The Bakerian lecture. On the theory of light and colours

Although the mode, much practised by the ancients, of accounting for a variety of phaenomena by a preconceived hypothesis be, if not wholly exploded, at least greatly discountenanced by modern philosophers; yet it must be owned, that when a number of facts have been collected and duly ascertained, it cannot but be conducive to the extensions of knowledge, to arrange them under certain heads, and, if possible, to ascribe them to some general cause; and that with men who are candid and not over-tenacious, even an error in such a proceeding may often be the means of eliciting some further information, which progressively may tend to the advancement of science. The immortal Newton has given us a striking example of this in his Theory of Light, which , should the principle he assumed prove ultimately erroneous, his investigation and mode of reasoning will yet remain an everlasting monument of acuteness and ingenuity, which it is likely will ever be found the best source of information to those who shall engage in this delicate branch of natural philosophy.