Use of the and Space by Chestnut-Backed Antbird

The Condor 87:187-191 0 The CooperOrnithological Society I98 5

USE OF TIME AND SPACE BY CHESTNUT-BACKED ANTBIRDS

PETER J. MARCOTULLIO AND FRANK B. GILL

ABSTRACT.-Foraging accounted for over 90% of the activity of two pairs of Chestnut-backed Antbirds (Myrmeciza exsul) during a one month study of their daytime time budgetsin the Pacific lowlands of Costa Rica. We also documented in detail their use of territorial space and preference for dense vegetation where foraging was most productive. Our results support previous findings that tropical insectivores may have to devote much effort to foraging and that dense cover is the best microhabitat for foraging by antbirds.

Daily time budgets reflect a bird’s allocation patterns of the Chestnut-backed Antbird of time to essential self-maintenance and to (Myrmeciza exsul). This species is common options such as reproduction and migration in lowland forest undergrowth and second- (King 1974). Breeding seasons,reproductive growth forest from Nicaragua to western Ec- effort, and pair bonds are shaped in part by uador (Willis and On&i 1972, Ridgely 1976). the foraging requirements of individuals. It typically inhabits old (3-6 years) gapsin the The proportion of the daylight hours de- forest with dense shrub and saplinggrowth, or voted daily to foraging ranges from less than edges of man-made clearings with similar 10% to nearly 100%. At one extreme, hum- vegetation (F. G. Stiles, in litt.). The bird does mingbirds and tropical frugivores obtain ad- not habitually associatewith antswarms. equate food in rather little time by feeding on Willis and Oniki (1972) suggestedthat in conspicuous,energy-rich foods (D. Snow 1963, Chestnut-backed Antbirds, conflicts between B. Snow 1970, Wolf and Hainsworth 197 1, self-maintenance and the demands of parental Wolf et al. 1975). At the other extreme, small care are resolved by active male participation landbirds in the far north feed almost contin- in all aspectsof nesting. Plentiful supplies of uously during short winter days to accumulate food in dense cover also may provide some the energy required for survival (Gibb 1956, relief. We wished to obtain more precise in- 1960). Remaining to be determined, however, formation on foraging effort and the impor- is how much time each day tropical insecti- tance of dense vegetation to this species. vores feed. Does the time they require to find enough energy and nutrition constrain other METHODS activities or somehow dictate seasonalcycles We conducted this study from 12 June to 5 in the availability of productive energy? July 198 1 in Corcovado National Park on the With the exception of the Mangrove Swal- Osa Peninsula of Costa Rica. Allen (1956) has low (Tachycineta albilinea; Ricklefs 197 1) and describedthe Tropical Wet Forests(Slud 1964: Checker-throated Antwren (Myrmotherula 17) of this park. Our study site, 100 x 200 m, fulviventris; J. Gradwohl and R. Greenberg, was at the edge of rain forest near the park unpubl. data), the daily foraging time budgets headquarters at La Sirena. The site, a recently and space use patterns of insectivorous trop- abandoned pasture edge, was second-growth ical birds remain unquantified. Mangrove vegetation, 3-4 years old with a canopy 4-8 Swallows with nestlings devote 80-100% of m tall, composed of Ochroma lagopus (Bom- the morning and late afternoon hours to the baceae), Trema micrantha (Ulmaceae), Piper capture of aerial insects.Mid-day foraging ap- auritum (Piperaceae), Heliocarpus appendi- parently is limited by heat stress. Checker- catus(Tiliaceae), and Ingu sp. (Leguminosae). throated Antwrens forage 80-90% of the day Heliconia imbricata (Musaceae), Heliconia most of the year and 30-50% of the day when curtispatha,Calathea sp. (Amaranthaceae),and food is most abundant. Studieson both of these densethickets ofAculypha sp. (Euphorbiaceae) birds reveal substantialdaily foraging effort as were the principal understory plants (D. H. well as seasonal variation. The widespread Janzen, pers. comm.). impression that tropical insects are abundant We spent 56 h observing one pair (A) of and, therefore, that finding adequateinsect food Chestnut-backed Antbirds color-marked with may not require much time seems erroneous. celluloid leg bands and 6 h observing a second We undertook to document further the im- pair (B), only one of which was color-marked. portance of foraging effort in the daily activity The antbirds were followed nearly continu-

11871 188 PETER J. MARCOTULLIO AND FRANK B. GILL

TABLE 1. Time budget of a pair of Chestnut-backedAntbirds. ’

N = 7%~; 890 min 5;;:;; 130 min 665 min 378 min 11:oo 13:oo 15:oo 07:oo gig; ll:oo 13:oo 15:oo 17:oo

Forage 90.4 + 11.6 98.3 ? 9.4 92.0 +9.4 80.3 ? 14.4 97.3 + 4.4 100.0 + 0.0 Restb 0.7 (1) 0.0 (0) 7.6 (2) 11.0 (0) 0.0 (0) 0.0 (0) Agonistic behavior 4.0 (4) 1.9 (1) 0.0 (0) 8.3 (1) 2.9 (2) 0.0 (0) Advertisement 4.9 (6) 0.0 (0) 0.0 (0) 0.0 (0) 0.0 (0) 0.0 (0) * Percentageof time devotedto the activity k I standarddeviation. bThe numberof times activity was observedper observationperiod is in parentheses. ously during each observation period in the The birds followed a daily routine throughout dense thickets of their territory via trails we the study with only minor variations. Both cut and also those of peccaries. pairs of antbirds were territorial and regularly Breeding preceded our observations. The engaged in disputes with their neighbors or principal pair (A) had a fledgling with them in with intruders. The territory boundaries did mid-June at the onset of our field work, but not change during the study. the fledglingdisappeared shortly before we col- Observation periods ranged from 30 to 285 or-marked the adults and started to monitor min, and usually started shortly after dawn at them. We did not see any signs of renesting, 0515 to 0545, when the antbirds began call- including feeding of the female by the male. ing. The percent of time foraging during each

JUNE 23

8:30

JUNE 25 JUNE 24

JUNE 29 JUNE 30

FIGURE 1. These maps show the movements of pair A during the mornings of June 23-25 and 29-30. Dotted lines connect points between observation periods and may not be actual routes of the birds. USE OF TIME AND SPACE BY ANTBIRDS 189

70 142 119 071 48

065 033 18 18 4

0101 @ @ 26 8

22 @ @ 18 5

FIGURE 2. Use of territorial spacein terms of total time (minutes) spent by pair A in each quadrat. Times spent in quadratswith dense vegetation are indicated within circles. The rest had open vegetation. Scale: 2 cm = 25 m. observation period proved to be independent marily an understory of Heliconia and scat- of the duration of that period. Days were di- tered Calathea. vided into 2-h periods. Two-thirds of our data were obtained before 1 l:OO. Therefore, we RESULTS treated the average foraging times for each period equally to estimate average daily foraging DAILY TIME BUDGETS effort. We recorded the birds’ activity, precise Considering the proportions of the six 2-h pe- location on the mapped territory, frequency of riods spent foraging as equal measures,despite peckingat possiblefood, and generalbehavior. the difference in total observation time in each, We classifiedthe antbirds’ activities as “For- our data show that the antbirds of pair A spent aging,” “Agonistic Behavior,” “Advertising,” an average of 93% of their time foraging (range and “Resting.” Foraging referred to time spent was 80-100% per 2-h period; Table 1). Non- actively procuring food as described for this foraging activities comprised lessthan 20% of species by Slud (1964), Skutch (1969), and any 2-h period and less than 6% of the total Willis and Oniki (1972). “Agonistic Behavior” daylight hours. Foragingeffort was leastat mid- accompanied territorial disputesand included day and was particularly intense in the late displays, attacks, chases,actual fights, and es- afternoon. The variations in foraging efforts caping. “Advertising” refers to singing on ex- among the 2-h periods, other than mid-day, posedbranches while clearly not foraging. The were minor. antbirds also sang throughout the day while foraging.“Resting” refers to sitting quietly with USE OF TERRITORIAL SPACE preening. Sometimes at mid-day the antbirds The antbirds used only part of their territory disappearedinto small, dense thickets only to each day. They usually started the day in the reappear ca. 20 min later; we assumed they center of the territory, moved out to the pe- were resting inside these thickets. riphery, and then returned to the center before We divided the territory of pair A into 42 mid-morning. The daily foraging paths them- quadratseach 20 m on a side. Total time, num- selves, however, varied greatly in coverage of ber of visits, and average time per visit were sectionsof the territory (Fig. 1). On some days computed for each quadrat. We also classified (e.g., June 23,26,28), the antbirds moved rap- the vegetation in each quadrat as “open” or idly over large sections of their territory, but “dense.” What we categorized as “dense vege- on other days (e.g., June 29), they restricted tation” consisted mostly of an understory of their activities to smaller sections. Acalypha thickets. “Open vegetation” was pri- The antbirds preferred certain parts of their 190 PETER J. MARCOTULLIO AND PRANK B. GILL territory; their time was not distributed evenly Goldcrests in England, and they do not face among the available quadrats (Fig. 2). Low- comparable metabolic demands on stored re- use quadrats were peripheral, whereas most serves, owing to long, cold nights. high-use quadrats were centrally located. In- We could not determine why the Chestnut- tensities of use in the morning were signifi- backed Antbirds spent most of the day for- cantly correlated with such intensities in the aging.Conceivably, they were eating more than afternoon (r = 0.5, P < 0.01). was necessaryfor maintenance in order to ac- Total time per quadrat reflected both the cumulate reserves for renesting. This possi- frequency of revisitation and the average bility seems unlikely because we did not see amount of time per visit to a quadrat. High- any courtshipfeeding of the female by the male, use quadrats usually were visited at least once as would be expected if this were happening daily, low-use quadrats only once a week. The (Willis 1972, Willis and Oniki 1972, Green- distribution of these visits was not random berg and Gradwohl 1983). Massesof the three (Chi-square for comparison to Poisson distri- individuals weighed at the beginning of this bution with mean of 2.43 = 14.65, df = 7, P < study were typical of the species-28, 28, 27 0.05). Average times per visit to high-use g (see also Willis and Oniki 1972). quadrats (14.3 + 5.6 min) and average-use Most likely, the foraging effort we docu- quadrats (12.2 + 5.2 min) were similar, but mented reflected the difficulties of finding prey average time per visit to low-use quadrats was in a lowland tropical forest where insects are less (7.6 + 5.1 min). exceedingly cryptic and often distasteful or “Open” vegetation covered 440 sq. m of the inedible (Fogden 1972, Janzen 1980). The territory whereas “dense” vegetation covered abundanceof litter-dwelling arthropods varies 400 sq. m. A two-way analysis of variance seasonally on Barre Colorado, Panama, in- showed that the antbirds averaged more time creasingduring the rainy season(Levings and per visit in quadratswith densevegetation than Windsor 1982). June, therefore, should not in quadrats with open vegetation, no matter have been a period of food scarcity at Cor- where these quadrats were located (Fig. 2) covado, although the seasonality of litter ar- (F1,37= 6.73; P < 0.01). Times per visit in thropod abundance has not been studied at “dense edge” versus “dense center” quadrats this site. Willis and Oniki (1972) suggestedthat were not significantly different (F1,3, = 1.05; the penchant of these antbirds for dense cover P < 0.31), but more total time was spent in indicatesbetter foragingin suchmicrohabitats. densecenter quadratsthan in denseedge quad- Our data show not only a bias of foraging time rats (Chi-square = 5.11, df = 1, P < 0.005). towards patchesof dense cover within the ter- Similarly, more total time was spent in open ritory, but also greater foraging activity in this center quadrats than in open edge quadrats microhabitat as indicated by frequency of (Chi-square = 39.8, df = 1, P < 0.005). pecking. The antbirds’ preference for quadrats with Confirmation of suchsubstantial foraging ef- dense vegetation reflected their apparent for- forts by tropical insectivores will complement aging success.The frequency of pecking at ac- the growing body of evidence that breeding by tual or potential prey was significantly higher tropical birds is proximately constrained by in dense areas (1.30 pecks/min + 0.44) than their ability to accumulate adequate protein in open areas (0.84 pecks/min + 0.48) (t = and fat reserves (Jones and Ward 1976, Fog- 4.17, df = 26, P -c 0.05). We are unable to tell den and Fogden 1979). Comparative studies from our data whether more frequent use of of the time budgetsof tropical birds may also center territories reflected better foraging or provide a productive route to understanding simply geometry of the territory. of the adaptive basis of monogamy and the evolution of pair bond diversity in tropical DISCUSSION birds. Our results, plus those of Gradwohl and ACKNOWLEDGMENTS Greenberg for Myrmotheruluantwrens, indi- cate that the foraging can be the principal day- We thank R. E. Ricklefs, M. B. Robbins, and D. Janzen for their comments on this manuscript. R. Greenbergand time activity of tropical insectivoreseven when J. Gradwohl generouslymade available their unpublished they are not feeding nestlings. Tropical ant- data on Checker-throatedAntwrens. We also expressour birds thus join tits (Purus spp.), Goldcrests appreciation to the National Park Service of Costa Rica (Regulusregulus), and Rock Pipits (Anthus and especiallyto the staff of Corcovado National Park for making our stay pleasant and productive. spinoletta)wintering in northern Europe (Gibb 1956, 1960) as examples of birds with limited LITERATURE CITED time for activities other than foraging. Such ALLEN,P. H. 1956. The rainforestsof Golfo Dulce. Univ. effort is the more surprising because antbirds of Florida Press, Gainesville. [Reissued 1977, Stan- have nearly 3 h more daylight each day than ford Univ. Press, Stanford, CA.] USE OF TIME AND SPACE BY ANTBIRDS 191

FOGDEN, M. P. L. 1972. The seasonalityand population RICKLEFS, R. E. 1971. Foraging behavior of Mangrove dynamics of equatorial forest birds in Sarawak. Ibis Swallowsat Barro Colorado Island. Auk 88:635-65 1. 114:307-343. RIDGELY, R. S. 1976. A guide to the birds of Panama. FOGDEN,M. P. L., AND P. M. FOGDEN. 1979. The role Princeton University Press, Princeton, NJ. of fat and protein reservesin the annual cycle of the SKUTCH.A. F. 1969. Life histories of Central American Grey-backed Camaroptera in Uganda (Aves: Sylvi- birds, III. Families Cotingidae, Pipridae, Formicari- idae). J. Zool. (Lond.) 189:233-258. idae, Furnariidae, Dendrocolaptidae,and Picidae. Pa- GIBB,J. A. 1956. Food, feeding habits, and territory of cific Coast Avifauna No. 35. the Rock Pinit (Anthus sainoletta). Ibis 98506-530. SLUD, P. 1964. The birds of Costa Rica: distribution and GIBB, J. A. 1960. Populations of tits’and goldcrestsand ecology. Bull. Am. Mus. Nat. Hist. 128. their food supply in pine plantations. Ibis 102:163- SNOW,B. K. 1970. A field study ofthe Bearded Bellbird 208. in Trinidad. Ibis 112:299-329. GREENBERG, R., AND J. GRADWOHL. 1983. Sexual roles SNOW,D. W. 1963. The evolution of manakin displays. in the Dot-winged Antwren (Microrhopiasquixensis), Proc. XIII Int. Omithol. Congr. (1962):553-561. a tropical forest passerine.Auk 100:920-925. WILLIS, E. 0. 1972. The behavior of Spotted Antbirds JANZEN, 6. H. 1980. Heterogeneity of potential food Omithol. Monogr. lO:l-162. abundancefor tropical small landbirds, p. 545-552. WILLIS, E. O., ANDY. ONIKI. 1972. Ecologyand nesting In A. Keast and E. S. Morton [eds.], Migrant birds in behavior of the Chestnut-backedAntbird (Myrmeciza the Neotropics. Smithsonian Institution Press,Wash- exsul). Condor 74:87-98. ington, DC. WOLF, L. L., AND F. R. HAINSWORTH. 1971. Time and JONES. P. J.. AND P. WARD. 1976. The level of reserve energy budgetsof territorial hummingbirds. Ecology proteinas the proximate factor controlling the timing 52:980-988. of breeding and clutch size in the Red-billed Quelea, WOLF, L. L., F. R. HAINSWORTH, AND F. B. GILL. 1975. Quelea quelea.Ibis 118~547-554. Foragingefficiencies and time budgetsin nectar feed- KING, J. R. 1974. Seasonalallocation of time and energy ing birds. Ecology 56: 117-l 28. resourcesin birds, p. 4-70. In R. A. Paynter, Jr. [ed:j, Avian enemetics. Publ. Nuttall Omithol. Club No. Department of Biology, The University of Pennsylvania, 15. - Philadelphia, Pennsylvania 19105, and The Academy of LEVINGS, S. C., AND D. M. WINDSOR. 1982. Seasonaland Natural Sciences,Philadelphia, Pennsylvania 19103. Re- annual variations in litter arthropod populations, p. ceived 8 December 1983. Final acceptance 26 January 355-388. In E. G. Leigh et al. [eds.], The ecology of 1985. a tropical forest. Smithsonian Institution Press, Washington, DC.

The Condor87~191 0 The CooperOrnithological Society 1985

RECENT PUBLICATIONS

Commonbird songs.Songs of easternbirds. Songsof west- The Peterson Field Guide Series. A field guide to bird ern birds.-Donald J. Borror. 1984. Three sets,each con- songsof eastern and central North America. Second edi- taining a 50-minute cassetteand a 64-pagebooklet. Dover tion.-Cornell Laboratory of Ornithology. 1983. Available Publications, Mineola, NY. $7.95 per set. These are cas- in albums of either two phonodiscs or two cassettes. sette versions of field recordingsthat were first issued as Houghton Mifflin Co., Boston. $19.95 per set. Here, in a phonodiscsin 1967-l 97 1. Their age notwithstanding, the completelyrevised edition, are field recordingsof the voices soundquality is good,though with a little more white noise of 250 speciesof birds. They are arrangedsystematically, background than one would like. Each cassettepresents i.e., following the sequencein the fourth edition of Peter- sixty species,with no duplication amongthem. The species son’s eastern Field Guide. Each speciesis announced by are arrangedmore or lessaccording to the generalcharacter name. Calls and/or songsare given, though not as sepa- of the vocalizations, from simple to complex. Birds with rately identified cuts.An index to the recordings,including similar songsare groupedtogether as an aid to comparison. their locality and a page reference to the speciesaccount Most speciesare representedby two or more cuts, thus a in the book, is given on the phonodiscalbum and a booklet fair sample of their vocal quality and characteristics.Each with the cassettes.Both formats have their advantages: speciesis introduced simply by name and its cutsby num- discsare easierfor finding particular species,whereas cas- ber. The accompanyingbooklet describesthe songsin the settescan be usedin the field. As comparedwith the Dover same sequenceas on the cassette;the accounts focus on sets mentioned at left, these offer many more species, features that serve for identification and are much fuller though some of them are questionably necessaryfor rec- than the remarks in field guides. A table showing the lo- ognition. The trade-off is that the samplesfor each species cality and month in which the songs were recorded is tend to be fewer and briefer, thereby presentingless of the appended. These cassettesare certainly effective aids for variability and the sometimes-characteristicpattern of learningbird songs,better yet for refreshingone ’s memory pausesin singing. The Cornell sets sound slightly clearer, before spring migration. They are good value for the mon- though they have a low backgroundhum in place of white ey, but eastern birders will have to buy the first two sets noise. Being keyed to the Field Guide, they enable an in order to get better coverage. observer to check a bird’s appearance,habits, and range along with its voice. Borror’s booklets, on the other hand, go into more detail about the vocalizations. A choice between them dependson one’s budget and the kind of help one needs in learning bird songs.