coal: 2, the chemical equivalent of air required: 3, the relative size and proportions of the furnace: 4, the proper dimensions of boilers: 5, the required extent of heating surface: 6, the relative evaporative results of the foregoing: and 7, and above all, the mechanical, mathematical, and even geometrical connexion between all these and the horse power of the steam! For as to any chemical connexion which might be supposed to exist between such details, quantities, and results, they are unimportant, and beneath his notice they are the mere "trammels" which science would interpose to "set bounds to practical results." "This furnace pot" with which he experimented, "was capable of holding eighteen or twenty gallons," (plus ou moins.) "The fire grate was six inches by eight, or one-third of a square foot in area; and the whole of the heating surface exposed was about three square feet. Into this boiler was measured two cubic feet of water, and after having caused it to boil, it was then found" (doubtless after much labour and calculation) "that by feeding the furnace with coal, and the boiler with water, and at the same time managing the draught of the chimney so as to keep the water boiling nearly at a uniform rate, the consumption of good coal was at the rate of four and a half pounds per hour; and the quantity of water boiled away, in that time, was exactly two gallons, or very nearly one-third of a cubic foot." Admirable and accurate manipulation! The whole of these very elaborate details dubbed "practical," are so scientifically determined, and withal, so very conclusive, that it is not to be wondered at, that the comprehensive mind which could embrace the whole of this complicated boiler-pot process, should also be satisfied with its claim to become the "unit measure of steam power"-the standard for all future calculations, and on which might safely be raised a superstructure to which the mechanical, practical, and scientific world might hereafter look with satisfaction and advantage. But let us see how beautifully and ingeniously these details are worked up into a regular formula for the guidance of engineers. "Now, as it is usual to reckon," says the author, "that the evaporative power of one cubic foot of water per hour is sufficient to furnish steam for one horse power, we have only to multiply each of the foregoing results by three, to obtain the following proportions, viz. :-one cubic foot of water evaporated per hour requires one square yard, or nine square feet, of heating surface, one square foot of fire-grate, and 134lbs. of good coal." How simple and scientific! Let the reader only consider that this formally arranged tissue is seriously laid down and published as the datum line, to which all calculations are to be referred, and which is to settle all disputed relations between combustion of fuel, evaporation of water, and powers of steam. The applicability of this arithmetical equation and scale of proportions to the complicated chemical and dynamical details of furnaces, boilers, and steam-engines, appears to the writer so important that he dwells on it with peculiar satisfaction as furnishing the "unit measure"- the great desideratum-the "be-all and endall" of the questio vexata, combustion with all that belongs to conduction, evaporation, steam power, and even "smoke burning." Can we be surprised, while such details as these pass current, how easily men are led astray, and so called "practical" rules obtain sanction among the unsuspecting class of "operative engineers.' With respect to the doctrine of relative proportions, the results, as shown in the tables inserted in your present volume, page 134, prove its utter fallacy, both as regards the areas of furnaces and the beating surface of boilers. To this, however, I will, with your permission, I am, yours, &c. C. W. WILLIAMS. return. Liverpool, August 6th, 1842. PROGRESS OF FOREIGN SCIENCE. Railway Accidents. The frightful accident of the 8th of May last, upon the Paris and Versailles Railway, has left an impression so durable upon the minds of all, and has been followed by so many discussions, memoirs, and propositions, of one sort or another, both in France and amongst ourselves, that it seems desirable to collect and arrange for future reference, the reports published in foreign journals, of the opinions and discussions of continental engineers upon the causes, &c. which led to this catastrophe, and generally upon those tending to railway accidents, and the means of avoiding them. We shall thus be better enabled to judge of the value of the several plans or opinions enunciated, and this seems the more desirable now, as a committee of the mechanical section of the British Association was appointed at its late meeting, to experiment and enquire into the supposed changes in the strength of axles upon railways, produced by their constant use-a change supposed by some to take place, and to be of sufficient importance to account for the fracture of axles in use, and consequent accident. The accident took place on the 8th of May, and on the next day, M. Cordier read to the Academy of Sciences of Paris a note by M. Combes, engineer of mines, and officially engaged as superintending government engineer of the steam engines existing in the department of the Seine. This note,-which may be considered as the official communication of the accident to the Academy-gives its details, which having been already published in various places, it is needless here to repeat. The foremost axle of the locomotive which broke is stated by M. Combes to have fallen between the rails, and hence broke off short at both ends-the fracture of the iron was lamillar in large plates. The driving axle, also, of the fourwheeled locomotive was found broken, and the fracture appeared to have been produced by torsion. The tender of this engine was upset and broken. The sixwheel engine of Messrs. Sharp and Roberts, which came after the first or fourwheel one, was also upset, the axles bent and detached, but not broken: neither boiler was burst. The tender of the second engine was also broken. The five carriages next the engine were thrown of a heap, and the hot coke thrown out of the fire-box of the large engine, fired first the jacketing of the boilers and the engines themselves, and immediately after the carriages, which burned with such extreme rapidity, that they were consumed in about ten minutes. M. Combes concludes by saying, "without entering into a discussion of the causes of the accident, it must be obvious to every one that the locomotive on four wheels was the principal one, and it seems proper that engines of this sort should no longer be employed." To this paper M. Biot adds a note, in which he points attention to the fact that the death of the victims of this accident was due to their being locked up, and mentions, that on the St. Etienne and Lyons railway the carriages of the passenger trains are always separated from the engine and tender by an empty waggon, which, moreover, carries a spare axle; and that, by means of a simple contrivance, the train can be instantly detached from the engine at the will of the conductor. This contrivance, it may be mentioned in passing, is due to M. Bergeron, a highly intelligent young French engineer, who gave an account of it at the late meeting of the British Association. M. Biot adds, that the structure in general of railway carriages presents only an adaptation of the common road carriage to railway purposes; and that those employed on the Belgian railways are of a very superior design. He suggests that there is much room for improvement in the construction of railway coaches, and instances, as showing the value of appropriate contrivance, that the freedom from any accident of the Parisian omnibuses is to be attributed chiefly to the ease of entrance and egress, &c. given by their peculiar construction. M. Elie de Beaumont adds, that on the Belgian railways the passengers are never locked up-and gives his opinion that the use of two engines to one train is dangerous, and should be abandoned. As each engine has its own chance of accident, we thus double the dangerbut still further, if one engine meets with an accident, it tends to derange the second, which, in its turn, by still forcing on the train, aggravates the danger due to the first. He considers that the mutual action of two engines, which cannot be made to work exactly with a coinciding action, tends to produce accident. He would not permit the use of more than one engine to each train, except to mount slowly an incline, and recommends the construction of brakes, that could be applied instantly and simultaneously to the whole train; and that one or more wagons, filled with some elastic material, should precede and follow each train of carriages, At the next sitting of the Academy, on the 16th May, M. Perdonnet (who, it will be recollected, was engineer of the line) read a note, or memoir, which is chiefly an endeavour to set aside the conclusions of M. Combes, previously stated. He proposes to consider three questions, bearing upon the causes of the accident, viz.- 1st, Are four-wheeled engines really more dangerous than those of six wheels? 2nd. When a four and a six-wheeled engine are coupled to one train, is it dangerous to place the four-wheeled one in advance, or should it be behind the other? 3rd. Would the use of small trains on the Paris and Versailles (left bank) railway be more or less dangerous than large ones, as at present? On the first question, he comes to the conclusion that four-wheeled engines are not more dangerous than six-wheeled, and are better for going round curves. His arguments are chiefly the authority of certain lines of railway in England, the non-condemnation of them by the Committee of Parliament on railway accidents the fact, that if the fore-axle break either in a four-wheeled or sixwheeled engine, the foremost end falls forward, and the whole is thrown over-and that in the four-wheeled, if the drivingaxle break it cannot let the engine down, as it is between six bearings-that in Stephenson's six-wheeled engines, without flanges to the driving. wheels, if the foreaxle break, the engines must run off the line-that on curves, the whole weight of the engine is sometimes borne by four of the wheels, and hence, the rails are broken by the enormous weight and the engine runs off the rails-and, finally, that experience on lines, where both sorts are used, or either exclusively, there is no difference in the amount of accidents. On the second question he concludes, that the four-wheeled engine ought to go first, because less liable to run off the rails; and because a broken fore axle is as bad for one as for the other; and, lastly, because in gradually gathering up the heavy train, the less powerful engine should be in advance. On the third question he concludes, that heavy trains are more safe and convenient than a greater number of lighter ones, on this particular railway, where lighter trains would have to start, at least on holidays, from each terminus, every quarter of an hour; and where any delay would be almost sure to cause the trains to run into each other; and would produce great danger in the very many crossings of roads on the level of the rails. He contends that two engines are better than one, applied to a heavy train, because by their weight they give the power of rapidly stopping, in case of an obstacle ahead; and he thinks a train of 30 carriages, drawn by three engines, would be safer than three trains of 10 carriages, each drawn by one engine, because it is possible that a broken axle of an intermediate engine might not, (by the traction of the others,) be permitted to stop or derange the motion of the train until it would be stopped. He considers a moderate speed essential, and that the serious nature of the accident was due to a peculiar and rare coincidence of unusual circumstances, viz., the breakage of a fore-axle, and just at the passage of a public road over the rails. Lastly, he recommends the adoption1. Of the attachment-hook of M. Bergeron, before mentioned. 2. The making the carriages of incombustible wood. (He says nothing as to the other materials composing them.) 3. The use of wagons before and after the train, loaded with inert matter. 4. The establishment of an examination as to knowledge and ability, &c., of all locomotive engine-drivers, plate-layers, &c., and giving them diplomas. Mr. Perdonnet entertains some doubts of the possibility, in a fiscal point of view, of adopting the third proposal. The views contained in this memoir are not to be neglected; but they are, many of them, flimsy in the extreme, and throughout there is manifest a special pleading to prove that the arrangements made on the line, at the time of the accident, were the best that human wit could devise, and that no blame is any where attributable. At the same sitting there was also read a memoir from a M. Manby, called "A Defence of four-wheeled Locomotives." The arguments are almost precisely the same as those preceding, with the addition of developing more fully the causes why six-wheeled engines fall forward, when the fore axle breaks. The only cause assigned, which has any pretension to novelty, is, that of the reaction of the issuing steam, &c., from the blast-pipe and funnel. The author concludes, that the fracture of a fore axle of any engine, of whatever sort, or of a carriage, must produce a sudden stoppage, and probably an accident; but denies that the number of wheels, &c., had any thing to do with the accident of the 8th of May, the real cause of which, however, he does not specify. Twenty-two other communications were addressed to the Academy, and are not published, but referred to the commission appointed upon the previous memoir of M. Perdonnet. The commission consists of Arago, Poncelet, Coriolis, and Leguier. A long letter was also read by M. Delessert to the Academy, from M. Prevost, who, it appears, is employed upon the London and Birmingham line, and describes the practice as regards engines, &c., on that line. There is in it nothing very much to the point, or new to English professional readers. M. Prevost is altogether in favour of four-wheeled engines, which alone are used on his line; and attributes the accident solely to excessive speed: he would prohibit any speed upon inclines beyond 30 miles per hour. On the 23rd of May, M. Franchot brought before the Academy his contrivance for preventing a shock to the train, whenever the locomotive preceding is stopped by any cause. The contrivance consists in connecting the engine to the train by a series of jointed parallelograms, like a "Lazy tongs," which are stretched at length while all goes on right; but on the stoppage of the locomotive, the parallelograms are forced together, and each is reacted on by a spring. It is, in fact, a plan of uniting, in succession, a great number of separate buffers, and is not devoid of ingenuity. Another plan was also brought forward by M. Jouffroy, for effecting the same object, by means of brakes, acting constantly upon each individual carriage, and only relieved by the traction of the engine upon the draw-bar in front of the carriage. As soon as the traction ceases, froin any cause, the brake of each carriage instantly begins to act. No details are given, but a plan is said to have accompanied the communication. It is hard THE DISC PHENOMENON-FIRST OBSERVED BY MR. ROBERTS, OF MANCHESTER, AND NOT BY M. CLEMENT DESORMES. Sir,-In your valuable Magazine for July 9th, 1842, is a communication from Mr. W. Wynn on what he calls "the disc problem," to which you have appended a note, stating that "The phenomenon alluded to by your correspondent was first noticed by M. Clement Desormes." This has been often said, but is nevertheless erroneous, as that gentleman first saw it exhibited at the works of Messrs. Sharp, Roberts, and Co. of Manchester, in company with the late Dr. Henry, of the same place. I herewith send you a copy of a paper read to the Literary and Philosophical Society of Manchester, many of the members of which were cognizant of the facts respecting my first observation of the phenomenon,-the visit of M. Clement Desormes, to whom it was shown, -and his subsequent statement respecting it. That statement did not, however, distinctly affirm that it was first noticed by him, but merely gave the fact that such phenomenon had been observed, leaving it to be inferred by the reader, that it was then observed for the first time. The paper which I send was published in the Memoirs of the Manchester Literary and Philosophical Society, with the note explaining that M. Clement Desormes first saw the phenomenon in Manchester; and I believe that that gentleman has never denied the correctness of the explanation given in the note. The cause of truth and fair dealing induces me to request that you will insert this in your Magazine and should it appear to you advisable, also some extracts from the paper. I am, Sir, your obedient servant, The printed paper which Mr. Roberts has been so good as to send us along with the preceding communication establishes incontrovertibly, that the merit of having first observed the remarkable phenomenon in question rests entirely with him. It is a pity that a philosopher of M. Desormes' eminence should not have had candour enough to spare Mr. Roberts the trouble of thus reclaiming his own. The paper is entitled "Experiments and Observations on Diverging Streams of Compressed Air," and was read before the Literary and Philosophical Society of Manchester, March 9, 1827. The author is Mr. T. Hopkins, but the experiments related, as will be seen by the following extracts, were made by Mr. Roberts, with the assistance (latterly) of the writer of the paper: "On the eleventh of October, in the year 1824, Mr. Roberts affixed a valve to the aperture of a pipe, used as a waste-pipe, for the purpose of regulating or equalizing the force of a blast of air which was blowing a furnace. To his surprise, however, he found that the valve, instead of being readily blown off by a strong blast, remained at a small distance from the aperture of the pipe, and was removed to a greater distance only by a considerable exertion of the power of the hand. This singular phenomenon was witnessed by many gentlemen, members of this society, in the same week, and appeared to be viewed by them all, as equally new and extraordinary.* "Mr. Roberts made some experiments on his air-valve at the time, and various theo-, ries were then suggested to account for the adherence of the valve to the pipe. It was not, however, until the month of September in the present year 1826, that I agreed to join him in making further experiments, a part of which I now proceed to give. * * * * * * "The valve was attached to one end of a scale beam by a string, and balanced by weights placed in a scale, attached to the opposite end of the beam. The valve being thus placed on the seat, without any weight of its own to press downward, the stream of compressed air was admitted into the pipe, when the valve rose from the flange or seat, 1-32nd of an inch, and there remained stationary. Thirteen ounces, avoirdupoise weight, were now put into the scale, which Mour. Clement, of Paris, was, I understand, in Manchester at this period, and saw the air-valve adhere to the pipe, yet he afterwards, it appears, represented the discovery to have been made in France long subsequent to the time he saw it at Mr. Roberts' works. raised the valve to 1-12th of an inch above the seat. Twenty-six ounces raised it to 1-8th of an inch, and thirty-two ounces raised it to 1-4th of an inch, but any weight beyond this last caused the valve to fly abruptly off. "It thus appeared, that when the valve was raised from its seat a quarter of an inch, there was the greatest difference between the force of the issuing current of air pressing against the under side of the valve, and of atmospheric pressure on the upper side of the valve. The pressure of the atmosphere was greater than the force of the issuing stream of previously compressed air, a weight of thirty-two ounces being requisite to establish an equilibrium. "That we might ascertain what was the state of the stream of air under the valve, in different parts of it, four double syphon tubes were procured, and proper quantities of mercury being put into them, they were inserted in holes made through the valve at certain distances from each other. The inserted limbs of these tubes being thus left exposed to the action of the stream of air, the compressed air was again admitted into the pipe, and the valve rose as before, 1-32nd of an inch. "From a general view of the results thus obtained, it appeared that while the valve adhered to the seat, and remained at but a small distance from it, a circular stripe or flat ring of attenuated air was found between the valve and its seat, and near to the aperture, the air at the same time in the parts further from the aperture becoming more dense, until close to the periphery it became nearly of common atmospheric density; but as the valve was raised, the ring of attenuated air approached the outer part or periphery of the valve. "To find the form and nature of this ring, it now appeared desirable that the different heights of mercury in the same tube, indicating degrees of vacuum should be ascertained at small and equal distances, beginning at the edge of the aperture, and proceeding along a radial line to the periphery of the valve. * * * * "These experiments showed, that until the valve was raised to a certain height above its seat, the under side of that part of the valve which was over the aperture, was exposed to a pressure of 14 inches of mercury more than atmospheric pressure; and the under side of all the rest of the valve, forming an outer stripe or ring, was exposed to a pressure less than atmospheric, or had a partial vacuum varying from one and 8-10ths of an inch of mercury up to atmospheric |