STATE GEOLOGIST. 161 
ing grains like pulverized glass. On adding water to 20 grams of the 
ground slag, the sandy portion settled first and then a heavy flocculent 
mass was precipitated. At the same time the volume of the powder con- 
stantly increased until it was about twice the original volume. The ap- 
pearance of this hardened powder did not differ essentially from that of 
the dry slag. But this action is much more energetic if alkalies, barium, 
or lime water is added. ‘The volume becomes three to five times the 
original volume, and assumes the colloid character. The mass hardens 
very rapidly, and the small grains of slag change to swollen, rounded 
particles. 
The conclusions of Zulkowski may be summarized as follows: 
1. The blast furnace slags, suitable for the manufacture of cement, 
are highly basic meta-silicates whose decomposition is prcyemed by rapid 
cooling or granulation. 
2. Owing to their anhydrous condition, they have a tendency to 
unite with water and to harden, which is assisted by the presence of 
alkalies. Such silicates may be called hydrolites. 
3. On slow cooling, an inter-molecular decomposition of the meta- 
silicate or hydrolite takes place in a greater or smaller measure. During 
this process the mass loses its uniformity, and the constituents are not 
dissolved by weak acetic acid in the proportion in which they are present 
in the slag. 
4. The hydration of the powdered basic meta-silicate is coincident 
with a change in volume, and a change in the shape of the grains, which 
now fill all the space, and gradually harden to a compact mass. 
5. slags cooled slowly, which contain only certain components or 
isomeric compounds of the above meta-silicates, do not possess the prop- 
erties of the hydrolite, mentioned under 2 and 4. | 
6. Portland cement, like slag cement, is a mixture of a hydrolite 
analogous to granulated furnace slag, and as much lime as remained un- 
combined on burning. The difference between the two is only that the 
necessary lime must be added to the slag in the shape of calcium hydrate. 
7. The calcium oxide of Portland cannot be separated directly. 
8. The calcium oxide may, however, be removed from Portland 
cement by a special treatment with dilute hydrochloric acid in greater or 
smaller amounts, and the hydraulic property is diminished accordingly. 
The residue remaining behaves like highly basic slag; that is, it hardens 
in the presence of the alkaline compounds without combining with them 
chemically. 
9g. But since the calcium oxide is thus removed in the wet way, the 
cement is sufficiently hydrated and the residual cement can never attain 
its original strength. The hydraulic agent of Zulkowski is represented 
by the structural formula, Fig. 14, which hydrates to Fig. 15. 
11—S. G. Bull. 3. 
