j. RaptorRes. 39(2):149-155 ¸ 2005 The Raptor ResearchFoundation, Inc.

BREEDING BIOLOGY AND FOOD HABITS OF THE MADAGASCAR KESTREL (FALCO NEWTONI) IN NORTHEASTERN MADAGASCAR

LILY-ARISONRENE DE ROLAND,JEANNENEY RABEARIVONY, HARILALAINAROBENARIMANGASON, AND GILBERTRAZAFIMANJATO ThePeregrine Fund's Project in Madagascar,B.P 4113, Antananarivo(101), Madagascar

RUSSELL THORSTROM 1 ThePeregrine Fund, 5668 WestFlying Hawk Lane, Boise,ID 83709 U.S.A.

ABSTRACT.--Westudied MadagascarKestrels (Falconewtoni) on Masoala Peninsula, northeastern Mada- gascarduring the 1997 and 1998breeding seasons. We locatedfive nestsites and observedeight nesting attemptsduring the two breeding seasons.All nestswere in tree cavitiesand averaged13.8 + 2.0 m (SE) above the ground in trees averaging22.8 +- 0.8 m (SE) in height (N = 5 nests). Egg laying took place from mid-Septemberto the first week of October.The modal clutch sizewas 4 + 0.9 (N = 6 nests, range 3-5 eggs).The incubationperiod averaged28 d, varying from 27-29 d (N = 5 nests).Hatching occurred from the middle of October to the first week of November with young fledging in late No- vember.Of 24 eggslaid in six nests,13 (54%) hatched,and seven(54%) of thosehatchlings fledged; thus, a total of 1.2 young fledged per breeding attempt were produced and overall nest successwas 50%. The MadagascarKestrel diet of 338 identifiedprey wascomposed of 93.8% lizards(N = 317), 2.6% insects(N = 9), 2.4% amphibians(N = 8), and 1.2% birds (N = 4). KEYWORDS: MadagascarKestrel; Falco newtoni; nestingbiology; food habits;nests; productivity.

BIOLOG• REPRODUCTIVA Y I-IJd3ITOS ALIMENTICIOS DE FALCO NEWTONI EN EL NORESTE DE MADAGASCAR

RESUMEN.-•Estudiamosindividuos de la especieP•tlco newtoni en la peninsulaMasoala en el norestede Madagascardurante las estacionesreproductivas de 1997y 1998.Localizamos cinco nidos y observamos ocho intentos de nidificaci0n durante los dos periodos reproductivos.Todos los nidos se encontraron en cavidadesen •rboles a una altura promedio de 13.8 + 2.0 m (EE) sobre el sueloyen •rboles con una altura promedio de 22.8 + 0.8 m (EE) (N = 5 nidos). La puestade huevosocurri6 desdemediados de septiembre hasta la primera semanade octubre. E1 tamafio modal de la nidada fue de 4 -+ 0.9 huevos (N = 6 nidos, rango 3-5 huevos).E1 periodo promedio de incubacitn fue de 28 dias, variando entre 27-29 dias (N = 5 nidos). La eclositn ocurri6 desdemediados de octubre hastala primera semanade noviembrey el abandonodel nido por parte de los polluelosocurri6 a fines de noviembre.De los 24 huevospuestos en seisnidos, 13 (54%) eclosionaron,y siete (54%) de los polluelosque eclosionaron abandonaron el nido. Por lo tanto, se produjeron un total de 1.2 juveniles que abandonaron el nido por intento de nidificaci0n,y el fixito de nidificaci0ngeneral fue del 50%. La dieta de F. newtoni,basada en 338 presasidentificadas, estuvo compuesta en un 93.8% por lagartijas(N = 317), en un 2.6% por insectos(N = 9), en un 2.4% por antibios(N = 8) yen un 1.2% por aves(N = 4). [Traduccitn del equipo editorial]

Madagascarhas three residentspecies of falcons: most common raptor in Madagascar(Siegfried and MadagascarKestrel (Falconewtoni newtoni), Banded Frost 1970), detailed information on its biology Kestrel (F. zoniventris),and Peregrine Falcon (F. pe- and natural history are lacking (Langrand and regrinusradama), and two wintering species,Eleo- Meyburg 1984, Langrand 1990). This speciesis nora's Falcon (F. eleonorae)and Sooty Falcon (F. widely distributed throughout Madagascar and is concolor).Although the MadagascarKestrel is the found in open grasslands,savannah habitat, de- graded forests,and in the vicinity of villagesand towns. This falcon has two distinctive color • Email address:[email protected] morphs:pale and rufous (Siegfriedand Frost1970,

149 150 RENE DE ROLAND ET AL. VOL. 39, NO. 2

Cade 1982, Langrand 1990). In this paper, we re- sions,nest depth, and internest distance) were measured port basic information on the breeding biology after the young had fledged.

and food habits of the MadagascarKestrel. RESULTS

STUDY AREA AND METHODS One pair of MadagascarKestrels was trapped This studywas conducted within the secondaryhabitat and marked during this study. The adult female and cultivated terrain surrounding Ambanizana Village measurements were wing chord, 200 mm; tail (49ø57'E, 15%7'S), situated at the western boundary of length, 129 mm; tarsuslength, 38.2 mm; and body the Masoala National Park (MNP; Robenarimangason 1999). mass, 122 g. The adult male measurementswere Most of the peripheral zones of the MNP, are com- wing chord, 195 mm; tail length, 110.5 mm; tarsus posed of degraded forest, secondaryforest, and fallow length, 37.0 mm; and body mass,110 g. land intermixed with cultivated areas (crops include Five nesting attemptswere observedduring the cloves, coffee, vanilla, and bananas). The MNP itself con- first season, 1997-98, and three were documented sistsof 230 000 ha of primary forest with a typicalcanopy height of 25 m and with some emergent trees exceeding during the second field session,1998-99. Of the 30 m (Guillaumet 1984). The forested terrain is moun- five nesting pairs observed during the two breed- tainous, and lacks roads and a trail system.The altitude ing seasons,three pairs were composedof a pale- in the park ranges from sea level to 1230 m (Nicoll and morph male and a rufous-morphfemale, one pair Langrand 1989). The current studywas undertaken in the lower altitude and cultivated areas ranging from sea was of a rufous-morphmale paired with a pale- level to 200 m above sea level. The annual mean rainfall morph female, and one pair included both rufous- at AndranobeField Station,7 km southof the studysite, morph individuals. is 6049 mm (Thorstrom et al. 1997). The dry seasonis Nest Characteristics. During this study, all nests from October to mid-December (Rene de Roland 2000). observed for MadagascarKestrels were placed in The climate is mild and the annual mean temperature varies from 18-31øC. natural-tree cavities,with a decayedwood substrate, The studyperiod wasfrom September1997-January in secondary and human-modified habitat. The 1998 and September1998-January 1999, which coincides trees identified for nestingwere mandrorofo (Tra- with the kestrel's breeding season. During September chylobiumrerrucosum), hintsia (Afzeliabijuga), lalona and October, we searchedfor nesting pairs by following vocalizing and flying adults. Observationswere made at (Weinmania sp.), and dead unidentified snags. least 50 m from nests, and five and three nests were mon- MadagascarKestrel nestsaveraged 13.8 -+ 4.5 m (N itored during the 1997 and 1998 breeding seasons,re- = 5) above the ground in trees averaging22.8 -+ spectively. 1.9 m (N = 5) in height. Nest cavitieswere oval- Nestswere observedfrom courtship up to post-fledg- shaped and measured 61 _+ 55.1 cm (N = 2) by ling period, and 521 hr 22 min of observationwere com- pleted. We recorded copulationfrequency (per hr and 33.5 _+ 16.2 cm (N = 2) with an interior depth of d) and duration (sec), clutch size (number), incubation 22.5 + 3.5 cm (N = 2). The distance between two period (d), nest attendance (percent of observation neighboringnests averaged 675 _+386.2 m (N = ume), dispersalof young (d), and productivity.Egg di- 4, range 300-1200 m). mensions (length and breadth), and egg and nestling Nesting Biology. Courtship activity involved vo- masswere measuredusing vernier calipers (with 0.01 mm of precision)and Pesolaspring-balance scales (0.1 g pre- calizations, nest site visits with food deliveries to cision), respectively.Daily nest observationswere made the female, and copulations.The courtshipperiod from 0600-1800 H with 10x binocularsand a zoom spot- wasmarked by flights (e.g., in open areasand over ting scope.While making observationsat nestsprey items trees) and accompaniedby moderate fluttering of were identified during prey deliveries. Kestrelswere trapped with a bal-chatri,a noosecarpet wings. During periods of inactivity,the kestrel pair fixed over the nest entrance, or a "wire hoop trap" (Ber- perched together in the top of dead branchesof ger and Mueller 1959, Thorstrom 1996). Morphometric trees. Courtship behavior,either flights or periods measurementstaken were: wing chord (mm), tail and of inactivity,was associatedwith loud calls "itsi, kit- tarsilength (mm), and body mass(g). One breeding pair si, kitsi, kitsi." During this period, the male's pri- of adult kestrelswas radiotaggedto estimate their rang- lng area. The kestrelswere radio-trackedfrom October- mary role seemed to be showingthe female poten- December 1998 by homing on foot, and locational fixes tial nesting sites. were recorded using an Eagle Explorer Global Position Copulationsusually occurred after prey deliver- System(GPS; Eagle Electronics,Catoosa, OK U.S.A.) with ies from the male to the female. Copulationsoc- 30-m of accuracy.The home range was estimatedby the minimum convex polygon (MCP) of the locationsusing curred on the top of dead branches.During cop- Ranges1V software (Kenward 1990). Severalnest site pa- ulations, the male emitted a "iiitsi, kitsi, kitsi, rameters (i.e., nest height, nest tree height, nest dimen- kitsi..." that continued to the end of this activity. JUNE 2005 BREEDINGBIOLOGY OF THE M•'d)^G^SC•V.KESTREL 151

Table 1. Number of prey items deliveredby male and female MadagascarKestrels at MasoalaPeninsula, Madagascar, during the breedingseasons of 1997 and 1998.

SEX COURTSHIP INCUBATION NESTLING POST-FLEDGING TOTAL

Male 13 110 130 18 271 Female 0 0 28 39 67 Total 13 110 158 57 338

The highest frequency of copulationsduring a 1 (Table 1). During this period, the adult female was hr period occurredbetween 0600-0700 H (N = 13 very aggressivetoward other cavity-nestingbirds of 91 observedcopulations). Copulationswere also (e.g., Madagascar Starlings [Hartlaubius aurata] frequent during the 0800-1000 H (N = 28) and and Broad-billed Rollers [Eurystomusglaucurus]), 1300-1500 H (N = 28). Ten daysprior to the onset and occasionallyattacked them. of incubation, the frequency of copulationsaver- Nestlings were observed wing exercising at the aged 9.1 - 0.4 (SE) times per day (N = 91 copu- cavity entrance at 3 wk of age. Two to three days lations, range 7-11). Copulationsaveraged 4.9 + before fiedging, two males weighed 112 g and 118 1.6 sec in duration (N = 44, range 3-8 sec). Three g, and one female was 128 g. Young fledged at 23- copulation attemptswere alsoobserved during the 24 d of age (N = 7 young). After fiedging, the late incubation and nestling period. young were observed perching together in trees Egg laying took place from mid-September to closeto the nest tree. During the secondweek after the first week of October. The earliest egg waslaid fiedging, they were observedcatching insectsand on 18 September1997 and the latestwas on 5 Oc- attackingprey, and the prey delivery rate by adults tober 1997. The mean clutch size was 4.0 + 0.9 (N decreased progressively.Young dispersed from -- 6 nests, range 3-5 eggs). Eggs measured 33.9 their natal areasat 44-45 d of age (N = 7 young). mm + 0.9 mm X 28 mm + 0.6 mm and their mass Reproductive Success.In six fully-documented was 14.3 + 0.6 g (N = 5 eggs)during the first week nestscontaining 24 eggs, 13 (54%) eggshatched of incubation. Eggs were pale white with dark and 7 (54%) of the hatchedyoung fledged. In to- brown spots. tal, seven young fledged from three successful Constant incubation seemed to be initiated fol- breeding attempts,for a productivityof 2.3 young lowing the laying of the second egg. The incuba- fledged per successfulnest. Overall productivity tion period averaged28 d and varied from 27-29 was 1.2 (7/6 pairs) young per nesting pair. Nest d (N= 5 nests: 27 d IN = 3], 29 d IN = 2]). successfor the 2 yr of this studywas 50% (3/6 Incubation was done by femalesonly, and they in- pairs; Table 2). cubated for 88.5% of the observation time and In the six nestingattempts during the two breed- were absent for 11.5% of the time (N = 153.1 hr). ing seasons,one nest failure occurred during in- The male's responsibilityduring this period was cubation and two during the nestling period. In providing food to the incubating female. Prey ex- 1997, we believe one nest failed during incubation changes occurred on branches of trees neighbor- due to nest competition with a pair of Broad-billed ing the nest tree. Rollers for the cavity,and the nest was abandoned Nestling and FledglingPeriod. Up to 6 d of age, by the kestrel pair prior to hatching. In another the young were constantlybrooded by the female instance in 1997, three nesdingswere killed and (96.5% of the observation time IN = 20.7 hr]). partially eaten by an unknown predator. In 1998, Broodingprogressively decreased to 38.7%just pri- one nest failed when the eggswere eaten and an- or to the time of fledging. Young were fed solely other nest failed during an intense 5-d rain storm by the female until about 15 d of age when they when the nestlingswere found dead in and below were able to feed themselves, were active in the the nest,possibly due to lack of sufficientbrooding nest, and completely covered with contour feath- by the adult female or due to water in the nest ers. After about 15 d of age, prey delivered by the cavity. adults was dropped into the nest cavityfrom the Food Habits. MadagascarKestrels used different entrance. Both the male and female provisioned hunting strategies,including hovering flight and the young with food during the nestling period perch hunting. Prey on the ground was hunted 152 RENV. Dr. ROkaND ET AL. VOL. 39, No. 2

Table 2. Reproductivesuccess of six fully-documentedbreeding attemptsof the MadagascarKestrel (Falconewtoni) during 1997 and 1998 at MasoalaPeninsula, Madagascar.

MEAN NUMBER NUMBER OF FLEDGLINGS/ NEST SUCCESS BREEDING NUMBER OF CLUTCH EGGS HATCHED YOUNG FLEDGED BREEDING PERCENT YEAR ATTEMPTS EGGS SIZE (%) (%) ATTEMPTS (N)

1997 4 16 4 10 (63) 4 (40) 1.0 (4/4) 50 (4) 1998 2 8 4 3 (38) 3 (100) 1.5 (3/2) 50 (2) Total 6 24 4 13 (54) 7 (54) 1.2 (7/6) 50 (6) from the air by stationary flight or "hovering," (malesaveraged 105 g [90-117 g, N -- 4] and fe- ending in the kestrelplunging down and capturing malesaveraged 144 [131-153 g, N -- 7]). We were the prey. The second method was scanningfor not able to document the extent of sexual dimor- prey from a high lookout, usually from a tree. phism, as we measured only one pair, and in this Once prey waslocated, kestrels bobbed their heads single instancethe male was 90% of the female's several times as if sighting in on the prey before mass. gliding down to seizetheir quarry. During the two All nestsfound in this studywere situatedin nat- study seasons,1997 and 1998, we recorded 370 ural-tree cavities. All cavities seemed to have devel- Madagascar Kestrel prey items, mainly those oped through decay,where a limb had broken or brought to nests;338 were identified. Platedlizards the tree's heartwood had a rotten opening the in- (Zonosaurusbrygooi and Z. madagascariensis)com- terior. Langrand (1990) reported that Madagascar prised93.8% (N = 317), insects2.6% (N = 9), Kestrel nests can also be found in cliff holes, under amphibians2.4% (N -- 8), and birds 1.2% (N = house roofs, and infrequently in old nestsof other 4). Adult males and females delivered different avian species.The SeychellesKestrel (E araea)and quantifies of prey during the different breeding Mauritius Kestrels(E punctatus)are known to use periods (Table 2). In total, of 338 identified prey, tree cavitiesfor nesting and also potholes in cliff 80.2% (N-- 271) and 19.8% (N-- 67) were deliv- faces(Temple 1977, Cade 1982, Watson1992). On ered by males and females,respectively. MasoalaPeninsula, rocky and potholed cliff faces During the 1998 breeding season,one nesting are extremely rare and are unavailablefor nesting pair of radio-taggedkestrels had an estimatedMCP sitesfor MadagascarKestrels. home range of 25.6 ha (N -- 30 locational fixes). In the central plateau region of Madagascar,this DISCUSSION speciesis known to occupybuildings for nestingas is the SeychellesKestrel (Watson 1992). However, The MadagascarKestrel is consideredto be one on the Masoala Peninsula humans are recent in- of the most common and widespread raptor spe- habitants in this region and large suitable build- cies throughout Madagascar(Rand 1936, Siegfried and Frost 1970, Langrand and Meyburg 1984). Fer- ings for nesting kestrelsare nonexistent.In south- guson-Leesand Christie (2001) stated "...this is western Madagascar,kestrels occupy abandoned probably the only Malagasyraptor to have gained nestsof other birds (e.g., Pied Crow [ Corvusalbus]; from deforestation, and is clearly under no Rene de Roland pers. obs.). In the northeastern threat." This speciesappears to be benefiting from region, old stick nestsare built (and usuallyoccu- deforestation (Langrand and Meyburg 1984), and pied) by larger raptorsinside forestedhabitat, and on the Masoala Peninsula, kestrels are occupying this tends to exclude kestrelsfrom using this type openings,secondary and human-modifiedhabitats of nesting structure. Consequently,the human adjacent to intact forest fragments and the primary degradation of forested habitat on Masoala Pen- forest (Rene de Roland 1994). insula has left isolated trees for nesting habitat for Cade (1982) commented that Madagascar Kes- MadagascarKestrels. Madagascar Kestrels do not trels show more pronounced sexual-size dimor- seemto be highly selectivein regard to tree species phism in comparisonto most kestrelspecies based or nest height they use,but are dependenton the on work by Siegfried and Frost (1970), who re- availabilityof suitable tree cavitieswithin the hu- ported males averaging72.9% of female mass man-modified habitat. On Masoala Peninsula, car- JUNE 2005 BREEDINGBIOLOGY OF THE MADAGASCARKESTREL 153

ity nesting by MadagascarKestrels differed mark- Like Seychelles Kestrels (Watson 1993) and edly from sympatricBanded Kestrels, which placed Mauritius Kestrels (Cade 1982), both female and their nests only inside arboreal-epiphytic ferns male MadagascarKestrels contributed to prey pro- (Thorstrom 1999, Rene de Roland et al. 2005). visioning during the nestling period. Madagascar During research conducted on raptors on Ma- Kestrelsdropped prey into the nest cavity during soalaPeninsula from 1991-97, no MadagascarKes- the latter stageof the nestling period, and this has trels were found nesting inside the intact primary alsobeen observedin American Kestrels(Balgooy- forest (Rene de Roland 1994, Robenarimangason en 1976). MadagascarKestrels were most aggres- 1999, Thorstrom and Rene de Roland 2000). In siveagainst other cavitynesters during the nestling contrast, the Mauritius Kestrel and SeychellesKes- period, and this has been observedwith other kes- trel did occupy and nest in dense forests, forest trel species(Balgooyen 1976). fragments and secondary forest patches (Cade MadagascarKestrels fledged at 23-24 d of age, 1982, Watson 1992, Cade and Jones 1993). much earlier than SeychellesKestrels at 35-42 d Most of the breeding biology of this speciesfol- (Watson 1993) and Mauritius Kestrels at 38-39 d lows the usual kestrel pattern and behavior, but (Cade 1982). The shorter nestling period for Mad- there are some exceptions.For instance, the court- agascar Kestrels might be attributed to their ad- ship period is marked by a distinctflight consisting aptation to nesting in open habitat, whereas the of a seriesof climbsand diveswith continuouspow- protracted breeding seasonof the Mauritius and erful wing beats as seen in American Kestrels (F. Seychelleskestrels are adapted to tropical forested sparverius;Willoughby and Cade 1964) and Euro- situations (Cade 1982, Watson 1992). pean Kestrels (E tinnunculus;Brown and Amadon Adult MadagascarKestrels continued to use the 1968), but we did not observe such courtship nest for prey deliveries to the fledglings. Young flights for MadagascarKestrels or the sympatric Madagascar Kestrels caught prey by 14 d after Banded Kestrelsin this region (Rene de Roland et fledging and young American Kestrelsalso were al. 2005). SeychellesKestrels do not display this successfulin catchingprey at 12-14 d after fledg- specializedcourtship flight either (Vesey-Fitzgerald ing (Balgooyen1976). The prey deliveredby adult 1940). Both sexesof MadagascarKestrels, Banded MadagascarKestrels decreased progressivelydur- Kestrels, as well as the SeychellesKestrel do emit ing the post-fledglingperiod until youngdispersed specificvocalizations during the courtship period and were independent at 44-45 d old. (Watson 1993, Rene de Roland et al. 2005). The In comparisonto an adjacent inland kestrel spe- copulation duration of 5 sec (range 3-8 sec) for cies, Seychellesand Mauritius kestrels,the nest suc- the MadagascarKestrel was slightly shorter than cessof 50% of MadagascarKestrels was lower due the sympatricBanded Kestrelwith a mean near 8 to the predation on eggsand nestlings,inclement sec (range of 5-10 sec; Rene de Roland et al. weather,and possiblythe smaller sample size. 2005). The MadagascarKestrel's diet was reported to Newton (1979) noted that the eggsof smaller comprise mainly of insects and some vertebrates raptor speciesare laid every couple of days,but in (Rand 1936, Cade 1982, Langrand 1990, Ferguson- this study we observed MadagascarKestrels laying Lees and Christie 2001). On Masoala Peninsula, eggsdaily. At one nest in 1997, five eggswere laid the kestrel's diet during the breeding seasonwas on five consecutivedays. The fresh egg massesfor vertebrates,and predominantly terrestrial plated SeychellesKestrels, 12.4 g or 14% of mean female lizards(Zonosaurus spp.; 93.8%, N = 317). This was body mass (N = 10 eggs;Watson 1993) and Amer- similar to the SeychellesKestrel, which fed mainly ican Kestrels,13.8 g or 11.2% of mean female body on skinks (Mabuyaseychellensis) and some day geck- mass(N = 53 eggs;Balgooyen 1976) are compa- os, (Phelsumaspp.) and the Mauritius Kestrel, rable to those of MadagascarKestrels (14.3 g or which consumedmostly day geckos (93%; Cade 10% of female body mass)from this study. 1982, Jones 1984, Watson 1992). The sympatric Cade (1960) noted that the genusFalco requires Banded Kestrel studied in the same area also 28-30 d of incubation. The incubation period for preyed on lizards, but more on arboreal species, SeychellesKestrels was 28-31 d (Watson 1992), for such as chameleons (Furciferand Calumna spp.) Mauritius Kestrels about 30 d (Cade 1982), and we and day geckos (Rene de Roland et al. 2005). found a similar period for the MadagascarKestrel In this study,almost half of the identified prey, rangingfrom 27-29 d. 158 of 338 (46.7%), were capturedduring the nest- 154 RENE DE ROLAND ET AL. VOL. 39, No. 2 ling period. The majority of prey was delivered by the world. Houghton Mifflin Co., New York, NY male MadagascarKestrels (82.3%, N = 130) rela- U.S.A. tive to females (17.7%, N = 28). Watson (1993) GUILLAUMET,J.-L. 1984. The vegetation:an extraordinary reported male SeychellesKestrels increased their diversity.Pages 27-54 in A. Jolly, P. Ober16, and R. Albignac lEDS.], Key environments:Madagascar. Per- hunting effort during nestling period, delivering gamon Press,New York, NY U.S.A. 92% of the food to the young. JONES,C.G. 1984. Feeding ecologyof the Mauritius Kes- In comparing the MadagascarKestrel with the trel. Page 209 in J. Mendelsohn and C.W. Sapsibrd two other closely-relatedinsular kestrels,the Sey- [EDS.], Proceedingsof the secondsymposium on Af- cheiles and Mauritius kestrels,and the larger sym- rican predatory birds. Natal Bird Club, Durban, South patric Banded Kestrel, this speciesseems to be Africa. adapted to open and disturbed habitats.Thus, the KENWARI•,R.E. 1990. RANGES IV: softwaretbr analyzing MadagascarKestrel may have benefited from de- animal location data. Institute of Terrestrial Ecology, forestation, human activities,and villages.Also, in Warham, U.K. comparison to the other two kestrels,the Mada- LANGRAND,O. 1990. Guide to the birds of Madagascar gascarKestrel frequently hovers for hunting, has Yale Univ. Press, New Haven, CT U.S.A. -- ANDB.-U. ME'mURG.1984. Birds of prey and owls two color morphs, and has a relativelyshort breed- in Madagascar:their distribution, status and conser- ing season.The Mauritius Kestrel is morphologi- vation. Pages3-13 inJ. Mendelsohn and C.W. Saps- cally similar to an accipiter and its habits and be- ford [EDS.],Proceedings of the secondsymposium on havior reflect this based on its occupancy of African predatory birds. Natal Bird Club, Durban, forested habitat, and the Seychelles Kestrel is South Africa. found both in forested and open habitat (Cade NEWTON,I. 1979. Population ecologyof raptors.Buteo 1982, Watson 1992). Books, Vermillion, SD U.S.A. N•co•.•., M.E. ANDO. LANGRAND.1989. Madagascar:revue ACKNOWLEDGMENTS de la conservationet des aires Protag6es.World Wide We would like to thank the Masoala technicians of The Fund tbr Nature International, Gland, Switzerland. Peregrine Fund for their assistancein data collection in RAND, A.L. 1936. The distribution and habits of Mada- the field. We are also indebted to the Association Nation- gascarbirds. Bull. Amez.Mus. Nat. Hist. 72:143-499. al pour la Gestion des Aires Proteg•es, Direction de la RENE DE ROIAND, L.-A. 1994. Metbode d'inventaire de Gestion Durable des Resources Foresti•res, and ga•tana- rapaces dans la presqu'ile de Masoala. M.S. thesis, narivoUniversity, The PeregrineFund's collaborators, tbr Univ. d'Antananarivo,Antananarivo, Madagascar. their invaluable administrative help, and especially,for permitting us to undertake this current studyat two dif- --. 2000. Contribution h l'&ude biologique, 6colo- ferent sites.We kindly thank L. Kiff; J. Bednarz, and two gique et 6thologique de trois espacesd'Accipiter dans anonymous reviewerstbr comments on the manuscript. la Presqu'Ile Masoala. Ph.D. dissertation, Univ. A special thanks the Liz Claiborne and Art Ortenberg d'Antananarivo, Antananarivo, Madagascar. Foundation, MacArthur Foundation, and Environment --, J. RABEARIVONY,G. RAZAFIMANJATO,H. ROBEN- Now for their financial support. ARIMANGASON,AND g. THORSTROM.2005. Breeding biology and diet of Banded KestrelsFalco zoniventns LITERATURE CITED on Masoala Peninsula,Madagascar. Ostrich. 76:32- BALGOO•N, T.G. 1976. Behavior and ecology of the 36. American Kestrel (Falcosparverius) in the Sierra Ne- ROBENARIMANGASON,H. 1999. Contribution •t l'&ude de vada of California. Univ. Calif. Publ. Zool. 103:1-83. la biologique,ecologique, et ethologiquede la famille BERGER, D.D. AND H.C. MUELLER. 1959. The bal-chatri: a de falconidae, faucon de newton, P•lco newtoni,et thu- trap for birds of prey. Bird-Banding30:18-26. con a ventre raye, E zoniventris,dans la partie occi- BROWN,L.H. AND D. AMADON.1968. Eagles, hawks,and dentale de la presqu'ile Masoala. M.S. thesis, Univ. falcons of the world. Country Lit• Books, London, d'Antananarivo,Antananarivo, Madagascar. U.K. SIEC,FR•ED, W.R. AND P.G.H. FROST. 1970. Notes on the C:•DE,T:J. 1960. Ecologyof the peregrine and Gyrthlcon MadagascarKestrel Falco newtoni. Ibis 112:400-402. populationsin Alaska. Univ. Calif. Publ. Zool.63:151- TEMPLE, S. 1977. The status and conservation of endemic 290. kestrelsin the Indian Ocean islands.Pages 78-82 in 1982. The falcons of the world. Cornell Univ. R.D. Chancellor [ED.], Proceedingsworld conference Press, Ithaca, NY U.S.A. of birds of prey. International Council for Bird Pres- --AND C.G. JONES.1993. Progressin the restoration ervation, Cambridge, U.K. of the Mauritius Kestrel. Conserv. Biol. 7:169-175. THORSTROM,R. 1996. Methods for capturing tropical for- FERGUSON-LEES,J. AND D.A. CHRISTIE.2001. Raptors of est birds of prey. Wildl. Soc.Bull. 24:516-520. JUNE 2005 BREEDINGBIOLOGY OF THE M^D^GASCARKESTREL 155

ß 1999. A description of nests,diet, and behaviour VESEY-FITZGERALD.,D.F. 1940. The birds of Seychelles. of the Banded Kestrel. Ostrich 70:149-151. Ibis, 14th S• 4:480-489. , A. ANDRIANARIMISA,AND L.-A. RENE DE ROLAND. WATSON,J. 1992. Nesting ecologyof the SeychellesKes- 1997. Weather at Andranobe Field Station, Western trel Falcoaraea on Mah•, Seychelles.Ibis 134:259-267 MasoalaPeninsula. Pages 127-130 in R. Watson[ED.], 1993. Breeding cycle of the SeychellesKestrel ProgressReport IV. The Peregrine Fund, Boise, ID Pages 73-79 in M.K. Nichols and R. Clarke lEDS], U.S.A. Proceedingsof the 1991 hawk and owl trust confer- , AND L.-A. RENE DE ROIAND. 2000. Status and con- ence, September 1991ß The Hawk and Owl Trust, servation of raptors on the Masoala Peninsula, Mad- London, U.K. agascar.Pages 35-41 in R.D. Chancellor and B.-U. W•LLOUC,HB¾, EJ. ANt)T.J. CAt)E.1964. Breeding behavior Meyburg [EDS.], Raptors at risk. World Working of the American Kestrel. Living Bird 3:75-96. Group on Birds of Prey and Owls, Hancock House, Berlin, Germany. Received7 October 2003; accepted 14 February 2005