The Occurrence and Ecology of the Golden-Cheeked Warbler in the Highlands of Northern Chiapas, Mexico ’

The Condor96fG-691 0 TheCooperOmithologimlSociety 1994

THE OCCURRENCE AND ECOLOGY OF THE GOLDEN-CHEEKED WARBLER IN THE HIGHLANDS OF NORTHERN CHIAPAS, MEXICO ’

ROSA MARIA VIDAL, CLAUDIA MAC~AS-CABALLERO* AND CHARLES D.DLJNCAN~,~ Pronatura Chiapas. A.C., Apartado Postal 219, San Cristdbalde las Casas, Chiapas 29250 Mt?xico =alsoCentro de InvestigacionesEcol6gicas de1 Sureste, San CristBbalde las Casas, Chiapas 29290 M6xico also‘ Institutefor Field Ornithology, Universityof Maine at Machias, 9 OBrien’ Avenue,Machias, ME 04654

Abstract. We detail 46 sightingsof Golden-cheekedWarbler in the Highlands of Northern Chiapas, representing wintering birds as well as spring and fall migrants. In winter, the specieswas approximately 1% as abundant as Townsend’s Warbler, the most numerous warbler. Among warblers detected, only Pink-headed Warbler was less numerous than Golden-cheeked Warbler. Sightingsof male Golden-cheeked Warblers outnumbered those of females by a factor of 2.4. Golden-cheeked Warblers used a variety of habitats in the study area, especially pine-oak and pine forests. They foraged by gleaningin the upper half of the treesin which they were found, and they occurredalmost exclusivelyin mixed-species flocks. Aggressiveinteractions involving Golden-cheeked Warblers were infrequent. Key words: Golden-cheekedWarbler; Dendroica chrysoparia;endangered; winter range; habitat; abundance;mixed-species flock; foraging; sex ratio; migrant: Mexico.

Resumen. Se documentan46 observacionesde Dendroicach ysoparia en las tierras alms de1Norte de1Estado de Chiapas, tanto de aves invemantes coma de migrantes de otofio y primavera. Durante el inviemo, la especie fue aproximadamente 1% tan abundante coma D. townsendi,el parulino m&s numeroso. Entre 10sparulinos detectadosincluyendo inver- nantes y residentes,solo Ergaticus versicolorfue menos abundante que D. chysoparia. Las observacionesde 10smachos de D. chysoparia sobrepasarona las de las hembras por un factor de 2.4. Esta especie utiliza una amplia variedad de habitats en el area de estudio, especialmente10s bosques de pino-encino y 10sde pino. Forrajean colectandoentre el follaje en la mitad superior de 10s arboles donde se encuentran, y ocurren principalmente en parvadas de especiesmixtas. Las interaccionesagresivas que involucran a D. chysoparia fueron muy poco frecuentes. Palabras claves: Dendroica chrysoparia;en peligro;rango invernal;hbbitat; abundancia; parvada de especiesmixtas; forrajeo; ratio de 10ssexes; migrante; Mc?xico.

INTRODUCTION duras, and north-central Nicaragua,” but not The Golden-cheeked Warbler (Dendroica chry- Mexico, the Sixth Edition of the Check-list of soparia) is listed by the United States Fish and North American Birds (AOU 1983) followed Pul- Wildlife Service as “endangered” (Jahrsdorler ich (1976) and not earlier statements by several 1990). The species breeds exclusively in “cedar authors (e.g., Alvarez de1 Toro 1980, Miller et brakes” in the Edwards Plateau of Texas (Pulich al. 1957, AOU 1957). The exclusion of Mexico 1976, Kroll 1980, AOU 1983), a highly restricted appears to have been based on Pulich’s rejection habitat in serious decline (Sexton 1992). Al- of several sight reports and of two old specimens, though the type specimen was taken in Guate- one lost and one misidentified, along with his mala in winter (Pulich 1976), the species is little failure to find the species during a nine-day win- known in the non-breeding season (Pulich 1976, ter search in southeastern Mexico. Pulich (1976) Kroll 1980, Sexton 1992). In describing the win- assumed that valid records from Chiapas, Mex- ter range as “the highlands of Guatemala, Hon- ico, as late as 11 October, were of fall migrants rather than wintering birds. This assumption, which we question, implies a remarkably pro- ’ ’ Received 5 October 1993. Accepted 9 March 1994. tracted southward movement. The vast majority 4 Correspondingauthor. of the species leaves Texas by the end of July,

[6841 GOLDEN-CHEEKED WARBLER IN CHIAPAS, MEXICO 685 and some individuals are known to reach Chia- March 1992. The shrub and pine-oak forest tran- pas by 9 August (Pulich 1976). sects were established during the second year, More recently, however, Braun et al. (1986) and were visited weekly from October 199 1 to described two unequivocally winter (January) the first of April 1992. When a Golden-cheeked sightings (1978 and 1983) each of one male, in Warbler was encountered, sex, habitat type, date, Chiapas. They also pointed out the difficulty of locality, and altitude, along with the bird’s meth- delimiting ranges of rare speciesin areas with a od of foraging, its height above the ground, the paucity of observations. To clarify the species’s height of the tree it used, whether it was a mem- status, we investigated the seasonality, abun- ber of a mixed speciesflock, the composition of dance, behavior, and use of habitat by Golden- the flock, and notes on any aggressionobserved cheekedWarblers in southern Mexico using three were recorded. We made no attempt to derive methods of observation. absolute population densities or determine “co- efficients of detectability” for Golden-cheeked Warblers (Emlen 197 1). STUDY AREA Casual observation.We use this term to de- We sought Golden-cheeked Warblers within the scribe accidental encounters with Golden- 484 km2 area of the municipality of San Cristobal cheeked Warblers in the course of other field de las Casas, Chiapas, Mexico (centered at work or simple bird watching. Data are ad libi- 16”44’N, 92”38’W), at elevations from 2,100 to turn (sensuAltmann 1974) and generally include 2,550 m. The seven most common habitat types sex, habitat type, date, locality, and altitude. In in this area, known as the Highlands of Northern many cases, the full set of data mentioned for Chiapas, are evergreencloud forest; pine-oak for- the transect counts above was recorded. est (see Breedlove 198 1 and Gonzalez-Espinosa Point counts. After preliminary results of the et al. 199 1 for detailed descriptions); pine forest two previous methods made it apparent that (artificially planted, mature Pinus spp., reaching Golden-cheeked Warblers were a rare-but-reg- 40 m); oak forest (essentially a special caseof the ular part of the Highlands avifauna, we began a pine-oak association described by Gonzalez-Es- series of 335 IO-min, 25-m radius point counts pinosa et al. 1991); shrub (including not only (Hutto et al. 1986) in eighteen non-contiguous shrubland, Gonzalez-Espinosa et al. 199 1, but localities. From 22 September to 11 December also areaswith early successionaltrees other than 1992, one or, more often, two observers visited Pinus spp. and Quercusspp., up to 4 m in height); localities within the study area. Each observer milpa (traditional, small-scale corn, vegetable, worked independently, making ten to twelve and flower farming); and mixed (comprisingthree counts at least 200 m apart, typically beginning or more of the previous types and characterized at 06:00-06:30 and finishing by 09:30-10:OO. by extensive vegetational “edges”). The pine-oak When a Golden-cheeked Warbler was encoun- forest and the mixed habitat share features iden- tered during a count, it was observed as long as tified by Kroll(l980) as important components possible, to a maximum of 15 min. Habitat and of the Golden-cheeked Warbler’s Texas breeding behavioral data recorded for the point counts grounds, namely edgesand open mosaics. were the same as those recorded for the line transects. On one occasion, two Golden-cheeked METHODS Warblers were encountered between point- counts. The same data were collected, and used CENSUSES in calculations concerning habitat use, but not Transect counts. As part of a broader study of abundance indices. Based on the previous re- habitat use by migrant birds, transects of 40 m sults, we concentrated on pine-oak (142 point by 1 km (Emlen 197 1, Franzreb 198 1) were es- counts, 42.4%), oak (67, 20%) and pine (31, tablished in the seven habitats described above. 9.3%) forests, and the mixed habitat type (32, Each transect-except those in shrub and pine- 9.5%) at elevations from 2,200 to 2,550 m. Oth- oak forest-was visited weekly beginning at er counts were in shrub (47, 14%), and milpas 06:OOfor two to two-and-a-half hours, from Oc- (16, 4.8%) as encountered at 200-m intervals. tober 1990 to mid-January 199 1; mid-February Evergreen cloud forest is a relatively scarceand to mid-April 199 1; October 199 1 to mid-Janu- generally inaccessible habitat in the Highlands ary 1992; and from mid-February to the end of of Northern Chiapas (Bubb, unpubl. data), and 686 R. M. VIDAL, C. MACiAS-CABALLERO AND C. D. DUNCAN was not sampled using point counts during this tected. Since all detections of Golden-cheeked study. Warblers using this point count method were in pine-oak forest, the same indices were calcu- IDENTIFICATION lated using only points within this habitat. The Since all members of the Dendroica virens su- absolute values for indices of abundance are dif- perspecies(i.e., Townsend’s Warbler, D. town- ficult to interpret since detectability is not iden- sendi,Hermit Warbler, D. occidentalis,and Black- tical for various taxa, nor from region to region throated Green Warbler, D. virens)winter in the (Hutto et al. 1986). For this reason,we offer com- Highlands of Northern Chiapas (AOU 1983) we parison abundance indices for certain other lo- were particularly cautious in identifying mem- cally-occurring species. bers of the complex, especially immature fe- For the transect studies, we calculated a fre- males. All identifications of Golden-cheeked quency per transect, f(t), dividing the number of Warblers were based on sightings using binoc- encounters (not individuals) of Golden-cheeked ulars by experienced persons, sometimes two or Warblers along a transect by the number of visits three together, familiar with the other speciesof made to the transect. We used the f(t)‘s to com- the complex, and aware of the rarity of Golden- pare the likelihood of encountering Golden- cheeked Warblers. Records always involved the cheekedWarblers acrosshabitats. Data from ca- observation of more than one of the various sual observations indicated the types and variety known field marks. In our study, adult males of habitats in which the speciesmay be found, were most easily identified by the combination though the amount of effort in each habitat is of white underparts, black throat and upper unknown. breast, and golden cheeks crossedby a well-de- Analysis of sex ratio was based on individuals fined black transocular line. Adult females and considered “certainly distinct” or “most likely immatures were identified on the basis of yellow distinct” to avoid bias from repeat observations cheeks, crossed by a well-defined black trans- of a single individual. Analysis of flock compo- ocular line, a streaked olive back, and starkly sition used all observations. We also used all white underparts, entirely without yellow tones observations for our analysis of habitat use. Chi- (Scott 1987). square analysis was used to determine signifi- cance of sex ratios. DATA ANALYSIS Individuals of a given sex were counted as “cer- RESULTS tainly distinct” if they were seenat localities more We made 46 observations of Golden-cheeked than 2 km apart, or if seen at a single locality in Warblers, from August 1990 to December 1992, different years.Individuals were counted as “most comprising not only presumed migrants but win- likely distinct” if they were recordedin decidedly tering birds as well (Appendix I). The observa- different habitats less than 2 km apart, or if they tions total 63 individuals. Of these, we consider were recorded two or more months apart, with there to have been 48 “certainly distinct” indi- no intervening sightings, at a single locality sur- viduals while three more were “most likely dis- veyed weekly. In a very few cases, individuals tinct.” Our earliest fall recordswere of one male seen at one locality within a single day were Golden-cheeked Warbler on 5 August 1990 and counted as “most likely distinct” when expressly one female, seen 11 August of the same year. considered so by the observer(s), based on dis- Our latest spring records were six observations, tance between the sightings in comparison with of three certainly distinct individuals, from 3 1 the warbler’s estimated rate of travel, as well as March to 13 April. on differing composition of mixed-species flocks involved. Birds failing to qualify for either cat- ABUNDANCE egory were considered “not distinct.” Golden-cheeked Warblers were nearly the least From the point-count data we were able to abundant of all resident or wintering warblers calculate relative abundance, though not an es- detected during the point counts (Table 1). Only timate absolute bird density (Hutto et al. 1986). the locally very rare Pink-headed Warbler (Er- We also calculated a frequency index f(25m), de- gaticus versicolor), was encountered less fre- fined by Hutto et al. (1986) as the proportion of quently. Townsend’s Warbler was the most 25-m radius counts in which the specieswas de- abundant warbler detected in point counts and GOLDEN-CHEEKED WARBLER IN CHIAPAS, MEXICO 681

TABLE 1. Mean number of selectedwarblers from point countsconducted in the Highlandsof Northern Chiapas,Mexico, Septemberto December1992. Townsend’sWarbler wasthe most abundantof all warblers and Pink-headedWarbler the leastabundant of all winteringor residentwarblers.

All sites(n = 335) Pine-oak(n = 142) Species Mean- i725 my MS%Ul’ f(25 m)b Townsend’sWarbler 1.15 0.49 1.32 0.54 Hermit Warbler 0.44 0.15 0.61 0.21 Black-throatedGreen Warbler 0.012 0.01 0.014 0.01 Green-cheekedWarbler 0.012 0.009 0.028 0.02 Pink-headedWarbler 0.006 0.006 0.014 0.01 * Mean numberof individuals per 25-m-radiuspoint count. b The proportionof 25-m-radiuscounts within which the specieswas detected. occurred in the largest proportion of counts (Ta- tions of Golden-cheeked Warblers (four individ- ble l), considering either all points or only those uals in three observations)during the point counts in pine-oak forests. Golden-cheeked Warblers were in pine-oak forest, which was also the hab- were approximately loo-times lessabundant than itat type with the greatestnumber of point counts. Townsend’s considering all points, and 50-times During the transect studies, however, there were less abundant in pine-oak forests. Of the other nine independent observations, totalling 22 members of the D. virenscomplex, Black-throat- Golden-cheeked Warblers, in five habitat types. ed Green Warblers were almost as scarce as The likelihood values, f(t), of encountering a Golden-cheeked Warblers. They occurred more Golden-cheeked Warbler were rather uniform frequently at lower elevations (Pulich 1976; Ma- acrosshabitat types. The likelihood was greatest cias-Caballero and Duncan, pers. observ.). Her- in pine forest (0.14) followed by pine-oak forest mit Warblers were intermediate in abundance. (0.09) and the mixed habitat (0.09). No Golden- checkedWarblers were found during transectsor DISTRIBUTION BY SEX point counts in shrub or milpa. Sightings of male Golden-cheeked Warblers (n = 36) significantly outnumbered those of females FORAGING, AGGRESSION, AND SOCIABILITY (n = 15) by a factor of 2.4 (x2 = 7.84, df = 1, P The mean estimated height and standard devi- 5 0.01). The true sex ratio, however, may be ation at which we observed Golden-cheeked masked by the fact that male Golden-cheeked Warblers foraging was 10 & 6 m (n = 26) while Warblers are more easily distinguished from the the mean estimated height of trees in which they sibling species than are females. Thus, female foraged was 15 f 6 m (n = 25). The mean ratio Golden-cheekedWarblers seenimperfectly might of foraging-height to tree-height was 0.72 f 0.17 not have been positively identified and would (n = 25), which is to say that thesewarblers were have gone unrecorded. This seems more likely found in the upper half of the trees in which they to have occurred before observers were aware foraged. Golden-cheeked Warblers foraged by that Golden-cheeked Warblers were a part of the gleaning from foliage and branches (n = 13) or region’s avifauna. Therefore we tested whether gleaning from foliage and branchesplus hanging- the results of the 1990-1991 season (18 males, to-glean from the underside of leaves (n = 2). 6 females)differed from the sightingsof the 199 l- We noted only two examples of aggressionin 1992 and 1992-1993 seasons(18 males, 9 fe- our sightings,even though in most cases,we spe- males). We found that the two subsets of our cifically looked for this behavior. In one case(11 sightings could have come from the same pop- December 199 l), a female Townsend’s Warbler ulation (x2 = 0.353, df = 1, 0.5 5 P 5 0.9), and displaced a male Golden-cheeked Warbler, and thus did not find a significant change through in the other (5 February 1992) a Crescent-chest- time in our ability to detect females relative to ed Warbler (Parula superciliosa) attacked a males. Golden-cheeked Warbler of unrecorded sex. Golden-cheeked Warblers occurred as mem- HABITAT USE bers of mixed-species flocks in 90% (n = 36) of We detected Golden-cheeked Warblers in all observations (seealso Braun et al. 1986, Johnson habitat types except milpas (Table 2). All detec- et al. 1988). In one instance, one male and one 688 R. M. VIDAL, C. MACIAS-CABALLERO AND C. D. DUNCAN

TABLE 2. Number of encounters(and individuals) of Golden-cheeked Warblers in different habitats of the Highlands of Northern Chiapas, Mexico, from August 1990 to December 1992, with measuresof effort where known.

Habitat Method Pine Pine-oak Oak Mixed Cloud Shrub Milpa

Casual observations l(l) 11 (9) 7 (6) 7 (7) l(1) 1 (1) 0 (0) Point counts 0 (0) 3 (4) 0 (0) 0 (0) -= 0 (0) 0 (0) no. of counts 31 142 61 32 41 16 Transects 5 (14) 4 (2) 2 (6) 3 (4) 2 (3) 0 (0) 0 (0) f(t) 0.14 0.09 0.05 0.09 0.08 0.00 0.00

mNot censused with this method.

female Golden-cheeked Warbler occurred to- than females. The true sexratio may be obscured, getherwithout other species,and in another case, however, by the presence of female-plumaged two males were seen together in the absence of immature males, and overlooked females. Our other species.For 34 cases,detailed data on flock results show that the sex ratio of the 1990-l 99 1 size and composition were recorded. Mean flock seasonwas homogeneouswith that of the 199 l- size was 16 f 13 (range 2-50). The mean number 1992 and 1992-1993 seasons when observers of Golden-cheeked Warblers per flock was 1.4 were more experienced. This suggeststhat fe- + 0.9. Speciesfound most commonly (more than males were not selectively overlooked. Indeed, 50% of the time) in flocks containing Golden- it is plausible that a greater percentageof males cheeked Warblers were Townsend’s Warbler, than females winter in Mexico, closer to the Hermit Warbler, Red-faced Warbler (Cardellina breeding grounds, and that a higher proportion rubrifons), and Solitary Vireo (Vireo solitarius) of females winter in the more southerly parts of (Table 3). the range, as occurs in other species(Ketterson and Nolan 1983, Morse 1989). DISCUSSION Habitat use, foraging and interspec@ inter- Distribution through the year. Our earliest fall actions. The diversity of habitats occupied by sightingsare consistentwith the earliest fall Chia- Golden-cheeked Warblers in the Highlands of passpecimen, 9 August 1950 (Pulich 1976). These Northern Chiapas is very different from their dates are noteworthy since other members of the specificityfor “cedar brakes” in the breeding sea- Black-throated Green complex do not arrive in son (cf. Kroll 1980) a situation known for other Chiapas until the end of August or early Septem- speciesas well (Askins et al. 1990, Hutto 1992, ber (Alvarez de1 Toro 1980; Macias-Caballero, Rabenold 1980). The discrepancy between the pers. observ.). Our late-March and April obser- small preference we observed for pine forests vations were somewhat surprising since the spe- (transect data) versus that for pine-oak forest cies arrives at the Texas breeding grounds by (point-count data) is probably unimportant be- mid-March (Pulich 1976). Nonetheless, Russell causethe number ofsightings is small. Moreover, (fide Pulich 1976) reported a male in northern the existence of differing micro-habitats within Mexico as late as 17 April. These late migrants, a broader forest type creates a basic incompati- the only April records outside Texas, may have bility between transect and point-count data hatched the previous year (Gauthreaux 1988, (Wiens 198 1). Morse 1989). Our other March records, from 3- Despite the increased diversity in habitat use, 9 March, represented birds that were probably the foragingbehavior we observedwas quite sim- able to reachthe breeding groundsby mid-March. ilar to that described by Pulich (1976) for for- These records concur with the observations and aging on the breeding grounds. We, too, found specimens from northeastern Mexico from 1 l- gleaning to be nearly the exclusive method em- 20 March (Pulich 1976, Johnson et al. 1988). ployed. Like Pulich, we did not observe flycatch- Distribution by sex. It is likely that the popu- ing or sallies-from-a-perch as foraging methods. lation of Golden-cheeked Warblers in the High- In fact, gleaning is the method employed by the lands of Northern Chiapas consistsof more males majority of Dendroica warblers during the winter GOLDEN-CHEERED WARBLER IN CHIAPAS, MEXICO 689

TABLE 3. Frequency as a percentageof occurrence, TABLE 3. Continued. and mean number (K f SD) of individuals of species in flocks(n = 34) containingGolden-cheeked Warblers Percent- in the Highlands of Northern Chiapas, Mexico, August Species s:: R 1990-December 1992. Ranking is by greatest percentageof flocks, followed by mean number of indi- Black Phoebe, Sayornis nigri- viduals. When these lead to equal rankings, species cans 3 0.03 k 0.17 sequenceis that of AOU (1983). Yellow-throated Vireo, Vireo flavtjrons 3 0.03 k 0.17 Percent- MacGillivray’s Warbler, Opor- ageof Species flocks R ornis tolmiei 3 0.03 t- 0.17 Golden-browed Warbler, Basi- Townsend’s Warbler, Dendroi- leuterusbelli 3 0.03 * 0.17 ca townsendi 91 4.4 k 5.1 Hepatic tanager, Piranga Java 3 0.03 k 0.17 Hermit Warbler, Dendroica occidentalis 53 2.7 ? 4.2 Red-faced Warbler, Cardellina rubrifrons 53 0.9 & 1.1 (Macias-Caballero 1993, Morse 1989). Our ob- Solitary Vireo, Vireo solitarius 53 0.7 ? 0.9 servation that wintering Golden-cheeked War- Wilson’s Warbler, Wilsonia pusilla 44 0.6 2 0.9 blers forage in the upper half of trees is at odds Crescent-chestedWarbler, Pa- with Kroll’s (1980) single sighting from Guate- rula superciliosa 41 0.6 ? 1.1 mala, but accordswell with Pulich’s (1976) find- Black-and-white Warbler, Mni- ings for the breeding season. While aggressive otilta varia 35 0.3 ? 0.6 Hutton’s Vireo, Vireo huttoni 29 0.4 * 0.7 competition for resourcesamong wintering mi- Olive Warbler, Peucedramus grant warblers is known (Macias-Caballero 1993, taeniatus 24 0.5 * 1.2 Greenberg et al. 1993) we observed very few Greater Pewee, Contopusperti- interspecific aggressiveinteractions and no in- nax 21 0.2 -c 0.4 traspecific ones. Black-throated Green Warbler, Dendroica virens 18 0.3 -t 0.8 Because we found the species almost exclu- Tufted Flycatcher, Mitrephanes sively among mixed flocks of passerines,we sug- phaeocercus 15 0.2 * 0.7 gest that searching such flocks may be the most TennesseeWarbler, Vermivora effective strategy for finding Golden-cheeked peregrina 15 0.2 * 0.5 Mountain Trogon, Trogon Warblers in other localities. If the density of mexicanus 15 0.1 * 0.4 Golden-cheeked Warblers in our study area is Hammond’s Flycatcher, Empi- typical of the rest of the wintering grounds, ob- donax hammondii 12 0.1 + 0.3 servers elsewhere will need to make the equiv- Painted Redstart, Myioborus alent of hundreds of point counts to assessthe pictus 9 0.1 2 0.4 Slate-throated Redstart, Myio- species’s abundance. The habitats in which we borusminiatus 9 0.1 * 0.3 found Golden-cheeked Warblers are not rare in Bushtit, Psaltriparusminimus 6 0.3 z!z1.3 the Highlands of Northern Chiapas, though con- Nashville Warbler, Vermivora version to milpa agriculture continues. The spe- r@icapilla 6 0.06 ? 0.24 Chestnut-sidedWarbler, Den- cies’s plasticity in wintering habitat use suggests, droica pensylvanica 6 0.06 * 0.24 however, that its current population decline is Gray Silky-flycatcher, Ptilogo- more likely a result ofhabitat losson the breeding nys cinereus 3 0.3 * 1.7 grounds than of changesin its wintering areas. Blackbumian Warbler, Den- droicajiica 3 0.06 * 0.34 White-eared Hummingbird, ACKNOWLEDGMENTS Hylocharis leucotis 3 0.03 * 0.17 We are grateful to the observers who participated in Garnet-throated Humming- the observations reported. The manuscript benefited bird, Lamprolaima rhami 3 0.03 rt 0.17 from helpful comments on earlier drafts by D. Becker, Spot-crowned Woodcreeper, D. Finch, and P. Vickery. We appreciatethe construc- Lepidocolaptesajinis 3 0.03 * 0.17 tive criticism offered by C. Thompson and A. Navarro. House Wren, Troglodytesae- Financial supportfor the transectstudies was provided don 3 0.03 f 0.17 by the Migratory Bird Center of the Smithsonian In- LeastFlycatcher, Empidonax stitution. The supportof the Centro de Investigaciones minimus 0.03 ? 0.17 Ecologicasde1 Sureste (Mexico) and the Consorciopara Empidonax sp. 0.03 + 0.17 la Conservacibny el Desarrollo Sustenablede1 Sureste 690 R. M. VIDAL, C. MACIAS-CABALLERO AND C. D. DUNCAN de Mexico is deeply appreciated.Duncan is pleasedto and conservation of Neotropical migrant land- birds. Smithsonian Institution Press,Washinaton, thank the Trusteesof the University of Maine System I for support during a sabbaticalleave. DC. Hurro, R. L., S. M. PLETSCHET,AND P. HENDRICKS. LITERATURE CITED 1986. A fixed-radiuspoint count method for non- breeding and breeding seasonuse. Auk 103:593- ALTMANN,J. 1974. Observational study of behavior: 602. samnling:methods. Behavior XLIX:227-267. JAHRSDORFER,S. 1990. Endangered and threatened ALVAR& DELTORO, M. 1980. Las aves de Chiapas, wildlife and plants; final rule to list the Golden- 2nd ed. Univ. Autonoma de Chiapas, Tuxtla Gu- cheekedWarbler as endangered.U.S. Dept. of In- titrrez, Mexico. terior, Fish and Wildlife Service.Fed Reg. 55 (249): AMERICANORNITHOLQGISTS UNION.’ 1957. Check- 53 153-53 160, 27 December. list of North American birds, 5th ed. American JOHNSON,K. W., J. E. JOHNSON,R. 0. ALBERT,AND Ornithologists’ Union, Washington, DC. T. R. ALBERT. 1988. Sightingsof Golden-cheeked AMERICANORNITHOLOGISTS UNION. 1983. Check- Warblers (Dendroicachrysoparia) in northeastern list of North American birds, 6th ed. American Mexico. Wilson Bull. 100:13 l-l 32. Omitholoaists’ Union. Washington. DC. KIZ~ZRSON,E. D., AND V. NOLAN. 1983. The evo- ASKINS,R. A., !. F. LYNCH,‘ANDR. &EE&FERG. 1990. lution of differential bird mieration. v. 357402. Population declines in migratory birds in eastern In R. F. Johnston [ed.], Cur&t ornithology, Vol. North America. Current Ornithology 7:1-57. 1. Plenum Press, New York. BRAUN,M. J., D. B. BRAUN,AND S. B. TERRILL. 1986. KROLL,J. C. 1980. Habitat requirementsof the Gold- Winter records of the Golden-cheeked Warbler en-cheekedWarbler: managementimplications. J. (Dendroicachrysoparia) from Mexico. Am. Birds Range Manage. 33:60-65. 40564-566. MAC?A.&ALIALLERO,C. M. 1993. Estudio de tres es- BREEDLOVE,D. E. 1981. Flora of Chiapas, part I: pecies de aves migratorias en una region de 10s introduction to the flora of Chiapas. California Altos de Chiapas:distribution, abundanciay con- Academy of Sciences,San Francisco. ductasocial. Licenciado en Biologiatesis, Instituto EMLEN,J. T. 1971. Population densities of birds de- de Ciencias v Artes de Chiapas, Tuxtla Gutierrez, rived from transect counts. Auk 88:323-342. Mexico. _ F~ANZ~EB,K. E. 1981. The determination of avian MILLER,A. H., H. FRIEDMANN,L. GRISCOM,AND R. J. densities using the variable-strip and fixed-width MOORE. 1957. Distributional checklist of the transect surveying methods. Stud. Avian Biol. birds of Mexico. Part 2. Pacific Coast Avifauna 6:139-145. 33. Berkeley, CA. GAUTHREAUX,S. A., JR. 1988. Age effectson migra- MORSE. D. H. 1989. American warblers. Harvard tion and habitat selection, p. 1106-l 115. In H. Univ. Press, Cambridge, MA. Ouellet [ed.], Acta XIX Congr. Intl. Omithol., PULICH,W. M. 1976. The Golden-cheeked Warbler. Univ. of Ottawa Press, Ottawa, Canada. Texas Parksand Wildlife Devartment. Austin. TX. GONZ&LEZ-ESPINOSA,M., P. F. QUINTANA-ASCENCIO,RA~ENOLD, K. N. 1980. The Black-throated Green N. RAM~W-MARCIAL, ANDP. GAYTAN-GUZMAN. Warbler in Panama:geographic and seasonalcom- 1991. Secondary successionin disturbed Pinus- parison of foraging, p. 297-307. In A. Keast and Quercusforests in the highlandsof Chiapas, Mex- E. S. Morton [eds.], Migrant birds in the Neo- ico. J. Vegetation Sci. 2:351-360. tropics. Smithsonian Inst. Press,Washington, DC. GREENBERG,R., C. MA&U CABALLERO, AND P. BICH- SCOTT,S. L. [ED.]. 1987. Field guide to the birds of IER. 1993. Defense of homopteran honeydew by North America, 2nd ed. National Geographic So- birds in the Mexican highlands and other warm ciety, Washington, DC. temperate forests. Oikos 63:5 19-524. SEXTON,C. 1992. A closer look: Golden-cheeked Hurro, R. L. 1992. Habitat distribution ofmigratory Warbler. Birding XXIV:373-376. landbird speciesin western Mexico, p. 221-239. WIENS, J. A. 1981. Scaleproblems in avian censusing. In J. M. Hagan and D. W. Johnston[eds.], Ecology Stud. Avian Biol. 6:5 13-52 1. GOLDEN-CHEEKED WARBLER IN CHIAPAS, MEXICO 691

APPENDIX I. Sightingsof Golden-cheeked Warbler in the Highlands of Northern Chiapas, Mexico, August 1990-December 1992.

Elev NIlI Date Habitat* (m) 8 P ?D Method’ Observe+ Commenr

1. 5Aug90 P-O 2,100 1 Cas PJB 2. llAug90 Oak 2,450 1 Cas PJB 3. 1 Sep 90 P-O 2,500 1 Cas PJB 4. 16 Ott 90 Oak 2,200 1 Cas CMC, RG 5. 31 Ott 90 Oak 2,450 1 Trans RMV 6. 24 Nov 90 Pine 2,250 5 Trans RMV 7. 27 Nov 90 Mix 2,200 1 1 Trans RMV 8. 12 Dee 90 Mix 2,200 1 Trans RMV Not distinct from #7 9. 17 Dec90 Pine 2,250 2 Trans RMV Not distinct from #6 10. 12 Jan 91 Mix 2,200 1 Cas CMC 11. 31 Jan 91 P-O 2,300 2 Cas CMC, JS 12. 22 Feb 91 Mix 2,200 1 Cas CMC 13. 26 Feb 91 Mix 2,200 1 Cas RMV Not distinct from # 12 14. 26 Feb 91 Cld 2,550 1 Cas CMC 15. 9Mar91 Mix 2,200 1 Cas CMC 16. 7Apr91 Oak 2,450 2 Cas CMC 17. 7Apr91 Oak 2,450 1 Cas CMC 18. 9Apr91 Oak 2,450 1 Cas CMC Not distinct from #17 19. 12 Apr 91 Cld 2,550 1 Trans CMC Not distinct from # 17 20. 13 Apr 91 Pine 2,250 1 1 Trans CMC d most likely distinct from #6 21. 24 Sep 91 Mix 2,100 1 Cas CMC 22. 28 Sep 91 P-O 2,400 1 Cas ERI, RG, CEG 23. 2Oct91 Pine 2,250 2 1 Trans CMC 24. 2 Ott 91 Mix 2,100 1 Cas CMC Not distinct from #21 25. 10 Ott 91 Mix 2,200 1 Trans CMC 26. 15 Ott 91 P-O 2,400 1 Trans RMV Not distinct from #22 27. 18 Ott 91 P-O 2,400 1 Cas ERI 28. 28 Ott 91 P-O 2,400 1 Trans CEG Not distinct from #22 29. 26 Nov 91 P-O 2,250 1 Trans CMC Not distinct from #23 30. 11 Dec91 P-O 2,400 1 Cas CMC Not distinct from #22 31. 10 Jan 92 Mix 2,200 1 Cas ERI 32. 22 Jan 92 Pine 2,250 1 Cas ERI 33. 22 Jan 92 Oak 2,200 1 Cas CMC 34. 31 Jan 92 P-O 2,300 2 Cas CMC, JS, PJB 35. 5 Feb92 Oak 2,150 1 1 Point ERI, jS 36. 3 Mar92 Cld 2,550 2 Trans CEG 37. 5 Mar92 P-O 2,400 1 Trans CEG Not distinct from #22 38. 31 Mar 92 Pine 2,250 1 1 Trans CMC 6 most likely distinct from #32 39. 21 Ott 92 P-O 2,100 1 Point CMC 40. 11 Nov 92 Oak 2,300 1 Point CDD, IB 41. 13 Nov 92 P-O 2,300 1 1 Point CMC 42. 29 Nov 92 P-O 2,500 1 Cas CDD, IB Not distinct from #39 43. 3 Dee 92 P-O 2,350 1 Point CMC 44. 4 Dee 92 P-O 2,310 1 1 Cas CDD 45. 12 Dee 92 P-O 2,310 1 Cas CDD, IB Not distinct from #44 46. 12 Dee 92 Shr 2,310 1 Cas CDD, IB Most likely distinct from #45 ’ ’ Habitat codes:P-O = pine-oakforest, Cld = evergreencloud forest, Mix = mixed vegetation,Shr = shrub.See text for morecomplete descriptions. b? = Sex unrecorded. EMethod codes:Gas = casualobservation, Tram = 40 m x 1 km transect,Point = IO-min, 25-m radiuspaint count. 1 Observers:Ilze Balodis,Peter Bichier, Philip J. Bubb,Claudia Ma&s-Caballero, CharlesD. Duncan,Claudia Elia GonAez, RussellGreenberg, EmestoRuelas-Inzunza, John Sterling,Rosa Maria Vidal. c Seetext for definitionsof “certainly distinct” and “most likely distinct” as applied to sightings.