be used with sand only, in making mortars for exposure to the air. When such limes are not to be found, instead of making artificial hydraulic limes, it will be preferable, as in the case of water mortars, to make the hydraulic mortar directly, by mixing fat lime with sand and hydraulic cement. The proportion of cement to be mixed in the mortar will depend on its quality, and the nature of the works to which it is to be applied.

Clay dust containing about one-fifth of lime appears to be less proper for mortars that are to be exposed to the air, than for those to be placed in water: but a few hundredths of lime, far from injuring, have an advanta geous effect in saving fuel, and facilitating the pulverization of burnt clays. Clay dust, and other analogous matters, should always be made very fine.

Arenes mixed with fat lime afford hydraulic mortars, which are very good in the air. When the arenes are feeble, they should be mixed with a little hydraulic cement.

When clay dust, ashes, scoriæ of forges, and arenes, have only weak hy. draulic properties, they give to air-mortars but feeble tenacity. Before using these matters in the air, therefore, they should be tried by mixing them with fat lime, and plunging them in water, according to the process given in the first section. It is the more important to make such trials, because these substances are very dear, and some varieties of them produce no better effect than so much sand.

Fat lime and hydraulic lime do not seem able to bear as much sand as is commonly thought. Fine sand affords much better results than coarse, both in air-mortars and in water-mortars. Earthy sands must be avoided. In places where earthy sands, only, can be procured, they must be washed: but before resorting to this operation, we should be satisfied that these sands are not arenes. Should they prove to be arenes, they must be used in the state in which they are found.

The process given in the first section for the manipulation of water-mortars applies equally to air-mortars. No ill consequences need be appre hended from wetting the mortars to the degree requisite to their being worked with ease; nor, when they have become too stiff, from exposure to the air, from adding a little water, on reworking them. An excess of trituration, beyond what is required to mix the ingredients thoroughly, is altogether useless.

It does not appear that the Romans had any particular process for making their mortars. No masonry has survived to our day but such as was made of hydraulic lime, or of fat lime and hydraulic cements, or arenes (I speak here of masonry made of small materials.) An inspection of their mortars shows that they were often made with little care, proving that their good quality is to be attributed solely to the quality of the lime, or of the substances mixed with it.

If, in general, no better results are obtained with fat lime, than those obtained by me, the practice of making mortars of fat lime and sand only, should be abandoned. A small quantity of hydraulic cement, or of some substance of similar nature, should always be mixed in the mortar; that is to say, all air-mortars should be hydraulic mortars. The expense will be a little greater it is true, but there will be full compensation in the duration of the masonry. There is no economy in putting up cheap masonry which will require to be rebuilt at the end of a few years; and will need costly repairs, annually: it is much better, and really more economical, to

encounter, at once, the expense which will secure to the work an indefinite duration, and exemption from all but trivial repairs.

Hydraulic mortars, whether made of hydraulic lime and sand, or of fat lime and hydraulic cement, or other similar substance, resist the inclemences of the seasons well, which makes them proper to form factitious stones. The art of making factitious stones is nothing more than the art of making good hydraulic mortars. This kind of stone may be easily made to possess, at the end of a year, a tenacity about equal to that of ordinary bricks, and this tenacity will go on augmenting, for several years. The solidity of hy draulic mortars is favoured, and of course the solidity of factitious stones also, by keeping them moist during the first year. They should, therefore, be buried under ground, or placed in water, if their dimensions will allow; and when too large to be thus disposed of, they should be formed on the spot they are to occupy, and be enveloped in some material which may be kept wet. It would be proper to make their dimensions a little in excess, in order to bring them, subsequently, to their true dimensions, by cutting away the surface that had been in contact with the air. In order to avoid disagreeable colours, cements should be taken, which are but little coloured by the oxide of iron.

In countries where building materials are of bad quality, and where energetic arenes, or good hydraulic limes, are to be had, concrete may be advantageously employed, in the construction of revetments, underground rooms, aqueducts, and various buildings: this mode may also be employed, even where no arenes or hydraulic limes are to be found, provided the materials for making cements, can be procured at a moderate price.

It will be important, wherever works in masonry are to be carried on, to make experiments in order to ascertain, 1st, the quality of the several kinds of lime to be found in the neighbourhood; 2nd, whether it be best to use fat lime fresh from the kiln, or to slake it into vats, and allow it to lie for some time wet, as is commonly done; 3rd, the proportion of sand that gives the best mortar; and 4th, the quality of various arenes and cements to be found in the vicinity.

To make these various experiments, both for water-mortars and airmortars, I think the essays should be submitted to trial, after about one year, with a machine like that used by me, and of which a drawing is added. It would be advantageous to make, in all places, the trials with prisms of mortar of the same dimensions, in order to compare the various results, and to ascertain the relative qualities of the materials, in different districts or countries. The great number of experiments I had to make, was the reason why I used prisms too small. I think that quadrangular prisms 14 inches long by 4 inches, (0.35m x 0.10m x 0.10m) would be of suitable size. To this end the mortar should be moulded in boxes 14 inches long by 4.80 in. wide by 4.80 high: after a year, four-tenths of an inch should be cut from each face, and the prism be placed in two stirrups of iron, like those in the plate-the stirrups being twelve inches clear distance apart, and be broken in the manner pointed out in the first section. Similar prisms should be made of the best brick earth in the vicinity, which, after being burned, should be cut down to the same dimensions as the mortar prisms; these brick prisms being broken, the average weight sustained by them will serve to appreciate the resistance of the mortar prisms. Stone prisms should also be subjected to the same trial: these, when of good stone, will show a strength much beyond that of the best bricks: I doubt if mortars can be

brought to afford a resistance equal to hard stone: but the strength of bricks may be easily attained, and that will give good masonry.

I must urge upon Engineers to study, in their several localities, the meterials most proper to make good mortars. The fabrication of mortars bas been, for a long period, abandoned to a routine which has produced perishable masonry, requiring frequent repairs; and thus consuming funds which might have been applied to the construction of new work or the amelioracion of the old. Engineers should not consider it beneath them to be occupied in this kind of research: and they should leave behind them at each place, a relation of the experiments they have made, and the results they have obtained. These operations require minute attention, certainly, but this will be recompenced by works of long duration.

SECT. III.-EXTRACTS FROM

"RESEARCHES AS TO LIME BURNING."

BY M. PETOT,
Engineer of Roads and Bridges.*

CHAPTER XVII.

On the Preparation of Factitious Puzzolana, and particularly of that afforded by Gneiss Sand.

§1. On different kinds of Puzzolana.

The important part that puzzolanas play in the improvement of hydraulic mortars, sufficiently explains the interest that belongs to an examination of this subject. As yet we know of no other means of forming mortars susceptible of hardening in water: and even since hydraulic limes have become the subject of particular study, that of puzzolanas has not been the less attended to, because hydraulic limes are not to be found in every locality, nor is it always possible, or economical, to manufacture them; and because in many circumstances they give mediocre results only, unless mixed with a certain dose of puzzolana.

Puzzolanas may be arranged in two principal classes, namely, natural puzzolanas, and artificial puzzolanas. Among the first, the most energetic are, generally, volcanic matters of a composition analogous to clays. These were discovered first in Italy, where their use goes back to time immemorial. Afterwards they were found in countries possessing extinct volcanoes, as Auvergne, Vivarais, Guadaloupe, &c. The matters which furnished them have sustained, by igneous action, a change in the primitive mode of combination in their elements; but as the intensity of this action was not every where the same, there resulted products of various degrees of cohesion: and it is not difficult to conceive that great differences may exist in their qualities, although none may exist in their chemical composition.

*Paris, from the Royal press, 1833.

It may, nevertheless, be possible that a puzzolana is no more than of medium quality, merely because, as is said in common parlance, it is still too young. In cases where there was at first too much cohesion, certain influences may, in the course of time, bring about a state of disaggregation. This phenomenon is not without example in nature: the greater part of the feldspathic rocks are in these circumstances, and produce a second variety of puzzolanas. Such, particularly, are the graywackes of Carhaix, the arenes of Perigord, and our gneiss sands. The geological position of these substances, at least of the gneiss which we have before us at Brest, does not admit the supposition of any igneous action. This disaggregation goes on little by little, and we take, as it were, nature in the act, for between the upper parts of the quarries which are in a state of sand, and those which are still in the condition of hard rocks, it is not rare to find a series of layers of every intermediate grade of cohesion.

These recent puzzolanas are but slightly energetic: but by torrifying them, they are rendered, if not equal to the first, at least applicable to almost all the same uses: they thereby pass into the class of artificial puzzolanas.

In the same class we must also arrange, after they have been highly calcined, compact schists, and even basalts. Calcination was, in fact, the means applied to these substances by M. Gratien Lepère, and the Swedish engineer Baggé. Lastly, all clays, considered either as proceeding from the decomposition of rocks, or as forming particular earths, lead to the same result by the application of heat.

Few localities are without clays; whence the general use, from very remote periods, of the dust of bricks or tiles, as cement. We may now substitute, with economy, the sands of graywacke, of gneiss, and of arenes; but these latter substances are less common. By their use the subject does not change its aspect; there is merely an enlargement of its boundaries.

In all puzzolanas, natural or factitious, the elements of clay, namely, silex and alumine, determine the hydraulic qualities: moreover, silex alone, in jelly, or lightly calcined, but in a state of extreme division, gives, by its mixture with fat lime, a good hydraulic mortar; while it is not the same with alumine; we may, therefore, say, definitively, that silex is the base of all puzzolanas; and that, reciprocally, all substances containing silex in a state of feeble cohesion are apt to become puzzolanas.

§ 2. On several methods employed in the preparation of Factitious

Puzzolanas.

The study of the quality and use of natural puzzolanas not having immediate connexion with the question before us, we shall restrict our remarks to artificial puzzolanas. And we shall first refer to the several methods applied to their preparation.

The material to which recourse is had, may be submitted to heat, first, in the state of powder; and second, in the state of fragments, either brick shaped, or of irregular forms. In the first case, all refractory, ochreous, or calcareous, clays may afford good puzzolanas by a torrefaction of a few minutes. It will be sufficient to spread the powder in a thin layer on an iron plate, and keep it incandescent from five to twenty-five minutes. M. Vicat, to whom we are indebted for the knowledge of this method, adds, however, that it has not been applied on a large scale, and that it offers some