Two groups of three horses each were fed either a control diet of ground hay and concentrates or a similar diet with 10.5% added fat. The horses were trotted at 2.9 m/sec to a heart rate (HR) of 160 beats per minute (BPM) 5 d/wk on an equine treadmill set at go for a training period of 12 wk. A standard exercise test (SET) was performed at wk 0, 6, and 12. Blood samples were taken before, during, and after exercise and analyzed for lactate, glucose, cholesterol, total serum protein, and free fatty acids (FFA). Muscle biopsies were taken before and after exercise during the SETs on wk 1 and wk 12. Differences between wk 0, 6, and 12 in blood levels were found: lactate (P < .01) and glucose (P < .001) decreased, while cholesterol (P < .01) and FFA (P < .05) for both groups, indicating a conditioning response. However, no differences were found in any of the blood factors between groups. In the final SET, pre-exercise muscle glycogen levels in the fat-fed horses were lower (13.61 vs 25.44 mg/g wet tissue). Post-exercise glycogen levels, however, were higher in the fat- fed horses (10.95 vs 8.27 mg/g wet tissue), indicating that there may have been a glycogen-sparing effect. The fat-fed horses avg .3 kg/d less in daily consumption, and avg .03 kg/d more in ADG (.36 kg/d for control vs .33 kg/d for fatfed). Differences in trotting times (time to HR 160 BPM) between groups were seen at the end of four (P < .05) and eight (P < .05) Wki control horses improved 39% while fatfed horses only 25% from wk 1 to wk 8. No differences were seen between groups from wk 9 to 12. Improvement over the 12 wk period was similar for both groups (control 61%, fatfed 67%). These results seem to suggest that although fat appears to be a reasonable source of energy for the horse, there may be a long adaptation period for horses fed higher levels of fat.