EXPANSION AND CONTRACTION OF CONCRETE. 23 
reading on the instrument. The total contraction over the uncracked 
portions of the pavement equaled 0.462 inch in a total length of 230 
feet, or 0.00017 inch per inch of length. The calculated contraction 
for the fall in temperature between the initial reading (82°) and this 
reading (50°) equals 0.000176 inch per inch of length, almost iden- 
tical with the actual measured contraction. 
On March 4, when the temperature of the pavement was 30° F., 
another set of readings was taken. In the interval the pavement 
had been subjected to the extreme cold of the winter. Here again is 
seen the effect of change in temperature upon the concrete. That 
part of it which remained free from cracks contracted 0.000307 inch 
per inch of length. The decrease in temperature during this period 
was 52°, and hence the theoretical contraction would be equal to 
0.00029 inch per inch of length, so that here again the temperature 
influence seems to be borne out by actual measurement. It will be 
seen that the cracks and expansion joints in this section opened con- 
siderably, and the colder the weather the greater the openings be- 
came. 
In June, 1915, not quite a year after the road was constructed, 
another set of readings was made and gave the results shown on the 
curve. The warm weather had decreased the contraction of the con- 
crete below the amount shown in the preceding March. This section 
was laid in warm weather when the temperature exceeded 82° and 
the temperature for the pavement on June 19 was only 68°, so that 
some contraction of the uncracked concrete should be expected. This 
amounted to 0.0167 inch, or 0.00006 inch per inch of length. The 
theoretical contraction for this fall in temperature is 0.000077 inch, 
just a little more than the actual measured contraction. A feature 
of this measurement is the great contraction at some of the expansion 
joints and a very large expansion at another joint and at two of the 
cracks. The large contraction or closing of some of the joints and 
the large opening at one may be accounted for by the relative move- 
ments of the slabs due to inequalities in the sub-base and to the fact 
that the section under test is at the bottom of a long 4 per cent grade. 
In February, 1916, the last set of measurements was taken in cold 
weather. Here again the contracting effect of low temperatures is 
seen on that part of the concrete which has remained intact. A con- 
traction of 0.00023 inch per inch of length occurred by actual meas- 
urement, whereas the figured temperature contraction was 0.000264 
inch per inch of length. Temperature again has played a prominent 
part in influencing the length of the concrete. Note that the 100-foot 
length cracked in two places, whereas the sections of smaller length 
did not crack. 
