keep the cloth, while undergoing cutting, from being lifted up by the action of the knife, and is therefore stationary. R represents a pile of cloth being cut into one side of a vest pattern.

The knife is of peculiar construction, and so is that portion of the table around which it plays. The lower end of the knife is a thin, narrow metal rod, p (fig. 3); it plays through a hole in the table, and down through the centre of the pinion, h, on the under side of table, B. The pinion is hollow, and has an opening for p to pass through it. J is a small circular plate or disk inserted into another rebate plate, I, in the table. [The left side of this plate is not distinctly shown in the figure]. The plate, J, is cast in one piece with pinion, h, and moves round with it. K is a small piece inserted at one side of plate, J, and secured with a screw, and answers for a cap.

It will easily be understood that when the cloth, R, say twenty plies of it, is laid upon the table, B, and the top piece chalked out for a certain piece of garment to be cut (all the pieces of the same size, like one side of a vest), that the reciprocating action of the knife will cut straight forward by pressing the cloth toward the knife; this action of the knife will be easily understood, but when an angle is to be formed, a sharp corner to be turned, how is this to be done? Simply by turning round the disk plate, J, in the table, and the knife, c1, simultaneously, to cut on the new line. By turning the hand-wheel, E1, the wheels, l, n, m, turn the spindle, K, which turns pinion, h, and the plate, J, and at the same time the spindle, E, moves the pinions c b, which turns the cutting knife, c1, and the holddown, S, all together. It is also necessary that the knife should turn in the smallest possible space, like a point. This is done by the knife being held up stationary while it is being turned above the cloth, and only the narrow thin guide rod, p, suffered to turn the cloth. The knife therefore is arrested and held suspended at the highest point of the stroke by a peculiar spring-brake in the part above crank plate, G; a pin being inserted into the periphery of said plate, which is caught between projections of the brake. [These are not shown in the figure, but are important devices for the perfect action of the machine.] U is the clutch lever for throwing the clutch collar and plate, G T, in and out of gear with the driving shaft, H. It is placed conveniently for the operator, so as to throw the cutting rod out of gear and motion instantaneously.

In the slot in the plate, J, through which the cutting knife plays, there is a metal edge at the one side (bearing against the knife), and a small flat spring on the other side; these prevent the cloth from being pushed or dragged down by the knife while cutting. By having the number of plies of cloth, R, set on the table as we have described, and as shown, twenty or more similar pieces for different garments, vests, or pants, &c., can be cut out at one operation. With this machine the inventor, Mr. Harraday, who is a practical tailor, can cut out 500 pairs of pantaloons in one day, and with more practice he has no doubt of being able to cut 1000 pairs in the same time. The advantage of this machine lies in cutting so many pieces of the same pattern at one time; it can cut on any line, straight or curved, and is altogether a most useful and ingenious machine.-Scientific American, July 22.

ART. V.-Resumé of Researches on the Resistance offered by Hydraulic Limes and Cements to the Destructive Action of Sea Water. By MM. MALAGUTI and DUROCHER. (Comptes Rendus de l'Academic des Sciences, 24th July, 1854.)

FOR several years past the attention of scientific men and of engineers has been occupied with the question of the destructive action exercised by sea water upon hydraulic mortars. M. Vicat, in seeking to explain this disastrous phenomenon, has shown that the sea water acts by its tendency to dissolve the lime of the mortars, which is then replaced by magnesia; but hitherto no efficacious means have been pointed out to prevent or neutralize this dissolving influence. We only know, in general, that the strongest hydraulic mortars, the cements or mixtures of lime and puzzolanas, which have set the most rapidly, are those which appear to best resist these causes of decomposition. Nevertheless, even amongst mortars and cements which set with equal rapidity, and are of nearly equal strength, it will be found that some possess very different powers of resistance, without its being possible a priori to distinguish, either by analysis or a quick trial, those on the stability of which complete confidence may be reposed.

In this state of uncertainty, we thought that, by the study of the cements which resist the decomposing action of sea water, conjointly with analyses of the hydraulic limes and cements incapable of withstanding its action, as well as of the products of the resulting decomposition, it might be possible to throw some light upon this question, whose difficulty is equal to its importance.

The samples, 16 in number, upon which we experimented, were the hydraulic limes of Paviers and of Doué, and the mortars made from them, Boulogne, Portland, Pouilly, Vassy, and Parker's cements. We are indebted for them to the kindness of those skilful engineers of the Ponts et Chaussées, MM. Jebuvier, Watier, and Bellanger, to whom we desire to express our thanks.

The mode which we have followed in our researches consisted in examining the modifications which were produced in the proportions of the different elements, in comparing the compositions of the substances plunged in sea water, with that of similar substances which were not immersed. But as we had no samples of mortars of lime and sand which had set under fresh water, to compare with those immersed in sea water, the examination of the latter could only be made by comparing their composition with that of the lime employed in their preparation. In these comparisons we had to deduct the sand, and to reduce the compositions found for the mortar, to those which they would have had if no sand had been used. We shall not detail here all the results which the discussion of our analyses revealed to us, and which are recorded in our complete memoir; we shall merely direct attention to the most prominent ones, which prove how complicated these phenomena of decomposition are.

Two cylinders of the bydraulic lime of Paviers were immersed, under similar conditions, in sea water during eighteen months. One of them lost

an enormous quantity of lime and gained very little magnesia; but to compensate for this, it fixed a quantity of carbonic acid almost sufficient to saturate the two earthy bases. An appreciable quantity of silicic acid was carried off with a little alumina. It appeared that a hydrated silicate of alumina was separated from the mortar at the same time with the lime, whilst carbonic acid was substituted for the constituents which disappeared. The alteration of the other cylinder was not so considerable; the loss in lime, and the gain in carbonic acid, were not so great; but, on the other hand, the quantity of magnesia substituted for the lime was double, and a little more silicate of alumina was abstracted. The same phenomena were observed with a mortar made with this lime.

Two prisms of this mortar were immersed during eighteen months in sea water. One of these prisms had no appearance of a well-marked alteration, whilst the second was in a very advanced stage of decomposition. It was found that some lime had, nevertheless, been eliminated, that a considerable proportion of carbonic acid had been fixed, and that the proportions of magnesia, silica, and alumina, had not undergone an appreciable change. The prism in which the alteration was considerably advanced had undergone a true transformation in respect to its composition. A considerable quantity of lime was replaced by more than an equivalent quantity of magnesia, and the carbonic acid had not sensibly changed; on the other hand, the silica and alumina had appreciably augmented.

Are we to seek for an explanation of these so different results in the non-homogéneity of the the hydraulic lime which had served to make the experiments? We may remark that, in the deposit of Paviers, the different beds of hydraulic limestone have not the same composition. The alteration which the mortar produced from the lime of Doué underwent, consisted in the loss of a considerable quantity of lime, without the substitution of magnesia, and by the fixation of a great quantity of carbonic acid.

With regard to the alterations of cements, that of Boulogne, previously moistened with fresh water, began to crack after an immersion of eight months in sea water; nevertheless, its chemical composition did not sensibly change. It was quite different with Portland cement, which cracked in every direction under the action of sea water, fixed as much carbonic acid as it contained in its normal state, and, finally, it had lost a little lime, which was substituted by a very small quantity of magnesia. Lastly, a mortar prepared with one volume of Portland cement, and two volumes of quartzose sand, immersed during a year in sea water, exhibited no trace of alteration, unless it gained some carbonic acid.

In fine, the facts which we have brought forward, and all those which are detailed in our memoir, prove that the decomposition of limes, cements, and mortars by sea water, do not constantly take place in the same manner; the substitution of magnesia for lime takes place sometimes, but not always, and as it is accompanied by the addition of carbonic acid, the altered mortar consists of a hydro-silicate of alumina, and of a double carbonate, which tends to approach dolomite in composition. But there are cases where the lime disappears without the introduction of magnesia, and

a

the phenomena appears then to occur as if it had been produced in water free from salt, but charged with carbonic acid. Further, in the alteration effected in mortars moderately hydraulic, there is a division of the constituents of the mortar into two compounds, the one rich in earthy carbonates, the other rich in alumina, coming to the surface and forming a snowy deposit, which the waves remove. This partition is not effected, or at least it only takes place very slowly, in the very hard and rapidly setting The alteration which is produced in the latter consists in a simple cracking of the mass, and in the disappearance of a small quantity of lime, with or without its being substituted by magnesia, and in both cases there is a tendency to diminish in volume, whence results the cracking of the mass.

cements or mortars.

In

It only remains for us to speak of those cements which are considered to best withstand the action of sea water. Hitherto the cements of Pouilly, Vassy, and Parker have been looked upon as the most stable. A circumstance struck us in the analysis of these three cements: it is, that they are very rich in oxide of iron, and that that of Parker, which is the best resisting, is exactly the richest. We have, in fact, found 7 per cent. of oxide of iron in the cements of Pouilly and Vassy, and nearly 14 per cent. in that of Parker. Hence we have been led to consider whether the presence of oxide of iron does not powerfully contribute to give to those cements the property of resisting the decomposing influence of sea water. order to justify this opinion, it became necessary to institute experimental researches, by making ferruginous cements and exposing them to the action of sea water. But before doing so, we had to ascertain whether oxide of iron contained in cements and mortars did not behave as an inert substance. Thus we had to examine how far this oxide was capable of forming, in the humid way, combinations with lime. With this object in view we formed directly kinds of puzzolanas, by making mixtures of silica and a little lime with alumina and oxide of iron, and then studied the action of lime water on these mixtures, previously heated to a dull redness. After immersion for some time these substances augmented in volume, and possessed the most remarkable characters. Each of them divided itself into two distinct compounds, one of which attached itself to the bottom of the flask, and had gained considerable cohesion and adherence; whilst the other assumed a flocculent aspect; it swelled out more and more, and rose to about 15 or 16 centimetres above the bottom. In analysing these different compounds, we have found that the quantity of lime precipitated is independent of the presence of alumina, whilst it is augmented by the presence of oxide of iron. Further, we have recognised that the flocculent compound was the richest in alumina, and that the concreted deposit was richest in oxide of iron.

These synthetical experiments having apparently demonstrated that oxide of iron is not an inert constituent of hydraulic cements, we believe that we may conclude that the presence of this oxide would contribute to give stability to mortars and cements immersed in sea water. It remains, in fact, to be ascertained whether cements or artificial hydraulic limes, formed by the addition of lime to ferruginous clays, or mixtures of clay

with hydrated peroxide of iron, or even mixtures of clay and substances capable of generating oxide of iron, will not be attacked by sea water. But these experiments require a considerable time, and in the meantime it may be useful to give publicity to the results which we have obtained, as they may be useful to those engaged in the construction of hydraulic works, and further, because it is of the greatest importance that they should be verified by experience. Whatever may be the future value which the test of experience may reserve to our inductions, two facts have at all events been well established:

1. Those cements which are reported as the best for resisting the destructive action of sea water always contain a notable quantity of peroxide of iron.

2. Certain combinations of silica, alumina, and lime, give, under otherwise similar conditions, very different reactions, according as they are deficient in, or contain large quantities of oxide of iron.

ART. VI. Notices of New Improvements in Mining, Metallurgy, Machinery, Chemical Manufactures, &c., and of Discoveries in general science bearing upon Industrial Arts.

MINING, METALLURGY, ETC.

Montigny's Chronometrical Anemometer for currents of Air in the Galleries of Mines. The Society of Sciences, Arts, and Literature of Hainault, lately proposed the following subject of inquiry:-"To discover an exact method of registering continuously, or at intervals extremely short, the velocity of air, especially in the shaft or gallery of a mine, during at least twelve consecutive hours." Professor Montigny, of Namur, has accordingly produced an instrument intended to fulfil these conditions. It consists chiefly of a clock indicating the hour, minute, and second. The pendulum is about one metre in length, and the whole is placed in an airtight box. The axis of rotation of the pendulum is prolonged beyond the box, and has attached to it a lever at right angles to the pendulum At the extremity of this lever is a flat rectangular disk, perpendicular to the lever, and consequently parallel to the pendulum rod. A counterpoise is attached to the axis within the box, so as to enable the whole apparatus to be in equilibrium, when not subjected to the action of currents. When the pendulum is vertical, the lever carrying the disk is horizontal, from which position it deviates more or less proportionally to the arcs of oscillation of the pendulum.

In order to ascertain the normal working condition of the instrument, it is first set going in still air, and its rate is compared to that of a chronometer, care being taken to adjust the different mechanical arrangements, so as ultimately to establish an exact synchronism between the anemometer and the chronometer.

This being effected, the instrument may be placed in a horizontal current, with the flat disk perpendicular to the direction of the current. The pressure thus received by the disk will evidently be transmitted to the pendulum, and will thus tend to diminish the time of each oscillation. This follows because the entire pressure on the disk can be decomposed into two, one tangent to the are described by the centre of the disk, the other in the direction of the lever. The tangential force tends to render the lever horizontal, consequently the pendulum, vertical, and it is thus an additional force added to those which, in the normal condition of