The Journal of Raptor Research Volume 35 Number 1 March 2001 Published by The Raptor Research Foundation, Inc THE RAPTOR RESEARCH FOUNDATION, INC. (Founded 1966) OFFICERS PRESIDENT: Michael N. Kochert SECRETARY: Patricia A. Hall VICE-PRESIDENT: Keith L. Bildstein TREASURER: Jim Fitzpatrick BOARD NORTH AMERICAN DIRECTOR #1: Philip Detrich NORTH AMERICAN DIRECTOR #2: Petra Bohall Wood NORTH AMERICAN DIRECTOR #3: Robert Lehman INTERNATIONAL DIRECTOR #1: Eduardo Inigo-Elias INTERNATIONAL DIRECTOR #2: Reuven Yosef OF DIRECTORS INTERNATIONAL DIRECTOR #3: Beatriz Arroyo DIRECTOR AT LARGE #1; Jemima Parry-Jones DIRECTOR AT LARGE #2: Robert Kenward DIRECTOR AT LARGE #3: Michael W. Collopy DIRECTOR AT LARGE #4; Miguel Ferrer DIRECTOR AT LARGE #5: John A. Smallwood DIRECTOR AT LARGE #6; Brian A. Millsap EDITORIAL STAFF EDITOR: Marc J. Bechard, Department of Biology, Boise State University, Boise, ID 83725 U.S.A. ASSOCIATE EDITORS Allen M. Fish Fabian JaksiC Juan Jose Negro Daniel E. Varland Charles J. Henny Cole Crocker-Bedford Ian G. Warkentin James C. Bednarz Marco Restani 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 Clark, 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 24-hour 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,”/. Raptor Res., Vol. 34(4), and are available from the editor. COVER: Light morph Booted Eagle {Hieraaetus pennatus) . Painting by Luis M. Curesma Gallardo. Contents Low Productivity of Bald Eagles on Prince of Wales Island, Southeast Alaska. Robert G. Anthony 1 Dispersal of Juvenile and Immature Bonelli’s Eagles in Northeastern Spain. Joan Real and Santi Manosa 9 Estimating the Breeding Population of Booted Eagles in the Cape Province, South Africa. David Pepler, Rob Martin, and HubertusJ. van Hensbergen 15 Sex Determination in Booted Eagles (Hieraaetus pennatus) using Molecular Proce- dures and Discriminant Function Analysis. Javier Balbondn, Miguel Ferrer, and Eva Casado 20 The Annual and Diel Cycles of Goshawk Vocalizations at Nest Sites. Vincenzo Penteriani 24 Breeding Rates of Eurasian Kestrels (Falco tinnunculus) in Relation to Sur- rounding Habitat in Southwest Spain. Jesus M. Aviles, Juan M. Sanchez, and Deseada Parejo 31 The Migration of Steppe Eagles {Aquila nipalensis) and Other Raptors in Central Nepal, Autumn 1999. Robert DeCandido, Deborah Allen, and Keith L. Bildstein 35 Density, Productivity, Diet, and Human Persecution of Golden Eagles {Aquila CHRYSAETOS) IN THE CeNTRAL-EaSTERN ITALIAN ALPS. Paolo Pedrini and Fabrizio Sergio .. 40 Management of Nonreleasable Raptors for Conservation Education Programs. Joe N. Caudell and Ken A. Riddleberger, Jr. 49 Short communications Winter Roosting Behavior of American Kestrels. Daniel R. Ardia 58 Diet and Prey Selection of Nonbreeding Peregrine Falcons in an Urban Habitat of Italy. Gian- luca Serra, Maria Lucentini, and Simona Romano 61 Bald Eagles Killed by Trains in New York State. Ward B. Stone, Peter E. Nye, and Joseph C. Okoniewski 64 Early Nesting by Great Horned Owls in Montana. Denver W. Holt and Stacy Drasen 66 Diet of the Short-eared Owl in Northwestern Argentina, Sebastian Cirignoli, Dario H. Podesta, and Ulyses F. J. Pardinas 68 A Survey of Raptors on Rhodes; An Example of Human Impacts on Raptor Abundance and Distri- bution. Arianna Aradis and Giuseppe M. Carpaneto 70 The iNaoENCE of Intestinal Parasites in British Birds of Prey. Nigel W.H. Barton and David C. Houston 71 Letters First Sight Record of the King Vulture in Baja California, Mexico. Russell B. Duncan and Jo Ann V. Lacroix 74 Probable Replacement Clutches by Booted Eagles (Hieraaetus iennatus) in the Tietar River Valley of Central Spain. Ignacio S. Garda Dios 75 Book Reviews. Edited by Jeffrey S. Marks 76 The Raptor Research Foundation, Inc. gratefully acknowledges a grant and logistical support provided by Boise State University to assist in the publication of the journal. THE JOURNAL OF RAPTOR RESEARCH A QUARTERLY PUBLICATION OF THE RAPTOR RESEARCH FOUNDATION, INC. VoL. 35 March 2001 No. 1 J. Raptor Res. 35(1) :l-8 © 2001 The Raptor Research Foundation, Inc. LOW PRODUCTIVITY OF BALD EAGLES ON PRINCE OF WALES ISLAND, SOUTHEAST ALASKA Robert G. Anthony U.S. Geological Survey, Oregon Cooperative Fish and Wildlife Research Unit, Department of Fisheries and Wildlife, Oregon State University, Corvallis, OR 97331-3803 U.S. A. Abstract. — I investigated reproductive success of Bald Eagles {Haliaeetus leucocephalus) on Prince of Wales Island, Alaska from 1991-93. Productivity (0.13 young produced per occupied territory) was the lowest recorded for the species throughout its geographic range. Productivity was not significantly dif- ferent among different habitats including remote roadless areas vs. roaded and logged areas, which suggested that habitat alterations were not the cause of low productivity. Because nesting densities were high and I observed some effects of proximity of nearest neighbor pairs, I suggest these densities (proximate factor) were affecting productivity through reduced food availability (ultimate factor). How- ever, I could not rule out the effects of environmental contaminants, although this seemed unlikely because of the distance of the island from industrial and agricultural areas. I discuss the various potential causes of this low rate of productivity. Key Words: Bald Eagle, Haliaeetus leucocephalus; nearest-neighbor epfecty, productivity; southeast Alaska. Baja productividad de aguilas calvas en la isla del Principe de Gales, sureste de Alaska Resumen. — Investigue el exito reproductivo de las %uilas calvas {Haliaeetus leucocephalus) en la Isla del Principe de Gales, Alaska 1991-93. La productividad (0.13 juveniles producidos por territorio ocupado) fue la mas baja registrada para la especie a lo largo de su rango geografico. La productividad no fue significativamente diferente entre los diferentes habitats incluyendo areas remotas y sin carreteras vs. areas de explotacion maderera con carreteras, lo cual sugiere que las alteraciones del habitat no fuera la causa de la baja productividad. Debido a que las densidades de los nidos fueron altas, observe algunos efectos producidos por la proximidad de las parejas vecinas mas cercanas. Sugiero que estas densidades (factor proximo) estaban afectando la productividad reduciendo la disponibilidad de comida (factor ultimo). Sin embargo, no pude medir los efectos de los contaminantes ambientales, aunque posible- mente no hubo debido a la distancia de la isla a las areas industriales. Discuto las causas potenciales de esta baja tasa de productividad. [Traduccion de Cesar Marquez] Bald Eagles {Haliaeetus leucocephalus) are consid- ered a Sensitive Species in Alaska and are managed under the Bald and Golden Eagle Protection Act (1940) and the Migratory Bird Treaty Act (1918). The most prominent factor for historical popula- tion declines in the state was the bounty system which was imposed on the species until 1953; how- ever, populations have increased since mortalities from this system have stopped (Hodges et al. 1979). Southeast Alaska currently has large popu- lations of breeding and wintering Bald Eagles. Alas- ka is considered a stronghold (Hodges et al. 1979, Hansen 1987) and provides high quality habitat for the species. However, high densities are not nec- essarily indicative of high quality habitat (Van Horne 1983) or sufficient demographic perfor- mance. Evaluating habitat quality should involve examination of the species’ reproductive success and/or survival within the area. Habitat quality may also be influenced by human activities, be- I I 1 2 Anthony VoL. 35, No. 1 cause high quality habitats are often avoided be- cause of the presence of humans (McGarigal et al. 1991). There were several human activities that may have disturbed Bald Eagles on Prince of Wales Is- land, including road construction, vehicular traffic, helicopter overflights, and habitat alteration in the form of timber harvest. The purpose of this project was to investigate Bald Eagle productivity on the eastern shoreline of Prince of Wales Island, Alaska, and examine both “natural” and human-related factors that may influence productivity. Specifically, I hypothesized that human activities as a result of logging and road construction were having a det- rimental effect on Bald Eagle productivity (An- thony and Isaacs 1989, McGarigal et al. 1991). Be- cause of dense breeding populations, I also predicted that nearest-neighbor interactions may affect productivity (Anthony et al. 1994). Hansen (1987) suggested that food availability has an influ- ence on productivity of Bald Eagles in southeast Alaska, so I was aware of this potential effect. How- ever, prey availability is next to impossible to de- scribe for this species because of their diverse diet of fish, birds, and mammals. Study Area I conducted surveys of Bald Eagles and described pro- ductivity on the east side of Prince of Wales Island, Alas- ka The coastline in this region is variable. Some areas are very convoluted with many small bays and peninsulas, as well as numerous offshore rocks and islets. Other areas are essentially straight shorelines with few prominent points and steep slopes descending directly to the shore- line. Vegetation consists of forests dominated by Sitka spruce (Picea sitchensis), western hemlock {Tsuga hetero- phylla) , and western red cedar ( Thuja plicata) . The study area consisted of shoreline between Mills Bay on the south and Lake Bay on the north, and small islands near the shoreline also were included in the study. In 1992, I expanded the study area to include Rose, Berry, Round, and the east side of Stevenson Islands (hereafter referred to as islets). Rivers or streams that support runs of salmon {Oncorhynchus spp.) within or near this area include Lake Bay, Coffman, Chum, Eagle, Ratz, Little Ratz, Sal, Cobble, Slide Creeks, and the Thorne River. Methods Surveys. Surveys were conducted from April through September 1991-93 by foot, from vehicles, sea (14' Gre- gor welded and 18' Alumaweld boats), and air (Hughes 500D helicopter). Occupancy and productivity of nests were determined with two aerial surveys each year with four people on board the helicopter. One person record- ed data while the others watched for eagles or nests. Lo- cations of all eagles and nests were plotted on U.S. Geo- logical Survey 1:63 360 maps or U.S. Forest Service maps of the study area. The first survey flight was conducted during the egg laying and incubation periods (7-8 May 1991, 6—7 May 1992, 3 May 1993) to determine occupan- cy of nesting territories. The second flight was conducted during the late nesding stage (3 and 8 July 1991, 28-30 July 1992, 3-4 August 1993) to determine productivity. In addition, a third survey flight was flown on 15 August 1991 to verify productivity at several nests. I also supple- mented aerial surveys at as many nest sites as possible throughout the breeding season using boats or foot trav- el. Terminology used in this paper follows that of Postu- palsky (1974). A territory was occupied when two adults were observed in association with a nest or when an adult was observed on a nest. If an adult was observed on a nest in incubating posture or with young nestlings, a breeding attempt took place in the territory (Steenhof 1987). A territory was successful if fledglings, or nestlings near fledgling age, were observed. For the purposes of analysis, any occupied site that failed to produce fledg- lings was considered a failure. Nearest-Neighbor Interactions. Because breeding pop- ulations were dense, I was interested in conspecific near- est-neighbor interactions and its potential influence on Bald Eagle productivity as described for Oregon (An- thony et al. 1994). Accordingly, 1 determined a Universal Transverse Mercator (UTM) location for each territory for each year. The location of occupied nests, if known, was used for the territory location. If no occupied nest was known for the territory, I used a nest within the area where eagles were frequently observed. If I failed to iden- tify an occupied nest or breeding attempt within the area, I approximated a central point for all eagle observations within the territory during that year. I calculated the dis- tance to the nearest occupied territory using the UTM locations for each territory occupied during a given year. I also calculated the distance to the second nearest ter- ritory, which was added to the nearest territory distance to provide the “total neighbor distance.” The nearest and total neighbor distances were used to evaluate the potential effect of nearest-neighbors on productivity. Statistical Analyses. For each year and for all years com- bined, I calculated three measures of productivity. Breed- ing success was defined as the percent of occupied sites that produced young, and productivity as the number of young fledged per occupied site. I also calculated the number of young fledged per successful site. Habitats surrounding nests were classihed as unroaded and un- logged, roaded but unlogged, unroaded but logged, roaded and logged, and islet nests. Unroaded and un- logged habitats were those that had no roads or past log- ging history within —1.6 km of the occupied nest or ter- ritory center. In contrast, roaded or logged habitats had roads or past logging history within 1.6 km of the nest or territory center. All logging activities were of the form of clearcut harvests. Chi-square tests or Fisher’s exact test were computed to determine if the proportion of suc- cessful nests was significantly different among years. Anal- ysis of variance was used to test for differences in the number of young produced per site among years and habitats. I used an alpha of 0.10 to determine signifi- cance for all statistical tests, because I wanted to minimize the probability of a Type II statistical error and maximize March 2001 Bald Eagle Productivity in Alaska 3 Table 1. Occupancy and productivity of Bald Eagle nests on Prince of Wales Island, Alaska, 1991-93. 1991 1992 1993 All Years Sites surveyed 109 109 109 327 Occupied^ 91 (84%) 98 (90%) 78 (72%) 267 (82%) Breeding attempt’^ 47 (52%) 62 (63%) 41 (53%) 150 (56%) Successful*’ 10 (11%) 10 (10%) 11 (14%) 31 (11%) # Young fledged 11 10 13 34 # Young/ occupied site 0.12 0.10 0.17 0.13 # Young/successful site 1.10 1.00 1.18 1.10 ^ Percent is based on number of sites surveyed. Percent is based on number of sites occupied. power to detect any possible differences (Sokal and Rohlf 1981). Nested analysis of variance (ANOVA) was used to test for differences in nearest and total neighbor distances using SAS (SAS Institute 1989). For both nearest neigh- bor distances, I compared active versus inactive nests and successful versus unsuccessful nests within years. Because sample sizes were unequal for many groupings of nest sites, Satterthwaite’s approximation to the F statistic was used for both tests of activity nested within years (Sokal and Rohlf 1981). The standard simple approximation of the F statistic was used for both tests of success nested within years. 1 transformed both nearest and total neigh- bor distances using the square root transformation so the frequency distributions were normally distributed. Pro- ductivity at territories on the more remote islets was ex- amined, because I hypothesized that the.se territories may not be influenced by nearest-neighbor interactions. A chi-square test of independence was used to compare breeding success at remote islets with that of nests in the remainder of the study area. Results Productivity. I identified 109 breeding territories within the study area and 267 breeding attempts (territory occupancy) during the three years. Of these, 62 (57%) were occupied all three years, 34 (31%) were occupied for two years, and 13 (12%) were occupied in only one year. Ninety-one terri- tories were occupied in 1991, and 47 (52%) breed- ing attempts were identified. Eleven young fledged from 10 successful nests. I identified 98 occupied territories in 1992, and 62 (63%) of these had breeding pairs. Ten young fledged from 10 suc- cessful nests. Seventy-eight territories were occu- pied in 1993, and 41 (53%) had breeding pairs. Thirteen young fledged from 11 successful nests. Productivity was extremely low for all years. The proportion of nests that were successful was not significantly different among years (x'^ = 0.695, df = 2, 7* — 0.7065) and averaged only 11% (Table 1). The number of young fledged per occupied site was not significantly different among years (x^ = 1.286, df = 2, P — 0.5257) and averaged only 0.13 (Table 1). For all years, an average of 1.1 young fledged per successful nest (Table 1). Timing of Nesting Failures. In 1991, 14 of the 29 breeding pairs (48%) failed by the early nestling stage. Of the 12 territories that were still occupied and could be monitored, eight succeeded in fledg- ing young. In 1992, 19 territories failed to produce young, and 12 (63%) were still occupied on 27 May. Of these 12, only three were still occupied on 10 June, and all three were successful in fledging young. Therefore, 47% of failures occurred within the two-week period at the beginning of June, which corresponded with the late incubation pe- riod. The cause of nesting failures was not deter- mined because I did not climb nest trees to inspect nests according to Anthony et al. (1994). Influence of Habitat Condition on Productivity. Because there were no significant differences in productivity among years, data for different years were combined for this analysis. Productivity for the different habitat conditions also was extremely low, but there were significant (P < 0.05) differ- ences among the habitat conditions. Productivity of islet territories was significandy higher (P < 0.05) than those in unroaded and unlogged, and in roaded and logged territories (Table 2). Overall, productivity of islet nests was the highest of all of the habitat conditions. Productivity of territories in unroaded and unlogged areas was not significantly (P > 0.05) higher than that of other habitats. When all territories that were either roaded or logged were combined (Table 2) , there was no sig- nificant (P > 0.17) difference in productivity of this group of territories and that of sites in un- roaded and unlogged sites. However, productivity of islet nests was significantly (P = 0.0031) higher 4 Anthony VoL. 35, No. 1 Table 2. Productivity of Bald Eagle nest sites on Prince of Wales Island in relation to habitat condition. Habitat Condition No. Occupied Sites Breeding Success (%) Young Fledged/ Occupied Site^ A. Separate analysis: Unroaded, unlogged 149 6 0.07^ Roaded, unlogged 5 0 0.00^ Unroaded, logged 22 18 0.18^ Roaded, logged 42 12 0.12=^ Newly roaded and logged 36 22 0.25 Islets (undisturbed) 13 38 0.38 B Combined analysis: Unroaded, unlogged 149 6 0.07=* Roaded or logged 105 16 0.17^ Islets (undisturbed) 13 38 0.38 ® Means with the same superscripts are not signihcantly (P > 0.10) different as determined by analysis of variance and a Bonferoni mean separation test. Table 3. Mean (±SE) nearest and total neighbor dis- tance (m) for all activity categories of Bald Eagle nests on Prince of Wales Island, 1991-93.' Nearest- Total Category N Neighbor Neighbor 1991 91 1130 (59) 2992 (135) Unoccupied 44 1172 (75)^ 3058 (173)“ Breeding attempt 47 1090 (92)“ 2929 (207)“ Failed 81 1089 (58)'’ 2918 (133)'’ Successful 10 1461 (259)^^ 3584 (586)'’ 1992 98 1048 (60) 2764 (127) Unoccupied 36 952 (92)“ 2560 (192)“ Breeding attempt 62 1104 (78)“ 2883 (165)“ Failed 88 1081 (65)'’ 2810 (137)'’ Successful 10 764 (117)<= 2360 (281)'’ 1993 78 1341 (69) 3547 (159) Unoccupied 37 1341 (112)“ 3467 (245)“ Breeding attempt 41 1341 (86)“ 3618 (208)“ Failed 67 1310 (77)'’ 3488 (173)” Successful 11 1528 (145)" 3904 (388)'’ All years combined 267 1162 (37) 3070 (82) Unoccupied 117 1158 (55)“ 3034 (121)“ Breeding attempt 150 1164 (50)“ 3098 (112)'’ Failed 236 1148 (39)'’ 3040 (86)" Successful 31 1260 (119)" 3303 (271)"* ^ Means for nests with breeding attempts vs. those without and successful vs. failed comparisons were significantly (P < 0.10) different when followed by different letters. All comparisons were made within a column. than that of territories in unroaded and unlogged areas for the combined analysis. Nearest-Neighbor Analysis. Nearest-neighbor distances between all territories averaged 1162 m (range = 175-3937) for all years combined (Table 3). There were no significant (P > 0.10) differ- ences between nearest-neighbor distances for nests with breeding attempts vs. those without or suc- cessful vs. unsuccessful nests for all years combined (Table 3). However, nearest-neighbor distance be- tween nest with breeding attempts vs. those with- out was significant (F = 7.50, P = 0.0795) within years after annual variation was removed. This was probably the result of nearest-neighbor distances increasing in 1993 because fewer territories were occupied than in 1991 or 1992. There was no sig- nificant (P = 2.74, P = 0.2124) difference in near- est-neighbor distances between successful and failed nests for all years combined; however, the difference was significant (P — 2.47, P = 0.0627) for within-year comparisons. In 1991 and 1993, suc- cessful nests had larger nearest-neighbor distances than failed nests (Table 3). However, successful nests had smaller nearest-neighbor distances than failed nests in 1992. Total neighbor distances averaged 3070 m (range = 355—8567) for all years (Table 3). Total neighbor distance was significantly different be- tween nests with breeding attempts vs. those with- out (P = 11.74, P = 0.0493) when data were com- bined over all years, but activity was not significantly (P = 0.72, P = 0.5399) different for ■ ■ March 2001 Bald Eagle Productivity in Alaska 5 Table 4. Productivity of Bald Eagle populations in North America. Region Occupied Sites Young Fledged/ Occupied Site Young Fledged/ Successful Site Study Period Source Colorado, Wyoming 85 1.21 1.92 1981-89 Kralovec et al. (1992) Saskatchewan, Canada 48 1.06 1.82 1984-87 Dzus and Gerrard (1993) Chesapeake Bay 1448 1.21 1.70 1981-90 Buehler et al. (1991) Wisconsin 1469 1.30 1.69 1983-88 Kozie and Anderson (1991) Northwest Ontario 1370 0.80^ 1.67 1970-80 Grier (1982) Arizona 45 0.80 1.63 1975-80 Grubb et al. (1983) Alaska Peninsula, Alaska 43 0.97 1.61 1970 Hehnke (1973) Kodiak Island, Alaska 312 1.00 1.59 1963-70 Sprunt et al. (1973) Oregon 606 0.92 1.52 1979-92 Isaacs and Anthony (1992) Wisconsin 492 1.00 1.52 1962-70 Sprunt et al. (1973) Texas 193 0.98 1.50 1981-90 Maybie et al. (1994) Gulkana River, Alaska 274 0.86 1.48 1989-93 Steidl et al. (1997) Florida 592 0.73 1.46 1961-70 Sprunt et al. (1973) Yukon Territory, 39 1.05 1.46 1980-82 Blood and Anweiler (1990) Canada California 140 0.81 1.45 1970-91 Jenkins (1992) Yellowstone Nat. Park, 107 0.41 1.43 1972-79 Alt (1980), Swenson (1975) Wyoming Amchitka Island, Alas- 71 0.86 1.42 1972 Sherrod et al. (1976) ka Maine 521 0.44^ 1.35 1972-78 Todd (1979) San Juan Islands, 275 0.84 1.35 1975-80 Grubb et al. (1983) Washington New Brunswick, Can- 55 0.73 1.33 1974-80 Stocek and Pearce (1981) ada Washington 866 0.87 1.32 1981-85 McAllister et al. (1986) Prince of Wales Island, 267 0.13 1.10 1991-93 This study Alaska ® The population in this study area has been influenced by organochlorine contaminants. within-year comparisons (Table 3) . This result cor- responded with the nearest-neighbor analyses and was probably the result of larger total neighbor dis- tances being observed in 1993, because fewer sites were occupied than in 1991 or 1992. Total neigh- bor distance for all years combined was significant {F = 6.17, P = 0.0864) when comparing successful vs. unsuccessful nests within years. However, total neighbor distance did not differ according to suc- cess within-year (P = 0.2571). I monitored nesting success of Bald Eagles on small islets in 1992 and 1993, because there was usually only one occupied territory per islet. Ter- ritories at these islets produced young more often than territories within the remainder of the study area (x^ = 90.6, df = 1, P = 0.0019). Of the 13 occupied territories on islets, 5 (38%) were suc- cessful in producing at least one young. At “non- islet” territories, only 26 of 252 (10%) occupied territories were successful. Mean nearest-neighbor distance at islet territories was 1369.9 m, as com- pared to 1151.3 m at “nonislet” territories. Mean total neighbor distance at islet territories was 3521.5 m, compared to 3046.4 m at “nonislet” ter- ritories. Discussion Productivity of Bald Eagles on Prince of Wales Island, Alaska, was extremely low for all three years and in all habitat conditions. The average number of young produced per occupied site (0.13) was the lowest reported for this species throughout its geographic range (Table 4) . In addition, the num- ber of young fledged per successful nest (110) was 6 Anthony VoL. 35, No. 1 also the lowest recorded tor bald Eagles, including other areas in Alaska. Hansen et al. (1984) studied productivity of Bald Eagles in the Chilkat River Val- ley from 1979-83 and found that 32% of occupied territories were successful in producing young with mean productivity rate of 0.42 young per occupied site. Steidl et al. (1997) reported higher mean pro- ductivity (0.86 young/occupied site) on the Gul- kana River of central Alaska. Potential causes of nesting failures that can reduce productivity of Bald Eagle populations include hu- man disturbance, contaminants, nesding mortality, infertile eggs, food stress, weather, nearest-neighbor effects, or the failure to lay eggs (Anthony et al. 1994) . Of these causes, human disturbance, contam- inants, nearest-neighbor interactions, and/or food stress were considered to be the most likely factors to cause the extremely low productivity of Bald Eagles on Prince of Wales Island. I found no evidence for human disturbance being a m^or influence on pro- ductivity, because nesting failures occurred along re- mote as well as human occupied shorelines during all three years. In addition, nest sites that were suc- cessful in producing young were associated with shorelines with human activities as frequendy as those that were associated with uninhabited shore- lines. Also, low productivity was prevalent in unroad- ed and unlogged as well as human inhabited areas. Consequendy, my data do not support the original hypothesis that human disturbance (i.e., logging or road construction) had an effect on productivity of Bald Eagles on Prince of Wales Island. My analysis of nearest-neighbor distances sug- gested that nearest-neighbor interactions may have influenced productivity. The higher nearest-neigh- bor distance for successful versus failed sites in 1991 and 1993, and the higher productivity of “is- let” versus “nonislet” territories (large vs. small nearest-neighbor distances) support this explana- tion. The extremely low success rates of eagles within our study area prevented us from conduct- ing nearest-neighbor analyses comparable to those of Anthony et al. (1994) for Oregon. However, they observed negative effects of nearest-neighbor pairs at greater distances (<3200 m) than those among most pairs on this study area (jc = 1162 m, range = 175-3967 m). Therefore, it is possible that Bald Eagles on Prince of Wales Island were nesting so densely that all nests were subjected to nearest- neighbor interactions, which acted as a proximate effect on productivity. Food stress is likely the ultimate factor influenc- ing productivity on Prince of Wales Island and may result in nearest-neighbor interactions. Our limit- ed data indicate that many of the nests failed dur- ing the egg-laying and incubation stages, which is a pattern associated with food-stressed populations (Newton 1979). Also, the spatially variable group- ings of successful nests that I observed each year suggested local prey availability in these areas. Some of these groupings were in close proximity to streams with abundant salmon runs, which may have provided the necessary prey resources for suc- cessful reproduction. However, salmon were not present in streams until later in the summer after nesting failures have occurred. Other anadromous fishes such as eulachon {Thaleichthys pacificus), sand lance {Ammodytes hexapterus), and herring ( Clupea pallasi) are some of the first foods available to eagles after the long winter, and their runs are highly variable spatially (P. Schempf pers. comm.). Consequently, the abundance of these fishes in space and time may influence Bald Eagle produc- tivity. This is a hypothesis for future work and test- ing. Several studies in Alaska have indicated that Bald Eagle productivity is controlled by prey abundance and/ or availability. In southeastern Alaska, Hansen (1987) found that placing prey within Bald Eagle nesting territories increased their productivity. Hansen and Hodges (1985) attributed variability in breeding rates of Bald Eagles to variability in prey abundance. Lastly, Steidl et al. (1997) suggested that most variation in reproductive success of Bald Eagles along the Gulkana River in central Alaska was attributable to prey availability. The low pro- ductivity of Bald Eagles on Prince of Wales Island may be due to low prey abundance or availability (ultimate factor) and is displayed through nearest- neighbor interactions (proximate factor) . I could not rule out the possibility of environ- mental contaminants having an effect on produc- tivity of Bald Eagles on Prince of Wales Island. This seemed unlikely because the area is remote from industrial and agricultural areas, the source of many pesticides that have been shown to effect Bald Eagle populations (Wiemeyer et al. 1984). However, elevated levels of DDE, PCBs, dioxins, or furans have been reported in waterfowl (White- head et al. 1990), seabirds (Elliott et al. 1989a), Great Blue Herons (Ardea herodias) (Elliott et al. 1989b), and Bald Eagles (Elliott et al. 1996) along the coast of British Columbia, Canada. The source of DDE and PCBs in these species of birds is un- March 2001 Bai.d Eagle Productivity in Alaska 7 known; however, the source of dioxins and furans is usually from pulp and paper mills that use bleaching processes to produce paper. These two compounds are reported to be some of the most toxic substances to birds, and effects on reproduc- tion have been documented in laboratory experi- ments on wood ducks (White and Seginak 1994) in concentrations of parts per trillion. In addition, Estes et al. (1997) and Anthony et al. (1999) have recently found elevated levels of DDE and PCBs in Bald Eagles in the western Aleutian Islands of Alas- ka. Consequently, environmental contaminants may be accumulating in food chains and affecting Bald Eagle reproduction on Prince of Wales Island. The possible effect of environmental contaminants on productivity of bald eagles should be investigat- ed by collecting eggs from nests over a 2-3 yr pe- riod (Wiemeyer et al. 1984, Anthony et al. 1993). I may have studied the Bald Eagles on Prince of Wales Island during a time when productivity was extremely low because productivity may have im- proved. Continued monitoring of reproductive success on the island would help answer this ques- tion. Monitoring trends of prey populations, par- ticularly salmon, herring, sand lance, eulachon, and smelt, could be important also. The timing, location, and magnitude of these runs may explain the clumped but sparse nature of successful breed- ing attempts of Bald Eagles that varies among years on the island. Data on eagle productivity and char- acteristics of anadromous fish runs over several years will be necessary to determine if any relation exists between the two, but such information could be valuable in determining the cause (s) of low pro- ductivity on Prince of Wales Island. Acknowledgments I thank the staff of Thorne Bay Ranger District, Ton- gass National Forest, for logistical support during the study. Their contributions in the way of housing, maps, technical advice, and especially, helicopter time for pro- ductivity surveys made this project possible. I thank C. Ford and E. Campbell for their support and advice dur- ing the project. B. Bibles, E. Bibles, F. Isaacs, and M. Van- der Heyden participated in the field surveys. R Schempf, E. Cambell, M. Jenkins, T. Grubb, and D. Buehler pro- vided valuable suggestions on an earlier draft of the man- uscript, and B. Matzke conducted the statistical analyses. This project was funded by the U.S. Forest Service through Contract No. 14-1 6-0009-1 5 77 to the Oregon Co- operative Wildlife Research Unit at Oregon State Univer- sity. Cooperators with the Oregon Cooperative Wildlife Research Unit at the time of the study were Oregon De- partment of Fish and Wildlife, Oregon State University, The Wildlife Management Institute, and U.S. Fish and Wildlife Service. Literature Cited Alt, K.L. 1980. Ecology of the breeding Bald Eagle and Osprey in the Grand Teton-Yellowstone National Parks Complex. M.S. thesis. Montana State Univ , Bozeman, MT U.S.A. Anthony, R.G. and F.B. Isaacs. 1989. Characteristics of Bald Eagle nest sites in Oregon. /. Wildl. Manage. 53. 148-159. , M.G. Garrett, and C.A. Schuler. 1993. Environ- mental contaminants in Bald Eagles in the Columbia River estuary./. Wildl. Manage. 57:10-18. , R.W. Frenzel, F.B. Isaacs, and M.G. Garrett 1994. Probable causes of nesting failures in Oregon’s Bald Eagle population. Wildl. Soc. Bull. 22:.576-582. , A.K. Miles, J.A. Estes, and F.B. Isaacs. 1999. Pro- ductivity, diets, and environmental contaminants m nesting Bald Eagles from the Aleutian Archipelago Environ. Toxicol. Chem. 18:2054-2062. Blood, D.A. and G.C. Anweiler. 1990. Nesting chronol- ogy and .success of Bald Eagles in southwest Yukon Northwest. Nat. 71:9—11. Buehler, D.A., J.D. Fraser, J.K.D. Seeger, G.D. Therres, and M.A. Byrd. 1991. Survival rates and population dynamics of Bald Eagles on Chesapeake Bay. f. Wildl. Manage. 55:608-613. Dzus, E.H. and J.M. Gerrard. 1993. Factors influencing Bald Eagle densities in northcentral Saskatchewan./ Wildl. Manage. 57:771-778. Elliott, J.E., D.G. Noble, RJ. Norstrom, and P.E. Whitehead. 1989a. Organochlorine contaminants m seabird eggs from the Pacific coast of Canada, 1971- 1986. Environ. Monit. A.ssess. 12:67-82. , R.W. Butler, RJ. Norstrom, and P.E. White- head. 1989b. Environmental contaminants and repro- ductive success of Great Blue Herons, Ardea herodtas, in British Columbia, 1986—87. Environ. Pollut. 59:91— 114. , RJ. Norstrom, A. Lorenzn, L.E. Hart, H. Phi- libert, S.W. Kennedy, J.J. Stegeman, G.D. Bellward, AND K.M. Cheng. 1996. Biological effects of poly-chlo- rinated dibenzo-p-dioxins, dibenzofurans, and biphe- nyls in Bald Eagle {Haliaeetus leucocephalus) chicks. En- viron. Toxicol. Chem. 15:782—793. Estes, J.A., C.E. Bacon, W.M. Jarman, RJ. Norstrom, R.G. Anthony, and A.K. Miles. 1997. Organochlo- rines in sea otters and Bald Eagles from the Aleutian Archipelago. Mar. Pollut. Bull. 34:1—5. Grier, J.W. 1982. Ban of DDT and subsequent recovery of reproduction in Bald Eagles. Science 218:1232-1235. Grubb, TG, R.L. Knight, D.M. Rubink, and C.H. Nash 1983. A five-year comparison of Bald Eagle productiv- ity in Washington and Arizona. Pages 35-45 in D.M. Bird [Ed.], Biology and management of Bald Eagles 8 Anthony VoL. 35, No. 1 Osjjicys. 1^1 css, olC, t Liiiic Cic Jz>ciicvU.C, Quebec, Canada. Hansen, A.J. 1987, Regulation of Bald Eagle reproduc- tion rates in southeast Alaska. Ecology 69:1387-1392. AND J.l. Hodges. 1985. High rates of nonbreeding adult Bald Eagles in southeastern Alaska. J. Wildl. Manage. 49:454-458. E.L. Boeker, J.L Hodges, and D.R. Cline. 1984. Bald Eagles of the Chilkat Valley, Alaska: ecology, be- havior, and management. Natl. Audubon Soc., New York, NY U.S.A. Hehnke, M.E. 1973. Nesting ecology and feeding behav- ior of Bald Eagles on the Alaska Peninsula. M.S. the- sis, Humboldt State University, Areata, CA U.S.A. Hodges, J.L, J.G. King, and EC. Robards. 1979. Resurvey of Bald Eagle breeding population in southeast Alas- ka. /. Wildl. Manage. 43:219-221. Isaacs, F.B. and R.G. Anthony. 1992. Bald Eagle nest locations and history of use in Oregon 1971 through 1992. Oregon Coop. Wildl. Res. Unit, Oregon State University, Corvallis, OR U.S.A. Jenkins, J.M. 1992. Ecology and behavior of a resident population of Bald Eagles. Ph.D. dissertation, Univ. California, Davis, CA U.S.A. Kozie, K.D. and R.K. Anderson. 1991. Productivity, diet, and environmental contaminants in Bald Eagles nest- ing near the Wisconsin shoreline of Lake Superior. Arch. Environ. Contam. Toxicol. 20:41-48. Kralovec, M.L., R.L. Knight, G.R. Craig, and R.G. Mc- Lean. 1992. Nesting productivity, food habits, and nest sites of Bald Eagles in Colorado and southeastern Wyoming. Southwest. Nat. 37:356-361. Maybie, D.W., M.T. Merendino, and D.H. Reid. 1994. Dis- persal of Bald Eagles fledged in Texas. J. Raptor Res. 28:213-219. McAllister, K.R., T.E. Owens, L. Leschner, and E. Cum- mins. 1986. Distribution and productivity of nesting Bald Eagles in Washington, 1981-1985. Murrelet 67: 45-50. Mc:Garigal, K., R.G. Anthony and E.B. Isaacs. 1991. In- teractions of humans and Bald Eagles in the Colum- bia River estuary. Wildl. Monogr. No. 115. Newton, I. 1979. Population ecology of raptors. Buteo Books, Vermillion, SD U.S.A. Postupalsky, S. 1974. Raptor reproductive .success: some problems with methods, criteria, and terminology. Raptor Res. Rep. 2:21-31. oAb iiNSiii UTE iyo'y. isAN/ SlAl usei s guide, version 6, 4th Ed. SAS Institute Inc., Cary, NC U.S.A. Sherrod, S.K., C.M. White, and E.S.L. Williamson. 1976. Biology of the Bald Eagle on Amchitka Island, Alaska. Living Bird 15:143-182. SOKAL, R.R. and E.J. Rohlf. 1981. Biometry: the princi- ples and practice of statistics in biological research, 2nd Ed. W.H. Freeman and Co., New York, NY U.S.A. Sprunt, a., IV, W.B. Robertson, Jr., S. Postupalsky, RJ. Hensel, C.E. Knoder, and FJ- Ligas. 1973. Compar- ative productivity of six Bald Eagle populations. Trans. N. Am. Wildl. Nat. Resour. Conf. 38:96-106. 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. Wildl. Fed., Washington, DC U.S.A. Steidl, R.J., K. Kozie, and R. G. Anthony. 1997. Produc- tivity of Bald Eagles in central Alaska, f. Wildl. Manag;e. 61:1313-1321. Stocek, R.F. and P.A. Pearce. 1981. Status and breeding success of New Brunswick Bald Eagles. Can. Eield-Nat. 95:428-433. Swenson, J.E. 1975. Ecology of the Bald Eagle and Os- prey in Yellowstone National Park. M.S. thesis, Mon- tana State Univ., Bozeman, MT U.S.A. Todd, C.S. 1979. The ecology of the Bald Eagle in Maine. M.S. thesis, Univ. Maine, Orono, ME U.S.A. Van Horne, B. 1983. Density as a misleading indicator of habitat quality./. Wildl. Manage. 47:893-901. White, D.H. and J.T. Seginak. 1994. Dioxins and furans linked to reproductive impairment in wood ducks. J. Wildl. Manage. 58:100-106. Whitehead, P.E., J.E. Elliott, RJ. Norstrom, and K. Vermeer. 1990. PCDD and PCDF contamination of waterfowl in the Strait of Georgia British Columbia, Canada: 1989-90. Organohalogen Compounds 1:459- 462. Wiemeyer, S.N., TG. Lamont, C.M. Bunck, C.R. Sinde- lar, FJ- Gramlich, J.D. Fraser, and M.A. Byrd. 1984. Organochlorine pesticide, polychlorobiphenyl, and mercury residues in Bald Eagle eggs — 1969-1979 — and their relationships to shell thinning and repro- duction. Arch. Environ. Contam. Toxicol. 13:529-549. Received 13 May 2000; accepted 18 November 2000 J. Raptor Res. 35(1):9-14 © 2001 The Raptor Research Foundation, Inc. DISPERSAL OF JUVENILE AND IMMATURE BONELLI’S EAGLES IN NORTHEASTERN SPAIN Joan Real and Santi Manosa Departament de Biologia Animal, Facultat de Biologia, Universitat de Barcelona, Avda. Diagonal 645, E-08028 Barcelona, Catalonia, Spain Abstract. — Between 1986-93, we wing tagged and banded 122 Bonelli’s Eagles {Hieraaetus fasciatus) in north- eastern Spain to analyze their dispersal before recruitment to the breeding population. By 1998, we had obtained 18 band returns and 42 incidental observations of juvenile and immature eagles ranging in age from 1-3 yr. These eagles were recorded from 1-1020 km from their nests with a geometric mean distance (95% C.I.) of 101 km (72-153). Up to 57% of the birds remained within a radius of 100 km of their nests, whereas 33% were found beyond 200 km. Short-distance dispersers were found mainly in northeastern Spain in central Catalonia (<200 km), while long-distance dispersers (>200 km) were found in central and south- eastern Spain. No significant difference in dispersal distance was found between males (101 km, 40-273, N = 15) and females (189 km, 86-419, N = 11), but males were mosdy recorded at shorter distances. Dispersal distances of juveniles (114 km, 68-193, N = 43) and immatures (77 km, 44—135, N = 16) also did not differ significantly, but immatures were mosdy recorded at shorter distances. No significant difference was found between sighting (82 km, 54-137) and band recovery (167 km, 87-323) distances, but the proportion of band recoveries to sightings was lower for short- than for long-distance dispersers, and it increased with distance. The recording rate declined sharply at the end of the first year of life, suggesting high mortality during this period. The main causes of death were electrocution and human persecution. Most long-distance dispersers were reported dead, suggesting that long-distance movements entailed some mortality costs. Keywords: Bondli’s Eagle; Hieraaetus fasciatus; subadult dispersal; Catalonia. Dispersion de juveniles de aguila perdicera en Cataluna Resumen. — Describimos el patron de recuperaciones y observaciones de 122 Aguilas Perdiceras {Hieraaetus fasciatus) equipadas con marcas alares y anillas en Cataluna, Espana entre 1986-93 con objeto de obtener informacion sobre sus movimientos antes de que scan reclutadas en territorios de cria. Hasta el final de 1998, se obtuvieron 18 recuperaciones de anillas y 42 observaciones de aguilas no adultas. Las aguilas se registraron entre 1 y 1020 km de sus respectivos nidos. La media geometrica de la distancia de registro (95% I.C.) fue de 101 km (72-153). Un 57% de las aguilas permanecieron dentro de un radio de 100 km del lugar de nacimiento, pero un 33% se alejo mas alia de 200 km. La zona central de Cataluna fue la principal area de acogida para las aves que se dispersaron a corta distancia (<200 km), mientras que las aves que efectuaron largos desplazamientos (>200 km) se dirigieron principalmente al centro y al sureste de Espana. No se encontraron diferencias significativas en la distancia de dispersion entre machos (101 km, 40-273, N = 15) y hembras (189 km, 86-419, N = 11), aunque una elevada proporcion de machos se encontraron cercanos a las areas de nidificacion. La distancia recorrida no difirio entre juveniles (114 km, 68-193, N = 43) e inmaduros (77 km, 44—135, N = 16), pero estos ultimos se registraron mas frecuentemente a cortas distancias de las areas de nidificacion. No se encontraron diferencias significativas en la distancia segun fueran observaciones (82 km, 54-137) o recuperaciones (167 km, 87-323), pero la proporcion de recuperaciones en relacion a las observaciones fue menor para cortas que para largas distancias de dispersion e incremento con la distancia. La tasa global de recuperaciones y observaciones disminuyo acusadamente al final del primer ano de vida, sugiriendo una elevada mortalidad durante este periodo. Las causas principales de mortalidad fueron la electrocucion y la persecucion. Los ejemplares dispersados a larga distancia fueron mas frecuentemente registrados muertos que vivos, sugiriendo que los movimientos a larga distancia conllevan costes de mortalidad. [Traduccion de Autores] In many large birds of prey, juveniles show no- These juvenile dispersal movements, as they are madic behavior soon after becoming independent called, may lead young birds to settle in juvenile (Gonzalez et al. 1989, Cugnasse and Gramm 1990). dispersal areas, which are seldom occupied by 10 Real and Manosa VoL. 35, No. 1 breeding conspecifics (Ferrer 1993a, 1993b, Man- osa et al. 1998). In most cases, these nomadic birds are philopatric and return to breed near their na- tal areas (Newton 1979). Long-distance juvenile dispersal has been associated with individuals in the best of health (Ferrer 1992, 1993b, Walls et al. 1999). It allows them to explore and settle in new optimal and unoccupied areas (Horn 1983, Nils- son 1989). However, it may also entail several costs in terms of increased mortality or reduced lifetime reproductive success (Belichon et al. 1996), which is associated with suboptimal areas. Populations of Bonelli’s Eagles (Hieraaetus fascia- tus) are declining in most parts of Europe (Roca- mora 1994), mainly as a result of high mortality (Real and Manosa 1997). A sharp decline in the population in northern Spain has been partially attributed to unbalanced patterns of dispersal that may favor more southern populations (Real and Manosa 1997). Although adults are sedentary (Cramp and Simmons 1980), juvenile and imma- ture Bonelli’s Eagles (1-3 yr old) show a wandering behavior (Cramp and Simmons 1980, Cugnasse and Cramm 1990, Manosa et al. 1998), which has been poorly described. The aim of our study was to obtain data on the location of nonbreeding ar- eas used by juvenile and immature Bonelli’s Eagles hatched in northeastern Spain. We also collected information on the average distances traveled and the proportion of individuals involved in these movements. Finally, we discuss the implications of these movements in terms of the life history, pop- ulation status, and conservation of Bonelli’s Eagles. Methods We conducted a wing tagging and banding project in northeastern Spain in Catalonia involving 36 pairs and 83 breeding attempts of Bonelli’s Eagles from 1986-93. Every nestling between 40-55 d of age was marked with a 6-g metal ring on one leg, a 6-g, 3-digit PVC ring on the other leg, and plastic wing tags (Kochert et al. 1983) on each wing. The tag on the right wing identified the bird’s area of origin and the color of the left tag denoted the year of tagging. Wing tags measured 6.7 X 13.5 mm when folded and weighed 11 g including rivets. We wrapped the wing tags around the humerus between the tertiaries and scapulars and secured them with two pop rivets and glue. Wing-tag components were supplied by Saflag (Safety Flag Co. of America, Pawtucket, R1 U.S.A.) between 1986—87, and by TXN-18 (Cooley Inc., Pawtuck- et, R1 U.S.A.) between 1988-93. The age and sex of every nestling was determined at the time of banding following Manosa et al.(1995). We compiled data on movements of tagged eagles from band recoveries of birds found dead or injured and from incidental sightings of tagged birds. We requested information on sightings of tagged eagles in Spain, France, Portugal, Italy, and Morocco by advertising in journals and newsletters of the main ornithological and conservation associations of southern and central Europe and northern Africa. We did not consider successive sightings of the same tagged bird in a given area. Since our study focused on dispersal movements of juvenile and immature eagles and not on natal dispersal, we an- alyzed only sightings and band recoveries involving non- breeding birds <3 yr old. According to published infor- mation about the postfledging period of Bonelli’s Eagle (Real et al. 1998), the birds sighted or recovered before 100 days after fledging and within a radius of 8 km from the nest were not dispersed, and were not considered. We computed distances traveled and compass direc- tions from natal nests, if known, or from the geometric center of marking areas if the exact identity of birds was unknown. In this case, however, compass direction of the movement was only reported for birds observed outside the circle centered on the geometric center of the mark- ing area and encompassing all the nests where young eagles were tagged. Bird ages were all computed from 31 May (age 0) of the hatching year, the approximate average fledging date for the study population (Real 1991), We established two age categories: juvenile (<366 d of age) and immature (366-1095 d of age). As dispersal distances showed a skewed distribution, re- sults are given as geometric mean distances with 95% C.I We used 2-tailed Mests (unless otherwise stated) to com- pare log-transformed dispersal distances between sight- ings and recoveries, between males and females, and be- tween juvenile and immature categories. We used Fisher exact probability tests to compare percentages. We con- ducted standard statistical analyses with the SPSS package for Windows (SPSS-Inc. 1990) and computed circular sta- tistics following Zar (1984). Results After accounting for tag loss in nests and mor- tality during the nestling period, we estimated that 122 tagged nestlings successfully fledged from nests. At least four of them died during the depen- dency period (3.3% of tagged birds). One was killed by an Eagle Owl {Bubo bubo), one starved, and two were electrocuted. We recorded 18 band recoveries (15%) and 42 sightings (35%) of tagged eagles between 1986-98. Most of these occurred in the first year after fledging and declined thereafter (Fig. 1). Juvenile and immature eagles dispersed to 1- 1020 km from natal areas (Fig. 2). Because these distances showed a bimodal distribution (Fig. 3), we classed the eagles as either short-distance (<200 km) or long-distance dispersers (>200 km). Some juvenile Bonelli’s Eagles moved >500 km from their natal areas within 1 mo of fledging (Fig. 2). The geometric mean dispersal distance (95% March 2001 Bonelli’s Eagle Dispersal 11 Figure 1. Number of sightings and band returns of ju- venile and immature Bonelli’s Eagles in relation to time after fledging. C.I.) was 101 km (72-153, N = 60), with a median of 70 km. Although most birds (57%) were re- ported within 100 km of their natal sites, 33% went farther than 200 km (Fig. 3). The proportion of long-distance dispersers was similar in both sexes (5 of 1 1 females vs. 7 of 15 males) , and the distance traveled by females (189 km, 86-419, N = 11) was not significantly different from that traveled by males (101 km, 40-273, A = 15; ^4 = -0.98, P = 0.34). However, a higher proportion of males con- centrated at shorter distances (Fig. 4a). The dis- tance traveled did not differ between juvenile (114 km, 68-193, N = 43) and immature eagles (77 km, 44-135, N = 16; ^57 = 0.99, P = 0.33); but, while the number of juveniles increased with distance, immatures showed the reversed trend (Fig. 4b) . The average azimuth to which eagles traveled was 258 ± 59'*E, N = 33) . Most short-distance dis- persals were recorded in central Catalonia, fol- lowed by the Ebre delta and the Aiguamolls de I’Emporda (Fig. 5a, b). Long-distance dispersers were mainly reported in central (Madrid, Toledo, and Extremadura) and southeastern (Alicante, Murcia, Albacete, and Eastern Andalusia) Spain, with fewer reports in northern Spain and France (Fig. 6 ). The distance at which sightings (82 km, 54—137) and band recoveries (167 km, 87-323) were reported did notvary significantly (^59 = 1 . 66 , P = 0.10), whereas the proportion of band recov- eries to sightings was lower for short-distance dis- persers (9 vs. 31) than for long-distance dispersers (9 vs. 11) (one-tailed Fisher exact probability test = 0.07), and increased with distance, as did the absolute number of recoveries (Fig. 4c) . Of the 18 banded birds found dead, 10 (55%) § V o 1100 1000 900 800 700 600 500 400 300 2OOH 100 0 nfo a o Q) o rt9 366 730 1085 Days after fledging Figure 2. Dispersal distances of juvenile and immature Bonelli’s Eagles after fledging. Circles represent sightings and triangles represent band recoveries. were electrocuted, 4 (22%) were shot, trapped or poisoned, 1 (6%) starved, and the cause of death for the remaining 3 (17%) was unknown. For short-distance dispersers, 5 (72%) were electrocut- ed, 1 (14%) was shot, and 1 (14%) starved. For long-distance dispersers, 5 (63%) were electrocut- ed and 3 (37%) were killed by people. Discussion A significant fraction of young Bonelli’s Eagles produced in northeastern Spain travel long dis- tances. Although banding and wing tagging are more likely to provide information on populated areas (Kochert et al. 1983, Young and Kochert 1987, Francis and Cooke 1993), we believe that the 30 Kilometers from Nest Figure 3. Distribution of the number of sightings and band recoveries of juvenile and immature Bonelli’s Ea- gles according to distance (km) traveled. 12 Real and Manosa VoL. 35, No. 1 0-45 46-70 71-431 Kilometers from Nest >431 16 14 12 QJjuveniles ^ immatures Kilometers from Nest 18 16 14 '•2 I 10 I « 6 4 2 0 Figure 4. Distribution of distance traveled by (top) male and female Bonelli’s Ez^les, (middle) juvenile and imma- ture Bonelli’s Eagles, and (bottom) band recoveries and sightings. Intervals are 25% percentiles of the global dis- tance distribution. 0-45 46-70 71-431 >431 Kilometers from Nest effect of such bias on our results was low. The areas with high observation rates (northeastern, central, and southeastern Spain) are not among the most frequently visited by ornithologists and birdwatch- ers. Moreover, the main dispersal areas we have identified agree with those recendy-identified by conventional or satellite radio telemetry, as well as with areas where concentrations of juvenile and immature Bonelli’s Eagles have been previously re- ported (Arroyo and Garza 1996, Cheylan and Mar- masse 1998, Manosa et al. 1998). Male and female eagles showed different dis- persal patterns. Although both genders moved long distances, male Bonelli’s Eagles remained near their natal areas more often as is typical of other species of raptors (Greenwood et al. 1979, Newton and Mearns 1988, Ferrer 1993a, Walls and Kenward 1995). The opposite trend shown by ju- venile and immature eagles may indicate that Bo- nelli’s Eagles return to their natal areas as they grow older (Gonz^ez et al. 1989, Walls and Ken- ward 1994) or that long-distance dispersers suffer high mortality (Belichon et al. 1996). Higher mor- tality associated with dispersal could have account- ed for the marked decline in the number of sight- ings and band recoveries with age. Our data suggest that power line casualties and illegal per- secution by people remain chief causes of mortality for juvenile and immature Bonelli’s Eagles. The fact that relative and absolute numbers of recoveries (dead eagles) increased with distance from natal areas indicated that long-distance move- ments entailed a cost for eagle survival. Moreover, all the eagles that had moved to northern and At- lantic France, far from the usual distribution range of the species, were reported dead or in poor con- dition. Juvenile Bonelli’s Eagles that fledge in north- eastern Spain undertake long-distance movements more often than eagles in France (Cheylan et al. 1996), and may thus experience additional mor- tality (Belichon et al. 1996). The population of Bo- nelli’s Eagles in northeastern Spain is at the edge of the distribution range of the species, and might thus be particularly exposed to the negative con- sequences of dispersal (Gadgil 1971, Walters et al. 1999) . Since a large fraction of the Bonelli’s Eagles produced in northeastern Spain move to close or distant dispersal areas, their conservation should rely on the effective management of these areas. This includes the reduction of power line mortality and illegal persecution, and management to main- March 2001 Bonelli’s Eagle Dispersal 13 Figure 5. Maps of the study area showing short-distance (<200 km) records of juvenile and immature Bonelli’s Eagles. The shaded area shows the tagging region. Lines join known natal sites and recovery sites. Figure 6. Map of long-distance (>200 km) band recov- ery and sightings of juvenile and immature Bonelli’s Ea- gles dispersing from northeastern Spain. Lines join known natal sites and recovery sites. Band returns ■ Immature □ Juvenile Sightings # Immature 0 Juvenile tain sufficient prey availability (Bustamante et al. 1997, Ferrer and Harte 1997, Manosa et al. 1998). These measures should be undertaken both in northeastern Spain and in distant dispersal areas in southern and central Spain. Acknowledgments We are grateful to J. Godina, R. del Amo, and D. Mo- lina for their help with banding and tagging. F. Hiraldo and J. Calderon from Estacion Biologica de Dohana pro- vided scientific support and PVC rings. M. Kochert helped us obtain the fabrics for the tags. The Fundacio Miguel Torres, the Diputacio de Barcelona, and the Fon- dation Suisse pour les Rapaces provided financial sup- port. We also thank all the people who sent us observa- tion or band return data, especially G. Cheylan, T. Sanchez, L. Rico, and the Oficina de Anillamiento del Ministerio de Medio Ambiente and the Grup Catala d’Anellament. Quercus Journal and the Assessoria Juri- dica of the Barcelona University are also acknowledged for their support and help. We are very grateful to M, Kochert, B. Arroyo, M. Bechard, and to an anonymous referee whose comments greatly improved the manu- script. Literature Cited Arroyo, B. and D.V. Garza. 1996. Movimientos disper- sivos y caracterizacion del habitat del aguila perdicera. Tech. rep. ICONA, Madrid, Spain. 14 Real and Manosa VoL. 35, No. 1 ljilKxHOjNi, i>., J, Cloberi, and M. Mabsoi. 1996. VVliai differences in fitness components between philopa- tric and dispersing individuals? Acta Oecol. 17:503-517. Bustamante, J.,J. A. Donazar, F. Hiraldo, O. Ceballos, AND A. Travaini. 1997. Differential habitat selection by immature and adult Grey Eagle-Buzzards Geranoae- tus melanoleucus. Ibis 139:322-330. Cramp, S. and K.E.L. Simmons. [Eds,]. 1980. The birds of the western Palearctic. Vol. 11. Oxford Univ. Press, Oxford, U.K. Cheylan, G. and a. Marmasse. 1998. Suivi par balises Argos de trois jeunes Aigles de Bonelli. Circulaire du Group de Travail Mondial sur les Rapaces (GTMR) 25— 28:22-24. , A. Ravayrol, and J.M. Cugnassee. 1996. Disper- sion des aigles de Bonelli Hieraaetus fasciatus juveniles bagues en France. Alauda 64:413-419, CUGNASSE, J.M. and P. Cramm. 1990. L’erratisme de I’aigle de Bonelli Hieraaetus fasciatus en France. Alau- da 58:59-66. Ferrer, M. 1992. Natal dispersal in relation to nutritional condition in Spanish Imperial Eagles. Ornis Scand. 23: 104-106. . 1993a. juvenile dispersal behaviour and natal philopatry of a long-lived raptor, the Spanish Imperial Eagle Aquila adalberti. Ibis 135:132-138. . 1993b. Ontogeny of dispersal distances in young Spanish Imperial Eagle Aquila adalberti. Behav. Ecol. So- ciobiol. 32:259-63. and M. Harte. 1997. Habitat selection by imma- ture Spanish Imperial Eagles during the dispersal pe- riod. /. Applied Ecol 34:1359-1364. Francis, C.M. and F. Cooke. 1993. A comparison of sur- vival rate estimates and dead recoveries of Lesser Snow Geese. Pages 169-183 mJ.D. Lebreton and P.M. North [Eds.], Marked individuals in the study of bird populations. Birkhauser Verlag, Basel, Switzerland. Gadgil, M. 1971. Dispersal: population consequences and evolution. Ecology 52:253-261, GonzAlez, L.M., B. Heredia, J.L. Gonzalez, and J.C. Alonso. 1989. Juvenile dispersal of Spanish Imperial Eagles. / Field Ornithol 60:369-379. Greenwood, P.J, P.H. Harvey, and C. Perrins. 1979. The role of dispersal in the Great Tit (Parus major)-, the causes, consequences and heritability of natal dispers- al./ Anim. Ecol 48:123-142. Horn, H.S. 1983. Some theories about dispersal. Pages 54-62 in R. Swingland and P.J. Greenwood [Eds.], The ecology of animal movement. Oxford Univ. Press, Oxford, U.K. 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. Manosa, S., J. Real, and J. Godina. 1995. Age estimation and growth patterns in nestling Bonelli s Eagles. / Raptor Res. 29:273-275. , , AND . 1998. Selection of settle- ment areas by juvenile Bonelli’s Eagle in Catalonia./. Raptor Res. 32:208-214. Newton, I. 1979. Population ecology of raptors. T & A.D. Poyser, Berkhamsted, U.K. AND R. MeARNS. 1988. Population ecology of per- egrines in South Scotland. Pages 651-665 mT.J. Cade, J.H. Enderson, C.G. Thelander, and C.M. White [Eds.], Peregrine Falcon populations. The Peregrine Fund, Inc. Boise, ID U.S.A. Nilsson, J.A. 1989. Causes and consequences of natal dis- persal in the Marsh Tit Parus palustris. J. Anim. Ecol 58:619-636. Real, J. 1991. L’aliga perdiguera Hieraaetus fasciatus a Ca- talunya: status, ecologia trofica, biologia reproductora i demografia. Ph.D. dissertation, Univ. Barcelona, Bar- celona, Catalonia, Spain. AND S. Manosa. 1997. Demography and conser- vation of western European Bonelli’s Eagle Hieraaetus fasciatus populations. Biol Conserv. 79:59-66. , , AND J. Codina. 1998. Post-nestling de- pendence period in the Bonelli’s Eagle Hieraaetus fas- ciatus. Ornis Fenn. 75:129-137. Rogamora, G. 1994. Bonelli’s Eagle Hieraaetus fasciatus Pages 184—185 in G.M. Tucker and M.F. Heath [Eds.], Birds in Europe, their conservation status. BirdLife International, Birdlife Conservation Series No. 3, Cambridge, U.K. SPSS-Inc. 1990. SPSS reference guide. SPSS Inc., Chica- go, IT U.S.A. Young, L.S. and M.N. Kochert. 1987. Marking tech- niques. Pages 125—156 in B.A. Giron Pendleton, B.A. Millsap, K.W. Cline, and D.M. Bird [Eds.], Raptor management techniques manual. Nat. Wildl. Fed Washington, DC U.S.A. Walls, S. and R.E. Kenward. 1994. The systematic study of radio-tagged raptors: II. Sociality and dispersal Pages 317-324 in B-U. Meyburg and R.D. Chancellor [Eds.], Raptor conservation today. WWGBP, Pica Press, U.K. and . 1995. Movements of radio-tagged Common Buzzards Buteo buteo in their first year. Ibis 137:177-182. , S. Manosa, R.M. Fuller, K.H. Hodder, and R.E Kenward. 1999. Is early dispersal enterprise or exile? / Avian Biol 30:407-415. Walters, J.R., H.A. Ford, and C.B. Cooper. 1999. The ecological basis of sensitivity of Brown Treecreepers to habitat fragmentation: a preliminary assessment Biol Conserv. 90:13-20. Zar, J.H. 1984. Biostatistical analysis. Prentice-Hall, En- glewood Cliffs, NJ U.S.A. Received 13 May 2000; accepted 18 November 2000 J. Raptor Res. 35(1);15-19 © 2001 The Raptor Research Foundation, Inc. ESTIMATING THE BREEDING POPULATION OF BOOTED EAGLES IN THE CAPE PROVINCE, SOUTH AFRICA David Pepler Department of Nature Conservation, Faculty of Agricultural and Forestry Sciences, University of Stellenbosch, Private Bag XI, Matieland 7602, South Africa Rob Martin Department of Forest Science, Faculty of Agricultural and Forestry Sciences, University of Stellenbosch, Private Bag XI, Matieland 7602, South Africa Hubertus j. van Hensbergen Department of Nature Conservation, Faculty of Agricultural and Forestry Sciences, University of Stellenbosch, private Bag XI, Matieland 7602, South Africa Abstract. — Data on the breeding range of Booted Eagles {Hieraaetus pennatus) were collected over 25 yr in the Northern, Western, and Eastern Cape Provinces, South Africa, to estimate the breeding pop- ulation. Based on the distribution of 150 known nest sites, we used information from digital terrain models to define topographical characteristics of nest sites. This information was used to identify the total suitable nesting habitat in the study area. By calculating the mean inter-nest distance, we estimated the total nesting population through extrapolation. With a mean inter-nest distance of 9.7 km, we arrived at an estimate of 702 nests. In core areas that we have studied intensively, we found even higher breeding densities and therefore consider our estimate to be conservative. Key Words: Booted Eagle, Hieraaetus pennatus; breeding density; South Africa; CIS habitat delineation; pop- ulation estimation. Estimacion de la poblacion reproductiva de Hieraaetus pennatus en la provincia del Cabo, Surafrica Resumen. — Recopilamos datos sobre el rango de reproduccion de Hieraaetus pennatus durante 25 anos en el norte, oeste y este de la Provincia del Cabo, Surafrica para estimar la poblacion reproductiva. Con base en la distribucion de 150 sitios de nidos, utilizamos informacion de modelos digi tales del terreno para definir las caracteristicas topograficas de los sitios de anidacion. Esta informacion fue utilizada para identificar el total del habitat propicio para anidacion en el area de estudio. A1 extrapolar el calculo de la media de la distancia entre nidos, estimamos el total de la poblacion anidante. Con una media de distancia entre nidos de 9.7 km, llegamos a un estimativo de 702 nidos. En las areas centrales que hemos estudiado intensivamente, encontramos densidades aun mas altas, por lo tanto consideramos que nuestro estimativo es conservador. [Traduccion de Cesar Marquez] In contrast to the Palearctic region, where the bi- ology of the Booted Eagle {Hieraaetus pennatus) is well-known (Cramp and Simmons 1980, del Hoyo et al. 1994, Suarez et al. 2000), only its breeding biology has been studied in southern Africa (Steyn and Grobler 1981, 1985). This breeding popula- tion was only recently discovered (Martin and Mar- tin 1974, Brooke et al. 1980) and the first modern breeding record was confirmed in 1973 (Martin and Martin 1974). The extent of the breeding range was clarified by models of seasonality and associated breeding by Boshoff and Allan (1997) and Harrison et al. (1997) but the range is com- plicated because there appear to be three separate populations of Booted Eagles in southern Africa (Boshoff and Allan 1997). These populations con- sist of nonbreeding summer migrants from the Pa- learctic region, a relict breeding population from the Waterberg in Namibia, and a breeding popu- lation in the Cape Province (del Hoyo et al. 1994, Brooke et al. 1980). To further complicate the sit- uation, some eagles overwinter in the southwestern Cape Province (Pepler and Martin 1997). To date, only one estimate of breeding population (400 15 16 Pepler et al. VoL. 35, No. 1 I / >■! Figure 1. The study area in southern Africa. pairs) has been made for the Cape Province (Mar- tin and Martin 1991). We collected data on the breeding range and density of the Booted Eagle population in the Cape Province over the past 25 yr through direct observations. We used elevation information from digital terrain models to define the topographical characteristics of Booted Eagle nesting sites based on the distribution of 150 known sites to deter- mine the total area of suitable habitat. To estimate the total nesting population, we established the nesting density within suitable habitat by calculat- ing the mean inter-nest distance following the method of Pepler et al. (1991). In this method a plot of the cumulative sum of deviations from the running mean (Lombaard 1989) indicates changes m density with distance. This method is more typ- ically used in the analysis of time-series data but is applied in this case to the distance series. A system- atic deviation from zero is indicative of a change in trend, in this case inter-nest distance. Study Area and Methods The study area covered the portion of southern Africa south of 31°S and west of 26°E (Fig. 1). The total land area was 263 532 krn^. The town of Stellenbosch (33°55'S, 18°52'E) was used as our base. Our study area covered most of the perceived breeding range as described by Steyn (1982) and Boshoff and Allan (1997). The study was conducted annually from 1975-93. Timing of field- work was restricted to the breeding season, which was typically from September to December. Within our study area, Booted Eagles hunted and bred in both hilly and open country, preferring habitats con- sisting of nama karoo, succulent karoo and fynbos (Low and Rebelo 1996) and, especially, the ecotones between these habitat types (Boshoff and Allan 1997). In recent years, however, we observed Booted Eagles hunting in suburban areas (Pepler and Martin 1996), and it is rea- sonable to assume that records of breeding from within these areas will be found in due course. We searched for nests from roads that gave access to iiiouiitaiiious aieas, and sLiategic vaniage pomes were se- lected that afforded the greatest possible field of view. The total distance covered during the course of the field- work was in excess of 500 000 km (approximately 2 km traveled per km^ of the study area or 0.1/km per yr). Occupied nest sites were confirmed when one or both adults were seen carrying nesting material to a specific site, prey items were seen being carried to sites, or young were observed at sites before they fledged. Nests were typically situated behind small trees or shrubs growing on cliffs. Whitewash around occupied nest sites has a unique streaked appearance that helped us locate them from greater distances. Breeding sites were plotted on 1:250 000 topographical maps and subsequently digitized into a format compati- ble with an ARC/INFO Geographic Information System (Environmental Systems Research Institute, Inc., 380 New York Street, Redlands, CA 92373-8100, U.S.A.). El- evation data for the study area were obtained from na- tional digital elevation data (Department of Land Infor- mation Systems, Private Bag XIO, Mowbray 7705, South Africa) and these data were also imported into the GIS. The vertical interval for the elevation data was 250 m. Booted Eagle nest sites are associated with mountain- ous country with broken terrain (Steyn 1982). Elevation alone was not a good indicator of suitable habitat because the eagles do not nest on high plateaus. Steep slopes were also indicators of suitable habitat but evenly sloping areas are not used for nesting. The broken terrain used by Booted Eagles was identified using the rate of change of slope which was determined from the second deriva- tive of the function describing the surface. The function was calculated for each point from the eight elevations immediately adjacent to a point as well as the elevation of the point itself. The value referred to a point at the center of a grid of nine points with a total dimension of 678 X 678 m. High values of this parameter indicated the rapidly changing slopes associated with broken hilly country while excluding plateaus and smooth inclines. Low values indicated constant slope. A number of values of this parameter were tried until one, which by inspec- tion of the area covered on the map just included the distribution of the majority of the known nests, was found. The term “mountainous,” in the context of this analysis, was taken as any area with a second derivative of height greater than, or equal to, 0.2. The cell size for the analysis was fixed at 226 X 226 m (51 076 m^) since this was the scale at which the second derivative was calculat- ed. Finally, we calculated the total number of cells in the mountainous and nonmountainous areas and the per- centage of the study area that was mountainous. Since the survey was carried out from roads, it was pos- sible that undetected nests in areas isolated from roads might cause an overestimate of the inter-nest distance and an underestimate of the density. Similarly, the survey was spread over a wide area and it was likely that nests closer to our base in Stellenbosch could have been more likely to be found. To determine if this was the case, we calculated the cumulative sum of deviations from the running mean (CUSUM, Lombaard 1989, Pepler et al. 1991) for inter-nest distances based on the observations ordered in increasing distance from a road and also on increasing distance from Stellenbosch. March 2001 Estimation of Booted Eagles in South Africa 17 * mi' Figure 2. Breeding distribution of Booted Eagles in southern Africa. Nest sites are flagged and areas with second derivatives of the surface function >0.2 are shaded. Figure 3. Breeding distribution of Booted Eagles in southern Africa. Nest sites are flagged and areas of steep slope with first derivative values >6..5 are shaded. 18 Pepler et al. VoL. 35, No. 1 0 200 400 600 800 Distance from Stellenbosch Figure 4. Cumulative sum of deviations from the running mean (CUSUM) plot for Booted Eagle intermest distances (km) and distances from Stellenbosch (km) . Results The breeding distribution of Booted Eagles cor- responded closely with areas where the second de- rivative of the surface function had values >0.2 (Fig. 2). This differed from areas of steep slopes where first derivative values were >6.5 (Fig. 3). The total area of the habitat identified is 61 663 km^. CUSUM showed a possible change in inter-nest distance for sites in excess of 200 km from Stellen- bosch (Fig. 4) , so we based our calculations of in- ter-nest distances on nests at distances <200 km from Stellenbosch. There was no evidence that nest detection was based on distance from roads since no change point was evident in the CUSUM plot. Therefore, we used observations within 200 km of Stellenbosch to calculate our estimate of breeding density. Calculation of this estimate was based on two crucial assumptions. First, that the area included in our survey was representative of the entire area in terms of the average inter-nest distance and, second, that the estimate of inter- nest distance was accurate. The mean inter-nest distance was 9.677 km (95% Cl = 9.17-10.18) and the estimate of the total breeding population for the study area was 702 pairs (95% Cl = 576-879). Discussion The results of our analysis were comparable to the atlas data of Harrison et al. (1997), especially with their models of breeding rate based on sea- sonality and breeding in zones four and eight. Our study was carried out over 25 yr and we as- sumed that all the recorded nests remained occu- pied throughout the study period. We made this assumption because we have, in a number of cases, observed the continuous occupation of particular nesting sites for periods ranging from 1972-99. Be- March 2001 Estimation of Booted Eagles in South Africa 19 cause our observations were made from roads, we made no attempt to sample many of the mountain massifs. Therefore, it was possible that densities within these massifs were lower than we estimated. Given the very high breeding density of four pairs of breeding Booted Eagles in a 3 km^ area that was recorded by Martin and Martin (1988, 1995) in parts of our study area, we considered our calculation of the total population to be a substan- tial underestimate. This did not imply that such a high density is evenly maintained throughout the entire breeding range, but the severe constraints of time and logistics placed on a study of this na- ture preclude saturation sampling. An example of such an undersampled area is the mountain range of the Great Escarpment between Beaufort West (32°20'S, 22°38'E) and Calvinia (3r2'7'S, 19°50'E). We believe that our data present the first attempt at an estimation of an entire breeding population of Booted Eagles in Africa. It has been suggested that this southern breeding population may be a subspecies of the northern Booted Eagle (R. Yosef and G. Verdoorn pers. comm.), but this needs con- firmation. Elsewhere in its range only “fairly ap- proximate estimates” of breeding density exist (del Hoyo et al. 1994). With time, we are confident that additional data will expand our database and in- crease the accuracy of our breeding density calcu- lations. Acknowledgments We would like to thank E. Martin and the late J. Martin for their invaluable contributions to this project. We ac- knowledge the support of the Division of Research Ad- ministration, University of Stellenbosch. We thank B. Op- perman for technical advice, for plotting distributions and producing the distribution maps. Litera'i URE Cited Boshoff, a. and D.G. Allan. 1997. Booted Eagle Hieraae- tus pennatus. In J.A. Harrison, D.G. Allan, L.G. Un- derhill, M. Herremans, A.J. Tree, V. Parker, and C.J. Brown. [Eds.], The atlas of southern African birds. Vol. 1. Non-passerines. BirdLife International, Johan- nesburg, South Africa. Brooke, R.K., R. Martin, J. Martin, and E. Martin. 1980. Booted Eagle, Hieraaetus pennatus, as a breeding species in South Africa. Gerfaut 70:297-304. Cramp, S. and K.E.L. Simmons. [Eds.]. 1980. Handbook of the birds of Europe, the Middle East and North Africa. Oxford Univ. Press, Oxford, U.K. DEL Hoyo, J., A. Elliot, andJ. Sargatal. [Eds.]. 1994 Handbook of the birds of the world. Lynx Edicions, Barcelona, Spain. Harrison, J.A., D.G. Allan, L.G. Underhill, M. Herre- mans, A.J. Tree, V. Parker, and C.J. Brown. [Eds.] 1997. The atlas of southern African birds. Vol. 1. Non- passerines. BirdLife International, Johannesburg, South Africa. Lombaard, F. 1989. Some recent developments in the analysis of changepoint data. S. Afr. Statist.]. 23.T-21 Low, A.B. and A.T. Rebelo. [Eds.]. 1996. Vegetation of South Africa, Lesotho and Swaziland. Dep. Environ Affairs and Tourism, Pretoria, South Africa. Martin, J.E. and R. Martin. 1974. Booted Eagle breed- ing in the southwestern Cape Province. Bokmakierie20. 21 - 22 . Martin, R. and J.E. Martin. 1988. Observations on Boot- ed Eagles. Promerops 182:9-11. and . 1991. Booted Eagles: an estimate of breeding numbers in the Cape. Promerops 198:8-9. and . 1995. Notes on two pairs of Booted Eagles breeding in the Boland. Promerops 2.11 :\2>. Pepler, D. and R. Martin. 1996. Suburban sightings of Booted Eagles. Promerops 226:10-11. AND . 1997. Booted Eagle overwintering m the Stellenbosch area. Promerops 231:12. , HJ. VAN HeNSBERGEN, AND R. Martin, 1991. Breeding density and nest site characteristics of Per- egrine Falcon Falco peregrinus minor in the south-west- ern Cape, South Africa. Ostrich 62:23—28. Steyn, P. 1982. Birds of prey of southern Africa. David Philip, Cape Town, South Africa. AND J.H. Grobler. 1981. Breeding biology of Booted Eagles in South Africa. Ostrich 52:108-118 AND . 1985. Supplementary observations on the breeding biology of Booted Eagles in Southern Africa. Ostrich 56:151-156. SuArez, S-, J. Balbotin, and M. Ferrer. 2000. Nesting hab- itat selection by Booted Eagles Hieraaetus pennatus and implications for management. J. Appl. Ecol. 27:215- 223. Received 5 December 1998; accepted 22 October 2000 J Raptor Res. 35(1) ;20-23 © 2001 The Raptor Research Foundation, Inc. SEX DETERMINATION IN BOOTED EAGLES {HIERAAETUS PENNATUS) USING MOLECULAR PROCEDURES AND DISCRIMINANT FUNCTION ANALYSIS Javier Balbontin, Miguel Ferrer, and Eva Casado Estacion Bioldgica de Donana (CSIC) Avda. Maria Luisa s/n, Pabellon del Peru, 41013 Sevilla, Spain Abstract. — ^We studied a breeding population of Booted Eagles {Hieraaetus pennatus) in Donana Na- tional Park (southwestern Spain) to develop a method of determining the sex of an individual based on the use of discriminant functions. Because there are size differences between age classes and sexes of eagles, we developed two different discriminant functions for each age group. Our discriminant function method approached 100% accuracy in correctly aging individuals using forearm length and body mass as predictor variables. Sex of young eagles was also determined with 98.8% accuracy using forearm, tail, bill, and tarsus lengths. Key Words; Booted Eagle, Hieraaetus pennatus; sex determination', morphometries', molecular sexing. Determinacion del sexo del aguila calzada Hieraaetus pennatus utilizando tecnicas de sexado molecular y analisis discriminates Resumen. — Una poblacion reproductora de aguila calzada ha sido estudiada en el Parque Nacional de Donana (Sudoeste de Espaha) con el objetivo de obtener un modelo de clasificacion de los sexos basados en analisis discriminantes apoyados en procedimientos de sexado molecular. Existen diferencias importantes en el tamaho entre aguilas adultas y polios, por lo que se han desarrollado dos funciones discriminantes de sexo diferentes para cada clase de edad. El sexo de los adultos se determina con una funcion discriminante que clasifica bien el 100% de los individuos, utilizando el antebrazo y el peso como variables predictoras. El sexo de los polios es determinado tambien correctamente con una fun- cion discriminate que clasifica bien el 98.8% de los individuos, utilizando cuatro variables predictoras: El antebrazo, la cola, el pico y el tarso. [Traduccion de Autores] Easy and reliable methods to identify the sex of individuals are useful for the study of many aspects of avian biology, including foraging ecology (An- derson and Norberg 1981), behavior, evolutionary ecology and genetics (Glutton-Brock 1986), survi- vorship (Newton et al. 1983), and dispersion and conservation genetics (Griffith and Tiwari 1995). Sex determination is also important in conserva- tion programs that concern the reintroduction of endangered birds when a fixed sex ratio is pre- ferred. Recently, Ellegren (1996) proposed molec- ular methods to sex birds based on chromosome differences but few studies have used this infor- mation to develop additional methods to sex birds based on biometric data. Field methods to sex rap- tors have several advantages over molecular tech- niques that require time and/or money. Despite the fact that the majority of raptors are highly di- morphic in size, which should allow for the devel- opment of sexing methods based on morphomet- ric data, only a few species have been utilized (Bortolotti 1984a, 1984b, Garcelon et al. 1985, Ed- wards and Kochert 1987, Ferrer and De Le Court 1992). The majority of these studies have been based on live individuals and museum skins. In most cases, both adults and immatures have been studied at museums or in private collections and few studies have been based on wild individuals. The objective of this study was to assess the differ- ences between young and adult Booted Eagles {Hi- eraaetus pennatus) and to develop predictive dis- criminant models to determine the sexes of adults and immatures of the species. Methods We used a sample of the breeding population of Boot- ed Eagles in Donana National Park. The park is located in southwestern Spain (37°N,6°30'W). It has a Mediter- ranean climate with an Atlantic Ocean influence. Marsh- es, Mediterranean scrubland mixed with scattered cork oak ( Quercus suber) or stone pine {Pinus pined) , and costal 20 March 2001 Sex Determination in Booted Eagles 21 Eagle. sand dunes are the main habitats found in the area. A more detailed description of this area is presented in Rogers and Myers (1980). Six morphometric measurements were taken from wild adult and immature eagles. To obtain measurements, we visited nests when young were 35-45 d old and their skel- etons were completely grown but their feathers were still growing. Young leave the nest when they are about 55 d old (Balbontin unpubl. data). A total of 100 young were measured between 1996-98. Adults were trapped using a 2 X 3 m dho gaza net and an unreleasable captive owl {Bubo bubo) lure. Forty-two adults were caught using this method, 12 in 1997 and 30 in 1998. We took measure- ments of wing, tail, bill with cere, and tarsus lengths using a digital caliper to the nearest 0.1 mm and metal rulers to the nearest 1 mm (Bortolotti 1984). We also measured the forearm length, or the length from the front of the folded wrist to the proximal extremity of the ulna using calipers (Fig. 1) (Ferrer and De Le Court 1992). All the individuals were weighed with 1 kg or 2.5 kg Pesola scales with precisions of 5 g and 10 g, respectively. We extracted 2 ml of blood from the brachial vein of each eagle and stored part of it (50 |jl1) in buffer and kept it refrigerated for later analysis. The cellular fraction was used to sex the eagles following Ellegren (1996). We used primers 2945F, cfR, and 3224R to amplify the W- chromosome gene following Ellegren’s (1996) recom- mendations. Using this technique, we identified the sexes of 81 immature (41 females, 40 males) and 41 adult ea- gles (16 males, 25 females) (Fig. 2). This sample of known-sex individuals was used to derive the discriminant function using morphometric data. Because young often differ in size from adults (Bor- tolotti 1984b), we used multivariate analysis of variance (MANOVA) to check for differences in size between males and females and young and adult eagles. Six mea- surements taken from all age and sex classes were com- pared using univariate analysis of variance ( ANOVA) and nonparametric statistics for those variables when homo- geneity of variance was not met. We used the SPSS pro- gram (Norusis 1992) to do this an^ilysis. We separated young from adults to examine differences between sexes. First, we checked for sexual differences for each of the six morphological characters using t-tests. We derived a discriminant function using DISCRIM procedure of the SAS System program (version 6.12). A jackknife proce- dure was applied to test the efficacy of the estimated dis- criminant function (Lachenbruch and Mickey 1968). Each individual in the sample was classified using a dis- criminant function derived from the total sample, ex- cluding the individual being classified (Chardine and Morris 1989, Amat et al. 1993). We chose the function which had the lowest percentage of misclassification based on the molecular determination of gender. Results and Discussion Our analyses of the morphometric data showed that adult Booted Eagles differed significantly in size from young eagles and that males were signif- icantly smaller than females (MANOVA; sex — F = 72.0, df = 6, 111, P < 0.001; age - P = 181.85, df = 6, 111, P < 0.001). Tail, wing, and culmen mea- surement showed the greatest difference between age groups, with the features of adult individuals larger than those of immatures (Table 1). There were no significant age- or sex-specific differences in bone measurements such as tcirsus and forearm FMFMFFFM FMFMFFMM F F F M F Figure 2. Gender identification using PGR test. A multiple amplification with 2945F and cfR specifically amplify a 210 bp fragment of the W chromosome in females and 2945F + 3224R that amplifies 630 bp fragments in both sexes. Females are indicated by the arrow. Table 1 . Morphometric measurements in mm of young and adult Booted Eagles. 22 Balbontin et al. VoL. 35, No. 1 t-H ^H I — ( ^H \Ti I— i o o o o O o o o o o d d d d d d A V V V V V j> i> E> d 1 d 1 00 CO CO Oi tL CM in d d d d rH CO I-H q CM GO a; CM d H Q d CO 1-H d d CM c/c II +1 +1 + 1 +1 +1 + 1 5 II > +1 00 CM CM IX d oo d d 00 OcS rAi CD oo 00 o J> u I-H oo CM CD j < S W o UO CO in OO 00 cso o 00 00 o p 00 d ^H CM CM 00 Z: P C/D 1 1 +1 + 1 + 1 + 1 +1 +1 5 > tI oo o ac CD Oi IX d d d d d 00 CD 00 'Cf I-H CM t-H CM 1-H 00 oo 00 t-H I-H 1-H CO lO o o o m ITS CD o o o o d d d d d d V V V A o o CO N GO oq irj 1 d 1 I-H 1 00 uo 00 CO I-H d d d lO CM CO in q ^H ov CD d H Q d d I-H d 00 f-H C/!l P 1 1 +1 + 1 + 1 +1 + 1 +1 II > + 1 CM *5 00 00 CD iX d d rH oo d d CD 00 OO CO Oi CD rH oo rH CO nH cn oc CO I> iD CD CM d 00 00 0 o Q d d rH CM rH CD z .... CZ) + 1 + 1 + 1 + 1 + 1 +1 0 s > +1 00 q q iX d I-H 00 d r-H d CD CO CM CM m 1-H CM I-H CO a a o; fU u a bC d XAl 0 U ^ 3 lengths but forearm length was significantly small- er in young female eagles (Table 1 ) . Booted Eagles show high sexual dimorphism in size and both adults and young differed significantly in the ma- jority of the variables we studied. Adult females were significantly larger than males for all mea- surements taken, with forearm and body mass the most dimorphic characters (Table 2). Young fe- males are also larger than males and they have also longer forearms and beaks, but similar-sized wings and tails. Our discriminant function analysis clas- sified 100% of the adult female and male eagles correcdy using body mass and forearm as predictor variables. The discriminant function equation for adults was: D - -178.885 + 0.05613(MASS) + 0.95937 (FOREARM) Young were classified most accuracy using the four variables forearm, tail, bill, and tarsus as predictors in the model. The discriminant function misclas- sified only one female. The discriminant function for young was: D - -197 + 0.6761 (FOREARM) - 0.19286 (TAIL) + 2.99438 (BILL) + 0.5858 (TARSUS) Values of D > 0 represent females and values of D < 0 represent males. By deleting tail and wing measurements which are highly variable from the model, young eagles were also classified with 84% accuracy using only tarsus and forearm measure- ments in the discriminant function: D = -33.815 + O.I47(FOREARM) + 0.207 (TARSUS) The equations we derived for sexing Booted Ea- gles should be useful for future work on the biol- ogy of this species. For immature eagles, measure- ments of wings and tails should be taken carefully if they are used to discriminate gender because the feathers of young birds keep growing after they first take flight. Adults were correctly classified to gender in 100% of cases examined by using the two variables, body mass and forearm. The latter is an easy measurement to take and repeated mea- surements taken by different observers showed low variances (Ferrer and De Le Court 1992). Gender discrimination for young eagles is valid at 35-45 d of age when nestlings have almost completed their growth. March 2001 Sex Determination in Booted Eagles 23 Table 2. Differences in morphometric measurements between male and female young and adult Booted Eagles. Adults Young Male (N= 16) (x ± SD) Female {N = 25) (x ± SD) t P Male {N = 40) (x ± SD) Female {N= 41) (x ± SD) t P Tarsus 64.1 ± 2.77 69.4 ± 3.23 -5.715 <0.001 64.4 ± 2.51 69.1 ± 3.30 -7.15 <0.001 Forearm 132.2 ± 2.64 143.5 ± 3.20 -12.40 <0.001 131.5 ± 4.72 140.0 ± 4.85 -7.95 <0.001 Culmen 31.5 ± 1.14 34.8 ± 1.32 -8.604 <0.001 28.8 ± 1.29 30.9 ± 1.09 -7.93 <0.001 Wing 355.0 ± 27.8 389.2 ± 9.41 -5.712 <0.001 244.4 ± 25.9 244.6 ± 28.8 -0.03 0.970 Tail 195.6 ± 8.41 208.7 ± 9.24 -4.763 <0.001 121.0 ± 18.6 112.9 ± 21.5 1.78 0.078 Mass 690.3 ± 40.9 973.2 ± 76.9 -13.46 <0.001 656.3 ± 68.7 828.7 ± 88.3 -9.81 <0.001 Acknowledgments We wish to thank to H. Le Franc and J. Cuesta for their help with the field work. J.M. Arroyo did the lab work and F. Moreno helped with the drawings. We are indebt- ed to L. Garcia for his help assisting, locating, and show- ing us eagle territories. We also wish to thank M. Berto- lotti for reviewing an early draft of this article. Literature Cited Amat, J.A., J. Vinuela, and M. Ferrer. 1993. Sexing Chin- strap Penguins {Pygoscelis antarctica) by morphological measurements. Colonial Waterbirds 16:213—215. Anderson, M. and R.A. Norberg. 1981. Evolution of re- versed sexual size dimorphism and role partitioning among predatory birds, with a size scaling of flight performance. Biol. J. Linn. Soc. 15:105-130. Bortolotti, G.R. 1984a. Sexual size dimorphism and age-related size variation in bald eagles./. Wildl. Man- age. 48:72-81. . 1984b. Age and sex size variation in Golden Ea- gles. J. Field Ornithol. 55:54-66. Chardine, J.W. and R.D. Morris. 1989. Sexual size di- morphism and assortative mating in the Brown Nod- dy. 91:868-874. Clutton-Brock, T.H. 1986. Sex ratio variation in birds. Ibis 128:317-329. Edwards, T.C. and M.N. Kochert. 1987. Use of body weight and length of footpad as predictors of sex in Golden Eagles./. Field Ornithol. 58:144—147. Ellegren, H. 1996. First gene on the avian W chromo- some (CHD) provides a tag for universal sexing of non-ratite birds. Proc. R. Soc. Land. B. 263:1635-1641 Ferrer, M. and De Le Court. 1992. Sex determination in the Spanish Imperial Eagle./. Field Ornithol. 62:359- 364. Garcelon, D.K., M.S. Martell, P.T. Redig, and L.C. Bud- EN. 1985. Morphometric, karyotipic and laparoscopies techniques for determinig sex in Bald Eagles. /. Wildl Manage. 49:593-599. Griffith, R. and B. Tiwari. 1995. Sex of the last wild Spix’s Macaw. Nature (Land.). 375—454. Lachenbruch, P.A. AND M.R. Mickey. 1968. Estimation of error rates in discriminant analysis. Technometrics 10' 1 - 11 . Newton, I., M. Marquiss, and P. Rothery. 1983. Age structure and survival in a Sparrowhawk population. /. Anim. Ecol. 52:591-602. Norusis, M.J. 1992. SPSS for Windows. Base System User’s Guide. Released 5.0, Chicago, IL U.S.A. Rogers, P.M. and H.M. Myers. 1980. Animal distribution, landscape classification, and wildlife management, Coto de Dohana, Spain./. Appl. Ecol. 17:545-565. Received 20 March 2000; accepted 22 October 2000 J Raptor Res, 35(1);24— 30 © 2001 The Raptor Research Foundation, Inc. THE ANNUAL AND DIEL CYCLES OF GOSHAWK VOCALIZATIONS AT NEST SITES Vincenzo Penteriani Estacion Biologica de Donana, CSIC, Avda. Marta Luisa s/n, Pahellon del Peru, 41013 Sevilla, Spain Abstract. — I attempted to quantify seasonal and daily Northern Goshawk {Accipiter gentilis) vocalizations at nest sites and identify their function. Both duration and number of calls showed significant differences among different periods of the year. The daily distribution of vocalizations differed through the breeding season both as a whole and in individual stages. My results suggested that the kek-kek-kek call may have two meanings as an alarm call and as a call to excite mates. During courtship, vocal activity was most intense in the early morning (female fertile period), but during other stages of the breeding season, vocal activity occurred throughout the day and was related to parental care. My results suggested that vocalizations of goshawks function primarily in territorial defense, intra-pair communication, and pro- tection of paternity of young. Young goshawks showed a rapid increase in the duration of their total daily vocalizations within the first 10 d after fledging. Afterward their vocalization rates decreased until 40 d after fledging. Key Words: Northern Goshawk, Accipiter gentilis; annual vocal activity, daily vocal activity, functions of vocalizations. Los ciclos de vocalizaciones diarias y anuales de Accipiter gentilis en sus sitios de anidacion Resumen. — Intente cuantificar e identificar la funcion de las vocalizaciones estacionales y diarias de Accipiter gentilis en los sitios de anidacion. Tanto la duracion como el numero de vocalizaciones mostra- ron diferencias significativas entre los diferentes periodos del aho. La distribucion diaria de las vocali- zaciones mostraron diferencias significativas entre los diferentes periodos del ano. La distribucion diaria de las vocalizaciones fne diferente a traves de la estacion reproductiva en su totalidad como por etapas individuales. Mis resultados sugieren que la vocalizacion del Kek-kek-kek puede tener dos significados, como sehal de alarma y para atraer a la pareja. Las vocalizaciones fueron mas intensas en la manana durante el cortejo (periodo fertil de la hembra) que durante las otras etapas de la actividad reproductiva, la actividad vocal durante el dia estuvo mas relacionada con el cuidado parental. Mis resultados sugieren que las vocalizaciones de los azores son parte principal de la defensa del territorio, de la comunicacion entre la pareja y de la proteccion y paternidad de los pichones. Los azores jovenes mostraron un rapido incremento en la duracion total de sus vocalizaciones diarias despues de diez dias de haber emplumado, luego su tasa de vocalizaciones disminuyo a partir del dia 40 despues de haber emplumado. [Traduccion de Cesar Marquez] Forest and woodland birds use songs to com- municate in visually-obstructed habitats and con- tact between mates may be maintained acoustically rather than visually (Keast 1985, 1994). Vocaliza- tion patterns and their evolution have been de- scribed and quantified in passerines (Kioodsma and Miller 1996) but, for other groups of birds, the literature is not as extensive. In particular, vocali- zations of raptors have rarely been quantified (Ro- senfield and Bielfeldt 1991). Falconids in the ge- nus Herpetotheres and Micrastur typically vocalize in the morning (Staicer et al. 1996) and their vocali- zations may serve as greeting displays between ma- tes (Cramp and Simmons 1980). In male Cooper’s Hawks {Accipiter cooperii), there is marked vocal ac- tivity at dawn (Stewart et al. 1996) and a specific dawn call is given (Rosenfield and Bielfeldt 1991). Similar behavior has been observed in the Spar- rowhawk, {Accipiter nisus) (Newton 1986). The vocalizations of Northern Goshawks {Accip- iter gentilis) have been described by various authors (e.g., Gromme 1935, Cramp and Simmons 1980, Squires and Reynolds 1997). These descriptions have sometimes been supplemented with remarks concerning associated behavior (Schnell 1958, Pac- teau 1989). However, the vocalizations of adult and young goshawks have never been quantified. In the vocabulary of goshawks, there are two basic types 24 March 2001 Goshawk Vocal Activity 25 of calls (Cramp and Simmons 1980, Squires and Reynolds 1997): a guttural, chattering call and a plaintive, wailing call. The chattering call has been described by Zirrer (1947), Schnell (1958), Cramp and Simmons (1980), and Squires and Reynolds (1997) as a slow kek . . . kek . . . kek call that is used in advertising and pair contact. There is also a fast kek-kek-kek call that is used as a threat and mobbing call, a subdued kai-kai-kai alarm call, and a soft and quick kew-keto-kew chattering call. Adult wailing calls (whee-oo . . . whee-oo) are categorized as food calls. In my analysis of adult goshawk vocalizations, 1 used only these major call types, because other calls are easily overlooked in the field (Penteriani unpubl. data, Squires and Reynolds 1997). The calls of young goshawks have been de- scribed by Schnell (1958) as a whee . . . whee . . . whee food and location call, a distress call consisting of a rapid, high-pitched twitter like chickens make, a contentment call similar to the previous call but consisting of more widely-spaced, single calls, an aggressive chatter ke-ke-ke call used by nestlings when they have food, and a kikk-kikk-kikk used by nestlings to express protest or alarm. The first objective of my study was to quantify seasonal and daily levels of goshawk vocalizations to determine if the frequency of these vocalizations varied seasonally and annually. My second objec- tive was to determine if goshawk calls are used mainly in territorial defense, intra-pair communi- cation (i.e., sexual attraction and sexual conflict when the presence and/or actions of one mate in- terfere with the adaptive interests of the other, Da- vies 1989) , or to protect paternity. If so, I predicted that they would continue throughout the breeding season, albeit at reduced rates, and they should peak during the fertile period of females during courtship and egg laying (Birkhead and Mpller 1992, Catchpole 1973, Bjorklund et al. 1990, Merila and Sorjonen 1994). If vocalizations only function to attract and retain mates, then I predicted that they would occur at higher rates before incubation and cease after the start of egg laying. Because in dense habitats, such as forests, limited visual sig- naling leads to the development of acoustic com- munication, as in the case of forest passerines (Kroodsma and Miller 1996), I predicted that gos- hawks would spend the majority of their time vo- calizing during the year and daily vocalization pat- terns would show similarities with song patterns of woodland passerines. Methods I noted goshawk vocalizations from 1 January 1996-31 December 1996 in a 5000-ha area of beech-forested hills of Cote d’Or (Burgundy region, France). I divided each month into three periods each consisting of 10 d. For each 10-d period and each solar day in the period, I com- puted the number of minutes per period and equally distributed them among eight neighboring goshawk pairs. Consequently, an hourly block was assigned to each pair on a rotation basis, and during each day of each 10-d period, 1 noted the vocalizations of the eight pairs. As significant changes in the breeding cycle might interfere with call data, each site was systematically monitored dur- ing the breeding period. Because each pair successfully reproduced, the annual analysis was carried out for all eight pairs selected at the start of the study. I made observations at each nest from a location where I did not disturb the pair (about 100 m away) and 1 noted calls without changing my position. From these points, I also made observations concerning the behavioral con- text of vocalizations. This position prevented me from hearing some calls of nestlings when they were in their first few weeks of life, vocalizations of adults outside the nest stand, and vocalizations of juvenile goshawks after they left the nesting area. In each 10-d period, I recorded the time of the first and last vocalizations of the day, choosing days with minimum interference from precipi- tation and wind. For each type of call, I recorded the time when the call began, and its duration from a series of single calls (e.g., kek, whee-oo and whee) or call-senes (e.g., kek-kek-kek, kai-kai-kai, kew-kew-kew, and kikk-kikk-kikk). I measured the duration of the vocalizations with a stop- watch, counting the seconds elapsed from the start to the last call given <60 sec from the previous one. I assumed one minute of silence between calls or between call se- quences indicated the end of a vocalization. An isolated single call received an arbitrary value of 1 sec. Sampling units were goshawk pairs for adult vocaliza- tions and nestlings and/or fledglings at each occupied nest for vocalizations of young. I analyzed the call data in relation to month and to the different periods of the annual breeding cycle: nonbreeding (September-Janu- ary), courtship (February-March) , incubation (April- early May), nestling period (early May-late June), and fledgling period (late June-August) . If a single vocaliza- tion implied multiple call numbers, I took the average value. Only nonparametric statistics were used in the analyses (Hollander and Wolfe 1973). Results Adult Vocalizations. Adult goshawk vocalizations showed one major peak during the year that co- incided with the courtship period (Fig.l). The du- ration of the vocal events differed significantly be- tween months {H — 73.11, P < 0.001, Kruskal- Wallis) and between periods {H = 54.86, P < 0.001). The duration of the vocal events increased during the courtship period and decreased during incubation (Table 1). During most of the annual cycle, the first call 26 Penteriani VoL. 35, No. 1 Tdble 1. FcaLuics oi adult and juvenile iNoi thern Gos- hawk vocal activity (N = 8) during the year. The duration (sec) represents the minimum, maximum, and average (±SD) call length of the monthly vocalizations by gos- hawks at nest sites. Month Adult Call Duration ( sec) min-max (x ± SD) Young Gall Duration ( sec) min-max {x ± SD) Jan 1-966 159.7 ± 194.5 Feb 1-749 272.5 ± 265.6 Mar 2-330 240.7 ± 297.2 Apr 2-59 110.1 ± 158.4 May 2-55 99.7 ± 88 Jun 1-112 1-1086 103 ± 130.3 161.5 ± 241.5 Jul 2-352 1-1491 85.8 ± 77.4 310.9 ± 273.6 Aug 4-42 1-560 15.9 ± 10 61.9 ± 66.1 Sept 2-30 12.2 ± 5.5 Oct 1-128 23.7 ± 48.3 Nov 1-136 32.2 ± 39.1 Dec 0-86 17.4 ± 12.9 was uttered at different times of the day, except during courtship, when it was always uttered before sunrise (range = 4-45 min before sunrise). The last call was always uttered prior to sunset, except in April, when it was recorded 22 min afterwards. The last call was less related to dusk than the first call was to dawn. During courtship, the vocalizations reached one major peak both in the first hour before sunrise and three hours after sunrise (Penteriani 1999). During incubation, nestling, and fledgling periods, the daily distribution of the vocalizations was more irregular. There was a significant difference {H = 23.54, P < 0.001) in the duration of vocalizations in various hours of the day. In the nonbreeding period, the diurnal distribution of vocalizations varied and their duration was shorter. An excep- tion was the month of January, when a peak in the same time period as in the courtship period was observed, although the duration of the vocaliza- tions was shorter (Fig. 1). I found a difference in the day-long distribution of vocal events between courtship and incubation periods {D^ — 0.78, P < 0 . 001 ), between courtship and nestling periods (/)„ = 0.86, P < 0 . 001 ), and between courtship and fledgling periods (D^ = 0.79, P < 0.001). The chattering call kek . . . kek . . . kek was the most common call throughout the year (37.6%). A common pattern between the duration of the vocal events and the call number characterizing them was observed for nonbreeding (r^ = 0.6, P < 0.01, Spearman rank), courtship (r^ — 0.4, P < 0.01), nestling (r^ = 0.46, P < 0.05), and fledgling peri- ods (75 — \, P < 0.001), but not for the incubation period (r^ = 0.24, P > 0.05). The chattering call was the second most frequently-used call through- out the year (34.6%) but there was no common pattern between the duration of the vocalizations and the number of calls that characterized them in the nonbreeding (r^ = 0.19, P> 0.05), courtship (r^ = 0.15, P > 0.05), incubation (r^ = 0.29, P > 0.05), nestling (r^ = 0.37, P> 0.05), and fledgling (r^ = 0.08, P > 0.05) periods. A total of 63.2% of chattering calls were preceded and/or followed at 0.05). Vocalizations of Young. The duration of the vo- cal events differed between months {H = 5.76, P < 0.05), but not between nestling and fledgling periods {H ~ 2.16, P > 0.05). The duration of vo- calizations by young goshawks during the nestling and fledgling periods increased rapidly until about the 10 th day after fledging and quickly declined until about the 40th day. No vocalizations were re- corded afterwards (Fig. 2). March 2001 Goshawk Vocal Activity 27 Figure 1. Annual vocalization pattern in adult Northern Goshawks at nest sites; sum of goshawk vocal activity (sec ± SD) per month {N = 8). During these two periods, the time of the first call was variable, whereas the last call was always uttered before sunset. During nestling, the peaks in vocalizations occurred during the 4th and 13th hr after sunrise; conversely, during fledgling, vo- calizations were clustered in the central hours of the day, between the 7th and 10th hr after sunrise. During these periods, prey delivering was related to vocal activity. A common pattern between the duration of the vocal events and number of vocalizations charac- terizing the whee . . . whee . . . whee call (frequency = 95.1%) was observed in both the nestling — 0.71, P< 0.05) and fledgling (r^ = 0.78, P< 0,001) periods. Discussion My study indicated that most adult goshawk vo- calizations occurred during the courtship period. Most vocalizations occurred in late winter and early spring, prior to breeding and corresponded to ini- tial courtship and territory establishment. Various studies of Northern Hemisphere birds have made similar observations (e.g., Kelsey 1989, Logan et al. 1990, Catchpole and Slater 1995). It was notewor- thy that the types of calls recorded from January onwards (about 3 mo before egg laying) were sim- ilar to those recorded during courtship. Starting in January, goshawks begin to become more territo- rial and they utter their alarm calls when people walk close to their nests. They are regularly ob- served displaying over nesting territories and woods (Anonymous 1990, Toyne 1997). Goshawks also intensified their vocalizations at the end of their reproductive cycle when young were dispers- ing from nest areas. The sole period when no vo- calizations were recorded near nests was from late November to the first 20 d of December. A com- parison of the durations and numbers of calls giv- en showed a similar pattern during the year with the period with the longest and most numerous vocalizations coinciding with the courtship period. Overall, the greatest number of daily vocalizations, the longest series of calls, and the most complex individual calls within each series of calls occurred during the courtship period. During the courtship period, vocalizations were clustered from 1 hr before sunrise through the 28 Penteriani VoL. 35, No. 1 Figure 2. Vocalization pattern in juvenile Northern Goshawks during the nestling and fledgling periods (from June to August) . three following hours and the first call was always uttered prior to sunrise. This finding was pertinent to the survey methodologies currently being used to locate goshawk nests (Penteriani 1999). This time period also corresponded to the daily copu- lation pattern of adults with a peak occurring around the time of egg laying early in the morning and when male goshawks spend most of their time near females (M 0 ller 1987) . The second peak in goshawk vocalizations dur- ing the breeding cycle might be ascribed to the intense activity of goshawks near their nests during the nestling period. The third and less intense peak during fall might be dependent on levels of gonadal hormones which are responsible for the activation of song behavior in most adult birds (Nottebohm et al. 1978, Brenowitz and Kroodsma 1996). Elevated levels of hormones might also be a nonfunctional cause of the singing peak at dawn (Staicer et al. 1996). The high frequency of the kek-kek-kek alarm call during the courtship period indicated that the meaning so far attributed to this call (e.g., Kimmel and Yahner 1990, Speiser and Bosakowski 1991, Kennedy and Stahlecker 1993) should not be con- fined to threat and mobbing, at least near the nest. The fact that up to 60% of these calls were always preceded and/or followed by pair contact, food, and greeting calls strongly supported this function. Nevertheless, I attributed a twofold meaning to this call related to alarm and high excitement during mate contacts. In most cases, this chattering call followed a series of kek . . . kek . . . kek or pair-con- tact calls, especially during territory establishment and until egg laying. Therefore, it was not neces- sarily related to stress or defense but also to breed- ing period. In a species, the same call may be ut- tered in different contexts, including conspecific territorial fights, defense against predators, and in- tra-pair interactions without the presence of any predator or territorial competitor (Logan 1994). It should also be emphasized that, during the repro- ductive season, aggressive calls may also express sexual conflicts (Davies 1989). I considered the whee-oo . . . whee-oo . . . whee-oo call to be an exclusive female food call. During the incubation period, it was given by females when food deliveries were made by males (Schnell 1958) March 2001 Goshawk Vocal Activity 29 and it denoted a very important stage in the breed- ing activity of the pair. In fact, during incubation and nestling (Schnell 1958, Penteriani unpubl. data) , females utter this call as soon as they see or hear males in the nest area (recognition scream), upon prey delivery (transfer scream), and if males remain in the nest area after food delivery (dis- missal scream). During the fledgling period, this call might be part of the communication between females and young when females continue to feed young after fledging (Schnell 1958). The vocalizations of young at nests increased from hatching until the first 10 d after they fledged. After that, vocalizations diminished and ended about 40 d after the young first left nests or approximately 80 d after hatching when dispersal from nesting areas typically occurs (Kenward et al. 1993a). This decrease in vocalizations after fledg- ing may have been due to the fact that, before dis- persing, juveniles can fly as far as 1 km from nests (Kenward et al. 1993b). Since the whee . . . whee . . . whee call of young goshawks is mainly interpreted as a food call, the day-long pattern I observed sug- gested that adults continued to make food deliv- eries to nests. Although food may be supplied to nestlings at all times of the day (Schnell 1958), peaks in vocalizations by young in my study, as well as the changes observed between the nestling and fledgling periods, fit the peaks reported by Schnell (1958) for food deliveries by male goshawks (74- 85% of total food deliveries). The day-long and annual patterns of goshawk vo- calizations suggested that the vocal activity in gos- hawks functions primarily in territorial defense and intra-pair communication. This finding was supported by the fact that vocalizations continued throughout the breeding season and the year, al- beit at different rates. Moreover, as vocalizations showed a peak during the fertile period of females, it also seemed that calls may be used to protect paternity. Acknowledgments My thanks go to F. Liberatori and M. Melletti for their assistance in the field. C. Carere, R.N. Rosenfield, S. Sar- aceni, D. Spector, and K. Titus provided useful comments on the first drafts of this manuscript. The Astronomic Society of Burgundy gave me data about sunrise/ sunset and the Atomic Research Centre of Valduc gave me the authorizations to work in its restricted area. Literature Cited Anonymous. 1990. Breeding biology of goshawks in low- land Britain. Brit. Birds 83:527-540. Birkhead, T. and A.R Moller. 1992. Sperm competition in birds. Evolutionary causes and consequences. Aca- demic Press, Cambridge, U.K. Bjorklund, M., B. Westman, and K. Allander. 1990 Song in Swedish Great Tits: intra- or intersexual com- munication? Behaviour 111:257-269. Brenowitz, E.A. AND D.E. Kroodsma. 1996. The neu- roethology of birdsong. Pages 285-304 in D.E. Kroodsma and E.H. Miller [Eds.], Ecology and evo- lution of acoustic communication in birds. Cornell Univ. Press, Ithaca, NY U.S.A. Catchpole, C.K. 1973. The functions of the advertising song in the Sedge Warbler {Acrocephalus schoenoha- enus) and the Reed Warbler {A. scirpaceus). Behaviour 46:300-320. AND P.J.B. Slater. 1995. Bird song: biological themes and variations. Cambridge Univ. Press, Cam- bridge, U.K. 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. Davies, N.B. 1989. Sexual conflict and the polygamy threshold. Anim. Behav. 38:226-234. Gromme, O.J. 1935. The goshawk (Astur atricapillus atn- capillus) nesting in Wisconsin. Auk 52:15-20. Hollander, M. and D.A. Wolfe. 1973. Nonparametnc statistical methods. Wiley, New York, NY U.S.A. Keast, a. 1985. Springtime song, periodicity and se- quencing, a comparison of a southern forest and northern woodland bird community. Pages 119-128 in A. Keast, H.F. Recher, H. Ford, and D. Saunders [Eds.], Birds of eucalyptus forests and woodlands- ecology, conservation, management. RAOU and Sur- rey Beatty and Sons, Sydney, Australia. . 1994. The annual cycle in a vocalization context- a comparison of the Eastern Yellow Robin Eopsaltna australis and Jacky Winter Microeca leucophaea. Emu 94: 230-238. Kelsey, M.B. 1989. A comparison of the song and terri- torial behavior of a long-distance migrant, the Marsh Warbler Acrocephalus palustris, in summer and winter Ibis 131:403-414. Kennedy, P.L. and D.W. Stahlecker. 1993. Responsive- ness of nesting Northern Goshawks to taped broad- casts of three conspecific calls. /. Wild! Manage. 57. 249-257. Kenward, R.E., V. Marcstrom, and M. Karlbom. 1993a. Post-nestling behaviour in goshawk, Accipiter gentilis: I The causes of dispersal. Anim. Behav. 46:365-370. , , AND . 1993b. Post-nesding behav- iour in goshawk, Accipiter gentilis: II. Sex differences m sociality and nest-switching. Anim. Behav. 46:371-378. Kimmel, J.T. and R.H. Yahner. 1990. Response of North- ern Goshawks to taped conspecific and Great Horned Owl calls./. Raptor Res. 24:107-112. Kroodsma, D.E. and E.H. Miller. 1996. Ecology and evo- 30 Penteriani VoL. 35, No. 1 lution of acoustic communication in birds. Cornell Univ. Press, Ithaca, NYU.S.A. Logan, C. 1994. Fluctuations in intra-pair calling across breeding phases in Northern Mockingbirds {Mimus polyglottos) . Behaviour 130:123-141. , L.L. Hyatt, and L, Gregorcyk. 1990. Song play- back initiates nest building during clutch overlap in Mockingbirds Mimus polyglottus. Anim. Behav. 39:943- 953. Merila, J. AND J. SOR[ONEN. 1994. Seasonal and diurnal patterns of singing and song-flight activity in Blue- throats (Luscinia svecica). Auk 111:556-562. M0LLER, A.P. 1987. Copulation behaviour in the goshawk Accipiter gentilis. Anim. Behav. 35:755-763. Newton, I. 1986. The Sparrowhawk. T. & A.D. Poyser, Staffordshire, U.K. Nottebohm, F., M.E. Nottebohm, L.A. Crane, and J.C. WiNGEiELD. 1978. Seasonal changes in gonadal hor- mone levels of adult male canaries and their relation to song. Behav. Neural Biol. 47:197-211. Pacteau, C. 1989. L’autour et I’epervier, du comporte- ment au sujet. Editions Hecate, Luyon, France. Penteriani, V. 1999. Dawn and morning goshawk court- ship vocalizations as a method for detecting nest sites. J. Wildl. Manage. 63:511-516. Rosenfield, R.N. andJ. Bielfeldt. 1991. Vocalizations of Cooper’s Hawks during the pre-incubation stage. Con- dor 93:659-665. Schnell, J.H. 1958. Nesting behavior and food habits of goshawks in the Sierra Nevada of California. Condor 60:377-403. Speiser, R. and T. Bosakowski. 1991. Nesting phenology, site fidelity, and defense behavior of Northern Gos- hawk in New York and New Jersey. J. Raptor Res. 25. 132-135. Squires, J.R. and R.T. Reynolds. 1997. Northern Gos- hawk {Accipiter gentilis) . In A. Poole and F. Gill [Eds J, The birds of North America, No. 298. The Academy of Natural Sciences, Philadelphia, PA, and The Amer- ican Ornithologists’ Union, Washington, DC U.S.A. Staicer, C.A., D.A. Spector, and A.G. Horn. 1996. The dawn chorus and other diel patterns in acoustic sig- naling. Pages 426-453 in D.E. Kroodsma and E.H Miller [Eds.], Ecology and evolution of acoustic com- munication in birds. Cornell Univ. Press, Ithaca, NY U.S.A. Stewart, A.C., R.W. Campbell, and S. Dickin. 1996. Use of dawn vocalizations for detecting breeding Cooper’s Hawks in an urban environment. Wildl. Soc. Bull. 24 291-293. Toyne, E.P. 1997. Nesting chronology of Northern Gos- hawks {Accipiter gentilis) in Wales: implications for for- est management. Forestry 70:121-127. Zirrer, E. 1947. The goshawk. Passenger Pigeon 9:79-94: Received 29 December 1999; accepted 29 October 2000 J Raptor Res, 35(1) :31-34 © 2001 The Raptor Research Foundation, Inc. BREEDING RATES OF EURASIAN KESTRELS (FALCO TINNUNCULUS) IN RELATION TO SURROUNDING HABITAT IN SOUTHWEST SPAIN Jesus M. Aviles, Juan M. SAnchez, and Deseada Parejo Conservation Research Group, Department of Zoology, University of Extremadura, Badajoz E-06071, Spain Abstract. — ^We studied breeding success of Eurasian Kestrels (Falco tinnunculus) in nest boxes in seven different habitat types in the southwest of Spain. A total of 567 nest boxes was installed on power pylons in fallow helds, cereal cropland, holm oak land, olive orchards, pastureland, irrigated cropland, and shrubland. Occupation of boxes did not vary among the habitats and there were no significant differ- ences among the seven habitat types in laying date, clutch size, or breeding success. When habitats with low numbers of breeding pairs were removed from analyses, we were able to detect significant differ- ences in mean laying dates, clutch sizes, and breeding success rates among the three habitat types with the highest sample sizes. Kestrels nesting in pastureland showed higher clutch sizes and higher breeding success that those nesting in the cerealland. A seasonal decline in clutch size was found in all three habitat types with the highest sample sizes. Our results suggested that habitat features influence the breeding biology of Eurasian Kestrels. Key Words; Eurasian Kestrel, Ealco tinunculus; habitat features', breeding success', agricultural intensification', Spain. Tasas de reproduccion de Falco tinnunculus en relacion al habitat circundante en el suroeste de Espaha Resumen. — Se ha estudiado la influencia del habitat de nidificacion sobre la biologia reproductora del Cernicalo Vulgar Falco tinunculus en una poblacion reproductora del sudoeste de Espana. Se instalaron en postes de Imeas de conduccion electrica 567 c^as-nido dentro de siete tipos diferentes de habitats: Barbechos, siembras de cereal, dehesas arboladas de encinas, olivares, pastizales, cultivos de regadio y areas con cobertura de matorral. No existieron diferencias significativas en los porcentajes de ocupacion de los nidales entre los siete habitats. No se detectaron diferencias entre los siete habitats en la fechas medias de puesta, tamanos de puesta y tasas reproductoras de los cernicalos. Sin embargo, cuando se extr^eron de los analisis aquellos habitats con menor numero de parejas nidificantes, existieron difer- encias entre habitats en el inicio de la reproduccion. Del mismo modo, el tamano de puesta y el exito reproductor variaron entre los tres habitats con mayores tamanos muestrales. Los cernicalos que nidi- ficaron en pastizales tuvieron mayores tamanos de puesta y mayor exito reproductor que los que lo hicieron en cultivos de cereal. En los tres habitats con mayor tamano muestral se detecto un descenso estacional del tamano de puesta que no vario entre los habitats. Los resultados sugieren la influencia de los rasgos del habitats sobre la biologia de reproduccion de la especie en nuestra zona de estudio. [Traduccion de Autores] European Kestrel {Falco tinnunculus) popula- tions are declining in the Palearctic because of the intensification of agriculture (Village 1990, Shrubb 1993). In Spain, the breeding population has remained stable since the 1970s at 25 000— 30 000 pairs (Aparicio 1997). Although this pop- ulation has been used as a tool in experimental studies (Aparicio 1994a, 1994b, 1998), there is little information on its basic biology. Some stud- ies have reported breeding rates (Aparicio 1994a, Gil-Delgado et al. 1995, Aviles et al. 2000) , but almost nothing is known about its breeding biology in different habitats in the Iberian Pen- insula. Nest boxes are readily accepted by kestrels (Vil- lage 1990). Although care should be taken when testing hypotheses related to fitness in such artifi- cial nesting situations (M 0 ller 1989), nest boxes of- fer an exceptional opportunity to conduct breed- ing studies in cavity-nesting species of birds (Glutton-Brock 1988). The aim of this study was to test the effect of habitat type on the breeding per- formance of Eurasian Kestrels in the Serena region in the southwest of Spain. 31 32 Aviles et al. VoL. 35, No. 1 VILLANUEVA 1C FL I ORELLANA CAMPANARIO PA CE HL ^ ■ CASTUERA CABEZADELBUEY I 1 1 1 05 10 15 km Figure 1. Map of the location of the study area and the length of electric power lines (lines) in each habitat type. Squares represent the major towns in the Serena region. Codes for habitat types are: fallowlands (FL) , cereal crop- lands (CE), holm oaklands (FIL), olive orchards (OL), pasturelands (PA), irrigated croplands (1C), and shrub- lands (SL). Study Area and Methods The study area was located in the Serena region of Spain (39°03'N, 5°14'W). The climate of the area is Med- iterranean and mean temperature and rainfall during May and June is 17.7°C and 11.6 mm, respectively (Aviles et al. 2000). In February and March of 1989, 567 nest boxes were installed on all the power pylons that crossed patches of seven different habitats in the study area: fal- lowlands {N = 26 nest boxes), cereal croplands (oats, wheat, and barley, A = 159 nest boxes), holm oaklands ( Quercus rotundifolia) (N = 63 nest boxes) , olive orchards (N = 14 nest boxes), pasturelands {N = 237 nest boxes), irrigated croplands (rice and maize, A= 18 nest boxes), and shrublands (mainly Retama sphaerocarpa, A = 50 nest boxes) (Fig. 1). Habitat patches with possible natural cav- ities (holm Oakland and olive orchards) and farmhouses with possible nesting sites were searched for pairs of breeding kestrels but we did not find any kestrels breed- ing in natural cavities. We considered a patch of habitat to accurately represent kestrel breeding parameters when dll the boxes in it were surrounded at least by 1 km of this same habitat type. The minimum distance between two patches was 1.5 km between holm oaklands and shrublands (Fig. 1). Although we did not make observa- tions of hunting activities of the kestrels, we considered 1 5 km to be a reasonable estimation of the hunting ter- ritories taking into account the fact that breeding kestrels forage at a maximum distance of 2 km from their nests m Spain (Veiga 1982). We did not expect any density- dependent effects on kestrel breeding performance be- cause the mean (±SD) density of nest boxes was 9.43 ± lU o < UJ o cc LU 0. Figure 2. Percent distribution of nest-boxes used by breeding Eurasian Kestrels in relation to habitat type. Codes for habitat types are: pasturelands (PA), cereal croplands (CE), shrublands (SL), holm oaklands (HL), fallowlands (FL), irrigated croplands (IC), and olive or- chards (OL). Habitats are ordered in relation to their nest box use. 0.26 boxes/km of power line independent of habitat type, and the percent of boxes occupied by kestrels in all seven habitats was <50% (Fig. 2) . All the boxes were monitored weekly from the first stages of breeding in 1989. We assumed that nest-box availability in each habitat type did not affect breeding of kestrels because the interhabitat distribution of nest boxes and breeding pairs of kestrels did not vary ( G-test Gs = 5.96, P = 0.42), Eurasian Rollers {Coracias garrulus) that also use nest boxes began egg laying in our study area at least 1 mo later than kestrels (Aviles et al. 1999), and there was no evidence of damage to kestrel eggs or young by rollers when kestrels were the first to breed m nest boxes (Cramp and Simmons 1980, Aviles pers. obs ) In boxes occupied by kestrels, nest visits were increased to one visit every 3-4 d during the nesting period to de- termine breeding success accurately. Laying dates were determined by subtracting the incubation period from the hatching date (Cramp and Simmons 1980). When determining hatching dates, we took into account the fact that the laying interval in the species is two days (Cramp and Simmons 1980). We measured percent hatching success as the percent of eggs in each clutch that hatched, the number of fledglings per successful nest with successful nests being those at which at least one young fledged, and breeding success as the number of fledglings per pair that laid at least one egg. We checked for interhabitat differences in the percent of boxes used vising a contingency table with a Chi-square test. Normality of the variables was checked with Kol- mogorov-Smirnov tests. Any nonnormality in differences in laying dates, clutch size, and breeding rates among habitat types was checked with Kruskal-Wallis tests. Dif- ferences between pairs of habitats were tested using non- parametric Tukey-type multiple comparisons. Seasonal declines in clutch size were analyzed using two-tailed Spearman correlations. We checked for interhabitat dif- ferences between clutch sizes and laying dates based on March 2001 Breeding Rates of Eurasian Kestrels 33 the comparison of correlation coefficients from indepen- dent samples (Zar 1996). Results The percent of nest boxes used by nesting kes- trels ranged from 47.1% in pasturelands to 2.17% in olive orchards, but it did not vary significantly among the seven habitats in our study (x^e ~ 7.55, P = 0.27) (Fig. 2). Likewise, no significant differ- ences between the seven habitat types were detect- ed in mean laying dates {Hq = 11.4, P = 0.08, N = 88), clutch sizes (PTg = 8.8, P = 0.83, AT = 136), hatching success — 3.9, P = 0.67, N = 123), breeding success (ifg — 8.8, P — 0.82, N = 125), or fledgling success {H^ = 7.65, P = 0.26, N — 115) (Table 1). Because our results were probably influenced by the low number of breeding pairs in fallowlands, olive orchards, holm oaklands, and irrigated crop- lands, we tested for differences in breeding param- eters in the three habitat types with the largest sam- ple sizes: cereal croplands, pasturelands, and shrublands. We found significant differences in mean laying dates (//g ~ 9.47, P = 0.008, N = 73), clutch sizes {H^ = 6.17, P = 0.04, N = 114), and breeding success rates {H 2 — 7.01, P = 0.02, N = 108), but not in hatching success {H^ — 0.91, P = 0.63, N = 106) and fledglings per successful nest (PTg = 4.14, P = 0.12, N = 100) (Table 1). Pairs nesting in boxes in pasturelands showed a higher clutch size and a higher breeding success than those nesting in cereal croplands (P < 0.05 in both cases. Table 1 ) . However, there were no significant differences between clutch sizes and breeding val- ues of kestrels nesting in shrublands and the other two habitats types (P > 0.05 in all cases. Table 1). Considering only the three habitats with the greatest numbers of breeding pairs, clutch size of kestrels decreased seasonally in cereal croplands (r^ = —0.65, P < 0.05, N = 19), and marginally in pasturelands (r^ = —0.30, P < 0.06, N — 39) and shrublands {r^ = —0.55, P < 0.06, N — 13); how- ever, correlation coefficients did not vary signifi- cantly among the habitat types (P > 0 .05 in all the cases) . Discussion Our results suggested that habitat features can influence the breeding biology of Eurasian Kes- trels. Previous studies in northern latitudes have reported effects of habitat on food preferences (Pettifor 1984) and breeding rates (Valkama et al. CO cj X u rS 1) cr urj 00 + 1 (M CO 05 05 05 0 CM d d + 1 CD I-H I> i-H + 1 00 I-H +1 no + 1 05 05 CM CO CO 00 00 00 'Hh !/) Q Z p, O Qi u 00 QQ oq »o j> 1-H 00 + 1 ^ +1 ^ +1 0+10+10 CO O UTJ CO J> CO 00 J> CO lO • ^ JD d C O •o H CO CO + 1 o no CO GO CO ^H cq cq 1-H CM ^H l-H + 1 to + 1 lO +1 O) 40 + 1 lO 00 05 " — ^ CM no CO 00 V Ui -1-1 fed a CS u w o 2^1 CO O CO cq 1 — I o o o ^ + 1 cr +1 CO +1 CO +1 GO +1 CO no .!>• O l> U-' o ^ ^ CO o Q C/D + 1 'H 00 Sm 0 '•w (J O Ch Oh o ~a a cS CM cq CD r- 05 CO CM 1-H 00 I-H I-H + 1 0 + 1 (6) + 1 QO + 1 00 +1 cq 0 CM 05 d 1-H 00 CO C] nH i-H + 1 I> 00 05 05 I> CO 05 + 1 ^ +1 CO CO ^ CO CO r- O) CM 10 CO OJ -l-l T3 be 01 t be be ; J> CO J> CO ■ a; tb T3 be J V N P U 00 b^) (/) bo 5/5 C ?! G ^ u K U rt OJ H oC H 4'^*^ year birds) (sensu Grimmett et al. 1999). Species diversity was highest on the three days (i.e., 30 and 31 October, and 3 November) when most individuals were counted. On these three days, migrant species totaled 6, 6, and 4, re- spectively (Table 1). Although a few raptors migrated east-to-west south of the site, most migrants passed east-to-west directly overhead, or within 750 m north of the March 2001 Raptor Migration in Nepal 37 Table 1. Daily counts of migrants at the Khare raptor-migration watchsite, central Nepal, 27 October-3 November 1999. October November All Species 27 28 29 30 31 1 2 3 4 Days Western Marsh Harrier (Circus aeruginosus) 0 0 0 0 0 0 0 1 0 1 Harriers (Cirrw5spp.) 0 0 0 0 1 0 0 1 1 3 Eurasian Sparrowhawk (Accipiter nisus) 0 0 0 1 1 1 0 0 0 3 Eurasian Buzzard (Buteo buteo) 0 1 0 2 1 0 0 0 0 4 Buzzard (Buteo spp.) 0 0 1 0 0 0 0 0 0 1 Steppe Eagle (Aquila nipalensis) 0 1 14 176 231 70 64 238 27 821 Booted Eagle (Hieraaetus pennatus) 0 0 0 0 0 1 1 0 1 3 Lesser Kestrel (Falco naumanni) 0 0 0 5 0 0 1 4 0 10 Eastern Red-footed Ealcon (F. amurensis) 0 1 4 34 0 4 0 0 0 43 Northern Hobby (F subbuteo) 0 0 3 9 6 0 0 1 0 19 Small falcons (Falco spp.) 0 0 0 30 10 0 0 0 0 40 Large falcons (Falco spp.) 0 0 0 0 1 0 0 0 0 1 All species 0 3 22 257 252 76 66 245 29 950 Hours of observation 5.33 6 5 6.5 7 6.5 5.75 5.83 4.25 54.16 site. Those passing overhead migrated at 25-75 m above the surrounding landscape. Most eagles were first detected soaring in thermals above the Yamdi Khola valley north of Naudanda and east of the site, and along a lower ridge between Naudan- da and the watchsite. Eagles seen thermaling and Eagles per hour Time of day Figure 1. Passage rates of Steppe Eagles {Aquila nipalen- sis) by time of day at the Khare raptor migration watch- site in central Nepal, 27 October-4 November 1999. riding on updrafts in the distance typically passed the watchsite within 5-7 min. Discussion Regional cloud cover appeared to have a pro- found affect on the numbers of eagles counted at the site. Two of the three highest daily counts for eagles, 30 October and 3 November, occurred on days when clouds began to build over the Anna- purna range earlier than on other days, and when the range’s foothills were covered in clouds by 0930 H. It appeared that heavy cloud cover forced eagles to shift their flight path south of the range and toward the hills and valleys near the watchsite (Fleming 1983, de Roder 1989). Indeed, our high- est hourly count of eagles (149 birds) occurred at 1100-1200 H on 31 October, the same hour in which clouds completely obscured both the massif and its foothills that day. Neither wind speed nor wind direction, both of which remained relatively constant at <20 kph from the south, varied sufficiendy during our ob- 38 DeCandido et al. VoL. 35, No. 1 servations to allow us to determine their effect, if any, on raptor migration at the site. Our observations of the diurnal periodicity of the flight, which indicate a decided peak at 1100- 1200 H (Fig. 1), followed by a second peak in late afternoon, are similar to those of de Roder (1989), who also noted that the greatest movements of ea- gles occurred between midday and late afternoon. Possible explanations for the diurnal periodicity in- clude the location of appropriate roosting areas several hours flying-time to the east, and the diur- nal shifts in soaring conditions at the site, both of which merit additional investigation. A large-scale, east-to-west, autumn movement of Aquila eagles along the southern foothills of the Himalayan Mountains was first observed by R.L. Fleming, Jr. in 1983 (Fleming 1983). The 821 ea- gles we counted over 9 d in late October-early No- vember 1999, together with the 1100 seen by Bijls- ma (1991) in approximately the same area over the same 9 d in 1984, and the 4907 seen by de Roder (1989) at Khare during the same 9 d in 1985, large- ly confirm Fleming’s initial description of the flight. The route, which apparently results when south-bound migrants from central and east Asia detour around the eastern flank of the Tibetan Pla- teau, enables the birds to soar on updrafts and thermals along the southern foothills of the Hi- malayan Mountains of Nepal, and presumably, northern India (Fleming 1983, Bijlsma 1991). Al- though the ultimate destination of the birds is thought to be southern Asia, including India and possibly Arabia, some of the birds may reach Africa (Welch and Welch 1991). Still unclear, however, is the seasonal magnitude of the flight, which to date has been observed for only relatively brief periods of time in any one au- tumn (i.e., 25 d in 1984, 18 d in 1985, and 9 d in 1999) . Fleming (1983) estimated a seasonal pas- sage of at least 45 000 eagles, while de Roder (1989) estimated the passage at between 10 000— 20 000 individuals. Our observations lead us to sug- gest that while >10 000 eagles probably do pass the site in most autumns (see Zalles and Bildstein 2000) , additional season-long observations (i.e., late September-early December) are needed to provide more accurate estimates of the magnitude of the flight. In addition to Steppe Eagles, we identified 18 additional species of raptors at the site, at least eight of which certainly included migratory indi- viduals (Table 1). Fleming (1983) initially reported Table 2. Raptors reported as migrants at or near the Khare raptor-migration watchsite in central Nepal. Species Source"* Eastern Honey Buzzard (Pernis ptilorhyn- T 3 chus)^ Black Kite {Milvus migrans) 1, 2, 3, this White-tailed Sea Eagle {Haliaeetus albi- study" 2, 3 cilla)^ Black Vulture {Aegypius monachus)^ 2, 3 Egyptian Vulture {Neophron percnopterus) 1, 2, 3 Short-toed Eagle {Circaetus gallicus)^ 2, 3 Western Marsh Harrier {Circus aerugi- This study nosus) ^ Hen Harrier (C. cyaneus) L 2, 3 Pallid Harrier {C. macrourus)^ 1, 3 Montagu’s Harrier (C. pygargus)^ 1, 3 Besra {Accipiter virgatus) 1, 3 Shikra (A. badius)^ 1, 3 Eurasian Sparrowhawk (A. nisus)^ 1, 2, 3, this Eurasian Buzzard {Buteo buteo) study 1, 2, 3, this Long-legged Buzzard {B. rufinus) study 1, 2, 3, this Steppe Eagle {Aquila nipalensis) study 1, 2, 3, this Imperial Eagle (A. heliaca) study 1, 2, 3 Booted Eagle {Hieraaetus pennatus) 2, this study Lesser Kestrel {Falco naumanni) 1, 2, 3, this Eurasian Kestrel {F. tinnunculus) study 2 Eastern Red-footed Falcon {F amuren- 1, 2, 3, this sis) study Northern Hobby {F. subbuteo) 2, this study Saker Falcon {F. cherrug)^ 1, 3 Peregrine Falcon {F. peregrinus)^^ 1, 3 Barbary Falcon {F. pelegrinoides)^ L 3 ® 1 — de Roder (1989), 2 — Bijlsma (1991), 3 — Zalles and Bildstein ( 2000 ). ^ Irregular or uncommon migrant. Seen in 1999, but not separated numerically from local resi- dents. that at least five species used the corridor, de Rod- er (1989) reported 18 species, and Bijlsma (1991) 16 species. Zalles and Bildstein (2000) summariz- ing these earlier efforts, suggested that 21 species regularly migrate at the site (Table 2) . Recommendations Our observations, together with those of Flem- ing (1983), de Roder (1989), and Bijlsma (1991), confirm a significant east-west movement of Steppe March 2001 Raptor Migration in Nepal 39 Eagles and smaller numbers of as many as 20 other species of raptors through the region. One of the species migrating at the site, the Lesser Kestrel, is a globally vulnerable raptor whose western Euro- pean populations recently have declined precipi- tously and whose central Asia populations are little studied (Collar et al. 1994) . It may merit additional monitoring at the site (Zalles and Bildstein 2000). The fairly-consistent nature of the flight (local weather conditions notwithstanding) suggests that the Khare watchsite has potential for monitoring regional populations of central and east Asia mi- gratory raptors, and for serving as an environmen- tal education center for local inhabitants and as a source of ecotourism revenue. Indeed, during our stay at Khare we were visited by a teacher from a local school, along with several of his students, all of whom were both curious and enthusiastic about our activities. Although local villagers were able to distinguish vultures from other raptors, none was aware of the substantial migration of Steppe Eagles and other birds of prey. Nepalese schools are in recess from mid-October through mid-November, presenting an ideal opportunity to train and use students to monitor the migration on a long-term basis. Ackn o wledgments We sincerely appreciate the encouragement and thoughtful advice of P. Kerlinger, as well as the comments and suggestions of A. Fish, J, Smith, and R. Yosef on an earlier version of the manuscript. M. Gauchan and her family provided a home-style atmosphere at the Gauchan Lodge during our stay in Naudanda, Nepal. We also gratefully acknowledge the many kind gestures of friend- ship shown to us by local teacher, S. Gurung, and his son, Sumit, in Khare. Those wishing to visit and stay at the site can contact Mr. Gurung at Balmandir English School, LARC, P.O. Box No, 1 Bokhara, Gandaki Zone, Nepal. This is Hawk Mountain Sanctuary contribution number 45. Literature Cited Ali, S. and S.D. Ripley. 1978. Handbook of the birds of India and Pakistan. Vol. 1. 2nd Ed. Oxford Univ. Press, Bombay, India. Bijlsma, R.G. 1991. Migration of raptors and Demoiselle Cranes over central Nepal. Birds Prey Bull. 4:73-80. Collar, N.J., M.J. Crosby, and A.J. Stattersfield. 1994 Birds to watch 2: the world list of threatened birds. BirdLife Conservation Series No. 4. Birdlife Interna- tional, Cambridge, U.K. DE Roder, F.E. 1989. The migration of raptors south of Annapurna, Nepal, autumn 1985. Forktail 4:9-1^ . Fleming, R.L., Jr. 1983. An east-west Aquila eagle migra- tion in the Himalayas. J. Bombay Nat. Hist. Soc. 80:58- 62. Forsman, D. 1999. The raptors of Europe and the Middle East: a handbook of field identification. T. & A.D. Poy- ser, Calton, U.K. Grimmett, R., C. Inskipp, and T. Inskipp. 1999. A guide to the birds of India, Pakistan, Nepal, Bangladesh, Bhutan, Sri Lanka, and the Maldives. Princeton Univ. Press, Princeton, NJ U.S.A. Inskipp, C. and T. Inskipp. 1991. A guide to the birds of Nepal, 2nd Ed. Christopher Helm, London, U.K. Porter, R.F., L. Willis, S. Christensen, and B.P. Niel- sen. 1986. Flight identification guide of European raptors, 3rd Ed. T. & A.D. Poyser, Calton, U.K. Welch, G. and H. Welch. 1991. The autumn migration of the Steppe Eagle Aquila nipalensis. Sandgrouse 13 24-33 Zalles, J.I. and K.L. Bildstein. 2000. Raptor watch: a global directory of raptor migration sites. BirdLife In- ternational, Cambridge, U.K.; and Hawk Mountain Sanctuary, Kempton, PA U.S.A. Received 3 March 2000; accepted 26 October 2000 J Raptor Res. 35(1) : 40-48 © 2001 The Raptor Research Foundation, Inc. DENSITY, PRODUCTIVITY, DIET, AND HUMAN PERSECUTION OF GOLDEN EAGLES (AQUILA CHRYSAETOS) IN THE CENTRAL-EASTERN ITALIAN ALPS Paolo Pedrini Raptor Conservation Research Unit, Museo Tridentino di Scienze Naturali, via Calepina 14, 38100 Trento, Italy Fabrizio Sergio 1 Edward Grey Institute of Field Ornithology, Department of Zoology, South Parks Road, Oxford 0X1 3PS, U.K. Abstract. — ^A Golden Eagle {Aquila chrysaetos) population of 46 pairs was regularly censused between 1982-92 in a 7800-km^ study plot in the central-eastern Italian Alps. Density was stable at 5.9 territorial pairs per 1000 km^. Mean nearest-neighbor distance was 8.7 km {N = 46), and nest areas were regularly dispersed. Sixteen percent of 70 pairs consisted of an adult and a nonadult individual. Mean laying date was 23 March {N = 27). The percentage of successful territorial pairs was 55% {N = 109). Mean number of fledged young was 0.61 per territorial pair {N = 109) and 1.10 per successful pair {N = 56). Diet was dominated by mammals (64%) belonging to the orders Artiodactyla, Roden tia, Lagomorpha, and Carnivora, and by birds (32%) belonging to the order Galliformes {N = 247 prey items). Productivity was affected by age of territory holders and the extent of woodland or grassland within the potential foraging range. Illegal shooting accounted for the deaths of 15 individuals between 1980-89. Compared to other alpine populations, the study population showed a low density, average nearest-neighbor dis- tance and productivity, and a typical frequency of nonadult territory holders. We suggest that the future long-term population trends of alpine Golden Eagles will be determined by the interactions among increasing food supply, declining availability of foraging habitat, decreasing human persecution, and increasing human disturbance. Key Words: Aquila chrysaetos; Golden Eagle; breeding success; density; diet, Italian Alps; land use changes. Densidad, productividad, dieta y persecucion humana de aguilas doradas {Aquila chrysaetos) en los alpes centro orientales Resumen. — Una poblacion de aguilas doradas {Aquila chrysaetos) de 46 pares fue censada regularmente entre 1982-92, en una parcela de 7800 km en el centro — oriente de Los Alpes. La densidad fue estable en 5.9 pares territoriales por 1000 km. La distancia media al vecino mas proximo fue de 8.7 km {N = 46), las distancias de los nidos fueron regularmente dispersadas. Diez y seis por ciento de las 70 parejas consistieron de un adulto y un individuo subadulto. La media de la fecha de postura fue el 23 de marzo {N = 27). El porcentaje de parejas exitosas fue del 55% (A^= 109). La media de pichones emplumados fue de 0.61 por pareja territorial {N = 109) y 1.10 por pareja exitosa {N = 56). La dieta estuvo repre- sentada por mamiferos en un 64% de los ordenes Artiodactila, Rodentia, Lagomorfa y Carnivora y por aves en un 32% pertenecientes al orden Galliformes {N = 247 presas). La productividad fue afectada por la distancia del vecino mas proximo, edad de los poseedores del territorio y por el tamaho de los bosques y pasturas dentro del area de forrajeo. La caza ilegal fue la causa de la muerte de 15 individuos entre 1980-89. Comparado con otras poblaciones alpinas, la poblacion estudiada mostro una baja de densidad, un promedio de distancia entre el vecino mas proximo, y ima frecuencia tipica de los posee- dores del territorio no adultos. Sugerimos que la tendencia poblacional a largo plazo de las aguilas doradas alpinas puede estar determinada por las interacciones entre el incremento de alimento, la declinacion de la disponibilidad del habitat de forrajeo, la disminucion de la persecucion humana y el incremento de la perturbacion humana. [Traduccion de Cesar Marquez] ^ Present address: Raptor Conservation Research Unit, Trento Natural History Museum, via Calepina 14, 38100 Tren- to, Italy. 40 March 2001 Golden Eagle Breeding Ecology 41 In Europe, populations of Golden Eagles {Aquila chrysaetos) are generally stable or increasing, but declines have been reported for Spain, Portugal, Albania, Romania, Greece, Belarus, and Ukraine (Watson 1994, Haller and Sackl 1997). In the Alps, Golden Eagles usually nest near tree line and hunt in alpine areas at higher elevations (Haller 1996). Alpine populations have recently recovered from a long history of human persecution and are cur- rently considered to be stable or slightly increasing (Fasce and Fasce 1992, Haller and Sackl 1997). Land abandonment in remote mountain valleys, however, is causing widespread woodland expan- sion (Potter 1997, Tucker and Dixon 1997, Pedrini and Sergio 2001), mainly at the expense of alpine pastures, which are the main foraging habitats of Golden Eagles in the Alps (Haller 1996). Land abandonment and afforestation are predicted to have a high impact on Golden Eagle populations in European mountainous areas, with declines of more than 20% of the overall population over the next 20 yr if these trends continue (Tucker and Dixon 1997). Such habitat loss emphasizes the need for quantitative monitoring of the density and productivity of these eagle populations. De- spite the intensive studies conducted on this spe- cies in the western-central Alps (Huboux 1985, Fas- ce and Fasce 1992, Haller 1996) , very little data are available for the eastern portion of the alpine chain (Tormen and Cimbien 1995). Here, we report the results of an 11-yr study on a Golden Eagle population in the central-eastern Italian Alps. We provide quantitative data on the density, diet, and productivity of the population, compare them with estimates from other alpine populations, and examine some of the factors po- tentially affecting reproductive success. Study Area Golden Eagles were surveyed within a 7800 km^ study area in the central-eastern Italian Alps and pre-Alps, with- in the administrative provinces of Trento, Vicenza, Bre- scia, and Belluno (45°N,11°E). Elevation of the study area ranged from 65-3764 m and 31% of the land was <1000 m (Tomasi 1962); these areas were rarely used by Golden Eagles for hunting. Twenty-eight percent of the area was between 1000-1500 m, 22% between 1500-2000 m, 13% between 2000-2500 m, and 6% was >2500 m (Tomasi 1962). The natural tree line is at 1800-1900 m, but was often lower because of human activities and sheep graz- ing (Piussi 1992). The landscape is characterized by cul- tivated valley floors, mountain slopes covered by forests interspersed with sparse pastures, and by montane grass- land, rocky outcrops, and permanently snow-covered ground above tree line. In particular, 52% of the area was covered by woodland, 18% by montane grassland and pastures, 6% by agricultural crops (mainly vineyards and apple groves), and 3% by human development. Forest composition varied from deciduous to coniferous de- pending on elevation, slope orientation, and local micro- climate. With increasing elevation, woodland tended to be dominated by Quercus pubescens, Quercus-Tilia- Acer spp., Fagus-Abies spp., Picea spp., and Larix-Cembra spp., respec- tively (P.A.T. 1995). Eighty-two percent of the woodland area was managed for wood production; 73% of this woodland was managed as mature forest and 27% by stool shoots regeneration (P.A.T. 1995). Woodland extent is currently increasing at a rate of 1.0% every three years, mainly at the expense of alpine and subalpine grassland (P.A.T. 1995). Methods The Golden Eagle population was censused annually from 1982-92 inclusive, and intensively monitored be- tween 1984—89. A nest area was defined as an area where more than one nest was found in the same or in different years, but where only one pair nested each year (e.g , Newton 1979, Sergio and Bogliani 1999, Sergio and Bo to 1999). Each Golden Eagle pair occupied a traditional nest area, containing 1-9 alternate nests. Nest-area oc- cupancy was checked each year during the two months before the average laying date (23 March, N = 27). Ter- ritorial and courtship displays, copulations, nest material transfers, or presence of at least one freshly refurbished nest were considered as minimum evidence of nest-area occupation. The center of each nest area was defined as the arithmetic center of the locations of all alternative nests within the nest area. Nearest-neighbor distance (NND) was calculated as the distance from the center of the nest area of a pair to the center of the nest area of the nearest pair (Tijernberg 1985). Nest-area dispersion was examined by means of the Gstatistic (Brown 1975), calculated as the ratio between the geometric and the arithmetic mean of the squared NNDs; the Gstatistic ranges from 0-1 and values >0.65 indicate a regular dis- persion of nest areas (Brown 1975). Statistical signifi- cance of the deviation from randomness towards regu- larity was tested according to Clark and Evans (1954), with the modification applied by Donnelly (1978). Details of the mathematical procedure can be found in Krebs (1989). Individual eagles within territorial pairs were aged on the basis of plumage characteristics, following Jollie (1947), Mathieu (1986), and Tijernberg (1988). Imma- ture and subadult eagles were both classified as non- adults, following Mathieu (1986), and Tijernberg (1988) Nest contents were checked at least three times during the breeding cycle: during incubation to assess whether egg laying had taken place, just after hatching to record hatching date, and when the nestlings were >60-d old to record fledging success (nestlings usually fledge at 70-80 d old. Cramp and Simmons 1980). To minimize distur- bance, nest contents were checked from a distance >800 m with a 20-60x telescope. Hatching dates were estimat- ed from feather development, following Cramp and Sim- mons (1980) and Mathieu (1985). Laying dates were es- timated by subtracting 44 d, the median incubation 42 Pedrini and Sergio VoL. 35, No. 1 period (Cramp and Simmons 1980), from hatching dates. To analyze the factors potentially affecting breeding output, we first calculated a mean estimate of productiv- ity for each nest area within the study period. This mea- surement of nest area productivity was expressed as the percentage of nesting attempts which were successful (>1 young raised until fledging), and as the mean num- ber of fledged young per nesting attempt. Only nest areas where productivity had been checked for at least three years between 1984-89 were used for such analyses. To examine the effect of land use on productivity, we mea- sured the percentage extent of woodland and grassland habitats within 5 km of the nest-area center of 22 pairs. The measure of 5 km was slightly higher than half the average NND of the study population, and the area with- in 5 km of the nest-area center was assumed to be an estimate of the potential hunting range of the resident pair (Watson 1992, Pedrini and Sergio 2001). To minimize disturbance, collection of prey remains was carried out at all accessible nests 7-15 d after the first flight of the young (Watson 1997). Prey remains were identihed to genus or species level by comparison to a reference collection of the Trento Museum of Nat- ural History. Prey weights were calculated using Mathieu and Choisy (1982), Haller (1996), and Macdonald and Barrett (1993). Because capture roe deer {Capreolus ca- preolus) were mainly juveniles, they were assigned a weight of 3700 g, following Haller (1996). A reproductive pair was one which laid eggs, a suc- cessful pair was one which raised at least one young until fledging, and breeding success was the percentage of ter- ritorial pairs which were successful (Steenhof 1987). To meet the assumptions of normality of parametric tests, variables were logarithmically, square root, or arc- sm-square root transformed as necessary (Sokal and Rohlf 1981). All proportions of land-use types were arc- sm-square root transformed. Differences in mean pro- ductivity values between or among groups were tested by means of <-tests, or ANOVA; differences in frequency of successful territorial pairs or in frequency of successful pairs raising two young to fledging were tested by means of x^-tests (Sokal and Rohlf 1981). The effect of laying date, NND, and habitat composition within the potential hunting range on productivity was assessed by means of parametric and nonparametric correlations, and by par- tial correlation analysis (Sokal and Rohlf 1981). Means are given ±1SE. All tests were two-tailed and statistical signihcance was set at P < 0.05. Results Density and Nest Dispersion. Between 1982-92, the study area supported a stable population of 46 territorial pairs of eagles. This corresponded to a density of 5.9 pairs per 1000 km^. NND ranged from 4-15 km, averaging 8.7 ± 0.4 km {N = 46). The G-statistic value of 0.82 indicated a regular dis- persion of nest areas. The pattern of nest disper- sion deviated significantly from randomness to- wards regularity (z = 17.26, P < 0.0001, Krebs 1989). Nest Sites, The number of alternate nests within a nest area ranged from 1—9, averaging 2.7 ± 0.3 (N — 46). All but one of 123 censused nests were positioned on cliffs. The only tree nest was in a spruce fir {Picea excelsd) . Mean nest elevation was 1445 ± 32 m (range — 750-2280 m, = 108). Eighty-six percent of the nests {N = 108) were be- tween 1000-2000 m. Age of Territory Holders, Breeding Season, and Productivity. We classified both partners of 70 pairs as adult or nonadult; 59 pairs (84.3%) were com- posed of adult individuals, 6 pairs ( 8 . 6 %) of one adult male and one nonadult female, and 5 pairs (7.1%) of one adult female and one nonadult male. Mean laying date was 23 ± 1.5 March (range = 1 March-7 April, N — 27). Between 1984—89, there were no significant differences among years in mean number of young fledged per territorial pair (ibg 4 = 0.74, P — 0.59, Table 1), mean num- ber of young fledged per successful pair 43 = 0.49, P — 0.78), or percentage of nesting attempts which were successful (x ^5 = 2.48, P — 0.79). Factors Affecting Breeding Output. The number of fledged young was not related to laying date (r^ = —0.26, N = 27, P = 0.19). Breeding success was not correlated with NND (square root trans- formed) (r = 0.289, N— 25, P = 0.161). The mean number of fledged young was not correlated with percentage woodland extent within the potential hunting range (Pedrini and Sergio 2001). When controlling for NND through partial correlation, the relationship between mean number of fledged young per territory was not related to percentage woodland extent within the potential hunting range (r = —0.41, N — 19, P = 0.063), or to per- centage amount of grassland habitats within the potential hunting range (r = 0.40, N = 19, P = 0.075), though both results neared significance. Mean number of fledged young was higher for pairs composed by two adults (0.71 ± 0.1, N = 55) than for pairs composed by one adult and one nonadult individual (0.11 ± 0.1, = 9) {p ,2 = 4.28, P = 0.0001). Diet. Diet was dominated by mammals and birds which accounted for 64% and 32% of 247 prey items collected between 1984—89 (Table 2). The importance of mammalian prey was even higher in the analysis by fresh weight, with mammals ac- counting for 85% of consumed food. In particular, diet was mainly composed by species belonging to the order Artiodactyla (20% by number and 40% by fresh weight), Roden tia (29% and 25%), Lago- Table 1. Mean (±SE) productivity estimates of a Golden Eagle population in the central-eastern Italian Alps between 1982-92. March 2001 Golden Eagle Breeding Ecology 43 1982-92 109 55 i-H d + 1 CO o d +1 o i-H d (M I-H 1989 65 d + 1 CO t> d +1 00 d o 1988 d d 22 45 + 1 xn +1 o q d t-H Oi CO 1987 rH 56 d + 1 cC t-H d 1-H cu CO 1985 d d 64 + 1 ^ — 1 j> + l t-H t-H d t-H Table 2. Diet of breeding Golden Eagles in the central- eastern Italian Alps (1984-89), as estimated by food re- mains collected in nests. Prey Category Number OF Items (%) % Fresh Weight Mammals 157 (64) 84.6 Roe deer {Capreolus capreolus) 38 (15) 30.6 Alpine marmot {Marmota marmota) 26 (11) 22.6 Edible dormouse ( Glis glis) 25 (10) 0.7 Hares {Lepus sp.) 19 (8) 11.6 Red squirrel (Sciurus vulgaris) 13 (5) 1.0 Red fox {Vulpes vulpes) 9 (4) 5.9 Chamois {Rupicapra rupicapra) 11 (4) 8.9 Other mammals'^ 16 (6) 3.3 Birds 80 (32) 15.0 Black Grouse {Tetrao tetrix) 31 (13) 7.1 Hazel Grouse {Bonasa bonasia) 9 (4) 0.6 Rock Ptarmigan {Lagopus mutus) 9 (4) 0.8 Other birds’^ ^ 31 (13) 6.4 Reptiles'^ 10 (4) 0.4 Total 247 — “ Includes: common dormouse (Muscardinus avellanarius, A^= 4), pine marten (Alartes martes, N = 3), beech marten {Martes foina, N = 1), domestic cat {Felis silvestris, N = 1), brown rat {Rattus norvegicus, N= 1), Mustela spp. (N = 3), Apodemus spp. (N = 1), unidentified rodents (N = 2). ^ Includes: domestic fowl {Gallus spp., N= T), Capercaillie (Tetrao urogallus, N = 3), Rock Partridge {Alectoris graeca, N ~ 3), Eur- asian Jay {Garrulus glandarius, N = 3), Eurasian Kestrel {Falco tinnunculus, N = 2), Alpine Chough {Pyrrocorax graculus, N = 2), Ring Ouzel {Turdus torquatus, N = 2), Ring-neck Pheasant {Phas~ ianus colchicus, N = I), Common Cuckoo (Cuculus canorus, N = I), Mistle Thrush {Turdus viscivorus, N = 1), Hooded Crow (Cor- vus corone cornix, N = 1), unidentified Galliform {N = 5). Includes: Aesculapian snake {Elaphe longissima, N = 4), western whip snake {Coluber viridiflavus, N = 1), Vipera spp. {N = 1), unidentified Colubridae {N = 4) . 00 cn cr> ^ CM d +1 d o d + 1 o o lU fX .b Oh ^ -O ^ u y u JJ u 0 bo O 0 Sh o Oh ■o be o C be . 0 ^ 0 u $ 'rt Oh^ Ch O Si be • ri O ■ • 2 g •z.&'S. S morpha (8% and 12%), Carnivora (7% and 9%), and Galliformes (28% and 15%). Main prey spe- cies by number were roe deer, alpine marmot {Marmota marmota), Black Grouse {Tetrao tetrix), and edible dormouse ( Glis glis) . Main prey species by weight were roe deer, alpine marmot, hares {Le- pus spp.), and chamois {Rupicapra rupicapra). Comparison of Breeding Parameters with Other Alpine Populations. We obtained estimates of den- sity, mean NND, and productivity of Golden Eagles in other alpine areas (Table 3) . It was possible to compare mean NND in our study area with one in the western Italian Alps (Fasce and Fasce 1984) and in the French pre-Alps (Mathieu and Choisy 1982) . Mean NND did not vary significantly among Table 3, Estimates of density, mean (±SE) nearest-neighbor distance (NND), and productivity for 13 Golden Eagle populations in the alpine chain (1978 44 Pedrini and Sergio VoL. 35, No. 1 w u 0 CO O z a Ph ^ C/5 m j> X 05 lO X 05 t-H o o rfa CO X CM CO lO CO GM o t-h CM ' — ' ' — ' ' — ' ' — ' ' — ' ' — ' ' — ' ' — ^ HH C/) o cO l> o o 05 X CO o CM q O U > d d d d d GM X CO d CM CM d 1-H 3- H X CM GM 1-H fa o d fa Q fa O Q fa fa (fa oi fa fa O Z D fa < n “ O fa fa fa H O Z 3 fa fa fa PQ Q z fa Oi GM 00 cn o ^ ^ fa. " r- pi -2 - o 3 o K o cc CTi O fa fa 'D •M CO Oi 3 a; ♦ ^ fa 2 z o - 1 Z fa S w u. . ^ ■ c fa o; r5 ffi ii c ^ cj fa 0 o o o o +1 +1 +l +1 +1 01 C.J lO 00 CM 00 00 CO o o o o +1 +1 +1 +1 lO OO lO oO ■fa' lO O M 00 05 00 X 00 I I I I I 00 M O O O J> 00 00 X 00 05 ^ 05 0^ 05 05 05 X X X 05 I I I J> O X X X X X 05 05 05 05 4^ !/) cx < < P3 (Z) a c/5 zA> PL PL I OJ u a fa fa tfa fa a; s o CO V Jp fa PS o a fa !U Jp fa 3 o ^ CTl 00 'o X I V w ffi Tj c -fa "fa 0 •■S 1 OJ c fa c/5 a c/5 a ■fa c 3 O P3 (X X 05 !U t> X CO o o o o o o o o o o o o o , — , lO f7 . — . S' sS 0? , — . oT "C t-h o o CO UO o o CO LO CO o V5 lO 1-H G7 !05 rfa CO faH CO T— 1 CO t-h 1 CO CM 1 ^H t O 1 1 1 X CO CM fac 1 1 1 1 1 o X CO 00 05 I> X CO o 05 X < ^ ^ ^ ^ 00 T-h CO 00 t-h cO >- 1 J> 1 1 M (M X X O) 05 I I I I O G^ 05 M X i> X X 05 05 05 05 o fa OJ 0 fa OJ OJ fa PlH .OJ .OJ ^ ^ I 0 fa 2 c/3 (73 V I 13 u w £3 0^ o £ s a GO ^ _L OJ X fa-i 00 o> CM s 2 fTi OJ > o o o u u 1-H :3 u (/3 U 1-1 u CI- T3 VS OJ faj .2 s i-. OJ cx S/J .s 1/3 'u U 0 T3 :3 .2 0 E *c ^ _5 (L faj _ (U 0 cx Lm .S 3 .2 ‘u OJ Dh ° T3 U OJ OJ Sfa V3 jp p c C fa 0) V u u 3 t/3 be c o o u Xi u . a c/i Ifa a OJ .3 I o *5 CJ ^ o Oi o CO q q o >o d d ,—4 d d CO CM U lo CM CM wo q CO q o CO q q q q W CM d l-H d d d CM CO d U HH + 1 + 1 +1 + 1 + l + l + l + 1 + 1 2 r- q q o q q lO 00 o d CO CO d t-h d J> d d C30 o cn CM X: X CO CO o CM CO O o TJH I— 1 CO CM i-H l-H d ci i/1 c/5 lO O ■Sd •Hi Co S 1 g u u w 0, C/D 5S s d, Q, U3 d t/3 ?s IT! D ■se ^ ■ CH r S &4 So d c/5 CJ d Qh • ini -Hi 'H.d s §D s a, V) C/> o ■Hi da < •Hi CO S d d CQ oq o 1— H CM q 00 o CO q o q o l-H CO d d d 1 — 1 d d CM d + 1 + 1 + 1 + 1 + 1 + 1 + 1 +1 + 1 + 1 +1 q CO ac q o H q q o 'Cf o CO d d d d rH CM rH d d d CM CM q O o O W) CO o q o o q d d d d d d d d d d d + 1 + 1 +1 + 1 +1 + 1 + 1 + 1 + 1 +1 +1 q o o o lO q q o q o o l-H d d d d d d d d q q q q q q q q q q q 00 d CO CO d d CO CM l-H l-H ^H +1 + 1 + 1 +1 + 1 + 1 +1 + 1 + 1 +1 + 1 q q q q q q q ^H q q cri 00 d d CM rH d cd CO CM CM CO rH l-H CM O q o O CM q Cv[ d d rH d d d •^H d l-H d d + 1 + 1 + 1 + 1 + 1 + 1 + 1 + 1 + 1 + 1 + 1 q q q o o CM q q q l-H tr d d CM d d d CM I-H CM 1-^ d CO o O q o CO o o 00 o q d d d CO d d CO d d d 00 + 1 + 1 + 1 + 1 +1 + 1 +1 + 1 + 1 +1 + 1 o o o o o o j> o l-H d d d d d d d d Oi d U l-H rH QO q q Th o CO q l-H WO CO q CM CM CO 00 d Th 00 d 00 + 1 + 1 + 1 +1 + 1 + 1 +1 + 1 + 1 +1 + 1 CM o q o o cn lO q 00 CD q CM d i-H d d 00 U q 00 00 00 rH l-H CM CO CM CM CM 00 00 CO 00 q o q O q d d CO d d CM d i> CO U + 1 +1 +1 + 1 + 1 + 1 + 1 + 1 + 1 + 1 + 1 q q O) CO o q q o lO q rH rH l-H d CO d hh d d 00 ^H CM CM CM CM CM CO WO q wo q o q q q CO q q d CM d CM d H d H d + 1 + 1 + 1 +1 + 1 + 1 + 1 + 1 +1 + 1 + 1 q q q q q q q q (O) q rH d d l-H d d d CM d d 00 CM co CM rH m -etc 1-H CO CM X rO X Th CM -f CM Th d m CO r- CM l-H CM I-H 1-H 1-H I-H CJ d i-H Oh 0 a; > s u u "•Hi * Hi n 3 q/5 "•ce H) 5^ ' Hi *G u ‘5b u o •1^ 1) e 0 OJ S ' Hi g O CS ci ‘<1 • Hi d cs d g 0 u s 0 a •Hi a i/1 Q r.i "Hi • 50 m) or a measuring tape (for distances <50 m), I revisited each roost site during daylight hours to verify roost type. I placed roost sites into four classes: tree branches, nat- ural cavity, eave (a human-made structure where kestrels were not inside a structure but were protected from the elements) and building (a human-made structure such as a barn or silo where kestrels physically entered the structure) . All kestrels classified as year-round residents were color-banded birds observed on winter territories before 15 September. Any bird that arrived after 15 Sep- tember and was not previously color-banded was classi- fied as a winter resident. Results Nest Box Use. I encountered no American Kestrels us- ing nest boxes either during nest box observations or during visits to boxes. A mcyority of nest boxes were emp- ty (65%); boxes were used by eastern gray squirrels (Sau- rus carolinensis; 20%), mice (Peromyscus spp.; 10%), and an Eastern Screech Owl (5%). Roosting Behavior. Wintering kestrels used human structures, tree branches, and tree cavities as roosts (Ta- ble 1). Kestrels roosted in maples {Acer spp.; N = 4), eastern white pines {Pinus strobus; N= 1) , and oaks ( Quer- cus spp.; N = 3). A majority of roost sites were human- made (68%). There was no difference between natural and human-made roosts in roost height, time before sun- set that kestrels entered the roost, nor in the distance that kestrels moved from their last perch to the roost (Table 2). I detected no difference between male and female kes- trels in roost-site types (x^ = 2.72, df = 3, P = 0.43), nor in roost height, time to roost, and distance traveled from their last perch to roost (Table 2). I also detected no difference between year-round residents and winter res- idents in roost-site types (x^ = 2.78, df = 3, P = 0.41), roost height, time to roost, and distance traveled from last perch to roost (Table 2). There was no interaction between sex and residency status for roost height, time to roost, and distance traveled from last perch to roost (Psi 2i < 2.56, ft > 0.12). Table 2. Characteristics of roost sites {x ± SD) used by American Kestrels wintering in southeastern Pennsylvania Height of Roost Site (m) Time to Roost (Min Before Sunset) Distance Traveled From Last Perch to Roost Site (m) All observations (N= 25) 5.7 ± 2.4 3.4 ± 1.3 34.4 ± 20.5 Natural roost sites (N= 8) 6.4 ± 1.5 3.3 ± 1.5 40.7 ± 27.6 Human-made roost sites (N= 17) 5.4 ± 2.7 3.5 ± 1.3 31.5 ± 16.3 lvalue ft, 21 = 0.99, ft, 21 = 0.25, ft, 21 = 1.08, P = 0.32 P = 0.62 P = 0.31 Male kestrels (N= 13) 5.0 ± 1.4 3.6 ± 1.0 37.7 ± 27.2 Female kestrels (N= 12) 6.4 ± 3.1 3.3 ± 1.5 30.9 ± 9.2 ft,2x = 2.34, lO o II ft,2i = 0.70, P = 0.14 P = 0.51 P = 0.42 Year-round residents (N= 11) 6.2 ± 3.1 3.3 ± 1.3 31.5 ± 16.7 Winter residents 5.1 ± 1.08 3.6 ± 1.31 37.6 ± 24.3 (N= 14) ft, 21 = 1.13, ft, 21 - 0.18, ft,2i = 0.42, P - 0.3 P = 0.67 P = 0.52 60 Short Communications VoL. 35, No. 1 ijlSOUSilON Patterns of kestrel roost use at my southeastern Penn- sylvania site were similar to those reported for a similar latitude (Mills 1975) and a more southerly site (Doody 1994). My results suggested that wintering kestrels at my study site may not use nest boxes as roost sites even though they are available. There is year-to-year variation in whether kestrels use nest boxes as winter roosts (Car- penter pers. comm.), and in some areas, nest boxes are never used as winter roosts (Bortolotti pers. comm.). Giv- en that my data are limited to a few observations in one season, it is premature to conclude that kestrels never use nest boxes at my study site; however, my results sug- gest that other roost sites may be more important to kes- trels than nest boxes. Because the winter energetics of kestrels are strongly influenced by convective heat loss (Hayes and Gessaman 1980, Ardia 1997), the lack of use of nest boxes is some- what surprising if the use of boxes has thermal benefits. Therefore, that nest boxes are not used suggested that nest boxes may provide no better thermal benefit than either natural or human-made roosts, or, more likely, that benefits of nest boxes are outweighed by potential costs. A possible cost of nest boxes may be increased risk of predation (Orell 1989); one cause of mortality in breed- ing kestrels is predation on females incubating or brood- ing at night (Kellner and Ritchison 1988, C.J. and S. Rob- ertson, D. Ardia unpubl. data). In winter, without the need to enter nest boxes to breed, females may avoid exposure to predation. Also, nest boxes may increase ex- posure to ectoparasites (Merila and Allander 1995), al- though winter weather conditions are often harsh enough to reduce ectoparasite activity. In my site and others, kestrels have readily adopted human-made roosts and use these sites over natural lo- cations, both of which were readily available. I observed no differences between males and females nor between year-round and winter residents, suggesting that if trade- offs exist in roost site selection, they may be similar for all kestrels. Further observations on winter roosting be- havior, especially the role of nest boxes, is needed across the range of the American Kestrel. Rfsumen. — Las perchas nocturnas pueden ser crfticas para reducir la perdida de energfa y los riesgos de de- predacion de las aves rapaces. Estudie el comportamien- to de Falco sparverius en las perchas de invierno en Penn- sylvania, para evaluar si utilizaron las cajas de anidacion como perchas nocturnas. No encontre evidencia que los cernicalos utilizaron las cajas. Sin embargo, al observar los individuos desplazarse a las perchas {N = 25), encon- tre que ambos machos y hembras tendfan a utilizar es- tructuras humanas (68% utilizaron cocheras y aleros como sitios de descanso en lugar de arboles y cavidades de arboles). No encontre diferencias entre machos y hembras en la altura de las perchas, hora antes del atar- decer para utilizar las perchas, o en la distancia de la ultima percha antes de utilizar la percha nocturna. Tam- poco encontre diferencias entre los residentes anuales y los de invierno en estas variables. Aunque los datos son limitados fue sorprendente encontrar que los cernicalos no utilizaron las cajas de anidacion como perchas noc- turnas. El posible incremento de los riesgos de depre- dacion y la exposicion a ectoparasites pueden sobrepasar a los beneficios “termales” que proveen las cajas de an- idacion. [Traduccion de Cesar Marquez] Acknowledgments I thank my late wife Melinda Ardia for intellectual in- spiration and field assistance throughout the study. I am grateful to K. Bildstein for advice and technical support. Additional field support was provided by B. and S. Rob- ertson. I thank the staff of Hawk Mountain Sanctuary Association for their assistance and generous use of their facilities. G. Bortolotti, T. Carpenter, B.J. Christman, S.M. Murphy, D. Varland, and an anonymous reviewer provid- ed helpful comments on the manuscript. I was supported by the Hawk Mountain-Zeiss Raptor Research Award and a Paul A. Stewart Award from the Wilson Ornithological Society. This is Hawk Mountain Sanctuary contribution number 42. Literature Cited Ardia, D.R. 1997. Sex-related differences in habitat selec- tion and foraging energetics in American Kestrels wintering in southeastern Pennsylvania. M.S. thesis. State Univ. of New York, College of Environmental Science and Forestry, Syracuse, NY U.S.A. and K.L. Bildstein. 1997. Sex-related differences in habitat selection in wintering American Kestrels Falco sparverius. Anim. Behav. 53:1305-1311. Atkinson, E.C. 1993. Winter territories and night roosts of Northern Shrikes in Idaho. Conpp. (Cresswell 1995, Tornberg 1997). There is also very little quantitative information on how the vulnera- bility of prey differs according to their behavior (Thiollay 1988, Cresswell 1995). Our study focused on a single pair of Peregrine Fal- cons that occurred regularly in urban habitat outside the breeding season. Our goal was to describe seasonal vari- ations in the composition of the diet relative to changes in the availability of prey and to assess prey selection ac- cording to the different heights at which prey typically flew. Methods Two adult Peregrine Falcons have regularly occurred during the nonbreeding season (from late May to begin- ning of February) within the historical center of Flor- ence, Italy, since at least 1993. Each year, they use the same habitual perches located on tops of the three his- torical monuments that tower over the city center. Using individual variations in morphological characters, we were able to distinguish the two individuals through a 60 X spotting scope. Observations on frequency with which they roosted close together at the same perches and their behavioral interactions, indicated that they were an established pair. 62 Short Communications VoL. 35, No. 1 pellets were collected below the peregrine roost sites be- tween January 1997-February 1999. We analyzed the prey remains and pellets according to Oro and Telia (1995) and Rosenfield et al. (1995) and identified a total of 46 different prey items taken by the falcons. Bird species identification was made through comparisons with spec- imens of the zoological collection of Istituto Nazionale Fauna Selvatica (INFS, Bologna, Italy). Bat hairs were identified through 40 X microscope analysis (Keller 1986). The number of prey individuals identified for each species represented the minimum number con- sumed by the falcons. The composition of the diet over the two nonbreeding seasons May 1997-February 1998 and May 1998-February 1999 was determined either as the relative frequency of capture or as the relative bio- mass consumed of each species over the total of individ- uals captured/biomass consumed by falcons. Average weights of bird species (Cramp and Simmons 1989) and bat species (Schober and Grimmberger 1993) were used as approximations for biomass consumed by falcons. The Friedman test for paired data was used to test for differ- ence in the rank of prey types (i.e., resident breeder, mi- grant breeder, passage migrant, and winter visitor) con- sumed during the three periods from May-July 1997 and 1998, August-November 1997 and 1998, December-Feb- ruary 1997-98 and 1998-99. An a level of 0.05 was used in all statistical tests. Selection of diurnal avian prey breeding within a 3-km radius of peregrine perches was also analyzed from May- July 1997 and 1998 according to Baker (1967), Treleaven (1977), Hunter et al. (1988), and Cramp and Simmons (1989). Hereinafter, this circular area is simply referred to as “Florence center.” Prey selection was assessed ac- cording to the following categories; (1) flight height - 162 hr of observations from elevated points enabled us to rank the bird community breeding within Florence center as either commonly observed flying high above the roof level of Florence center (high-fliers), sometimes observed flying high above the roof level (mid-fliers), birds rarely or never observed flying high above the roof level (low fliers); (2) species — the five most abundant high-flying species with relative abundance estimates > 1 0 % of the total breeding pairs counted within Florence center, N = 9533 (House Martin [Delichon urbica] , Swift [^Apus apus], European Starling ^Sturnus tiM /g'am] , Jack- daw [Corvus monedula\, and Pigeon \Columba livia]); (3) weight — ^we recognized three weight classes (1-100, 101- 200, and >201 g) within the five most abundant high- flying species, whose average weight ranged between 35- 300 g. We tested two null hypotheses: that relative prey con- sumption (expressed either by frequency and by bio- mass) occurred in proportion to prey abundance, consid- ering all the prey classes simultaneously and that prey consumption occurred in proportion to prey abundance within the same prey class. To test the former hypothesis, we used either Chi-squared goodness-of-fit or Kolmogo- rov-Smirnov goodness-of-fit tests according to data and to test the latter hypothesis we used a Bonferroni’s test. An a level of 0.05 was adopted as the minimum value for rejection the null hypotheses. We identified a total of 18 species of prey in the non- breeding diet of the peregrine pair we studied. Pigeons were the most important prey both by frequency of cap- ture (30.4%) and by biomass (54.0%). Two species of bats (Savi’s pipistrelle [Hypsugo savii] and Kuhl’s pipis- trelle [Pipistrellus kuhlii] ) and Swifts were also frequently- captured prey (15.1 and 8.7% of total prey, respectively). Wintering Black-headed Gulls {Larus ridibundus) were also important prey both by frequency (8.7%) and bio- mass (13.1%). The average weight of prey taken ranged from 7 g (Savi’s pipistrelle) to 305 g (Woodcock [Scolopax rusticola\), but 53% were <150 g. Prey taken by the peregrines varied seasonally, al- though this variation was not statistically significant either by frequency or biomass (Friedman test = 0.37 and 0.87, respectively, A = 3, P > 0.1), During May-July, the pair fed mainly on resident breeders (70% by frequency and 88% by biomass) and to a lesser extent on migrant breed- ers (30% by frequency and 12% by biomass). The per- cent of resident breeders in the diet decreased progres- sively during August-November and December-February (53 and 43% by frequency, 58 and 39% by biomass), while winter visitors increased to 16 and 57% by frequen- cy and 20 and 61% by biomass. Moreover, during Au- gust-November passage migrants became a considerable component of the diet (31% by frequency and 22% by biomass) . Among the 54 bird species breeding in Florence cen- ter (both resident and migrant), those that we com- monly observed flying high above roof level were taken by peregrines significantly more frequently (frequency of capture = 0.800) than expected based on their rel- ative frequency of occurrence (0.512). The opposite was true for those species sometimes or rarely observed fly- ing high above roof level which were far less frequently captured than expected (x^ with Yates’s correction = 4.56, df = 1, P < 0.05). Likewise, consumption of prey types within categories differed from that which was ex- pected with more high-flying species taken than mid- to low-flying species Bonerroni confidence interval = 0.569-1.0 vs. an expected frequency of 0.512 for high- flying prey and Bonferroni confidence interval = 0- 0.431 vs. an expected frequency of 0.489 for mid- to low- flying species. Among most abundant bird species commonly flying high above the roof level of Florence, peregrines showed no significant preference for any species (Kolmogorov-Smirnov goodness-of-fit test = 0.06, A = 5, P > 0.1) or weight (Kolmogorov-Smirnov goodness-of-fit test = 0.06, A= 3, P> 0.2). Our results were the same when we analyzed the data expressed as biomass of prey consumed. Discussion There is little information about the diet of Peregrine Falcons outside the breeding season (Baker 1967, Mearns 1982), mainly due to its variable use of numerous perch- March 2001 Short Communications 63 es and roosts during this period (Ratcliffe 1993). Despite the relatively small size of the sample of prey that we collected, we felt that it was representative of the non- breeding diet of Florentine peregrines. We identihed a total of 18 prey species and this fell within the range of 10-22 species in the diets of other European Peregrine Falcons in urban habitats which were based on larger sample sizes (Mocci Demartis and Murgia 1986, Ranazzi 1995). Our study confirmed that peregrines fed oppor- tunistically during the nonbreeding season adjusting their diet in accordance with season variations in prey abundance and the height at which they flew in the Flor- ence city center. Although not significantly different and likely due to our small sample size, the proportion of prey captures characterized by different residence status did not vary from that which was expected. Resident breeders were important prey, especially during the period May-July, both by frequency and by biomass. During this period, the migrant breeder, the Swift, was an important prey species by frequency (Ranazzi 1995). Between August and January, resident breeders were conspicuously re- placed by passage migrants and winter visitors in the diet. In particular, between December and January, the pro- portion of winter visitors was greater than that of resident breeders, especially when captures of Black-headed Gulls increased. Considering all the bird species breeding in Florence center and their relative abundance, high-fliers appeared to be significantly more vulnerable to peregrine preda- tion than mid- and low-fliers. Among high-flying birds, no significant preference for any one prey species was found. Even prey that were most important in peregrine diets were captured in proportion to their relative abun- dance. Our findings were not consistent with findings of Hunter et al. (1988) and Rosenfield et al. (1995) who found significant preferences of breeding peregrines for certain taxa. We also found no significant preference ac- cording to weight classes of prey. Nutritional require- ments of young and daily temporal constraints may force Peregrine Falcons to be more selective with respect to prey classes during the breeding rather than the non- breeding season. Resumen. — La dieta de una pareja de halcones peregri- nes {Falco peregrinus brookei) que estacionalmente habitan en el centre de Florencia, Italia, fue investigada durante la epoca no reproductiva (desde finales de mayo hasta principles de febrero). Los halcones peregrines ajustaron su dieta en forma oportunista de acuerdo a la variacion estacional en la disponibilidad de presas. Las compara- ciones entre el consume y la disponibilidad relativa de las diferentes clases de presas desde mayo^ulio mostra- ron que en un habitat urbano con construcciones altas la altura a la cual volaban las presas fue la mayor limitan- te en la dieta de este cazador aereo. Las aves que tipi- camente volaban por encima del nivel del techo de Flo- rencia fueron capturadas mas frecuentemente que lo esperado de acuerdo con su abundancia relativa. Lo con- trario sucedio con las aves que ocasionalmente volaban mucho mas alto que el nivel del techo. Los halcones per- egrines no mostraron ninguna preferencia entre clases y pesos de clases. [Traduccion de Cesar Marquez] Acknowledgments We are indebted with G. Benini for giving us access to her balcony on top of Hotel Medici in Florence Center We also thank P. Agnelli (Zoological Museum “La Spe- cola,” Florence University), S. Cannicci and G. Santini (Florence University), N. Baccetti (INFS), A. De Faveri (INFS), L. Serra and M. Zenatello (INFS), M. Chiavetta (Bologna University), U. Muccini (Florence Municipali- ty) , and P. Osticresi (Florence Cathedral) . Literature Cited Baker, J.A. 1967. The peregrine. Collins, London, U K Cade, T.J., M. Martell, P. Redig, G. Septon, and H. Tor- DOFF. 1994. Peregrine Falcons in urban North Amer- ica. J. Raptor Res. 28:45-46. Cramp, S. and K.E.L. Simmons. 1989. The birds of the western Palearctic. Vol. 2. Oxford Univ. Press, Oxford, U.K. Cresswell, W. 1995. Selection of avian prey by wintering sparrowhawks Accipiter nisus in southern Scotland. Ar~ dea 83:381-389. Hunter, R.E., J.A. Crawford, and R.E. Ambrose. 1988 Prey selection by Peregrine Falcons during the nes- tling stage./. Wildl. Manage. 52:730-736. Keller, A. 1986. Etude comparative de la structure fine des polls des pipistrelles d’ Europe (Mammalia: Chi- roptera). Rev. Suisse Zool. 93:409-415. Mearns, N. 1982. Winter occupation of breeding terri- tories and winter diet of peregrines in south Scotland Ornis Scand. 13:79-83. Mocci Demartis, A. and C. Murgia. 1986. Contributo alia conoscenza dello spettro alimentare del Falco pel- legrino, Falco peregrinus, in autunno-inverno. Riv. Ital Ornitol. 56:95-105. Oro, D. and J.L. Tella. 1995. A comparison of two meth- ods for studying the diet of the Peregrine Falcon J Raptor Res. 29:207-210. Ranazzi, L. 1995. Dati preliminari sul regime alimentare del Falco pellegrino {Falco peregrinus) a Roma. Avocetta 19:122. Ratcliffe, D.A. 1993. The Peregrine Falcon. T. &: A D Poyser, London, U.K. Rosenfield, R.N., J.W. Schneider, J.M. Papp, and WS. Seegar. 1995. Prey of Peregrine Falcons breeding in west Greenland. Cowfior 97:763-770. SCHOBER, W. and E. Grimmberger. 1993. Bats of Britain and Europe. Hamlyn Guide, Hamlyn Publishing Group, Ltd., London, U.K. Thiollay, J.-M. 1988. Prey availability limiting an island population of Peregrine Falcons in Tunisia. Pages 64 Short Communications VoL. 35, No. 1 701—710 in T.J. Cadc,J.H. Eiiderson, C.G. i heiander, and C.W. White [Eds.], Peregrine Falcon populations, their management and recovery. The Peregrine Fund Inc., Boise, ID U.S.A. Tornberg, R. 1997. Prey selection of the Goshawk Accip- iter gentilis during the breeding season: the role of prey profitability and vulnerability. Ornis Fenn. 74:15-28. TreleaVEn, R.B. 1977. Peregrine: the private life of the Pergrine Falcon. Penzance, U.K. Headland Publica- tions. Received 17 May 2000; accepted 28 October 2000 J Raptor Res. 35(1) :64— 65 © 2001 The Raptor Research Foundation, Inc. Bald Eagles Killed by Trains in New York State Ward B. Stone Wildlife Pathology Unit, New York State Department of Environmental Conservation, 1 08 Game Farm Road, Delmar, NY 12054 U.S.A. Peter E. Nye Endangered Species Unit, New York State Department of Environmental Conservation, 1 08 Game Farm Road, Delmar, NY 12054 U.S.A. Joseph C. Okoniewski Wildlife Pathology Unit, New York State Department of Environmental Gonservation, 108 Game Farm Road, Delmar, NY 12054 U.S.A. Key Words; Bald Eagle, Haliaeetus leucocephalus; mor- tality, trains. High-speed passenger trains were first introduced into the Hudson River valley rail corridor in eastern New York State in 1980. As of October 2000, there were 20 train trips per day during daylight hours between Albany and New York City on a set of tracks along the eastern shore of the river. Over the same period, use of the lower Hud- son River valley by nesting, migrating, or overwintering Bald Eagles {Haliaeetus leucocephalus) greatly increased. Sightings of Bald Eagles during annual January aerial sur- veys of the lower Hudson River have risen from none in the early 1980s to 28 in 2000 (Nye unpubl. data). These two developments have resulted in the deaths of Bald Eagles killed by high-speed trains. Between 1986 and Oc- tober 2000, we examined the carcasses of 10 Bald Eagles apparently struck and killed by trains, eight in the last four years. At least eight of the incidents occurred on the Albany-New York City tracks in the mid-to-lower Hudson River valley, A ninth bird was recovered from the front of a high-speed locomotive which traveled this route, but which also traveled to and from Vermont during the pe- riod in question. The tenth eagle was killed along a rail- bed that parallels the western shore of Lake Champlain and supports high-speed train traffic to Montreal, Cana- da. Based on two reports by rail personnel, it is likely that one other eagle (an adult) was struck (presumed killed) in the lower Hudson River valley in August 1996. Most of the train mortalities we examined occnrred during periods of fall migration or overwintering in Sep- tember (1 mortality), October (2), November (1), De- cember (1), and January (3). The two exceptions were immatures killed in late June 1997 and late July 2000 Nine of the eagles were immatures (five female, three male, one unknown sex) , and one was an adult male. All were in good nutritional condition except for one im- mature struck in September. Analyses for organochlorine pesticides and polychlori- nated biphenyls (PCBs) completed at Hazleton Labs, Madison, WI U.S.A.; EnChem, Inc., Madison WI U.S.A.; Illinois Animal Disease Laboratory, Centralia, IL U.S.A , and University of Mississippi, MS U.S.A. on various tissues from eight eagles (brain [4 birds], liver [3], subcutane- ous fat [3]) did not reveal levels that might implicate them as predisposing factors in these train mortalities. DDE was detected in .seven birds (0.02-0.73 ug/g in brain, 0.13-0.19 ug/g in liver, 1.4-3.73 ug/g in fat). PCBs were present in six birds (0.13-13.8 ug/g in brain, 0.45- 4.3 ug/g in liver, 3.7-71 ug/g in fat). Highest levels were found in the immature from June 1997 supporting sus- picions that it fledged from a nest near the PCB-contam- inated Hudson River. Other organochlorines detected less frequently and at much lower levels included DDD March 2001 Short Communications 65 (max = 0.084 ug/g in brain, 0.023 ug/g in liver, and 1.2 ug/g in fat), DDT (max = 0.04 ug/g in brain, 0.52 in fat), and chlordane-related compounds (max = 0.92 ug/g cis-nonachlor in fat, 0.019 ug/g in liver). Analyses for lead and mercury in the livers of seven of the eagles similarly excluded those contaminants as pre- disposing factors with the possible exception of an im- mature male that showed 23 ug/g mercury. This bird was m excellent nutritional condition, in contrast to what might be expected in an animal overtly intoxicated with mercury. Levels of mercury in the other six eagles ranged from 0.76-6.0 ug/g. Levels of lead (0.03-0.42 ug/g) were not notably elevated. Rail hazards to wildlife have almost certainly been greatly increased by the newer high-speed passenger trains. Traveling at speeds up to 200 km/h, these trains are also remarkably quiet, further increasing the hazard. Eagles and other wildlife may have difficulty in effectively responding to objects moving at speeds greatly exceeding those ordinarily encountered in nature. Animals killed by trains attract scavengers that are subsequently imper- iled. Two of the eagles in this report were observed stand- ing near carrion before they flew into the path of the train, and it is likely that most of the others were struck in similar circumstances. The preponderance of immature birds among the train casualties contrasts with the frequency of adults (63%, N = 141) observed during the January aerial sur- veys of the lower Hudson over the past 14 yr (Nye un- publ. data). This suggests that immatures are more vul- nerable to this hazard, possibly because young eagles are more dependent on scavenging (Stalmaster 1987). Win- ter trapping efforts in the same region using deer or duck carcasses consistently attract more immatures (Nye unpubl. data) . Young eagles may be further jeopardized by being both less wary and less agile than adult birds. The significance of these train mortalities for eagles utilizing resources near these high-speed rail corridors is uncertain. The casualties reported here comprise 10 of 16 post fledged eagles examined from this part of New York State since 1986 (24% of 41 examined statewide). It has been suggested (Harmata pers. comm.) that high- speed trains, by providing a food base at critical periods (post fledging, first winter) , might actually promote over- all eagle survival and production despite causing occa- sional collision losses. At present, there are no data on amounts of carrion provided by the trains or use of the carrion by eagles that can be used to quantitatively eval- uate this hypothesis. However, given the abundance and availability of fish in certain stretches of the lower Hud- son River, the substantial amount of carrion linked to highway traffic in this region, and the frequency of col- lisions in relation to the number of eagles in this area. we suspect that the net effect of the high-speed trains is most likely to be decidedly negative. In any event, the greatest impact, positive or negative, will probably be on fledglings of local nesting pairs. Five nesting territories have become established in the lower Hudson River val- ley since 1992, and at least one of the train casualties was believed to have been a local juvenile. The immediate proximity of this high-speed rail cor- ridor to one of New York’s most important eagle habitats suggests that the number of eagle-train collisions can be expected to increase if eagle use of the lower Hudson Valley continues to rise, or if there are further increases in train speed or trip frequency. As faster passenger trains appear to be expanding in both numbers and destina- tions nationwide, further investigation of their impor- tance to Bald Eagles and other scavengers is warranted In the interim, ways to possibly reduce the hazards of these high-speed trains to scavengers, or to wildlife in general, should be explored. Carcass removal in certain locations and seasons might be beneficial, as might some sort of audible or visual signal that would scare avian scav- engers well ahead of the train. Such signals might be more effectively located along the track than on the tram itself. Resumen. — La parte baja del Valle del Rio Hudson ha crecido en importancia en las dos ultimas decadas como sitio de invierno y de anidacion de las aguilas calvas {Hal- iaetus leucocephalus) . A1 mismo tiempo ha habido una in- troduccion y expansion de los trenes rapidos de pasajeros en el corredor ferreo a lo largo del Rio Hudson. La in- troduccion de trenes de alta velocidad ha estado acom- panada de colisiones con aguilas calvas. Entre 1986-2000, 10 aguilas calvas fueron golpeadas y muertas por los tre- nes, ocho en los ultimos cuatro anos. Nueve de estas eran aguilas juveniles. Estos accidentes represen taron 10 de las 16 aguilas emplumadas a las que se les practice el post- mortem y que pertenecian a esta region durante este per- lodo de tiempo. El aumento en el uso de trenes de alta velocidad a nivel nacional puede aumentar la mortalidad de aguilas calvas y otras aves rapaces atraidas a la carrona ocasionada por los trenes. [Traduccion de Cesar Marquez] Acknowledgments We thank employees of Amtrak, Canadian Pacific, CSX, and Metro North Railroads for reporting eagle ca- sualties and/or freely discussing problems regarding wildlife mortality along railbeds. Literature Cited Stalmaster, M.V. 1987. The Bald Eagle. Universal Books, New York, NY U.S.A. Received 20 March 2000; accepted 18 November 2000 66 Short Communications VoL. 35, No. 1 J Raptor Res. 35(1):66— 67 © 2001 The Raptor Research Foundation, Inc. Early Nesting by Great Horned Owls in Montana Denver W. Holt and Stacy Drasen Owl Research Institute, RO. Box 39, Charlo, MT 59824 U.S.A. Key Words; Early nesting; Great Horned Owl; Bubo virgi- nianus; Montana. Great Horned Owls {Bubo xnrginianus) begin nesting earlier than any other owl species across its range in the United States and Canada (Austing and Holt 1966, Hous- ton et al. 1998). A correlation is evident between egg laying and latitude, with northern populations laying eggs later than southern ones (Houston et al. 1998). For example, in southern Florida, eggs are often laid in De- cember and rarely in November (Bailey 1925). In North and South Carolina, Great Horned Owls lay eggs in late December (Houston et al. 1998). In Ohio, a total of 903 breeding owls began nesting in January and February (Holt 1996). Craighead and Craighead (1956) found that the earliest egg laying occurred in Michigan was 12 Feb- ruary. In Saskatchewan and the Yukon Territory, Great Horned Owls lay eggs from late February through mid- May (Houston et al. 1998). In west-central Montana, Great Horned Owls lay eggs from mid-February to early April (Baumgartner 1938, Holt pers. obs.). In addition to their latitudinal variation in the timing of laying, it seems that Great Horned Owls may also lay their eggs later at higher elevations. Occasionally Great Horned Owls lay earlier than nor- mal in a given location. Watson (1933) reported finding a Great Horned Owl nest with one egg at Andover, New York on 29 January 1933, and, on 20 January 1935, Elder (1935) found a female incubating one egg near Madison, Wisconsin. Herein, we describe an unprecedented early nest record for the Great Horned Owl in Montana. Observations We have monitored nesting sites of Great Horned Owls near Missoula, Montana, to determine breeding since 1985. Monitoring begins in February, and only once have we seen evidence that nesting began before February. On 19 February 1996, while checking for nesting Great Horned Owls, we observed a nest with a female and two well-developed nestlings at the entrance to the Missoula International Airport, Missoula, MT (46‘’54'48"N, 114°05'02"W; elevation 960 m). The nest was a platform 7 m above the ground, and was originally constructed by a Black-billed Magpie {Pica pica) . The nest was in the middle tree of three, medium-sized Rus- sian olive {Elaeagnus angustifolia) trees, which were 7.5 m from each other. The trees lined the entrance to the airport, bordering a highway. The male owl was 74 m away in a densely-twigged Norway spruce {Picea abies) The surrounding area is relatively flat, and consisted of a mixture of open lands used for ranching, agriculture, and the airport. There were few trees in the area and it was exposed to winds. Thereafter, the nest was moni- tored daily until the nestlings moved onto nearby branches in the nest tree. The nestlings were almost identical in size, suggesting similar ages. Great Horned Owls usually hatch two days apart (Houston et al. 1998). After comparing our obser- vations with descriptions and photographs of known-aged Great Horned Owl nestlings (Hoffmeister and Setzer 1947, Austing and Holt 1966), we estimated the nestlings to be 4 wk old on 19 February. Considering a mean in- cubation period of 33 d (range = 30-37) (Hoffmeister and Setzer 1947, Austing and Holt 1966, Peck and James 1983), then egg laying occurred approximately 22 De- cember and hatching likely occurred about 22 January The young fledged on 11 and 14 March, almost two months earlier than usual for Montana (D. Holt pers. obs.). In fact, all of the breeding pairs that we monitored began to breed in mid-to-late February in 1996 {N =17) Most Great Horned Owls in west-central Montana typi- cally lay their eggs from mid-February to early April (Baumgartner 1938). The adult Great Horned Owls successfully fledged their young, even though the winter of 1996 was unusually cold Shortly after the eggs would have hatched on about 22 January, there were bitter Arctic winds in Missoula and the average daily temperatures and wind chills (in parenthe- ses) were -UC (-1°C); -2°C (-4°C); -4°C (-12°C), -8°C (-18°C); -10°C (-12TJ; -9°C (-23°C); -15°C (-30°C); -18°C (-23°C); -2.3°C (-23°C); -25°C (-25°) for 22-31 January and -24°C (-25°C); -26°C (-26°C); -25°C (-26°C); -15°C (-16°C) for 1-4 February. These data were calculated from readings of temperatures and wind speeds taken hourly 800 m from the nest by the Na- tional Weather Service. Discussion As in all owls, Great Horned Owl nestlings are ptilo- peadic (covered in white protoptile down) , semi-altricial, and nidicolous (Holt et al. 1999). They are essentially poikilothermic for the first five days after hatching, and are unable to maintain a body temperature >3°C above ambient temperature (Turner and McClanahan 1981) March 2001 Short Communications 67 In comparison with adults, the capacity of nestlings to regulate body temperature is 25% at 15 d, 69% at 20 d, 90% at 25 d, and 95% at 47 d (Turner and McClanahan 1981). During the cold period the owls were forced to endure in January and February 1996, nestlings were de- pendent on the female for thermoregulation. Female Great Horned Owls brood almost continuously for the first two weeks of the brood period (Houston et al. 1998) while males provide most of the food. Pakpahan et al. (1989) reported that Great Horned Owls have a low standard metabolic rate (SMR) relative to sym- patric raptor species, and that female Great Horned Owls have a lower SMR than males. This low SMR undoubtedly maximizes the efficient use of energy by female Great Horned Owls allowing them to begin nesting earlier than other raptor species. Although our observations help in our understanding of how early nesting can occur in the Great Horned Owl, they leave us wondering what might have in- fluenced this pair to begin nesting two months earlier than the species’ usual initiation date in Montana. Both ultimate and proximate factors interact to influ- ence the onset of breeding; ultimate factors tune a spe- cies phenology to the best average time for reproduction while proximate factors act as cues during any one sea- son. Temperature and photoperiod are purported to be important modifiers of annual gonadal cycles (Farner and Mewaldt 1952, Gwinner 1996). The months of No- vember and December had average monthly tempera- tures of 2.6 and — 2.8°C, which were 2.3 and 2.1°C above normal, respectively. It is conceivable that such warm weather could have influenced early gonadal and ovarian development; however, if temperature alone was respon- sible, then it seems other pairs should have also bred early. Artificial lighting for the airport and its entrance may have influenced the effective photoperiod, but whether this was a contributing cause is not known. Analysis of the pellets found around the nest and the male roost indicated that the pair relied on voles {Microtus montanus and M. pennsylvanicus) , European Starlings {Stur- nus vulgaris), and House Sparrows {Passer domesticus) for food. Unfortunately, we did not have local population es- timates of the prey species. Still, small mammal snap-trap- ping results in the Mission Valley, approximately 80 km away, resulted in no trapped voles in 500 trapnights (100 traps X 5 nights). Although we are unable to adequately answer the question why this pair initiated its nest so early, the fact that it successfully raised young provides compel- ling evidence for the hardiness of the Great Horned Owl. Resumen. — Observe un nido de Bubo virginianus con pi- chones de cuatro semanas en febrero 19 de 1996. Los buhos estaban utilizando un nido viejo de Pica pica en un arbol de un olivo ruso (Elaegnus angustifolia) . La pos- tura de huevos y la eclosion ocurrieron en este nido en- tre diciembre 20 y enero 22, dos meses antes de lo que usualmente ocurre en Montana. El promedio de tem- peraturas mensuales en noviembre y diciembre fue de 2.3 y 2.1 grados C por encima de lo normal, quizas el clima mas caliente pudo haber influenciado el desarrollo gonadal temprano en esta pareja. [Traduccion de Cesar Marquez] Acknowledgments We thank C.S. Houston, J.B. Holt, R. Austing, and MJ. Bechard for helpful comments on the manuscript. Literature Cited Austing, G.R. and J.B. Holt, Jr. 1966. The world of the Great Horned Owl. J.B. Lippincott Co., Philadelphia, PA U.S.A. Bailey, H.H. 1925. The birds of Florida. Williams and Wilkins, Baltimore, MD U.S.A. Baumgartner, F.M. 1938. Territory and population in the Great Horned Owl. Auk 56:274-282. Craighead, JJ- and F.C. Craighead. 1956. Hawks, owls and wildlife. Stackpole Co., Harrisburg, PA U.S.A. Elder, W.H. 1935. Early nesting of the Great Horned Owl. Auk 52:309-310. Farner, D.S. and L.R. Mewaldt. 1952. The relative roles of photoperiod and temperature in gonadal recru- descence in male Zonotrichia leucophrys gambelii. Anat Rec. 113:612-613. Gwinner, E. 1996. Circannual clocks in avian reproduc- tion and migration. Ibis 138:47-63. Hoffmeister, D.F. and H.W. Setzer. 1947. The postnatal development of two broods of Great Horned Owls {Bubo virginianus). Univ. Kansas Mus. Nat. Hist. 1:157-173. Holt, J.B., Jr. 1996. A banding study of Cincinnati Great Horned Owls./. Raptor Res. 30:194—197. Holt, D.W., R. Berkley, C. Deppe, P.L. Enriquez-Rocha, P.D. Olsen, J.L. Petersen, J.L. Rangel-Salazar, KP. Segars, and K.L Wood. 1999. Family Strigidae (spe- cies accounts). Pages 153-242 in], del Hoyo, A. Elli- ott, and J. Sargatal [Eds.], Handbook of the birds of the world. Vol. 5. Lynx Edicions, Barcelona, Spain. Houston, G.S., D.G. Smith, and C. Rohner. 1998. Great Horned Owl {Bubo virginianus). In A. Poole and F. Gill [Eds.], The birds of North America, No. 372. The Academy of Natural Sciences, Philadelphia, PA, and The American Ornithologists’ Union, Washington, DC U.S.A. PAitPAHAN, A.M., J. B. Haufler, AND H.H. Prince. 1989 Metabolic rates of Red-tailed Hawks and Great Horned Owls. Conrfor 91:1000-1002. Pegk, G.K and R.D. James. 1983. Breeding birds of On- tario: nidiology and distribution. Vol. 1. Nonpasseri- nes. Life Sci. Mus. Publ., Royal Ontario Museum, To- ronto, Canada. Turner, J.C., Jr. and L. McClanahan, Jr. 1981. Physi- ogenesis of endothermy and its relation to growth m the Great Horned Owl, Bubo virginianus. Comp. Bioch- em. Physiol. 68A: 167-1 73. Watson, C.H. 1933. Early nesting of the Great Horned Owl. Auk 50:220-221. Received 3 March 2000; accepted 28 October 2000 68 Short Communications VoL. 35, No. 1 J Raptor Res. 35(1) ; 68-69 © 2001 The Raptor Research Foundation, Inc. Diet of the Short-eared Owl in Northwestern Argentina Sebastian Cirignoli and Dario H. Podesta Facultad de Ciencias Naturales y Museo, Universidad Nacional de La Plata, Paseo del Bosque s/n, 1 900 La Plata, Argentina Ultses F.J. Pardinas Departamento Cientifico Paleontologia Vertebrados, Museo de La Plata, Paseo del Bosque s/n, 1 900 La Plata, Argentina Key Words: Short-eared Owl; Asio flammeus; diet; north- western Argentina. Despite its widespread distribution of the Short-eared Owl {Asio flammeus) in South America, there is very little known about its feeding habits (Martinez et al. 1998). In Argentina, there is very little dietary information (see Dalby 1975, Massoia 1985, Dieguez 1996). Here, we re- port on the results of an analysis of Short-eared Owl pel- lets from northwestern Argentina and compare our re- sults to those from other areas of South America. We collected the Short-eared owl pellets in April 1993 at Cerrito Leones, near Pan de Azucar mine, in the Man and Biosphere Reserve in Laguna de Pozuelos (22°15'- 22°27'S, 65°56'-66°03'W; 3600-4800 m elevation) in the Province of Jujuy, Argentina. The reserve is an endor- rheic basin with a central lake surrounded by two ridges. The landscape includes a complex mosaic of mountain and highland systems with grasslands and shrublands be- longing to the Puna Phytogeographic Province (Cabrera 1971) and slopes of bare rocks. The climate is cold and dry with a mean annual precipitation of 350 mm. The 72 fresh pellets that we collected averaged 44.4 ± 7 8 mm (± SD) in length (range = 29-62 mm) and 26.1 ± 3.4 mm in width (range = 16-35 mm). The mean number of prey in each pellet was 3.11 ± 1.12 individuals (range = 1-6) and it was not correlated with pellet length (r = 0.03, P > 0.50). The diet consisted mainly of native sigmodontine rodents (97.7%) with very low fre- quencies of marsupials, birds, and insects (Table 1). Two small rodents ( Calomys lepidus and Eligmodontia puerulus) accounted for >90% of the prey. Levins’ index of food- niche breadth (NB, after Marti 1988) was 2.30 for all prey categories and 2.20 for mammals. The geometric mean prey weight was 20.1 g. A one year trapping study of the rodents in Laguna de Pozuelos found that the community was dominated by Phyllotis darwini, C. lepidus, and Akodon albiventer (Bona- ventura et al. 1999; Table 1). P. darwini inhabits rocky slopes with shrubs of Fabiana densa and Baccharis holivien- sis, while C. lepidus is restricted to shrub areas of Parastre- phta lepidophylla on less rocky slopes. A. albiventer m found ubiquitously throughout the study area. According to Marks et al. (1999), the daily activity pattern of the Short- eared Owl is likely dictated by the activity of its main prey In Laguna de Pozuelos, high predation on crepuscular and nocturnal rodents such as C. lepidus and E. puerulus and the underrepresentation of diurnal rodents such as A. albiventer, suggested that the Short-eared Owls we stud- ied had mainly a crepuscular to nocturnal hunting peri- od. Low frequencies of Phyllotis in our pellet sample were probably due to its large size and not its availability since it does occur commonly in the diet of sympatric Great Horned Owls {Bubo virginianus) (Massoia 1994). Our results agreed with those of Massoia (1985) who studied the diet of Short-eared Owls in temperate and humid Pampean grasslands. There, small native sigmo- dontine rodents also dominated the diet of Short-eared Owls and birds and marsupials were negligible. Never- theless, his measure of NB was much higher (5.86 for all categories, 5.60 for mammals) and his geometric mean weight was even higher still (24.0 g). The harsh environ- mental conditions of Laguna de Pozuelos may have caused the apparent lower species richness of the Short- eared Owl diet there. Analyses of Short-eared Owl diets in continental Chile were also similar to our results. Rau et al. (1992) and Martinez et al. (1998) found that mice {Abrothrix olivaceu.s) were the most important prey of Short-eared Owls in northern Chile. Resumen. — Se documentan los primeros datos sobre la dieta de Asio flammeus en el noroeste de Argentina, sobre la base de 72 egagropilas recolectadas en la Reserva del Hombre y la Biosfera Laguna de Pozuelos (22°15'- 22°27'S, 65^56'-66“03'W, 3600-4800 m, Jujuy, Argenti- na). Mas del 90% de las presas consumidas fueron roe- dores sigmodontinos nativos de pequeho tamafio ( Calomys y Eligmodontia) . Los resultados obtenidos con- cuerdan con los registros previos para Argentina y Chile que indican una depredacidn centrada en micromami- feros de <30 g. [Traduccion de Autores] Acknowledgments Special thanks are extended to M.S. Bonaventura who collected pellets, C. Galliari for his identification of prey March 2001 Short Communications 69 Table 1. Diet of the Short-eared Owl in Man and the Biosphere Reserve in Laguna de Pozuelos, Province ofjujuy, Argentina. Prey Number Percent by Number Adult Body Mass'' Percent by Biomass Percent in THE Field’’ Rodents Calomys lepidus 122 54.7 15.1 43.9 22.5 Eligmodontia puerulus 81 36.3 21.4 41.3 — Auliscomys sublimis 10 4.5 38.5 9.2 — Phyllotis spp. 2 0.9 57.0 2.7 26.8 Cavia spp. 0 0.0 nd 0.0 1.4 Akodon albiventer 3 1.3 25.3 1.8 47.9 Marsupials Thylamys spp. 2 0.9 23.0 1.1 1.4 Birds 2 0.9 — — — Coleopterans Total 1 223 0.5 — — — ■* Mean adult masses in g were obtained from Redford and Eisenberg (1992). Data from Bonaventura et al. (1999). remains, and L. Shirlaw and M. Cremonte for helpful suggestions regarding the English translation. F. Jaksic and two anonymous reviewers improved the manuscript. This study was supported by a grant from Consejo Na- cional de Investigaciones Ciendficas y Tecnicas to U. Par- dihas. Literature Cited Bonaventura, S.M., R. Tecchi, V. Cueto, and M.I. Sanchez Lopez. 1999. Capitulo 9. Patron de uso de habitat en roedores cricetidos en la Reserva del la Biosfera Laguna de Pozuelos. Pages 127-137 mJ.L. Cajal, J.J. Garcia Fernandez, and R. Tecchi [Eds.], Ba- ses para la conservacion y manejo de la Puna y Cor- dillera Frontal. El Rol de las reservas de la Biosfera. UNESCO, Uruguay. Cabrera, A.L. 1971. Fitogeografia de la Republica Argen- tina. Bol. Soc. Argent. Bot. 16:1-42. Dal,by, P. 1975. Biology of Pampa rodents, Balcarce Area, Argentina. Museum Natural History, Michigan State Univ., Biol. Ser. 5:149-272. Dieguez, A.J. 1996. Aves depredadas por Asio flammeus suinda en Saladillo, Provincia de Buenos Aires. Bol. dent. Asoc. Protecc. Natur. 330:25-26. Marks, J.S., R.J. Cannings, and H. Mikkola. 1999. Family Strigidae (typical owls). Pages 76-242 in], del Hoyo, A. Elliott, and J. Sargatal [Eds.], Handbook of the birds of the world. Vol. 5. Barn Owls to humming- birds. Lynx Edicions, Barcelona, Spain. Marti, C.D. 1988. A long-term study of food-niche dy- namics in the Common Barn-Owl: comparisons with- in and between populations. Can. J. Zool. 66:1803- 1812. Martinez, D.R., R.A. Figueroa, C.L. Ocampo, and FM Jaksic. 1998. Food habits and hunting ranges of Short- eared Owls {Asio flammeus) in agricultural landscapes in southern Chile./. Raptor Res. 32:111-115. Massoia, E. 1985. Analisis de regurgitados de Asio flam- meus del arroyo Chasico. Acintacnia (INTA) 2:7-9 . 1994. Analisis de regurgitados de Bubo virginianus de Laguna de Pozuelos, Provincia de Jujuy. Bol. Cient Asoc. Protecc. Natur. 26:13-16. Rau, R.J., M.C. ViLLAGRA, MJ. Mora, D.R. Martinez, and M.S. Tilleria. 1992. Food habits of the Short-eared Owl {Asio flammeus) in southern South America J Raptor Res. 26:35-36. Redford, K.H. and J.F. Eisenberg. 1992. Mammals of the Neotropics. The southern cone: Chile, Argentina, Uruguay, Paraguay. Univ. Chicago Press, Chicago, IL U.S.A. Received 6 February 2000; accepted 16 October 2000 70 Short Communications VoL. 35, No. 1 J Raptor Res. 35(1) :70-7l © 2001 The Raptor Research Foundation, Inc. A Survey of Raptors on Rhodes: An Example of Human Impacts on Raptor Abundance and Distribution Arianna Aradis and Giuseppe M. Carpaneto Universitd degli Studi “Roma Tre, ” Dipartimento di Biologia, Viale G. Marconi, 446, 00146 Roma, Italy Key Words: raptor survey; Rhodes; human impacts. The Dodecanese includes more than 200 relatively-un- disturbed islands, only 27 of which are inhabited by peo- ple (de Grissac et al. 1994). Rhodes Island is a strategic area for bird conservation because it is close to important migration routes along the Turkish coast and may be a stopover for spring and autumn migrants. Because loca- tions on the island have experienced varying degrees of human exploitation, it is also a place where the impacts of human activities on raptor populations can be evalu- ated. The aim of our research was to verify the impact of high levels of tourism on raptor abundance and to de- termine the effects of road building and settlements on raptor abundance. We hoped to provide documentation for the impacts of people on raptor abundance in the Mediterranean region, where raptors and people are fre- quently in conflict. Study Area and Methods Rhodes is located in the southeastern part of the Ae- gean Sea, <20 km from the Turkish coast. The island is 80 km long and has a surface area of about 1400 km^. There are four habitat types on the island; coniferous forest {Cupressus sempervirens, Pinus brutia, Pinus halepen- sis) , maquis {Arbutus andrachne, Erica arborea, Quercus coc- cifera), phrygana {Thymus capitatus. Erica manipuliflora, Sarcopoterium spinosum, Cistus spp., Lithodora hispidula), and wetlands (de Grissac et al. 1994). Based on previous surveys, we divided the island into two zones of human impact: the northern zone with high levels of habitat degradation caused by touristic exploi- tation, and by the presence of an electric power station, and the southwestern zone which is still not impacted by tourism. Ten survey routes, five in the northern zone and five in the southern zone, were selected and each route was surveyed in August for three years (1997-99). The routes were approximately 20 km in length and each route was treated as a line transect (Fuller and Mosher 1987) to estimate the relative abundance of each species of raptor. Carriage roads were driven at 20-40 km/hr during the morning (0730-1230 H) and afternoon (1500-1930 H), alternating the time period in order to not bias the data. Relative abundance indexes were com- puted (Woffinden and Murphy 1977) as follows: R.A. = [(Number of species ^ total individuals observed) Number of km traveled] X 1000 Data for the northern and southern zones were analyzed separately for each year using Mann-Whitney G-tests. A Kruskal-Wallis test was used to compare the mean num- ber of each species of raptor observed in the two areas each year. All tests were two-tailed. Results A total of eight species of raptors was observed. Four species, Eurasian Kestrel {Ealco tinnunculus) , Long-legged Buzzard {Buteo rufinus), Eurasian Buzzard {Buteo buteo), and Eurasian Sparrowhawk {Acdpiter nisus) were resi- dents of the island. Eleonora’s Falcon {Ealco eleonorae) was a summer resident, and there were three migrants; Northern Hobby {Ealco subbuteo), Merlin {Ealco columbar- ius), and Booted Eagle {Hieraaetus pennatus). In all, 165 individuals were observed over the 600 km driven. In the northern zone, relative abundance estimates for raptors were similar (R.A. = 20) for all three years. The only raptors observed were Eurasian Kestrels (70.3%, N= 27) and Eurasian Buzzards (29.6%, N = 27). In the southern zone, eight species of raptors, Eurasian Kestrels, Eleo- nora’s Falcons, Northern Hobbies, Merlins, Booted Ea- gles, Long-legged Buzzards, Eurasian Buzzards, and Eur- asian Sparrowhawks, were observed (R.A. = 80). Kestrels and Eleonora’s Falcons were the most frequently ob- served (R.A. = 180 and 150, respectively). Long-legged Buzzards were more common (R.A. = 100) than Eur- asian Buzzards (R.A. = 76), and Eurasian Sparrowhawks were infrequently observed (R.A. = 20). We detected a difference in the relative abundances of raptors between the northern and the southern zones in each year of our surveys {U = 0, P < 0.05), but no sig- nificant differences were detected in the number of rap- tor sightings between northern and southern portions of the island. Discussion Our results indicated that the two zones of human im- pact on Rhodes have resulted in a differential distribu- tion of raptors on the island. Only kestrels and buzzards, species that are highly adaptable to humans (Newton 1979), were observed in the northern part of the island where development has been the greatest. Those species were observed in the southern portion of the island are uncommon in areas with large amounts of human devel- opment. Southern portions of the island are not as at- March 2001 Short Communications 71 tractive for tourism because the rocky seashore is often inaccessible and the sea is always rough owing to domi- nant easterly winds. Nevertheless, there was two potential threats to the area’s future suitability for raptors. The southern part of the island was increasingly being settled by people and there were projects designed to build new residences and several villages. Perhaps the greatest threat was a project to build a second electric power sta- tion on the island. The project would involve the destruc- tion of 8 ha of native habitat along the coast where the majority of raptor sightings were made. Because we observed so many migratory raptors during our surveys, Rhodes seems to play an important role in the migration of several raptor species. Rhodes is prob- ably important because it offers undisturbed resting ar- eas, water, and food resources. Due to this, future devel- opments on the island should take into consideration the importance of native habitats on the island to migratory species of raptors. Resumen. — En agosto desde 1997-99, censamos las aves rapaces en la Isla de Rodas para determinar los impactos de carreteras y construcciones asociadas al aumento de turismo en .su abundancia. Hubo dos zonas diferentes de impacto, la zona norte con altos niveles de turismo y la zona sur con impacto relativamente poco de la gente. Diez rutas de investigacion, cinco en cada zona de im- pacto fueron utilizadas. Un total de ocho especies de aves rapaces fueron observadas: Cuatro especies residentes Falco tinnunculus, Buteo rufinus, Buteo buteo, Accipiter nisus, un residente de verano Falco deonorae y tres migratorios Falco subbuteo, Falco columbarius y Hieraaetus pennatus. Las especies ocurrieron en forma desigual con especies aso- ciadas a la gente {Falco tinnunculus y Buteo buteo) mas fre- cuentemente observadas en la parte norte mas altamente impactada de la isla. [Traduccion de Cesar Marque/] Ackn o weed gments We thank Chelon — Marine Turtle Conservation and Re- search Program. We are very grateful to S. Fioretti for lo- gistic support and G. Gaibani for statistical advice. We also thank two anonymous referees for their helpful comments A particular thanks to V. Penteriani for his critical advice Literaiure Cited DE Grissac, A.J., V. Tilot, J.M. Sinnassamy, and P. Pan- AYOTIS. 1994. Preliminary study on conservation of the environment of the island of Rhodes. Regional Activity Centre for specially Protected Areas/IUCN Marine Programme, Athens, Greece. 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 techniques manual. Natl. Wildl. Fed , Washington, DC U.S.A. Newton, I. 1979. Population ecology of raptors. T. & A.D. Poyser, Berkhamsted, U.K. WOFFiNDEN, N.D. andJ.R. Murphy. 1977. A roadside cen- sus in the eastern Great Basin, 1973-1974. Raptor Res 11:62-66. Received 2 December 1999; accepted 26 October 2000 J. Raptor Res. 35(l):7l-73 © 2001 The Raptor Research Foundation, Inc. The Incidence of Intestinal Parasites in British Birds of Prey Nigel W.H. Barton^ and David C. Houston Ornithology Group, Institute of Biomedical and Life Sciences, Graham Kerr Building, Glasgow University, Glasgow, G12 8Qff Scotland, U.K. Key Words: raptors', intestinal parasite.s-, Britain. During studies on the comparative morphology of the digestive tract of British birds of prey, we examined the gut contents of 379 individuals of six raptor species. Methods of postmortem examination and the sources of these birds are given in Barton and Houston (1991, 1992, ^ Present address: The Falcon Facility, Penllynin Farm, College Road, Carmarthen, SA33 5EH, U.K. 1993a, 1993b, 1994, 1996). To obtain weights and mea- surements of empty digestive tracts, we removed the di- gesta from the whole gut by cutting the gut open along its entire length and carefully scraping out the gut con- tents, followed by washing. We took this opportunity to examine carefully the gut wall and gut contents for par- asitic worms. We only recorded those clearly visible to the naked eye, and some species or microscopic individuals too small to be seen may have been missed. We counted the total number of worms removed from each individ- ual. All those recovered were nematodes. This group of 72 Short Communicl\tions VoL. 35, No. 1 worms are extremely difficult to identify. Because their taxonomy is still very poorly known (D.W.T. Crompton and J.D. Ewald pers. comm.), we were not able to identify them further. Most of the carcasses examined had been found dead and snbmitted for pesticide analysis. The cause of death was assessed during the dissection and based on criteria routinely used by the Institute of Ter- restrial Ecology (Newton et al. 1982). Many birds had died from starvation or collisions. The body condition of all birds was estimated from pectoral muscle weight by dissecting the muscles, drying them to constant weight at 60°C, and extracting the fat by chloroform Soxhlet ex- traction to determine the fat-free dry muscle weight and fat content. A condition index, which accounted for the body size differences, was calculated from the residuals of the regression of dry muscle weight against a body size factor developed from factor loadings derived from Prin- cipal Components Analysis of five body measurements (Barton and Houston 1994). In this note, we report on the incidence of intestinal parasites and consider wheth- er there was any correlation between the number of par- asitic worms and the body condition of the birds. We examined 135 Sparrowhawks {Accipiter nisus), 23 Peregrine Ealcons {Falco peregrinus) , 76 Eurasian Kestrels {Falco tinnunculus) , 77 Common Buzzards (Buteo buteo), 9 Red Kites {Milvus milvus), and 59 Goshawks {Accipiter gen- tihs). The incidence of intestinal parasites was low, with only 20% of individuals having one or more parasites. There was no significant difference between species in the incidence of parasitic infection. The number of worms per individual was very variable, with up to 70 recorded from some individuals. It might be expected that birds in poor condition would have the greatest number of parasites and there was some slight evidence for this. Goshawks showed a significant negative correla- tion between the amount of pectoral fat and parasite numbers (r |3 = —50, P < 0.05), so birds with the least fat reserves had the greatest parasite load. There were no significant relationships between pectoral fat and para- sites for the other species. Our small sample sizes might have accounted for this, so we also examined whether those individuals with no parasites were in better condi- tion than those individuals with parasites. In buzzards, those individuals with little fat were also likely to be in- dividuals with parasites {U = 76.0, N = 39, P < 0.005, Mann-Wliitney G-test) as were those with the smallest lean dry muscle weights {U = 71.0, N = 39, P < 0.003), but there was no significant relationship for the other species. We were surprised that the incidence of parasite infec- tion was so low. Our sample may have been biased be- cause we discarded all birds that had not been freshly killed but, even with this precaution, most birds would have been dead for a day or so before collection, and frozen and thawed before examination. Few tapeworms were found. This may have been due to the fact that they disintegrated before the digesta was examined. Neverthe- less, the cuticle of nematodes is remarkably resilient (Bird 1971, Lee 1972) and all of tbe nematodes removed from the gut were in good condition. Therefore, we think it likely that most nematodes were recovered. There was some weak evidence of a correlation be- tween body condition and parasite load, although it is impossible to establish cause and effect. Birds in poor condition might have been more susceptible to parasit- ism. Alternatively, parasites could have been the cause of the poor condition. There is very little literature on the incidence of intestinal parasites in birds of prey. They appear uncommon in captive birds (Greenwood et al. 1984) but, apart from some references to unusual indi- vidual case histories (Simpson and Harris 1992), there have been few surveys of wild birds. Mclnnes et al. (1994) surveyed 109 Tawny Owls {Strix aluco) and found nema- todes were present in the small intestine of only 18% of individuals, although 68% were infected with an Acan- thocepahalan worm, and Houston and Cooper (1975) examined the intestines of 18 Rueppell’s Griffon Vul- tures ( Gyps rueppellii) and found nematode infections in 16 individuals. Parasites often use predators at the top of food chains as dehnitive hosts (Crompton and Nickol 1985) . There is a considerable lack of information about the role of birds of prey in parasite transmission and this is an area of study which might repay further examina- tion for those with the opportunity to carry out postmor- tem examinations. Resumen. — Examinamos 379 individuos de seis especies de aves rapaces britanicas de las cuales solo el 20% tenian nematodos intestinales. Hubo una pequefia evidencia que los individuos de Accipiter gentilis y Buteo buteo en con- diciones fisicas pobres tuvieron una carga mayor de par- asitos. [Traduccion de Cesar Marquez] Acknowledgments We are extremely grateful to I. Newton and I. Wyllie of the Institute of Terrestrial Ecology for kindly supplying the majority of the birds used in this study. Professor D.W.T. Crompton and Dr. J.A. Ewald kindly advised on our unsuccessful attempts to identify the species of nem- atodes found. Literature Cited Barton, N.W.H. and D.C. Houston. 1991. The use of titanium dioxide as an inert marker for digestion studies in raptors. Comp. Biochem. Physiol. 100A:1025- 1029. AND . 1992. Post-mortem changes in avian gross intestinal morphology. Can. J. 7x>ol 70:1849- 1851. AND . 1993a. The influence of gut mor- phology on digestion time in raptors. Comp. Biochem, Physiol. 105:571-578. AND . 1993b. A comparison of digestive ef- hciency in birds of prey. Iths 135:363-372. March 2001 Short Communications 73 AND . 1994. Morphology adaptation of the digestive tract in relation to feeding ecology in rap- tors. /. Zool. Land. 232:133-150. AND . 1996. Factors influencing the size of some internal organ systems in raptors. J. Raptor Res, 30:219-223. Bird, A.F. 1971. The structure of nematodes. Academic Press, London, U.K. Crompton, D.W.T. AND B.B. Nickol. 1985. The biology of the Acanthocephala. Cambridge Univ. Press, Cam- bridge, U.K. Greenwood, A.G., C.W. Furley, and J.E. Cooper. 1984. Intestinal nematodiasis in falcons (order Falconifor- rnes). Vet. Rec. 114:477-478. Houston, D.C. and J.E Cooper. 1975. The digestive tract of the White-backed Vulture and its role in disease transmission among wild ungulates. J. Wildl. Dis. 1 1 306-313. Lee, D.L. 1972. The structure of the helminth cuticle. In B. Dawes [Ed.], Advances in parasitology. Vol. 10. Ac- ademic Press, New York, NYU.S.A. McInnes, F.J., D.W.T. Crompton, and J.A. Ewai.d. 1994 The distribution of Centrorhynchus alucornis (Acantho- cepahala) and Porrocaecum spiale (Nematoda) in Taw- ny Owls Strix aluco from Great Britain. J. Raptor Res 28:34-38. Newton, L, A.A. Bell, and 1. Wyllie. 1982. Mortality of Sparrowhawks and kestrels. Br. Birds 75:195-204. Simpson, V.R. and E.A. Harris. 1992. Cyathostoma Ian (Nematoda) infection in birds of prey. J. Zool. Lond 227:655-659. Received 17 May 1998; accepted 21 October 2000 Li±j 1 lERS / Raptor Res. 35(1):74 © 2001 The Raptor Research Foundation, Inc. First Sight Record of the King Vulture in Baja California, Mexico On 31 October 1999 from 0900-1130 H, we observed a solitary adult King Vulture (Sarcoramphus papa) at San Jose del Cabo Estuary, Baja California Sur, Mexico (23°03'N, 109°41'W; elevation just above sea level). The estuary is located just east of the Presidente Forum Los Cabos resort about 1.0 km southeast of San Jose del Cabo and is in a tropical and semiarid portion of the Baja California peninsula where the Rio San Jose meets the Pacific Ocean. Maya et al. (1997, pages 5-25 in L. Arriaga and R.R. Estrella [Eds.], Los oasis de la Peninsula de Baja California. Centro de Investigaciones Biologicas del Noroeste, S.C., La Paz, B.C.S., Mexico) provides a more complete description of this estuary. The King Vulture was observed roosting along with numerous Turkey Vultures ( Cathartes aura) in a dense grove of Mexican fan palms {Washingtonia robusta) that bordered a part of the estuary. According to Howell and Webb (1995, A guide to the birds of Mexico and northern Central America. Oxford Univ. Press, Oxford, U.K.), adult King Vultures are “unmistakable” and “usually seen singly,” and the species “associates with other vultures.” Although not photographed or collected, the King Vulture that we saw was easily separable from the Turkey Vultures roosting m the area. It was definitely larger than the Turkey Vultures and was aggressive towards them, apparently to acquire better sunning sites. The white wing feathers except for the main flight feathers that were black were easily observed while the bird extended its wings to thermoregulate in the morning sun. After returning from the field and reviewing Eitniear (1996, /. Raptor Res. 30:35-38), we determined that the King Vulture was an adult in definitive plumage, approximately 6-7 yr old. Its multicolored head was also easily seen, especially the portions that were orange. Our viewing distance was less than 150 m with the sun to our backs, and we used 8 X 42 binoculars. It was impossible to reduce the viewing distance for photographic purposes because of an intervening wetland. To our knowledge, this represents the first report of the King Vulture in Baja California. Brewster (1902, Bull. Mus Comp. Zool. 41:1-241), Grinnell (1928, Univ. Calif. Publ. Zool. 32:1-300), Wilbur (1987, Birds of Baja California. Univ of Calif. Press, Berkeley, CA U.S.A.), Howell and Webb (1995), and American Ornithologists’ Union (1998, Check- list of North American birds, 7th Ed. Am. Ornithol. Union, Washington, DC U.S.A.) did not list the King Vulture from Baja California. Howell and Webb (1995) state that the King Vulture is a rare lowland species that is decreasing in numbers, and their range map depicts the species’ former distribution in western Mexico to as far north as central Sinaloa near the Culiacan area. The 300 km distance between the mainland in Sinaloa, Mexico and the Cape region of Baja California Sur is not all that far for a species with the presumed flight range of the King Vulture. We could not relocate it on a brief visit to the same site two days later and biologists from the Centro de Investigaciones Biologicas del Noroeste, La Paz, Baja California Sur did not observe it there some weeks following our initial obser- vation (R.R. Estrella pers. comm.). It is not known whether this King Vulture observation represents a natural “ac- cidental” record or an escapee from captivity. It seems unlikely that it could have been the latter because no one knew of a King Vulture held in captivity in the Santiago Zoo or elsewhere in Baja California (R.R. Estrella and A.C Vera pers. comm.) . It was also unlikely that it could have escaped from a zoo in California because none were reported and most are kept in “double-door” facilities and California has one of the most restrictive state wildlife regulations in the U.S. (J. Bellinger pers. comm.). It is, of course, possible the bird originated from an area in the west outside of California where wildlife regulations are less restrictive. However, we feel that this was not an escaped King Vulture that we observed in Baja California but was, in fact, a free living bird well away from its usual range in mainland Mexico. We thank M. Bechard, J. Bellinger, J. Clinton Eitniear, R. Rodriguez Estrella, L. Kiff, S. Speich, P. Unitt, A. Castel- lanos Vera, and S. Wilbur for reviewing the draft manuscript and making useful suggestions. — Russell B. Duncan and Jo Ann V. Lacroix, R.B. Duncan 8c Associates, Biological Consultants, 6111 Bobcat Lane, Tucson, AZ 85743 U.S.A. 74 March 2001 Letters 75 J. Raptor Res. 35(1):75 © 2001 The Raptor Research Foundation, Inc. Probable Replacement Clutches by Booted Eagles {Hieraaetus pennatus) IN the Tietar River Valley of Central Spain The Booted Eagle {Hieraaetus pennatus) is a summer resident species in Spain whose breeding ecology is poorly known (Brown and Amadon 1968, Eagles, hawks, and falcons of the world, Vols. 1 and 2, McGraw-Hill Book Company, New York, NY U.S.A.; Cramp and Simmons 1980, Handbook of the birds of Europe, the Middle East, and North Africa, Vol. 2, Oxford Univ. Press, Oxford, U.K.; Brown et al. 1982, The birds of Africa, Vol. 1, Academic Press, London, U.K.). Available information is limited to scarce data on clutch sizes, laying dates, productivity, and food habits (Labitte 1955, Alauda 23:249-253; Suetens and Van Groenendael 1969, Ardeola 15:19-36; Iribarren 1975, Ardeola 21:305-330; Steyn and Grobler 1981, Ostrich 52:108-118). To my knowledge, replacement clutches have not been reported for this species. Here, 1 report the first records of replacement clutches in Booted Eagles. The data were obtained during a long-term study of breeding ecology of the species in the Tietar River valley (Avila Province, central Spain, 40°40'N, 4°42'W) from 1995-2000. This is a mountainous area (300-2594 m) with large tracts of pine forest {Pinus pinaster) interspersed with smaller clearings of cultivated lands and scrubland. Booted Eagles arrive from their wintering grounds in late March and early April. Known territories were checked every two days to estimate arrival dates of each pair. The rest of the study area was surveyed intensively each year during March- April to locate new pairs. Nests were checked first after the female was observed in incubation posture. Eggs were measured and marked with felt pens, and I noted if they were warm or cold. When single eggs were found in nests during the first nest visits, 1 made a second visit two days later, or as soon as weather conditions permitted. Only two replacement clutches were found among 82 breeding attempts with accurately-known clutch sizes. The first replacement clutch was recorded in May 1998. This pair nested on a platform used by a pair of Black Kites {Milvus migrans) during the previous year. Aggressive interactions between the pair of Black Kites and the Booted Eagles were frequendy observed before the first nest check on 25 April. On this first visit, I found a recently-broken Booted Eagle egg on the ground below the nest and another egg on the nest. On 29 April the female was not incubating, but was seen perched near the nest tree. I revisited the nest again on 2 May and a new Booted Eagle egg was found and marked. On that visit, I did not find remains of the original eggs. Three days later the nest was checked again, and only the marked egg was found. On 13 June, I visited the nest again, but the egg was not found. The second replacement clutch was observed in May 1998 at a newly-occupied nesting territory. I had surveyed this area since 1995 but did not find any evidence that Booted Eagles occupied the site. On 13 April 1998, I found the nest of this pair and, on 17 April, I found the female incubating one egg in the nest. The nest was checked again two days later, and I found only the marked egg. The female was incubating when the nest was visited again 38 d later (the typical incubation period). The marked egg was in the nest, dirty and only slightly warm. Three days later, I visited the nest again. There were eggshell remains of the original egg, and a new recently-laid egg, cold, clean, and completely white was found. A Booted Eagle was flying over the nest site. On 5 July, the nest was checked again, but only eggshell remains were found. The adult Booted Eagles were not marked, so I was not absolutely sure that these were replacement clutches. They could have been new clutches of different pairs of eagles. If these were cases of pair replacement, I would have observed more than two birds in the same territory, aggressive interactions between individuals, courtship displays, and/or territorial flights as I observed in other cases of suspected pair changes. Therefore, I concluded that these were replacement clutches laid by Booted Eagles whose initial clutches failed early in the nesting season. After failing in the early stages of the breeding cycle. Booted Eagles may produce one replacement egg, supporting the suggestion that mid-size raptors that usually lay only one clutch can lay replacement eggs if they have lost the original clutch at an early stage (Newton 1979, Population ecology of raptors, T, & A.D. Poyser, London, U.K.). I am grateful to J. Gomendio who allowed me to do this study. This work could not have been carried out without the help by M. Garcia Tornero and J. Munoz Eamiliar. M. Ferrer, G. Bortolotti, C. McIntyre, and J. Vinuela made critical evaluations of previous drafts. This paper is dedicated to Warden Marcos, from the town of Guisando, as a model of what exploiting and preserving our surrounding nature should be. — Ignacio S. Garcia Dios, Departamento de Ecologia Evolutiva, Museo Nacional de Ciencias Naturales, Jose Gutierrez Abascal 2, 28006 Madrid, Spain. Book Reviews Edited by Jeffrey S. Marks J Raptor Res. 35(1) ; 76-7'7 © 2001 The Raptor Research Foundation, Inc. Owls: A Guide to the Owls of the World. By Claus Konig, Friedhelm Weick, and Jan-Hendrick Becking. 1999. Yale University Press, New Haven, CT. 462 pp., 64 color plates, numerous maps and line drawings. ISBN 0-300-07920-6. Cloth, $50.— Claus Konig and his team have produced an ex- cellent book on the owls of the world. Owing to new research findings, and to the fact that Pro- fessor Konig strongly believes that differences in vocalizations serve to separate even closely related taxa of owls, no other book has recognized so many species of owls. This book treats 212 species, versus, for instance, the 205 species treated in vol- ume 5 of the Handbook of the Birds of the World (del Hoyo et al. 1999). Burton’s (1973) Owls of the World recognized 134 species, and in that same year Eck and Busse proposed a new taxonomic revision of the world owls with only 108 species. So, in less than 30 years we have found or created 104 new species of owls on this planet. This is a remarkable taxonomic achievement for which Professor Konig and his team deserve special con- gratulations. Although this book has named several new owls and reclassified many old taxa, I was disappointed that it did not attempt to harmonize English names. For example, for many years in Africa we have called Glauddium perlatum the Pearl-spotted Owl and G. capense the Barred Owl. Now, Konig et al. have joined others (e.g., del Hoyo et al. 1999) in calling these species Pearl-spotted Owlet and Af- rican Barred Owlet, respectively. In the Dictionary of Birds, Campbell and Lack (1985) define “owlet” to stand for a young owl, so to my mind no species should he called “owlet.” Especially strange is that we have owlets not only in Glauddium (11 species), but in Xenoglaux (1) and Athene (2). This is very confusing, to say the least. If in the genus Otus (as this book does correctly) we call the Old World species “scops-owls” (41) and the New World species “screech-owls” (19), can’t we similarly rename all Glauddium “pygmy- owls,” all Athene “little-owls,” and Xenoglaux loweryi (Long-whiskered Owlet) simply the Long-whis- kered Owl? With the same logic, we should not call Pyrroglaux podarginus and the two species of Ptilopsis “scops-owls,” but rather Palau Owl and white-faced owls (Northern and Southern) , respectively. In the case of long tradition, and to respect the AOU’s role in naming North American birds, we could make exceptions, for instance by using Burrowing Owl instead of Burrowing Little-Owl, and Pearl- spotted Owl could easily be called Pearl-spotted Pygmy-Owl. Similarly, G. capense should be called Barred Pygmy-Owl to distinguish it from the North American Strix varia, which also is well-known as the Barred Owl. My point is that a dire need exists for some entity, perhaps the Raptor Research Foundation, to organize a conference in which owl researchers could agree on the English names of owls, once and for all. Illustrations in the book are detailed enough in line drawings, but the numerous color plates are very monotonous. The overall quality of the color plates does not help what is an otherwise largely enjoyable work. I am not sure whether the quality of the plates was dictated by the small page format, or whether the publisher was operating within a tight budget. It is true that owls are not too colorful, but with- out question Great Gray Owls {Strix nebulosa) have very bright yellow eyes, and Milky Eagle-Owls {Bubo lacteus) have bright fleshy eyelids. Somehow, these two species (among several others) have lost their brightness, at least in the copy that I received. May- be this was the fault of the color separation during printing. Any book of this magnitude takes years to write, hut still an author is hound to miss some infor- mation among the wealth of data from recent re- search on owls. Taxonomy and vocalizations are ex- tremely well covered throughout, but distribution and ecology are not. Coming from northern Eu- rope and living in Africa, it is readily apparent that the authors did not put adequate effort into get- 76 March 2001 Book Reviews 77 ting the distribution maps correct. Certainly Great Gray Owls in the north and Pharaoh Eagle-Owls {Bubo ascalaphus) in the south, again only as ex- amples, occur much farther south than illustrated in the maps. The latest research on the ecology of owls has been almost totally neglected. The most productive owl ecologist in recent years, Erkki Korpimaki, is mentioned only once in the text (under Teng- malm’s Owl [Aegolius funereus]), but in the Bibli- ography none of his excellent works are listed. The converse is true in my case, in which the Bibliog- raphy lists some of my works, although the citations are not mentioned in the text; e.g., Mikkola (1986) on the Barn Owl (Tyto alba). My paper on the Northern Hawk Owl {Surnia ulula) also is in the Bibliography but not in the text, and the year should be 1971 instead of 1973. The lack of infor- mation on ecology is partly understandable given that the authors concentrated on taxonomy and vocalizations, but the focus on the latter topics should have been mentioned somewhere in the book. Despite my negative remarks, I warmly recom- mend this unique taxonomic work to anyone in- terested in knowing more about diversification of owls on our planet. This book is a must for serious “owlers” and academic owl researchers from throughout the world. The book also points out issues for further study, such as the serious need for additional research on vocalizations of differ- ent owls, the lack of DNA samples for many taxa, and the paucity of data on the ecology of many species. Only after more studies have been con- ducted and synthesized can one hope to produce a truly comprehensive work on the owls of the world. — Heimo Mikkola, Institute of Applied Bio- technology, University of Kuopio, P.O. Box 1627, FIN-70211 Kuopio, Finland. Literature Cited Burton, J.A. [Ed.]. 1973. Owls of the world. Peter Lowe, London, U.K. Campbell, B. and E. Lack. [Eds.]. 1985. A dictionary of birds. T. & A.D. Poyser, Calton, U.K. DEL Hoyo, J., A. Elliott, and J. Sargatal. [Eds.]. 1999. Handbook of the birds of the world. Vol. 5. Lynx Ed- icions, Barcelona, Spain. Eck, S. and H. Busse. 1973. Eulen Die rezenten und fos- silen Formen: Aves, Strigidae. Ziemsen Verlag, Witten- berg-Lutherstadt, Germany. Mikkola, H. 1971. Znr Ernahrung der Sperbereule {Sur- nia ulula) zur Brutzeit. Angew. Ornithol. 3:133-141. . 1986. Barn Owl Tyto alba in Bali. Kukila 2:95. / Raptor Res. 35(1) : 77-78 © 2001 The Raptor Research Foundation, Inc. First Symposium on Steller’s and White-tailed Sea-Eagles in East Asia. Edited by Mutsuyuki Ueta and Michael J. McGrady. 2000. Wild Bird Society of Japan, Tokyo, Japan. 127 pp., numerous figures and tables. Paper, $20.00. — The current offering is a proceedings of a workshop organized by the Wild Bird Society of Japan, Shiretoko Museum, and the Lead Poisoning Network of Eagles, that was held in Tokyo and Hokkaido, Japan, from 9 to 15 Feb- ruary 1999. Participants included specialists from Japan, Russia, and the United States. Despite its title, the proceedings focuses clearly on the Stell- er’s Sea-Eagle {Haliaeetus pelagicus) , the largest and one of the least-studied of the world’s eight Hal- iaeetus eagles. The species, described by Pallas in 1811, was named in honor of its collector, arctic naturalist Georg Wilhelm Steller, who considered it to be “bold,” “cunning,” and of “savage dispo- sition.” The Steller’s Sea-Eagle’s massive nature, unmistakable plumage, and oversized yellow bill have long attracted the admiration of raptor biol- ogists. Until recently, however, the species’ rugged and remote breeding grounds in easternmost Rus- sia have made it difficult to study the bird in the field. Not surprisingly, then, this proceedings offers much in the way of new information. The work includes 11 journal-style chapters, nine of which concern Steller’s Sea-Eagles, and two of which deal with lead poisoning and chlorinated hydrocarbon contamination in White-tailed Eagles {H. albicilla) and Steller’s Sea-Eagles. Papers focus- ing only on the Steller’s Sea-Eagle include treat- ments of bill structure, molt, diet, migration, post- natal development, distribution and abundance in the Magadan and Khabarovsk districts and on Sa- khalin Island, and habitat use in Northern Okho- tia, as well as information on population trend analysis and techniques for determining sex and 78 Book Reviews VoL. 35, No. 1 age. A 95-entry reference section on Steller’s Sea- Eagles is included as an appendix. Proceedings can be notoriously uneven in plac- es, and although this is the case here, it is so only because substantial differences in the state of our knowledge regarding various aspects of the biology of Steller’s Sea-Eagles have made it necessary. In- deed, and in spite of several minor lapses, the ed- itors are to be congratulated on a job well done, as well as on a job most rapidly done. In addition to the work’s judicious editing, Michiko Shige- hara’s artwork conveys a welcome sense of place for the birds whose biology is described. Alexander Ladyguin’s chapter on bill structure in Steller’s Sea-Eagles is particularly enlightening and well written. Having described the species’ oversized bill in detail, Ladyguin goes on to discuss the function of its massive and powerful nature, arguing persuasively that both owe their origins not only to the exceptional toughness of fishes’ skin, but also to the rapidity with which the spe- cies — like Old World vultures — devours its prey while feeding within groups of a dozen or more aggressive conspecifics. Thus, the Steller’s Sea-Ea- gle is capable of consuming 900 g of fish in 3-4 min, whereas the White-tailed Eagle takes almost 18 min to accomplish this task. The complexities of molt in Steller’s Sea-Eagles — the species replaces one-half to one-third of all feathers, and one- fourth to one-third of all flight feathers annually — are capably described by Teruaki Morioka. Vladi- mir Masterov’s offering reveals that nestlings hatched early in the season grow more slowly than those that hatch later, and that nestling males reach adult size 7-10 days earlier than nestling fe- males. In two exceptionally well-written papers, Po- tapov et al. suggest that Steller’s Sea-Eagles breed almost entirely within a 100-km wide strip along the coastline of the Sea of Okhotsk, that those breeding in regions with low sloping coastlines and broad littoral zones are more tolerant of human disturbance than those breeding along more steep- ly sloped coastiines, and that diets of nesting pairs differ depending on coastline structure. In yet an- other excellent paper by the same authors, Utek- hina et al. document seasonal shifts in diet, with carrion being particularly important in spring, and fish and colonial-nesting waterbirds being more important in summer. McGrady et al.’s satellite te- lemetry studies of movements in the species detail fledgling dispersal to premigratory summering sites, as well as eventual migratory journeys to the south, the latter at rates of approximately 50 km per day. Most eagles from Amur and Magadan mi- grate down the western shorelines of the Sea of Okhotsk, whereas individuals from Kamchatka move south through the Kuril Islands. One excep- tional bird undertook a minimal overwater cross- ing of 730 km from Magadan southeast to south- ern Kamchatka. Finally, and on a more poignant note, Iwata et al. detail the extent of chlorinated hydrocarbon contamination and lead poisoning in both species of eagles. Lead poisoning, in partic- ular, appears problematic, with pellets in deer shot in wintering areas in Hokkaido, Japan, playing a major role. Overall, the work provides a thorough and re- markably up-to-date summary of state-of-the-art knowledge of the biology of Steller’s Sea-Eagles. The symposium’s organizers and proceedings’ ed- itors are to be congratulated on their efforts, which should serve the region’s conservation biologists and the world’s sea-eagle specialists for some time. This slim volume belongs on the shelves of all rap- torphiles. — Keith L. Bildstein, Hawk Mountain Sanctuary, 1700 Hawk Moiuitain Road, Kempton, PA 19529 U.S.A. J Raptm-Res. 35(1):79 © 2001 The Raptor Research Foundation, Inc. MANUSCRIPT REFEREES The following people reviewed manuscripts for the Journal of Raptor Research in 2000. Peer review plays a vital role in the publishing process and in improving the quality of the Journal. The editorial staff would like to thank the following for reviewing manuscripts this past year. The names of those who reviewed two or more manuscripts are indicated with an asterisk. D. Andersen, J. Aparicio, B. Arroyo*, J. Aviles, T. Balgooyen, R. Banks, J. Belthoff, I. Bellocq, M. Bertolotti, D. Bird, R. Bierregard*, K. Bildstein*, P. Bloom*, C. Boal*, C. Bock, P. Bohall Wood, G. Bortolotti*, T. Bosakowski, F. Bozinovic, D. Buehler, J. Bustamente*, T. Cade*, R. Cannings*, X Carpenter*, A. Carey, W. Clark*, N. Clum, H. Cofre, C, Cornelms, S. Destefano, L. Diller, J. Donazar, D. Ellis*, J. Enderson, S. England, J. Feijoo, M. Forero, E. Eorsman,J Ganey*, J. Gessaman, S. Goad Henry, M. Goldstein, T. Grubb*, R. Gutierrez, A. Harmata*,0. Hatzoffe, G. Hayward, L. Hicks, J.B. Holt, G. Holyroyd, C.S. Houston, D. Houston, G. Hunt, G. Janss, A. Jenkins, J. Jimenez, M. Kochert*, E. Korpimaki, I. Lazo, M. Lopez- Calleja, S. Lutz, R. Mannan, A. Margalida, J. Marks*, P. Marquet, M. Martell, C. Marti*, D. Martinez, J. Marzluff, R. McClelland, M. McGrady*, C. McIntyre*, B.-U. Meyburg, H. Mikkola, B. Millsap, F. Mougeot, H. Mneller*, I. Newton*, M. Pandolfi*, J. Parry-Jones, E. Pavez, V. Penteriani*, S. Petty*, J. Phillips, A Poole, P. Radley, J. Rau, M. Restani, G. Ritchison*, A. Rodriguez, E. Rodriguez Estrella*, R. Rosenfield, T. Schultz, J Schmutz, F. Sergio*, D. Serrano, S. Sheffield, J. Simonetti, D. Smith, N. Snyder, P. Stacey, D. Stahlecker, S. Swengel, J.-M. Thiollay*, K. Titus, H. Tordoff, J. Torres-Mura, A. Travaini, I. Warkinton, S. Wilbur*, J. Wiley, R. Yosef. 79 Recently published RAPTORS AT RISK Proceedings of the 5th World Conference on Birds of Prey & Owls R.D. Chancellor & B.-U. Meyburg (Eds.) 90 original papers, with diagrams, maps, drawings and photographs covering current studies on virtually every aspect of raptor biology, ecology and conservation. Size 22 X 15.5cm - 900 pages, cover in colour. Price $48.50 post free. Visa & Mastercard accepted. Orders to: robin.chancellor@virgin.net WWGBP/Hancock House P.O.B 0 X 52, Towcester NN12 7ZW, UK. email: WWGBP@aol.com internet: www.raptors-intemational.de Other books still available: Raptors in the Modern World, Eagle Studies, Raptor Conservation Today. A Telemetry Receiver Designed with The Researcher in Mind What you've been waiting for! 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 S1SOO. The features and performance of the new R*1000 pocket sized telemetry receiver will impress you, and the price will convince you. Other Features Indude: • Factory tuned to any 4HHz wide segment in the U8-174MHz Band • Very high sensitivity of -UBdBm to •ISOdBm • Illuminated display and keypad for use in tow light or darkness • User selectable scan rates from 1*30 seconds in 1 second steps • Rechargeable batteries operate the receiver for 1? hours and can be replaced with standard AA Alkaline batteries in the held. Both 12vdc and llOvac chargers are included. • 6.r (15.5cm) high. 2.6* (6.6cm) wide. 1.5* (3.8cm) deep. • 3 year warranty • 1 day delivery S695.00 Please specify desired 4MHz wide segment in the 148-174MHZ band Visit our website for complete specihcations. operating manual and information on the R'lOOO or call our toll-free number to order your receiver now. Try the New R-1000 and You'U Be Impressed! COMMUNICATIONS SPECIALISTS. INC. <26 W«t Taft Avenut • 0tan9«. CA 92866 <296 • l•7l<•998-^0^■ • Fax l-7U-97<-3<20 Entire U.S.A. (800) 854-0547 • Fax (800) 850-0547 • http://www.com-spec.com More than just Video Cameras * — V Solar cells Parabolic mics r The auto-color camera includes IR and color cameras with an automatic light- sensitive switching device. -r. 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. • BORROW 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 LGX 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 , ^ • . . , : . v- 5 15 W Voscmitc Avt-nue Mantc< a L,A Q5 Jli' C,):l 239 7460 • fax: :209;. 239 ' 57 i • c-mail- Ann'.-^peopcrp^'opU- vom Catalog pt-e;>t'rpeoplc corn J 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 Barn Owl (1). Carl D. Marti. 1992. 16 pp. Boreal Owl (63). G.D, Hayward and RH. 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. Ferruginous Hawk (172). MarcJ. Bechard and Josef K. Schmutz. 1995. 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. Clum and Tom J. Cade. 1994. 28 pp. Harris’ Hawk (146). James C. Bednarz. 1995. 24 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. Red-shouldered Hawk (107). Scott T. Crocoll. 1994. 20 pp. Red-tailed Hawk (52). C.R. Preston and R.D. Beane. 1993. 24 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. 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 first five volumes of this projected 12-volume work have been published including; Volume 2: New World Vultures to Guineafowl (1994) covering the diurnal raptors and Volume 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. Fine in slipcase. $300.00 and other editions at lesser prices. 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 1 1 7th 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. 0 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 Mumesota, St. Paul, A/TM KK1 ns TT C A Ancrnst 1 5 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.SA. Deadline: Deadline established for meeting paper abstracts. Grants^ The Stephen R. Tully 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.S.A. 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.