When we turn to the contents of our manuscripts, the first problem which awaits us is that of weeding out from the whole collection what is dubious or certainly spurious. We may start with the fact that certain insignificant items of the collection were already recognized as spurious when the arrangement of the dialogues which we find in our oldest Plato MSS. was made. By counting each dialogue great or small as a unit, and reckoning the collection of Epistles also as one dialogue, a list of thirty-six works was drawn up, arranged in " tetralogies " or groups of four. It is not abso­ lutely certain by whom or when this arrangement was made, though it certainly goes back almost to the beginning of the Christian era and perhaps earlier. It is commonly ascribed by later writers to a certain Thrasylus or to Thrasylus and Dercylides. The date of

contents which is not thoroughly Platonic. In fact, it forms, as the Neoplatonic commentators saw, an excellent introduction to the whole Platonic ethical and political philosophy. It is just this character which is really the most suspicious thing about the dialogue. It is far too methodical not to suggest that it is meant as a kind of "textbook," the sort of thing Plato declared he would never write. And the character-drawing is far too vague and shadowy for Plato even in his latest and least dramatic phase. In the interlocutors, though they bear the names Socrates and Alcibiades, there is no trace of any genuine individuality-far less than there is even in the anonymous speakers in the Laws. It is a further difficulty that on grounds of style and manner the dialogue, if genuine, would have to be assigned to a late period in Plato's life when he is hardly likely to have been composing such work. On the whole, it seems probable that Alcibiades I is the work of an immediate disciple, probably written within a generation or so of Plato's death and possibly even before that event.