x. W. E. COOK. 
The above results show clearly the great advantage of 
grading the sand carefully, so as to produce a solid aggre- 
gate. 
When the specification for the main outfall sewer was 
being prepared in Melbourne, in 1891, the difficulty of 
obtaining sand suitable for concrete forced itself on the 
attention of the Kngineer in Chief, and accordingly a 
number of experiments were made with bluestone toppings 
in lieu of sand. The results of both tensile and crushing 
tests showed that the substitution was not only feasible, 
but advantageous. 
TENSILE STRAINS, 
Material. Proportion. Age. Breaking Strain. 
Bluestone Toppings ... 2tol ... lidays ... dips 
Beaconsteld Sand v.65 Stood a ee «4 LOD 
Frankston Sand oa 2) 80 lero: ate ts oa Denes 
Bluestone Toppings... 3totl ... 15 ,, sin ey: ALOT coe 
Beaconsfield Sand ~~... 3-tod. “ss;'. 7 +3 ron! OSU 
The age was in favour of the briquettes made with 
toppings, but the gain between 8 or 9 days and 15 days 
would not bring the sand briquettes up to anything like 
the toppings. 
As the concrete was to be practically always in 
compression, the crushing tests were made much more 
numerous. All proportions from 1 to 1 of sand or toppings, 
to 1 to 15 of sand or toppings, were tried, and it was found 
that the strength of the sand blocks varied from 845 tons 
for 1 to 1, to 3 tons per square foot for 1 to 15; while in 
the case of toppings the variation was 168 to 12 tons, 7.e., 
toppings blocks were twice as strong as sand blocks in the 
1 to 1 mixture, and 4 times as strong in the 1 to 15 mixture. 
Various mixtures of cement, sand or toppings, and screen- 
ings were then tried, with the result that the toppings 
still showed a superiority, but not so marked as in the first 
