NESTING BIOLOGY AND BEHAVIOR OF THE MADAGASCAR
Transcript
NESTING BIOLOGY AND BEHAVIOR OF THE MADAGASCAR
J RaptorRes.34(2):120-125 ¸ 2000 The Raptor ResearchFoundation, Inc. NESTING BIOLOGY AND HARRIER-HAWK BEHAVIOR OF THE MADAGASCAR (POLYBOROIDES RADIATUS) IN NORTHEASTERN MADAGASCAR RUSSELL THORSTROM ThePeregrine Fund, 566 WestFlyingHawk Lane, Boise,1D 83709 U.&A. GIUSEPPE LA MARCA Via Plinio, 9, 20129-Milano, Italy ABSTRACT.--We studied the endemic MadagascarHarrier-Hawk (Polyboroides radiatus)during the 1998 breeding seasonon MasoalaPeninsula,northeastern Madagascar.One nestingpair wasobservedfrom courtshipto 1 wk into the post-fiedging period (September-January). Copulationsstarted25 d prior to egg layingand peaked7 d before egg laying.Two eggswere laid and the female and male incubated for 61.6% and 35.5% of 189.8hr of nestobservations, respectively. Incubationlasted39 d from laying to hatchingof the first egg.Both eggshatchedbut the second-hatched youngdied from siblicideat 7 d of age.The survivingyoungfledgedat 50 d of age.Frogs(Boophis and Platypelis spp.),geckos( Uroplatus and Homopholis spp.)and nestlingbirdswerethe mostnumerouspreytaken,representing 82% of 160 identified prey items. K•Y WORDS: MadagascarHarrier-Hawk;Polyboroidesradiatus;nestingbiology; behavior;, siblicid•, foodhabits. Biologiade anidacitn y comportamientode Polyboroides radiatusen el norestede Madagascar RESUmEN.--Estudiamos al endtmico Polyboroides radiatusdurante la estacitn reproductivade 1998 en la Peninsulade Masoala,noreste de Madagascar.Una pareja en anidaci6n fue observadadesdeel cortejo hastauna semanadesputsde que los pichonespudieron volar (Septiembre-Enero).Las ctpulas com- enzaron25 diasantesde la posturade huevosy su pico ocurri6 7 diasantesde la postura.La nidada fue de doshuevos,la hembray el machoincubaronel 61.6% y el 35.5% respectivamente de las 189.8 horas de observacitn en el nido. La incubacitn tttvo una duraci6n de 39 dias desdela posturahastala eclositn del primer huevo. Los dos huevoseclosionaron,el segundopichtn fue eliminado por su hermano a los sietedias de edad. E1pichtn sobrevivientepudo volar a los 50 diasde edad. Las ranas (Boaphis y Platypelis spp.)geckos(Uroplatus y Homopholis spp)y pichonesde avesfueron las presasmas abundantesrepresentando el 82% de las 160 presasidentificadas. [Traducci6n de Ctsar Marquez] Langrand 1990, del Hoyo et al. 1994). Existinginformation comesmainly from incidental and anecdotal observations(Dee 1986, Milon et al. 1973) and data on food habits (Goodman and Pidgeon 1991, Karpanty and Goodman 1999), but litfie is known of its breeding biology (Langrand 1990, del (Thorstrom et al. 1995, Rent de Roland et al. 1996, Hoyo et al. 1994). Here, we report on the nesting Berkelman 1996, Rent de Roland in press,Thor- biologyof the MadagascarHarrier-Hawk in an unstrom 1999, Thorstrom and Rent de Roland in disturbed lowland tropical forest of northeastern press). The genus Polyboroides containstwo species, Madagascar. the Africa Harrier-Hawk (P. typhus)which is con- STUDY AREA AND METHODS fined to mainland Africa and the MadagascarHarThe west coast of the Masoala Peninsula is roadless and rier-Hawk (P. radiatus)which occursonly on Mad- composedof a mosaicof slash-and-burnclearings,secagascar.The latter speciesis common throughout ondary growthand primary rain forests.The maturelowMadagascarbut has been little studied (Dee 1986, land rain forest of the Masoala Peninsula has a canopy Our knowledge of African tropical raptors,mainly from the southern subtropics,has increasedin recent yearsbut the raptors endemic to the western Indian Ocean islands are still poorly known (Virani and Watson 1998). During the last decade several studies have focused on Malagasyraptors 120 JUNE2000 MADAGASCAR HARRIER-HAWK NESTINGBIOLOGY 121 height near 30 m with emergent trees, high fioristic diversity and steep mountainous topography (Guillaumet 1984). Elevationsrange from 0-1200 m. Averageannual period with dry branchesmaking up the remaining 10.3% (21). The female contributed 79.8% (162) rainfall recorded at Andranobe Field Station (AFS) was and the male 20.2% (41) of the greenery through6049 mm from 1991-96 (range = 4468-7002 mm) out the breeding period. Both the male and female (Thorstromet al. 1997). Monsoonrainsand cyclonesoc- were observedconstructingthe nestduring all daycur betweenDecember and March, but rain falls steadily light hours (0500-1700 H), but most of the conbetween April and August. September to November are the driest monthsreceiving,on average,481 mm (8%) struction took place in the morning hours between of the annual rainfall recorded during 1991-96 (Thor- 0600-1100 H; 82% of the nesting material wasaddstrom et al. 1997). ed to the nest before Daily observationswere made at a harrier-hawknest from 0500-1700 H, initially from the ground and later the last deliveryof greenery to the nest on 7 January 1999, two daysafter the young fledged from from a tree hide, with 10X binoculars from distances of 25-45 m. Observations from the hide were also made using a 16-48X spottingscope.Data collectedincluded notes on adult and nestling behavior, nest attendance and identification of prey and provisioning rates. The length, breadth and depth of the nest were measured with a metric tape and nest height wasmeasuredwith a plumb line from the nest to the ground. Nest tree height was estimated in meters above the nest height. Egg volume was estimated from the equation V = 0.51LB2, whereL is the length and B the breadthof the the 1200 H. The female made nest tree. We observed 167 prey items delivered by the male and female to the nest from courtship to the post-fiedging period. The harrier-hawks had a broad diet delivering at least 15 different taxa during the nesting period. Prey wasalwayscarried and exchanged at the foot of the nest by the male to the female. Of these, 160 were identified at least to classlevel. Frogs (Boophisand Platypelisspp.) toegg (Hoyt 1979).The equation(V•gest- Vsmanest/Vsm•n½st) was used to calculate the percent difference in volume between the two eggsin the nest (Edwardsand Collopy 1983). RESULTS The pair of MadagascarHarrier-Hawks nestedin a tree near one of the main research trails, approximately 2.0 km north of AFS, from 1991-98 (Borge 1992). The nest was a large, stick structure approximately18 m above the ground within the canopyand slightlybelow a ridge on a north facing slope. The pair fledged one young every year the taled 31.3% (50), lizards (Uroplatus and Homophohs spp.) 23.8% (38), birds ( Otusrutilusand Couaspp.), mainly nestlings (Phyllastrephusspp. and Coua spp.), 26.9% (43), and rodents (Eliurusspp. and Rattus spp.) and lemurs (Lepilemurmustelinusand Microcebus spp.) 17.5% (28) of the identified prey items. Of interest was the occurrence of 7 lemurs and the pair constructeda new nest 50 m north- delivered during the nestling period by both the male and female. Nestling birds accounted for 86% of the 43 avian prey items delivered during the nestling period. The male delivered 131 prey items from courtship to the post-fledgingperiod and the female delivered 36 prey items from the beginning of incubation to the post-fledgingperi- east of the old nest site. In 1996, this nest measured od. 76 X 72 cm, with 7 cm interior and 47 cm exterior We observed the pair for 152.4 hr during the courtship period which lasted from the first week nest was observed. In 1995, the nest tree fell down depths. Two other nestswere located in 1998 2530 km to the east of AFS. They had lengths and widthsmeasuring85 X 50 cm and 90 X 70 cm and interior depths of 5 and 9 cm and exterior depths of 45 and 46 cm. Nest heights were 30.8 and 31.9 m and nest tree heights were 33 and 36 m. In 1998, we made observationstotaling 677 hr from courtship (September) to the post-fledging period (January) at the nest nearest to AFS. We observed 203 items of nesting material delivered to the nest throughout the nesting period. Of them, 90 (44.3%) were deliveredduring courtship, 31 (15.3%) during incubation, 81 (39.9%) during the nestlingperiod and 1 (0.5%) during the postfledgling period. Green branchesmade up 89.7% (182) of the greenery delivered during the nesting of Septemberto 7 October 1998. During this period, the female made 64 deliveriesof nesting ma- terial consistingof 57 green sprigs and 7 dead branches. The male delivered 24 green sprigsand 3 dead branches.We observed80 copulationsduring the 4-wk courtship period. The earliest copulation wason 13 September1998, 25 d before egg laying,and copulationswere mostfrequent 7 d before egg laying on 1 October 1998. On that day, we observed the pair copulate eight times during a 10.5 hr nest observationperiod. A total of 76 copulations totaling 1634 sec was observed for an averagecopulation time of 21.5 sec (range = 10.534.0 sec). The duration of copulations gradually increasedfrom 13.0 sec 25 d prior to egg laying to 122 THORSTROM AND LA MARCA 34.0 sec 5 d prior to egg laying and subsequently tapered off with no copulationswere observedafter the first egg waslaid. Copulationsoccurred at the nest and on perches of the nest tree and nearby trees. During copulations both sexesexhibited the red facial flushing described by Thurow and VOL. 34, No. 2 519 pieces of food while the younger nestling received only 59 pieces.After the younger nestling died in the nest, the female ate most of it and fed some to the survivingnestling. The female was the sole attendant at the nest during the nestling period and she brooded the Both nestling,fed it and deliveredgreenery76 times(66 sexesgavea high, shrill call during copulations.We green sprigsand 10 dry branches). The female also observed the male feed the female 24 prey items. delivered 28 prey items consistingof 14 (50%) Fifty percent (12) were frogs, 12.5% (3) lizards, geckos (Uroplatusspp. and Homopholisspp.), 5 12.5% (3) rodents, 8.33% (2) birds, 4.17% (1) (18%) birds,4 (14%) frogs,3 (11%) lemursand 2 snake and 12.5% (3) unidentified items. (7%) rats.During this period, the male only delivIn 1994 and 1996, 2 two-eggclutcheswere mea- ered prey to the female and nestling. The female sured and weighed. In 1994, the first egg weighed alwaysvocalized a high "pe-ee-ee-ee" (Langrand 46.0 g and measured57.0 X 42.3 mm and the sec- 1990) at the nest or near the nest as the male arond eggweighed47.0 g and measured54.4 X 41.2 rived with food. The male delivered greenery only min. In 1996, the first egg weighed 48.0 g and mea- once. The male delivered 80 prey items that consured49.6 X 38.8 and the secondeggweighed59.0 sisted of 33 (41%) birds, 15 (19%) rodents, 14 g and measured54.5 • 41.6 min. The percent dif- (18%) lizards,12 (15%) frogs,4 (5%) lemurs and ference in volume between the two eggs in 1994 2 (2.5%) unidentifieditems.The youngfledgedon was 10.5% and in 1996 was 26.1% (i = 18.4% + 5 January 1999, at 50 d of age. 0.05, -1SE). The eggs were buff, marbled and We observed the young for 40.4 hr during the spotted with reddish-brown post-fledging period from 6-14 January. During Nest observationstotaled 189.8 hr during incu- this period, the female delivered I green branch bation, from 8 October-16 November. One egg and spent2.9% of the time at the nest.The female was laid on 8 October and the second egg waslaid delivered7 preyitemsconsisting of 5 geckos,I frog Black (1981) for the African Harrier-Hawk. between 10-12 October. The female delivered and I unidentified item. The male delivered 2 greeneryto the nest 18 times (17 green and I dry) frogs, I gecko and I unidentified item during this and the male 13 times (all green branches) during period. the incubation period. During incubation, the DISCUSSION male delivered 23 prey items to the female, of which 82.6% (19) were frogs. The female also The Madagascar Harrier-Hawk occurs throughbrought a lizard to the nest that she ate herself out Madagascar,except on the high plateau. Deafter an incubation exchange with the male. The spite the fact that it is the fourth most common total incubation time for the female and male was raptor on the island (Langrand and Meyburg 116.8 (61.5%) and 67.6 (35.6%) hr, respectively 1984), it hasnot been studiedin great detail. Thorand both adults were absent from the nest for only strom and Watson (1997) found the Madagascar 2.9% (5.3 hr) of the time during incubation. The Harrier-Hawk at sevenof eight avian inventory sites incubation period from laying to hatching of the throughout the Masoala Peninsula confirming its first-laid egg was 39 d. widespreadoccurrence in the rainforest of this reNest observationstotaled 294.7 hr during the gion. nestling period from 16 November 1998, when we For diurnal raptors in general, nest building observedthe first hatchednestlingat 0719 H, to 5 takes place in the morning and males are mainly January 1999. We documented asynchronous responsiblefor nest construction(Newton 1979). hatching with at least a 2-d interval between hatch- Our results showed that harrier-hawks added most ing of the first and second eggs.The secondegg nesting material to the nest before 1200 H but the hatched on 18 November 1998. From 18-25 No- vember, we observed the oldest young strike the younger nestling822 timeswith its beak, eventually leadingto its death on 25 Novemberat 7 d of age. During the one week period of 18-25 November, the female was observedfeeding the oldestyoung female was the main nest builder. Thurow and Black (1981) reported that breeding activity of African Harrier-Hawks in South Africa begins in late spring before summer rains begin. Tarboton and Allan (1984) were more specific indicating that hatching coincideswith peak pro- JUNE 2000 •AG^SCo;m HARRIER-HAWK NESTINGBIOLOGY duction of passerinenestlingsfrom October to December. Breeding activity of the MadagascarHarrier-Hawk was similar beginning in August and 123 eggs,e.g., the American Swallow-tailedKite (Elanoidesforficatus)in the northern Neotropics (Gerhardt et al. 1997), the Augur Buzzard(Buteoaugur) ending in January (Tarboton 1978). Nestlings in mainlandAfrica (Gargett 1970) and the African hatched during November when conditionswere Harrier-Hawk (Tarboton et al. 1990, del Hoyo et at their driest and most passerineswere breeding, al. 1994). Tarboton and Allan (1984) recorded 10 and fiedging occurred during January when the 2-egg clutches of the African Harrier-Hawk in rainy seasonhad begun. Transvaal but seven nests contained only one The three nestsof the MadagascarHarrier-Hawk young. A 2-eggclutch wasdocumented at the study we studiedwere similarbut averagedslightlylarger nest of the harrier-hawkand we suspectedthat sity nests ob(75 cm in diameter X 20 cm wide) and higher licide occurred at 8 other harrier-hawk (10.3 m) than nestsdescribedby Tarboton and Al- served from 1991-98. lan (1984) for the African Harrier-Hawk. Similar Edwardsand Collopy (1983) describedtwo types to the Transvaal region where many nests were of siblicide,obligateand facultative,in large eagles used in consecutiveyears,the pair of harrier-hawks with asychronoushatching and a high percentage we observedoccupied one nest site from 1991-95 of volume difference within two-egg clutches and a newly constructedsite from 1996-98. ()10%). In a specieswith obligatesiblicide,the MadagascarHarrier-Hawks broke off branches second young to hatch alwaysdies, e.g., Aquila powithin a 20-50 m radius of the nest tree. After demarina, A. verreauxiiand Stephanoaetus coronatus livering greenery to the nest, they frequently cut (Edwardsand Collopy 1983). In facultativesiblicioff the leavesand placed them in the nest bowl de species,the second-hatchedyoung sometimes and then positionedthe leaflessbranch to the ex- dies, usually during food restriction, and egg volterior to form a nest rim. Green branches ap- ume differenceis (10%, e.g, A. chrysaetos, A. aupeared to be preferred over dry branches,perhaps dax, A. rapax,A. clanga,Hieraaetusfasciatus,H. penThe African becausethey offered both material for construct- natus, and Haliaeetus leucocephalus. ing the nest and fresh leavesfor lining the nest Harrier-Hawk appearsto exhibit faculativesiblicide bowl. Perhaps, the greenery provides insulation (Tarboton et al. 1990). We feel that the Madagasand protection for eggs and nestlings,and posi- car Harrier-Hawk may be more like the Swallowtioning of twigsin a nest rim createsa border for tailed Kite (Gerhardt et al. 1997) and Augur Buzmaintaining them in the center of the nest. zard (Gargett 1970) and has obligate sublicide. At 7 d before egglaying,we observedthe highest The mean egg volume of 18.4% for the Madagasfrequency (8 times) of copulationsduring a nest car Harrier-Hawk fell into the category associated observation period and the average copulation with the large obligate siblicide eagles (Edwards time graduallyincreasedto the longestduration5 and Collopy1983). We found the MadagascarHarrier-Hawk to have d prior to egg laying. Basedon our observations, we suspectthat the optimum time for insemination similar reproductive parameters to the African of harrier-hawksoccursfrom 5-7 d before egglay- Harrier-Hawk, with an incubation period of 39 d ing. Optimum inseminationsfor a captiveNorth- compared to 35-36 d, a nestling period of 50 d ern Goshawk (Accipitergentilis)via artificial insem- compared to 45-55 d and occurrence of siblicide, inafion occurred 4 d before the first egg (Berry though two young may sometimesfledge (Tarbo1972). Thurow and Black (1981) observed one ton and Allan 1984, Tarboton et al. 1990, and del pair still copulafingafter egg laying and until the Hoyo et al. 1994). Previousinformation reported on food habitsof nestlingwas8 d Of age. In this pair of Madagascar Harrier-Hawks,no copulationswere observedafter the MadagascarHarrier-Hawk comes from Rand egg laying. (1936), who found carrion (Tenrececaudatus),spiSiblicide has been documented for some marine der, insects,locusts,frog, lizards,snakeand a small birds (Proctor 1975, Anderson 1990) and large ea- mammal (Mus musculus)in the diet. Recently, in gles,includingthe MadagascarFishEagle (Haliaee- the dry climate of extreme southeasternMadagastusvociferoides), with a clutch sizeof two eggs(Mey- car, Goodman and Pidgeon (1991) documented burg 1974, Edwardsand Collopy 1983, Gargett the harrier-hawkpreying on flying foxes (Pteropus 1993, Watson et al. 1996). It has been documented rufus) and Karpanty and Goodman (1999) docufor only a few smaller birds of prey that lay two mented it feeding on adult and infant Verreaux's 124 THORSTROM AND. LA MARCA Sifaka (Propithecus verreauxi)during the courtship period. In the courtship period of September 1997, we observedone partiallyeaten dwarf lemur (Cheirogaleus major) lying at the base of the nest tree and, during nest observations,we documented lemurs (Microcebus and Lepilemur)in the diet only in the nestling period. Frogs were the predominant prey type during the courtshipand incubationperiodswhile nestlingbirdswere the major prey typestaken during the nestlingperiod. In 85 prey itemsof the AfricanHarrier-Hawk,nestling birds composed31% of them (Thurow and Black VOL. 34, No. 2 and ANGAP for their collaboration with The Peregrine Fund's Project in Madagascar.This work was supported by grants from the Disney Corporation. LITF, RATUPm CITED ANDERSON, DJ. 1990. Evoloution of obligate siblicidein boobiesß1. A test of the insuranceegg hypothesis.Am. Nat. 135:334-350. BF,RKELMAN, J. 1996. Breeding biology of the Madagascar Bnzzard in the rain forest of the Masoala Peninsula Condor 98:624-627. BF,Pmy,R.B. 1972. Reproductionby artificialinsemination in captive Northern Goshawks.J. Wildl. Manageß36. 1981). 1283-1288. The high percentage of nestling birds, Scops BORGF,,L. 1992. First studies on Madagascar HarrierHawk and Henst'sGoshawk.Pages73-74 in R. Watson Owls (Otus rutilus), nocturnal geckosand frogs in [ED.], Madagascarproject progressreport I, 1991 and the diet may be explained by the unique foraging 1992. The Peregrine Fund, Boise, ID UßS.Aß strategyused by MadagascarHarrier-Hawks.They have an unusualmorphologicaladaptation,an intertarsaljoint that permits them to flex their legs backward, aswell as forward (Burton 1978), which allows them to use their feet to probe and extract prey from crevices,holes in trees, rotting trunks and epiphytes.Such an adaptationmay enable harrier-hawksto extract nestling birds from difficultto-reachplaces.We had only one piece of evidence indicating that these harrier-hawksrobbed bird nests. It occurred when the male arrived BURTON, PJ.K. 1978.The intertarsaljoint of the harrier- hawks Polyboroides spp. and the Crane Hawk Geranospiza caerulescens. Ibis120:171-177. DEE, T.J. 1986. The endemic birds of Madagascar.ICBP, Cambridge, U.K. EDWARDS, T.C., JR.ANDM.C. COLLO?Y. 1983. Obligate and faculative brood reduction in eagles:an examination of factors stuck to its face. Further evidence on nes- fratricide. Auk 100:630-635. gur Buzzard.Ostrich 41:256-257. at the nest ß1993. The Black Eagle. AcademicPress,London, U.K. with a very small nestlingbird and egg shell fragments that influence GARGETT,V. 1970. The Cain and Abel conflict in the Au- GERHARDT, R.P., D.M. 1997. Siblicide in GERHARDT AND M.A. VASQUEZ Swallow-tailed Kites. Wilson Bull fling predation occurredwhen Ren• de Roland et 109:112-120. al. (1996) observed Madagascar Harrier-Hawks GOODMAN,S.M. ANDM. PmC;EON.1991. MadagascarHarfeeding on nesfiing Henst's Goshawks(Accipiter rier Hawk Polyboroides radiatuspreying on flying fox Pteropusrufus. Ostrich62:215-216. GUILLAUMET, J.-L. 1984. The vegetation:an extraordinary diversity.Pages27-54 in A. Jolly,P. Oberl• and R. A1bignac lEDs.I, Key environments:Madagascar.Perga- henstii) on two occasions. In the dry southeasternpart of Madagascar, Karpanty and Goodman (1999) found twopairsof harrier-hawksfeeding mainly on mammals and birds, thus differing from our observationsin the rainmon Press, New York, NY UßS.A. forestwhere they fed predominatelyon frogsand DELHOYO,j., A. ELLIOTANDJ. SARGATAL. 1994. Handbook of the birds of the world. Vol. 2. New World birds. Our studyoccurred in the wettestregion of Vultures to Guineafowl. Lynx Edicions, Barcelona, Madagascar,the northeastern coastal rainforest, Spain. and although it wasbased on observationsat only one nest, it concurred with Karpanty and Good- Hovr, D. 1979. Practicalmethodsof estimatingvolume and freshweightof bird eggs.Auk96:73-77. man (1999) in showing that the MadagascarHarKARPANTY, S.M. ANDS.M. GOODMAN.1999. Prey profile of rier-Hawk is a generalist and its diet can vary bethe MadagascarHarrier-Hawk (Polyboroides radiatus)in tween regionswithin Madagascar. southeasternMadagascar.J. RaptorRes33:313-316. LANGRAND, O. 1990. Guide to birds of Madagascar.Yale ACKNOWLEDGMENTS We thank R. Watson and W. Burnham of The Pere- grine Fund for makingthis studypossible.Specialthanks to R. Watson and L. Kiff for comments on earlier dr',fits and M. Marin of Western Foundation of Vertebrate Zo- ologyfor an obscurecitation.We thank the Direction des Eaux et For•ts, CommissionTripartite, Projet Masoala Univ. Press, New Haven, CT U.SßAß -- ANDB-U. MEYBURG. 1984. Birds of prey and owls in Madagascar:their distribution, statusand conservation. Pages3-13 inJ.M. Mendelsohn and C.W. Sapsford lEDS.], Proc. 2nd symposiumon At•ican Predatory Birds. Natal Bird Club, Durban, South Africa. JUNE2000 MAr)AGASCaX HARRIER-HAWK NESTINGBIOLOGY MEYBURG, B-U. 1974. Sibling aggressionand mortality among nestling eagles.Ibis 116:224-228. MILON,P.,JJ. PETTER ANDg. RANDRIANASOLO. 1973. Faune de Madagascar.XXXV. Oiseaux. ORSTOM and ANDR.T. WATSON.1995. Repeated sightingsand first capture of a live MadagascarSerpent-EagleEutriorch•' astur confirms the survival of this species. Bull. Bnt Ornithol. haviour NEWTON,I. 1979. Populationecologyof raptors.Buteo --AND SD U.S.A. --, and habits of Mada- Kestrel. Ostrich 70:149-151 R.T. WATSON. 1997.Avianinventoryand key serv. Int. 7:99-115. South Polar Skua Catharacta maccormicki. Ibis 117:452- RAND, A.L. 1936. The distribution of the Banded speciesof' Masoala Peninsula, Madagascar.Bird Con- PROCTOR, D.L.C. 1975. The problem of chick lossin the 459. Club. 115:40-45. 1999. A description of the nests,diet, and be- CNRS, Tananarivo, Madagascarand Paris, France. Books, Vermillion, 125 A. ANDRIANARIMISAAND L.-A. RENt DE ROLAND. 1997. Weather at Andranobe Field Station, western MasoalaPeninsula.Pages127-130 in R. WatsonlED ], gascarbirds. Bull. Ame•:Mus. Nat. Hist. 72:143-499. Masoala Prqject progressreport IV. The Peregrine RENI•DEROLAND,L.-A. 2000. Breeding biologyof the FrFund, Boise, ID U.S.A. ances'sSparrowhawkAccipiter francesiiin a lowland --AND L.-A. RENI• DE ROLAND. 2000. First nest derainforest of northeastern Madagascar.Proceedings scription, breeding behaviour and distribution of the IX Pan At¾icanOrnithological Congress:inpress. MadagascarSerpent-EagleEutriorchis astur.Ibis 142 in , R. THORSTROM ANDR.T. WATSON.1996. Breeding press. records and nestling predation of Henst's Goshawk THUROW,T.L. ANDH.L. BLACK.1981. Ecologyand behavon Masoala Peninsula, Madagascar.Ostrich67:168iour of the Gymnogene. Ostrich52:25-35. 170. VIRANI, M. AND R.T. WATSON.1998. Raptors in the east TAm3OTON, W.R. 1978.Avian populationsin TransvaalsaAti-ican tropics and western Indian Ocean islands: vanna. Proceedings IV Pan Atkican Ornithological stateof ecologicalknowledgeand conservationstatus Confkrence. OstrichSuppl. J. RaptorRes.30:28-39. --AND D.G. ALk4N. 1984. The status and conserva- tion of birds of prey in the Transvaal.TransvaalMuseum Monograph No. 3. --, P. PICKFORDAND B. PICKFORD. 1990. At¾ican birds WATSON, R.T., S. THOMSETT, D. O'DANIEL AND R. LEWIS 1996. Breeding, growth, development, and managemant of the MadagascarFish-Eagle(Haliaeetusvoc•feroides). J. RaptorRes.30:21-27. of prey. Cornell Univ. Press,Ithaca, NY U.S.A. THORSTROM, R., B. DAMATARY,E TOTO, M. BABA, V. BABA Received30 July 1999;accepted7 February2000