114 BEISTOL BUILDING STONES. 



}iand, with Pennant, the moan of fonr experiments with 

 cubes (17-20) gives a resistance of 21-24 tons ; while the 

 mean of three experiments (21-23) with prisms gives 20-46 

 tons, a slight advantage on the side of the cube. It will be 

 seen that the differences between the means in each case are 

 far less than those between the extremes for cube and prism 



respectively. 



With Dundry stone one-inch cubes give about one-tlurd 

 of the resistance of two-inch cubes. 



With Millstone Grit the resistance of the one-inch, cubes 

 is out of all proportion to that of the two-inch cubes. With 

 so hard and brittle a rock it is possible that tho larger cubes 

 had invisible cracks. The one-inch cube also indents the 

 iron deeply, and is thus more firmly gripped in proportion 

 than the two-inch cube. 



Experiments 3 and 6 were made with the object of testing 



. whether stones saturated with water offered more or less 



resistance to crushing than tlio dry stone. Tho results do 



not show any departure from tho average of those obtained 



with dry prisms. 



Experiments 36-41, and 31, 32, were made on rooks which 

 had been repeatedly frozen and thawed. They do not seem 

 to indicate any wcaiconing. 



Experiments 37-41, with Old Eed Sandstone bring out 

 clearly the fact tliat stones are weaker when tho incidence 

 of the stress is on the edge of tlie bedded layers ( + ), than 

 when it is on the bed ( = ). 



Old Red Sandstone and Pennant gave very fairly definite 

 pyramids after fracture, with an angle of about 62° at the 

 apex. Millstone Grit, Black Rock, and Dundry stone gave 

 very irregular pyramids, or irregular splintery forms. In 

 some cases these rocks, and Draycott stone in three cases 

 out of four, went into a heap of fragments. 



