150 REIBLING AND REYES. 
These figures show that unless the slag is cooled very suddenly from a 
high fusion temperature it will contain very little strong cementive material 
and also that there is too little lime present to produce great strength 
under any conditions. Although this analysis is more or less theoretical 
it corroborates general practical experience that the more basic the slag, 
the better suited it is for cement making. 
Portland cement.—Generally speaking, rotary-kiln Portland 
cement, and more so fused Portland cement, surpasses similar 
set-kiln products in strength and uniformity. Schott® showed 
that the crystalline structure of Portland cement clinker changed 
with the burning temperature independent of the percentage 
of lime which was the same in every instance, that Portland 
cement obtains its full development only at very high tem- 
peratures, and that at low temperatures, certain intermediary . 
compounds are formed, namely, belite, celite, and felite, which 
do not harden to the same degree and which with increasing 
heat unite and form double compounds and ultimately repre- 
sent a uniform mass (alite). 
Owing to the unequal degree of burning to which the raw 
materials in different parts of a set kiln are subjected, and to 
fuel contamination, the aggregate produced by this process 
usually consists of a mixture of all five classes of hydraulic 
cements in which the well-sintered Portland cement clinker, 
typical of good rotary practise, predominates. For example, 
the weathered set-kiln products, the chemical and physical prop- 
erties of which were described in Tables XVIII and XIX of 
Part II ** of this paper, contained at least 15 per cent of under- 
burned cement as well as some vitrified clinker. In Table XL 
we show the class and the source of its hydraulic constituents. 
We attribute the characteristic lower eariy strength of set-kiln 
products almost entirely to the presence of underburned mate- 
rial. This is at variance with the usual belief that the rapidity 
of the rotary process—more particularly the cooling down of 
the clinker—is the apparent cause of the greater activity of 
rotary cement. However, it already has been shown that there 
is no radical difference between the hardening properties of 
the pure well-burned product of these two kilns. Foreign matter 
from the fuel in the cement made by the older process often 
* Cement & Eng. News (1910), 22, 377. « 
* This Journal, Sec. A (1910), 5, 410. 
