I then poured about three and a quarter pounds of peas into the tube, and finding that the wooden cylinder which was perfectly free, did not fall, I added a weight of two and a quarter pounds and other weights, without even moving it. It might still be objected, however, that the small cylinder adhered to the sides of the tin tube. To answer this objection, and to render this experiment more striking, I removed the cylinder, and made use of a simple disk of wood of greater diameter than the tube, and supported against its bottom by placing in the balance just weight enough to keep the two in contact. This weight was commonly from ten to twelve grammes.

I then filled the large tube with from three to four pounds of sand, and placed additional weights upon the top of the column, nevertheless the disk, retained by the small counterpoise of ten or twelve grammes, did not move. If this same weight of a few grammes had been laid on that part of the disk which projected beyond the tube, it would without doubt have caused it to fall, for it alone retained the disk in its place. A slight touch of the finger, caused the sand to pour from the lower end of the tube, and fall into a basin placed below to receive it. The disk was therefore instrumental in retaining the sand, but did not sustain the weight of it, which was all transferred to the sides of the large tin tube. Ten grammes would have caused this disk to separate from the tube, and since it remained adhering to it, the disk was not loaded with the mass of the sand.

6. To remove all kind of doubt, I gave up the use of the balance, and placing a tub of water near the large fixed tube, floated the disk of wood on the water with the smooth side upwards; I then brought the end of the tube down upon the disk, and poured water into the tub. The disk was

pressed by the weight of the water against the end of the tube. I next filled the tube with dried peas but the disk did not move. It, however, was essential in retaining the peas, which without it would have fallen through the tube; but the peas did not press upon it, since a very small force would have sufficed to make them fall from the tube and thus derange the whole apparatus. 7. Leaving every thing in the same condition, I poured water into the large tube; it was kept there with the peas, for a considerable time, until an unforseen motion produced by the compressed air, which was disengaged from the bottom of the tube, caused the machine to incline. The peas then escaped into the tub, and the water flowed out at the same time. The same trial was made with sand; a considerable quantity of water was poured on the sand, fully impregnating it, and during a very long time it was supported without flowing out.

In another trial made a little differently, the sand took such a consistence with the water that it caused much trouble to get them out of the tube, which therefore entirely supported the weight of the sand and of the water, together with the force necessary to expel them.

8. We can make these experiments by simply causing the large tube to rest on a small conical heap of sand, whilst it is still suspended from the disk of the balance. The sand does not escape when the weight put into the other disk is nearly equivalent to the weight of the tube and its contents.

The same trials succeeded with grain: I have repeated them with shot with equal success, although this has a very great weight. They may also be made with a simple roll of paper tied with two small strings; they are then much more striking as the weight acquired by the paper tube contrasts better with its original lightness.

9. I have repeated these experiments with a tin tube widened at the bottom and much larger than the great tube; the result was the same, although there can be no doubt that there is a limit beyond which the sand would receive no further support from the sides of the tube. This will be the case when the inclination of these sides to a horizontal plane is the same as the slope assumed by sand in a heap, that is to say about thirty degrees. I have also repeated several of these trials with a cylindrical tube four inches in diameter, with the same success.

10. From all that I had seen I presumed that it would be very difficult to force sand through a tube even by means of a direct pressure. I made the trial in the following manner. I filled the great tube with sand and laid it in a horizontal position, and with a cylinder of wood, several feet in length, and a little less in diameter than the tube, endeavored to force out the sand at one end by pressing it at the other, but without success. It appeared to me that it would be easier to burst the tube than to move the sand a single inch. The tube being inclined to the horizon about twenty degrees, and the effect being thus aided by the weight of the body, the sand still could not be expelled; the same result followed in inclining the tube in the contrary direction. This explains very clearly why a blast confined with sand is as effectual as-any other.

Ynerduv, 15th January 1829. P. S. If in the experiment in section 2, under the head of the pressure of sand, we pour water into the tube which contains the peas, the mercury will rise in the glass tube one-fourteenth of the height of the water; a proportion which corresponds with that of the specific gravities of those liquids. The water acts as usual, but the peas exert no pressure.

2nd. There is another way of making the experiment with the tube which is within the reach of every body. Procure a tin tube an inch in diameter and as long as is desired, open at both ends. Take a sheet of fine paper and apply it against the end of the tube pressing up the edges with the hand so as to make it take its form; then moisten the edges of the paper with water and cause them to adhere to the sides of the tube. Place the end on a table and fill the tube with sand. Raise it with care, and notwithstanding the slight adherence of the paper, the sand will be sustained while the tube is freely moved about.

3rd. It would be desirable to place a vessel of sand provided with an orifice for its escape, under an air pump, in order to determine whether the velocity would be affected by its flowing in a vacuum.

(TO BE CONTINUED.)

[Biblioth. Univ. XL, 22.

FOR THE JOURNAL OF THE FRANKLIN INSTITUTE.

On the Manufacture of Military Projectiles, Translated from the French of F. J. Culmann, Chef d'escadron d'artillerie, &c. &c. by ALFRED MORDECAI, Captain United States Ordnance Department.

The principal objects of this article are to point out certain faults in the manufacture of projectiles, and to indicate the means of giving them an even surface, an accurate eye, a thin seam, exact dimensions and perfect sphericity: on these points no detail will be neglected, but we shall not dwell on the description of processes which are well known in founderies.

Of the Iron used for casting Projectiles.

The kind of cast iron of which the best hollow projectiles are made is that obtained from very fusible ore, reduced with charcoal in furnaces of small elevation, at a medium beat, or by working the furnace in such a manner that the metal may be well mixed, inclining more towards a lamellar white metal than to grey, so that the lamina, marked with greyish spots, may still be distinguished in it. The surface of a projectile made of this metal, which is very liquid, is perfectly smooth and free from flaws and holes, which is not the case with those made of grey metal, particularly of that which does not run freely. Metal inclining to white cannot be obtained with certainty from refractory ores, nor even from fusible ores if reduced with coke, or in furnaces of a certain height: this metal is moreover unsuitable for the manufacture of other articles, even for that of solid projectiles. In general, therefore the production of it is not desirable, and when accidentally obtained, it can seldom be used, because the projectiles made from it are too small; white cast iron, or that which inclines to white, shrinks more in cooling, or else at the instant of becoming solid, it expands less than the grey metal. In order to employ it usefully, therefore, the dimensions of the mould must be adapted to the properties of this kind of iron. It may also be doubted whether this brittle iron presents a sufficient resistance to the force of the powder, to prevent the projectile from being broken in leaving the piece, and to enable it to give, in certain cases, large fragments moving with sufficient velocity. It is used however in one of the iron districts of France, and with excellent results.

For the casting of hollow projectiles it is of little consequence whether the metal be good or bad, with reference to the quality of the fine iron obtained from it. It may even be said that the metal which produces a brittle iron, and which is generally very liquid, is better suited for this purpose than that which produces tough iron, provided that its bad quality does not proceed from the presence of too much silex, which would cause cracks and rents.

Castings which are to be very dense and solid, and of a medium thickness should not be made from the crude iron of coke furnaces, when it contains a large proportion of earthy minerals: a portion of the latter is thrown out when the metal is cooled by exposure to the air, and this causes flaws in the interior resembling rents; and when this metal is cooled without exposure to the air, interior crystallizations are formed, which also produce flaws. These phenomena, arising from unequal cooling, seldom occur if the castings are either very large or very thin; in the former case the metal being very liquid, heats the mould and then cools nearly at the same time throughout the mass; in the latter, the cooling is almost instantaneous.

Metal which does not contain a large proportion of earthy minerals has no tendency to form these crystallizations; an appearance which resembles them at first sight is sometimes produced by lamina of graphite which in the cooling of the metal, collect in the interior of the mass. Crystallizations are frequently met with in the white grained metal, (when it is not produced by an overcharge of ore,) as well as in all kinds of grey-iron which contain earthy minerals.* Unless we have the exclusive use of a furnace, the di

*These crystalline forms, so common in bombs, rarely occur in twenty-four pound howitzes, or in flasks. Out of thirty-eight ten inch and twelve inch shells, rejected for other reasons, which I have had broken, one third presented, in the fracture, crystallizations coloured yellow, crimson, &c. In six inch howitzes, we find rents produ

mensions of the models, or globes, should be regulated according to the quality of the metal which in the particular foundry employed is best adapted for casting in sand, or for making most of the common cast iron utensils, and this is generally a mixed metal. The grey metal may also be used if it have the property of remaining liquid, which will be the case when the mixture of ore and fluxes is somewhat refractory; but if the grey metal should become thick and throw out a large quantity of graphite, it would give the projectile a very porous, wrinkled surface, covered with dross, and of an unseemly appearance. What we have said of the kind of metal best adapted for hollow projectiles, does not apply to that which should be used for making shot; white metal, or that which inclines to white, gives very ugly shot. The best is a slightly mixed metal, inclining rather to grey than to white, or else the clear grey metal, very liquid and having a pure slag. Such metal is easily obtained in furnaces fed with coke or charcoal. It is to be observed that the ore which furnishes brittle iron, whatever may be its colour, does not give as good shot as some of the ores from which medium, or tough iron is obtained; but the latter are generally of too much value for the manufacture in question. To obtain shot of an even surface, a certain quantity of the better quality should, however, always be added to the former kind.

We are at no loss to understand that white metal which, when poured into moulds, presents a very even surface, may furnish good shells, and at the same time be unfit for the fabrications of shot; because the latter must be rolled and hammered, and this metal is not adapted to either of those operations; the same may be said of almost all the ores of very brittle iron; they are not sufficiently ductile to take a smooth surface after having been hammered.

Grey metal which is a little thick, occasions, around the superior pole of the projectile, small cavities, very narrow and deep; especially if the metal has been reduced with coke of a bad quality, or from impure ores. In that case it contains a large quantity of silex, a part of which is separated from the metal by oxidation and cooling; if the ore is, besides, very fusible, the metal throws out graphite in cooling. This graphite and the silex thrown out are collected about the superior pole, where, mixed again with a certain quantity of metal, they form a soft spongy matter which gives a very bad appearance to the shot, and should cause its rejection-when metal

ced by the expulsion of the earthy minerals; but these substances are not entirely crystallized, because the metal of this projectile is thin and cools quickly. In the twentyfour pound howitzes these rents are for the same reason' very rarely found, and never in grenades. The pellicles which so often appear on the surface of projectiles, are produced only by the crystallization of the earthy minerals. These troublesome accidents may be prevented by keeping the metal for some minutes in the ladles; when poured into the moulds it then becomes well mixed, and the tendency of the foreign substances to separate from the mass is counteracted; as the metal cools more quickly this separation becomes less easy, and the flaws are neither so great nor so numerous. This precaution should not be neglected in the fabrication of projectiles; if the metal be used too hot, depressions and cavities occur in cooling. These depressions, which are found about the eye, on the interior surface of six inch and eight inch howitzes, and ten inch and twelve inch shells, are caused by rents which often extend from the centre of the thickness of the metal to its interior surface. In some foundries most of the rejections are caused by the faults we have just mentioned, and we cannot too strongly recommend to those charged with the manufacture of hollow projectiles to allow the grey metal, when very hot, to remain a short time in the ladles; especially metal obtained, as it generally is for this purpose, from impure ore.

entirely grey has been obtained by means of charcoal from refractory and rather pure ores, it becomes more liquid, throws out less graphite, and is more suitable for making shot of an even surface; but small cavities may still be seen at the superior pole.

Of casting hollow Projectiles.

The moulds for hollow projectiles are made of sand; clay was formerly used for the cores, but they are now also made of sand; at least it is to be hoped that this improvement will be generally adopted. Pit sand should be used for moulds; river sand has too little adhesiveness. It should be of a fine grain, and of such a consistence that it may stick together when pressed in the hands. If it contain too much earth it adheres to the casting, and gives it a rough surface: if too pure it has not sufficient consistence, and the moulds are easily broken and spoiled. The sand should however be as pure as it is possible to use it, in order that the surface of the casting may be more readily cleaned.

Sand which is too earthy may be easily corrected by the addition of dust from charcoal, coke or mineral coal, a very refractory substance which may be obtained perfectly fine, and which resists, in the strongest manner, the tendency to vitrification, and consequently to the adhesion of the sand to the metal. The dust of coke or of mineral coal is preferable to that of charcoal, and should always be used to give projectiles a fine surface. Calcination also furnishes the means of preventing the sand from adhering too strongly to the metal; and this method is naturally employed, by making use of the sand in which other castings have been made. It is necessary to mix it with. fresh sand in order to give it greater consistence, and at the same time a certain proportion of the dust of coke or of pulverized coal is added.

Before using sand it is dried, then sifted, and properly worked and moistened: the quantity of water added should be the least possible to make it fit for use, because too much moisture may cause the casting to fail; there is however no danger to the workman in an excess of moisture, because the steam finding little resistance, passes easily through the sand without causing explosions, which often occur, in using clay moulds.

Of Sand for Cores.

Sand for cores should of course contain more clay than that used for moulds, in order that after having been dried the cores may be so hard as not to be easily injured, and that they may adhere properly to the spindles. If the sand contain too much clay the core would not dry thoroughly without long exposure to a very high temperature. This inconvenience may be remedied by the addition of pure sand, or of coke dust, and by drying a second time.

In general, the quality of the sand and the degree of heat to which the cores should be exposed are dependent on each other. It is easy to succeed by subjecting them a sufficient length of time to a high heat; but the results are more certain, the operation is quicker and less expensive, when the sand is of the proper quality, having sufficient consistence not to be easily separated, and at the same time not retaining water with so much force as to require exposure to a very high degree of heat. If sand of this quality is not to be found on the spot, it may be composed by mixing the different kinds, or even by adding clay, provided however that it does not contain too great a proportion of calcareous matter, the proportions of the mixture are