LIME AND CEMEI^T INDUSTEIES 717 



sand to cement is commonly 3 to 1 in case of Portland and two 

 to one in case of natural rock cement. 

 Johnson states: 



For special purposes 4 to 5 parts of sand may also be employed, 

 specially with finely ground cements, such as give a residue of 

 less than lOfo and a sieve with 14,400 meshes per square inch. 

 Since in the sand mixtures a standard sand must be employed, it 

 has become necessary to use clean sharp sand which has passed 

 a no. 20 sieve and stopped on no. 30. 



In order farther to insure the identity of the sand, the Ameri- 

 can society of civil engineers has recommended that crushed 

 quartz be used, such as is used in the manufacture of sandpaper. 

 Johnson does not favor this practice; for the material has fully 

 50/^ of voids, while the ordinary sands wuth roughly rounded 

 grains have but 33^ of voids. Any good sharp sand therefore of 

 the size, 20-30, should give very nearly uniform results, which 

 will average much higher than those obtained with crushed quartz, 

 unless the quartz briquets be thoroughly compacted by hard 

 hammering. 



The amount of water added will vary somewhat with the kind 

 of cement, but it should be very little, in fact just enough to 

 produce a mixture resembling damp sand. Jameson gives the 

 approximate amounts (p. 55) as 20^ to 25/"^ for neat cement, 15/^ 

 for one part sand, and 10/^ to 12/^ for three parts sand. It is 

 always w^ell to note the amount of water used. The temperature 

 of the water and also of the laboratory should be between 60 '^ 

 and 70° F. 



The mixing should always be done on a non-absorbent surface, 

 and the sand and cement should be mixed dry, and then the water 

 added. 



Compressive strength 



The test for compressive strength is seldom carried out, the 

 reason being that the results are apt to be uncertain even though 

 care be taken in the preparation of the specimens. They must 



