SAMPLING AND TESTING HIGHWAY MATERIALS. 91 
2. The heat treatment should he such as will result in making 
the balls aa hard as possible consistent with the ability to r< 
the pressure without cracking or crushing. 
Note. — While a long series of careful experiments would be needed 
to justify the specification of more exact conditions in these regards, 
and while som- users of these tests think that the chemical eomposi 
tion (within limits) plays a very small part, if any. in the problem, 
it is believed that the above requirements will be found satisfactory 
for commercial work until such time as suitable experiments shall 
have furnished the data necessary for making the conditions more 
precise. 
DiA METER AIS'D FORM OF BAIJiS. 
3. The standard diameter of balls should be 10 millimeters, with Balls. 
a permissible variation of 0.0025 millimeter (0.0001 inch), plus or 
minus; no ball, either new or old, showing a greater variation is 
to be employed. The standard diameter should always be em- 
ployed except in very rare cases when some other is absolutely 
necessary. If at any time in testing a hardness of No. 60<» be 
exceeded, the balls should be remicrometered. 
Xote. — Thus far there is not sufficient evidence to show that thp 
hardness numbers will be the same when different diameters of balls 
are used, and some of the evidence indicates that the reverse is the 
case; hence the importance of adhering to one size of ball. 
Pressure. 
4. The standard pressures used should be 3,000 kilograms for Pressure. 
steel and 500 kilograms for softer metals. Departure from these 
pressures should never be tolerated except in rare cases where 
it is unavoidable. The time of pressure should be at least 30 
seconds. 
Note. — The fact that, with our present light on the subject, 
we can only regard the results as comparative renders it important 
to employ as few different pressures as possible. 
Measurement of Diameter or Depth of Indentation. 
5. Whether the diameter or the depth of the indentation is Measurement. 
measured, apparatus should be used that will give results as 
accurately as a microscope mounted on. and moved by. a micro- 
meter screw. 
Note. — As to the choice between the two, there exists a very 
decided difference of opinion, some thinking one and some the 
other more conducive to accuracy. The source of error in either 
case (assuming the measuring apparatus to be accurate) is the de- 
pression or the elevation of the metal immediately surrounding 
the indentation. 
TESTS OF WIRE ROPE (GUARDRAIL). 
1. The report of the physical test of wire rope shall include the 
following information: Diameter of the rope; diameter of the 
wire; number of strands and wires; length of lay of strands and 
wires; cross-sectional area in square inches, breaking load in 
pounds, and a note as to whether the failure was in the body of 
the rope or at the socket. 
DEFINITIONS. 
2. (a) The "lay of the rope" is the length expressed in indies 
for each complete turn of a strand around the axis of the rope 
measured along the axis. 
(&) The "lay of the strand" is the length expressed in inches 
for each complete turn of a wire around the axis of the strand 
measured along the axis. 
(c) The diameter of a wire rope is the diameter of a circle 
inclosing it. 
3. The cross-sectional area of the rope shall be determined as 
follows: Measure the diameter of the component wires of a 
