(ISSN 0892 101 The Journal OF Raptor Research Volume 28 December 1994 Number 4 Contents Reevaluating Delineated Bald Eagle Winter Roost Habitat in Lava Beds National Monument, California. Thomas J. Stohlgren and Chris A. Farmer 205 Dispersal of Bald Eagles Fledged in Texas. David W. Mabie, M. Todd Merendino and David H. Reid 213 Abundance and Distribution of Nesting Golden Eagles in Hudson Bay, Quebec. Francois Morneau, Serge Brodeur, Robert Decarie, Suzanne Carriere and David M. Bird 220 Blood Parasites of Raptors in Florida. Donald J. Forrester, Sam R. Telford, Jr., Garry W. Foster and Gordon F, Bennett 226 Ectoparasites of the Spotted Owl. John E. Hunter, R. J. Gutierrez, Alan B. Franklin and David Olson 232 Effects of Raptors on the Activity of Sandgrouse. Peter N. Ferns and Shelley A. Hinsley ; 236 Seasonal Abundance of Black Kites Associated with the Rubbish Dump of Madrid, Spain. Guillermo Blanco 242 Diet of Urban and Suburban Tawny Owls {Strixaluco) in the Breeding Season. Andrzej Zalewski 246 Diet Composition of the Long-eared Owl in Central Slovenia: Seasonal Variation in Prey Use. Davorin Tome 253 Short Communications Breeding Density and Brood Size of Rough-legged Hawks in Northwestern Quebec. Serge Brodeur, Francois Morneau, Robert Decarie, Juan J. Negro and David M. Bird 259 Effects of Prescribed Fires on Habitat Use by Wintering Raptors on a Texas Barrier Island Grassland. Felipe Chavez-Ramirez and Felipe G. Prieto 262 Winter Diet of Long-eared Owls {Asio otus) in the Po Plain (Northern Italy) . Paolo Galeotti and Luca Canova 265 Refinements to Selective Trapping Techniques: A Radio-controlled Bow Net and Power Snare for Bald and Golden Eagles. Ronald E. Jackman, W. Grainger Hunt, Daniel E. Driscoll and Frank J. Lapsansky 268 Letter 274 Book Reviews Edited by Jeffrey S. Marks 276 The Raptor Research Foundation, Inc. gratefully acknowledges a grant and logistical support provided by Weber State University to assist in the publication of the journal. 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 ad- dress tojim Fitzpatrick, Treasurer, 14377 117th Street South, Hastings, Minnesota 55033, U.S.A. The Journal of Raptor Research (ISSN 0892-1016) is published quarterly and available to individuals for $24.00 per year and to libraries and institutions for $30.00 per year from The Raptor Research Foundation, Inc., 14377 ll7th Street South, Hastings, Minnesota 55033, U.S.A. (Add $3 for destinations outside of the continental United States.) Second class postage paid at Hastings, Minnesota, and additional mailing of- fices. POSTMASTER: Send address changes to The Journal of Raptor Research, 14377 117th Street South, Hastings, Minnesota 55033, U.S.A. Printed by Allen Press, Inc., Lawrence, Kansas, U.S.A. Copyright 1994 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). THE JOURNAL OF RAPTOR RESEARCH A QUARTERLY PUBLICATION OF THE RAPTOR RESEARCH FOUNDATION, INC. VoL. 28 December 1994 No. 4 J Raptor Res. 28(4):205-212 © 1994 The Raptor Research Foundation, Inc. REEVALUATING DELINEATED BALD EAGLE WINTER ROOST HABITAT IN LAVA BEDS NATIONAL MONUMENT, CALIFORNIA Thomas J. Stohlgren^ National Park Service Cooperative Parks Studies Unit, University of California, Davis, CA 91616 U.S.A. Chris A. Farmer Department of Biology, University of California, Santa Barbara, CA 93106 U.S.A. Abstract. — We reevaluated bald eagle (Haliaeetus leucocephalus) winter roost habitat in the Ctddwell Butte area in southern Lava Beds National Monument, California. A gradient of five forest habitat strata was evaluated in previously delineated “primary” and “secondary” roost habitat to quantify the size and age structure of ponderosa pine (Pinus ponderosa) along the habitat gradient, and to reevaluate current and potential eagle roost habitat. Results indicated that primary roost habitat was considerably smaller in area and more concentrated in the Caldwell Butte roost area than previously reported. Forest plots centered on randomly selected, known eagle roost trees in the high basal area forest (>18 m^/ha of ponderosa pine) had significantly greater past eagle use (136.4 ± 39.7 castings/ha; mean ± 1 SE) than adjacent high basal area forest plots (1.8 ±1.2 castings/ha) whose locations were randomly selected. Previously identified primary roost areas of moderate and low Ponderosa pine basal area (<18 m^/ha) and previously identified secondary roost areas lacked signs of past eagle use. On average, the trees with the greatest diameters in these strata were >200 yr younger than known roost trees. Stratum A plots had 47 trees > 1 50-yr-old (7.47t> of the trees) compared to just 6 trees >1 50-yr-old (0.7% of the trees) in Stratum B plots. Strata C, D and E had very few trees >1 50- yr-old. We found that future bald eagle habitat in the area may be severely limited by current forest stand structures, forest dynamics, and the probability of producing old-growth trees. Key Words: bald eagle habitat', forest age structure', forest demography', Haliaeetus leucocephalus. Reevaluacion de los habitat de descanso invernal de Haliaeetus leucocephalus en el Monumento Nacional de Lava Beds, California. Resumen. — Reevaluamos el habitat de descanso invernal de Haliaeetus leucocephalus en el area de Caldwell Butte, al sur del Monumento Nacional de Lava Beds, California. Un gradiente de cinco estratos de habitat de bosque fue evaluado previamente, como habitat de descanso “primario” y “secundario,” cuantificando el tamano y la estructura de edad de Pinus ponderosa a lo largo del gradiente. Los resultados indicaron que el habitat de descanso primario fue considerablemente mas pequeno en area y mas concentrado en el area de descanso de Caldwell Butte que lo previamente informado. Parcelas centradas en el bosque y seleccionadas azarosamente, con arboles de descanso conocidos en el area basal alta del bosque (>18 m^/ha de P. ponderosa) tenian un uso pasado significativamente mayor (136.4 ± 39.7 observaciones/ha; media ± 1 E.S.) que en las parcelas adyacentes del area basal alta del bosque (1.8 ± 1.2 observaciones/ha) y cuya localizacion fue seleccionada azarosamente. Areas de descanso primarias, previamente identificadas, de moderadas y bajas areas basales de P. ponderosa (<18 m^/ha) y areas de descanso secundaria, previamente identificadas, perdieron las sehales del uso pasado por esta aguila. En promedio, los arboles con el mayor diametro en estos estratos eran 200 anos mayores que los mas jovenes. El Estrato A tenia 47 arboles sobre 150 ahos de edad (7.4%), comparado ' Current address: National Biological Survey, Natural Resource Ecology Laboratory, Colorado State University, Ft. Collins, CO 80523 U.S.A 205 206 Thomas J. Stohlgren and Chris A. Farmer VoL. 28, No. 4 con los seis arboles mayores a 150 anos (0.7%) del Estrato B de la parcela. Los Estratos C, D y E tenian muy pocos arboles mayores de 150 anos de edad. Encontramos que futures habitat de H. leucocephalus en el area pueden ser severamente limitados por la estructura de los bosques actuales, dinamica de bosques y la probabilidad de crecimiento de los arboles viejos. [Traduccion de Ivan Lazo] The exact boundaries of wildlife habitats, forest types, and ecosystems are often difficult to delineate (Kerr 1986). Yet, biologists are often asked to map primary, secondary, or potential habitat for rare and threatened, politically sensitive, or locally important species. This was the case for bald eagle (Haliaeetus leucocephalus) winter roost habitat in the Klamath Basin of north- eastern California. In the 1990s, an interagency agree- ment between the Modoc National Forest (USDA Forest Service) and Lava Beds National Monument (USD I National Park Service) required both parties to assess current and potential bald eagle winter roost habitat on their respective lands in the Caldwell/Cou- gar roost area. Some information on the roost area was available. Krauss (1977) reported that the Caldwell Butte portion of the roost area covered about 12 ha in old-growth ponderosa pine {Pinus ponderosa) forests “but the exact boundaries were not yet determined.” However, Krauss (1977) and others (Keister 1981, Sogge and Sydoriak 1990) delineated large areas in the roost site as either infrequently used secondary roost area or “heavy use daytime perching area” (“primary roost area” [Fig. 1]). Keister (1981) compared general forest stand char- acteristics of five roost areas in the Klamath Basin with the point-quarter method (Mueller-Dombois and El- lenberg 1974). He found considerable differences be- tween the Caldwell and Cougar roost stands in density (25.6 vs. 44.3 trees/ha) and number of stumps (0.3 vs. 12.2 stumps/ha), but similar mean tree diameters (50.4 vs. 56.0 cm dbh) and heights (24.6 vs. 27.6 m). How- ever, the methods used provided only wide-area aver- ages and no information on potential habitat, the forest age structure and dynamics of ponderosa pine, the principle roost tree species. Keister and Anthony (1983) reported that bald ea- gles roosted primarily in trees averaging 76.5 cm di- ameter at breast height (dbh), 24.7 m tall, and 289 yr of age on the Caldwell Butte roost area. Stohlgren (1993) systematically surveyed 4 km^ of forested areas in Lava Beds National Monument to locate and de- scribe the physical characteristics of roost trees. He also described ponderosa pine size distributions and basal area, and the distribution of areas of high (>18 m^/ ha) and low (<18 m^/ha) basal area of ponderosa pine. Areas of high basal area did not always contain eagle roost trees. Only about 32 ha of the 250 ha primary and secondary roost areas surveyed in the Caldwell Butte area contained roost trees (Stohlgren 1993). Consequently, a higher resolution analysis of forest structure was needed. The primary objectives of our study were to; (1) evaluate forest characteristics, particularly stand age structure, along a potential bald eagle habitat gradient, and (2) reevaluate previously delineated primary and secondary roost habitat to em- phasize the role of forest demography in perpetuating, and managing, bald eagle winter roosts. Study Area The Caldwell/Cougar Roost is between Caldwell and Cougar Buttes near the southeastern boundary of Lava Beds National Monument (Fig. 1). Vegetation in the area has been classified as mixed eoniferous dominated by ponderosa pine, western juniper (Jum- perus occidentalis) , white fir {Abies concolor), incense cedar {Calocedrus decurrens)^ and some curl-leaved mountain mahogany {Cercocarpus ledifolius) (Erhard 1979). The understory is composed primarily of bit- terbrush (Purshia tridentata) ^ currant {Ribes sp.), man- zanita {Arctostaphylos sp.), and elderberry {Sambucus sp.) (Erhard 1979). Bald eagles typically arrived in the area in Novem- ber and departed in March with peak use in January (Keister 1981). Stohlgren (1993) found castings under 58 of 103 previously used bald eagle roost trees (see Keister 1981). The annual maximum winter daily count of bald eagles in Caldwell/Cougar Roost de- clined from 278 eagles in 1983/84 to fewer than 60 eagles in 1991 (P. Toops pers. comm.). The reasons for this decline are unknown. The recent land use history of the area is complicated because the study site straddles national park and na- tional forest land. Lava Beds became a National Forest Reserve in 1925, and incidental livestock grazing oc- curred until 1933 when the area was placed under National Park Service jurisdiction. No logging has oc- curred in Lava Beds National Monument, but typical of ponderosa pine forests, there have been several fires; a 73-ha wildfire in 1911, a 480-ha wildfire in 1924 in the northern half of the primary roost habitat (in- December 1994 Reevaluating Bald Eagle Habitat 207 LAVA BEDS NATIONAL MONUMENT Figure 1. Map of Lava Beds National Monument and adjacent lands showing previously delineated areas of primary and secondary bald eagle winter roost habitat (Keister 1981, Sogge and Sydoriak 1990), and areas of productive (>18 m^/ha ponderosa pine basal area) and marginal habitat (<18 m^/ha ponderosa pine basal area; Stohlgren 1993). eluding some of Strata A, B and C described below), and two prescribed fires in 1976 (26 ha) and 1977 (64 ha) in the secondary roost habitat (including some of Strata C, D, and E described below). Twelve lightning- caused fires of <0.1 ha have occurred throughout the study area between 1934 and 1990 (Martin and John- son 1979, Olsen and Martin 1982). Methods A stratified-random study design was used with five habitat strata based on previous information on bald eagle roost tree locations (Keister 1981), and known areas of high (>18 m^/ ha), moderate (12-18 m^/ha), and low basal area (<12 m^/ ha) of ponderosa pine (Stohlgren 1993). All five study strata were selected within areas previously delineated as primary or secondary roost habitat (Krauss 1977, Sogge and Sydoriak 1990, Stohlgren 1993). Eleven variable-radius forest structure plots (described below) were established in each stratum. The forest structure plots in Stratum A were centered on randomly selected roost trees in a 0.5 km^ site of high basal area forest southwest of Caldwell Butte (Fig. 1). Another set of 11 plots were randomly selected throughout the same high basal area forest (Stratum B) irrespective of the locations of known bald eagle roost trees. Eleven forest structure plots were randomly located in a 0.5 km^ area due south of Caldwell Butte (Fig. 1) presumed to have moderate basal area (12-18 m^/ha) of ponderosa pine within the previously delineated primary roost habitat (Stratum C). The previously delineated secondary roost habitat in a 0.5 km^ southeast of Caldwell Butte (Fig. 1) was lower in basal area (<12 m^/ha ponderosa pine) than the primeuy roost habitat (Stohlgren 1993). Eleven forest structure plots also were randomly placed in each of these two strata. The strata were selected to differentiate between areas that contained at least one large (>25 cm dbh) tree (Stratum D) and areas that contained no large trees (Stratum E). Thus, the strata represented a continuum of potential roost sites, all about equidistant from the primary bald eagle feeding area (Tule Lake) 18 km north (Keister et al. 1987). Variable radius circular plots were randomly located in 208 Thomas J. Stohlgren and Chris A. Farmer VoL. 28, No. 4 Figure 2. Relationship of mean number of castings per tree and tree age class for 43 trees with castings. Number of trees by age class in parentheses. Vertical bars are standard errors of mean castings per tree. each stratum to determine the size and age structure of pon- derosa pine. Within a radius of 12.6 m of each plot center (0.05 ha), small ponderosa pines (< 1 .4 m tall) were measured in height (to the nearest centimeter) and diameter at ground level (to the nearest mm), and the number of branch whorls counted. Other tree species <1.4 m tall in the 0.05-ha plot were tallied. Within a 17.6-m radius of each plot center (0.1 ha), all trees > 1 .4 m tall were measured (dbh in centimeters); assigned a code for species, height class, and canopy structure; and the number of castings were tallied. Height classes were <10 m, 11-20 m, 21-30 m, and >30 m. Structure classes were based on the tree canopy and crown classification system of Keen (1943) as modified by Keister and Anthony (1983) and Stohlgren (1993). To provide more information on local eagle use (i.e., castings), all eagle roost trees within a 28.2 m radius of the plot center (0.25 ha) were recorded similarly. Also, a 100% survey of previously used or currently used roost trees was conducted throughout the Caldwell Butte roost area. The age of each tree with the greatest dbh on each 0.1- ha plot was determined by coring (two perpendicular cores per tree) near ground level and carefully counting the annual rings. Tree ages were estimated only to the nearest 20-yr interval and no attempt was made to cross-date rings or estimate missing or multiple rings. Age/diameter relation- ships were determined within each stratum (except for Stra- tum E) for a minimum of 30 randomly selected ponderosa pines. Standard linear regression (SAS 1988, Neter et al. 1990) was used to relate diameter to age for ponderosa pine in each stratum. Log transformations were used on all data to nor- malize the highly skewed distributions. The residuals of each regression were graphed and analyzed to ensure that the error terms conformed to the assumption of normality and that the linear model chosen was indeed the best fit. If the resulting regression equations were highly significant (P < 0.0001), the regression equations for each stratum were applied to the other uncored trees to estimate their ages. To reduce the error associated with estimating ages of individual trees, broad 50- yr age classes were used to display the age distribution of ponderosa pine in each stratum. Trees with estimated ages >275-yr-old were lumped into a single class. Estimating the ages of ponderosa pine seedlings (trees <1.4 m tall) was the focus of another study in the same area and all seedlings were found to be <50-yr-old (C. Farmer unpubl.). The number of trees (and basal area) of each species was summed to assess the species composition of the strata. Two methods were used to compare forest structure characteristics (i.e., tree densities by size class, basal area per plot, largest tree dbh, largest tree age, and number of castings/ha) among strata. Standard analysis of variance was performed on log- transformed data. Since castings were found only on two strata and the data were not normally distributed, the non- parametric Wilcoxon rank-sum test was used (SAS 1988) Where the AN OVA revealed a significant difference (at a = 0.05) among groups (i.e., strata), Ryan-Einot-Gabriel- Welsch (SAS 1988) multiple range tests were used to detect significance differences (a = 0.05) among means. Results Ponderosa pine was the dominant tree species in Strata A, B, C and D (>92% of the basal area; Table 1). Western juniper was the dominant tree in Stratum E (69% of the basal area) and total basal area was far less than in the other strata. Total basal area of Stratum C, the presumed moderate basal area site (Fig. 1), was greater than the 12-18 m^/ha anticipated. Thus, pon- derosa pine basal area in Stratum C was similar to Strata A and B (high basal area stands, and somewhat similar to Stratum D that included clumps of trees in the low basal area stand in the secondary roost habitat (Fig. 2). All variables tested were significantly different among strata (i.e., at least one stratum was different from the others; Table 2). However, means of many stand struc- ture variables were not significantly different among strata due to the variability within strata. For example, the mean density of trees was not significantly different among Strata A, B, C and D for trees in the <25 cm dbh and 25-50 cm dbh classes. Stratum A had a sig- nificantly greater density of >50 cm dbh trees than Strata B, C, D and E but the mean basal area of trees was not significantly different among Strata A, B, C and D (Table 2). Several significant differences among strata were found when the largest diameter trees in each plot were compared in size and age. Largest tree diameters in Stratum A were significantly larger than the di- ameters in all other strata (Table 2). The mean di- ameter of the largest trees in Stratum B was signifi- cantly greater than that for the largest trees in Strata D and E. The mean age of the largest diameter trees on each plot also varied among strata (Table 2). The largest December 1994 Reevaluating Bald Eagle Habitat 207 LAVA BEDS NATIONAL MONUMENT Figure 1. Map of Lava Beds National Monument and adjacent lands showing previously delineated areas of primary and secondary bald eagle winter roost habitat (Keister 1981, Sogge and Sydoriak 1990), and areas of productive (>18 m^/ha ponderosa pine basal area) and marginal habitat (<18 m^/ha ponderosa pine basal area; Stohlgren 1993). eluding some of Strata A, B and C described below), and two prescribed fires in 1976 (26 ha) and 1977 (64 ha) in the secondary roost habitat (including some of Strata G, D, and E described below). Twelve lightning- caused fires of <0.1 ha have occurred throughout the study area between 1934 and 1990 (Martin and John- son 1979, Olsen and Martin 1982). Methods A stratihed-random study design was used with five habitat strata based on previous information on bald ta^t roost tree locations (Keister 1981), and known areas of high (>18 w?/ ha), moderate (12-18 m^/ha), and low basal area (<12 m^/ ha) of ponderosa pine (Stohlgren 1993). All five study strata were selected within areas previously delineated as primary or secondary roost habitat (Krauss 1977, Sogge md Sydoriak 1990, Stohlgren 1993). Eleven variable-radius forest structure plots (described below) were established in each stratum. The forest structure plots in Stratum A were centered on randomly selected roost trees in a 0.5 km^ site of high basal area forest southwest of Caldwell Butte (Fig. 1). Another set of 11 plots were randomly selected throughout the same high basal area forest (Stratum B) irrespective of the locations of known bald eaigle roost trees. Eleven forest structure plots were randomly located in a 0.5 km^ tirea due south of Caldwell Butte (Fig. 1) presumed to have moderate basal area (12-18 m^/ha) of ponderosa pine within the previously delineated primary roost habitat (Stratum C). The previously delineated semndary roost habitat in a 0.5 km^ southeast of Caldwell Butte (Fig. 1) was lower in basal area (<12 m^/ha ponderosa pine) than the primary roost habitat (Stohlgren 1993). Eleven forest structure plots also were randomly placed in each of these two strata. The strata were selected to differentiate between areas that contained at least one large (>25 cm dbh) tree (Stratum D) and areas that contained no large trees (Stratum E). Thus, the strata represented a continuum of potential roost sites, all about equidistant from the primary bzdd eagle feeding area (Tule Lake) 18 km north (Keister et al. 1987). Variable radius circular plots were randomly located in 208 Thomas J. Stohlgren and Chris A. Farmer VoL. 28, No. 4 lU UJ oc I- ■>. CO C5 z H CO < o < 111 2 AGE CLASS MIDPOINT (yrs) Figure 2. Relationship of mean number of castings per tree and tree age class for 43 trees with castings. Number of trees by age class in parentheses. Vertical bars cU'e standard errors of mean castings per tree. each stratum to determine the size and age structure of pon- derosa pine. Within a radius of 12.6 m of each plot center (0.05 ha), small ponderosa pines (< 1 .4 m tall) were measured m height (to the nearest centimeter) and diameter at ground level (to the nearest mm), and the number of branch whorls counted. Other tree species <1.4 m tall in the 0.05-ha plot were tallied. Within a 17.6-m radius of each plot center (0.1 ha), all trees > 1 .4 m tall were measured (dbh in centimeters); assigned a code for species, height class, and canopy structure; and the number of castings were tallied. Height classes were <10 m, 11-20 m, 21-30 m, and >30 m. Structure classes were based on the tree canopy and crown classification system of Keen (1943) as modified by Keister and Anthony (1983) and Stohlgren (1993). To provide more information on local eagle use (i.e., castings), zdl eagle roost trees within a 28.2 m radius of the plot center (0.25 ha) were recorded similzu-ly. Also, a 100% survey of previously used or currently used roost trees was conducted throughout the Caldwell Butte roost area. The age of each tree with the greatest dbh on each 0.1- ha plot was determined by coring (two perpendicular cores per tree) near ground level and carefully counting the amnual rings. Tree ages were estimated only to the nearest 20-yr interval and no attempt was made to cross-date rings or estimate missing or multiple rings. Age/diameter relation- ships were determined within each stratum (except for Stra- tum E) for a minimum of 30 randomly selected ponderosa pines. Standard linear regression (SAS 1988, Neter et al. 1990) was used to relate diameter to age for ponderosa pine in each stratum. Log transformations were used on all data to nor- malize the highly skewed distributions. The residuals of each regression were graphed and analyzed to ensure that the error terms conformed to the assumption of normality and that the linear model chosen was indeed the best fit. If the resulting regression equations were highly significant (P < 0.0001), the regression equations for each stratum were applied to the other uncored trees to estimate their ages. To reduce the error associated with estimating ages of individual trees, broad 50- yr age classes were used to display the age distribution of ponderosa pine in each stratum. Trees with estimated ages >275-yr-old were lumped into a single class. Estimating the ages of ponderosa pine seedlings (trees < 1 .4 m tall) was the focus of another study in the same area and all seedlings were found to be <50-yr-old (C. Farmer unpubl.). The number of trees (and basal area) of each species was summed to assess the species composition of the strata. Two methods were used to compare forest structure characteristics (i.e., tree densities by size class, basal area per plot, largest tree dbh, largest tree age, and number of castings/ha) among strata. Standard analysis of variance was performed on log- transformed data. Since castings were found only on two strata and the data were not normally distributed, the non- parametric Wilcoxon rank-sum test was used (SAS 1988). Where the ANOVA revealed a significant difference (at a = 0.05) among groups (i.e., strata), Ryan-Einot-Gabriel- Welsch (SAS 1988) multiple range tests were used to detect significance differences (a = 0.05) among means. Results Ponderosa pine was the dominant tree species in Strata A, B, C and D (>92% of the basal area; Table 1). Western juniper was the dominant tree in Stratum E (69% of the basal area) and total basal area was far less than in the other strata. Totcil basal area of Stratum C, the presumed moderate basal area site (Fig. 1), was greater than the 12-18 m^/ha anticipated. Thus, pon- derosa pine basal area in Stratum C was similar to Strata A and B (high basal area stands, and somewhat similar to Stratum D that included clumps of trees in the low basal area stand in the secondary roost habitat (Fig. 2). All variables tested were significantly different among strata (i.e., at least one stratum was different from the others; Table 2). However, means of many stand struc- ture variables were not significantly different among strata due to the variability within strata. For example, the mean density of trees was not significantly different among Strata A, B, C and D for trees in the <25 cm dbh and 25-50 cm dbh classes. Stratum A had a sig- nificantly greater density of >50 cm dbh trees than Strata B, C, D and E but the mean basal area of trees was not significantly different among Strata A, B, C and D (Table 2). Several significant differences among strata were found when the largest diameter trees in each plot were compzired in size and age. Largest tree diameters in Stratum A were significantly larger than the di- ameters in all other strata (Table 2). The mean di- Eimeter of the largest trees in Stratum B was signifi- cantly greater than that for the largest trees in Strata D and E. The mean age of the largest diameter trees on each plot also varied among strata (Table 2). The largest December 1994 Reevaluating Bald Eagle Habitat 211 Table 4. Number of trees sampled by age class in each habitat stratum. Percent trees by strata in age class. Age Glass (yr) Strata <50 50-99 100-149 150-199 200-249 250+ Total A 453 90 37 18 11 18 627 (72.2%) (14.4%) (5.9%) (2.8%) (1.8%) (2.8%) B 692 124 17 5 1 0 839 (82.5%) (14.8%) (2.0%) (0.6%) (0.1%) C 349 139 21 1 1 0 511 (68.3%) (27.2%) (4.1%) (0.2%) (0.2%) D 197 208 17 0 0 0 422 (46.7%) (49.3%) (4.0%) E 15 13 8 1 0 0 37 (40.5%) (35.1%) (21.6%) (2.7%) ever become roost trees (i.e., >275-yr-old and attain the necessary structure). The small percentages of the trees in the 150-199 and 200-249-yr-old-age classes in Strata B, C, and D suggests that ponderosa pine survivorship may be lower in those strata than in Stra- tum A. That is, recruitment might be high (i.e., high percentages of <50-yr-old trees) but survivorship to old age may be limited, and dependent on local soil development, resistance to fire (e.g., due to size, bark thickness, or location in rocky outcrops), or other rea- sons. Except as a buffer-area to protect roost trees from unwanted disturbance from humans, it is difficult to classify the types of forests in Strata C, D, or E as primary or secondary roost habitat. Furthermore, the size and age structure of ponderosa pine in those stands suggests they should not be considered as potential roost area in a 100-yr period. Bald eagles will roost in younger trees (Fig. 2). However, they may do so only when large, old trees are in the immediate vicinity (Stratum A). We found no evidence of eagle use in younger stands in Strata C, D, and E. Anthony et al. (1982) and Keister and Anthony (1983) also stressed the need for large (old) trees in multi-layered (uneven-aged) stands for com- munal roosting by bald eagles in the Klamath Basin. Managing bald eagle winter roost habitat requires detailed, quantitative information of forest age struc- ture and stand dynamics. Bald eagle winter roost hab- itat can be defined quantitatively at the tree level and defined qualitatively at the stand level (DellaSala et al. 1987). Bald eagles in the Caldwell Butte area pre- ferred roosting in large ponderosa pines averaging > 80 cm dbh (Keister and Anthony 1983, Stohlgren 1993) many of which are >200-yr-old (Fig. 2). A strict def- inition of primary roost habitat might include only these trees that dot the landscape. At the stand level, a multi-tiered canopy (uneven-aged stand) containing some >200-yr-old ponderosa pines (Table 4; Stratum A) best defines primary roost habitat. Because these primary resources may be clumped or scattered in distribution or concentrated in space when viewed at the landscape level, it may be misleading to delineate primary roost habitat as a large, continuous area. This study showed that much of the area previously delin- eated as primary roost habitat (Krauss 1977, Keister 1981, and Sogge and Sydoriak 1990) contained no sign of recent eagle use and no large, old trees preferred by eagles for roosting (Tables 2 and 4; Fig. 2). However, there may be other reasons for delineating and protecting larger areas around scattered primary resources (i.e., any previously used roost trees). Buffer zones might be established to protect the eagles from human disturbance, particularly while the birds are roosting. Alternatively, buffer areas may include sec- ondary or potential habitat. Secondary habitat might include areas adjacent to roost trees having moderately large, old trees that could serve as overflow areas when eagle numbers are high. Using this definition, the areas we identified in Strata D and E should probably not be considered as secondary roost habitat. Defining areas as potential roost habitat should in- corporate aspects of eagle use (behavior and population size), forest development (tree population dynamics), and time. The communal nature of winter roosting in bald eagles influences eagle use at the tree level while eagle population numbers influence eagle use at the stand, landscape and regional levels (DellaSala et al. 1987, Keister et al., 1987). The differences in age 212 Thomas J. Stohlgren and Chris A. Farmer VoL. 28, No. 4 structure of ponderosa pine among habitat strata (Ta- ble 4) suggest strongly that future bald eagle habitat in the area may be severely limited by current forest stand structures, forest dynamics, and the probability of producing old-growth trees. That is, we are not confident that Strata B, C, D and E areas can produce significant numbers of large, old ponderosa pines in the next 100 yr. Meanwhile, existing roost trees may be threatened by wildfire, drought, and pathogens. Maintaining, perpetuating, and protecting primary bald eagle roost habitat requires an understanding of forest development and tree population dynamics as well as changing bald eagle use patterns, population sizes, and behavior. The gradient approach used here to assess bald eagle winter roost habitat could be used to delineate habitat gradients for other species. Acknowledgments We thank C. Raider for assisting in the extensive field work and data entry, and P. Toops, B. Stouffle and other staff at Lava Beds National Monument for their assistance throughout the project. The National Park Service provided the funding for the study. Literature Cited Anthony, R.G., R.L. Knight, G.T. Allen, B.R. Mc- Clelland AND J.I. Hodges. 1982. Habitat use by nesting and roosting bald eagles in the Pacific Northwest. Trans. North. Am. Wildl. Nat. Resour. Conf. 47:332-342. DellaSala, D.A., R.G. Anthony and T.A. Spies. 1987. A habitat management plan for bald eagle Haliaeetus leucocephalus) communal roost sites at the Bear Valley National Wildlife Refuge, Oregon. Rep. to USDA Fish Wildl. Serv., Klamath Basin National Wildlife Refuge, Contract No. 14-16-0009-1533. RWO #7. Klamath Falls, OR U.S.A. Erhard, D.H. 1979. Plant communities and habitat types in the Lava Beds National Monument, California. M.S. thesis. Oregon State Univ., Corvallis, OR U.S.A. Keen, F.P. 1943. Ponderosa pine tree classes redefined. J. For. 41:249-253. Keister, G.P., Jr. 1981. Characteristics of winter roosts and population of bald eagles in the Klamath Basin. M.S. thesis. Oregon State Univ., Corvallis, OR U.S.A. and R.G. Anthony. 1983. Characteristics of bald eagle communal roosts in the Klamath Basin, Oregon and California. J. Wildl. Manage. 47:1072-1079. , R.G. Anthony and E.J. O’Neill. 1987. Use of communal roosts and foraging areas by bald eagles win- tering in the Klamath Basin. J. Wildl. Manage. 51:415- 420. Kerr, R.M. 1986. Habitat mapping. Pages 49-72 in A.Y. Cooperrider, R.J. Boyd and H.R. Stuart [Eds.], Inven- tory and monitoring of wildlife habitat. USDI Bur. Land Mdinage., Denver CO U.S.A. Krauss, G.D. 1977. A report on the 1976-1977 Klamath Basin bald eagle winter use area investigation. USDA For. Serv., Klamath National Forest. Tule Lake, CA U.S.A. Martin, R.E. and A.H. Johnson. 1979. Fire manage- ment of Lava Beds National Monument. Pages 1209- 1218 in R.M. Linn [Ed.], Proceedings of the first con- ference on scientific research in the national parks. Vol. II. USDI Natl. Park Serv. and Amer. Inst. Biol. Set., Washington, DC U.S.A. Mueller- Dombois, D. and H. Ellenberg. 1974. Aims and methods of vegetation ecology. John Wiley and Sons, New York, NY U.S.A. Neter, j., W. Wasserman and M.H. Kutner. 1990 Applied linear statistical models. Irwin, Inc., Boston, MA U.S.A. Olsen, C. and R.E. Martin. 1982. Recommended pro- gram for fire management at Lava Beds National Mon- ument, California. USDI Natl. Park Serv. Western Re- gion Final Rep., Contract CX-8000-9-0015. Tule Lake, CA US A. SAS. 1988. SAS for personal computer. Release 6.03 edi- tion. SAS Institute, Inc. Cary, NC U.S.A. SOGGE, M. AND C. Sydoriak. 1990. Caldwell/Cougar bald eagle winter roost management plan. Lava Beds National Monument and Modoc National Forest. Tule Lake, CA U.S.A. Stohlgren, T.J. 1993. Bald eagle winter roost charac- teristics in Lava Beds National Monument, California Northwest Sci. 67:44-54. Received 8 February 1994; accepted 26 May 1994 J Raptor Res. 28(4):213-219 © 1994 The Raptor Research Foundation, Inc. DISPERSAL OF BALD EAGLES FLEDGED IN TEXAS David W. Mabie Texas Parks and Wildlife Department, 715 S. Highway 35, Rockport, TX 78382 U.S.A. M. Todd Merendino Texas Parks and Wildlife Department, Matagorda County Courthouse, Bay City, TX 77414 U.S.A. David H. Reid Texas Parks and Wildlife Department, Rt. 3 Box 247, Bay City, TX 77414 U.S.A. Abstract. — One hundred and thirty-eight bald eagle {Haliaeetus leucocephalus) nestlings were banded and color-marked in Texas, 1985-91. Survival of marked eaglets to fledging was 97%. Three recoveries and 61 sightings were verified from 1985-93. Recoveries were from Texas and southeast Louisiana. Beginning in April, most marked eagles showed a gradual spring-summer northward migration that was centered between the Rocky Mountains and the Mississippi River Valley, extending north into Canada. However, two sightings occurred on the Atlantic Coast, and two sightings occurred west of the Rocky Mountains. Forty-six of the 138 marked eaglets have attained breeding age. Nine of these eagles are known to be nesting in Texas and two others have been reported as nesting in Arizona and Mexico. Based on sight records we feel that bald eagles fledged in Texas may enter breeding populations throughout the southern United States. Key Words: bald eagle; banding; color-marking; Haliaeetus leucocephalus; migration; survival; Texas. Dispersion de individuos volantones de Haliaeetus leucocephalus en Texas Resumen. — Entre 1985 y 1991 se marcaron 138 polluelos de la especie Haliaeetus leucocephalus en el Estado de Texas. El 97% de los polluelos alcanzo el estado de volanton. Tres recuperaciones y 61 avistamientos se registraron entre 1985 y 1993. Las recuperaciones se hicieron en Texas y al sudeste de Louisiana. Comenzando en abril (primavera-verano), la mayoria de las aguilas marcadas mostraron una migracion gradual hacia el norte, centrada entre las Montanas Rocosas y el Valle del Rio Mississippi, extendiendose por el norte hasta el interior de Canada. Sin embargo, dos avistamientos ocurrieron en la Costa Atlantica y otros dos al oeste de las Montanas Rocosas. Cuarenta y seis de los 138 individuos han alcanzado la edad reproductiva. Nueve de estas aguilas han nidificado en Texas y otras dos han sido reportadas nidificando en Arizona y Mexico. Basados en los registros de avistamiento, creemos que los volantones de Haliaeetus leucocephalus pueden ingresar a la poblacion reproductiva desde el sur de los Estados Unidos. [Traduccion de Ivan Lazo] Most migration studies of bald eagles {Haliaeetus leucocephalus) have documented southward post- breeding movement from northern nesting territories to southern wintering grounds (Dunstan 1973, Reese 1973, Dunstan et al. 1975, Postupalsky 1976, Ger- rard et al. 1978, Griffin et al. 1980); however, eagles nesting at southern latitudes of the United States exhibit northward migration during summer (Bro- ley 1947, Hunt et al. 1992). Bald eagles are reported as nonmigratory in some areas (Stalmaster 1987). From 1975-91, the number of active bald eagle nests increased from seven to 39 in Texas. Infor- mation on eaglet survival to fledging, fledging dis- persal, and recruitment into the breeding population was necessary to properly manage this increasing bald eagle population. Such data were also needed to document status (threatened/endangered) for the bald eagle recovery plan. This study presents mi- gratory movements, survival of eaglets to fledging, and recruitment of eagles banded and color-marked as nestlings in the southeastern area of Texas, Methods Fixed-wing aerial surveys have been conducted to mon- itor nesting activities of bald eagles in Texas since 1971. These surveys determined number of nests, number of active nests, and number of fledged young. From 1985- 91 , we located and climbed 85 eagle nest trees, representing 26 nesting territories in 16 counties. One hundred and 213 214 Mabie et al. VoL. 28, No. 4 thirty-eight eaglets, age 6-11 wk, were banded and color- marked (Fig. 1). All eaglets were leg banded with a U.S. Fish and Wildlife Service pop-rivet band and a pop-rivet plastic color band on the opposite leg. Red, black, green, and white bands, with each color representing a specific year, were used from 1985-89. Anodized aluminum bands were used in 1990 and 1991 due to reported high loss rates (94%) of plastic leg bands (C. Todd unpubl. data). Vinyl patagial wing markers (Herculite 20) were attached to both wings of all eaglets (Kochert 1973, Gerrard et al. 1974, 1992). Yellow patagial markers, which designated an eagle as being from the southeastern region, were used on the right wing (D. Bystrak pers. comm.). Alpha nu- meric codes were painted with black NAZDAR vinyl ink on yellow patagial markers to individually mark each ea- glet. Blue patagial markers with a large yellow dot were used on the left wing to designate a bird as being fledged in Texas. Kochert et al. (1983) indicated that patagial markers did not affect breeding success of golden eagle {Aquila chrysaetos), red-tailed hawk {Buteo jamaicensis) and common raven {Corvus corax). Letters and requests for information on observed or recovered marked eagles were sent to all U.S. state and Canadian province nongame/endangered species divisions following each banding season. Reports or sightings of wing-marked eagles were evaluated on the basis of cor- respondence or telephone conversations with the observer. Consecutive sightings of a single marked bird within one general area and in the same year were recorded as one sighting. Marked nesting birds observed in consecutive years at nest sites were recorded each year as one sighting. Survival to fledging of color-marked eaglets was deter- mined through fixed-wing aircraft surveys (Nesbitt et al. 1988) and ground checks of all nest sites. Fledging success was determined when marked birds were observed in flight or away from the nest tree. Recruitment of marked eaglets into the breeding population was determined when marked eagles were observed incubating or brooding during aerial and ground surveys. Results and Discussion Eaglet Survival to Fledging. Survival of color- marked eaglets to fledging was 97%, with produc- tivity from those nests calculated at 1.6 eaglets per nest, 1985-91. Productivity values were similar to those reported in Florida (Broley 1947, McEwan and Hirth 1979), and are as high as any recorded values (Sprunt et al. 1973). The increasing bald eagle population in Texas is likely tied to the high fledging success and high fledgling survival of this population. Four marked eaglets died during this study. Of these, two were killed by bobcats (Felis rufus), and two were found <100 m from the nest tree (cause of death was unknown). There was no indication that our banding operation caused this mortality. Similarly, other studies (Broley 1947, Grier 1969, Fraser et al. 1985) report little mortality due to banding activities. Band Recoveries. Three band recoveries were reported from 1985-93 (Fig. 2). A juvenile female found dead on 1 June 1987 in Red River County, Texas, was banded and color-marked on 2 April 1987, 525 km to the south, in Brazoria County, Texas; the bird had fledged on 29 April 1987. Only bones and feathers were found and the cause of death was unknown. A 2-yr-old male banded on 31 March 1988 in Matagorda County, Texas, was hit and injured by a vehicle on 17 March 1990 in LaFourche Parish, Louisiana, approximately 600 km from the nest site. The third recovery was of a 3-yr-old banded on 16 April 1990 in Colorado County that was shot and killed on 9 March 1993 in Liberty County, Texas, approximately 177 km northeast of the nest site. Sightings. Sixty-one sightings (29 adult, 32 im- mature) of color-marked eagles were verified from 1985-93. Forty-one sightings were reported from within state (Fig. 3), with 70% occurring from No- vember to March. A marked eagle observed in Cam- eron County is the most southerly documentation of bald eagles in Texas. Twenty sightings were outside of Texas (Fig. 2), with 74% occurring from May to August. Observations of color-marked eagles indi- cated that a gradual spring-summer northward mi- gration begins in April. Marked birds moved across a broad region from the Rocky Mountains to the Mississippi River and north into Canada (Fig. 2). Two sightings occurred on the Atlantic Coast (South Carolina, New York), one in Mexico (Sonora), and one in Arizona (Fig. 2). In 1989, a nesting eagle with a yellow wing tag and black lettering was re- ported in Sonora, Mexico. The alpha-numeric code was not fully legible, but the first two numbers in- dicated that the bird had been banded as an eaglet in Texas in 1985. The eagle sighted in Arizona has recently been confirmed as nesting there. We suspect that bald eagles fledged in Texas may enter breeding populations throughout the bald eagles’ southern breeding range. Most color-marked eagles were observed only on one day. However, some eagles were observed on consecutive days. For example, a color-marked bird was observed for 10 consecutive days in Nebraska, another bird was observed for 19 consecutive days at a catfish farm in Arkansas, and another bird was observed for 30 consecutive days in Cameron Coun- ty, Texas. December 1994 Bald Eagle Dispersal 215 Figure 1. Locations of 138 bald eagles banded and color-marked as nestlings in Texas, 1985-91. 216 Mabie et al. VoL. 28, No. 4 BC VvA C'l? NT AB MB SK r.'K Vv- ■•JL- ;D MN lA Wl h r I V r-iv 1 1 II \N V ‘ kS 1 L • 9 NM M( A \ TX TN A! NF PQ NB V I NY PA M*-- A > Rl Vv V . > /A ' CT NJ DE MD -yrs M''. ,Y"a ■ \ Banding Location MFXIO' NS A Recovery • Sightings Figure 2. Location of out-of-state sightings and recoveries of bald eagles banded and color-marked as nestlings in Texas, 1985-93. December 1994 Bald Eagle Dispersal 217 Jut Wehi Nuecpr. I Trini->' Afvg^iini '™ r^bmc Waikef Polk Tyi«^ San *»Jacirif Moolgonery -ibcMv Mardm 1 JeV'crsc ■ ;>j"nDer3 ► ® Wharton 0 i fort Rer (b -2 3 ■CL a; II CL CM s 3 I " 'S Cm lc o s' .y g s i " 8 S 3 o a 3 Cl to 3 S g II § o 3 CM §. - P ■S ^ ^ II 3 LJ i: 3 II £ *o .L) X •-T >4 £ II II ^ « a E 3 ciT § Q N 3 JS O 3 O ^ K "i cw II ^ c I I " tb c ^ Pm "K* ^ £ S 3 ^ fc k 2 ^3 o u V u fl a ‘u V B < Lh O u V u ^ i a a December 1994 Raptor Blood Parasites 229 Table 2. Prevalences of the common genera of blood parasites in raptors from North America and the Neotropical region. Number OF Birds Examined Per- cent Posi- TIVE Percent Positive pa jjb LC FOR ■yd Source Strigiformes North America 118 51 7 21 37 3 Greiner et al. (1975) Maryland-New Jersey 28 18 4 7 7 4 Williams and Bennett (1978) Florida 54 63 17 56 2 2 This study Louisiana 34 68 0 73 12 0 Olsen and Gaunt (1985) Neotropics 44 25 9 18 White et al. (1978) Falconiformes North America 309 40 3 23 21 3 Greiner et al. (1975) Maryland-New Jersey 33 33 0 33 9 0 Williams and Bennett (1978) New Jersey area^ 249 59 1 22 43 1 Kirkpatrick and Lauer (1985) Florida 93 33 11 26 8 0 This study Louisiana 21 24 0 14 9 0 Olsen and Gaunt (1985) Neotropics 119 30 8 19 1 2 White et al. (1978) Plasmodium. ^ Haemoproteus. Leucocytozoon. ** Trypanosoma. ^ Prevalences not given by author, f New Jersey, Pennsylvania, Delaware, and Virginia. H. noctuae have been reported previously from east- ern screech-owls (Bishop and Bennett 1989). Trypanosoma confusum is known from several stri- giforms (Bennett et al. 1982, Bishop and Bennett 1992), but has not been reported previously from the barred owl. An unidentified trypanosome found by Wetmore (1941) in the barred owl was probably T. confusum. Although Baker (1976) has defined the trypanosome from the New World as T. confusum, the parasite is morphologically indistinguishable from Trypanosoma avium of Old World strigids. Bennett (1961, 1970) showed that the trypanosome origi- nating from the northern saw-whet owl {Aegolius acadicus) could be transmitted via the mosquito Aedes aegypti to a number of different avian species of several orders and families, thus indicating a total lack of host specificity. Measurements and derived indices from trypanosomes from Old and New World boreal owls {Aegolius funereus) are identical to those from the northern saw-whet owl and the two species of trypanosomes are likely to be synonymous (G. Bennett unpubl. data). Falconiforms were parasitized more commonly by Leucocytozoon than were strigiforms. This may re- flect a greater opportunity for contact with the si- muliid vectors by falconiforms because of their broader ranges, in comparison to the more sedentary strigiforms. It is possible also that strigiform behav- ior patterns (foraging at night and using secluded, shady perches during day) expose them more fre- quently to a variety of vector species (and hence a larger variety of hemoparasites) than do those of falconiforms (daytime activity period and elevated perches at night). The overall prevalence of Plasmodium species in both strigiforms and falconiforms in Florida was greater than that recorded in other comparable sur- veys (Table 2). These prevalences probably would have been even higher if isodiagnostic techniques had been used since Plasmodium infections in birds are known to virtually disappear from the circulating blood when infections become chronic (Herman et al. 1966). Haemoproteus species were also more com- mon in strigiforms in Florida than in other areas except Louisiana (Olsen and Gaunt 1985), but in falconiforms their prevalence in Florida was similar to the overall estimate for North America (Greiner et al, 1975) and the Neotropics (White et al. 1978). Leucocytozoon species, however, had a lower prev- alence in strigiforms from Florida than generally in 230 Forrester et al. VoL. 28, No. 4 North America (Greiner et al. 1975) or on the east- ern seaboard (Williams and Bennett 1978). In fal- coniforms, Leucocyte zoon species had a lower prev- alence in Florida than reported in the North Amer- ican survey (Greiner et al. 1975) and the 1985 New Jersey survey (Kirkpatrick and Lauer 1985), but was comparable to that found in the 1978 Mary- land-New Jersey survey (Williams and Bennett 1978) and the Louisiana survey (Olsen and Gaunt 1985). Leucocytozoon species are present in the Neo- tropics (White et al. 1978), but there is little infor- mation on prevalences. The data on Trypanosoma from all areas surveyed may be misleading because of the inefficiency in detecting infections by using blood films; more trypanosome infections would have been detected if bone marrow had been examined and if culture methods had been used. Acknowledgments We thank Noha Abou-Madi and Ramiro Isaza for pro- viding access to raptors in the Wildlife Clinic of the Vet- erinary Medical Teaching Hospital at the University of Florida. Vicki L. Clyde prepared some of the blood films and Edward B. Rabin furnished us with slides from the Clinical Pathology files. We appreciate their cooperation. Dyanne D. Singler gave us access to raptors at The Con- servancy in Naples, Florida. Petra B. Wood supplied us with slides from nestling bald eagles. Ellis C. Greiner generously permitted us to include some of his material from bald eagles in our sample, and both he and Martin D. Young reviewed the manuscript. Marilyn G. Spalding also examined an early draft of the manuscript and offered some helpful suggestions. This research was supported by contracts from the Florida Game and Fresh Water Fish Commission’s Federal Aid to Wildlife Restoration Pro- gram, Florida Pittman-Robertson Project W-41 to DJF and by an operating grant from the Natural Sciences and Engineering Research Council of Canada to GFB. This IS Florida Agricultural Experiment Station’s Journal Series No. R-03155. Literature Cited Baker, J.R. 1976. Biology of the trypanosomes of birds. Pages 131-174 in W.H.R. Lumsden and D.A. Evans [Eds.], Biology of the Kinetoplastida. Vol. 1. Academic Press, New York, NY U.S.A. Bennett, G.F. 1961. On the specificity and transmis- sion of some avian trypanosomes. Can. J. Zool. 39:17- 33. . 1970. Trypanosoma avium Danilewsky in the avian host. Can. J. Zool. 48:803-807. AND M.A. Peirce. 1988. Morphological form in the avian Haemoproteidae and an annotated check- list of the genus Haemoproteus Kruse, 1890. J. Nat Hist. 22:1683-1696. , M. Whiteway and C. Woodworth- Lynas 1982. A host-parasite catalogue of the avian haema- tozoa. Mem. Univ. Newfoundland Occ. Pap. Biol. 5, St. John’s, Newfoundland, Canada. , R.A. Earle, M.A. Peirce, F.W. Huchzermeyer AND D. Squires- Parsons. 1991. Avian Leucocyto- zoidae: the leucocytozoids of the Phasianidae sensu lato J. Nat. Hist. 25:1407-1428. Bishop, M.A. and G.F. Bennett. 1989. The haemo- proteids of the avian order Strigiformes. Can. J. Zool 67:2676-2684. AND . 1992. Host-parasite catalogue of the avian haematozoa. Supplement 1 and bibliography of the avian blood-inhabiting haematozoa. Supplement 2. Mem. Univ. Newfoundland Occ. Pap. Biol. 15, St John’s, Newfoundland, Canada. Dooris, G.M., P.M. Dooris and W.D. Courser. 1981 The absence of hematozoa in burrowing owls of the Tampa Bay area, Florida. Fla. Field Nat. 9:9. Forrester, D.J., E.C. Greiner, G.F. Bennett and M.K. Kigaye. 1977. Avian Haemoproteidae. 7. A review of the haemoproteids of the family Ciconiidae (storks) and descriptions of Haemoproteus brodkorbi sp. nov. and H. peircei sp. nov. Can. f. Zool. 55:1268-1274. Greiner, E.C. and A. A. Kocan. 1977. Leucocytozoon (Haemosporidia; Leucocytozoidae) of the Falconi- formes. Can. f. Zool. 55:761-770. , D.J. Black and W.O. Iverson. 1981. Plas- modium in a bald eagle (Haliaeetus leucocephalus) in Florida, f. Wildl. Dis. 17:555-558. , G.F. Bennett, E.M. White and R.F. Coombs 1975. Distribution of the avian hematozoa of North America. Can. f. Zool. 53:1762-1787. Herman, C.M., J.O. Knisley and E.L. Snyder. 1966. Subinoculation as a technique in the diagnosis of avian Plasmodium. Avian Dis. 10:541-547. Kimsey, R.B. 1992. Host association and the capacity of sand flies as vectors of lizard malaria. Int. f. Parasitol 22:657-664. Kirkpatrick, C.E. AND D.M. Lauer. 1985. Hematozoa of raptors from southern New Jersey and adjacent areas, f. Wildl. Dis. 21:1-6. Olsen, G.H. and S.D. Gaunt. 1985. Effect of hemo- protozoal infections on rehabilitation of wild raptors /. Am. Vet. Med. Assoc. 187:1204-1205. Peirce, M.A., G.F. Bennett and M. Bishop. 1990 The haemoproteids of the avian order Falconiformes /. Nat. Hist. 24:1091-1100- Sykes, P.W., jR AND D.J. Forrester. 1983. Parasites of the snail kite in Florida and summary of those re- ported for the species. Fla. Field Nat. 11:111-116. Telford, S.R., Jr. 1988. A contribution to the system- atics of the reptilian malaria parasites, family Plas- December 1994 Raptor Blood Parasites 231 modiidae (Apicomplexa: Haemospororina). Bull. Fla. State Mus. Biol. Set. 34:65-96. AND D.J. Forrester. 1992. Morphometric comparisons of the Plasmodium (Novyella) species re- ported from North American birds, with comments on a species from the barred owl (Strix varia Barton). System. Parasitol. 22:17-24. , M.G. Spalding and D.J. Forrester. 1991. Hemoparasites of wading birds (Ciconiiformes) in Florida. Can. J. Zool. 70:1397-1408. Wetmore, P.W. 1941. Blood parasites of birds of the District of Columbia and Patuxent Research Refuge vicinity. /. Parasitol. 27:379-393. White, E.M., E.C. Greiner, G.F. Bennett and G.M. Herman. 1978. Distribution of the hematozoa of neotropical birds. Rev. Biol. Prop. 26:43-102. Williams, N. A. AND G.F. Bennett. 1978. Hematozoa of some birds of New Jersey and Maryland. Can. J Zool. 56:596-603. Received 10 March 1994; accepted 2 July 1994 J. Raptor Res. 28(4):232-235 © 1994 The Raptor Research Foundation, Inc. ECTOPARASITES OF THE SPOTTED OWL John E. Hunter, RJ. Gutierrez, Alan B. Franklin^ AND David Olson Department of Wildlife, Humboldt State University, Areata, CA 95521 U.S.A. Abstract. — We conducted a survey of spotted owl (Strix occidentalis) ectoparasite richness by examining live and museum specimens of the three subspecies of spotted owl. Seven ectoparasite species from five arthropod orders were collected. A tick (Dermacentor occidentalis) , a mite {Euschoengastia sp.), and a flea iPpisodasys vesperalis) were collected, but were presumed to be accidental strays from prey. Strigiphilus lice were found on all the subspecies of spotted owl, while the louse Kurodaia magna was only collected from the northern spotted owl {S. o. caurina). The hippoboscid fly Icosta americana was found on the California spotted owl {S. o. occidentalis)', this species had previously been well documented in the northern subspecies. The only hippoboscid fly found infesting the Mexican spotted owl {S. o. lucida) was Ornithoica vicina. Key Words: ectoparasites; Icosta americana; Kurodaia magna; Ornithoica vicina; spotted owl; Strigiphilus syrnii; Strix occidentalis. Ectoparasites de Strix occidentalis Resumen. — Realizamos un estudio sobre la riqueza de ectoparasites de tres subespecies de Strix occidentalis, a traves del examen de especimenes vivos y de museo. Se colectaron siete especies de ectoparasites correspondientes a cinco ordenes de artropodos. Entre las especies que se colectaron estan Dermacentor occidentalis, Euschoengastia sp. y Opisodasys vesperalis, aunque presumimos que son accidentalmente arras- trados por las presas mayores de S. occidentalis. Strigiphilus se encontro en todas las subespecies de S. occidentalis, mientras que Kurodaia magna se colecto solamante en S. o. caurina, la especie del norte. Icosta americana se encontro en S. o. occidentalis de California; previamente tambien ha sido bien documentada en la subespecie del norte. Ornithoica vicina se encontro infectando a S. o. lucida. [Traduccion de Ivan Lazo] Ectoparasites are a potentially important yet rel- atively unstudied aspect of avian biology. Ectopar- asites can impair thermoregulatory ability (Booth et al. 1993), reduce nestling body mass and survivor- ship (Moller 1990), influence sexual selection (Clay- ton 1990a), and transmit endoparasites and patho- gens (Baker 1967, Clayton 1990a). Ectoparasites and their hosts also offer unique opportunities to study coevolution and community ecology, and can help elucidate phylogenetic relationships among re- lated bird taxa (Marshall 1981). The spotted owl {Strix occidentalis) has generated considerable scientific and political interest (USDI 1992). Consequently, all aspects of its biology are potentially important to biologists and managers. Recent surveys have described some endoparasites ^ Present address: Colorado Cooperative Fish and Wildlife Research Unit, 201 Wagar Building, Colorado State Uni- versity, Fort Collins, CO 80523 U.S.A. which infect spotted owls (Gutierrez 1989, Hoberg et al. 1989, Hoberg et al. 1993, Young et al. 1993). We initiated a survey of spotted owl ectoparasite richness, including each of the three subspecies (AOU 1957). Methods Between 1987 and 1993 approximately 1000 spotted owls were captured and released during demographic stud- ies in northwest California (Franklin et al. 1990), the central Sierra Nevada of California (Bias and Gutierrez 1992), and central Arizona. Birds from these areas rep- resented the northern {S. o. caurina), California (-S', o. oc- cidentalis), and Mexican {S. o. lucida) subspecies, respec- tively. Ectoparasites from these birds were collected incidentally when they were encountered during routine banding and data collection. Because no systematic meth- odology or special equipment were used, we did not at- tempt to quantify the prevalence or intensity of ectopar- asites using data from live owls. During 1993, however, we specifically examined 1 8 live Mexican spotted owls for hippoboscids. These birds were examined for about 5 min each, with each body region (front, back, head, wings) being searched by looking at the surface and by deflecting 232 December 1994 Spotted Owl Ectoparasites 233 feathers. No magnification or forceps were used. We used these data to estimate prevalence as the proportion of Mex- ican spotted owls infested, and mean intensity as mean number of hippoboscids per infested owl. We also examined 13 northern, 13 California, and 28 Mexican spotted owl museum skins for phthirapteran chewing lice. Many chewing lice will not abandon dead hosts, and will remain attached to feathers after the bird has been processed into a museum specimen (Marshall 1981). The host specificity and high mortality away from hosts exhibited by chewing lice (Marshall 1981) reduced the probability that secondary transfer occurred between museum specimens. Owl specimens examined were housed in the Museum of Vertebrate Zoology; Humboldt State University Collection; American Museum of Natural His- tory; Smithsonian Institution, Southwest Forest Science Complex; and the Museum of Northern Arizona. Owl specimens were originally collected at locations scattered throughout their respective ranges, except for northern spotted owl specimens which came solely from northwest California. Each museum specimen was searched with the naked eye for about 20 min by looking at the surface and by deflecting feathers, and any lice found were removed with forceps. Estimates of louse prevalence were calculated using data from museum specimens only. All ectoparasites collected were placed in 70% ethyl alcohol. Representative specimens of each taxon were iden- tified by cooperating taxonomists of the United States De- partment of Agriculture’s Systematic Entomology Labo- ratory in Beltsville, Maryland. Results and Discussion Seven species from five arthropod orders were collected: one tick (Parasitiformes), one mite (Acar- iformes), one flea (Siphonaptera), two chewing lice (Phthiraptera), and two hippoboscid flies (Diptera). Given the limited scope of our survey, we are con- fident that other ectoparasite taxa are yet to be col- lected from spotted owls. We considered three of the seven ectoparasite spe- cies as accidental on spotted owls, probably origi- nating from prey items. A larval Dermacentor occi- dentalis (Acari: Ixodidae) and a nymphal D. occidentalis were collected from live northern spotted owls. Small mammals act as hosts for immature D. occidentalis throughout western California and Or- egon (Kohls 1937). We also collected a larval Eus- choengastia sp. (probably E. numerosa) mite (Acari: Trombiculidae) from a Mexican spotted owl mu- seum specimen that was collected in Phoenix, Ari- zona. Euschoengastia show a preference for rodents; E numerosa larvae parasitize a wide variety of bird and mammal taxa (Wrenn and Loomis 1974). We observed, but were unable to collect, an unidentified flea on a live Mexican spotted owl, and we collected an Opisodasys vesperalis (Ceratophyllidae) from a live northern spotted owl. This flea is a common parasite of northern flying squirrels (Glaucomys sabrinus) in the Pacific Northwest (Lewis et al. 1988). True hosts for each of these ectoparasite species were listed by Forsman et al. (1984) as spotted owl prey. Accidental ectoparasites are of interest because colonization of new host species may be a result of repeated transfers from true to accidental hosts (Marshall 1981, Clay- ton 1990b). We found lice on five (38.5%) of the northern, two (15.4%) of the California, and one (3.6%) of the Mexican spotted owl museum skins. A possible ex- planation for the greater percentage of northern spotted owl museum specimens with lice was that based on specimen labels, most of these owls were collected after being found dead or injured. The sin- gle Mexican spotted owl museum specimen with lice was found dead, while all the remaining Mexican and California specimens were apparently healthy when they were collected. Unhealthy hosts often have higher louse loads, presumably as a result of de- creased grooming activity (Marshall 1981). Any ef- fect of Phthiraptera on spotted owls is unknown. We collected Strigiphilus syrnii (Ischnocera: Phil- opteridae) from both live and museum specimens of northern spotted owls, and Strigiphilus sp. lice from museum specimens of California spotted owls that were collected in Mariposa County, California, but none from live California spotted owls. Strigiphilus sp. were found on a Mexican spotted owl museum skin that was collected in the Sacramento Mountains of New Mexico. S. syrnii were found on live Mexican spotted owls. Northern spotted owls are known hosts of this species (Clayton and Price 1984, Clayton 1990b). The previously published records of S. syrnii from spotted owls were from unspecified locations in California and British Columbia, Canada (Clay- ton and Price 1984). Other known hosts for S. syrnii are the great horned owl (Bubo virginianus) , barred owl (Strix varia), great gray owl (Strix nebulosa), and rufous-legged owl (Strix rufipes; Clayton and Price 1984, Clayton 1990b). We collected Kurodaia magna (Amblycera; Men- oponidae) from live and museum specimens of north- ern spotted owls from northwest California, but none from California or Mexican spotted owls. This spe- cies had been previously collected from northern spotted owls from western Oregon (R.D. Price pers. comm.). The great horned owl and the barred owl are the other known hosts of K. magna (Price and Beer 1963). 234 Hunter et al. VoL. 28, No. 4 Young et al. (1993) described in some detail the occurrence of hippoboscid flies on northern spotted owls from northwest California. Seventeen percent of the owls that they examined were infested with hippoboscid flies; mean intensity was 2.4. They col- lected one individual of Ornithomya anchineuria, while all additional flies were Icosta americana. We found /. americana on California spotted owls. Of the 18 Mexican spotted owls we examined for hippobos- cids, 'h'h.'iJo were infested with Ornithoica vicina. Mean intensity was 4.2 (range 2-6, SD = 1.5). Bequaert (1956) noted that subspecific distinctions are not known to influence host choice in hippobos- cids, and cited verified records of I. americana from northern and Mexican spotted owls, and O. vicina from California and Mexican spotted owls (Be- quaert 1954, 1955). While these hippoboscids may occur on any of the three subspecies, we found that northern spotted owls were primarily infested with /. americana, while Mexican spotted owls were pri- marily infested with O. vicina. The infestation of northern and Mexican spotted owls by two different species of hippoboscid may be explained by different climatic tolerances. I. americana has been found pri- marily in temperate northern climates, but not in subtropical climates (Bequaert 1952, Bennett 1961) whereas 0. vicina is considered primarily a tropical and subtropical species (Bennett 1961). Other hosts of I. americana include Accipiter spp., red-tailed hawks {Buteo jamaicensis), ruffed grouse {Bonasa umbellus), barn owls {Tyto alba), great horned owls, and barred owls. O. vicina is one of the most common hippo- boscids, and has been found on members of 10 avian orders including red-tailed hawks, ruffed grouse, great horned owls, barred owls, and a variety of passerines (Bequaert 1956). Hippoboscids may affect spotted owls in several ways. Given the importance of hearing in owl for- aging (USDI 1992), larviposition in the ears by 0. vicina could reduce the fitness of individual owls. Bequaert (1952) cited two records of larviposition by O. vicina, one of which was in the ears of a great horned owl; we observed O. vicina exclusively within the ears of Mexican spotted owls. Hippoboscids also may affect spotted owls by acting as vectors for he- matozoa (Gutierrez 1989, Young et al. 1993) or other pathogens (Baker 1967). Increasing interac- tions between barred and spotted owls as a result of barred owl range expansion (USDI 1992) may ex- pose spotted owls to new, more virulent pathogens. Finally, northern spotted owl nestlings have been observed with severe trauma as a result of hippo- boscid infestation (USDI 1992). Acknowledgments We thank J.E. Keirans for identifying ticks, W.J. Wrenn for mite identification, R. Traub for flea identification, R.D. Price for identifying chewing lice and reviewing drafts, and R.V. Peterson for identifying hippoboscid flies. We are grateful to C. Moen and D. Oram for collecting ectoparasites. We also thank R. Bank, G.F. Barrowclough, W.M. Block, M.R. Browning, S. Cutler, J.L. Ganey, and N.K. Johnson for providing access to museum specimens. D.H. Clayton provided a thoughtful review of an earlier draft. Partial funding was provided by the California De- partment of Fish and Game (Contract FG0271), USDA Forest Service (PSW-90-0012CA, PSW-93-0020CA, and Contract 52-82FT-3007), and USDI Fish and Wildlife Service (Coop. Agreement 14-16-009-1547). Literature Cited American Ornithologists’ Union. 1957. Check-list of North American birds. 5th. Ed. Am. Ornithol. Union, Washington, DC U.S.A. Baker, J.R. 1967. A review of the role played by the Hippoboscidae (Diptera) as vectors of endoparasites. J. Parasitol. 53:412-418. Bennett, J.C. 1961. On three species of Hippoboscidae (Diptera) on birds in Ontario. Can. J. Zool. 39:379- 406. Bequaert, J.C. 1952. The Hippoboscidae or louse-flies (Diptera) of mammals and birds. Part I. Structure, physiology and natural history. Entomol. Am. 32:1- 209. . 1954. The Hippoboscidae or louse-flies (Dip- tera) of mammals and birds. Part II. Taxonomy, evo- lution and revision of American genera and species Entomol. Am. 34:1-232. . 1955. The Hippoboscidae or louse-flies (Dip- tera) of mammals and birds. Part II. Taxonomy, evo- lution and revision of American genera and species Entomol. Am. 35:233-416. . 1956. The Hippoboscidae or louse-flies (Dip- tera) of mammals and birds. Part II. Taxonomy, evo- lution and revision of American genera and species. Entomol. Am. 36:417-611. Bias, M.A. and R.J. GutiErrez. 1992. Habitat asso- ciations of the California spotted owl in the central Sierra Nevada. /. Wildl. Manage. 56:584-595. Booth, D.T., D.H. Clayton and B.A. Block. 1993 Experimental demonstration of the energetic cost of parasitism in free-ranging hosts. Proc. R. Soc. Lond. Ser. B. Biol. Sci. 253:125-129. Clayton, D.H. 1990a. Mate choice in experimentally parasitized rock doves: lousy males lose. Am. Zool. 30 251-262. . 1990b. Host specificity of Strigiphilus owl lice December 1994 Spotted Owl Ectoparasites 235 (Ischnocera: Philopteridae), with the description of new species and host associations. /. Med. Entomol. 27:257- 265. AND R.D. Price. 1984. Taxonomy of the Stri- giphilus cursitans group (Ischnocera: Philopteridae), parasites of owls (Strigiformes). Ann. Entomol. Soc. Am. 77:340-363. Forsman, E.D., E.C. Meslow AND H.M. Wight. 1984. Distribution and biology of the spotted owl in Oregon. Wildl. Monogr. 87. Franklin, A.B., J.P. Ward, R.J. Gutierrez and G.I. Gould, Jr- 1990. Density of northern spotted owls in northwest California. /. Wildl. Manage. 54:1-10. GutiErrez, R.J. 1989. Hematozoa from the spotted owl. /. Wildl. Dis. 25:614-618. Hoberg, E.P., G.S. Miller, E. Wallner-Pendleton AND O.R. Hedstrom. 1989. Helminth parasites of northern spotted owls {Strix occidentalis caurina) from Oregon. /. Wildl. Dis. 25:246-251. , R.J. Cawthorn and O.R. Hedstrom. 1993. Enteric coccidia (Apicomplexa) in the small intestine of the northern spotted owl (Strix occidentalis caurina). J. Wildl. Dis. 29:495-497. Kohls, G.M. 1937. Hosts of the immature stages of the Pacific coast tick Dermacentor occidentalis Neum. (Ix- odidae). Public Health Rep. 52:490-496. Lewis, R.E., J.H. Lewis and C. Maser. 1988. The fleas of the Pacific Northwest. Oregon State Univ. Press, Corvallis, OR U.S.A. Marshall, A.G. 1981. The ecology of ectoparasitic in- sects. Academic Press, London, U.K. Moller, A.P. 1990. Effects of parasitism by a hae- matophagous mite on reproduction in the barn swal- low. Ecology 71:2345-2357. Price, R.D. and J.R. Beer. 1963. The Kurodata (Mal- lophaga: Menoponidae) parasitic on the Strigiformes, with a key to the species of the genus. Ann. Entomol Soc. Am. 56:849-857. U.S. Department of the Interior. 1992. Draft re- covery plan for the northern spotted owl. USDA Fish Wildl. Serv., Portland, OR U.S.A. Wrenn, W.J. and R.B. Loomis. 1974. The Euschoen- gastia radfordi species complex (Acarina: Trombiculi- dae) from western North America, with descriptions of five new species. Ann. Entomol. Soc. Am. 67:241- 256. Young, K.E., A.B. Franklin and J.P. Ward. 1993. Infestation of northern spotted owls by hippoboscid (Diptera) flies in northwestern California. J. Wildl Dis. 29:278-283. Received 11 April 1994; accepted 18 July 1994 J Raptor Res. 28(4);236-241 © 1994 The Raptor Research Foundation, Inc. EFFECTS OF RAPTORS ON THE ACTIVITY OF SANDGROUSE Peter N. Ferns and Shelley A. Hinsley School of Pure and Applied Biology, University of Wales, College of Cardiff , P.O. Box 915, Cardiff, CF1 STL, United Kingdom Abstract. — Raptors accounted for 55% of the disturbance incurred by two species of sandgrouse (Pterocles alchata and P. orientalis) in breeding habitat in La Mancha, Spain, during spring and summer. The main species of raptors were the golden eagle {Aquila chrysaetos) in spring and Montagu’s harrier (Circus pygargus) in summer. During the breeding season, Montagu’s harriers carried out the majority of their foraging at those times of day when male sandgrouse collected water for their chicks. This presumably enabled the harriers to locate chicks more easily. Sandgrouse lost no more than 0.5 hr of feeding time per day as a result of disturbance by raptors, and yet were still able to rest for at least 3 hr a day at times when they could have been feeding. Predators as a whole were probably the major factor responsible for the different dispersion patterns adopted by sandgrouse at different times of year. Key Words: golden eagle; Montagu’s harrier; Pterocles alchata; Pterocles orientalis; raptor disturbance; sandgrouse; vigilance. Resumen. — -Aves rapaces fueron responsables por 55% de las perturbaciones sufridas por dos especies de gangas (Pterocles alchata and P. orientalis) en sus areas de reproduccion en la Mancha, Espana, durante la primavera y el verano. Las especies de aves de rapina mas prominentes fueron el aguila real (Aquila chrysaetos) en la primavera y el aguilucho cenizo (Circus pygargus) en el verano. Durante la estacion de reproduccion los aguiluchos cenizos llevaron a cabo la mayoria de su forrajeo en las horas del dia en que las gangas machos estaban recogiendo agua para sus polios. Esto supuestamente les permitio a los aguiluchos localizar los polios mas facilmente. Las gangas perdieron no mas de media hora del tiempo de alimentacion diario debido a las perturbaciones por aves de rapina sin embargo estuvieron inactivos por lo menos durante 3 horas al dia cuando hubieran podido estar alimentando. Las aves de rapina en general fueron probablemente el factor mas importante en determinar los distintos partrones de dispersion de las gangas en distintas epocas del ano. [Traduccion Autores] Raptors can exert a significant influence on the flocking behavior of birds, both in the short term and the long term. In response to attacks by spar- rowhawks (Accipiter nisus), redshanks (Tringa tot- anus) deserted their winter feeding territories for up to an hour and formed flocks, while turnstones (Ar- enaria interpres) reduced their individual spacing from three bird lengths to one bird length (Whitfield 1988). Myers (1984) found that winter feeding territories were abandoned by sanderlings (Calidris alba) in years when merlins (Falco columbarius) were more abundant. These studies focused on the influence of raptors on the flocking and vigilance of potential prey (see also Metcalfe 1989). The amount of time for which target organisms are prevented from car- rying out other activities has seldom been measured. In one early study, Morse (1973) found that spar- rowhawks attacked particular tit (Parus spp.) flocks about once per day on average, and disturbed the birds for only about 2 min. Longer disturbances can be inferred from detailed studies of merlins and spar- rowhawks attacking wintering waders (Page and Whitacre 1975, Boyce 1985, Cresswell and Whit- field 1994), though no actual estimates are available in these cases. In order to derive estimates of the total time lost by pin-tailed sandgrouse (Pterocles alchata) and black- bellied sandgrouse (P. orientalis) as a result of dis- turbance by raptors, we recorded the amount of time the birds spend being vigilant, the number of inci- dents of disturbance and the factors responsible. This was done as part of the compilation of time and energy budgets for these two species of sandgrouse in Spain. 236 December 1994 Raptor Disturbance of Sandgrouse 237 Study Area and Methods Observations were made in semi-arid farmland in the province of La Mancha, Spain, in spring (29 March to 21 April) and summer (22 July to 10 August) 1986. Sandgrouse occurred in flocks during the spring, com- prising on average 26 (SD = 23, N = 31) individuals in pin-tailed sandgrouse, with a maximum flock size of 75, Most daytime flocks of black-bellied sandgrouse contained 2-7 birds, though slightly larger groups (maximum 28) roosted together overnight. In summer, both species oc- curred in scattered family groups. When compiling time budgets, we scan-sampled bird activity at 1 min intervals. In both seasons, the day was divided into 10 equal time periods. These time periods were longer in summer (92 min) than in spring (82 min). The field area and study methods are described elsewhere (Hinsley 1994, Hinsley and Ferns 1994). A total of over 46 000 bird observations were made, involving on average 29 pin-tailed sandgrouse and 6 black-bellied sandgrouse each day in spring, and family groups of 2-5 pin- tailed sandgrouse in summer (at least two different families were watched each day, and over 40 families were studied in total). Every potential predator that was observed close enough to the focal birds to disturb them was recorded. If the same source of dis- turbance caused more than one group of birds to react, this was treated as a single incident, even if the events occurred a considerable time apart (actual maximum = 7 min). The observations in the spring were undertaken at a range of 300-1000 m (typically 700 m) to avoid disturbing the birds, and it was not possible to distinguish gender at this time of year. Observations in the summer were un- dertaken at a range of 200-800 m (typically 400 m). In the summer, groups split up into family parties in which males, females and juveniles (1 -2-mon-old at the time of the observations) were distinguishable. To estimate the amount of time that sandgrouse spent being vigilant when disturbed, we performed stepwise re- gression on the percentage of time spent being vigilant (arcsine transformed) in each of the 10 daily time periods against the number of records of disturbance caused by all factors that we were able to identify. These were divided into five categories comprising the numbers of the most common raptors in each season (golden eagle [Aquila chry- saetos] in spring, Montagu’s harrier [Circus pygargus] in summer); other raptors; other birds (mainly red-legged partridges [Alectoris rufa] and great grey shrikes [Lanius excubitor]); mammals (rabbits, sheep and dogs); and hu- mans (including noises caused by human activities). This method of measuring the affect of disturbance has the advantage of taking into account losses of feeding time through increased alertness even when this lasted for some time after the disturbance ended. Results In the absence of disturbance, sandgrouse vigi- lance was typically restricted to short bouts with the head up. The duration of these bouts was longest in male pin-tailed sandgrouse accompanying chicks (.x = 6.5 sec, SD = 3.9, N = 47). When disturbed, Table 1. Number of raptors observed while compiling sandgrouse time budgets. Number of Sightings OF Each Raptor SUM- Spring MER N = N = Raptor Species 23 da 20 da Black kite {Milvus migrans) 4 Red kite (M. milvus) 6 Montagu’s harrier {Circus pygargus) 1 23 Common buzzard {Buteo buteo) 3 4 Imperial eagle {Aquila heliaca) 2 Golden eagle {A. chrysaetos) 11 Booted eagle {Hieraaetus pennatus) 1 Common kestrel {Falco tinnunculus) 1 Peregrine falcon {Falco peregrinus) 1 2 Little owl {Athene noctua) 1 sandgrouse adopted alert upright postures or crouched close to the ground (these behaviors were also recorded as vigilance). The most common identified source of disturbance was raptors. During a total of 43 d of observations, the number of disturbances due to raptors was 60. A further 49 disturbances were attributed to other causes. The species of raptors involved (Table 1) differed significantly between spring and summer, though the number of incidents was very similar. The most important species in spring was the golden eagle. The most important species in summer was Montagu’s harrier (golden eagles were absent from the area during the summer). Because the same in- dividual raptor may have caused disturbance on more than one day, there may have been some unavoidable pseudoreplication. However, the proportion of dis- turbances caused by the most abundant raptor (com- pared with all other birds) in the two seasons was significantly different (Fisher’s exact test, P < 0.005). A greater diversity of raptors was thus responsible for disturbance during the spring (Table 1). Red kites (Milvus migrans) and black kites (M. milvus) were recorded in spring but not summer, both being migrants in the area. A flock of 106 black kites that passed over on migration on the evening of 3 August was not included since it caused no disturbance. Golden eagles were observed circling quite high (about 30-50 m) over the fields and hill slopes, and Number of raptors, and percentage of sandgrouse time budget 238 Peter N. Ferns and Shelley A. Hinsley VoL. 28, No. 4 Figure 1. Selected raptor activity (numbers) and sand- grouse vigilance (percentage of time budget) throughout the daylight period. The total number of raptors is shown in spring (a), the total number of raptors other than Mon- tagu’s harriers (b), and number of Montagu’s harriers in summer (c). also perched on vantage points overlooking the fields. Both golden eagles and red kites regularly take birds the size of adult sandgrouse (Cramp and Simmons 1979). In spring, a juvenile imperial eagle {Aquila heliaca) took a taxidermic mount of a sandgrouse (used to measure operative temperature) from the ground and was only prevented from flying off with it by the attached data-logger. Montagu’s harriers and red kites quartered the fields at a height of only a few meters, and occasionally circled at a greater elevation. In summer, Montagu’s harriers were ob- served attacking young sandgrouse, but not adults. The young of ground-nesting birds form a consid- erable part of this harrier’s diet during the summer months (Cramp and Simmons 1979). Vigilance of adult sandgrouse in each time period (Fig. 1) peaked during the middle of the day in spring, and during the morning and evening in sum- mer. The gap between the summer peaks coincided with a period of reduced activity of both raptors and sandgrouse in the middle of the day, induced by high temperature. The vigilance of juvenile sandgrouse reached a peak earlier in the morning than that of adults. Adult males spent more time being vigilant in summer than in spring (19% of daylight hours compared with 37o), presumably because of the vul- nerability of their ofTspring during the former pe- riod. This difference was statistically significant (Kruskal-Wallis test, H = \\.?>, P = 0.001. While male and female vigilance in the 10 daily time periods were correlated significantly during summer (r = 0.69, P < 0.02, N = 10), juvenile vigilance was not correlated with that of either par- ent (r = 0.46 with males and r = 0.12 with females, P > 0.10 in both cases). This is largely because juveniles spent most of the day feeding or resting, relying on their parents to act as sentinels (especially their fathers). They showed most alertness when he was away collecting water for them during the morn- ing and when he returned. Juveniles were less sen- sitive to disturbance by Montagu’s harriers in the evenings (Fig. 1), probably because the latter spent most of their time hunting rabbits during this period of the day. Sandgrouse responded to raptors differently in spring and summer, and this reflected the different dispersion pattern they adopted in the breeding sea- son because of the vulnerability of their young (Lima and Dill 1990). In neither spring nor summer did the sandgrouse form denser groups when disturbed; if anything the flocks tended to split up. In spring. December 1994 Raptor Disturbance of Sandgrouse 239 they merely flew to another of the several fields in which they normally foraged during the course of the day. In summer, the first response of both adults and young to the presence of a raptor was to crouch and freeze. If a family group was located by a raptor, the adults usually mobbed it before flying off, while the young ran or flew into the nearest dense vege- tation. Consequently, their activity was affected mainly by a direct loss of time available for feeding, preening, drinking and other activities. We saw three direct attacks by Montagu’s harriers on juvenile sandgrouse, all unsuccessful, and adults with young mobbed Montagu’s harriers on four additional oc- casions. Male vigilance showed no significant cor- relation with disturbance factors, mainly because they spent such a high proportion of the time alert, even in the absence of any obvious disturbance (19% of daylight hours spent alert, compared with 1 2% in females). Since disturbance often caused sandgrouse to fly in spring, there was significantly more sandgrouse alertness in time periods in which the birds spent more time flying (r = 0.80, P < 0.01, N ^ 10 in all cases). In summer, reduced levels of alertness in both sexes of pin-tailed sandgrouse were associated with periods of resting (males, r = —0.68, P < 0.05; females, r = —0.67, P < 0.05). Differences in flock sizes were not significant in different time periods, nor was alertness significantly correlated with flock size in either species. Therefore, alertness within seasons was not significantly correlated with any variable that might have confounded its correlation with disturbance, including flock size or distance to cover (Elgar 1989). Stepwise regression analysis indicated that a high proportion of the time spent being vigilant was ac- counted for by the observed sources of disturbance, despite the small sample size (TV = 10 in all cases). In spring, 48% of the variation in pin-tailed sand- grouse vigilance between time periods was accounted for by the overall number of raptors recorded {t = 3.2, P < 0.02) and 20% by human disturbances {t = 2.4, P = 0.05). In summer, 46% of female vigilance was accounted for by raptors other than Montagu’s harriers {t = 2.5, P < 0.05) and 257o by mammals {t = 2.5, P < 0.05). Also in summer, 68% of juvenile vigilance was accounted for by the number of Mon- tagu’s harriers {t = 4.1, P = 0.003). Fifty-seven percent of black-bellied sandgrouse vigilance in spring was accounted for by disturbance caused by dogs, sheep and rabbits (^ = 5.6, P = 0.001), and 19% by that caused by humans {t = 4.8, P = 0.003). The larger size of this species apparently made it less vulnerable to disturbance by most raptors. Discussion While the spring flocks of sandgrouse may have been reasonably easy for golden eagles and other raptors to locate, they did not provide a particularly easy target as indicated by the rarity of the attacks we witnessed. Young sandgrouse during the breed- ing season were much more vulnerable, though they must have been more difficult to locate in the well- dispersed family groups. It is thus not surprising that Montagu’s harriers concentrated most of their foraging activity in the period when male sandgrouse returned with water from the waterholes, since this provided them with a potential clue to the where- abouts of young. Such predation pressure may have led in turn (c.f. Lima and Dill 1990) to a wider spread of watering times amongst males in this area, including evening watering. The two species of sandgrouse we observed tended to forage in difTerent types of habitat (Hinsley 1994), but we do not believe that this choice was influenced by raptor activity. The pin-tailed sandgrouse was the more vulnerable species, but it foraged in open fields where it was more visible than on the hill slopes favored by black- bellied sandgrouse. As suggested by Lima and Dill (1990), vigilance costs were higher during the breed- ing season. Adults with young were also prepared to run greater risks by mobbing raptors during the summer. Predation of young sandgrouse probably repre- sents a major source of mortality (Maclean 1985). However, young are likely to be even more vulner- able to raptors at water holes (Cade 1965) and this may explain why males continued to carry water to them even after they had become capable of flying to water holes for themselves. Raptors recorded at- tacking or consuming sandgrouse in other areas in- clude pallid harrier (Circus macrourus), golden eagle, greater kestrel (Falco rupico hides), sooty falcon (F. concolor), lanner falcon (F. biarmicus) and Barbary falcon (F. pelegranoides) (Ali and Ripley 1983, Cramp 1985, Urban et al. 1986). The amount of time males with young spent being vigilant (19% of daylight hours) was much less than that of male barnacle geese (Branta leucopsis) es- corting their broods (48%) (Forsland 1993). In both species, the males devoted more time to vigilance 240 Peter N. Ferns and Shelley A. Hinsley VoL. 28, No. 4 than the females. Walters (1990) found no difference in the response of long-toed lapwings {Vanellus cras- sirostris) and blacksmith plovers (V. armatus) ac- cording to whether or not they had young, but they did respond more strongly to harriers than to eagles. The intensity of predatory disturbance at our study site was relatively low compared with some others. For example, at Bolinas Lagoon in California, a single merlin mounted about 19 attacks per day on shorebirds (Page and Whitacre 1975). However, the 5000 potential prey in this area were divided into groups of at most several hundred, and therefore the disturbance incurred per bird was much lower. Cal- vo (1993) recorded more than 10 disturbances per day (an order of magnitude more than that caused by raptors at our site) at breeding colonies of collared pratincoles {Glareola pratincola) in southwest Spain. Black kites and common kestrels {Falco tinnunculus) were responsible for most pratincole disturbance, though Montagu’s harriers were also seen taking chicks. Most of the time sandgrouse spent being vigilant as a result of the activities of raptors, humans, and other mammals was time they would otherwise have spent feeding. In the middle of the day, when birds rested in the shade of shrubs, alertness and distur- bance levels were relatively low and the small amount of disturbance that did occur then was usually caused by red-legged partridges. None of the results described here suggest that disturbance by raptors presented sandgrouse with any difficulty in meeting their daily energy require- ments. The time of the year when the highest pro- portion of the day was occupied by foraging was the spring, and even then black-bellied sandgrouse were still able to devote over 3 hr, and pin-tailed sand- grouse over 4 hr, to resting at times when they could presumably have been feeding. The above regres- sions enable us to estimate that raptors in spring caused pin-tailed sandgrouse to lose about 24 min of foraging time per day. Mammals caused black- bellied sandgrouse to lose 10 min per day at the same time of year. In summer, Montagu’s harriers caused juvenile pin-tailed sandgrouse to lose 9 min per day. Other raptors caused adult females to lose about 25 min. Raptors may be the largest single source of disturbance, but man and other mammals also caused some, even in this thinly populated area. Predators as a whole were probably a major factor responsible for the different dispersion patterns adopted by pre- breeding and breeding sandgrouse. Acknowledgments We are most grateful to the Fundacion Jose Maria Blanc for allowing us to work on their estate in Spain and to D.H. Thomas for locating such a good place to study sandgrouse. SAH was supported by a SERC research studentship. We would also like to thank three referees for numerous improvements to the manuscript. Literature Cited Ali, S. and S.D. Ripley. 1983. Handbook of the birds of India and Pakistan. Oxford Univ. Press, Oxford, U.K. Boyce, D.A. 1985. Merlins and the behavior of win- tering shorebirds. Raptor Res. 19:94-96. Cade, T.J. 1965. Relations between raptors and colum- biform birds at a desert water hole. Wilson Bull. 77: 340-345. Calvo, B. 1993. Influences of agricultural land-use and habitat modification on the breeding biology and con- servation of collared pratincoles Glareola pratincola in SW Spain. Ph.D. dissertation, Univ. Glasgow, U.K. Cramp, S. [Ed.]. 1985. The birds of the western Pale- arctic. Vol. 4. Oxford Univ. Press, Oxford, U.K. AND K.E.L. Simmons. [Eds.]. 1979. The birds of the western Palearctic. Vol. 2. Oxford Univ. Press, Oxford, U.K. Cresswell, W. AND D.P. Whitfield. 1994. The effects of raptor predation on wintering wader populations at the Tyninghame estuary, southeast Scotland. Ibis 136: 223-232. Elgar, M.A. 1989. Predator vigilance and group size in mammals and birds: a critical review of the empirical evidence. Biol. Rev. Camb. Philos. Soc. 64:13-33. Forsland, P. 1993. Vigilance in relation to brood size and predator abundance in the barnacle goose, Branta leucopsis. Anim. Behav. 45:965-973. Hinsley, S.A. 1994. Daily time budgets and activity patterns of sandgrouse (Pteroclididae) in contrasting arid habitats in Spain and Israel. /. Arid Environ. 26: 373-382. AND P.N. Ferns. 1994. Time and energy bud- gets of breeding males and females in sandgrouse Pter- ocles species. Ibis 136:261-270. Lima, S.L. and L.M. Dill. 1990. Behavioral decisions made under the risk of predation: a review and pro- spectus. Can. J. Zool. 68:619-640. Maclean, G.L. 1985. Sandgrouse: models of adaptive compromise. S. Afr. J. Wild! Res. 15:1-6. Metcalfe, N.B. 1989. Flocking preferences in relation to vigilance benefits and aggression costs in mixed- species shorebird flocks. Oikos 56:91-98. Morse, D.H. 1973. Interactions between tit flocks and sparrowhawks Accipiter nisus. Ibis 115:591-593. Myers, J.P. 1984. Spacing behavior of nonbreeding shorebirds. Pages 271-321 in J. Burger and B.L. Olla December 1994 Raptor Disturbance of Sandgrouse 241 [Eds.], Behavior of marine animals. Vol. 6. Shorebirds. Migration and foraging. Plenum Press, NY U.S.A. Page, G. and D.F. Whitacre. 1975. Raptor predation on wintering shorebirds. Condor 77:73-83. Urban, E.K., C.H. Fry and C.S. Keith. [Eds.]. 1986. The birds of Africa. Vol. 2. Academic Press, London, U.K. Walters, J.R. 1990. Anti-predatory behavior of lap- wings: field evidence of discriminative abilities. Wilson Bull. 102:49-70. Whitfield, D.P. 1988. Sparrowhawks Accipiter msus affect the spacing behavior of wintering turnstone Ar- enaria interpres and redshank Tringa tetanus. Ibis 130. 284-287. Received 22 February 1994; accepted 30 July 1994 / Raptor Res. 28(4):242-245 © 1994 The Raptor Research Foundation, Inc. SEASONAL ABUNDANCE OF BLACK KITES ASSOCIATED WITH THE RUBBISH DUMP OF MADRID, SPAIN Guillermo Blanco Departamento de Biologia Animal, Universidad de Alcala de Henares, Alcala de Henares, 28871 Madrid, Spain Abstract. — This study describes the seasonal variations in number of black kites (Milvus migrans) foraging at the rubbish dump of Madrid, Spain. Seasonal changes in the abundance of kites using the dump showed two peaks corresponding to both pre- and post-breeding migratory passages. The autumn passage was more lengthy and produced the highest number of kites (847). During the breeding season the population of kites foraging in the dump showed no seasonal trends, suggesting that it consisted mainly of nonbreeding birds, both adults and subadults. Age-related differences were observed in the use of the dump by black kites during post-breeding which may simply reflect the existence of age-related differences in migration. A possible advantage for kites to forage in the dump is access to an abundant and predictable food supply. On the other hand, kites may face some dangers from entanglement in plastic materials and other rubbish in the dump. Key Words: black kite, Madrid; Milvus migrans; rubbish dump; seasonal abundance; Spain. Abundancia estacional de Milanos Negros {Milvus migrans) asociados al basurero de Madrid, Espana Resumen. — Este estudio describe las variaciones numericas estacionales de Milanos Negros {Milvus migrans) en relacion a su periodo de permanencia y paso migratorio sobre el basurero de Madrid, Espana. Los cambios estacionales en la abundancia de milanos reflejaron dos maximos correspondientes a los pasos migratorios. El paso migratorio postreproductor fue mas prolongado y en el se alcanzo el mimero mas alto de milanos (847). Durante la epoca de erfa la poblacion de milanos asociada al basurero no mostro variaciones numericas, sugiriendo que estuvo compuesta principalmente por individuos no repro- ductores, tanto adultos como subadultos. La proporcion de cada clase de edad vario durante el paso postreproductor en el basurero, lo cual podria estar relacionado con diferencias migratorias entre las distintas clases de edad. Se discuten las posibles ventajas y desventajas de alimentarse en basureros para los Milanos Negros. [Traduccion Autor] Black kites {Milvus migrans) often gather in large communal roosts and breeding colonies. Large ag- gregations also occur where food is abundant and during migration (Bernis 1980, Cramp and Sim- mons 1980). Their highly gregarious and opportu- nistic foraging behavior leads them to eat the most abundant and available prey, especially slow-moving and injured animals as well as food obtained by scavenging (Valverde 1967, Arroyo 1978, Cramp and Simmons 1980, Shiraishi et al. 1990). As scav- engers, black kites have been recorded in large num- bers on waste accumulations generated by human activities, including rubbish dumps, markets, fish- ing-ports and abattoirs all over their world range (e.g., Bernis 1973, Pomeroy 1975, Roberts 1991, Shiraishi et al. 1990). Despite the widespread range and abundance of the black kite (Brown and Amadon 1968), limited information is available concerning the proportion of age classes in resident groups and in migration (Bernis 1980, Cramp and Simmons 1980). Knowl- edge about the age distribution of kites using rubbish dumps, and the timing of residency and migration associated with these sites is almost nonexistent (Pomeroy 1975, Gomez-Tejedor and De Lope 1993). The objective of this study was to determine the seasonal variations in numbers of black kites at the Madrid, Spain rubbish dump. I also present data concerning the age distribution of black kites asso- ciated with the dump through the summer, and dis- cuss the possible advantages and disadvantages for black kites of foraging on refuse. 242 December 1994 Black Kites in the Madrid Dump 243 Study Area and Methods The domestic refuse generated by Madrid, a city of about 4 000 000 people, is deposited daily at the Valdem- ingomez municipal dump, about 12 km southeast of Ma- drid. Several bird species have been recorded feeding on the more than 1 000 000 tons of refuse deposited there annually (Cantos and Asensio 1990). The major scaven- gers recorded at this dump were white storks iCiconia ciconia), black kites, black-backed gulls (Larusfuscus), black- headed gulls (Larus ridibundus) , magpies {Pica pica) and jackdaws {Coruus monedula) numbering in the hundreds and thousands of birds in each case (Gomez and De Juana 1984, pers. obs.). During 1993, I monitored the number of black kites using the Madrid dump by counting them when they arrived at the only communal roost in the area. The com- munal roost was situated in the bank vegetation of the Manzanares River, about 4 km east of the dump. The adjacent area was used for agriculture and cattle grazing (Blanco et al. 1991). Roost trees were mainly large cot- tonwoods {Populus alba) and elms (Ulmus minor). About 2 hr before sunset I began observing and counting the kites perched on the roost trees from about 50 m with binoculars and a 20-60 x spotting scope. Later I noted the number of kites entering the communal roost from the dump. Observations of arriving kites were conducted from a fixed vantage point located between the roost and the dump, and lasted until all the kites were settled on the trees. When conditions of light and time allowed, I attempted to age as many kites as possible on the basis of plumage coloration, shape (Sylven 1977, Cramp and Simmons 1980) and plumage features derived from molt. Black kites were considered either adult (definitive plumage), juvenile (first summer), or immature (second summer). Birds in their second summer were easily recognized by remnants of juvenile plumage especially in coverts and scapulars, as well as by the coloration of old primaries. In addition, kites in their second summer molting their flight feathers for the first time, were easily recognized in flight by their shape — an irregular appearance because of the loss of several flight feathers or irregular molt patterns. This clearly contrasted with the synchronized molt pattern of adults. Because of the impossibility of exactly aging a large number of kites by these criteria, I focused on recording whether they were molting (adults and immatures) or not (juveniles), which was easily accomplished at long dis- tances. This partial ageing method was used from June, when the first juveniles appeared in the roost, to the end of July, when the development stage of molt in adults makes it difficult to assign each individual to an age class. After the end of July I considered only kites aged by means of plumage coloration, shape, and molting pattern. Results Seasonal changes in abundance of kites using the Madrid dump clearly peaked in correspondence with both pre-breeding and post-breeding migration (Fig. 1). Black kites usually arrived on the study area from Africa about the first week of February. Spring NUMBER OF BLACK KITES Figure 1. Seasonal abundance of black kites foraging m the Madrid dump, 1993. migration rapidly increased the number of kites to nearly 300. Numbers were lowest during the breed- ing season (March-July), and increased from June as the first juveniles joined the roost. The proportion of juveniles increased to the end of July although the highest proportion was reached later (Fig. 2) when the major influx of migrants occurred (Fig. 1) . The autumn migration was more lengthy and the numeric increase of kites was not always con- stant, suggesting the arrival and/or departure of suc- cessive surges of migrants. By recording if black kites were molting, I was able to assign 31-86% {N = 134-235) of the birds present each day to one of the two age classes defined by this method. By the more precise ageing technique, I successfully aged only about 13% {N = 2-109). Adults were the highest proportion of the population throughout the period in which the more precise method was used (Fig. 2) . The percentage of adults increased to 100% by the end of period when black kites were resident in the area. On the other hand, the sightings of im- mature individuals peaked in early August and then declined steadily (Fig. 2). Departure of nearly 850 kites occurred in the first week of August, and black kites disappeared entirely by the end of that month (Fig- !)• During the breeding season kites foraging at the dump showed no seasonal trends (Fig. 1). Most kites observed arriving from the dump used the communal roost, and breeding individuals usually roosted close to their nesting sites. Hence, it is unlikely that the presence of active breeders in the roost may have been influenced by the lack of seasonal variations in the size of the population during the breeding season. 244 Guillermo Blanco VoL. 28, No. 4 Molting Adults LI ] Second summer T.-.J Juveniles DATE Figure 2. Percentage distribution of black kites by age classes using the Madrid dump in summer. Molting kites include birds in their second summer and adults. This suggests that the communal roost contained mainly nonbreeding birds. Observations on the age and behavior of kites using the communal roost showed that both breeding and nonbreeding (im- matures and nonbreeding birds with definitive plum- age) foraged in the dump throughout the breeding season. However, it was impossible to obtain a large enough sample of aged birds to assess the proportion of each age class. Discussion The results of this study indicate that black kites used the Madrid dump as a feeding station before and during their migratory travels. The increase of black kites foraging at refuse dumps during migra- tion has been reported at several other Spanish lo- calities (Sanchez et al. 1990, Donazar 1992, Gomez- Tejedor and De Lope 1993), suggesting that such sites could act as stopover and/or congregating cen- ters for migrant black kites throughout their migra- tory range in Spain. The number of black kites using the Madrid dump may be much larger than the maximum abundance peak indicated. In fact, my results suggested that successive surges of migrants used the dump at dif- ferent dates. Moreover, the highest count in the Ma- drid dump in early August was larger than counts recorded at other dumps in Spain. These numbers were only comparable with roosting and migrating aggregations of black kites passing southward over the Strait of Gibraltar (Bernis 1980). Such a cir- cumstance could be explained by the superabundant and permanently available food supply found in Ma- drid dump, but especially by the lack of other abun- dant sources of food in the surrounding area. The use of the Madrid dump by black kites has been known for several decades (Bernis 1973, pers. obs.) which suggests that the high predictability and abun- dance of food could have influenced the recruitment of birds from year to year. No previous studies attempted to examine sea- sonal age-dependent variations in the use of dumps by black kites. Black kites in their first summer did not constitute a major part of population foraging in the dump. This study revealed that timing of the use of the dump by the different age classes simply reflected age-related differences in the timing of mi- gration. By foraging in dumps, black kites could benefit from the availability and predictability of food found there (Pomeroy 1975, Donazar 1992). In contrast, several negative effects associated with the exploi- tation of refuse by birds have been pointed out (Inigo 1987, Donazar 1992). But, no electric power lines dangerous to black kites were in the vicinity of the Madrid dump, and there was no indication that toxic materials were ingested by the kites. Five kites were observed with pieces of rope, plastic bags, and pieces of plastic (up to 40 cm long) entangled with their legs, and one immature was seen with an empty can (approximate weight of 850 g) caught on a leg. The effects of these and other potential dangers to kites foraging in dumps remains unknown. Further research is needed to determine how much of a trade-off occurs between the costs and benefits of feeding on refuse for kites. More studies are also needed to understand the long-term influence of for- aging in dumps on the population size and life- history strategies of the black kite and other species. Acknowledgments I am grateful to Mari C. Blanco and Mario Herrera for their company during some counts and to Jose L. Telia, Jose A. Donazar, Fernando Hiraldo, and Jaime Potti for comments and improvements on the manuscript. Literature Cited Arroyo, B. 1978. La alimentacion del Milano negro {Milvus migrans) en una localidad de Espana Central. Ardeola 25:47-58. Bernis, F. 1973. Algunos dates de alimentacion y de- predacion de Falconiformes y Estrigiformes ibericas. Ardeola 19:225-247. . 1980. La migracion de las aves en el Estrecho de Gibraltar. Universidad Complutense de Madrid, Madrid, Spain. December 1994 Black Kites in the Madrid Dump 245 Blanco, G., J.A. Cuevas and J.A. Fargallo. 1991. La poblacion de Chova Piquirroja (Pyrrhocorax pyr- rhocorax) en el Sureste de Madrid (Centro de Espana). Ardeola 38:91-99. Brown, L.H. and D. Amadon. 1968. Eagles, hawks and falcons of the world. Country Life Books, London, U.K. Cantos, F.J. and B. Asensio. 1990. Evolucion de la invernada de gaviota reidora {Larus ridibundus) en Ma- drid. Ardeola 37:305-308. Cramp, S.C. and K.E.C. Simmons (Eds.). 1980. The birds of the western Palearctic. Vol. 2. Oxford Univ. Press, Oxford, U.K. Donazar, J.A. 1992. Muladares y basureros en la biol- ogia y conservacion de las aves en Espana. Ardeola 39: 29-40. G6mez, L. and E. De Juana. 1984. Aspectos de la invernada de Larus ridibundus en Madrid. Ardeola 31: 123-128. G6mez-Tejedor, H. and F. De Lope. 1993. Sucesion fenologica de las aves no passeriformes en el vertedero de Badajoz. Ecologia 7:419-427. I:Rigo, E.M. 1987, Feeding habits and ingestion of syn- thetic products in a black vulture population from Chiapas, Mexico. Acta Zool. Mex. Nueva Ser. 22:1-15 Pomeroy, D.E. 1975. Birds as scavengers of refuse in Uganda. Ibis 117:69-81. Roberts, T.J. 1991. The birds of Pakistan. Vol. 1. Oxford Univ. Press, Karachi, Pakistan. Sanchez, J.M., C. Gonzalez, A. Romero and A. SAnchez. 1990. Noticiario Ornitolbgico: Milano Ne- gro (Milvus migrans). Ardeola 37:333, Shiraishi, S., K. Koga and N. Kawaji. 1990. Food habits of the black-eared kite, Milvus migrans lineatus, in Nagasaki airport and its adjacent areas. /. Fac. Agric Kyushu Univ. 34:247-254. Sylven, M. 1977. Age determination of red kite Milvus milvus and black kite Milvus migrans. Vdr Fdgelvdrld 36:33-37. Val VERDE, J.A. 1967. Estructura de una comunidad de vertebrados terrestres. Estacion Biologica de Donana, Sevilla, Spain. Received 28 December 1993; accepted 26 July 1994 J Raptor Res. 28(4):246-252 © 1994 The Raptor Research Foundation, Inc. DIET OF URBAN AND SUBURBAN TAWNY OWLS (STRIX ALUCO) IN THE BREEDING SEASON Andrzej Zalewski Mammal Research Institute, Polish Academy of Sciences, 17-230 Biaknvieza, Poland Abstract. — The diet of tawny owls {Strix aluco) was studied during the breeding seasons 1988-90 in an urban and a suburban area in Toruh, Poland. Two amphibian, 18 bird, and 14 mammal species were recorded as prey in a sample of 312 pellets. From 600 prey items found in both sites, the house sparrow {Passer domesticus) was the most frequently taken bird prey and the common vole (Microtus arvalis) the most frequently taken mammal prey. Significantly more mammals than birds were taken at the suburban site than at the urban site {P < 0.001). At the urban site, the proportion of birds (except for tits [Parus spp.] and house sparrows) increased over the course of the breeding season, while the proportion of Apodemus spp. decreased. Similarly, at the suburban site the proportion of all birds increased and the proportion of Microtus spp. decreased. House sparrows at the urban site and Eurasian tree sparrows {Passer montanus) and tits at the suburban site were taken in higher proportion than their availability. An examination of dietary studies from elsewhere in Europe indicated that there was a positive correlation between mean prey size and increasing proportion of birds in tawny owl diets {P — 0.003). Key Words: tawny owl; diet; urban and rural area; bird selection; prey-size selection. Dieta de Strix aluco urbano y suburbano en la estacion reproductiva Resumen. — Se estudio la dieta de Strix aluco durante las estaciones reproductivas de 1988 a 1990 en un area urbana y suburbana en Torun, Polonia. Dos anfibios, 18 aves y 14 especies de mamiferos fueron registrados como presa en una muestra de 312 egagropilas. De 600 categorias de presas encontradas en ambos sitios. Passer domesticus fue el ave-presa mas frecuente junto al mamifero Microtus arvalis. Se consumieron significativamente mas mamiferos que aves en el sitio suburbano que en el urbano {P < 0.001). En el sitio urbano, la proporcion de aves (excepto para Parus y P. domesticus) incremento a medida que transcurria la estacion reproductiva, mientras que la proporcion de Apodemus spp. disminuyo. Si- milarmente, en el sitio suburbano la proporcion de todas las aves incremento y la proporcion de Microtus spp. disminuyo. Tanto P. domesticus en el sitio urbano como Passer montanus y Parus spp. en el sitio suburbano fueron consumidos en una proporcion maor a su disponibilidad. Un examen de estudios dietaries en Europa indico que hubo una correlacion positiva entre le tamano promedio de presas y el incremento de la proporcion de aves en la dieta de buhos {P = 0.003). [Traduccion de Ivan Lazo] Increasing numbers of animal species are adapting to urban environments; among birds, omnivorous and granivorous species most frequently inhabit towns (TomiaJoje and Profus 1977, Zalewski 1994). Predatory birds are also increasingly colonizing ur- ban areas, and thus become important links in the urban food webs. Such birds have to either find a habitat containing natural food resources or change their feeding ecology. The tawny owl {Strix aluco) is an example of a polyphagous species (Mikkola 1983), that can inhabit many environments and adapt to preying on the most abundant species. In this paper I compare tawny owl diets during the breeding season between an urban and a sub- urban area. I focus on diet changes throughout the breeding period, and diet changes in relation to bird prey availability in both study areas. Study Area and Methods The study took place in and around the city of Torun, central Poland (53“01'N, 18°35'E). The human population of Torun is 200 000 and the city covers an area of 115.8 km^. Forests of pine {Pinus sylvestris) predominate on sandy soil around the city. Oaks (mainly Quercus robur) and birches {Betula verrucosa) form admixtures in the pine forest. The urban site consisted of a park (2.1 ha) and a cem- etery (3.7 ha) located near the city center. The eastern side of the urban area Joined a small villa district, and on the northern side it was bordered by an open grassland area with residential estates behind it (the Chrobry resi- dential district). The suburban site (17 ha) was located 3 246 December 1994 Diet of Tawny Owls 247 km from the city center. It was a 60-70-yr-old pine forest with deciduous trees (black alder [Alnus glutinosa], maple [Acer sp.], and horsechestnut [Aescuius hippocastanum]) and a rich shrub layer. A small stream flowed through this site which resembled the forest surrounding Torun. On the eastern side it was flanked by buildings, and on the western side by a large open grass area. Tawny owl pellets were collected between February and June in 1988, 1989, and 1990 in the urban site, and in 1989 and 1990 in the suburban one. At both sites, pellets were collected from one or two breeding pairs of owls in the same territories in each year. These pellets were gath- ered regularly from roosts of adults and young owls and also near nests. The pellets were analyzed by standard methods. Prey remains were identified using keys for mammals (Pucek 1981), birds (Moreno 1985, 1986), and amphibians (Bohme 1977). The number of vertebrate prey were calculated separately for each sample of pellets found at the same time on the basis of skulls, jaws, or other bone elements. Insects were identified and counted based on exoskeleton remains. For each site the frequency of oc- currence and percentage biomass were calculated. Con- version factors were applied (Table 1), using 20 g as the standard weight for small mammals (Southern 1954). Thus, total weight of prey was converted to “prey units.” Pellets were collected during three periods: (1) egg laying and hatching (from 1 February to 15 March); (2) nestling (from 16 March to 30 April); and (3) fledgling — when young owls were fledged but still dependent on their par- ents (from 1 May to 30 June). This division was based on observations of breeding and young tawny owls around Torun. The G-test was used to test the proportions of prey Items in the owls’ diet (Sokal and Rohlf 1981). For both study sites, the proportion of each bird species in the diet was then compared with its relative abundance in the bird community (Ozga 1990, Zalewski and Przystalski 1993, Zalewski 1994) close to where the pellets had been col- lected. In all the sites, breeding birds were censused by the territorial mapping method. In urban green areas cen- suses were carried out in 1988-89, in the Chrobry district in 1989, and in the suburban forest in 1989-90. Results In the urban site 223 pellets were found (30, 96, and 97 in consecutive years). In the suburban forest 89 pellets were collected (11 and 78 in consecutive years). In all, 18 species of birds, 14 species of mam- mals, and two species of amphibians (common frog [Rana temporaria]^ and spadefoot toad [Pelobates fus- cus]) were recorded. Comparison of Owl Diets Between the Urban and Suburban Areas. At the urban site, tawny owls preyed mainly on birds (66.6% by numbers and 68.2% by biomass). The most important prey were the house sparrow and the Eurasian tree sparrow (Table 1). Larger birds such as the collared dove {Streptopelia decaocto) and European starling (Sturnus vulgaris) contributed a high percentage by biomass (23.8%). Significantly more mammals than birds were taken at the suburban site than at the urban site (G = 125.64, df = 1, P < 0.001). These were mainly common voles (Microtus arvalis) and Apodemus spp. which together comprised 39.8% prey by numbers and 36.7% of the prey biomass. At the urban site, mammals composed only 30.5% of tawny owl prey, and rabbits were also important food items (11.4% of the prey biomass). Amphibians were not a sig- nificant component of the owls’ diet, especially in town. Invertebrates were recorded in the diet during the study at both sites (Table 1). The mean weight of prey in the urban sites was 37.6 g, and 28.6 g in the suburban site. Little change occurred in the diet composition of tawny owls at the urban site through the breeding season (Table 2). However, the proportion of Apo- demus spp. decreased from the first stage of the breeding season (1 February to 15 March) to the next two stages. Likewise, the proportion of tits and mammals decreased. By contrast, except for tits and sparrows, the proportion of birds and house mice {Mus musculus) increased, although in the latter case to a minor degree. Amphibians from the genus Rana were important in the diet of the suburban owls in March and April when they migrated from the places of hibernation. In later months, their frequency decreased (Table 2). In contrast, the contribution of spadefoot toads increased as the breeding season progressed, asso- ciated with this species’ shift to terrestrial life m May and June (Juszczyk 1987). The proportion of birds increased greatly in the later stages of the breeding season, although the increase in house spar- rows was minor. The proportion of insectivorous mammals, especially the common shrew (Sorex ara- neus), also increased. The frequency of Microtus spp., however, dropped noticeably from February to June. Nevertheless, throughout the study the proportion of mammals in the diet was invariably above 50%. Bird Prey. Comparison of the proportions of bird species in the tawny owl diet with their relative availability indicated that the most abundant bird species were taken (Table 3). Ivlev’s electivity index (modified by Jacobs 1974) was calculated to quantify tawny owls’ selectiveness for a few bird species. House sparrows were clearly preferred by the owls in the urban site (D = 0.39). Eurasian tree sparrows com- prised a large proportion of the diet, although they were not common birds. Tawny owls preyed on house martins {Delichon urbica) to a greater extent than 248 Andrzej Zalewski VoL. 28, No. 4 Table 1. Tawny owl diet composition in the breeding season (1 February to 30 June) in 1988-90. (Cf = conversion factor in ‘prey units’ [see text], N = number of individuals, 7oN = percent of prey specimens, %B = percent of prey biomass.) Remains of insects were not included in biomass calculations. Prey Species Urban Site Suburban Site Total Cf N %N N 7 c N 7 oB N 7 oN 7 oB Rana spp . 1.6 10 2.6 2.3 67 30.1 33.6 77 12.8 12.0 Pelobates fuscus 0.9 1 0.3 0.1 9 4.0 2.5 10 1.7 0.9 Anura subtotal 11 2.9 2.4 76 34.1 36.1 87 14.5 12.9 Streptopelia decaocto 10.0 12 3.2 16.9 — — — 12 2.0 11.7 Delichon urbica 0.8 14 3.7 1.6 — — — 14 2.3 1.1 Sturnus vulgaris 4.0 12 3.2 6.9 — — — 12 2.0 4.7 Turdus merula 4.0 2 0.5 1.1 2 0.9 2.5 4 0.7 1.6 Phoenicurus phoenicurus 0.8 2 0.5 0.2 — — — 2 0.3 0.1 P. ochruros 0.8 2 0.5 0.2 — — — 2 0.3 0.1 Erithacus rubecula 0.8 — — — 1 0.5 0.3 1 0.2 0.1 Parus major 1.0 13 3.5 1.8 4 1.8 1.3 17 2.8 1.7 P caeruleus 0.5 3 0.8 0.2 — — — 3 0.5 0.1 P palustris / ater 0.5 — — — 1 0.5 0.2 1 0.2 0.1 Pica pica 9.5 1 0.3 1.3 — — — 1 0.2 0.9 Garrulus glandarius 8.5 — — — 1 0.5 2.7 1 0.2 0.8 Fringilla coelebs 1.0 — — — 2 0.9 0.6 2 0.3 0.2 Passer domesticus 1.5 146 38.7 30.9 9 3.9 4.2 155 25.9 22.6 P montanus 1.0 27 7.2 3.8 2 0.9 0.6 29 4.8 2.8 Carduelis chloris 1.5 5 1.3 1.1 — — — 5 0.8 0.7 Serinus serinus 0.5 2 0.5 0.1 — — — 2 0.3 0.1 Emberiza citrinella 1.2 1 0.3 0.2 — — — 1 0.2 0.1 Unidentified passerines 1.5 9 2.4 1.9 1 0.5 0.5 10 1.7 1.5 Aves subtotal 251 66.6 68.2 23 10.4 12.9 274 45.7 51.0 Talpa europaea 5.0 1 0.3 0.7 2 0.9 3.1 3 0.5 1.5 Sorex araneus 0.5 1 0.3 0.1 9 3.9 1.4 10 1.7 0.5 Eptesicus serotinus 1.2 1 0.3 0.2 — — — 1 0.2 0.1 Oryctolagus cuniculus 20.0 4 1.0 11.4 — — — 4 0.7 7.8 Mus musculus 0.8 34 9.0 3.8 4 1.8 1.0 38 6.3 3.0 Rattus norvegicus 5.0 1 0.3 0.7 1 0.5 1.6 2 0.3 1.0 Micromys minutus 0.4 7 1.8 0.4 1 0.5 0.1 8 1.3 0.3 Apodemus agrarius 1.2 19 5.0 3.2 4 1.8 1.5 23 3.8 2.7 A sylvaticus 1.2 2 0.5 0.3 12 5.3 4.5 14 2.3 1.6 A flavicollis 1.2 1 0.3 0.2 — — — 1 0.2 0.1 Apodemus spp. 1.2 14 3.7 2.4 21 9.4 7.9 35 5.8 4.1 Arvicola terrestris 4.5 1 0.3 0.6 — — — 1 0.2 0.4 Clethrionomys glareolus 1.2 9 2.4 1.5 13 5.8 4.9 22 3.7 2.5 Microtus oeconomus 1.4 — — — 4 1.8 1.8 4 0.7 0.5 M. arvalis 1.4 19 5.0 3.7 52 23.3 22.8 71 11.8 9.7 Microtus spp . 1.4 1 0.3 0.2 1 0.5 0.4 2 0.3 0.3 Mammals subtotal 115 30.5 29.4 124 55.5 51.0 239 39.8 36.1 Total vertebrates 377 223 600 Melolontha spp. 28 — 28 Garabidae 6 2 8 Geotrupes spp. — 13 13 Coleoptera 3 6 9 Total invertebrates 37 21 58 December 1994 Diet of Tawny Owls 249 Table 2. Variation in tawny owl diet between the three stages of the breeding period in the urban and suburban sites in 1988-90. (N = number of individuals, %N = percent occurrence of prey.) Prey Species 1 Feb to 15 March 16 March to 30 April 1 May to 30 June G-test N %N N %N N %N G P Urban site Rana spp. 5 4.2 4 2.5 1 1.0 2.05 ns Pelo bates fuscus — — — — 1 1.0 — Anuran subtotal 5 4.2 4 2.5 O 2.0 0.83 ns Parus major 7 5.8 6 3.8 — — 5.00 ns Passer domesticus 47 39.2 65 40.9 34 34.7 0.54 ns Other birds 19 15.8 34 21.4 39 39.8 11.79 b Birds subtotal 73 60.8 105 66.1 73 74.5 1.41 ns Insectivora 1 0.8 1 0.6 — — 0.27 ns Mus musculus 5 4.2 17 10.7 12 12.2 4.60 ns Apodemus spp. 21 17.5 11 6.9 4 4.1 10.09 b Microtus spp. 7 5.8 8 5.0 4 4.1 0.31 ns Other mammals 8 6.7 12 8.2 3 3.1 2.54 ns Mammal subtotal 42 35.0 50 31.4 23 23.5 2.40 ns Total prey 120 100.0 159 100.0 98 100.0 Number of pellets 61 86 76 Suburban site Rana spp. 50 35.0 14 28.0 3 10.0 15.49 C Pelobates fuscus 4 2.8 1 2.0 4 13.3 12.37 C Anuran subtotal 54 37.8 15 30.0 7 23.3 3.44 ns Parus major 2 1.4 1 2.0 1 3.3 0.85 ns Passer domesticus 3 2.1 3 6.0 3 10.0 5.61 ns Other birds 3 2.1 5 10.0 2 6.7 7.69 ns Birds subtotal 8 5.6 9 18.0 6 20.0 9.79 b Insectivora 4 2.8 3 6.0 4 13.3 7.87 a Mus musculus 2 1.4 1 2.0 1 3.3 0.85 ns Apodemus spp. 18 12.6 13 26.0 6 20.0 4.78 ns Microtus spp. 50 34.9 3 6.0 4 13.3 24.66 C Other mammals 7 4.9 6 12.0 2 6.7 3.36 ns Mammal subtotal 81 56.6 26 52.0 17 56.7 0.26 ns Total prey 143 100.0 50 100.0 30 100.0 Number of pellets 45 21 12 ^ P < 0.05. < 0 . 01 . ^P < 0 . 001 . predicted from this species’ abundance in the Chrob- ry residential district (D = —0.40; Table 3). The yellowhammer {Emberiza citrine Ha), a species absent from the city of Toruh, was also a prey item, sug- gesting these owls also hunted in nonurban areas. In the suburban site, house sparrows breeding close to the forest contributed the largest proportion to the diet. Among forest bird species, tawny owls preferred the Eurasian tree sparrow (D = 0.56) and tits (D = 0.53) which nested only in the forest. Although the chaffinch {Fringilla coelebs) was a dom- inant species in the suburban forest, it was not taken in proportion to its availability by tawny owls (D = —0.41). The owls did not prey on chiffchaff (Phyl- 250 Andrzej Zalewski VoL. 28, No. 4 Table 3. Proportion of birds in tawny owl diet {%N = percent by number) and their relative abundance in the bird community in various habitats. (UGA = urban green areas [Zalewski and Przystalski 1993], OCh = the Chrobry residental district [Ozga 1990], SF = suburban forest [Zalewski 1994], + = occurring in the forest and breeding in the surrounding areas.) Urban Site Suburban Site Species %N UGA OCh %N SF Passer domesticus 58.2 + 37.7 39.1 + Passer montanus 10.7 2.9 1.6 8.7 2.4 Parus spp. 6.4 9.2 1.8 21.6 7.8 Fringilla coelebs — 10.3 1.2 8.7 19.1 Delichon urbica 5.6 — 12.7 — — Streptopelia decaocto 4.8 2.9 28.4 — — Sturnus vulgaris 4.8 19.5 2.8 — 3.4 Turdus merula 0.8 0.6 0.4 8.7 9.0 Erithacus rubecula — — — 4.4 4.3 Carduelis chloris 2.0 1.7 0.4 — 0.8 Phoenicurus phoenicurus 0.8 2.9 — — 0.8 P ochruros 0.8 — 0.6 — — Pica pica 0.4 2.3 0.6 — — Garrulus glandarius — — — 4.4 0.8 Serinus serinus 0.8 5.7 — — 3.5 Emberiza citrinella 0.4 — — — — Phylloscopus collybita — — — — 9.8 Sylvia atricapilla — — — — 6.2 Other passerines Density (pairs/10 ha) 3.5 42.0 150.0 11.8 183.4 4.4 32.1 75.4 loscopus collybita) and the blackcap {Sylvia atricap- illa), despite the fact that they were the most abun- dant species (D = —1; Table 3). Discussion In rural areas the diet of the tawny owl is com- posed mainly of small mammals (Bochehski 1990, Goszczyhski et al. 1993). In towns, mammals are less important dietary components (Manganaro et al. 1990, Galeotti et al. 1991, Goszczyhski et al. 1993). However, in both these environments the propor- tional contribution of mammals is lower in spring and summer while higher in autumn and in winter (Bochehski 1990, Manganaro et al. 1990, Galeotti et al. 1991). In Toruh, between February and June the proportion of Apodemus spp. (especially of A. agrarius) decreased. Similarly the proportion of Apo- demus spp. decreased in the diet in other studies of tawny owl diet in nonurban environments (Southern 1954, Yalden 1985, Kirk 1992). Apart from seasonal changes, annual variation in diet was recorded de- pending on rodent availability (Gruzdev and Lik- hachev 1960, Gbrner and Kramer 1973, Galeotti et al. 1991). Often, the alternative food source in the years of low rodent number was birds. This adapt- ability to alternative food may enable tawny owls to penetrate areas with few rodents but many birds. In Toruh, as in many other European cities, the major food of tawny owls was birds (Manganaro et al. 1990, Galeotti et al. 1991, Goszczyhski et al. 1993), particularly the house sparrow and other syn- anthropic species (collared dove and the house mar- tin). In urban areas house sparrows are also im- portant prey for sparrowhawks {Accipiter nisus; Primer 1989) and long-eared owls (Asia otus; von Dathe 1988). A considerable increase in avian prey in tawny owl diets was recorded in May and June, which coincided with arrival of migrants on their breeding territories in spring. Similar increases in the relative contribution of birds in the diet of these owls have been noted in other habitats (Gruzdev and Likhachev 1960, Glutz and Bauer 1980, Manganaro et al. 1990, Kirk 1992). In Toruh, the population density of house sparrows, a resident species, did not change during this study. The density of tits did change because these birds were more frequent in December 1994 Diet of Tawny Owls 251 0.122X+ 29.2 r= 0.545 P= 0.003 A/ =28 Percent of birds in diet Figure 1. Relationship between the mean prey size and proportion of birds in tawny owl diets in Europe (Schnurre 1961, Corner and Kramer 1973, Glutz and Bauer 1980, Wendland 1980, Yalden 1985, Cerveny and Obuch 1988, Bochehski 1990, Manganaro et al. 1990, Wasilewski 1990, Gramsz 1991). Only studies with at least 200 prey items were included. Torun in February and March than in later stages of the breeding season (D. Czeszczewik pers. comm,). However, in the suburban forest their importance increased in the diet from April to June. In suburban forests, Eurasian tree sparrows and tits were preferred by the owls, and were frequent in this owl’s diet in other European areas (Schnurre 1961, Wendland 1980, Bochehski 1990, Manganaro et al. 1990). Interestingly, tawny owls did not prey on chaffinches although this species is dominant in many forest areas (e.g., in the suburban forest it constitutes 19.2% of the bird community). The chaf- finch has been the predominant prey species of the tawny owl in only a few places (Southern 1954). Because tawny owls rely on hearing to locate prey (Southern 1954, Mikkola 1983), higher noise levels in urban habitats may interfere with prey capture. Also, because there are fewer rodents in town, the owls focus mainly on birds. Birds are probably more difficult to catch than the majority of nocturnal, ter- restrial, and vocally noisy small mammals. Other dietary studies of tawny owls in Europe show that increasing proportions of birds were correlated with an increase in mean size of prey items (r = 0.545, P = 0.003, Fig. 1). Hunting large prey may be an energetic compensation for diminished hunting time. In Torun, wild rabbits were an important addition to the owls’ diet, as were larger birds such as pigeons and ducks (Harrison 1960, Bogucki 1967, Galeotti et al. 1991). Thus, because the tawny owl can adapt to preying on alternative food sources and can tolerate the close presence of humans, the species may become in- creasingly abundant in towns. Acknowledgments I am indebted to Z. Pucek and B. J^drzeJewska for their comments on an earlier draft. Reviews by S.J. Petty and D. Kirk are gratefully acknowledged. Literature Cited Bochenski, JUN. Z. 1990. The food of suburban tawny owls on the background of birds and mammals occur- ring in the hunting territory. Acta Zool. Cracov. 33: 149-171. Bohme, G. 1977. Zur Bestimmung quartarer Anuren Europas an Hand von Skelettelmenten. Wiss. Z. Hum- boldt-Univ. Berl. Math.-Naturwiss. Reihe 26:283-300. Bogucki, Z. 1967. On the food of the tawny owls (6’^nA: aluco L.) nesting in the city of Poznan. Przegl. Zool 11:71-74. Cerveny, J. and J. Obuch. 1988. Small mammals in a diet of tawny owls in Sumava piedmont. Lynx 24:5- 14. Frimer, O. 1989. Food and predations in suburban sparrowhawks Accipiter nisus during the breeding sea- son. Dan. Ornitol. Foren. Tidsskr. 83:35-44. Galeotti, P., F. Morimando and G. Violani. 1991 Feeding ecology of the tawny owls {Strix aluco) in urban habitat (northern Italy). Boll. Zool. 58:143-150 Glutz von Blotzheim, U.N. and K. Bauer. 1980. Strix aluco Linnaeus 1758 — Waldkauze. Pages 579- 610 in Handbuch der Vogel Mitteleuropas. Akadem- ische Verlagsgesellschaft, Wiesbaden, Germany. Gorner, M. and V. Kramer. 1973. Ergebnisse einer elfjahrigen Nahrungskontrolle des Waldkauzes (Strix aluco L.) im Gebiet des Kottmar bei Ebersbach (Lau- sitzer Bergland). Abh. Ber. Naturkundemus. Gorlitz 48 1 - 8 . Goszczynski, j., P. Jablonski, G. Lesinski and J Romanowski. 1993. Variation in diet of tawny owls Strix aluco L. along an urbanization gradient. Acta Ornithol. 27:113-123. Gramsz, B. 1991. Tawny owls Strix aluco diet in oak horbean forest near Olawa (south-western Poland) Acta Ornithol. 26:3-13. Gruzdev, L.V. and G.N. Likhachev. 1960. Contri- bution to feeding habits of Strix aluco in the Tula Za- seki. Zool. Zh. 39:624-627. Harrison, C.J.O. 1960. The food of some urban tawny owls. Bird Study 7:236-240. Jacobs, J. 1974. Quantitative measurements of food 252 Andrzej Zalewski VoL. 28, No. 4 selection; a modification of the forage ratio and Ivlev’s electivity index. Oecologia 14:413-417. JuszczYK, W. 1987. Plazy i gady krajowe, PWN, War- szawa, Poland. Kirk, D.A. 1992. Diet changes in breeding tawny owls (Strix aluco). J. Raptor Res. 26:239-242. Manganaro, a., L. Ranazzi, R. Ranazzi and A. Sorace. 1 990. The diet of tawny owl, Strix aluco, in the urban park of Villa Pamphili (Rome). Riv. Ital. Ornitol. 60: 37-52. Mikkola, H. 1983. Owls of Europe. T. & A.D. Poyser, London, U.K. Moreno, E. 1985. Clave osteologica para la indentifi- cacion de los Passeriformes Ibericos. I Aegithalidae, Remizidae, Paridae, Emberizidae, Passeridae, Frin- gillidae, Alaudidae. Ardeola 32:295-377. Moreno, E. 1986. Clave osteologica para la indentifi- cacion de los Passeriformes Ibericos. II. Hirundinidae, Prunellidae, Sittidae, Cerothidae, Troglodytidae, Cin- clidae, Lanidae, Oriolidae, Strunidae, Motacillidae. Ardeola 33:69-129. OZga, W. 1990. Badania ilosciowe dwoch roznych sro- dowisk miejskich Torunia: Zespohi Staromiejskiego (1987-1989) oraz Osiedla Chrobrego (1989). M. Sc. thesis, Nicolaus Copernicus Univ., Torun, Poland. PucEK, Z. 1981. Keys to vertebrates of Poland, Mam- mals. PWN, Warszawa, Poland. ScHNURRE, O, 1961. Lebensbilder markischer Wald- kauze (Strix aluco L.). Milu 1:83-124. SOKAL, R.R. AND F.J. Rohlf. 1981. Biometry. W.H. Freeman, New York, NY U.S.A. Southern, H.N. 1954. Tawny owls and their prey. Ibis 96:384-410. ToMiAEOjfi, L. AND P. Profus. 1977. Comparative analysis of breeding bird communities in two parks of Wroclaw and in an adjacent Querco-Carpinetum for- est. Acta Ornithol. 16:117-177. VON Dathe, H. 1988. Uber die Ernahrung einer Wal- dohreule, Asio otus, inmitten der GroBtadt Berlin. Beitr Vogelkd. 34:41-46. Wasilewski, j. 1990. Dynamics of the abundance and consumption of birds of prey in the Niepolomice Forest. Acta Zool. Cracov. 33:173-213. Wendland, V. VON. 1980. Der Waldkauze (Strix aluco) im bebauten Stadtgebiet von Berlin (West). Beitr. Vo- gelkd. 26:157-171. Yalden, D.W. 1985. Dietary separation of owls in the Peak District. Bird Study 32:122-131. Zalewski, A. and A. Przystalski. 1993. Quantitative investigations on the breeding avifauna of green areas of Torun in 1987-1989. Acta Univ. Nicolai Copernici Biol. 43:105-124. Zalewski, A. 1994. A comparative study of breeding bird populations and associated landscape character, Torun, Poland. Landscape Urban Plann. 29:31-34. Received 28 January 1994; accepted 7 July 1994 J Raptor Res. 28(4):253-258 © 1994 The Raptor Research Foundation, Inc. DIET COMPOSITION OF THE LONG-EARED OWL IN CENTRAL SLOVENIA: SEASONAL VARIATION IN PREY USE Davorin Tome Institute of Biology, University of Ljubljana, Karlovska 19 - POB 141, 61000 Ljubljana, Slovenia Abstract. — The diet of long-eared owls {Asio otus) was examined in central Slovenia during 1989-91. Mammals were the most common prey group (97% by number), with the common vole {Microtus arvalis) as the most frequent prey species (46%); its proportion in the diet varied yearly and seasonally (range 11-90%) according to its population density. When this vole’s densities were low, owls shifted their predation pressure to alternate prey. Mice from the genus Apodemus were the most common alternate prey. Their proportion in the diet increased during winter and during year-round shortages of main prey as well. Other prey species contributed significantly only during winter (Microtus agrestis and Pitymys subterraneus) or only during year-round shortages (Arvicola terrestris and birds). Body size of the three prey species with highest proportions in the pellets (M. arvalis, M. agrestis, and Apodemus sylvaticus) varied with the season, the largest being during summer and the smallest during winter. Key Words: Asio otus; diet composition; long-eared owl; prey size; seasonal variation; Slovenia. Composicion de la dieta de Asio otus en Eslovenia Central: variacion estacional en el uso de presas Resumen. — Durante 1989 a 1991 se examino la dieta de Asio otus en Eslovenia Central. Los mamiferos constituyeron el grupo de presas mas comun (97% por numero) y cuya especie-presa mas frecuente fue Microtus arvalis (46%). La proporcion de M. arvalis en la dieta varia anual y estacionalmente (rango: 11 a 90%) de acuerdo a su densidad poblacional. Cuando sus densidades fueron bajas, A. otus incrementaba la depredacion sobre presas alternativas; su proporcion en la dieta se incrementaba durante el invierno y en los periodos de baja densidad intra-anual de la presa principal. Otras especies-presa contribuyeron significativamente a la dieta solamente en invierno (Microtus agrestis, Arvicola terrestris y aves). El tamaho corporal de tres especies-presa, con altas proporciones en las egagropilas (M. arvalis, M. agrestis y Apodemus sylvaticus, variaba con la estacion, siendo mas grandes en verano y mas pequehas durante el invierno. [Traduccion de Ivan Lazo] The diet of the long-eared owl (Asio otus) has been studied extensively throughout Europe and North America (summarized in Schmidt 1974, Marti 1976, Mikkola 1984, Cramp 1985). Due to difBculties in finding pellets during the breeding season, however, many studies have a bias toward the winter diet. Some authors tried to overcome this by presenting results of pellet analyses separately for winter and summer or breeding and nonbreeding season (Arm- strong 1958, Sulkava 1965, Glue and Hammond 1974, Goszczinsky 1981). Only Nilsson (1981) and Wijnandts (1984) evaluated long-eared owl diet year round. I present data on long-eared owl diet derived from pellets collected throughout the year in central Slo- venia. My major goal was to evaluate seasonal dif- ferences in prey species composition and prey size. Methods and Materials Long-eared owl pellets were collected systematically from January 1989 to December 1991 at three localities on Ljubljana moor (south of the city of Ljubljana, Slovenia) The study area (about 160 km^) was homogenous farm- land. In localities where pellets were collected forest patch- es of predominantly Pino sylvestris-Betulletum and Betulo- Ojaercetum roboris types (for a more detailed description see Tome 1991). Sampling at each locality took place at least once a month. Small mammal remains in the pellets were identified to species according to Krystufek (1985) Birds and insects were not identified beyond class because of their low numbers. Biomass was calculated using av- erage mass of prey species (see Tome 1991). For study of year-round variation in the diet, prey (by number) were pooled for each year into six 2-mo periods beginning with January. Due to their small proportions and low variability in the diet, species in the genera Mt- cromys, Rattus, and Apodemus were pooled as “mice,” and species in the genera Sorex, Neomys, Crocidura as “shrews.” 253 254 Davorin Tome VoL. 28, No. 4 Table 1. Diet composition of the long-eared owl in central Slovenia {N% = dietary percent by number, B% = dietary percent by biomass). 1989 1990 1991 1989-91 Prey N% N% N% N% B% Microtus arvalis 66.6 52.3 21.2 44.2 45.6 Microtus agrestis 13.3 17.8 11.6 14.3 20.2 Pitymys subterraneus 7.1 8.3 10.2 8.7 6.0 Clethrionomys glareolus 2.5 1.2 3.8 2.5 2.4 Arvicola terrestris tr.^ 0.4 5.7 2.3 5.0 Apodemus sylvaticus 5.1 7.1 14.4 9.4 7.7 Apodemus flavicollis tr. 2.1 0.3 0.9 0.8 Apodemus spp. 2.2 5.6 15.5 8.5 6.9 Micromys minutus 0.9 2.9 7.5 4.1 1.2 Rattus norvegicus tr. tr. 0.1 tr. 0.1 Sorex araneus tr. 0.3 2.4 1.0 0.4 Neomys fodiens tr. tr. 0.1 tr. tr. Crocidura leu codon 0.9 0.3 0.1 0.4 0.2 Crocidura suaveolens tr. 0.1 0.2 0.1 tr. Crocidura spp. 0.1 tr. 0.2 0.1 tr. Muscardinus avellanarius tr. 0.2 0.4 0.2 0.3 Talpa europaea tr. 0.1 0.4 0.2 0.8 Total mammals 98.7 98.7 94.1 97.0 97.7 Birds 1.2 1.1 4.8 2.5 2.3 Insects 0.1 0.2 1.1 0.5 tr. Total number Total biomass (g) 691 921 999 2611 60 726 » <0.1%. Dormice {Muscardinus avellanarius) , moles {Talpa euro- paea), and insects were grouped together as “other.” Food-niche breadth (FNB) was calculated according to Levins (1968). In these calculations unidentified individ- uals in the genera Apodemus and Crocidura were assigned to species in the same proportions as their identified coun- terparts. Birds and insects were regarded as only two tax- ons. An index of the size of prey individuals of the three most frequent prey species in the diet (common vole [Mi- crotus arvalis], field vole [Microtus agrestis], and wood mouse [Apodemus sylvaticus]) was obtained by measuring the dis- tance between the upper incisor and the third molar (IM3) on one side of each unbroken skull. Measurements were taken to the nearest 0.1 mm using a caliper. To assess relative abundance of dominant prey species in the field, snap traps were set in spring and early sum- mer. Traps were placed 5 m apart in lines of 30 and left for one night. Altogether 1290 trap nights were accu- mulated on the grasslands in 3 yr. A snap-trap index (STI = number of animals caught per 100 traps) was used to determine density. Results Diet Composition. Fifteen species of small mam- mals were found in pellets of the long-eared owl from Ljubljana moor with the common vole being the most frequent (>40% by number). Among major prey groups, voles {Microtus, Pitymys, Clethrionomys, and Arvicola) were dominant, constituting 72% of prey items by number, followed by mice {Apodemus, Micromys, Rattus; 23%), birds {Aves; 3%) and shrews {Sorex, Neomys, Crocidura', 2%). The proportion of insects was negligible (<1%). Proportions of prey species by biomass were similar to proportions by number, because of similar average weights of the most frequent prey species (Table 1). Year-to-year diet of the owls changed consider- ably (x^ = 564, P < 0.01). Species most variable in the diet were common voles and mice from the genus Apodemus, followed by water voles {Arvicola terres- tris), harvest mice {Micromys minutus), and birds. Proportions of field voles, common pine voles {Pi- tymys subterraneus) , and bank voles {Clethrionomys glareolus) were more stable (Table 1). Seasonal Variation in Prey Use. Most promi- nent in the seasonal variation in prey use were the summer-autumn peaks of the common vole, which December 1994 Long-eared Owl Prey Use 255 ■ Birds M. arvalis M M. agrestis P. subterraneus C. g/areo/us A. terrestris m Mice iii Shrews r •: Other 100 % 80% 60% 40% 20 % 0 % 1989 1990 1991 JF MA MJ JA SO ND JF MA MJ JA SO ND JF MA MJ JA SO ND 162 233 98 110 19 69 210 279 238 96 50 48 125 248 202 147 95 182 Figure 1. Seasonal variation in the diet (proportions by number) of the long-eared owl during 1989-91 in central Slovenia in 2-mo intervals. Sample size is given below x-axis intervals. constituted up to 90% of food intake by number in that period. The winter diet shifted notably from common voles to mice, field voles, and common pine voles. In 1991, mice were taken more frequently throughout the year, but field and common pine voles did not surpass the proportions found in pellets in previous years. During the summer of 1991, the proportions of water voles and birds markedly in- creased (Fig. 1). The relationships between main and alternate prey in the diet were investigated using correlations be- tween proportions of the species in the diet and the FNB. Optimal foraging theory predicts that FNB should expand when the density of the main prey species decreases and shrink when the main prey increases (Pyke 1984). This means that the most important main prey species have the largest neg- ative correlation coefficient and the most important alternate prey have the highest positive correlation. FNB was usually low during the summer and high during the winter, but in 1991 it was high throughout the year (Fig. 2). Correlation between the proportion of species in the diet and the FNB revealed that the common vole was the only main prey for long-eared owls. Wood mice were the most important alternate prey, followed by water voles and shrews (Table 2). Population Density of Common Voles. Density of common voles, the main prey species and the most Table 2. Spearman rank correlations between propor- tions of species found in pellets of long-eared owls and the food-niche breadth of the owls’ diet. Species Microtus arvalis -0.97a Microtus agrestis 0.21 Pitymys subterraneus 0.48 Clethrionomys glareolus 0.52 Arvicola terrestris 0.68a Apodemus spp. 0.84a Micromys minutus 0.56 Soricidae 0.66a Birds 0.56 < 0 . 01 . 256 Davorin Tome VoL. 28, No. 4 JF MA MJ JA SO ND JF MA MJ JA SO ND JF MA MJ JA SO ND Figure 2. Seasonal variation in the food-niche breadth (FNB) in the long-eared owl in central Slovenia in 2-mo intervals. variable one in the long-eared owls’ diet, varied widely in the field. It was highest in 1989 (STI = 21.03), intermediate in 1990 (STI = 11.71), and the lowest in 1991 (STI = 0.26). Seasonal Variation in Prey Size. Body size as estimated by the average IM3 distance varied sig- nificantly between 2-mo periods in the three most common prey species (ANOVA; common vole, F = 9.50, P < 0.01; field vole, F = 6.84, P < 0.01; and wood mouse, F = 5.95, P < 0.01). This measure- ment was greater during the summer than during the winter, being shortest in the last third of the year (September to December; Fig. 3). Discussion Diet of the long-eared owl in this study was sim- ilar to diets elsewhere in Europe (summarized in Schmidt 1974, Marti 1976, Mikkola 1984, Cramp 1985). Small mammals contributed a majority of prey items by number (97%) and by mass (98%). The rest were birds (2%) and insects (1%). FNB index was, in contrast to Sweden (Nilsson 1981) and the Netherlands (Wijnandts 1984), higher during winter than during the summer. In years of its abundance, the common vole was by far the most important species in the summer and autumn diet. During the winter and spring, as well as during the summer and autumn in 1991 when densities of the common vole were low, proportions of alternate prey in the diet increased notably. Mice were the most important alternate prey because their proportion increased during winter as well as during year-round shortages of common voles. Other species contributed significantly only during winter (field vole and common pine vole) or only during year- round shortages of common voles (water vole and birds) clearly diminishing their importance as al- ternate prey. In Sweden, dense vegetation in summer presum- ably reduced the availability of voles in relation to other prey which resulted in increased proportions of alternate prey and also increased FNB in the long- eared owl (Nilsson 1981). Open habitats on Lju- bljana moor (main hunting habitat of long-eared owls; Tome 1991) regularly had low vegetative cover during winter. Snow cover during this study was practically nonexistent. Consequently heavy vege- tative or snow cover could not have been the reason for increased FNB on Ljubljana moor during winter. It is well-known that populations of small mam- mals are lowest during winter (Petrusewicz 1983, Tamarin 1985). On the other hand, the proportion of the common vole in the diet of the long-eared owl is dependent on the abundance of this species in the owls’ hunting habitat (Korpimaki 1992). I suggest that low density of common voles in open habitats during winter, as well as during most of 1991, was the main reason for decreased proportion of this December 1994 Long- EARED Owl Prey Use 257 Micro tus arvafis (N = 781) E E 16 15.5 15 14.5 14 13.5 Micro tus agrestis (N = 299) JF MA MJ JA SO ND A. sy/vaticus (N = 191) Figure 3. Average distance between the upper incisor and the third upper molar in small mammals found in pellets of long-eared owls during 1989-91. Results are pooled according to date of origin into 2-mo periods. Vertical lines show standard deviation. 258 Davorin Tome VoL. 28, No. 4 species in the diet and in consequence for increased FNB. How the variable body sizes of prey species influences the diet of the long-eared owl is still to be evaluated. Literature Cited Armstrong, W.H. 1958. Nesting and food habits of the long-eared owl in Michigan. Mich. State Univ. Publ. Mus. Biol. Ser. No. 1. Cramp, S. [Ed.] 1985. The birds of the western Pale- arctic. Vol. 4. Oxford Univ. Press, New York, NY U.S.A. Glue, D.E. and G.J. Hammond. 1974. Feeding ecology of the long-eared Owl in Britain and Ireland. Br. Birds 67:361-369. Goszgzinsky, J. 1981. Comparative analysis of food of owls in agrocenoses. Ekol. Pol. 29:431-439. Korpimaki, E. 1992. Diet composition, prey choice, and breeding success of long-eared owls: effects of mul- tiannual fluctuations in food abundance. Can. J. Zool. 70:2373-2381. Krystufek, B. 1985. Mali sesalci. Nasa rodna gruda. Prirodosl. drustvo Slovenije. Ljubljana, Slovenia. Levins, R. 1968. Evolution in changing environments. Princeton Univ. Press, Princeton, NJ U.S.A. Marti, C.D. 1976. A review of prey selection by the long-eared owl. Condor 78:331-336. Mikkola, H. 1984. Owls of Europe. T. & A.D. Poyser, Staffordshire, U.K. Nilsson, I.N. 1981. Seasonal changes in food of the long-eared owl in southern Sweden. Ornis Scand. 12. 216-223. Petrusewicz, K. [Ed.] 1983. Ecology of the bank vole. Acta Theriol. 28:1-242. Pyke, G.H. 1984. Optimal foraging theory: a critical review. Annu. Rev. Ecol. Syst. 15:523-575. Schmidt, E. 1974. Die ernahrung der Waldohreule (Asio otus) in Europa. Aquila 81:221-238. Sulkava, P. 1965. Vorkommen und Nahrung der Wal- dohreule, Asio otus (L.) in Ilmajoki (EP) in den Jahren 1955-1963. Aquilo Ser Zool. 2:41-47. Tamarin, R.H. [Ed.] 1985. Biology of new world Mi- crotus. Am. Soc. Mammal., Spec. Publ. 8. Stillwater, OK U.S.A. Tome, D. 1991. Diet of the long-eared owl (Asio otus) in Yugoslavia. Ornis Fenn. 68:114-118. Wijnandts, H. 1984. Ecological energetics of the long- eared owl (Asio otus). Ardea 72:1-92. Received 7 March 1994; accepted 6 July 1994 Short Communications J. Raptor Res. 28(4):259-262 © 1994 The Raptor Research Foundation, Inc. Breeding Density and Brood Size of Rough-legged Hawks in Northwestern Quebec Serge Brodeur G.R.E.B.E. Inc., 2045 Stanley, Montreal, QJUE H3A 2V4, Canada, and Avian Science and Conservation Centre, Macdonald Campus of McGill University, 21,111 Lakeshore Road, Ste. Anne de Bellevue, QJUE H9X 3V9, Canada Francois Morneau and Robert DfecARiE G.R.E.B.E. Inc., 2045 Stanley, Montreal, QJUE H3A 2V4, Canada Juan J. Negro and David M. Bird Avian Science and Conservation Centre, Macdonald Campus of McGill University, 21,111 Lakeshore Road, Ste. Anne de Bellevue, QUE H9X 3V9, Canada Key Words: Buteo lagopus; habitat; Hudson Bay; nest; reproduction; rough-legged hawk; survey. Rough-legged hawks {Buteo lagopus) breed in northern Canada and are especially common around Hudson Bay, Ungava Bay, the Labrador coast and the Arctic (Todd 1963, Palmer 1988). However, most nesting observations in eastern Canada are anecdotal and no information is available on breeding densities and reproduction of the species in the province of Quebec. Proposed hydroelectric development in northwest Quebec has increased interest in surveying raptor populations in that area. Field surveys are needed to accurately assess the potential impacts of these projects. We report the results of surveys of nesting rough-legged hawks in the Hudson Bay region of north- western Quebec. Our aim is to provide baseline data on the distribution, density, brood size, and nest-site char- acteristics of rough-legged hawks in the Hudson Bay re- gion of Quebec. Study Area and Methods The surveys were conducted in parts of the basins of the rivers Grande Riviere de la Baleine and Petite Riviere de la Baleine, and Guillaume-Delisle Lake (55-57°N, 74- 78“W), covering approximately 9850 km^ (Fig. 1). The study area encompassed boreal forest, boreal forest-tundra ecotone, and a coastal strip of arctic tundra (Payette 1983). The dominant vegetation was black spruce {Picea mariana) with the density of trees decreasing from south to north. The rugged landscape has numerous inland rocky outcrops and coastal cliffs. A more thorough description of the study area is given in Morneau et al. (1994). Prior to the surveys, cliffs were located on 1:50 000 scale maps of the study area by finding where contour lines were very close to one another. During the surveys all vertical rock faces, including those not identified on the maps, were visited. Steep rocky hills were not surveyed No attempts were made to locate nests in forests because tree nesting by rough-legged hawks, while occurring in other regions (Palmer 1988) has never been reported in the area. The cliffs were surveyed aboard an A Star 350 helicopter at a distance of 20 m at 30-70 km/hr. Several passes were made along high cliffs, the helicopter being flown successively lower at each pass (Kochert 1986, Ful- ler and Mosher 1987). Sixty hours were spent flying be- tween 10-20 July 1990, with two to three observers plus the pilot. Flights were only conducted when weather con- ditions were favorable (i.e., no fog or precipitation). We plotted nest locations on a map and recorded number of eggs or nestlings present as well as nest orientation and height. In estimating mean brood size, only nests with chicks over 1 wk old and no eggs were considered. Nest- lings were aged according to criteria established by Mor- itsch (1983). We also recorded the nests of other cliff- nesting raptors (golden eagle [Aquila chrysaetos], peregrine falcon [Ealco peregrinus], red-tailed hawk [Buteo jamaicen- jA] and great horned owl [Bubo virginianus]). Results Rough-legged hawks were the most abundant cliff-nest- ing bird of prey in the study area. During the surveys, 49 nests containing at least one nestling or nestlings and eggs {N = 4) were located. Fifteen golden eagle (see Morneau et al. 1994), one peregrine falcon, six red-tailed hawk, and two of great horned owl nests were found. The minimum density of rough-legged hawks was thus one breeding pair per 201 km^. In addition, 70 empty Buteo nests were found, some of which may have been alternate nests and nests 259 260 Short Communications VoL. 28, No. 4 LEGEND A Nest location 1^] Study area Hudson Bay OKT Hudson Bay ' ( Umiujaq ^a Gmfiaume'Dehslo Lake 0 Ukm 1 * j I December 1994 Short Communications 261 used in previous years or by pairs that failed early in the season (Palmer 1988). Some empty nests may have been built by red-tailed hawks, although cliff nests built by the latter were less abundant than rough-legged hawk nests and were in the boreal forest south of the Petite Riviere de la Baleine. The distribution of nests was not homo- geneous within the study area, and most of them were located along the Hudson Bay coastline (see Fig. 1). The mean nearest-neighbor distance between rough-legged hawk nests was 4.95 km (SD = 3.82, = 48 nests). The minimum distance between two active nests was 250 m, and another eight nests were less than 1 km apart. Min- imum distances to active nests of other cliff-nesting raptors were 5.75 km for the peregrine falcon, 4.00 km for the golden eagle, 1.00 km for the great horned owl and 7.75 km for the red-tailed hawk. Rough-legged hawk nests were located at 6-90 m from the bottom of the cliff (mean = 19, SD = 15.4, N = 49 nests). Thirty-four (69.4%) nests faced south, southeast, or southwest. Of the rest, 13 (26.5%) faced north or northeast and two (4.1%) faced east. Five (10.27o) of the nests had overhangs. The mean brood size was 3.41 nestlings (SD = 1.09, range 1-6, N = 44 nests). The frequencies of brood sizes were as follows; one nestling (N = one nest), two {N = 9), three {N — 12), four {N — 16), five {N = 5), six (N =1). During the surveys, most nestlings (36 of 48 broods) were covered in down and were estimated to be 1-2-wk-old. In addition, 1 1 nests contained young 3-wk-old and one nest contained young 4-wk-old. Discussion Local populations of rough-legged hawks are known to fluctuate greatly in numbers from year to year (Mindell et al. 1987, Mindell and White 1988, Palmer 1988), and a single year survey did not allow us to assess whether the density (i.e., one breeding pair per 201 km^) is typical for the area. Another rough-legged hawk population in the central Canadian Arctic varied from one breeding pair per 62.5 km^ to one pair per 333.3 km^ in a 4-yr study (Poole and Bromley 1988). Higher densities (1 pair/30-50 km^) have been reported for forest dwelling rough-legged hawks in Finland (Pasanen 1972 in Palmer 1988). These com- parisons, on the other hand, can be misleading, as the area that we surveyed was much larger than in the other studies. Moreover, the distribution of nesting pairs occurred in clusters in our study, and densities were much higher within those clusters. Most rough-legged hawk nests were confined to the cliffs of the Hudson Bay coastline and adjacent islands, even though there were numerous cliffs in other parts of the study area. Irregular spacing of nests was also observed in the central Canadian Arctic by Poole and Bromley (1988), who suggested the distribution of rough-legged hawks was affected by the patchy distribution of microtine rodents. Nest-site characteristics that we observed were similar to those of previous studies (White and Cade 1971, Poole and Bromley 1988). Most nests in our study area lacked overhangs and most faced in a southerly direction. Orientation of the coastal relief and adjacent islands (SW- NE) provides abundant southeast- and east-facing rock faces in the western part of the study area but that pattern of relief is not present inland where cliffs face all directions. Nonetheless, other factors, such as the tempering effect of the Hudson Bay on temperature may also be involved in the irregular distribution of rough-legged hawks in our study area. In our study, the minimum distance between nests of rough-legged hawks and other raptor species was larger than the minimum distance between nests of rough-legged hawks. This differs from Poole and Bromley (1988) and Schmutz et al. (1980) who found greater evidence for competition for space within than among species of raptors The mean brood size of 3.41 nestlings was higher than those found in other areas (2.8 [Sealy 1966] and 2.0-3 5 [Poole and Bromley 1988] in the Northwest Territories, Canada; 3.0 in Norway [Hagen 1952 in Palmer 1988], 2.0 in Finland [Pasanen 1972 in Palmer 1988]). According to Poole and Bromley (1988), the decline in brood size and in other reproductive parameters was correlated with the decrease in microtine numbers. Therefore, our data may have been collected in a good food year. Because the survey was conducted when the nestlings were relatively young, mortality of young may have occurred before fledg- ing and we cannot estimate nest success or productivity. Resumen. — Utilizando un helicoptero se prospecto un area de 9850 km^ en el noroeste de Quebec para censar nidos de Ratonero Calzado (Buteo lagopus). Se observaron 49 nidos que contenian al menos un joven (1 nido ocupado por 201 km^). Los Ratoneros Calzados eran las aves de presa mas abundantes en la zona. La mayoria de los nidos se concentraban en los acantilados de la Bahia de Hudson y en los de las islas adyacentes, a pesar de que tambien habia numerosas paredes rocosas en otras partes del area de estudio. El 69.4% de los nidos se orientaba hacia el sur y solo un 10% estaba protegido por extraplomos. El tamano medio de nidada, 3.41 polios, fue similar al encontrado en otras poblaciones. [Traduccion Autores] Acknowledgments This study was part of the Grande-Baleine hydroelectric project studies undertaken by Hydro-Quebec, and part of the North American Peregrine Falcon five-year survey coordinated by M. Lepage (Min. de I’Environnement et de la Faune). We wish to thank R. McNicoll, P. Lafon- taine, J. Carrier, R. Dulong and R. Julien for helping in the fieldwork. Gary Bortolotti, Josef Schmutz, and Ted Swem helped to improve the manuscript by their careful review. Literature Cited Fuller, M.R. AND J. A. Mosher. 1987. Raptor survey techniques. Pages 37-65 in B.A. Giron Pendleton, B.A. Millsap, K.W. Cline, and D.M. Bird [Eds.], Raptor Figure 1. Study area and location of occupied nests of rough-legged hawks in 1990. 262 Short Communications VOL. 28, No. 4 management techniques manual. Natl. Wildl. Fed., Washington, DC U.S.A. Kochert, M.N. 1986. Raptors. Pages 313-349 m A. Y. Cooperrider, R.J. Boyd and H.R. Stuart [Eds.], In- ventory and monitoring of wildlife habitat. USDI. Bur. Land Manage., Denver, CO U.S.A. Mindell, D.P., J.L.B. Albuquerque and C.M. White. 1987. Breeding population fluctuations in some rap- tors. Oecologia 72:382-388. AND C.M. White. 1988. Fluctuations of ob- served breeding rough-legged hawks and gyrfalcons: regularity reconsidered. Oecologia 77:14-18. Moritsch, M.Q. 1983. Photographic guide for aging nestling red-tailed hawks. USDI Bur. Land Manage., Boise, ID U.S.A. Morneau, F., S. Brodeur, R. D^carie, S. Carri^;re AND D.M. Bird. 1994. Abundance and distribution of nesting golden eagles in Hudson Bay, Quebec. /. Raptor Res. 28:220-225. Palmer, R.S. [Ed.]. 1988. Handbook of North Amer- ican birds. Vol. 5. Yale Univ. Press, New Haven, CT U.S.A. Payette, S. 1983. The forest tundra and present tree- lines of the northern Quebec- Labrador Peninsula. Pages 3-23 in P. Morisset and S. Payette [Eds.], Tree-line ecology. Proc. northern Quebec tree-line conference. Centre d’etudes nordiques Quebec, Qc, Canada. Poole, K.G. and R.G. Bromley. 1988. Interrelation- ships within a raptor guild in the central Canadian Arctic. Can. J. Zool. 66:2275-2282. Sealy, S.G. 1966. Notes on the rough-legged hawk in the Perry River region. Northwest Territories. Blue Jay 24:127-128. ScHMUTZ, J.K., S.M. SCHMUTZ AND D.A. Boag. 1980. Coexistence of three species of hawks (Buteo spp.) in the prairie-parkland ecotone. Can. J. Zool. 58:1075- 1089. Todd, W.W.C. 1963. Birds of the Labrador Peninsula and adjacent areas. A distributional list. Univ. Toronto Press. Toronto, ON, Canada. White, C.M. and T.J. Cade. 1971. Cliff-nesting rap- tors and ravens along the Colville River in arctic Alas- ka. Living Bird 10:107-150. Received 11 April 1994; accepted 18 July 1994 J Raptor Res. 28(4):262-265 © 1994 The Raptor Research Foundation, Inc. Effects of Prescribed Fires on Habitat Use by Wintering Raptors on a Texas Barrier Island Grassland Felipe Chavez- Ramirez Department of Wildlife and Fisheries Sciences, Texas A&M University, College Station, TX 77843-2258 U.S.A. Felipe G. Prieto Matagorda Island National Wildlife Refuge, P.O. Box 100, Austwell, TX 77950 U.S.A. Key Words: American kestrel-. Circus cyaneus; Falco sparverius; fire; habitat use; northern harrier; Texas. The structure of vegetation at ground level in grasslands appears to be of greater importance than prey abundance in selection of hunting sites by birds of prey (Janes 1985, Preston 1990). Land management practices that affect raptor habitat will, therefore, be those that affect vege- tation structure rather than impact solely prey abundance (Millsap et al. 1987). Fire can affect raptors primarily by altering the quality of habitat parameters such as cover and prey availability. Burning and maintaining grasslands provides habitat for northern harriers {Circus cyaneus) and short-eared owls {Asio flammeus) (Hamerstrom 1974) dur- ing the breeding season. Considerably less is known about the effects of prescribed fires on winter habitat use by raptors. In monitoring the effects of fire on animal communities researchers have generally considered periods of one or more years post-burn (Peterson and Best 1987, Pylypec 1991). Considerably less effort has been made to monitor the immediate effects of fire on wildlife communities. As suitable wintering habitats become limiting, understand- ing the immediate effects the management practices that December 1994 Short Communications 263 Table 1. Raptor species and number of individuals observed during winter road surveys on Matagorda Island, Texas during 1992-93 winter. December January February 2 26 17 29 7 Crested caracara 6 3 5 5 3 {Caracara plancus) American kestrel 9 12 15 6 21 {Falco sparverius) Peregrine falcon 1 0 0 0 0 (P. peregrinus) Merlin 0 0 1 0 0 (P. columbarius) Northern harrier 31 30 38 37 52 {Circus cyaneus) Black-shouldered kite 17 7 6 2 2 {Elanus caeruleus) White-tailed hawk 5 4 7 12 4 {Buteo albicaudatus) Red-tailed hawk 2 2 3 2 6 {B. jamaicensis) Total 71 58 75 64 88 drastically alter vegetation structure (e.g., prescribed fire or shredding) have on habitat use patterns by migrant birds becomes increasingly important. We studied raptor use of a natural grassland immediately before and after winter prescribed burns were applied. Our objective was to identify changes in species composition and degree of raptor use shortly after the burning of a grassland. Methods The study was conducted on Matagorda Island Na- tional Wildlife Refuge and State Natural Area in Calhoun County, Texas. This barrier island is approximately 4 x 61 km and has an area of 22 934 ha. Vegetation types on the island include beach dune complex, upland barrier flats, and saltmarsh (Blankenship 1990). The interior up- land barrier flats of the island is a grassland dominated by marshhay cordgrass {Spartina patens), gulfdune pas- palum (Paspalum monostachyum), and seacoast bluestem {Schizachyrium scoparium) with sparse woody vegetation composed of mesquite (Prosopis juiiflora) and false willow {Baccharis neglecta). As part of ongoing burn studies, two 140-ha plots were burned on 4-5 January 1993. Raptor use at each of the two plots was evaluated by conducting 1 -hr counts at week- ly intervals, between 13 December 1992 and 14 February 1993 — 4 wk pre-burn to 4 wk post-burn. Observers walked slowly through the middle of each plot during the count and all raptors noted within the study plot were recorded. Care was taken to track raptors while in the plots to avoid double counting individuals. Five road surveys were con- ducted biweekly, from 2 December to 7 February, to de- termine presence and relative abundance of raptors win- tering on the island during the study period, Road surveys were 58 km long and were conducted along the single road that runs lengthwise through the middle of the island Surveys were conducted during favorable weather (good visibility, no rain or extreme winds) and started between 0800 and 0900 H. Kendall’s tau measure of rank correlation (Conover 1980) was used to evaluate changes in numbers of total raptors, northern harriers, and American kestrels (Falco sparverius) observed in road surveys. Differences in raptor use (ex- cluding Cathartidae) between plots and pre- and post- burn within plots were evaluated with Mann-Whitney t/-tests (Conover 1980). Similarity in raptor species com- position was compared between plots pre-burn, post-burn, and pre- and post-burn within plot using the chord dis- tance method (Ludwig and Reynolds 1988). Chord dis- tance (CD) values range from 0 (maximum similarity) to 1.41 (maximum dissimilarity). Results Vegetative cover on study plots was reduced from about 100% ground cover pre-burn to nearly 0 after burns. Based on road surveys (Table 1), the total number of raptors increased positively (t = 0.5) throughout the study period, however, not significantly so {P > 0.05). Total number of raptors observed at both plots pre- and post-burn were similar (Table 2) with slight and moderate decreases ob- served at plots B and A, respectively. Non-significant dif- ferences were observed pre- and post-burn between plots (pre-burn U = 18.5, df = 15, P > 0.05, post-burn U = 24, df = 15, P > 0.05) so data from both plots were pooled. Number of total raptors observed post-burn were not sig- nificantly different from pre-burn numbers (6/ = 72.5, df = 15, P > 0.05). Pre- and post-burn species composition 264 Short Communications VoL. 28, No. 4 Table 2. Raptor species and number of individuals observed during pre-burn (Pre) and post-burn (Post) surveys and percent change at two study plots on Matagorda Island, Texas. Plot A Plot B Pre Post % Change Pre Post % Change Crested caracara 0 1 0 0 American kestrel 1 4 + 300 1 6 + 500 Peregrine falcon 0 0 0 1 Merlin 1 0 1 1 Northern harrier 17 8 -53 22 14 -36 White-tailed hawk 0 0 1 1 Short-eared owl 1 0 0 0 Total 20 13 -35 25 23 -8 of communities at each plot were the most dissimilar (CD = 0.42 plot A, CD = 0.36 plot B), while raptor com- munities were most similar between plots pre-burn (CD = 0.08) and post-burn (CD = 0.17). Only northern harriers and American kestrels were ob- served in sufficient numbers to evaluate changes post-burn and were the most common raptors observed in road sur- veys and study plots (Tables 1 and 2). Harriers and kes- trels showed positive increases (t — 0.6 and 0.4, respec- tively) throughout the study period, but increases were not statistically significant {P > 0.05). In study plots northern harriers decreased significantly (67=85, df=15,P< 0.05), by 53% (plot A) and 367o (plot B) post-burn, while American kestrels increased significantly (6/ = 49, df = 15, P < 0.05), by 3007o (plot A) and 500% (plot B). Discussion Winter burns on Matagorda Island did not appear to have short-term effects overall on wintering raptor num- bers. The total number of wintering raptors using an area pre- and post-burn did not change significantly; however, species composition did, primarily due to changes in num- ber of harriers and kestrels. The decreases in northern harriers on burned plots is contrary to the increasing numbers observed in road survey counts. The decrease in harriers post-burn was likely due to the elimination of vegetative structure, since harriers are surprise hunters flying low over vegetation, then pounc- ing on prey (Schipper et al. 1975). Harrier hunting success is lowered in areas with dense vegetation compared to areas with sparse vegetation (Collopy and Bildstein 1987). However, as observed in this study, complete elimination of vegetative cover significantly decreased habitat use by harriers. Harriers, and other raptors, have been previously reported to avoid bare ground for hunting (Preston 1990). The increase in American kestrels post-burn may cor- respond to the slight increases observed on road surveys, but may be due to the elimination of vegetative structure, since they prefer to hunt in open habitats with low veg- etative structure (Smallwood 1987). Hunting success by kestrels decreases significantly with increasing vegetation height (Toland 1987). Increased use by kestrels post-burn on Matagorda Island may have been in response to the recent disturbance created by the fire. Kestrels have re- sponded positively to human-related disturbances, such as crop harvesting (Toland 1987), irrigation (Rudolph 1982), and controlled fires (Smallwood et al. 1982). Our results show that wintering harriers and kestrels can be affected by prescribed fires. The effect of burns on habitat use patterns, however, needs further study. For example, do harriers and kestrels change territory size and/ or shape after habitat modifications occur from burn- ing? Do raptors alter hunting and foraging behavior in response to burned areas? More intense studies are nec- essary to evaluate changes in habitat use patterns, before and after burns. Resumen. — Estudiamos los efectos de quema controlada en el uso de habitat por rapaces invernantes en un pastizal natural en la costa de Texas. Dos areas de 140 ha se selecionaron para estudio y se quemaron el 4 y 5 de Enero de 1993. Ocho de diez rapaces invernantes y residentes utilizaron las areas de estudio. Los numeros de rapaces totales sufrieron cambios moderados en ambas areas des- pues de la quema. La composicion de especies cambio en las dos areas despues de la quema con la aguililla rastrera {Circus cyaneus) y el halcon cernicalo {Falco sparverius) mostrando los cambios mas significantes. El numero de aguilillas rastreras disminuyo significantemente {P < 0.05) 537o y 367o en las dos areas, mientras el numero de cer- nicalos aumento significantemente {P < 0.05) 3007? y 5007>. [Traduccion Autores] Acknowledgments We are grateful to Chris S. Pease for valuable logistical support. Comments by C.S. Pease, T. Stehn, and J. Van’T Hull to earlier drafts of this manuscript were appreciated We appreciate comments and suggestions provided by three reviewers. Literature Cited Blankenship, T. 1990. Matagorda Island comprehen- sive management plan. Texas General Land Office, Austin, TX U.S.A. December 1994 Short Communications 265 COLLOPY, M.W. AND K.L. BiLDSTElN. 1987. Foraging behavior of northern harriers wintering in southeastern salt and freshwater marshes. Auk 104:11-16. Conover, W.J. 1980. Practical nonparametric statistics. 2nd ed. John Wiley, New York, NY U.S.A. Hamerstrom, F.N. 1974. Raptor Management. Pages 5-8 in F.N. Hamerstrom, B.E. Harrell, and R.R. Olendorff [Eds.], Management of raptors. Raptor Res. Rep. 2. Janes, S.W. 1985. Habitat selection in raptorial birds. Pages 159-188 in M.L. Cody [Ed.], Habitat selection in birds. Academic Press, New York, NY U.S.A, Ludwig, J.A., and J.F. Reynolds. 1988. Statistical ecology. Wiley and Sons Inc., New York, NY U.S.A. Millsap, B.A., K.W. Cline and B.A. Giron Pendleton. 1987. Habitat Management. Pages 215-237 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. Peterson, K.L. and L.B. Best. 1987. Effects of pre- scribed burning on nongame birds in a sagebrush com- munity. Wildl. Soc. Bull. 15:317-329. Preston, C.R. 1990. Distribution of raptor foraging in relation to prey biomass and habitat structure. Condor 92:107-112. Pylypec, B. 1991. Impacts of fire on bird populations in a fescue prairie. Can. Field-Nat. 105:346-349. Rudolph, S.G. 1982. Foraging strategies of American kestrels during breeding. Ecology 63:1268-1276. ScHiPPER, W.J.A., L.S. Burma and P. Bossenbroek. 1975. Comparative study of hunting behavior of win- tering hen harriers (Circus cyaneus) and marsh harriers (Circus aeruginosus) . Ardea 63:1-29. Smallwood, J.A. 1987. Sexual segregation by habitat in American kestrels wintering in southcentral Florida: vegetative structure and responses to differential prey availability. Condor 89:842-849. , M. WooDREY, M.J. Smallwood and M.A. Kettler. 1982. Foraging by cattle egrets and Amer- ican kestrels at a fire’s edge. /. Field Ornithol. 53:171- 172. Toland, B.R. 1987. The effect of vegetative cover on foraging strategies, hunting success, and nesting dis- tribution of American kestrels in central Missouri. J. Raptor Res. 21:14-20. Received 24 January 1994; accepted 6 August 1994 /. Raptor Res. 28(4):265-268 © 1994 The Raptor Research Foundation, Inc. Winter Diet of Long-eared Owls (Asio otus) IN THE Po Plain (Northern Italy) Paolo Galeotti and Luca Canova Dipartimento di Biologia Animate Universitd di Pavia - Piazza Botta 9, 27100 Pavia, Italy Key Words: food habits; diet; Italy; long-eared owl; Asio otus. The winter diet of the long-eared owl (Asio otus) has been well documented in northern and central Europe and North America (see Cramp 1985 for review), but there are few data for southern Europe and particularly Italy. In northern Europe, the long-eared owl specializes in hunting voles (Microtus spp.) in open fields (Herrera and Hiraldo 1976), and it is considered a restricted feeder (Marti 1976, Kallander 1977, Nilsson 1981). Likewise, in the Mediterranean region voles are the most important component in the diet of long-eared owls (Araujo et al. 1973, Tome 1991). Only three studies (Gerdol and Perco 1977, Casini and Magnaghi 1988, Canova 1989) have reported the diet and prey selection of long-eared owls in the Po Plain of northern Italy, which is the most important part of their Italian wintering and breeding range (Ga- leotti 1990). In this paper, we review the diet of long-eared owls in the Po Plain to: (1) provide new information on the trophic niche of southern wintering populations of long-eared owls, (2) compare local diets, and (3) determine the hunting habitats most utilized by the species. Study Areas and Methods This area has a sublittoral continental temperate climate with two peaks of rainfall in spring and autumn; during winter the mean temperature ranges from 0-1 0“C and precipitation averages 50-60 mm monthly. Pellets were collected between January and March from eight winter roosts located in the Po Plain. Roosts 1, 2, and 3 (Table 1) were close to the eastern edge of the study area, a few kilometers from the Adriatic Sea, while all other roosts were located in Lombardy, the central part 266 Short Communications VoL. 28, No. 4 feS pa (-T u B 3 C XI cj C u 3 cr u ,<1 >- rt ■ 4 -' o q CN o 00 q eg eg q m d cO NO o d d CM o O o NO o o o o NO d d od o o e*- •^* d On m eg B% oo CN q o q q q q q q o id o o o O o o o CN] d d o d ni o o ei o ON cn (N 00 q m q cn GO q q q m cn o id o o o o o o o oi d 00 o in lO o^ d o o eg o 1 d ON to o o b« ^ • i2 ■*-4 3 1 ■i-4 3 3 to rt c ■w X >- u Pi c . ^ o 3 5 2 s o 3 s savii '•-^a Q k 3 d a VI 3 to 3 .3 C 's 3 to «o 3 ■*o ti) 3 3 d a CA B "O o; cx: u OJ weig *5 Ph +-* fj u VI VI V > .2 15 B B a s o 3 C 2 C! a 2 ~i Ci a .2 S a C W ■ 2 -4.J ^ 2 ;2 Microtus 3 2 U § to 3 s u § S to 1 s 2 tj § 3 s o 3 E O 2 «o 3 3 0^ O 3 to 3 3 s ■§ 1 fc— 3 s *■*-> B OJ -o ’S e 0 c >-> u u cx B qj QJ U XI w 4J C HH < s (iH PM Q Ricefields/woodlots. 8 Farmland/ricefields. December 1994 Short Communications 267 of the Po Plain. Roosts were between 25-450 km apart. The habitat types surrounding each roost (2.5 km radius) are reported in Table 1. Although the long time span over which pellets were collected (16 yr) could be responsible for variation in local diets, we found no evidence of fluctuations in small rodent populations in northern Italy among years (L. Canova unpubl. data). Moreover, there is evidence that some vole species living in Mediterranean regions are not cyclic (Par- adis and Guedon 1993). Finally, the diet of long-eared owls can be more variable between seasons than between years (Nilsson 1981). Therefore, we believe that, taken together, these findings justify our tentative comparison among local diets irrespective of the years in which pellets were collected. All pellets were examined using standard techniques (Yalden 1977). Prey remains were identified using taxo- nomic keys (Erome and Aulagnier 1982), and reference specimens collected in the study areas. Because few work- ers have identified birds to species in dietary studies we considered them as a single category. Because they were all small passerines (mainly Passer spp.), we used a mean mass of 20 g for biomass calculations. Small mammal biomass was derived from the literature (Di Palma and Massa 1981, Galeotti et al. 1991). Frequency by number and biomass for each prey species and the mean mass of prey were calculated for diets at each roost. Diet breadth (based on frequencies of ecological prey- categories) was determined by the index: B = 1//? (Feinsinger et al. 1981) where p is the proportion of the prey i in the total sample and R is the number of prey categories collected at each station; the B-index varies from 0 (no use of any resource) to 1 (full use of total resources). Results and Discussion Overall Diet. We identified 3499 prey items that amounted to a total biomass of 75 466 g. Prey included mammals (89,6%), birds (10.2%) and insects (0.2%). Al- though long-eared owls preyed on 16 mammal species from four families, three species (Apodemus sylvaticus, Mi- crotus savii, and Micromys minutus) accounted for 64.8% of the diet and murid rodents predominated overall (Table 1) The mean mass of prey (21.5 g) was lower than that calculated (32.2 g) by Marti (1976) from a number of European studies; consequently, the average meal was also slightly lower (51.8 g) than that reported from a number of diets in northern Europe and North America (55-60 g) . Local Diets and Feeding Habitats. Diets differed in type and proportion of mammal prey species. Murids were the most important prey category at all but two roosts. Microtids were the second most common prey used by long-eared owls at all roosts but one, and they were preyed on more heavily at eastern sites. Birds were preyed on at all roosts, but predominated in the diet of long-eared owls in the urban roost and in willow/farmland. The impor- tance of birds in owl’s diet at one roost was related to weather: the frequency of bird prey increased in relation to the snow cover, ranging from 13-14% when snow was absent to 42% when snow covered the ground (Canova 1989). Niche breadth increased in relation to an increase m woodland species (r^ = 0.92, P < 0.01) which were rare prey in the diets. The frequency of farmland species in the diet (Rattus rattus, Apodemus agrarius, and Microtus arvalis) increased toward the east, while woodland species (Clethrionomys glareolus, and Microtus multiplex) were preyed on more often at western sites. This was consistent with the localized distribution of Clethrionomys glareolus which is locally abundant in wooded habitats in the west- ern portion of the Po Plain, but scattered or absent in the East (Canova et al. 1991). Hygrophilous species were also mainly exploited in western locations (roosts 4, 6, and 8), where numerous ditches and canals were associated with ricefields. Conclusions The diet of wintering long-eared owls in the Po Plain consisted mainly of mice, whereas elsewhere in Europe long-eared owls prey mainly on voles (Saint Girons and Martin 1973, Araujo et al. 1973, Glue and Hammond 1974, Marti 1976, Kallander 1977, Tome 1991). This difference was probably due to both a lower richness in vole species (only eight of the 20 European Microtidae species occur in Italy) and low population levels in south- ern Europe compared to the North (Herrera and Hiraldo 1976). Moreover, the number of vole species decreases across the Po Plain from east to west, with only four species being present in the west (Niethammer and Krapp 1982). Voles are scarcer than mice in most habitats except in scrubs and woods (Canova and Fasola 1991, P. Galeotti and L. Canova unpubl. data). The apparently close re- lationship between diet composition and prey availability confirms the trophic plasticity of long-eared owls in their Italian winter range (Canova 1989). Resumen. — La dieta de Asio otus invernantes fue estu- diada en Po Plain al norte de Italia, a traves de analisis de egagrapilas obtenidas desde ocho perchas de descanso invernal. En contraste con lo que ocurre en el norte y centre de Europa, los ratones predominaron por sobre las ratas de aqua. La composicion dietaria sugirio que las tierras de cultivo eran los habitat de alimentacion preva- lecientes alrededor de las perchas. La variacion dieteria desde presas de mamiferos a aves puede ocurrir en con- diciones climaticas adversas (nieves otonales) o en habitat urbanos. [Traduccion de Ivan Lazo] Acknowledgments We wish to thank D.A. Kirk and J.S. Marks for their useful suggestions and criticism on earlier drafts of the paper, and M.T. Forlini for her help in pellet analysis. Literature Cited Araujo, J., J.M. Rey, A. Landin and A. Moreno. 1973 Contribucion al estudio del buho chico (Asio otus) en Espana. Ardeola 19:397-428. Canova, L. 1989. Influence of snow cover on prey se- lection by long-eared owls Asio otus. Ethol. Ecol. Evol. 1:367-372. and M. Fasola. 1991. Communities of small 268 Short Communications VoL. 28, No. 4 mammals in six biotopes of northern Italy. Ada Theriol. 36:76-86. , P. Galeotti and M. Fasola. 1991. Distri- bution of the bank vole Clethrionomys giareolus in plain habitats of northern Italy. Mammalia 55:435-439. Casini, L. and a. Magnaghi. 1988. Alimentazione in- vernale di gufo comune Asia otus in un’area agricola dell’Emilia orientale. Avocetta 12:101-106. Cramp, S.C. 1985. The birds of the western Palearctic. Vol. V. Oxford Univ. Press, Oxford, U.K. Di Palma, M.G. and B. Massa. 1981. Contributo me- todologico per lo studio deH’alimentazione dei rapaci. Atti Conv. Ital. Orni. 1:69-76. ErOME, G. and S. Aulagnier. 1982. Contribution a I’identification des proies des Rapaces. Bievre 2:129- 135. Feinsinger, P., E.E. Spers and R.W. Poole. 1981. A simple measure of niche breadth. Ecology 62:27-32. Galeotti, P. 1990. Gufo comune Asio otus. Page 107 in P. Brichetti and M. Fasola [Eds.], Atlante degli Uccelli nidificanti in Lombardia, Editoriale Ramperto, Brescia, Italy. , F. Morimando AND C. VIOLANI. 1991. Feeding ecology of the tawny owl (Stnx aluco) in urban habitats (northern Italy). Boll. Zool. 58:143-150. GeRDOL, R. and F. PerCO. 1977. Osservazioni ecolo- giche sul gufo comune (Asio otus otus L.) nell’Italia Nord-Orientale. Boll. Soc. Adriat. Sci. 61:37-59. Glue, D.E. AND G.J. Hammond. 1974. Feeding ecology of the long-eared owl in Britain and Ireland. Br. Birds 67:361-369. Herrera, C.M. and F. Hiraldo. 1976. Food-niche and trophic relationships among European owls. Ornis Scand. 7:29-41. Kallander, H. 1977. Food of the long-eared owls Asio otus in Sweden. Ornis Fenn. 54:79-84. Marti, G.D. 1976. A review of prey selection by the long-eared owl. Condor 78:331-336. Niethammer, J. and F. Krapp. 1982. Handbuch der Saugetiere Europas. Rodentia. Akad. Verlag, Wies- baden, Germany. Nilsson, I.N. 1981. Seasonal changes in food of the long-eared owl in southern Sweden. Ornis Scand. \2- 216-223. Paradis, E. and G. Guedon. 1993. Demography of a Mediterranean microtine: the Mediterranean pine vole, Microtus duodecimcostatus. Oecologia 95:47-53. Saint Girons, M.C. and C. Martin. 1973. Adaptation du regime de quelques rapaces nocturnes au paysage rural. Les proies de I’Effraie et du Moyen-duc dans le department de la Somme. Bull. Ecol. 4:95-120. Tome, D. 1991. Diet of the long-eared owl in Yugoslavia. Ornis Fenn. 68:114-122. Yalden, D.W. 1977. The identification of remains in owl pellets. Occas. Publ. Mammal. Soc., London, U.K. Received 21 April 1993; accepted 29 April 1994 /. Raptor Res. 28(4):268-273 © 1994 The Raptor Research Foundation, Inc. Refinements to Selective Trapping Techniques: A Radio-controlled Bow Net and Power Snare FOR Bald and Golden Eagles Ronald E. Jackman, W. Grainger Hunt, Daniel E. Driscoll and Frank J. Lapsansky BioSystems Analysis, Inc., 303 Potrero Street Suite 29-203, Santa Cruz, CA 95060 U.S.A. Key Words: Aquila chrysaetos; bald eagle; bow net; cap- ture techniques; golden eagle; Haliaeetus leucocephalus; power snare. Research and management of raptors often requires the capture of specific individuals for radiotagging or color- marking. Bloom (1987) reviewed raptor trapping tech- niques, including several selective methods used for eagles: cannon and rocket nets (see also Grubb 1988), the pit trap, and our bow net. Meng (1963) was first to develop a radio- controlled bow net and Bryan (1988) modified it for use with American kestrels (Falco sparuerius). The power snare, a “manually-operated, single noose system,” was devel- oped for the selective capture of white-bellied sea-eagles (Haliaeetus leucogaster) by Hertog (1987). During studies of wintering and breeding bald eagles (Haliaeetus leucocephalus) and golden eagles (Aquila chry- saetos) in Washington, California, and Arizona (Hunt et al. 1992a, 1992b, 1992c, 1992d), we constructed a radio- controlled bow net to selectively capture eagles (Fig. la). We were able to completely conceal it in loose soil and operate it from distances up to 400 m. In addition, we December 1994 Short Communications 269 Figure 1. Radio-controlled eagle bow net: (a) bow net opening, showing position of principal components, (b) top view, no springs, (c) detail of spring-hinge-bow-channel attachment, (d) cross section detail of channel at trigger mount, (e) interior detail of trigger box. 270 Short Communications VoL. 28, No. 4 Figure 2. Application of radio-controlled trigger box to Hertog’s power snare. Distance relationships between bait, snare, forked stick, and hoop are from Hertog (1987). modified Hertog’s (1987) manually-operated power snare with a radio-controlled trigger device (Fig. 2). Materials and Assembly Bow Net. The physical dimensions of the bow net, shown in Fig. lb, allowed the hoop and net to safely capture a single eagle. A larger version was used by A. Harmata (pers. eomm.) to capture two eagles simulta- neously. Table 1 provides a list of the principal materials for the bow net; a metal fabrication (welding) shop can supply and construct many of the components. We formed the aluminum bow into a semicirele with a pipe bender. Support bars inside the bow/pipe fitting connection helped distribute the force from the springs to prevent the bow from snapping off (Fig. Ic). We attached the net loosely by placing a large cardboard box (about 1.1 m^) in the middle of the trap, draping the net over the box and open bow, and tying/taping the net evenly at 5- 10 cm intervals to the bow and to a stiff wire secured to the bottom of the channel. Foam pipe insulation protected the bow. A channel liner of plastic sheeting (Fig. Id) protected the net from snagging inside the channel. We tightened each spring approximately eight revolu- tions, both springs in the same direction. If correct, they decreased in diameter and expanded lengthwise when tightened. A plastic “boot” of loosely wrapped plastic sheeting taped around the garage door springs kept grit from fouling the spring action (Fig. la). We tested the completed bow net by placing an eagle- sized cardboard box in the middle of the trap. If constructed correctly, the bow and net released immediately, flew up and over the box without disturbing it, and contacted the other side in about 0.5 sec. Trigger. The bow net trigger system (Table 1, Fig. le) was separately contained in a military-surplus ammuni- tion box. The remote unit consisted of a two-channel radio control system of the type used in model airplanes. The actual trigger was an archery bowstring release, capable of holding extreme resistance yet easily set off with the action of the remote servo unit. The trigger mount on the bow net channel (Fig. Id) supported the head of the re- lease; a trigger loop of cable was just long enough to reach into the release arm and, when the trap was set, hold the bow down as low as possible in the channel. Power Snare. We modified Hertog’s (1987) power snare with a radio-controlled release system using the trigger described above. Hertog’s (1987) dimensions (snare dis- tance relationships) for white-bellied sea-eagles were ef- fective in capturing bald eagles. The materials necessary to construct a power snare are listed in Table 1; our modified power snare is depicted m Fig. 2. We used a heavier snare material (27 kg breaking strength (test) vs. 18 kg for the smaller white-bellied sea- eagle) and multifilament extension line to connect the snare material to the shock cord. We tied overhand loops to each end of the 5 m nylon-jacketed shock cord. Into one over- hand loop we connected the trigger loop and the shock cord loop (Fig. 2, Table 1). Field Use Bow Net. We chose a site frequented by the target bird in a relatively open area which could be viewed from above We buried the channel up to the rims, the springs just subsurface, and covered the (activated) trigger box after wrapping the antenna around a twig pushed into the ground. We staked the trap down in two places along the crosspiece and cleared debris from around the hinge area We set the trigger loop into the release arm, tested it, and, to conserve batteries and avoid premature triggering, we turned the transmitter unit off until it was time to trigger the trap. We secured the bait solidly with two pieces of baling wire attached to the middle of the tubular shaft. We laid December 1994 Short Communications 271 Table 1. List of principal materials needed to construct bow net, trigger system, and power snare. Number Needed Item Dimensions Materials 1 set garage door tor- Bow Net size used on 5 m (16 ft) sectional garage set = right & left springs 1 sion springs tubular shaft door (approx. 0.6 m long) 190 cm X 2.5 cm (1 in) O.D. steel torsion bar 1 cross piece 113 cm X 2.5 cm (1 in square) iron square stock 1 channel 8 cm wide x 8 cm deep 14 gauge steel 1 reinforcement plate 8 cm X 10 cm 14 gauge steel 1 bow approx. 3.1 m x 1.6 cm (0.625 in) 0.09 cm (0.035 in) thickness aircraft 1 set hinge brackets O.D. 0.6 cm Q/yi in) thick (see Fig. Ic) aluminum type 6061-T6, WW-T- 700/6 strap iron 2 pipe fittings 15 cm, 1.6 cm (0.625 in) I.D. iron pipe 2 support rods 30 cm of 1.3 cm (0.5 in) dia hardwood dowel several shims 2.5 cm (1 in) I.D., 3.8 cm O.D. steel washers 2 spring boots approx. 80 cm x 30 cm 3 ml plastic sheeting 2 channel liners approx. 3.1 m x 20 cm 3 ml plastic sheeting 1 net approx. 3.5 m x 3.7 m of 9 cm mesh multifilament salmon gill-netting 1 trigger mount 1.9 cm (0.75 in) I.D. steel washer 1 trigger loop approx. 13 cm of 0.16 cm (Vi6 in) plastic coated cable & connector sleeves 1 bow insulation 3.1 m of 1.3 cm (0.5 in) thickness foam pipe insulation 20 channel covers 20 cm X 10 cm plastic contact paper over cardboard 1 bow release Trigger archery bow release 1 radio control system servo unit, receiver, transmitter two-channel radio control set 1 trigger box 26 cm X 18 cm x 9 cm watertight ammunition box 1 snare Power Snare 9 m spool of 27 kg (60 lb) test (black) plastic coated multi-strand fishing lead- 1 extension line spool of 54 kg (120 lb) test er cable & connector sleeves dark, braided dacron fishing line 1 shock cord 5 m of 5 mm dia nylon-jacketed shock cord 1 forked stick approx. 75 cm, 18 cm thick straight forked branch 1 hoop approx. 45 cm long, 6 cm opening dog tie-out (auger) stake 1 anchor 4.5 kg barbell weight or sand auger 1 trigger loop 10 cm, 113 kg (250 lb) test plastic coated multi-strand fishing lead- 1 shock cord loop 15 cm, 113 kg (250 lb) test er cable & connector sleeves plastic coated multi-strand fishing lead- er cable & connector sleeves channel covers (sections of thin cardboard sandwiched be- tween brown plastic contact paper) along the channel at the surface of the soil and used a sifter to sprinkle sand or soil lightly over the covers; thus, the trap was completely hidden. We further camouflaged it with grass and leaves. The entire process required about one hour, longer if soil was compacted or muddy. A. Harmata (pers. comm.) used the bow net in snow, taking precautions to prevent the net from freezing together or snow from freezing over the top. We chose a blind with an elevated view of the trap site, a factor which also improved radio reception. We placed a recognizable marker (e.g., rock or small bush) just out- side the perimeter of the trap to help verify the eagle did not move the bait. The strength of the springs needed to operate this large bow net could injure or kill an eagle if used incorrectly, so we made sure that the eagle was in the center of the trap and taking bites with its head down before triggering. When eagles refused carrion bait, we used live bait. We installed the trap prior to first light to avoid alerting the eagle. Power Snare. We used the power snare in remote areas (where the heavier bow net could not be easily trans- ported), on rocky substrate, and along wet shoreline areas. When the snare reached partially into the water, we po- 272 Short Communications VoL. 28, No. 4 sitioned the trigger (water-resistant but not waterproof) on higher ground. Black fishing leader cable formed the noose around the staked-down bait, extended up and over the forked stick, through the hoop and out toward the trigger box (Fig. 2). We used a simple overhand knot with a 2 cm free end for the noose slip knot. We positioned the trigger box 9 m from an anchor (tree, shrub, buried weight, or sand auger) to which we secured one end of the 5 m shock cord. We used a dog tie-out (auger) stake twisted into the ground for the retaining hoop. All components of the set (bait, forked stick, hoop, trigger, anchor) were in a straight line, cleared of debris. We activated, buried, and staked the trigger box so the trigger tilted slightly aboveground and pointed at the an- chor (Fig. 2). We elevated the antenna wire by wrapping it on a twig pushed into the ground. We set tension on the trap by stretching the shock cord 4 m and setting the trigger loop into the trigger. The amount of stretch in the shock cord was critical to effective operation: too little tension made the snare weak and slow, too much could break the snare cable. We tied one end of the extension line to the shock cord loop; the other end was tied to an overhand loop at the end of the snare line. We used two rocks to anchor the snare at the head of the bait. The snare operated effectively in shallow water if we used larger rocks to force the far end of the snare to leave the ground last. Two smaller rocks placed at the tail end of the snare (Fig. 2) helped prevent small birds from disturbing the integrity of the noose. We tested each ap- plication by placing a hand on the bait to simulate an eagle capture. About one hour was required to set up the power snare. In contrast to the bow net, we set off the power snare when the eagle’s head was up, following a bite or two on the carcass (Hertog 1987), A large carp {Cyprinus carpio) carcass placed partially in the water was most effective at bringing bald eagles to the power snare. Discussion The radio-controlled bow net was successful in 16 of 19 attempts to capture bald eagles (84% success rate). Failures occurred when; (1) the wires of the trigger box loosened after testing, leaving the trigger inoperable, (2) the water level rose on a reservoir shoreline set, flooding (and shorting out) the trigger mechanism, and (3) the eagle’s stoop moved the (small) bait off the retaining wires and outside the center of the trap, thus rendering the trap unsafe for use. We caution the reader to use larger baits, attached securely. Our attempts to trap golden eagles with this bow net were also successful; we captured 26 of 30 (87%) eagles that approached the bait. Failures included: (1) weak bat- teries in the trigger box, (2) the transmitting signal did not reach the trigger, and (3) eagles walking toward the bait stepped on the channel covers, became suspicious, and flew off. We corrected the latter problem by using slightly stiffer and wider channel covers. The radio-controlled power snare was effective five of seven times bald eagles came to bait (71% success). We missed once when the eagle disturbed the snare prior to hopping up on the bait. On another occasion, the snare was set on a relatively steep shoreline, reducing the effect of the forked stick (in bringing the snare up around the legs of the eagle). We later experimented with a taller forked stick which offset the slope’s effect. In our opinion, the eagle capture techniques described in this paper have certain advantages over other selective traps. The bow net is very reliable when properly installed, is less dangerous and more easily camouflaged than a rocket or cannon net, and is mobile, unlike the pit trap On the other hand, the quick action of this spring-powered bow is potentially hazardous, and we advise adherence to our precautions, particularly with regard to the eagle’s position at the moment of triggering. The radio-controlled power snare is extremely mobile and safe, but probably less reliable, because the snare can be disturbed by the eagle or other birds prior to triggering. Resumen. — Capturamos individuos especificos de Hal- laeetus leucocephalus, usando dos sistemas de trampeo ra- dio-controlados. Una red “bow” de dos m de diametro, activada por resortes de torsion de puertas de cochera, fue altamente confiable {N =16 capturas/19 intentos, 84% de exito) y puede ocultarse completamente en sustratos sueltos. Tambien capturamos 26 de 30 (87% de exito) individuos de Aquila chrysaetos con este sistema. Modifi- camos un sistema de lazo tensado y manualmente operado por nuestro sistema radio-controlado de activacion. Este sistema fue mas adecuado para remotas localizaciones de trampeo, capturandose cinco individuos de H. leucoceph- alus en siete intentos (71% de exitos). [Traduccion de Ivan Lazo] Acknowledgments We developed capture techniques while studying bald and golden eagle ecology under separate contracts with Seattle City Light, Pacific Gas and Electric Co., Sacra- mento Municipal Utility District, the U.S. Bureau of Land Management, the U.S. Bureau of Reclamation, Kenetech/ U.S. Windpower, and U.S. Department of Energy (Na- tional Renewable Energy Laboratory). We are grateful to the following individuals for their encouragement and advice during trapping efforts: M. Lockhart, R. Olendorff, P. Bloom, C. Thelander, M. Jenkins, T. Gatz, and R, Mesta. A. Harmata contributed design improvements and knowledge of field implementation of the bow net. Field assistance was provided by B, Johnson, T. Miller, L Small- Jackman, P. Hunt, P, Detrich, R. Lehman, S Spangle, W. Lehman, M. Cross, J. Didonato, R. Spauld- ing, F. Hein, J. Driscoll, B. Latta, and T. Brown. Literature Cited Bloom, P.H. 1987. Capturing and handling raptors Pages 99-123 in B.A. Pendleton, B.A. Millsap, K.W. Cline, and D.M. Bird [Eds.], Raptor management techniques manual. Natl. Wildl. Fed. Sci. Tech. Series 10, Washington, DC U.S. A. Bryan, J.R. 1988. Radio controlled bow-net for Amer- ican kestrels. North Am. Bird Bander 13:30-31. Grubb, T.G. 1988. A portable rocket-net system for December 1994 Short Communications 273 capturing wildlife. USDA For. Serv., Rocky Mountain Forest and Range Exper. Sta. Res. Note RM-484. Fort Collins, CO U.S.A. Hertog, A.L. 1987. A new method to selectively capture adult territorial sea-eagles. J. Raptor Res. 21:157-159. Hunt, W.G., B.S. Johnson, and R.E. Jackman. 1992d. Carrying capacity for bald eagles wintering along a northwestern river. /. Raptor Res. 26:49-60. , D.E. Driscoll, E.W. Bianchi and R.E. Jack- man. 1992a. Ecology of bald eagles in Arizona. Rep. to U.S. Bur. Reel., Contract 6-CS-30-04470. Bio- Systems Analysis, Inc., Santa Cruz, CA U.S.A. , R.E. Jackman, J.M. Jenkins, C.G. Thelander AND R.N. Lehman. 1992c. Northward post-fledging migration of California bald eagles. J. Raptor Res. 26: 19-23. , J.M. Jenkins, R.E. Jackman, C.G. Thelander AND A.T. Gerstell. 1992b. Foraging ecology of bald eagles on a regulated river. J. Raptor Res. 26:243- 256. Meng, H. 1963. Radio controlled hawk trap. East. Bird Band. Assoc. News 26:185-188. Received 9 March 1994; accepted 24 July 1994 J. Raptor Res. 28(4):274-275 © 1994 The Raptor Research Foundation, Inc. Letter First Record of Interspecific Cartwheeling Between Large Raptors: Buteo poecilochrous and Geranoaetus melanoleucus Two raptors engaging each other by the feet or talons and mutually rotating about a central axis while falling earthward has been variously and inconsistently referred to as “cartwheeling,” “whirling,” and “talon-grappling,” (L. Brown and D. Amadon 1968, Eagles, hawks and falcons of the world, Vol. 1, Country Life Books, London, U.K., I. Newton 1979, Population ecology of raptors, Buteo Books, Vermillion, SD U.S.A.; W.S. Clark 1984, Condor 86:488; R.E. Simmons and J.M. Mendelsohn 1993, Ostrich 64:13-24). Although these terms often are used more or less interchangeably, not all cases of foot engagement are followed by rotation; sometimes only brief contact is made (G. Blaine 1970, Falconry, Neville Spearman, London, U.K.; R. Sierra 1985, Distribucion, Presas y Ecologia Reproductiva de Geranoaetus melanoleucus en los Andes Equinocciales, Provincia de Pichincha, Ecuador, Tesis de Lie., Pontificia Univ. Catolica del Ecuador, Quito, Ecuador; D.H. Ellis 1992, J. Raptor Res. 26:41-42). We use the term “cartwheeling,” and recommend it to describe the combination of behaviors involving initially the interlocking of feet known as “talon- grappling” followed by a “cartwheel-like” rotation. On 24 May 1993, while observing a pair of breeding Gurney’s buzzards {Buteo poecilochrous) at a known nesting area (78“15'W, 0°15'S) approximately 35 km east-southeast of Quito, Ecuador, we witnessed a spectacular example of interspecific cartwheeling between the female of the Gurney’s buzzard pair and an adult black-chested buzzard- eagle {Geranoaetus melanoleucus), presumed to be a male. We were watching the buzzards as they patrolled the western side of a ridge crest in their nesting area when, at 1500 H, the black-chested buzzard-eagle appeared from the northeast slowly soaring somewhat lower than the pair of buzzards. The buzzards began giving alarm calls and flew toward the intruder. The male buzzard attacked the buzzard-eagle first, making three or more short stoops at the intruder, but pulling up before making contact. On each stoop the buzzard-eagle rolled over and presented its talons to the buzzard. During these exchanges the female Gurney’s buzzard circled above and then dove straight toward the buzzard-eagle with her wings tucked in. Unlike the male, however, she continued her attack until she made contact with the buzzard-eagle which again flipped upward presenting Its talons. The buzzard and buzzard-eagle locked their feet together and, with their wings outspread, rotated downward in 6-8 moderately rapid “cartwheels.” The cartwheeling episode lasted only 3-4 sec, and the birds broke free of their grasp about 60 m above the ground. Neither participant seemed to be in danger of striking the ground nor of losing aerodynamic control, and neither appeared to be injured. The buzzard-eagle then turned back in the opposite direction from which it came, and glided down to perch on a large boulder some 100 m away. The female Gurney’s buzzard rejoined the male, which had been circling overhead, and both flew to perch on some low-lying rocks within their nesting area. Both cartwheeling and talon-grappling behaviors have been widely reported, but the majority of these cases have been within species, typically between defender and intruder, parent and offspring, siblings, or mates (e.g., D. Watson 1977, The hen harrier, T. & A.D. Poyser, Berkhamsted, U.K.; M.A. Springer 1979, J. Raptor Res. 13:19; Simmons and Mendelsohn 1993). In contrast, there is very little documentation of either of these behaviors occurring between species, suggesting they are quite rare. The interaction we observed was clearly a case of agonistic rather than courtship behavior, and is thus consistent with the suggestion by Simmons and Mendelsohn (1993) that instances of cartwheeling are generally agonistic in nature. In their comprehensive review of cartwheeling in the Falconiformes, Simmons and Mendelsohn (1993) cite just one definite interspecific account, between a hen harrier {Circus cyaneus) and a common kestrel {Falco tinnunculus) in Scotland (Watson 1977). Other published records of interspecific encounters in which cartwheeling was implied all involved European falcons: hobby {F. subbuteo) and common kestrel (S.J. Hayhow 1988, Br. Birds 81:324); red-footed falcon {F. vespertinus) and common kestrel (M. Coath 1992, Br. Birds 185:496); and, merlin {F. columbarius) and peregrine falcon {F. peregrinus; M.S. Wallen 1992, Br. Birds 85:496). Jimenez and Jaksic (1989, Condor 91:913-921; 1993, J. Raptor Res. 27:143-148; and pers. comm.) reported no incidences of talon-grappling or cartwheeling in their lengthy field studies on black-chested buzzard-eagles and other raptors in Chile. We thank D. Amadon, R.H. Barth, N.J. Farquhar, F. Jaksic, J. Jimenez, J.M. Mendelsohn, and B. Whitney for constructive comments on earlier drafts. We are also grateful to T. de Vries and L.E. Lopez, Pontificia Universidad * Present address: Texas Natural Heritage Program, Texas Parks and Wildlife Department, 3000 1-35 South, Suite 100, Austin, TX 78704 U.S.A. 274 December 1994 Letter 275 Catolica del Ecuador, Quito. This work was part of a study by CCF on the ecology and systematics of B. polyosoma and B. poecilochrous supported by a Frank M. Chapman Research Fellowship from the American Museum of Natural History, New York. — C. Craig Farquhar,^ Department of Ornithology, American Museum of Natural History, Central Park West at 79th, New York, NY 10024 U.S.A.; William S. Clark, 7800 Dassett Court, Apt. 101, Annandale, VA 22003 U.S.A.; Robert G. Wright, Wisconsin Department of Natural Resources, 3550 Mormon Coulee Road, La Crosse, WI 54601 U.S.A.; Monica Coello, Departamento de Ciencias Biologicas, Pontificia Universidad Catolica del Ecuador, Carrion y 12 de Octubre, Quito, Ecuador. BOOK REVIEWS Edited by Jeffrey S. Marks J. Raptor Res. 28(4):276-277 © 1994 The Raptor Research Foundation, Inc. The Peregrine Falcon, By Derek Ratcliffe. 1 993. 2nd edition. T. & A.D. Poyser, London, U.K. Unit- ed States edition by Academic Press, San Diego, GA. xxxiii + 454 pp., 57 photos, 4 color plates, 22 figures, 31 tables. ISBN 0-85661-060-7. Cloth, $39.95.— In 1980, Derek RatclifFe came out with the first edition of The Peregrine Falcon. The appearance of the sec- ond edition invites a comparison between the two because so much has happened in those intervening years. I was privileged to review the first edition {Nature 288:519-520). In that review, I remarked “The book is truly the definitive work on a species that has become a cause celebre for the environmen- tally conscious.” While there is now another “de- finitive” book {Peregrine Falcon Populations: Their Management and Recovery, by T.J. Cade et al.) to vie for that label, the second edition of Ratcliffe’s book has nonetheless enlivened and heightened the discussion concerning the environmental issues ex- emplified by the peregrine falcon {Falco peregrinus). It is still one of the definitive works on peregrines. Another review of Ratcliffe’s first edition, and one that gives a more extensive critique, was by F. Pres- cott Ward (/. Wildl. Manage. 45:1084-1086). As Ward pointed out, the book was more properly a book about the peregrine in the British Isles than about peregrines as a species overall. So it is with the second edition, although this one contains some major revisions reflecting not only the status change in the peregrine in the British Isles since 1980, but also the entire pesticide issue and population trends worldwide. New material notwithstanding, Rat- cliffe’s book title would have conveyed more precisely the information it contains were it along the lines of “Biology of peregrines in the British Isles: with com- ments from elsewhere over its range.” (The same comment applies equally well to the other T. & A.D. Poyser books on raptors, viz. The Kestrel, The Hen Harrier, The Sparrow hawk, and The Barn Owl.) The changes and increases in information in the second edition are reflected in its size. The first edi- tion contained 416 pages, whereas the second has 454 pages. In the 16 chapters of this new edition, several increased by only a single page, and one, “Ecological Relationships with Other Birds,” re- mained the same length. Several chapters had either a major increase in data or more involved discussions of the data as indicated by page increases in the following chapters: “The Peregrine’s Country” (4- page increase), “Distribution and Numbers in Brit- ain” (3 pages), “Food and Feeding Habits” (5 pages), “Nesting Habits” (5 pages), “The Breeding Cycle. Laying and Fledging” (11 pages), “Movements and Migration” (5 pages), and “The Pesticide Story” (5 pages). In addition to the aforementioned chapters, there are also chapters entitled: “The Peregrine and Man,” “Population Trends in Britain,” “The Breeding Cycle: Pairing and Courtship,” “Breeding Density and Territory,” “Population Dynamics and Regulation,” “Other Enemies,” “Appearance, Form and Geographical Variation,” and “Conservation and the Future.” In the preface, Ratcliffe indicates that two key events have made this second edition justified. First has been the remarkable and continuing increase in numbers of falcons in Britain and Ireland and the dramatic recovery elsewhere in the world. Second is the large amount of new material published on the species. Ratcliffe further indicates that the most im- portant additions to knowledge of the peregrine’s natural history in the British Isles concern popu- lation dynamics, movements, and nesting adapta- tions. Much of the data presented did not result from Ratcliffe’s original investigations in the British Isles but represent a compilation of numerous research reports. In the first edition, it was estimated that Britain and Ireland together contained around 1050 pairs just prior to World War II. This figure was taken to represent a “baseline” number for carrying ca- pacity in that region. By 1991, it was estimated that at least 1600 pairs occurred in those same areas. Britain and Ireland combined are about the same size as the state of New Mexico; 314 300 km^ com- pared with 314 925 km^, respectively. The density, then, in that little corner of Europe is truly re- markable. (New Mexico does not contain “suitable” peregrine habitat throughout the state and has about 276 December 1994 Book Reviews 277 55 locations that peregrines are known to have oc- cupied at some time; there may never have been more than 100-300 pairs during the best of conditions.) Recovery in the British Isles has not been uniformly distributed, however, even in “good” habitat. Rat- cliffe draws attention (page 71) to a puzzling pattern; no reoccupation in some areas, colonization of areas not known to have had peregrines historically, lower local numbers in 1991 than in 1981, and enormous increases over large regions (e.g., 173% over 1930s baseline numbers in the Lakeland district). Ratcliffe reckons that the doubling of numbers in some regions since the early 1980s represents a combination of three factors: increased observer effort, reduced gamekeeper persecution, and occupation of com- pletely new territories. Additionally, part of the over- all increase in falcons throughout the British Isles may have resulted from an increase in a major prey item, the rock dove (Columba livia). Between 1977 and 1991, the number of racing pigeon bands issued increased from 1.5 to 2.5 million. Ratcliffe suggests that, based on the prey availability-peregrine density relationship, food supplies are available to support a substantial additional increase in falcon numbers. The population increase since 1980 has been accom- panied by increased use of “marginal” cliffs (earlier thought to be unacceptable) and other structures such as stone quarries, especially in Ireland, where 60-65 quarries were occupied in 1991. Who can predict what level this population will reach? At whatever point it finally stabilizes, however, it will certainly remain one of the densest populations in the world. Banding data have produced some interesting re- sults. Between 1921 and 1990, 357 falcons have been recovered of the 4476 that were banded. There is movement between Britain and Ireland, although none of the banded adults has been recovered farther than 200 km from their banding sites. The only peregrine recovery from a British or Irish breeding site outside of those two countries is of one moving from northern Ireland to Portugal. Some peregrines banded in Scandinavia have wintered in the British Isles, but there is no movement in the opposite di- rection with British-raised falcons. The species is remarkably sedentary considering the geographic size and location of that region. This new edition of The Peregrine Falcon has a wealth of interesting material, even though it applies principally to that part of Europe mentioned. Any- one interested in the peregrine, especially post-pes- ticide population trends, should have this book. The reasonable price makes it a good purchase. — Clay- ton M. White, Department of Zoology, Brigham Young University, Provo, UT 84602 U.S.A. /. Raptor Res. 28(4);277-278 © 1994 The Raptor Research Foundation, Inc. The Black Eagle: A Study. By Valerie Gargett. 1993. Academic Press, San Diego, CA. 279 pp., 7 color plates, 70 color photographs, 4 black-and-white photographs, 53 figures, 60 tables. ISBN 0-12- 275970-2. Cloth, $59.95. — This well-written and well-illustrated book presents the results of a mon- umental 20-yr study (1964-1983) of the life history of the black (Verreaux’s) eagle, Aquila verreauxii, in Zimbabwe. This study ranks up there with other great single-species studies such as those by Mar- garet Morse Nice on the song sparrow {Melospiza melodia) and Ian Newton and colleagues on the Eur- asian sparrowhawk (Accipiter nisus). Besides being well written and chock full of information, the book is nicely illustrated with color plates and many color photographs. The delightful and accurate pen-and- ink drawings by Rob Davies are, by themselves, reason enough to buy the book. In the Introduction, Gargett explains how the study began and outlines various research methods. The study area in the Matobo Hills of Zimbabwe is described in detail in Chapter 1. Chapter 2 de- scribes in equal detail the black eagle itself, including information on distribution in Zimbabwe, plumages, molt, vocalizations, display flights, and sex differ- ences in wing shape. The meat of the book, Chapters 3 through 9, contain detailed descriptions of various aspects of reproduction, including pairing, courtship displays, nest sites, incubation, reproductive success, hunting, and prey species. Chapter 10 discusses dis- persal of young birds, and Chapter 1 1 provides data on annual survivorship and longevity. A wonderful description of the project’s careful study of obligate siblicide, the so-called Cain and Abel syndrome, is presented in Chapter 12. Chapter 13 is a personal account of the author’s relationship with two female 278 Book Reviews VoL. 28, No. 4 eagles. Chapters 14 and 15 give accounts of other raptors in the study area and their relationships with black eagles. The final chapter is a look into the future of black eagles in the Matobo Hills. The three appendices contain a list of the 88 in- dividuals who participated in the research, a dis- cussion of the problems of egg collecting in the Ma- tobo Hills, and data from the continuation of the eagle breeding survey from 1984-1988 (during which the author was absent from Zimbabwe). One of the participants was later convicted of collecting eagle eggs. After stealing the eggs, he had falsified data on the nests he was observing! The book ends with a glossary and bibliography. The former is necessary in order for non- Africans to understand terms such as “dassie,” the local name for a hyrax that is the main prey of black eagles. This book is so attractive that it could easily be misconstrued as just another coffee table book. But It is much more than that — it is a detailed account of the life history of a raptor that will serve as a model for single-species monographs. The Black Ea- gle is a must for all serious raptor biologists, es- pecially those who study eagles. I recommend it also for any non-biologists with a serious interest in birds of prey and for anyone who collects raptor art; the drawings by Rob Davies are superb. — William S. Clark, 7800 Dassett Court, Apartment 101, An- nandale, VA 22003 U.S.A. J Raptor Res. 28(4):278-279 © 1994 The Raptor Research Foundation, Inc. Birds of Prey. By Floyd Scholz. 1993. Stackpole Books, Mechanicsburg, PA. 318 pp., 470 color pho- tographs, 39 figures. ISBN 0-8117-0242-1. Cloth, $59.95. — This is yet another book with the simple and general title “Birds of Prey.” There is a nice photo of a golden eagle (Aquila chrysaetos) head on the dust cover. One must look inside and read the introduction to realize that the purpose of this book is to provide detailed reference material on diurnal birds of prey for bird carvers. The back of the book’s dust cover does contain “An Artist’s Guide to Un- derstanding Raptors.” The reference material presented in this large for- mat book consists of hundreds of close-up color pho- tographs of live raptors and specimens, a few photos of raptors in the wild, many detailed illustrations, and accompanying explanations and descriptions in text and captions. Many of the live raptors were photographed from multiple aspects to show differ- ent details. Seventeen of the 34 species of North American diurnal raptors are covered; these include all three accipiters, osprey {Pandion haliaetus), both eagles, five falcons, and six buteonines. One widespread buteo, the Swainson’s hawk {Buteo swainsoni), is not covered, whereas the Harris’ hawk {Parabuteo uni- cinctus), with a limited distribution in North Amer- ica, is covered. Also omitted are vultures, kites, and the northern harrier {Circus cyaneus), as well as the peripheral species from the southern United States. The first chapter is entitled “What is a Raptor?” and is a general description of diurnal birds of prey. This is accomplished with a combination of many close-up photos, detailed drawings with dimensions, and descriptions of the eyes, feet, nostrils, skeletons, and other anatomical details. Next follow 17 chap- ters, one for each of the species described. The cov- erage of each species is very detailed for at least one age or sex category but does not include all variations for any species. For most species, only the light- morph adults are illustrated and described. The next section is entitled “Techniques for the Artist and Carver” and includes short chapters on “Painting the American Kestrel,” “Sculpting a Raptor Head,” and “Making Eyes.” The last chapter, called simply “Gallery,” consists of color photographs of eight carved raptors. Each carving is shown in two or three photos. The carvings are stunning, with fine details noticeable and accurate. The text, for the most part, is useful and accurate, but the author’s lack of experience with raptors re- sults in many errors of omission, as well as a few outright mistakes. An example is that the female merlin (Falco columbarius) is not “slightly larger,” as written, but considerably larger than the male. It would have been helpful to mention that the adult rough-legged hawk {Buteo lagopus) in the photos is a male. A drawback of the book is the steep price; however, considering the vast amount of reference material in the many photos and drawings, it should be worth the price for artists and carvers. December 1994 Book Reviews 279 I recommend this book for bird carvers and artists who aspire to create accurate, detailed works of art depicting birds of prey. It will serve equally well as a reference source for taxidermists. It should be con- sidered as a helpful reference work for raptor re- habilitators, raptor biologists, field guide illustrators, falconers, and others with an interest in diurnal birds of prey. — William S. Clark, 7800 Dassett Court, Apartment 101, Annandale, VA 22003 U.S.A. J Raptor Res. 28(4):279 © 1994 The Raptor Research Foundation, Inc. Raptor Conservation Today. Edited by B.-U. Meyburg and R.D. Chancellor. 1994. Proceedings of the IV World Conference on Birds of Prey and Owls. Pica Press, East Sussex, U.K. Distributed in the United States by Buteo Books, Shipman, VA. XIV -1- 799 pp., numerous figures, tables, and line drawings. ISBN 1-873403-33-X. Paper, $49.95. — The IV World Conference on Birds of Prey and Owls was held in Berlin, Germany from 10-17 May 1992. Organized by the World Working Group on Birds of Prey and Owls, the conference was attended by more than 500 participants from dozens of coun- tries. Of the 240 oral and poster presentations, 100 were published in the proceedings. This volume of material, which is both enormous and diverse, pre- cludes a detailed review. The first paper is the keynote address by Claus Konig on taxonomic problems in New World Glau- cidium and Otus. Next are 99 papers in 10 sessions on the following subjects: “Population Studies: As- pects of Long-Term Changes” (nine papers); “Rare and Declining Raptors” (23); “Tropical Rain For- ests and Raptors” (10); “Trapping, Marking & Ra- dio Tagging” (10); “Biology & Conservation of Large Falcons” (six);“Reintroductions” (six); “Population Ecology of Owls” (nine); “Extirpated, Rare or Less- er Known Owls” (four); “Systematics & Taxono- my” (five); and “Environmental Contaminants and Raptors” (17). The book concludes with a list of resolutions on various conservation concerns. The cover photograph of a Steller’s sea-eagle {Haliaeetus pelagicus) in flight is stunning. All of the papers are in English. As one would expect, quality varies markedly, with contributions ranging from brief status summaries to 1 2-page re- search papers packed with original data. The ses- sions on rare and declining raptors and environ- mental contaminants are especially noteworthy for their breadth of coverage. The symposium on pop- ulation ecology of owls marks the first time that a major session at a World Conference has been de- voted to nocturnal raptors. I suspect that much of the information in the proceedings has heretofore been unavailable to North American biologists, but there is at least one replicate publication (the paper on distribution of Mexican owls appeared in similar form in J. Raptor Res. 27:154-160). Considering that English was not the first language for many of the contributors, the editing is impressive. So, too, is the range of topics considered. Truly, there is something that will appeal to anyone who studies raptors. This book is a must for university libraries and for those whose interest in raptors extends beyond the borders of their respective countries. — Jeff Marks, Coop- erative Wildlife Research Unit, University of Montana, Missoula, MT 59812 U.S.A. J. Raptor Res. 28(4):279-280 © 1994 The Raptor Research Foundation, Inc. The Birds of North America. Edited by A. Poole, P. Stettenheim, and F. Gill. The Academy of Nat- ural Sciences, Philadelphia, PA and the American Ornithologists’ Union, Washington, DC. Individual accounts are now available from the American Bird- ing Association, P.O. Box 6599, Colorado Springs, CO 80934-6599 U.S.A. Accounts have been pub- lished for the following raptors: No. 1 Barn Owl, by C.D. Marti No. 10 Snowy Owl, by D.F. Parmelee No. 30 White-tailed Hawk, by C.C. Farquhar No. 41 Great Gray Owl, by E.L. Bull and J.R. Duncan 280 Book Reviews VoL. 28, No. 4 No. 42 Northern Saw-whet Owl, by R.J. Can- nings No. 62 Short-eared Owl, by D.W. Holt and S.M. Leasure No. 44 Merlin, by N.S. Sodhi, L.W. Oliphant, P.C. James, and I.G. Warkentin No. 63 Boreal Owl, by G.D. Hayward and P.H. Hayward No. 52 Red-tailed. Hawk, by C.R. Preston and R.D. Beane No. 75 Cooper’s Hawk, by R.N. Rosenfield and J. Bielefeldt No. 61 Burrowing Owl, by E.A. Haug, B.A. Millsap, and M.S. Martell No. 93 Flammulated Owl, by D.A. McGallum THE JOURNAL OF RAPTOR RESEARCH A QUARTERLY PUBLICATION OF THE RAPTOR RESEARCH FOUNDATION, INC. (Founded 1966) EDITOR IN CHIEF Carl D. Marti ASSOCIATE EDITORS Keith L. Bildstein Fabian Jaksi6 Gary R. Bortolotti Patricia L. Kennedy Charles J. Henny Erkki Korpimaki BOOK REVIEW EDITOR Jeffrey S. Marks EDITOR OF RRF KETTLE Richard J. Clark CONTENTS FOR VOLUME 28, 1994 Number 1 Productivity and Nesting Chronology of the Cooper’s Hawk and Sharp-shinned Hawk in Missouri. Ernie p. Wiggers and Kevin j. Kritz 1 Tree Nesting by the Gyrfalcon {Falco rusticolus) in the Western Canadian Arctic. Joachim Obst 4 Winter Roost-site Use by Female American Kestrels (Falco SPARVERIUS) IN LOUISIANA. J. Sean Doody 9 Nest-site Selection by the Hobby (Falco subbuteo) in Poplar Plantations in Northern Italy. Giuseppe Bogliani, Francesco Barbieri and Eugenio Tiso 13 Releasing Montagu’s Harrier (Circus pygargus) by the Method of Hacking. Manel Pomarol 19 Sexual Differences in Conspecific Territorial Defense of Marsh Harriers (Circus AERUGINOSUS) . Carmelo Fernandez and Paz Azkona 23 The Effect of Exercise on Plasma Activities of Lactate Dehydrogenase and Creatine Kinase in Red-tailed Hawks (Buteo JAMAICENSIS) . Shannon T. Knuth and Susan B. Chaplin 27 The Distribution of Centrorhynchus aluconis (Acanthocephala) and PORROCAECUM SPIRALE (NEMATODa) IN TaWNY OwLS (StRIX ALUCO) FROM Great Britain. F. j. Mclnnes, D. W. T. Crompton and J. A. Ewald 34 Diet, Breeding Success, and Nest-site Selection of the Short-toed Eagle (Circaetus gallicus) in Northeastern Greece. Christos G. viachos and Nikolaos K, Papageorgiou 39 Short Communication Peromyscus Carcass in the Nest of a Flammulated Owl. Brian D. Linkhart and Richard T. Reynolds 43 Abstracts of Presentations Made at the Annual Meeting of the Raptor Research Foundation, Inc 45 Manuscript Referees 72 Number 2 Bone Digestion and Intestinal Morphology of the Bearded Vulture. David C. Houston and Jamieson A. Copsey 73 Behavior of Colonial Common Kestrels {Falco tinnunculus) During the Post-Fledging Dependence Period in Southwestern Spain. Javier Bustamante . . 79 Goshawk Diet IN A Mediterranean Area OF Northeastern Spain. SantiManosa .. 84 The Antipredator Vocalizations of Adult Eastern Screech-Owls. Thomas McKell Sproat and Gary Ritchison 93 Behavior and Activity of Rehabilitated Buzzards {Buteo buteo) Released in Northern Italy. Davide Csermely and Carlo Vittorio Corona 100 Short Communications Parathion Poisoning of Mississippi Ejtes in Oklahoma. J. Christian Franson 108 Status and Reproduction of the Peregrine Falc;on at a Coastal Lagoon in Baja California Sur, Mexico. Aradit Castellanos, Federico Salinas-Zavala and Alfredo Ortega-Rubio 110 Successful Nesting by a Pair of Bald Eagles at Ages Three and Four. Daniel W. Mulhern, Michael A. Watkins, M. Alan Jenkins and Steve K. Sherrod 113 Traps for Capturing Territorial Owls. S. M. Redpath and I. Wyllie 115 Letters 118 Dissertation Abstracts 123 Number 3 A Symposium on Using Nest Boxes to Study Raptors: Do the Boxes Provide Virtual Reality? Frederick R. Gehlbach, Coordinator 125 Effect of Nest-box Size on Nest-site Preference and Reproduction in American Kestrels. Gary R. Bortolotti 127 Value of Nest Boxes for Population Studies and Conservation of Owls in Coniferous Forests in Britain. Steve J. Petty, Geoff Shaw and David I. K. Anderson 134 Facts and Artefacts in Nest-box Studies; Implications for Studies of Birds of Prey. Anders Pape M0ller 143 Selection and Use of Nest Sites by Barn Owls in Norfolk, England. Paul N. John.son 149 Nest-box Versus Natural-cavity Nests of the Eastern Screech-owl: An Exploratory Study. Frederick R. Gehlbach 154 The Post-fledging Dependence Period of the Lesser Kestrel {Falco naumanni) In Southwestern Spain. Javier Bustamante and Juan Jose Negro 158 Habitat Characteristics of Great Horned Owls in Southcentral Pennsylvania. Thomas E. Morrell and Richard H. Yahner 164 Parent-offspring Relations During the Post-fledging Dependency Period in THE Black Kite {Milvus migrans) in Japan. Kimiya Koga and Satoshi Shiraishi I7l Hematology and Occurrence of Hemoparasites in Migrating Sharp-shinned Hawks {Accipiter striatus) During Fall Migration. Lauren v. Powers, Mark Pokras, Kim Rio, Cathy Viverette and Laurie Goodrich 178 A Raptor Roadside Survey in Western Turkey and Eastern Greece. WadeL. Eakle 186 Short Communications Preparation of Avian Material Rec^overed erom Pellets and as Prey Remains. Beth Ann Sabo and Roxie C. Laybourne 192 Use of Mist Nets and a Live Great Horned Owl to Capture Breeding American Kestrels. Karen Steenhof, George P. Carpenter and James C. Bednarz 194 Letter 197 Book Reviews Edited by Jeffrey S. Marks 199 In Memoriam; Richard R. Olendorff, 1943-1994. Robert Lehman 204 Number 4 Reevaluating Delineated Bald Eagle Winter Roost Habitat in Lava Beds NatiONAI. Monument, California. Thomas J. Stohlgren and Chris A. Farmer 205 Dispersal of Bald Eagles Fledged in Texas. David W. Mabie, M. Todd Merendino and David H. Reid 213 Abundance and Distribution of Nesting Golden Eagles in Hudson Bay, Quebec. Francois Morneau, Serge Brodeur, Robert Decarie, Suzanne Carriere and David M. Bird 220 Blood Parasites of Raptors in Florida. DonaldJ. Forrester, Sam R. Telford, Jr., Garry W. Foster and Gordon F. Bennett 226 Ectoparasites of the Spotted Owl. John E. Hunter, R. J. Gutierrez, Alan B. Franklin and David Olson 232 Effects of Raptors on the Activity of Sandgrouse. Peter N. Ferns and Shelley A. Hinsley 236 Seasonal Abundance of Black Kites Associated with the Rubbish Dump of Madrid, Spain. Guillermo Blanco 242 Diet of Urban and Suburban Tawny Owls {Strix aluco) in the Breeding Season. Andrzej Zalewski 246 Diet Composition of the Long-eared Owl in Central Slovenia: Seasonal Variation IN Prey Use. DavorinTome 253 Short Communications Breeding Density and Brood Size of Rough-legged Hawks in Northwestern Quebec;. Serge Brodeur, Francois Morneau, Robert Decarie, Juan J. Negro and David M. Bird 259 Effects of Prescribed Fires on Habitat Use by Wintering Raptors on a Texas Barrier Island Grassland. Felipe Chavez-Ramirez and Felipe G. Prieto 262 Winter Diei of Long-eared Owi.s {Asia otus) in the Po Plain (Northern Italy) . Paolo Galeotti and Luca Canova 265 Refinements to Selective Trapping Techniques: A Radio-controlled Bow Net and Power Snare for Bald and Golden Eagles. Ronald E. Jackman, W. Grainger Hunt, Daniel E. Driscoll and Frank J. Lapsansky 268 Letter 274 Book Reviews Edited byJeffreyS. Marks 276 THE RAPTOR RESEARCH FOUNDATION, INC. (Founded 1966) OFFICERS PRESIDENT: Michael W. Collopy SECRETARY: Betsy Hancock VICE-PRESIDENT: David M. Bird TREASURER: Jim Fitzpatrick BOARD OF DIRECTORS EASTERN DIRECTOR: Brian A. Millsap CENTRAL DIRECTOR: Thomas Nicholls MOUNTAIN & PACIFIC DIRECTOR: Karen Steenhof CANADIAN DIRECTOR: Paul C. James INTERNATIONAL DIRECTOR #I : Jemima Parry-Jones INTERNATIONAL DIRECTOR #2: M. Isabel Bellocq DIRECTOR AT LARGE #1: Jim Bednarz DIRECTOR AT LARGE #2: Robert E. Kenward DIRECTOR AT LARGE #3: Keith L. Bildstein DIRECTOR AT LARGE #4: Josef K. Schmutz DIRECTOR AT LARGE #5: Paul F. Steblein DIRECTOR AT LARGE #6: Katherine McKeever EDITORIAL STAFF JOURNAL EDITOR: Carl D. Marti, Department of Zoology, Weber State University, Ogden, UT 84408-2505 U.S.A. ASSOCIATE EDITORS Keith L. Bildstein Fabian Jaksk GaryR. Bortolotti Patricia L. Kennedy CharlesJ. Henny Erkki Korpimaki BOOK REVIEW EDITOR: Jeffrey S. Marks EDITOR OF RRF KETTLE: Richard J. Clark The Journal of Raptor Research is distributed quarterly to all current members. Original manuscripts dealing with the biology and conservation of diurnal and nocturnal birds of prey are welcomed from throughout the world, but must be written in English. Submissions can be in the form of research arti- cles, 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'/2 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 nu- merals. 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 ( 6 th ed., 1983) or an- other authoritative source for other regions. Subspecific identification should be cited only when per- tinent 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 poli- cy are outlined in “Information for contributors,”/. Raptor Res., Vol. 27(4), and are available from the editor. 1995 ANNUAL MEETING The Raptor Research Foundation, Inc. 1995 annual meeting will be held on 1-4 November at the Duluth Entertainment and Convention Center in Duluth, Minnesota. Details about the meeting and a call for papers will be mailed to Foundation members in the summer, and can be obtained from Dan Varland, Scientific Program Chairperson, ITT Rayonier, Inc., Northwest Forest Re- sources, P.O. Box 200, Hoquiam, WA 98550, (telephone 206 533-7000; e-mail DAN_VARLAND@IT- TRAYNR.CCMAIL.COMPUSERVE.COM), and Gerald Niemi, Local Chairperson, Natural Re- sources Research Institute, University of Minnesota Duluth, Duluth MN 55811 (telephone 218 720-4279; e-mail GNIEMI@SAGE.NRRI.UMN.EDU). For information about the associated sympo- sium “A Comparison of Forest Raptor Responses to Forest Management — A Holarctic Perspective,” contact Gerald Niemi. Raptor Research Foundation, Inc., Awards Recognition for Significant Contributions^ The Dean Amadon Award recognizes an individual who has made significant contributions in the field of systematics or distribution of raptors. Contact: Dr. Clayton White, 161 WIDE, Department of Zoology, Brigham Young University, Provo, UT 84602 U.S.A. Deadline August 15. The Tom Cade Award recognizes an individual who has made significant advances in the area of captive propagation and reintroduction of raptors. Contact: Dr. Brian Walton, Predatory Bird Research Group, Lower Quarry, University of California, Santa Cruz, CA 95064 U.S.A. Deadline: August 15. The Fran and Frederick Hamerstrom Award recognizes an individual who has contributed significantly to the understanding of raptor ecology and natural history. Contact: Dr. David E. Andersen, Department of Fisheries and Wildlife, 200 Hodson Hall, 1980 Folwell Avenue, University of Minnesota, St. Paul, MN 55108 U.S.A. Deadline: August 15. Recognition and Travel Assistance The James R. Koplin Travel Award is given to a student who is the senior author of the paper to be present- ed at the meeting for which travel funds are requested. Contact: Dr. Petra Wood, West Virginia Cooper- ative Fish and Wildlife Research Unit, P.O. Box 6125, Percival Hall, Room 333, Morgantown, WV 26506-6125 U.S.A. Deadline: Deadline established for conference paper abstracts. The William C. Andersen Memorial Award is given to the student who presents the best paper at the annual Raptor Research Foundation Meeting. Contact: Ms. Laurie Goodrich, Hawk Mountain Sanctuary, Rural Route 2, Box 191, Kempton, PA 19529-9449 U.S.A. Deadline: Deadline established for meeting paper abstracts. Grants^ The Stephen R. 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: Alan Jenkins, George Miksch Sutton Avian Research Center, Inc., P.O. Box 2007, Bartlesville, OK 74005- 2007 U.S.A. Deadline: 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, 9384 Hito Court, San Diego, CA 92129-4901 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.