In order to assess the effect of tsetse control on livestock productivity, 21 herds of cattle of N'Dama, Baoule and Zebu cross breeds in the region of Boundiali, northern Cote d'Ivoire, were monitored monthly for trypanosome prevalence, packed red cell volume (PCV), body weight and reproductive performance between January 1987 and December 1989. Mean calf growth rate over the wet season of 1987, prior to tsetse control, was 243 + 48 (standard deviation among herds) g per day; mean cow body weight in 1987 was 223 + 15 kg; mean cow PCV, 29.8 + 2.5 percent; mean conception rate by 6 months postcalving, 28 + 22 percent; median monthly trypanosome prevalence was 21.4 percent (range 3.9-32.0 percent) in animals under 24 months of age and 12.6 percent (range 0.0-44.7 percent) in cows. A tsetse- control campaign using alpha-cypermethrin-impregnated traps was introduced in December 1987 to control trypanosomosis. Regression analyses were used to relate changes in mean PCV, calf growth rate, and cow body weight and conception rate between 1987 and 1988/1989 (the period of tsetse control) to corresponding reductions in trypanosome prevalence. An average 13.6 percentage unit reduction in mean trypanosome prevalence in calves between 1987 and 1988/1989 was associated with an increase of 12.2 + 3.6 (standard error, SE) kg in calf liveweight gain over the 7-month wet season. In cows, a corresponding 8.7 percentage unit reduction in trypanosome prevalence was associated with a 4.9 + 2.3 (SE) kg increase in body weight and a 1.3 + 0.3 percentage unit increase in PCV. An 18.4 + 8.0 (approximate SE) percentage unit increase in the number of cows conceiving within 6 months of calving resulted from a mean decrease of 6.8 percentage units in trypanosome prevalence from 1987 to 1988. The difficulties in obtaining estimates of effects of health interventions on livestock productivity that are both precise and unconfounded with other effects over time are discussed.