GENERAL NOTES

Bald EaglePredation on NocturnalSeabirds.--Bald Eagles (Haliaeetus leucocephalus) are opportunisticpredators and scavengers(Retfalvi 1970, Dunstanand Harper 1975, Shetrodet al. 1975, Grubb and Coffey 1982).In Alaskamost eagles depend primarily on fish (Ofelt 1975, Grubb and Hensel 1978), but thoseresiding near large seabirdcolonies rely heavilyon marinebirds for food (Murie 1940, Krog 1953, Shetrodet al. 1976). In thisnote we describethe contentsof 24 pelletsregurgitated by Bald Eaglesand speculate on the methodsused by eaglesto obtain nocturnal,burrow-dwelling seabirds. Our observationswere made from 10-12 June 1977 on Petrel Island,which is part of the ForresterIsland NationalWildlife Refuge,about 140 km southwestof Ketchikan, Alaska.Petrel Island is about65 ha, risesto an elevationof 100 m, and is heavilyforested with Sitka spruce(Picea sitchensis). The islandis fiddled with nestingburrows of Fork- tailed and Leach'sstorm-petrels (Oceanodromafurcata, O. leucorhoa), and Cassin'sAuklets (Ptychoramphusaleuticus). The number of activeburrows averaged 2.7/m 2 over much of the islandin 1976. Estimatesof breedingpopulations for thesespecies on the islandin 1976 were: 689,000 Leach's Storm-Petrels, 89,000 Fork-tailed Storm-Petrels, and 23,000 Cassin'sAuklets (Sowls et al. 1978).Other nestingspecies which were potentialBald Eagle prey included PelagicCormorants (Phalacrocorax pelagicus), Black Oystercatchers(Hae- matopusbachmani), Glaucous-winged Gulls (Larusglaucescens), Common Murres (Uria aalge), Pigeon Guillemots (Cepphuscolumba), Ancient Murrelets (Synthliboramphusantiquus), Rhinoceros Auklets (Cerorhincamonocerata), Tufted Puffins (Fraterculacirrhata), Horned Puffins (Fraterculacorniculata), and severalpasserines (Willett 1915). We observedup to 10 Bald Eagleson Petrel Island on 10 June 1977; most were immatures.A pair nestedon the islandin 1977. On 10June 1977,we collected24 Bald Eaglepellets from hundredsthat lay scattered over the island.Analysis of the pelletssuggested that nocturnalseabirds, notably Cassin's Auklets,were importantprey of Bald Eagles(Table 1). Superficialexamination of addi- tionaldecomposing pellets suggested that all were composedprimarily of feathersand smallbones. Bald Eaglesalso were observedcapturing or carryingGlaucous-winged Gull chicks(2), a Leach'sStorm-Petrel, a black rockfish(Sebastes melanops), and a Northern Fulmar (Fulmarusglacialis). We suspectedthat eagleswere excavatingseabirds from burrowsbecause we found severalareas where surfacesoil was disturbed by digging.No land mammalslarger than a mouse(Peromyscus sitkensis), a vole (Microtus coronarius), and a shrew(Sorex obscurus) occur on thisisland group and it seemslikely that birdswere responsible. The diggingsdid not seemcharacteristic of the burrowingspecies of seabirds.Our suspicionswere reinforced when,on severaloccasions, we flushedup to 6 eaglesfrom the groundin the forestduring the day. D. J. Forsell(pers. comm.) confirmed that Bald Eaglesare capableof digging whenhe observedan immatureeagle excavating a Tufted Puffinburrow in the eastern Aleutian Islands.G. V. Byrd (cf. Trapp 1979) reportedGlaucous-winged Gulls inspecting storm-petrelburrow entrances for unwaryadults and chicks.Bald Eaglesmay combine a similarstrategy with excavationof shallowburrows to capturestorm-petrels. Bald Eagleson PetrelIsland were activejust beforedark and werefrequently ob- servedflying aboveand within the forest at this time. On the night of 12 June two immatureBald Eagleswere discoveredroosting 10 m from the groundin dead spruce trees.At 01:30 a third immatureeagle was found roostingon a 1 m high sprucestump. Severalincriminating storm-petrel feathers were stuckto its head and bill. The eagle appearedunwilling to fly in the darkness,but uponour closeapproach it crashedclumsily throughthe underbrush.Two freshlykilled Leach'sStorm-Petrels were found near the roost.Presumably the eaglehad capturedthe storm-petrelsshortly after they beganto arrive at the colonyor emerge from their burrows.Storm-petrels and aukletsare so numerouson Petrel Island that eagleson the ground could probablycapture them at night through chanceencounter alone. E. P. Baileyand D. R. Nysewander(pers. comm.) independentlyobserved Bald Eagleschasing storm-petrels on the groundat night on St. Lazaria Island near Sitka, Alaska.

407 408] GeneralNotes J.Field Ornithol. Autumn 1982

TABLE1. Relativeimportance of preyin 24 BaldEagle pellets from PetrelIsland in June 1977.

No. & % of pelletsin No. & % of prey whichprey occurred in pellets Species No. % No. %

Fork-tailed Storm-PetreP 2 6 2 5 Leach's Storm-Petrel I 6 17 6 15 Unid. Storm-PetreP 4 11 4 10 Common Murre 1 3 1 3 Cassin's Aukle0 14 40 18 46 Rhinoceros Auklet • 2 6 2 5 Tufted Puffin 1 3 1 3 Unid. Alcidae 2 6 2 5 Unid. Bird 2 6 2 5 Unid. Mammal 1 3 1 3

i Nocturnalat breedingcolony.

Bald Eagles may use severalmethods to secureseabirds on Petrel Island. Ancient Murrelets, Cassin'sAuklets, RhinocerosAuklets, and Tufted Puffins may occasionallybe taken directlyfrom the water.Auklets are alsocaptured after they are injured by striking trees during their incomingflights at night (Heath 1915). We observedboth Cassin's Auklets and RhinocerosAuklets sitting stunnedon the ground after forcefullystriking treesat night. Storm-petrelscould only be taken by Bald Eagleson land as they are too small and agile for an eagle to catch in flight and are only seennear their coloniesat night. The abundanceof bird remainsin Bald Eagle pelletson Petrel Island as well as our observationsof eaglesat night suggestthat nocturnalseabirds are the principalprey of eagleson Petrel Island in the summer.We suggestthat the large number of Bald Eagles attractedto this smallisland, particularly immatures which are not restrictedto nestsites, and their unique feeding strategiesare directly related to the abundanceof nocturnal seabirds. We thank the U.S. CoastGuard for logisticalsupport. W. Clark, C. J. Lensink, W. E. Southern,and J. L. Trapp providedmany helpful commentson the manuscript.

LITERATURE CITED DUNSTAN,T. G., ANDJ. F. HARPER.1975. Food habitsof Bald Eaglesin north-central Minnesota.J. Wildl. Manage. 39:140-143. GRUBB,T. G., ANDR. J. HENSEL. 1978. Food habitsof nestingBald Eagleson Kodiak Island, Alaska. Murrelet 59:70-72. , ANDM. A. COFFEY.1982. Evidenceof Bald Eaglesfeeding on freshwatermussels. Wilson Bull. 94:84-85. HEATH, H. 1915. Birds observedon Forrester Island, Alaska during the summer of 1913. Gondor 17:20-41. KROC.,J. 1953. Notes on the birds of Amchitka Island, Alaska. Gondor 55:299-304. MURIE, O.J. 1940. Food habits of the northern Bald Eagle in the Aleutian Islands, Alaska. Gondor 42:198-202. OFELT,G. H. 1975. Foodhabits of nestingBald Eaglesin southeastAlaska. Gondor 77: 337-338. RETFALVI,L. 1970. Food of nestingBald Eagleson SanJuan Island, Washington.Con- dor 72:358-361. SHERROD,S. H., J. A. ESTES,AND C. M. WHITE. 1975. Depredationof seaotter pups by Bald Eaglesat AmchitkaIsland, Alaska.J. Mammal. 56:701-703. Vol..•:•,,o 4 GeneralNotes [409

, C. M. W•I•rE, AND F. S. L. WILLIAMSON. 1976. Biology of the Bald Eagle on Amchitka Island, Alaska. Living Bird 15:143-182. SowLs,A. L., S. A. HA•rc•, ANDC. J. LE•S•NV, 1978. Catalog of AlaskanSeabird Col- onies.U.S. Dept. Interior, Fish and Wildl. Serv.,FWS/OBS-78/78. TRAPP,J. L. 1979. Variation in summer diet of Glaucous-wingedGulls in the Western Aleutian Islands:an ecologicalinterpretation. Wilson Bull. 91:412-419. WXL•q•r, G. 1915. Summer birds of Forrester Island, Alaska. Auk 32:295-305. Axa•o•,'¾ R. D•GANC,•A•D JAY W. NELSON,U.S. Fishand WildlifeService, 1011 E. Tudor Rd., At•chorage,Alaska 99503. Received24 July 1981, accepted4 Apr. 1982.

A Scale for Weighing Birds at Habitual Perches.--Repeatedweight measurements of adult birds in the wild are often difficult to obtainbecause of problemsassociated with trapping and disturbingthe birds. Body weightsfor only 2 speciesof the world'sraptors havebeen followed through the breeding season (Newton 1979); daily and seasonalweight changefor other bird speciesare only slightlybetter known (see Palmer 1963). As weight is usuallyconsidered a measureof a bird'scondition, weight changes are usefuldata since they may reflect the energeticcosts associated with a particular activityand/or a bird's foragingefficiency (Drent and Daan 1980). The simpleand non-disturbingmethod for obtainingaccurate weights of wild birds describedhere shouldthus prove helpful to field ornithologists. We designedand testeda compact,weatherproof, and relativelyinexpensive elec- tronic scalefor birds at perches.This scaletakes advantage of the fact that many birds use habitualperches and will often shift to artificialperches if theseare placedat their habitualsites (Hall et al. 1981; pers. obs.).Our balanceuses a transducerwhich is easily incorporatedinto an artificialperch and from whichremote readingsof weightcan be made. We used 2 suchscales during studiesof breedingOspreys (Pandion haliaetus), and found the method accurate and effective. We think the method could also be used with other birds. Although Siblyand McCleery (1980) have developedan electronicbalance for remoteweighing of ground-nestingbirds, their methoddoes not appearreadily mod- ified to weigh birds at perches. Designand construction.--Theperch-scale consists of 2 main units: (1) the perch, in- corporatingthe transducer(Fig. 1); and (2) the batteries,amplifier (Fig. 2), and recorder unit, whichare easilycontained in a 14 x 28 x 17.5 cm weatherproofarmy surplusam- munition box. A bird landing on the perch deflectsthe tranducer,a metal beam with 4 deflectionsensing strain gauges bonded to it, via the plunger.As the tranducerdeflects, the straingauges change resistance which develops a voltageproportional to weight.This voltageis then read remotely.The 2 unitsare portable(ca. 8 kg with batteries),and easily installedat a site (Fig. 3). The scale'saccuracy is improved if perch length is kept short so a bird remains centeredover the plungerwhen it landson the perch;such centering minimizes friction betweenthe plunger and its guide. Silicon spray or a hard wax applied to the plunger alsohelps reduce friction. A 24-cm perch gave accurateweights and was adequatefor Ospreysto feed and reston. We experimentedwith a more expensiveplunger mechanism (a machined stainlesssteel plunger containedby teflon bushingswithin the guide), but foundthat precisionwas not improvedenough to justifythe addedcost and effort. The perch unit basecan be quicklyattached to a perch site usingC-clamps or duct tape. It helpsto modifya perchsite so that the perchbase rests on a firm, flat surface whichkeeps the perch steadyand perpendicular.We installedwooden perch sitesbeside 18 Ospreynests (Fig. 3), alongwith "mock"perch-scales (wooden perches that roughly resembledthe scales),to ensurerapid installationand acceptanceof our scalesas we moved them amongnests. However Ospreys often acceptedperch-scales quickly without being previouslyexposed to "mock"perches. The perch unit can be weatherproofedby sealingthe strain gaugeswith a film of