SMAIE TGHOLOGISI: Pie 
the main constituent, the hydro-silicate of lime crystallization cannot be 
proven notwithstanding the claims of Chatelier and others. The pro- 
cess of hardening depends upon an increase in volume on the part of 
some of its constituents, or the formation of gelatinous silicic acid. On 
letting a large quantity of water act on Portland cement, the volume of 
the latter is changed to a colloid mass occupying 33 times the original 
volume. On treating cement with lime water, the volume is increased 
21 times. Even quartz, when treated with milk of lime, is converted 
into the colloid variety in which it is greatly increased in volume. The 
same is true of alumina and iron. Michaelis cites the manufacture of 
sand brick, by the action of calcium hydrate on crystalline silica as a 
proof of his theory. His claims, however, have found no approval by 
most of the cement chemists and investigators. | 
Of late, since the study of colloids has assumed new importance, 
his claims deserve more attention than they have received so far. It 
cannot be denied that colloidal conditions have something to do with the 
hydration and hardening of cements. 
SUMMARY OF THE CONTENDING VIEWS. 
It might seem difficult to draw general conclusions from the array 
of theories presented to the reader in the foregoing paragraphs, but the 
conditions are not as complex as they seem if we put our main reliance 
on microscopic evidence and do not attempt to write precise structural 
formulas. Portland cements are basic slags produced by vitrification, 
belonging to the subsilicate group, with the tri-calcium silicate as the 
main hydraulic factor, assisted by certain calcium alumina compounds 
which are spinel-like in character or more basic. The main silicate crys- 
tallizes out from a more fusible magma, enough of which must be present 
to produce solution, just as in a porcelain body the fluxes, feldspar and 
lime have the power to bring into fusion comparatively large quantities 
of the refractory materials, like kaolin and flint. The percentage com- 
position of the cement requires certain, but by no means narrow, limits — 
of lime and silica, in order to produce a sufficient amount of hydrolite, 
yet not too much, as otherwise the cement would require too high a 
temperature for vitrification, since not enough of the magma could be 
produced at lower temperatures. In nature the only mineral we know 
as approaching the type of the sub-silicate of an alkaline earth is chon- 
drodite, Mg,Si,O,,, which, however, invariably carries iron and fluorine. 
The different stages of the Portland cement reaction may be illus- 
trated by considering the components of the raw mixture. It is made up 
of calcium carbonate, clay substance, quartz (more or less fine), felds- 
pathic and iron carrying minerals, like mica, augite, hornblende, etc. 
I. At about goo° C. the chemical water of the clay substance is 
expelled as well as the carbon dioxide of the calcium carbonate. 
2. At the same temperature the chemical water is replaced by lime 
forming with the clay substance Al,O,, 2SiO, 2CaO. 
*Ton. Ztg., 1899, p. 785. 
