Chemoecology 10:049 – 050 (2000) 0937–7409/00/010049–02 $1.50 +0.20 © Birkha ¨user Verlag, Basel, 2000 Short communication An experiment on the ability of free-ranging turkey vultures (Cathartes aura ) to locate carrion by chemical cues William J. McShea, Elizabeth G. Reese, Thomas W. Small and Paul J. Weldon Conservation and Research Center, Smithsonian Institution, 1500 Remount Road, Front Royal, VA 22630, U.S.A., e-mail: wmcshea@crc.si.edu The nature of the cues to which vultures attend in locating carrion has been debated for more than a century (reviewed in Stager 1964). Some authors claim that vultures of the genus Cathartes, at least, use chem- ical cues to locate food (e.g. Coles 1938; Owre & Northington 1961; Stager 1964; Graves 1992). This suggestion is based primarily on observations that these birds find carcasses under conditions where visual cues are obstructed, as when food is concealed in dense vegetation. Few controlled studies of the involvement of chemoreception in foraging by vultures have been re- ported. Houston (1986) described a test of free-ranging turkey vultures (C. aura ) in Panama in which chicken carcasses were either covered by dry leaves or left uncovered on forest trails while vulture visitations were monitored. No difference was observed in the time taken by birds to discover concealed versus exposed carcasses, indicating their use of chemoreception. We report here an additional field experiment on the ability of C. aura to locate carrion by chemical cues. This species, which ranges from Canada through North, Central, and South America to Tierra del Fuego, is a scavenger that subsists primarily on verte- brate carcasses. Our results provide further evidence that free-ranging C. aura uses chemoreception to ex- ploit this food resource. Our study was conducted from 23 September to 10 October, 1997, and from 31 March to 14 July, 1998, at the Conservation and Research Center of the Smithso- nian Institution in Front Royal, Virginia, which con- sists of 1300 ha of deciduous forest and open grass fields on hilly terrain. Four experimental stations were established on grassy hills 600 to 1100 m from a central observation site. Each station consisted of four bales of hay (1.0 ×0.8 ×0.5 m) arranged at right angles and spaced to create an inner cavity measuring 1 m 3 . The area of each station, as viewed from above, was ap- proximately 4 m 2 . Each station was situated a mean distance of 375.3 921.8 m from adjacent stations; the greatest distance between stations was 1150 m. The carcasses of adult white-tailed deer (Odocoileus 6irginianus ) obtained as traffic casualties (in varying states of decomposition) were used as bait. A carcass was placed into the cavity of one of the four stations at between 08:30 and 10:00. A 15-cm layer of hay was then loosely scattered over the carcass to conceal it visually. The cavities of the other three stations were filled only with hay. The station containing the carcass in the first trial was selected at random; it was then randomly assigned to one of the three stations that had not contained a carcass in the previous trial. The area around our study site was monitored using binoculars until 18:00 during the first day of a trial, and from 07:00 up to 18:00 during each of up to three consecutive days. Once vultures were sighted in the area, stations were monitored on a discontinuous schedule with intervals of up to 30 min. Trials were terminated when a vulture landed at a station, or after 96 h had elapsed. We also noted the presence of black vultures (Coragyps atratus ) in the area. Seventeen trials were conducted, seven of which were discounted from our analysis. Four trials were discounted because carcasses were removed overnight or uncovered and partially consumed by scavengers, probably black bears (Ursus americanus ). Three other trials were discounted because they were abandoned after 96 h due to lack of a response. A failed response to a station baited on 10 October, and the absence of C. aura from our study site at this time, may have been due to the seasonal migration of C. aura out of the area or, alternatively, to the greater availability of deer carcasses elsewhere (possibly due to the onset of hunt- ing season in early October). Hence, we postponed our study and resumed it the following March, when C. aura was again observed and weather conditions per- mitted us to conduct additional trials. During each of the remaining 10 trials, C. aura was observed to have arrived at experimental stations 8.5 to 72 h (x ¯ =38.5 h) after baiting. The arrival times ranged from 07:00 to 17:30, with four occurring before 12:00, and six afterwards. The failure of vultures to visit control stations suggests that they responded directly to cues arising from carcasses, rather than visiting stations as a learned response to associated food reward. C. atratus also was observed only at baited stations, but this species always appeared subsequent to the arrival of C. aura ; this is consistent with the suggestion that C. Correspondence to : William J. McShea