U.S. Department of Agriculture: Animal and Plant Health Inspection Service


Date of this Version



Proc. 19th Vertebr. Pest Conf. (T.P. Salmon & A.C. Crabb, Eds.) Published at Univ. of Calif., Davis. 2000.


As part of efforts to develop The Livestock Protection Collar (U.S. EPA Reg. No. 56228-22), we videotaped sheep-predation events by 23 (15 males and 8 females) wild-caught, confined coyotes (Canis latrans) in a 31 x 41 m enclosure. Coyotes were paired individually with a sheep (Ovis aries) during 1 h daily trials. Nineteen (13 males and 6 females) of the coyotes made 75 fatal attacks of 1 to 7 sheep each; 4 coyotes (2 males and 2 females) made no fatal attacks despite 19 to 39 daily pairings. Of coyotes that made fatal attacks, 13 (9 males and 4 females) always attacked at the neck of sheep; 5 (4 males and 1 female) always attacked by nipping at the legs/head/back of sheep; and 1 attacked at the legs/head/back of sheep during two initial events, but subsequently attacked at the neck of sheep. Greater time in captivity was not correlated with trials preceding a fatal attack (rho= +0.23). Among coyotes making ≥ 2 fatal attacks, subsequent predation events occurred after fewer intervening pairings with sheep. Initial feeding sites occurred most frequently at the flanks/ribs of sheep. Although collected between 1976 to 1980, these observations represent a never-to-be-acquired-again data set that remains timely. Data showed that not all coyotes display sheep-predation behaviors or kill sheep efficiently. Instrumental learning and stimulus-habituation models of coyote predation behavior are discussed.