Reproductive Behavior of White-Breasted Nuthatches I

REPRODUCTIVE BEHAVIOR OF WHITE-BREASTED NUTHATCHES I. DISTRACTIONDISPLAY, BILL-SWEEPING, AND NEST HOLE DEFENSE

LAWRENCE KILI-IAM

THIS report describestwo forms of behavior related to defenseof nest- holes among White-breastedNuthatches (Sitta carolinensis)which, as thesebirds nestin comparativelylarge natural cavitiesin trees,are particularly exposedto competitionfrom tree-livingsquirrels. Observations have involved20 breedingpairs of nuthatch'es,of which 7 were observedin Bethesda,Maryland between 1952 and 1960 and 13 in Tamworth and Lyme, New Hampshire,between 1963 and 1967. Th'ey are preliminary in nature; more definitive data are being soughtin attempts to breed hand- reared White-breastedNuthatches in captivity. The problemsinvolved have not been describedelsewhere in print, so far as I am aware.

SWEEPING AND NESTING t)ERIOD The habit of bill-sweepingis confinedto the nestingseason. It is similar to wipingthe bill as manybirds may do after feeding,except that a White- breastedNuthatch of eithersex usually holds some object such' as an insect in its bill, and swingsits whole body in an arc (Figure 1) as it sweeps back and forth over areasof bark by the nest entrancefor many minutes at a time. It will occasionallysweep with the bill alone and at other times poke into an object, suchas the body of an insect,then dab at the bark as if applyingwhatever adheres to its bill. An overall view of bill-sweeping and objectsused is given in Table 1. Period of nest building.---Apair of nuthatchesbegan inspectinga squirrel box in Bethesda,Maryland on 10 April 1959. The female was carryingin bits of bark and other nest material,but sheinterrupted these

TABLE 1 SUMMARIZING VIEW OF TIlE BILL-SWEEPING HABIT OF WIIlTE-BREASTED NUTIlATCttES IN TER•IS OF GENERALAND SPECIFIC T12VlES INVOLVED AND OBJECTS USED

Objects Times involved Intensity used Extent

A. General Nest-building q- q- q- q- Insect Incubation q- Fur q-q- Nestling period q- Plant material B. Spedtic and occasional Feather (single observation) Capturespecial insect q-q-q- Bill alone q-q-q- Threat of squirrel q-q-q-

477 The Auk, 85: 477-492. July, 1968 478 LAWRE•C• KILa• [Auk, Vol. 85

I I I !

Figure 1. White-breasted Nuthatch sweepingwith insect. activitiesoccasionally to sweepthe roof of the box and adjacenttree trunk for 5 or more minutes at a time. She paid no attention when the male offered her food. He arrived shortly afterward with an insect in his bill. Using it in the manner of a broom, he swept inside, then outside the entrance until almost nothing of the insect remained. The actual building of a nest may occupy female nuth'atchesin an almost exclusivefashion, but sweepingbefore or after this activity is often a prominentfeature of the prenestingperiod, as becameapparent in the courseof observationsmade in New Hampshire. Sweepingof female FB in Lyme, for example,led to the discoveryof her nest cavity at 0600'hours on 29 April 1967. She was sweepingwhen visited at 1300 and again at 1800 when her mate, MB, was also sweeping.Areas swept includedthe insideof the nestcavity, the spacearound the nest entrancewith particular attentionto the baseof a smallbranch • just belowthe hole, and the upper July, 1968] White-breasted Nuthatch Behavior 479 slopeof the leaningmaple trunk in which the cavity was located. A fork formedwith a large branch 6 metersbelow the hole also receivedspecial attention. FB continuedto sweepfor the next few weeks.and did so at mosttimes of day exceptwhen it rained. I sometimesfound MB sweeping at the sametime, but he wasnearly always the first to stopand fly off. An impressionone had from watching the nuthatcheswas that they were somehowpreparing or adaptingthe maplestub by sweepingthe bark again and againas one might paint a houseprior to movingin. Incubationperiod.--More extensiveobservations on sweepingwere on a pair of nuthatchesthat nested in a squirrel box in Bethesdabetween 28 April and 11 June 1957. On 2 May the male sweptso vigorouslyat 1300 hoursthat I couldhear the swish,swish 60 feet away. The swishesheard in the next 7 minuteswere too rapid to count, but they were obviously in the hundreds.The femalealso swept for part of this period. I believe shelaid her first eggon this day for there were 5 eggsin the nest when I looked in 5 days later. On 6 May both nuthatchesswept continuouslyfrom 1908 to 1918 hours. They swept particularly on the roof and cornersof the nest box as well as on knob-like growthson the tree trunk, paying little attention to each other and making no vocalizations.At 1925 they both returned to sweepfor another 5 minutes even more vigorouslythan before. The female then entered the box for the night and her mate, after sweeping with almost frantic vigor for an additional 2 minutes, flew off to roost elsewhere. Nestling period.--The eggsof this pair hatchedon 23 May. On the eveningof 31 May the femaleswept the roof and tree trunk in seemingly frantic fashion for 13 minutes,then entered the nest box for the night at 2021. Both nuthatcheswere sweepingon the eveningof 8 June but inde- pendentlyof each other. At one time, for example,he was 8 feet above the nestbox while shewas sweeping5 feet belowit. I noted bill-sweeping in this pair for the last time on 10 June,which was the day the youngleft their nest. The intensivesweeping of this pair prior to roostinghas been noted in other pairs during incubation and nestingwhen little other bill- sweepingwas manifest.

SITUATIONS PRECIPITATING BILL-SWEEPING The followingobservations made in Lyme between1965 and 1967 illus- trate more specificsituations leading to bill-sweeping: Bill-sweepingas reactionto squirrels.--Bill-sweepingcan have a special relation to the proximity of squirrels. While the majority of nest trees observedhad no squirrelsliving particularly closeto them, a few of the nest trees were in favorablesquirrel habitats. It was at these that most of the sweepingepisodes, such as the following,took place: 480 LAWRENCEK•Lxxx•r [Auk, Vol. 85

PATHWAY OF RED SQUIRREL oooo MARKEDBY SCENT AREAS SWEPT REPEATEDLY xxx x BY NUTHATCHES NESTHOLEOF•"• t NUTHATCHES"'••

Figure 2. Diagramatic representation of interrelations of White-breasted Nuthatch and red squirrel. Squirrel has pathways up adjacent trunks and branches, used routinely and presumablymarked by scent. No such odor trails became established on the nest maple of pair W becauseof the repeated sweeping of protuberancesand branch junctions by the nuthatches (for conceptsinvolved see Text).

When a red squirrel (Tamiasciurushudsonicus) ran up an adjacent maple at 0800 on 5 May 1966, female nuthatch FW emergedfrom her nest hole immediately. She first made a distractiondisplay (see Figure 3), then beganto sweepas the squirrelmoved away. FW had nothingin her bill initially, but after her mate arrived to feed her, she swepthard for 15 minuteson specialparts of the sugarmaple usingfragments of an insect. Areas swept included protruding knobs and ridges, junctions of largerbranches and th'emain trunk, as well as the vicinity of the nest hole as shown in Figure 2. July, 1968] White-breasted Nuthatch Behavior 481

FmURE 3. Diagram of events observed on nest maple of nuthatch pair W on 20 June 1966. Red squirrel ascendingtrunk paused,then made detour of nuthatch's nest, seeminglydeterred by distraction display of female W at nest rim and possibly, as an additive effect, by confusing odors laid down by sweeping activities of the nuthatches.Detail at C showshow raising of auricular feathersgives appearanceof large, false eyes,to nuthatch as viewed from behind.

I observedno further sweepingat the sugarmaple until 14 May when a gray squirrelsuddenly began a loud scolding,"chuck, chuck,c-w-aan" 30 metersfrom the nest. FW emergedfrom her nest hole and swept for 4 minuteswhile the squirrelcontinued to scold. 482 LAWRENCEKrLaa•X [Auk, Vol. 85

Female W was incubatingwithin the nest hole a week later when a juvenilered squirrelran up the adjacentmaple. This time FW remained inside,but her matereacted almost immediately with' 2 minutesof sweeping, centeredlargely around the nestentrance. This was the first and only time I saw MW sweeping. I notedno furthersweeping nor any closeapproaches of squirrelsto nest W until 20 June, when a red squirrelwhose territory was closeby ran up the trunk of the nuthatches'nest tree. FW disappearedwithin her hole immediately.It wasas if her immediateconcern was to guardthe entrance and her nestlingsfrom within. By the time FW emerged5 minuteslater the red squirrelhad passed,but she now caught sight of a chipmunk (Tamia striatus) directly below on the ground. FW remainedmotionless while eyeingit intently. Then, when the chipmunkcame no closer,she took a piece of unidentifiedmaterial from storageand swept for several minutes.This reactionto chipmunkswas not unusual,for I sawit at two other nests. Captureof specialinsects and sweeping.--Asidefrom the approach'of a squirrel,the only other situationI notedthat servedto precipitateimme- diate sweepingwas capture of somespecial insect. Examplesat two nests may be illustrated by events at Nest F: FemaleFF flew to her nestat 1219on 11 June 1967with a singleinsect, larger and hencemore recognizablethan what she had been in the habit of carrying to feed her young. The insectwas a blue-blackbeetle with a prominent abdomen,and was about 2 cm in length. FF exhibited an intensedrive to sweepthe walls of the nest cavity in an almost frantic mannerfor 9 minutes. Sheemerged a few times to sweepbriefly about the entrance. After her last bit of sweepingon the outside she stored the remnantsof the beetlein a crevice. She now appearedas if exhaustedby her efforts and restedmotionless for 6 minutesjust below the nest hole. I saw no further sweepingat this nest until 1045 on 16 June,when FF flew to the nest tree carrying the same type of blue-blackbeetle as the oneused 11 June. My immediatethought was, she is goingto sweep,which provedcorrect. FF beganto sweepimmediately on alighting,concentrating her activitieson the insideof the nest cavity as before. It was only when she came outsidefor brief periodsthat I could see additional aspectsof her sweepingbehavior. Once she stoppedsweeping to depositthe insect in a crevice,then poked her bill into it repeatedlyprior to sweepingfor a few momentswithout it. It appearedas thoughshe were applyingsome- thing, as one might dip a brush into a can of paint. FF storedthe rem- nants after 5 minutes. I had noticed that she ate the viscera when these had protrudedduring her activities. When throughwith her sweeping shewent back over her course,devouring a few smallbits of material that had clung to theebark here and there. July, 1968] White-breasted Nuthatch Behavior 483

A numberof pointsregarding these episodes are worthyof comment.First is the fact that FF gave no signs,physical or vocal, that she was excited by the threat of any immediate danger. To an observertherefore, the stimulusto sweepappeared to come from the nature of the insect itself, whichpossibly had someirritating property. A secondpoint of interestis the specialattention given to the insideof the nest cavity. As this sweeping was doneat a time when the youngwere halfway throughthe nestling periodit may have servedsome imprinting function, either by letting the nestlingssee the parent perform in an intensive fashion within the nest itself, or chemicallyby giving them a chanceto pick up someof the taste of the insect as they poked about the nest cavity. One might note here that the mother nuthatch had picked off and eaten bits of insect adhering to bark, a fact supportingthis idea that the nestlingsmight do the same. OBJECTSUsEn IN SWE•mNG Insects.--Usually the nuthatchessweep with a mere fragment of material whichmight not be recognizableeven at closerange and is much'less so to an observeron the ground and many meters distant. Insects are more readily recognizedby their chitenousouterparts than most other object. While nuthatches usually take them from storage prior to sweeping,in severalexamples above insects,possibly of a specialnature, were used immediatelyafter capture. Plant materials.--As plant materials generally lack the hard parts characteristicof insects,it does not take much manipulating to render them amorphousand unrecognizable.Two types of examplessuggested that nuth'atchesuse them in sweeping.First of thesewere direct observations, suchas one made on female FB on 12 May 1967 as sheswept about her nesthole for 8 minuteswith a wad of bright green,exceedingly moist- looking fibrousmaterial a full centimeteror more in diameter. She lodged the material several times in crevices,then poked her bill into its mass before sweepingfor a time with her bill alone. A secondline of evidencefor use of plant materialswas purely circum- stantial. This was the collectionand storageof buds from sugarmaples, a phenomenonobserved in threenesting pairs. Male C, for example,attacked six suchbuds on 8 and 9 May 1965, usuallyclinging to them upsidedown to h'ammerat the attachment of bud to twig from below. I never saw a nuthatch eat such buds; in each case the buds were simply stored in some other part of th'e nest tree and left there. Bussman (1946) has describeda similar type of behavior in the European Nuthatch (Sitta europaea). In his case,however, gum from the bud was incorporatedin the nest entrance. Fur.--Fur is, after insects, the most frequently recognizableof the materials the nuthatchesuse. The various ways they may store and/or 484 LAWRENCEKILItA• [Auk, Vol. 85 sweepwith it was illustratedby the activitiesof pair B in 1967. At 1730 hours on 13 May FB took a wad of gray fur from storageclose to the nest entranceand sweptwith it for 5 minutes,bits of the fur comingloose and clingingto the bark here and there as she did so. The fur might well have comefrom a gray squirrel. On 18 May her mate, MB, took from a crevicea smallwad of red fur suchas might have comefrom a red squirrel, then re-stored it closer to the entrance hole without sweeping. I then noticed that FB was sweepingwith a feather, the only time I ever saw a nuthatch do so. This feather was curved and about 2.5 cm in length. I suspectedthat it might have come from a Ruffed Grouse (Bonasa um- bellus) . Sweepingwith bill alone.--Nuthatchesappear to store a considerable supply of objectsboth inside and outsideof nest holes. They can thus start sweepingwithout lossof time wheneverseized by the impulseto do so. When incited to sweepabruptly by somespecial stimulus, they may start sweepingwith nothing but their bills alone.

BIOLOGICAL SIGNIFICANCE OF BILL-SWEEPING At least severalanswers are conceivableto the questionof whether the habit of bill-sweepinghas survival value for White-breastedNuthatches. One is that the habit servesto strengthenthe pair bond as well as attach'- ment to the nest site. Yet it appearsdoubtful for a numberof reasonsthat the habit couldbe for suchpurposes alone. First is the fact that courtship activities are usually at a low level during actual nesting and would not be expectedto rise to an intense level when an intruder threatened the nest. A further reasonis that female nuthatchessweeping alone seeminglypay no attention to their mates. If both membersof a pair sweepat the same time, they may be many feet apart and even out of sight of one another in seemingindifference, all of which is unlike behaviorprimarily associated with a pair bond. An alternateexplanation of bill-sweepingis that the habit serveschiefly as a first line of defenseof the nesthole againstsquirrels. This hypothesis has been suggestedby field observations. Findings presentedbelow are thus regardedas beingonly suggestivein nature and are being followedup by a long term study involving attempts to breed hand-raised White- breastedNuthatches in captivity, then to test variousinteractions between them and hand-raisedgray squirrels. These projectsare already underway. Meanwhile the biologicalsignificance of the bill-sweepinghabit as it appearsat present is outlined by means of the following concepts: first that red as well as gray squirrelsuse the same type of nest cavity in old sugar maplesand otker trees as White-breastedNuthatches and are thus nest hole competitors; secondthat these squirrels leave scent trails on branchesand trunks of treeswhich they customarilyuse in travelingabout; July, 1968] White-breasted Nuthatch Behavior 485 and third that the bill-sweepingof nuthatcheswith various objects is primarily a meansof defenseof their nest holesagainst mammalian competitors. Types of holesfor which squirrelsand nuthatchescompete.--The range of cavitiesnuthatches use in old sugarmaples have roughlytwo sizesof entranceholes; thoseof 3 to 4 cm in diameterattracting red squirrelsand thoseof 7 to 10 cm, the larger gray squirrels. In 1964 I found two nuthatch nestswith the smallerentrances. On returninga year later to find out if the birdsmight be reusingthe samecavities for nesting,I foundred squirrelsoccupying both of them. On other occasionsit has taken many visits to reveal whethera squirrelwas actually occupyinga hole. While makingobservations on nuthatchpair D in two successiveseasons with repeatedvisits in April, May, and June,I notedwhat appearedto be an ideal, but apparentlyvacant cavity in an old sugarmaple near where I usuallystood. On only a singleday in eachyear did I discoverthat it was occupiedby nestinggray squirrels.Thus in 1966 the mother squirrel rested some moments before entering and in 1967, three half-grown squirrelssuddenly emerged from the hole, then reenteredit a few moments later. Scentstrails of squirrels. In a discussionof intraspecificsignals used by mammalsin general,Wynne-Edwards (1962) pointsout th'at there are "no instrumentsby which odorscan be detected,amplified, analyzed or defined." I have not found much specificinformation on scenttrails in published accountsof squirrels. The following observationssuggested that squirrelsmay lay downsuch trails on old sugarmaples, among other trees, in the breedingseason of nuthatches: A femalered squirrelwith a centerof activitiesin a brushpile 8 meters from the nestmaple of nuthatchpair W in the springof 1966 (see Figure 2) had youngmoving about neighboringtrees by 7 June. On 18 June the squirrelrubbed her chin a numberof times on a main log of the brush pile. She then made a circular tour of adjacent trees and branchesgiving what I have interpretedas a territorial song (Kilham, 1954) about every 3 minutes. The squirrel stayed away from the nuthatch maple on this as on previousoccasions. On 20 June, however,the mother red squirrel startedup the mapleand was halfway to the nesthole beforethe female nuthatch looked out, then spreadwings and tail in distractiondisplay (Figure 3). The squirrel movedon up the maple, without comingclose. She then returned, paying attention to various ridges of bark and pro- tuberanceson the way down, all of which were placesI had noticedthe nuthatchessweeping on previousoccasions (Figure 2). At 0645 on 7 June 1967 I watched an adult male gray squirrel as it walked slowly acrossbranches to an old sugarmaple where it sniffed the bark, particularlywhere large branches formed the main trunk (Figure 4). 486 LAWRENCEK•LHA• [Auk, Vol. 85

Cornelia

Figure 4. Types of gray squirrel activitiesinvolving parts of old sugar maples such as are swept with particular intensity by nesting White-breastedNuthatches. Male squirrel sniffs at place where large branch joins trunk on tree where female gray squirrel,resting on protuberance,had young.

It even retraced its steps to do so twice. While the squirrel was thus engaged,three half-growngray squirrelsemerged from a cavity in the hole below,then quickly reentered.The tree was thus occupiedby a family of squirrels.The mothersquirrel might, expectedly, be cominginto a breeding conditionagain as her youngreached the juvenilestage. Both Flyger (1955) and Taylor (1966) observedthat male gray squirrelsmay not only sniff but also eat the bark wh'ere females in estrus have rested. While I have encounteredno extensiveaccounts of squirrelslaying scenttrails, my own observationshave receivedsupport in personalcom- municationfrom Peter Weigl, who is studying squirrelsof various.species at Duke University, in the followingpersonal communication: "The tree and flying squirrelswhich I have hand-rearedhabitually marked their trails over new surfaceswith minute drops of urine. The positions of animalsat suchtimes with their stomachclose to th'eground would facili- tate the spreadingof glandularand other substancesfrom the urogenital region." Weigl has also noted that rubbing of the chin on a log is "a common form of marking behavior." TransJeroJ youngsquirrels and night movements.--Squirrelshave some July, 1968] White-breasted Nuthatch Behavior 487 ability to moveabout at night. As red and gray squirrelsboth have pure- cone retinae (Tansley et al., 1961) they are unable to see well at such times,and onecan presume that they possiblyfollow scent trails laid down during the day. At 2300 on 6 September1952 I locatedan adult female gray squirrelby a peculiarwhine as she movedback and forth betweena hickoryand a tulip poplarwhich had a leafy nestat the top in our yard in Bethesda.Hearing the samewhine at 0630 the followingmorning, I was able to watch the motheras she transferredfour one-third-grownyoung, one at a time, from the leaf nest to a squirrel box up on a pole. Gray squirrelsmay have a closefamiliarity with all potential holeswithin their homerange. Intrusion by sucha mothersquirrel in the night might repre- senta specialhazard to a pair of nestingnuthatches and may explainwhy the birds sweepwith special intensity just before dark. I have had no observationsthat flying squirrels(Glaucomys) are also a hazard. They seemto prefer to occupywoodpecker holes in straighter trees rather than the types of cavities White-breasted Nuthatches use, which may explainwhy they were not encounteredin presentstudies. Sweepingas chemical deJenseagainst squirrels.--Chemical defense of nest holesis not unknownamong birds. Nestling Hoopoes(Upupa epops epops) produce a foul-smellingsecretion from special glands (Sutter, 1946). It is conceivablethat White-breasted Nuthatches accomplisha similartype of defenseby usingthe specialand analogousglands of insects as broomsin bill-sweeping.The crushedbodies of variousarthropods can be very potentin this regard. Eisnerand Meinwald (1966), amongothers, have recently describedhow many speciesof insectsproduce defensive secretionswhich are not only odorous,irritating, or otherwiserepellent in nature, but are also persistentin effects,and are producedin relatively large amounts and in a highly concentratedform within integumental glands. It is often difficult for man to appreciatethe sensitivityof other mammals to odors in even trace amounts. One may ask, if this is true in regard to. insects,why shouldnuthatches use plant materialson occasionas well? The related Red-breastedNut- hatch (S. canadensis)applies plant materialsin the form of resinoussap from coniferoustrees to the bark surroundingnest holes. The usefulness of theseresins in defenseis probably in their stickiness.Plant materials have a wide variety of other properties and the fact that the various alkaloids, essentialoils, and other so-called "secondary substances"are primarily defensivein servingto ward off attacks of phytophagusinsects and herbivoreshas only been appreciatedin recent years (see Fraenkel, 1959). Eisner (1964) has pointed out, furthermore,that defensivesub- stancesof plants and arthropodsmay even be identical in chemicalstruc- ture in some instances. Thus White-breastedNuthatches could, conceiw ably, sweepwith bits of plants or of insectsinterchangeably. 488 LAWRENCEKn•uA•r [Auk, Vol. 85

An important questionstill remainingis why do thesebirds sweep,and not infrequently with no more than a few whisps of weathered fur, a feather,or evenwith the bill alone? One presumesthat this latter habit also has survivalvalue. A possibleexplanation is that White-breasted Nuthatches use saliva as well as juices of plants and insectsin sweeping. The conceptof saliva with specialproperties is not new among birds. Welty (1964: 89) givesa goodsummary of suchspecies as the European GreenWoodpecker (Picus viridis) with salivaryglands 7 cm in lengththat supply an insect-holdingfilm for its protrusible tongue and of the EuropeanHouse Martin (Delichonurbica) which builds a nest of mud mixedwith saliva. The salivaryglands of thislatter speciesare especially enlargedduring th'ebreeding season. One wonderswhether similar condi- tions may not obtain among somespecies of nuthatches. The European Nuthatch (S. europaea),for example,builds a nestentrance of mudwhile the Rock Nuthatch (S. neumayer)builds an all mud nest much like that associatedwith somespecies of swallowsbut embedsthe bodiesof insects within it (seeL/Shrl, 1967). L/Shrl(1958) raisesthe issueof whetherS. europaeauses saliva in nest construction,but leavesit unanswered.One can do the same with as much reason for S. carolinensis.

DISTRACTION DISPLAYS A White-breasted Nuthatch of either sex can transform itself in what to a humanobserver is a moth-likeobject when,holding itself in a rigid pose with wings and tail outspreadand bill pointed upward, it sways slowlyfrom s/deto sidein the faceof an intruder approachingits nesthole (Figure 5). Each sway may take i0 to 20 seconds,and such displays are much alike when producedat full intensity. The ability of nuthatches to cling to tree trunks in odd positionsmakes them look somewhatdiffer- ent at different times. Examplesillustrating the range of this performance and the contextsin which it may be performedare as follows: Bethesda1957.--Pair A. Male A was sweepingnear his nest box in our yard at 0730 on 7 May, when he suddenlywent into a distractiondisplay in which he pointed straight out from th'e tree trunk as he faced a gray squirrel approachingfrom the ground. The squirrel came to within a meter of the displayingnuthatch, paused, then turned and left. MA then continuedsweeping. The nest box contained 5 eggsat the time. I observedsimilar episodeson 17 May and 7 and i i June. In eachcase the nuthatchesexhibited the same interplay of sweepingand distraction displays. They had to contendwith a relatively concentratedgray squirrel populationin the suburbanarea in which they lived. Experiencewith aviary pair, Lyme 1967.---An opportunity for close observationof nest hole defenseamong White-breasted Nuthatches was affordedby a hand-raisedpair that went throughmuch' of their breeding July, 1968] White-breasted Nuthatch Behavior 489

Figure 5. White-breasted Nuthatch in distraction display. behaviorin captivity. A chipmunk began entering their cage in early April. On 7 April the malenuthatch was resting by his nestbox watching the groundintently, occasionallyfanning his tail and partly raising his wingsas he did so. The chipmunkhad enteredtile cagefrom the ground. It suddenlyran up the side of the cage and over onto the roof of the nest box, wherethe male nuthatchmet it in full distractiondisplay. The chipmunk,possibly from daily closeexposure to the nuthatches,was not deterred. As it came even closer the nuthatch swungbackward, still in display, until its body, with belly upward, projectedoutward from the edgeof the nestbox at an angleof 90 degrees.After a momentarypause it droppedaway. The chipmunk now came over the edge of the box and looked into the entrance. Here it faced the female nuthatch. She drove at the intruder head on and, from the rapid departureof the chipmunk, her sharpbill musthave delivered an effectiveblow. This episodeclarified onepoint--that a nuthatchin distractiondisplay may hold its rigid pose, without evidenceof either fear or of making of any motion of threat to attack, right up to the momentof beingalmost touched by an intruder.

DISCUSSION Distraction displays.--Armstrong(1964) has describedinjury-feigning and mammalsimulation as the only types of distractiondisplays among birds. Both, however,apply to ground-nestingspecies. and thus appear 490 LAWRENCEKILI-IA3•[ [Auk, Vol. 85 to differ in origin and nature from what I have termed the "distraction" displaysof White-breastedNuthatches, which are deliveredin one place rather than while runningabout. The only parallel I have encounteredis that Ferguson-Lees(1964) describesfor nesting bitterns (Botaurinae spp.), "Most striking is the concealmentposture shared only with the tiger herons.In this the bird becomescompletely rigid with bill and neck pointed stiffly skyward . . . the bird keeps its undersidetowards the sourceof danger,watching with unwinkingyellow eyes swivelleddown- wards, and revolvesslowly . . . if th'e observeror other threat moves round.... In a light breeze,at leastsome species will swaywith the reeds. Ixobrychusmay evenlet itself be pickedup." This sudden"mental freez- ing" of a bittern on its nest suggestsan additional parallel with White- breastedNuthatches. Thus when I openeda nest box of this. speciesin June 1957, I found that the female sat so closeover her nestlingsthat I had to push her aside witk my finger to see them. Grinnell and $torer (1924) describethe same phenomenonas follows: "The female was on the nest and as she refused to leave even during the hubbub incident to enlargingthe entrance,the observerhad to lift her from the nest in order to examinethe eggs. She seemedto be in a sort of lethargy and did not struggleuntil actually taken in hand." It is th'isaspect of lethargywith its hypnoticquality, which may be a basicpart of phenomenaunderlying the distractiondisplay. Theseappear to have muchin commonwith those Hartman (1952) describesin his chapteron the death-feigningof opossums (Didelphis virginiana)and otheranimals. In referenceto the stateof tonic immobility induced,Hartman speaksof "reactionsto a suddenstimulus too rapid to be controlledby consciousemotions." Bill-sweeping.•A generalizationapplying to most hole-nestingbirds is that suitablenest holesare in short supply and competitionfor them is usually intense (yon Haartman, 1956). The nature of the competition varieswith differentspecies. While avian competitorsare of chief impor- tance for most hole-nestingbirds, tree squirrelsappear to be the chief competitorsfor natural cavities with nesting White-breastedNuthatches in New Hampshirewoodlands. The extent of this competitionis not easy to assess. Squirrels are wary about entering their nest holes, a mother squirrelcan spendlong hourswithin nursingher young,and squirrels,like mammalsin general (seeWynne-Edwards, 1962), live to a large extent in a world of olfactory signalsbeyond direct comprehensionof a human observer. Bill-sweepingmay have developedfrom the simplehabit of wiping the bill a few times on bark to clean it of fluids or bits of tissue remaining after poundingsome insect. Organicsubstances would then remain in trace amountson the bark. Retentionof the practicemight then follow if such July, 1968] White-breasted Nuthatch Behavior 491 bits happened,by chance,to have survivalvalue in deterringtree squirrels by meansof disruptingolfactory signals. The habit bearssome analogies to that of antingamong other passerines, in which insectfluids are applied to feathers (for descriptionsee Simmons, 1964). In bill-sweepingsimilar defensive fluids are applied to bark of nest trees. The habits may be defensivein either case. If anting is a parallel habit, its effect couldbe to hide or disguisebody odor by anointing contour feather with repugnatory or repellent organic substancessuffi- ciently long-lastingin nature to protect the bird from mammalianpredators during the especiallyvulnerable hours of roostingat night. Even a slightly confusingscent might be sufficient to deter or to deflect intruders. Bill-sweepinghabits of White-breastedNuthatches, like their distraction displays, are possibly unique among hole-nestingbirds. References to either type of behavior appear to be few. Hans LShrl has observedthe bill-sweepingof S. carolinensis(pers. comm.) and F. H. Allen (quoted in Bent, 1948: 3) while observinga nesting pair, noted that "the wiping was too regularand long-continuedto be merely for the purposeof cleaning the bill."

SUMMARY This report describestwo behavior patterns used by White-breasted Nuthatchesin defenseof their nest-holes,one a "distraction"display and the other bill-sweeping. The unusual nature of these performancesappears related to the fact that White-breastedNuthatches nest in natural cavitiesin trees,a situation exposingthem to competitionfrom tree squirrels. In bill-sweepinga nuthatch usually holds some object such as crushed insect in its bill. The sweepingmay go on inside and outside of nest cavities and inten- sively for many nfinutesat a time. Although limited to the breedingseason, nothing indicates that it is primarily related to either courtship or the pair bond. Periods when bill-sweepingcan be observedwith greatest regularity are at the time of nest-building,especially in a new cavity, and during the last half hour before roostingduring periodsof incubation and rearing of nestlings. Two sets of circumstancesmay precipitate bill-sweepingat almost any time: (1) the threat posedby a possibleintrusion of the nest tree by a red squirrel, a gray squirrel, or a chipmunk,and (2) capture of certain insects that appear to have special properties when crushed. The hypothesisis proposedthat White-breastedNuthatches make use of the chenfical defense secretions of insects to divert or to distract tree squirrelsfrom approachingor entering nest cavities. 492 LAWRENCEKILHAM [Auk, Vol. 85

When a squirreldoes come close to a nest,a parent nuthatchof either sex may spreadits wings fully and hold a fixed pose, swayingslowly back and forth, in what is termeda "distractiondisplay." As nothing in this behavior suggestsimpulses to attack or escape,it is regardedhere as being more related to "death feigning"phenomena observed in various animal species. The presentreport, which is preliminaryin natureand is basedon field observation,is beingused as a basisfor an experimentalapproach with White-breastedNuthatches hand-raisedin captivity.

LITERATURE CITED ARMSTRONC,E.A. 1964. Distraction display. P. 205 •n A new dictionary of birds (A. L. Thomson, ed.). New York, McGraw-Hill. B•N% A. C. 1948. Life histories of North American nuthatches, wrens, thrashers and their allies. U.S. Natl. Mus., Bull. 195. BUSSMANN,J. 1946. Beitrag zur Kenntnis der Brutbiologie des K]eibers, $•tta europaeacaes{a. Orn. Beob., 43: 1-5. EISNER,T. 1964. Catnip: Its raison d'•tre. Sdence, 146.. 1318-1320. EIS•rER,T., A•rD J. MEINWAnD. 1966. Defensive secretionsof arthropods. Science, 153: 1341-1350. FERCUS0•r-LEES,I. J. 1964. Bittern. Pp. 100-101 in A new dictionary of birds (A. L. Thomson, ed.). New York, McGraw-Hill. FLY•ER, V.A. 1955. Implications of social behavior in grey squirrel management. Trans. N. Amer. Wildl. Conf., 20: 281-289. FRAE•rKEn,G. S. 1959. The raison d'•tre of secondaryplant substances.Science, 129: 1466-1470. GRINmSnn,J., ANDT. I. STORER.1924. Animal life in the Yosemite. Contrib. Mus. Vert. Zool., Univ. California. HAAR•MAN, L. 1957. Adaptation in hole-nesting birds. Evolution, 2: 339-347. HAR•XVrA•r,C. G. 1952. Possums. Austin, Univ. Texas Press. Kn;•AM, L. 1954. Territorial behavior of red squirrel. J. Mamm., 35: 252-253. L6•RL, H. 1958. Das Verhalten des Kleibers. (Sitta europaea caesia Wolf.) Z. Tierpsychol., 15: 191-252. L6•R•, H. 1967. Die Kleiber Europas. Wittenberg Lutherstadt, A. Ziemsen Verlag. SIMMONS,K. E.L. 1964. Feather maintena.nce.Pp. 278-286 in A new dictionary of birds (A. L. Thomson, ed.). New York, McGraw-Hill. SU•ER, E. 1946. Das Abwehrverhalten Nestjunger Wiedehopfe. Orn. Beob., 43: 72-81. TANSLEY, K., R. M. COPENI{AVER,AND R. D. GUNKEL. 1961. Some aspects of the electroretinographicresponse of the American red squirrel, Tamiasciurus hudsoni- cusloquax. J. Cell. Comp. Physiol., 57: 11-19. TAYLOR,J.C. 1966. Home range and agonisticbehavior in the grey squirrel. Symp. Zool. Soc., London, no. 18: 229-235. WELTY, J.C. 1963. The life of birds. New York, Alfred A. Knopf. WYmVE-EDWAV,DS, V. C. 1962. Animal dispersion in relation to social behavior. New York, Hafner Publishing Co. Department of Microbiology, Dartmouth Medical School, Hanover, New Hampshire 03755.