BULLETIN 537, U. S. DEPARTMENT OF AGRICULTURE. 
20 
and properly faced, are bedded in plaster of Paris. The cylinder 
then is crushed in a 200,000-pound universal testing machine. A 
small 2-inch spherical bearing block is placed between the moving 
head of the machine and the upper surface of the specimen. The 
average of at least two determinations is reported as the crushing 
strength, calculated in pounds per square inch. Crushing strength 
tests are made upon samples of road-building rock only when especi- 
ally requested. Of a total of 282 compression tests made up to 
January 1, 1917, Table II shows that 97 were made on granites, 
13 on gneisses, 78 on limestones, 42 on dolomites, 28 on sandstones, 
and 24 on various other types of material. 
The percentage variation in the strength of the 110 granites and 
gneisses and the 120 limestones and dolomites is shown graphically 
in figure 1. In this 
chart the per cent of 
total samples tested 
having various values 
for crushing strength 
are plotted as indi- 
cated. For instance, 
the chart shows that 
10 per cent of all 
granites and gneisses 
tested show, a crush- 
ing strength of 20,000 
pounds per square 
inch. Likewise, by 
summing up all the 
per cents to the left 
of the 20,000-pound 
line we find that 48 per cent of all samples of granite and gneiss tested 
have a crushing strength of less than 20,000 pounds per square inch, 
which shows that the average crushing strength of this type of 
material lies between 19,000 and 20,000 pounds per square inch. 
Likewise, the average crushing strength of the limestones and dolo- 
mites lies between 18,000 and 19,000 pounds per square inch. 
INTERPRETATION OF RESULTS OF PHYSICAL TESTS. 
A discussion of the interpretation of the results of physical tests 
was given in Bulletin 370, to which reference has been made. Since 
the publication of that bulletin, however, a table of general limiting 
test values for broken stone for various types of road construction 
has been adopted by the office and printed on the back of its form 
for reporting tests. For general reference these limiting values, 
together with comments upon limits shown, are given in Table IV. 
( 
:r.a^ivtt 
\\ ^ 
(1! 
<F.TF 
,S 
(HO r 
estsj 
I 
1 
III 
1 
ill 
, 
I i 
UMf 
^STflNJF 
■ * r 
^01 1 
1MI1 
T, 
(120 
TesT= 
) 
J 11 
, II 
ill 
1 
il. 
1 1 ■ 
CRUSHING STRENGTH - LBS PER SQ. IN. 
Fig. 1.— Variations in the crushing strength of rock. 
