490 REPORT ON NATIONAL MUSEUM, 1886 



chanical separation of the mineral constituents, which may in their 

 turn be in part or wholly subjected to analysis. 



Strength and ratio of absorption. — The test of compressive strength 

 is at the present time the principal test to which a stone is put to ascer- 

 certain its adaptibility to any particular kind of structural application. 

 The value of the results are, it seems to the author, greatly overesti- 

 mated. It is a rule among builders never to place a stone where it will 

 be subject to more than one-tenth the pressure it has show^n itself capa-- 

 ble of bearing by actual experiment. Even under these circumstances 

 there is scarcely a stone in the market that would not be found when 

 freshly quarried strong enough for all ordinary purposes of construc- 

 tion. The problem is not what will a selected and carefully prepared 

 sample of the stone bear to-day, but what will it bear after many sea- 

 sons' exposure to heat and frost ? For all ordinary purposes of con- 

 struction the excess of strength of any stone over 15,000 pounds per 

 square inch is of little value excepting so far as it denotes density, and 

 hence greater resistance to atmospheric influences. 



The size of the cubes tested and the methods used in their prepara- 

 tion are matters that need consideration in making comparisons of results 

 in any series of experiments. General Gillmore found* that within 

 certain limits u the compressive resistance of cubes per square inch of 

 surface under pressure increases in the ratio of the cube roots of the 

 sides of the respective cubes, expressed in inches." Thus a series of 

 cubes varying in size from one-fourth inch to 4 inches square were found 

 to give results varying from 4,992 pounds to 11,720 pounds per square 

 inch of surface. It naturally follows that ambitious dealers desiring 

 any stone to show great power of resistance would select the larger sized 

 cubes to be experimented upon. That the method of preparing a cube 

 to be experimented upon is of moment will become apparent when we 

 consider that in the process of dressing a small sample by hammer and 

 chisel it becomes filled to a greater or less extent with small fractures 

 and hence will break under less strain than though carefully sawn out 

 and ground down to a smooth and even surface, t 



* Report on compressive strength, etc., of building stone, Ann. Rep. Chief of Engi- 

 neers, 1875. 



tThe author ventures to submit the following scheme for testing stone in addition 

 to the chemical methods already alluded to. It aims to accomplish in the course of 

 a few weeks results such as would be brought about by natural weathering in per- 

 haps as many years. 



Let six samples of the stone, all from the same bed and so far as can be determined 

 all exactly alike, be selected and dressed by sawiug and grinding (never by hammer 

 and chisel) to a uniform size, say 2-inch cubes. From not less than three of these 

 let the ratio of absorption be ascertained by weighing, immersing in water for not 

 less than twenty-four hours, and reweighing. Take two of the cubes and ascertain 

 their crushing strength when dry ; two more and in like manner ascertain their crush- 

 ing strength when saturated with water, say after twenty-four hours' immersion 

 though a longer time would be preferable. Take the two remaining cubes and, 



