It is not the purpose to enter here upon a history of the evolution of the art of masonry construction, however interesting such a history might be; nor to describe ancient masonry structures however creditable and instructive; but to discuss present practice, the changes of recent times and those changes which are likely to be made in the near future.
It is true that in recent years dams have been built under very primitive conditions. Isolated locations remote from transportation facilities, with consequent prohibitive expense of cement and modern plant, have not prevented the building of some important dams. Labor if abundant and cheap can be used in place of machinery, and adequate if not high-class cement can be made by improvised methods from materials quite generally occurring in nature. Admitting the justification and necessity for such methods, and recognizing not only the satisfactory result but the high character and skill of their engineers, still the development of the cement and transportation industries renders it less liable that future masonry dams will be constructed on any radical departure from standard practice. The time up to and including the construction of the New Croton and Wachusett dams, may be called the age of mortar, and the time since then the age of concrete. Each will be described.
Sand should be free from organic matter with sound particles, graded in size and with not more than 15 per cent, of material fine enough to be classed as clay. When the cost is about equal, the matter of quality would probably more often favor natural pit sand than sand crushed from rock. To obtain pit sand free from organic matter it is usually necessary only to strip the surface of the pit to a sufficient depth. The soundness of the particles need not be argued, for a material whose particles can be readily divided, crushed or rubbed down is not sand, and is out of the question. The size of the particles should be graded from coarse to fine for the same reason that gravel or crushed stone should be graded for concrete, i.e., in order to get the most dense mixture possible with the same quantity of cement.
In 1900 and 1901, when the desirability and results of proper grading of sand and gravel had not been as completely demonstrated and were not as commonly appreciated as at present, quite extensive experiments in that direction were carried out in connection with the preparations for the Wachusett dam. These experiments were designed to show the strength and permeability of mortar composed of various percentages of coarse, medium and fine sand, the divisions between those arbitrary classifications being the 30- and 100-mesh sieves. Coarse sand was that which stayed on the 30-mesh sieve; medium sand that which passed the 30 and stayed on the 100 and fine sand that which passed the 100 sieve.
The numerous experiments showed most conclusively that mortar made with a graded sand was stronger and more impermeable. A large portion of the pit that furnished sand for the Wachusett dam contained fine material in excess of the proportion which the experiments indicated as being ideal; however, by daily testing of samples and by scraping from different parts of the pit, the standard of 50 per cent, coarse was followed fairly closely. It seemed to be practically immaterial how the remaining 50 per cent, was divided between the medium and the fine, that is, within the possible range of that pit.
As to the effect of material in the sand fine enough to pass the 200-mesh sieve, it seems possible that the effect would be the same as if this material were mixed with the cement as is done in the manufacture of sand cement. The U. S. Reclamation Service is experimenting extensively in this direction. (See Engineering News for June 19, 1913).
The method of measuring the sand should be specified because it makes some practical difference whether the sand is shoveled into a measuring box or drawn into it through a chute from a bin. With sand falling into the box from a height of 2 ft. or 3 ft., the sand spread, leveled and struck off with a straight edge, the box is apparently filled by a less quantity of sand than would be required if it were to be thrown into the box a shovelful at a time. At the Wachusett dam this was quite carefully determined to be 3.61 per cent., i.e., a box measuring 103.61 per cent, when filled from a chute will hold 100 per cent, of sand shoveled in.