Monsoon as the discharge fell to 750 cubic feet a second the surface level would be raised to +12 by a temporary dam across the 45 foot gap, the water flowing through the interstices in the stone dam. In the comparatively still water thus induced a front bund (a) was to be made by depositing earth from boats to a level of +8. Four syphons 4 feet diameter were then to be put up and filled, to take the whole discharge of the river; the temporary dam across the gap was to be removed, and the surface level in front would fall to +8 leaving the top of bund (a) dry, it being protected during the fall of the water by a covering of halved bamboos lashed together and weighted with stones. A rear bund of earth (b) could then be completed and the enclosed space pumped dry.
The bottom layers of puddle lining and of the main earthen dam could then be carried on inside, and at the same time the front and rear bunds (a) and (b) and the stone dam raised. When the front and rear bunds reached +20 and +15 two of the syphons were to be raised, followed by the other two, and their former positions made good. The whole was thus to be raised in two foot layers, the bunds (a) and (b) always 5 feet to 10 feet higher than the enclosed space, and the stone dam 15 feet higher, with a corner always left open for the discharge of casual freshes. By the middle of February Major Ryves calculated that the earth and puddle would have reached a level of +25, and the stone dam would be continued across the gap and be at a level of + 42. In this condition he intended to leave the work till the next season, the rear bund (b) being covered with halved bamboos loaded with stone, his computation being that the proportion of the river discharge which percolated through the stone dam would have a velocity of only 14 feet a second, while that flowing over the top would run at about 6 feet a second. A certain amount of damage would, he allowed, occur to the rear slope of the stone dam, but not enough to materially endanger it.
So far as Major Ryves observations showed the stone dam at this level was capable of passing through its interstices the whole discharge of the river from June to November for 9 days out of 10. He therefore considered that this dam might be continued throughout the whole of the working season, and also the earth and puddle on one flank. During the dry season the latter would be brought up to a uniform level the whole way across, and in this manner he calculated that by the fourth season the work would be brought up to the level of a saddle at +110, across which all the water would then be passed, and from this point onwards the stone dam in rear could be discontinued. On the completion of the main
