STRENGTH OF CONCRETE. XXV. 



is noted, and afterwards increased until fracture occurs. The 

 fractured cube has the form of a pyramid, the base of which is 

 that of the cube, and nearly equal to it in height, the four corners 

 of the cube are sheared away. 



The following experiments were made by the writer on specimens 

 6 inch by 6 inch by 6 inch, the concrete was mixed and filled into 

 the moulds by hand ; the broken stone and the gravel used were 

 separated into three sizes by means of screens, between 2 inch and 

 1 inch, between 1 inch and \ inch, and between £ inch and ^inch, 

 these were then mixed together in the proportion of 5 : 2 : 1, and 

 the volume of the voids carefully measured. It was found that 

 the proportion of the voids in the broken stone was 39*5% and in 

 the gravel 31-6%, so that sand was added to fill the voids in each 

 case, and the cement was added in the proportion of 1 part of 

 cement to 6, 8 and 10 of stone. The proportions of cement and 

 sand entering into the composition of the mortars in the various 

 concretes was therefore nearly as follows : — 



Broken stone concrete 1 to 2 -4, 1 to 3 2, and 1 to 4 

 Gravel concrete 1 to 1*9, 1 to 2*5, and 1 to 3. 



The gravel and blue metal (Basalt) was found to be practically 

 non absorbent, but the sandstone absorbed 2*9% of water, so that 

 more water was necessary to make sandstone concrete than in the 

 case of gravel or blue metal. The results obtained by testing 

 these cubes at 7 days, 28 days, and 84 days are recorded in the 

 Table Series I. and plotted in figs. 1 to 3 inclusive. 



A series of compressive tests were also made on cubes of con- 

 crete 12 inches by 12 inches by 12 inches prepared by ordinary 

 workmen from concrete mixed in the usual way on the Sewerage 

 Works, Arncliffe, near Sydney, and filled into timber moulds under 

 instructions received from the Ensineer-in-Chief for Sewerage 

 Construction, Mr. J. Davis, M. Inst. c.E. The specimens were pre- 

 served in moist soil until they had reached the age required, then 

 sent in carts to the Engineering Laboratory where they were 

 tested. These specimens represent concrete made in the ordinary 

 way, and were more or less rough, so that the uniformity so 



