Abstract The experiment was performed at the Waggoner Ranch, Laramie, Wyoming. Experimental animals were commercial cattle of mixed breeds born in the spring of 1993. Cattle were housed in 4 separate pens at the rate of 5 animals per pen. Each pen was approximately 20 X 60 ft. Cattle were fed at the rate of 2 lbs of corn per animal per d, plus free choice alfalfa/grass hay. Salt, mineral mix, and well water were provided ad libitum. Efficacy was determined by counting all species of lice present on the animals. Lice infestations were naturally acquired. At the time of the pretreatment counts (8, 9 Mar) all cattle were infested with Linognathus vituli, 18 were infested with Bovicola bovis and two were infested with Haematopinus eurysternus. Treatments were applied, following pretreatment lice counts on 9 Mar 1994. Post treatment counts were performed 6, 14, 21, 28, 35, 42, 49 and 56 DAT. For purposes of examination, cattle were restrained in a squeeze chute with a head catch. A high intensity lamp provided illumination. Lice were counted in eight examination sites on each animal. Examination sites and dimensions were as follows: topline 5 X 15 cm, withers 5X15 cm, around right eye 10 X 15 cm, around left eye 10 X 15 cm, right cheek 5 X 10 cm, left cheek 5 X 10 cm, muzzle 5 X 25 cm and dewlap 5X15 cm. Sample sites included predilection areas of L. vituli, B. bovis, and H. eurysternus. Following pretreatment lice counts, all animals were ranked by total counts of L. vituli. The two animals with the highest counts were randomly assigned to either replicate 1 or 2. The next two animals were then assigned to the other replicate. This procedure was repeated with each group of 4 similarly infested animals. Treatment was randomly assigned to each animal within a pair of similarly infested animals. Treatment was based on body wt: 15 ml per animal 750 lb or greater; 10 ml per animal 500 to 750 lb; and 5 ml per animal 250 to 500 lb. The cattle weighed between 302 and 846 lb. Insecticide was applied as a pour-on from a graduated container along the face, neck, shoulder and midline of each animal as far back as the dosage would allow. Total lice counts were transformed using In (total count + 1). A general linear model procedure (SAS) was used to analyze the transformed counts. Total variation in the lice count was partitioned into that attributable to the treatment, animal within treatment, sampling d, treatment X sampling d interaction, and residual. Geometric means for lice counts were estimated for each treatment group on each sampling d.