Structural Changes in Song Ontogeny in the Swamp Sparrow Melospiza Georgiana

STRUCTURAL CHANGES IN SONG ONTOGENY IN THE SWAMP SPARROW MELOSPIZA GEORGIANA

PETER MARLER AND SUSAN PETERS RockefellerUniversity Field Research Center, Tyrrel Road,Millbrook, New York 12545USA

ABS?RAC?.--Sixteenmale Swamp Sparrowstrained with recordedsongs in infancywere recordedon a weekly basis throughoutthe processof song development.Analyses of the 15,000 songsrecorded from them provided the basis for a seven-stageclassification of the stepsin songontogeny, with the four major divisions of subsong,subplastic song, plastic song,and full song.The individuallyisolated males all followeda basicallysimilar developmental progressionthrough the seven stagesof song development.Accounts of other speciessuggest that the pattern seen in the Swamp Sparrowis a rather generalone, when one allowsfor speciesdifferences in songstructure. The abrupt reductionof songduration and its variability was a marker for the crystallizationof plasticsong into full song.At this time there was also a reduction in the number of syllabletypes used and the number of partsper song.Swamp Sparrows produced considerably more syllable types in plasticsong than necessaryto generatethe normal,species-specific song. Received 10 June 1981, accepted 19 October 1981.

A GENERALpicture is availableof the overall Four males,designated as the "prehatchtreated" structural properties of subsong and plastic group,were injected with CI628on 4 successivedays songand the transformationsto which they are between8 daysbefore hatching and two daysafter subjectprior to the emergenceof full song hatching. Three males, the "posthatch treated" (Nice 1943, Marlet 1956, Thorpe and Pilcher group, were injectedon 6 successivedays between 3 and 8 days after hatching.A third group of five 1958, Nottebohm 1972a). Despite the plethora males,the "late treated"group, were injectedon 4 of studies on mature birdsong, there has been to 6 successivedays, betweenthe agesof 9 and 24 little descriptive analysis of song ontogeny. daysof age.Four males were not treated at all. There The phenomenonof subsongis nevertheless were no lossesas a consequenceof the treatments, of greatpotential interest because of its distri- and no differencesin song could be detected be- bution in birds, which seems to coincide with tween treatmentgroups and controls. the incidence of song learning (Nottebohm All birds were trained for 40 daysbetween 24 June 1972a, 1975; Marlet and Peters in press a). No 1977 and 26 August 1977 startingbetween the ages one, however, has yet conducteda longitudi- of 16and 26 daysof ageand endingwhen the birds were between55 and 67 daysof age. The birds were nal study of songdevelopment in individual trained with one of two setsof tape-recordedsongs, birds to determinethe timing and manner of A and B, designedto exploresome of the species emergenceof learnedthemes. As a first step, differencesin song learning in Swamp and Song we presenthere the resultsof an investigation (Melospizamelodia) sparrows. The detailsof this ex- on the structureof subsongand plasticsong in perimentwill be presentedelsewhere, but we de- the Swamp Sparrow(Melospiza georgiana). scribethe training songsbriefly here. Wild male Swamp Sparrowsongs typically consist METHODS of identical repetitions of one distinctive syllable Sixteenmale SwampSparrows hatched in the wild type at a steadytempo. This patternand the rarely- between30 May 1977and 24 June1977 were brought occurring(<5%) two-partedtrill servedas the basis into the laboratorybetween 2 and 10 daysof ageand for our training songs (for examples of natural rearedby hand to independence.Younger nestlings Swamp Sparrowsong, see Marler and Peters1977, were brooded by canariesbetween feedingsuntil Peterset al. 1980). ScheduleA, used for training six about 10 days of age. The projecton songdevelop- males, contained 16 one-parted trills composed of ment was grafted onto anotherstudy of possibleef- either normal Swamp Sparrow syllablesor artificial fectsof an anti-estrogen,CI628, on the development syllablescreated from notesof Swampand Song of songlearning abilities. The experimentalmanip- sparrowsyllables (see Zoloth et al. 1980for proce- ulationsinvolved had no effecton singing behavior dural details). The remaining 10 maleswere trained and will not be consideredfurther in this paper be- with ScheduleB, which contained11 different songs, yond thesedetails. 6 one-partedtrills, and 5 two-parted trills. These 446 The Auk 99: 446-458.July 1982 JvL¾1982] SparrowSubsong and Plastic Song 447

TABLE1. Criteria for a seven-stateclassification of stagesof songdevelopment in the Swamp Sparrow.

Syllables Songs Stage Morphology Repetitions Morphology Duration Crystallizedsong I Stereotyped Clear trills Stableorder Short II Stereotyped Clear trills Variable order Short Plasticsong III Minor variations Clear trills Stableorder Longer IV Variable Clear trills Variable order Long and V Rudiments Some Variable order variable Subplasticsong VI Rudiments None Variableorder Variable Subsong VII None None None Variable

songswere composedof a total of 16 different intact tographed, sound-spectrographicanalysis of songs Songand Swampsparrow syllables. Songs for both with the frequencyspectrum displayed against time. scheduleswere organized into 3-min bouts by re- In early stageswe analyzeda 5-10-min samplefor peatinga singlesong type onceevery 10 s, a normal eachrecording date for eachbird. With later singing songdelivery rate for Swamp Sparrows.These were all edited material was analyzed. played to subjectstwice per day, moming and eve- For detailed analysisof the songs,we adopted a ning, during the periodsalready indicated. definition of a "song unit" suitable for all stages. After songtraining was completed,each male was Songsof adult Swamp Sparrowsin nature are about housed in a sound-proofchamber, with the lights 2.3 s in duration with a typical intersonginterval of kept on a normal photoperiod by a time-clock more than 5 s. The averageintersyllable interval for changedevery week. We first determined the peak full songis 40 ms and the averageinternote interval singing period during the day. A Crown Interna- is 12 ms. The timing of earlysinging behavior is very tional 800 seriestape deck was set up with a time different, however. Songsseparated by intervals of delayand resetsystem. Once per weekfrom 8 Au- more than 5 s often contain within them intervals gust 1977 to 2 September1977, we sampleda subset greater than 80 ms. Some of the note sequences of 3-7 birds from each training schedule, starting at bounded by 5-s intervalsare extremelyshort. After about 50 days of age and selecting1 out of every 5 someexperimentation, we adoptedthe followingcri- min for 24 h. Most singingoccurred just after dawn, teria for song-likeunits. Any sequenceof notes qual- when the lights were switched on in the chambers. ified as a song if it lastedmore than 0.5 s with no This information provided the basis for the recorded intervals of silence longer than 0.25 s between samplingof singing of all 16 males. Beginningat an notes.Thus, a 1.3-s sequenceof notesthat contained averageof 99 daysof age and continuinguntil about an interval of 0.3 s occurringafter the first 0.5 s would 380 days of age, we recordedeach male for 1 h once be analyzedas two songs.Applying thesecriteria to per week. TandbergSeries 15 tape recorderson time the output of all 16 subjects,we defined a total of clockswere used,with a tape,speedof 33/4inches per 15,062songs for all songstages. Song durations were second.From 593 recordingsessions for all birds, 235 measuredon a Summagraphicstablet and a PdP 11/ h of recordingswere obtained that containedsong. 10 DEC computer,programmed to registertime du- All recordingswere monitoredand then editedfor rations to the nearest 12.8 ms. analysis.The editing procedurediffered slightly for early and later stagesof singing.With early singing, we edited out up to 15 min of continuoussinging RESULTS and estimatedthe remaining amount in the 1-h re- PATTERNS OF SONG ONTOGENY cording. In later stages,song occurredin more dis- creteunits, and all songswere edited out, with two As a first step in the interpretationof song exceptions.Occasionally, recording quality was poor development, we createda seven-stageclassi- becauseof microphoneor tape-recordermalfdnction. fication, based primarily on the presenceor In a few of these casesa small sample of song was absenceof syllabicstructure and on the relative edited out to documentthe stageof singing. Late in stability of syllable morphology and the se- songdevelopment we occasionallyedited only alter- nate songs,when it was clear that a large sampleof quencesin which syllableswere delivered (Ta- the male's crystallizedsongs had already been ob- ble 1). Figure 1 representssamples of songpro- tained. duction from one male Swamp Sparrow, The edited tapeswere run on a PrincetonApplied illustrating the sevenstages in the emergence ResearchSpectrum Analyzer, which providesa pho- of one of this male'stwo crystallizedsyllable 448 MARLER AND PETERS [Auk, Vol. 99

Training syllables

2 3 4

Stage Subsong Stage Plastic song ',"h'l'11,t ' ,','1

syll'5 syll'6 syll3 Subplasticsong

syll 2 syll 4

,,!It I Izttt 1t 71/I ,'t

Stage Crystallizedsong

syll 2 .5 sec

Fig. 1. Songdevelopment in a SwampSparrow. Samples of eachof the sevenstages of developingsong in a male SwampSparrow that was trainedwith ScheduleB. The stageis indicatedat the left and major stagesabove. Imitations of thesix training syllables portrayed at thetop of thefigure are identified by number. Trainingsyllable 6 is a SongSparrow syllable; the remainderare from SwampSparrows. Syllables were presentedin one-partedsongs, except for syllables5 and6, whichappeared together in the same"hybrid" song.In nature,Swamp Sparrows typically have a repertoireof threeto foursong types, and, although there is somesyllable-sharing in local populations, most individual repertoires are unique. The experimental male shownhere had a final repertoireof two songtypes, shown in stages! and II. This crystallizedsong is virtuallyidentical to thesong of thelocal wild malethat served as the original source of thetraining syllable.

types. The upper row of the figure illustrates repertoire back to more rudimentary syllabic syllablesfrom someof the training songsim- forms. We also identified additional syllable itated by this male. It is convenientfor most typesnot presentin the maturerepertoire but purposesto collapsethis classificationinto four common at earlier stages. stages,crystallized song (stage 1), plasticsong In the very earliest stages,song units con- (stages2-4), subplasticsong (stages 5-6), and tained sequencesof undassifiablenotes, iden- subsong (stage 7). tified as subsong.Subplastic song is composed Using this classification,we inspectedthe of unidentifiablenote sequencesas well as syl- internal structure of each sound-spectro- lablesin a very rudimentaryform. Thus, it is graphed songunit. We developeda catalogue the stageat which we candetect a bird making of syllabletypes for eachmale by workingback his very first attemptsat recallingmemorized from crystallizedsong into earlier stages.The material. Becausethe syllable morphologyis organizationof syllabletypes within songswas only incipient,however, and becauseuniden- alsorecorded. By this procedurewe were able tiffablenote sequencespredominate, we have to tracesyllables present in a bird's crystallized combined subsongand subplasticsong for JULY1982] SparrowSubsong and Plastic Song 449

/1 •, ...... 235 255 275 295 315 335 355 375 395 Age (days) Fig. 2. Numbersof male SwampSparrows singing during weeklyrecording sessions from about95 to about405 daysof age. Note the midwinterbreak from 165 to 235 days.The 16 maleswere individually isolated in sound-insulated chambers. many of our analyses.For stageslater than a meansong duration of 2.53 s (SD = 1.82).For subplasticsong, eachsong unit was analyzed plastic song, 56 recordingsyielded 2,826 songs for the syllabletypes present, their order, and with a mean duration of 2.50 s (SD = 1.29). For the number of consecutiverepetitions within crystallizedsong, representedin 114 record- the phrase. ings,9,814 songs gave a meanduration of 1.96 This analysisprovided us with a basis for s (SD = 0.38). There are thus indications of a describingthe stagesof songdevelopment and decreasein songduration as developmentpro- relatingthem to otherparameters, such as age and the emergenceof imitations.

Numbersof birdssinging.--Figure 2, a graph 5.0- of the numbers of birds representedin eachof the weekly recordingsessions from 95 to 400 days of age, is one index of the incidenceof singing behavior at variousages under our experimental conditions. Note that the scale is broken between 165 and 235 days of age. Dur- 3.5 ing this intervalno singingwas recorded.Pre- vious to this there was sporadicsinging, but, 3.0 as can be seenfrom Fig. 2, only two or three birds were likely to be representedin any given week. This was alsotrue of singingprior to 95 daysof age.Because birds were not system- atically recorded prior to 95 days, we cannot make a quantitative statementon the early in- 1.5- cidenceof singing, but it was heard sporad- ically from all malesbeginning at 20-30 days I.O' of age. After about 240 days, the number of individuals representedbegan to climb, until, 0.5' by 280days of age,between 50 and 75% of the 0 maleswere representedin eachrecording ses- 50 I(•O I,•O 2•O 2•O 3•)O 3•O sion. For the next 100 days all of the 16 males Age (days) remained in full song until singing waned at Fig. 3. Changesin averagesong duration in de- about 400 days. veloping Swamp Sparrows.Vertical lines indicate Developmentalchanges in song duration.- standard deviations around the means. Song crys- Mean songlengths were calculatedfor eachof tallization occurred, on average, at 334 days of age. the majorstages of songdevelopment. For sub- The two final points representmean recordingsfor songand subplasticsong, represented in 64 of one late-seasonsinger, which happened to have the total of 234 recordings,2,422 songs yielded shortersongs than average. 450 MARtœRAtaD PœtœRS [Auk, Vol. 99

6.0- 6.0

4.0-

• 2.0-

0- ,* • 0 i;o a4o soo s;oo soo

-• 6.0- 6.0-

0 O3 4.0- 4.0-

2,0. 2.0'

a;,o soo 0 2•0 300 360 Onsetof plasticsong/ Onsetofcrystallized song ,•cJe(days) Fig. 4. Averagesong duration/recording day duringthe songdevelopment of four individuals.Vertical linesindicate standard deviations around the means.Onset dates of plasticand crystallizedsong are shown. Numbersof songssampled per day for thesefour birds rangedfrom 1 to 190. Averagesamples per bird were 67.1 + SD 52.1 songsper hour of recording. ceedsand a strikingreduction in the variability in the age at which singing began after the of song duration as crystallization occurs. mid-winter hiatus. Despite the variation in These trends are seen more clearly if song du- time of onset, eachbird went through a regular ration is simply plotted againstage (Fig. 3). progression.Crystallized song emerged at an There are signsof a drasticchange in songor- averageage of 334 days, precededby a period ganizationoccurring at about330 days of age, of plastic song, comprisingstages 2-4. Plastic with the mean duration dropping to a very sta- songwas in turn precededby subplasticsong ble value of 1.5-2.0 s. As shown below, this with an averageage of 299days for the change- suddenchange in the temporalcharacteristics over. The averageonset age of subplasticsong of songis accompaniedby a numberof other was 285 days. transformations. It was not uncommon for more than one Songduration data for individual malesre- stageto be recordedon a given day, although, vealeda changeequally dramatic to that in the oncecrystallized song began, a male rarely re- averageddata in song duration and its vari- verted to an earlier stage. Thus, in only 7 out ability. Mean songdurations during develop- of the 114 samplescontaining stage I crystal- ment for four individuals are shown in Fig. 4. lized song were any stage 2 or 3 songs repre- Mean songdurations for subsong,plastic song, sented. Crystallizationappears to constitutea and crystallizedsong were calculatedfor each stepwise change in development, tending to bird. Table 2 summarizes the results. Each of be irreversible for that reason. Thus, if one the 16 birds averagedlonger songs in both considersthe onsetdates for crystallizedsong subsongand in plasticsong than in crystallized indicated in Fig. 5, for 14 out of the 16 birds song, and for 12 birds thesedifferences were this was the first record of stage i song. significant.For two birds only plastic songs The onset of plastic song was designated as were significantly longer than crystallized the first occurrenceof stage4 and was the first song,and one bird onlyhad significantlylong- date on which the majority of songswere com- er subsongs(t-tests, see Table 3). posed of syllabletypes that could be reliably Theorder of stagesin song.--Asshown in Fig. identified. Songsof all classesoccurred during 5, there was considerableindividual variation the plastic song periods designatedin Fig. 5. JULY1982] SparrowSubsong and Plastic Song 451

TABLE2. Durationsof subsong,plastic, and crystallizedsong in 16 male SwampSparrows.

Songduration Subsong Plasticsong Crystallizedsong Bird n :• SD n • SD n • SD

i 103 2.48 2.17 210 2.81 1.53 585 2.21 0.26 2 213 2.42 1.84 219 1.84 0.79 418 1.52 0.21 3 227 2.43 1.66 52 3.42 2.63 396 2.10 0.24 4 184 2.77 1.79 118 3.71 1.78 379 2.28 0.26 5 44 2.39 1.49 140 2.18 1.07 622 1.90 0.19 6 198 2.47 1.86 149 2.54 1.62 949 1.85 0.22 7 156 3.34 2.93 53 3.40 1.44 479 1.88 0.14 8 125 2.49 0.99 421 2.21 0.53 474 1.95 0.17 9 204 2.66 1.86 56 2.33 1.24 535 1.78 0.13 10 156 2.43 1.71 375 2.33 0.77 754 2.11 0.20 11 353 1.99 1.21 199 2.19 1.11 203 1.86 0.33 12 64 3.14 1.75 181 2.22 1.10 1,032 2.19 0.17 13 41 2.52 1.35 366 2.51 0.92 1,054 1.91 0.18 14 138 2.17 1.17 59 2.28 1.09 1,032 2.08 0.29 15 128 2.47 1.64 24 3.78 1.56 377 1.76 0.17 16 88 3.80 2.39 204 3.32 1.84 525 1.89 0.24

The most commonwas plasticsong, occurring for n = 16). Likewise, the onset date of plastic in 42 of the 56 samplesrepresenting this stage. songis negatively correlatedwith the duration Stage6 was the criterionfor the onsetof sub- of the plastic song period (P ( 0.05, Spearman plastic song. Thirty-two recording dates were rank correlation coefficient rs = -0.535 for n = designated as subplastic song, and all con- 16). tained stage6 songs,23 containedstage 7, and Occasionally,though rarely, birds reverted 18 contained stage 5. All 32 recording dates from one developmentalstage to an earlier one. prior to the onsetof subplasticsong consisted The progressivenature of developmentis dis- almost entirely of stage 7 songs. Thirteen of playedmore clearlyin Fig. 6, which showsthe theseoccured in the prewinter singing period, and two of them containedsongs of one male that were classifiedas subplasticsong. All oth- TABLE3. Statisticalcomparisons of song durations er prewinter song was stage 7 subsong.Note in threestages of developmentin individualbirds. that the first hints of syllabicstructure appear Asterisksindicate significantdifferences (t-test). in subplasticsong. Thus, a considerableperiod elapsesbetween training and the explicit re- Subsongvs. Plasticsong vs. hearsalof learned song syllables,as analyzed crystallizedsong crystallizedsong in detail elsewhere (Marler and Peters in press Bird t P t P b). i 1.26 P > 0.2 5.65 P < 0.001' With the onset of spring song, 7 of the 16 2 7.10 P < 0.001' 5.89 P < 0.001' males sang stage 7 subsongexclusively. The 3 2.98 P < 0.01' 3.62 P < 0.001' first recordsof spring singing of the nine re- 4 3.69 P < 0.001' 8.70 P < 0.001' 5 2.18 P < 0.05* 3.09 P < 0.01' maining birds qualifiedas subplasticsong, in- 6 4.68 P < 0.001' 5.19 P < 0.001' cludedwith subsongin Fig. 4. The onsetdate 7 6.22 P < 0.001' 7.68 P < 0.001' of spring songwas significantlylater for these 8 6.07 P < 0.001' 9.63 P < 0.001' nine birds by an averageof about 2 weeks 9 6.75 P < 0.001' 3.32 P < 0.001' 10 2.33 P < 0.05* 5.44 P < 0.001' (Mann-Whitney U-test, P (0.05). There is 11 1.90 P > 0.05 4.02 P < 0.001' evidencethat the later a bird startsto sing, the 12 4.34 P < 0.001' 0.37 P > 0.2 more rapidly he develops.The onset date of 13 2.89 P < 0.01' 12.39 P < 0.001' subsongis negatively correlatedwith the du- 14 0.90 P > 0.2 1.41 P > 0.1 ration of the subsongperiod (P ( 0.05, Spear- 15 4.89 P < 0.001' 6.34 P < 0.001' 16 7.49 P < 0.001' 11.06 P < 0.001' man rank correlation coefficient rs = -0.453 452 MARLERAND PETERS [Auk, Vol. 99

Subsong Plastic Crystallized ( ...... ) song song (_ _ _) :272 ':299 :334

I I I 235 285 335 385 Age(days) Fig.5. A diagramof songdevelopment in the spring in 16male Swamp Sparrows. Birds are arranged with the earliestto startat the bottom.Subplastic song is includedwith subsong(dotted line). The mean agesfor theonset of subsong,plastic song and crystallized song are indicated by verticallines.

least advanced of the seven developmental cally startedduring the first week of April stagesrepresented on a given recordingdate. (medianstarting date 4 April), with the earliest Despite occasionalregressions through one on 10 March and the latest on 9 May. The first stage,each male tends to progressthrough the delivery of crystallizedsong was recordedon seriesin the same order. In only 6 out of 234 21 April. The lastbird crystallized36 dayslater, samplesdid a bird revertto an earlierstage of and the median date of onset was 10 May. development.The sevenstages represent a rel- These dates indicate that the singing programs atively smoothprogression, except for stages for thesecaged and individuallyisolated males 2 and 3 (Table 1). As can be seen in Fig. 6, did not depart too greatlyfrom thoseof wild malesvary in how they proceedfrom stage4 birds in the area. to stage1, and it maybe that stages2 and 3 are As can be seen in Fig. 5, a bird that starts alternativepathways to crystallization. springsubsong and subplasticsong early also Thus, despitethe greatvariability of the de- tends to be ahead with the transitions to plastic velopmentalstages of song, Swamp Sparrows songand crystallizedsong. The onsetof later kept in captivityunder theseconditions dis- developmentalstages is no more synchronous play a regularand repeatablepattern of ontog- with regardeither to ageor to timeof yearthan eny leadingup to the emergenceof full song. is the onsetof subsong.On the otherhand, the Songdevelopment in relationto seasonand duration of the subsongand subplasticsong age.--With the exceptionof one bird, all fall periods combinedwas negativelycorrelated singingrecorded (from 50 to 165 days) con- with duration of the plastic songperiod, hint- sistedof stage7 subsong.Although not tape- ing at a possiblecompensatory relationship recorded, subsongwas noted as early as 20 between these two developmental stages daysin somebirds. After the silentmid-winter (P < 0.05, Spearmanrank correlationcoeffi- period, vigoroussubsong and subplasticsong cient rs = -0.511 for n = 16). A searchfor pos- began,with a medianonset date of 9 March sible correlationsbetween age and develop- 1978. The earliest bird began on 14 February ment also revealed a negative relationship and the latest on 28 March. Plastic song typi- betweenhatching date and the agesof onset JULY1982] SparrowSubsong and Plastic Song 453

'• I Crystallizedsong

o 2 '• Plasticsong

.•- o Subplasticso

? Subsong '"

Age (days) Fig. 6. Songdevelopment in the SwampSparrow as revealedby plotting the leastadvanced stage representedin any given sample.Sixteen birds are represented.The categoriesof songdevelopment are as defined in Table 1. of three major stages of song development ture. We identified 199 different plastic song (rs= -0.672 for subsongand subplasticsong, syllabletypes for the 16 males. A comparison -0.737 for plastic song, -0.680 for crystallized with the training songsrevealed that 59 of song). Thus, the earlier in the seasona bird these were imitations (29.6%). The remaining was hatched, the later the age at which sub- 140 were classified as inventions or improvis- song or subplasticsong, plastic song, and full ations(70.3%). A detailedanalysis of thesewill songbegan. The differencesbecome amplified be presented elsewhere (Marler and Peters in in the process,so that one bird hatched24 days prep.). Evidently, nonimitated syllablesmake beforeanother started spring subsong or sub- up an appreciablepart of the plastic songrep- plastic song 65 days later. The functional sig- ertoire.Twelve sanga combinationof imitated nificanceof this correlateof early- and late-sea- and nonimitated syllables,and four sang the son hatching is unclear and merits further latter only. investigation. Despite this large syllable vocabularyem- ployed during plastic song, the males only CHANGES IN SONG STRUCTURE crystallizeda smallpart of their potentialrep- DURING ONTOGENY ertoire (22.6%). As a group, they crystallized Syllablestructure.--We have yet to attempt 45 syllabletypes, of which 19 (42%) were im- a comprehensiveanalysis of the structureof itations. Four of the 16 birds produced no im- subsong.As indicatedin the samplein Fig. 1, itationsin either plasticor full song, and two it is exceedinglyvariable, with a wide range of the 12 malesthat sangimitated syllablesin of frequenciesrepresented and virtually no re- plasticsong failed to includethem in their final peated elements. Only with the birds' attain- repertoire.No bird produceda syllablein crys- ment of stage 6 subplasticsong do we occa- tallized songthat was not representedin some sionallysee the first hints of syllabicstructure. form in plastic song.Excluding the four with- By stage4, mostnote sequencescould be iden- out imitations in plasticsong, there was some tified, and all plasticsong syllables were then tendencyto favor imitatedsyllables in the crys- classified on the basis of their acoustic struc- tallization process [39.6% in plastic song, 454 MARLERANy PETERS [Auk, Vol. 99

TABLE4. Numbers of syllabletypes used by male Swamp Sparrowsup to the time of songcrystallization.

Number of birds I 8 10 12 10 9 6 16 Averagesyllables per bird 13 12 11.6 10.1 7.3 8.3 5.8 3.1 Range 13 8-15 3-16 5-18 1-12 2-19 3-10 1-5 Standard deviation -- +2.4 +4.3 +3.8 +3.9 +3.9 +3.1 +1.3 Days from songcrystallization -49 -42 -35 -28 -21 -14 -7 0

57.6% in crystallizedsong, but the difference phrases. Figure 7 presentsthe mean number is not significant (X2 test: X2= 2.87, of phrasesper songfor all 16 birds, plotted in 0.05 < P < 0.1)]. relation to the day of song crystallization.In The 45 crystallizedsyllable types were rep- addition, the averageminimum and maximum resented in 35 songs,with an averagereper- number of phrasesare alsoindicated. It is clear toire of 2.2 songtypes per bird. Of thesecrys- that, just as the number of syllable types used tallized songs,25% of them were two-parted. by each bird decreasesin plastic song as the Wild males, with a repertoire of 3-4 typically time for crystallizationapproaches, so the same sing one-partedsongs. Two-parted songs do is true of the number of phrasesper song. In occurin nature, althoughrarely. Three out of early plastic song there may be as many as 58 (5%) of the songtypes recorded from 25 wild eight phrasesper song, with a maximum av- males were composedof more than one part. eraging 3.5, decreasingto an average of 1.2 In the laboratory-rearedbirds, one-parted with crystallization.Both phrasestructure and songsstill predominated,although less so than syllabletype number undergo progressivere- in nature. duction as mature song emerges. There is a decline in the number of syllable Further developmentaltrends.--One might typesuttered by a bird as it proceedsfrom the hypothesize that the complexity of mature first stagesof plastic song through to crystal- singing relates to the time spent in certain de- lization. The mean number of syllable types velopmental stages, such as plastic song. We per bird per recordingsession during plastic could find no correlation between the duration song was 9.5 (SD = 4.25:56 recording ses- of any phase and the number of plastic or crys- sions).In crystallizedsong, the mean number tallized syllable types developed. There is, of syllabletypes per bird was 2.9 (SD = 1.22: however, a positive correlation between the 114 recording sessions),a highly significant number of syllabletypes used in plasticsong difference(t-test, P < 0.001, t: 11.4). Early in and the number of songs crystallized in the plastic song the average number of syllable final repertoire (P < 0.05, Spearmanrank cor- typesper malewent up to 12-13, and onemale relation coefficient rs = 0.463 for n = 16). The peakedat 19 (Table4). It is thus clearthat the different training scheduleshad no effect on male Swamp Sparrows develop many more the number of plastic or crystallizedsyllable syllablesin plastic song than are needed to types, whether imitated or not. generatespecies-specific song. The processof To what extentdoes knowledge of the struc- developmentalattrition that accompaniesthe ture of plastic song permit prediction of the progressionthrough plastic song is discussed syllablesdestined to be crystallized?For each in more detail elsewhere (Marler and Peters in bird's plastic song recordingswe countedthe prep). number of songscontaining each syllabletype, Phrasestructure.--If we use the term "phrase" then groupedthe syllabletypes and combined for structuralunits within the song larger than the numbers of songsin which they occurred individual syllables, the typical crystallized under the following four categories:imitated song of a male Swamp Sparrow consistsof a and nonimitated syllables,syllables eventually single phrase (e.g. Fig. 1). Although two- included in the final repertoire, and syllables phrase songsdo occurboth in nature and in excludedfrom the final repertoire. For an ab- our captive birds, songswith more than two solutemeasure of syllableusage, we calculated phraseshave never been recorded from Swamp the averagenumber of songsper hour in which Sparrows after crystallization. By contrast, a copy or noncopysyllable and a final reper- plasticsong often incorporatesmore than two toire syllable or a discontinuedsyllable oc- JuLY1982] SparrowSubsong and Plastic Song 455

X•x ß • phrases/songfor16 birds • x• maxandmin for 16 birds X•x

ß • \ X•X'""'"' X X X X

Daysfrom onsetof crystallizedsong Fig. 7. Developmentalchanges in SwampSparrow song structure. The meannumbers of phrasesper songare plotted in relationto the dayof songcrystallization. Average maxima and minima are also shown. curredin eachof the recordingdates during ther division of ontogeneticstages of songde- the plasticsong period. The resultswere asfol- velopment,in which the term "subsong"was lows. restricted to the very earliest stages,as with For the groupof 16 birds,the averagenum- Nice's "warbling" definition, and the term ber of songsper hour in which an imitated "plastic song" was coined for a more highly syllableoccurred during plastic song was 9.3 structured, intermediate stage, distinct both (SD= 10.6)and for a nonimitatedsyllable 7.4 from subsongand from the fully crystallized (SD= 4.7). We had39 plasticsong recordings adult song (Marler 1956).This classification,or for the 12 birds that used imitations. An imi- othersequivalent to it, has been applied and tated syllableoccurred in an averageof 13.3 extendedby others (e.g. Lanyon 1960; Notte- (SD = 10.4) plastic songsand a nonimitated bohm 1968, 1972a, b, 1975). To judge from syllablein 7.2 (SD = 4.9), a significantdiffer- published descriptions,it is applicable to a ence (t-test, P • 0.01). A comparisonof the wide variety of songbirds(e.g. Lanyon 1957, averagefrequency of occurrenceof a final rep- 1960; Poulsen 1959; Armstrong 1963; Lemon ertoire syllable (12.5, SD = 10.8) and a discard- and Scott 1966; Kroodsma 1974). We have ed syllable(7.6, SD = 5.3) in plasticsong re- found it instructiveto add a further stageof vealed that the birds sang those syllables subplasticsong, transitionalbetween subsong destinedfor the final repertoiresignificantly and plasticsong. It has many of the qualities more often (t-test, P • 0.01). of subsongdiscussed below but also contains some suggestionsof syllabic structure. DISCUSSION Given the great variation in the structureand complexityof the full, primary song of differ- The term subsongseems to have originated ent species,there appearsto be a surprising with Nicholson (1927, Nicholson and Koch degree of conformity to a general pattern of 1936). Its use was extended by various other development.Singing tends to begin with an ornithologists(e.g. Lister 1953), but Thorpe amorphous,highly variable and unstructured was mostresponsible for formalizinguse of the subsong. term and defining the basicproperties of sub- The subsong of the Swamp Sparrow illus- song (Thorpe 1955, Thorpe and Pilcher 1958). trates several of the general features outlined An apparently equivalent term, "warbling," by Thorpe and Pilcher(1958) for the subsong was coined by Nice (1943) but has been dis- of thrushes,finches, and some other species. placedby the earlierterm. Analyses of subsong The differencesbetween subsong and full song in the Chaffinch (Fringillacoelebs) led to a fur- that recur include the volume, with subsong 456 MAr{LEt{A•qt• PETEr{S [Auk, Vol. 99 quieter than full song; patterning, with sub- the Red-winged Blackbird (Agelaiusphoeni- song having a very different structure; dura- ceus) (Marler 1970, Marler et al. 1972), indicat- tion, which tendsto be longer in subsong;fre- ing that this may be a rather generalphenom- quency range, inclined to be greater in enon. subsong;and seasonaltiming, with subsong There are indications that the classification occurringearlier in the season. of ontogeneticstages of Swamp Sparrow song There are some striking parallels between presentedhere may be applicableto a variety the Swamp Sparrowand the Song Sparrowas of other songbirds,with adjustmentsfor dif- reported by Nice (1943). She listed five stages, ferencesin species-specificsong structure. De- beginning with "continuous warbling" and spite these similarities, it would probably be culminating in "adult song." After stage 1, a mistaketo concludethat subsongand plastic continuous warbling, stage 2 included some song serve a similar function in all songbirds. shortsongs and muchwarbling, with intervals Although we can do little more than speculate between songs much shorter than the song at this stage,it seemslikely that there are mul- length. Stage 3 included some warbling but tiple functions even within a species. consisted predominantly of short songs, not One further finding in the Swamp Sparrow yet crystallizedinto adult form but with inter- that complicatesfunctional interpretation con- val lengthsabout equal to thoseof songs.Stage cerns the occurrenceof plastic song in older 4 songs were practicallyadult in form, with birds. In four males we followed the process intervals usually twice as long as the songs of song development through a secondyear. themselves.This was describedas a period of There was virtually no subsongor subplastic trying out songs,adding some, and deleting song, but plastic song did occur, though for a others. With Stage 5, fully adult song was shorter period than in the first year. Without achieved, stereotyped except for some varia- exception, all of the second-yearplastic song tion in endings, restricted to the final reper- syllablescould be matched from the first year toire. of that individual. Fewer were used, however. It is notable that variationsin songduration The four birds used 51 syllabletypes in first- servedas usefulmarkers for the five SongSpar- year plastic song and only 25 in the second row stages. We have shown that an abrupt year. Although there were no additions, some decreasein song duration and a dramatic con- shifts of emphasis occurred,favoring different traction of the variability are striking accom- syllablesfrom those in year 1. The four birds panimentsof the processof songcrystallization crystallized the same 12 syllable types as in in the Swamp Sparrow, coinciding with the year 1, all well rehearsedin plastic song. The first emergenceof the full song. 13 plastic song syllables that were rejected in Nice also described in the Song Sparrow a second-yearcrystallization had also been re- processthat may correspondto the syllableat- jected in the first year: trition we have describedin the Swamp Spar- Thus, in Swamp Sparrowsthe excessof syl- row in its progressionthrough the plasticsong lable production does not provide a bank for period. She describesyoung malesas they es- generating new songs in the future. In fact, tablish territories replying to rivals with sim- Swamp Sparrow song remains virtually un- ilar songs, even though many of these were changed from year to year. In our 16 captive subsequentlydropped from the repertoire. She males,for example,31 of the 33 songtypes pro- concludedthat "each young song sparrowhas duced in the first year persistedunchanged in a large fund of potential songs,as is clearfrom the second. One song was dropped, and in listening to the rambling warblings of juve- another some high frequency components niles." She suggeststhat what may appear to were lost, as though perhapsfrom a syringeal be a processof imitation occurringduring early abnormality. Indications from field data are territorial establishment may actually be "a that no new songsare added to the adult rep- calling forth of songsalready in the potential ertoire after the first year (D. Kroodsmaand R. repertoire, most of them being later lost Pickert, in litt.). throughdisuse" (Nice 1943:139).A lossof song In species that change the song repertoire material toward the end of ontogenyhas been from year to year, such as the canary (Serinus noted in other species, such as the White- canarius) (Marler et al. 1973, Nottebohm and crowned Sparrow (Zonotrichialeucophrys) and Nottebohm 1978), the Red-winged Blackbird JvI•¾1982] SparrowSubsong and Plastic Song 457

(Marler et al. 1972, Yasukawa et al. 1980), the cation (W. E. Lanyon and W. N. Tavolga, Eds.). Saddleback (Philesturnuscarunculatus) (Jenkins Washington, D.C., Amer. Inst. Biol. Sci. 1977), the Indigo Bunting (Passerinacyanea) LEMON, R. E., & D. M. SCOTT. 1966. On the development of song in young Cardinals. Can. J. (Riceand Thompson1968), and the indigo bird Zool. 44: 191-199. [Vidua(= Hypocherina)spp.] (Payne 1973),an LISTERßM. D. 1953. Secondarysong; a tentativeclas- excess of plastic song material, if it occurs, sification. Brit. Birds 46: 139-143. might provide a sourcefor additions to the full MARLER, P. 1956. Behaviour of the chaffinch. Be- song repertoire in subsequent years. This is haviour Suppl. 6: 1-186. one of many questionsraised by this studythat --. 1970. A comparativeapproach to vocal learn- deservesfurther investigation. ing: song developmentin White-crownedSpar- Perhaps the most surprising result is the rows. J. Comp. Physiol. Psych. 71: (2, Part 2) high incidenceof songinvention in our captive 1-25. ß 1981. Birdsong:the acquisition of a learned Swamp Sparrows. This occurred despite the motor skill. Trends in NeuroSciences 4: 88-94. large set of training songs used, though it is ß & S. PETERS.1977. Selective vocal learning hard to be sure what would constitute a surfeit in a sparrow. Science198: 519-521. of acceptablemodels. A comparisonwith the , & --.. In press a. Subsong and plastic simple syllables of innate Swamp Sparrow song:their role in the vocallearning process.In songs(Marler 1981) suggeststhat exposureto Ecology and evolution of acousticcommunica- song actually stimulatesvocal invention in this tion in birds (D. Kroodsma and T. Millerß Eds.). species. It remains to be determined whether BloomingtonßIndiana Univ. Press. the use of live tutors rather than tape record- ß & --. In press b. Long-term storage of learned birdsongsprior to production. Anim. ings affectsthe amount of songinvention. Un- Behav. like some emberizines, such as the White- ß M. KONISHI, A. LUTJEN, (g• M. S. WASER. crowned Sparrow, local Swamp Sparrow pop- 1973. Effects of continuous noise on avian hear- ulations exhibit considerableheterogeneity in ing and vocal development.Proc. Natl. Acad. their songs, with a high incidence of rare song Sci. USA 70: 1393-1396. types. A high invention rate is one means by ß P. MUNDINGER,M. S. WASER, (g•A. LUTJEN. which male Swamp Sparrows may generate 1972. Effects of acoustical stimulation and de- this trend toward song diversity in local pop- privation on song development in the Red- ulations. winged Blackbird (Agelaiusphoeniceus). Anim. Behav. 20: 586-606.

ACKNOWLEDGMENTS NICEßM. 1943. Studiesin the life history of the song sparrow. II. Trans. Linnaean Soc. New York, This research was supported by PHS research Vol. VI. grant S07 RR-076512to the RockefellerUniversity NICHOLSON, E.M. 1927. How birds live. London, and MH 14651 to Peter Marler. Valuable contribu- Williams and Norgate. tions to the projectwere made by P. PriceßC. Marler, ß & L. KOCH.1936. Songsof wild birds. Lon- J. Marler, M. Searcy, and G. Sherman. L. Baptista don, H. F. & G. Witherby. and D. Heinemann made useful comments on the NOTTEBOHM,F. 1968. Auditory experienceand song manuscript. development in the chaffinch Fringilla coelebs. Ibis 110: 549-568. ß 1972a. Neural lateralization of vocal control LITERATURE CITED in a passerinebird. II. Subsong,calls and a the- ARMSTRONC,E. A. 1963. A study of bird songßLon- ory of vocal learning. J. Exp. Zool. 179: 35-49. don, Oxford Univß Press. --. 1972b. The origins of vocal learning. Amer. JENKINSßP. F. 1977ßCultural transmission of song Natur. 106: 116-140. patternsand dialectdevelopment in a free-living ß 1975. Vocal behavior in birds. Pp. 287-332 bird population. Anim. Behav. 25: 50-78. in Avian biology,vol. 5 (D. Farner, Ed.). New KROODSMA,D. E. 1974ßSong learning, dialectsßand York, Academic Press. dispersalin the Bewick'swren. Z. Tierpsychol. ß & M. NOTTEBOHM.1978. Relationship be- 35: 352-380ß tween song repertoire and age in the canary, LANYON,W. E. 1957ßThe comparativebiology of the Serinuscanarius. Z. Tierpsychol.46: 298-305. meadowlarks (Sturnella) in Wisconsinß Publ. PAYNEßg. 1973. Behaviorßmimetic songsand song Nuttall Ornithol. Club No. 1: 1-67ß dialectsßand relationshipsof the parasitic indi- --. 1960. The ontogeny of vocalizations in birds. gobirds(Vidua) of Africa. Ornithol. Monogr. 11: Pp. 321-347 in Animal sounds and communi- 1-33. 458 MARLERAND PETERS [Auk, Vol. 99

PETERS, S., W. A. SEARCY, & P. MARLER. 1980. , & P.M. PILCHER.1958. The nature and char- Species song discrimination in choice experi- acteristicsof subsong.Brit. Birds 51: 509-514. ments with territorial male Swamp and Song YASUKAWA,K., J. L. BLANK,C. B. PETERSON.1980. sparrows. Anim. Behav. 28: 393-404. Songrepertoires and sexualselection in the Red- POUrSEN,H. 1959. Song-learningin the domestic winged Blackbird. Behav. Ecol. Sociobiol. 7: canary.Z. Tierpsychol.16: 173-178. 233-238. RtCE,J. O., & W. L. THOMPSON.1968. Song devel- ZOrOTH,S., R. J. DOOrING,R. MIrLER, & S. PETERS. opmentin the Indigo Bunting. Anim. Behav.16: 1980. A minicomputersystem for the synthesis 462-469. of animal vocalizations.Z. Tierpsychol.54: 151- THORPE,W. H. 1955. Comments on "The bird fan- 162. cyer'sdelight," togetherwith noteson imitation in the subsongof the chaffinch.Ibis 97: 247-251.

The Board of Directors of the Hawk Mountain SanctuaryAssociation announcesits fifth annual award for raptor research.To apply for the $500.00 annual award, students should submit a description of their researchprogram, a curriculumvitae, and two lettersof recommendationby 31 October 1982to: Mr. JamesJ. Brett, Curator, Hawk Mountain Sanctuary Association, Route 2, Kempton, PA 19529. The final decision by the Board of Directors will be made in February, 1983. Only studentsenrolled in a degree granting institution are eligible. Both undergraduateand graduate studentsare invited to apply. Projectswill be picked completelyon the basisof their potentialcontribution to improve understandingof raptor biology and their ultimate relevanceto conservationof North American hawk populations.

The North American Loon Fund (NALF) announcesthe availability of two grant programs for support of new or currentresearch, management, or educationprojects that may yield usefulinformation for Common Loon conservationin North America. The first of theseprograms, the Robert J. LurtsemaResearch Award, consistsof a $1,000 stipend available annually for a suitable researchproject focusedon a member of the Family Gaviidae. Preferencewill be given to studentsand independent researcherswith limited availability of other funding. The secondprogram offers modest grants in supportof research,management, or educa- tional projectsdirectly related to the conservationof Common Loons as a breeding species.Proposals in the range of $500 to $3,000 are most likely to be consideredfor funding. Further guidelines for prospective applicants are available upon request from the NALF Grants Com- mittee. Deadline for submissionof proposalsis 31 January1983. Funding awardswill be announcedby 15 March 1983.Please submit guideline requests to: North AmericanLoon Fund GrantsCommittee, North American Loon Fund, Meredith, NH 03253.

The Center for Field Researchannounces grants ranging from $5,000 to $50,000 (average$16,000) for field researchprojects in the humanities and sciences,anywhere in the world. Eligibility: postdoctoral(or the equivalent)scholars of any nationality; researchteams must employ qualified membersof EARTHWATCH, whose contributions (av. $500 per participant) provide the funds. Letters of intent should be submitted as early as possible. Full proposalsare invited upon favorable review of preliminary proposals,and are due no later than 9 months before project date. Decisionsare made after both internal and externalpeer review. For information and applicationscontact The Center for Field Research,10 Juniper Road, Box 127, Belmont, MA 02178. (617) 489-3032.