renewing them, I and the three others at last consented to take part in the venture. As soon, therefore, as the London season was over we began our preparations, which would necessarily be somewhat lengthy. From the beginning of August up to the end of October we met again and again at Beckett's house in Yorkshire, our proceedings being shrouded in serio-comic secrecy. In order that we might perfect ourselves in the use of our mathematical weapons, each day after breakfast the dining-room table was cleared and covered with a large green cloth divided into numbered spaces, like the green roulette board at Monte Carlo. In the middle of this was placed a large roulette. Rakes were provided of the true Monte Carlo pattern. One of us played the part of croupier, while to each of the others was allotted a certain number of counters indistinguishable in aspect from twenty-franc gold pieces. Each of us made his own calculations on cards provided for the purpose; each day we played solemnly for four hours on end, and then examined the results. We sometimes varied this routine by taking one of the Monte Carlo records, our croupier not turning the wheel, but calling out the numbers or colors seriatim which had actually turned up at the tables on this day or that. The general results were, I must say, most extraordinary. On only two occasions did the operations of an entire day leave any of the players bankrupt or without a substantial gain, though they all started their work at different moments, and the actual details of the staking were in no two cases the same. Throughout a long series of these experimental meetings, the winnings were as a whole about 20 per cent. daily on the total capital risked. Encouraged by these results, we had no sooner mastered the system as a mathematical scheme than we promptly made arrangements for beginning the work in earnest. We all thought it desirable that, until it was crowned with success, our enterprise should remain unknown to anybody except ourselves. It was therefore settled that our journey should take place at once--that is to say, about the end of October, at which time Monte Carlo would be nearly empty, and we should run least risk of encountering loquacious acquaintances or of having our secret stolen from us by inquisitive and sinister rivals. We accordingly secured in advance--since all the great hotels at the time in question were closed--a suite of four bedrooms and a sitting room at a small establishment called the Hotel de Russie. Its appointments, when we arrived, proved to be so simple that the floor of the restaurant was sanded; but the rooms upstairs were comfortable; and not even at the Hotel de Paris could anything better have been found in the way of wine or cooking.