PHYSICAL AND CHEMICAL PROPERTIES. PARTS Iv, V. 189 
TODAYS. 28DAYS. 2MOS. - . f 7 . 2YRS. 3YRS. 
Fic. 23.—Typical curve of the tensile strength of Portland cement. 
eliminate this characteristic depression almost at will. There- 
fore, it was necessary to search in other directions for the cause 
of this phenomenon. 
THE INFLUENCE OF ULTIMATE CHEMICAL COMPOSITION AND EARLY 
GAIN ON STRENGTH. 
W. A. Aiken * produced results and deductions from a large 
number of tests, the figures of which for five years— 
corroborated his original assumption that an early stage, low-pulling 
cement which undoubtedly will show best results in the long run, can be 
made in the rotary kilns notwithstanding the general tendency of such 
commercial products to show abnormally high early results with a con- 
sequent retrograde movement later on. 
He also conducted a long series of careful analyses and phys- 
ical tests on the Lehigh Valley cements from which he fixed 
a theoretical analysis of material giving the highest ultimate 
results. However, he did not prove why this should be so, and 
his results have not been verified in tests on other than the few 
fresh commercial products with which he worked. 
It was thought that an investigation on the variety of com- 
mercial Portland cements at our disposal would supply the data 
which were lacking. It soon became evident that Aiken’s con- 
clusions could have no general application. Sooner or later the 
* Cement Age (1908), 1, 75. 
