362 STATE BOARD OF AGRICULTURE. 



structure; in fact the process of manufacture should be such that the 

 tile wall will be impervious iu the usual service conditions. 



To show the effect of soil water upon the cement in concrete tile, 

 considerable laboratory work was done in preparing experimental porous 

 cups from pure quartz sand and cement, then causing different solutions 

 to pass through their structure. These experiments were so conducted 

 that a considerable quantity of the solutions used was passed through 

 the cups; and in all cases, including experiments in which pure water 

 was used, a portion of the cement was dissolved and the cups showed 

 disintegration in a degree sufficient to loosen a small amount of sand 

 on the inside surface of each. The water caused to pass through the 

 cup structure formed small channels through the concrete, that gradually 

 became larger as percolation continued. 



These experiments convince us that any solution percolating through 

 concrete tile will dissolve some of the cement; in other words, some of 

 the material Avhich is necessary to maintain the tile structure intact, 

 and points conclusively to the necessity for an impervious tile wall 

 stiiicture. Experiments also showed that solutions would percolate 

 through incompletely hardened tile much more rapidly than through tile 

 that had been properly and completely hardened; so the action of solu- 

 tions upon partially hardened concrete tile is greater than upon those 

 which have been properly hardened. 



In a tile which has been well made of properly proportioned materials, 

 and, after formation properly hardened, the wall structure will be so 

 nearl}- impervious that it seems probable the small pores which may 

 exist will be filled very quickly with fine material from the soil in which 

 such tile is laid, and thus afford added protection against the possi- 

 bility of soil waters dissolving the cement. 



Our first experiment with concrete cups was directed toward making 

 a cup that would be porous rather than impervious. The next experi- 

 ments along similar lines were directed toward producing an experi- 

 mental concrete cup of structure so dense that the concrete would resist 

 the passage of solutions, even against a small amount of water pressure 

 such as might exist, for instance, in a drain. Conclusions reached from 

 these preliminary experiments, seem to point to the following as the 

 most important factors for consideration in. manufacturing concrete tile 

 to fulfill desired requirements. 



(1) A Portland cement, meeting the requirements of the standard 

 specifications for Portland cement of the American Society for Testing 

 Materials, as revised to date. 



(2) Glean and preferably siliceous sand, graded in size from the 

 finer particles to those which will just pass a 14-inch mesh screen, for 

 all tile ten inches and less in diameter. 



(3) Proper and accurate proportioning of the cement and sand. The 

 method of measuring materials should be one which will insure separate 

 and uniform proportions of each of the materials at all times. 



(4) Thorough, mixing of materials, preferably by a power-operated 

 batch mixer, and continuing for at least one minute after all materials, 

 including water, are in the mixer. 



(5) Using a machine that accomplishes thorough packing of ma- 

 terials at proper consistency. 



