A Review of Some Aspects of Avian Field Ethology 1

Home , Agonistic behaviour, Begging in animals, Displacement activity

A REVIEW OF SOME ASPECTS OF AVIAN FIELD ETHOLOGY 1

ROBERT W. FICKEN AND MILLICENT S. FICKEN

T•E last 30 yearshave seen the developmentof a new approachto the study of behavior, which has increasinglyinterested ornithologists as they have attemptedto understandavian biologyfully. This new ap- proachis now known as ethology. Both amateur and professionalorni- thologistshave contributedto its progressfrom the beginning. A basictenet of ethologyis that all behaviorof animalscan eventually be understood.The ethologistattempts to explainbehavior functionally (what doesit do for the animal?), causally(what are the internal and externalfactors responsible for each given behavior?),and evolutionarily (what is the probablephylogeny of the behavior,how doesit contribute to th'esurvival of the species,and what are the selectivepressures acting uponit?) (seeTinbergen, 1951, 1959; Hinde, 1959b). Somebasic proceduresare essentialin an ethologicalstudy (Hinde, 1959b: 564). The first step is to describeand classifyall the behavior of an animal, at the same time, if possible,dividing the behavior into logicalunits which can be dealt with furtherin other disciplines--particularly physiologyand ecology(see Russellet al., 1954). Althoughstudies often beginqualitatively, they eventuallyshould be quantified. Generali- zations,which are to be valid for many species,must be basedon work using closelyrelated speciesfirst, and more distantly related ones later. The animal is studied as an integrated whole. Some workers, notably Konrad Lorenz, keep and breed animalsin captivity under closescrutiny. Most of this paper dealswith communicationof birds. We have stressed particularly the analysisof displays,their evolution,and their relation to taxonomy--topicsof great interest to systematicornithologists as well as to ethologists.In addition, we discusscertain maintenanceactivities.

LITERATURE OF ET•OLOC¾

Much of the early ethologicalliterature appearedin Europeanpublications, but it is now also appearingin this country frequently. The principal journalsare: Animal Behavioar (formerly British Journal of Animal Behaviour), Symposiumof the ZoologicalSociety of London,Journal of Comparativeand PhysiologicalPsychology, Zeitschrlftfiir Tierpsychologie,and Behaviour(with monographsappearing as supple- ments in the last two). Papers on avian ethology also appear frequently in the major ornithologicaljournals. Standardreference books on ethologyare The study of instinct (Tinbergen,1951), Learning and instinct in animals (Thorpe, 1963), and Social behaviorand organizationamong vertebrates(Etkin, 1964). Recent monographson individual speciesinclude those by Marler (1956a) on the Chaffinch (Fringilla coelebs);Hinde (1952), Great Tit (Parus major); Tinbergen

• Written at the request of the A.O.U. Committee on Research.

637 The Auk, 83: 637-661. October, 1966 638 FI(2KENAND FICKEN, Avian Field EthoIogy [ Vol.Auk 83

(1953), Herring Gull (Larus argentatus); Moynihan (1955b), Black-headed Gull (L. ridibundus); Curio (1959a), Pied Flycatcher (Muscicapa hypoleuca); Simmons (1955), Great Crested Grebe (Podiceps cristatus); and Immelmann (1959), Zebra Finch ( Taeniopygia castanotis). Some outstanding examples of comparative ethological studies are those by Lorenz (1952) on dabbling ducks, Meyerriecks (1960) on herons, Tinbergen (1959) and Moynihan (1955b, 1956, 1958a, b, 1959, 1962) on gulls, Andrew (1957a, b) on em- berizine finches and (1961a) other passerines,Itinde (1955-1956) on carduelines, Dilger (1960) on parrots (Agapornis), McKinney (1961) on eiders (Somateria), Curio (1959b) on flycatchers (Muscicapa), and L•ihrl (1960-1961) on nuthatches ( Sitta) .

ETHOLOGICAL CONCEPTS

Sincethere are severalrecent reviews of ethology (Emlen, 1955; Thorpe, 1963; Hinde, 1959b, 1961; Eibl-Eibesfeldtand Kramer, 1958), we shall discussonly a few critical conceptsand terms of particular interest to field students.Many of the theoreticalconcepts are still controversialand will undoubtedlybe modified substantiallyin the future. The glosseries of Verplanck (1957) and ThOrpe (1951) contain many definitionsof ethologicalterms. Someactions of animalsare variable, othersmore stereotyped,and the two kinds often occur togetherin a behavioralsequence. "The variable introductoryphase of an instinctivebehaviour pattern or sequence"is calledappetitive behavior and the act terminatinga behaviorpattern or sequenceis the consummatoryact (Thorpe, 1951). Tinbergen (1951: 106) givesan exampleof a sequencein which a PeregrineFalcon (Falco peregrinus)searching for food beginsby randomlyroaming over its hunt- ing territory. This appetitive behaviorcontinues until a stimulussituation is encountered(e.g., a flock of teal executing fligh't maneuvers). Such situations may cause the falcon to abandon its random searching,but what follows is still appetitive behavior of a more specializedkind--the flock of teal may releasesham attacks by the falcon, resulting in the isolation of a member of the flock. The final stoop, capture, kill, plucking, and eating form a relatively simpleand stereotyped"chain of consumma- tory acts." Stereotypedmovements, many of which serve as consummatoryacts, have been termedfixed patternsor fixed action patterns (Thorpe, 1956). Investigationof such patterns led to the modern conceptsof instinctual behavior. Instinct as exemplifiedby these actions is a stereotypedpat- tern of behaviorwhich is inherited and specific (Hinde, 1959b). In addition, instinctualbehavior may be characterizedby the accumu- lation of reaction specificenergy (now often referred to as specificaction potential or S.A.P.) which is defined by Thorpe (1951) as "the state of the animal responsiblefor its readinessto perform the behaviourpatterns 190•t•] F•½•:E•^•) F•½•:•,Arian Field JEthology 639 of oneinstinct in preferenceto all otherbehaviour patterns." He points out that this specificreadiness "diminishes or disappearswhen the consummatoryact of the chargedinstinct h'as been performed."However, if the actionis not released,the S.A.P. accumulates.The thresholdfor releaseof the particular act is then loweredand the action may occur without any externalstimulus. Such action is often called a vacuum activity,overflow activity, or "Leerlaufreaktion"(Lorenz, 1950; Thorpe, 1956). Lorenzobserved this in a captiveStarling (Sturnus vulgaris) whichrepeatedly performed all the behaviorpatterns of insecthunting, catching,killing, and swallowing,without any discerniblestimulus (Tin- bergen,1951: 61). Lorenzpostulated a mechanismwhich would prevent the occurrenceof an act before the animal encounteredthe specific environmental stimuli. This "block" has been termed the innate releasing mechanism(I.R.M. or, more recently,R.M.) (Lorenz, 1950; Thorpe, 1956). The I.R.M. is "selectivelyresponsive to a specialstimulus situation"(Tinbergen and Perdeck, 1951: 2). Thisconcept was introduced to accountfor the factsthat animalsoften respondto a particularstimulus situationwith specificresponses even though they have never encountered the situationbefore, and that, in suchcases, the animaloften responds to only a very limited part of the total stimulussituation. Releasers are structures or actions that emanate sign stimuli which are perceivedby an animal. Releasersoften reach a high degreeof specialization,and their evolutionmay parallel that of the associated specificresponse (Lorenz, 1935). Tinbergen (1953: 197), using models, showedthat in Herring Gulls the red spot on the parent's bill is the primary releaser of the pecking responseof the chick. Neither the back- groundcolor of the bill or head color are releasers;in fact, the presence of the head is not even necessary. Although most of the examples of releasersand their sign stimuli are derived from experimentalstudies of visual stimuli, the conceptsare also applicableto olfactory, auditory, and tactile ones (see Lorenz, 1950). Supernormalsti•nuli are those that are even more effective than the natural stimuli; for example, Tinbergen (1951: 43) foundthat oystercatchersprefer a clutchof five eggsto a natural clutch of three and an abnormally large egg to one of their own. Drive has been defined by Thorpe (1951) as "the complexof internal and external statesand stimuli leading to a given behaviour." Thus, this term may includethe stimuli, specificaction potential, and innate releasing mechanismsinvolved in a given behavioralact. Becauseof the difficulties of determining internal states without experimentation, the term tendencyhas been used recently to indicate"the readinessto showa particular type of behaviouras observedunder natural conditions"(Hinde, 1955-1956,as givenby Marler, 1956a: 5). Causalfactors include both 640 F•½izE3r^•DFmizE3r, Arian Field Ethology [ Vol.Auk 83 externaland internal stimuli leadingto a givenbehavior (Tinbergen, 1959: 29) and thus causationis approximatelysynonymous with the term drive or motivation. Displacementactivity has been used in two ways. It sometimesmeans behaviorwhich is thought to have been activated by one or more drives which are not the drives normally associatedwith it. In other use it refers to the performanceof a behaviorpattern out of the contextof behavior with which it is normally related (Thorpe, 1951). The first definitionis highly controversial(see lersel and Bol, 1958; Rowell, 1961; Sevenster,1961); the secondis more useful for the field observer.One exampleof displacementactivity is that of fighting Great Tits which "may suddenlystart to peck at a bud or pick up leavesand then throw them over their shoulders--activitiesnormally used in feeding" (Hinde, 1959b). Somedisplacement activities, however, are not entirely irrelevant. They are seen frequently when an animal seeminglyhas tendenciesto behave in two different ways or when one strong tendencyis prevented from expression.For example,song in birds which have just escapeda dangeroussituation is probablya manifestationof attack-escapemotiva- tion (seeMoynihan, 1965a: 242). Redirectionoccurs in contextssimilar to those of displacement,but is fundamentallydifferent in causationaccording to Moynihan (1955a: 242) who has defined redirection as "autochthonous activities of a drive di- rected toward an object or animal other than the one releasingand usually directingthem (althoughthe releasingobject or animal remainsavailable, or partly available,as a potentialgoal at the time)." Both redirected attack (Moynihan, 1955a) and redirectedsexual behavior (Ficken and Dilger, 1960) have been noted in birds. Young (1949) describeda male Robin (Turdus migratorius) which tried to copulatewith a female but she threatenedhim each time; the male then "copulated"with a mound of earth and a crumplednewspaper before the female respondedsexually. Animals sometimesstart to perform a movementbut do not complete it. Heinroth termed theseintention movementsbecause they indicate to the observerwhat the animal is going to do (Marler, 1956a: 4). Such acts often occur when an animal is in a situation in which two tendencies are in conflict, and they are consideredto indicate ambivalence.

M^rNTF•N^NCE ACTrv•X•F•S

Maintenance activities are those "concerned with locomotion and the generalhealth and efficiencyof the body, mostly occurringthroughout the year" (Marler, 1956a: 8). Of these,motor patterns involved in the care of the body surface have been called comfort movementsor toilet behavior. Oct.] F•cx• ANDF•CX•, Avian Field F•thology 641 1966 •

Maintenanceactivities are often easilyobserved, yet they remainvirtu- ally unrecordedfor somefamilies. Theseactivities are describedfor various species,particularly by Nice (1943), Andrew (1956), Heinroth and Heinroth (1958), Wicklet (1961b), Ficken (1962b), Meyerriecks(1960), Coutlee (1963), and McKinney (1965). The study of maintenanceactivities is important,aside from the fact that th'eyare necessaryfor the survivalof the animal. Sincesuch move- mentsoften evolveinto componentsof display (see p. 647), knowledge about them is essentialto an understandingof the evolutionof bird displays. Further, they are usually stereotypedand of potential value in taxonomy,particularly at the ordinal and familial levels (see Table 1, below). We shall describebriefly some of the more significant movements with the hope of encouragingfurther observations. Stretchingmovements of two types occur in all carinate birds (Nice, 1943). One involvesthe stretchingdown of one wing and the corresponding leg, typically also with a spreadingof the tail. However,Kilham (1959) notesthat woodpeckersdo not stretch'the correspondingleg. The otheris a stretchingof the wings(closed or perhapssomewhat extended) over the back. During this latter, somespecies spread their tails sym- metricallyat the sametime (Heinroth and Heinroth, 1958: 112). Some, suchas herons,stretch their neck forwardduring both typesof stretching (Meyerriecks,1960: 11). The stretchingdownward of both wingswhich occursin some young passerines,has also been reported in some adult parrots (Nice, 1943) and in adult emberizines(Andrew, 1956). The straighteningof both legswhile the bird is standing(usually without in- volvementof the wings)occurs in youngbirds but lessfrequently in adults (Nice, 1943; Ficken,1962b: 159). Comparativestudies of thesemove- ments,giving the finerdetails of form,sequence (see Ficken, 1962b), and ontogeny,may be of taxonomicvalue. Preeninginvolves manipulating feathers with the bill. Preeningof the tarsi (Whitaker, 1957a) hasalso been noted in severalspecies of passerines. Preeningmovements and sequencesare unknownin mostbirds, but they may be of taxonomicvalue. For instance,oscines and anatidsdistribute oil over the plumagein different ways. Oscinestake oil from the uropygial gland with the tip of the bill, then scratch,first touchingthe tip of the bill with the claw of the second toe and subsequentlyrubbing the claw all over the top of the head. Anatids bite their oil gland and pressoil into the feathersgrowing on or about the gland; they then rotate the head against these feathers (Lorenz, 1956). Anting was describedby Whitaker (1957b: 195) as "the applicationof foreignsubstances to the plumageand possiblyto the skin." Ants are typicallyused, but a varietyof otherobjects may stimulatethe behavior. 642 Fm•:•rAND FICKEN, Avian Field Ethology [ Vol.Auk 83

Simmons (1959) distinguishedtwo types: direct (the bird annointing itself) which occursin over 100 speciesfrom many passerinefamilies, and indirect or passiveanting (the ants swarm over the bird) which occurs in certain larger birds such as corvids, turdids, grallinids, and cracticids. Recent reviews include those by Whitaker (1957b), Simmons (1957a, 1959), and Poulsen (1956). Recently Kelso and Nice (1963) reviewed the experimentalevidence of Dubinin (1951-1956) that anting functions to destroy external parasites. Bathing movementsare describedby Nice (1943). Morris (1955) de- scribedthe variationsin the form and sequenceof bathing and drying movementsin severalbirds. Recordsof specieswhich bathe in water, dew, snow, and dust are scanty. In body shaking,the contourfeathers are first evenly and slowlyruffled. The shaking begins with the torso, and then usually involves the neck and head. In owls, this progressionis reversed (Heinroth and Heinroth, 1958). Sunbathinginvolves a ruffling of the body feathers,spreading the wings and tail, and often leaningto one side, as describedfor many speciesby Hauser (1957) and Gibb (1947). The stimulus for this behavior is not known, although Hauser suggeststhat humidity is more important than high air temperatureand Lanyon (1958) thinks that a suddenwarming of the bird's immediate environment may initiate it. Its function is un- certain. In head-scratching,birds scratchin one of two distinctiveways. Some lower one wing and bring the correspondingleg over the shoulder(in- directlyor "over the wing"). Othersbring the foot directly to the head, with no wing lowering (directly or "under the wing") (Heinroth and Heinroth, 1958; Simmons,1957b, 1961). Head-scratchingprobably functions in arranging and oiling the feathers of the head, and possiblyin relieving irritating stimuli (Simmons, 1961). Observationson the methodof head-scratchingof many commonspecies remain unpublished.Since the type of head-scratchingis almost always consistentwithin a species,it is usefulas a taxonomiccharacter (Table 1). Some nonpasserinefamilies scratchindirectly, others directly; most passerines scratch indirectly. In a few cases there is variation within a family, e.g., Psittacidae (Simmons, 1957b; Brereton and Immelmann, 1962). There is also occasionalvariation within a genus, as in Seiurus (Ficken and Ficken, 1958). In a few known casesthere is a changewith' age. For example,some waterthrushes(Seiurus) scratchdirectly for a few dayswhen the behavior first appearsand then switchto the indirect method (Ficken and Ficken, 1958). The reverseof this processis unknown (Wicklet, 1961b). 1966Oct. ] FICI

In sleepingsome birds (e.g., passetines)place the bill under the scapu- lars while many others (e.g., pigeons) draw their heads betweentheir shoulders.Diverse sleepingpostures are found in other groups,some of them apparentadaptations to peculiaranatomical features. For example, heronsplace the bill betweenthe bend of the wing and the body (Hein- roth and Heinroth, 1958). The hangingparakeets (Loriculus) sleephang- ing upsidedown, and the related Agapornispullaria also sometimessleep in this position(Dilger, 1960). Locomotionobviously includes several types of movement.Since flight is so often modifiedin displays,a knowledgeof the normal form of flight in a speciesis particularly important. Flight patternsare often diagnostic of groups and may separate superficially similar groups such as swifts and swallows. The legsare used for locomotionin walking, running,hopping, sidling, and swimming. Although there are many exceptions,walking and running are characteristicof groundforaging birds. Somespecies, such as the Song Sparrow(Melospiza melodia), employ both, dependingon the environmental conditions.Many specieswhich walk as adults, hop when they first leave the nest (Nice, 1943). The only bird known to do the oppositeis the Wheatear (Oenantheoenanthe) (Neal Smith, pets. comm.). There is variation even within a genus; all speciesof Carduelishop, except C. flavirostris,a tundra dwellerwhich walks. Walking and hoppingare thus rather plasticbehaviors and vary accordingto habitat (seeWicklet, 1961b). The motor patterns of feeding for even a single species,are poorly known although they are obviouslyof great importance. A few feeding patterns which showinteresting taxonomicdistributions are discussedhere. A prying movement (sometimescalled gaping or "Zirkeln") occurs in at least eight passefinefamilies. The birds stick their bills into the ground, a soft substance,or a crack, and then open them with considerableforce. Some ground foraging birds scratch the vegetational substrate, but othersapparently never do (e.g., the Ovenbird,Seiurus aurocapillus, uses its bill in moving leavesaside). Two different scratchingprocedures have evolved. Gallinaceous birds scratch backward first with one foot and then the other,while othergroups (e.g., someemberizines, thrushes, timaliids) useboth feet at once(Wicklet, 1961b: 323). There is no apparentrela- tionshipbetween type of locomotionon the groundand type of scratching, althoughWicklet (1961b) suggestedthat scratchingarose from an existing locomotorytype. The practiceof holding food with one or both feet showsa spotty distribution in both nonpasserinesand passetines.For example,among nine- primaried oscinesit occursin at least some icterids and carduelinesbut is absentin emberizinesand parulids (except for the Yellow-breastedChat, 644 FICKEIVA1N'D FICKEIV, Arian Field ;Ethology [ Vol.Auk 83

Icteria virens [Neal Smith,pers. comm.]). Wickler (1961b) discussesthe peculiardistribution of this motorpattern and suggestsit is at leastpartly related to tarsal structure.

COMMUNICATION

Communication,which occurs"whenever the activities of one animal influencethe activities of another animal" (Alexander, 1960: 38), is a major field of research.Birds are an especiallyfavorable group for study becauseof the similarity of their sensorysystems to our own and the consequentpredominance o[ signalsinvolving color vision and hearing. Communicationis characteristicof all birds and is mostly by display, that is, "those peculiarly standardizedand often exaggeratedperform- ances, including all vocalizationsand many movementsand postures, whichhave become specialized and modifiedas socialsignals or releasers" (Moynihan, 1955a: 240). Displaysusually are fixed actionpatterns (see p. 638). The processof ritualization resultsin inter- and intra-specificsignals or displays. The evolutionof displaysystems involves simultaneous modification of the motor patterns and releasersof the actorsand the patterns of responseand readinessto respondof the reactors. However, certain movementsof birds may communicateinformation without being peculiarly standardizedor exaggeratedto serve a communicatoryfunction-- for instance,the coordinatedflight of birds in a flock. In somecases it is difficult to differentiate between ritualized and non-ritualized movements, e.g., some pre-flight movements.The analysis of displays has been of primary concernto ethologists,perhaps even to the neglectof other areas. Causation.--Recent studies have indicated that displays are produced by dual or multiplemotivational factors (see Tinbergen, 1959). Agonistic displays (the term "h'ostile"was formerly used) are those where pre- sumablyboth attack and escapetendencies are primarily involved. Court- shipdisplays usually involve sexual, attack, and escapetendencies (Mor- ris, 1956), thusanalysis of motivationin thesecases is especiallydifficult. There are as yet no ideal studies which can be used as models for the analysis of causationand it is obvious that many attempts at causal analysisare neededbefore the best methodologycan be established. Threeclosely related approaches to the analysisof causationhave often been used (see Tinbergen, 1959). First, the specific social and environmentalsituations evoking certain behaviorsare often determined. For example, in territorial species,the bird usually exhibits "pure" attack to an intruder within his territory, while the intruder exhibits"pure" escape.Most displaysoccur at or near Oct.] F•c•z•r ^m) F•c•z•r, Avian Field Ethology 645 1966 ] territorialboundaries, where it mightbe expectedthat escapeand attack tendencies are in conflict. Second,study of the separateelements of a complexdisplay may sug- gestwhat tendenciesare involved.Tinbergen (1959: 11) hasshown that in the "Upright"display of gulls,which occurs in agonisticsituations, the attacktendency is indicatedby theraising of thecarpal joints and stretching of the neckwith the bill pointeddownward. These movements are characteristicallyfound in intentionmovements of two typesof attack, i.e., deliveringwing beatswith the foldedwing and peckingat an op- ponentfrom above. However, other components of the Uprightdisplay are indicativeof the escapetendency. The relatedGreat Skua (Stercorarius skua)does not raise th'e carpals in a similardisplay, a factcorrelated with the absenceof wing beating in fights. Stokes'study (1962) of the agonisticbehavior of the Blue Tit (Parus caeruleus),however, somewhat contradicts the ideathat differentelements of a displayindicate different tendencies. He foundthat a givenelement rarelyled to a highlypredictable subsequent behavior, since individual elementsoften simultaneously represented tendencies to attack,stay, and escape.Also the relationshipbetween elements of the displayand subsequent behaviorchanged seasonally. Finally, the useof "time scores"involves analyzing the sequentialpat- terningof behavioracts over time. This yieldsquantitative information on the relative strengthsof the tendenciesinvolved since they represent "the relativeproportion of observedattacks and withdrawals linked with eachdisplay" (Tinbergen, 1959: 33). This methodis properlyused when behavioralpatterns alternate in quicksuccession. An animalin an un- changingenvironment usually does not have rapid motivationalshifts. Thus if a displayalternates rapidly and consistentlywith other movements,it may be concludedthat the display and these movementshave roughly the samemotivational state. This method is difficult to employ in field studiesbecause of the interactionbetween the displayingbird and the reactor(s). This difficulty can be overcomeby analyzingonly those caseswhere the recipient's behavior is relatively constant. Models or mountedspecimens are sometimesuseful as "constantstimuli" (Tinbergen, 1959: 32-33). Althoughthe precedingexamples of causalanalysis have dealt with visualdisplays, the motivationunderlying vocalizations can be determined in muchthe samemanner. Examplesof suchstudies are thoseby Andrew (1961b) on the Blackbird(Turdus merula) and Ficken (1962a) on the AmericanRedstart ( Setophagaruticilla). Function.--Displaysare releasersand their functionsor "signalvalues" are determinedby observingthe reactionof the recipientof the display. 646 FICKENAND FICKEN, Arian Field Ethology [ Vol.Auk 83

The reactionsmay be covert, overt, or both. Covert reactionsare usually more gradual physiologicalchanges, e.g., stimulation of ovulation in budgerigarsby the calls of males (Vaugien, 1951; Ficken et al., 1960; Brockway, 1966). Overt responsesusually occur immediatelyafter the communicatedsignal is received. For example,observations by Tinbergen (1959: 23) and his co-workersshowed that a female Black-headedGull approacheda hiddenmale immediately after an "Oblique-cum-Long-Call" in 13 out of 16 cases;when she failed to do so she was engagedin squabbleswith anotherbird. The femaledid not approachthe male when he was not calling. Thus one can concludethat the functionof the call is to inducethe femaleto approach.Assessing function is often difficult and large numbersof examplesare needed(e.g., Stokes,1962), and some displays, such as "song," may have more than one function. Marler (1956b) discussesin detail the functionsof the vocalizationof the Chaf- finch. Tinbergen (1959: 24) has, from the functionalstandpoint, classified displaysinvolved in agonisticinteractions and pair formation into two groups: spacingout, or distance-increasing,and distance-reducingdisplays. The first groupincludes threat displayswhich tend to make the opponent retreat or cease advancing (Tinbergen, 1959; Moynihan, 1955c). In most speciesseveral threat displayshave evolvedwith slightly different functions--someare more offensive,some defensive,some for long-range signalling(e.g., song), and others for short-rangesignalling (Tinbergen, 1959: 25). Distance-reducingdisplays are also widespreadin birds. Appeasement or submissivedisplays inhibit aggressionby the opponent. Many of these displaysare composedof "reversedmovements" of attack, i.e., opposite movementsfrom those indicating a strong attack tendency (see Marler, 1956a). For example,many passerinesadopt a fluffed submissiveposture which is very unlike the postureof an attacking bird, which'sleeks its feathers and directs its head toward the opponent. Distance-reducing displays are particularly important in promoting social bonds between membersof a pair, youngand parent, etc. Origin o/ displays.--Displaycomponents seem to be "derived" activities,having ultimately evolved from precursorswhich did not have signal value (Tinbergen,1952). The originof vocalizationsis very poorlyunder- stood,but Spurwayand Haldane (1953) suggestthat they are derived frombreathing movements. Moynihan (1955a) hasgiven four sourcesof visualdisplays. First, displacementactivities are oftenincorporated into displays.An exampleis the mock preeningof theeMallard (Anas platyrhynchos)--asimilar display also occursin many other ducks•in which the male, while courtingthe female,reaches his bill behind his Oct.] FICKEN^•) F•CX•N,Avian Field Ethology 647 1966.1 slightlylifted wing, movingthe bill over the undersideproducing a "Rrr" sound. In somespecies specially modified feathersare exhibitedduring this display(Lorenz, 1951-53). Mock preeningseems clearly derived from real preening,which, in a courtshipsituation, is "out of context." Displays may also be derived from redirectedactivities, but examples are rarer. Moynihan (1955a: 243) and Tinbergen (1959) suggestthat the peckinginto the groundand pullingof vegetationin somegull displays during territorial clashesprobably had their origin in redirectedaggression. Third, intention movementsare commonlyritualized into display components;Daanje (1950) givesmany examples.Both sexualand agonistic displaysmay have componentsderived from the lowering of the breast and the wing and tail movementswhich precede jumping up or flying. Finally, higher intensity movements characteristic of the tendencies producingthe display may becomeritualized. Moynihan (1955a: 241) cites examplesof the "Swoop" and "Soar" displaysof the Black-headed Gull which include attack and escape elements of greater "vigor and elaboration" than intention movementswhich are low intensity movements. Sincemany displaysare composedof more than one postureor movement, often with different evolutionaryhistories, it is necessaryto con- sider each componentseparately. In many casesthe same component occursin more than one display and this often facilitatesanalysis. De- terminingthe derivationof componentsmay be relativelysimple when they are similarto an unritualizedbehavior pattern (see Daanje, 1950; Marler, 1956a; Tinbergen,1959). However,in other cases,e.g., the complexcourtship displays of manakins(Pipridae) (Sick, 1959) andbirds of paradise(Paradiseidae) (Armstrong, 1942), the movementshave under- gone so much elaborationthat their precursorsare unrecognizable.In thesecases comparative studies of a group of related species,some of whichpossess the movementin questionin a lessritualized form, provide many usefulclues to derivation. The study of the ontogenyof displays is an important but seldomused method of obtaining evidenceabout derivation and th'e developmentalintegration of form and function (Moynihan, 1959). The types of changesinvolved in the ritualization of a movementhave been discussedby Daanje (1950), Morris (1957), Blest (1961), and Hinde and Tinbergen(1958). We mentionhere a few of the moreobvious changeswhich occur. Conspicuousstructures which are correlatedwith a movementoften are developed.As Hinde and Tinbergen (1958: 258) point out, "there has probably alwaysbeen a parallel elaborationof structure and movement."For example,warblers which habitually spread their 648 FICKENAND FICKEN, Arian Field Ethology [ Vol.Auk 83 tails have strikingtail markingswhile thosethat wag or flick their tails do not (Ficken and Ficken, 1962a). Changesin the nature of the movement,such as exaggeration(Lorenz, 1952: 7; Daanje, 1950), and changesin speedof performanceas comparedwith the precursor(Blest, 1961) alsooccur. Rhythmicrepetition maydevelop, as in the"dancing" movements of estrildinefinches (Morris, 1957). There are also often changesin coordination(Daanje, 1950) as, for example,the simultaneousoccurrence of a componentof the first phaseof take-off (crouching)and one from the second(tail lowering). In ritualization a variable behavior often becomes more constant in form. Wide ranges in the strengthsof the eliciting factors are accom- paniedby slight changesin form of the display, giving it a typical intensity (Morris, 1957; Hinde and Tinbergen,1958). The courtshipdis- play of the male CutthroatFinch (Amadinafasciata) remainssimilar in form despitemotivational shifts (Morris, 1957). In contrast,graded displaysare oneswhere slight motivational shifts are reflectedin display movements(Marler, 1961a: 107). Emancipationoccurs when ritualized responsesbecome freed from the causal factors of their precursors. There are few examplesof this phe- nomenonand it may not be as widespreadas once thought (Hinde and Tinbergen,1958: 259). Blest (1961) discussesthe topicat length. Naming displaysand the problemof homology.--Displaysshould be givendescriptive names, not onesthat suggesttheir causationor function, particularlybecause concepts about these may change(Tinbergen, 1959; Dilger, 1962: 85). Displaysare usuallynamed for their most obvious componentor combinationof components(Tinbergen, 1959). Namesof displaysare usuallycapitalized to distinguish'them from namesof other behaviors. Behavioralhomology, as with morphologicalhomology, implies common phylogeneticdescent. However, homologiesare rarely known with cer- tainty. Tinbergen (1959: 7) suggeststhat similarity in form, wide and continuousdistribution through the group, and similar motivations,functions, and derivations,are indicativeof homology. Difficulties arise when the same name is applied to similar displays in two speciessince this implies hornology. The number of descriptive terms to use for a given type of display is limited, however. It may be best at this time to use the most useful descriptivewords and to state that no homologiesare implied with behaviorpatterns, of other species,that might bear the

same name. Oct.] F•cxE• Am)F•cx•, Avlan Field l•hology 649 1966 J

BEHAWOR A•) EVO•.UT•O•

Intraspecificand interspecificvariation in behavior.--The evolutionof behaviorat the intraspecificlevel has attracted attention only relatively recently. In a few casesthe selectivepressures responsible for geographic variation are known. Curio (1961: 47) demonstratednumerous differencesbetween two subspeciesof Pied Flycatchers;these were manifested mainlyin changesof thresholdand motivation.One differencewhich has obvioussurvival value is that the northernsubspecies mobbed a predator, the Red-backedShrike (Lanius collurio), strongly,while the southern subspecies,which is outsidethe rangeof shrikes,did somuch more weakly. Adaptive geographicvariation is also shownin the beggingnotes of nestlingSnow Buntings (Plectrophenax nivalis). On Bylot Island nests are accessibleto weaselsand the nestlingsare usuallysilent (Drury, 1961). However,in Greenlandthere are no weaselsand the young have loud and frequent beggingcalls (Tinbergen, 1939). In addition to extensiveindividual differencesin song (Weeden and Falls, 1959; Marler, 1961b), there is also geographicvariation in some cases(e.g., Mayr, 1942; Benson,1948; Lanyon and Fish, 1958; Thorpe, 1961; Borror, 1961). In many casessong "dialects" (interpopulational differences) are phenotypic rather than genotypic differences. Local dialectsmay be signsof incipientspeciation (Marler and Tamura, 1962). Behaviormay also be remarkably similar betweenisolated populations of the same species. Dilger (1960) found that the four subspeciesof Agapornispersonata are very much alike in such features as courtship feeding,precopulatory displays, the mannerof carryingnest material, and nestbuilding, although they differ distinctlyin color. In somebirds the songsof two isolatedsubspecies are remarkablysimilar, e.g., the songsof the British and Tenerife Island forms of both the Corn Bunting (Em- berizacalandra) and Blackbird(Marler, 1960). Polymorphismin behavior has been infrequently described. A well studiedcase is that of the White-throatedSparrow (Zonotrichia albicollis). Someindividuals have a white median stripe on the crown and others a tan stripe; matingsare usually of birds of oppositemorphs (Lowther, 1961). White-stripedfemales sing a nearly normal songand react to song playback, but tan-striped femalesdo neither. White-striped males react to playback longer and stronger than do tan-striped males. The assortative mating is probably a result of these behavioral differences (Lowther, 1962). Huxley (1955) cites examplesof behavioral polymorphism in birds in host preference,vocalizations, nest-construction, and breeding season. Sexualdimorphism in behavioris commonand is often associatedwith dimorphismin color pattern. In the Common Grackle (Quiscalusquis- 650 Fxc•:ExA2•D FXC•:EX, Avian Field Ethology [ Auk 1_Vol. 83 cula) both sexesgive all but one of the displaysof the other, but differ- encesoccur in the relative frequencywith which they give these (R. Ficken, 1963: 70). Such differencesbetween the sexesare probably due to differencesin threshold,as are somedifferences in the frequencyof homologousdisplays among related species(Hinde, 1959a: 94) and among populationsof the samespecies (Curio, 1961: 47). Thresholddifferences also may be responsiblefor individual variation and the rare occurrence of a behavior.For example,a displaycommon in onespecies of Agapornis wasseen only twicein anotherspecies despite years of observation(Dilger, 1960: 682). Threshold changesare a convenientway of retaining the potential for possiblerapid shifts of behavior in responseto changing selectivepressures (R. Ficken, 1963: 70). Displays and associatedreleasers, such as color patterns, are often involved in maintainingreproductive isolation among closelyrelated sympatric species.Thus, there often is selectionfor differencesin behavior patterns,particularly those involved in pair formationand courtship. In birds with a long period betweenpair formation and copulation, displayspreceding copulation need not be as specificas thosein birds in which copulationoccurs at theeinitial meeting (e.g., Marler, 1957: 28; Hinde, 1959a: 89-90). In most speciesmales are more active in courtship than femalesand their displaysare more speciestypical. Hinde (1959a) discussesat length the nature of speciesdifferences in courtship.These includedifferences in the actual and relative strengths of the tendenciesto attack,escape, and behavesexually toward the mate. For example,Chaffinch and Bullfinch (Pyrrhula pyrrhula) malesare less aggressivetoward their matesthan are the Goldfinch(Carduelis carduelis) and Canary (Serinus canarius). The Chaffinch and Bullfinch are also strongly sexually dimorphic in color pattern, while the Goldfinch and Canary are not (Hinde, 1959a: 93-94). The femalesof non-dimorphic speciesmay release more aggressionbecause of their plumage color (see also Hamilton, 1961). In someclosely related sympatricspecies colors and patterns are more divergentthan the display movements(Hinde, 1959a: 96; Ficken and Ficken, 1962a). A goodexample of modificationof courtship,as a result of selectionon some other feature, occurs among the Ploceinae (Crook, 1959). The nest of primitive membersis globularand malesdisplay just below the entrance,which' is near the top. In other speciesthe nest is modifiedand the entrance is underneath; the birds display with the same orientation to the entrance,but hang upsidedown. Behavior and speciation.--The role of behavior patterns as isolating mechanismsbetween two sympatric populations has recently been re- 19•6t•] Fm•E•^• Fm•E•,Avian Field EthoIogy 651 viewedby Hinde (1959a), Spieth (1958), and Mayr (1963). Although there are severalother typesof mechanismsreducing the probabilityof interspecificcrosses (premating mechanisms) or of their success(post- matingmechanisms), behavioral isolating mechanisms are probablythe mostimportant (Mayr, 1963:95). Vocalizationsand visualbehavior are effective in various combinationsin reducingmixed pairings (Dilger, 1956a;Stein, 1958, 1963; Lanyon, 1960a; and Hinde, 1959a, give many examples).It hasbeen suggested that learnedbehavior may be important as an isolatingmechanism (see Cushing, 1941), but Hinde (1959a: 100) suggeststhat this is improbablealthough he statesthat "learnedmating preferencescould tide the speciesover a periodin whichthe courtship displayshad altered somewhat but therehad been no corresponding genetic changein responsivenessto these displays by otherindividuals." In some casesalthough the signalsthemselves may be innate,the responseto them may be learned (Marler, 1961a: 116). Habitat preferenceand feedingbehavior differencesmay reducecom- petition and enable reproductively isolated populations to coexist sym- patrically (Hinde, 1959a; Dixon, 1961, gives examplesin Parus). Al- though there are often morphologicaldifferences which supplementthese behavioraldifferences, MacArthur (1958) found that in five speciesof sympatricDendroica, which did not differ appreciablyin body lengthor bill size,behavioral differences in the methodof feedingwere of prime importancein allowingcoexistence in the samehabitat. Brown and Wilson (1956) give examplesof characterdisplacement in bird behaviorincluding feeding habits, habitat preference,and song. Actions of selectivepressures on behavior.--Single environmental factors may produce a multiplicity of effects on behavior. This has been well demonstratedby Cullen (1957) in her study of the adaptationsof the Black-leggedKittiwake (Rissa tridactyla) to cliff nesting. Such nesting situations,when comparedwith the open ground nesting of most gulls, result in relaxation of predator pressureand restrictionsimposed by the limited area and dangerto the youngof falling off the cliff. Both factors have producednumerous changes in behavior; the alarm call is rare, and the submissivedisplay is exaggerated,thus obviatingfalling. Most often many forcesshape a singlebehavior pattern. For example, if song (and certain call notes) has the primary functionsof attracting conspecificfemales and repelling rival males, there are many variables which would influenceits evolution. Songswith propertieswhich promote ease of location by conspecificswould be advantageousbut, at theesame time, would render the singingindividual more vulnerable to predation. Furthermore,individual recognition by conspecificsof the songof males on neighboringterritories reduces aggression (Weeden and Falls, 1959). 652 F•c•rE•^•D F•c•rE•, Arian Field Ethology [ Vol.Auk 83

TABLE 1 EXAMPLES OF TI•IE USE OF BEI•IAVIORAL CItARACTERS 12• TAXOIgOlVIY

Activity Level used Taxonomicdiscrimination Method of head Familial Recurvirostra, Himantopus, and Haematopus related scratching to charadriids rather than scolopacids (Simmons, 1957b) Subfamilial SuggestPsittacinae may be polyphyletic (Brereton and Immelmann, 1962) Holding food with Familial Separates Icteria virens from Parulidae (Fieken and the foot Fieken, 1962b) Dust bathing Subfamilial Supports relationship of Passer to ploceids (Mayr, et al., 1953: 120) Water bathing Familial Separatestimaliids from most other passerines(Sim- method mons, 1963) Method of oiling Familial Separatestimaliids from most other passerines(Sim- feathers mons, 1963) Visual agonistic Generic Splits Hylocichla mustelina from Catharus spp. (Dil- displays ger, 1956b) Generic Often useful at this level in Anatidae (JohnsBard, 1961) Song Tribal Useful at this level in estrildine finches (Delacour, 1943) Intergeneric Shows relationship of Setophaga ruticilla and Den- droica (Fieken and Fieken, 1962a) Intrageneric Shows close affinities of Vermivora pinuc and V. chrysopteraand of V. peregrinaand V. ruficapilla (Saunders, 1951) Flight call notes Subfamilial Suggest relationship of Fringilla and carduelines (Marler, 1957) Length of pair bond Subfamilial Separatesestrildines from ploceids (Steiner, 1955 in Mayr, 1958) Sub familial Separates Anatinae from some other subfamilies (JohnsBard,1961) Bower construction Subfamilial Subdivides Ptilonorhychidae into two subfamilies and familial and places cat-birds in a separate family (Ailuroe- didae) (Marshall, 1954: 183) Form of courtship Subfamilial Shows affinity between Frlngilla and carduelines display (Andrew and Hinde, 1956) Interfamilial Presence of tail quivering suggestspossible relationship among corvids, estrildines,and ploceids (An- drew, 1961a: 561) Nest construction Ordinal Evidence for relationship between hummingbirds and swifts (Pearson, 1953; Amadon, 1959) Intergeneric Indicates close relationship between Hirundo and Ptyonoprogne (Mayr and Bond, 1943) Familial Separates Peucedramus from Parulidae (George, 1962) Time of initiation Subfamilial Separates estrildines from ploceids (Steiner, 1955 of incubation in Mayr, 1958) Participation of Subfamilial Separates Anserarias semipalmata from other Anat- sexesin parental idae (JohnsBard,1961)

Feeding of nestlings Sub familial Separates estrildines from ploceids (Steiner, 1955 in Mayr, 1958) Nest sanitation Familial Separates Peucedramus from Parulidae (George, 1962) Oct.] F•cxEN ANDF•CXEN, Avian Field E•hology 653 1966 J

However,this advantage is counterbalancedby the stereotypyneeded for speciesrecognition (Marler, 1957,1960). Localdialects may helpinduce the return of malesand femalesto their birthplaces(see Marlet and Tamura,1962: 375) wherethey may be bestable to survive.The physi- cal characteristicsof the habitatmay have importanteffects on the form of song.Th'us, limited visibility within dense habitats may resultin a highdegree of dependenceon vocalsignals (Dilger, 1956b: 350) andcan favorthe development of low frequency songs which carry farther (Ficken and Ficken, 1962a). Speciesspecificity is stronglyselected for in areas with rich avifaunaswhere. each song must stand out conspicuouslyfrom the general backgroundof sound (Marler, 1960: 350). Other factorsmay influencethe evolutionof bird vocalization.Certain distantlyrelated, but ecologicallycompeting, species have developedsimi- lar threatcalls (Dilger, 1956a: 196; Dixon, 1961: 199-200) whichcould result in someincrease in spacing. Socialstructure is also influential. Dueting (antiphonalsinging or calling) is associatedwith long-termor closecontact between members of a pair and occursin sometropical and temperatezone species which remain mated through the year and alsoin a few migratorybut colonialspecies (see Armstrong, 1942: 156-159). Songs of estrildinefinches are of low volumeand specificity,qualities correlated with their useonly as closerange sexual signals; in thesebirds pair formation occursin the flock so the female seesa potential mate before shehears him and malesdo not sing in territorial defense(Hall, 1962). The whole subjectof "song" is further complicatedby the wide range of differences among speciesin the degree to which a song is modified by learning (Lanyon, 1960b; Thorpe, 1961; Marler et al., 1962) and also by the difficulty of definingit. Convergence.--Behavioralconvergence is widespread,probably as a result of various factors. One is the limited number of possibilitiesavail- able, a fact seemingto accountfor the similarity of appeasementpostures of distantlyrelated species(Tinbergen, 1959: 61). Functionalnecessity may also lead to similarity, as with the female solicitingdisplay, most of the componentsof which are functionalin facilitating copulation(Marlet, 1956a: 116-117). Similarity in habitats is often associatedwith convergence.Thus, scratchingthe groundby hoppingbackwards has appeared in eight passerinefamilies (Wickler, 1961b: 323) and flycatchingin at least five (Hinde and Tinbergen, 1958: 262). Hole-nestingbirds of different families may exhibit similarities which are adaptive to this mode of life, e.g., gapingof the young at the darkeningof the nest hole as occurs when the parents arrive (Haartman, 1957: 341), loud begging calls (Heinroth in Tinbergen, 1939: 35-36), and notesresembling the hissing of a snake (Sibley, 1955). Similar precursorsof two behaviorsmay lead 654 Fxci•rA>m Fm)c•r, Avian Field Ethology [ Vol.Auk 83 to convergence,as in courtshipfeeding displays of different specieswhich are derivedfrom similarjuvenile begging displays (Andrew, 1961a: 562). Behavior and taxonomy.--A taxonomiccharacter is "any attribute of an organismor of a groupof organismsby whichit differs from an organism belongingto a different taxonomiccategory or resemblesan organismbe- longing to the same category" (Mayr et al., 1953: 105). Tinbergen (1959), Hinde and Tinbergen(1958), Mayr (1958), Amadon(1959), and Wickler (1961a, b) reviewthe subjectof the use of ethologicalchar- acters in taxonomy. Examples of the use of behavioral charactersin taxonomy are listed in Table 1. This table is by no meanscomprehensive but is designedto give an idea of the variety of typesof behaviorwhich have beenused and the taxonomiclevels at which they have proven to be useful. In many of thesecases morphological characters substantiate conclusions drawn from behavior. The use of behaviorpatterns as taxonomicch'aracters presents many of the sameproblems (such as convergence)encountered in usingmorpho- logicalones as well assome additional difficulties (e.g., modification through experience). Some patterns, such as call notes, visual displays, and maintenanceactivities, develop nearly identicallyunder a wide range of conditions. Others, such as song and reactionsto environmentalstimuli (Mayr, 1958), are often easilymodified (see Tinbergen,1959: 4). The functionalsignificance of behaviorsmust be known if they are to be usefultaxonomically. For example,displays and morphologicalfeatures servingas releasers,which have beenselected for specificdistinctiveness, will be of little use abovethe specieslevel (Marler, 1957). In casesin which there is interspecificcommunication, as often occurswith certain mobbing calls, there is often little specific distinctivenessand the calls may evenbe convergent(Marler, 1957: 25), thus of little taxonomicuse. Behavior charactersare sometimesmore useful than others since they may be more conservative(Mayr et al., 1953: 119) or permit finer dis- criminations(Mayr, 1958). In morphologicallysimilar groups they are particularlyimportant since even the slightestclues must be used(Mayr, 1958: 348). Behaviorhas been of greatvalue in detectingsibling species. Thus, Empidonaxbrewsteri was first detectedas beingdifferent from a sibling(E. trallii) by its song(Stein, 1963). Behaviorpatterns are generallyuseful as taxonomiccharacters in direct proportionto their improbability. Improbability increaseswith complexity. However,beh'avior patterns which are performedin oneof two ways,neither seemingto be more advantageous,are often useful taxonomiccharacters. This is true of the form of the vertical componentin tail movementsand type of headscratching movement. Oct.] Fm•cE•r^•rt) FrancEs,, Avian Field Ethology 655 1966 •

Comparisonof "likes" betweenspecies must be donewhen individuals are similarly motivated. Thus, only the highestintensities of particular behaviorpatterns should be comparedand the behaviorsshould be studied in similar contexts. Comparisonsshould, of course,be made during the samestages of the life cycle. The useoJ behavioralclues in reconstructingevolution.--The retention of characteristicswhich seem to be primitivemay give cluesto the origin of a group. For example,some wood warblers nest low (but feedhigher), singflight songsin the forest,and have distractiondisplays. These behavioralpatterns are characteristicof ground-adaptedspecies and suggest that parulidsmay have originatedfrom sucha group (Ficken and Ficken, 1962a). Retention of apparently "useless"characters may indicate the recency of an adaptation; for example, the eggs of the h01e-nesting ProthonotaryWarbler (Protonotariacitrea) are heavily spotted. Within a speciesthere are two types of rare behavior. Some are found in all or almostall individualsbut occuronly with unusualor supernormal stimulussituations. Others are found in only a small part of the population but are not dependenton unusualstimuli. If they are commonin closelyrelated species it is highly probablethat they were alsoat one time of more frequent occurrencein the speciesin question. Likewise, if a particular rare display is complex,it is also very likely that it was once more common. It is usually impossibleto know whether such behaviors are beingselected for or against,but they conceivablycould be important to the future evolutionof the group.

ACKNOWLEDGMENTS

We wish to thank the A.O.U. Committee on Research for inviting us to write this paper and for guidance during its preparation. In addition, we wish to thank Robert Beck, Walter Bock, Carl Carlson, William C. Dilger, Lester L. Short, Francis G. Scheider, George Watson, and Richard Zusi for their criticisms. This work was supported by National Salerice Foundation Grants GB 891 and GB 3226.

LITERATURE CITED

An•XA•mER, R.D. 1960. Sound communication in Orthoptera and Cicadidae. Pp. 38-92 in Animal soundsand communication(W. Lanyon and W. Tavolga, eds.). A.I.B.S. Publication No. 7. Washington, D.C. A•At)O•, D. 1959. Behavior and classification: some reflections. Naturforschenden Gesellschaftin Zijrich, 104: 73-78. A•rtm•w, R. J. 1956. Normal and irrelevant toilet behaviour in Emberiza spp. Brit. J. Anim. Behav.,4: 85-91. Am)R•W, R.J. 1957a. The aggressiveand courtship behaviour of certain emberizines. Behaviour, 10: 255-308. A•mR•W, R.J. 1957b. A comparativestudy of the calls of Emberiza spp. (buntings). Ibis, t)O: 27-42. 656 FICKENAND FICKEN, Avian Field Ethology [ Vol.Auk 83

A•DREW,R.J. 1961a. The displaysgiven by passerinesin courtshipand reproductive fighting: a review. Ibis, 103a: 315-348, 549-579. ANDPa•W,R. J. 1961b. The motivational organisation controlling the mobbing calls of the Blackbird (Turdus merula). I. Effects of flight on mobbing calls. II. The quantitative analysis of changesin the motivation of calling. III. Changes in the intensity of calling due to changes in the effect of the owl or to the progressive waning of mobbing. IV. A general discussion of the calls of the Blackbird and certain other passerines. Behaviour, 17: 224-246, 288-321; 18-. 25-43, 161-176. A•REW, R. J., ^• R. A. I-I•E. 1956. The systematic position of Fringilla• someethological evidence. J. f. Orn., 97: 263-273. Am•sT•o•G, E. A. 1942. Bird display and behaviour. New York, Oxford Univ. Press. BE•so>r, C.W. 1948. Geographic voice-variation in African birds. Ibis, 90: 48-71. B•;•sT, A.D. 1961. The conceptof "ritualisation." Pp. 102-124 in Current problems in animal behaviour (W. H. Thorpe and O. L. Zangwill, eds.). Cambridge, Cambridge Univ. Press. BorRow, D. J. 1961. Intraspecific variation in passerine bird songs. Wilson Bull., 73: 57-78. BR•E•O•, J., ^• K. I•rM•L•^•r. 1962. Head-scratching in the Psittaciformes. Ibis, 104: 169-175. B•oc•cw^¾, B. F. 1965. Stimulation of ovarian development and egg laying by male courtship vocalization in Budgerigars (Melopsittacus undulatus). Anim. Behav., 13: 575-578. B•ow•, W. L., ^•D E. O. WinsoN. 1956. Character displacement. Syst. Zool., 5: 49-64. Cotr•]•, E. 1963. Maintenance behavior of the American Goldfinch. Wilson Bull., 75: 342-357. C•oox, J.H. 1959. Behaviour study and the classificationof west African weaver birds. Proc. Linn. Soc. London, 170 Section, 1957-58, pt. 2: 147-153. Ctrn•;•, E. 1957. Adaptations in the kittiwake to cliff-nesting. Ibis, 99: 275-302. Ctrmo, E. 1959a. Verhaltensstudien am Trauerschn•ipper. Zeits. f. Tierpsych., 3: 1-188. Ctrmo, E. 1959b. Beobachtungen am I-Ialbringschniipper, Ficedula sernitorquata, im mazedonischenBrutgebiet. J. f. Orn., 1OO: 176-209. Ctr•o, E. 1961. Zur geographischenVariation von Verhaltensweisen.Die Vogelwelt, 82: 33-48. CtrsRr•G, J. E. 1941. Non-genetic mating preference as a factor in evolution. Condor, 43: 233-236. D^^•J•, A. 1950. On locomotory movements in birds and the intention movements derived from them. Behaviour, 3: 48-98. D•;^cotr•, J. 1943. A revision of the subfamily Estrildinae of the family Ploceidae. Zoologica,gS: 69-86. Dn•½E•, W. C. 1956a. Adaptive modifications and ecologicalisolating mechanisms in the thrush genera Catharus and Hylocichla. Wilson Bull., 68: 171-199. Dn;c.E•, W. C. 1956b. Hostile behavior and reproductive isolating mechanisms in the avian genera Catharus and Hylocichla. Auk, 73.' 313-353. Dma•, W. C. 1960. The comparative ethology of the African parrot genus Agapornis. Zeits. f. Tierpsych., 17: 649-685. Dn•½ER,W. C. 1962. Methods and objectives of ethology. Living Bird, 1: 83-92. Drxo•, K.L. 1961. Habitat distribution and niche relationshipsin North American Oct.1966 ] F•CXEN^ND F•CX•N, Avian Field Ethology 657

speciesof Parus'.Pp. 179-216 in Vertebratespeciation (W. F. Blair, ed.). Austin, Univ. of Texas Press. DRUR¾,W.H. 1961. Studiesof the breedingbiology of Horned Lark, Water Pipit, Lapland Longspur,and Snow Bunting on Bylot Island, NorthwestTerritories, Canada. Bird-Banding, 32: 1-46. D•JmNn%V. B. 1951-56. Fauna, USSR. Paukoobrasnye (Arachnida). 1951. Vol. 6, no. 5. Feather mites (Amalgesoidea). Ch. I. An introduction to their study. Acad. of Sciences,USSR, Moscow-Leningrad:363 pp. 1953. Ibid., vol. 6, no. 6. Ch. II. Families Epidermoptidae and Freyanidae. 1956. Ibid., vol. 6, no. 7. Ch. III. Family Pterdichidae. 814 pp. E•BL-EmEs•LDr, I., ^t,m S. KR^•r•R. 1958. Ethology, the comparative study of animal behavior. Quart. Rev. Biol., 33: 181-211. E•rL•N, J.T. 1955. The study of behavior in birds. Pp. 105-153 in Recent studies in avian biology (A. Wolfson, ed.). Urbana, Univ. of Illinois Press. ErKr•, W. (ed.). 1964. Social behavior and organization among vertebrates. Chicago, Univ. of Chicago Press. FmK•N, M. S. 1962a. Agonistic behavior and territory in the American Redstart. Auk, 79: 607-632. Fmx•N, M. S. 1962b. Maintenance activities of the American Redstart. Wilson Bull., 74o: 153-165. FmXEN, M. S., ^N• W. C. D•R. 1960. Comments on redirection with examples of arian copulationswith substitute objects. Anim. Behar., 8: 219-222. F•cK•, M. S., ^N• R. W. F•cK•m 1962a. The comparative ethology of the wood warblers: areview. Living Bird, 1: 103-122. FmK•N, M. S., ^N• R. W. F•CKEN. 1962b. Someaberrant charactersof the Yellow- breasted Chat, lcteria virens. Auk, 79: 718-719. F•cK•, R. W. 1963. Courtship and agonisticbehavior of the Common Grackle, Quiscalusquiscula. Auk, 8'0: 52-72. F•CK•N, R. W., ^N• M. S. F•cK•. 1958. Head-scratchingin Seiurus (Parulidae) and other passerines.Ibis, 100: 277-278. FICKEN,R. W., A. V^NTIENHOVEN, M. S. FICEEN,AND F. C. SIBLEY. 1960. Effect of visual and vocal stimuli on breeding in the Budgerigar (Melopsittacus undulatuy). Anita. Behav., 8: 104-106. GEORGE,W.G. 1962. The classificationof the OliveWarbler, Peucedramus taeniatus. Amer. Mus. Novit., no. 2103: 1-41. GraB,J. 1947. Sunbathingby birds. Brit. Birds,4,0: 172-174. H^^RTM^N,L. VON. 1957. Adaptation in hole-nestingbirds. Evolution, 11: 339-347. H^LL, F. 1962. Evolutionary aspectsof courtshipsong in estrildine finches (abstract). Anita. Behav., 10: 178. H•LTON, T. H. 1961. On the functions and causes of sexual dimorphism in breeding plumage characters of North American species of warblers and orioles. Amer. Nat., 95: 121-123. H^•JSER,D. 1957. Some observationson sun-bathing in birds. Wilson Bull., 69: 78-90. HmNROra, O., ^ND K. HmNROta. 1958. The birds. Ann Arbor, Univ. of Michigan Press. H•N•, R.A. 1952. The behaviour of the Great Tit (Parus major) and some related species.Behaviour Suppl., 2. Hr•E, R. A. 1955-56. A comparativestudy of the courtship of certain finches (Fringillidae). Ibis, 97: 706-745; 98: 1-23. 658 FICKENAND FICKEN, Avian Field Ethology [ Vol.Auk 83

I-I•n•, R.A. 1959a. Behaviour and speciationin birds and lower vertebrates. Biol. Rev., 34: 85-128. I-I•n•, R.A. 1959b. Some recent trends in ethology. Pp. 561-610 in Psychology, a study of a science. Study I, vol. 2 (S. Koch, ed.). New York, McGraw-Hill. I-I•ng, R.A. 1961. Behavior. Pp. 373-411 in Biology and comparativephysiology of birds, Vol. 2 (A. J. Marshall, ed.). New York, Academic Press. I-I•n•, R. A., A•O N. T•B•C•. 1958. The comparativestudy of species-specific behavior. Pp. 251-268 in Behavior and evolution (A. Roe and G. G. Simpson, eds.). New Haven, Yale Univ. Press. I-IuxLg¾, J.S. 1955. Morphism in birds. Acta XI Int. Orn. Congr., Basel: 309-328. I•asgn, J. J. A. VAN,AND A. C. A. BoL. 1958. Preening of two tern species.A study on displacementactivities. Behaviour, 13: 1-88. I•r•Z•rA•,K. 1959. Experimentelle Untersuchungeniiber die biologischeBedeutung artspezifischer Merkmale beim Zebrafinken (Taeniopygia castanotis Gould). ZoologischeJahrbiicher, 116: 437-592. JoabscAmp,P. 1961. The taxonomy of the Anatidae--a behavioural analysis. Ibis, 103a: 71-85. Kgzso, L., Am) M. M. Nm•. 1963. A Russian contribution to anting and feather mites. Wilson Bull., 75: 23-26. K•L•A•r, L. 1959. Head-scratchingand wing-stretchingof woodpeckers.Auk, 76: 527-528. LArynx, W. E. 1958. The motivation of sun-bathingin birds. Wilson Bull., 70: 280. L.a•vo•, W. E. 1960a. The Middle American population of the CrestedFlycatcher Myiarchus tyrannulus. Condor, 62: 341-350. LArYnx, W. E. 1960b. The ontogeny of vocalizationsin birds. Pp. 321-347 in Animal soundsand communication (W. Lanyon and W. Tavolga, eds.). A.I.B.S. Publ. No. 7. Washington, D.C. LANYON,W. E., ANDW. R. F•s}t. 1958. Geographicalvariation in the vocalizations of the WesternMeadowlark. Condor, 60: 339-341. LSar•L, H. 1960-61. Vergleichende Studien iiber Brutbiologie und Verhalten der Kleiber Sitta whiteheadiSharpe und Sitta canadem{sL. J. f. Orn., 101: 245-264; 102: 111-132. LORENZ,K. 1935. Der Kumpan in der Urnwelt des Vogels. J. f. Orn., 83: 137-213, 289-413. Lom•NZ, K. 1950. The comparative method of studying innate behaviour patterns. Pp. 221-268 in Physiologicalmechanisms in animal behaviour. Soc. Exp. Biol. Sympos.4, Cambridge. Loa•NZ, K. 1951-53. Comparative studies on the behaviour of the Anatinae. Avic. Mag., 57: 157-182; 58: 8-17, 61-72, 86-94, 172-184; 59: 24-34, 80-91. Lom•NZ,K. 1956. The objectivistictheory of instinct. Pp. 51-76 in L'instinct dans le comportementdes animaux et de l'homme. Paris, Masson et Cie. LOWT•Ea, J. K. 1961. Polymorphism in the White-throated Sparrow Zonotrichia albicollis (Gmelin). Canadian J. Zool., 39: 281-292. LOWT•Ea, J.K. 1962. Behavior in relation to polymorphicvariation in the White- throated Sparrow, Zonotrlchia albicollis. Abstracts, 13th Intern. Orn. Cong., Ithaca, New York. MAcAa•ua, R. H. 1958. Population ecology of some warblers of northeastern coniferousforests. Ecology, 39: 599-619. M^an•a, P. 1956a. Behaviourof the Chaffinch,Fringilla coelebs.Behav. Suppl.,5. Oct.1966 ] FIc•c•^•I) FIC•CE•, Avian Field l•thology 659

M^RLER, P. 1956b. The voice of the Chaffinch and its function as a language. Ibis, 98: 231-261. M^RL•a, P. 1957. Specific distinctivenessin the communication signals of birds. Behaviour, 11.- 13-39. MâLER, P. 1960. Bird songs and mate selection. Pp. 348-367 in Animal sounds and communication (W. Lanyon and W. Tavolga, eds.). A.I.B.S. Publ. No. 7, Washington, D.C. Mâ•a, P. 1961a. The evolution of visual communication.Pp. 96-121 in Vertebrate speciation (W. F. Blair, ed.). Austin, Univ. of Texas Press. M^R•a, P. 1961b. The logical analysisof animal communication.J. Theoret. Biol., 1: 295-317. Mâ•a, P., M. Ka•Ta, ^>;DM. T^•rum. 1962. Song developmentin hand-raised Oregon Juncos. Auk, 79: 12-30. M^•,•a, P., ^•D M. T^•ruR^. 1962. Song "dialects"in three populationsof White- crowned Sparrows. Condor, 64: 368-377. Mâs•-•L, A. J. 1954. Bower-birds. Oxford, Clarendon Press. M^¾a, E. 1942. Systematicsand the origin of species.New York, Columbia Univ. Press. M^¾a, E. 1958. Behavior and systematics. Pp. 341-362 in Behavior and evolution (A. Roe and G. G. Simpson,eds.). New Haven, Yale Univ. Press. M^¾a, E. 1963. Animal speciesand evolution. Cambridge, Harvard Univ. Press. M^¾a, E. ^•D J. Bo•. 1943. Notes on the classification of the swallows, Hirundinidae. Ibis, 85: 334-341. M^¾a, E., E.G. L•s•¾, ^•D R. L. Us•c•R. 1953. Methods and principles of systematiczoology. New York, McGraw-Hill. McK•¾, F. 1961. An analysis of the displays of the European eider Somateria mollissima mollissima (Linnaeus) and the Pacific eider Somateria mollissima v. nigra Bonaparte. Behaviour Suppl. 7. McK•>;>;•¾,F. 1965. The comfortmovements of Anatidae. Behaviour,25: 120-220. M•¾•aaI•CXS, A.J. 1960. Comparative breedingbehavior of four speciesof North American herons. Publ. Nuttall Orn. Club, 2: 1-158. Moams, D. 1955. Birds' cleaningbehaviour. Cage Birds, 108: 257, 415, 418. Mom•Is, D. 1956. The function and causationof courtship ceremonies.Pp. 261- 286 in L'instinct dansle comportementdes animaux et de l'homme. Paris, Masson et Cie. Moams,D. 1957. "Typicalintensity" and its relationto the problemof ritualisation. Behaviour, 11: 1-12. MoYmm•, M. 1955a. Remarks on the original sourcesof displays. Auk, 72: 240-246. MoYma^•, M. 1955b. Some aspectsof reproductive behavior in the Black-headed Gull (Larus ridibundusridibundus L.) and related species.Behaviour Suppl. 4. Moyma^•, M. 1955c. Types of hostile display. Auk, 72: 247-259. MoYma^•, M. 1956. Notes on the behavior of some North American gulls. I. Aerial hostile behavior. Behaviour, 10: 126-178. MoYma^•, M. 1958a. Notes on the behavior of some North American gulls. II. Non-aerial hostile behavior of adults. Behaviour, 12: 95-182. Mo¾•a^•, M. 1958b. Notes on the behavior of someNorth American gulls. III. Pairing behavior. Behaviour, 12: 112-130. Mo•im•, M. 1959. Notes on the behavior of some North American gulls. IV. The ontogeny of hostile behavior and display patterns. Behaviour, 14: 214-239. 660 FICIrENAND FICIrEN, Avian Field Ethology [ Vol.Auk 83

MoYm>r^•, M. 1962. Hostile and sexualbehavior patterns of South Americanand Pacific Laridae. Behaviour Suppl., 8. NicE, M.M. 1943. Studiesin the life history of the Song Sparrow II. Trans. Linn. Soc. New York, 6: 1-329. PE^RSO•,O. P. 1953. Use of caves by hummingbirds and other speciesat high altitudes in Peru. Condor, $$: 17-20. Po•JLsE•,H. 1956. A study of anting behaviour in birds. Dansk Orn. Foren. Tids., $0: 267-298. ROWELS,C. H. F. 1961. Displacement grooming in the Chaffinch. Anim. Behav., 9: 38-63. R•JSS•LL,W. M. S., A. P. M•^•), ^m) J. S. H^¾•s. 1954. A basis for the quantitative study of the structure of behaviour. Behaviour, 6: 153-205. SA•J•)•}•s, A. A. 1951. A guide to bird songs. New York, Doubleday and Co. Sew•sTe•, P. 1961. A causal analysis of a displacement activity (fanning in Gasterosteusaculeatus L.). Behaviour Suppl., 9. Sm•e¾, C. G. 1955. Behavioral mimicry in the titmice (Paridae) and certain other birds. Wilson Bull., 67: 128-132. SIcx, H. 1959. Die Balz der Schmuckv6gel(Pipridae). J. f. Orn. 100: 269-302. Sru•ro•s, K. E.L. 1955. Studieson Great CrestedGrebes. Avic. Mag., 61.' 3-13, 93-146, 181-253,294-316. S•M•O•S, K. E. L. 1957a. A review of the anting-behaviour of passerinebirds. Brit. Birds, $0: 401-424. S•r•ro•s, K. E. L. 1957b. The taxonomic significance of the head-scratching methodsof birds. Ibis, 99: 178-181. S•M•O•S, K. E.L. 1959. Anting movements and their relationship to certain other behaviourpatterns. Ibis, 101: 368-373. S•M•O•S, K. E.L. 1961. Problemsof head-scratchingin birds. Ibis, 103a: 37-49. Stupors, K. E. L. 1963. Some behaviourcharacters of the babblers (Timaliidae). Avic. Mag. 69: 183-193. SPI•:r>r, H. T. 1958. Behavior and isolating mechanisms. Pp. 363-389 in Behavior and evolution (A. Roe and G. G. Simpson, eds.). New Haven, Yale Univ. Press. SPm•w^v, H., ^m) J. B. S. H^•.•)^•. 1953. The comparative ethology of vertebrate breathing. Behaviour, 6: 8-34. S:r•, R. C. 1958. The behavioral, ecological, and morphological characteristics of two populationsof the Alder FlycatcherEmpidonax traillii (Audubon). New York State Mus. and Sci. Serv., Bull. 371. S•m•, R.C. 1963. Isolating mechanismbetween populations of Traill's Flycatchers. Proc. Am. Philosoph.Soc., 107: 21-50. S:r•m•, H. 1955. Das Brutverhaltender Prachtfinken,Spermestidae, als Ausdruck ihres selbsfiindigenFamiliencharakters. Acta XI Congr. Internat. Ornithol., Basel: 350-355. S•roxEs, A. W. 1962. Agonistic behaviour among Blue Tits at a winter feeding station. Behaviour, 19: 118-138. Tao•E, W. 1951. The definition of terms used in animal behaviour studies. Bull. Anim. Behav., 9.' 34-40. T>rol•e•, W. H. 1956. Learning and instinct in animals. London, Methuen. T>rol•e•, W. H. 1961. Bird-song. Cambridge, Cambridge Univ. Press. Tao•es, W. H. 1963. Learning and instinct in animals. London, Methuen. T•m•lm•, N. 1939. The behavior of the Snow Bunting in spring. Trans. Linn. Soc. New York, $: 1-95. 1966Oct.] Fm•:E•Am) F•½•:EN, Avian Field Ethology 661

TIl•BERGEIg,N. 1951. The study of instinct. Oxford, Clarendon Press. T•BEROEN,N. 1952. Derived activities,their causation,biological significance, origin and emancipation during evolution. Quart. Rev. Biol., 27: 1-32. TIIqBEROEIq,N. 1953. The Herring Gull's world. London, Collins. TIIgBERGEIg,N. 1959. Comparative studies of the behaviour of gulls (Laridae): a progressreport. Behaviour, 15: 1-70. TINBERGEIg,N., AIgDA. C. PERDECK. 1951. On the stimulus situation releasing the begging responsein the newly hatched Herring Gull chick (Larus argentatus argentatusPont.). Behaviour,3: 1-39. VAUO•E•r,L. 1951. Ponte induite chez la perruche ondul6emaintenue /t l'obscurit6 et dans l'ambiance des voli•res. Comp. Rend. Acad. Sci., 232: 1706-1708. VERI'LAIqCIr,W. S. 1957. A glossaryof some terms used in the objective science of behavior. Amer. Psychol. Assn., Washington, D.C. WEEDE•, J. S., AIqDJ. B. FALLS. 1959. Differential responsesof male Ovenbirds to recordedsongs of neighboringand more distantindividuals. Auk, 76: 343-351. WH•TA•:Ea,L.M. 1957a. Lark Sparrowoiling its tarsi. Wilson Bull., 69: 179-180. WHITAKER,L. •f. 1957b. A r6sum6of anting, with particular referenceto a captive OrchardOriole. Wilson Bull., 69: 195-262. WlCKLER,W. 1961a. 0kologie und Stammesgeschichteyon Verhaltenweisen.Fort. der Zool., 13: 303-365. Wm•LEa, W. 1961b. t•ber die Stammesgeschichteund den taxonomischenWert einiger Verhaltensweisender ViSgel. Zeit. f. Tierpsych., 18: 320-342. Youno, It. 1949. Atypicalcopulatory behavior of a Robin. Auk, 66: 94. Departmentof Zoology,University of Maryland,College Park, Maryland.