Abstract: Eight pairs of pony mares were observed. Members of a pair were housed in adjacent stalls and fed hay ad libitum. The behavior of both ponies was recorded simultaneously in the morning (1000 to 1200 h) and afternoon (1400 to 1600 h) for a total of 117 h. The time budget was: 70.1 {+/-} 8.6% eating; 17.8 {+/-} 7.4% standing (including stand rest, stand alert and stand nonajert); 5.2 {+/-} 7.0% pushing hay; 2.9 {+/-} 1.2% walking; 1.9 {+/-} 2.9% drinking; 1.3 {+/-} 1.1% self-grooming; .2 {+/-} .3% defecating; .06 {+/-} .1% chewing nonfood items; .06 {+/-} .03% urination; .06 {+/-} .1% licking salt; .07 {+/-} .1% pawing hay; .6 {+/-} .7% lying and .07 {+/-} .08% stretching the neck over the stall wall dividing the ponies. While eating, the ponies lifted their heads 25.4 {+/-} 11.0 times/h. In less than one-half of the occasions when urination or defecation was observed, the ponies walked away from the spot where they had been eating to eliminate. During one-half of the observations, visual contact between the ponies was prevented by a solid partition between the stalls. The ponies spent significantly more time standing nonalert when the partition prevented visual contact (12 {+/-} 7%) than when visual contact could take place (6 {+/-} 3%, P<.05). When fresh hay was supplied in the mornings, the ponies spent similar amounts of time eating whether visual contact was allowed or not, but in the afternoon significantly more time was spent feeding when visual contact was allowed (73 {+/-} 4%) than when it was not (60 {+/-} 7%). Less time was spent eating, in the absence of visual contact, despite the presence of auditory and olfactory contact. Apparently social facilitation is important in maintaining feeding behavior in ponies. N1 -