INTRODUCTION
final nu in both P and A. It may be due partly to pronunciation, partly to the use of abbreviation in the original, sometimes it is clearly caused by a misunderstanding of the imperatival infinitive. The omission of rubricated capitals or the rubrication of erroneous capitals is very common in A, especially in the later Books. This need not deceive us; it is akin to our modern misprint. More important is the evidence of the existence in the archetype of marginal or interlinear variants. Here and there, these have crept into our manuscripts; there are also a few cases where glosses appear to have been embodied. These are, however, very few, I believe, though many more have been suspected by critics of the Cobet school. These presumed glosses were supposed by Nauck to be the work of a needy schoolmaster (iv. 30); Schenkl discovers occasionally the bowdlerism of a prudish scribe or the reverence of a Christian monk. More than one critic has detected the work of a physician, who doctored his copy with scraps of medical lore. Dr. Rendall has indicated many glosses which he presumes to have arisen in this way. He even suggests (no doubt half playfully) that the great Galen may sometimes have been at his imperial patient's elbow as he worked. One recent critic, presuming these to be glosses, traces the hand of the young doctor Toxites upon the manuscript which he brought to Conrad Gesner. He forgets that A contains the same additions, and that Toxites had no access to that manuscript.
I have spoken above of dislocations of certain places in the text. Gataker expressed the same suspicion in more than one of his notes. He and the great Saumaise in the seventeenth, and Morus in the eighteenth century, suggested transpositions of some shorter passages. Leopold and Schenkl, following Coraes, have made such a change in x. 1, and Dr. Kronenberg has lately proposed one such change, a change which had occurred to myself indepen-
