly for about 1 min, but rarely flew from nests rapidly reentering nests after this display. Males seldom appeared at nest entrances throughout the study period. When males were inside nest-holes, they showed behaviors similar to those shown by females, appearing at nest entrances and excitedly calling, but rapidly reentering nest cavities. At all nests where young birds were observed, young kestrels never responded to taped calls. Instead, they always hid them- selves in an internal corner of the hole during playbacks. No significant correlations were found between re- sponse time and the following variables: date (r^ = 0.28, P= 0.124, iV= 31), time (r^ = 0.08, P= 0.663, N= 31), temperature (r^ = 0.32, P = 0.076, N = 31), wind (r^ = -0.02, P= 0.921, N= 31), nnd (r^ = -0.13, P= 0.480, December 2000 Short Communications 321 N = 31), and breeding success (% = 0.29, P = 0.205, N = 21). Likewise, no significant correlation was found be- tween these variables and the response index: = —0.17, P = 0.349, = 31 for date; = -0.24, P = 0.184, N = 31 for time; r<^ = —0.18, P = 0.326, = 31 for temper- ature; Tff = 0.16, P = 0.384, A^ = 31 for wind; = 0.04, P = 0.829, A^ = 31 for nnd; or = -0.30, P = 0.185, N = 21 for breeding success. For a total of 20 nesting pairs, occupation of 16 nest sites (80%) was direcdy confirmed by playback stimula- tion. Discussion The broadcasting of taped calls is a useful tool in lo- cating nesting raptors in woodland settings (Fuller and Mosher 1981). The technique used in this study may rep- resent a first time such a technique has been used to detect nesting pairs of nonforest species. We found that, after occupation, both male and female kestrels defend- ed nest sites from neighboring and intruding kestrels. Because of this, vocalizations of breeding kestrels could be easily elicited by broadcasting a taped call, such as the “signal call,” especially in the first stages of the nesting period. Although kestrels are not highly territorial (Vil- lage 1990), their response to playbacks was relatively high compared to other diurnal raptors (Mosher et al. 1990, Cerasoli and Penteriani 1992), indicating that the play- back method may integrate other field techniques in lo- cating breeding pairs of Eurasian Kestrel. The playback method may be particularly useful in high density pop- ulations where observers must check the occupation of two or more neighboring nest sites. As kestrels are very versatile in their choice of nest sites and their identifi- cation can be very difficult (Shrubb 1993), this technique may also be a practical tool in low density situations. For example, in cases where kestrels have a scattered distri- bution, this technique can be used to cover relatively large areas in a short time and it is a faster method of surveying for kestrels when they nest in uncommon sites and habitats (e.g., crow nests in pine plantations). Nev- ertheless, because playback methods are invasive, re- searches should minimize disturbance to the pairs stud- ied by performing playbacks only in the first stage of the breeding season and each nest should not be visited more than twice during the breeding season. Finally, playbacks should not be used in counting fledglings be- cause they do not respond to playbacks and seem to be disturbed when adults respond to calls. Resumen. — La respuesta de parejas de Falco tinnunculus a las vocalizaciones grabadas fue estudiada en una pob- lacion urbana en Roma, Italia. Los Cernicalos respondi- eron con varios despliegues cerca de sus nidos. La efi- ciencia de los “playbacks” y la latencia individual fueron similares entre machos y hembras. No hubo una corre- lacion entre las tasas de respuestas y el clima, o entre las variables de poblacion. Los “playbacks” pueden aumen- tar el numero de tecnicas de campo para la localizacion de parejas en reproduccion. [Traduccion de Cesar Marquez] Acknowledgments We thank V. Penteriani for reviewing and greatly im- proving a first version of the manuscript. We are also grateful to two anonymous referees for their criticism and suggestions. We thank F. Mangianti (Ufficio Centrale di Ecologia Agraria, Roma) for kindly providing meteo- rological data. E. Gizzi made a critical reading of the En- glish language. This research was supported by a grant from the Italian Ministero delFUniversita e della Ricerca Scientifica e Tecnologica (“Variazione geografica e div- ersita a livello di .specie, faune e zoocenosi: cause storiche ed ecologiche”) . Literature Cited Cerasoli, M. and V. Penteriani. 1992. Effectiveness of censusing woodland birds of prey by playback. Avocetta 16:35-39. Fuller, M.R. and J.A. Mosher. 1981. Methods of detect- ing and counting raptors: a review. Pages 235—246 in C.J. Ralph andJ.M. Scott [Eds.], Estimating numbers of terrestrial birds. Stud, Avian Biol. 6. Morrell, T.E., R.H. Yahner, and W.L. Harkness. 1991 Factors affecting detection of Great Florned Owls by using broadcast vocalizations. Wildl. Soc. Bull. 19:481- 488. Mosher, J.A., M.R. Fuller, and M. Kopeny. 1990. Sur- veying woodland raptors by broadcast of conspecific vocalizations./. Field Ornithol. 61:453-461. Piattella, E., L. Salvati, A. Manganaro, and S. Fatto- rini. 1999. Spatial and temporal variations in the diet of the Common Kestrel {Falco tinnunculus) in urban Rome, Italy./. Raptor Res. 33:172-175. Redpath, S.M. 1994. Censusing Tawny Owls Strix alucohy the use of imitation calls. Bird Study 41:192-198. Shrubb, M. 1993. Nest sites in the Kestrel Falco tinnun- culus. Bird Study 40:63-73. Viliv\.ge, a. 1990. The kestrel. Poyser, London, U.K. Received 2 December 1999; accepted 22 July 2000 J Raptor Res. 34(4):322-326 © 2000 The Raptor Research Foundation, Inc. The Breeding Success of Tawny Owi.s {Strix aluco) in a Mediterranean Area: A Long-Term Study in Urban Rome Lamberto Ranazzi Dipartimento di Medicina Spmmentale e Patologia, Policlinico “Umberto /” Universitd di Roma “La Sapienza” Viale Regina Elena 324, I-OOl 61 Rome, Italy Alberto Manganaro Via di Donna Olimpia 152, L00152 Rome, Italy Luca Salvati Piazza F. Morosini 12, 1-00136 Rome, Italy Key Words: Tawny Owl; Strix aluco; breeding success; an- nual fluctuations; Mediterranean areas; urban Rome. The breeding biology of the Tawny Owl (Strix aluco) was studied in northern and central Europe specihcally focusing on the influence of food abundance on clutch size and productivity of young (Southern 1970, Wend- land 1984). Baudvin (1990) found a remarkable positive correlation between the reproductive output of Tawny Owls and the percentage of woodland rodents in the diet of pairs in central France. Annual fluctuations in Tawny Owl breeding success were directly linked to the abun- dance of woodland rodents, e.specially the yellow-necked mouse (Apodemus flavicollis) , the main prey of this owl in woods and mixed farmlands (Wendland 1984, Baudvin 1990, Jedrzejewski etal. 1994). Moreover, alternative prey (e.g., birds and amphibians) increased in diet in low mouse years (Plesnik and Dusik 1 994) . Cyclic fluctuations in populations of rodent prey is probably the main factor affecting Tawny Owl productivity, but other factors, such as weather conditions, could be also involved (Kostrzewa and Kostrzewa 1990, Gil-Delgado et al. 1995, Penteriani 1997). In Mediterranean areas, very few studies have focused on the study the annual variations in the breeding suc- cess of any raptor (Gil-Delgado et al. 1995). The aim of this study was to assess the long-term breeding success of Tawny Owls in a Mediterranean urban area, checking for variations, if any, in productivity of young, and comparing them to the breeding performance of other areas in Eu- rope. Methods The study was carried out from 1984-99 in hve urban census plots, including developed areas, small gardens, and city parks (mean density of Tawny Owl territories = 3 0/km^) and from 1989-99 in three suburban plots of Rome, including open land and deciduous woodland patches (mean density of Tawny Owl territories = 5.6/ km^, Ranazzi et al. 2000). Vegetation in small gardens included pines (Pinus pinea ) , cypresses ( Cupressus semper- virens), cedars (Cedrus spp.), as well as isolated oaks (Quer- cus spp.). Vegetation of city parks as well as suburban woodlands was generally composed of strands of mixed deciduous wood predominated by oaks (e.g., Quercus ilex). Nests were generally located in natural cavities of old oaks and pines. The rate of territory occupation was remarkably high in both habitats, so the population den- sity did not show significant variations among years (Ran- azzi et al. 2000). Procedures for mapping territories and locating nest- ing sites followed Ranazzi et al. (2000). Although the bulk of the data on breeding success were obtained from pairs consecutively studied throughout the census peri- od, some pairs, especially those of urban parks, were not continuously censused due to the impossibility to visit their territories in some breeding seasons (e.g., occur- rence of summer events or public works in parks and gardens and Tawny Owls nesting sometimes on private property). Data for 1984-85 seasons were not considered due to the small number of records available. The num- ber of occupied territories censused each year was 14.3 ± 5.9 (N = 200) from 1986-99 in urban plots, and 10.1 ± 5.8 (N = 111) from 1989-99 in suburban plots. Estimates of the number of young in nests were made by broadcasting calls of male Tawny Owls on a SANYO portable stereo with 6 W loudspeakers within the nesting area at a distance of about 50 m from known nest sites (see Ranazzi et al. 2000 for details) and listening for re- sponses. Generally all young responded to calls with their persistent ‘ptzie’ begging calls, so this method was used to evaluate the number of successful pairs and fledgling production (Wendland 1984). Nest site disturbance was reduced by limiting visits to each territory to only two in May-August. This period was chosen to census young based on a preliminary assessment in 1984-85 of young vocal activity at eight known nests. All fledglings began to utter the ‘pt^i^ call in May and they remained in their parents’ territories at least until early August, continu- ously uttering their begging calls. The.se data were con- sistent to those in non-Mediterranean zones (Southern 1970, Wendland 1984). Data gathered in early May or in August were included only if a control visit was made in June or July. When we were uncertain of the exact num- ber of young that begged due to the many calls contem- poraneously uttered, we omitted them from analyses. We agree with Wendland (1984) that this method can 322 December 2000 Short Communications 323 3’ "O fl I o r- cn C D p: o c/3 O cn cn PQ o f-H 00 1-H rr o »-H to 00 CM i-H oo CO oo u u U cfl 1 1 1 Gl G 1 1 CM O 1-H 50 CO 50 CO O to o 50 d 7 d rr- d d H rH 1-H 1-H od +1 lA +1 1 + 1 J + 1 1 + 1 1 05 o o o oo CM CM CM oo H hQH ID q 00 d 7 d 05 d I> q 1-H o O ^ d d CM u d 7 d *7 d o d 9 d q CM CM CM CM + 1 J + 1 1 +1 1 + 1 + 1 1 J> o iCl o o CO rH CO O on CM 00 ^ q 00 rH 7 d on 1-H 7 q CM 1-H 1-H ^ I-H 1-H CM -Q C/2 ns G * G J> r-v OO cn ^ o d ? rP ®2 X on' d R d rH OO rH GO 1-H d CM O ?-H d to 00 J> J> X + 1 1 CM +1 1 + 1 1 + 1 i X o lO CM od 00 1-H 50 ■r-^ ID i> CO CM r-H CM (M H d lO ^ 'f tD> ' ^ ■ TtH rH 1-H oo o !— 1 1 — 1 (M 1-H 1-H ' " '' — ^ ' — ' 05 05 cn on on 05 05 1 1 50 1 X 1 1 05 1 cn 1 i> 1 o 00 00 lO oo 05 05 cn 05 on rH 1-H ^H U U u U + PQ PP 1 ? 'a o p: G Xi G ITi i7 ? Q ? Q TO 0, S-I p D QJ C G S-i s 0 D G G 0 "u V t2 X QJ 0 pej P^ PQ 0 u CT) cn 'i/3 P Q t3 G Pi ’S sn OJ oo 00 05 00 cn OJ IPh o i-H c^r * 05 c^^ I O'! IT) fii CM + 1 CM CO CM 00 CM CM cn c Ifi 00 I on 05 00 ci 00 05 PQ % O !-i 1) lU -o T3 K a i> 00 I t-H 00 05 PQ 8 fU u fD •s be a !h 0 U u Ij C _o u 0 ”0 o S-I o, bC G G O g o 3 o c3 w QJ >- ^ D u ' O ^ II O a. X II 0 * X iO c« o QJ a o be C 0 1 be 0 c ^ ■p c G ■= G5 !L> 3 O O, PP 7- II s c a i sS ° ^ I— < II 'S Pm 02 IJ > OJ M § E M a 02 O d 5- 0 -M dj OJ S O u u 1. ^ (U 15 3 o u u 'be w be X o 9 he G E G U T3 ^ QJ 2 li > H OJ fl .. C3 OJ M )-• B S -Q Q OJ .P QJ U X a C QJ II P>-N Sh Q X 0 Sh PLi rt d 0 a J5 u •0 0/] T3 g C QJ 326 Short Communications VoL. 34, No. 4 Literature Cited Baudvin, H. 1990. Bilan de 10 annees d’etude sur la chouette hulotte Strix aluco en Bourgogne (France). Uccelli d’ltalia 15:30—38. Capula, M., L. Luiselli, and L, Rugiero. 1993. Compar- ative ecology in sympatric Podarcis muralis and Podards Simla (Reptilia; Lacertidae) from the historical centre of Rome: what about competition and niche segre- gation in an urban habitat? Boll. Tool. 60:287-291. Galeotti, P. 1994, Patterns of territory size and defence level in rural and urban Tawny Owl {Strix aluco) pop- ulations. /. Zool, London 234:641-658. Gil-Delgado, J.A., J. Verdejo, and E. Barba. 1995. Nes- tling diet and fledgling production of Eurasian Kes- trels {Falco tinnunmlus) in Eastern Spain./. Raptor Res. 29:240-244. Jedrzejewski, W., B. Jedrzejewska, K. Zub, A.J. Ruprecht, AND C. Bvstrowski. 1994. Resource use by Tawny Owls Strix aluco in relation to rodent fluctuations in Bialow- ieza National Park, Poland./. Avian Biol. 25:308-312. Kostrzewa, a. AND R. Kostrzewa. 1990. The relationship of spring and summer weather with density and breeding performance of the Buzzard Buteo buteo. Goshawk Acdpiter gentilis and Kestrel Falco tinnunculus. Ibis 132:550-559. Manganaro a., L. Salvati, S. Fattorini, and L. Ranazzi. 1999. Predation on geckos by urban Tawny Owls {Strix aluco). Avocetta23:'73-75. Penteriani, V. 1997. Long-term study of a Goshawk breeding population on a Mediterranean mountain (Abruzzi Apennines, central Italy); density, breeding performance and diet./. Raptor Res. 31:308-312. Petty, S.J. 1989. Productivity and density of Tawny Owls Strix aluco in relation to the structure of a spruce for- est in Britain. Ann. Zool. Fenn. 26:227-233. Piattella, E., L. Salvati, A. Manganaro, and S. Fatto- rini. 1999. Spatial and temporal variations in the diet of the Common Kestrel {Falco tinnunculus) in urban Rome, Italy. /. Raptor Res. 33:172-175. Plesnik, J. and M. Dusik. 1994. Reproductive output of the Tawny Owl Strix aluco in relation to small mammal dynamics in intensively cultivated farmland. Pages 531-535 in B.-U. Meyburg and R.D. Chancellor [Eds.], Raptor conservation today. World Working Group on Birds of Prey and Owls and Pica Press, Ber- lin, Germany. Ranazzi, L., A. Manganaro, R. Ranazzi, and L. Salvati. 2000. Density, territory size, breeding success and diet of a Tawny Owl {Strix aluco) population in a Mediter- ranean urban arean (Rome, Italy). Alauda 68:133- 143. Rizzo, M.C., L. Migliore, and B. Massa. 1993. Insects, small mammals and breeding performance of farm- land populations of the Common Kestrel {Falco tin- nunculus). Pages 11-18 in M.K. Nicholls and R. Clarke [Eds.] , Biology and conservation of small falcons. Pro- ceedings of the 1991 Hawk and Owl Trust Confer- ence, The Hawk and Owl Trust, London, U.K. Selas, V. 1998. Does food competition from red fox {Vul- pes vulpes) influence the breeding density of Goshawk {Acdpiter gentilis)'^ Evidence from a natural experi- ment./ Zool., London 246:325—335. SoDHi, N.S., PC. James, I.G. Warkentin, and L.W. Oli- PHANT. 1992. Breeding ecology of urban Merlins {Fal- co columbarius) . Can.]. Zool. 70:1477-1483. Southern, H.N. 1970. The natural control of a popula- tion of Tawny Owls {Strix aluco). J. Zool., London 162. 197-285. Tella, J.L., F. Hiraldo, J.A. DonAzar-Sancho, and JJ Negro. 1996. Costs and benefits of urban nesting in the Lesser Kestrel. Pages 53-60 in D. Bird, D. Varland, and J.J. Negro [Eds.], Raptors in human landscapes. Raptor Research Foundation, Academic Press, Lon- don, U.K. Wendland, V. 1984. The influence of prey fluctuations on the breeding success of the Tawny Owl Strix aluco Ibis 126:284-295. Received 31 December 1999; accepted 22 July 2000 J. Raptor Res. 34(4):327-329 © 2000 The Raptor Research Foundation, Inc. Noctural Activity of Lesser Kestrels Under Artificial Lighting Conditions in Seville, Spain Juan Jose Negro and Javier Bustamante Department of Applied Biology, Estacion Biologica de Donana, CSIC, Avda. Maria Luisa s/n, 41013 Sevilla, Spain Giro Melguizo and Jose Luis Ruiz Sociedad Protectora de Animales y Plantas de Sevilla, C/Santa Ana 10, Sevilla, Spain Juan Manuel Grande Department of Applied Biology, Estacion Biologica de Donana, CSIC, Avda. Maria Luisa s/n, 41013 Sevilla, Spa, in Key Words: Lesser Kestrel', Falco naumanni; nocturnal ac- tivity; foraging behavior, prey deliveries. Lesser Kestrels {Falco naumanni) are migratory, colo- nial small falcons. Kestrels in western European popula- tions breed mainly in holes and crevices in large historic buildings within towns and villages, or often in aban- doned farm houses scattered across the countryside (Cramp and Simmons 1980, Gonzalez and Merino 1990, Negro 1997). The species is considered Endangered in western Europe (Biber 1994). In the city of Seville in southern Spain, three Lesser Kestrel colonies remain in the downtown area. To our knowledge, no other city in western Europe as large as Seville (population 750 000) currently has Lesser Kestrel colonies. In Seville, the main colony of about 70 pr is located in the Cathedral. This Gothic building is the larg- est cathedral in Spain and third largest in the Christian world. One smaller colony is located in El Salvador (25 pr), a Baroque church nicknamed Seville’s second cathe- dral located about 500 m away from the Cathedral itself, and another is at Montesion (7-10 pr), a small chapel about 1.5 km from the Cathedral (C. Melguizo and J.L. Ruiz unpubl. data). The size of the city’s population of Lesser Kestrels has not changed significantly in the last 10 yr (J.J. Negro, C. Melguizo, and J.L. Ruiz unpubl. data), although the kestrels were surely more abundant in the past, when numerous breeding colonies thrived in different city buildings (Gonzalez and Merino 1990). Reports from the early 1970s (Andrada and Franco 1974) indicate that Lesser Kestrels were active at night around the Cathedral, where they apparently took advan- tage of the powerful ornamental illumination that high- lighted this historic building. However, no systematic study was ever conducted to determine whether Lesser Kestrels were active at night on a regular basis at the Cathedral or at other locations in the city. The fraction of birds involved in nocturnal behavior was also un- known, although Andrada and Franco (1974) suggested that up to 50% of individuals in the Cathedral could be active on any given night. The goals of our work were to: (1) determine if Lesser Kestrels were active every night under ornamental lights, (2) determine which fraction of the colony was active at night, (3) determine the func- tion of nocturnal activity, and (4) describe this unusual behavior in a typically diurnal species. In addition, prey deliveries by adults were recorded at selected nests dur- ing day- and night-time hours to determine the relative contribution of nocturnal activities in raising nestlings. Methods We monitored seven nests in the Cathedral, which were clustered in two groups of three and four nests, respectively. In El Salvador, where there is also ornamen- tal illumination at night, we observed a group of hve nests and a group of four nests. Nests within each group were close enough to permit simultaneous observation from vantage points in the street, at an average distance from the nests of about 40 m. Surveys were conducted on days 3, II, 17, 20, and 21 June 1998, coincidental with the period when young were in nests. Focal nests were monitored with lOX binoculars and 20-(50X spotting scopes during three time periods: midday (1200-1400 FI), late afternoon (1800-2000 H), and night (2200-2400 H). The first two periods occurred during full daylight Sunset took place at about 2130 FI during the study pe- riod, and ornamental illumination was on between 2200- 2400 H, coincidental with our nighttime observations Total observation time amounted to 30 hr/nest. Obser- vations were not extended after 2400 FI because, prior to this study, we had observed that kestrels roosted as soon as the lights were turned off. During observations, we recorded instances of young being fed by adults and the sex of the feeder by plumage characteristics (Cramp and Simmons 1980). We tried to identify prey, but deliveries happened so quickly that it was impossible to identify them in most cases. Between 2300-2400 H, we also recorded the maximum number of kestrels that were flying together over the Cathedral and El Salvador, respectively. Observations were carried out with the help of 40 stu- dents of the Faculty of Biological Sciences, University of Seville. They were trained on species recognition and dif- ferent aspects of breeding biology prior to taking obser- vations. They carried out observations in groups of two and were randomly assigned to the four observation spots. Each group observed for an average of 6 hr At 327 328 Short Communications VoL. 34, No. 4 Table 1. Maximum number of Lesser Kestrels seen flying together each night (2300-2400 H) of the study period in 1998 at the Cathedral and El Salvador in Seville, Spain. No. OF Number in Flight Breeding Pairs 03 June 1 1 June 17 June 20 June 21 June Cathedral 70 26 23 22 55 33 El Salvador 25 5 10 19 15 21 least one of us was present with each student group to supervise fieldwork. Differences in feeding rates were not significant be- tween the two churches (one-way analysis of variance [ANOVA], P > 0.05), so data for the 12 focal nests were pooled for analysis. Differences in feeding rates (feed- mgs/hr) during the time periods were also tested using a one-way ANOVA. For this analysis, we used all observed feedings, including those instances in which the sex of the feeder was unknown. To test for differences in feed- ing rates between males and females along the different time periods, a two-way ANOVA was used. Period and sex were used as factors, and the analysis was restricted to those observations where the sex of the feeding parent was known. Results Lesser Kestrels were active every night we made obser- vations. The number of kestrels that we observed simul- taneously while flying at night ranged between 22-55 in the Cathedral, and 5-21 in El Salvador (Table 1). There- fore, a large fraction of birds from each colony was active every night. Kestrels typically soared together in circles over the illuminated buildings. The flock of soaring birds would suddenly disperse and individuals would chase and eat flying insects. Nocturnal flights took place at different heights over both the Cathedral and El Salvador. Often the birds circled and hunted above the Giralda, the Ca- thedral’s bell tower, which is the tallest structure in the Figure 1. Prey delivery rates (x ± 2 SE) by male and female Lesser Kestrels during three periods of observa- tion at the Cathedral and El Salvador in Seville, Spain. downtown area with a height of about 98 m. Some kes- trels descended to the nests from time to time, although prey deliveries were rarely observed. A total of 411 prey deliveries were observed at the focal nests, 44 of them during night observation periods. Feed- ing of nestlings varied significantly between the three dai- ly periods {F = 25.56, df = 2, P < 0.001). On average, we observed 1.3 prey deliveries/hr/nest during midday, 2.1 prey deliveries/hr/nest in the afternoon, and 0.4 prey deliveries/hr/nest during the nocturnal period. A two-way ANOVA showed significant differences between sexes {F = 7.451, df = 1.66, P — 0.008), periods {F — 31.52, df = 2.66, P< 0.0001), and the interaction effect between these two factors {F = 3.30, df = 2.66, P = 0.042). Males provided more food items to the young than females during the day (Fig. 1), as previously re- ported (Donazar et al. 1992). Nocturnal feedings to young by both males and females were very uncommon Discussion We present the first description of nocturnal activity in Lesser Kestrels and examine its possible contribution to the successful raising of young. At the Cathedral and El Salvador, lights used to illuminate the buildings at night attracted large quantities of insects making them both visible and accessible to kestrels. The insects also attract- ed significant numbers of pipistrelle bats (Pipistrellus pip- istrellus), which were sometimes found among prey re- mains of kestrels nesting at the Cathedral (Negro unpubl. data). Bat hunting by Lesser Kestrels is uncom- mon, and the few published records involved individuals that were hunting at dusk or in daylight (Carbajo and Ferrero 1981, Paterson 1991). The unusual nighttime activity of Lesser Kestrels in- cluded a fairly large fraction of the adults in each colony In fact, we believe that, at some time, all adult birds in the two colonies were active at night. Kestrels remained active until midnight, when the lights were turned down, approximately 2.5 hr after sunset. Unfortunately, lights did not stay on for the whole night, and we do not know whether the kestrels would be able to extend their activity period even further. Compared to daytime hours, nest provisioning was minimal at night, so we inferred that the main purpose of the adult kestrels’ activity at night was to feed them- December 2000 Short Communications 329 selves. Daytime feeding rates at nests were similar to those found in other areas of southern Spain (2.0 prey deliveries/hr) and Portugal (2.2 prey deliveries/hr), but lower than in rural areas of northern Spain (3.9 feed- ings/hr) where distances to foraging areas were shorter (Negro 1997). Lesser Kestrels in Seville breed in colonies in the old downtown area, far from foraging places in the city out- skirts. While their nocturnal activity may not result in a significant increase in the prey delivery to the nestlings, it could facilitate feeding of adults during the breeding season perhaps making their urban existence easier (Tel- ia et al. 1995). Urban sprawl around the city is already affecting the hunting areas of the kestrel and it could jeopardize the future of this population. Therefore, noc- turnal feeding by the breeding population could make the difference that permits the large kestrel colonies to continue to thrive despite being encroached by many ki- lometers of apartment blocks in every direction. Although Lesser Kestrels are typically migratory, some birds remain all year round in southern Spain (Andrada and Franco 1975, Gonzalez and Merino 1990) and, spe- cifically in Seville (Negro et al. 1991) . Andrada and Fran- co (1974) suggested that nocturnal activity could go on through the year at the Cathedral but it remains to be seen whether wintering kestrels are active at night in Se- ville. Further studies are needed to assess the actual effect of nocturnal activity on adult kestrel foraging strategy, not only during the breeding season, but also during the remainder of the year. It is also important to clarify the role of nocturnal foraging in long-term survival in this population. Resumen. — Estudiamos la actividad nocturna del Cerni- calo Primilla {Falco naumanni) en dos colonias de cria situadas en edificios historicos del centre de Sevilla (Sur de Espana). Nuestro objetivo fue describir este compor- tamiento inusual asi como su importancia relativa para la alimentacion de los polios. Se observaron 12 nidos dur- ante tres periodos diarios (1200-1400 H, 1800-2000 H y 2200-2400 H) en Junio de 1998. La mayoria de los cer- nicalos adultos de ambas colonias estuvieron activos cada noche a lo largo del periodo de estudio mientras fun- cionaba la iluminacion ornamental. La actividad cesaba cuando era apagada la iluminacion a medianoche. Los cernicalos capturaban y comian insectos en vuelo llevan- do pocas presas a los nidos. Los Cernicalos Primillas, por tanto, permanecen activos por la noche para incremen- tar su ingesta diaria de alimento y no para alimentar a sus polios. El desarrollo urbanistico de Sevilla esta redu- ciendo los territorios tradicionales de caza de los Cerni- calos en el entorno de la ciudad. Es posible que la ex- tension de la caza a horas nocturnas permita que los Cernicalos perduren aun en el centre de Sevilla. [Traduccion de los autores] Acknowledgments We are very thankful to the volunteers who helped us with the fieldwork: A. Ruiz, A. Crespo, M. Diaz, C. Cha- morro, M.E. Fernandez, P. Ibarra, M.C. Aroca, D. Ciudad, Leonidas, A. Iglesias, P. Aranda, J. Chaves, M.C. Torres, C. Garcia, J. Renfel, Zurina, Sonia, R. Balbontin, L. Fer- nandez, P. Alvarez, M. Narvaez, J. de la Rosa, G. Ortega, Susana, A. Alonso, J.R. Garceta, L. Morales, J. Hernandez, Antonio, Reyes, Aura, E. Gomez, A. Gomez, M. Martin, I. Moreno, D. Doblas, Monica, Rocio, and J. Esquivias. The manuscript benefitted from the comments of T.W. Carpenter and O. Hatzoffe. Literature Cited Andrada, J. and A. Franco. 1974. Actividad nocturna en Falco naumanni. Ardeola 19:471. AND . 1975. Sobre el area de invernada de Falco naumanni en Espana. Ardeola 21:321-324. Biber, J.P. 1994. Lesser Kestrel in G.M. Tucker and M.E. Heath [Eds.], Birds in Europe: their conservation sta- tus. BirdLife Conservation Series No. 3. BirdLife In- ternational, Cambridge, U.K. Carbajo, F. and j. Ferrero. 1981. Cernicalo Primilla {Fal- co naumanni). Ardeola 28:155. Cramp, S. and K.E.L. Simmons. 1980. Handbook of the birds of Europe, the Middle East and North Africa. Vol. 2. Oxford Univ. Press, Oxford, U.K. DonAzar, J.A., JJ- Negro, and F. Hiraldo. 1992. Func- tional analysis of mate-feeding in the Lesser Kestrel Falco naumanni. Ornis Scand. 23:190-194. Gonzalez, J.L. and M. Merino. 1990. El Cernicalo Prim- illa {Falco naumanni) en la Peninsula Iberica. Sene Tecnica, ICONA, Madrid, Spain. Negro, JJ- 1997. The Lesser Kestrel. Birds of Western Pa- learctic Update, 1:49-56. , M. de la Riva, and j. Bustamante. 1991. Patterns of winter distribution and abundance of Lesser Kes- trels {Falco naumanni) in Spain. J. Raptor Res. 25.30- 35. Paterson, A.M. 1991. Lesser Kestrel hunting bats Br. Birds 84:151. Tella, J.L., F. Hiraldo, J.A. DonAzar-Sancho, andJ.J. Ne- gro. 1995. Costs and benefits of urban nesting in the Lesser Kestrel. Pages 53-60 in D.M. Bird, D.E. Varland, and J.J. Negro [Eds.], Raptors in human landscapes. Academic Press, London, U.K. Received 3 March 2000; accepted 3 August 2000 J Raptor Res. 34(4):330-333 © 2000 The Raptor Research Foundation, Inc. Nest-Site Characteristics oe Crested Caracaras in La Pampa, Argentina Michael I. Goldstein^ Department of Wildlife and Fisheries Sciences, Texas A&’M University, 210 Nagle Hall, College Station, TX 77843-2238 US. A. Key Words: Crested Caracara; Caracara plancus; nest char- acteristics', grasslands, Argentina. Crested Caracaras {Caracara plancus) range from the southern United States to southern South America and, although the natural history of North American popula- tions has been described (Bent 1938, Brown and Ama- don 1968, Morrison 1996), there has been little in-depth study of the species elsewhere. Nest characteristics for Crested Caracaras have been documented in B^a Cali- fornia (Rivera-Rodriguez and Rodriguez-Estrella 1998), Texas (Dickinson and Arnold 1996), Florida (Morrison 1996), and Argentine Patagonia (Travaini et al. 1994). Knowledge of the Argentine population is limited to Tra- vaini et al. (1994) and several brief notes. In this paper, I describe characteristics of 17 Crested Caracara nests found during December 1998 and January 1999 in north- ern La Pampa, Argentina, and compare these character- istics with those observed in other populations. Study Area The landscape in the northern La Pampa panhandle (approximately 35°14'S, 63°57'W) is dominated by a mix of cattle ranching and row-crop agriculture. Common summer crops are alfalfa, sunflower, sorghum, and corn. Uncommon and isolated native mesquite {Prosopis spp.) trees and introduced Chinese elm ( Ulmus parvifolia) are found along dirt and paved roads. Mesquite trees in fields have generally been removed for agriculture and few remain. Forested areas are generally groves of intro- duced eucalyptus {Eucalyptus spp.) trees, planted as shade areas for cattle, for wind breaks between fields, and as entrance corridors to estate houses. High winds and strong rains are common during the summer, particularly from the southeast. Annual rainfall is 828 mm, with 50% occurring from November through February. Summer temperatures range from 23-38°C. Methods Surveys for caracara nests were conducted by vehicle in December 1998 and January 1999. Surveys were con- ducted once, over approximately 420 km along a 2000 km^ grid of public dirt and paved roads. Roads were buff- ered by 10 m of public land on each side, were fenced, and lay adjacent to private ranches. Private lands were not surveyed. All trees and shrubs along the roads were checked for caracara nests. Nests showing disrepair or no sign of occupancy during the breeding season were not ^ Present address: Mendocino Redwood Company, P.O. Box 489, Fort Bragg, CA 95437. used in this analysis. For each occupied nest found (eggs, nestlings, fledglings, or adults at the nest), substrate height (base to the top of substrate), nest height (base to the egg level of the nest), and nest orientation (devi- ation from magnetic north grouped in 45° octants) were recorded. Nest and nest cup diameters were measured for five nests, two in mesquite and three in elm. Differ- ences in nest heights between tree species were analyzed using a t-test. Nest orientation was examined using Ray- leigh’s test for circular uniformity (Zar 1996). All mean values are presented as mean ± standard error. A critical value of 0.05 was used for all analyses. Results and Discussion Seventeen caracara nests were found during road sur- veys. Thirteen nests were found in mesquite and four in elm. Nests were not found in other vegetation types Nests were generally located in isolated trees and in the tallest vegetation in the immediate area. When nests were found in elms, no mesquite trees were in the vicinity. For two of the nests found in mesquite, a neighboring mes- quite tree was located approximately 40 m away. Agricul- tural fields surrounded all nest trees. Nests were loeated more than 1 km from the nearest building, although buildings were not generally found near roads, but well inside private estate boundaries. Two nests were located 0.4 km apart. The average height of the nest tree was 7.4 ± 0.4 m. Mesquite trees containing nests averaged 7.0 ± 1.4 m {N — 13) in height; elm trees with nests averaged 8.6 ±1.7 m {N = 4) . Mesquite trees were significantly shorter than elms {t = 1.870, df = 15, P < 0.05). Nests were located in the top third of trees, or the upper structural canopy, for both tree species. Nest height in mesquite trees av- eraged 5.5 ± 1.4 m; nest height in elm trees averaged 6.0 ± 0.8 m. Nest heights between tree species were not significantly different {t = 0.636, df = 15, P> 0.05). The average nest height was 5.6 ± 0.3 m. Eight nests were oriented towards the northwest and northeast and six nests were oriented towards the south- west (Fig. 1). However, nest orientation did not differ from a uniform distribution (Rayleigh’s R = 3.808, N = 17, P> 0.05). Nests were typically constructed from twigs from the same species of nest tree, woven together with grass stalks. External diameter of nests averaged 58 ± 4 cm in width and 7l ± 6 cm in length. Nest cups averaged 31 ± 1 cm in width and 42 ± 4 cm in length. Internal area accounted for an average of 33 ± 4% of the nest area. The nest cup was lined with wool, grass, black baling 330 December 2000 Short Communications 331 T North storms ^ ^ Figure 1. Nest orientation of Crested Caracaras in La Pampa, Argentina. Values are nests open within a 45° sec- tor. twine, and feathers. Black baling twine was the most com- mon item used to line the nest cup. In one nest, a nes- tling was found entangled in black baling twine and re- quired extrication. The same nest contained the corpse of a dead nestling with legs entangled in twine. An ad- ditional nest contained a dead nesding, also entangled in twine. The landscape surrounding caracara nests in La Pam- pa was similar to that described for caracaras nesting in Florida and Texas (Dickinson and Arnold 1996, Morrison 1996); generally open grassland with low ground cover, a low density of tall vegetation, few trees, and scattered brush. Trees in this part of La Pampa have been typically removed from fields; they are found scattered along roadsides, or have been planted for aesthetic beauty, shade or as windbreaks near houses, barns, and corrals. Mesquite trees are more common in the landscape 100- km west in the province of San Luis. ^Although mesquite trees used by caracaras were signif- icantly shorter than elms, I obtained neither more de- tailed measurements of nest tree structure nor data to compare random trees that were not used for nesdng. If roadside landscapes are fairly homogenous across this re- gion, choice of the nest tree may be related primarily to structure and cover, not height. In general, mesquite branches are stouter, branch structure is more inter- twined, and branches maintain their thorny configura- tion, thereby providing better structure and cover for ca- racara nests. Use of elm trees may be related to other structural characteristics or may be a result of the lack of mesquite trees in the vicinity. Nest dimensions are difficult to measure because nests are frequently reused in consecutive years and become layered in structure and quite large (Bent 1938, Dickin- son and Arnold 1996, Morrison 1996). At the end of the nestling period, nestlings stomp the nest flat and it be- comes difficult to measure the nest cup or to be sure the peripheral edges of the nest have not deteriorated (Mor- rison pers. comm.). Nest length in Texas averaged 59 ± 5 cm and nest width averaged 50 ± 2 cm {N = 5; Dick- inson and Arnold 1996). Florida nests measured 71,1 cm in diameter {N = 12; Layne 1996). Nests measured in La Pampa were similar in size. In both Florida and Patagonia, Crested Caracaras show a strong tendency to orient their nests away from pre- vailing winds (Travaini et al. 1994, Morrison 1996). Dur- ing the breeding season in La Pampa, cold storms come from the southeast and warm humid winds come from the subtropical north. Although I found no significant dispersion from random for caracara nests in La Pampa, most nests were generally oriented away from the south- east and towards the north. Since in the Southern Hemi- sphere the sun crosses the northern sky during the day, the nests facing north are generally facing towards the warm part of the sky and warmer winds. Across their range. Crested Caracaras use a variety of structural supports for their nests, but typically choose the tallest vegetation available and construct their nests near the top of the nest structure. In Patagonia, nests in aspen {Populus tremuloides) were in the lower half of the tree, although nests in Nothofagus spp. and Berberis spp were found on top of the substrate (Travaini et al. 1994). In Florida, Texas, and La Pampa, nests were built below the nest-support canopy but tended toward the maxi- mum structural height for the tree species. This suggest- ed some preference for structure or cover in the nest tree, either large enough to support their bulky nests or with cover suitable for protecting young. In La Pampa, urban growth has proceeded rapidly dur- ing the last two decades, resulting in fragmentation of private estates and landscape changes that include per- sistent loss of native flora. It is reasonable to suspect that the loss of mesquite and other trees resulting from con- version of native vegetation to pasture and agriculture represents loss of breeding structures suitable for use by Crested Caracaras. Although no population data are available for Crested Caracaras in La Pampa, such loss of suitable nest structures may ultimately have a negative influence on the population. Impacts of human activities on these caracaras were also indicated by observations of young becoming entangled in baling twine used to line nests. Resumen. — Hice un reconocimiento de aproximada- mente 2000 km^ atraves de carreteras asfaltadas y de tier- 332 Short Communications VoL. 34, No. 4 Table 1. A comparison of Crested Caracara nest and nest tree heights (mean ± SE) from several locations across the species’ range. Tree Species Tree Ht ( m) Nest Ht ( m) N/T^ N SOURCE‘S Pachycereus spp. 9.7 ± 2.2 4.7 ± 0.8 0.50 18 1 Yucca valida 6.0 4.2 0.71 2 1 Olneya tesota 4.5 4.0 0.89 1 1 Cercidium microphyllum 6.0 5.3 0.88 1 1 Washingtonia robusta 9.0 8.5 0.94 1 1 Rosa bracteatcf 4.3 ± 1.4 3.7 ± 1.2 0.86 6 2 Populus tremuloides 19.9 ± 7.8 8.0 ± 3.6 0.43 14 3 Maytenus boatia 6.0 4.5 0.76 1 3 Araucaria araucaria 13.0 9.0 0.70 1 3 Berberis darwinii 4.0 1 3 Salix humboldtiana 8.0 4.0 0.50 1 3 Nothofagus spp. 5.0 5.0 1.00 1 3 Sabal palmetto 7.3 ± 0.2 6.4 ± 0.2 0.88 83 4 Taxodium distichum 8.5 11 0.77 1 4 Quercus virginiana 8.6 9.3 0.92 1 4 Prosopis spp. 7.0 ± 1.4 5.5 ± 1.4 0.79 13 5 Ulmus parvifolia 8.6 ± 1.7 6.0 ± 0.8 0.70 4 5 Mean 0.76 ^ Includes one nest located in yaupon {Ilex vomitoria). •^N/T indicates the ratio of nest height to tree height. ‘ Sources: (1) Rivera-Rodriguez and Rodriguez-Estrella 1998, (2) Dickinson and Arnold 1996, (3) Travaini et al. 1994, (4) Morrison pers. cornm., and (5) this study. ra en el norte de La Provincia de La Pampa, Argentina. Encontre 17 nidos de carancho {Caracara plancus) , 13 en Prosopis spp. y 4 en olmos ( Ulmus parvifolia) . El promedio de la altura de los arboles con nidos fue 7.4 m; el pro- medio de la altura de los nidos fue 5.6 m. Los olmos fueron mas altos que los de Prosopis spp. {Ulmus 8.6 m; Prosopis 7.0 m) aunque la altura de los nidos no fue sig- mficativamente diferente {Ulmus 6.0 m; Prosopis 5.5 m). En general, la orientacion de los nidos fue distinta a la direccion prevaleciente de los vientos, aunque con la prueba de Rayleigh, no encontre una dispersion signifi- cativamente diferente de la normal (Rayleigh’s R = 3.808, P > 0,05). Encontre dos polluelos muertos en ni- dos en Ulmus, con las patas enredadas en cordel de nylon utilizado para embalar pacas de heno. En este reporte, compare las caracteristicas de los nidos de carancho en el norte de La Pampa, Argentina, con otras poblaciones de carancho atraves de rango de distribucion de esta es- pecie. [Traduccion del autor] Acknowledgments A. Lanusse provided lodging and transportation in La Pampa, and with T. Hibbitts, provided field assistance during the study. J. Sarasola (Universidad Nacional de La Pampa, Santa Rosa) provided logistical support in Argen- tina, J. Morrison kindly provided unpublished data and sound advice. T. Lacher, T. Hibbitts, J. Morrison, A. Tra- vaini, and R. Rodriguez-Estrella reviewed earlier versions of the manuscript. Additional funding was provided by the Frank M. Chapman Memorial Fund and the Depart- ment of Wildlife and Fisheries Sciences at Texas A&M University. Literature Cited Bent, A.C. 1938. Life histories of North American birds of prey. Part 2. U.S. National Museum, Bull. 170. U.S. Government Printing Office, Washington, DC U.S.A. Brown, L. and D. Amadon. 1968. Eagles, hawks, and fal- cons of the world. McGraw-Hill Book Co., New York, NY U.S.A. Dickinson, V.M. and K.A. Arnold. 1996. Breeding biol- ogy of the Crested Caracara in south Texas. Wilson Bull. 108:516-523. Layne, J.N. 1996. Crested Caracara. /raJ.A. Rodgers, Jr., H.W. Kale, 11, and Henry T. Smith [Eds.], Rare and endangered biota of Florida. Vol. V. Birds. Univ. Press Florida, Gainesville, FL U.S.A. Morrison, J.L. 1996. Crested Caracara {Caracara plan- ms). In A. Poole and F. Gill [Eds.], The birds of North America, No. 249. The Academy of Natural Sciences, Philadelphia, PA, and The American Ornithologists’ Union, Washington, DC U.S.A. December 2000 Short Communications 333 Rivera-Rodriguez, L.B. and R. Rodriguez-Estrella. 1998. Breeding biology of the Crested Caracara in the Cape Region of B^a California, Mexico. J. Field Orni- thol 69:160-168. Travaini, a, J.A. Donazar, O. Ceballos, M. Funes, A. Rodriguez, J. Bustamante, M. Debiles, and F. Hir- ALDO. 1994. Nest-site characteristics of four raptor spe- cies in the Argentinian Patagonia. Wilson Bull. 106. 753-757. Zar, J.H. 1996. Biostatistical analysis, Brd Ed. Prentice Hall, Princeton, NJ U.S.A. Received 2 September 1999; accepted 23 July 2000 J Raptor Res, 34(4) ;334-338 © 2000 The Raptor Research Foundation, Inc. Diet of the Barn Owl ( Tyto alba tuidara) in Northwestern Argentine Patagonia Maria S. Pillado and Ana Trejo Centro Regional Universitario Bariloche, Unidad Postal Universidad, 8400 San Carlos de Bariloche, Argentina Key Words: Bam Owl; Tyto alba; diet, Patagonia', Argen- tina. Given its wide distribution and sedentary habits, the diet of the Barn Owl ( Tyto alba) has been studied in more detail and more extensively than that of any other bird of prey (Everett et al. 1992). Rodents and other small mammals are the main prey in the diet of Barn Owl in all of its range along with variable proportions of birds, reptiles, amphibians, fish, and arthropods (Taylor 1994). The Barn Owl {T. alba tuidara) is widespread in conti- nental Argentina and occasionally on islands (Canevari et al. 1991). Food habits of the Barn Owl have been thor- oughly studied in agrosystems in Argentina (Bellocq 1990, Bellocq and Kravetz 1994), but little is known about its diet in southern Argentina. In Patagonia, most studies have focused on the arid eastern steppes (De Santis and Pagnoni 1989, De Santis et al. 1993, 1996, Garcia Espon- da et al. 1998). Our aim was to provide information on the diet of the Barn Owl in a somewhat different area with more mesic vegetation features and a small mammal fauna mixing typical steppe species with others more characteristic of humid forests nearby (Monjeau 1989). Study Area and Methods The study site was located in the Reserve Area of Na- huel Huapi National Park, in northwestern Argentine Pa- tagonia (7r07'25"W, 40°47T4"S) at 700 m elevation above sea level. The area is an ecotone between the arid Patagonian steppe to the east and the southern beech (Nothofagus spp.) forests to the west. The site was domi- nated by bunchgrasses {Stipa speciosa) and cushion bushes (Mulinum spinosurn) with scattered trees (Austrocedrus chi- lensis, Maytenus boaria, and Populus nigra). At times, wil- lows (Salix fragilis) formed small gallery forests. Owl roosts were located by observing areas of white- wash or recording places where pellets were found. Pel- lets were collected every two weeks from June 1993-May 1994 at two known roost sites. Pellets were grouped into calendar seasons, oven-dried in 70°C for 72 hr, and pro- cessed following standard methods (Marti 1987). Most prey were identified to species. Mammalian prey were identified and quantified on the basis of skulls and den- taries using reference collections and keys (Pearson 1995). Insects were quantified by counting head capsules and mandibles. Biomass of each rodent species in the total biomass of the diet was calculated by multiplying mean body mass of individuals by the number of individuals in pellets and expressed as a percentage of total rodent biomass con- sumed. We calculated the geometric mean of weight of prey (Marti 1987): GMW = antilog (S rajog nj. where n^ was the number of individuals of the ith species and Wi was the mean weight. We also determined the mean length of rodents consumed after Jaksic et al. (1977): MLR = S fiXi/m, where p was the frequency of the i species in the diet, x^ was mean body length, and m the total number of identified rodents. Mean weight of mammals and mean body length of rodents were taken from the literature (Redford and Eisenberg 1992, Pear- son 1995). Food-niche breadth (FNB) was estimated using Levins’ (1968) index: FNB — 1/(S p^), where was the propor- tion of prey taxon i in the diet. A standardized niche breadth value (FNB^J was then calculated, which ranged from 0-1: FNBgt = (FNB — l)/(n — 1), where « was the total number of prey categories (Colwell and Futuyma 1971). Evenness (J') was calculated by the Shannon-Wie- ner function as follows: J' = H'/H'max, where H' was the Shannon-Wiener function and H'max was the maxi- mum value of H'; that is, the logarithm of the number of species in the sample (Krebs 1989). Results and Discussion A total of 425 prey items was identified from 229 pel- lets. The mean number of prey/pellet was 1.9 ± 0.9 (±SD, range = 1-4) and the mean number of rodents/ pellet was 1.8 ± 0.9 (range = 1-4). Barn Owls preyed mainly on rodents (95.1%). Hares and insects made up 0.5% and 4.4% of prey, respectively. The two European hares {Lepus europaeus) found in the diet were newborns. Insects were all in the family Scarabaeidae (Table 1 ) . By percent frequency, the most consumed sigmodon- tine rodent species were Abrothrix longipilis, Loxodontomys micropus, and Oligoryzomys longcaudatus. In terms of bio- mass, Loxodontomys micropus was the most important prey in the diet, followed by Abrothrix longipilis ‘And Oligoryzomys longicaudatus (Fig. 1). The Barn Owl feeds almost exclusively on small mam- mals throughout its range, although the proportions of other prey may vary slightly (Taylor 1994). Barn Owls in our study preyed almost exclusively on rodents with ju- venile hares and insects rarely appearing in the diet, mostly in spring. We did not find birds, reptiles, nor am- phibians to be important prey as was the case in La Pam- pa, Argentina (Noriega et al. 1993). Based on the literature, the most important prey of Barn Owls in Argentine Patagonia are Eligmodontia mor- gani and Reithrodon auritus (De Santis and Pagnoni 1989, De Santis et al. 1993, Tiranti 1996, Travaini et al. 1997). In our study area, neither species represented >4% and 8% of total prey items, respectively. This was not surpris- ing because the habitat characteristics of our study area 334 December 2000 Short Communications (33 ;z: d oi V a, ;h U rO S d d •i-i o js ^-l i-i V > o t 3 u "d jj u SA (U I+-I o S-i !U rO s d d ci • i-H d o PLH d '-d d OJ be d u J-i 1 ) I o d O d cd ii jd j-i t+H o -M u T 3 d o 1/3 CD CTi OOCMODOi-HCMlDCMe^OO C^T CM ' — I OirHTtloDCO^S'^Oi^QO ^ I— I ^ O CM CD CD iDiDOqCMCMt'- r -4 00 CM in ye CD m CM in O CD I> 05 CM CM 1 — i nj CD ^b m CM CD CM ^ O CM CD CD 00 CM J> o 6 CD CM m yb CM CD CM CD CM CO 00 CM Tt yb CM I— I CM m CM CO CO O 'f d CM m CM CO CD in CD 1 1 'd 03 03 P 1 1 CM 1 CM j-H d 1 -H GO cd 00 ^ 1 yb d P ;& U 53 Pi -S 3 '2 “= a I o ? d >i H o u o: Oh bi o a o CS «o P 05 CO in CM o CM in o> ^ CM CM ^ CM d _ 03 00 00 m CD <0 f-H 1 — I in- CD o o -cj^ 00 03 p i> in o d d P P CD in ^ p CM 03 CM CD 00 03 i> CO CD CD i> i> CM i-H d d P P 00 o ^ 1 I-H 03 in oq CO 1 1 1 CM P P P 00 00 ^ 1 1 00 m CO CD CO CD 00 ^ I-H 'Tfi 00 03 CM CM d d P P P I-H Tf CM I-H e— O CD 00 m 00 03 OJ Si fd 6 d V t 3 O u Mh O be id 'C d C 5 i p a bo d d d d a o V O 335 336 Short Communications VoL. 34, No. 4 Ct A1 Ax Em Ech It Lm 01 Px Ra Species □ Frequency (%) ■ Biomass (%) Figure 1. Frequency and biomass of rodent prey species in the diet of the Barn Owl. Biomass is expressed as the percentage of biomass of each species calculated on total rodent biomass. Ct — Ctenomys haigi, M — Abrothrix longipilis. Ax — A. xanthorhinus, Em — Eligmodontia morgani, Ech — Euneomys chinchilloides. It — Irenomys tarsalis, Lm — Loxodontomys micwpus, 01 — Oligoryzomys longicaudatus, Px — Phyllotis xanthopyga, Ra — Reithrodon auritus. were not optimal for these rodents, which prefer the more xeric and open habitats of the Patagonian steppe (Pearson 1995). The most common species in the diet, both in fre- quency and biomass, Abrothrix longipilis, Oligoryzomys lon- gicaudatus, and Loxodontomys micwpus are good climbers and prefer brushy places, although the latter is also found in shallow wet grasslands (Pearson 1983). Taking this into account, we inferred that the most frequently- used habitats in the Barn Owls’ hunting range were those with good vegetation cover and ample water. Hares were only occasionally eaten by Barn Owls de- spite their relative abundance (approximately 4-18 hares/ha; Novaro et al. 1992), their crepuscular or noc- turnal habits, and their open nests (Bonino and Monte- negro 1997), all traits which might make them vulnerable to an aerial nocturnal predator like the Barn Owl. There is only one citation for the Argentine Patagonia record- ing predation by Barn Owls on rabbits ( Oryctolagus cunic- ulus, 0.1% of total prey, Travaini et al. 1997). In central Chile, the proportion of rabbits in the diet of Barn Owls IS also very low (0.03% of total prey, Herrera and Jaksic 1980). In Chilean Patagonia, Iriarte et al. (1990) did not record predation on hares by Barn Owls although they were eaten by Great Horned Owls {Bubo virginianus) in variable proportions (Donazar et al. 1997, Trejo and Gri- gera 1998). Even juvenile hares may not be very suitable prey for Barn Owls since they are much smaller than Great Horned Owls (Everett et al. 1992). According to Jaksic (1986), this is a common situation in southern South America where some predators hunt mainly the more abundant native rodents, often ignoring abundant introduced lagomorphs. Jaksic (1986) attributed this fact to an “escape in size.’’ Maximum weight of juvenile hares is about 300 g (Bonino and Montenegro 1997), which puts them beyond the size of prey more frequently con- sumed by Barn Owls. Rabbits, although smaller than hares, were probably not in our study area. Mean weights and sizes of rodents were approximately the same during the four seasons, suggesting that in our study area the Barn Owl preyed more upon medium- sized {A. longipilis, L. micropus, and O. longcaudatus) than on the smaller-sized {A. xanthorhinus and E. morgani) ro- dents in the area. Mean weight of prey of the Barn Owl in Chilean Patagonia is smaller (29.9 g), due to a greater consumption of smaller species (Iriarte et al. 1990). Food-niche breadth is intermediate, as has been shown in Chilean Patagonia (Iriarte et al. 1990). This indicated that the Barn Owls in our study behaved essentially as specialized rodent predators. Diets of sympatric Great December 2000 Short Communications 337 Horned Owls have been studied in two sites in north- western Patagonia (Don^ar et al. 1997, Trejo and Gri- gera 1998) and, in both cases, a lower food niche breadth (0.20) was found to be due to lower species evenness in the diet. Resumen. — En el presente trabajo se estudio la dieta de Tyto alba tuidara en el noroeste de la Patagonia argentina. Los roedores sigmodontinos fueron el componente prin- cipal de la dieta, tanto en numero como en biomasa. Las liebres y los insectos fueron poco consumidos. Las espe- cies de roedores mas consumidas fueron Abrothrix longi- pilis, Loxodontomys micropus y Oligoryzomys longicaudatus. De los datos de la dieta y teniendo en cuenta los habitats de las presas se inhere que la actividad de caza de T. alba en el area de estudio se desarrollo preferentemente en am- bientes humedos o mesicos con buena cobertura vegetal. [Traduccion de los autores] Literature Cited Bellocq, M.I. 1990. Composicion y variacion temporal de la dieta de Tyto alba en ecosistemas agrarios pam- peanos, Argentina. Vida Silv. Neotrop. 2:32-35. AND F.O. Kravetz. 1994. Feeding strategy and pre- dation of the Barn Owl ( Tyto alba) and the Burrowing Owl (Speotyto cunicularia) on rodent species, sex, and size, in agrosystems of central Argentina. Ecol. Aust. 4: 29-34. Bonino, N. and a. Montenegro. 1997. Reproduction of the European hare in Patagonia, Argentina. Acta Ther- iol. 42:47-54. Canevari, M.P., P. Canevari, G.R. Carrizo, G. Harris, J. Rodriguez Mata and R.J. Straneck. 1991. Nueva guia de las aves argentinas. Fundacion Acindar, Buen- os Aires, Argentina. Colwell, R.K. and D.J. Futuyma. 1971. On the measure- ments of niche breadth and overlap. Ecology 52:567- 576. De Santis, L.J.M. and G.O. Pagnoni. 1989. Alimentacion de Tyto alba (Aves: Tytonidae) en localidades costeras de la Provincia del Chubut (Republica Argentina). Neotropica 35:43-49. , I.M. Pena Cozzarin, and M.F. Grossman. 1993. Vertebrados depredados por Tyto alba (Aves, Tytoni- dae) en las proximidades del rio Corintos (Provincia del Chubut, Argentina) . Neotropica 39:53-54. , C.M. Garcia Esponda, and G.J. Moreira. 1996. Vertebrados depredados por Tyto alba (Aves: Tytoni- dae) en el sudoeste de la provincia de Chubut (Ar- gentina). Neotropica 42:123. Donazar, J.A., A. Travaini, O. Ceballos, M. Delibes, and F. Hiraldo. 1997. Food habits of the Great Horned Owl in northwestern Argentine Patagonia: the role of introduced lagomorphs. /. Raptor Res. 31: 364-369. Everett, M., I. Prestt and R. Wagstaefe. 1992. Barn and Bay Owls Tyto, Pholidus. Pages 36-50 in J.A. Burton [Ed.], Owls of the world. Eurobook, Italy. Garcia Esponda, C.M., L.J.M. De Santis, J.I. Noriega, G.O. Pagnoni, G.J. Moreira, and N.M. Bertellotti 1998. The diet of Tyto alba (Strigiformes: Tytonidae) in the lower Chubut valley (Argentina) . Neotropica 44- 57-63. Herrera, C.M. and F.M. Jaksic. 1980. Feeding ecology of the Barn Owl in central Chile and southern Spain. A comparative study. Auk 97:760-767. Iriarte, J.A., W.L. Franklin, and W.E. Johnson. 1990. Diets of sympatric raptors in southern Chile./. Raptor Res. 24:41-46. Jaksic, F.M. 1986. Predation upon small mammals in shrublands and grasslands of southern South Ameri- ca: ecological correlates and presumable consequenc- es. Rev. Chil. Hist. Nat. 59:209-221. , R. Persico, and J. Torres. 1977. Sobre la parti- cion de recursos por las Strigiformes de Chile central. An. Mus. Hist. Nat. Valparaiso 10:185-194. Krebs, C.J. 1989. Ecological methodology. Harper and Row, New York, NYU.S.A. Levins, R. 1968. Evolution in changing environments: some theoretical explorations. Princeton Univ. Press, Princeton, NJ U.S.A. Marti, C.D, 1987. Raptor food habit studies. Pages 67- 80 in B.A. Giron Pendleton, B.A. Millsap, K.W. Cline, and D.M. Bird [Eds.], Raptor management tech- niques manual. Sci. Tech. Ser. 10. Natl. Wildl. Fed., Washington, DC U.S.A. Monjeau, J.A. 1989. Ecologia y descripcion geografica de los pequenos mamiferos del Parque Nacional Nahuel Huapi y areas adyacentes. Ph.D. dissertation, Univ. Nac. de La Plata, La Plata, Argentina. Noriega, J.I., R.M. Aramburu, E.R. Justo, and L.J.M. De Santis. 1993. Birds present in pellets of Tyto alba (Stri- giformes, Tytonidae) from Casa de Piedra, Argentina. J. Raptor Res. 27:37-38. Novaro, a, a. Capurro, A. Travaini, M. Funes, and J. Rabinovich. 1992. Pellet-count sampling based on spatial distribution: a case study of the European hare in Patagonia. Ecol. Aust. 2:11-18. Pearson, O.P. 1983. Characteristics of a mammalian fau- na from forests in Patagonia, southern Argentina. J Mammal. 64:476-492. . 1995. Annotated keys for identifying small mam- mals living or near Nahuel Huapi National Park or Lanin National Park, Southern Argentina. Mastozool. Neotrop. 2:99-148. Redford, K.H. and J.F. Eisenberg. 1992. Mammals of the Neotropics, the southern cone. Vol. 2. Univ. Chicago Press, Chicago, IL U.S.A. Taylor, I. 1994. Barn Owls. Predator-prey relationships 338 Short Communications VoL. 34, No. 4 and conservation. Cambridge Univ. Press, Cambridge, U.K. Tiranti, S.I. 1996. Small mammals from Chos Malal, Ne- uquen, Argentina, based upon owl predation and trapping. Tex. J. Sci. 48:303-310. Travaini, a., J.A. Donazar, O. Ceballos, A. RodrIguez, F. Hiraldo, and M. Delibes. 1997. Food habits of common Barn Owls along an elevational gradient in Andean Argentine Patagonia./. Raptor Res. 31:59-64. Trejo, A. and D. Grigera. 1998. Food habits of the Great Horned Owl {Bubo virginianus) in a Patagonian steppe in Argentina./. Raptor Res. 32:306-311. Received 10 February 2000; accepted 21 July 2000 Letters J. Raptor Res. 34(4):339-340 © 2000 The Raptor Research Foundation, Inc. Diet OF Breeding Northern Goshawks in the Coast Range of Oregon Northern Goshawks {Accipiter gentilis) were not known to breed in the Coast Range of western Oregon until June 1995, when two breeding pairs were reported (Thrailkill and Andrews 1996,/. Raptor Res. 30:248-249). Research suggests that Northern Goshawks generally nest in areas with high prey densities, that they forage opportunistically, and that their diets reflect the diversity of available prey species (Opdam 1975, Ardea 63:30-54; Hantage 1980, J Ornithol. 121:200-201; Widen et al. 1987, 49:233— 235; Kenward and Widen 1989, Pages 561-567 Meyburg and R.D. Chancellor [Eds.], Raptors in the modern world. World Working Group on Birds of Prey and Owls, London, U.K.; Boal and Mannan 1994, Stud. Avian Biol. 16:97—102). Thirty or more species of birds and mammals are known to be preyed upon by nesting Northern Goshawks; however, no dietary information exists for the Coast Range of Oregon (Block et al. 1994, Stud. Avian Biol. 16). The rarity of nesting goshawks in the Coast Range may be attributed to the dense vegetative structure of the area and its negative influence on prey availability (DeStefano and McCloskey 1997,/. Raptor Res. 31:34-39). We believe, therefore, it is important to document the diet of these goshawk pairs. We studied two goshawk nests that were approximately 16.1 km apart. There were three young in one nest and two young in the other. We collected prey remains (i.e., fur, feathers, skeletal parts) from plucking posts (stumps and tree branches in the nest stands), and nests from 7 June through mid-July 1995. Remains were combined and analyzed as a single sample for each nest site. Skeletal keys (Verts and Carroway 1984, Keys to the mammals of Oregon, 3rd Ed. Oregon State Univ. Book Stores Inc., Corvallis, OR U.S.A.) and museum collections (Dept, of Fisheries and Wildlife, Oregon State University, Corvallis, OR U.S.A.) were used to identify prey. We identihed 39 prey items, of which 84% were birds and 16% were mammals. Ruffed grouse {Bonasa umbellus) remains comprised 45% of total prey and 64% of bird prey items. Other bird species included Steller’sjay {Cyanoatta stelleri) (13%), American Robin (Turdus migratorius) (13%), Ring-necked Pheasant {Phasianus colchicus) (8%), and Mountain Quail (Oreortyx pictus) (5%). Mammalian species included Douglas’ squirrel {Tamiasciurus douglasii) (13%) and mountain beaver {Aplodontia rufa) (3%). We made no attempt to calculate biomass composition of the prey. Our data were consistent with that of other investigations. The variety of prey species was similar to other goshawk studies in the western United States. Avian prey comprised >50% of the diet during the breeding season. Our sample contained a relatively large percentage of avian prey, as much as 20% greater than some other studies (Meng 1959, Wilson Bull. 71:169-174, Opdam 1975; Reynolds and Meslow 1984, Auk 101:761-779; Bloom et al. 1986, Calif. Dept. Fish and Game, Wildl. Manage. Branch, Adm. Rept. 85-1; Widen et al. 1987; Bull and Homann 1994, Stud. Avian Biol. 16:103-105; Reynolds et al. 1994, Stud. Avian Biol. 16:106-113; Watson et al. 1998,/. Raptor Res. 32:297-305) We opportunistically collected prey items from plucking posts and nests after fledging took place, but we did not use direct observations to identify prey delivered to nests. The latter method can indicate a larger proportion of mammals in goshawk diets (Boal and Mannan 1994). The most accurate assessment of raptor diets is accomplished through a combination of direct visual observations and collection of prey remains (Marti 1987, Raptor food habit studies, Pages 67-80 in B.G. Pendleton, B.A. Millsap, K.W. Cline, and D.M. Bird [Eds.], Raptor management tech- niques manual. Nat. Wild. Fed., Sci. Tech. Ser. No. 10). Thus, our results may overestimate the proportion of avian prey. We suggest that the Ring-necked Pheasant in the diet of one of the two pairs of goshawks studied demonstrates the opportunistic foraging behavior of the goshawk. Although it was not observed, we suspect goshawks may have also foraged in agricultural/pastureland approximately 1.2 km from their nest where a hunting club released pheas- ants. We occasionally observed pheasants on forest roads near the site, so it is possible the pheasants were captured in the forest. Studies conducted in Europe show that pheasants are important goshawk prey. Habitat there consists of small woodlots surrounded by agricultural helds and pastures where pheasants are released (R. Kenward and P. Widen 1989). The relative absence of breeding goshawks from the Coast Range is well-documented (DeStefano and McCloskey 1997). They suggested three hypotheses to explain the absence of breeding goshawks from the Coast Range of Oregon. The hypothesis we believe most plausible is that the structure of the vegetation may limit prey availability to goshawks and thus prevent nesting, despite potentially suitable nesting substrate and adequate populations of prey. Two observations support this hypothesis. First, the similarity in the variety of prey species we found in this sample 339 340 Letters VoL. 34, No. 4 relative to other studies suggests that prey diversity is not a factor which precludes Northern Goshawk nesting in the Coast Range. Second, our field observations and those of other researchers indicate that dense understories are common throughout the Coast Range (Franklin and Dyrness 1973, Natural vegetation of Oregon and Washington. USDA For. Serv. Tech. Rep. PNW-8, Portland, OR U.S.A.; Reynolds and Meslow 1984; DeStefano and McCloskey 1997). This dense understory may decrease prey vulnerability to Northern Goshawks, reducing foraging efficiency and breeding success. We suspect that the hypothesized lack of suitable foraging habitat may be a limiting factor contributing to the low breeding density of goshawks in the Coast Range of Oregon. We thank J. Crawford, R. Jarvis, S. DeStefano, M. Collopy, E. Forsman, and R. Anthony for their contributions. — James A. Thrailkill, Lawrence S. Andrews, and Rita M. Claremont, Oregon Cooperative Fish and Wildlife Research Unit, Department of Fisheries and Wildlife, Oregon State University, Corvallis, OR 97331 U.S.A. J Raptor Res. 34(4):340-341 © 2000 The Raptor Research Foundation, Inc. First Dark Morph Brood of Montagu’s Harriers {Circus pygargus) in 14 Years in Italy The dark morph of the Montagu’s Harrier {Circus pygargus) is rare in Eurasia and its ecology and adaptive value is unclear. From 1986-99 (14 yr), I studied an Italian population of Montagu’s Harriers in a 120 000 ha study area on the eastern slopes of the Central Apennines. It is the southernrnost population on the Adriatic coast with the nearest population 70-80 km to the north. The size of this population has varied between 12-32 pairs annually (x = 21.7 pairs). The dark (melanistic) morph of Montagu’s Harrier is relatively frequent in the western portion of the European range. In France and Spain, it varies between 2-5% of all individuals in the passage at Gibraltar (Clarke, 1996, Montagu’s Harrier, Harlequin Press, Chelmsford, Essex, U.K.) and, in a population in the province of Burgos in northern Spain, as much as 10% of the population (88 nests examined in 1994-96, Sancho and Ansola Aristondo, 1998, Regulacion del caracter “plumaje oscuro” en una poblacion de Aguilucho cenizo {Circus pygargus) , Abstract, 5° Reunion Iberica sobre Aguiluchos, Evora, Portugal:! 1) is melanistic. The same is true in Portugal where as much as 20% of the population is the dark morph in Castroverde (Franco pers. comm.). Dark morphs occur less frequently in eastern and central Europe. Shirihai (1996, The birds of Israel, Academic Press, London, U.K.), for example, reported only 10 dark morphs among thousands of Montagu’s Harriers migrating through Israel in the 1970s and 1980s. We have no data for the number of dark morphs migrating at the Messina Strait but, in the migration along the Adriatic coast, no melanistic birds have been observed in three years of observations nor have they been observed at the neighboring Monte Conero site where 286 Montagu’s Harriers have ben observed (Pandolfi et al., 1998, Migrazione primaverile dei rapaci diurni nel Parco Naturale Regionale del San Bartolo (PS), “59° Congresso Na- zionale dell’U.Z.L,” Abstract:43) . The Italian breeding population of Montagu’s Harriers appears to belong to the eastern portion of the range in this respect. Indeed, in 13 yr of observations through 1999, I encountered only one melanistic male in 1989, and it was not a breeder but a vagrant seen only for a few days in a breeding site of three pairs. During this time, >1000 adult and young Montagu’s Harriers were observed in about 280 nests where nearly 400 young fledged. About 20% of these were nonbreeders and each observation/year was independent. Nevertheless, only this single case of a dark morph was observed (<0.1%). In the 1999 breeding season, while on a late visit to band and tag harriers, three melanistic, young Montagu’s Harriers were observed at a nest on 17 July at Monte della Mattera near Mombaroccio, a known breeding site occupied by the harriers almost every year since 1987. The parents of the three melanistic young were light morphs but all three young in the brood were completely dark, smokey brownish-black, except for the upper tail coverts, which were white as in normal brown juveniles and adult females. This last feature, a white mark at the upper end of the tail, is not normally depicted nor described in most field guides or general ornithological books. Indeed, even the two recent books (Clark, 1999, A field guide to the raptors of Europe, the Middle East and North Africa, Oxford Univ. Press, Oxford, U.K. and Forsman, 1999, The raptors of Europe and the Middle East, T. & A.D. Poyser, London, U.K.), make no mention of this white mark on the upper tail coverts. Sancho and Ansola Aristondo (1998 and pers. comm.) found no melanistic young produced from normal plumage adults. For this reason, they felt the dark plumage character was dominant. Sage (1962, Br. Birds 55:201-225) felt December 2000 Letters 341 that melanism was a heterozygous dominant trait. The three dark Montagu’s Harrier young we observed conflicted with this assessment. Perhaps, the nestlings in the brood resulted from extra pair copulations with a nonbreeding, melanistic male. During the 1999 breeding season, the Monte della Mattera site was monitored three times/wk with a mean of 5.5 hr of observation/d. Useful copulations can occur only during a female’s fertile period which, for the Montagu’s Harrier, is about 10 d; 7 d before the deposition of the first egg and ending about 1 d before deposition of the last egg (Pandolfi et al., 1998,/. Raptor Res. 32:269-277). During the 4-wk courtship period, which includes the fertile period, we made 65 hr of direct, on-site observations of the pair, but no melanistic bird was observed. In all the other breeding sites of the population (a total of 17 pairs recorded and 8-10 nonbreeding harriers), no melanistic individual was observed up to the departure of all birds on migration (first half of August). Extra-pair copulations in Montagu’s Harriers seem to be low. In fact, in the population we studied, they were recorded during only 3.4% of the copulations we observed (Pandolfi et al. 1998), while Arroyo has recorded 4-8% in Spain (Arroyo, 1999, Condor 101:340-346). Considering these data, it does not seem very likely, although it is remotely possible, that a single vagrant, dark male arrived in this heavily-monitored area during the breeding period, copulated with the female of an existing pair, and immediately left the territory vanishing. I would like to thank B. Arroyo, K. Bildstein, and an anonymous referee for improving the earlier version of this manuscript and for their helpful suggestions. — Massimo Pandolfi, Istituto di Scienze Morfologiche, Laboratory of Zoology, University of Urbino, 61029 Urbino, Italy. BOOK REVIEW Edited by Jeffrey S. Marks J. Raptor Res. 34(4) ;342 © 2000 The Raptor Research Foundation, Inc. Raptors at Risk. Edited by R.D, Chancellor and B.-U. Meyburg. 2000. Proceedings of the V World Conference on Birds of Prey and Owls. World Working Group on Birds of Prey and Owls, Berlin, Germany, and Hancock House Publishers, Blaine, WA. 895 pp., numerous figures and tables. ISBN 0- 88839-478-0. Paper, $50. The V World Conference on Birds of Prey and Owls was held in Johannes- burg, South Africa, from 4 to 11 August 1998. Or- ganized by the World Working Group on Birds of Prey and Owls (WWGBP), the Raptor Conserva- tion Group, and the Vulture Study Group of the South African Endangered Wildlife Trust, the con- ference was attended by more than 250 partici- pants. Of the 130 oral presentations and 35 posters presented at the meeting, 88 were published in the proceedings. The papers are organized into 13 sections: “Cur- rent Studies of African Raptors” (12 papers); “Bi- ology & Conservation of the Vultures of the World” (nine); “Falcons in Asia and the Middle East” (eight); “Conservation Models for Raptors of the World” (11); “Raptors in Urban Environ- ments” (six); “Understanding Distribution: the Whys and Wherefores of Geographical Ranges of Raptors” (three); “Predation and Feeding Ecolo- gy” (seven); “Conservation Biology of the World’s Migratory Raptors” (five); “Islands and Raptors” (10); “Impact of Electricity Utility Structures on Raptors” (five); “Biology of Owls with Emphasis on Vocalisations” (five); “Taxonomy, Phylogeny, De- velopments in Raptor DNA Studies and Other The- oretical Aspects” (four); and “General Studies” (three) . As with previous proceedings published by WWGBP, all of the papers are in English, and the covers are graced with beautiful color photographs (Indian Vulture {Gyps indicus indicus} on front, Lesser Spotted Eagle [Aquila pomarina] on back) . Space limitations prevent an in-depth review of the proceedings, so I will highlight only a small number of papers. The section devoted to African raptors contains significant new information on Bat Hawks {Macheiramphus alcinus) based on time- lapse video recording at a nest in South Africa (T. Harris, A. Kemp, and J. Dunning) and on Henst’s Goshawks {Accipiter henstii) gleaned from seven nests observed during three breeding seasons in Madagascar (Lily-Arison Rene de Roland) . Also of note are the papers by Bill Clark and R.A.G. Davies on taxonomic problems in African falconiforms and by Michael Wink and Hedi Sauer-Gurth on molecular systematics. The section on vultures con- tains detailed status reviews of species in Africa (P.J. Mundy), Asia (S.M, Satheesan), and Latin America (Marsha Schlee) plus Lloyd Kiff s review of the sta- tus of cathartids in North America. Six papers are devoted to Saker Falcons (Falco cherrug) in Asia and the Middle East. Owls receive little attention in this volume. The paper on molecular systematics by Mi- chael Wink and Petra Heidrich adds several species to the growing list of taxa whose phylogeny has heen evaluated based on mitochondrial DNA. Wink and Heidrich estimate that New World Otus (screech-owls) have been separated from Old World Otus (scops-owls) for 6-8 million years and suggest tbat the two groups should be placed in different genera. They also advocate merging Nyc- tea, Ketupa, and Scotopelia with Bubo. This volume continues the worthy series of pro- ceedings on the world’s raptors produced by Chan- cellor and Meyburg over the years and thus de- serves a place in university libraries and in the personal collections of raptor biologists. — Jeff Marks, Montana Cooperative Wildlife Research Unit, University of Montana, Missoula, MT 59812 U.S.A. 342 J. Raptor Res. 34(4) :343— 346 © 2000 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”) con- taining a single file that is identical with the print- ed copy. The electronic copy should be supplied as an IBM-compatible text file (ASCII, Word, or WordPerfect). Include information on the type of computer and word processor used. Manuscripts that are accepted upon condition of revision must be returned to the editor within 60 days. Manuscripts held longer will lose their pri- ority and may be treated as new submissions. The editor should be notified if extenuating circum- stances prevent a timely return of the manuscript. Authors will receive proofs of their articles prior to publication. Proofs must be read carefully to cor- rect any printer errors and returned by the fastest mail within two days of receipt TO THE EDITOR. Changes in typeset text are expensive and authors making extensive changes will be billed for the costs. A reprint order will accompany page proofs to enable 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 $100.00 U.S. per page). Authors who are not as- sociated with a research institution or simply do not have access to such grants may request the page charges be waived. Such a request can only be approved if the author is a member of the Foun- dation and the article is short. Payments of amounts less than the full page charges will be ac- cepted. Authors of long manuscripts are expected to pay publishing costs. It is unlikely that articles longer than 10 printed pages or 18 typewritten pages including tables and illustrations can be pub- lished without full payment. Authors employed by government agencies, universities, or firms that will meet reprint and page charges may forward a statement to the editor indicating intent to pay. Upon receipt of such a statement, reprints will be mailed to the author and the agency will be billed with the understanding that payment will be made within 30 days. All checks should be made payable to the Raptor Research Foundation, Inc. All per- sonal payments toward publication costs are tax- deductible in the United States. 343 344 Information for Contributors VoL. 34, No. 4 Journal of Raptor Research CHECKLIST FOR PREPARATION OF MANUSCRIPTS {check items and submit xvith manuscript) I. General Instructions (Consult recent issues for additional guidance on format) I I Type manuscripts on one side of either 216 x 278 mm (8.5 x 11”) or standard international size (210 X 297 mm) good quality paper (do not use erasable or lightweight paper) . Word-processor-generated manuscripts must be done with a letter-quality or near-letter- quality printer, DOUBLE SPACE THROUGHOUT including title, text, tables, figure legends, and literature cited. □ 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 regions. Do not give subspecific identification unless it is pertinent. Use lower case for all common names, including birds. □ Use American spelling and Webster’s Ninth New Colle- giate Dictionary (1983, Merriam-Webster, Inc.) as a spelling authority. □ Leave at least a 25 mm (1”) margin on all sides. Avoid hyphens or dashes at ends of lines; do not divide a word at the end of a line. □ Use a nonproportional font of at least elite size (4.7 characters/cm = 12 characters/inch) or 12 point, preferably Courier. DO NOT USE RIGHT JUSTIFI- CATION-LEAVE RIGHT MARGIN RAGGED. □ Use italic type for addresses, scientific names, journal names and third level headings. □ 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, 11 May to 11 June). □ Use 24-hour clock (e.g., 0800 H, 1345-1400 H) □ Write out numbers one to nine unless a measurement (e.g., four birds, 3 km, 40 sites, 6 yr). Use 1000 and 10 000; 0.15 instead of .15; % instead of percent. □ Each reference cited in text must be listed in the Lit- erature Cited section, and vice versa. Double check the accuracy of all entries — THE EDITORIAL STAFF CANNOT DO THIS FOR YOU. □ Literature citations in the text are as follows: a. One author-Jones (1993) or (Jones 1993) b. Two authors-Smith and Jones (1991) or (Smith and Jones 1991) c. Three or more authors-Hernandez et al. (1990) or (Hernandez et al. 1990) d. Manuscripts accepted for publication but not yet publish ed-Howard (in press) or (Howard in press) e. Unpublished materials-K. Jacobson (unpubl. data); (K. Jacobson pers. comm,); or K. Jacobson (pers. comm.) -do not place in the Literature Cited section. f. When citing several references within parentheses, separate with commas and put in chronological or- der, oldest first) . g. For manuscripts submitted as letters, place cita- tions in text in abbreviated form, e.g., (I.C. Birds 1993, J. Raptor Res. 27:45-50). □ Assemble manuscripts for regular articles in this or- der: (1) title page, (2) abstract page, (3) text, (4) ta- bles, (5) figure legends, (6) figures. DO NOT STA- PLE. II. Title Page □ Place full title 6-8 lines below top of page in all capital letters. Below title, center author’s name(s) and address (es) followed by a running title (short title) not to exceed 30 characters. If the author (s) is/are currently at another loca- tion from where the work was done, use super- script number(s) following author(s) name(s) to indicate current address in footnote at bot- tom of the page. In multiauthored papers, in- dicate the author responsible for correspon- dence and requests for reprints. Give phone number and, if possible, FAX number and e- mail address of the corresponding author. III. Abstract/summary □ For regular articles, include an abstract of about 250 words in one paragraph that is completely 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. December 2000 Information for Contributors 345 □ Include five to seven key words for indexing af- ter the abstract. □ Short communications will be printed with a Spanish summary only. Authors must provide an English summary to be translated into Span- ish unless they are fluent in Spanish. □ Avoid citing references in the abstract. If they must be cited, include journal name, volume, pages, and year, all in parentheses. IV. Text □ Follow instructions in section I. □ Main headings are all capital letters and flush with left margin. n Typical main headings for regular articles are; METHODS, RESULTS, and DISCUSSION. An introduction begins the text but does not have a heading. D 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. □ Short communications and letters may or may not have headings within the text depending upon the need. V. Literature Cited □ 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 first au- thor’s surname. References by a single author precede multiauthored works by the same se- nior author regardless of date. □ List works by the same author (s) chronologi- cally, beginning with the oldest. n Use a long dash when the author is the same as in the preceding citation. □ “In press” citations must have been accepted for publication and must include date, volume number and the name of the Journal or pub- lisher. □ Initials of second, third, and. . . authors precede their surname. □ Abbreviate journal names according to the Se- rial Sources for the BIOSIS Data Base (published annually by the BioSciences Information Ser- vice). □ Do not list personal communications and un- published reports. VI. Tables (Tables are expensive to print— keep them to a min- imum and put each on a separate page-try to de- sign them to fit a single column.) □ Double space throughout. Assign each table an Arabic number followed by a period. □ Table titles must be complete sentences. □ Use same size of type as in text. □ Indicate footnotes by lowercase superscript let- ters. □ Do not use vertical lines. VII. Figure Legends □ Print all figure legends on one page, double spaced. □ Number using Arabic numbers consecutively in the same order the figures appear in the text (i.e.. Figure 1, Figure 2., etc.). VIII. Preparation of Illustrations (Illustrations are referred to as figures and include drawings, graphs, and black and white half-tones [photographs] . CONSULT THE EDITOR IN AD- VANCE ABOUT COLOR.) □ Use professional standards in preparing figures; their reproduction in the JRR is virtually iden- tical to what is submitted. Consult issues of JRR for examples and Steps Toward Better Scientific Il- lustrations (Allen Press, P.O. Box 368, Lawrence, KS 66044) for more information. 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) . □ Submit drawings as original art (undiluted In- dia ink on heavy-weight, smooth-finish drafting paper) or as photomechanical transfers (PMTs) . Submit graphs as mechanical drawings or as high-resolution laser prints. Typed or handwritten text or symbols are not acceptable. Add text and symbols with press-on symbols and letters or mechanical lettering. Review copies of figures can be photographic copies but must ap- proach the quality of the original. □ Figure text must be a plain (sans serif) typeface (e.g., Helvetica), not compressed, and large enough so that it will be as large as the text type (8-10 point) when in print. 346 Information for Contributors VoL. 34, No. 4 D 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. O Use the same style of lettering and presentation for all figures. 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 and tables (if the manuscript was prepared using a word processor) . □ Submit to: Marc J. Bechard, Editor Department of Biology Boise State University Boise, Idaho 83725 U.S.A. More information? Telephone: 208 426-3530 FAX: 208 426-4267 E-mail: MBECHARD@BOISESTATE.EDU J. Raptor Res. 34(4):347-352 © 2000 The Raptor Research Foundation, Inc. Index to Volume 34 BY Elise Vernon Schmidt The index includes references to general, species, common names, key words and authors. Reference is also made to book reviews, dissertation and thesis abstracts, letters and reviewers, Taxa other than raptors are included where referenced by authors. A Abundance, raptor, 133-136 Acarina, 210-231 Accipiter badius, 249-261 brevipes, 249-261 soloensis, 249-261 Activity, 93-101 Aegolius, acadicus, 42—44, 299—304 Anderson, Stanley H., see Buhler, Matt L. Andrews, Lawrence S., see Thrailkill, James A. Annual fluctuations, 322-326 Aposematic coloration, 311-318 Appetite suppressant, 311-318 Applegate, Roger D., see Williams, Christopher K. Aquila chrysaetos, 48-52 Argentina, 108-119, 235-237, 237-241, 330-333, 334- 338 Arizona, 270-278 Arkansas, 26-32 Arsenault, David R, see Olson, Chad V. Asio clamator, 235—237 otus, 93-101 Attie, Carole, see Bretagnolle, Vincent Thomas Aviles, J. M., J. M. Sanchez and A. Sanchez, Breeding biology of the Eurasian Kestrel in the steppes of southwestern Spain, 45-48 B B^a California, 187-195 Bakaloudis, Dimitris E., Christos G. Vlachos, and Graham J. Holloway, Nest features and nest-tree characteris- tics of Short-toed Eagles {Circaetus gallicus) in the Dadia-Lefkimi-Soufli forest, northeastern Greece, 293-298 Balaquit-Ibanez, Gliceria A., see Miranda, Hector C., Jr. Banding and marking, 262-269 Barton, Nigel W. H., Trapping estimates for Saker and Peregrine Falcons used for falconry in the United Arab Emirates, 53-55 Bednarz, James C., see Garner, Heath D. Behavior, 120-125 Bellocq, M. Isabel, A review of the trophic ecology of the Barn Owl in Argentina, 108-119 Bird, David M., see Nicholls, Michael K. Block, William M., see Ganey, Joseph L. Bo, Maria S., Sandra M. Cicchino, and Mariano M. Mar- tinez, Diet of breeding Cinereous Harriers {Circus dnereus) in southeastern Buenos Aires Province, Ar- gentina, 237-241 Bo, Maria S., see Isacch, Juan P. Bonin, 241-243 Book Reviews, 153-155, 247-248, 342 Bortolotti, Gary, see Murza, Gillian L. Bortolotti, Gary, see Miller, Michael J. R. Breeding, 56-57 biology, 299-304 diet, 237-241 dispersal, 262-269 success, 37-41, 67-74, 322-326 Bretagnolle, Vincent, Thomas Ghestemme, Jean-Marc Thoiollay, and Carole Attie, Distribution, population size and habitat use of the Reunion Marsh Harrier, Circus m. maillardi, 8-17 Bubo bubo, 232-235, 305-310 Buhler, Matt L., Jake H. Powell, and Stanley H. Anderson, Golden Eagle pair kills Ferruginous Hawk in Wyo- ming, 245-246 Bustamante, Javier, see Juan Jose Negro Buteo albigula, 143-147 buteo toyoshimai, 241-243 jamaicensis, 26—32, 203-209 lagopus, 157-166 lineatus, 18-25 Buzzard, Ogasawara, 241-243 C California, 187-195 northwestern, 75-84 Call, food and contact, 232-235 practice, 232-235 Calling activity, diurnal, 232-235 Calvo, Jose Francisco, see Carrete, Martina Canyonlands, 1-7 Caracara, Crested, 330-333 Caracara plancus, 330-333 Carpenter, Leslie B., see Lehman, Robert N. Carrete, Martina, Jose Antonio Sanchez-Zapata, and Jose Francisco Calvo, Breeding densities and habitat at- tributes of Golden Eagles in southeastern Spain, 48- 52 347 348 Index to Volume 34 VoL. 34, No. 4 Cazassus, Helene, see Penteriani, Vincenzo Census, occupied nest, 232-235 Chile, 143-147 Chubbs, Tony E., Bruce Mactavish, Keith Oram and Per- ry G. Trimper, First confirmed breeding records and other incidental sightings of Northern Harriers in Labrador, 56—57 Cicchino, Sandra M., see Bo, Maria S. Cirraetus gallicus, 293-298 Circus cinereus, 237-241 cyaneus, 56-57, 203-209 m. maillardi, 8-17 Claremont, Rita M., see Thrailkill, James A. Clutch size, 45-48 Commentary, 62-63 Condition, 137-142 Condor, Andean, 33—36 Contreras-Balderas, Armando J., see Ruiz-Campos, Gor- gonio Cover type selection, 203-209 Cropland, 203-209 CTA, 311-318 D Dawson, Russell D., see Murza, Gillian L. De Vries, Tjitte, and Cristian Melo, First nesting record of the nest of a Slaty-backed Forest-Falcon {Micrastur mirandollei) in Yasuni National Park, Ecuadorian Am- azon, 148-150 Density, 203-209, 241-243 Desert, Great Basin, 133-136 Mojave, 133-136 Diet, 42-44, 108-119, 287-292, 334-338 Diller, Lowell V, see Folliard, Lee B. Dispersal, natal, 1-7 timing, 1-7 Distribution, 56-57 winter, 157-166 Dobler, Frederick C., see Wilson, Ulrich W. Doremus, John H,, see Marks, Jeffrey S. Dzus, Elston H., see Miller, Michael J. R. E Eagle, Bald, 167-174, 287-292 Golden, 48-52 Philippine, 37-41 Short-toed, 293-298 Eagles, 210-231 Ecuador, 33-36, 148-150 El Nino, 67-74 Endangered species, 126-132 Endemic, 241-243 Erickson, Richard A., see Patten, Michael A. Eucalyptus, 18-25 European rabbit, 305-310 Exotic trees, 18-25 F Falco cherrug, 53-55 mexicanus, 262-269 naumanni, 327-329 peregrinus, 53-55, 67-74, 126-132 sparvenus, 137-142, 203-209, 311-318 tinnunculus, 45—48, 319-321 Falcon, Peregrine, 53-55, 67-74, 126-132 Prairie, 262-269 Saker, 53-55 Falconiformes, 210-231 Falconry, 53-55 Falcons, 210-231 Fattorini, Simone, see Salvati, Luca Fish-Owl, Tawny, 102-107 Folliard, Lee B., Kerry P. Reese, and Lowell V. Diller, Landscape characteristics of Northern Spotted Owl nest sites in managed forests of northwestern Cali- fornia, 75-84 Food choice, 311-318 habits, 120-125, 167-174, 196-202, 235-237, 287-292 stress, 305-310 Foraging, 137-142 behavior, 327-329 habitat, 175-186 Forest, managed, 175-186 temperate, 143-147 Forest-Falcon, Barred, 196-202 Collared, 196-202 Slaty-backed, 148-150 Forest fragmentation, 37-41 Freeman, Pamela L., Identification of individual Barred Owls using spectrogram analysis and auditory cues, 85-92 G Gallardo, Max, see Penteriani, Vincenzo Ganey, Joseph L., William M. Block, and Rudy M. King, Roost sites of radio-marked Mexican Spotted Owls in Arizona and New Mexico: Sources of variability and descriptive characteristics, 270—278 Garner, Heath D. and James C. Bednarz, Habitat use by Red-tailed Hawks wintering in the delta region of Arkansas, 26-32 Ghesterame, Thomas, see Bretagnolle, Vincent Goldstein, Michael I., Nest-site characteristics of Crested Caracaras in La Pampa, Argentina, 330-333 Grande, Juan Manuel, see Negro, Juan Jose Grasslands, 330-333 Greece, 293-298 Grubb, Teryl G., and Roy G. Lopez, Food habits of Bald Eagles wintering in northern Arizona, 287-292 December 2000 Index to Volume 34 349 H Habitat, 48-52, 279-286 selection, 102-107 use, 8-17, 26-32, 93-101 Haliaeetus leucocephalus, 167—174, 287-292 Handicaps, 137-142 Harness, Richard, a review of: Birds and Power Lines, Edited by Miguel Ferrer and Guyonne F. E. Janss, 1999, 154-155 Harrier, Cinereous, 237-241 Northern, 56-57, 203-209 Reunion Marsh, 8-17 Harrier-Hawk, Madagascar, 120-125 Hawk, Grey Frog, 249-261 Harris’, 187-195 Red-shouldered, 18-25 Red-tailed, 26-32, 203-209 Rough-legged, 157-166 White-throated, 143-147 Hawks, 210-231 Henrioux, Fabienne, Home range and habitat use by the Long-eared Owl in northwestern Switzerland, 93- 101 Herremans, Marc, and Michel Louette, A partial post- juvenile molt and transitional plumage in the Shikra {Accipiter badius) and Grey Frog Hawk {Accipiter so- loensis), 249—261 Herter, Dale R. and Lorin L. Hicks, Barred Owl and Spot- ted Owl populations and habitat in the central Cas- cade Range of Washington, 279-286 Hicks, Lorin L., see Herter, Dale R. Holloway, Graham J., see Bakaloudis, Dimitris E. Holt, Eric A., see Lederle, Patrick E. Home range, 93-101 I Ibanez, Jayson C., see Miranda, Hector C., Jr. Idaho, 299-304 Indian Ocean, 8-17 Individual identification, 85-92 Interbreeding, 279-286 Irwin, Larry L., Dennis F Rock, and Gregory P. Miller, Stand structures used by Northern Spotted Owls in managed forests, 175-186 Isacch, Juan R, Maria S. Bo, and Mariano M. Martinez, Food habits of the Striped Owl (Asio clamator) in Buenos Aires Province, Argentina, 235-237 J Johnston, David W., see Whalen, David M. K Kato, Yuka, see Suzuki, Tadashi Kennedy, Patricia L., a review of: The Northern Goshawk: Ecology, Behavior and Management in North Amer- ica, By Thomas Bosakowski, 1999, 153-154 Kestrel, American, 137-142, 203-209, 311-318 Eurasian, 45-48, 319-321 Lesser, 327-329 Ketupa Jlavipes, 102-107 King, Rudy M., see Ganey, Joseph L. Kochert, Michael N., see Lehman, Robert N. Koenen, Marcus T, Koenen, Sarah Gale, and Norma Yanez, An evaluation of the Andean Condor popu- lation in northern Ecuador, 33-36 Koenen, Sarah Gale, see Koenen, Marcus T. L La Marca, Guiseppe, see Thorstrom, Russell Labrador, 56-57 Landscape pattern, 75-84 Latitudinal segregation, 157-166 Laying date, 45-48 Lead exposure, 167-174 Lederle, Patrick E., James M. Mueller, and Eric A. Holt, Raptor surveys in southcentral Nevada, 1991-1995, 133-136 Lee, Ching-Feng, see Sun, Yuan-hsun Lehman, Robert N., Karen Steenhof, Leslie B. Carpenter, and Michael N. Kochert, Turnover and dispersal of Prairie Falcons in southwestern Idaho, 262-269 Letters, 58-61, 151-152, 244-246, 339-341 Lopez, Roy G., see Grubb, Teryl G. Louette, Michel, see Marc Herremans Love, Oliver R, see Nicholls, Michael K. Lutz, R. Scott, see Williams, Christopher K. M Maceda, Juan Jose, see Sarasola, Jose Hernan Mactavish, Bruce, see Chubbs, Tony E. Managed forest, 75-84 Manganaro, Alberto, see Ranazzi, Lamberto Manganaro, Alberto, see Salvati, Luca Manuscript referees, 64 Margalida, Antoni, see Juan Jose Negro Marks, Jeffrey S., Book Reviews, 153-155 Marks, Jeffrey S. and John H. Doremus, Are Northern Saw-whet Owls nomadic?, 299-304 Marks, Jeffrey S., a review of Raptors at Risk, Ed. By R D. Chancellor and B.-U. Meyburg, 2000, 342 Martell, Mark S., Jennifer L. McNicoll and Patrick T. Re- dig, Probable effect of delisting of the Peregrine Fal- con on availability of urban nest sites, 126-132 Marti, Carl D., a review of Handbook of the Birds of the World, Volume 5. Barn-owls to Hummingbirds, Ed. by Josep del Hoyo, Andrew Elliott, and Jordi Sarga- tal, 1999, 247-248 Martinez, Mariano M., see Bo, Maria S. Martinez, Mariano M., see Isacch, Juan R 350 Index to Volume 34 VoL. 34, No. 4 Mason, J. R., Golden Eagle attacks and kills adult male coyote, 244-245 McMillan, Anita, see Wilson, Ulrich W. McNicoll, Jennifer L., see Martell, Mark S. Mediterranean, 48-52, 305—310 areas, 322—326 Melguizo, Giro, see Negro, Juan Jose Melo, Criatian, see De Vries, Tjitte Micrastur, mirandollei, 148-150 ruficollis, 196-202 semitorquatus, 196-202 Middle East, 53-55 Migration, 42-44, 143-147 differential, 157-166 Miller, Gregory R, see Irwin, Larry L. Miller, Michael J. R., Mark E, Wayland, Elston H. Dzus, and Gary R. Bortolotti, Availability and ingestion of lead shotshell pellets by migrant Bald Eagles in Sas- katchewan, 167-174 Miranda, Hector C., Jr., Salvador, Dennis L, Jayson C. Iba- nez, and Gliceria A. Balaquit-Ibanez, Summary of Philippine Eagle reproductive success, 1978—98 Mites, 210-213 Molt, contour, 249-261 Mueller, James M., see Lederle, Patrick E. Murza, Gillian L., Gary R. Bortolotti and Russell D. Daw- son, Handicapped American Kestrels: Needy or pru- dent foragers? 137-142 N Natal dispersal, 262-269 Negro, Juan Jose, and Antoni Margalida, How Bearded Vultures ( Gypaetus barbatus) acquire their orange col- oration: A comment on Xirouchakis (1998), 62-63 Negro, Juan Jose, Javier Bustamante, Giro Melguizo, Jose Luis Ruiz, and Juan Manuel Grande, Nocturnal ac- tivity of Lesser Kestrels under artificial lighting con- ditions in Seville, Spain, 327-329 Nest, 56-57 characteristics, 330-333 features, 293-298 first record, 148-150 site fidelity, 262-269 site selection, 75-84 stick, 148-150 tree characteristics, 293-299 Nesting biology, 120-125 habitat, 175-186 period, 319-321 Nevada, 133-136 New Mexico, 270-278 Niche breadth, 196-202 overlap, 196-202 Nicholls, Michael K., Oliver P. Love and David M. Bird, An evaluation of methyl anthranilate, aminoaceto- phenone, and unfamiliar coloration as feeding re- pellents to American Kestrels, 311-318 Nocturnal activity, 327-329 Nomadism, 299-304 O Olson, Ghad V, and David P. Arsenault, Differential win- ter distribution of Rough-legged Hawks {Buteo lago- pus) by sex in western North America, 157-166 Olson, Ghad V., and Sophie A. H. Osborn, First North American record of a melanistic female Northern Harrier, 58-59. Oram, Keith, see Ghubbs, Tony E. Oregon, 175-186 Oryctolagus cuniculus, 305-310 Osborn, Sophie A. H., see Olson, Ghad V. Owl, Barn, 108-119, 334-338 Barred, 85-92, 279-286 Eagle, 232-235, 305-310 Long-eared, 93-101 Mexican Spotted, 1-7, 270-278 Northern Saw-whet, 42-44, 299-304 Northern Spotted, 75-84, 175-186, 279-286 Striped, 235-237 Tawny, 322-326 Owls, 210-231 P Palacios, Gesar-Javier, Decline of the Egyptian Vulture {Neophron percnopterus) in the Ganary Islands, 61 Pandolfi, Massimo, First dark morph brood of Montagu’s Harriers {Circus pygargus) in 14 years in Italy, 340- 341 Parabuteo unicinctus, 187-195 Paramo, 33-36 Parasites, 210-231 Parrish, John W., Possible prevention of European Star- ling nesting by Southeastern American Kestrels at a power substation in southern Georgia, 152 Patagonia, 334—338 Patten, Michael A., and Richard A. Erickson, Population fluctuations of the Harris’ Hawk {Parabuteo unicinc- tus) and its reappearance in Galifornia, 187-195 Pavez, Eduardo F., Migratory movements of the White- throated Hawk {Buteo albigula) in Ghile, 143—147 Pellet analysis, 42-44, 287-292 Pellets, 287-292 Penteriani, Vincenzo, Max Gallardo and Helene Gazas- sus, Diurnal vocal activity of young Eagle Owls and its implications in detecting occupied nests, 232-235 Philips, James R., A review and checklist of the parasitic mites (Acarina) of the Falconiformes and Strigifor- mes, 210-231 Pillado, Maria S., and Ana Trejo, Diet of the Barn Owl {Tyto alba tuidara) in northwestern Argentine Pata- gonia, 334-338 December 2000 Index to Volume 34 351 Pithecophaga jefferyi, 37-41 Playback, 319-321 Plumage, transitional post-juvenile, 249-261 Polyboroides radiatus, 120-125 Population decline, 37-41 fluctuations, 187-195 size, 8-17 structure, 33—36, trend, 67-74 turnover, 262-269 Populations, 279-286 Powell, Jake H., see Buhler, Matt L. Predation, intraguild, 305—310 Prey abundance, 26-32 deliveries, 327-329 Productivity evaluation, 232-235 R Radio telemetry, 93-101, 102-107, 270-278 Ranazzi, Lamberto, Alberto Manganaro, and Luca Salvati, The breeding success of Tawny Owls {Strix aluco) in a Mediterranean area; A long-term study in urban Rome, 322-326 Rangeland, 203-209 Redig, Patrick T., see Martell, Mark S. Redwoods, 75-84 Reese, Kerry P., see Folliard, Lee B. Repeated measures, 270-278 Reproductive rates, 45-48 success, 18-25 Riparian, 18-25 Rock, Dennis F., see Irwin, Larry L. Roost sites, 270—278 Rottenborn, Stephen C., Nest-site selection and repro- ductive success of urban Red-shouldered Hawks in central California, 18-25 Ruiz, Jose Luis, see Negro, Juan Jose Ruiz-Campos, Gorgonio and Armando J. Contreras-Bald- eras, New northern nesting record of the Peregrine Falcon in B^a California, Mexico, 151 Rusch, Donald H., see Williams, Christopher K. S Salvador, Dennis L, see Miranda, Hector C., Jr. Salvati, Luca, Alberto Manganaro and Simone Fattorini, Responsiveness of nesting Eurasian Kestrels Falco tin- nunculus to call playbacks, 319-321 Salvati, Luca, see Ranazzi, Lamberto Sanchez, A., see Aviles, J. M. Sanchez, J. M., see Aviles, J. M. Sanchez-Zapata, Jose Antonio, see Carrete, Martina Sarasola, Jose Hernan, Ramon Alberto Sosa, and Juan Jose Maceda, A case of nest predation on Turkey Vultures nesting in Argentina, 60 Saskatchewan, 137-142, 167-174 Seabird colonies, 67-74 Seavy, Nathaniel E., Observations at an Ayres’ Hawk-Ea- gle nest in Kibale National Park, Uganda, 59-60 Serrano, David, Relationship between raptors and rabbits in the diet of Eagle Owls in southwestern Europe Competition removal or food stress?, 305-310 Shikra, 249-261 Short Communications, 37-57, 148-150, 232-243, 319- 338 Shotshell, lead pellets, 167-174 Siblicide, 120-125 Songs, 85-92 Sosa, Ramon Alberto, see Sarasola, Jose Hernan Spain, 45-48 Sparrowhawk, Levant, 249-261 Steenhof, Karen, see Lehman, Robert N. Steppe habitat, 45-48 Stomach-content analysis, 42-44 Strigiformes, 210-231 Strix aluco, 322-327 occidentalis, 279-286 0 . caurina, 75-84, 175-186 0 . lucida, 1-7, 270-278 varia, 85-92, 279-286 Sun, Yuan-Hsun, Wng Wang, and Ching-Feng Lee, Hab- itat selection by Tawny Fish-Owls {Ketupa Jlavipes) in Taiwan, 102-107 Survey technique, 319-321 Surveys, helicopter, 67-74 roadside, 133-136 roadside raptor, 157-166 Suzuki, Tadashi and Yuka Kato, Abundance of the Oga- sawara Buzzard on Chichijima, the Pacific Ocean, 241-243 Switzerland, 93—101 T Taste aversion, conditioned, 311-318 Telemetry, 1-7 Territory, 102-107 Thiollay, Jean-Marc, see Bretagnolle, Vincent Thorstrom, Russell, The food habits of sympatric forest- falcons during the breeding season in northeastern Guatemala, 196—202 Thorstrom, Russell and Guiseppe La Marca, Nesting bi- ology and behavior of the Madagascar Harrier-Hawk {Polyboroides radiatus) in northeastern Madagascar, 120-125 Thrailkill, James A., Lawrence S. Andrews and Rita M Claremont, Diet of breeding Northern Goshawks in the Coast Range of Oregon, 339-340 Transect, line, 203-209 Trapping, 53-55 Trejo, Ana, see Pillado, Maria S. Trimper, Perry G., see Chubbs, Tony E. Trophic ecology, 108-119 niche breadth, 237-241 352 Index to Volume 34 VoL. 34, No. 4 Tyto alba, 108-119, 334-338 U Urban nesting, 126-132 Rome, 322-326 V Van Riper, Charles, see Willey, David W. Variation, sources of, 270-278 Visual perception, 311-318 Vlachos, Christos G., see Bakaloudis, Dimitris E. Vocalizatons, 85-92 Vultur gryphus, 33-36 Vulture, 33-36 Vultures, 210-231 W Wang, Ymg, see Sun, Yuan-hsun Washington, 279-286 Watts, Bryan D., see Whalen, David M. Wayland, Mark E., see Miller, Michael J. R. Whalen, David M., Bryan D. Watts, and David W. John- ston, Diet of autumn migrating Northern Saw-whet Owls on the eastern shore of Virginia, 42-44 Williams, Christopher K., Roger D. Applegate, R. Scott Lutz and Donald H. Rusch, A comparison of raptor densities and habitat use in Kansas cropland and rangeland ecosystems, 203-209 Willey, David W., and Charles van Riper III, First-year movements by juvenile Mexican Spotted Owls in the canyonlands of Utah, 1-7 Wilson, Ulrich W., Anita McMillan, and Frederick C Dobler, Nesting, population trend and breeding suc- cess of Peregrine Falcons on the Washington outer coast, 1980-98, 67-74 Winter, 26-32 habitat, 287-292 roosts, 287-292 Y Yanez, Norma, see Koenen, Marcus T. THE JOURNAL OF RAPTOR RESEARCH A QUARTERLY PUBLICATION OF THE RAPTOR RESEARCH FOUNDATION, INC. (Founded 1966) EDITOR IN CHIEF MarcJ. Bechard ASSOCIATE EDITORS Gary R. Bortolotti Allen M. Fish Charles J. Henny Juan Jose Negro Ian G. Warrentin Fabian Jaksic Daniel E. Varland Cole Crocker-Bedford James C. Bednarz Marco Restani BOOK REVIEW EDITOR Jeffrey S. Marks CONTENTS FOR VOLUME 34, 2000 Number 1 First-year Movements by Juvenile Mexican Spotted Owls in the Canyonlands of Utah. David W. Willey and Charles van Riper III 1 Distribution, Population Size and Habitat Use of the Reunion Marsh Harrier, Circus M. MAILLARDI. Vincent Bretagnolle, Thomas Ghestemme, Jean-Marc Thiollay and Carole Attie 8 Nest-site Selection and Reproductive Success of Urban Red-shouldered Hawks IN Central California. Stephen c. Rottenbom 18 Habitat Use by Red-tailed Hawks Wintering in the Delta Region of Arkansas. Heath D. Garner and James C. Bednarz 26 An Evaluation of the Andean Condor Population in Northern Ecuador. Marcus T. Koenen, Sarah Gale Koenen and Norma Yanez 33 Short Communications Summary OF Philippine Eagle Reproductive Success, 1978-98. Hector C. Miranda, Jr., Dennis 1. Salvador, Jayson C. Ibanez and Gliceria A. Balaquit-Ibahez 37 Diet of Autumn Migrating Northern Saw-whet Owls on the Eastern Shore of Virginia. David M. Whalen, Bryan D. Watts and David W. Johnston 42 Breeding Biology of the Eurasian Kestrel in the Steppes of Southwestern Spain. J.M. Aviles, J.M. Sanchez and A. Sanchez 45 Breeding Densities and Habitat Attributes of Golden Eagles in Southeastern Spain. Martina Carrete, Jose Antonio Sanchez-Zapata and Jose Francisco Calvo 48 Trapping Estimates for Saker and Peregrine Falcons Used for Falconry in the United Arab Emirates. Nigel W.H. Barton 53 First Confirmed Breeding Records and Other Incidental Sightings of Northern Harriers in Labrador. Tony E. Chubbs, Bruce Mactavish, Keith Oram and Perry G. Trimper 56 Letters 58 Commentary 62 Number 2 Nesting, Population Trend and Breeding Success of Peregrine Falcons on the Washington Outer Coast, 1980—98. Ulrich W. Wilson, Anita McMillan and Frederick c. Dobler 67 Landscape Characteristics of Northern Spotted Owl Nest Sites in Managed Forests of Northwestern California. Lee b. Foiiiard, Kerry p. Reese and Loweii v. Diiier .... 75 Identification of Individual Barred Owls Using Spectrogram Analysis and Audi- tory Cues. Pamela L. Freeman 85 Home Range and Habitat Use by the Long-eared Owl in Northwestern Switzer- land. Fabienne Henrioux 93 Habitat Selection by Tawny Fish-Owls {Ketupa flavipes) in Taiwan. Yuan-Hsun Sun, Ying Wang and Ching-Feng Lee 102 A Review of the Trophic Ecology of the Barn Owl in Argentina, m. Isabel Beiiocq .... 108 Nesting Biology and Behavior of the Madagascar Harrier-Hawk (Polyboroides RADIATUS ) IN NORTHEASTERN MADAGASCAR. Russell Thorstrom and Guiseppe La Marca 120 Probable Effect of Delisting of the Peregrine Falcon on Availability of Urban Nest Sites. Mark S. Martell, Jennifer L. McNicoll and Patrick X Redig 126 Raptor Surveys in Southcentral Nevada, 1991-95. Patrick e. Lederie James m. Mueller and Eric A. Holt 133 Handicapped American Kestrels: Needy or Prudent Foragers? Gillian l. Murza, Gary R. Bortolotti and Russell D. Dawson 1 37 Migratory Movements of the White-throated Hawk {Buteo albigula) in Chile. Eduardo F. Pavez 143 Short Communications First Nesting Record of the Nest of a Slaty-Backed Forest-Falcon {Micrastur mirandollei) in Yasuni National Park, Ecuadorian Amazon. Tjitte de Vries and Cristian Melo 148 Letters 151 Book Reviews. Edited by Jeffrey S. Marks 153 Number 3 Differential Winter Distribution of Rough-Legged Hawks {Buteo lagopus) by Sex in Western North America, chad v. oison and David p. Arsenault 157 Availability and Ingestion of Lead Shotshell Pellets by Migrant Bald Eagles in Saskatchewan. MichaelJ.R. Miller, MarkE.Wayland, Elston H.Dzus, and Gary R. Bortolotti 167 Stand Structures Used by Northern Spotted Owls in Managed Forests. Larry l. Irwin, Dennis F. Rock, and Gregory P. Miller 175 Population Fluctuations of the Harris’ Hawk {Parabuteo unicinctus) and its Reappearance in California. Michael A. Patten and Richard A. Erickson 187 The Food Habits of Sympatric Forest-Falcons During the Breeding Season in Northeastern Guatemala. Russell Thorstrom 196 A Comparison of Raptor Densities and Habitat Use in Kansas Cropland and RaNGEIAND Ecosystems. Chrisopher K. Williams, Roger D. Applegate, R. Scott Lutz, and Donald H. Rusch 203 A Review and Chegkitst of the Parasitic Mites (Acarina) of the Falconiformes AND STRIGIFORMES. James R. Philips 210 Short Communications Diurnal Vocal Activity of Young Eagle Owls and its Impucations in Detecting Occupied Nests. Vincenzo Penteriani, Max Gallardo, and Helene Gazassus 232 Food Habits of the Striped Owl {Asio clamator) in Buenos Aires Province, Argentina. Juan P. Isacch, Maria S. B6, and Mariano M. Martinez 235 Diet of Breeding Ginereous Harriers ( Circus cinereus) in Soui heastern Buenos Aires Province, Argentina. Maria S. Bo, Sandra M. Cicchino, and Mariano M. Martinez 237 Abundance of the Ogasawara Buzzard on Chichijima, The Pacific Ocean. Tadashi Suzuki and Yuka Kato 241 Letters 244 Book Review. Edited by Jeffrey S. Marks 247 Number 4 A Partial. PostJuvenile Molt and Transitional Plumage in the Shikra {Accipiter BADIUS) AND GrEY FrOG HaWK {AcCIPUER SOLOENSIS) . Marc Herremans and Michel Louette .. 249 Turnover and Dispersal of Prairie Falcons in Southwestern Idaho. Robert n. Lehman, Karen Steenhof, Leslie B. Carpenter, and Michael N. Kochert 262 Roost Sites of Radio-Marked Mexican Spotted Owls in Arizona and New Mexico: Sources of Variability and Descriptive Characteristics. Joseph l. Ganey, WiiUam m. Block, and Rudy M. King 270 Barred Owi. and Spotted Owl Populations and Habitat in the Central CascvM)E Range of Washington. Dale R. Herter and Lorin L. Hicks 279 Food Habits of Bald Eagles Wintering in Northern Arizona. Teryi g. Gmbb and Roy G. Lopez 287 Nest Features and Nest-Tree Characteristics of Short-Toed Eagles ( Circaetus GALLicus) in the Dadia-Lefkimi-Soufli Forest, Northeastern Greece. Dimitris e. Bakaloudis, Christos G. Vlachos, and Graham J. Holloway 293 Are Northern Saw-Whet Owls Nomadic? Jeffrey s. Marks and John h. Doremus 299 Relationship Between Raptors and Rabbits in the Diet of Eagle Owls in Southwestern Europe: Competition Removal or Food Stress? David Serrano 305 An Evaluation of Methyl Anthranii ate, Aminoacetophenone, and Unfamiuar Coloration as Feeding Repellents to American Kestrels. Michael k. Nichoiis, Oliver P. Love, and David M. Bird 311 Short Communications Responsiveness oe Nestinc; Eurasian Kestrels Falco tinnuncmlus to Call Playbacks. Luca Salvati, Alberto Manganaro, and Simone Fattorini 319 The Breeding Sui^cess of Tawny Owls (Strix ai.uco) in a Mediterranean Area: A Long- Term Study in Urban Rome. Lamberto Ranazzi, Alberto Manganaro, and Luca Salvati 322 Nocturnal Activity of Lesser Kestrels Under Artificial Lighting Conditions in Seville, Spain. Juan Jose Negro, Javier Bustamante, Giro Melguizo, Jose Luis Ruis, and Juan Manuel Grande 327 Nest-So e Characteristics of Crested Caracaras in La Pampa, Argentina. Michael I. Goldstein 330 Diet of the Barn Owl {Tyto alba tutdara) in Northwestern Argentine Patagonia. Maria S. Pillado and Ana Trejo 334 Letters 339 Book Review. Edited by Jeffrey S. Marks 342 BUTEO BOOKS The following Birds of North America Species Accounts are available through Buteo Books, 3130 Laurel Road, Shipman, VA 22971. TOLL-FREE ORDERING: 1-800-722-2460; FAX: (804) 263-4842. E-mail: alien® buteobooks.com Bald Eagle (506). David A. Buehler. 2000. 40 pp. Barn Owl (1). Carl D. Marti. 1992. 16 pp. Barred Owl (508). Kurt M. Mazur and Paul C. James. 2000, 20 pp. Black Vulture (411). NeilJ. Buckley. 1999. 24 pp. Boreal Owl (63). G.D. Hayward and P.H. Hayward. 1993. 20 pp. Broad-winged Hawk. (218). L.J. Goodrich, S.C. Crocoll and S.E. Senner. 1996. 28 pp. Burrowing Owl (61). E.A. Haug, B.A. Millsap and M.S. Martell. 1993. 20 pp. Common Black-Hawk (122). Jay H. Schnell. 1994. 20 pp. Cooper’s Hawk (75). R.N. Rosenheld and J. Bielefeldt. 1993. 24 pp. Crested Caracara (249). Joan L. Morrison. 1996. 28 pp. Eastern Screech-owl (165). Frederick R. Gehlbach. 1995. 24 pp. Elf Owl (413). Susanna G. Henry and Frederick R. Gehlbach. 1999. 20 pp. Ferruginous Hawk (172). MarcJ. Bechard and Josef K. Schmutz. 1995. 20 pp. Ferruginous Pygmy-Owl (498), Glenn A. Proudfoot and R. Roy Johnson. 2000. 20 pp. Flammulated Owl (93). D. Archibald McCallum. 1994. 24 pp. Great Gray Owl (41). Evelyn L. Bull and James R. Duncan. 1993. 16 pp. Great Horned Owl (372). C. Stuart Houston, Dwight G. Smith, and Christoph Rohner. 1998. 28 pp. Gyrfalcon (114). Nancy J. Glum and Tom J. Cade. 1994. 28 pp. Harris’ Hawk (146). James C. Bednarz. 1995. 24 pp. Hawaiian Hawk (523). Kenneth E. Clarkson and Leona P. Laniawe. 2000. 16 pp. Long-eared Owl (133). J.S. Marks, D.L. Evans and D.W. Holt. 1994. 24 pp. Merlin (44). N.S. Sodhi, L. Oliphant, P. James and I. Warkentin. 1993. 20 pp. Mississippi Kite (402). James W. Parker. 1999. 28 pp. Northern Saw-whet Owl (42). Richard J. Cannings. 1993. 20 pp. Northern Goshawk (298). John R. Squires and Richard T. Reynolds. 1997. 32 pp. Northern Harrier (210). R. Bruce MacWhirter and Keith L. Bildstein. 1996. 32 pp. Northern Hawk Owl (356). James R. Duncan and Patricia A. Duncan. 1998. 28 pp. Northern Pygmy-Owl (494). Denver W. Holt and Julie L, Petersen. 2000. 24 pp. Red-shouldered Hawk (107). Scott T. Crocoll. 1994. 20 pp. Red-tailed Hawk (52). C.R. Preston and R.D. Beane. 1993. 24 pp. Sharp-shinned Hawk (482). Keith L. Bildstein and Ken Meyer. 2000. 28 pp. Short-eared Owl (62). D.W. Holt and S.M. Leasure. 1993. 24 pp. Snail Kite (171). P.W. Sykes, Jr., J. A. Rodgers, Jr. and R.E. Bennetts. 1995. 32 pp. Snowy Owl (10). David F. Parmelee. 1992. 20 pp. Spotted Owl (179). R.J. Gutierrez, A.B. Franklin and W.S. Lahaye. 1995. 28 pp. Swainson’s Hawk (265). A. Sidney England, MarcJ. Bechard and C. Stuart Houston. 1997. 28 pp. Swallow-tailed Kite (138). Kenneth D. Meyer. 1995. 24 pp. Turkey Vulture (339). David A. Kirk and Michael J. Mossman. 1998. 32 pp. Whiskered Screech-Owl (507). Frederick R. Gehlbach and Nancy Y. Gehlbach. 2000. 24 pp. White-tailed Hawk (30). C. Craig Farquhar. 1992. 20 pp. White-tailed Kite (178). Jeffrey R. Dunk. 1995. 16 pp. Buteo Books stocks all published species accounts, not only those covering raptors. The current list in taxo- nomic order may be viewed at: http://www.buteobooks.com Buteo Books stocks the Handbook of the Birds of the World. The Rrst 6ve volumes of this projected 12-volume work have been published including: Volume 2: New World Vultures to Guineafowl (1994) covering the diurnal raptors and Voliune 5: Barn Owls to Hummingbirds (1999) covering owls. These volumes are priced at $185 each plus shipping and handling. Usually available from Buteo Books, the classic reference on diurnal birds of prey: Brown, Leslie and Dean Amadon. Eagles, Hawks and Falcons of the World. Country Life Books, 1968. Two volumes. First English edition in brown cloth. Eine in slipcase. $300.00 and other editions at lesser prices. A Telemetry Receiver Designed with The Researcher In Mind What you've been waiting for! COMMUNICATIONS SPECIALISTS. INC. ® 426 W«t Taft Avfnue • Orange. CA 9286S-4296 • 1-714-998-3021 • Fax 1-714-974-3420 Entire U.S.A. (800) 8S4-0547 • Fax (800) 850-0547 • http://www.com-spec.com Finally, a highly sensitive 999 channel synthesized telemetry receiver that weighs less than 13 ounces, is completely user programmable and offers variable scan rates over all frequencies. For each animal being tracked, the large LCD display provides not only the frequency (to lOOHz) and channel number, but also a 7 character alphanumeric comment field and a digital signal strength meter. Stop carrying receivers that are the size of a lunch box or cost over S1500. The features and performarKe of the new R-IOOO pocket sized telemetry receiver will impress you. and the price will convince you. Othtr Featum Include: • Factory tuned to any 4MHz wide segment in the 148-174MHz Band • Very high sensitivity of •148dBm to -150dBm • Illuminated display and keypad for use in low light or darkness • User selectable scan rates from 1-30 seconds in 1 second steps • Rechargeable batteries operate the receiver for 12 hours and can be replaced with standard AA Alkaline batteries in the held. Both 12vdc and llOvac chargers are included. • 6.1' 2 . 6 ' 1.5' (15.5cm) high. (6.6cm) wide. (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 specihcations, operating manual and information on the R-1000 or call our toll-free number to order your receiver now. Try the New R-1000 and You'll Be Impressed! More than just Video Cameras Solar cells Parabolic mics The auto-color camera includes IR and color cameras with an automatic light- sensitive switching device. BURROWING OWL WITH EGGS Image captured with the Peeper Video System Copyright Dan Rosenberg 1999 Systems from Sandpiper Technologies • ELEVATED: Pole-mounted video cameras extend up to 50 feet. • BURROW VIDEO PROBES: 5/16” dia. fiber-optic sensors,!” and 2 . 3 ” dia. goosenecks, and remote control systems. All systems include head mounted video display and are interchangeable with other probe extensions. • SURVEILLANCE: Time-lapse, color, IR, digital, zoom, combination IR and color, nestbox cameras. Note: Our ultra -lowlight camera is 0.00015 LUX and captures images in shadow or moonlit situations. • VOCALIZERS: MP3 format reduces battery requirements. • SOLAR: We offer multiple solar options and remote powered energy devices. SANDPIPER TECHNOLOGIES, INC. . r 33!^) 'A 'Vn-.fit It- -Xv.-nij*- r L.A 95337 (209i2Jy M6C • ^ 5 • e -mail; com Catal -I [ . p|«' ( ;:,m 2001 ANNUAL MEETING The Raptor Research Foundation, Inc. 2001 annual meeting will be held on 25-30 October in Winnipeg, Manitoba, Canada. For information about the meeting contact Jim Duncan, Biodiversity program. Wildlife Branch, Manitoba Natural Resources, Box 24, 200 Saulteaux Crescent, Winnipeg, MB R3J 3W3 Canada. Email jduncan@nr.gov.mb.ca. Persons interested in predatory birds are invited to join The Raptor Research Foundation, Inc. Send requests for information concerning membership, subscriptions, special publications, or change of address to OSNA, P.O. Box 1897, Lawrence, KS 66044-8897, U.S.A. The Journal of Raptor Research (ISSN 0892-1016) is published quarterly and available to individuals for $33,00 per year and to libraries and institutions for $50.00 per year from The Raptor Research Foundation, Inc., 14377 117th 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 2000 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., Awards Recognition for Significant Contributions^ The Dean Amadon Award recognizes an individual who has made significant contributions in the field of systematics or distribution of raptors. Contact: Dr. Clayton White, 161 WIDE, Department of Zoology, Brigham Young University, Provo, UT 84602 U.S.A. Deadline August 15. The Tom Cade Award recognizes an individual who has made significant advances in the area of captive propagation and reintroduction of raptors. Contact: Dr. Brian Walton, Predatory Bird Research Group, Lower Quarry, University of California, Santa Cruz, CA 95064 U.S.A. Deadline: August 15. The Fran and Frederick Hamerstrom Award recognizes an individual who has contributed significantly to the understanding of raptor ecology and natural history. Contact: Dr. David E. Andersen, Department of Fisheries and Wildlife, 200 Hodson Hall, 1980 Folwell Avenue, University of Minnesota, St. Paul, MN 55108 U.S.A. Deadline: August 15. Recognition and Travel Assistance The James R. Koplin Travel Award is given to a student who is the senior author of the paper to be presented at the meeting for which travel funds are requested. Contact: Patricia A. Hall, 5937 E. Abbey Road, Flagstaff, AZ 86004 U.S.A. The William C. Andersen Memorial Award is given to the student who presents the best paper at the annual Raptor Research Foundation Meeting. Contact: Ms. Laurie Goodrich, Hawk Mountain Sanctuary, Rural Route 2, Box 191, Kempton, PA 19529-9449 U.S.A. Deadline: Deadline established for meeting paper abstracts. Grants^ The Stephen R. TuUy Memorial Grant for $500 is given to support research, management and conservation of raptors, especially to students and amateurs with limited access to alternative funding. Contact: Dr. Kimberly Titus, Alaska Division of Wildlife Conservation, P.O. Box 20, Douglas, AK 99824 U.S.A. Dead- line: September 10. The Leslie Brown Memorial Grant for $500-$l,000 is given to support research and/or the dissemination of information on raptors, especially to individuals carrying out work in Africa. Contact: Dr. Jeffrey L. Lincer, 1220 Rosecrans St. #315, San Diego, CA 92106 U.SA. Deadline: September 15. ^Nominations should include: (1) the name, title and address of both nominee and nominator, (2) the names of three persons qualified to evaluate the nominee’s scientific contribution, (3) a brief (one page) summary of the scientific contribution of the nominee. ^Send 5 copies of a proposal (^5 pages) describing the applicant’s background, study goals and methods, anticipated budget, and other funding.