These words of Rigaud indicate the fame deservedly acquired by an earnest and simple-minded devotion to science: but can we learn anything from the study of Bradley's work to guide us in further research? The chief lessons would seem to be that if we make a series of careful observations, we shall probably find some deviation from expectation: that we must follow up this clue until we have found some explanation which fits the facts, not being discouraged if we cannot hit upon the explanation at once, since Bradley himself was puzzled for several years: that after finding one _vera causa_, and allowing for the effect of it, the observations may show traces of another which must again be patiently hunted, even though we spend nineteen years in the chase: and that again we may have to leave the complete rectification of the observations to posterity. But though we may admit the general helpfulness of these directions, and that this patient dealing with residual phenomena seems to be a method capable of frequent application, we cannot deduce any universal principle of procedure from them: witness the difficulty of dealing with meteorological observations, for instance. It is not always possible to find any orderly arrangement of the residuals which will give us a clue to start with. When such an arrangement is manifested, we must certainly follow up the clue; it would almost seem that no expense should be prohibitive, since it is impossible to foresee the importance of the result. CHAPTER IV ACCIDENTAL DISCOVERIES [Sidenote: The Oxford new star found during work on Astrographic Chart.]