Various marketed chloramphenicol ophthalmic solutions were compared and various dilutions of Tobrex Ophthalmic Solution were tested for effectiveness in a Staphylococcus aureus rabbit keratitis model. Anesthetized rabbits were each infected intracorneally with 10(4) Staphylococcus aureus ATCC 29737 cells. Treatment groups consisted of five or six rabbits (10 or 12 eyes) per group. One group of rabbits was infected but not treated (Positive Control Group). Topical dosing of commercially available ophthalmic solutions was accomplished by depositing 0.1 mL of a color-coded test solution into the lower cul-de-sac of each eye. Dosing begin one hour after the mid-infection time and continued for a total of nine hourly treatments. Twenty-four hours after infection the rabbit eyes were graded (masked) using standard slit-lamp scoring procedures. The slit-lamp scores for five of eight ocular parameters were used to calculate an eye score value for each rabbit eye. The five ocular parameters were selected, based on previous Stepwise Discriminant Computer Analysis of over 300 rabbit eyes infected with Staphylococcus aureus ATCC 29737 and treated with various antibiotics. The eye score values for each group were averaged and the treatments were compared for significant differences in efficacy using the nonparametric, Wilcoxon-Mann-Whitney Rank Sum Test.(ABSTRACT TRUNCATED AT 250 WORDS)