904 
Journal of Agricultural Research 
Vol. V, No. 19 
specimen rests. The end of the specimen is ground away in inverse ratio to 
its hardness, so that the hardness may be computed by determining the 
loss in weight after any given number of revolutions of the disk. The 
coefficient of hardness discussed later is obtained by subtracting one- 
third of the loss in weight in grams from 20, after 1,000 revolutions of 
the disk. 
The degree of toughness is determined by the Page impact method. 
A cylinder 1 inch in diameter and 1 inch high, cut from the rock speci¬ 
men, is subjected to the impact caused by the free fall of a 2-kgm. weight 
droppecf from successively increasing heights until the energy of the 
blow is sufficient to fracture the test specimen. The test consists of a 
1-cm. fall for the first blow, followed by falls increased by 1 cm. after 
each blow until failure occurs. The height from which the weight 
drops when failure takes place is used as a measure of the toughness of 
the material. 
Since the establishment of the Road-Material Laboratory by the 
United States Government, upwards of 3,000 samples, representing 
every known variety of road-building rock, and obtained from every 
State in the Union, as well as from foreign countries, have been sub¬ 
jected to the tests outlined above. The results of these tests are plotted 
in graphic form in figure 1. The coefficients of hardness are plotted as 
abscissae and the factors of toughness as ordinates. Each small circle 
represents the corresponding hardness and toughness of an individual 
rock sample. The large circles represent the average of all the coeffi¬ 
cients of hardness for each value of toughness. Hardness values range 
from o to 20 and toughness values from 1 to 47. 
A study of this curve brings out the following points: 
(1) That the average toughness for all tests made is about 9. 
(2) That the average hardness increases with toughness, and that the 
rate of increase becomes less as the toughness values become larger. 
(3) That individual values of hardness vary through wide limits for 
low values of toughness, and that the variations from the average decrease 
uniformly with the increase in toughness up to a certain point, about 20, 
after which they remain constant with very little variation from the 
average. 
(4) That, when any given value for toughness falls within certain 
limits, which define the suitability of the material for macadam-road 
construction under different traffic conditions, the corresponding value 
for hardness will f 5 .ll within similar limits for hardness. 
The first three facts are clearly indicated, but in order to substantiate 
the last deduction it will be necessary to define the limiting values of 
hardness and toughness which experience has shown should be applied 
when judging the fitness of stone for use in macadam construction under 
different traffic conditions. Such limiting values for toughness are 
shown on the curve in the ordinates at 4.5, 9.5, and 18.5, and the 
