Wilson Bull., 11 l(2), 1999, pp. 195-209

BEHAVIOR AND VOCALIZATIONS OF THE CAURA AND THE YAPACANA

KEVIN J. ZIMMERx*’

ABSTRACT-The first detailed information on the vocalizations (including the first sound spectrograms) and natural history of the Caura (Percnostolu caurensis) and the Yapacana Antbird (Myrmeciza disjuncta) are presented. The Caura Antbird was studied in the Serrania de la Cerbatana, edo. Bolivar, Venezuela, where it inhabits humid foothill forest dominated by large rocks. Caura Antbirds specialized in foraging on or beneath rocks, a behavior unusual among the Thamnophilidae. The Yapacana Antbird was studied at a site along the south bank of the Rio Ventuari, edo. Amazonas, Venezuela. These antbirds were locally abundant in a specialized stunted woodland that grows on white sand soils. Based on newly described vocal characters, the closest relatives of P. caurensis appear to be P. leucostigma and P. schistacea, whereas M. disjuncta has no apparent close relatives and probably merits placement in a monotypic . Received 9 July 1998, accepted 5 Jan. 1999.

Among the least known members of the STUDY AREAS AND METHODS large antbird family Thamnophilidae are the I observed Caura Antbirds lo-15 February, 1998 in Caura Antbird (Percnostola caurensis) and the Serrania de la Cerbatana near Hato Las Nieves the Yapacana Antbird (Myrmeciza disjuncta). (6” 34’ N, 66” 12’ W), edo. Bolivar, Venezuela (Fig. Both are nearly endemic to south- 1). The Serrania de la Cerbatana rings a large valley western Venezuela and have remained rela- vegetated mostly by a mixture of savanna and tropical dry forest, transected by narrow bands of gallery forest tively unobserved by modem field omitholo- along occasional streams, and dotted with groves of gists. The most extensive collections of both Mauritiu palms in poorly drained areas. Three main species (36 specimens of P. caurensis and 5 rivers, CaAo Las Nieves, Rio Agua Fria, and Rio Dan- specimens of M. disjuncta) reside in the Co- ta, drain the valley. The Serrania mountain range rises lecion Omitol6gia Phelps (COP), Caracas, dramatically from the valley, and on the southern and western borders is covered with humid forest. The Venezuela. Single specimens of each species south facing slopes of the mountains to the north of collected near Pica Neblina (edo. Amazonas, the valley are noticeably drier. The tallest peak in the Venezuela) in 1984 by Field Museum of Nat- chain is Pica Las Nieves at 2080 m. I studied Caura ural History (FMNH) personnel represent the Antbirds along a 1.5 km trail that began in humid for- only specimens of P. caurensis and M. dis- est at 280 m elevation and extended up the side of a juncta collected anywhere since 1972 and ridge to 400 m. I studied Yapacana Antbirds 22-27 February 1998 198 1 respectively. There is essentially no pub- along a trail (hereafter the “Picua Trail”) near the set- lished information on habitat or behavior of tlement of Picua (4” 5’ N, 66” 45’ W) in Yapacana Na- the two species, and nothing is known of their tional Park, edo. Amazonas, Venezuela (Fig. 1). Picua vocalizations (Ridgely and Tudor 1994). is a small settlement of Piaroa and Mako Indians, on In February 1998 I observed the habitats the south (left) bank of the Rio Ventuari at about 150 and behaviors and tape-recorded the vocali- m elevation. The surrounding area contains a mosaic of different soil types that support a patchwork of dis- zations of Caura Antbirds in the Serrania de tinct vegetation types. The Ventuari and its many small la Cerbatana, edo. Bolivar, Venezuela, and of tributaries are flanked by bands (of varying width) of Yapacana Antbirds in Yapacana National tall forest (15-25 m) that grow on yellow clay soils Park, edo. Amazonas, Venezuela. This is the and are seasonally flooded. This varzea forest is char- first detailed information on the natural his- acterized by a closed canopy with a fairly open un- tory and vocalizations of these species and al- derstory and an abundance of vines and lianas. On higher banks above the river the v&rzea grades into a lows a more informed assessment of their pos- taller transitional forest with a denser understory. sible generic affinities. Much of the area farther removed from the river is dominated by white sand soils on which grow lower stature woodlands and grassy savannas. The savannas ’ ’ Los Angeles County Museum of Natural History, range in size from less than 1 ha to about 1 km*, with 900 Exposition Blvd., Los Angeles, CA 90007. scattered shrubs and small trees. Larger savannas in ZCorrespondence address: 1665 Garcia Rd., Atas- the area contained stands of Mauritia palms. Large, cadero, CA 93422; E-mail: [email protected] isolated patches of red clay soils support tall (>30 m),

195 196 THE WILSON BULLETIN l Vol. I I I, No. 2, June 1999

mens, tape recordings, or photographs (Isler 1997; Fig. 1).

PERCNOSTOLA CAURENSIS N E U E L A ,b Distribution and habitat.-The Caura Ant- ri is known only from the western portion ff of the “Pantepui” region (Mayr and Phelps 1967) south of the Rio Orinoco in the Vene- r zuelan states of Bolivar and Amazonas and in COLOMBIA, extreme northern Brazil (Fig. 1). Percnostola caurensis is well represented in museum col- lections, with 36 specimens (the most recent collected in 1972) in the Colecidn Omitoldgia Phelps (Caracas, Venezuela) alone (C. Rodner, pers. comm.). The most recent substantiated record was of a male collected in 1984 at 1250 FIG. 1. Distributions of Caura Antbird (Percnos- m near the base of Pica Neblina (edo. Ama- tola caurensis) and Yapacana Antbird (Myrmeciza dis- zonas, Venezuela; Willard et al. 1991). The junctu): open circles = confirmed sites for P. caurensis occurrence of P. caurensis in the Serrania de (confirmation based on specimen records); open la Cerbatana represents a slight range exten- squares = confirmed sites for Myrmeciza disjuncta; sion to the northwest (Fig. 1). open star = site near the base of Pica de la Neblina where both species have been collected in close prox- The forest along the first 500 m of the trail imity. The black circle and #l indicates the site of the was tall (ca 30 m), with an open understory present study in the Serrania de la Cerbatana, edo. Bo- dominated by slender palms. The terrain was livar, Venezuela where P. caurensis was studied. The flat and nearly devoid of large rocks. I could black square and #2 indicates the site of the present locate only one territory of P. caurensis along study near Picua, in Yapacana National Park, edo. this portion of the trail, and it abutted the bot- Amazonas, Venezuela where M. disjunctu was studied. tom of the hill. The hillside forest beyond 500 m was also fairly open, with an intermittent canopy of about 20 m. Few trees were larger lush “islands” of humid tropical forest that are not than 30 cm dbh, and woody vines were abun- seasonally flooded (= terra firme forest). Quartzite dome-like sandstone hills (cerros) and low outcrop- dant. Large stands of a narrow-leaved, non- pings are scattered throughout the region. spiny bamboo (l-2 m in height) occupied Whenever individuals or pairs of P. caurensis or M. most light gaps. The entire slope was extreme- disjuncta were located, I followed them for as long as ly rocky, with numbers of boulders up to 8 m possible, tape-recording as many vocalizations as I tall and 1.5 m along their longest axis. These could and summarizing foraging and other behaviors boulders were typically moss and fern cov- on cassette tape. Some of these behaviors were also documented on videotape. I used tape-playback of ered, with terrestrial bromeliads, cacti, and these recordings to assess presence or absence of ant- bamboo growing over their tops and in the in places where no spontaneously vocalizing crevices (Fig. 2). Many were topped with birds were heard and to determine the limits of terr- small trees, the gnarled roots of which draped torial boundaries. All measurements included in such off the sides of the rocks like tendrils, trapping summaries (height above ground, territory size, dis- leaf litter and organic debris. I located 6 pairs tances, times, etc.) are estimates. Terminology for for- of P. caurensis along about 1 km of trail aging behavior follows Remsen and Robinson (1990). Tape recordings were made with a Sony TCM-5000 through this rocky, hillside forest. recorder and Sennheiser MKH-70 microphone. All re- The Serrania de la Cerbatana was extremely cordings have been or will be archived at the Library dry in February 1998. Typical dry season con- of Natural Sounds, Cornell Univ., Ithaca, New York. ditions appeared exacerbated by ongoing El Spectrograms were made by Phyllis Isler on a Power Nifio related events. Leaf litter throughout the Macintosh 7500 computer using Canary version 1.2.1 forest was extremely dry and many trees had (Bioacoustics Research Program, Cornell Laboratory of Ornithology, Ithaca, New York). Morton Isler com- shed large numbers of leaves. This was par- piled a comprehensive list of distributional records of ticularly evident in the hillside forest, where the two species of antbirds as documented by speci- large patches (0.5-l ha) of deciduous vege- Zimmer l CAURA AND YAPACANA ANTBIRDS 197

FIG. 2. Rocky hillside forest in the Serrania de la Cerbatana, edo. Bolivar, Venezuela. (A) Relatively open forest, with a broken canopy of about 20 m. The rocks were 2-3 m tall and 4 m in diameter. (B) A rocky alluvial fan along which were located two Caura Antbird territories. Note the highly deciduous state of the vegetation in this light gap, and the abundance of leaf litter trapped in the roots and vines overtopping the rocks. tation were conspicuously scattered across the individual antbirds. The loudsong of P. cau- slope, usually coincident with the rock strewn rensis is a far carrying series of 7-15 modu- alluvial fans at the bottom of ravines. lated and well spaced notes (Fig. 3A). The Morphology.-Soft-part colors were iden- first notes are widely spaced, and the terminal tical for both sexes. The iris was reddish- notes are closer together and drop in pitch. brown, the legs and feet were slate gray (a Female songs (Fig. 3B) were similar in pattern shade paler than the bill), and the bill was to male songs, but differed in other character- blackish. Plumage was as described by Ridge- istics such as mean number of notes, mean ly and Tudor (1994). frequency, etc. Females sang less frequently Vocalizajions.-Caura Antbirds were gen- than males. erally quiet during my fieldwork, as were In response to tape playback and during ter- most other species of insectivorous birds. ritorial encounters, both sexes gave loud, Dawn choruses were both unremarkable and buzzy “zhew” calls at varying levels of fre- short, suggesting a low level of breeding ac- quency modulation (Figs. 3C-E). On a few tivity for most species during the height of the occasions, birds involved in territorial dis- dry season. There was a sustained rain during putes uttered sharp “quip” notes (Fig. 3F) the early morning hours of 13 February; the when neighboring birds approached closely. two following mornings I noted increased The most frequently heard vocalization, and spontaneous song from Caura Antbirds. one given by birds startled along the trail, was I recorded over 120 loudsongs (as defined an abbreviated loud rattle (Figs. 3G, H), sim- by Isler et al. 1997) and 900 calls from 12 ilar to the alarm calls of P. leucostigma and 198 THE WILSON BULLETIN - Vol. III, No. 2, June 1999

7 A 6-

i 0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0

blc F

5-

41

3-

2-

l-

” I I I 0 oi5 110 115 2.0 2 3.0 3.5

7 H 6- G

5- I! 4-

3-

2-

l-

0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 0

Time (seconds)

FIG. 3. Spectrograms of Cam-a Antbird (Percnostola caurensis) vocalizations: (A) male loudsong, (B) female loudsong, (C, D, E) “zhew” calls, which probably function as aggression calls, at various levels of frequency modulation, (F) “quip” call, (G) male alarm rattles, and (H) female alarm rattles.

several species of Myrme&u antbirds (pers. ence gave this call immediately before flush- obs.). A loud “chikit” (not tape recorded) was ing, as did birds chasing one another about in often given by Caura Antbirds at the moment an aggressive boundary dispute. they took flight. Birds alarmed by my pres- Behavior.--Caura Antbirds were encoun- Zimmer - CAURA AND YAPACANA ANTBIRDS 199 tered singly or in pairs, but did not associate casional longer wing-assisted hops or short, with mixed-species flocks. Mates foraging to- abrupt flights, often to low, overhanging gether were typically within 15 m of one an- branches or ledges. They frequently hopped other, and alarm calls from one bird elicited from a rock up to a low branch or sapling to an immediate vocal response from the other. scan for l-5 s before dropping back to the Foraging birds maintained a nearly horizontal rock. Birds often dropped l-5 m from the posture, with the head held higher than the rocks to the ground. There, they hopped be- axis of the body. All individuals continuously neath the overhangs, probed in leaf litter and raised their tails a few degrees above the plane inspected rock surfaces. Antbirds spent many of the body and then wagged them .slowly minutes inspecting fissures, crevices, and gaps downward in an arc 20-30” below the plane within and between rocks, often retracing their of the body. This tail movement was most ex- routes. aggerated when the antbirds clung laterally to One male observed foraging for more than elevated perches or responded to tape play- 30 minutes spent the bulk of this time forag- back, but was also used during terrestrial for- ing over, under, and between rocks, probing in aging. Mostly the tail appeared to be slightly moss and vine tangles. On three occasions the fanned. Tail movements were occasionally ac- bird dropped to the ground at the base of large companied by a simultaneous wing-flick. rocks and spent l-5 minutes vigorously toss- Caura Antbirds foraged mostly on rocks or ing dead leaves in the manner of a leaftosser the ground. Both sexes spent long times (Sclerurus spp.). Large leaves (many larger creeping over the large boulders, often cling- than the bird) were picked up with the bill and ing laterally to nearly vertical rock faces and tossed. Smaller leaves were frequently flipped going in and out of the numerous crevices (of- by inserting the bill beneath the leaf and then ten for minutes at a time) in the manner of giving a quick upward flaking motion. He also Slaty Bristlefronts (Merulaxis ater; M. Isler made an upward sally of 20-25 cm from the and I? Isler, pers. comm.), Rock Wrens (Sal- ground to take an unidentified arthropod from pinctes obsoletus) or Canyon Wrens (Cather- the underside of a green leaf. Other individ- pes mexicanus). While creeping about the uals occasionally made similar short upward rocks, the antbirds frequently probed in the sallies to glean prey from overhanging rocks. mosses and small ferns covering the surface, One female antbird spent several minutes but spent most of their time inspecting the leaf hopping around the periphery of a large emer- litter trapped between the roots and vine tan- gent swarm of small, winged ants. Although gles of trees overtopping the boulders. Arthro- she picked at the ground several times, it was pod prey (primarily orthopterans and hemip- not clear whether she was feeding on the ants. terans, as well as many arthropods too small The bird flew off and returned to the ant- to be identified) were gleaned from root and swarm twice. The male antbird was foraging vine surfaces with quick stabbing motions. nearby, but did not attend the swarm. Curled dead leaves were carefully probed with Territories appeared to be about 150-200 m the bill. Some antbirds inspected dead leaves in diameter. I witnessed only one territorial without tossing them, others picked up leaves conflict. This was a prolonged encounter be- with the bill before tossing them aside. The tween pairs whose territories bounded a large antbirds routinely squeezed themselves into rock-slide along the center of a ravine. The small spaces between the rock surface and conflict was marked by several advances and overlying roots and vines, remaining in these retreats by both pairs. Both members of each somewhat “canopied” niches to forage for up pair countersang at length while gradually ap- to 60 s. When foraging on rocks, the antbirds proaching their counterparts. Whenever the tended to spend most of their time in some- pairs approached to within about 20 m of one what protected locations, within the interior of another, they tended to substitute “zhew” vine tangles and root masses overtopping the calls for songs. When they were in visual con- rocks, within crevices in the rocks, or beneath tact or close auditory contact, the pairs ex- rocky ledges and overhangs. The antbirds changed harsh calls for a few minutes before moved steadily over open rock faces, pro- one individual or pair retreated, often with its gressing by short hops of 5-10 cm and oc- rival in close pursuit. These abrupt retreats 200 THE WILSON BULLETIN * Vol. 111, No. 2, June 1999 were always preceded by the loud, hard base of Cerro Yapacana (edo. Amazonas, Ven- “chikit” call. The dispute lasted for more than ezuela) in April 193 1. Five more specimens 30 minutes. Eventually the pairs retreated to were collected from the same general locality opposite ends of the rockslide, periodically in April-May 1947 (specimens COP). Cerro singing. Yapacana is an isolated outlier of the western Most species of besides P. cau- Tepuis, rising steeply above the Rio Orinoco rensis were encountered in mixed-species can- to an elevation of 1340 m. Meyer de Schauen- opy and mid-level flocks. The apparent ab- see and Phelps (1978:218), perhaps describing sence of terrestrial or semi-terrestrial antbirds the general habitat surrounding Cerro Yapa- (other than P. caurensis) or furnariids is note- cana, listed the habitat of M. disjuncta as worthy, although my failure to detect such “High rain forest at about 100 m in under- species could be an artifact of depressed song growth and low bushes.” Subsequent to its activity during the dry season. The taller, less- description, M. disjunctu has been document- deciduous forest below the mountains had a ed from only three additional sites (Fig. 1). In more diverse avifauna. March 198 1, while working in sandy-belt for- The presence of large rocks appeared to be est near Puerto Inirida, depto. Guainia, Co- a critical component of P. caurensis habitat. lombia, J. Dunning mist-netted an antbird lat- Of the 7 territories, 6 were in areas with abun- er identified from photos as a female Yapa- dant large rocks. The seventh territory was cana Antbird (Hilty and Brown 1986; ANJP close enough to the bottom of the hillside to specimen 175723, R. Ridgely, pers. comm.; possibly have included some rocky terrain as photo on file at VIREO). In February 1984 a well. The rocks provided microhabitats in single female M. disjuncta was collected at which a diversity of plants flourished, and 140 m near the left bank of the Rio Barfa on these plants, in turn, provided a wealth of po- the Venezuelan-Brazilian border (Willard et tential foraging strata for a small terrestrial in- al. 1991). This extended the known range of sectivore such as P. caurensis. Caura Antbirds M. disjunctu about 350 km south. There is no spent more than 80% of their foraging time published description of the habitat in which on rocks and overtopping vegetation, or on the this Yapacana Antbird was collected, but the ground directly beneath overhanging rocks. nearby base camp was in “tall seasonal rain The importance of rocks as foraging sites for forest drained by both black-water and white- Caura Antbirds may be increased during the water streams” (Willard et al. 1991). J. Coons dry season, when favorable moisture-retaining and D. Stejskal (pers. comm.) were the first to microclimates are created along rock edges or find M. disjuncta near Picua in January 1997. crevices. Furthermore, the tendency for over- They reported seeing or hearing several indi- topping root masses and vine tangles to trap viduals in savanna woodland on white sand leaf litter is accentuated during the dry season, soils along the Picua Trail. when many trees drop their leaves. Such ac- The only habitat of the Ventuari in which I cumulations of organic litter may provide at- found M. disjuncta was what the local people tractive sites for arthropods when the forest is water-stressed. It is interesting that no other refer to as “monte cerrado.” I found this hab- bird species was observed to exploit the extra itat only on the south bank of the Rio Ventuari resource dimensions created by the large along the Picua trail. This is a stunted, virtu- rocks. The presence of Guianan Cock-of-Rock ally impenetrable woodland that grows on (Rupicolu rupicola) in the area is almost cer- fine, compacted white sand soils that are sea- tainly dependent on the availability of large sonally saturated (Fig. 4). It is similar to the rocks for nest sites (Snow 1982), but only the “savanna woodland” described from Cam- Caura Antbirds seemed to use the rocks as pamento Junglaven located farther north (up- foraging substrates. stream) along the Ventuari (Zimmer and Hilty 1997) but has a greater density of vines, along MYRMECIZA DISJUNCTA with abundant sawgrass and bamboo scattered Distribution and habitat.-The Yapacana through the understory. The canopy varies Antbird was described in 1945 by H. Fried- from 6-10 m and is of generally uniform mann from two specimens collected near the height with only occasional emergent trees of Zimmer l CAURA AND YAPACANA ANTBIRDS 201

FIG. 4. (A) Monte cerradowoodland along the PicuaTrail, YapacanaNational Park, edo. Amazonas, Ven- ezuela. (B) Interior of the monte cerrado, depicting a typical Yapacana Antbird territory. lo-15 m. Few trees in this habitat have trunks Antbirds along more than 150 m of trail tran- thicker than 10 cm dbh. secting this more grassy woodland. A different type of “sandy belt forest” or Morphology.-There has been some con- savanna woodland occurred on the north bank. fusion in the literature regarding plumage This woodland was partly deciduous, less di- characters of M. disjuncta. Central to the con- verse, and even more stunted than the monte fusion is the type specimen of M. disjuncta, cerrado. The understory was more open and an immature male molting into adult plumage lacked both bamboo and sawgrass. This forest (Friedmann 1945, 1948), which displayed a grew on coarser, well-drained white sand soils combination of adult and immature plumage atop low ridges or rocky outcroppings. Myr- characters. Thus, Meyer de Schauensee (1970: meciza disjuncta and many other species typ- 249) wrote that the male has “upperparts ical of the monte cerrado were absent from blackish-gray, crown and nape tinged brown” this scrub woodland. and “sides of head gray, chin white; rest of Using tape playback I located at least 24 underparts white, strongly tinged ochraceous pairs of M. disjuncla along 1350 m of the Pi- buff; center of abdomen white.” Similarly, cua Trail. No birds were detected farther than Meyer de Schauensee and Phelps (1978:218) 50 m from the trail. Territories were evenly described the male plumage as having the spaced along both sides of the trail and ap- “throat and breast white suffused with ochra- peared to be no more than 50-75 m in di- ceous tawny, strongly so on breast and sides ameter. Near the savanna edge the monte cer- of throat, middle of abdomen white, undertail rado was particularly stunted, with a more coverts dark gray.” They described the female open canopy and more sawgrass in the under- as differing from the male by “dark ochra- story. I found only two territories of Yapacana ceous buff spots on wing coverts and ochra- 202 THE WILSON BULLETIN l Vol. 111, No. 2, June 1999

ceous under tail coverts.” This description neous singing, suggesting that the level of was essentially repeated in Hilty and Brown breeding activity was low. However, birds (1986). Ridgely and Tudor (1994:332) were were highly responsive to tape playback, much more nearly correct in both their illus- which often elicited singing from one or more tration of the male and in the description of neighboring pairs in addition to the resident the plumage, but still suggested that the un- pair. Both males and females responded vo- derparts of the male are “tinged with creamy cally to playbacks, although males were much buff.” more aggressive in approaching the speaker. None of the 20 or more male Yapacana All types of vocalizations that I recorded were Antbirds that I saw showed any hint of tawny, given by both sexes of Yapacana Antbirds. I ochraceous buff, or creamy buff color on the tape recorded over 350 loudsongs and 230 underparts, nor did they have the crown or calls from 34 individuals. nape tinged brown. Instead they were uni- The typical loudsong of M. disjuncta con- formly dark gray (with an almost steely blue sisted of two prolonged, harsh, heavily fre- cast) above (except for the usually concealed quency modulated notes separated by a short white interscapular patch) and on the sides of pause, into which were inserted one or two the face, with the chin, throat, breast, belly, “pip” notes (Fig. 5A). The first harsh note and undertail coverts white. The sides and was the longest, and increased in intensity flanks were washed with gray, but these areas while rising in pitch. The second harsh note were often concealed by the wings. The wing was shorter and had a more uniform ampli- coverts were blackish, with the lesser and me- tude. A typical loudsong could be transcribed dian wing coverts fringed white. The tail was as “cchhhhhhh pipizhhhh”. On many occa- blackish and the rectrices were either narrowly sions birds sang. songs with no discernible tipped or fringed white. This character is not “pip” notes between the harsh elements, al- mentioned in any of the previous descriptions, though spectrograms of such songs reveal a possibly because the narrow white tips/fringes distinct spike at the beginning of the second were lost to wear in the few existing speci- harsh note (Fig. 5B). Less frequently birds mens. The legs were pinkish gray, the iris sang songs with three harsh elements instead blackish brown, the maxilla blackish, and the of two (Figs. 5C, D). The third harsh note in mandible whitish. Females differed in having such series was usually the shortest. Loud- a slightly brownish cast to the upperparts, par- songs varied in duration, depending largely on ticularly the crown and nape, which contrasted the number of harsh elements included. I with the gray sides of the face. The underparts could find no consistent differences between were a bright ochraceous buff (almost pale or- male and female songs, although males more ange) from the throat to the undertail coverts, frequently inserted two “pip” notes between and were only slightly paler on the chin and harsh elements and females more frequently upper throat. I did not note pale tips or fringes sang songs with no “pip” notes. on the rectrices of females, nor could I con- The most commonly heard calls were long, firm the presence of a contrastingly colored harsh, single notes, at a somewhat higher pitch interscapular patch. than the harsh elements of the song, and with All birds that I saw had proportionately a peculiar, slightly nasal quality (Figs. 5E, F). short, broad tails that appeared graduated. These “harsh calls” were given by both sexes, This could have been influenced by molt, but, although those of females were higher if so, all of the individuals in this area were pitched. “Harsh calls” seemed to be aggres- highly synchronous in the regrowth of their sion calls given in response to tape playback outer rectrices. or by a bird disturbed by my presence. Birds Vocalizations.-J. Coons and D. Stejskal occasionally gave a soft rattle (Fig. 5G) in a (pers. comm.) tape recorded some single-note similar context. When Yapacana Antbirds of harsh calls of M. disjuncta during their Janu- either sex were suddenly startled or strongly ary 1997 visit to Picua, but did not encounter agitated, they gave one or more sharp and singing birds. Yapacana Antbirds were also somewhat squeaky “squip” notes (Fig. 6A). generally quiet during the period of my field These notes were similar in tonal quality to work. Even at dawn there was little sponta- the notes inserted between the primary harsh Zimmer l CAURA AND YAPACANA ANTBIRDS 203

0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5:5 6.0

7 C 6

31 m

50, I I I I I I / , I I e 0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0

5-

4-

3-

2-

lm

0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0

7 E F G 6-

5-

4-

0 / , I I 0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 Time (seconds)

FIG, 5. Spectrograms of Yapacana Antbird (M.vrmecizu rli.~j~~-fa) vocalizations: (A) male loudsong with two “pip” notes inserted between the primary harsh elements, (B) loudsong variant with no discernible “pip” notes between harsh elements (sex of singing bird unknown), (C) loudsong variant with three harsh elements and single “pip” notes (sex of singing bird unknown), (D) loudsong variant with three harsh elements and no “pip” notes (sex of singing bird unknown), (E) male harsh call, probably an aggression call, (F) female harsh call, probably an aggression call, and (G) soft rattle call, given in an aggressive context. 204 THE WILSON BULLETIN l Vol. 1 II, No. 2, June 1999

7

0 I I I I I I I , I / I 0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0

Time (seconds)

FIG. 6. Spectrograms of Yapacana Antbird (Myrmeciza disjuncta) calls: (A) “squip” notes given by a startled or alarmed bird, (B) rarely heard complex call (sex of calling bird unknown), and (C) variant complex call (sex of calling bird unknown).

elements of most loudsongs. Occasionally what jerky manner. Such motions involved the birds of either sex gave a more complex call entire tail being swung a few degrees away that began with several “squip” notes and from the axis of the body, held briefly in that ended with a descending series of soft, whis- position, then swung still further in the same tled “wheee” or “whew” notes (Figs. 6B, C). direction before being swung back into align- These complex calls had a distinct tailing off ment. This jerky motion is similar to some of quality, as in “squip squip squip wheee wheee the tail movements employed by the Silvered wheee whew whew.” The function of these Antbird (Sclateria naevia; pers. obs.). Wheth- calls was not clear. er foraging or singing, Yapacana Antbirds Behavior and sociality.-Yapacana Ant- flicked their wings at least once during virtu- birds were encountered singly or in pairs and ally every pause between hops. Wing-flicks did not associate with mixed-species flocks. occurred both independent of and in synchro- Given the relatively small size of their terri- ny with tail movements. Singing birds rarely tories, members of pairs were rarely far from sang consecutive songs from the same perch. their mates. An alarm call or song from one In response to tape playback males often ex- bird almost invariably brought an immediate posed a white interscapular patch. vocal response from its mate. I did not witness Yapacana Antbirds foraged mostly on or any confrontations between neighboring pairs near the ground, always lower than 1.5 m. of antbirds, although on several occasions a They were restless, active foragers, moving by singing pair of birds stimulated an adjacent short hops (often wing assisted) and seldom pair to approach the apparent boundary and pausing for more than 2 s in one spot. When countersing for several minutes. moving above the ground they clung laterally Yapacana Antbirds typically maintained a to slender vertical saplings or perched across horizontal posture, with the head held higher horizontal limbs and vines, progressed in an than the axis of the body. Singing birds usu- often erratic, zigzag course, and frequently ally maintained a more upright posture. The moved up and down. They were adept at tail was held within a few degrees above or clinging to the thinnest stems, including slen- below horizontal, and often was kept slightly der bamboo stalks and vines. On a few oc- fanned. Foraging birds often quickly flicked casions I saw birds hop headfirst down nearly the tail up and down in a shallow arc of less vertical stems or branches, almost in the man- than lo”, but just as frequently dipped the tail ner of a nuthatch (Sitta spp.). Birds frequently slowly downward at about a 30” angle before took several hops on the ground before jump- flicking it back up more rapidly. Singing birds ing up to a low perch and then back down to frequently shivered the tail up and down more the ground. Small arthropod prey were rapidly throughout a song. Some individuals gleaned from stems and from tops and bot- wagged their tail sideways in a slow, some- toms of live leaves by reaching out, up, or Zimmer * CAURA AND YAPACANA ANTBIRDS 205

down on extended legs and with neck craned. turned to a low perch to beat the prey on a Prey were captured with a quick stabbing mo- branch before swallowing. Slightly more time tion of the bill and swallowed entire. Larger was spent hopping on the ground between the prey items were bashed against the perch and columns of ants and tossing dead leaves in the mandibled one or more times before being manner of a leaftosser (Sclerurus spp.). Most swallowed. Antbirds typically wiped their bill leaves were tossed by inserting the bill be- on the perch after swallowing prey. Most prey neath the leaf and lifting it with a quick flak- items that could be identified were small or- ing motion. Occasionally the bird picked up a thopterans (katydids and crickets), hemipter- leaf in its bill and tossed it aside. David Wolf ans, and geometrid larvae. (pers. comm.) observed another female ant- I encountered Yapacana Antbirds foraging bird (away from ants) that remained in one in the open along the main trail on only three spot tossing leaves in a similar manner for l- occasions. Two of these encounters involved 2 minutes. pairs, and the other involved a lone female I observed no other species of terrestrial or plumaged bird. In each case the birds were semi-terrestrial antbirds in the monte cerrado. working the edge of the dense monte cerrado The Black-throated Antbird (Myrmeciza atro- vegetation, as well as shrubs and clumps of thorax) and the Black-chinned Antbird (Hy- grass growing in parts of the trail. All foraged pocnemoides melanopogon), both of which mostly on the ground, progressing by a series routinely forage below 1.5 m (Hilty and of short hops with minimal pauses in between, Brown 1986; pers. obs.), were locally com- and always with wings (and frequently the mon in nearby forest or edge habitats, but tail) flicking. The most frequent attack ma- were not found in the monte cerrado. The only neuvers were gleans from the surface of the other (besides M. disjuncta) in this leaf litter or brief probes with the bill beneath habitat that I found foraging below 1.5 m was the leaf litter. The next most frequent tech- the Buff-breasted Wren (Thryothorus leuco- nique was reaching up to glean from the un- tis), which foraged everywhere from the dersides of overhanging green leaves and ground to the canopy. grass blades. On several occasions birds jumped 6-15 cm upward to glean prey from DISCUSSION the undersides of leaves. The two pairs of ant- Habitat and conservation.-My field work birds encountered in the open were found in indicates that both the Caura Antbird and the the early morning before the sun had illumi- Yapacana Antbird are habitat specialists, oc- nated the trail. The lone female plumaged bird curring in subtypes of more widely distributed (possibly a subadult male) was found in mid- macrohabitats. Percnostola caurensis has morning, when the entire trail was sunlit and been recorded over a broad elevational range, temperatures were already above 30” C. I fol- 100-1300 m (Meyer de Schauensee and lowed this bird as it foraged steadily at the Phelps 1978). In the Pantepui region this edge of the woodland for more than 20 min- range of elevations often spans the distance utes and covered more than 50 m. It crossed between tall, seasonal humid forest and elfin at least one known territorial boundary, but cloud forest. An antbird that occurs across remained silent and did not attract attention such a spectrum of habitats might normally be from any other antbird. considered an ecological generalist. However, On one occasion I found a female Yapacana if the critical ecological factor determining its Antbird attending a foraging swarm of army distribution is the presence of large rocks ants (Eciton sp.) within the monte cerrado. I within forest regardless of elevation, then P. observed this bird over 20 minutes during caurensis is very much a specialist. This which it was the only bird attending the could, in part, account for the absence of the swarm. The female antbird employed two species from so many seemingly suitable low- strategies in the vicinity of the ants. Part of land sites in Bolfvar and Amazonas (Campa- the time she scanned the swarm from perches mento Junglaven: Zimmer and Hilty 1997; within 0.3 m of the ground, dropped to the Brazo Casiquiare: Paynter 1982). ground to seize fleeing arthropods (orthopter- Percnostola caurensis may have evolved as arts, hemipterans, and spiders) and then re- something of a rock-specialist to occupy a 206 THE WILSON BULLETIN l Vol. Ill, No. 2, June 1999

niche that is locally abundant in parts of the nerable than most other birds of the Guianan highly eroded Guianan Shield. Mayr and lowlands. Stotz and co-workers (1996:4) have Phelps (1967:277) described the tepuis (table noted that “. . . the first major waves of ex- top mountains) of this region: “many are ac- tinctions in the Neotropics are not occurring tually strongly dissected and strewn with iso- in centers of diversity such as the Amazon. lated blocks, some more than 100 m high, and Rather, extinctions are occurring within cen- with a large variety of other rock forms.” ters of local endemism, especially among spe- Most Caura Antbirds have been collected cies that have evolved ecological specializa- from the slopes of tepuis (Zimmer and Phelps tions that limit their ability to adapt to human 1947, Phelps and Phelps 1963, Meyer de modifications of their habitats.” Schauensee and Phelps 1978). The most re- Before the conservation threats facing P. cently collected specimen, from 1250 m at the caurensis and A4. disjuncta can be adequately base of Pica Maguire (edo. Amazonas, Ven- assessed, we must first confirm their depen- ezuela), was from a site described as “A dence on or preference for the microhabitats hanging valley in dense cloud forest with in which I found them, then attempt to quan- moderately tall trees, and rocky forest floor tify just how much appropriate habitat exists covered with thick moss” (Willard et al. 1991: within their ranges. More comprehensive sur- 6). The apparent absence of P. caurensis from vey work within the region is clearly needed. much of the eastern portion of the tepui region Continued protection of existing parks or re- of Bolivar is a mystery. serves, such as Yapacana National Park, is vi- Myrmeciza disjuncta appears to be restrict- tal, particularly with regard to threats posed ed to woodlands growing on white sand soils. by illegal gold-mining. White sand habitats are widely but patchily Intrafamilial relationships.-The Caura distributed throughout Amazonia, with their Antbird was described by Hellmayr (1906) center of distribution in the upper Rio Negro and placed in the genus Sclateria. The Caura region (Pires 1974, Stotz et al. 1996). White Antbird was subsequently transferred to Schis- sand soils support many different types of tocichla (Zimmer and Phelps 1947), which vegetation, from scrub to tall forest (Anderson was later subsumed into Percnostola without 1981). Several distinctly different types of elaboration by Peters (195 1). Subsequent au- vegetation were found growing on white sand thors (e.g., Meyer de Schauensee 1966, 1970; soils in the Picua region but I found M. dis- Sibley and Monroe 1990; Monroe and Sibley juncta only in the monte cerrado. Similarly, 1993) have continued this treatment, recog- surveys of two nearby sites in Amazonas nizing five species: P. rufifrons, P. schistacea, (Campamento Junglaven and Pto. Ayacucho) P. leucostigma, P. caurensis, and P. lophotes. have failed to record M. disjuncta in spite of Ridgely and Tudor (1994: urrected the ge- the prevalence of white sand woodlands (Zim- nus Schistocichla for schista ‘a, leucostigma, mer and Hilty 1997). The Yapacana Antbird and caurensis on the basis of their rounder, may therefore be restricted only to a particular uncrested heads (crested in P. rufifons and P. type of white sand woodland, the monte cer- lophotes) and spotted rather than fringed wing rado. This would indicate an even patchier coverts. distribution than previously suspected and On purely morphological grounds, P. cau- would help explain how this species has es- rensis, P. leucostigma, and P. schistacea caped detection for so long. would appear to comprise a natural grouping. Both the Caura Antbird and the Yapacana Plumage differences between P. caurensis and Antbird appear to be locally common within P. leucostigma are especially subtle, with size their preferred habitats. The distributions of and soft part coloration being the most im- both species are centered in the lowlands or portant field characters for visually distin- foothills of Amazonas and western Bolfvar, guishing the two species (pers. obs.). Vocal which are among the least populated regions similarities are less apparent, in part because in Amazonia. Thus, neither species is under of pronounced geographic variation in the vo- immediate threat of extinction. However, their calizations of the various named subspecies of patchy distributions and apparent restriction to P. Zeucostigma (pers. obs.). Indeed, some vo- particular microhabitats make them more vul- cal differences within the P. leucostigma com- Zimmer l CAURA AND YAPACANA ANTBIRDS 207

plex are as great as the between species dif- they are distant in all respects from M. dis- ferences in the Schistocichlu group. Given junctu (pets. obs.). In some morphological re- this, I feel that resolution of the intrageneric spects, M. disjuncta is reminiscent of the relationships of the five species currently in- monotypic Sclateria, as suggested by Ridgely cluded in Percnostola should await molecular and Tudor (1994), but it has fringed rather comparisons, as well as a closer evaluation of than spotted wing coverts, white tail tips, an vocal and morphological differences as they interscapular patch, and differs greatly in both relate to the P. leucostigma group. vocal and behavioral characters. The song of Almost since its description, there has been M. disjuncta is mildly reminiscent of that of speculation regarding the placement of the Ya- various members of the Cercomacra nigricans pacana Antbird in Myrmeciza. Friedmann group (as defined by Fitzpatrick and Willard (1948:478) offered that “the species is not too 1990), and the plumage pattern of males, distantly related to Myrmeciza atrothorax but largely gray with white-fringed wingbars and is clearly specifically distinct from that form.” white tail-tips, and females ochraceous below, He went on to note that Zimmer had examined fits several members of the C. tyrunnina the type and the paratype and had pointed out group. However, Cercomacra antbirds tend to that “the general plumage has about the tex- be slender and proportionately long-tailed ture of Cercomacra carbonaria” (Friedmann (Ridgely and Tudor 1994), whereas M. dis- 1948:478). Friedmann (1948) further noted juncta is relatively compact and short-tailed. Zimmers’ suggestion that a fully adult male No Cercomacra approaches the white under- M. disjuncta might show a closer relationship parts of male M. disjuncta; the genus as a to Cercomacra than was suggested by the type whole has gray or blackish underparts. More specimen. He also commented that “The pat- importantly, male-female antiphonal duets are tern of the markings of the upperwing coverts an important component of the vocal reper- is very like that of some forms of Cercomacra toires of virtually all species of Cercomacra (sewa for example), but the bill is that of (Zimmer et al. 1997), but are not found in M. Myrmeciza” (Friedmann 1948:478). Peters disjuncta. (195 1) alluded to the seemingly polyphyletic In size, proportions, and some aspects of nature of Myrmeciza as he defined it and made plumage Myrmeciza disjuncta is somewhat several recommendations for the placement of suggestive of Hypocnemoides. However, nei- various species, but did not mention M. dis- ther species of Hypocnemoides is nearly as junctu. Ridgely and Tudor (1994:333) also sexually dimorphic as is M. disjuncta, and noted the heterogeneous nature of Myrmeciza they lack any suggestion of the ochraceous and suggested specifically that M. disjuncta coloration found in female M. disjuncta. Myr- may not belong in the genus and “perhaps is meciza disjuncta lacks the black throat and more closely allied to Sclateria”. They also pale eye found in Hypocnemoides. Vocal dif- seemed to suggest somewhat indirectly, that ferences between M. disjuncta and Hypocne- M. disjuncta was or should be included in the moides are much greater than the morpholog- formerly recognized genus Myrmoderus. ical differences (pers. obs.). By themselves, However, I can find no evidence that M. dis- the two species of Hypocnemoides form a nat- juncta was included in the various shifts of ural grouping, with great similarities in plum- species between Myrmeciza and Myrmoderus age, voice, foraging behavior, and habitat use. by Hellmayr (in Cory and Hellmayr 1924), Almost none of these characters are shared Todd (1927), or Peters (1951). with M. disjuncta. Morphological, vocal, and behavioral char- I feel that Myrmeciza as currently con- acters offer contradictory clues to the possible structed is paraphyletic, with various sub- generic affiliations of the Yapacana Antbird. groupings that do not appear to be closely al- Myrmoderus is not currently recognized, but lied on the basis of morphological, vocal, or both Todd (1927) and Peters (195 1) advocated behavioral characters (e.g., Peters 195 1, that it be reserved for [Myrmeciza] Zoricatu Ridgely and Tudor 1994). None of these sub- and squamosa. The latter are clearly sibling groups is a good fit for M. disjuncta. Myr- species that share several distinctive morpho- meciza atrothorax has been suggested as a logical, vocal, and behavioral characters, and close relative of M. disjuncta by Friedmann 208 THE WILSON BULLETIN *Vol.‘ 111, No. 2. June 1999

(1948), but the males of the two species differ Alechiven Lodge, my two bases of operation for this dramatically in plumage, and the two species research. L. Oleaga was of particular help in cutting trails and facilitating my work at Hato Las Nieves. C. share no vocal similarities that I can detect. Herrera, M. Ruiz and D. Forbes of Caracas provided No compelling morphological similarities invaluable assistance with trip logistics. clearly ally M. disjuncta with any other Myr- Special thanks must go to D. Wolf, who accompa- mecizu, nor with any other antbird. Similarly, nied me on my trip to Yapacana National Park. Dave I have compared songs and calls of M. dis- shared in the excitement of finding the Yapacana Ant- junctu to the other 18 species currently placed birds and taping their vocalizations; his good humor in Myrmeciza (Isler and Whitney 1999; Zim- and knowledgeable insights made the heat and the bit- ing jejenes more tolerable. Final thanks go to V. Eman- mer, unpubl. data) and can find nothing to sug- uel and Victor Emanuel Nature Tours, Inc., for provid- gest a close relationship between A4. disjuncta ing me with travel opportunities that made much of and any of the other species. Relationships my research possible. suggested by one or two morphological char- acters in one sex are contradicted by morpho- LITERATURE CITED logical characters in the other sex, vocal char- ANDERSON, A. B. 198 1. White-sand vegetation of Bra- acters, behavioral characters, or by some com- zilian Amazonia. Biotropica 13: 199-2 10. bination of the three. Although past descrip- CORY, C. B. AND C. E. HELLMAYR. 1924. Catalogue of birds of the Americas. Pteroptochidae, Conopo- tions of genera have been based largely on phagidae, Formicariidae. Field Mus. Nat. Hist. plumage characters which may or may not (Zool. Series) 13(3): l-369. have phylogenetic relevance, the addition of FIT~PA~UCK, J. W. AND D. E. WILLARD. 1990. Cercom- vocal and behavioral data adds important ev- acra manu, a new species of antbird from soutl- idence in redefining these relationships (e.g, western Amazonia. Auk 107:239-245. comments in Remsen 1997, Remsen and FRJEDMANN,H. 1945. A new ant-thrush from Venezue- Schulenberg 1997). In the absence of a mo- la. Proc. Biol. Sot. Wash. 58:83-84. FRIEDMANN, H. 1948. Birds collected by the National lecular based phylogeny the most conserva- Geographic Society’s expeditions to northern Bra- tive approach would be to leave M. disjuncta zil and southern Venezuela. Proc. U.S. Nat. Mus. where it is, as yet another poor fit in a genus 97:373-569. understood to be heterogeneous. However, in HELLMAYR, C. E. 1906. [Descriptions of new South my opinion, the sum of morphological, vocal, American birds.] Bull. Brit. Ornithol. Club. 19: and behavioral evidence would suggest that 8-9. HILTY, S. L. AND W. L. BROWN. 1986. A guide to the the Yapacana Antbird is monotypic, deserving birds of Colombia. Princeton Univ. Press, Prince- of its own genus. ton, New Jersey. ISLER, M. L. 1997. A sector-based ornithological geo- ACKNOWLEDGMENTS graphic information system for the Neotropics. Primary thanks go to M. and I? Isler for generously Omithol. Monogr. 48:345-354. lending their time and talents to the many figures in ISLER, M. L., I? R. ISLER, AND B. M. WHITNEY. 1997. this paper. Phyllis produced the many spectrograms af- Biogeography and systematics of the Tfzamnophi- ter reviewing several hours of my tape recordings, and Zus punctatus (Thamnophilidae) complex. Omi- Mort compiled the list of distributional records and thol. Monogr. 48:355-382. produced the map used in Figure 1. The Islers made ISLER, I? R. AND B. M. WHITNEY. 1999. Voices of the many helpful comments on an early draft of this man- antbirds: Thamnophilidae, Formicariidae, and uscript and have been an invaluable source of advice Conopophagidae. Library of Natural Sounds, Cor- and ideas with all aspects of my research on antbirds. nell Laboratory of Ornithology, Ithaca, New York. M. Lentino, C. Rodner, and R. Restall of the Colecidn MAYR, E. AND W. H. FHELPS, JR. 1967. The origin of Omitologica Phelps generously supplied important the bird fauna of the south Venezuelan highlands. collection information. ‘I Schulenberg provided valu- Bull. Am. Mus. Nat. Hist. 136:269-328. able advice as well as access to several references. J. MEYER DE SCHAUENSEE, R. 1966. The species of birds V. Remsen reviewed an early draft of the manuscript of South America with their distribution. Living- and made numerous helpful comments that improved ston Press, Narbeth, Pennsylvania. it. I am most grateful to J. Coons for sharing infor- MEYER DE SCHAUENSEE,R. 1970. A guide to the birds mation on his experiences in the Picua region, partic- of South America. Livingston Press, Narbeth, ularly for bringing Alechiven Lodge to my attention. Pennsylvania. D. Cooper was the first to report the probable presence MEYER DE SCHAUENSEE,R. AND W. H. PHELPS, JR. of Caura Antbirds in the Serrania de la Cerbetana and 1978. A guide to the birds of Venezuela. Princeton his advice was helpful in planning my trip to that re- Univ. Press, Princeton, New Jersey. gion. Thanks also to the staffs of Hato Las Nieves and MONROE, B. L., JR. AND C. G. SIBLEY. 1993. A world Zimmer * CAURA AND YAPACANA ANTBIRDS’ 209

checklist of birds. Yale Univ. Press, New Haven, RIDGELY, R. S. AND G. TUDOR. 1994. The birds of Connecticut. South America. Vol. 2. Univ. of Texas Press. Aus- PARKER, T. A., III, D. E STOTZ, AND J. W. FITZPATRICK. tin. 1996. Ecological and distributional databases. Pp. SIBLEY, C. Cl. AND B. L. MONROE. 1990. Distribution 115-436 in Neotropical birds: ecology and con- and of birds of the world. Yale Univ. servation (D. E Stotz, J. W. Fitzpatrick, T A. Press, New Haven, Connecticut. Parker, III, and D. K. Moscovits, Eds.). Univ. of SNOW, D. 1982. The cotingas. Cornell Univ. Press, Ith- Chicago Press, Chicago, Illinois. aca, New York. PAYNTER, R. A. 1982. Ornithological gazetteer of Ven- STOTZ, D. E, J. W. FITZPATRICK,I‘ A. PARKER,III, AND ezuela. Museum of Comparative Zoology, Cam- D. K. MOSKOVITS. 1996. Neotropical birds: ecol- bridge, Massachussetts. ogy and conservation. Univ. of Chicago Press, PETERS, J. L. 195 1. Checklist of birds of the world. Chicago, Illinois. Vol. VII. Harvard Univ. Press, Cambridge, Mas- TODD, W. E. C. 1927. New gnateaters and antbirds sachussetts. from tropical America, with a revision of the ge- nus Myrmeciza and its allies. Proc. Biol. Sot. PHELPS, W. H. AND W. H. PHELPS, JR. 1963. Lista de Wash. 40:149-178. las aves de Venezuela con su distribucibn. Parte WILLARD, D. E., M. S. FOSTER, G. E BARROWCLOUGH, II. Passeriformes, second ed. Bol. Sot. Venez. R. W. DICKERMAN, P. E CANNELL, S. L. COATS, J. Cienc. Nat. 24:1-479. L. CRACRAFT,AND J. I? ONEILL.’ 1991. The birds PIRES, J. M. 1974. Tipos de vegetacao da Amazonia. of Cerro de la Neblina, Territorio Federal Ama- Publ. Avul. Mus. Goeldi. 20: 179-202. zonas, Venezuela. Fieldiana Zool. 65: l-90. REMSEN, J. V., JR. 1997. A new genus for the Yellow- ZIMMER, J. T. AND W. H. PHELPS. 1947. Seven new shouldered Grosbeak. Omithol. Monogr. 48:89- subspecies of birds from Venezuela and Brazil. 90. Am. Mus. Novit. 1338:1-7. REMSEN, J. V., JR. AND S. K. ROBINSON. 1990. A clas- ZIMMER, K. J. AND S. L. HILTY. 1997. Avifauna of a sification scheme for foraging behavior of birds in locality in the upper Orinoco drainage of Ama- terrestrial habitats. Stud. Avian Biol. 13:144-160. zonas, Venezuela. Omithol. Monogr. 48:865-886. REMSEN, J. V., JR. AND 'I S. SCHULENBERG.1997. The ZIMMER, K. J., A. WH~TTAKER,AND D. E STOTZ. 1997. pervasive influence of Ted Parker on Neotropical Vocalizations, behavior and distribution of the Rio field ornithology. Ornithol. Monogr. 48:7-20. Branco Antbird. Wilson Bull. 109:663-678.