STATE GEOLOGIST. L99 
not find the exact analogies of Portland cement. From what we know of 
Portland cement at present we can say that it is, mineralogically speak- 
ing, a basic rock and its closest approach in the natural group of rocks 1s 
the peridotite group whose principal minerals are olivine, pyroxenes, 
hornblende and garnet. 
REVIEW OF THE INVESTIGATIONS ON THE CONSTITUTION OF 
PORTLAND CEMENT. 
The question as to what Portland cement really is has received a great 
deal of attention from many investigators, but it seems so far that the 
closest approach to an accurate conception of the structure of Portland 
cement has been made by mineralogical examinations while chemical 
investigations have failed more or less completely. It is interesting to 
examine the various methods by which investigators have approached the 
subject and, hence, in the following paragraphs a brief review of some of 
the work done by various men will be given. Much has also been 
written by men who base their conclusions on theoretical reasoning 
entirely. Such contributions are not considered. 
The work done in this direction may be classified under several 
headings indicating the chief method by which it has been endeavored to 
determine the structure of Portland cement. 
1. Mineralogical (microscopic) examination. 
2. study of synthetic mixtures. 
32. study of hydration reactions. 
4. Determination of calcium oxide. 
5. study of heat reactions. 
Of course, no investigator has confined himself to any single method, 
but most of the experimenters have paid special attention to one line of 
attack. 
MINERALOGICAL INVESTIGATIONS. 
The Work of Chatelier—The first investigator to approach the 
subject along these lines was Chatelier,* who attacked the problem in 1887. 
In this year he published his classical work on the constitution of 
Portland cement based on microscopic analysis, in which he states that 
he found the main constitutents of Portland cement to consist of (1) col- 
orless double-refracting, cubical crystals; (2) between these a darker 
substance, double-refracting, but without crystalline structure. In addi- 
tion he found several accessory constitutents. 
(a) Slightly yellowish crystals, opaque and showing’ striation. 
(b) Very small crystals with rather strong double-refraction. 
(c) Finally, zones of matter without influence on polarized light. 
Constitutent No. 1 was called, later, alite; No. 2, celite; a, belite; 
and b, felite. Chatelier considers the alite as 3CaOSiO,, this being the 
*Recherches Experimentales Sur la Constitution Des Mortiers Hydrauliques, Annales Des 
Mines, 1887. 
