[Illustration]

SCIENTIFIC AMERICAN SUPPLEMENT NO. 647

NEW YORK, MAY 26, 1888

Scientific American Supplement. Vol. XXV. , No. 647. 

Scientific American established 1845

Scientific American Supplement, $5 a year. 

Scientific American and Supplement, $7 a year. 

 * * * * *

TABLE OF CONTENTS. 

 PAGE. 

I. ARCHITECTURE. --Elements of Architectural Design. --By H. H. Statham. --Continuation of this important contribution to building art, Gothic, Roman, Romanesque, and Mediæval architecture compared. --26 illustrations. 10339

 The Evolution of the Modern Mill. --By C. J. H. Woodbury. --Sibley College lecture treating of the buildings for mills. 10329

II. CHEMISTRY. --An Automatic Still. --By T. Maben. --An improved apparatus for making distilled water. --1 illustration. 10335

 Testing Indigo Dyes. --Simple and practical chemical tests of indigo products. 10342

III. CIVIL ENGINEERING. --Railway Bridge at Lachine. --Great steel bridge across the St. Lawrence near Montreal. --2 illustrations. 10333

IV. ELECTRICITY. --Influence Machines. --By Mr. James Wimshurst. --A London Royal Institution lecture, of great value as giving a full account of the recent forms of generators of static electricity. --14 illustrations. 10327

V. HYGIENE. --The Care of the Eyes. --By Prof. David Webster, M. D. --A short and thoroughly practical paper on the all important subject of preservation of sight. 10341

VI. MECHANICAL ENGINEERING. --Economy Trials of a Non-condensing Steam Engine. --By Mr. P. W. Winans, M. I. C. E. --Interesting notes on testing steam engines. 10331

 The Mechanical Equivalent of Heat. --By Prof. De Volson Wood. --A review of Mr. Hanssen's recent paper, with interesting discussion of the problem. 10331

VII. METEOROLOGY. --The Meteorological Station on Mt. Santis. --A new observatory recently erected in Switzerland, at an elevation of 8, 202 feet above the sea. --1 illustration. 10341

VIII. NAVAL ENGINEERING. --Improved Screw Propeller. --Mr. B. Dickinson's new propeller. --Its form and peculiarities and results. --4 illustrations. 10333

IX. PHOTOGRAPHY. --Manufacture of Photographic Sensitive Plates. --Description of a factory recently erected for manufacturing dry plates. --The arrangement of rooms, machinery, and process. --10 illustrations. 10336

X. TECHNOLOGY. --Cotton Seed Oil. --How cotton seed oil is made, and the cost and profits of the operation. 10335

 Improved Dobby. --An improved weaving apparatus described and illustrated. --1 Illustration. 10333

 Sulphur Mines in Sicily. --By Philip Carroll, U. S. Consul, Florence. --How sulphur is made in Sicily, percentage, composition of the ore, and full details. 10334

 The Use of Ammonia as a Refrigerating Agent. --By Mr. T. B. Lightfoot, M. I. C. E. --An elaborate discussion of the theory and practice of ammonia refrigerating, including the hydrous and anhydrous systems, with conditions of economy. 10337

 * * * * *

INFLUENCE MACHINES. [1]

 [Footnote 1: Lecture delivered at the Royal Institution, April 27, 1888. For the above and for our illustrations we are indebted to _Engineering_. ]

BY MR. JAMES WIMSHURST. 

I have the honor this evening of addressing a few remarks to you uponthe subject of influence machines, and the manner in which I proposeto treat the subject is to state as shortly as possible, first, thehistorical portion, and afterward to point out the prominentcharacteristics of the later and the more commonly known machines. Thediagrams upon the screen will assist the eye to the general form ofthe typical machines, but I fear that want of time will prevent mefrom explaining each of them. 

In 1762 Wilcke described a simple apparatus which produced electricalcharges by influence, or induction, and following this the greatItalian scientist Alexander Volta in 1775 gave the electrophorus theform which it retains to the present day. This apparatus may be viewedas containing the germ of the principle of all influence machines yetconstructed. 

Another step in the development was the invention of the doubler byBennet in 1786. He constructed metal plates which were thicklyvarnished, and were supported by insulating handles, and which weremanipulated so as to increase a small initial charge. It may be betterfor me to here explain the process of building up an increased chargeby electrical influence, for the same principle holds in all of themany forms of influence machines. 

This Volta electrophorus, and these three blackboards, will serve forthe purpose. I first excite the electrophorus in the usual manner, andyou see that it then influences a charge in its top plate; the chargein the resinous compound is known as negative, while the chargeinduced in its top plate is known as positive. I now show you by thiselectroscope that these charges are unlike in character. Both chargesare, however, small, and Bennet used the following system to increasethem. 

Let these three boards represent Bennet's three plates. To plate No. 1he imparted a positive charge, and with it he induced a negativecharge in plate No. 2. Then with plate No. 2 he induced a positivecharge in plate No. 3. He then placed the plates Nos. 1 and 3together, by which combination he had two positive charges withinpractically the same space, and with these two charges he induced adouble charge in plate No. 2. This process was continued until thedesired degree of increase was obtained. I will not go through theprocess of actually building up a charge by such means, for it wouldtake more time than I can spare. 

In 1787 Carvallo discovered the very important fact that metal plateswhen insulated always acquire slight charges of electricity; followingup those two important discoveries of Bennet and Carvallo, Nicholsonin 1788 constructed an apparatus having two disks of metal insulatedand fixed in the same plane. Then by means of a spindle and handle, athird disk, also insulated, was made to revolve near to the two fixeddisks, metallic touches being fixed in suitable positions. With thisapparatus be found that small residual charges might readily beincreased. It is in this simple apparatus that we have the parent ofinfluence machines (see Fig. 1), and as it is now a hundred yearssince Nicholson described this machine in the Phil. Trans. , I think itwell worth showing a large sized Nicholson machine at work to-night(see Fig. 11, above). 

[Illustration: Figs. 1-9. ]

In 1823 Ronalds described a machine in which the moving disk wasattached to and worked by the pendulum of a clock. It was amodification of Nicholson's doubler, and he used it to supplyelectricity for telegraph working. For some years after these machineswere invented no important advance appears to have been made, and Ithink this may be attributed to the great discoveries in galvanicelectricity which were made about the commencement of this century byGalvani and Volta, followed in 1831 to 1857 by the magnificentdiscoveries of Faraday in electro-magnetism, electro-chemistry, andelectro-optics, and no real improvement was made in influence machinestill 1860, in which year Varley patented a form of machine shown inFig. 2. It also was designed for telegraph working. 

In 1865 the subject was taken up with vigor in Germany by Toepler, Holtz, and other eminent men. The most prominent of the machines madeby them are figured in the diagrams (Figs. 3 to 6), but time will notadmit of my giving an explanation of the many points of interest inthem; it being my wish to show you at work such of the machines as Imay be able, and to make some observations upon them. 

In 1866 Bertsch invented a machine, but not of the multiplying type;and in 1867 Sir William Thomson invented the form of machine shown inFig. 7, which, for the purpose of maintaining a constant potential ina Leyden jar, is exceedingly useful. 

The Carre machine was invented in 1868, and in 1880 the Voss machinewas introduced, since which time the latter has found a place in manylaboratories. It closely resembles the Varley machine in appearance, and the Toepler machine in construction. 

In condensing this part of my subject, I have had to omit manyprominent names and much interesting subject matter, but I must statethat in placing what I have before you, many of my scientific friendshave been ready to help and to contribute, and, as an instance ofthis, I may mention that Prof. Sylvanus P. Thompson at once placed allhis literature and even his private notes of reference at my service. 

I will now endeavor to point out the more prominent features of theinfluence machines which I have present, and, in doing so, I must aska moment's leave from the subject of my lecture to show you a smallmachine made by that eminent worker Faraday, which, apart from itsvalue as his handiwork, so closely brings us face to face with theimperfect apparatus with which he and others of his day made theirvaluable researches. 

The next machine which I take is a Holtz. It has one plate revolving, the second plate being fixed. The fixed plate, as you see, is so muchcut away that it is very liable to breakage. Paper inductors are fixedupon the back of it, while opposite the inductors, and in front of therevolving plate, are combs. To work the machine (1) a specially dryatmosphere is required; (2) an initial charge is necessary; (3) whenat work the amount of electricity passing through the terminals isgreat; (4) the direction of the current is apt to reverse; (5) whenthe terminals are opened beyond the sparking distance, the excitementrapidly dies away; (6) it does not part with free electricity fromeither of the terminals singly. 

It has no metal on the revolving plates, nor any metal contacts; theelectricity is collected by combs which take the place of brushes, andit is the break in the connection of this circuit which supplies acurrent for external use. On this point I cannot do better than quotean extract from page 339 of Sir William Thomson's "Papers onElectrostatics and Magnetism, " which runs: "Holtz's now celebratedelectric machine, which is closely analogous in principle to Varley'sof 1860, is, I believe, a descendant of Nicholson's. Its great powerdepends upon the abolition by Holtz of metallic carriers and metallicmake-and-break-contacts. It differs from Varley's and mine by leavingthe inductors to themselves, and using the current in the connectingarc. "

In respect to the second form of Holtz machine (Fig. 4) I have verylittle information, for since it was brought to my notice nearly sixyears ago I have not been able to find either one of the machines orany person who had seen one. As will be seen by the diagram, it hastwo disks revolving in opposite directions, it has no metal sectorsand no metal contacts. The "connecting arc circuit" is used for theterminal circuit. Altogether I can very well understand and fullyappreciate the statement made by Professor Holtz in _Uppenborn'sJournal_ of May, 1881, wherein he writes that "for the purpose ofdemonstration I would rather be without such machines. "

The first type of Holtz machine has now in many instances been made upin multiple form, within suitably constructed glass cases, but when somade up, great difficulty has been found in keeping each of the manyplates to a like excitement. When differently excited, the one set ofplates furnished positive electricity to the comb, while the next setof plates gave negative electricity; as a consequence, no electricitypassed the terminal. 

To overcome this objection, to dispense with the dangerously cutplates, and also to better neutralize the revolving plate, throughoutits whole diameter, I made a large machine having twelve disks 2 ft. 7in. In diameter, and in it I inserted plain rectangular slips of glassbetween the disks, which might readily be removed; these slips carriedthe paper inductors. To keep all the paper inductors on one side ofthe machine to a like excitement, I connected them together by a metalwire. The machine so made worked splendidly, and your late president, Mr. Spottiswoode, sent on two occasions to take note of my successfulmodifications. The machine is now ten years old, but still worksperfectly. I will show you a smaller sized one at work. 

The next machine for observations is the Carre (Fig. 8). It consistsessentially or a disk of glass which is free to revolve without touchor friction. At one end of a diameter it moves near to the excitedplate of a frictional machine, while at the opposite end of thediameter is a strip of insulting material, opposite which, and alsoopposite the excited amalgam plate, are combs for conducting theinduced charges, and to which the terminals are metallicallyconnected; the machine works well in ordinary atmosphere, andcertainly is in many ways to be preferred to the simple frictionalmachine. In my experiments with it I found that the quantity ofelectricity might be more than doubled by adding a segment of glassbetween the amalgam cushions and the revolving plate. The current inthis type of machine is constant. 

The Voss machine has one fixed plate and one revolving plate. Upon thefixed plate are two inductors, while on the revolving plate are sixcircular carriers. Two brushes receive the first portions of theinduced charges from the carriers, which portions are conveyed to theinductors. The combs collect the remaining portion of the inducedcharge for use as an outer circuit, while the metal rod with its twobrushes neutralizes the plate surface in a line of its diagonaldiameter. When at work it supplies a considerable amount ofelectricity. It is self-exciting in ordinary dry atmosphere. It freelyparts with its electricity from either terminal, but when so used thecurrent frequently changes its direction, hence there is no certaintythat a full charge has been obtained, nor whether the charge is ofpositive or negative electricity. 

I next come to the type of machine with which I am more closelyassociated, and I may preface my remarks by adding that the inventionsprang solely from my experience gained by constantly using andexperimenting with the many electrical machines which I possessed. Itwas from these I formed a working hypothesis which led me to make myfirst small machine. It excited itself when new with the firstrevolution. It so fully satisfied me with its performance that I hadfour others made, the first of which I presented to this Institution. Its construction is of a simple character. The two disks of glassrevolve near to each other and in opposite directions. Each diskcarries metallic sectors; each disk has its two brushes supported bymetal rods, the rods to the two plates forming an angle of 90 deg. With each other. The external circuit is independent of the brushes, and is formed by the combs and terminals. 

[Illustration: Fig. 10. ]

The machine is self-exciting under all conditions of atmosphere, owingprobably to each plate being influenced by and influencing in turn itsneighbor, hence there is the minimum surface for leakage. Whenexcited, the direction of the current never changes; this circumstanceis due, probably, to the circuit of the metallic sectors and the makeand break contacts always being closed, while the combs and theexternal circuit are supplemental, and for external use only. Thequantity of electricity is very large and the potential high. Whensuitably arranged, the length of spark produced is equal to nearly theradius of the disk. I have made them from 2 in. To 7 ft. In diameter, with equally satisfactory results. The diagram, Fig. 9, shows thedistribution of the electricity upon the plate surfaces when themachine is fully excited. The inner circle of signs corresponds withthe electricity upon the front surface of the disk. The two circles ofsigns between the two black rings refer to the electricity between thedisks, while the outer circle of signs corresponds with theelectricity upon the outer surface of the back disk. The diagram isthe result of experiments which I cannot very well repeat here thisevening, but in support of the distribution shown on the diagram, Iwill show you two disks at work made of a flexible material, whichwhen driven in one direction close together at the top and the bottom, while in the horizontal diameter they are repelled. When driven in thereverse direction, the opposite action takes place. 

I have also experimented with the cylindrical form of the machine (seeFig. 10). The first of these I made in 1882, and it is before you. Thecylinder gives inferior results to the simple disks, and is morecomplicated to adjust. You notice I neither use nor recommendvulcanite, and it is perhaps well to caution my hearers against theuse of that material for the purpose, for it warps with age, and whenleft in the daylight it changes and becomes useless. 

[Illustration: Figs. 11 & 12. ]

I have now only to speak of the larger machines. They are in allrespects made up with the same plates, sectors, and brushes as wereused by me in the first experimental machines, but for conveniencesake they are fitted in numbers within a glass case. One machine haseight plates of 2 ft. 4 in. Diameter; it has been in the possession ofthe Institution for about three years. A second, which has been madefor this lecture, has twelve disks, each 2 ft. 6 in. In diameter. Thelength of spark from it is 13-5/8 in. (see Fig. 12). During theconstruction of the machine every care was taken to avoid electricalexcitement in any of its parts, and after its completion severalfriends were present to witness the fitting of the brushes and thefirst start. When all was ready the terminals were connected to anelectroscope, and the handle was moved so slowly that it occupiedthirty seconds in moving one-half revolution, and at that pointviolent excitement appeared. 

The machine has now been standing with its handle secured for abouteight hours. No excitement is apparent, but still it may not beabsolutely inert. Of this each one present must judge, but I willconnect it with this electroscope (Figs. 13 and 14), and then move thehandle slowly, so that you may see when the excitement commences andjudge of its absolutely reliable behavior as an instrument for publicdemonstration. I may say that I have never, under any condition, foundthis type of machine to fail in its performance. 

[Illustration: Fig. 13. ]

[Illustration: Fig. 14. ]

I now propose to show you the beautiful appearances of the discharge, and then, in order that you may judge of the relative capabilities ofeach of these three machines, we will work them all at the same time. 

The large frictional machine which is in use for this comparison is sowell known by you that a better standard could not be desired. 

In conclusion, I may be permitted to say that it is fortunate I hadnot read the opinions of Sir William Thomson and Professor Holtz, asquoted in the earlier part of my lecture, previous to my own practicalexperiments. For had I read such opinions from such authorities, Ishould probably have accepted them without putting them to practicaltest. As the matter stands, I have done those things which they said Iought not to have done, and I have left undone those which they said Iought to have done, and by so doing I think you must freely admit thatI have produced an electric generating machine of great power, andhave placed in the hands of the physicist, for the purposes of publicdemonstration or original research, an instrument more reliable thananything hitherto produced. 

 * * * * *

VIOLET COPYING INK. --Dissolve 40 parts of extract of logwood, 5 ofoxalic acid and 30 parts of sulphate of aluminium, without heat, in800 parts of distilled water and 10 parts of glycerine; let standtwenty-four hours, then add a solution of 5 parts of bichromate ofpotassium in 100 parts of distilled water, and again set aside fortwenty-four hours. Now raise the mixture once to boiling in a brightcopper boiler, mix with it, while hot, 50 parts of wood vinegar, andwhen cold put into bottles. After a fortnight decant it from thesediment. In thin layers this ink is reddish violet; it writes darkviolet and furnishes bluish violet copies. 

 * * * * *

SIBLEY COLLEGE LECTURES. --1887-88. 

BY THE CORNELL UNIVERSITY NON-RESIDENT LECTURERS IN MECHANICALENGINEERING. 

THE EVOLUTION OF THE MODERN MILL. [1]

 [Footnote 1: The lecture was illustrated by about fifty views on the screen, which cannot be reproduced here, showing photographs of mills and mechanical drawings of the methods of construction alluded to in the lecture. ]

BY C. J. H. WOODBURY, BOSTON, MASS. 

The great factories of the textile industries in this country arefashioned after methods peculiarly adapted to the purposes for whichthey are designed, particularly as regards the most convenient placingof machinery, the distribution of power, the relation of the severalprocesses to each other in the natural sequence of manufacture, andthe arrangement of windows securing the most favorable lighting. Thefloors and roofs embody the most economical distribution of material, and the walls furnish examples of well known forms of masonryoriginating with this class of buildings. 

These features of construction have not been produced by a stroke ofgenius on the part of any one man. There has been no Michael Angelo, no Sir Christopher Wren, whose epitaph bids the reader to look aroundfor a monument; but the whole has been a matter of slow, steadygrowth, advancing by hair's breadth; and, as the result of continualefforts to adapt means to ends, an inorganic evolution has beeneffected, resulting in the survival of the fittest, and literallypushing the weaker to the wall. 

This advance in methods has, like all inventions, resulted in theimpairment of invested capital. There are hundreds of mill buildings, the wonder of their day, now used for storage because they cannot beemployed to sufficient advantage in manufacturing purposes to competewith the facilities furnished by mills of later design. Thus theirowners have been compelled to erect new buildings, and, as far as theoriginal purpose of manufacturing is concerned, to abandon their oldmills. 

In the case of a certain cotton mill built about thirty years ago, andused for the manufacture of colored goods of fancy weave, the ownersadded to the plant by constructing a one story mill, which proved tobe peculiarly adapted to this kind of manufacture, by reason of addedstability, better light, and increased facilities for transferring thestock in process of manufacture; and they soon learned not only thatthe old mill could not compete with the new one, but that they couldnot afford to run it at any price; the annual saving in the cost ofgas, as measured by the identical meter used to measure the supply tothe old mill, being six per cent. On the cost of the new mill. 

In another instance, one of two cordage mills burned, and a new millof one story construction was erected in its place. The advantage ofmanufacture therein was so great that the owners of the propertychanged the remaining old mill into a storehouse; and now, as theywish to increase their business, it is to be torn down as a cumbererof the ground, to make room for a building of similar construction tothe new mill. 

It is true that such instances pertain more particularly to industriesand lines of manufacture where competition is close and conditions areexacting. Still they apply in a greater or less degree to nearly everyindustrial process in which a considerable portion of the expense ofmanufacture consists in the application of organized labor to machinesof a high degree of perfection. 

These changes have been solely due to the differences in theconditions imposed by improvement in the methods of manufacture. Theearly mills of this country were driven by water power, and situatedwhere that could be developed in the easiest manner. They weretherefore placed in the narrow valleys of rapid watercourses. Themethod of applying water power in that day being strictly limited toplacing the overshot or breast wheel in the race leading from thecanal to the river, the mill was necessarily placed on a narrow stripof land between these two bodies of water, with the race-way runningunder the mill. 

To meet these conditions of location, which was limited to this stripof land, the mill must be narrow and short, and the requisite floorarea must be obtained by adding to the number of stories. It wasessential that the roof of such a mill should be strong and wellbraced in order to sustain the excessive stress brought to bear uponit. The old factory roof was a curious structure, with eaves springingout of the edge of hollow cornices, the roof rising sharply untilabout six feet above the attic floor, with an upright course of aboutthree feet, filled with sashes reaching to a second roof, which, at amore moderate pitch than the first slope, trended to the ridge. 

The attic was reduced to an approximately square room, by placingsheathing between the columns underneath the sashes, and ceilingunderneath the collar beams above; thus forming a cock-loft above andconcealed spaces at the sides which diminished the practicallyavailable floor space in the attic. This cock-loft and these concealedspaces became receptacles for rubbish and harbors for vermin, both ofwhich were frequent causes of fire. 

The floors of such a mill were similar in their arrangement to thoseof a dwelling. Joists connecting the beams supported the floor; andthe under side was covered over by sheathing or lath and plaster, thusforming, as in the case of the roof, hollow spaces which were a sourceof danger. This method caused at the same time an extravagantdistribution of material, by the prodigal use of lumber and theunnecessary thickness of such floors, and entailed an excessive amountof masonry in the walls. 

Mills built after this manner were frequently in odd dimensions; andthe machinery was necessarily placed in diversified arrangement, calling forth a similar degree of wasted skill as that used in makinga Chinese puzzle conform to its given boundaries. Their area dependedupon the topography of the site, and their height upon the owner'spocket book. There was in Massachusetts a mill with ten floors, builton land worth at that time ten cents or less per square foot, whichhas been torn down and a new mill rebuilt in its place, because, sincethe advent of modern mills, it has failed every owner by reason of theexcessive expenditure necessary for the distribution of power, forsupervision, and for the transfer of stock in process, in comparisonwith the mills of their competitors, built with greater ground areaand less number of stories. 

With the advent of the steam engine as prime mover in mills, and theintroduction of the turbine wheel with its trunk, affording greaterfacilities in the application of water power, the character of thesebuildings changed very materially, though still retaining many oftheir old features. One of the first of these changes may be noticedin the consideration which millwrights gave to the problem of fixingupon the dimensions of a mill so as to arrange the machinery in themost convenient manner. Although the floors were still hollow, therewas a better distribution of material, the joists being deeper, oflonger span, and resting upon the beams, thus avoiding the perniciousmethod of wasting lumber, and guarding against fracture by tenoningjoists into the upper side of beams. 

But this secondary type of mills was not honest in the matter ofdesign. The influence of architects who attempted effects notaccordant with or subservient to the practical use of the property isapparent in such mills. The most frequent of these wooden efforts atclassic architecture was the common practice of representing adiminutive Grecian temple surrounding a factory bell perched in midair. There were also windows with Romanesque arches copied fromchurches, and Mansard roofs, exiled from their true function ofdecorating the home, covering a factory without an answering lineanywhere on its flat walls. 

I do not mean to criticise any of these elements of design in theirproper place and environment; but utility is the fundamental elementin design, and should be especially noticeable in a buildingconstructed for industrial purposes, and used solely as a source ofcommercial profit in such applications. Its lines therefore fulfilltheir true function in design in such measure as they suggeststability and convenience; and this can be obtained in such structureswithout the adoption of bad proportions offensive to the taste. Infact, certain decorative effects have been made with good results; butthese have been wholly subordinate to the fundamental idea of utility. 

The endurance with which brick will withstand frost and fires, and thedisintegrating forces of nature, in addition to its resistance tocrushing and the facility of construction, constitute very importantreasons for its value for building purposes. But the use of this hasbeen too often limited to plain brick in plain walls, whose monotonyportrayed no artistic effect beyond that furnished by a fewgeometrical designs of the most primitive form of ornament, andfalling far short of what the practice of recent years has shown to bepossible with this material. 

Additions of cast iron serve as ornaments only in the phraseology oftrade catalogues; and the mixture of stone with brick shows results inflaring contrasts, producing harsh dissonance in the effect. Thefacades of such buildings show that this is brick, this is stone, orthis is cast iron; but they always fail to impress the beholder withthe idea of harmonious design. The use of finer varieties of clay interra cotta figures laid among the brickwork furnishes a field ofarchitectural design hardly appreciated. The heavy mass of brick, divided by regular lines of demarkation, serves as the groundwork ofsuch ornamentation, while the suitable introduction in the properplaces of the same material in terra cotta imparts the mostappropriate elements of beauty in design; for clay in both forms showsalike its capacity for utility and decoration. The absorption of lightby both forms of this material abates reflection, and renders itsproportions more clearly visible than any other substance used inbuilding construction. 

The modern mill has been evolved out of the various exactingconditions developed in the effort to reduce the cost of production tothe lowest terms. These conditions comprise in a great measurequestions of stability, repairs, insurance, distribution of power, andarrangement of machinery. 

In presenting to your attention some of the salient features of modernmill construction, I do not assume to offer a general treatise uponthe subject; but propose to confine myself to a consideration of sometopics which may not have been brought to your notice, as they arestill largely matters of personal experience which have not yet foundtheir way into the books on the subject. Much of this, especially thedrawings thrown on the screen, is obtained from the experience of themanufacturers' mutual insurance companies, with which I am connected. By way of explanation, I will say that these companies confine theirwork to writing upon industrial property; and there is not amechanical process, or method of building, or use of raw material, which does not have its relation to the question of hazard by fire, byreason of the elements of relative danger which it embodies. 

It is indeed fortunate that it has been found by experience that thosemethods of building which are most desirable for the underwriter arealso equally advantageous for the manufacturer. There is no pretensemade at demands to compass the erection of fireproof buildings. Infact, as I have once remarked, a fireproof mill is commerciallyimpossible, whatever effort may be made to overcome the constructivedifficulties in the way of erecting and operating a mill which shallbe all that the name implies. The present practice is to build a millof slow burning construction. 

FOUNDATIONS. 

In considering the elements of such buildings, I wish to devote a fewwords to the question of foundations, because in the excessive loadsimposed by this class of buildings, and in the frequent necessity ofconstructing them upon sites where alluvial drift or quicksands formcompressible foundations, there is afforded an opportunity for thewidest range of engineering skill in dealing with the problem. In suchinstances, a settling of the building must be foreseen and providedfor, in order that it may be uniform under the whole structure. Thisis generally accomplished by means of independent foundations underthe various points of pressure, arranged so as to give a uniformintensity of pressure upon all parts of the foundation. It isconsidered important to limit the load upon such foundations to twotons a square foot, although loads frequently exceed this amount. 

There is a large building in New York City which has recently beenreconstructed, and the foundations rearranged, where the load reachedto the enormous amount of six to ten tons per square foot. It was afrequent occurrence in the class of high mills spoken of to imposeloads of so much greater intensity upon the wall foundation than uponthe piers under the columns of the mill, that the floors became muchlower at the walls than at the middle. 

The stone for such foundations should be laid in cement rather than inmortar, not merely because cement offers so much greater resistance tocrushing, but because its setting is due to chemical changes occurringsimultaneously throughout the mass. The hardening of mortar, on theother hand, is due to the drying out of the water mechanicallycontained with it, and its final setting is caused by the action ofthe carbonic acid gas in the air. 

Although quicksands are never to be desired, yet they will sustainheavy loads if suitably confined. When inclined rock strata are metwith, all horizontal components of stress should be removed by cuttingsteps so that the foundation stones shall lie upon horizontal beds. 

Foundations are frequently impaired by the slow, insidious action ofsprings or of water percolating from the canal which supplies thewater power for the mill; and the proper diversion of such streamsshould be carefully provided for. 

In the question of foundations, there is much of a general naturewhich is applicable to all structures; but, at the same time, eachcase requires independent consideration of the circumstances involved. 

WALLS. 

In addition to what has been said, there is but little for me to offeron the subject of walls beyond the general question of stability. Inmill construction, walls of uniform thickness have been displaced bypilastered walls, about sixteen inches thick at the upper story, andincreasing four inches in thickness with each story below. 

The remainder of the walls is from four to six inches less inthickness than at the pilasters. Frequently the outside dimensions ofthese pilasters are somewhat increased, giving greater stability andartistic effect. By leaving hollow flues within them, and using theseflues as conductors for heated air which may be forced in by a blower, such pilasters afford a means for the most efficient method of warmingthe building. 

Consideration must be given to the contraction of brick masonry, especially when an extension or addition is to be made to an olderbuilding. This shrinkage amounts to about three-sixteenths of an inchto the rod, an item which is of considerable importance in the floorsof high buildings, where the aggregate difference is very appreciable. Some degree of annoyance is caused by neglect to consider this elementof shrinkage in reference to the window and door frames, which shouldhave a slight space above them allowing for such contraction. Thiscontraction is often the source of serious trouble in brick buildingswith stone faces, the shrinkage of the brick imposing excessive stresson the stone. Instances of this are quite frequent, especially inlarge public buildings, notably the capitol at Hartford and the publicbuilding at Philadelphia, where the shivering of the joints of thestone work gave undue alarm, on the general assumption that itindicated a dangerous structural weakness. The difficulty has, Ibelieve, been entirely remedied in both cases. 

The limit of good practice on loads upon brickwork is eight to tentons per square foot, although it is true that these loads are largelyexceeded at times. It is not to be shown, however, that the limits ofsafety in regard to desirable construction should be confined to theuse of masonry for any low buildings. Structures which may be said tobe equal to those of brickwork, as far as commercial risk isconcerned, can be built wholly or in part of wood so as to conform toall practical conditions of safety. This statement does not applyexcept to low buildings of one or possibly two stories in height, where the timber cannot be subjected to the intense blast of flameoccurring when a high building is on fire. 

Mr. George H. Corliss, the eminent engine builder, of Providence, first built a one-story machine shop, with brick walls extending onlyto the base of the windows, above this the windows being very closetogether, with solid timber construction between them. 

Another method is to place upright posts reaching from the sill to theroof timbers, and to lay three-inch plank on the outside of such postsup to the line of the windows. A sheathing on the outside plankbetween the timbers is laid vertically and fastened to horizontalfurring strips. In some instances a small amount of mortar is placedover each of the furring strips. The reason for this arrangement is toprevent the formation of vertical flues, which are such a potentfactor in the extension of fires. 

WINDOWS. 

Light is often limited or misapplied on account of faulty position orsize of windows. The use of pilastered walls permits the introductionof larger windows, which are in most instances virtually doublewindows, the two pairs of sashes being set in one frame separated by amullion. A more recent arrangement, widely adopted in Englishpractice, is to place a swinging sash at the top of the window, whichcan be opened, when necessary, to assist in the ventilation, while themain sashes of the window are permanently fixed. 

Rough plate glass is used in such windows, because it gives a softerand more diffused light, which is preferred to that from ordinaryclear glass. White glass may be rendered translucent by a coat ofwhite zinc and turpentine. 

The top of a window should be as near the ceiling as practicable, because light entering the upper portion of a room illuminates it moreevenly, and with less sharply marked shadows, than where the windowsare lower down. 

The walls below the windows should be sloped, in order that there maybe no opportunity to use them as a resting place for material whichshould be placed elsewhere. 

FIRE WALLS. 

Brick division walls should be built so as to constitute a fire wallwherever it is practicable to do so. Such walls should project atleast three feet above the roof, and should be capped by stone, terracotta, or sheet metal. They must form a complete cut-off of allcombustible material, especially at the cornices. 

FIRE DOORS. 

All openings in such walls must be provided with such fireproof doorsas will prove reliable in time of need. Experience with iron doors ofvarious forms of construction show that they have been utterlyunreliable in resisting the heat of even a small fire. They will warpand buckle so as to open the passageway and allow the fire to passthrough the doorway into the next room. 

A door made of wood, completely enveloped by sheets of tinned iron, and strongly fastened to the wall, has proved to resist fire betterthan any door which can be applied to general use. I have seen suchdoors in division walls where they had successfully resisted the flamewhich destroyed four stories of a building filled with combustiblematerial, without imposing any injury upon the door except the removalof the tin on the sheet iron; and the doors were kept in furtherservice without any repairs other than a coat of paint. 

The reason for this resistance to fire is that the wood, being a poorconductor of heat, will not warp and buckle under heat, and cannotburn for lack of air to support combustion. A removal of the sheetmetal on such a door after a fire in a mill shows that the surface ofthe wood is carbonized, not burned, reduced to charcoal, but not toashes. 

Many fire doors are constructed and hung in such a manner that it isdoubtful whether they could withstand a fire serious enough to requiretheir services. 

The door should be made of two thicknesses of matched pine boards ofwell dried stock, and thoroughly fastened with clinched nails. Itshould be covered with heavy tin, secured by hanging strips, and thesheets lock-jointed to each other, with the edge sheets wrappingaround, so that no seam will be left on the edge. 

Sliding doors are preferable to swinging doors for many reasons, especially because they cannot be interfered with by objects on thefloor. But, if swinging doors are used, care should be taken that thehinges and latches are very strong, and securely fastened directly tothe walls, and not to furring or anything in turn attached to thewalls. The portion of the fixtures attached to the doors must befastened by carriage bolts, and not by wood screws. 

Sliding on trucks is the preferable method of hanging sliding doors, inclined two and one half inches to the foot, and bolted to the wall. The trucks should be heavy "barn door hangers, " bolted to the door;and a grooved door jamb, of wood, covered with tin similar to thedoor, should receive it when shut. A step of wood will hold the dooragainst the wall when closed. A threshold in the doorway retards firefrom passing under the door, and also prevents the flow of water fromone room to another. 

These doors are usually placed in pairs, and sometimes an automaticsprinkler is placed between them. 

Fire doors should always be closed at night. In some well orderedestablishments there is a printed notice over each door directing thenight watchmen to close such doors after them. In a storage warehousein Boston, the fire doors are connected with the watchman's electricclock system, so that all openings of fire doors are matters of recordon the dial sheet. 

Fire doors should certainly be closed at times of fire; yet, that suchdoors are open at night fires, or left open by fleeing help at dayfires, is an old story with underwriters. A simple automatic devicecan be used to shut such doors. It consists of two round pieces ofwood with a scarfed joint held by a ferrule, forming a strut which isplaced on two pins, keeping the door open, as other sticks have longsince served like purposes. 

The peculiarity of this arrangement is that the ferrule is nothomogeneous, but is made up of four segments of brass solderedtogether with the alloy fusible at 163 degrees Fahr. , which is widelyknown for its use in automatic sprinklers. When the solder yields, therod cripples, and the door rolls down the inclined rail and shuts. Atany time the door can be closed by removing one end of the rod fromone of the pins and allowing it to hang from the other pin. 

MILL TOWERS. 

Because of economic reasons for preserving the space within the wallsof the mill so that it may be to the greatest extent available for thebest arrangement of machinery, the stairways should be placed outsideof the building. Such stairways should not be spiral stairways, butshould be made in short straight runs with square landings, because inthe spiral stairway the portion of the stairs near the center is of somuch steeper pitch that it renders them dangerous when the help arecrowding out of the mill. 

The wear of stairs from the tread of many feet presents a difficultproblem. A very common practice consists in covering each tread with athin piece of cast iron marked with diagonal scores, and generallyshowing the name of the mill. These treads wear out in the course oftime, but for this use they answer very well, although somewhatslippery. 

A wood tread gives a more secure foothold upon the stairway; and insome instances stairs have been protected by covering the treads withboards of hard wood, containing grooves about three-eighths of an inchdeep, and of similar width, with a space of half an inch between them. These boards are grooved on both sides and placed on the stairs. Afterthe front edge is worn, they are turned around so as to present theother edge to the front, and, in course of time, turned from theexposed side to do service in two positions on the other side. In thismanner these tread covers are exposed to wear in four differentpositions. 

Mill towers, besides containing the stairways, also serve otherpurposes, as for cloak rooms for the help. They often contain a partof the fire protective apparatus, carrying standpipes with hydrants ateach floor. For this use they are easily available, and furnish a lineof retreat in case a fire spreads to an extent beyond the ability ofthe apparatus to cope with it. These towers also furnish an excellentfoundation for the elevated tank necessary for the supply of water forthe fire apparatus in places unprovided with an elevated reservoir. 

In view of the terrible and deplorable accidents which have occurredby reason of lack of proper stairway facilities at panics caused intime of fire, I would repeat the words of the late Amos D. Lockwood, the most eminent mill engineer which this country has yet produced, when he said to the New England Cotton Manufacturers' Association, "You have no moral right to build a mill employing a large number ofhelp, with only one tower containing the stairways for exit. "

The statute laws of several of the States require fire escapes; but itis a matter of fact that they are rarely used, because people are notoften cool enough to avail themselves of that opportunity of escape. Iknow of one instance where a number of girls jumped out of a fourthstory window, because they did not think of the stairways, and did notdare to use the fire escape. In that instance, none of the groupreferred to tried to go down the stairs, which did furnish a perfectlysafe means of exit to a number of others. 

Most of the fire escapes are put up so as to conform to the letter ofthe law; and in such manner that no one but a sailor or an acrobatwould be likely to trust himself to them. In crowded city buildings, and in other places where the ordinary means of escape are not induplicate, it is essential that fire escapes should be provided; butit is a great deal better to make a mill building so that they shallnot be necessary as a matter of fact, even if they are put up toconform to the requirements of statute law. 

REAR TOWERS. 

In addition to stairways, towers are placed at the rear of the mill, for the purpose of accommodating the elevators and sanitaryarrangements. It is not desirable that elevators should be boxed orsurrounded with anything that would result in the construction of aflue; but it is preferable that they pass directly through the floors, with the openings protected by automatic hatchways which closewhenever the elevator car is absent. In the washroom, etc. , in thesetowers, it is desirable to protect the wood floors by means of a thinlayer of asphalt. 

BASEMENT FLOORS. 

There are difficulties connected with the floors on or near theground, by reason of the dry rot incident to such places. Dry rotconsists in the development of fungus growth from spores existing inthe wood, and waiting only the proper conditions for theirgermination. The best condition for this germination is the exposureto a slight degree of warmth and dampness. There have been manymethods of applying antiseptic processes for the preservation of wood;but, irrespective of their varying degrees of merit, they have notcome into general use on account of their cost, odor, and solubilityin water. 

It is necessary that wood should be freely exposed to circulation ofair, in order to preserve it under the ordinary conditions met with inbuildings. Whenever wood is sealed up in any way by paint or varnish, unless absolutely seasoned, and in a condition not found in heavymerchantable timber, dry rot is almost sure to ensue. Whitewash isbetter. 

There has recently been an instance of a very large building in NewYork proving unsafe by reason of the dry rot generated in timberswhich have been completely sealed up by application of plaster ofParis outside of the wire lath and plaster originally adopted as aprotection against fire. Wire lath and plaster is one of the bestmethods of protecting timber against fire; and, if the outside is notsealed by a plaster of stucco or some other impermeable substance, themortar will afford sufficient facilities for ventilation to preventthe deposition of moisture, which will in turn generate dry rot. 

Where beams pass into walls, ventilation should be assured by placinga board each side of the beam while the walls are being built up, andafterward withdrawing it. In the form of hollow walls referred to, itis a common practice to run the end of the beam into the flue thusformed, in order to secure ventilation. 

I am well acquainted with a large mill property, one building of whichwas erected a short time before the failure of the corporation, whichresulted in the whole plant remaining idle several years. After thelapse of about five years this establishment was again put intooperation; but before the new mill could be safely filled withmachinery, it was necessary to remove all the beams which enteredwalls and to substitute for them new ones, because the ends were sothoroughly rotted that it would have been dangerous to impose anyfurther loads upon the floors. When floors are within a few feet ofthe ground, unless the site be remarkably dry, it is essential toprovide for a circulation of air, which can be done very feasibly in atextile mill by laying drain pipe through the upper part of theunderpinning, forming a number of holes leading into this space, andthen making a flue from this space to the picker room or any otherplace requiring a large amount of air. The fans of the picker room, drawing their supply from underneath the building, produce acirculation of air which keeps the timber in good condition. 

It is supposed by some that there is a difference in the quality oftimber according to the season in which it is felled, preference beinggiven to winter timber, on account of the greater amount of potash andphosphoric acid which it is said to contain at that time. In someparts of Europe it is a custom to specify that the lumber should havebeen made from rafted timber, on account of the action of the water inkilling certain species of germs. Whatever may be the merits of eitherof these two theories, the commercial lumber of the northern part ofthis country is generally felled in winter and afterward rafted. 

The action of lime in the preservation of wood has always beenattended with the most excellent results; although not suited toplaces subject to the action of water, which dissolves the lime, leaving the timber practically in its original condition. Thepreservative action of lime upon wood is readily shown by theadmirable condition in which laths are always found. I doubt if anyone ever found a decayed lath in connection with plaster. 

As an example of the action of lime as a preservative of lumber. I cancite an instance of a mill in New Hampshire where the basement floorwas placed in 1856, the ledge in the cellar having been blasted outfor the purpose. The rock was very seamy, and abounded in waterissuing from springs or percolating from the canal supplying water tothe mill. The rock was blasted away to a grade two feet below thefloor, and most of the space filled up again by replacing the smallpieces of stone, so arranged as to form blind drains for the removalof any water which might find its way under the floor. 

Toward the top of this filling, finer stones were used, then aboutthree inches of gravel, which was covered with two inches of sand andlime. Two years ago I was at this mill when some alterations requiringthe removal of the floor were in progress, and found that the lumberwas still in good, sound condition, except for a superficial decay onthe under side of the floor plank. 

But there are frequent instances where it is necessary to place thefloor directly upon the earth, without any space or loose fillingunderneath it, in order to save room, or to secure a firm support formachinery. By way of information upon what has actually beenaccomplished in this direction, I will cite instances of three floorsin such positions, all of which have to my knowledge fulfilled thepurpose for which they were designed. 

The first instance is that of a basement floor laid twenty-one yearsago, a portion of which was made by excavating one foot below thefloor, six inches of coarse stone being filled in, then five inches ofcoal tar concrete made up with coarse gravel, and finally about oneinch of fine gravel concrete. Before the concrete was laid, heavystakes were driven through the floor about three feet apart, to whichthe floor timbers were nailed and leveled up. The concrete was thenfilled in upon the floor timbers, and thoroughly tamped and rolled outto the level of the top of the floor timbers. The under side of thefloor timbers was covered with hot coal tar. 

This floor is still in good condition, and has not needed repairscaused by the decay of the timber. Another portion of the floor laidat the same time and in the same manner, with the exception thatcement concrete was used in the place of the coal tar, was entirelyrotted out in ten years. 

Another floor was made in quite a similar manner. All soil and loamwas removed from the interior of the building; the whole surface wasbrought up to the grade with a puddle of gravel and ashes; stakes twoand a half by four inches, and thirty inches in length, were drivendown; and nailing strips were secured to them. Over this puddledsurface a coat of concrete eight inches thick was laid, the top beingflush with the upper surface of the nailing strips. This concrete wasmade of pebbles about two inches in diameter, well coated with coaltar, and laid in place when hot. It was then packed together by beingtamped and rolled, and a thin covering of the tarred sand placed uponthe top, forming a smooth, hard surface. The first floor consisted oftwo inches of matched spruce, grooved on both sides, and fitted withhard pine splines, five-eighths by one and one-fourth inches. On thetop of this a hard pine 1¼ inch floor was laid over a course ofbuilding paper. 

Another method, which is certainly more novel than either of theothers, consists in supporting a floor upon a bed of resin. Theunderlying earth was removed, and replaced with spent moulding sand, leaving trenches for the floor timbers, which were placed upon brickslaid without mortar. Melted resin was poured into the space alongsideand underneath the timbers. The floor planks were then laid upon thetimbers, the tops of which were about half an inch above the level ofthe sand. Holes were bored into the floor plank about four feet apart, and melted resin then poured into the holes, so as to interpose alayer of resin underneath the floor plank and beams. Upon this floor atop floor of hard wood was laid in the usual manner. This floor hasbeen used for a number of years to support a large quantity of heavymachine tools, principally planers, without yielding or depreciationdue to decay, and has proved to be most satisfactory. 

In some instances asphaltum or coal tar concrete floors are notcovered with wood, although it is much more agreeable for the help tostand upon wooden floors. It should be remembered that all thesecompounds are readily softened by means of oil, and they should beprotected from oil by a coat of paint when not covered with wood; thepreferable method being to first apply a priming containing verylittle oil, or a coat of shellac, and follow with some paint mixed upwith boiled linseed oil. 

(_To be continued. _)

 * * * * *

THE MECHANICAL EQUIVALENT OF HEAT. 

BY DE VOLSON WOOD, PROFESSOR OF ENGINEERING IN STEVENS INSTITUTE OFTECHNOLOGY. 

It is clearly intimated by Mr. Hanssen, in his determination of themechanical equivalent of heat, published in the Scientific AmericanSupplement, No. 642, April 21, 1888, that his object is to determinethe _absolute_ value of this constant. With his data he finds it to be771. 89 foot pounds. But the determination by direct experiment gives alarger value. Thus, the most reliable experiments--those of Joule andRowland--give values exceeding by several units that found by Hanssen. A committee of the British Association, appointed for this purpose, reported in 1876 that sixty of the most reliable of Joule'sexperiments gave the mean value 774. 1. The experiments were made withwater at a temperature of about 60° F. , according to the mercurialthermometer, and reduced to its value at the temperature of meltingice, according to the formula given by Regnault for the variation ofthe specific heat of water at varying temperature under the constantpressure of one atmosphere. According to this formula the specificheat of water increases with the temperature above the melting pointof ice, so that the equivalent would be somewhat less at 32° F. Thanat 60° F. It will be found in Regnault's _Relation des Experiences_that he experimented on water at high temperatures, but more recentlyProfessor Rowland has found that the specific heat of water is_greater_ at 40° F. Than at 60° F. , thus reversing between theselimits the law given by Regnault; the increase, as given by the mostprobable values, being, roughly, about 1/250 of its value at 60° F. The proper correction due to this cause would make the equivalent over777 foot pounds, instead of 774. 1. Professor Rowland's experiments, when reduced to the same thermometer, same temperature, and samelatitude as Joule's, agreed very nearly with those of the latter, being about 1/1000 part larger; so that the chief difference in theultimate values consists in the reductions for temperature andlatitude. The force of gravity being less for the lower latitudes, thenumber representing the mechanical equivalent will be greater for thelatter, since the unit pound mass must fall through a greater numberof feet to equal the same work; so that the equivalent will be greaterat Paris than at Manchester. Professor Rowland also found that thedegrees on the air thermometer from 40° F. Upward to above 60° F. Exceeded those on the mercurial thermometer throughout thecorresponding range, and that from 40° to 41° the degree was between1/150 and 1/200 of a degree larger on the air thermometer than on themercurial. Although this fraction is too small to be observed byordinary means, yet, if it exists, it cannot be ignored if absolutevalues are sought. Regnault employed the air thermometer in hisexperiments, while Joule used the mercurial thermometer, and ifJoule's value 774. 1 be increased by 1/200 of itself in order to reduceit from the equivalent of the degree on the mercurial thermometer tothat on the air thermometer, we get 778 foot pounds, nearly. Rowlandfound from his experiments that when reduced to the air thermometerand to the latitude of Baltimore, the equivalent was nearly 783, subject to small residual errors. 

Nearly all writers upon this subject--except Rankine--have consideredthat the mechanical equivalent of heat, in British units, was theenergy necessary to raise the temperature of one pound of water from32° F. To 33° F. , but Rankine defines it as the heat necessary toincrease the temperature of one pound of water one degree Fahrenheitfrom that of maximum density, or from 39° F. To 40° F. For ordinarypractice it is immaterial which of these definitions is used, for theerrors resulting therefrom are much less than those resulting fromordinary observations. But when the value is to be determined bydirect experiment at the standard temperature, Rankine's limits aremuch to be preferred; for it is so very difficult to determine exactvalues by observation when the substance is near the state borderingon a change of state of aggregation, as that of changing from water toice. Observations made at about 60° F. Were reduced by means ofRegnault's law for the specific heat of water, as has been stated, which is expressed by the formula

 4 9 c = 1 + ------ t + ------ t^{2} 10^{5} + 10^{7}

in which t denotes the temperature according to the Centigrade scale. According to this law, the mechanical equivalent would not be 0. 2 of afoot pound greater at 5° C. (41° F. ) than at 0° C. (32° F. ); hence, ifthis law were correct, it would make no practical difference whetherthe temperature were at 0° C. Or 5° C. This law makes the _computed_value at 32° F. About 0. 95 of a foot pound less than that determinedby experiment at 60° F. ; whereas Rowland's experiments make it_greater_ at 40° F. By more than four foot pounds, for the airthermometer. In determining a _fixed_ value to be used for scientificpurposes, it is necessary to fix the place, the thermometer, and theparticular degree on the thermometer. The place may be known by itslatitude if reduced to the level of the sea. The air thermometeragrees most nearly with that of the ideally perfect gas thermometer, while the mercurial thermometer differs very much from it in somecases. Thus, Regnault found that when the air thermometer indicated630° F. Above the melting point of ice (or 662° F. ), the mercurialthermometer indicated 651. 9° above the same point (683. 9° F. ), adifference of 22° F. It is apparent that the air thermometer furnishesthe best standard. As for the particular degree on the scale to beused for the standard, it is apparent, from the observations abovemade, that the temperature corresponding to that at or near themaximum density of water is more desirable than that at the meltingpoint of ice. The fact, also, that the specific heats at constantpressure and at constant volume are the same at the point of maximumdensity, as shown by theory, is an additional argument in favor ofselecting this point for the standard. It thus appears that thesolution of this problem, which appears simple and very definite byMr. Hanssen's method, becomes intricate and, to a limited degree, indeterminate when subjected to the refinements of direct experiment. If the constants used by Hanssen are absolutely correct, then hisresult must be unquestioned; but since physical constants are subjectto certain residual errors, one would as soon think of finding thespecific heat of air at constant volume, by using the value of themechanical equivalent as one of the elements, and trusting the result, as he would to trust to the computed value of the mechanicalequivalent without subjecting it to the test of a direct experiment. We will, therefore, examine the constants used to see if they are theexact values of the quantities they represent. 

He says they are universally accepted as correct; and this may betrue, when used for general purposes, and yet not be scientificallyexact. He uses 0. 2377 as the specific heat of air. This is the value, to four decimals, found by Regnault. Thus, Regnault gives for the meanvalue of the specific heat of air

 Between -30° C and + 10° C. 0. 23771 " 0° C " 100° C. 0. 23741 " 0° C " 200° C. 0. 23751

And we know of no reason why one of these values should be used ratherthan another, except that the mean of a large range of temperaturesmay be more nearly correct than that of any other; and if this reasondetermines our choice, the number 0. 2375 would be used instead of0. 2377. Although this difference is small, yet the former value wouldhave reduced his result about 0. 7 of a foot pound. 

Again, he uses 0. 1686 for the specific heat of air at constant volume. The value of this constant has never been found to any degree ofaccuracy by direct experiment, and we are still dependent upon themethod established by La Place and Poisson, according to which theconstant ratio of the specific heat of a gas at constant pressure tothat at constant volume is found by means of the velocity of sound inthe gas. The value of the ratio for air, as found in the days of LaPlace, was 1. 41, and we have 0. 2377 ÷ 1. 41 = 0. 1686, the value used byClausius, Hanssen, and many others. But this ratio is not definitelyknown. Rankine in his later writings used 1. 408, and Tait in a recentwork gives 1. 404, while some experiments give less than 1. 4, andothers more than 1. 41. 

An error of one foot in a thousand in determining the velocity ofsound will affect the third decimal figure one or two units. A smalldifference in the assumed weight of a cubic foot of air also affectsthe result. M. Hanssen gives 0. 080743 pound as the weight at 32° F. Under the pressure of one atmosphere; while Rankine gives 0. 080728pound. In my own computations I use 1. 406 as a more probable value ofthe constant sought. This will give for the specific heat of air atconstant pressure

 0. 2375 ÷ 1. 406 = 0. 1689

This is only 0. 0003 of a unit greater than the value used by Hanssen, but it would have given him nearly 775, instead of 771. 89. 

Again, he uses 491. 4° F. For the absolute temperature of melting ice. The exact value of this constant is unknown; but the mean value asdetermined by Joule and Thomson, in their celebrated experiments withporous plugs, was 492. 66° F. This value would slightly change hisresult. It will be seen from the above that a small change in theconstants used may affect by several units the computed value of themechanical equivalent. I have computed it, using 1. 406 for the ratioof the specific heat of air at constant pressure to that at constantvolume, 491. 13° F. As the temperature of melting ice above the zero ofthe _air_ thermometer, 26, 214 feet for the height of a homogeneousatmosphere, and 0. 2375 for the specific heat of air, and I find, bymeans of these constants, 778. If computed from the zero of theabsolute scale, 492. 66° F. , I find 777 to the nearest integer. Recently I have used 778. If the value given by Rowland, about 783according to the air thermometer at 39° F. , should prove to becorrect, it seems probable that the constant 1. 406 used above would bereduced to about 1. 403, or that the other constants must be changed bya small amount. The height of the homogeneous atmosphere used above, 26, 214 feet, is the value used by Rankine as deduced from Regnault'sfigures, and only one foot less than the value used by Sir WilliamThomson; but the figures used by Mr. Hanssen give 26, 210½ feet. 

The method above called Hanssen's is really that of Dr. Mayer (theGerman professor), who in 1842 used it for determining the mechanicalequivalent; but on account of erroneous data, the value found by himwas much too small. 

 * * * * *

ECONOMY TRIALS OF A NON-CONDENSING STEAM ENGINE--SIMPLE, COMPOUND, ANDTRIPLE. [1]

 [Footnote 1: Abstract of paper read before the Institution of Civil Engineers, March 13. ]

BY MR. P. W. WILLANS, M. I. C. E. 

The author described a series of economy trials, non-condensing, madewith one of his central valve triple expansion engines, with onecrank, having three cylinders in line. By removing one or both of theupper pistons, the engine could be easily changed into a compound orinto a simple engine at pleasure. Distinct groups of trials were thuscarried out under conditions very favorable to a satisfactorycomparison of results. 

No jackets were used, and no addition had, therefore, to be made tothe figures given for feed water consumption on that account. Most ofthe trials were conducted by the author, but check trials were made byMr. MacFarlane Gray, Prof. Kennedy, Mr. Druitt Halpin, ProfessorUnwin, and Mr. Wilson Hartnell. The work theoretically due from agiven quantity of steam at given pressure, exhausting into theatmosphere, was first considered. 

By a formula deduced from the [theta] [phi] diagram of Mr. MacFarlaneGray, which agreed in results with the less simple formulas of Rankineand Clausius, the pound weight of steam of various pressures requiredtheoretically per indicated horse power were ascertained. (See annexedtable. )

A description was then given of the main series of trials, all at fourhundred revolutions per minute, of the appliances used, and of themeans taken to insure accuracy. A few of the results were embodied inthe table. The missing quantity of feed water at cut off, which, inthe simple trials, rose from 11. 7 per cent. At 40 lb. Absolutepressure to nearly 30 per cent. At 110 lb. And at 90 lb. Was 24. 8 percent. , was at 90 lb. Only 5 per cent. In the compound trials. In thelatter, at 160 lb. , it increased to 17 per cent. , but, on repeatingthe trial with triple expansion, it fell to 5. 46 per cent. Or to 4. 43per cent. In another trial not included in the table. 

On the other hand, from the greater loss in passages, etc. , thecompound engine must always give a smaller diagram, considered withreference to the steam present at cut-off, than a simple engine, and atriple a smaller diagram than a compound engine. Nevertheless, even at80 lb. Absolute pressure, the compound engine had considerableadvantage, not only from lessened initial condensation, but fromsmaller loss from clearances, and from reducing both the amount ofleakage and the loss resulting from it. These gains became moreapparent with increasing wear. The greater surface in a compoundengine had not the injurious effect sometimes attributed to it, andthe author showed how much less the theoretical diagram was reduced bythe two small areas taken out of it in a compound engine than by thesingle large area abstracted in a simple engine. The trials completelyconfirmed the view that the compound engine owed its superiority toreduced range of temperature. At the unavoidably restricted pressuresof the triple trials, the losses due to the new set of passages, etc. , almost neutralized the saving in initial condensation, but withincreased pressure--say to 200 lb. Absolute--there would evidently beconsiderable economy. The figures of these trials showed that the lossof pressure due to passages was far greater with high than with lowpressure steam, and that pipes and passages should be proportionedwith reference to the weight of steam passing, and not for aparticular velocity merely. 

The author described a series of calorimetric tests upon a large scale(usually with over two tons of water), the results of which werestated to be very consistent. After comparing the dates of initialcondensation in cases where the density of steam, the area of exposedsurface, and the range of temperature were all variables, with othercases (1) where the density was constant and (2) where the surface wasconstant, the author concluded that, at four hundred revolutions perminute, the amount of initial condensation depended chiefly on therange of temperature in the cylinder, and not upon the density of thesteam or upon the extent of surface, and that its cause was probablythe alternate heating and cooling of a small body of water retained inthe cylinder. The effect of water, intentionally introduced into theair cushion cylinder, corroborated the author's views, and he showedhow small a quantity of water retained in the cylinder would accountfor the effects observed. At lower speeds surface might have moreinfluence. The favorable economical effect of high rotative speed, _per se_, was very apparent. 

In a trial with a compound engine, with 130 lb. Absolute pressure, themissing quantity at cut-off rose from 11. 7 per cent. At 405revolutions to 29. 66 per cent. At 130 revolutions, the consumption offeed water increasing from 20. 35 lb. To 23. 67 lb. This saving of 14per cent. Was due solely to increase of speed. Similar trials had beenmade with a simple engine. In one simple trial at slow speed themissing quantity rose to 44. 5 per cent. Of the whole feed water. 

------------------------------+-----+------------+-------------+------+-------------+-------------+------Intended mean | | | | | | | admission pressure Lb. | 40 | 90 | 110 | 130 | 150 | 160 | 170------------------------------+-----+------+-----+------+------+------+------+------+------+------+------ Simple, Compound, or Triple. | S. | S. | C. | S. | C. | C. | C. | T. | C. | T. | T. Actual mean | | | | | | | | | | | admission pressure Lb. |40. 88| 92. 65|87. 54|106. 3 |109. 3 |130. 6 |149. 9 |151. 9 |158. 5 |158. 1 |172. 5Percentage ratio of actual | | | | | | | | | | | mean pressure, referred to | | | | | | | | | | | low pressure piston, to | | | | | | | | | | | theoretical mean pressure |98. 2 |100 |91. 3 |100. 7 | 94. 8 | 94. 2 | 94. 6 | 84. 54| 95. 9 | 85. 3 | 85. 2Indicated horse power |16. 51| 31. 61|28. 14| 33. 5 | 33 | 36. 31| 38. 59| 35. 69| 39. 55| 35. 56| 38. 45------------------------------+-----+------+-----+------+------+------+------+------+------+------+------Feed water actually used per | | | | | | | | | | | indicated H. P. H. -- | | | | | | | | | | | Simple Lb. |42. 76| 26. 89| ... | 26 | ... | ... | ... | ... | ... | ... | ... Compound Lb. | ... | ... |34. 16| ... | 21. 37| 20. 35| 19. 45| ... | 19. 19| ... | ... Triple Lb. | ... | ... | ... | ... | ... | ... | ... | 19. 68| ... | 19. 19| 18. 45Steam required theoretically | | | | | | | | | | | per 1 H. P. H. Lb. |34. 67| 19. 24|19. 86| 17. 9 | 17. 65| 16. 25| 15. 23| 15. 16| 14. 87| 14. 9 | 14. 36Percentage efficiency |81. 1 | 71. 5 |82. 2 | 68. 8 | 82. 5 | 80 | 78. 3 | 77 | 77. 4 | 77. 6 | 77. 8------------------------------+-----+------+-----+------+------+------+------+------+------+------+------Percentage of feed water | | | | | | | | | | | missing at cut off in | | | | | | | | | | | high pressure cylinder | ... | ... | ... | ... | ... | ... | ... | 5. 33| ... | 6. 84| 5. 01Ditto high pressure cylinder | ... | ... | 5 | ... | 9. 5 | 11. 7 | 15. 1 | 14. 84| 17 | 12. 06| 15. 33Ditto low pressure cylinder |11. 7 | 24. 8 |15. 2 | 29. 56| 16. 25| 19. 1 | 20. 6 | 22. 12| 21. 3 | 22. 11| 24. 21Percentage of feed water | | | | | | | | | | | missing at end of stroke | | | | | | | | | | | in low pressure cylinder |10. 4 | 18. 83|14. 25| 21. 53| 16. 59| 17. 55| 20. 69| 18. 01| 19. 55| 18. 81| 19. 25------------------------------+-----+------+-----+------+------+------+------+------+------+------+------

The author compared a series of compound trials, at different powers, with 130 lb. Absolute pressure, and various ratios of expansion, witha series giving approximately the same powers at a constant ratio ofexpansion, but with varying pressures, being practically a trial ofautomatic expansion against throttling. Starting with 40 indicatedhorse power, 130 lb. Absolute pressure, four expansions, and aconsumption of 20. 75 lb. Of water, the plan of varying the expansion, as compared with throttling, showed a gain of about 7 per cent. At 30indicated horse power, but of a very small percentage when below halfpower. If the engine had an ordinary slide valve, the greaterfriction, added to irregular motion, would probably neutralize thesaving, while if the engine were one in which initial condensationassumed more usual proportions, the gain would be probably on the sideof variable pressure. Even as it was, the diagrams showed that themissing quantity became enormously large as the expansion increased. Judging only by the feed water accounted for by the indicator, theautomatic engine appeared greatly the more economical, but actualmeasurement of the feed water disproved this. The position of theautomatic engine was, however, relatively more favorable when simplethan when compound. 

In conclusion, the author referred to a trial with a condensingengine, at 170 lb. Absolute pressure, in which the feed water used was15. 1 lb. , a result evidently capable of further improvement, and to anefficiency trial of a combined central valve engine and Siemens'dynamo, made for the Admiralty, at various powers. At the highestpower the ratio of external electrical horse power to indicated horsepower in the engine was 82. 3 per cent. Taking the thermo-dynamicefficiency of the engine at 80 per cent. , that of the combinedapparatus would be nearly 66 per cent. 

 * * * * *

RAILWAY BRIDGE AT LACHINE. 

The subject of our large illustration this week is a large steelbridge carrying the Central Pacific Railway over the St. LawrenceRiver at Lachine, near Montreal. The main features of this reallymagnificent structure are the two great channel spans, each 408 feetlong. It will be noticed that the design combines, in a very ingeniousmanner, an upper and a lower deck structure, the railway track beinglaid on the top of the girders forming the side spans, and on thelower flanges of the channel spans, which are crossed by continuousgirders, 75 feet deep, over the central pier, and supported bybrackets as shown. The upper of our two engravings shows the method ofconstructing the principal spans, which were built outward from theside piers, while the work on the center pier was extended on eachside to meet. It was built at the works of the Dominion BridgeCompany, Montreal, from the design of Mr. C. Shaler Smith, thewell-known American bridge engineer. --_Engineering. _

[Illustration]

 * * * * *

IMPROVED SCREW PROPELLER. 

While the last few years have seen great advances made in the designsof steamships and of their engines, little or nothing has been done inthe way of improving the screw propeller. As a general rule it wouldappear to be taken for granted that no radical improvement could bemade in the form of the propeller, although various metals have beenintroduced in its manufacture with the view of increasing itsefficiency. For sea-going steamers, however, the shape remains thesame, the variation chiefly relating to the number of blades employed. A striking departure from ordinary practice, however, has of late beenmade by Mr. B. Dickinson, who has invented a screw propeller which, onpractical trial, has given an efficiency far in advance of theordinary screw. This new propeller we illustrate here in Figs. C andD, while Fig. A shows an ordinary propeller. The Dickinson propellerillustrated has six blades, giving a surface of 30 square feet; it isright handed, and has pitch of 15 ft. And a diameter of 10 ft. 6 in. The ordinary screw propeller shown at Fig. A is right handed and twobladed, with a pitch at the boss of 13 ft. 6 in. And at the tip of 15ft. It has a diameter of 10 ft. 9 in. And 32 square ft. Of surface. The projected area looking forward is 22 square ft. And the projectedarea looking athwartship 22. 84 square feet. The most graphic way ofillustrating the principle of Mr. Dickinson's propeller is to take atwo bladed propeller of the ordinary type as shown at Fig. A in theannexed cuts, and divide into three sections as in Fig. B, then movesection No. 1 to the line position on the shaft of No. 3, and No. 3 tothat of No. 1, No. 2 remaining stationary. The effect of thisinterchange will be that (having regard to the circle of rotation) No. 3, the rearmost section, will rotate in advance of No. 2, and No. 2 inadvance of No. 1 (see Fig. C). By this arrangement the water operatedon escapes freely astern from every blade--that from No. 1 passing inthe wake of No. 2, while that from Nos. 2 and 1 passes in the wake ofNo. 3. Fig. D represents the blades with a wider spread as practicallyused. The advantages claimed by Mr. Dickinson for his propeller, andwhich are sufficiently important to be given in detail, are:

[Illustration: Figs. A-D. ]

1. That the blades of each section, when the vessel is in motion, necessarily cut solid, undisturbed water, each blade operating uponprecisely the same quantity of water as an individual broad bladewould do, though, of course, it parts with it in one-third of thetime. 

2. That each sectional blade exerts the equivalent efficiency of thefirst or entering third portion of the breadth of an ordinarypropeller blade, and that consequently the combined sections havegreater effective power. It is now regarded by experts as anascertained fact that the after or trailing portion of the broad bladeis relatively non-effective as compared with the forward or enteringportion. 

3. When three blades are fitted, the spent water from No. 2 beingdelivered immediately in the wake of No. 3, and that from No. 1 in thewake of No. 2, has the effect of destroying or reducing to a minimumthe back draught of sections Nos. 2 and 3, No. 1 alone being subjectto this drawback. This is of greater importance than might at firstthought appear, as in cases where there are three or four bladesrevolving in one plane, the water is drawn after the retreating blade, lessening the resistance to the face of the advancing one. 

4. That by the subdivision of the blades, as arranged spirally, thewater passing through within the radius of the propeller has itsresisting capacity more thoroughly worked out than is possible withany propeller whose blades are all on the same plane. This view isconfirmed by the visibly increased rotation of the water in the wakeof the vessel. 

5. That by broadening the blades or increasing the number of sections, the diameter of the propeller may be proportionately diminishedwithout the sacrifice of engine power. This is often desirable withvessels of light draught, the complete immersion of the screw being atall times necessary to avoid waste of power. 

6. The propeller being made and fitted on the shaft in sections, allthat is necessary in case of accident is to replace the brokensection. This in many cases could be done afloat. 

7. The blades being arranged to take their water at different planes, there is the greater certainty of one or other of the sectionsoperating upon what is termed the water of friction. This isconsidered an advantage. 

8. Where it is desirable, the blades of the different sections can bemade of varying breadth or pitch. 

9. The principle of division into two or more sections applies equallyto two, three, or four bladed ordinary propellers. 

10. The adoption of this principle does not entail any alteration orenlargement of the screw space or bay as usually provided. 

11. As a consequence of the freedom and rapidity with which the wateroperated upon escapes from the narrow blades, the depression at thestern of the vessel caused by the action of the ordinary propeller isgreatly reduced. 

12. The vibration caused by this propeller is so slight as to behardly noticeable, thereby effecting a saving in the wear and tear ofthe engine and machinery. This may also be a consideration inpromoting the comfort of passengers. 

From a practical and working point of view we take Mr. Dickinson'schief claims to be, in the first place, the yielding of a greaterspeed per power employed, or an economy in obtaining an equal speed;in the second, increased, rapidity in maneuvering and stopping avessel; and in the third, a reduction of vibration. In order to putthese claims to a practical and reliable comparative test, Messrs. Weatherley, Mead & Hussey, of Saint Dunstan's Hill, London, placed atthe inventor's disposal two of their new steamers, the Herongate andthe Belle of Dunkerque. These are in every respect sister boats, andwere built in 1887 by Messrs. Short Brothers, and engined by Mr. JohnDickinson, of Sunderland. The Herongate was fitted about four monthsago with the largest propeller yet made on Mr. B. Dickinson'sprinciple, the Belle of Dunkerque having an ordinary four-bladedpropeller of the latest improved type. Every precaution was taken toplace the two vessels on the same footing for the purpose of acomparative test, which was recently carried out. Both vesselspreviously to the trial were placed on the gridiron, cleaned andpainted, their boilers opened out and scaled, their steam gaugesindependently tested, and both vessels loaded with a similar cargo ofpitch, the only difference being that the Herongate carried 11 tonsmore dead weight and had one inch more mean draught than the Belle ofDunkerque, while the former had been running continuously for ninemonths against the latter's two and a half months. On the day of thetrial the vessels were lying in the Lower Hope reach, and it wasdecided to run them over the measured mile there with equal pressureof steam. The order of running having been arranged, the Herongate gotunder way first, the Belle of Dunkerque following over the samecourse. Steaming down against tide, the Herongate is said to have comeround with remarkable ease and rapidity, and in turning on eitherhelm, whether with or against tide, to have shown a decided advantage. Equally manifest, it is stated, was the superiority shown in bringingup the vessel by reversing, when running at full speed, thusconfirming the very favorable reports previously received by theowners from their captains since the Dickinson propeller was fitted tothe Herongate. Those who were on board her state that the vibrationwas scarcely noticeable. From a statement submitted to us it is clearthat the Herongate had the turn of the scale against her in deadweight and draught, vacuum, and diagrams taken, but notwithstanding(making allowance for one faulty run due to the variations in tide)she appears to have more than held her own in the matter of speed, with a saving of 4½ and 3¼ revolutions per minute at 140 lb. And 160lb. Steam pressure respectively. This is further confirmed by theresults of a run made after the experiments were concluded, the twovessels being placed in line, and fairly started for a half hour's runover the flood with 150 lb. Steam pressure. At the expiration of thattime the Herongate was judged to be leading by at least half a length, her revolutions being 76, as against 80 in the Belle of Dunkerque. Itwas agreed by all present at these trials that the propeller hadrealized in full the three main working advantages claimed for it. This being the first Dickinson propeller fitted to a sea-going vesselof this size, it is quite within the limits of possibility that thepresent results may be improved upon in further practice. In any casewe can but regard this propeller as a distinct and original departurein marine propulsion, and we congratulate Mr. Dickinson on his presentsuccess and promising future. Messrs. Weatherley, Mead & Hussey alsodeserve credit for their discernment, and for the spirited manner inwhich they have taken up Mr. Dickinson's ingenious invention. Weunderstand that they are so satisfied with the results that theyintend having one of their larger ocean-going steamers fitted with theDickinson propeller. --_Iron_. 

 * * * * *

IMPROVED DOBBY. 

[Illustration]

At the Manchester Royal Jubilee Exhibition, Messrs. Butterworth &Dickinson, Burnley, showed Catlow's patent dobby, which is illustratedabove, as applied to a strong calico loom. This dobby is a double liftone, thus obtaining a wide shed, and the use of two lattice barrelsconnected by gearing so that they both revolve in the same direction. The jack lever is attached to the vertical levers, the top and bottomcatches being worked respectively by the two barrels, and connectedwith the ends of the levers. To each of these catches a light bladespring is attached, which insures them being sprung upon the top ofthe knife, and thereby obtaining a certain lift. A series of woodenjacks or levers are employed, so as to give a varying lift to thefront and back healds, in this way keeping the yarn in even tension, and preventing slack sheds. The healds are drawn down by means of aseries of levers adjoining one another, and worked by means of arocking bar driven from the tappet shaft. When the shed is beingformed, the jacks are pushed down until it is fully open, and the warpis thus drawn down with the same certainty as the upward movement ismade. --_Industries_. 

 * * * * *

[UNITED STATES CONSULAR REPORTS. SPECIAL ISSUE NO. 10. ]

SULPHUR MINES IN SICILY. 

BY PHILLIP CARROLL, U. S. CONSUL, PALERMO. 

Sulphur, or brimstone, is a hard, brittle substance of various colors, from brilliant yellow to dark brown, without smell when cool, of amild taste, and burns with a pale blue flame, emitting pungent andsuffocating fumes. Its specific gravity is from 1. 9 to 2. 1. 

Sulphur exists more or less in all known countries, but the island ofSicily, it is thought, is the only place where it is produced on alarge scale, and consequently that island appears to command themarket. Small quantities have been found in the north of Italy, theGrecian Archipelago, Russia, Austria, Poland, France, Spain, easternshores of Egypt, Tunis, Iceland, Brazil, Central America, and theUnited States. Large quantities are said to exist in various countriesof Asia, but it is understood to be impracticable to utilize the same, consequent upon the distance from any commercial port and the absenceof rail or other roads. 

Sulphur is of two kinds, one of which is of volcanic emanation, theother being closely allied to sedimentary rocks. The latter is foundin Sicily, on the southern and central portions of the island. MountEtna, situated in the east, seems to exert no influence in theformation of brimstone. There are various hypotheses relative to itsnatural formation. Dr. Philip Swarzenburg attributes it to theemanations of sulphur vapor expelled from metallic matter existing inthe earth, consequent upon the fire in the latter, while ProfessorsHoffman and Bischoff ascribe it to the decomposition of sulphuretedhydrogen. Hoffman believes the sulphureted hydrogen must have passedthrough the fissures of stratified rocks, but Bischoff is of opinionthat the sulphureted hydrogen must have been the result of thedecomposition of sulphate of lime in the presence of organic matter. The theory of others is that sulphur owes its origin to thecombination of lacustrine deposits with vegetable matter, and othersagain suppose that it is due to the action of the sea upon animalremains. The huge banks of rock salt often met with in the vicinity ofsulphur mines, and which in some places stretch for a distance ofseveral miles, seem to indicate that the sea has worked its way intothe subsoil. Fish and insects, which are frequently found in strata oftripoli, which lie under sulphur beds, induce the belief that lakesformerly existed in Sicily. 

The following is a list of the various strata which form part of thecrust of the earth in Sicily, according to Professor Mottura, anItalian geologist:

_Pliocene. _--Sandstone; coarse calcareous rock; marl. 

_Upper Miocene. _--Calcareous marl; gypsum, etc. ; sulphur embedded incalcareous limestone; silicious limestone; tripoli, containing fossilsof fish, insects' eggs, etc. 

_Middle Miocene. _--Sandstone containing quartz, intercalated with marlof a saltish taste. 

_Lower Miocene. _--Rock salt; blue marl, containing petroleum andbitumen; flintstone; ferruginous clay, mixed with aragonite andbituminous schists; ferruginous and silicious sandstone. 

_Eocene. _--Limestone, containing diaspores and shells. 

At times one or another of the strata disappears, while the order ofsome is slightly reversed on account of the broken state of the crust. Upon the whole, however, the above has been generally observed in thevarious mines by the author referred to. 

Sulphur mines have been operated in Sicily over three hundred years, but until the year 1820 its exportation was confined to narrow limits. At present the number of mines existing in Sicily is about threehundred, nearly two hundred of which, being operated on credit, are, it is understood, destined to an early demise. It is said that thereare about 30, 000, 000 tons of sulphur in Sicily at present, and thatthe annual production amounts to about 400, 000 tons. If this should betrue, taking the foregoing as a basis, the supply will becomeexhausted in about seventy-five years. 

In 1819 a law was passed in Italy, which is still in force, governingmining in Sicily, which provides that should a land owner discover orein his property he would be the owner thereof, and should have theright to mine, operate, or rent the property to others for thatpurpose, but if he should decline to operate his mines or to rent themto others to be operated, the state would rent them on its ownaccount. 

Royalties vary from 12 to 45 per cent. They are paid according to thequality of the ore and the facilities for producing sulphur; 25 percent. May, however, be taken as an average. There is a land tax of 36per cent. Of the net income, which is usually paid by the owners andlessees of the mines, in proportion to the quantity of sulphur whichthey produce. The export duty is 10 lire per ton. All mines areinspected by government officials once a year, and the owners arerequired to furnish the state with plans of the works and theirprogress, with a view to insure the safety of the workmen and toascertain the extent of the property. 

Those who rent their mines receive from 10 to 40 per cent. Of thesulphur produced. Leases are valid for such period as the contractingparties may stipulate therein. The general limit, however, is nineyears. The average lease is 25 per cent. , 40 per cent. Being paid onlywhen the mines are very favorably situated and the production good. Some lessees prefer paying a considerable sum in cash in advance, atthe beginning of the term of the lease, and giving 15 or 20 per cent. In sulphur annually thereafter, instead of a higher percentage. 

The external indications of the presence of sulphur are the appearanceof gypsum and sulphurous springs. These are indubitable signs of thepresence of sulphur, and when discovered the process resorted to here, in order to reach the sulphur, is to bore a hole sufficiently large toadmit a man, after which steps are constructed in the passage in orderto facilitate the workmen in going to and fro. These steps extendacross the passage, and are about 25 centimeters high and 35 broad. The inclination of the holes or passages varies from 30 to 50 degrees. Upon attaining the depth of several meters water is often met with, and in such considerable quantity that it is impossible to proceed. Hence it becomes necessary to either pump the water out or retreat inorder to bore elsewhere. It is often necessary to bore severalpassages in order to discover the ore or seam of sulphur. When, however, it has been discovered the passages are made to follow itsdirection, whether upward or downward. As the direction of seams is inmost cases irregular, that of the passages or galleries is likewise. Where the ore is rich and the matrix yielding, the miners break it bymeans of pick-axes and pikes, but when such is not the case gunpowderis resorted to, the ore in this case being carried to the surface byboys. The miners detach the ore from the surrounding material, and thecavities which ensue in consequence assume the appearance of vastcaves, which are here and there supported by pillars of rock and orein order to keep them from falling or giving way. In order tostrengthen the galleries sterile rock is piled upon each side andcemented with gypsum. In extensive mines, however, these supports andlinings are too weak, and not infrequently, as a result, the galleriesand caverns give way, occasionally causing considerable havoc amongthe miners. Sulphur is found from the surface to a depth of 150meters. The difficulties met with in operating mines are numerous, andamong the greatest in this category are water, land slides, irregularity of seam, deleterious gases, hardness of rocks andmatrices. Of these difficulties, water is the most frequently metwith. Indeed, it is always present, and renders the constant use ofpumps necessary. At one time miners were allowed to dig where theypleased so long as sulphur was extracted, the consequence being thatin groups of mines, the extent and direction of which being unknown totheir respective owners, one mine often fell into or upon another, thus causing destruction to life and property. It was largely for thisreason, it is understood, that the government determined to requireowners and lessees of mines to furnish plans thereof to properauthority, and directed that official inspection of the mines shouldbe made at stated periods. In order to comply with the decree of thegovernment it became necessary to employ mining engineers to draw theplans, etc. , and those employed were generally foreigners. In thesystem of excavation described no steam power is employed. Pumping isperformed by means of primitive wooden hand pumps, and when sufficientore has been collected it is conveyed on the backs of boys to thesurface--a slow, costly, and difficult procedure. This system may, however, be suitable to small mines, but in large mines there is noeconomy in hand labor; indeed, much is lost in time and expense by it. For this reason steam has been introduced into the larger and moreimportant mines. The machinery employed is a hoisting apparatus, witha drum, around which a coil is wound, with the object of hoisting andlowering trucks in vertical shafts. Steam pumps serve to extract thewater. The force of the hoisting apparatus varies from 15 to 50 horsepower. The fuel consumed is English and French coal, the former beingpreferred, as it engenders greater heat. The cost of a ton of coal atthe wharf is $4. 40, whereas in the interior of the island it costsabout $10. The shafts or pits are made in the ordinary way, great carebeing taken in lining them with masonry in order to guard against landslides. In level portions of the country vertical shafts arepreferred, but where the mine is situated upon a hill a debouch mayoften be found below the sulphur seam, when an inclined plane ispreferred, the ore being placed in trucks and allowed to run down theplane on rails until it reaches the exterior of the mine, where itsuddenly and violently stops, and as a result the trucks are emptiedof their load, when they are drawn up the plane to be refilled; andthus the process goes on indefinitely. In these mines a gutter is madein the inclined plane which carries off the water, thus dispensingwith the necessity of a pump and the requisites to operate it. Thegalleries and inclined shafts are lined with beams of pine or larch, which are brought hither from Sardinia, as Sicily possesses verylittle timber. The mines are illuminated by means of iron oil lamps, the wicks of which are exposed. The lamps are imported from Germany. In certain cases an earthenware lamp, made on the island, and said tobe a facsimile of those used by the Phoenicians, is employed. Thislamp is made in the shape of a small bowl. It is filled with oil and awick inserted, which hangs or extends outward, and is thus ignited, the flame being exposed to the air. Safety lamps are unknown, andthose described are generally secure. Few explosions take place--onlywhen confined carbonic hydrogen is met with in considerablequantities, and when the ventilation is not good. In this case themine is easily ignited, and once on fire may burn for years. The onlypractical expedient for extinguishing the fire is to close all inletsand outlets in order to shut off the air. This, however, is difficultand takes time. Notwithstanding the closing of communications, thegases escape through the fissures and openings which obtaineverywhere, and the ingress of air makes it next to impossible toextinguish the fire; hence it burns indefinitely or until the mine isexhausted. Occasionally the burning of a mine results beneficially toits owners, in that it dispenses with the necessity of smelting, andproduces natural, refined sulphur. 

Galleries in extent are usually 1. 20 by 1. 80 meters, and when ore isnot found and it becomes necessary to extend the galleries, laborersare paid in accordance with the progress they may make and thecharacter of the rock, earth, etc. , through which it may be necessaryto cut, as follows:

Silicious limestone, 60 lire per meter; daily progress, 0. 20 meter. 

Gypsum, 50 lire per meter; daily progress, 0. 30 meter. 

Marl, 30 lire per meter; daily progress, 0. 50 meter. 

Clay, 15 lire per meter; daily progress, 1 meter. 

Laborers working in the ore are paid 4. 30 lire per ton. This includesdigging, extracting, and illumination. In some mines, however, thelaborers are paid when the sulphur is fused and ready for exportation. One ton of sulphur, or its equivalent (say from 40 to 50 lire), is theamount generally paid. In mines where this system obtains theadministration is only responsible for their maintenance. Each minerproduces on an average about 1½ tons of ore daily, and when the worksare not more than 40 meters in depth he employs one boy to assist him, two boys when they reach 60 meters, and three when under 100 meters. These boys are from seven to sixteen years of age, and are paid from0. 85 to 1. 50 lire per day by the miner who employs them. They carryfrom 1, 000 to 1, 500 pounds of ore daily, or in from six to eighthours. The food consumed by miners is very meager, and consists ofbread, oil, wine, or water; occasionally cheese, macaroni, andvegetables are added to the above. 

Mining laborers generally can neither read nor write, and whenemployed in mines distant from habitations or towns, live and sleeptherein, or in the open air, depending on the season or the weather. In a few mines the laborers are, however, provided with suitabledwelling places, and a relief fund is in existence for the succor ofthe families of those who die in the service. This fund is greatlyopposed by the miners, from whose wages from 1 to 2 per cent. Isdeducted for its maintenance. In the absence of a fund of thischaracter, the sick or infirm are abandoned by their companions andleft to die. Generally miners are inoffensive when fairly dealt with. They are said to be indolent and dishonest as a rule. The managers ofmines receive from 3, 000 to 5, 000 lire per annum; chief miners from1, 500 to 2, 500 lire; surveyors, 700 to 1, 000 lire; and weighers andclerks, from 1, 000 to 2, 000 lire per annum. The total number of mininglaborers in Sicily is estimated at about 25, 000. 

The ore for fusion of the first grade as to yield contains from 20 to25 per cent. Of sulphur, that of the second grade from 15 to 20 percent. , and of the third grade 10 to 15 per cent. The usual meansadopted for extracting sulphur from the ore is heat, which attains theheight of 400 degrees Centigrade, smelting with the kiln, which inSicilian dialect is called a "calcarone. " The "calcarone" is capableof smelting several thousand tons of ore at a time and is operated inthe open air. Part of the sulphur is burned in the process of smeltingin order to liquefy the remainder. "Calcaroni" are situated as closelyto the mouth of a shaft as possible, and if practicable on the side ofa hill, in order that when the process of smelting is complete, thesulphur may run down the hill in channels prepared for the purpose. The shop of a "calcarone" is circular and the floor has an inclinationof from 10 to 15 degrees. A design of a "calcarone" is herewithinclosed. The circular wall is made of rude stone work, cementedtogether with gypsum. The thickness of the wall at the back is 0. 50meter, and from this it gradually becomes thicker until in front, where it is 1 meter, when the diameter is to be 10 meters. In front ofthe thickest part of the wall an opening is left, measuring 1. 20meters high and 0. 25 meter broad. 

Through this opening the liquid sulphur flows. Upon each side of thisopening two walls are built at right angles with the circular wall, inorder to strengthen the front of the kiln. These walls are 80centimeters thick each and are roofed. A door is hinged to thesewalls, thus forming a small room in front of each kiln in which thekeeper thereof resides from the commencement to the termination of theflow of sulphur. The inclined floor of the kiln is made of stone workand is covered with "ginesi, " the name given to the refuse of a formerprocess of smelting. The stone work is 20 centimeters thick, and the"ginesi" covering 25 centimeters, which gradually becomes thicker asit approaches its lowest extremity. The front part of the circularwall is 3. 50 meters high and the back 1. 80 meters. The interior of thewall is plastered with gypsum in order to render it impermeable. 

The cost of a "calcarone" of about 500 tons capacity is 800 lire. Thecapacity varies from 40 to 5, 000 tons, or more, depending uponcircumstances. If a mine is enabled to smelt the whole year round, thesmaller "calcaroni, " being more easily managed, are preferred; theinverse is the case as to the larger "calcaroni, " when this isimpracticable. When a "calcarone" is situated within 100 meters of acereal farm, its operation is prohibited by law during the summer, lest the fumes of the sulphur should destroy the crop. 

When, however, the distance is greater from the farm or farms than 100meters, smelting is permitted; but should any damage ensue to thecrops as a result of the fumes, the owners of the "calcaroni" arerequired to liquidate it. Therefore the mines which are favorablysituated smelt the entire year, and employ "calcaroni" of from 40 to500 tons, as there is less risk of a process failing, whichoccasionally happens, and for the reason that the ore can be smeltedas soon as it is extracted; whereas, when kilns or "calcaroni" aresituated within or adjacent to the limit adverted to, they can only beoperated five or six months in the year, consequent upon which the oreis necessarily stacked up all through the summer or until such time assmelting may be commenced without endangering the crops, when itbecomes necessary to use "calcaroni" whose capacity amounts to severalthousand tons. As intimated, these large "calcaroni" are not somanageable as those of smaller dimensions, and as a result manythousands of tons of sulphur are lost in the process of smelting, besides perhaps the loss of an entire year in labor. Again, the oredeteriorates or depreciates when long exposed to the air and rain, allof which, when practicable, render the kilns or "calcaroni" of thesmaller capacity more advantageous and lucrative to those operatingsulphur mines in Sicily. Smelting with a "calcarone" of 200 tonscapacity consumes thirty days, one of 800 tons 60 days, and with a"calcarone" of 2, 000 tons capacity from 90 to 120 days are consumed. 

In loading or filling the "calcaroni, " the larger blocks of ore areplaced at the bottom as well as against the mouth, in order to keepthe lower part of the kiln as cool as possible with a view ofpreventing the liquid sulphur from becoming ignited as it passes downto where it makes its exit, etc. The blocks of ore thus first placedin position are, for obvious reasons, the most sterile. After thefoundation is thoroughly laid the building of the "pile" is proceededwith, the larger blocks being placed in the center to form, as itwere, the backbone of the pile; the smaller blocks of ore are arrangedon the outside of these and in the interstices. The shape or form ofthe pile when completed is similar to a truncated cone, and whenburning the kiln looks like a small volcano. When the kiln has beenfilled with ore, the whole is covered with ginesi with a view ofpreventing the escape of the fumes. The ore is then ignited by meansof bundles of straw, impregnated or saturated with sulphur, being heldabove the thin portion of the top of the kiln, which is at once closedwith ginesi, and the "calcarone" is left to itself for about a week. During the burning process the flames gradually descend, and thesulphur contained in the ore is melted by the heat from above. Inabout seven or eight days sulphuric fumes and sublimed sulphurcommence to escape, when it becomes necessary to add a new coat ofginesi to the covering and thus prevent the destruction of vegetationby the sulphur fumes. The mouth of the kiln, which has been left openin order to create a draught, is closed up about this time with gypsumplaster. When the sulphur is all liquefied it finds its way to themost depressed part of the kiln, and there, upon encountering thelarge sterile blocks, quite cold, already referred to, solidifies. Itis again liquefied by means of burning straw, whereupon an iron troughis inserted into a mouth made in the kiln for the purpose, and thereliquefied sulphur runs into it, from which it is immediatelycollected into wooden moulds, called "gadite, " and which have beenkept cool by being submerged in water. Upon its becoming thoroughlycool the sulphur is taken out of the moulds referred to, and is now insolid blocks, each weighing about 100 weight. Two of these blocksconstitute a load for a mule, and cost from 4 to 5 francs. 

The above is the result when the operation succeeds; but this is notalways the case. At times the sulphur becomes solidified before itreaches the mouth of the kiln, because of the heat not beingsufficient to keep it liquid in its passage thereto, and othermisfortunes not within control, and consequent upon the use of thelarger kilns, or "calcaroni. "

When the sulphur ceases to run from the kiln, the process is complete. The residue is left to cool, which consumes from one to two months. The cooling process could be accomplished in much less time bypermitting the air to enter the kiln, but this would be destructive tovegetation, and even to life, consequent upon the fumes of thesulphur. The greatest heat at a given time in a kiln is calculated tobe above 650 degrees Centigrade--that is, at the close of the process. This enormous heat is generally allowed to waste, whereas it isunderstood it could be utilized in many ways. A gentleman of the nameof Gill is understood to have invented a recuperative kiln, whichwill, if generally adopted, utilize the heat of former processesnamed. A ton of ore containing about 25 per cent. Of sulphur yields300 pounds of sulphur. This is considered a good yield. When it yields200 pounds it is considered medium, and poor when only 75 pounds. Laborers are paid 0. 40 lire per ton for loading and unloading kilns, and from thirty to forty hands are employed at a time. The keeper of akiln receives from 2 to 2. 50 lire per day. 

Notwithstanding the "calcarone" has many defects, it is the simplestand cheapest mode of smelting, and is preferred here to any othersystem requiring machinery and skilled labor to operate it. 

The following are the principal furnaces in use here: Durand's;Hirzel; Gill and Kayser's system of fusion; Conby Bollman process;Thomas steam process of smelting; and Robert Gill's recuperativekilns. 

There are seven qualities or grades of sulphur, viz. :

1. Sulphur almost chemically pure, of a very bright and yellow color. 

_Second Best. _--Slightly inferior to the first quality; bright andyellow. 

_Second Good. _--Contains 4 to 5 per cent. Of earthy matter, but is ofa bright yellow. 

_Second Current. _--Dirty yellow, containing more earthy matter thanthat last named. 

_Third Best. _--Brownish yellow; this tint depends on the amount ofbitumen which it contains. 

_Third Good. _--Light brown, containing much extraneous matter. 

_Third Current. _--Brown and coarse. 

These qualities are decided by color, not by test. The difference ofprice is from 3 to 10 francs per ton. Manufacturers prefer the thirdbest, because of its containing more sulphuric acid and costing lessthan the sulphur of better quality. 

Sulphur is conveyed to the seaboard by rail, in carts, or on mules ordonkeys. Conveyance by cart, mule, or donkey is only resorted to whenthe distance is short or from mines to railroad stations. The tariffin the latter case is understood to be 1 lire per ton per mile. Therailroad tariff is 0. 12 per ton per kilometer; but it is contemplated, it is understood, to reduce this to 7 centimes in a short time. Theprice per ton of sulphur is as follows:

 At Porto At At Grade. Empedocle. Licata. Catania. Lire. Lire. Lire. Second best 86. 60 87. 00 90. 70 Second good 84. 42 84. 50 90. 30 Second current 83. 90 83. 90 88. 40 Third best 79. 00 79. 90 86. 90 Third good 77. 80 77. 80 83. 00 Third current 76. 80 76. 70

Sulphur free on board, brokerage, shipment, export duty, and all otherexpenses included, costs 20 lire per ton in excess of the aboveprices. Nearly all the sulphur exported from Palermo emanates from theLercara mines, in the province of Palermo, the price per ton being asfollows: first quality, 91. 60 lire; second quality, 88. 40. Sulphur isusually conveyed in steamers to foreign countries from Sicilian ports. The average freight per ton to New York is about as follows: FromPalermo, 8. 70 lire; from Catania, 13. 50 lire; from Girgenti, 16 lire. An additional charge of 2. 50 lire is made when the sulphur may bedestined for other ports in the United States. 

Liebig once said that the degree of civilization of a nation and itswealth could be seen in its consumption of sulphuric acid. Now, although Italy produces immense quantities of sulphur, it cannot, onaccount of the scarcity of fuel, and other obvious reasons perhaps, compete with certain other countries in the manufacture andconsumption of sulphuric acid. 

Sulphur is employed in the manufacture of sulphuric acid, and thelatter serves in the manufacture of sulphate of soda, chloridic acid, carbonate of soda, azodic acid, ether, stearine candles, purificationof oils in connection with precious metals and electric batteries. Nordhausen's sulphuric acid is employed in the manufacture of indigo. Sulphate of soda is employed in the manufacture of artificial soda, glassware, cold mixtures, and medicines. Carbonate of soda is used inthe manufacture of soap, bleaching wool, coloring and paintingtissues, and in the manufacture of fine crystal ware and thepreparation of borax. Chloric acid is used in the preparation ofchlorides with bioxide of manganese, and with chlorides in thepreparation of hypochlorides of lime, known in commerce under the nameof bleaching powder, and improperly called chloride of lime, which isused as a disinfectant in contagious diseases, in bleaching stuffs, and in the manufacture of paper from vegetable fibers, and in themanufacture of gelatine extracted from bones, as well as in fermentingmolasses and in the manufacture of sugar from beet root. Sulphur isalso used in the preparation of gunpowder and oil of vitriol, and inthe manufacture of matches and cultivation of the vine. 

In the year 1838 the Neapolitan government granted a monopoly to aFrench company for the trade in sulphur. By the terms of the agreementthe producers were required to sell their sulphur to the company atcertain fixed prices, and the latter paid the government the sum of$350, 000 annually in consideration of this requirement. This, however, was not a success, and tended to curtail the sulphur industry, and thegovernment, discovering the agreement to be against its interests, annulled it, and established a free system of production, charging anexport tax per ton only. At that time sulphuric acid was derivedexclusively from sulphur. Hence the demand from all countries wasgreat, and the prices paid for sulphur were high. It was about thisperiod that the sulphur industry was at its zenith. The monopolyhaving been abolished, every mine did its utmost to produce as muchsulphur as possible, and from the export duty exacted by thegovernment there accrued to it a much larger revenue than that whichit received during the period of the monopoly. The progress of sciencehas, however, modified the state of things since then, as sulphur cannow be obtained from pyrite or pyrite of iron. This discoveryimmediately caused the price of sulphur to fall, and the great demandtherefore correspondingly ceased. In England, at the present time, itis understood that two-thirds of the sulphuric acid used ismanufactured from pyrites. The decrease in prices caused many of themines to suspend operations, and as a result the sulphur remained idlein stock. In 1884 an association was formed at Catania with a view tobuying up sulphur thus stored away at the mines and various ports atlow prices, and store it away until a favorable opportunity shouldpresent itself for the sale thereof. This had the effect of increasingthe prices of sulphur in Sicily for some time, and the producers, discovering that the methods of the association increased the foreigndemand for their produce as well as its prices, exported it directlythemselves, thus breaking up the association referred to, as it was nolonger a profitable concern. 

The railroad system, which in later years has placed the mostimportant parts of Sicily in communication with the seaboard, has beenmost beneficial to the sulphur industry. A great saving has been madein transporting it to the ports. This was formerly (as stated)accomplished by carts drawn by mules at an enormous expense, as theroads were wretched, and unless some person of distinctioncontemplated passing over them, repairs were unknown. 

Palermo, March 20, 1888. 

 * * * * *

AN AUTOMATIC STILL. 

BY T. MABEN. 

The arrangement here described is one that may readily be adapted to, and is specially suited for, the old fashioned stills which are infrequent use among pharmacists for the purpose of distilling water. The idea is extremely simple, but I can testify to its thoroughefficiency in actual practice. The still is of tinned copper, twogallon capacity, and the condenser is the usual worm surrounded withcold water. 

The overflow of warm water from the condenser is not run into thewaste pipe as in the ordinary course, but carried by means of a benttube, A, B, C, to the supply pipe of the still. The bend at B acts asa trap, which prevents the escape of steam. 

[Illustration]

The advantages of this arrangement are obvious. It is perfectlysimple, and can be adapted at no expense. It permits of a continuoussupply of hot water to the still, so that the contents of the lattermay always be kept boiling rapidly, and as a consequence it condensesthe maximum amount of water with the minimum of loss of heat. If thesupply of water at D be carefully regulated, it will be found that acontinuous current will be passing into the still at a temperature ofabout 180° F. , or, if practice suggest the desirability of running inthe water at intervals, this can be easily arranged. It is necessarythat the level at A should be two inches or thereabout higher than thelevel of the bend at C, otherwise there may not be sufficient head toforce a free current of water against the pressure of steam. It willalso be found that the still should only contain water to the extentof about one-fourth of its capacity when distillation is commenced, asthe water in the condenser becomes heated much more rapidly than thesame volume is vaporized. By this expedient a still of two gallonscapacity will yield about half a dozen gallons per day, a much greaterquantity than could ever be obtained under the old system, whichrequired the still to be recharged with cold water every time one anda half gallons had been taken off. 

The objection to all such continuous or automatic arrangements is, ofcourse, that the condensed water contains all the free ammonia thatmay have existed in the water originally, but it is only in caseswhere the water is exceptionally impure that this disadvantage willbecome really serious. The method here outlined has, no doubt, occurred to many, and may probably be in regular use, but not havingseen any previous mention of the idea, I have thought that it might beuseful to some pharmacists who prepare their own distilledwater. --_Phar. Jour. _

 * * * * *

COTTON SEED OIL. 

"Cotton seed oil, " said Mr. A. E. Thornton, of the Atlanta mills, "isone of the most valuable of oils because it is a neutral oil, that is, neither acid nor alkali, and can be made to form the body of any otheroil. It assimilates the properties of the oil with which it is mixed. For instance, olive oil. Cotton seed oil is taken and a little extractof olives put in. The cotton oil takes up the properties of theextract, and for all practical purposes it is every bit as good as thepure olive oil. Then it is used in sweet oil, hair oil, and, in fact, in nearly all others. A chemist cannot tell the prepared cotton oilfrom olive oil except by exposing a saucerful of each, and the oliveoil becomes rancid much quicker than the cotton oil. The crude oil isworth thirty cents a gallon, and even as it is makes the finest ofcooking lard, and enters into the composition of nearly all lard. "

A visit to the mills showed how the oil is made. From the platformwhere the seed is unloaded it is thrown into an elevator and carriedby a conveyor--an endless screw in a trough--to the warehouse. Then itis distributed by the conveyor uniformly over the length of thebuilding--about 200 feet. The warehouse is nearly half filled now, andthousands and thousands of bushels are lying in store. Anotherelevator carries the seed up to the "sand screen. " This is a revolvingcylinder made of wire cloth, the meshes being small enough to retainthe seed, which are inside the cylinder, but the sand and dirt escape. Now the seeds start down an inclined trough. There is something elseto be taken out, and that is the screws and nails and rocks that weretoo large to be sifted out with the sand and dirt. There is a hole inthe inclined trough, and up through that hole is blown a current ofair by a suction fan. If it were not for the fan, the cotton seed, rocks, nails, and all would fall through. The current keeps up thecotton seed, and they go on over, but it is not strong enough to keepup the nails and pebbles, and they fall through. Now the seed, free ofall else, is carried by another elevator and endless screw conveyor tothe "linter. " This is really nothing more than a cotton gin with anautomatic feed. 

"HULLER" AND "HEATERS. "

Then the seed is carried to the "huller, " where it is crushed orground into a rough meal about as coarse as the ordinary corn "grits. "The next step is to separate the hulls from the kernels, all the oilbeing in the kernel, so the crushed seed is carried to the"separator. " This is very much on the style of a sand screen, being arevolving cylinder of wire cloth. The kernels, being smaller than thebroken hulls, fall through the broken meshes, and upon this principlethe hull is separated and carried direct to the furnace to be used asfuel. The kernels are ground as fine as meal, very much as grist isground, between corrugated steel "rollers, " and the damp, reddishcolored meal is carried to the "heater. "

The "heater" is one iron kettle within another, the six inch steamspace between the kettles being connected direct with the boilers. There are four of these kettles side by side. The meal is brought intothis room by an elevator, the first "heater" is filled, and for twentyminutes the meal is subjected to a "dry cook, " a steam cook, the steamin the packet being under a pressure of forty-five pounds. Inside theinner kettle is a "stirrer, " a revolving arm attached at right anglesto a vertical shaft. The stirrer makes the heating uniform, and thehigh temperature drives off all the water in the meal, while theinvolatile oil all remains. 

In five minutes the next heater is filled, in five minutes the next, etc. 

Now there are four "heaters, " and as the last heater is filled--at theend of twenty minutes--the first heater is emptied. Then at the end offive minutes the first heater is filled, and the one next to it isemptied, and the rotation is kept up, each heater full of meal being"dry-cooked" for twenty minutes. 

Corresponding to the four heaters are four presses. Each pressconsists of six iron pans, shaped like baking pans, arranged one abovethe other, and about five inches apart. The pans are shallow, andaround the edge of each is a semicircular trough, and at the lowestpoint of the trough is a funnel-shaped hole to enable the oil to runfrom one pan to the next lowest, and from the lowest pan to the"receiving tanks" below. 

PRESSING OUT THE OIL. 

As soon as a "heater" is ready to be emptied, the meal is taken outand put into six hair sacks, corresponding to the six pans in thepress. There are six hair mats about one foot wide and six long, oneside of each being coated with leather. The hair mat is about an inchthick. Now the hair sack, containing ten and a half to eleven poundsof heated steaming meal, is placed on one end of the mat, and the mealdistributed so as to make a pad or cushion of uniform thickness. Thepad of meal is not quite three feet long, a foot wide, and threeinches thick, and the hair mat is folded over, sandwiching the pad andleaving the leather coating of the pad outside. In this form the sixloads are put into the six pans, and by means of a powerful hydraulicpress the pans are slowly pressed together. The oil begins tricklingout at the side, slowly at first, and then suddenly it begins runningfreely. The pressure on the "loads" is 350 tons. After being pressedabout five minutes, the pressure is eased off and the "loads" takenout. What had been a mushy pad three inches thick is a hard, compactcake about three-quarters of an inch thick, and the sack is literallyglued to the cake. The crude oil has a reddish muddy color as it runsinto the tanks. 

To one side were lying great heaps of sacks of yellowish meal--thecakes which have been broken and ground up into meal. That, asexplained above, forms the body of all fertilizers. The following is asummary of the work for the eight months' season at the Atlanta mills:

Fifteen thousand tons of seed used give:

Fifteen million pounds of hull. 

Ten million three hundred and thirty-one thousand two hundred andfifty pounds of meal. 

Four million six hundred and sixty-eight thousand seven hundred andfifty pounds of oil. 

Three hundred thousand pounds of lint cotton. 

The meal is worth at the rate of $6 for 700 pounds, or $88, 603. 58. 

The oil is worth thirty cents a gallon, or seven and a half pounds, or$186, 750. 

The lint is worth $18, 000, making a total of $293, 353, and thatdoesn't include the 15, 000, 000 pounds of hull. --_AtlantaConstitution. _

 * * * * *

MANUFACTURE OF PHOTOGRAPHIC SENSITIVE PLATES. 

Quite recently Messrs. Marion & Company, London, began on their ownaccount to manufacture sensitive photographic plates by machinery, andthe operations are exceedingly delicate, for a single minute airbubble or speck of dust on a plate may mar the perfection of apicture. Their works for the purpose at Southgate were erected in thesummer of 1886, and were designed throughout by Mr. Alexander Cowan. 

[Illustration: Fig. 1. ]

Buildings of this kind have to be specially constructed, because someof the operations have to be carried on in the absence of daylight, and in that kind of non-actinic illumination which does not act uponthe particular description of sensitive photographic compoundmanipulated. Glass and other materials have therefore to pass fromlight to dark rooms through double doors or double sliding cupboardsmade for the purpose, and the workshops have to be so placed inrelation to each other that the amount of lifting and the distance ofcarriage of material shall be reduced to a minimum. Moreover, thefinal drying of sensitive photographic plates takes place in absolutedarkness. Fig. 1 is a ground plan of the chief portion of the works. In this cut, A is the manager's private office, B the counting house, C the manager's laboratory, and D his dark room for privateexperiment, which can thus be conducted without interfering with theregular work of the establishment. E is the carpenter's shop andpacking room, F the albumen preparation room, G the engine room, withits two doors; the position of the engine is marked at H. The mainbuilding is entered through the door, K; the passage, L, is used forthe storage of glass, and has openings in the wall on one side topermit the passage of glass into the cleaning room, M; this room isilluminated by daylight. The plates, after being cleaned, pass intothe coating rooms, N and O, into which daylight is never admitted; thecoating machine is in the room, N, and three hand coating tables inthe room, O; both these rooms are illuminated by non-actinic light. 

[Illustration: Fig. 2. ]

[Illustration: Fig. 3. ]

The walls of N and O are of brick, to keep these interior rooms ascool as possible in hot weather, for the making of photographic platesis more difficult in summer time, because the high temperature tendsto prevent the rapid setting of the gelatine emulsion upon them. Atthe end of these rooms and communicating with both is the lift, P, bywhich the coated plates are carried to the drying rooms above, whichthere cover the entire area of the main building; they consist of tworooms measuring 60 ft. By 30 ft. , and are each 30 ft. High at thehighest part in the center of the building; these rooms arenecessarily kept in absolute darkness, except while the plates arebeing stored therein or removed therefrom, and on such occasionsnon-actinic light is used. After the plates are dry, they come downthe lift, Q, into the cutting and packing room, R, which isilluminated by non-actinic light. In the drying rooms the batches ofplates are placed one after the other on tram lines at one end of theroom, and are gradually pushed to the other end of the building, sothat the first batches coated are the first to be ready to be takenoff when dry, and to be sent down the lift, Q. The plates in R, whensufficiently packed to be safe from the action of daylight, are passedthrough specially constructed openings into the outside packing room, S, where they are labeled. The chemicals are kept in the room, U, where they are weighed and measured ready for the making of thephotographic emulsion in the room, U. The next room, V, is for washingsmall experimental batches of emulsion, and W is the large washingroom. The emulsion is then taken into the passage, X, communicatingwith the two coating rooms. A centrifugal machine in the room, Y, isused for extracting silver residues from waste materials, also forfreeing the emulsion from all soluble salts. Washing and cleaning ingeneral go on in the room, Z. 

[Illustration: Fig. 4. ]

The glass for machine coating is cut to standard sizes at thestarting, instead of being coated in large sheets and cut afterward--apractice somewhat common in this industry. The disadvantage of theordinary plan is that minute fragments of glass are liable to settleupon the sensitive film and to cause spots and scratches during thepacking operations; any defect of this kind renders a plate worthlessto the photographer. When any breakages take place in the cutting, itis best that they should occur at the outset, and not after the platehas been coated with emulsion. The cutting when necessary is effectedby the aid of a "cutting board, " Fig. 2, invented by Mr. Cowan, andnow largely in use in the photographic world. This appliance is usedto divide into two equal parts, with absolute exactness, any platewithin its capacity, and it is especially useful in dimly lightedrooms. It consists of four rods pivoted together at the corners andswinging on two centers, so that in the first position it is trulysquare, and in other positions of rhomboid form, the two outer barsapproaching each other like those of a parallel ruler. The hinge flapcomes down on the exact center of the plate, minus the thickness ofthe block holding the diamond. By this appliance plates can be cut ineither direction. Fig. 3 represents a similar arrangement for cuttinga number of very small plates out of one large one; in this the hingeflap is made in the form of a gridiron, and the bars are spaced ataccurate distances, according to the size of the plate to be cut, sothat a plate 10 in. Square, receiving four cuts in each direction, will be divided into twenty-five small plates. 

[Illustration: Fig. 5. ]

Before being cleaned all sharp edges are roughly taken off thoseplates intended for machine coating by girls, who rub the edges andcorners of the plates upon a stone; the plates are then cleaned by anysuitable method in use among photographers. The plates, now ready forthe coating room, have to be warmed to the temperature of theemulsion, say from 80 deg. F. To 100 deg. F. , before they pass to thecoating machine, the inventor of which, Mr. Cadett, having come to theconclusion that, if the plates are not of the proper temperature, thecoating given will be uneven over various parts of the surface. Theplate-warming machine is represented in Fig. 4; it was designed by Mr. A. Cowan, and made by his son, Mr. A. R. Cowan. It consists of atrough 7 ft. Long by 3 in. Deep, forming a flat tank, through whichhot water passes by means of the circulating system shown in theengraving. To facilitate the traveling of the glass plates withoutfriction the top of the tank is a sheet of plate glass bedded on asand bath. An assistant at one end places the glasses one after theother on the warm glass slab, and by means of a movable slide pushesthem one at a time under the cover, which cover is represented raisedin the engraving to show the interior of the machine. After having putone glass plate on the slide, another cannot be added until the man inthe dark room at the other end of the slide has taken off the farthestwarmed plate, because the slide has a reciprocating movement. Thisheating apparatus is built at right angles to the coating machine inthe next room, in order to be conveniently placed in the presentbuilding; but it is intended in future to use it as a part of thecoating machine itself, and to drive it at the same speed and with thesame gearing, so that the cold plates will be put on by hand at oneend, get warmed as they pass into the dark room, at the other end ofwhich they will be delivered by the machine in coated condition. Underneath the heating table is a copper boiler, with its Bunsen'sburner of three concentric rings to get up the temperature quickly andto give the power of keeping the water under the heating slab at adefinite temperature, as indicated by a thermometer. The cold watertank of the system is represented against the wall in the cut. 

[Illustration: Fig. 6. ]

Fig. 5 represents the hot water circulating system outside the coatingrooms for keeping the gelatine emulsions in these dimly lightedregions at a given temperature, without liberating the products ofcombustion where the emulsion is manipulated. The temperature isregulated automatically. It will be noticed where the pipes enter thetwo coating rooms, and Fig. 6 shows the copper inside one of themheated by the apparatus just described. The emulsion vessel in thecopper is surrounded by warm water, and the copper itself is jacketedand connected with the hot water pipes, so forming part of thecirculating system. 

[Illustration: Fig. 7. ]

Fig. 7 is a general view of the coating machine recently invented byMr. Cadett, of the Greville Works, Ashtead, Surrey. The plates warmedin the light room, as already described, are delivered near the end ofthe coating table, where they are picked off a gridiron-like platform, represented on the right hand side of the cut, and are placed by anassistant one by one upon the parallel gauges shown at the beginningof the machine proper; they are then carried on endless cords underthe coating trough described farther on. After they have been coatedthey are carried onward upon a series of four broad endless bands ofabsorbent cotton--Turkish toweling answers well--and this cotton iskept constantly soaked with cold water, which flows over sheets ofaccurately leveled plate glass below and in contact with the toweling;the backs of the plates being thus kept in contact with fresh coldwater, the emulsion upon them is soon cooled down and is firmly set bythe time the plates have reached the end of the series of four wettables. They are then received upon one over which dry towelingtravels, which absorbs most of the moisture which may be clinging tothe backs of the plates; very little wet comes off the backs, so thatduring a day's work it is not necessary to adopt special means toredry this last endless band. What are technically known as "wholeplates, " which are 8½ in. By 6½ in. , are placed touching each otherend to end as they enter the machine, and they travel through it atthe rate of 720 per hour; smaller sizes are coated in proportion, thesmaller the plates the larger is the number coated in a given time. The smaller plates pass through the machine in two parallel rows, instead of in a single row, so that quarter plates, 4¼ in. By 3¼ in. , are delivered at the end of the machine at the rate of 2, 800 per hour, keeping two attendants well employed in picking them up and placingthem in racks as quickly as they can do the work. The double row ofcords for carrying two lines of small plates through the machine isrepresented in the engraving. Although the plates touch each other attheir edges on entering the machine, they are separated from eachother by short intervals after being coated; this is effected bydifferential gearing. The water flowing over the tables for coolingthe plates is caught in receptacles below and carried away by pipes. Between each of the tables is a little roller to enable small platesto travel without tilting over the necessary gap between each pair ofbands. 

[Illustration: Fig. 8. ]

The feeding trough of Cadett's machine is represented in Fig. 8. Theplates, cleaned as already described, are carried upon the cords undera brass roller, the weight of which causes sufficient friction to keepthe plates from tilting; they next pass under a soft camel's hairbrush to remove anything in the shape of dust or grit, and are thencoated. They afterward pass over a series of accurately leveled wheelsrunning in a tank of water kept exact by an automatic regulator at atemperature of from 80 deg. Fah. To 100 deg. Fah. , by means of a smallhot water circulating system. The emulsion trough is jacketed with hotwater at a constant temperature. This trough is silver plated inside, because most metals in common use would spoil the emulsion by chemicalaction. The trough is 16 in. Long; it somewhat tapers toward thebottom, and contains a series of silver pumps shown in the cut; thewhole of this series of pumps is connected with one long adjustablecrank when plates of the largest size have to be coated; when coatingplates of smaller sizes some of the pumps are detached. A chief objectof the machine is to deliver a carefully measured quantity of emulsionupon each plate, and this is done by means of pumps, in order that thequantity of emulsion delivered shall not be affected by changes in thelevel of the emulsion in the trough; the quantity delivered is thusindependent of variations due to gravity or to the speed of themachine. These pumps draw the emulsion from a sufficient depth in thetrough to avoid danger from the presence of air bubbles, and thebottom of the trough is so shaped that should by chance anysedimentary matter be present, it has a tendency to travel downward, away from the bottoms of the pumps. There is a steady flow of emulsionfrom the pumps to the delivery pipes, then it passes down a guideplate of the exact width of the plate to be coated. Immediately infront of the guide plate is a fixed silver cylinder, kept out ofcontact with the plate by the thickness of a piece of fine and veryhard hempen cord, which can be renewed from time to time. These cordskeep the cylinder from scraping the emulsion off the plate, and theyhelp to distribute it in an even layer. There would be two lines uponeach plate where it is touched by the cords, were not the emulsion sofluid as to flow over the cut-like lines made and close them up. 

[Illustration: Fig. 9. ]

The silver cylinder to a certain extent overcomes the effects ofirregularities in the glass plates, for the cylinder is jointedsomewhat in the cup and ball fashion, and is made in two or moreparts, which parts are held together by lengths of India rubber. 

The arrangement is shown in section in Fig. 9, in which A is the hotwater jacket of the emulsion vessel; B, the crank driving the pumps;C, a pump with piston in position; D, delivery tube of the pump; E, the silver guide plate to conduct the emulsion down to the glass; F, the spreading cylinder; G, the cords regulating the distance of thecylinder from the glass plates; H, soft camel's hair brush; K, friction roller; L L L, three plates passing under the emulsion tank;M, knife edged wheels in the hot water tank, N; the "plucking roller, "P, has a hot water tank of its own, and travels at slightly greaterspeed than the other rollers; R is the beginning of the cooling bands;T, the driving cords; and W, a level of the emulsion in the trough. Yrepresents one of the bucket pistons of the pumps, detached. Theconstruction of the crank itself is such that, by adjustment of theconnecting rods, more or less emulsion may be put upon the plates. Mr. Cowan, however, intends to adjust the pumps once for all, and toregulate the amount of emulsion delivered upon the plates by means ofdriving wheels of different diameters upon the cranks. 

[Illustration: Fig. 10. ]

Fig. 10 is a section of the hollow spreading cylinder, made of sheetsilver as thin as paper, so that its weight is light. For coatinglarge plates it is divided in the center, so as to adapt itselfsomewhat to irregularities in the surface of each plate. In this caseit is supported by a third and central thread, as represented in thecut. Otherwise the cylinder would touch the center of the plate. Itstwo halves are held together by a slip of India rubber. --_TheEngineer_. 

 * * * * *

THE USE OF AMMONIA AS A REFRIGERATING AGENT. [1]

 [Footnote 1: Paper lately read before the Civil and Mechanical Engineers' Society. ]

BY MR. T. B. LIGHTFOOT, M. I. C. E. 

Within the last few years considerable progress has been made in theapplication of refrigerating processes to industrial purposes, and thedemand for refrigerating apparatus thus created has led to theproduction of machines employing various substances as therefrigerating agent. In a paper read by the author before theInstitution of Mechanical Engineers, in May, 1886, these systems wereshortly described, and general comparisons given as to theirrespective merits, scope of application, and cost of working. In thepresent paper it is proposed to deal entirely with the use of ammoniaas a refrigerating agent, and to deal with it in a more full andcomprehensive manner than was possible in a paper devoted to theconsideration of a number of different systems and apparatus. In theUnited States and in Germany, as well as to some extent elsewhere, ammonia has been very generally employed for refrigerating purposesduring the last ten years or so. In this country, however, itsapplication has been extremely limited; and even at the present timethere are but few ammonia machines successfully at work in GreatBritain. No doubt this is, to a large extent, due to the fact that inthe United States and in Germany there existed certain stimulatingcauses, both as regards climate and manufactures, while in thiscountry, on the other hand, these causes were present only in amodified degree, or were absent altogether. The consequence was thatup to a comparatively recent date the only machine manufactured onanything like a commercial scale was the original Harrison's ethermachine, first produced by Siebe, about the year 1857--a machinewhich, though answering its purpose as a refrigerator, was both costlyto make and costly to work. In 1878 the desirability of supplementingour then existing meat supply by means of the large stocks in ourcolonies and abroad led to the rapid development of the special classof refrigerating apparatus commonly known as the dry air refrigerator, which, in the first instance, was specially designed for use on boardship, where it was considered undesirable to employ chemicalrefrigerants. Owing to their simplicity, and perhaps also to theirnovelty, these cold air machines have very frequently been applied onland, under circumstances in which the same result could have beenobtained with much greater economy by the use of ammonia or some otherchemical agent. Recently, however, more attention has been directed tothe question of economy, and consideration is now being given to theapplicability of certain machines to certain special purposes, withthe result that ammonia--which is the agent that, in our present stateof knowledge, gives as a rule the best results for largeinstallations, while on land at any rate its application for allrefrigerating purposes presents no unusual difficulties--promises tobecome largely adopted. It is hoped, therefore, that the followingpaper respecting its use will be of interest. 

In all cases where a liquid is employed, the refrigerating action isproduced by the change in physical state from the liquid to thevaporous form. It is, of course, well known that such a change canonly be brought about by the acquirement of heat; and for the purposeof refrigeration (by which must be understood the abstraction of heatat temperatures below the normal) it is obvious that, other thingsbeing equal, that liquid is the best which has the highest heat ofvaporization, because with it the least quantity has to be dealt within order to produce a given result. In fact, however, liquids vary, not only in the amount of heat required to vaporize them (this amountalso varying according to the temperature or pressure at whichvaporization occurs), but also in the conditions under which suchchange can be effected. For instance, water has an extremely highlatent heat, but as its boiling point at atmospheric pressure is alsohigh, evaporation at such temperatures as would enable it to be usedfor refrigerating purposes can only be effected under an almostperfect vacuum. The boiling point of anhydrous ammonia, on the otherhand, is 37½° below zero F. At atmospheric pressure, and therefore forall ordinary cooling purposes its evaporation can take place atpressures considerably above that of our atmosphere. Some other agentsused for refrigerating purposes are methylic ether, Pictet's liquid, sulphur dioxide, and ether. In this connection it should be statedthat Pictet's liquid is a compound of carbon dioxide and sulphurdioxide, and is said to possess the property of having vapor tensionsnot only much below those of pure carbon dioxide at equaltemperatures, but even below those of pure sulphur dioxide attemperatures above 78° F. The considerations, therefore, which chieflyinfluence the selection of a liquid refrigerating agent are:

1. The amount of heat required to effect the change from the liquid tothe vaporous state, commonly called the latent heat of vaporization. 

2. The temperatures and pressures at which such change can beeffected. 

This latter attribute is of twofold importance; for, in order to avoidthe renewal of the agent, it is necessary to deprive it of the heatacquired during vaporization, under such conditions as will cause itto assume the liquid form, and thus become again available forrefrigeration. As this rejection of heat can only take place if thetemperature of the vapor is somewhat above that of the cooling bodywhich receives the heat, and which, for obvious reasons, is in allcases water, the liquefying pressure at the temperature of the coolingwater, and the facility with which this pressure can be reached andmaintained, is of great importance in the practical working of anyrefrigerating apparatus. Ammonia in its anhydrous form, the use ofwhich is specially dealt with in this paper, is a liquid having atatmospheric pressure a latent heat of vaporization of 900, and aboiling point at the same pressure of 37½° below zero F. Water beingunity, the specific gravity of the liquid at a temperature of 40° F. Is 0. 76, and the specific gravity of its vapor is 0. 59, air beingunity. In the use of ammonia, two distinct systems are employed. Sofar, however, as the mere evaporating or refrigerating part of theprocess is concerned, it is the same in both. The object is toevaporate the liquid anhydrous ammonia at such tension and in suchquantity as will produce the required cooling effect. The actualtension under which this evaporation should be effected in anyparticular case depends entirely upon the temperature at which theacquirement of heat is to take place, or, in other words, on thetemperature of the material to be cooled. The higher the temperature, the higher may be the evaporating pressure, and therefore the higheris the density of the vapor, the greater the weight of liquidevaporated in a given time, and the greater the amount of heatabstracted. On the other hand, it must be remembered that, as in thecase of water, the lower the temperature of the evaporating liquid, the higher is the heat of vaporization. It is in the method ofsecuring the rejection of heat during condensation of the vapor thatthe two systems diverge, and it will be convenient to consider each ofthese separately. 

_The Absorption Process. _--The principle employed in this process isphysical rather than mechanical. Ordinary ammonia liquor of commerceof 0. 880 specific gravity, which contains about 38 per cent. By weightof pure ammonia and 62 per cent. Of water, is introduced into a vesselnamed the generator. This vessel is heated by means of steamcirculating through coils of iron piping, and a mixed vapor of ammoniaand water is driven off. This mixed vapor is then passed into a secondvessel, in order to be subjected to the cooling action of water. Andhere, owing to the difference between the boiling points of water andammonia, fractional condensation takes place, the bulk of the water, which condenses first, being caught and run back to the generator, while the ammonia in a nearly anhydrous state is condensed andcollected in the lower part of the vessel. 

This process of fractional condensation is due to Rees Reece, andforms an important feature in the modern absorption machine. Prior tothe introduction of this invention, the water evaporated in thegenerator was condensed with the ammonia, and interfered veryseriously with the efficiency of the process by reducing the power ofthe refrigerating agent by raising its boiling point. In the improvedform of apparatus, ammonia is obtained in a nearly anhydrouscondition, and in this state passes on to the refrigerator. In thisvessel, which is in communication with another vessel called theabsorber, containing cold water or very weak ammonia liquor, evaporation takes place, owing to the readiness with which cold wateror weak liquor absorbs the ammonia, water at 59° Fahr. Absorbing 727times its volume of ammonia vapor. The heat necessary to effect thisvaporization is abstracted from brine or other liquid, which iscirculated through the refrigerator by means of a pump. Owing to theabsorption of ammonia, the weak liquor in the absorber becomesstrengthened, and it is then pumped back into the generating vessel tobe again dealt with as above described. 

The absorption apparatus, as applied for cooling purposes, consists ofa generator, which is a vessel of cast iron containing coils of ironpiping to which steam at any convenient pressure is supplied; ananalyzer, in which a portion of the water vapor is condensed, and fromwhich it flows back immediately into the generator; a rectifier andcondenser, in the upper portion of which a further condensation ofwater vapor and a little ammonia takes place, the liquid thus formedpassing back by a pipe to the analyzer and thence to the generator, while in the lower portion the ammonia vapor is condensed andcollected; and a refrigerator or cooler, into which the nearlyanhydrous liquid obtained in the condenser is admitted by a pipe andregulating valve, and allowed to evaporate, the upper portion being incommunication with the absorber. 

Through this vessel weak liquor, which has been deprived of itsammonia in the generator, is continually circulated, after being firstcooled in an economizer by an opposite current of strong cold liquorpassing from the absorber to the generator, while, in addition, theliquor in the absorber, which would become heated by the liberation ofheat due to the absorption and consequent liquefaction of the ammoniavapor, is still further cooled by the circulation of cold water. Asthe pressure in the absorber is much lower than that in the generator, the strong liquor has to be pumped into the latter vessel, and forthis purpose pumps are provided. Though of necessity the variousoperations have been described separately, the process is a continuousone, strong liquor from the absorber being constantly pumped into thegenerator through the heater or economizer, while nearly anhydrousliquid ammonia is being continually formed in the condenser, thenevaporated in the refrigerator and absorbed by the cool weak liquorpassing through the absorber. 

Putting aside the effect of losses from radiation, etc. , all the heatexpended in the generator will be taken up by the water passingthrough the condenser, less that portion due to the condensation ofthe water vapor in the analyzer, and plus the amount due to thedifference between the temperature of the liquid as it enters thegenerator and the temperature at which it leaves the condenser. In therefrigerator the liquid ammonia, in becoming vaporized, will take upthe precise quantity of heat that was given off during its cooling andliquefaction in the condenser, plus the amount due to the differencein heat of vaporization, owing to the lower pressure at which thechange of state takes place in the refrigerator, and less the smallamount due to the difference in temperature between the vapor enteringthe condenser and that leaving the refrigerator, less also the amountnecessary to cool the liquid ammonia to the refrigerator temperature. When the vapor enters into solution with the weak liquor in theabsorber, the heat taken up in the refrigerator is imparted to thecooling water, subject also to corrections for differences of pressureand temperature. The sources of loss in such an apparatus are:

a. Radiation and conduction of heat from all vessels and pipes abovenormal temperature, which can, to a large extent, be prevented bylagging. 

b. Conduction of heat from without into all vessels and pipes that arebelow normal temperature, which can also to a large extent beprevented by lagging. 

c. Inefficiency of economizer, by reason of which heat obtained by theexpenditure of steam in the generator is passed on to the absorber andthere uselessly imparted to the cooling water. 

d. The entrance of water into the refrigerator, due to the liquidbeing not perfectly anhydrous. 

e. The useless evaporation of water in the generator. With regard tothe amount of heat used, it will have been seen that the whole of thatrequired to vaporize the ammonia, and whatever water vapor passes offfrom the generator, has to be supplied from without. Owing to the factthat the heating takes place by means of coils, the steam passedthrough may be condensed, and thus each pound can be made to give upsome 950 units of heat. With the absorption process worked by anefficient boiler, it may be taken that 200, 000 thermal units per hourmay be eliminated by the consumption of about 100 lb. Of coal perhour, with a brine temperature in the refrigerator of about 20° Fahr. 

_Compression Process. _--In this process ammonia is used in itsanhydrous form. So far as the action of the refrigerator is concerned, it is precisely the same as it is in the case of the absorptionapparatus, but instead of the vapor being liquefied by absorption bywater, it is drawn from the refrigerator by a pump, by means of whichit is compressed and delivered into the condenser at such pressure asto cause its liquefaction at the temperature of the cooling water. Itmust be borne in mind, however, that allowance must be made for therise of temperature of the water passing through the condenser, andalso for the difference in temperature necessary in order to permitthe transfer of heat from one side of the cooling surface to theother. In a compression machine the work applied to the pump may beaccounted for as follows:

a. Friction. 

b. Heat rejected during compression and discharge. 

c. Heat acquired by the ammonia in passing through the pump. 

d. Work expended in discharging the compressed vapor from the pump. 

But against this must be set the useful mechanical work performed bythe vapor entering the pump. The heat rejected in the condenser is theheat of vaporization taken up in the refrigerator, less the amount dueto the higher pressure at which the change in physical state occurs, plus the heat acquired in the pump, and less the amount due to thedifference between the temperature at which the vapor is liquefied inthe condenser and that at which it entered the pump. An ammoniacompression machine, as applied to ice making, contains ice-makingtanks, in which is circulated a brine mixture, uncongealable at anytemperature likely to be reached during the process. This brine alsocirculates around coils of wrought iron pipes, in which the liquidammonia passing from the condenser is vaporized, the heat required forthis vaporization being obtained from the brine. A pump draws off theammonia vapor from the refrigerator coils, and compresses it into thecondenser, where, by means of the combined action of pressure andcooling by water, it assumes a liquid form, and is ready to be againpassed on to the refrigerator for evaporation. The ammonia compressionprocess is more economical than the absorption process, and with agood boiler and engine about 240, 000 thermal units per hour can beeliminated by the expenditure of 100 lb. Of coal per hour, with abrine temperature in the refrigerator of about 20° Fahr. 

GENERAL CONSIDERATIONS. 

From what has been said, it will have been seen that, so far as themere application is concerned, there is no difference whatever betweenthe absorption and compression processes. The followingconsiderations, therefore, which chiefly relate to the application ofrefrigerating apparatus, will be dealt with quite independent ofeither system. The application of refrigerating apparatus may roughlybe divided into the following heads:

a. Ice making. 

b. The cooling of liquids. 

c. The cooling of stores and rooms. 

_Ice Making. _--For this purpose two methods are employed, known as thecan and cell systems respectively. In the former, moulds of tinnedsheet copper or galvanized steel of the desired size are filled withthe water to be frozen, and suspended in a tank through which brinecooled to a low temperature in the refrigerator is circulated. As soonas the water is completely frozen, the moulds are removed, and dippedfor a long time into warm water, which loosens the blocks of ice andenables them to be turned out. The thickness of the blocks exercisesan important influence upon the number of moulds required for a givenoutput, as a block 9 in. Thick will take four or five times as long tofreeze solid as one of only 3 in. In the cell system a series ofcellular walls of wrought or cast iron are placed in a tank, thedistance between each pair of walls being from 12 to 16 in. , accordingto the thickness of the block required. This space is filled with thewater to be frozen. Cold brine circulates through the cells, and theice forms on the outer surfaces, gradually increasing in thicknessuntil the two opposite layers meet and join together. If thinnerblocks are required, the freezing process may be stopped at any timeand the ice removed. In order to detach the ice it is customary to cutoff the supply of cold brine and circulate brine at a highertemperature through the cells. Ice frozen by either of the abovedescribed methods from ordinary water is more or less opaque, owing tothe air liberated during the freezing process, little bubbles of whichare caught in the ice as it forms, and in order to produce transparentice it is necessary that the water should be agitated during thefreezing process in such a way as to permit the air bubbles to escape. With the can system this is generally accomplished by means of armshaving a vertical or horizontal movement. These arms are eitherwithdrawn as the ice forms, leaving the block solid, or they are madeto work backward and forward in the center of the moulds, dividing theblock vertically into two pieces. With the cell system agitation isgenerally effected by making a communication between the bottom ofeach water space and a chamber below, in which a paddle or wood pistonis caused to reciprocate. The movement thus given to the water in thechamber is communicated to that in the process of being frozen, andthe small bubbles of air are in this way detached and set free. Theice which first forms on the sides of the moulds or cells is, as arule, sufficiently transparent even without agitation. The opacityincreases toward the center, where the opposing layers join, and itis, therefore, more necessary to agitate toward the end of thefreezing process than at the commencement. As the capacity for holdingair in solution decreases if the temperature of the water is raised, less agitation is needed in hot than in temperate climates. Experiments have been made from time to time with the view ofproducing transparent ice from distilled water, and so dispensing withagitation. In this case the cost of distilling the water will have tobe added to the ordinary working expenses. 

_Cooling of Liquids. _--In breweries, distilleries, butter factories, and other places where it is desired to have a supply of water orbrine for cooling and other purposes at a comparatively lowtemperature, refrigerating machines may be advantageously applied. Inthis case the liquid is passed through the refrigerator and thenutilized in any convenient manner. 

_Cooling of Rooms. _--For this purpose the usual plan is to employ acirculation of cold brine through rows of iron piping, placed eitheron the ceiling or on the walls of the rooms to be cooled. In this, asin the other cases where brine is used, it is employed merely as amedium for taking up heat at one place and transferring it to theammonia in the refrigerator, the ammonia in turn completing theoperation by giving up the heat to the cooling water duringliquefaction in the condenser. The brine pipes cool the adjacent air, which, in consequence of its greater specific gravity, descends, beingreplaced by warmer air, which in turn becomes cold, and so the processgoes on. Assuming the air to be sufficiently saturated, which isgenerally the case, some of the moisture in it is condensed and frozenon the surface of the pipes; and if the air is renewed in whole or inpart from the outside, or if the contents of the chamber are wet, thedeposit of ice in the pipes will in time become so thick as tonecessitate its being thawed off. This is accomplished by turning acurrent of warm brine through the pipes. Another method has beenproposed, in which the brine pipes are placed in a separatecompartment, air being circulated through this compartment to therooms, and back again to the cooling pipes in a closed cycle by meansof a fan. This plan was tried on a large scale by Mr. Chambers at theVictoria Docks, but for some reason or other was abandoned. Onedifficulty is the collection of ice from the moisture deposited fromthe air, which clogs up the spaces between the pipes, besidesdiminishing their cooling power. This, in some cases, can be partiallyobviated by using the same air over again, but in most instancesspecial means would have to be provided for frequent thawing off, thepipes having, on account of economy of space and convenience, to beplaced so close together, and to be so confined in surface, that theyare much more liable to have their action interfered with than whenplaced on the roof or walls of the room. 

In addition to the foregoing there are, of course, many otherapplications of ammonia refrigerating machines of a more or lessspecial nature, of which time will not permit even a passingreference. Many of these are embraced in the second class, cold wateror brine being used for the cooling of candles, the separation ofparaffin, the crystallization of salts, and for many other purposes. In the same way cold brine has been used with great success forfreezing quicksand in the sinking of shafts, the excavation beingcarried out and the watertight tubing or lining put in while thematerial is in a solid state. In a paper such as this it would bequite impracticable to enter into details of construction, and theauthor has therefore confined himself chiefly to principles ofworking. In conclusion, however, it may be added that in ammoniamachines, whether on the absorption or compression systems, no copperor alloy of copper can be used in parts subjected to the action of theammonia. Cast or wrought iron and steel may, however, be used, provided the quality is good, but special care must be taken in theconstruction of those parts of absorption machines which are subjectedto a high temperature. In both classes of apparatus first-classmaterials and workmanship are most absolute essentials. 

 * * * * *

[Continued from Supplement, No. 646, p. 10319. ]

ELEMENTS OF ARCHITECTURAL DESIGN. [1]

 [Footnote 1: Delivered before the Society of Arts, London, December 13, 1887. From the _Journal_ of the Society. ]

BY H. H. STATHAM. 

III. --CONTINUED. 

The Romans, in their arched constructions, habitually strengthened thepoint against which the vault thrust by adding columnar features tothe walls, as shown in Fig. 108; thus again making a false use of thecolumn in a way in which it was never contemplated by those whooriginally developed its form. In Romanesque architecture the columnwas no longer used for this purpose; its place was taken by a flatpilaster-like projection of the wall (plan and section, Fig. 109), which gave sufficient strength for the not very ambitious vaultedroofs of this period, where often in fact only the aisles werevaulted, and the center compartment covered with a wooden roof. Atfirst this pilaster-like form bore a reminiscence of a classic capitalas its termination; a moulded capping under the eaves of the building. Next this capping was almost insensibly dropped, and the buttressbecame a mere flat strip of wall. As the vaulting became bolder andmore ambitious, the buttress had to be made more massive and ofgreater projection, to afford sufficient abutment to the vault, moreespecially toward the lower part, where the thrust of the roof iscarried to the ground. Hence arose the tendency to increase theprojection of the buttress gradually downward, and this was done bysuccessive slopes or "set-offs, " as they are termed, which assisted(whether intentionally or not in the first instance) in further aidingthe correct architectural expression of the buttress. Then thevaulting of the center aisle was carried so high and treated in sobold a manner, with a progressive diminution of the wall piers (as thetaste for large traceried windows developed more and more), that aflying buttress (see section, Fig. 110) was necessary to take thethrust across to the exterior buttresses, and these again, under thisadditional stress, were further increased in projection, and were atthe same time made narrower (to allow for all the window space thatwas wanted between them), until the result was that the masses ofwall, which in the Romanesque building were placed longitudinally andparallel to the axis of the building, have all turned about (Fig. 110, plan) and placed themselves with their edges to the building to resistthe thrust of the roofing. The same amount of wall is there as in theRomanesque building, but it is arranged in quite a new manner, inorder to meet the new constructive conditions of the complete Gothicbuilding. 

[Illustration: Figs. 108-114. ]

It will be seen thus how completely this important and characteristicfeature of Gothic architecture, the buttress, is the outcome ofpractical conditions of construction. It is treated decoratively, butit is itself a necessary engineering expedient in the construction. The application of the same principle, and its effect uponarchitectural expression, may be seen in some other examples besidesthat of the buttress in its usual shape and position. The wholearrangement and disposition of an arched building is affected by thenecessity of providing counterforts to resist the thrust of arches. The position of the central tower, for instance, in so many cathedralsand churches, at the intersection of the nave and transepts, is notonly the result of a feeling for architectural effect and thecentralizing of the composition, it is the position in which also thetower has the cross walls of nave and transepts abutting against it inall four directions: if the tower is to be placed over the centralroof at all, it could only be over this point of the plan. In theNorman buildings, which in some respects were finer constructions thanthose of later Gothic, the desire to provide a firm abutment for thearches carrying the tower had a most marked effect on thearchitectural expression of the interior. At Tewkesbury, for instance, while the lower piers are designed in the usual way toward the northand south sides (viz. , as portions of a pier of nearly squareproportion standing under the angle of the tower), in the east andwest direction the tower piers run out into great solid masses ofwall, in order to insure a sufficient abutment for the tower arches. On the north and south sides the solid transept walls were availableimmediately on the other side of the low arch of the side aisle, buton the east and west sides there were only the nave and choir arcadesto take the thrust of the north and south tower arches, and so theNormans took care to interpose a massive piece of wall between, inorder that the thrust of the tower arches might be neutralized beforeit could operate against the less solid arcaded portions of the walls. This expedient, this great mass of wall introduced solely forconstructive reasons, adds greatly to the grandeur of the interiorarchitectural effect. The true constructive and architecturalperception of the Normans in this treatment of the lower piers isillustrated by the curious contrast presented at Salisbury. There thetower piers are rather small, the style is later, and the massivebuilding of the Normans had given way to a more graceful but lessmonumental manner of building. Still the abutment of the tower archeswas probably sufficient for the weight of the tower as at first built;but when the lofty spire was put on the top of this, its verticalweight, pressing upon the tower arches and increasing their horizontalthrust, actually thrust the nave and choir arcades out of theperpendicular toward the west and east respectively, and there theyare leaning at a very perceptible angle away from the center of thechurch--the architectural expression, in a very significant form, ofthe neglect of balance of mass in construction. 

But while the buttress in Gothic architecture has been in process ofdevelopment, what has the vault been doing? We left it (Fig. 92) inthe condition of a round wagon vault, intersected by another similarvault at right angles. By that method of treatment we got rid of thecontinuous thrust on the walls. But there were many difficulties to befaced in the construction of vaulting after this first step had beentaken, difficulties which arose chiefly from the rigid andunmanageable proportions of the circular arch, and which could not beeven partially solved till the introduction of the pointed arch. Thepointed arch is the other most marked and characteristic feature ofGothic architecture, and, like the buttress, it will be seen that itarose entirely out of constructive difficulties. 

These difficulties were of two kinds; the first arose from thetendency of the round arch, when on a large scale and heavilyweighted, to sink at the crown if there is even any very slightsettlement of the abutments. If we turn again to diagram 77, andobserve the nearly vertical line formed there by the joints of thekeystone, and if we suppose the scale of that arch very much increasedwithout increasing the width of each voussoir, and suppose it built intwo or three rings one over the other (which is really theconstructive method of a Gothic arch), we shall see that these jointsin the uppermost portion of the arch must in that case become stillmore nearly vertical; in other words, the voussoirs almost lose thewedge shape which is necessary to keep them in their places, and avery slight movement or settlement of the abutments is sufficient tomake the arch stones lose some of their grip on each other and sinkmore or less, leaving the arch flat at the crown. There can be nodoubt that it was the observance of this partial failure of the roundarch (partly owing probably to their own careless way of preparing thefoundations for their piers--for the mediæval builders were very badengineers in that respect) which induced the builders of the earlytransitional abbeys, such as Furness and Fountains and Kirkstall, tobuild the large arches of the nave pointed, though they still retainthe circular-headed form for the smaller arches in the same buildings, which were not so constructively important. This is one of theconstructive reasons which led to the adoption of the pointed arch inmediæval architecture, and one which is easily stated and easilyunderstood. The other influence is one arising out of the lengthenedconflict with the practical difficulties of vaulting, and is a rathermore complicated matter, which we must now endeavor to follow out. 

[Illustration: Figs. 93-107. ]

Looking at Fig. 92, it will be seen that in addition to theperspective sketch of the intersecting arches, there is drawn under ita plan, which represents the four points of the abutment of the arches(identified in plan and perspective sketch as A, B, C, D), and thelines which are taken by the various arches shown by dotted lines. Looking at the perspective sketch, it will be apparent that theintersection of the two cross vaults produces two intersecting arches, the upper line of which is shown in the perspective sketch (marked _e_and _f_); underneath, this intersection of the two arches, which forms afurrow in the upper side of the construction, forms an edge whichtraverses the space occupied by the plan of the vaulting as twooblique arches, running from A to C and from B to D on the plan. Although these are only lines formed by the intersection of two crossarches, still they make decided arches to the eye, and form prominentlines in the system of vaulting; and in a later period of vaultingthey were treated as prominent lines and strongly emphasized bymouldings; but in the Roman and early Romanesque vaults they weresimply left as edges, the eye being directed rather to the vaultingsurfaces than to the edges. The importance of this distinction betweenthe vaulting surfaces and their meeting edges or _groins_[2] will beseen just now. The edges, nevertheless, as was observed, do formarches, and we have therefore a system of cross arches (A B and C D[3]Fig. 95), two wall arches (A, D and B C), and two oblique arches (A Cand B D), which divide the space into four equal triangular portions;this kind of vaulting being hence called _quadripartite_ vaulting. Inthis and the other diagrams of arches on this page, the cross archesare all shown in positive lines, and the oblique arches in dottedlines. 

 [Footnote 2: A _groin_ is the edge line formed by the meeting and intersection of any two arched surfaces. When this edge line is covered and emphasized by a band of moulded stones forming an arch, as it were, on this edge, this is called a _groin rib_. ]

 [Footnote 3: The "D" seems to have been accidentally omitted in this diagram; it is of course the fourth angle of the plan. ]

We have here a system in which four semicircular arches of the widthof A B are combined with two oblique arches of the width of A C, springing from the same level and supposed to rise to the same height. But if we draw out the lines of these two arches in a comparativeelevation, so as to compare their curves together, we at once find weare in a difficulty. The intersection of the two circular archesproduces an ellipse with a very flat crown, and very liable to fail. If we attempt to make the oblique arch a segment only of a largecircle, as in the dotted line at 94, so as to keep it the same levelas the other without being so flat at the top, the crown of the archis safer, but this can only be done at the cost of getting a queertwist in the line of the oblique arch, as shown at D, Fig. 93. Thelike result of a twist of the line of the oblique arch would occur ifthe two sides of the space we are vaulting over were of differentlengths, i. E. If the vaulting space were otherwise than a square, aslong as we are using circular arches. If we attempt to make theoblique arches complete circles, as at Fig. 96, we see that they mustnecessarily rise higher than the cross and side arches, so that theroof would be in a succession of domical forms, as at Fig. 97. Thereis the further expedient of "stilting" the cross arches, that is, making the real arch spring from a point above the impost and buildingthe lower portion of it vertical, as shown in Fig. 98. This device ofstilting the smaller arches to raise their crowns to the level ofthose of the larger arches was in constant use in Byzantine and earlyRomanesque architecture, in the kind of manner shown in the sketch, Fig. 99; and a very clumsy and makeshift method of dealing with theproblem it is; but something of the kind was inevitable as long asnothing but the round arch was available for covering contiguousspaces of different widths. The whole of these difficulties wereapproximately got over in theory, and almost entirely in practice, bythe adoption of the pointed arch. By its means, as will be seen inFig. 100, arches over spaces of different widths could be carried tothe same height, yet with little difference in their curves at thespringing, and without the necessity of employing a dangerously flatelliptical form in the oblique arch. A sketch of the Gothic vault inthis form, and as the intersection of the surfaces of pointed vaults, is shown in Fig. 101. 

But now another and most important change was to come over the vault. The mediæval architects were not satisfied with the mere edge left bythe Romans in their vaults, and even before the full Gothic period theRoman builders had emphasized their oblique arches in many cases byponderous courses of moulded or unmoulded stone in the form ofvaulting ribs. These, in the case of Norman building, were probablynot merely put for the purpose of architectural expression, but alsobecause they afforded an opportunity of concealing behind the lines ofa regularly curved groin rib the irregular curves which were reallyformed by the junction of the vaulting surfaces. But when the vaultbecome more manageable in its curves after the adoption of the pointedarch, the groin rib became adopted in the early pointed vaulting as ameans of giving expression and carrying up the lines of thearchitectural design. On its edge were stones moulded with the deepundercut hollows of early English moulding, defining the curves of theoblique as well as of the cross arches with strongly marked lines, and, moreover, falling on a level with each other in architecturalimportance; the oblique vault of the arch is no longer a secondaryline in the vaulting design; on the contrary, the cross arches areusually omitted, as shown in Figs. 102 and 103 (view and plan of anearly Gothic quadripartite vault); so that the cross rib, which, inthe early Romanesque wagon vault (Fig. 90), was the one marked line onthe vaulting surface, has now been obliterated, and the line of theoblique arch (E F, Figs. 102, 103) has taken its place. 

The effect of the strongly marked lines of the groin ribs, radiatingfrom the cap of the shaft which was their architectural support, seemsto have been so far attractive to the mediæval builders that they soonendeavored to improve upon it and carry it further by multiplying thegroin ribs. One of the stages of this progress is shown in Figs. 104, 105. Here it will be seen that the cross rib is again shown, and thatintermediate ribs have been introduced between it and the oblique rib. The richness of effect of the vault is much heightened thereby; but avery important modification in the mode of constructing it has beenintroduced. As the groin ribs become multiplied, it came to be seenthat it was easier to construct them first, and fill in the spacesafterward; accordingly the groin, instead of being, as it was in theearly days of vaulting, merely the line formed by the meeting of twoarch surfaces, became a kind of stone scaffolding or frame work, between which the vaulting surfaces were filled in with lightermaterial. This arrangement of course made an immense difference in thewhole principle of constructing the vault, and rendered it much moreductile in the hands of the builder, more capable of taking any formwhich he wished to impose on it, than when the vault was regarded andbuilt as an intersection of surfaces. There was still one difficulty, however, one slight failure both practical and theoretical in thevault architecture, which for a long time much exercised the minds ofthe builders. The ribs of the vaulting being all of unequal length, they had to assume different curves almost immediately on rising fromthe impost; and as the mouldings of the ribs have to be run into eachother ("mitered" is the technical term) on the impost, there not beingroom to receive them all separately, it was almost impossible to getthem to make their divergence from each other in a completelysymmetrical manner; the shorter ribs with the quicker curves partedfrom each other at a lower point than the larger ones, and the"miters" occurred at unequal heights. The effort to get over thisunsatisfactory and irregular junction of the ribs at the springing wasmade first by setting back the feet of the shorter ribs on the impostcapping, somewhat in the rear of the feet of the larger ribs, so as tothrow their parting point higher up; but this also was only amakeshift, which it was hoped the eye would pass over; and in fact itis rarely noticeable except to those who know about it and look forit. Still the defect was there, and was not got over until the ideaoccurred of making all the ribs of the same curvature and the samelength, and intercepting them all by a circle at the apex of thevault, as shown in Figs. 106, 107; the space between the circles atthe apex of the vault being practically a nearly flat surface or_plafond_ held in its place by the arches surrounding it; though, foreffect, it is often treated otherwise in external appearance, beingdecorated by pendants giving a reversed curve at this point, but whichof course are only ornamental features hung from the roof. If we lookagain at Fig. 104, we shall see that this was a very naturaltransition after all, for the arrangement of the ribs and vaultingsurfaces in that example is manifestly suggestive of a form radiatinground the central point of springing, though it only suggests that, and does not completely realize it. But here came a further and verycurious change in the method of building the vault, for as the ribswere made more numerous, for richness of effect, in this form ofvaulting, it was discovered that it was much easier to build the wholeas a solid face of masonry, working the ribs on the face of it. Thusthe ribs, which in the intermediate period were the constructiveframework of the vault, in the final form of fan vaulting came back totheir original use as merely a form of architectural expression, meantto carry on the architectural lines of the design; and they perform, on a larger scale and with a different expression, much the same kindof function which the fluting lines performed in the Greek column. Thefan vault is therefore a kind of inverted dome, built up in courses onmuch the same principle as a dome, but a convex curve internally, instead of a concave one, the whole forming a series of invertedconoid forms abutting against the wall at the foot and against eachother at their upper margins. This form of roof is wonderfully rich ineffect, and has the appearance of being a piece of purely artisticwork done for the pleasure of seeing it; yet, as we have seen, it isin reality, like almost everything good in architecture, the logicaloutcome of a contention with structural problems. 

We have already noticed the suggestion, in early Gothic or Romanesque, of the dividing up of a pier into a multiple pier, of which each partsupports a special member of the superstructure, as indicated in Fig. 90. The Gothic pier, in its development in this respect, affords astriking example of that influence of the superstructure on the planwhich has before been referred to. The peculiar manner of building thearch in Gothic work led almost inevitably to this breaking up of thepier into various members. The Roman arch was on its lower surface asimple flat section, the decorative treatment in the way of mouldingsbeing round the circumference, and not on the under side or _soffit_of the arch, and in early Romanesque work this method was stillfollowed. The mediæval builders, partly in the first instance becausethey built with smaller stones, adopted at an early period the plan ofbuilding an arch in two or more courses or rings, one below andrecessed within the other. As the process of moulding the arch stonesbecame more elaborated, and a larger number of subarches one withinanother were introduced, this characteristic form of subarches becamealmost lost to the eye in the multiplicity of the mouldings used. Butup to nearly the latest period of Gothic architecture this form maystill be traced, if looked for, as the basis of the arrangement of themouldings, which are all formed by cutting out of so many squaresections, recessed one within the other. This will be more fullydescribed in the next lecture. We are now speaking more especially ofthe pier as affected by this method of building the arches in recessedorders. If we consider the effect of bringing down on the top of asquare capital an arch composed of two rings of squared stones, thelower one only half the width (say) of the upper one, it will beapparent that on the square capital the arch stones would leave aportion of the capital at each angle bare, and supporting nothing. [4]This looks awkward and illogical, and accordingly the pier is modifiedso as to suit the shape of the arch. Figs. 111, 112, 113, and 114, with the plans, B C D, accompanying them, illustrate this developmentof the pier. Fig. 111 is a simple cylindrical pier with a coarselyformed capital, a kind of reminiscence of the Doric capital, with aplain Romanesque arch starting from it. Fig. 112, shown in plan at B, is the kind of form (varied in different examples) which the pierassumed in Norman and early French work, when the arch had beendivided into two recessed orders. The double lines of the arch areseen springing from the cap each way, in the elevation of the pier. Ifwe look at the plan of the pier, we see that, in place of the singlecylinder, it is now a square with four smaller half cylinders, one oneach face. Of these, those on the right and left of the plan supportthe subarches of the arcade; the one on the lower side, which we willsuppose to be looking toward the nave, supports the shaft whichcarries the nave vaulting, and which stands on the main capital with asmall base of its own, as seen in Fig. 112--a common feature in earlywork; and the half column on the upper side of the plan supports thevaulting rib of the aisle. In Fig. 113 and plan C, which represents apier of nearly a century later, we see that the pier is broken up byperfectly detached shafts, each with its own capital, and eachcarrying a group of arch mouldings, which latter have become moreelaborated. Fig. 114 and plan D show a late Gothic fourteenth centurypier, in which the separate shafts have been abandoned, or ratherabsorbed into the body of the pier, and the pier is formed of a numberof moulded projections, with hollows giving deep shadows between them, and the capitals of the various members run into one another, forminga complete cap round the pier. This pier shows a remarkable contrastin every way to B, yet it is a direct development from the latter. Inthis late form of pier, it will be observed that the projection, E, which carries the vaulting ribs of the nave, instead of springing fromthe capital, as in the early example, Fig. 111, springs from thefloor, and runs right up past the capital; thus the plan of thevaulting is brought, as it were, down on to the floor, and theconnection between the roofing of its building and its plan is ascomplete as can well be. In Fig. 113 the vaulting shaft is supposed tostop short of the capital and to spring from a corbel in the wall, situated above the limit of the drawing. This was a common arrangementin the "Early English" and "Early Decorated" periods of Gothic, but itis not so logical and complete, or so satisfactory either to the eyeor to the judgment, as starting the vaulting shaft from the floorline. The connection between the roofing and the plan may be furtherseen by looking at the portion of a mediæval plan given under Fig. 110, where the dotted lines represent the course of the groin ribs ofthe roof above. It will be seen how completely these depend upon theplan, so that it is necessary to determine how the roof in a vaultedbuilding is to be arranged before setting out the ground plan. 

 [Footnote 4: This was illustrated by diagrams on the wall at the delivery of the lecture. ]

Thus we see that the Gothic cathedral, entirely different in its formfrom that of the Greek temple, illustrates, perhaps, even morecompletely than the Greek style, the same principle of correct andtruthful expression of the construction of the building, and that allthe main features which give to the style its most striking andpicturesque effects are not arbitrarily adopted forms, but are theresult of a continuous architectural development based on thedevelopment of the construction. The decorative details of the Gothicstyle, though differing exceedingly from those of the Greek, are, likethe latter, conventional adaptations of suggestions from nature; andin this respect again, as well as in the character of the mouldings, we find both sides illustrating the same general principle in thedesign of ornament, in its relation to position, climate, andmaterial; but this part of the subject will be more fully treated ofin the next lecture. 

We have now arrived at a style of architectural construction andexpression which seems so different from that of Greek architecture, which we considered in the last lecture, that it is difficult torealize at first that the one is, in regard to some of its mostimportant features, a lineal descendant of the other. Yet this isunquestionably the case. The long thin shaft of Gothic architecture isdescended, through a long series of modifications, from the singlecylindrical column of the Greek; and the carved mediæval capital, again, is to be traced back to the Greek Corinthian capital, throughexamples in early French architecture, of which a tolerably completeseries of modifications could be collected, showing the gradual changefrom the first deviations of the early Gothic capital from itsclassical model, while it still retained the square abacus and thescroll under the angle and the symmetrical disposition of the leaves, down to the free and unconstrained treatment of the later Gothiccapital. Yet with these decided relations in derivation, what adifference in the two manners of building! The Greek building iscomparatively small in scale, symmetrical and balanced in its maindesign, highly finished in its details in accordance with apreconceived theory. The Gothic building is much more extensive inscale, is not necessarily symmetrical in its main design, and thedecorative details appear as if worked according to the individualtaste and pleasure of each carver, and not upon any preconceivedtheory of form or proportion. In the Greek building all thepredominant lines are horizontal; in the mediæval building they arevertical. In the Greek building every opening is covered by a lintel;in the Gothic building every opening is covered by an arch. No twostyles, it might be said, could be more strongly contrasted in theirgeneral characteristics and appearance. Yet this very contrast onlyserves to emphasize the more strongly the main point which I have beenwishing to keep prominent in these lectures--that architecturaldesign, rightly considered, is based on and is the expression of planand construction. In Greek columnar architecture the salient featureof the style is the support of a cross lintel by a vertical pillar;and the main effort of the architectural designer is concentrated ondeveloping the expression of the functions of these two essentialportions of the structure. The whole of the openings being bridged byhorizontal lintels, the whole of the main lines of the superstructureare horizontal, and their horizontal status is as strongly marked aspossible by the terminating lines of the cornice--the whole of thepressures of the superstructure are simply vertical, and the whole ofthe lines of design of the supports are laid out so as to emphasizethe idea of resistance to vertical pressure. The Greek column, too, has only one simple office to perform, that of supporting a singlemass of the superstructure, exercising a single pressure in the samedirection. In the Gothic building the main pressures are oblique andnot vertical, and the main feature of the exterior substructure, thebuttress, is designed to express resistance to an oblique pressure;and no real progress was made with the development of the arched styleuntil the false use of the apparent column or pilaster as a buttresswas got rid of, and the true buttress form evolved. On the interiorpiers of the arcade there is a resolution of pressures whichpractically results in a vertical pressure, and the pier remainsvertical; but the pressure upon it being the resultant of a complexcollection of pressures, each of these has, in complete Gothic, itsown apparent vertical supporting feature, so that the plan of thesubstructure becomes a logical representation of the main features andpressures of the superstructure. The main tendency of the pointedarched building is toward vertically, and this vertical tendency isstrongly emphasized and assisted by the breaking up of the reallysolid mass of the pier into a number of slender shafts, which, bytheir strongly marked parallel lines, lead the eye upward toward theclosing-in lines of the arcade and of the vaulted roof which forms theculmination of the whole. The Greek column is also assisted in itsvertical expression by the lines of the fluting; but as the object ofthese is only to emphasize the one office of the one column, they arestrictly subordinate to the main form, are in fact merely a kind ofdecorative treatment of it in accordance with its function. In theGothic pier the object is to express complexity of function, and thepier, instead of being a single fluted column, is broken up into avariety of connected columnar forms, each expressive of its ownfunction in the design. It may be observed also that the Gothicbuilding, like the Greek, falls into certain main divisions arisingout of the practical conditions of its construction, and which form akind of "order" analogous to the classic order in a sense, though notgoverned by such strict conventional rules. The classic order has itscolumnar support, its beam, its frieze for decorative treatment. TheGothic order has its columnar support, its arch (in place of thebeam), its decoratively treated stage (the triforium), occupying thespace against which the aisle roof abuts, and its clerestory, orwindow stage. All these arise as naturally out of the conditions andhistorical development of the structure in the Gothic case as in theGreek one, but the Greek order is an external, the Gothic an internalone. The two styles are based on constructive conditions totallydifferent the one from the other; their expression and character aretotally different. But this very difference is the most emphaticdeclaration of the same principle, that architectural design is thelogical, but decorative, expression of plan and construction. 

 * * * * *

THE METEOROLOGICAL STATION ON MT. SANTIS. 

[Illustration]

At the second International Meteorological Congress, in 1879, theerection of an observatory on the top of a high mountain wasconsidered. The Swiss Meteorological Commission undertook to carry outthe project, and sent out circulars to different associations, governments, and private individuals requesting single or yearlycontributions to aid in defraying the expense of the station. InDecember, 1881, an extra credit of about $1, 000 was granted by theBundesversammlung for the initial work on the station, which wastemporarily placed in the Santis Hotel, and a telegraph was put upbetween that place and Weisbad in August, 1882, so that on September 1of the same year the meteorological observations were begun. 

At the end of August, 1885, this temporary arrangement expired, andthe enterprise could not be carried on unless the support of the samewas undertaken by the Union. On March 27, 1885, the Bundesversammlungdecided to take the necessary steps. Mr. Fritz Brunner, who died May1, 1885, left a large legacy for the enterprise, making it possible tobuild a special observatory. 

For this purpose the northeast corner of the highest rocky peak wasblasted out and the building was so placed that the wall of rock atthe rear formed an excellent protection from the high west winds. Bythe first of October, last year, the building was ready for occupancy, and there was a quiet opening at which Mr. Potch, director of the BlueHill Observatory, near Boston, and others were present. 

The building is 26 feet long, 19 feet deep, and 30 feet high, and isvery solid and massive, having been built of the limestone blastedfrom the rock. It consists of a ground floor containing the telegraphoffice, the observers' work room, and the kitchen and store rooms; thefirst story, in which are the living and sleeping rooms for theobservers and their assistants; and the second story, living andsleeping rooms for visiting scientists who come to make specialobservations, and a reserve room. The barometer and barograph areplaced in the second story, at a height of about 8, 202 feet above thelevel of the sea, whereas in the hotel they were only about 8, 093 feetabove the sea level. The flat roof, of wood and cement, which extendsvery little above the plateau of the mountain top, is admirablyadapted for making observations in the open air. All the rooms in thehouse are ceiled with wood, and the walls and floors of the groundfloor and first story and the ceilings of the second story are coveredwith insulating material. The cost of the building, including theequipments, amounted to about $11, 200. 

The fact that since the erection of the Santis station there has beena still higher station constructed on Sonnblick (10, 137 feet high)does not decrease the value of the former, for the greater the numberof such elevated stations, the better will be the meteorologicalinvestigations of the upper air currents. The present observer atSantis is Mr. C. Saxer, who has endured the hardships and privationsof a long winter at the station. 

The anemometer house, which is shown in our illustration, is connectedwith the main house by a tunnel. Several times during the day recordsare taken of the barometer, the thermometer, the weather vane, as wellas notes in regard to the condition of the weather, the clouds, fallof rain or snow, etc. A registering aneroid barometer marks thepressure of the atmosphere hourly, and two turning thermometersregister the temperature at midnight and at four o'clock in themorning. --_Illustrirte Zeitung. _

 * * * * *

THE CARE OF THE EYES. [1]

 [Footnote 1: From a paper by David Webster. M. D. , professor of ophthalmology in the New York Polyclinic and surgeon to the Manhattan Eye and Ear Hospital, New York. ]

BY PROF. DAVID WEBSTER, M. D. 

"The light of the body is the eye. " Of all our senses, sight, hearing, touch, taste, and smell, the sight is that which seems to us the mostimportant. Through the eye, the organ of vision, we gain moreinformation and experience more pleasure, perhaps, than through any orall our other organs of sense. Indeed, we are apt to depreciate thevalue of our other senses when comparing them to the eyesight. It isnot uncommon to hear a person say, "I would rather die than be blind. "But no one says, "I would rather die than lose my hearing. " As amatter of fact, the person who is totally blind generally appears tobe more cheerful, happier, than one who is totally deaf. Deaf mutesare often dull, morose, quick tempered, obstinate, self-willed, anddifficult to get along with, while the blind are not infrequentlydistinguished for qualities quite the reverse. It is worthy of remarkthat the eye is that organ of sense which is most ornamental as wellas useful, and the deprivation of which constitutes the most visibledeformity. But it is unnecessary to enter into a comparison of therelative value of our senses or the relative misfortune of our loss ofany one of them. We need them all in our daily struggle for existence, and it is necessary to our physical and mental well-being, as well asto our success in life, that we preserve them all in as high a degreeof perfection as possible. We must not lose sight of the fact that allour organs of sense are parts of one body, and that whatever we do toimprove or preserve the health of our eyes cannot do harm to any otherorgan. We shall be able to "take care of our eyes" more intelligentlyif we know something of their structure and how they perform theirfunctions. The eye is a hollow globe filled with transparent materialand set in a bony cavity of the skull, which, with the eyelids andeyelashes, protect it from injury. It is moved at will in everydirection by six muscles which are attached to its surface, and islubricated and kept moist by the secretions of the tear gland andother glands, which secretions, having done their work, are carrieddown into the nose by a passage especially made for the purpose--thetear duct. We are all familiar with the fact that our eyes are "to seewith, " but in order to be able to take care of our eyes intelligently, it is necessary to understand as far as possible how to see with them. 

THE BACK WALL OF THE EYE. 

It is a remarkable fact that every object we see has its pictureformed upon the back wall of our eyes. The eye is a darkened chamber, and the whole of the front part of it acts as a lens to bring the raysof light coming from objects we wish to see to a focus on its backwall, thus forming a picture there as distinct as the picture formedin the camera obscura of the photographer. This has not only beenproved by the laws of optics, but has been actually demonstrated inthe eyes of rabbits and other animals. Experimenters have held anobject before the eye of a rabbit for a few moments, and have thenkilled the animal and removed the eye as quickly as possible, and laidits back wall bare, and have distinctly seen there the picture of theobject upon which the eye had been fixed. It is a truly wonderful factthat these pictures upon the back wall of the eye can be changed sorapidly that the picture of the object last looked at disappears in aninstant and makes way for the picture of the next. We know that thepicture formed on the back wall of the eye is carried back to thebrain by the optic nerve, but there our knowledge stops. Sciencecannot tell us how the brain, and through it the mind, completes theact of seeing. It is there that the finite and the infinite touch, and, as our minds are finite, we cannot comprehend the infinite. 

But there is enough that we can understand, and it shall be myendeavor in this paper to make some plain statements that will help asa guide in the preservation of those wonderful and useful organs. 

FAR AND NEAR SIGHTEDNESS. 

We have to use our eyes for near and far distant vision. In gatheringpictures of distant objects the normally shaped eye puts forth littleor no effort. It is the near work, such as reading, sewing, ordrawing, that puts a real muscular strain upon the eyes. There arecertain rules that apply to the use of the eyes for such near workregardless of the age of the person. 

READING. 

1. In reading, a book or newspaper should be held at a distance offrom ten to fifteen inches from the eyes. It is hardly necessary tocaution anybody not to hold the print further away than fifteeninches. The only objection to holding ordinary print too far away isthat in so doing the pictures formed on the back wall of the eye aretoo small to be readily and easily perceived, and the close attentionconsequently necessary causes both the eyes and the brain to tire. Most persons quickly find this out themselves, and the tendency israther to hold the book too near, for the nearer the object to theeye, the larger its picture upon the retina, or back eye wall. Buthere we encounter another danger. The nearer the object the eyes areconcentrated upon, the greater the muscular effort necessary; so thatby holding the book too near, the labor of reading is greatlyincreased, and the long persistence in such a habit is likely toproduce weak eyes, and may, in some instances, lead to realnear-sightedness. When children are observed to have acquired thishabit and cannot be persuaded out of it, they should always be takento a physician skilled in the treatment of the eye for examination andadvice. A little attention at such a time may save them from a wholelifetime of trouble with their eyes. Of course, the larger the print, the farther it may be held from the eyes. 

POSITION. 

2. The position of the person with regard to the light should be sothat the latter will fall upon the page he is reading, and not uponhis eyes. It is generally considered most convenient to have the lightshine over the left shoulder, so that in turning the leaves of thebook, the shadow of the hand upon the page is avoided. It is notalways possible to do this, however, and, at the same time, to getplenty of light upon the page. When one finds himself compelled toface the light in reading, or in standing at a desk bookkeeping, heshould always contrive to shade his eyes from a direct light. This maybe done with a large eye shade projecting from the brow. A friend ofmine, a physician, is very fond of reading by a kerosene lamp, thelamp being placed on a table by his side, and the direct light keptfrom his eyes by means of a piece of cardboard stuck up by the lampchimney. 

PROPER LIGHT. 

3. The illumination should always be sufficient. Nothing is moreinjurious to the eyes than reading by a poor light. Many personsstrain their eyes by reading on into the twilight as long as theypossibly can. They become interested and do not like to leave off. Some read in the evening at too great a distance from the source oflight, forgetting that the quantity of light diminishes as the squareof the distance from the source of light increases. Thus, at fourfeet, one gets only one-sixteenth part of the light upon his page thathe would at one foot. It is the duty of parents and others who havecharge of children to see to it that they do not injure their eyes byreading by insufficient light, either daylight or artificial light. There is a common notion that electric light is bad for the eyes. Theonly foundation I can think of for such a notion is that it is tryingto the eyes to gaze directly at the bright electric light. It is badto gaze long at any source of light, and the brighter the source oflight gazed at, the worse for the eyes, the sun being the worst ofall. I have seen more than one person whose eyes were permanentlyinjured by gazing at the sun, during an eclipse or otherwise. As amatter of fact, nothing short of sunlight is better than theincandescent electric light to read by or to work by. 

READING IN BED. 

As to reading while lying down in bed or on a lounge, I can see noobjection to it so far as the eyes are concerned, provided the book isheld in such a position that the eyes do not have to be rolled downtoo far. Unless the head is raised very high by pillows, however, itwill be found very fatiguing to hold the book high enough, not tomention the danger of falling asleep, and of upsetting the lamp orcandle, and thus setting the bed on fire. Many persons permanentlyweaken their eyes by reading to pass away the tedious hours duringrecovery from severe illness. The muscles of the eyes partake of thegeneral weakness and are easily overtaxed. Persons in this conditionmay be read to, but should avoid the active use of their own eyes. 

READING IN RAIL CARS. 

Reading while in the rail cars or in omnibuses is to be avoided. Therapid shaking, trembling or oscillating motion of the cars makes itvery difficult to keep the eyes fixed upon the words, and is verytiresome. I have seen many persons who attributed the failure of theireyes to the daily habit of reading while riding to and from the city. Children should be cautioned against reading with the head inclinedforward. The stooping position encourages a rush of blood to the head, and consequently the eyes become congested, and the foundations fornear-sightedness are laid. 

(_To be continued. _)

 * * * * *

TESTING INDIGO DYES. 

The author deals with the question whether a sample of goods is dyedwith indigo alone or with a mixture of indigo and other blue coloringmatters. His method may be summarized as follows: Threads of thematerial in question should give up no coloring matter to boilingwater. Alcohol at 50 and at 95 per cent. (by volume) ought to extractno color, even if gently warmed (not boiled). Solution of oxalic acidsaturated in the cold, solution of borax, solution of alum at 10 percent. , and solution of ammonium molybdate at 33-1/3 per cent. Oughtnot to extract any coloring matter at a boiling heat. The boraxextract, if subsequently treated with hydrochloric acid, should notturn red, nor become blue on the further addition of ferric chloride. Solutions of stannous chloride and ferric chloride with the aid ofheat ought entirely to destroy the blue coloring matter. Glacialacetic acid on repeated boiling should entirely dissolve the coloringmatter. If the acetic extracts are mixed with two volumes of ether andwater is added, so as to separate out the ether, the water shouldappear as a slightly blue solution, the main bulk of the indigoremaining in suspension at the surface of contact of the ethereal andwatery stratum. This acid watery stratum should be colorless, andshould not assume any color if a little strong hydrochloric acid isallowed to fall into it through the ether. No sulphureted hydrogenshould be evolved on boiling the yarn or cloth in strong hydrochloricacid. On prolonged boiling, supersaturation with strong potassa inexcess, heating and adding a few drops of chloroform, no isonitrileshould be formed. --_W. Lenz_. 

 * * * * *

THE SCIENTIFIC AMERICAN

ARCHITECTS AND BUILDERS EDITION. 

$2. 50 a Year. Single Copies, 25 cts. 

This is a Special Edition of the Scientific American, issuedmonthly--on the first day of the month. Each number contains aboutforty large quarto pages, equal to about two hundred ordinary bookpages, forming, practically, a large and splendid MAGAZINE OFARCHITECTURE, richly adorned with _elegant plates in colors_ and withfine engravings, illustrating the most interesting examples of modernArchitectural Construction and allied subjects. 

A special feature is the presentation in each number of a variety ofthe latest and best plans for private residences, city and country, including those of very moderate cost as well as the more expensive. Drawings in perspective and in color are given, together with fullPlans, Specifications, Costs, Bills of Estimate, and Sheets ofDetails. 

No other building paper contains so many plans, details, andspecifications regularly presented as the Scientific American. Hundreds of dwellings have already been erected on the various planswe have issued during the past year, and many others are in process ofconstruction. 

Architects, Builders, and Owners will find this work valuable infurnishing fresh and useful suggestions. All who contemplate buildingor improving homes, or erecting structures of any kind, have beforethem in this work an almost _endless series of the latest and bestexamples_ from which to make selections, thus saving time and money. 

Many other subjects, including Sewerage, Piping, Lighting, Warming, Ventilating, Decorating, Laying out of Grounds, etc. , are illustrated. An extensive Compendium of Manufacturers' Announcements is also given, in which the most reliable and approved Building Materials, Goods, Machines, Tools, and Appliances are described and illustrated, withaddresses of the makers, etc. 

The fullness, richness, cheapness, and convenience of this work havewon for it the LARGEST CIRCULATION of any Architectural publication inthe world. 

MUNN & CO. , PUBLISHERS, 361 BROADWAY, NEW YORK. 

 * * * * *

A Catalogue of valuable books on Architecture, Building, Carpentry, Masonry, Heating, Warming, Lighting, Ventilation, and all branches ofindustry pertaining to the art of Building, is supplied free ofcharge, sent to any address. 

 * * * * *

BUILDING PLANS AND SPECIFICATIONS. 

In connection with the publication of the Building Edition of theSCIENTIFIC AMERICAN, Messrs. Munn & Co. Furnish plans andspecifications for buildings of every kind, including Churches, Schools, Stores, Dwellings, Carriage Houses, Barns, etc. 

In this work they are assisted by able and experienced architects. Full plans, details, and specifications for the various buildingsillustrated in this paper can be supplied. 

Those who contemplate building, or who wish to alter, improve, extend, or add to existing buildings, whether wings, porches, bay windows, orattic rooms, are invited to communicate with the undersigned. Our workextends to all parts of the country. Estimates, plans, and drawingspromptly prepared. Terms moderate. Address

MUNN & CO. , 361 BROADWAY, NEW YORK. 

 * * * * *

THESCIENTIFIC AMERICAN SUPPLEMENT. 

PUBLISHED WEEKLY. 

Terms of Subscription, $5 a year. 

Sent by mail, postage prepaid, to subscribers in any part of theUnited States or Canada. Six dollars a year, sent, prepaid, to anyforeign country. 

All the back numbers of The Supplement, from the commencement, January1, 1876, can be had. Price, 10 cents each. 

All the back volumes of The Supplement can likewise be supplied. Twovolumes are issued yearly. Price of each volume, $2. 50 stitched inpaper, or $3. 50 bound in stiff covers. 

COMBINED RATES. --One copy of SCIENTIFIC AMERICAN and one copy ofSCIENTIFIC AMERICAN SUPPLEMENT, one year, postpaid, $7. 00. 

A liberal discount to booksellers, news agents, and canvassers. 

MUNN & CO. , 361 BROADWAY, NEW YORK, N. Y. 

 * * * * *

A New Catalogue of Valuable Papers

Contained in SCIENTIFIC AMERICAN SUPPLEMENT during the past ten years, sent _free of charge_ to any address. MUNN & CO. , 361 Broadway, NewYork. 

 * * * * *

USEFUL ENGINEERING BOOKS

Manufacturers, Agriculturists, Chemists, Engineers, Mechanics, Builders, men of leisure, and professional men, of all classes, needgood books in the line of their respective callings. Our post officedepartment permits the transmission of books through the mails at verysmall cost. A comprehensive catalogue of useful books by differentauthors, on more than fifty different subjects, has recently beenpublished, for free circulation, at the office of this paper. Subjectsclassified with names of author. Persons desiring a copy have only toask for it, and it will be mailed to them. Address, MUNN & CO. , 361 Broadway, New York. 

 * * * * *

PATENTS. 

In connection with the SCIENTIFIC AMERICAN, Messrs. MUNN & CO. Aresolicitors of American and Foreign Patents, have had 42 years'experience, and now have the largest establishment in the world. Patents are obtained on the best terms. 

A special notice is made in the SCIENTIFIC AMERICAN of all inventionspatented through this Agency, with the name and residence of thePatentee. By the immense circulation thus given, public attention isdirected to the merits of the new patent, and sales or introductionoften easily effected. 

Any person who has made a new discovery or invention can ascertain, free of charge, whether a patent can probably be obtained, by writingto MUNN & CO. 

We also send free our Hand Book about the Patent Laws, Patents, Caveats, Trade Marks, their costs, and how procured. Address

MUNN & CO. , 361 BROADWAY, NEW YORK. BRANCH OFFICE, 622 AND 624 F ST. , WASHINGTON, D. C. 

 * * * * *

TRANSCRIBER'S NOTES

1. Simple and obvious typographical errors have been corrected. 

2. In the article "Manufacture of Photosensitive Plates", the original text referred to room U twice. The first instance has been changed to room T. 

3. In the article "An Improved Screw Propeller", the text refers to the propeller in figure A as being four bladed and also two bladed. It is clearly two bladed and the reference to it being four-bladed has been corrected.