623
parts of broken stone. The second and third quality mortars seem less likely to be homogeneous. The best and only certain plan is to determine the proportions experimentally, according to the voids actually found in the matrix.
yellow earth
5. I attach so much importance to complete solidity that, if concrete propor- Use of red tioned as I have described, contains more cement than can be afforded, I should try with cement. the effect of adding good red earth, which is of the nature of Puzzolana. Recent experiments made in Germany, show that the addition of Puzzolana to Portland cement, improves its setting properties. Portland cement, in setting, liberates a certain proportion of hydrate of lime. When Puzzolana is present it combines forthwith, with the lime liberated, thus preventing effervescence and obviating porosity. In certain water-works in Mauritius a mixture of burned coral-lime, coral sand, red earth and a small proportion of cement, was used with success. It was also found that the addition of red earth to the cement and sand mortar, used for jointing pipes, completely obviated porosity, which was most difficult to avoid when sand alone was used. I think that it would be well worth while to experiment with various mixtures of stone, sand and cement, with and without red earth, the voids in any case being filled. If 9" cubes, of the several samples, were made, and sent home to me at the University College, I shall be happy to have them crushed, in the testing machine, belonging to the Engineering Laboratory. The exact strength of each mixture would then be known, and it could be allotted to the part of the work for which it proves suitable.
foundations.
6. In my original report, I recommended that foundations should not be Stepping stepped. I then referred to stepping in the longitudinal section of the dam. In the cross-section, at right angles to its length, stepping may, and in many cases should be resorted to.
drainage of
7. It is desirable to construct a regular Drain or, Culvert running along the Construction whole of the length of the dam, at the lowest point of the foundation and located of dam immediately behind the water-tight mass, forming the water-front or inner lining foundations. of the dam. This Drain will serve to collect and carry off any spring-water, that may be found during construction. It should have a free outlet at the lowest level possible, so that, should there be any failure to produce an absolutely water-tight junction with the soil or rock, it will effectively prevent any accumulation of water under full pressure, beneath the foundation, so as to exert a prejudicial upward pressure. If there be any other water-bearing springs or fissures, in any part of the foundation, their water should be collected and conveyed to the main drain or to the outlet channel from it, which should always be left open. If there be any leakage it is better that it should be visible and escape freely than that it should accumulate and perhaps do unseen and unknown mischief.
superst ruc-
8. I trust that if the suggestions which I have made as to the composition of Drainage of concrete be adopted the whole mass of the dam will be homogeneous and water- fure of dam. tight. Nevertheless I think that it would be a prudent precaution to provide open channels or drains, extending through the whole mass from the outer-face to with- in about 4 feet of the inner water-tight face. These drains might be of 3" Chinese stoneware pipes, of the commonest quality, spaced vertically and horizontally, about ten feet apart. These pipes can do no harm, they will cost little, and they may, if any local mistake be made in construction, be the means of relieving a congestion of water, which might be prejudicial. They will, at least, afford the means of localising any leakage, should one exist. I am fully aware that the pro- vision of drains, through the mass of a masonry dam, is not a common practice. Mr. MANSERGH is not doing so at the Birmingham water-works. But it is to be remembered that these works are being carried out in England, departmentally, and with an abundance of highly skilled supervision, and not by a Chinese con- tractor, and limited skilled inspection.
9. After discussion with Messrs. CHATHAM and GIBBS, I see no reason for The outlet altering the dimensions of the outlet culvert.
culvert.