290 EXPERIMENT STATION RECORD. [Vol. 38 



"The design of a slab of any width can be accomplished by using the for- 

 mulas for narrow rectangular beams and substituting for the breadth (b) the 

 value obtained from the above table." 



The flow of concrete under sustained loads, E. B. Smith {Proe. Amer. Con- 

 crete Inst., IS (/9i7), pp. 99-102, fiys. 2).— Experiments conducted by the Office 

 of Public Iloads and Rural Engineering of the U. S. Department of Agriculture 

 on the flow of concrete in cylinders and beams under load are reported. 



It was found that " the law of the flow of concrete is asymptotic. The flow 

 continues at a gradually decreasing rate and yet is an appreciable amount 

 during three to four weeks. It then continues more slowly for an indeflnite 

 period, but this additional change is very small. 



" The natural total shrinkage of dry concrete is about 0.05 per cent in three 

 months. The total net flow under load, exclusive of natural shrinkage, may be 

 as great as 0.15 per cent, depending upon the time, material, and load. The 

 total combined effects of shrinkage and flow in compression may amount to as 

 much as 0.2 per cent. This would give, in a 20-ft. column not reinforced, if 

 loaded to about 800 lbs. per square inch, nearly 0.5 in. of deformation ; and may 

 produce, in a reinforced beam of 20-ft. span fully loaded, a sag of nearly 0.3 in. 

 Even the much smaller deformations, which are inevitable, may produce, if not 

 anticipated and provided for, serious results in the setting of apparatus and 

 machinery, and in the alignment of shafting, and inay easily cause other parts 

 and members of the structure to be overloaded. 



" The effect of flow wuthin the material itself is either to relieve the stress 

 condition, if the construction and loading make this possible, or to gradually 

 change the length or position of the member. 



" The maximum amount of flow or the flow for any particular period is almost 

 directly proportional to the magnitude of the stress up to 1,000 lbs. per square 

 inch. It is, therefore, only necessary to decide upon the allowable flow defor- 

 mations in designs before determining the allowable dead- and live-load stresses. 



" The measurement of stress conditions in concrete structures can not be made 

 directly by deformation readings, unless all the flow constants as to time, 

 material, and loading are known. . . . Deformation readings taken only a 

 short time apart will indicate apparently different stress values. 



" The magnitude of the flow deformations vary quite largely with the kind of 

 aggi'egate and the mixture. ... It is shown in these experiments that gravel 

 concrete has about 20 per cent njore deformation than limestone concrete. 



" The modulus of elasticity of concrete is different for each mixture and for 

 each different aggregate. It changes and decreases in value with time and as 

 the flow deformations increase. If the modulus of elasticity could be ascer- 

 tained for any particular concrete with due respect to the time factor, stress 

 values could then be determined by simple deformation readings. 



" In the case of a reinforced beam, the effect of flow in the concrete is to 

 lower the position of the neutral axis, thus enlarging the compressive cross- 

 sectional area and relieving the unit stress value. More stress is also thrown 

 into the steel." 



Friction tests of concrete on various sub-bases, A. T. Goldbeck {Proc. Atner. 

 Concrete Inst.. IS (1917), pp. 239-245, figs. 2; Good Roads, 51 {1911), No. 15, 

 pp. 229-231, figs. 2).— Further experiments (E. S. R.. 37, p. 88) are reported 

 in which the ground was muddy following a thaw. The results are given in 

 the following table; 



