Aim of the Work: The aim of this study is to evaluate and compare the epi-on versus the epi-off techniques of corneal collagen cross linking as regards their safety and efficacy for treatment of mild to moderate degree keratoconus.   Design: Prospective, non randomized study.   Patients and Methods: Thirty eyes in 15 patients (9 males and 6 females) with bilateral mild to moderate degree keratoconus were included in this study. Their mean age ± SD was 26.2 ± 3.9 years. Diagnosis of keratoconus was based on clinical evaluation as well as pentacam examination. All patients were subjected to corneal collagen cross linking. According to the technique used, eyes were classified into 2 groups:   Group I: Included 15 eyes in 15 patients, where cross linking was performed after removal of the corneal epithelium (Epi-off). Group II:  included the other 15 eyes of the same patients, where cross linking was performed with intact epithelium (Epi-on). Comparison between the two groups as regards visual outcome, keratometric readings, least corneal thickness, refraction outcome and corneal haze was done.   Results: No intra operative complications were reported in our study. Re-epithelialization in eyes of group I was reported within a week except three eyes were re- epithelialization was reported after 10,14and 21 days. There was a statistically significance improvement in best corrected visual acuity in both groups (0.36 before versus 0.60 six months after surgery) but the difference between both groups was not significant. As regards refraction, there was reduction in spherical error 6 months after surgery in both groups (-6.14D before,-5.22D 6 months after surgery) as well as cylindrical error (4.87D before 3.79 D 6 month after surgery). Differences between both groups were statistically not significant. There was increase in the least corneal thickness after surgery (448.6 microns before surgery versus 451.9 microns 6 months after surgery) but statistically the difference was not significant (P=0.75) and the difference between both groups was also not significant. Changes in keratometric readings were statistically not significant, and the difference between both groups was also not significant. Corneal haze was observed one month post operatively in four eyes in group one versus 3 eyes in group II. Persistent haze at the end of follow up was reported in three eyes in group I versus two eyes in group II. The incidence of haze as well as its density was higher in group I than group II with statistical significant difference.            Conclusion: Both epi-on and epi-off techniques of corneal collagen cross linking are safe and effective in stabilization or even improvement of mild to moderate degree keratoconus as regards best corrected visual acuity, refraction, keratometric readings, and least corneal thickness. The epi-on technique is easier and more tolerable by the patient with less postoperative corneal haze.