20 BULLETIN 532, U. S. DEPARTMENT OF AGRICULTURE. 
and it is an observable fact that as soon as the wet-earth covering is 
removed from the surface of the concrete it begins to dry out and, 
necessarily, to shrink. The maximum average shrinkage shown by 
the concrete in any of the sections of different mixtures was approxi- 
mately 0.0001 inch per inch of length, and this occurred about three 
months after the pavement was laid. The temperature in this time, 
however, had dropped 20°, and this fall in temperature accounts 
almost exactly for the shrinkage. It has been shown by means 
of laboratory specimens stored out of doors and subjected to all the 
changes, of the atmosphere that very little change in length takes place 
under such conditions. It seems probable that the moisture content 
of the concrete at Chevy Chase changed so little that the length of tjie 
concrete was very little affected thereby. A hard rain, thoroughly 
soaking the concrete, will have no immediate great effect upon its 
]ength. It has been pointed out that moisture changes are rather slow 
and progressive, and therefore it is unlikely that hard rains of even 
several days' duration will have great effect on the expansion of the 
concrete in the road. 
EXPANSION AND CONTRACTION OF OHIO POST ROAD.i 
These measurements were taken over three 300-foot experimental 
sections of the concrete Ohio Post Road constructed in 1911 under 
the supervision of the Office of Public Roads. The Ohio Post Road is 
located partly in Muskingum County and partly in Licking County, 
Ohio, and runs west from Zanesville, Ohio, for 24 miles over the old 
National Pike to the Moscow Bridge over the South Fork of the 
Licking River. This road was constructed on a sub-base composed 
mainly of a stiff red clay. The total width of concrete surfacing is 
16 feet, with a thickness of 6 inches at the edges and 8 inches at the 
center. Expansion joints made of one thickness of 2-ply tar paper 
were spaced 30 feet apart throughout the length of the road, with the 
exception of the three experimental sections, on each of which the 
spacing varied from 20 feet to 100 feet, intervening sections being 
40, 60, and 80 feet long. Expansion joints were placed at an angle 
of 15° across the road. Both gravel and crushed stone were used 
as coarse aggregate, the proportions in the former case being 1 : 1-J : 3 
and in the latter 1 : If : 3. Crushed stone used in the experimental 
sections was obtained from limestone quarried on the Scioto River at 
Marble Cliff. The sand and gravel were obtained from a washing- 
plant at Dresden, Ohio. 
As soon after laying as possible the concrete surface was covered 
with canvas, which was removed when the concrete was 24 hours old 
and replaced by a 2-inch earth covering. This was allowed to remain 
1 The authors desire to acknowledge the assistance of the Ohio State Highway Depart- 
ment In conducting the tests on this road. 
