In September 1913 some two hundred teachers of Latin assembled at Cambridge to study the Direct Method. There was much demonstration, experiment, and discussion: there was great enthusiasm. I best remember one thing. W. L. Paine had brought a class of about a dozen Whitgift boys who had one year's Latin behind them. I was interested in their keen enjoyment of the work, their eagerness to speak and answer. But what surprised me was their accuracy. A boy was called out at the end of the lesson to write up on the board the story which, mainly by question and answer, had been put before the class: he did it with almost no mistakes, and quite quickly. Any such blunder as the confusion of nominative and accusative had met, as the lesson proceeded, with a chorus of immediate correction: but I was still surprised at the final proof of a far higher degree of accuracy at that stage than I had ever seen or believed to be possible. This shows that Dr. Rouse was not a lone magician, of infinite resource and sagacity, who had invented an instrument which only he could use. The Whitgift boys were quite as good as the Perse boys. Paine was a scholar, but of nothing like Dr. Rouse's quality in scholarship. He was a well-qualified Latin master, who had taught on old lines at Oundle and found the whole business abominably dull. After experimenting with oral work in a less degree, he was encouraged at Whitgift to adopt the Direct Method: there he found himself.