slaked to a thick paste and left in a vessel; another portion was slaked to a thin paste, and, as it thickened, I added a little water so as to keep it in the consistence of sirup; the third portion was slaked to a dry powder with one-fourth of its bulk of water, and put, like the others, in an open vessel. I, immediately, made the mortar in the first column, which may serve as a term of comparison. The others were made, at the periods expressed in the table, of lime slaked in the several modes mentioned. In proportioning the parts, I added, when I made the mortar, a little water to the lime which had been slaked to a thick paste and also to that which had been slaked to powder, so as to bring all to the condition of that slaked to thin paste. These experiments were begun in the month of November, 1823, that is to say seven months after the experiments of table No. XIV. In table No. XV, I have followed the same mode of expressing the time required to harden, as in table No. XIV. The mortars of three series were made of the same quantities of lime, sand and trass; and I took care, as in the preceding table, to use the same trass, and the same sand, for all. In the first series of the table, the mortar made immediately gave a result rather weaker than those made afterwards: the hardening was more prompt in summer than in winter.

I left the lime of the second series for six months in a state of clear paste before using it. We see that the results were not so good as those obtained from the thick paste. There are, certainly, several anomalies in the results, but, as before remarked, we must look at the whole.

The third series was likewise commenced at the end of six months. In 'table No. XIV, the mortars of the first series, made of lime slaked to powder, were formed at appropriate periods, which were not extended beyond seven months. The resistances obtained in table No. XIV are not the same as those obtained in Table No. XV, because the proportions were different: but if we compare the experiments of the first series of table No. XIV with those of the first and third series of table No. XV, we shall see that instead of making hydraulic mortar of fat lime, sand and trass, immediately after slaking the lime, it will be more advantageous, after slaking the lime, with a little water, as it comes from the kiln, to leave it exposed to the air for some time. We see, here again, that the hardening was more slow in winter than in summer. If we compare the results obtained in the preceding tables, we shall see that there is a great difference in the effects of mortars made of natural or artificial hydraulic lime and sand, and of these made of fat lime, sand and trass. When the first are made of lime which has for some time slaked to powder, or which has been airslaked, they generally lose much of their force. There is not the same disadvantage with fat lime: whether the mortars are made as soon as the lime comes from the kiln, or after it has been slaked with a little water and left exposed for some time in the air, or after the lime has been air-slaked, good results are always obtained: but we have seen that the best are got by slaking the lime with a little water as soon as it is burned and leaving it exposed for some time to the air in a covered place. Experiments, to be given by and by, will show that I obtained good results, also, by making hydraulic mortars of sand, trass, and fat lime which had been lying wet in basins for four or five years.

We have also seen by comparison of the preceding table, that mortars made of hydraulic lime, natural or artificial, without trass or puzzalona, did not harden, with sand, until from eight to fifteen days, although giving resistances; while those made of fat lime, sand and trass, hardened at the

same season of the year, in the space of from four to six days, and, on the average, gave much greater resistances.

These results induced me to make researches in order to the fabrication of factitious trass or puzzalona. In the following chapter I shall give the results I obtained-commencing with several essays that have previously been made by others, with this view.

CHAPTER VI.

Of Artificial Trass and Puzzalona.

Mr. Baggé, a Swedish Engineer, was, I think, the first who attempted to make artificial puzzalona. This Engineer used, in his experiments, a species of black, and quite hard, schistus; he heated it highly several times; afterwards reduced it to powder, and having mixed it with lime, announced having obtained an excellent mortar, having all the properties of mortar made of puzzalona.

I do not at all doubt the success of Mr. Baggé: but his experiments having been repeated elsewhere, a less satisfactory result was secured. This was owing to using schists of a different composition, and which, in lieu of being heated highly, like those of the Swedish Engineer, required to be heated moderately.

Mr. Faujas de Saint-Fond made several researches in 1778 with the puzzalonas of Vivarais, which he found to be equal to the puzzalonas of Italy. He showed, also, that the trass of Andernach was a true puzzalona.

In 1786 Mr. Chaptal repeated the experiments of Mr. Faujas de SaintFond, on the puzzalonas of Vivarais, and found that they were inferior to those of Italy. This contradiction between the results obtained by Mr. Faujas de Saint-Fond and Mr. Chaptal is easily explained, for one used hydraulic lime in his experiments, and the other used fat lime.

Mr. Chaptal published in 1787 a memoir on the use of the ochreous earths of the south of France. These earths were calcined in a kiln like those used in some countries for burning lime. The kilns are reversed cones from about eight feet eight inches to about ten feet ten inches in height, and are from six to eight feet in diameter at the base: an opening is left near the apex of the cone, through which to withdraw the products of calcination: these kilns are filled by placing alternately, a layer of seacoal or turf, and a layer of the ochreous earth; the fire is kindled, after a few layers have been placed, and, when it is in full action, other alternate layers are added until the kiln is full. When the lower portion, which was the first heated, is sufficiently calcined, it is withdrawn as fast as necessary, and other new layers of clay and fuel are successively added above. Thus the burning is continual; the clays heat gently near the top of the kiln, and are subject to a much greater heat in the middle: they gradually cool towards the bottom, by the action of the current of air, after the fuel is consumed. This manner of burning earths, possesses great advantages as I shall have occasion to show. It would be equally advantageous to calcine hydraulic limes in these kilns, because, as we have seen above, it is very important to use these limes soon after they are burned. By constructing several kilns of this kind, there might be as great a supply of fresh lime as could be needed.

1

Mr. Chaptal attributed to iron, a great effect in improving puzzalonas;

and he appears to have attributed a very feeble one to alumine. Mr. Vicat says, on this subject, in his memoir: "If, as Mr. Chaptal assures us, clays deprived of iron, and calcined, cannot be employed as puzzalonas, it must be the oxide of iron that acts principally on the silex and modifies it, by the aid of fire in the ochreous earths, as the lime does in the hydraulic limestones: the alumine, therefore, appears to take the least part in these reactions; it does, nevertheless, make part of good puzzalonas."

Experiments which follow will show that iron plays no part in the preparation of puzzalonas, while alumine is very active. We shall see, also, that there is another substance which is very active and which has occasioned the divergencies of opinion on the preparation of puzzalona; but I will not anticipate; and will go on with the statement of the various attempts made to produce factitious puzzalona.

From experiments made at Cherbourg in 1787 by Mr. de Cessart, it appeared that basalts, obtained in the Department of the Haute-Loire, and pulverized after having been calcined, produced a mortar possessing all the qualities of those made with Italian puzzalona. Analysis shows this basalt to contain in one hundred parts, the following; Alumine, 16.75; Silex, 44.50; Oxide of iron, 20.00; Lime, 9.50; oxide of Manganese, 2.37; Soda, 2.60; Water, 2.00; loss, 2.28.

The works in the port of Cherbourg required a great quantity of puzzalona; but the war which broke out with England on the rupture of the treaty of Amiens, raised the substance to an exorbitant price: it was calculated that the price of a cubic metre (35.34 cubic feet) cost, in 1803, more than 400 francs (about $76.00.)

They might easily have procured on the spot, clays which would have given results absolutely the same as those afforded by Italian puzzalona, at the cost of about 30 francs (about $6.00.) And had they understood the manner of making puzzalona, there would have been a great saving. Considerations of this sort induced Mr. Gratien, Senior, Engineer of roads and bridges, to occupy himself with the subject. He made a few essays with the porcelain clay of Valognes, and made many with the schists of Haineville, which he calcined, in repetition of the experiments of the Swedish engineer. The Valognes clay calcined, gave but feeble results; the Haineville schists gave better; analysis showed this schist to consist, in 100 parts, of the following: alumine, 26.00; silex, 46.00; magnesia, 8.00; lime, 4.00; oxide of iron, 14.00; loss and water, 2.00. The results of the experiments of Mr. Gratien, Senior, are given in two memoirs which he published, one in 1805, the other 1807. A commission of the Institute was charged with examining experiments, comparative of mortars made of puzzalona, of trass, and of the Cherbourg schist; the judgment of this commission is inserted by Mr. Gratien in his memoir of 1807, page 9; it declares "that after taking from the water the twelve boxes of beton of different compositions: "1st. All had acquired a certain consistence, but very different amongst themselves.

"2nd. The difference was striking between the betons composed of puzzalona and trass, and those into which these substances did not enter. "3rd. The two compositions of burnt and pulverized schistus offered a resistance quite satisfactory, but not so great as it probably would be after a longer immersion.

The above report shows that Mr. Gratien obtained with the schists, results inferior to those given by puzzalona.

In 1806 Mr. Le Masson, Engineer of roads and bridges, at Rouen, at

tempted, in concert with Mr. Vitalis, to make factitious puzzalona by calcining yellow ochreous earths, according to the process of Mr. Chaptal. The betons made with the calcined earth, having acquired a remarkable consistence, Mr. Le Masson, in 1807, repeated the experiments on a larger scale: he immersed, in the Seine, casks filled with beton made of the calcined ochreous earth. After six months, the casks were taken out, and it was found that the betons had acquired such hardness that it was necessary to strike it two hundred times with a mass of iron weighing 26,4 lbs. to break in a depth of ten to twelve inches; the tenacity was so great, that the entire mass weighing 2800 lbs. was suspended by means of a tire fond (lewis?) Mr. Gratien and Mr. Vitalis judged that the masonry had acquired a hardness greater than could ever be absolutely necessary, even in constructions that require the greatest solidity in their foundations. These experiments are reported in the memoir of Mr. Gratien 1807-page 46 and following.

Mr. Vicat in his memoir of 1818 confined himself to reporting the different attempts that had been made, up to that time, to form artificial puzzalona: but in 1819 he sent to the Institute a memoir on that subject. I do not know whether this memoir was printed, but an extract is contained in the annales de chimie et de physique of 1820, Vol. XV, page 365 and following. After several observations on limestone, on the action of fire on calcareous stones, and the combination of water with lime, the author gives a succinct history of puzzalona. We find afterward the following passage.

"Since the quality of natural hydraulic lime, depends solely on the presence of a certain quantity of clay, combined by fire with the calcareous matter, it was natural to think that by mixing clay in suitable proportions with fat lime, slaked, no matter how, and then submitting the mixture to calcination, a similar result would be obtained: experiments made on a large scale, and in many places, have confirmed this idea in a manner so complete, that it is now possible to fabricate any where, and at a very moderate price, artificial lime superior to the natural analagous lime."

"In like manner, since chemical analysis gave, for the constituents of natural puzzalona, silex, alumine, oxide of iron, and a little lime; it was easy to suppose that our clays, of which the composition is altogether similar, might be transformed, by burning, into artificial puzzalonas. This idea was already old at the period of the experiments of the author of these researches; but by a remarkable fatality it was as if stricken with sterility: the circumstances on which the quantities of good puzzalona depend, had not been determined with sufficient precision. There was a persuasion, for example, that iron was very active therein; that, therefore, only ochreous clays should be used; that in order to a more perfect imitation of nature, there should be a high degree of heat, 'because,' said they, 'the fire of volcanos is much more intense than the fire of our kilns.' (Several geologists are of a contrary opinion.) Certain puzzalonas, came, beyond doubt, in the form of lava, from volcanos: they were in fact subject to a high degree of heat; but since the very distant epoch of their formation, they have sustained diverse decompositions, either from intestine modifications or from the action of acid vapours, or from other causes, and these decompositions have totally changed the mode of combination of their principles. As to the red puzzalonas of the neighbourhood of Rome, every thing shows that they were only vast beds of ochreous clay variously burned, either by subterranean fires, or by currents of lava which covered them and broke them up in every direction. Thus their quality is very variable according to the disposition

and depth of the beds. But whatever may be the process of the formation of these substances, it is demonstrated that all the mystery of their properties, resides, not in the presence of iron or lime, but in a particular state of combination of silex and alumine-a state to which all clays, soft and greasy to the touch, may be brought, with the greatest facility, by a light calcination. The means which hitherto appear to have succeeded best, consists in reducing the dry clay to a very fine powder, and calcining it, for some minutes, on metallic plates heated to an obscure (brun) red. The truth is, that practice has not yet fully matured the process, and it is probable that full success requires the contrivance of a mode of calcination more expeditious and convenient than the above: but the problem is not the less resolved."

Such is the mode proposed by Mr. Vicat for making factitious puzzalona; and he teaches us nothing new, for he does not tell us what is the particular condition in which the silex and alumine should be found to afford good results. Opinions were for a long time divided on the question whether the clays, which it was desired to convert into good cements, should be heated much or little-these cements being, really, artificial puzzalonas.* Mr. Vicat has adopted the opinion of those who thought they should be burned but little; but experiments which are to follow will show that this (as a general principle) is erroneous--it being necessary to burn more or less, according to the composition of the clay. The problem, therefore, was not resolved by Mr. Vicat, though he says it was; and the experiments which follow will show in what it really consists.

I made several essays, substituting brick and tile dust for trass. To this end I composed a number of mortars of fat lime and the dust of bricks or tiles taken from all the kilns of the neighbourhood. A part of the mortars were made of brick dust, and a part of tile dust. I obtained many results: -sometimes very good, sometimes indifferent, and sometimes very bad. What struck me much, at first, was, that mortars made of different dusts coming from the same burning gave very different results: notwithstanding that the dusts were of the same burning, were all made of the same clay, were used with the same lime, and that all other circumstances were the same. I saw from this, that great risk was run of making bad mortar, by taking brick or tile dusts without discrimination. I know that the great majority of constructors preferred highly burned dusts, and that, although made of the same clay, they much preferred dust of tiles to that of bricks. To settle my opinion on these two points, and to explain up contradictory results that I had obtained, I made the experiments reported in the following table.

By the term ciment-translated, in the above sentence, cement, the French often mean, simply brick dust or tile dust; and it will be so rendered wherever it is supposed to bear that meaning.-Trans.