Fluconazole, an experimental azole antifungal agent with good tissue penetration following oral administration, offers the possibility of a new approach to the treatment of keratomycosis. Its efficacy as an orally administered agent was investigated in two models of experimental fungal infection in Dutch-belted rabbits. The study proceeded in three stages. In the first, a model of keratitis due to Aspergillus fumigatus was developed, the suitability of quantitative isolate recovery techniques for the evaluation of the disease caused by this organism was confirmed, and the correlation between the severity of clinical disease scored nonparametrically and the isolate recovery rate was established. The model was found to be most useful for study during the first 5 days of infection. The natural course of experimental Candida alibcans keratitis was evaluated and, on the basis of quantitative isolate recovery techniques, this model was found to be appropriate for studies lasting up to 1 week. In the second stage, corneal uptake following oral administration of fluconazole was studied in Dutch-belted rabbits. The drug was found to readily penetrate the cornea in amounts that correlated with serum levels (R = 0.89). Eight hours following a single 20 mg/kg dose, the corneal level was 7.4 mg/gm, almost double the amount when a 10 mg/kg dose was administered. When given in a twice daily divided dose, fluconazole accumulated steadily in the corneas over a period of 5 days. The presence of inflammation induced by fungal infection did not influence corneal uptake. In the final stage, the efficacy of orally administered fluconazole in the treatment of keratomycosis was evaluated. Overall, a significant therapeutic effect was observed with both infections. Treatment of the animals with oral fluconazole for 1 day prior to inoculation with Candida albicans led to a significant decrease in isolate recovery 1 day later (P = 0.01). However, when treatment was continued for 5 days following inoculation, no additive effect of pretreatment was noted. Pretreatment for 1 day followed by 5 days postinoculation treatment led to a significant decrease in clinical disease (P less than 0.05) and isolate recovery (P = 0.05). A beneficial effect of pretreatment compared to treatment begun 1 day postinoculation, as measured by a reduction in clinical severity and isolate recovery, was also noted. On the basis of these short-term therapeutic studies and the excellent corneal penetration of fluconazole, further investigation of oral therapy of keratomycosis appears warranted.