3. The assumption of short marginal notes into the text, appears to have been a frequent source of various readings. When such notes supplied ellipses in the author’s language, or when they conduced much to the perspicuity of an obscure passage, the copyist would be very likely to incorporate the exegetical phrase, rather than that it should either be lost to the reader, or should deform the margin. 4. Transcribers frequently thought themselves free to substitute modern for obsolete words or phrases; and sometimes they consulted the wishes of their customers, by exchanging the forms of one dialect (of the Greek) for those of another; or, more often, for the common forms of the language. Alterations of this kind have often been the occasion of bringing authentic works under needless suspicion; for when the text has contained words or phrases which are known to belong to a later age than that of the supposed author, such incongruities have seemed to afford proofs of spuriousness. 5. Intentional omissions, interpolations, or alterations, were unquestionably sometimes ventured on by transcribers. But so many are the means we possess for detecting any such wilful corruptions--drawn from a comparison of different manuscripts, or from the incongruity of the interpolated passage, that there is perhaps, altogether, more probability that, from some accidental peculiarity of style, genuine passages of ancient authors should fall under suspicion, than that any actually spurious portions should entirely escape suspicion and detection.