Water from the reservoir enters under pressure the masonry of a dam and also the rock foundation underneath. Under certain circumstances it may produce an uplift pressure sufficient to affect seriously the design of the dam. The proper solution of the problem will be a matter of debate as long as certain engineers are more conservative than others. Again, the nature of the solution will depend more upon the engineer's natural cast of mind than it will upon the circumstances of the particular case.

Whether or not prevention is better than cure; whether to eliminate the pressure by taking means to prevent the entrance of water or to counterbalance the pressure by adding masonry, will be decided largely according to the degree to which the engineer is willing to trust his reputation to his judgment. The limit of possible pressure is known, i.e., full reservoir head applied to the entire area of the plane under consideration. Further, an entirely adequate remedy lies in making the masonry section heavy enough. That safe and expensive method will always have its advocates. Not only is this an entirely adequate remedy but, if adopted, the necessary masonry section is susceptible of precise determination by the designer in advance of any data obtained during the preparation of the foundation and entirely independent of the judgment of any person as to nature of foundation or quality of masonry. Some reason and excuse for the conservative design lies in the fact that the judgment of the constructing engineer is often a questionable or unknown factor. It should at least always be possible to secure the services of a man of competent judgment to oversee and advise, and of a resident engineer who may be depended upon to carry out the prescribed measures faithfully and intelligently.

A rational view of the entire situation has been obscured by various loose and confusing ways of thinking about and stating the conditions and by a misapprehension as to just what it is desirable and possible to accomplish by certain preventive measures. For instance, there is the old familiar and perfectly correct proposition that any amount of leakage, no matter how infinitesimal, may produce a pressure equal to the reservoir head. In conjunction with this proposition it is pointed out how practically impossible it is to prevent some leakage through either masonry or foundation. Whereas the only pertinent conjunctive statement is the equally true one, namely that be the leakage "to" a certain point or plane of application a drop, a gallon or a thousand gallons per minute, day or year, the pressure at that point can be only that which is necessary to force such quantity away to an outlet at an equal rate. In other words, if leakage "to" exceeds leakage "from," pressure will result; if leakage "from" is relatively free and easy, pressure is impossible. Substitute money for water and affluence for pressure and we have Micawber's formula for happiness or misery.

Another source of error lies in considering a dam (masonry or foundation or both) to be a homogenous mass, or rather of equal permeability throughout; also in considering such a condition desirable and attempting to realize it even approximately. We may indeed treat an entire foundation in such a manner as to cut off 99 per cent, of the leakage, but when it has been accomplished we do not know whether the leakage has been cut off at upstream side, centre or downstream side, and the remaining 1 per cent, may have just as much destructive effect as the original 100 per cent. It is the same with the masonry. We cannot exclude the last drop of water, and in proportion as we succeed in making every part equally impermeable we succeed in making it uncertain where the water is cut off and whether or not any pressure can exist. We may realize a very high degree of impermeability but if we utilize the entire width of the dam in doing so we have accomplished absolutely nothing in the way of limiting or controlling the pressure. The actual amount of the leakage (barring streams of such a size as to cause erosion) is of small moment compared to the location of the limiting section of the channel upstream from which the pressure exists. With the problem thus stated it will be seen that the solution lies not in aiming for a tight dam or foundation but for a dam or foundation one part of which shall be much tighter than the remainder. One condition is uncertain and impracticable if not impossible; the other condition is perfectly feasible and certain of realization. We may be uncertain as to absolute tightness but we can be certain as to relative tightness. Possible horizontal seams below the masonry present no difficulties but rather tend to facilitate a satisfactory treatment.

In connection with the subject of uplift pressure some observations as to actual existing pressures will be of interest, although naturally they will settle no controversies. Indeed, arguments in support of widely differing views may be drawn by different engineers from the same observations.

In Engineering News for July 31, 1913, is an abstract of a paper which appeared in Zeitschrift fur Bauwesen, Vol. 63, 1913, page 102, giving some observations as to pressure under two German dams, the Oester and the Neye. The Oester has a maximum height of 131 ft. and a thickness of about 90 ft. Upon each of four sections across the dam three pipes led from the foundation to galleries where a pressure gage was applied. The foundation was of graywacke and shale varying from firm to rotten, with some clay crevices. The stratification was nearly in vertical planes, up and downstream, i.e., crossing the dam. Excavation in the bed rock was 13.1 ft. to 14.8 ft., with no sign or indication of a cut off trench and no mention of any special treatment like grouting. The gates were closed Feb. 16, 1907. On May 6 water was going over spillway and pressures were observed.