The Card Trick To the Editor of the SSCIENTIFIC AMERICAN: In reference to the article Another Card Puzzle, which appeared in the October 5th, 1912, issue of the SSCIENTIFIC AMERICAN, I would advise those interested that the trick, as I call it, can be done with any number of piles of cards from one to five inclusive, the latter number being the best, as there would be fewer cards left, and if the face cards are valued at ten each, three added to the number of cards left will give the total of the bottom cards on the five piles. To make this trick seem more mysterious, ask the party counting to put two cards on each pile, then one on each, or two on the first and second piles, and so on, in any manner that you wish, until the cards remaining have all been used. This is done to find out the number of cards without asking for them or without letting anyone know that you must have the number. I find that this adds immensely to the effect of the trick. New York city. R. W. HOEL. Required, a Machine for Shaping Granite Cubes To the Editor of the SCIENTIFIC AMERICAN: A pavement somewhat similar to that which has been in use in the streets of Liverpool for a number of years past has recently been laid in Pulton Street in this city. This pavement, which consists of small granite cubes some three or four inches across, accurately cut so as to admit of laying with close joints, seems to be better adapted than any other to meet the requirements of heavy street traffic, and at the same time affords a good footing for animals. The chief obstacle to its extensive use lies in the fact that the cost is considerably greater than most other forms of pavement. It has occurred to me that these granite cubes could be produced at a lesser cost if labor-saving machinery were employed in shaping them. Would you kindly call this requirement to the attention of inventors throughout the country, in the hope that they may be able to solve the difficulty? HENRY BERGH, Chairman Committee on Improved Pavement. New York, N. Y. [If the cubes are shaped by hand, their relatively high cost is readily understood. It should not be difficult to devise suitable machines for doing this work.--EDITOR.] Determining Differences in Longitude by Wireless To the Editor of the SCIENTIFIC AMERICAN: The paragraph in your issue of October 5th regarding the proposed use of wireless by the Belgian government. for the exact determination of difference of longitude, induces me to remark that this device was successfully employed last summer by Canadian government vessels. The work was performed by the "Arctic" and "Minto," and consisted in the charting of the Button Islands at the entrance to Hudson Strait. While one vessel remained at anchor, the other visited the various islands which compose the group. Each day (the weather permitting) observers from each vessel made careful shore observations for local time, and then, as soon as possible, these times were compared by wireless. The distances involved ranged from thirty to fifty miles, and it was found that by pressing the key at intervals of a minute for four or five minutes, an accuracy in transmission considerably greater than that of the observation itself could readily be attained. The method proved satisfactory in every particular, and its rapid extension seems assured. Berlin, Ontario. W. B. WIEGAND. The Carob Bean Tree To the Editor of the SCIENTIFIC AMERICAN: Your interesting article on "The Carob Bean Tree" in the January 11th issue of your highly esteemed paper reminded me of my observations of this tree. It grows here in a wild state, both the bean-bearing and the male tree. The latter is used for posts of various kinds, is very durable and hard, also rapid growth. The female develops less rapidly and has thorns from eight inches long down to very small ones on the smaller branches. The beans or pods are flat and black and measure as long as sixteen inches and about one and one half inches wide; the honey lying in the thicker portion of the pod and the beans in the thin. I have seen them over eighty feet in height, and the lumbermen call them "black locust.\ I know of a row of the male locusts near here, about one quarter mile long which were originally posts for a wire fence, and were allowed to grow into trees now about fifty feet high. I have also seen telephone posts made of this variety, which developed into beautiful trees. They bloom in early spring, with great clusters of pendulous, creamy white and very fragrant blossoms; the foliage is also very graceful. Cold does not affect this tn) here. The Louisiana orange is grown some four hundred miles south of us, and does not stand our climate. This tree is very hard to kill by cutting down, as it will invariably put out not only suckers from its most abundant roots, but will also sprout from the stump. The soil here is rich and sandy below the surface, with a good layer of rich black topsoil; the lowest stratum is gravel of an exceptional quality. In my younger days we would bind the long hard thorns to a lance of bamboo cane and use them for snake spears, and with their many tough prongs they answered admirably for this purpose. Incidentally I have used a smaller thorn in place of a lost suspender button, the utility of which can be imagined. The male locust is also known here as "locust and wild-honey" tree. Negro children sometimes eat the honey portion, but it is never used as a food. Neither the male nor female trees are ever green here, and we are perfectly bare of trees now. I have seen these trees in Miami, Florida, in Arkansas, Tennessee, Texas, and Louisiana, besides my home State, Mississippi. It is interesting to know that there is more value attached to this tree than was supposed here. I send you this little article for what you may wish to do with it. I have had the benefit and pleasure of being a subscriber to and reader of your valuable paper for twenty years, and I still look for its weekly advent with pleasant anticipation. P. P. BESSAC. Natchez, Miss. Misuse and Failure of Metals and Alloys To the Editor of the SCIENTIFIC AMERICAN: In order to bring about a better understanding between the users and manufacturers of metals and alloys, I venture to request the favor of the insertion of this letter. I hope it may provoke discussion, and if possible help us to arrive at the truth. Users, including the non-expert, public, of metals and alloys in general do not sufficiently realize that their corroded metallic wares, for example copper pans and boilers, mysterious breakages of for instance chains, railway accessories, and sudden failures of condenser tubes and copper pipes, are due to two easily preventable causes described below. In order to satisfy the demand for beauty of form or the modern craving for cheapness, many ingenious mechanical devices have been evolved by manufacturers' works staffs, which put upon the metallic articles, while being made, uneven strains, or introduce in antiquated furnacing appliances deleterious compounds, which tend to and, in varying times, finally render the articles useless in the hands of the purchasers or users. Often enough they'are dangerous to human life and property. An attempt is made to remove these strains by annealing in unevenly heated furnaces. Why this state of things? One must reply, "general ignorance," perhaps "apathy"--the major causes of most human troubles. A few manufacturers, whose products are subject to constant physical tests, are quite alive to the situation. It must also be conceded that manufacturers of ferrous and non-ferrous metals and alloys spare no pains to free them while in the molten state from deleterious substances. They employ expensive nostra known as de-oxidizers--ingenious devices to prevent contact of the liquid metal with the air during the casting operations, all tending to produce soundness in the ingots or castings, i. e., to free them from blowholes and segregations. No sooner is this desirable end attained than the metal or alloy is introduced into furnaces wherein hot gases containing free oxygen, sulphur, and other objectionable elements are allowed to impinge upon or envelop the metal. While hot, they are brought out of the furnace into the air and mechanically treated in an atmosphere containing oxygen. This operation is often repeated several times. In cases of cold working, the metal, with few exceptions, is annealed between processes in furnaces to which the air has access. After this heat treatment they are withdrawn and allowed to cool, sometimes in approximately closed receptacles, oftener in the air. Not only is the surface of the metal oxidized or tarnished, but during the whole time of heating, and partly in cooling, solution of oxygen and other gases takes place, with formation of compounds in intergranular spaces, or in the body of the metal or alloy, which form centers or areas of corrosion when in use. In the last year or two, exact investigations and experience have proved beyond doubt that both the above defects in heat treatment are prolific causes of corrosion and the other breakdowns. I therefore with all respect beg leave to submit that the time has arrived for all engineers and users of metals and alloys to insist and specify that, at least, two causes of failure of metals and alloys under the control of the makers shall be removed, namely: (a) Uneven heating appliances, (b) Antiquated heat treatment in chemically and physically active atmospheres. Incidentally the public health will benefit, because all the appliances on the market capable of bringing about the above results are smokeless when in operation. The present unnecessary and wicked waste of one of our natural resources, viz., coal, will also be curtailed. T. VATJGHAN HUGHES, A.R.S.M. Birmingham, England. Bursting of 13.5-inch and 14-inch Guns To the Editor of the SCIENTIFIC AMERICAN: In your number of December 14th last, when speaking of the bursting of an English 13.5-inch gun, which had taken place in Shoeburyness a few days previously, you alluded to the fierce controversy which had taken place a few years ago, respecting the question of the relative strength of the wire-wound guns and hooped guns. Starting from the justified assumption that the English gun which burst was wire-wound and of the most recent pattern, you recalled to mind that "the advocates of wire-wound construction claim, or did claim, before the recent improvements in hooped guns, that the wire-wound gun, because of the absolute inspection to which every part of it could be subjected, was proof against the kind of accident which recently happened at the proving ground." Prom the wording of the article it results clearly that you had doubts as to whether the claims of the manufacturers of wire-wound guns were well founded, and I presume that these doubts will have been further strengthened by the bursting of a 14-inch wire-wrapped gun which took place at the Sandy Hook proving ground on December 9th. As appears from the Army and Navy Journal of December 14th, the gun, after having fired a first shot with reduced charge, burst at the second discharge with a normal charge of 320 pounds powder and a projectile of 1,660 pounds, producing a pressure of 42,000 pounds per square inch; while the contract strength of the gun called for a minimum of 55,000 pounds. This is quite an extraordinary event, which, taken together with that which had taken place in England, where the 13.5-inch gun burst at the seventh discharge, shows that the criticism against the wire-wound guns is well founded, and that it is, in fact, not true that the latter have a circumferential strength greater than that of the hooped guns. That the wire-wound guns are very defective as regards longitudinal strength is a matter which is now so well known that there is no need to demonstrate it, and in connection therewith it will be sufficient for me to refer to the important article which was published on this subject in the February-March number of the Journal of the United States Artillery. As regards the circumferential strength, I beg to call your attention to the two important essays which were published in the "Memorial de l'Artillerie Navale" of 1912 by the ordnance and naval engineers Messrs. Leon Coupaye and Pierre Malaval. These two gentlemen have exhibited, by different methods and procedures, the following principle: "Whatever may be the system according to which a compound gun is constructed, whether it is hooped or wire-wound, the internal pressure which it can resist without altering its shape permanently has as its upper limit the value of the limit of elasticity of the metal of which its internal tube is manufactured." Thus, for instance, a gun the inner tube of which is manufactured of a metal with a limit of elasticity of 40 kilos, cannot resist, without a permanent deformation, an internal pressure exceeding 4,000 kilos per square centimeter (56,891 pounds per square inch). This limit of internal pressure can, however, be reached only in case the internal tube is of infinite thickness; it is less in practice, and decreases with the thickness of the tube. In the wire-wound guns, the internal tube is of small thickness and strongly compressed by the steel ribbon surrounding it externally. The limit of elasticity of the ribbon is superior to that of the tube, and consequently, under a given pressure, the latter suffers an elongation greater than that of the ribbon. When the entire arrangement returns to its state of repose, the tube can no more retake its former dimensions, because it does not find the necessary space to do so and, consequently, contracts or breaks. The contraction of the internal diameter of the wire-wound guns is a well-known phenomenon which I have personally ascertained; it has been, may be, and will be the cause of the jamming of the projectile in the bore, and of the consequent bursting of the gun. I think that to this cause must be attributed the bursting of the English 13.5-inch gun, as well as that of the American 14-inch gun, and it seems to me that these two accidents, which took place within a few days of each other, must give rise to serious thoughts. Torino, Italy. ETTORE BRAVETTA, Captain Italian Royal Navy (letired). [The theory of our correspondent regarding the cause of the bursting of the 13.5-inch gun is decidedly interesting. He is wrong with regard to the American 14-inch gun, whose failure was due to other causes which we are not at liberty to disclose.--EDITOR.]