4. Buffalo Natural Portland (analyzed by writer).

Where analyses of cement or cement rock do not give the amount of the alkalies, they must be regarded as incomplete. Gillmore says in his book on "Limes, Hydraulic Cements and Mortars" that all cement rocks contain alkalies, which are important in determining their cementforming qualities. It is also known that in the manufacture of Portland cement from a mixture of clay and limestone a certain per cent. of alkali if not present in these materials, must be added. The amount is small, but it is essential.

The causes of the setting of a cement have been much discussed and a number of theories devised. Probably no one cause explains all that happens, and it may be that different cements bring into action dif ferent chemical actions; but for practical purposes, the following points may be considered as established, and as they have direct bearing upon the recognition and treatment of cement rocks, it will be necessary to discuss them in some detail.

In the first place, the principal constituent of all these rocks is lime. There can be no cement made from any shale, clay, or slate, alone, which does not contain this element in excess of others; and as the lime always occurs as carbonate, it is evident that the greater part of all such substances will be dissolved by warm, dilute muriatic acid. There will be, in all cases, a residue left undissolved by the acid, which is very fine and not sandy, and when separated and washed, is found to be like very fine clay. It contains the silica and almost all of the alumina of the stone but only a trace of the lime, that having been taken out by the acid. This clay-like character of the residue is characteristic of the cement limestones. Those which have a sandy, gritty residue, will have

no hydraulic properties when burned, even though the amount of such residue be considerable. This constitutes an easy approximate test for cement stone. When pulverized, it must dissolve in warm muriatic acid, giving a muddy liquid, the more residue the better, if it is not sandy and gritty.

After such a rock is burned at a bright red heat for some hours, it changes entirely in character. When powdered, it dissolves entirely or nearly so in acid, giving an almost clear liquid; and in the solution will be found the silica and the alumina of the stone. Chemistry shows that in this case the silica and alumina were combined with lime, while in the original clay residue, they were combined with each other in the clay. Therefore, a cement contains silicate and aluminate of lime. These compounds absorb water, uniting with it to form hard, crystalline minerals. Evidently this, then, is part of the action of the "setting." The extent to which this formation of aluminate and silicate takes place in the burning probably depends upon the stone used. It certainly does not have to be complete. If a clay containing some free lime be strongly burned, so as to cause a partial combination between it and the alumina and silica, and then, after finely pulverizing, be mixed with ordinary slaked lime, this mixture will "set," and the resulting mass contains crystalline hydro-silicates, exactly as in the case of a cement. Again, if some cement rocks be heated to the point where complete combination occurs, the mass may melt and give a hard, glassy "slag," having no cement properties whatever.

What the small amount of alkali has to do with the setting, it is difficult to state. It may aid in promoting the combination during burning, and it may also make the silicate formed more easily attacked by water, and so facilitate the hydration, or finally, it may act to transfer silica to lime, forming silicate of potash, which is continually decomposed by the lime.

A cement rock, then, consists of a limestone, either magnesian or not, having intimately mixed with it fifteen to thirty-five per cent. of clay, and a little potash or soda. The hydraulic energy given to such a rock by burning will vary with the temperature and length of time it is heated. Where it is burned too hard, it may, as before stated, lose entirely its property of setting, forming even a slag-like mass like the furnace cinder in iron-making. This danger of over-burning is probably greater in the more magnesian stones, as it is well known that lime and magnesia form a more fusible combination with silica than either alone. This may be the reason that, while the artificial cements, having about the same composition as the natural ones, except that they contain only lime, are best and strongest when highly burned, the others are only able to stand a limited burning without becoming vitrified and ruined.

From the table of analyses given before, the composition of an average cement rock may be stated to be

and it may be expected that any one having a composition between these limits will probably furnish a cement when calcined for a certain time at a certain temperature, points which may be ascertained by experiment. Gillmore gives full directions in his work before referred to for ascertaining the proper degree of calcination. His method is essentially as follows:

The stone, after breaking into pieces of approximately the same size, which should be about as large as a walnut, is placed in a large fire-clay crucible provided with a hole in the bottom, like an ordinary flower-pot. This hole serves to permit a current of air to pass up through the crucible and thus assist the calcination of the stone. The crucible with its contents is then placed in a good coke-fire (best in a large grate) and heated to bright redness. After forty-five minutes, a few pieces are removed, and again after an additional forty-five minutes, some others, and so on at intervals of three-quarters of an hour, until some have been heated for at least six hours. These various fragments are separately ground to powder, and made into a stiff paste with water. The time and energy of setting together with the subsequent behavior in water, will indicate the quality of the stone as developed at each degree of burning.

Limestones, of course, slack and fall to powder. The slightly hydraulic limes will set and hold their shape when very lightly burned, but the longer heated portions crack and disintegrate. The true cements set and gradually harden under water, though those portions hardest burned may lose their value. Where the greatest strength is developed by a moderate burn, the cement will be found to set rapidly, as a rule.

There are some cements, which, as the temperature and time of burning increase, become slower-setting and stronger. Such a rock may furnish two kinds of cement, a quick-setting, light-burned, and a slowersetting, heavier burned, the latter resembling the artificial Portland cement. The "Buffalo" cement rock is of this character; and the burned stone from the kiln is partially pulverized, then separated into powder, constituting the light cement and hard lumps, which, being heavier burned, resist the first grinding. These, subsequently ground again, furnish the "Buffalo Portland" cement.

The preparation of the cement on a large scale is simply the application of proper machinery for doing in a cheap and efficient manner the same thing that was described before in the laboratory testing of the rock, the two steps being a careful burning of the rock and its subsequent pulverization. The goodness of the resulting material will, however, largely depend upon the skill with which these operations are performed. The proper temperature being ascertained for the burning of the particular rock, it must be carefully maintained by regulating the amount of fuel, and in the product drawn from the kiln, all under-burned or overburned parts being separated by the workmen before it goes to the mills.

The pulverization is one of the most important points. It has been found by careful experiment that the coarser particles of the ground cement are of no more value than so much admixed sand, only the finest powder having the power of setting; and as the cement is always used in mortar with sand, it is evident that the more of the material which is in the dusty state, the more sand the cement will carry to make a mortar of a given strength. Hence, in testing a cement prepared for use, it should be sifted through a very fine sieve, and the amount of coarse-ground material noted. The smaller this is, the better.

The kilns used in burning are oval or round in section, and from twenty-five to forty-five feet in height. They are contracted toward the bottom, and are there provided with arched openings through which the burned stone is drawn. The shell or support of the kiln is masonry or sheet-iron, as in the case of blast-furnaces. The natural cement rocks are usually burned "continuously," that is, the material is continually added at the top as it settles in the furnace, and is drawn at small intervals from the bottom. This plan, of course, takes less fuel than where a kiln full of material is allowed to cool, and then a fresh one heated. This latter plan, however, seems to be preferred in burning the artificial cements.

The fuel is, in many cases, mixed with the stone and the whole burned together, coal slack or small coke being used. In other kilns, there are furnaces at the sides in which wood, coal or coke can be burned, and the hot flames then pass into the body of the stack, heating the stone thus kept from direct contact with the fuel. This plan has one great advantage. As the temperature and extent of burning will be regulated by the rapidity with which the material is drawn from the bottom, the faster the stone is taken out, the more rapidly it passes through the kiln, and the less it will be burned.

Where the fuel is mixed with the stone, the heat will be regulated by the proportion added. This must be determined by experiment; and should a variation in the quality of the rock necessitate a change in the temperature, it can only be learned when the kiln has burned through, and when it is too late to correct the damage, while in the former class of furnaces, the burner simply changes the rate of drawing, and so raises or lowers the heat, as the appearance of the clinker indicates.

At the Utica mills, described in the "Transactions of the American Institute of Mining Engineers," vol. 13, where four kilns with side furnaces are used, the largest having a capacity of 1,000 barrels, the drawing is done every two hours, the material being in the kiln four days.

To ascertain the quality of the rock, a small "try kiln" a couple of feet square and six or eight feet high is used, and it should be kept running pretty constantly where the regulating of the burning depends ubon the varying of the fuel mixed with the rock, or an undetected change in the character of the stone may cause the loss of a kiln full of cement. A simple chemical expedient for controlling the character of the material would be the determination of the carbonic acid given off by a certain weight when dissolved in acid. This could be done on the stone as it came from the quarry, and any variation in the quality would appear at once. This method of control, when applied to the manufacture of artificial cement, has given perfect satisfaction in securing regularity of product, the mixture being tested three times daily.

The arrangements for handling the stone must be of the simplest and most efficient description. At one of the principal works at Louisville, the rock is quarried from a bed twenty feet thick which lies under five feet of soil and shale (stripping), is thrown into cars which are drawn by a wire rope up an inclined plane, and dumped into the top of the kiln with the requisite fuel. The kilns are in rows of four. The "clinker," or burned rock, is drawn from the bottom, as usual. One man will do the drawing for a large kiln. Where kilns of the side-furnace type are used, a man can fire one or two of them. As this work is necessarily continuous, double-turns, night and day, must be worked.

The kilns are lined with a single course of fire-brick, supported as previously stated, by masonry or an iron shell of No. 12 to 16 tank iron, with an intervening course of red brick and a space of two inches, more or less, filled with sand, loam, or ashes, to allow expansion. The kilns should face the prevailing winds to accelerate the draft.

The burned stone must be, as already noted, carefully picked over to remove under-burned or over-burned portions which are worse than merely useless, being in some cases dangerous to the quality of the cement, causing it to swell and crack, owing to their containing uncombined lime. The clinker is next crushed to the size of small gravel, the