2 BULLETIN 532, U. S. DEPARTMENT OF AGRICULTURE. 
An extended series of laboratory and field tests was begun in 1910 
by the Office of Public Eoacls to make a close study of expansion and 
contraction movements of concrete pavements. These included de- 
tailed attention to the spacing, design, and movement of expansion 
joints. It is the purpose of this bulletin to present the results of these 
experiments in the hope that they will be of some value to those in- 
terested in work of this nature. No attempt will be made herein to 
apply the results obtained to the practical side of road construction, 
although certain broad conclusions will be drawn from the results 
which will be available for immediate application by the engineer. 
LABORATORY MEASUREMENTS OF EXPANSION AND CON- 
TRACTION. 
It has been established by other investigators 1 that concrete ex- 
pands on being heated and contracts on cooling by an amount differ- 
ing very little from that of steel. A coefficient of 0.0000055 per de- 
gree Fahrenheit seems to express accurately the effect of temperature. 
Therefore, with change in temperature of 100° F., a 100-foot length 
of concrete road, if unrestrained by friction at the base, would ex- 
pand or contract 0.0000055X100° X100 feet= +0.055 feet or 0.66 
inch. This phenomenon alone, without considering any other influ- 
ences, probably accounts in many instances for the cracking of con- 
crete pavements. 
In view of the probable reliability of the already established tem- 
perature coefficient of expansion, all efforts of the laboratory were 
directed toward obtaining the change in length of concrete due to 
other causes. It has been shown, originally by Bauschinger and sub- 
sequently by other investigators, that both neat cement and mortar 
contract to a considerable degree upon hardening in air, while, on the 
other hand, they show considerable expansion when placed in water. 
Some of the values of contraction of neat cements and mortars, re- 
ported by White in the 1911 Proceedings of The American Society 
for Testing Materials, are as follows : 
Shrinkage of neat cement kept in air. 
Per cent. 
7 days (average of 6 specimens) 0.109 
28 days (average of 6 specimens) . 190 
6 months (average of 6 specimens) .236 
1 year (average of 5 specimens) . 270 
2 years (average of 5 specimens) . 2S9 
4 years (average of 5 specimens) . 322 
1 Kellar, " Volumanderungen von Cementen verscbiedener Miscbung bei wecbseldner 
Temperatur and wechseldner Feuchtigheitsgrad," Tonind. Zeit. 19, 469 and 4S7, 1S94. 
Pence, W. D., "The Coefficient of Expansion of Concrete," Eng. News, 46,3S0 (1901). 
Norton, Chas. I., " Some Thermal Properties of Concrete," J. Am. Soc. Mech. Eng., 
June, 1913, p. 1012. 
Rudeloff and Sieglerschmidt, " Untersuchungen liber die Langenanderungen von Beton- 
prismen beim Erharten und in folge Temperaturwechsel," Deutsch. Ausehuss fur Eisen- 
beton, Heft 23, 1913. 
