- BULLETIN of the FLORIDA STATE MUSEUM Biological Sciences Volume 22 1977 Number 2 GEOGRAPHIC VARIATION IN THE SONG OF BELDING'S SAVANNAH SPARROW (PASSERCULUS SANDWICHENSIS BELDING.4 RICHARD ALAN BRADLEY e -. , I . .r - .* r S ., 4* f S 2 / a . 1. 1 UNIVERSITY OF FLORIDA GAINESVILLE Numbers of the BULLETIN OF THE FLORIDA STATE MUSEUM, BIOLOGICAL SCIENCES, are published at irregular intervals. Volumes contain about 300 pages and are not necessarily completed in any one calendar year. OLIVER L. AUSTIN, JR., Editor RHODA J. RYBAK, Managing Editor Consultants for this issue: Luis F. BAPTISTA JOHN WILLIAM HARDY j Communications concerning purchase or exchange of the publications and all manuscripts should be addressed to: Managing Editor, Bulletin; Florida State Museum; University of Flor- ida; Gainesville, Florida 32611. This public document was promulgated at an annual cost of $1,667.26 or $1.667 per copy. It makes available to libraries, scholars, and all interested persons the results of researches in the natural sciences, emphasizing the circum-Caribbean region. Publication date: June 17, 1977 Price: $1.70 GEOGRAPHIC VARIATION IN THE SONG OF BELDING'S SAVANNAH SPARROW (PASSERCULUS SANDWICHENSIS BELDINGI) RICHARD ALAN BRADLEY' SYNopsis: The song of Belding's Savannah Sparrow, resident from Santa Barbara, California, south to El Rosario, Baja California del Norte, was studied during the spring of 1973. Songs were recorded from individuals at 14 of the 15 remaining breeding localities of this subspecies and analyzed with an audiospectrograph. Recordings of color-banded birds indicated that each male sang a single highly stereotyped song pattern. Detailed analysis of intrapopulational and inter- populational variation was made. The study of interpopulational variati6n yielded a mosaic pattern of song "dialects." The occurrence of such song dialects has been substantiated else- where for a variety of species. The present study involves very small adjacent, but isolated pop- ulations. TABLE OF CONTENTS INTRODUCTION . 58 ACKNOWLEDGEMENTS. 59 METHODS 59 Recordings 59 Analysis..... 61 Terminology . 61 REsuLTs ... 61 Call Repertoire ....._ 61 Features of Song Delivery The Structure of the Primary Song.. Ontogeny of the Song . Individual Variation Intrapopulational Variation Interpopulational Variation Summary of Interpopulational Variation by Location . @ ~ S E R R ~ & 65DISCUSSION. LITERATURE CITED.. The author is an Associate in Natural Sciences at the Florida State Museum, University of Florida, Gainesville 32611. This paper was submitted in partial fulfillment for the degree of Master of Arts at California State University, Long Beach 90804 Manuscript accepted 18 December 1975. BRADLEY, RtcHARD ALAN. 1977. Geographic variation in the song of Belding's Savannah Sparrow (Passerculus sandwichensis beldingi). Bull. Florida State Mus., Biol. Sci. 22(2):57-100. 58 BULLETIN FLORIDA STATE MUSEUM Vol. 22, No. 2 INTRODUCTION Geographic variation in bird song has been the subject of recent studies and has taken several forms. The early emphasis was on documentation of variation within species having wide distributions. More recently compre- hensive studies have been conducted using recordings from many localities or analyzing variation between local contiguous or isolated populations. Per- haps the best summary of the work on geographic variation in bird song is by Thieleke (1969). Early work on the Chaffinch (Fringilla coelebs) and other species revealed a distinctive mosaic pattern of variation (Thieleke 1969). Song patterns used by most of the birds in one place are similar to their neighbors' songs, while individuals from other populations have different themes. Often an entire system of such variants is found when several different populations are studied. The term "song dialect" refers to this type of variation. Many authors discussed the evolutionary significance of song dialects. Mayr (1963) stated that geographic variation of song is a widespread isolating mechanism in birds. Other workers expressed similar ideas about song as an isolating mech- anism (Thorpe 1961, Marler and Tamura 1964, Lanyon 1969, Nottebohm 1969, Thieleke 1969, Armstrong 1973). Nottebohm (1972) summarized the proposed mechanisms of such isola- tion: "It is necessary to assume that female birds develop a pref- erence for the song dialect of the area where they are born, and that this preference is revealed in their choice of partner. Males in turn must learn to sing the dialect of their birth area. If birds of both sexes migrate or wander between fledgling age and their first breeding season, they must return to breed in the same gen- eral area where they were raised. In this fashion an assortative mating system based on song preferences reinforces a more or less loose philopatry." The Savannah Sparrow (Passerculus sandwichensis) is one of the most widely distributed of North American sparrows. The species breeds from 71°N (Barrow, Alaska) south to 16°N (Hacienda Chancol, Guatemala) and is divided into 18 subspecies that vary considerably in size, proportions, and coloration (American Ornithologists' Union Check-list 1957, supplement 1973). Of the 18 subspecies 8 are restricted to the coastal marshes of California and northwestern Mexico. Van Rossem (1947) presented an excellent review of the seven races that breed in western Mexico. The Savannah Sparrow form that breeds along the coast from Santa Bar- bara south to El Rosario was originally named as a full species, Passerculus beldingi, by Ridgway (1885). This race, now known as Passerculus sand- 1977 BRADLEY: BELDING'S SAVANNAH SPARROW 59 wichensis beldingi (American Ornithologists' Union Check-list 1957) is re- ferred to here as Belding's Sparrow. The name Savannah Sparrow is used when referring to the species in general. Belding's Sparrow is restricted to the tidal salt marshes of southern Cali- fornia and northwestern Baja California. In a preliminary census conducted in conjunction with this study, I estimated a population of fewer than 3,000 singing males (Bradley 1973). The restriction of this subspecies to local patches of salt marsh has produced small isolated populations ideal for the study of geographic variation. The pattern of song variation within the spe- cies, as outlined by Borror (196la) is well suited to the development of song dialects: (1) each bird sings only one song pattern, (2) variation occurs between different individuals, and (3) song patterns used in any particular location show considerable overlap. Because of restrictions of distance and time, recordings could be made of only a limited number of populations. As a result, a somewhat arbitrary de- cision was made to restrict the study to the coastal populations now assigned to P. s. beldingi. Recordings from another coastal subspecies (P. s. alaudinus) in central and northern California indicate a continuum of similar patterns of variation in that subspecies, and perhaps throughout the species. This study was conducted primarily in the spring of 1973, although some additional ob- servations and recordings were made later in 1973 and early in 1974. Tape recordings were made of 280 different individuals at 14 of the 15 known re- maining breeding sites of Belding's Sparrow. No recordings were made at Ensenada Bay, Baja California del Norte, because of time limitations and the small, widely-scattered nature of that population. ACKNOWLEDGEMENTS I thank David R. Bontrager, Marjorie J. Bradley, John William Hardy, and Graeyer Mans- field-Jones for reading all or parts of the manuscript and making many helpful suggestions. I give special thanks to Charles T. Collins and Stuart L. Warter for guiding my research as well as suggesting corrections and improvements of this paper. In addition to the above named per- sons, many others helped me in the field, or obtained permission for my studies in restricted areas. For such help I express my appreciation to Commander Robert Baker, Frank DeVore, L. E. Fellows, Barbara Massey, Larry Pomeroy, Jene St. Germain, Deane Swickard, Margaret Tay- lor, and Bruce Wales. I also thank David Bradley for preparation of Figure 16, assistance in the field, and innumerable hours spent conducting computer analysis of the data. METHODS RECORD]NGS Most of the recordings in this study were made with a Uher 4000 Report S tape recorder and a Uher M 512 microphone mounted on either a 61 cm or a 76 cm fiberglass parabolic reflector. Some additional recordings were made with a Sony TC 800 recorder and a Sony ECM 19 B mi- crophone mounted on the 61 cm reflector. All recordings were made at a tape speed of 19 cm/s. Recordings were made in the field with a hand-held reflector during the peak period of song activity in the morning. Copies of the recordings are on file in the Bioacoustic Archive of the Florida State Museum, Gainesville. 60 BULLETIN FLORIDA STATE MUSEUM Vol. 22, No. 2 Belding's Sparrow is a bird of the coastal salt marshes of southern California and is largely , restricted to Salicorniq-covered tidal flats..Each singing individual, presumed to be a male, de- fends an exclusive territory -in this habitat and regularly sings from one or more exposed perches within his territory. When I surveyed a particular location, I made recordings beginning with an individual at one edge of the marsh and continued in a regular pattern from territory to territory until the entire nesting area had been traversed. Usually I recorded 5 t6 10 songs from each individual. As males of this subspecies defend very small territories (Bradley 1973),each singing bird could be iden- tified by its song perch location. Great care was taken to record songs from distinct individuals. In those cases where a large number of birds were singing from a relatively small area, only a few indivjduals were recorded if any probability of confusion existed. With this method only small samples could be obtained from some sites, but these samples are free from duplication. At some locations the samples were recorded on two or more dates. In each case additional recordings for the sample were made in parts of the marsh where birds had not been previously recofded. A list of locations visited and sample sizes obtained is presented in Table l. Information on the approximate population size at each location is presented elsewhere (Bradley 1973). TABLE 1.-LOCATION DATA, Sample Letter Number Location and City . Size Code 0 0 8 @ 9 8 @ 2 8 2 2 Goleta Slough, Santa Barbara 7 GS El Estero, Carpinteria 28 EE Mugu Lagoon, Point Mugu 28 PM Playa del Rey, Los Angeles 4 PR Anaheim Bay, Seal Beach 61 AB Huntington Harbour 5 HH Bolsa Chica Lagoon, Orange County 0 BC Upper Newport Bay, Newport Beach 30 NB Santa Margarita Lagoon, Oceanside 29 SM Agua Hedionda Lagoon, Carlsbad 10 AH San Elijo Lagoon, Cardiff 4 SE 12 Los Penasquitos Estuary, Del Mar 21 LP 13 Imperial Beach, Imperial Beach 8 IB 14 Ensenada Bay, Ensenada 0 EB 15 San Quintin Bay, San Quintin 34 SQ 16 Laguna El Rosario, El Rosario 11 ER Playback of recorded song was successful in elieiting close approach and song in several in- dividuals tested at Anaheim Bay. Although the song of each individual proved to be constant, no playback was used during the sampling portion of this study to avoid any possible influence on the songs to be recorded. To determine the amount of individual variation in song, and to verify information about the territoriality of males of this form, a number of individuals were color-banded. Birds were cap- tured in mist nets set across the surface of the marsh in the evening. In addition to a Fish and Wildlife Service numbered aluminum band, three colored plastic bands were used on each bird to produce individually recognizable combinations. Songs of some additional known, but not color-banded individuals were recorded in the study period. These unbanded individuals were chosen because they occupied solitary or readily identifiable territories where they could be found repeatedly. All work on individual variation was carried out at the main study location at Anaheim Bay. This site was relatively free from outside interference. 1977 BRADLEY: BELDING'S SAVANNAH SPARROW 61 ANALYSIS One or more songs of each individual recorded were analyzed on a Kay Elemetrics Corp. Sona-Graph, model 7029 A, employing the 160-16, 000 Hz scale, HS (high-shape) equalization, and narrow band filter. Each audiospectrogram was then analyzed in detail. Each song was transcribed into an alphabetic sequence and every distinguishable note given a paired code of two letters, A total of 86 distinct note types were named from recordings made during this study (Figs. 1 and 2). The first letter Of the paired code (upper case) designates the general note type (e.g. B notes are usually high-pitched introductory notes). The second letter (lower case) indicates the specific variant of that particular note t*e. When one note is repeated two to several times in sequence, a number follows the two-letter code indicating the number of repe- titions (e.g. the note Ca repeated three times would be coded as Ca3). A complex string of letters and numbers is then constructed asa representation of the specific song pattern of an individual. For example, the song pictured in Figure 3 would be coded as AaBb4PaDaDbDcKaFdFe~a. In a few cases a phrase (composed of many similar notes or syllables) was named with only one pair of letters. Sequences so named are rapidly modulated buzzes or buzz phrases. It would be diffi- cult to count each modulation in a buzz, and I believe that enumerating each as a separate note would be meaningless. Differences in the rate of modulation and tonal quality of the buzz phrases are indicated by the different lettered codes assigned to them. Recordings of the songs from several known individuals were analyzed for diurnal and sea- sonal variation. Finally, comparisons were. made of inter- and intrapopulational variation in the songs. TERM[NOLOGY The nomenclature of bird song is nearly as variable as the number of studies conducted. In this paper I have tried to use universally recognizable terminology, although in some cases it may differ considerably from some other published works. SONG,-The term song is here applied to the basic species-specific or "primary" song as dis- tinguished from the other call notes of a species. SONG BOWT.-A series of songs given in sequence at a more or less regular rate, ending with a much longer pause or with the cessation of song. NoTE.-A note has been almost universally defined as the smallest unit of bird song. It is rep- resented on an audiospectrogram as a continuous trace. Schwartz (1972) gave a similar defini- tion for the term "figure." PHRASE.-A group of notes that combine to form a recognizable unit is termed a phrase (the "note complex" of some authors). The individual notes that combine to form a phrase were termed syllables by Marler and Tamura (1962). Buzz.-A phrase made up of many similar notes uttered in rapid succession is a buzz. Borror (1965) used the term buzz for sounds consisting of notes repeated at a rate greater than 40 per second, or a single note that fluctuates at this rate. It should be noted that audiospectrograph analysis yields little information about the nature of the modulation in a buzz. It is nearly im- possible to tell if a sound is frequency or amplitude modulated (Greenewalt 1968). For an accu- rate determination of the nature of the modulation within a buzz or other complex note, one should use an oscilloscope as Greenewalt (1968) demonstrated. SONG PATTERN.-The particular sequence of notes and phrases that comprise the song of an individual is considered as one song pattern (equals "song type" or "theme" of other authors). In some cases several individuals may sing songs of the same song pattern. RESULTS CALL REPERTOIRE In addition to the primary song of the species, the chief subject of this study, the Savannah Sparrow has a number of other typical vocalizations. Gobiel (1970) described the various calls in this species' repertoire. I heard 16-- e2 B U LLE TIN FLO R ID A S TATE M U SEU M Vol. 22, N o. 2 12- 8- **4- 4~4 -0 -1 .,4 *==4 -1 9~ -0141: 4- Aa Rb Bc Be Bf Bg Sh N Bk Bl am en Ca Cb Cd Ce 111111111111111 16- 12- 8- S ; 1 d f 1 1 44 4 * 7 4* =~ 01 T , < -1 4- 4 r 7 14- C, Cg Ch C' CA Cli Cl ~ Da Db Dc Di Di Ea Eb Ec Fa Fb Fc Fd Fe Ff z Ga ~ Gb Gc 16- 12- 4 -4 .r-4- 1.-I .--I -- Gd Ha Hb Hc Hd He Hf Hh Hi la 16 Ic Ja Jb Jc Jd Je Jf Jh Illllllllllllll 16-- 12- 8- -- .b,=-- 3230*1|i:42*6*44**h!!681 1. ifit. J Jk JI Jm Jn Ka (short) Ka Oong) Kb La 111111111111111 1977 BRADLEY: BELDING'S SAVANNAH SPARROW 63 and recorded many of the same call notes he described, as well as a few addi- tional ones. Dwight (in Chapman 1896) described the typical call note of' the species as a vigorous "chipping." Gobiel referred to this note as a softer version of the "tsip" alarm note. Spectrograms of the typical call note recorded in the pres- ent study are presented in Figure 2. This note ranges in frequency from 8.5 to 10 kHz and is rather weak with a duration of less than 0.05 seconds. Both sexes uttered this "chip" note in a variety of contexts. A slight variant of the typical call note described above was also recorded. This variant was associated with copulation. Although I have heard both sexes using this copulatory note, it is more frequently uttered by the female. The structure of the copulatory note is illustrated in Figure 2. The frequency and temporal characteristics of this note are quite similar to those of the typical call note. The copulatory note is given just prior to copulation and is repeated in rapid succession with intervals of less than 0.5 seconds. The alarm note of the species is much louder and harsher than the typical call note. Gobiel described the alarm note as "tsup" or "tsip." A spectrogram of this note is included in Figure 2. The frequency range of the alarm note (7 to 11 kHz) is much greater than that of the typical call note and is consider- ably louder. The time interval between successive alarm notes depends upon the emotional state of the calling bird. In general the rendition increases in rate, intensity, and volume as an intruder moves deeper into the nesting terri- tory. The functions of the alarm note are to warn other sparrows of impending danger and probably to distract a potential predator from the nest or young. The intraspecific agonistic note is actually a loud buzz, which Gobiel termed the hostile note. Both wide and narrow band spectrograms of this buzz are presented in Figure 2..The hostile buzz is usually uttered twice in close succession. Gobiel described the resultant sound as "psst-psst" or "buzt-buzt." When delivered on the ground this buzz is usually accompanied by stereo- typed posturing. The calling bird crouches down with its tail slightly cocked and the wings lowered and fluttering out to the sides. On a few occasions I saw actual attacks made by two adjacent territorial males that were display- ing and buzzing to each other. In addition this buzz is frequently given dur- ing an aerial chase. The last and most complex call is presented in Figure 2. I have termed this vocalization the flight slur; it is actually a complex sequence of notes lasting FiGURE 1.-Spectrograms of the different note types. Each of the note types from Aa to La are illustrated here in alphabetical order. A few of the combinations that might be expected (Bd, Bj, Ce, etc.) were not present and have either been superseded by new naming schemes or proved indistinguishable from existing note types. Two examples of the Ka buzz are included to demon- strate the continuous variability in the length of this phrase. In all figures frequency is plotted on the ordinate in kHz. In Figures 1 and 2 time is plotted on the abscissa in 0.2 seconds, in Figures 3 through 15 in seconds. 64 BULLETIN FLORIDA STATE MUSEUM Vol. 22, No. 2 «6r .E a -2= *ID„ Alifi0 0 , !·Kit . ilf -ltrllit# 6 W 2.1//M- '' 43 '6' #)11 liFiltL 9 0 , 1. =B *2- - ae# ~il~4/ 2 9 48 - * 8- - 18 8 - 1 - 11!11111 zm !?-: Z ·,m u* mm ,#W "W , 16 - A 1 2 - 1 1 111 . 91. M.91'16 - B 0 -9, -EL -9, -EL 1977 BRADLEY: BELDING'S SAVANNAH SPARROW 65 about 3.2 seconds. Although I noted this sound given most frequently at the end of an airborne chase sequence, occasionally it is given from within the vegetation (Barbara Massey, pers. comm.). The flight slur may be analagous to the "flight song" described by Gobiel, who stated that the flight song was used· in communication between two members of a mated pair. I am not certain of the function of the flight slur, but I believe it is associated with aggression. FEATURES OF SONG DELIVERY ~ To me, the delivery of the Savannah Sparrow's primary song is perfunc- tory, almost seeming mechanical at times. The singer usually perches in an upright stance with the head held slightly back and utters the song with the bill wide open. The song of Bel(ling's Sparrow is usually given while the bird is perched atop a sprig of Salicornia. I have also seen birds singing from higher perches, such as the tip of small mounds of soil, fence posts, and other wooden structures, as well as an occasional utility line. In addition to these typical singing perches I have several times seen birds singing from open mudflats. During intense territorial disputes males even uttered the primary song in flight. Although several authors have written of the male's song there is little direct evidence that only the male is capable of song. I made observations on 10 individuals that I caught and color-banded. Of these 10 birds, 6 were known to be males (cloacal protuberances present), 1 was a known female (brood patch present), and 4 were not sexable. In the term of this study, I saw the known female only twice after banding. She did not sing, but she did utter the typical call note of the species. Of the six color-banded males only four were seen subsequently, and all four sang regularly. In addition to these observations of color-banded individuals, twice I saw mating. Both times the male engaged in song bouts before and after the copu- latory act. The female was silent except for the copulatory note. Additional studies of more color-banded individuals need to be under- taken, but considering closely related species (Bent 1968) and the observations made in this study, song is most probably limited to the male. Herein singing birds are assumed to be males. The singing rate (number of songs per minute) varies with the time of day, season, and the emotional state of the singing bird. This rate usually ranges FIGURE 2.-Spectrograms of the different note types and calls. Line A and the first portion of B represent a continuation of the list of note types from Ma through Zb. The last part of B illus- trates the typical call notes of the species: a-alarm note, b-agonistic buzz (narrow band), c- agonistic buzz (wide band), d through f-typical call notes (three repetitions), and g-precopu- latory note. Line C is the flight slur: Line D shows different buzz phrases, first a narrow band, then a wide band spectrogram of the same buzz. The phrase pairs in sequence are Pa, Pf, Pb, Ka, and Ma buzzes. 66 BULLETIN FLORIDA STATE MUSEUM Vol. 22, No. 2 partl ~ part 2 ~ part 3 ~ part 4 16- A 12- - •4 04 4- - 1 1 16- 8 12- 4 04 4- 05 10 1.5 <0 25 FIGURE 3.-The four parts of a song. A is a narrow band spectrogram of bird song number 05-50, B a wide band spectrogram of the same song. from six to eight songs per minute. Singing may begin just before sunrise, but the peak of song occurs most often between 30 and 90 minutes after sunrise. In general, song activity decreases throughout the day with a lull during the midday hours. Song activity makes a secondary peak in the evening and may continue until well after dark. For example, on 27 March 1973 the evening song peak was at 1830 (about 20 minutes after sunset) and singing continued until about 1900. I heard sporadic songs as late as 2200. King (1972) also re- ported hearing night song from the Rufous-collared Sparrow (Zonotrichia capensis). In addition to diurnal variation, the rate of singing changes with the sea- son. I first heard song at the main study area (Anaheim Bay) on 26 January, when a few males were singing sporadically. On 2 February there was con- siderably more song activity, and I saw several territorial chases. The height of singing in 1973 lasted from mid-February through mid-May. Song activity was noticeably lower in late May and dropped off rather quickly in June. By the end of July adult song activity was very low, but I began to hear subsong from birds of the year. In late August subsong activity was high, and only oc- casional adult songs were heard. The following winter (1973-1974) song ac- tivity began earlier in the year. By 24 December adult song was heard occa- sionally, and it was noticeably more regular on 9 January (Barbara Massey, pers. comm.). The rate of singing can be increased artificially by recording an individ- ual's song and then playing it back through a loudspeaker. With this technique I could increase the rate of song to approximately 12 songs per minute. That one individual rarely sings alone is interesting to note. The song of one bird tends to stimulate all of his territorial neighbors, which frequently transforms relative silence into a peak of song activity by a chain reaction of singing and countersinging. 1977 BRADLEY: BELDING'S SAVANNAH SPARROW 67 THE STRUCTURE OF THE PRIMARY SONG The primary song of the Savannah Sparrow is a rather unmusical series of thin high-pitched chips and buzzes. One of the most accurate general descrip- tions of the song of the Savannah Sparrow is given by Borror (196lb). Al- though the Belding's Sparrow songs recorded in this study have qualities simi- lar to those Borror described, several differences are apparent. The high pitched and rather weak nature of the primary song of the Sa- vannah Sparrow may be related to the territory size and habitat preferences of the species. As I have already mentioned, males of this species command very small territories. Pairs nest primarily in open grassland or marshland habitats. As distances between neighbors are quite small with little or sparse intervening vegetation, no necessity exists for' a louder or lower pitched song that would carry farther. In fact, the weak nature of the song may well render the singing males less conspicuous to predators. The characteristics of Savan- nah Sparrow song conform well to the theory, discussed by Emlen (197la), Jilka and Leisler (1974), and Morton (1975), relating the quality of a species' song to its habitat. The songs of individuals recorded in this study ranged in length from 1.7 to 3.1 seconds, and most songs were between 2.0 and 2.5 seconds long. The notes range in frequency from 2 to 12 or 13 kHz (certain harmonics may reach 16 kHz or higher). Most of the sound energy is concentrated between 4 and 11 kHz. The structure of the primary song is best described by treating it in four sections (Fig. 3). The first section (Part I) is composed of three to eight similar notes. These introductory notes average higher in pitch than the remainder of the song. Most are named as variants of note B (Fig. 1 and 2). As the spectro- grams of notes Bb through Bn show, the structure of the note includes a hori- zontal portion and a vertical portion. The horizontal portion represents a con- tinuous tone of one frequency and the vertical portion a sharp staccato ending. This ending makes the sound audible as a "chip." In contrast, a note that lacks this sharp ending has a more melodious sound (like note ja). In addition to the introductory notes of type B, Part I often includes the very short and high- pitched note Aa, which is quite similar to the typical call note of the species. In fact a calling bird occasionally broke directly into song, incorporating the last call note or two into the introductory phrase. I heard note Aa from nearly every sparrow recorded, yet no bird included it consistently. Therefore, I have assigned very little importance to the presence or absence of note Aa. The second part of the primary song varies in composition. Among the 50 different note types that occur in Part II many are shorter than 0.1 seconds. Individual birds include from 1 to 11 different note types. In some cases a par- ticular note is repeated 2 to 17 times in sequence. Notes of the general types C, D, and E are frequently found in Part II. In many cases these notes are ar- 68 BULLETIN FLORIDA STATE MUSEUM Vol. 22, No. 2 ranged in a specific pattern that may recur in songs of several other members of the same population. For example the sequence DaDbDc is common among birds recorded at Anaheim Bay. In addition to these notes, individuals often include a very short buzz in Part II. Buzzes found in Part II, unlike most of the buzzes that compose Part III, are modulated approximately 50 to 100 times per Second. The buzzes that occur in Part II are classified as P, S, and Z phrases (Fig. 2). The third part of the song is the main buzz section. It usually includes one or two K-type buzzes or an L-type buzz. Buzzes occurring in Part III are much more rapidly modulated than the type of buzz phrase found in Part II. The rate of modulation is frequently as high as 180 times per second. On the nar- row band spectrograms these buzzes appear as indistinet bands, ranging in.fre- quency from 5 to 10 kHz. With the wide band filter setting, spectrograms are produced that appear as a series of vertical streaks (Fig. 1). In addition to the buzz phrases in Part III, a few shorter notes are found occasionally between two successive buzzes. Many of the note types found between buzzes are the same as those found in Part II. Rarely a sequence of notes characteristic of Part II will also lie between the two buzzes. The final section of the primary song is the shortest of the four parts. This "terminal flourish" is composed of one to five short high-pitched notes fol- lowed by a single low-pitched note. The high-pitched notes are occasionally complex in structure. In some cases they even appear to possess the dual- source quality described by Greenewalt (1968). Examples of such complex notes are classified as types F, I, and R. The final note is usually between 0.15 and 0.4 seconds in length and has a pure, almost whistled quality (note Ja, Fig. 1). This tonal quality is attributable in part to the narrow frequency range of the note (between 5 and 6 kHz). As will be shown later, Part IV is one of the most consistent phrases found within the songs of a particular population. In addition, most populations have one or more unique variants of this terminal flourish. ONTOGENY OF THE SONG To my knowledge no study on the development of song in the Savannah Sparrow has been conducted to date. A complete study, probably involving Kaspar Hauser isolates, is needed before definitive conclusions about song learning can be made, but the following observations may throw some light on the process as it may occur in the wild. I first heard subsongs, presumably given by birds of the year, on 28 August 1973 in the marsh at Anaheim Bay. A hatching-year individual can usually be identified in the field by the grayish color of the superciliary line and loral spot, which are yellowish in an adult. The plumage of the young bird appears looser and the markings less distinct. Subsongs were delivered from the top of Salicornia bushes in the same general manner as the definitive songs of adult 1977 BRADLEY: BELDING'S SAVANNAH SPARROW 69 males. Young birds rarely throw back the head or open the bill and they ap- pear to assume a more relaxed posture. Subsong at first is a rather low volume series of buzzes and chips with little organization. Another type of subsong, which presumably represents a later stage of development, resembles the "random warbling" described for many other oscines (Lanyon 1960). Even within this phase of song development some typical call notes are still included. One of the call notes most frequently heard was a buzz similar to the hostile buzz of adult Savannah Sparrows. I term the third type of subsong "discrete subsong." This type of song was heard more frequently late in the summer and in the fall. It consists of a shortened "warbling" song about 2.5 to 3.0 seconds long, and contains some notes reminiscent of definitive primary song..In Lanyon's (1960) scheme, this phase is probably "rehearsed song." Unfortunately I was unable to visit the study area throughout much of the fall and winter, but young birds continued to sing subsong well into December (Barbara Massey, pers. comm.). In the peak season of subsong (late summer and fall) few adults were heard, and the primary song of adult males was sporadic and irregular. Structure of adult primary song heard at this time was typical of birds recorded in the spring. Although Mulligan (1966) found that song development in Song Sparrows (Melospiza melodia) was essentially innate, studies of other fringillids indicate that young birds must learn their songs from adults. Chaffinches for example produce abnormal songs when reared in isolation (Poulsen 1951, Thorpe 1958). In similar studies summarized by Marler (1967), normal song develop- ment in members of the genera Junco and Zonotrichia was also dependent upon previous auditory experience. The role of learning and other aspects of the ontogeny of the song in Savannah Sparrows remains to be worked out. INDIVIDUAL VARIATION The song of each Belding's Sparrow is remarkably stereotyped. A typical song bout involves 15 to 20 droning repetitions of a particular song pattern. A short pause is followed by another monotonous song bout. With occasional pauses for feeding and maintenance activities, this pattern may last an entire morning during the seasonal peak of song activity. In fact, aside from minor omissions at the beginning or ending, the song pattern of each known bird re- corded remained constant throughout this study. As an example of this constancy, I have included spectrograms of bird 05-58 in Figure 4. As shown, the song pattern (AaBb3Ce3HiDaDbDcKaFd- FeJa) is the same in all this bird's songs. These recordings were made during the period 14 February-21 April. Variation seen in recordings of this individu- al's song includes the occasional omission of an Aa, Bb, or Ce note, as well as the length of the buzz phrase. One color-banded male (XR-RR), first recorded on 2 April 1973, was recorded singing the same song pattern 10 February 1974. Although this is the only bird I was able to record in both nesting seasons 16- A 12- 70 BU LLETIN FLO R ID A STATE M USEUM Vol. 22, No. 2 8- - - - -4 - -4--7.Vin~" 4 1 .--4- 1 1 1 1 1 1 16- B 12-- 8- 6 ./ .0/ 0% 0% .:. .' I.NA../- . I --/Ilt/.19%"".4. : *4*.Ill/:"-I" 4- 1 1 1 1 1 1 16- C 12- 8- 4- - 1 lilli 16- D 12- 8- '* 0% 0% 04 /// 1 /* 1 LP-10 AaBc3 EcZEbCgEbCa3Ee Ka EcFa~a 1 LP-11 AaBc4 CdEa-EbOcObDe Ka Ec~a 1 LP-12 AaBe6 PaEaEbHc2Dc Ka EefaJe 3 LP-13 AaBe4 EdEbDaCe2Dc Ka Ec~e 1 Imperial Beach (N = 8) IB-01 Bm3 PfHdDa Ka DaFeCaEaFgJa 1 IB-02 AaBm2 Ca2Hd Ka DaFeCaDaFgja 2 IB-03 AaBm4 PfDaHd Ka EaFeCaEaFgJa 3 IB-04 AaBm4 Ca4HdDa Ka EaFeCaEaFgJa 2 San.Quintin Bay (N = 34) SQ-01 AaBh3 EbeCgEbeaOe KaObKb EaFe~b 1 SO-02 AaBh3 EaEbCgEbCa30c Kb Eafbjb 3 SQ-03 AaBh5 fbEaCa KaOcKb Eafbjb 1 SQ-04 AaBh3 EaEbCgEb KaOcKb EaFbJb 6 SQ-05 Aa(2)Bh3 DaEbCgEbCgjm KaOcKb DaF'bjb 2 SQ-06 AaBh3 OaEbCgEbCeOc Ka EcfbJa 5 SQ-07 Aa2814 Eb2CgEb LaOcKa Ecfbja 3 SQ-08 AUB12 EaEb2CgEb LaOcKa EePa 2 SQ-09 AaB13 EcEaEbCgEb LaOcKa EcFbJb 3 SQ-10 B16 CeEaJa Ka Ecfblb 3 SQ-11 AaB13 CaDeEb2CgEb LaOcKb EcFbJb 1 SQ-12 AaB13 CeEbCeEbCgEbjb Ka EcfbJa 1 SQ-13 AaB13 Eb2CgEbCdCa20c Ka EcFbjb 1 SQ-14 B14 EaEbCiCaOc Ka EcFb~b 1 SQ-15 B14 Ea2CiCeOc Ka Ecjb 1 El Rosario (N = 11) ER-01 Aa2Bn5 La Da~a 1 ER-02 AaBn6 La Da~dCkDaJa 4 ER-03 AaBn5 La DiHdCkDaCkDa~a 6 'Notes or phrases occasionally omitted and numbers of repetitions that varied were placed in parentheses. VARIATION IN PART III.-Variation within the main buzz section (Part III) was relatively simple. Although many individuals from the various popu- lations included a number of notes between successive buzzes, the basic pat- tern of one or two K-type buzz phrases was maintained in most of the birds sampled. Within most populations the number of buzzes was constant, but 16- A 1977 B R A D LE Y: B E LD IN G 'S S AVA N N A H SPAR R O W 87 12-- 8- 4- .- 1 1 1 1 1 1 16- B 12-8-ill Ill *1 0/.51'.M#40 4- 1 1 1 1 1 1 16- C 12-- 8- * 4 4 4- ...//i"- 1 1 1 1 1 1 16- D 12- 8- . '4 wi *-~ 4..4 -¥!I'.--g--IT#p,q4- - 1 1 1 1 1as 1.0 15 2.0 2.5 3:0 FIGURE 12.-Spectrograms recorded at San Elijo Lagoon and El Rosario. A-pattern SE-01, B-pattern SE-02, C-pattern ER-02, and D-pat- tern ER-03. 88 BULLETIN FLORIDA STATE MUSEUM Vol. 22, No. 2 there were several exceptions to this general rule. In a few localities the typi- cal Ka or Kb buzzes were replaced by La, Ma, or Na buzz phrases. In the main buzz sections of the entire sample 24 I found different note types, but most birds included only two to four notes between any two successive buzzes. VARIATION IN PART IV.-Variation between the different populations is best demonstrated by analysis of Part IV. This terminal flourish portion of the songs was rather constant at a given location. Each population had one or a few characteristic patterns. Considerable variation existed among the several recording localities. This diversity took the form of variation in the number of notes, the general note sequence, and in some cases the structure of certain notes. The number of notes included ranged from two to eight. Most of the notes were of the general types D, E, F, I, and J. All but 7 of the 280 birds re- corded in this study terminated their songs with a J-type note. SUMMARY OF INTERPOPULATIONAL VARIATION BY LocATION 01 GOLETA SLOUGH.-~nta Barbara, Santa Barbara County, California, close to the Santa Barbara municipal airport. The marsh is fairly large with a deep channel running down the center. The population of sparrows was probably fewer than 50 pairs. Recordings were made on 25 May of only eight individuals (Fig. 5). The songs of these birds had many short introductory notes of type Bg. The two buzzes present were separated by a single note and the terminal flourish was consistently of four notes. 02 EL ESTERO.-Less than 1 mile west of Carpinteria, Santa Barbara County, California. Although this marsh was being channelized, a large num- ber of sparrows were still singing there. The population was estimated at 100 breeding pairs. Recordings were made on 11 May of a total of 28 individuals (Fig. 6). The introductory notes were similar to those recorded from birds at Goleta Slough, with from four to seven Bg notes followed by a series of C notes or a P-type buzz. The number of K-type buzzes varied. The terminal flourish usually consisted of three simple notes followed by a short Jb note. 03 POINT MuGu.-In the Pacific Missile Range, about 5 miles southeast of Port Hueneme, Ventura County, California. This area was visited on 4 May. Recordings were restricted to the eastern finger of the estuary, near Point Mugu State Park. Songs of 28 of the estimated 175 singing males were re- corded (Fig. 7). The initial pattern BhBi2 gave their songs a recognizable opening. The main buzz was quite long (up to 0.7 seconds). Part IV consisted of three to four notes usually with an Ib note and ending with a Jb note. Three individuals used a very distinctive harsh buzz constructed of numerous Cl notes. 04 PLAYA DEL REY.-Just east of Marina del Rey, Los Angeles County, California. Recordings were made on July 21. Although only 4 of the esti- mated 25 males were recorded, several others that I heard were using similar 16- A 1977 B R A D LE Y: B E LD IN G 'S S AVA N N A H SPAR R O W 89 12- - 4 4 4 411 4 3:E· 6 - <<<~~~~~~~<~~~~~~ 4,9. 4- 1 1 1 1 1 1 16- B 12-- 4- ...I-- 1 1 1 1 1 1 16- ( 12- 8- 4, ; a U- a-f 4- • - 1 1 1 1 1 1 16- D 12- 8- » I .4 #4** ..1,'.. h.- bl./- .0 4- 1 0.5 1.~0 1.5 2.0 2.5 3.0 Fic:URE 13.-Spectrograms recorded at Los Penasquitos Estuary. A-pattern LP-01, B-pattern LP-04, C-pattern LP-08, and D-pattern LP-12. 90 BULLETIN FLORIDA STATE MUSEUM Vol. 22, No. 2 song patterns (Fig. 5). The songs recorded here had no B notes, and several short C, D, and E notes formed the introduction; Parts II and III consisted of a P buzz followed by a single note then a K buzz. The terminal flourish was composed of only three notes ending with a short J-type note. 05 ANAHEIM BAY.-Seal Beach National Wildlife Refuge, Seal Beach, Orange County, California. Anaheim Bay was the main study area where I made observations of individual variation and seasonal variation. In addition to recording 61 individuals, I color-banded a number of birds for individual recognition. I estimated the population here to be approximately 125 singing males. Recordings were made on the following dates: 2, 7, 14, 17, and 26 February, 3 March, 2 and 21 April, 12 and 20 May, 29 August 1973, and 10 February 1974 (Fig. 8). The songs of birds recorded at this locality began with three or four Bb notes followed by a P buzz or a series of C notes forming a buzzlike phrase. Part III had only one K-type buzz. The terminal flourish usually consisted of the pattern FdFeja with a rather long Ja note. 06 HUNTINGTON HARBOUR.-SUnSet Beach, Orange County, California. As mentioned earlier this area and Bolsa Chica Lagoon (07) to the east were both once part of the Anaheim Bay marsh complex. The combined population of Huntington Harbour and Bolsa Chica Lagoon is about 50 pairs. Five birds were recorded at Huntington Harbour on 6 June (Fig. 9). The patterns of their songs were similar to those recorded at Anaheim Bay. 08 NEWPORT BAY.-Less than 1 mile northeast of Newport Beach, Orange County, California. Although this is the largest estuary that I visited in Cali- fornia, the habitat was mostly Spartina and open mudflats. The portion of the estuary where Salicornia occurred supported about 130 breeding pairs of Belding's Sparrows. Recordings were made on 2 February and 16 March of a total of 30 males on the two visits (Fig. 10). The songs of these birds bore some resemblance to those recorded at Anaheim Bay. They had a variable number of introductory notes (usually four to seven), and either one or two K-type buzzes. In the song patterns containing only one K buzz it was usually pre- ceded by a P buzz in Part II. The distinctive four-note terminal flourish EaFaFdja was characteristic of this location. 09 SANTA MARGARITA LAGOON.-On Camp Joseph H. Pendleton just north of Oceanside, San Diego County, California, this estuary supported approxi- mately 125 singing males. A sample of 29 birds was recorded on 28 April (Fig. 11). Most of the song patterns included three of the relatively long Bk in- troductory notes. Two K-type buzzes were present in the songs. The first was usually a Ka buzz about 0.4 seconds long; the second was a Kh buzz about 0.25 seconds long. The song ended with three notes usually of the pattern IcDaJa. Three individuals used the highly distinctive Na buzz in their songs. 10 AGUA HEDIONDA LAGooN.-Carlsdad, San Diego County, California. Recordings were made at this area on 24 March (Fig. 9). Although only 10 of 16- A 1977 B R A D LE Y: B E LD IN G 'S S AVA N N A H SPAR R O W 91 12-- 8- d --/ 4 4,1 fit-·--/ 4- 111111 16- B 12-- 8- * a /11//b-1 4- ...... 111111 16- C 12-- 1 8- . ZE; 4# .4 9*444 41-=~ lb-* 80% - 4- il jb~*h 1 '1" ~ - *W471 - 0 04 -.. 06 10 el 1 13 4 California N 15 FIGURE 16.-Map localities and type B note variants. Boxes near each recording locality show basic structure of B-type note variant or variants used by members of that population. Numbers refer to the location numbers listed in Table 1. Again numbers 07 and 14 are omitted as no birds were recorded at Bolsa Chica Lagoon or Ensenada Bay. No illustration is indicated for location 04 (Playa del Rey) because B-type notes were absent from songs of birds recorded there. 96 BULLETIN FLORIDA STATE MUSEUM Vol. 22, No. 2 posed is that of the male parent. If young females respond most strongly to this song pattern, the most prolific (fit) males would sing the most popular song patterns. The occurrence of clustering in males with similar song patterns, as observed in this study, would tend to enhance this process. Both male and fe- male juveniles would thus be exposed predominately to one song pattern. That higher passerines can distinguish between songs of different individ- uals has been shown by several authors (Marler 1956, Weeden and Falls 1959, Falls 1969, and Emlen 197lb). These studies involved responses of territorial males to recorded songs. To my knowledge, only two field studies of differen- tial responses in adult females have been conducted. Bertram (1970) demon- strated that individual female Indian Hill Mynahs (Gracula religiosa) recog- nized their mates by vocal characteristics. Milligan and Verner (1971) men- tioned that female White-crowned Sparrows responded more strongly to the local dialect. Differential responses to songs of different dialects have also been shown in territorial males (Lemon 1967, Milligan and Verner 1971, Thieleke 1973, and Harris and Lemon 1974). Again little field data are avail- able for the responses of mature females. The presumed function of song in attracting the female is critical to any discussion of song as an isolating mechanism. This advertising function of song has been shown for several species (Quaintance 1938, Nice 1943, Woolfenden 1956, Smith 1959, Catchpole 1973). In addition, Payne (1973a) demonstrated a high degree of assortative mating between Indigobirds imprinted upon spe- cific mimetic songs. Positive responses have also been elicited in captive fe- males to recordings of the advertising songs of males (Payne 1973b). Record- ings of song stimulate approach in female Chaffinches (Marler 1956). Female White-throated Sparrows may even assume a soliciting posture and utter pre- copulatory vocalizations in response to male song (Falls 1969). Specific functions of the song in the Savannah Sparrows are undocu- mented. At least part of the function includes the establishment and reten- tion of discrete territories by the singing males. There seems little doubt that the behavior of the male during song makes him conspicuous to females seek- ing a mate. While these sparrows spend most of their time on or near the ground, song is nearly always given from an exposed perch, often well above the surrounding vegetation. If song dialects are truly an important mechanism of behavioral isolation, evidence should exist of morphological or genetic variation between different dialect groups. The interaction of a dialect system with the occurrence of certain genes was shown by Nottebohm and Selander (1972) in Zonotrichia capensis. King (1972), in another study of Z. capensis, showed a correlation between song themes and the specific habitat in which the sparrows occurred. He also included some morphological data on the birds from each area. Bake'r (1975) found genetic differences between dialect groups in White-throated Sparrows. In the same paper he presented data showing allelic variation in 1977 BRADLEY: BELDING'S SAVANNAH SPARROW 97 another population where no dialects exist. Preliminary work that I have con- ducted on Orange-crowned Warblers (Vermiuora celata) has revealed that the darker insular race (sordida) has a distinctly slower trill than the brightly colored mainland race (leutescens). In studies of very closely related species. differences in their vocalizations are often cited as possible isolating mech- anisms (Lanyon 1957, Stein 1958, Schwartz 1972, Thieleke 1973). Adaptive radiation within the coastal species of North American emberi- zine sparrows is extensive. Several marsh nesting species such as the Seaside Sparrow (Ammospiza maritima) and the Sharp-tailed Sparrow (A. caudacuta) have a number of different morphological variants. Along the west coast of North America, no fewer than 11 different subspecies of the Savannah Spar- row are currently recognized (American Ornithologists' Union Check-list 1957). Along the eastern coast of the continent are many marsh-inhabiting sparrows including several distinct species. Beecher (1955) believed that the Seaside Sparrows and the Sharp-tailed Sparrows were originally derived from Savannah Sparrow stock. It seems evident that speciation in coastal popula- tions of North American sparrows is progressing at a relatively rapid rate. If song dialects are indeed an important isolating mechanism, they may play a significant role in such rapid speciation. LITERATURE CITED American Ornithologists' Union. 1957. Check-list of North American birds, 5th ed. Lord Balti- more Press, Baltimore, Md. 691 pp. American Ornithologists' Union. 1973. Thirty-second supplement to the American Ornithologists' Union Check-list of North American birds. Auk 90:411-419. Armstrong, E. A. 1973. A study of bird song, 2nd ed. Dover Publications, Inc., New York. 343 pp Baker, M. C. 1975. Song dialects and genetic differences in White-crowned Sparrows (Zono- trichia leucophrys). Evol. 29:226-241. Baptista, L. F. 1974. The effects of songs of wintering White-crowned Sparrows on song devel- opment in sedentary populations of the species. Z. Tierpsychol. 34:147-171. Beechet, W. J. 1955. Late-Pleistocene isolation of saltmarsh sparrows. Ecology 36:23-26. Bent, A. C. (0. L. Austin, Jr., ed.) 1968. Life histories of North American Cardinals, Grosbeaks, Buntings, Towhees, Finehes, Sparrows, and allies. Order Passeriformes: Family Fringilli- dae. Genera Pipilo (part) through Spizelia. U.S.N.M. Bull. No. 237. Smithsonian Inst. Press. Bertram, B. 1970. The vocal behavior of the Indian Hill Mynah, Gracula religiosa. Anim. Behav. Monogr. 3:80-192. Borror, D. J. 1956. Variation in Carolina Wren songs. Auk 73:211-229. _ . 196la. Intraspecific variation in passerine bird song. Wilson Bull. 73:57-78. 196lb. Songs of finches (Fringillidae) of eastern North America. Ohio J. Sci. 61:161- 174. 1965. Song variation in Maine Song Sparrows. Wilson Bull. 77:547. Bradley, R. A. 1973. A population census of the Belding's Savannah Sparrow, Passerculus sand- wichensis beldingi Western Bird Bander 48:40-43. Catchpole, C. K. 1973. The functions of advertising song in the Sedge Warbler (Acrocephalus schoenobaenus) and the Reed Warbler (A. scirpaceus). Behaviour 46:300-319, Chapman, R. M. 1896. Handbook of birds of eastern North America, 3rd ed. D. Appleton and Co., New York. 427 pp. 98 BULLETIN FLORIDA STATE MUSEUM Vol. 22, No. 2 Conrads, K. and W. Conrads. 1971. Regionaldialekte des Ortolans (Emberiza hortulana) in Deutschland. Die Vogelwelt 92:81-100. DeWole, B. B., D. D. Kaska, and L. J. Peyton. 1974. Prominent variations in the songs of Gambel's White-crowned Sparrows. Bird-banding 45:224-252. Emlen, S. T. 197la. An experimental analysis of the parameters of bird song eliciting species recognition. Behaviour 41:130-171. 197lb. The role of song in individual recognition in the Indigo Bunting, Z. Tierpsy- chol. 28:241-246. Falls, J. B. i969. Functions of territorial song in the White-throated Sparrow. Pages 207-232 in R. A. Hinde, ed. Bird vocalizations. Cambridge University Press, London. Gobiel, R. E. 1970. Vocalizations of the Savannah Sparrow. Bird-banding 41:18-21. Greenewalt, C. H. 1968. Bird song: acoustics and physiology. Smithsonian Institute Press, Wash- ington, D.C. 194 pp. Grimes, L. G. 1974. Dialects and geographical variation in the song of the Splendid Sunbird Nectarina coccinigaster. Ibis 116:314-329. Harris, M. A. and R. E. Lemon. 1972. Songs of Song Sparrows (Melospiza melodia): Individual variation and dialects. Canad. J. Zool. 50:301-309. . 1974. Songs of Song Sparrows: reactions of males to songs of different localities. Condor, 76:33-44. Jilka, A. and B. Leisler. 1974. Die Einpassung dreier Rohrsangerarten (Acrocephalus schoeno- baenus, A. scirpaceus, A. arundinaceus) in ihre Lebensraume in bezug auf das Frequenzspec- trum ihrer Reviergesange. J. Ornithol. 115:192-212. King, J. R. 1972. Variation in the song of the Rufous-collared Sparrow (Zonotrichia capensts) in northwestern Argentina. Z. Tierpsychol. 30:344-373. Kroodsma, D. E. 1974. Song learning, dialects and dispersal in the Bewick's Wren. Z. Tierpsy- chol. 35.352-380. Lanyon, W. E. 1957. The comparative biology of the Meadowlarks (Sturnella) in Wisconsin. Publ. Nuttall Omithological.Club 1:1-67. 1960. The ontogeny of vocalizations in birds. Pages 321-347 in W. E Lanyon and W. N. Tavolga, eds. Animal sounds and communication. American Institute of Biological Science, Washington, D.C. 1969. Vocal characters and avian systematies. Pages 291410 in R. A. Hinde, ed. Bird vocalizations. Cambridge University Press, London. Lemon, R. E. 1966. Geographic variation in the song of Cardinals. Canad. J. Zool. 44:413-428. 1967. The response of Cardinals to songs of different dialeets. Anim. Behav. 15:538- 545. Marler, P. 1952. Variations in the song of thv Chaffinch, Fringilla coelebs. Ibis 94:458472. 1956. The voice of the Chaffinch and its function as a language. Ibis 98:231-261. 1967. Comparative study of song development in sparrows. Proc. 14th Int. Ornith. Congr. Blackwells, Oxford, pp. 231-244. Marler, P. and M. Tamura. 1962. Song -dialects" in three populations of White-crowned Spar- rows. Condor 64:368-377. 1964. Culturally transmitted patterns of vocal behavior in sparrows. Science 146:1483-1486. Mayr, E. 1963. Animal species and evolution. The Belknap Press. Cambridge, Mass. 797 pp. Milligan, M. M. and J. Verner. 1951. Inter-populational song dialect discrimination in the White- crowned Sparrow. Condor 73:208-213. Morton, E. S. 1975. Ecological sources of selection on avian sounds. Amer. Naturalist 108:17-34. Mulligan, J. A. 1966. Singing behavior and its development in the Song Sparrow, Melospiza melodia. Univ. Calif. Publ. Zool. 81:1-76. Nice, M. M. 1943. Studies in the life history of the Song Sparrow. II. The behavior of the Song Sparrow and other passerines. Trans. Linnaean Soc. of New York 6:1-328. Nottebohm, F. 1969. The song of the Chingolo, Zonotrichia capensis, in Argentina: Descrip- tion and evaluation of a system of dialects. Condor 71:299-315. 1972. The origins of vocal learning. Amer. Naturalist 106:116-140. Nottebohm, F. and R. K. Selander. 1972. Vocal dialects and gene frequencies in the Chingolo Sparrow (Zonotrichia capensis) Condor 74:137-143. 1977 BRADLEY: BELDING'S SAVANNAH SPARROW 99 Orejuela, J· E. and M. L. Morton. 1975. Song dialects in several populations of Mountain White- crowned Sparrows (Zonotrichia leucophrys oriantha) in the Sierra Nevada. Condor 77:145- 153. Payne, R. B. 1973a. Behavior, mimetic songs and song dialects, and relationships of the parasitic Indigobirds ( Vidua) of Africa. Orn. Monogr. 11:1-333. 1973b. Vocal mimicry of the paradise whydas (Vidua) and response of female why- dahs to the songs of their hosts (P!,tilia) and their mimics. Anim. Behav. 21:762-771. Poulsen, H. 1951. Inheritance and learning in the song of the Chaffinch (Fringilla coelebs L.). Behaviour 3:216-228. Quaintance, C. W. 1938. Context, meaning and possible origin of males song in the Brown Tow- hee. Condor 40:97-101. Ridgway, R. 1885. Some emended names of North American birds. Proc. U. S. Nat. Mus. 8:354- 356. Schwartz, P. 1972. Micrastur giluicollis, a valid species sympatric with M. mficollis in Amazonia. Condor 74:399-415. Smith, R. L. 1959. The songs of the Crasshopper Sparrow. Wilson Bull. 71:141-152. Stein, R. C. 1958. The behavioral, ecological and morphological characteristics of two popula- tions of the Alder Flycatcher, Empidonax traillii (Audubon). Bull. New York State Mus. Sci. Serv. 371:1-63. Thielcke, G. 1969. Geographic variation in bird vocalizations. Pages 311-339 in R, A. Hinde, ed. Bird vocalizations. Cambridge University Press, London. 1973. On the origin of divergence of learned signals (songs) in isolated populations. Ibis 115:511-516. Thompson, W. L. 1970. Song variation in a population of Indigo Buntings. Auk 87:58-71. Thorpe, W. H. 1958. The learning of song-patterns by birds, with especial reference to the song in the Chaffinch Fringilla coelebs. Ibis 100:535-570. 1961. Bird-song. Cambridge University Press, London. 143 pp. van Rossem, A. J. 1947. A synopsis of the Savannah Sparrows of northwestern Mexico. Condor 49:97-107. Ward, R. 1966. Regional variation in the song of the Carolina Chickadee. Living Bird 5:127-150. Weeden, J. S. and J. B. Falls. 1959. Differential responses of male Ovenbirds to recorded songs of neighbouring and more distant individuals. Auk 76:343-351. Woolfenden, G. E. 1956. Comparative breeding behavior of Ammospiza caudactita and A. mari- tima. University of Kansas Publ. 10:45-75. Contributions to the BULLETIN OF THE FLORIDA STATE MUSEUM, BIOLOGICAL SCl- ENCES SERIES, may be in any field of biology. Manuscripts dealing with natural history of systematic problems involving the southeastern United States or the New World tropics are solicited especially. Manuscripts should be of medium length-circa 35 to 200 pages (10,500- 16,000 words). Examination for suitability is made by an Editorial Board. The BULLETIN is distributed worldwide through institutional subscriptions and exchanges. It is considered the responsibility of the author to distribute his paper to all interested individ- uals. To aid in this the auth6r(s) receiv«s) 50 copies free, and he may purchase additional separates at cost if ordered when page proof is returned. The author is also responsible for any charges incurred for alterations made by him on galley or page proofs. The Museum will send an invoice to the author for this amount upon completion of publication. PREPARATION OF MANUSCRIPT Contributors should consult recent numbers of the BULLETIN for preferred style and format. Highly recommended as a guide is the CBE Style Manual, 3rd Edition, 1972 (Washington, D.C., Amer. Inst. Biol. Sci.). MSS must be submitted in duplicate (please no onionskin) and satisfy the following minimal re- quirements: They should be typewritten, double-spaced (especial/g tables, figure captions, and "Literature Cited"), on one side of numbered sheets of standard (8-1/2 x 11 in.) bond paper, with at least one-inch margins all around. Figure legends and tables should be typed on separate sheets. All illustrations are referred to as figures. They must comply with the following stan- dards: Photographs should be sharp, with good contrast, and printed on glossy paper. Drawings should be made with dense black waterproof ink on quality paper or illustration board. All il- lustrations should have a cover sheet. All lettering will be medium weight, sans-serif type (e.g. Futura Medium, News Gothic) in eutout, dry transfer, or lettering guide letters. Make allow- ance so that after reduction no lowercase letter will be less than 1 mm high (2 mm is preferred) nor any capital letter greater than 5 mm high. The maximum size for illustrations is 8-5/8 x 14 in. (twice typepage size); illustrations should not be less than typepage width (46/16 in.). Designate the top of each illustration and identify on the back with soft pencil by author's name, MS title, and figure number. All manuscripts not submitted in BULLETIN format will be returned to the author for retyping. Manuscripts and all editorial matters should be ackiressed to: Managing Editor of the BULLETIN Florida State Museum Museum Road University of Florida Gainesville, FL 32611