18S 
ME. KOBEET MALLET ON VOLCANIC ENEEGV. 
ancl the differences are thus obtainable. These vary much for different classes of 
rock. 
If, however, we take a mean of the pressures at which first disintegration occurred, 
and of the differences between those and the pressures of complete crushing, we shall 
find that — -- g , or about J added to the pressure of first disintegration, gives that of final 
crushing to powder — a result of some interest to the engineer and architect, and one 
which enables us in some degree to compare our results with the largest and most accu- 
rate series of experiments that the writer believes had previously been made on the 
resistance to crushing of rocks. 
124. He refers to those of Mr. George Wilkinson, Architect, given in his work 4 On 
the Practical Geology and Ancient Architecture of Ireland’ (Murray, 1845), which were 
conducted by that gentleman upon almost the entire series of the rocks of Ireland from 
the highest to the lowest. The reliability of these experiments is known to the writer, 
as he himself devised and constructed the lever-apparatus by which they were performed ; 
and the experiments were made in the engineering works of the writer’s late firm by 
Mr. Wilkinson. As Mr. Wilkinson’s object was to obtain constructive data only, the 
crushing-pressures recorded are seldom those at which the specimen (which in every case 
was 1 cubic inch in size) was reduced to powder, but that at which it was disintegrated 
only, and hence the results are all, as regards our point of view, here too low. If, how- 
ever, we add to the mean of each class one-fourth part, we shall approximate to results 
more nearly comparable with those we have here obtained. 
The following Table gives Wilkinson’s highest, lowest, and average results, together 
with the number of different quarries and total number of experiments from which each 
average was ascertained. This summary has been made by the writer, and is not to be 
found in Mr. Wilkinson’s work. 
Class of rock. 
Crushing- weights in lbs. per square inch. 
Number 
of 
quarries. 
Number 
of 
experiments. 
Maximum. 
Minimum. 
Average. 
Limestones 
27510 
1344 
15053 
125 
210 
Sandstones 
26670 
1239 
8183 
31 
82 
Sandstones across beds 
18790 
1680 
8864 
33 
42 
Sandstones in line of beds 
20650 
2940 
9824 
29 
36 
Slates across beds 
27370 
5040 
13930 
9 
13 
Slates in line of beds 
21770 
6160 
11285 
11 
18 
Granites 
13440 
2310 
6657 
8 
20 
Basalts, metamorphic 
48020 
7140 
19025 
12 
25 
The granites of Ireland generally are very friable. 
125. It may be interesting to add the following volcanic rocks on the authority of 
Prudhomme (‘ Cours pratique de Construction ’) : — 
Ancient lava ( Volvic stone ) of Auvergne . 28,446 lbs. per square inch. 
Vesuvian lava (date not given) .... 8,392 „ „ 
Granitoid porphyry (Bazoche) .... 21,072 „ „ 
