On triple objectives with complete color correction

Abstract

THE prime defect in the large refractors of tbe present day is the secondary spectrum. This, arising from the irrationality in the spectra. produced by the crown and flint glass, hardly noticeable in small apertures, detrimental in telescopes of medium power, is positively obnoxious in the large instl"Uments and will speedily put an end to farther increase in dimensions. On this account thel·e have been many efforts to produce two kinds of glass differing sufficiently in dispersive power, which would still yield mutually rational spectra. As far as I know we are now no nearer success in this direction than when Brewster investigated the subject fifty years ago. Can we secure the same end by increasing the number of glasses in the objective? Theoretically, since a new disposable constant for color change is introduced with each lens in the system, tbe answer is evidently affirmative; but if we limit ourselves by the condition that the construction shall be practicable, i. e., that there shall not be too many lenses and the curvatures shall be moderate, the conclusion is not so read.y. On entering the discussion we will assume three as the limiting number of lenses and n the focal length as the minimum radius of curvature. The formula Tor tbe focal lengtb F of three thin lenses in