Proceedings.]AND THE WATER-SUPPLY OF W O . 481 supply or closed. If the demand from the cultivators for water sir Thomas was not sufficient to keep the canal or distribution channel runningWed* full, it was closed for a few days ; the demand was thus stimulated, and the channel could be run full for a time. If, on the other hand, the river did not give enough water t o fill all the channels, the water was divided among the channels in such a way that whatever channel was running flowed full. The problem of the regime channel was thus solved very simply. It became, in effect, subconscious knowledge. Mr. R. G. Kennedy, who was many years senior to himself, doubtless had that, as it were, in his make-up, and considered regime only from the point of view of a channel running full. Mr. Lacey discussed in his Paper a channel working at lower stages. Sir Thomas Ward had found that it was not possible to get a channel to work satisfactorily a t different stages, but in Madras the experience was different, perhaps because there was no silt trouble there. One looked for evidence of silt trouble on the Lanks of the channel. If the cultivators had never had to be collected to clear out the silt, then it had not been troublesome. In the flood season, when the water from the river was muddy, the silt was deposited on the bed of the channel, and increased the slope until a regime for that quality or grade of silt was arrived a t ; later on, when the clearer water came in the season when the river was low, that silt was scoured out. In Madras, it might be remarked, the constant supply demanded by the rice irrigation necessitated the water in the river being equitably divided out amongst all the channels, and they were run with whatever supply was available. When an engineer was designing a canal in a new country, and had to find out what gradients he might use in his channel, he had no traditional knowledge to draw on, such as was available in Northern India, and in such a case it was now usual to follow the advice given by Mr. Kennedy in his Paper 1 and to examine existing channels to obtain the curve for the particular class of silt. That had been done by the late Dr. A. A. Stoddard in Siam ; he selected from among the effluents of the Bangkok River, from which he was about to design canals, those channels which he found were in regime, and made observations on them, designing his canals from the data thus obtained. Incidentally, the results agreed very closely with those of the formula which was already in use in Egypt. That was more or less to be expected ; by the time a river reached its delta, the silt was very fine, and it would probably be found that in the deltas of all rivers the grade was much the same. With reference to the discussion by the Author on p. 426 of the regime of rivers, in the case of tidal rivers-for example, the Thames-the Mr. J. M. LACEY congratulated Mr. Gourley on his excellent Paper and on its careful treatment of details, but wished to ask a few questions. He noticed that the river meandered a ...
