Purpose: To compare the efficacy of 0.1%, 0.18%, and 0.3% hyaluronic acid (HA) artificial tear in the treatment of experimental dry eye (EDE).Methods: EDE was established in female C57BL/6 mice through an air draft and subcutaneous scopolamine injection. The mice were divided into 5 groups according to topical treatment regimens (n = 5 each): EDE control, balanced salt solution (BSS), preservative-free 0.1% HA, 0.18% HA, and 0.3% HA. The tear film break-up time (TBUT) and corneal fluorescein staining scores were measured 5, 10, 14, 21, and 28 days after treatment. The corneal smoothness scores were measured. In addition, periodic acid–Schiff (PAS) and terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) staining were performed.Results: The values for TBUT and corneal fluorescein staining showed greater improvements in all the HA groups (P < 0.05) than in the EDE and BSS groups after 10 days of treatment. Mice treated with 0.3% HA showed a more significant improvement in all clinical parameters than did those in the EDE control, BSS, 0.1% HA, and 0.18% HA groups (all P < 0.05) after 28 days of treatment. The goblet cell counts were higher in the 0.3% and 0.18% HA groups than in the 0.1% HA group. The number of TUNEL-positive cells was the lowest in the 0.3% HA group.Conclusions: In EDE, 0.3% HA artificial tears are more effective than the 0.1% and 0.18% HA in improving tear film instability and ocular surface staining and irregularity, in increasing the number of conjunctival goblet cells, and in decreasing corneal epithelial apoptosis.