It is only within the past ten years that the value of retinoscopy as an objective test in the estimation of errors of refraction, has been fully recognized. The fact that, after a little practice, the kind and amount of refractive error can be ascertained quickly, accurately and independently of the patient, has made it deservedly popular. This popularity is largely due to the labors of Philadelphia men, particularly of Drs. Jackson and Thorington, and I therefore feel that it is quite appropriate that any additions I may have to offer may receive your friendly criticism. The mirror which I use is plane, 2.2 cm. in diameter, with a central aperture 3 mm. in diameter cut through the glass. My objection to a mirror with the sight-hole obtained by merely scraping away the silvering, is that vision is obscured by the dust which unavoidably collects on the glass and which