Corella,2006, 30(2): 29-34 BEHAVIOUROF THE BLACK.FACEDWOODSWALLOW Artamus cinereus

IAN ROWLEY

53 Swan Street, Guildford. Western 6055

Received; 18 Mav 2005

A of the Black{aced WoodswallowArtamus cinereus was observed lor 22.5 hours during incubation and 32.5 hours duringthe nestlingstage, from a hide placedthree metresaway. The five attendingwoodswallows were individually colour-banded.Continuous recordings of vocalizations at the nest were made; sound spectrographsof significantvocalizations are presented.A restrictedrange of vocalizationswas heard at the nest.These includeda contact call (qur),nestling calls, greetingcalls (qua), at high intensityincluding a rapid 'quas 'qua' series of as a rattle segment.Alert, alarm and mobbing calls based on the syllable at increasing rates and intensities and shorter durations were also recorded. Other asoects of their behaviour sucn as locomotion,foraging, roosting, body care, courtshipand reproductionare described.

INTRODUCTION Manmanning, in the wheatbelt of Western Australia and followed from 1973-19'76. In 1975 a canvas hide was established within 'parent', three metres The family has recently becomethe of a nest (number 501) where the five attending were taxonomically, of an assemblagethat, besidesthe wood- individually recognizable.Observations of behaviour were made during hide-watches (n = , includesthe ,, , totalling 22.5 hours 7) during incubation and for 32.5 hours (n = 7) during the nestling stage. Where relevant these observations pitohuis and bristleheads(Schodde and Mason 1999).The have been supplemented with others from the literature. genusArtam&.r gives its name to this varied A continuous recording from just family that is largely united on the strength of a shared a Beyer 100 microphone placed below the nest was stored on a National Panasonic Cassette Recorder blue-greybill colour. Six of the elevenArtamus species RQ-42lDS and vocalizationsfrom the resulting 33 cassetteshave been occur in Australia, but relatively little is known about their analysed in this paper. Calls were analysed on a Macintosh iMac lifestyles, largely becausethey spend most of their time computer. Sounds were digitized at 16-bit resolution and edited using either flying high and fast when travelling, in pursuit of Soundedit 16, version 2 software (Macromedia Inc.); backgroundnoises and irrelevant sounds were removed as far as possible. their prey, or quietly Sound spectro- else sitting on a branch. This graphs were prepared using Canary 1.2.4 (Bioacoustics Res. Prog., does not leave much time for interestingstudy and, apart Cornell Laboratory of Ornithology). from sighting recordsand generalcomments, the literature is sparse. Keast (1958) sorted out the geographical RESULTS AND DISCUSSION distribution and movement of the Australian speciesand more recently Schoddeand Mason (1999) realignedthe Locomotion subspecies.Between 1959-60 Klaus Immelmannspent a very productive year in Australia and wrote extensive Mayr (1945) rated Artamur as the best fliers amongst accountsof his observations,but unfortunatelymost of the Oscines.Their flight is characterizedby short glides those on the woodswallows are in German and not easily interspersedwith rapid wing-beats. Unfortunately this available(Immelmann 1963, 1966).Clunie (1973, 1916) appearssuperficially similar to the flight of the Common described the behaviour of the White-breastedWood- Starling Sturnus vulgaris, which do not occur in Western swaflow A. leucopterusin and Recher and Schulz Australia.Several woodswallows have beenshot in mistake (1983) that of White-browedWoodswallow A. superciliosus for this invadingexotic. Black-facedWoodswallows fly fast in New South Wales.For the rest, the literatureconsists and with great agility when hunting close to the ground of brief snippetsof information (Chisholm 1909; Dove or when up high, but if the wind is very strong they make 1909;Heathcote 1931; Cameron 1933; D'Ombrain 1934; only brief sorties from convenientlow perchessuch as Coleman 1945; Rowley 1951, 1976; Hindwood 1956; fence lines or stray weeds left standing in stubble. Immelmann 1960, 1963, 1966;Austin 1972l'Bourke 1972.. On the ground woodswallowshop with both feet together Lowe and Lowe 1972; Sharland 1972). and appearclumsy since their short tarsi are placed to the This paper describesaspects of woodswallow behaviour rear of the body. They usually perch on exposedbranches seen during thirty years of fieldwork in most parts of from which they can easily launch and fly sorties.If they Australia,including an opportunisticstudy of Black-faced need to move along a branch, they shuffle sideways by WoodswallowsA. cinereusfrom 1973-1976(Rowlev 1999. moving one foot after the other. Long strong claws enable 2002). them to cling to the trunk of a all night when roosting, to shelter from a rainstorm, or to stealthily retreat to the METHODS under or rear surface of a branch to avoid being seen by As described in detail elsewhere(Rowley 1999), a group of Black- an approaching raptor or other predator. This later faced Woodswallows A. cinereus had been colour-banded at behaviourappears to be unique, as it is employedto avoid

29 30 l. Rowley:Behaviour of the Black-facedWoodswallow Artamus cinereus Corella30(2) being seen in what would otherwise be a very exposed where they chattered and preened.After 1G-15 minutes position and where flight might be even more dangerous they all took off and flew silently to the roost site for that (pers.obs.; G. S. Chapman,pers. comm.). night. At Manmanning severalof these sites were located with a view to netting the but only two birds were Foraging caught in this way. Most roost siteswere in verticalhollow in dead , but they also used the scars left after Barker and Vestjens (1990) list a wide range of insect spouts limb had broken off. On one night they roosted in families consumedby woodswallowsas well as three sorts a of an old White-browed Babbler of plant seeds(Xanthorrhoea, Enchylaena, and Rhagodia). the hollow centre supercilioJuJ nest, and on another in a Observation from a hide placed near a nest enabled 96 Pomatostomus underneath an Australian Raven Corvus food items to be identified to order or family (Table 4 in depression nest.The mean size of roosting assemblieswas Rowley 2002), although these data were biased in favour coronoides ll (7-20) birds with roosting sites being recorded in of the larger items. Sixty-eight were presumedto have been February(l), April (l), July (1), September(8), October taken from on or near the ground (a skink, centipedes, (4), (1). grasshoppers,crickets, phasmids,caterpillars and spiders) and November (blowflies, and 28 whilst in flight lacewings, dragonflies Immelmann (1960) descnbedBlack-faced Woodswallows and moths). Grasshoppersprovided more than a third of arriving at a pre-roosting gathering each carrying a small (1966) identifiable prey, but since Immelmann describes twig, which was dropped on arrival; we did not observe Black-faced Woodswallows in the Kimberley catching this behaviourat Manmanning. jumping grasshopperswhile flying close to the ground,the Manmanning birds may have been hunting in the same This sort of sociable nocturnal roosting scrum also has way, which would increase the proportion of prey taken been observed for Dusky Woodswallows and Little on the wing. Black-faced Woodswallows are obviously WoodswallowsA. minor (Coleman 1945; Hindwood 1956; specialistpredators of theseagricultural pests, as are other Bourke 1972; Rowley 2000 and pers. obs.). artamids (McGilp 1935). Even when they were less than sevendays old nestlingswere fed whole grasshoppers:the Bodycare legs protrudedfrom the bill until digestion made room for them to be swallowed. Preening of the body and flight feathers took place regularly whenever the birds were perched. Their A wide variety of techniquesare employed by foraging sociability was demonstratedby birds preeningeach other, Black-faced Woodswallows: particularly those areasthat were hard for an individual to was i. Trawling high above the vegetation for aerial reach, such as the crown and nape. This allopreening (pers.obs.). not confined to mated pairs but extended to all group members including fledglings. When such help was not 'lookout ii. By flying swift sorties from exposed perches' available woodswallows scratchedtheir headswith a foot to snatch flying or stationary prey and return to the passedover the lowered wing. Their wings are long and sameperch (pers. obs.). require special attention; in turn each primary was drawn through the bill, presumablyreconnecting any barbulesthat iii. By hovering before snatchinginsects and larvae from had become unhooked. This required space and was the ground or foliage (Cameron 1933). usually done by the itself, away from the others. 'following iv. By the plough' in cultivation paddocks, Sunbathing or rain-bathing has been reported by pouncing on prey exposed by the machinery (pers. Immelmann (1966) to involve ruffling the feathersso as obs.). to allow sun or rain to reach the skin of the bird. Nestlings v. Flying close to the ground and catching jumping do this by stretching their necks and orienting their bare grasshoppers(Immelmann 1966). throatsto the sun (seeReproductive behaviour, vi, below).

vi. Searchingeucalypt and other flowers for nectar,which Vocalizations they are thought to lap with their brush tongues, is recorded for Black-faced Woodswallows (Johnstoneand From analysis of 33 cassettesrecorded during hide- Storr 2004), and is reported also for several other watchesat the nest in 1975, the vocalizetionsof attending species(Chisholm 1909,White-browed Woodswallow; Black-facedWoodswallows are summarizedas follows: Marshall 1935, A. cyanopterus; Sedgwick 1947, Masked WoodswallowA. personatus; i. Contact calls were usually brief with variation in mood McKean 1969). or intent achieved by changesto the intensity or rate of repetition. At the lowest intensity, when a solitary 'talking vii. Occasionally,large prey are carried in the foot when sitting bird appearedto be to itself', this call the woodswallow flies to a place where it can complete could be rendered as qui (Fig. la), scarcely audible dismemberment;such prey may be held in the foot, more than five metres from the nest. More frequently torn to pieces and eaten (Sedgwick 1947). heard was a stronger,louder call renderedas qua (Fig. lb) that was used as a basic contact call given by Roosting individuals in a flock flying high, by a sitting bird Throughout the years, each evening shortly before sunset, communicating with others nearby or by birds visiting Black-faced Woodswallows gathered in a particular tree the nest to feed unbrooded young. June,2006 L Rowley:Behaviour of the Blac*-facedWoodswallow Artamus cinereus 31

.tr ; *fr{* $' $, *r ii ii .i i*$ tr***fr$t *,*i$S*:$T ta i Sf r $&,'

Figure l. (a). qui call given by bne sitting bird. (b) basic contact call qua. (c) Alarm call, a series of short slnrp qtas. (d). MobbinS call by several birds uttering loud, slnrt, harsh quas. The horizontal axis measures time (seconds) and tlrc vertical axis is frequency (kiloHerz). 32 l. Rowtey:Behaviour of the Black'facedWoodswallow Artamus cinereus Corella30(2)

{tt 2 'J|t kHz ililrilililililililil$il*[ ffiffiffiffift 1.5 2.0 2.5 3.0 4.0 5.0

relief bird. In this afier 1.5 seconds, the call develops Figure-into 2. Greeting call given by an incubating or brooding bird towards an approaching figure, with a a rattle oj increasing amptitude (as siown in the lower waveform plot) as short, sharp qt s are repealed louder and loude4 and ends series of quas. In the waie1orm plot oj sound pressure level (lower), the ienical axis conveys the relative loudness of each part of the call.

ii. Greeting Calls were given by the incubating or brooding intensity greeting calls. As the intensity of the vocalization bird as soon as an approaching relief was sighted. increases,either in a greeting situation or alarm/mobbing, These loud repeated quas became more and more the individual syllables are repeatedat a faster rate and excited, especially if the arriving bird had food to offer, become louder and much shorter (<0.1 s), and in the case in which casethe sitter beganvigorous wing-fluttering. of the mobbing call, harsher.The literaturementions vocal Frequently such seriesincorporated a distinctive rattle mimicry by woodswallows (Chisholm 1946; Frith 1969), consistingof a rapid seriesof short qua-type syllables but none was heard in the course of theseobservations' increasing in amplitude. The rattle was generally followed by a seriesof loud, longer quas (F\9. 2). Reproductiveb ehaviour rii. Nestling Calls were a series of briei high-pitched, loud The following observations supplement those already syllables in expectationof food, but subsidedrapidly describedin Rowley (1999,2002). if an alarm was given. i. Pairing. Pairs remained together throughout the year, so iv. Alert, Alarm andMobbing Calls were loud short sharp long as both remained alive. Outside the breeding qucs usually given by a bird on guard duty at a nest. seasonseveral groups might coalesceand forage,perch They were graded in intensity from Alert through Alarm and roost together.With the onset of breeding,groups to Mobbing. A single sharp Alert qua (Fig. lb) was separatedand defended,even from conspecifics,an area directedusually at the nestlingsto tell them to be quiet. within five metres of the nest-site. At Manmanning An Alarm (Fig. lc) was a seriesof shorter quas given were seldom closer than 100 metres.Elsewhere, when the sentry chasedan intruder that did not really in more arid parts of Australia, Immelmann (1966) pose any threat, such as a Western Gerygone Gerygone found Black-faced Woodswallows nesting much more colonially, with nests as close as 1.8 metres. In fusca, Brown-headed Honeyeater Melithreptus brevi- rostris, or Brown Honeyeater Lichmera indistincta, or 1983, G. Johnston (pers. comm.) saw Black-faced an intruder overhead. Mobbing occurred when the Woodswallows breeding in loose groups or colonies on intruder posed a real threat - such as a raven, a the northern Eyre Peninsula, . Nests perched raptor, an approaching human, a fox Vulpes were in small Hakeas (c. 2 m high) in Myall vulpes,or a reptile such as a monitor Varanussp. (Fig. and were spaced 1-5 metres apart. This spacing 1d). Mobbing consistedof a seriesof vigorous swoops appeared unnecessary,as there were more Hakeas accompanied by harsh, short, sharp qaas,' usually farther afield that were not used for nesting. This several birds joined in, resulting in an effective breeding event followed the break of a drought. tenifying cacophony. ri. Courtship. As with severalother woodswallows,Black- The vocalizationsdescribed here were heard in a much faced Woodswallows perform a beautiful pre-copulatory more restricted situation than those described for Dusky courtship display. Either member of the pair started Woodswallows (Rowley 2000). Both species have alarm proceedings,which usually began when the birds were and mobbing calls, and specific calls heard when a bird perched three to ten metres apart. The initiator part- approaches the nest. These are based on a short syllable spread both wings and waved them up and down in (0.2-0.3 s) that is termed qua from its sound in the low time with the rotation of the part-spreadtail. Sometimes June,2006 L Rowley:Behaviour of the Blac*-facedWoodswallow Artamus cinereus 33

the partner ignored this, and the display went no alarm calls from their parents. Fledglings were able to further; usually the partner responded with a like fly competently four days after leaving the nest and display and the two birds gradually achieved synchrony. were difficult to follow after this. In a very strong wind, After two or three minutes the male flew to the female one brood was found perched on a log and still being and mounted her, briefly. The entire display was fed by adults 14 days after fledging; three weeks later conductedin silence. they were still being allopreenedby adults. 11i.Copulation. This was invariably precededby the above CONCLUSION display and lastedtwo to three seconds.The male then dismountedand resumedperching nearby.The female Even though the details in this paper iue from one nest settled her ruffled feathers and usually remained where the participants had been colour-banded,and offered perchedwhere she was. a unique opportunity for observation, much of the behaviour described also applies to species of wood- iv. Nest building. After they had chosen the site, the pair swallows other than the Black-faced Woodswallow. In sometimes perched together nearby for several days speciesin which the basic social unit is often more than a before they started building. Although the nest is simple pair, with young birds remaining with their parents unsubstantialit took both birds several days to complete for more than a year and helping to rear young that are as single stems of grass, twigs and rootlets were not their own, a well-developed repertoire of social gatheredand placed in position. It is not known what behaviour and vocalizations would be expected. Some role, if any, supernumerarygroup members played at evidenceof this was seenin the complex vocal exchanges this stage because,like most birds, woodswallows do between a sitting bird and its replacement. Further studies not like to be watchedbuilding. on vocalizations in a ranse of contexts are needed to v. Incubation was performed by all group members.At a establishthis. nest observedfor 22 hours the senior male and female incubatedfor 25 per cent and 38 per cent of the time, ACKNOWLEDGMENTS one helper in its third year sat for 20 per cent, while a I thank Graeme Chapman and Craig Bradley for their help outside bird just enteringits secondyear appearedto be refused normal working hours and my then chief the late Harry Frith, who drd access to the nest on several occasions and only not know but encouraged me. I also thank Peter Fullagar and Eleanor managedto incubate for 5 per cent of the time. The Russell for their help with the spectrographs, Michael Brooker for his comments on an earlier version of this paper and Alan Burbidge and eggs were covered for 92 per cent of the observation Greg Johnston as referees. period (Table I in Rowley 1999). REFERENCES vi. Brooding was also performed by all group members when the nestlings were incompletely feathered (the Austin. C. N. (1972). Woodswallows in Western Victoria. Aust. Bird Watcher4:2ll-212. first seven days) and before the day warmed up. 'The Barker, R. D. and Vestjens,W. J. M. (1990). Food of Australian Nestlings of this age, when uncovered, sometimes Birds. II. .'(CSIRO: Melbourne.) stretched their necks and heads up as far as they would Bourke, P A. (1972).'Clustering'of woodswallows.Aust. BirdWatcher go and, oriented towards the sun, presumably seeking 4:220-222. additional warmth. Cameron, A. C. (1933). Some notes on the Black-faced Woodswallow. Emu32:157-158. vii. Shading.When the sun was directly on the nest and Chisholm, A. H. (1909). Woodswallows as honeyeaters. Emu 8': very hot, all membersof the group spent time perched 219-220. Chisholm,A. H. (1946).'Nature's Linguists.' (Brown Prior Anderson: on the edge of the nest with their wings part-open, Melbourne.) effectively providing shade.In four hours observation Clunie, F. (1973). Nest helpers at a White-breastedWoodswallow nest. when the temperature exceeded 38'C, the nestlings Notornis 20: 378-380. were shadedfor 60 per cent of the time. Clunie, F. (1976). Behaviour and nesting of Fijian White-breasted Woodswallows. Notornis 232 61--75. viii. Feeding. All group members shared in feeding the Coleman, E. (1945). Clustering of the Sordid Woodswallow. Vic. Nat. nestlingswith the proportion of feedsvarying from day 6l| 218-219. D'Ombrain, A. F. (1934). The White-browed Woodswallow. Emu 332 to day (Table 2 in Rowley 1999). 292-297. Dove, H. S. (1909). Woodswallows ix. Faecal Jacr were removed by all attendants.When the clustering. Emu 9z 30. Frith, H. J. (Ed.). (1969).'Birds in the Australian High Country.' nestlings were less than seven days old, attendants (A. H. and A. W. Reed: Melboume.) swallowed the sacs; later, attendantscarried the sacs Heathcote,W. (1931). Notes on Dusky Woodswallows.Emu 31: 168. away and dropped them at a distancefrom the nest.A Hindwood, K. A. (1956). Clusteringof woodswallows.Emu 56: 165-166. sac was removed about once every hour, which with Immelman, K. (1960). Behavioural observationson several species of Western Australian bftds. Emu 60:237-244. three nestlings voiding, gives an individual excretion Immelman, K. (1963). Drought adaptationsin Australian desert birds. 'Prcrceedings rate of once every three hours. In of the XIII International Omithological Congress.' (Ed. C. G. Sibley). Pp. 649457. (AOU: Ithaca, N.Y.) young. x. Care of the All attendantspreened the rapidly Immelmann, K. (1966). Beobachtungen an Schwalbenstren. J. fur growing nestlings,until they were more than ten days Ornith. lOTz 3749. 'Handbook old and began to preen themselves.Towards the end Johnstone,R. E. and Ston, G. M. (2004). of the Birds of of their time in the nest they exercisedtheir wings with Western Australia. Vol. 2. Passerines'. (Western Australian Museum: Perth.) vigorous waving. Sometimes they left the nest Keast, J. A. (1958). Seasonal movements and geographic variation in prematurely, before they could fly, when prompted by the Australian woodswallows Artamidae. Emu 58: 207-218. 34 l. Rowley:Behaviour of the Black-facedWoodswallow Artamus cinereus Corella30(2)

Lowe, V. T. and Lowe, T. G. (1972). Woodswallowsin mid-northern Rowley,I. (1999).Co-operative breeding by Black-facedWoodswallows Victoria.Aust. Bird Watcher4:205-210. Artamw cinercus.Corella 2326346. Marshall,A. J. (1935).On the birds of the McPhersonRanges, Mt. Rowley, I. (2000). Co-operativebreeding by Dusky Woodswallows Warningand contiguouslowlands. Emu 35,.3648. Artamuscyanopterus. Canberra Bird Notes2Sz 49-58. Mayr, E. (1945). 'Birds of the SouthPacific.' (Macrnillan: New York.) Rowley, l. (2002). Nesting of the Black-facedWoodswallow Artamus McGilp,J. N. (1935).Birds and grasshoppers. Emu35:127-129. cinercusin the Wheatbeltof WestemAusffalia. Corella 26:.6549. 'Directory McKean,J. L. (1969).The brush-tongueof Artamidae.BulI. Brit. Orn. Schodde,R. and Mason, I. J. (1999). of Ausralian Birds.' Club E9: 129-130. (CSIRO:Melboume.) Recher,H. F. and Schulz,M. (1983). Observationson the breedingof Sedgwick,E. H. (1947).Northem Territory Bird Notes.Part 2. Emu 46:. White-browedWoodswallows. Corella 7: 14. 349-378. Rowley,I. (1951).Courtship feeding of Dusky Woodswallows.Emr 51: Sharland,M. (19'12).The lure of the woodswallow.Azsr. Bid Watcher 80. 4z2l3Jl9. Rowley, I. (1976). Artamus spp. In: 'ReadersDigest CompleteBook Wigan, M. L. (1931).The Dusky Woodswallow.Emu 3l:. 93-95. of AustralianBirds.'(Ed. H. J. Frith.)Pp.56'l-571. (Readers Digest: Sydney).