No method for the objective measurement of corneal surface roughness has thus far been available for experimental and clinical investigations. The surface roughness of photoablated porcine corneas was investigated with the silicone-cast method and automatic confocal topometry. Photoablation was performed with the 193-nm excimer laser. Surface roughness was determined as a function of the repetition rate (2-60 Hz; fluence, 180 mJ/cm2; beam diameter, 2 mm; 300 laser pulses/experiment). Immediately after photoablation, a silicone replica of the ablated surface was made and analyzed later using confocal topometry. The surface roughness for excimer laser ablation was found to be frequency-dependent: 0.4-0.6 +/- 0.08 micron (2-30 Hz) and 1.2 +/- 0.12 micron (60 Hz). The silicone cast of the cornea of a healthy eye gave an exact image of the profile and curvature: the difference between radius measurements of the cornea (Zeiss ophthalmometer) and of the silicone cast (laser topometer) was about 0.5%. As the silicone-cast method can also be used for in vivo animal experiments, the correlation between surface roughness and laser repetition rate can be studied with regard to photorefractive keratectomy.