XVIII. W. E. COOK. 
particles in the one case were of a certain diameter, and 
in the other case half that diameter, the interstices in the 
sand contained in a fixed volume would be the same in both 
cases, but the surface in the latter case would be six times 
as great as in the former. 
Turning back to the figures showing volume of slime 
washed out of samples of crushed sandstone, it will be 
seen that the resulting slime washed out was 17% of the 
original volume, though the washed sand was 91% of that 
volume, taking the mean of the two sets of figures. 
The material washed away from crushed sandstone we 
commonly call pipeclay, a material that is detrimental to 
sand, but at the same time it evidently serves to fill up 
the interstices, as, adding 125 cubic inches to 663 cubic 
inches results in a mixture only 729 cubic inches in volume 
in place of 788 cubic inches. The slight gain in strength 
obtained by washing crushed sandstone would probably be 
increased considerably if, in place of 125 cubic inches of 
slime washed out, we added an equal bulk of fine sharp 
sand, so as to have a graded mixture of good material, 
equal in bulk to the original sample. 
It is quite clear that the clay and loam mentioned inthe 
American experiments are better materials than the slime 
washed out of crushed sandstone, as washing the latter 
material clearly improves it, though only to the extent of 
4 to 7 per cent. For most purposes the trouble and 
expense would more thancounter balance this gain, but 
for works where the strength of the concrete is important, 
washing should be specified in the case of crushed sand- 
stone. 
In August, 1901, the Engineer-in-Chief for the Water 
and Sewerage Board had experiments made to determine 
the voids in bluestone and sandstone of certain gauges, 
and in certain mixtures of them. In the hope that the 
