SHORT COMMUNICATIONS 251

The Condor 90:25 l-253 0 The Cooper Ornithological Society 1988

NEST-SITE SELECTION, NEST, AND EGGS OF THE STOUT-BILLED CINCLODES (CINCLODES EXCELSIOR), A HIGH ANDEAN FURNARIID ’

GARY R. CRAWS Departmentof VertebrateZoology, National Museum of Natural History, SmithsonianInstitution, Washington.DC 20560

GONZAL~ ARANCO Unidad de Ornithologia,Institute de CienciasNaturales, UniversidadNational de Colombia, A.A. 7495, Bogotd,Colombia

Key words: Cinclodesexcelsior; Colombia; Nevado parison (Table I). As noted previously, texture and del Ruiz; :volcanism: cinclodes. hardnessof strata were highly correlated to color. The pale pumiceouslayers were porousand friable and eas The Stout-billed Cinclodes (Cinclodesexcelsior) has a ily excavatedwith a trowel. Paleosolswere harder and relictual distribution in the paramo and puna grass- heavier and more difficult to work. lands of the Andes. Subspecificallydifferentiated pop- In January 1986, Graves counted cinclodes’ bur- ulations are known from three widely separatedareas rows, and the type of strata (dark vs. light) they were (Peters I95 I, unpubl. observ.): (1) C. e: columbiana excavated from, along a 2-km section of the Murillo- from the / vol- Guali Road (3,900 to 4,100 m) about 6 km east of canic complex in the Central Cordillera of Colombia; Aguacerales(Table I). The threat of volcanic eruption (2) C. e. excelsiorfrom southernColombia on Nevado preventedthe measurementofburrow placement.This del Cumbal south through northern and central Ec- locality is lesssheltered from wind and lesshumid than uador to the province of Chimborazo; (3) C. e. aricom- Aguacerales. ae from southernPeru in the departmentsof Puno and Although there were highly significantdifferences in Cuzco. Despite its extensive distribution in open hab- strata selection between sites (x2 = 74.9, df = I, P < itat formations, almost nothing has been published on 0.001) these data show that cinclodes in both areas its biology. The eggsand nest have not been previously prefer pumiceousstrata as burrow sites. Both surveys described (Vaurie 1980; L. Kiff, pers. comm.). In this included abandoned burrows as well as those in use. paper, we combine our observations made indepen- Roadcutswere constructedin the 1950s and 1960s at dently in April 1982 (Arango) and January 1986 Aguacerales,and duringthe 1970seast of the Rio Guali. (Graves) of nestingcinclodes on Nevado del Ruiz, De- Burrows persist for many years until the cliff or bluff partment of Caldas, Colombia. faceerodes away, thusdocumenting generations of nest- site selection. We hypothesize that pumiceous strata NEST-SITE SELECTION were chosen not only for ease of excavation, but also The Stout-billed Cinclodes is a common and conspic- becausedrainage there is better than in paleosols.We uousinhabitant of the paramo from 3,550 to 4,300 m. note parentheticallythat the known distribution of C. In January and April, pairs of cinclodeswere actively excelsiorin the northern Andes coincideswith volcan- digging and attending burrows (nesting and roosting) ically active regions. in vertical banks and cliff facesformed by roadcuts.A The majority of burrows were excavatedwithin I m majority of the burrows seemed to be located in pale- colored strata composedof decomposedpumice from pastvolcanic eruptions, rather than in the darker,dense, alternating layers of paleosol (Fig. I). To verify our initial impressions,we independently categorized the placementof burrows on different sectionsof the road crossingthe northwesternRank ofthe Ruiz strato cone. In April 1982, Arango censusedcinclodes ’ burrows along the roadcut between Arenales and Aguacerales (3,800 to 4,300 m), and estimated the color, texture. and hardness of the stratum from which the burrow was excavated,as well as its vertical placement on the roadcut. Color was originally classifiedinto one of six categories(ranging from cream-coloredto black), How- ever, becauseGraves later distinguishedonly “dark” paleosolsand “light” pumiceousstrata, Arango’s data have been reducedto two categoriesfor statisticalcom- FIGURE I. Section of roadcut on Nevado del Ruiz showingalternating pumiceous strata (pale) and paleo- I Received 30 March 1987. Final acceptance8 July sols (dark). The entrancesof three cinclodes burrows 1987. can be seen in upper pumiceousstratum. 252 SHORT COMMUNICATIONS

TABLE I. Placement of burrows by Cinclodesexcel- sior on Nevado de1 Ruiz. Burrows in paleosols and those from paleosol/pumiceousboundaries are pooled for within-locality binomial tests (large sample ap- proximation).

AglLXeG4eS East of Rio Guali) Stratum ( 1982) (1986)

Paleosol 143 6 Paleosol/pumiceous 5 Pumiceous 171 154 Binomial test 2 = 1.58 z= 11.13 P i 0.057 P < 0.0001

FIGURE 2. Cracked eggand thin mat of dried grass chamber contained an egg with hairline cracks and a taken from slightly enlargedterminal chamber of bur- crushedegg (Fig. 2). The crackedegg, which could not row. be salvaged,was cream-colored, immaculate, slightly glossy,and measured22.0 x 28.5 mm. It contained a decayed,well-developed embryo. Becauseone eggwas of the top of the roadcut (Aguacerales;n = 285, K = well-incubated, and the remains of only two eggswere 0.5 f 0.3 m, range = 0.05-2.3 m); three to six alter- observed, we suspectthat the clutch may have been nating layers of pumice and paleosolwide enough for two eggs. Abandonment of the nest may have been burrow excavation occur in this section. The average causeddirectly (death of the birds) or indirectly (em- height of burrows above the foot of the roadcut was igration) by the eruption. 2.3 + 1.2 m (Aauacerales:n = 285, range = 0.8-10.0 The excavated burrows were identical in external m). In many p&es tussockgrass overhangs the bank appearanceto active nests along the same roadcuts. edge above the burrow entrancesproviding some de- Judgingfrom the uniformity of burrow diameter and gree of protection from rain. Becauseof the steepness placement,we suspectthat nearly all the burrows were of the roadcuts(usually greater than 70”), burrows are excavated by cinclodes. (The Bar-winged Cinclodes, somewhatinaccessible to terrestrialpredators. We sus- Cinclodesficus, is widespread in the Colombian An- pect that the consistentselection of the upper quartile des, but does not occur in the Ruiz/Tolima volcanic of the roadcut for burrow excavation is influenced by complex in the Central Cordillera.) The only other bur- both factors. We noted only two burrows excavatedon row or cavity-nestingspecies observed above 3,800 m natural slopesor cliffs away from roadcuts. This sug- on Ruiz were Brown-bellied Swallows (No- geststhat a limited amount of road building may ac- tiochelidonmurina) and the rare Rufous-fronted Par- tually be beneficial to this species.At Aguacerales,the akeet (Bolborhynchusferrugineifons)(see Graves and minimum horizontal distance between burrows was Giraldo 1987). Active swallow nests, however, were 0.65 m. There was no vertical overlap (i.e., one directly placed in small crevicesin rocks or in rounded cavities above another) of burrows in either locality. We spec- at the baseof tussockgrass overhanging cliff edges.No ulate that the greaterusage of denserpaleosols at Agua- signsof rodent usagewere seen in burrows excavated ceralesmay be due to the saturationof “prime” pumi- from the face of vertical roadcuts. Other bird species ceous strata with old burrows. Patterns of soil commonly observed in paramo on Nevado de1 Ruiz stratification did not appear to differ grosslybetween build cup-like or enclosednests in shrubs,at the base the two sites. of tussockgrass, or in shallow cavities (e.g., Leptas- thenura andicola, Asthenesjlammulata.Grallaria qui- NEST AND EGGS tensis, Ochthoecafumicolor, Turdus &cater, Cate- Arango excavated five burrows from previous nesting menia inornata, Phrygilus unicolor, Zonotrichia seasons(i.e., no signs of recent use). Park rules pre- capensis). ventedthe examinationofactive nests.Varying in length All species of Cinclodes,Geositta, and Upucerthia from 0.65 to 1.10 m, they sloped gently upward and for which information existsconstruct burrows, utilize lacked a noticeable terminal chamber. A thin bed of natural cavities or crevices, or adopt rodent burrows decayed vegetable material was observed at the ter- for nesting (Vaurie 1980). minus of the burrows. The presenceof a low lip at the burrow entrance was variable. We thank INDERENA for permission to work in While censusingbirds on January29 in an area dev- Parque National 10sNevados, and Ing. Pablo Medina astated by the 13 November 1985 eruption of Ruiz, of the Comitt de EstudiosVulcanol6zicos for orovid- Graves excavated a burrow (0.95 m in length) that ingjeep and helicoptersupport in Jam&y 1986.Graves ’ appeared to have been recently occupied. The con- researchwas supported by a grant from the Research struction was similar to those examined by Arango Opportunity Fund, Smithsonian Institution. exceptthat the burrow terminated in a slightlyenlarged nest chamber floored with a thin mat of dried grass. RESUMEN The internal diameter of the burrow varied from 70 El barranquerode paramo o cinclodesde pica gruezo to 90 mm (burrow entrancesare somewhatlarger). The (Cinclodesexcelsior) es un habitante conspicuo y co- SHORT COMMUNICATIONS 253 mun de 10sparamos de la Cordillera Central de 10s donada se encontraron dos huevos (uno agrietado y Andes Colombianos entre 3550 y 4300 metros de al- otro roto). titud. A lo larao de1talud de torte de 10scarreteables de la zona seobserva su preferencia por excavar las LITERATURE CITED galeriasde anidacion y descansoen 10sestratos claros formados por mantos de cenizas volcanicas de erup- GRAVES,G. R., AND J. A. GIRALDO. 1987. Population cionespasadas, mas que en 10sestratos oscuros y den- statusof the Rufous-fronted Parakeet(Bolborhyn- SOSde 10spaleosoles. Los mantos de cenizasvolcanicas chusferrugineifrons),a Colombian endemic. Ger- son porososy friables siendo facilmente excavadospor faut. 17189-92. cinclodes. La mayorla de las galerlas han sido exca- PETERS,J. L. 1951. Check-list of birds of the world. vadas cercade1 borde superior de1talud (n = 285; K = Vol. 7. Museum of Comnarative Zooloav.__ Cam- 0.5 + 0.3 m desde arriba). La altura media desde la bridge, MA. base de1 carreteable es de 2.3 + 0.3 m. Las galerlas VAURIE,C. 1980. Taxonomy and geographicdistri- comprendidas entre 0.6 y 1.1 metros de longitud as- bution of the Furnariidae (Aves, Passeriformes). cienden suavementey terminan en una camara liger- Bull. Am. Mus. Nat. Hist. 166:1-357. amente mayor. En una galerla recientemente aban-

The Condor 90:253-256 0 The Cooper Omithologml Society 1988

A HIGH INCIDENCE OF BROWN-HEADED COWBIRD PARASITISM OF WILLOW FLYCATCHERS’

JAMESA. SEDGWICKAND FRITZ L. KNOPF U.S. Fish and Wildlife Service,National Ecology Center, 1300 Blue SpruceDrive, Fort Collins, CO 80524-2098

Key words: Willow Flycatcher; Brown-headed cated in North Park, a high elevation (2,500 m) inter- Cowbird;broodparasitism; Colorado;Empidonax trail- glacialbasin. We studiedWillow Flycatchers lii; Molothrus ater. along a 7-km stretch of the floodplain of the Illinois River. Willow Flycatchersare common in North Park, Brown-headed Cowbirds (Molothrus ater) parasitize occurringwherever there are extensivestands of healthy both Willow (Empidonax traillii) and Alder (E. al- shrub willows, Salix spp. (Knopf and Sedgwick, un- norum)flycatchers(Friedmann et al. 1977, Friedmann publ. data). Eight speciesof willows characterizedthe and Kiff 1985). These two flycatcherswere considered woodycommunity at ANWR (Cannonand Knopf 1984) a single species until 1973-(AOU 1973) which has with Woods rose (Rosa woodsii)and golden current masked information about the frequency with which (Ribesaureum) being minor componentsof the woody each is parasitized. Whereas several studies of the su- community. Common timothy (Phleumpratense), blue perspecies(Traill ’s Flycatcher) have focusedon or in- flag (Iris versicolor),and severalspecies of sedges(Car- cludeddetails ofcowbird parasitism,most were ofeast- ex spp.) dominated the herbaceouslayer. ern populations, and most reported frequencies of As part of a larger study of avian populations and parasitism ~21% (Hicks 1934, Berger195 1,Bergerand habitats, we surveyedthe avifauna of the willow com- Parmalee 1952. Walkinshaw 1966. Holcomb 1972). munity usingpointtransects after Reynoldset al. (1980) Friedmann et al. (1977: 13) suggestedthat westernpop- during June of 1985 and 1986 (see Knopf et al. 1988). ulations(Willow Flycatchers)are parasitizedonly about Population densities of common species, including half as much (ca. 10%) as easternpopulations (Traill’s Willow Flycatchersand Brown-headedCowbirds, were Flycatcher superspecies).This note describes a high calculatedusing nroaram TRANSECT (Bumham et al. rate of cowbird parasitismwithin a population of Wil- 1980) adapted?or point transectdata (R. P. Bumham, low Flycatchersin northcentral Colorado. We include pers. comm.). details of responsesto parasitism and host vs. cowbird We located most nests during constructionor early fledging success. in the egg-layingperiod. Intensive nest searcheswere conducteddaily following avian censuses.After several STUDY AREA AND METHODS censuses,virtually all territories were identified and we The study was conductedin northcentral Colorado on narrowed the scope of our search to activity centers the Arapahoe National Wildlife Refuge (ANWR), lo- within territories. We followed females to locate most nests; other nests were found by searchingindividual busheswithin territories. Nests were visited every 1 to I Received 3 April 1987. Final acceptance27 August 2 days during the period of maximum susceptibilityto 1987. parasitism (nest building through egg laying) and at