The shoe containing a groove (Fig. 6), which we shall see later, made its appearance in Germany in the fifteenth century. From this time, according to our present knowledge, ceases the period of the Roman horseshoe. Its influence, however, lasted a great deal longer, and has even remained until our present day.



Its successor became partly the Arabo-Turkomanic and partly the Southwest European horseshoe.

For the descendants of the Numidian light cavalry, the Roman and old Spanish horseshoe was evidently too heavy for their sandy, roadless deserts, so they made it thinner and omitted the bent-up rim, because it prevented the quick movement of the horse. For the protection of the nail heads the outer margin of the shoe was staved, so as to form a small rim on the outer surface of the shoe, thus preventing the nail heads from being worn and the shoe lost too soon.



A horseshoe of that kind is shown by Fig. 8, which was used in North Africa in the twelfth century, and became the model for all forms of horseshoes of the Mahometan tribes. Even now quite similar shoes (Fig. 9) are made south and east from the Caspian Sea, at the Amu-Darja, in Samarkand, etc., which were probably introduced under Tamerlane, the conqueror of nearly the whole of Asia Minor in the fourteenth century.

The so-called "Sarmatische" (Sarmatian) horseshoe (Figs. 10 and 11), of South Russia, shows in its form, at the same time, traces of the last named shoe, however, greatly influenced by the Mongolian shoe, the "Goldenen Horde," which at the turn of the sixteenth to the seventeenth century played havoc at the Volga and the Aral. The unusual width of the toe, and especially the lightness of the iron, reminds us of the Turkomanic horseshoe, whereas, on the contrary, the large bean-shaped holes, as well as the calks, were furnished through Mongolian influence.



The Sarmatian tribes were principally horsemen, and it is not surprising, therefore, that the coat of arms of the former kingdom of Poland in the second and third quadrate shows a silver rider in armor on a silver running horse shod with golden shoes, and that at present about 1,000 families in 25 lineages of the Polish Counts Jastrzembiec Bolesezy, the so-called "Polnische Hufeisen Adel" (Polish Horseshoe Nobility), at the same time also carried the horseshoe on their coats of arms. The silver horseshoe in a blue field appears here as a symbol of the "Herbestpfardes" (autumnal horse), to which, after the christianization of Poland, was added the golden cross. The noblemen participating in the murder of the holy Stanislaus in 1084 had to carry the horseshoe reversed on their escutcheon.



From the African and Turkomanic horseshoe, through the turning up of the toes and heels, originated later the Turkish, Grecian and Montenegrin horseshoe of the present as shown by Fig. 12.



By the Moorish invasion in Spain, the Spanish-Gothic horseshoeing was also modified, through which the shoe became smooth, staved at the margin, very broad in the toe, and turned up at toe and heel, and at a later period the old open Spanish national horseshoe (Fig. 13) was developed. As we thus see, we can in no way deny the Arabian-Turkish origin of this shoe.



As France had received her whole culture from the south, and as the crusades especially brought the Roman nation in close contact with them for centuries, so it cannot appear strange that the old French horseshoe, a form of which has been preserved by Bourgelat and is represented by Fig. 14, still remained in the smooth, turned up in front and behind, like the shoe of the southern climates, with Asiatic traces, which hold on the ground, the same as all southern shoeing, by the nail heads.



The transit of the German empire, in order to keep up the historical course, once more brings us back to the middle of the fifth century. At this time Attila, the "Godegisel" (gods' scourge), left his wooden capitol in the lowlands near the river Theis, to go to the Roman empire and to the German and Gallican provinces, there to spread indescribable misery to the horrors of judgment day.

The following is a prayer in those days of horror:

"Kleiner Huf, kleines Ross, Krummer Sabel, spitz Geschoss— Blitzesschnell und sattlefest: Schrim uns Herr von Hunnenpest."

We are at present reminded of those times of fright, when during the clearing and tilling of the soil, a small roughly made horseshoe is found in Southern Germany, about as far as the water boundary of the Thuringian forest, and occasionally on, but principally around Augsburg, and in France as far as the Loire.

These shoes, covering the margin or wall of the foot, show slight traces of having been beveled on the lower surface, and contain two bent calks very superficially placed. Occasionally they are sharpened and turned in two directions. The characteristic wide bean-shaped nail holes are conical on the inside, and are frequently placed so near the outer margin of the shoe that from the pressure the hoofs were likely to split open. The nail heads were shaped like a sleigh runner, and almost entirely sunk into the shoe. It evidently was not bent up at the toe, like the old form of these kinds of shoes.

These shoes, according to our conception of to-day, were so carelessly finished that in the scientific circles of historical researches they were, until very recently, looked upon as saddle mountings or something similar, and not as horseshoes.

This shoe was for some time, while it was plentifully found in France, regarded as of Celtic make; but this is certainly not the case, as it is of Hunish and Hungarian "nationalitat" (nationality). An exactly scientific proof, it is true, according to our present knowledge, cannot be furnished; however, it will stand well enough until the error is proved.

This peculiar kind of horseshoe has been found in South Germany and Northeast France, as far as the region of Orleans, where, as it has been proved, the Huns appeared. This, therefore, speaks for their descendants: 1st, the far extended and yet sharply limited places of finding the shoe; 2d, the small size corresponds to the historically proved smallness of the Hunish horse; 3d, the hasty and careless make, which does not indicate that it was made by settled workmen; 4th, the horseshoe (Fig. 15) bespeaks the Hunish workmanship of the present Chinese shoe, which, in making of the nail holes, shows to-day related touches of the productions of the Mongolian ancestors.



Aside from the peculiar shaped nail holes, the characteristic of the Hunish shoe consists in the changes of the calks for summer and winter shoeing, as well as in the sinking of the nail heads. The Huns, therefore, aside from the indistinctly marked attempts of the Romans in this direction, which are the only ones known to me, must be regarded as the inventors not only of the calks, but partly, next to the Normans, also of the sharpened winter shoeing, and of the not unimportant invention of sinking the nail heads observed in Fig. 15.

The Hunish shoeing was therefore an important invention for the Germans. After centuries later, wherever horseshoeing was practiced, it was done solely according to Hunish methods; whereby the shoe was very possibly made heavier, was more carefully finished and in course of time showed an attempt to bend the toe (Fig. 16a).



In the Bomberg Dom we find an equestrian statue, not unknown in the history of art, which was formerly held to be that of Emperor Conrad III. At present however the opinion prevails generally that it represents "Stephen I., den Heiligen" (Stephen I., the Saint).

Stephen I., the first king of Hungary, formerly was a heathen, and was named "Najk." He reigned from 997 to 1038. His important events were the many victorious wars led against rebellious chieftains of his country, and he was canonized in 1087. His equestrian monument in Bomberg Dom was, in consequence, hardly made before the year 1087. Notwithstanding that the Huns had been defeated 500 years before on the plains of Catalania, the horse of the above mentioned monument carries, as I have convinced myself personally, Hunish horseshoes, modified, however, by blade-shaped calks just then coming into use. This is proof that, at least in Hungary, the Hunish method of shoeing was preserved an extraordinary long time. By this it has not become improbable that at least the many shoes of this kind which were found on the Lechfield come, not directly from the Huns, but from their successors, the Hungarians, whose invasions took place in the first half of the tenth century.

About the same time of the Hungarian invasions, the Normans began to disturb the southwestern part of Europe with their Viking expeditions. Their sea kings seem to have been equestrians at very early times, and to have had their horses shod, although perhaps only in winter; at least the excavation of the Viking ship in 1881 disclosed the remains of a horse which was shod. The shoeing consisted only of a toe protection—"Brodder" (Bruder, Brother)—provided with a small sharp calk, and fastened by two nails.

When later, in the year 1130, the Norwegian king Sigard Yorsalafar, during his journey to Jerusalem, entered Constantinople, his horse is said to have carried only the small toe-protecting shoes.

The art of horseshoeing, immediately after the migration of the nations, came near our improvement of the same to-day; especially near the reputed discoveries met with, which consist simply of iron protection for the margin of the hoof, fastened by nails. The heads were sunk into the shoe so as to increase its firmness. Special consideration was given to local and climatic conditions through the introduction of toes and heels.

The mechanism of the hoof also found remarkable consideration, inasmuch as they apparently avoided driving nails too close to the heel end of the shoe. Notwithstanding this early improvement in the art of horseshoeing, the Huns (as stated before) took a prominent part. It appears to have taken a long time after the migration of the nations for shoeing to become general, as is shown by various descriptions of tournaments, pictures of horses, etc.



We will mention in the first place the "Percival des Wolfram von Eschenbach," as well as "Christ von Troies," where there is a great deal said about horses, horse grooms, and tournaments, but nowhere in those works is any mention made of horseshoeing. Likewise is found the horse on the coat of arms of Wolfram von Eschenbach, in the Manessi collection in Paris, which was begun in Switzerland in the fourteenth century; but, although we find this horse most beautifully finished, it was not shod.



During the time of the crusades, 1096-1291, however, there appeared suddenly in Germany a plate-like horseshoe of southern character (Figs. 18 and 19), which was occasionally bent upward at the heel end, and was very heavy. The toe was very broad sometimes, and was also bent upward. In this form we have seen the shoes of the Balkan and Pyrean peninsula. The shoe was remarkably narrow at the heel, and was supplied with calks, which accounts for the highness of the back part of the shoe. Frequently we find one calk set diagonally, but the other drawn out wedge shaped, and sharp; so that there existed a great similarity between this iron shank and that used by Count Einsiedel for winter shoeing. Sometimes both shanks were sharpened in this way, or were provided with blade-shaped calks well set forward. The form of nail holes used was very characteristic of that of the Huns, but they were decidedly smaller and square, as were seen in the African shoe of the twelfth century. The nail heads were slightly sunk, which was according to southern customs.

That this shoe really belongs to the period of the crusades is proved by the numerous horse pictures which have been preserved from that time; of which we will mention the manuscript of Heinrich von Veldecka ("Eneidt")[4] in the year 1180, which belongs to the most valuable parts of German history of art.

[Footnote 4: "Wanderungen des Aeneas" (Travels of Aeneas).]

This south European Hunish horseshoe had remained the standard form during the middle ages and until the thirty years war, at least in South Germany. The shoe was continually improved, and reached its highest point of perfection about the time of the "Bauern-krieg" (Revolution of the Peasants), at a time when, under the leadership of the Renaissance, the whole art of mechanics, and especially that of blacksmithing, had taken an extraordinarily great stride (Figs. 20 and 21).



The shoe (Figs. 22 and 23) is found in Franconia, in all places where, in the sixteenth century, battles had been fought with the rebellious peasants. We may, therefore, be justified in fixing its origin mainly from that period, for which also speaks its high perfection of form. We find here still the bent-up heel and toe (the latter broad and thin) of the south European form.

The staved rim of the Spanish Arabic Turkomanic shoe is observed to be undergoing a change to that of a groove. The broad surface of the shoe evidently led to the beveling of the same, so as to lessen sole pressure. The size of the nail holes remains still like that of the Huns; but the unsunk southern nail heads yet serve to improve the hold on the ground. The calks were next placed forward, perhaps from an uncultivated sense of beauty, or from the high bending up of the hind part of the shoe, which would necessitate a high and heavy unsightly calk.

From this time on horseshoeing in south Germany fell back very quickly, and loses all scientific holds of support after the thirty years war. In the mean time toe protection in the form of a calk had spread from the colder north over southern Germany; whereas this north German invention did not find favor in England in consequence of her mild oceanic climate.



Also, the calks in England, as well as in the southern countries, on the same ground, therefore, with good reason, could at no time be adopted. This did, however, not interfere with the use of the calk in the colder south Germany, where after a use of nearly 1,500 years it has maintained its local and climatic adaptation. Notwithstanding the occasional aping by foreigners, it has remained victorious in its original form, and has been chosen in many countries.

The historical development of the horseshoe in general, from about the time of Emperor Maximilian until the seven years war, furnishes a true picture of the confused condition of things at that period of time, which, to make intelligible, would require a separate and complete treatise. Interesting as it is to the scientist to follow up this development and mode of present German horseshoeing, which, aside from the national toe and calk, is the English form and has become influential, and with full right, for a periodical of this kind further, more comprehensive, statement would under all circumstances take up too much room; therefore I must drop the pen, although reluctantly.



* * * * *



SHEET GLASS FROM MOLTEN METAL.

The present practice in making metal sheets is to cast ingots or slabs and then reduce these by repeated rollings and reheating. Attempts have been previously made to produce sheets directly from molten metal by pouring the metal: (1) between two revolving rollers; or (2) between a revolving wheel and the surface of an inclosing fixed semicircular segment. By these means none but very thin plates could be satisfactorily produced. In this invention by C.M. Pielsticker, London, the machinery consists of a large receiving roller of 5 ft. diameter more or less, and of a length equal to that of the plate to be produced. With this are combined small forming rollers arranged in succession part way round the periphery of the large roller, and revolving at the same rate as the large roller. The rollers can be cooled by a current of water circulating through them. The molten metal flows on to the surface of the large roller and is prevented from escaping sideways by flanges with which the large roller is provided. These flanges embrace the small rollers and are of a depth greater than that of the thickest plate which it is proposed to roll. The distance between the large roller and the small rollers can be adjusted according to the desired thickness of the plate. When dealing with metals of high melting point, such as steel, the first small roller is made of refractory material and is heated from inside by the flame of a blow pipe. The rollers are coated with plumbago or other material to prevent adhesion to the molten metal. In the case of metals of high melting point the machine is fed direct from a furnace divided into two compartments by a wall or bridge in which is a stopper which can be operated so as to regulate the flow of metal. When applied to forming sheets of glass, the rollers should be warmed by a blow pipe flame as above described, and the sheet of glass stretched and annealed as it leaves the last roller.

* * * * *



WELDLESS STEEL CHAINS.

At the Royal Naval Exhibition, London, Messrs. William Reid & Co. are exhibiting their weldless steel chains, which we now illustrate.

Of the many advantages claimed for steel chains, it may be prominently noted that a very important saving of weight is effected on account of their possessing such a high breaking strain, compared with the ordinary welded iron chains. To illustrate this, it may be stated that a given length of the weldless steel chain is 35 to 40 per cent. less in weight than an equivalent length of iron chain, will stand the same breaking strain as the latter, and indeed, where steel of special quality is used in making the weldless chains, this difference can be increased as much as 70 to 80 per cent. Whereas superior iron chains break at a strain at 17 tons per square inch, these weldless steel chains will stand a strain of 28 to 30 tons, with 20 to 26 per cent. elongation.



Again, there is greater security in their use from the fact that there are no welds, and they give warning of the limit of strain to which they can bear being approached, by elongation, which can be carried to a considerable extent before the chain breaks. Moreover, over, in chains made by this process, the links are all exactly alike. Though the life of a weldless steel chain is said to be twice that of an ordinary one, the price per length is little more than that of best iron chains.

They are made in lengths of from 40 to 50 feet, being compressed from a solid rolled steel bar, the section of which is shaped like a four-pointed star. In the first place holes are pierced at intervals down the length of the bar, thus determining the length of the several links. Then the bar is notched between the holes so as to give the external form of the links. The next step is "flattening out," which presses the links into shape on their inner side, but leaves the openings still closed by a plate of metal. They are then stamped out so as to round them up, and the metal inside them is punched out, and the edges "cleaned," or trimmed off. The links are now parted from one another and stamped again, to insure equal thickness in all parts of the chain. The only processes now to be gone through are dressing and finishing. According to the die used, the shape of the links can be varied to suit any required pattern. The lengths of chain thus made are joined by spiral rings made of soft steel, the convolutions being afterward hammered together till they become solid. A ring of this description, 3/4 inch diameter, underwent a strain of 46,200 lb., that is, 23 tons to the square inch, its elongation being 21 per cent.

These chains have passed satisfactorily the tests of the Bureau Veritas, and both that association and Lloyd's have accepted their use on the same conditions and under the same tests as ordinary chains.

So much for the general idea of punching steel chains. We will now describe a recent invention by which superior steel chains are produced, the author of which is Mr. Hippolyte Rongier, of Birmingham, Eng. He says:

My invention has for its object the manufacture of weldless stayed chains, whereof each link, together with its cross strut or stay, is made of one piece of metal without any weld or joint; and the invention consists in producing a chain of stayed links from a bar of cruciform section by the consecutive series of punching, twisting and stamping operations hereinafter described, the punching operations being entirely performed on the metal when in the cold state.

Figs. 1 to 10 show the progressive stages in the manufacture of the chain, and the remaining figures show the series of tools that are employed.

The general method of operation of making stayed chains according to my invention is so far similar to the methods heretofore proposed for making unstayed chains from the bar of cruciform section that the links are formed alternately out of the one and the other pair of diametrically opposite webs of the rod, the links, when severed and completed, being already enchained together at the time of their formation. The successive operations differ, however, in many important practical respects from those heretofore proposed, as will appear from the following detailed description of the successive steps in the process illustrated by Figs. 1 to 10.

I will distinguish the one pair of diametrically opposite webs of the bar and the notches and mortises punched therein and the links formed therefrom from the other pair by an index figure 1 affixed to the reference letters appertaining thereto.

a a are one pair of diametrically opposite webs, and a' a' the other pair of webs of the bar.



The first operation illustrated in Fig. 1 is to punch out of the edge of one of the webs, a, a series of shallow notches, b, at equal intervals apart, corresponding to the pitch of the links to be formed out of that pair of webs and situated where the spaces will ultimately be formed between the ends of that series of links. The notches are made with beveled ends, and are no deeper than is absolutely necessary (for the purpose of a guide stop in the subsequent operations, as hereinafter described), so as to avoid, as far as possible, weakening the bar transversely. This operation is repeated upon one of the pairs of webs a'; but whereas in the first operation of notching the web the "pitch" of the notches is determined by the feed mechanism, in this second operation of notching the notches, b, cut in the web, a, serve as guides to influence and compensate for any inaccuracy of the feed mechanism, so that the second set of notches, b', shall be intermediate of and rigorously equidistant from the first set of notches, b. This compensation is effected by the notches, b, fitting on to a beveled stop on the bed of the punching tool by which the notches, b', are cut, the beveled ends of the notches, b, causing the bar under the pressure of the punch to adjust itself in the longitudinal direction (if necessary) sufficiently to rectify any inaccuracy of feed. These notches, b b', similarly serve as guides to insure uniformity of spacing in the subsequent operations of punching out the links.

The second operation (illustrated in Fig. 2) is to punch out of the pair of opposite webs, a a, pairs of oblong mortises—two pairs, c c, and one pair, d d. These three pairs of mortises (which might be punched at separate operations, but are preferably punched at one stroke of the press) are situated as close as possible up to the faces of the other pairs of webs, a' a', the pairs of mortises, c c, being so spaced as to correspond in position to the eyes of the links to be formed, to which they correspond approximately in form, while the pair, d, correspond in position to the notches, b, and therefore to the intervals by which the links formed out of the same pair of webs, a a, will be separated when completed. This operation is continued along the whole length of the pair of webs, a. It will be observed that a considerable thickness of metal is left at a* between the notches, b, and the mortises, d. This is of primary importance and is one of the essential features of my method of manufacture, inasmuch as by first punching out the mortises, d, the subsequent removal of the metal from between the outer ends of the links is greatly facilitated, while by leaving the solid metal, a*, the transverse strength of the webs, a a, is not materially diminished, so that when the operation of punching the mortises, c and d, in the other pair of webs, a', is performed the bar will not be bent and crippled, as would inevitably be the case were the whole of the metal opposite the notches, b, which is ultimately to be removed, to be punched out at so early a stage of the manufacture. The operation of punching the pairs of mortises, c' and d, having been repeated along the other pair of webs, a', it will be observed that like the notches, b, the mortises, c d, in the one pair of webs alternate with those, c' d', in the other pair of webs.

The third operation (illustrated in Fig. 3) is to elongate the mortises, c d, and bring the mortises, c c', more nearly to the final form. This is performed by punches similar to but larger (in the direction of the length of the rod) than those used in the second operation.

The third operation, which is repeated upon both pairs of webs, a a a' a', may be considered as a second stage of the second operation, it being preferable to punch out the mortises in two stages in order to remove sufficient metal without unduly straining the bar.

The fourth operation (illustrated in Fig. 4) consists in roughly shaping the ends of the links externally by punching out the portions, a*, of the webs, a, between the links lying in the same plane or formed out of the same pair of webs. This operation is repeated on the other pair of webs, a'. Up to this point a continuous core of metal has been left at the intersection of the two pairs of webs.

The fifth operation (illustrated in Fig. 5) consists in punching out the portions, e, of the core at each side of the cross stay of the link, so as to separate the cross stay from the outer ends of the adjacent links. This operation is performed by removing a portion only of the metal of the core which intervenes between the cross stay and the outer ends of the adjacent links enchained with the link under operation—that is to say, portions, e*, of the core are temporarily left attached to the outer ends of the links in order to avoid crippling or bending the bar, which might occur were the whole of this metal, which is ultimately to be removed, to be punched out at once, these portions, e*, being supported by the bed die in the operation of punching out the spaces, e, as hereinafter described. This operation having been repeated upon both pairs of webs, it will be observed that the rod-like form of the chain is now only maintained by the portion of the core at the points, f, where the inner side of the eye or bow of one link is united with that of the next one. The severing of these intervening portions of the core and the breaking up of the rod into the constituent links of the chain constitute the sixth operation.

The sixth operation (illustrated in Fig. 6) is performed by torsion, and for this purpose one end of the rod is held fixed while the other is twisted once or twice in opposite directions, until by fatigue of the metal at the points, f, the whole of the links are severed almost at the same instant, and a chain of roughly formed stayed links is produced.

The seventh operation (illustrated in Fig. 7) is to remove the superfluous projecting pieces of metal both from the inside and outside of the ends of the links. For this purpose the two ends of each link are operated on at the same time by two pairs of punches corresponding to the outline of the ends of the link.

The eighth operation (illustrated in Fig. 8) is to bring the ends of the links to their finished rounded form. This is performed by stamping both ends of each link at the same time between pairs of shaping dies or swages.

The ninth operation (illustrated in Fig. 9) is to bring the middle portion of each link—that is to say, the side members and the cross stay—to the finished rounded form, which is also performed by means of a pair of dies or swages.

The tenth and last operation (illustrated in Fig. 10) is to contract the link slightly in the lateral direction in order to correct any imperfections at the sides left by the two previous operations and bring the link to a more perfect and stronger form, as shown. This operation has the important result of strengthening the link considerably by contracting or rendering more pointed the arched form of the bow or end of the link, and also by thickening the metal at that part where the wear is greatest, this thickening of the metal at the ends of the link occurring in the direction of the line of strain (as indicated by x in Fig. 10) and being brought about by the compression or "upsetting" of the metal at the end of the link. It may be preferable to perform this operation immediately after the seventh operation, and I reserve the right to do so.

In the case of large cables only the metal is preferably heated for the eighth, ninth, and tenth operations.

I will now refer to the figures which illustrate the series of tools whereby the above mentioned operations are performed.

Fig. 1a shows a plan (the punch being in section) and Fig. 1b an elevation of the bed die of the tool by which the notches b of the first operation are performed. The feed mechanism is not shown, but might be of any ordinary intermittent kind. g is a groove in the bed, in which lies the lower vertical web of the rod, of cruciform section, the two horizontal webs lying upon the bed with the edge of the web to be notched lying just over the die, in which works the punch, B, of which B' is the cutting edge. The punch is operated in the usual way, its lower end, which does not rise out of the die, acting as a guide. B* is the beveled stop in the groove, g, which by fitting in the notches, b or b', corrects inaccuracies of the feed.

Fig. 2a is a sectional plan and Fig. 2b an elevation of the tool by which the second operation is performed, the same tool being also used for performing the third operation. (Illustrated in Fig. 3a.) h h are a pair of bed-dies having a space h' between them to receive the lower web of the bar, and having notches, C C and D D, in their inner ends, forming counterparts of the punches by which the pairs of mortises, c d, Fig. 2, are punched in the pair of webs lying upon the bed-dies, h. These bed-dies are fitted to slide a little in opposite directions upon a suitable bed plate and are caused by the inclined cams, i', on the guides, i, of the press head (which pass through corresponding apertures in the bed-dies, h) to approach each other at the moment the punches come down on the work, so as to grip the lower web of the rod and support the pair of webs being operated on close up to the sides of the lower web lying in the space h', while when the punches rise the bed-dies move apart, so that the web is quite free in said space h' and the rod may be easily fed forward for a fresh stroke of the press. B* is the beveled stop in the space, k', as in the tool first described. The bed-dies h have a second set of notches C' D' at their outer ends, similar to but longer than those C D, so that by reversing the bed-dies they will form counterparts for a second set of punches corresponding thereto for performing the third operation—i.e., enlarging the mortises, c d, as represented in Figs. 3 and 3a; or, instead of adapting the dies, h, to perform the two operations, separate tools may be used for the second and third operations.

Fig. 4a is an elevation and Fig. 4b a sectional plan of the tool for performing the fourth operation—namely, removing the portion a*, Figs. 3, 3a, 4a, and 4b. This is done by a pair of punches, A*, corresponding in shape to the ends of the link in the rough and to the aperture shown in the bed-die, k, Fig. 4b, which has a groove, k', to admit the lower web of and to guide the rod. The beveled stop, B*, used in operating on the pair of webs, a, corresponds to the notches, b'; but in operating on the webs, a', the stop must be replaced by one corresponding to the aperture left by the removal of the portion, a*.

Fig. 5a is an elevation, Fig. 5b a plan, and Fig. 5c a longitudinal vertical section of the tool for performing the fifth operation, the work being shown in section in the latter figure. It consists of a bed-die, l, with groove, m, to receive the lower web, but terminating at a distance from the die apertures, so as to leave supports, n, for the parts, e*, of the rod to resist the downward pressure of the punches, E, which remove the portions, e, from each side of the cross stay, as shown in Figs. 5b and 5c. The correct position of the work in regard to the punches is insured by these supporting parts, n, which terminate the grooves, m.

Fig. 6a is an elevation of the winch for performing the sixth operation.

Fig. 7a is an elevation and Fig. 7b a plan of the tool for performing the seventh operation. P P are the punches for trimming the outside and Q Q those for trimming the inside of the ends of the links. The links adjacent to the one to be operated on are brought together into the position shown in dotted lines, the bed-die having an aperture in it to admit of this, so that both ends of the link to be trimmed may be operated on together.

The tool for performing the eighth operation consists of a pair of swages, the bottom one only being shown in Fig. 8a. The swages correspond to the intended rounded sectional form of the ends of the link, which is placed in position between the swages in a similar manner to that described for Fig. 7b, so that both ends are rounded or finished off at once.

Fig. 9a is a plan of the bottom swage of the tool for performing the ninth operation, the upper swage corresponding thereto at least in so far as the middle part of the link to be operated on is concerned.

The tool for performing the tenth operation is represented in elevation and plan in Figs. 10a and 10b. It consists of a pair of bed-dies, R, fitted to slide together and operated by the cams, s, on the guide rods, S, the operation being similar to that of the tool shown in Figs. 2a and 2b, except that there are no punches, and that the link which lies in the cavity of the dies is merely compressed in the lateral direction by the inward motion of the bed-dies.

My invention further comprises a modification of the above described process, which has for its object to enable the weldless stayed links to be made as short and particularly as narrow as may be necessary in order to adapt the chain to run over the sheaves of pulley blocks and to suit other purposes for which short-link welded chain has heretofore only been available.



In the manufacture of chains by the aforesaid process of punching there is a practical minimum limit for the dimensions of the punches which cannot be reduced without compromising their efficiency, and consequently the width (and therefore the length) of the link must necessarily bear a certain proportion to the thickness of the web of metal out of which it is formed, since the breadth of the link depends on the length of the cross stay, which is determined by the breadth of the mortises forming the eyes of the link. The present modification enables these dimensions to be reduced without reducing the dimensions, and consequently the efficiency, of the punches which form the eyes of the link. The modification applies to what I have designated the fifth operation of the above described process; and it consists in punching out the middle of the cross stay (so as to leave only two short stumps jutting inward from the side members of the link), this operation serving to interrupt the continuity of the core, which was the object of the fifth operation. For this purpose I substitute for the pair of punches illustrated in Figs. 5a and 5c a single punch, which removes that part of the "core" of the cruciform bar which is situated at the middle of the strut. This tool is represented in Fig. 11, and the effect of its operation is shown in Fig. 12. The subsequent operations, herein designated the sixth, seventh, eighth, and ninth operations, are performed as hereinbefore described; but the tenth operation has the effect of closing together the two stumps, g g, until they abut together at the middle of the link and together constitute a cross strut or stay, which prevents any further lateral collapse of the link. In the operation of closing up the gap between the stumps, g g, the link is brought to the narrow form shown in Fig. 12, the eyes of the links being only just wide enough to receive the end of the adjacent link enchained therewith without gripping it. This operation is performed by a tool similar to that shown in Figs. 10a and 10b, above referred to.

* * * * *



AN ENGLISH STEAM FIRE ENGINE.

The steam fire engine of which we give an engraving is one specially built for the Indian government by Messrs. Shand, Mason & Co., London. It has the distinction of being the first steam fire engine supplied for the province of Upper Burma, having been purchased primarily for the royal palace, and to serve for the protection of the cantonment of Mandalay. The engine is placed vertically in front of the boiler, and consists of a double acting pump with valves which can be taken out for renewal or examination in two or three minutes. The capacity is 200 gallons per minute, and the height of jet 140 ft. As shown in the engraving, the fore part of the machine forms a hose reel and tool box, and can be instantly separated from the engine to allow of the independent use of the latter at a fire.



The engine is constructed with wrought iron side frames, fore carriage and wheels, and steel axles, springs, etc. The tool box, coachman's seat, and other parts are of teak. It is provided with Messrs. Shand, Mason & Co.'s quick steaming boiler, in which 100 lb. pressure can be raised from cold water in from five to seven minutes, an extra large fire box for burning wood, with fire door at the back, feed pump, and injector, fresh water tank, coal bunker, and other fittings and arrangements for carrying the suction pipe. A pole and sway bars are fitted for two ponies, and wood cross bars to pass over the backs of the animals at the tops of the collars. Two men are carried on the machine, a coachman on the box seat and a stoker on the footboard at the rear of the engine. The whole forms a very light and readily transportable fire engine.—The Engineer.

* * * * *



THE SYSTEM OF MILITARY DOVE COTES IN EUROPE.[1]

[Footnote 1: Continued from Scientific American of July 11, p. 23.]

France.—The history of the aerial postal service and of the carrier pigeons of the siege of Paris has been thoroughly written, and is so well known that it is useless to recapitulate it in this place. It will suffice to say that sixty-four balloons crossed the Prussian lines during the war of 1870-1871, carrying with them 360 pigeons, 302 of which were afterward sent back to Paris, during a terrible winter, without previous training, and from localities often situated at a distance of over 120 miles. Despite the shooting at them by the enemy, 98 returned to their cotes, 75 of them carrying microscopic dispatches. They thus introduced into the capital 150,000 official dispatches and a million private ones reduced by photo-micrographic processes. The whole, printed in ordinary characters, would have formed a library of 500 volumes. One of these carriers, which reached Paris on the 21st of January, 1871, a few days previous to the armistice, carried alone nearly 40,000 dispatches.

The pigeon that brought the news of the victory of Coulmiers started from La Loupe at ten o'clock in the morning on the tenth of November, and reached Paris a few minutes before noon. The account of the Villejuif affair was brought from Paris to Tourcoing (Nord) by a white pigeon belonging to Mr. Descampes. This pigeon is now preserved in a stuffed state in the museum of the city. The carrier pigeon service was not prolonged beyond the 1st of February, and our winged brothers of arms were sold at a low price at auction by the government, which, once more, showed itself ungrateful to its servants as soon as it no longer had need of their services. After the commune, Mr. La Perre de Roo submitted to the president of the republic a project for the organization of military dove cotes for connecting the French strongholds with each other. Mr. Thiers treated the project as chimerical, so the execution of it was delayed up to the time at which we saw it applied in foreign countries.

In 1877, the government accepted a gift of 420 pigeons from Mr. De Roo, and had the Administration of Post Offices construct in the Garden of Acclimatization a model pigeon house, which was finished in 1878, and was capable of accommodating 200 pairs.

At present, the majority of our fortresses contain dove cotes, which are perfectly organized and under the direction of the engineer corps of the army.

The map in Fig. 1 gives the approximate system such as it results from documents consulted in foreign military reviews.

According to Lieutenant Grigot, an officer of the Belgian army, who has written a very good book entitled Science Colombophile, a rational organization of the French system requires a central station at Paris and three secondary centers at Langres, Lyons and Tours, the latter being established in view of a new invasion.

As the distance of Paris from the frontier of the north is but 143 miles at the most, the city would have no need of any intermediate station in order to communicate with the various places of the said frontier. Langres would serve as a relay between Paris and the frontier of the northeast. For the places of the southeast it would require at least two relays, Lyons and Langres, or Dijon.



As Paris has ten directions to serve, it should therefore possess ten different dove cotes, of 720 birds each, and this would give a total of 7,200 pigeons. According to the same principle, Langres, which has five directions to provide for, should have 3,600 pigeons.

Continuing this calculation, we find that it would require 25,000 pigeons for the dove cotes as a whole appropriated to the frontiers of the north, northeast, east, and southeast, without taking into account our frontiers of the ocean and the Pyrenees.



A law of the 3d of July, 1877, supplemented by a decree of the 15th of November, organized the application of carrier pigeons in France.

One of the last enumerations shows that there exist in Paris 11,000 pigeons, 5,000 of which are trained, and, in the suburbs, 7,000, of which 3,000 are trained. At Roubaix, a city of 100,000 inhabitants, there are 15,000 pigeons. Watrelos, a small neighboring city of 10,000 inhabitants, has no less than 3,000 carrier pigeons belonging to three societies, the oldest of which, that of Saint-Esprit, was founded in 1869.

In entire France, there are about 100,000 trained pigeons, and forty-seven departments having pigeon-fancying societies.

Germany.—After the war of 1870, Prussia, which had observed the services rendered by pigeons during the siege of Paris, was the first power to organize military dove cotes.

In the autumn of 1871, the Minister of War commissioned Mr. Leutzen, a very competent amateur of Cologne, to study the most favorable processes for the recruitment, rearing, and training of carrier pigeons, as well as for the organization of a system of stations upon the western frontier.

In 1872, Mr. Bismarck having received a number of magnificent Belgian pigeons as a present, a rearing station was established at the Zoological Garden of Berlin, under the direction of Dr. Bodinas.

In 1874 military dove cotes were installed at Cologne, Metz, Strassburg, and Berlin. Since that time there have been organized, or at least projected, about fifteen new stations upon the frontier of France, upon the maritime coasts of the north, or upon the Russian frontier.

Berlin remains the principal rearing station, with two pigeon houses of 500 pigeons each; but it is at Cologne that is centralized the general administration of military dove cotes under Mr. Leutzen's direction. The other stations are directly dependent upon the commandant of the place, under the control of the inspector of military telegraphy. The Wilhelmshaven dove cote, by way of exception, depends upon the Admiralty. In each dove cote there is a subofficer of the engineer corps and an experienced civil pigeon fancier, on a monthly salary of ninety marks, assisted by two orderlies. In time of war, this personnel has to be doubled and commanded by an officer.

The amount appropriated to the military dove cotes, which in 1875 was about 13,000 francs, rose in 1888 to more than 60,000 francs.

As a rule, each dove cote should be provided with 1,000 pigeons, but this number does not appear to have been yet reached except at Thorn, Metz, and Strassburg.

Germany has not confined herself to the organization of military dove cotes, but, like other nations, has endeavored to aid and direct pigeon fancying, so as to be able, when necessary, to find ready prepared resources in the civil dove cotes. The generals make it their duty to be present, as far as possible, at the races of private societies, and the Emperor awards gold medals for flights of more than 120 miles.

On the 13th of January, 1881, nineteen of these societies, at the head of which must be placed the Columbia, of Cologne, combined into a federation. At the end of the year the association already included sixty-six societies. On the 1st of December, 1888, it included seventy-eight, with 52,240 carrier pigeons ready for mobilization.

The first two articles of the statutes of the Federation are as follows:

"I. The object of the Federation is to unite in one organization all societies of pigeon fanciers in order to improve the service of carrier pigeons, which, in case of war, the country must put to profit.

"II. The Federation therefore proposes: (a) To aid the activity of pigeon-fancying societies and to direct the voyages of the societies according to a determined plan; (b) to form itinerent societies and on this occasion to organize expositions and auction sales of pigeons; (c) to maintain relations with the Prussian Minister of War; (d) to obtain diminutions and favors for transportation; (e) to make efforts for the extermination of vultures; (f) to obtain a legal protection for pigeons; and (g) to publish a special periodical for the instruction of fanciers."

Italy.—The first military dove cote in Italy was installed in 1876 at Ancona by the twelfth regiment of artillery. In 1879, a second station was established at Bologna. At present there are in the kingdom, besides the central post at Rome, some fifteen dove cotes, the principal ones of which are established at Naples, Gaeta, Alexandria, Bologna, Ancona and Placenza. There are at least two on the French frontier at Fenestrella and Exilles, and two others in Sardinia, at Cagliari and Maddalena. The complete system includes twenty-three; moreover, there are two in operation at Massoua and Assab.

The cost of each cote amounts to about 1,000 francs. The pigeons are registered and taken care of by a pigeon breeder (a subofficer) assisted by a soldier. The head of the service is Commandant of Engineers Malagoli, one of the most distinguished of pigeon fanciers.

We represent in Fig. 2 one of the baskets used in France for carrying the birds to where they are to be set free.—La Nature.

* * * * *



THE ISLE OF MAN TWIN SCREW STEAMER TYNWALD.

We place on record the details of the first high speed twin screw steamer built for the service. Of this vessel, named the Tynwald, we give a profile and an engraving of stern, showing the method of supporting the brackets for propeller shafting.



The Tynwald is 265 feet long, 34 feet 6 inches beam, and 14 feet 6 inches depth moulded, the gross tonnage being 946 tons. The desire of the owners to put the vessel alternately on two distinct services required special arrangement of the saloons. Running between Liverpool and the island there was no necessity for sleeping accommodation, as the passage is made in about three hours; and the ship had to be suited to carry immense crowds. But as the owners wished on special occasions to run the vessel from Glasgow to Manxland it was necessary to so arrange the saloons as to admit of sleeping accommodation being provided on these occasions. On the Liverpool run the vessel will carry from 800 to 900 passengers. A spacious promenade is an indispensable desideratum, and the upper or shelter deck has been made flush from stem to stern, the only obstructions in addition to the engine and boiler casings, and the deck and cargo working machinery, being a small deck house aft with special state rooms, ticket and post offices, and the companion way to the saloons below. On the main deck forward is a sheltered promenade for second class passengers, while on the lower deck below are dining saloons, the sofas of which may be improvised for sleeping accommodation. At the extreme after end of the main deck is the first class saloon, with the ladies' room forward on the starboard side, and, there being no alley way forward, the ladies' lavatories are provided on the starboard side of the engine casing. On the port side are the gentlemen's lavatories, and smoking saloon and bar. The dining saloon is aft on the lower deck, with ladies' room forward. In the two saloons and ladies' rooms sofa berths can be arranged to accommodate 252 passengers. The crew and petty officers are accommodated in the forward part of the ship. As the profile shows, the vessel is divided by transverse bulkheads into seven watertight compartments, and there are double bottoms. She has six large boats and several rafts.



The twin screws are revolved by separate triple expansion engines, steam being supplied by two double-ended boilers. Each boiler is placed fore and aft, and each has a separate uptake and funnel. There are three stokeholds, and to ventilate them and supply sufficient air for the furnaces there is in each a 6 foot fan driven by an independent engine running at 250 revolutions. These have been supplied by Messrs. W.H. Allen & Co., London. The boilers are of steel and adapted for a working pressure of 160 lb. to the square inch. They are 16 feet in diameter and 18 feet long, and there are eight furnaces in each boiler, sixteen in all, the diameter of each furnace being 3 feet 41/2 inches.

The cylinders of the main engines are 22 in., 36 in., and 57 in. in diameter respectively, with a piston stroke of 3 ft. The high-pressure cylinders are each fitted with a piston valve, and the intermediate and low-pressure cylinders with double-ported slide valves, all of which are worked by the usual double eccentric and link motion valve gear, by which the cut-off can be varied as required. All the shafting is forged of Siemens-Martin mild steel of the best quality, each of the three separate cranks being built up. The condensers are placed at the outsides of the engine room, and the air, feed, and bilge pumps are between the engines and the condensers and worked by levers from the low-pressure engine crosshead. There are two centrifugal pumps, each worked by a separate engine for circulating water through the condenser, and these are so arranged that they can be connected to the bilges in the event of an accident to the ship. In the engine room there is fitted an auxiliary feed donkey of the duplex type and made by the Fairfield Company.

This pump has all the usual connections, so that it can be used for feeding the boilers from the hot well, for filling the fresh water tanks, for pumping from the bilges, or from the sea as a fire engine. The engines are arranged in the ship with the starting platform between them; and the handles for working the throttle valves, starting valves, reversing gear (Brown's combined steam and hydraulic), and drain cocks are brought together at one end of the platform, so that the engineer in charge can readily control both engines. The two sets of engines are bound together by two beams bolted to the framing of each engine. This feature was introduced into the design for steadiness.

The method of supporting the propeller shaft brackets is interesting, and we reproduce a photograph that indicates the arrangement adopted. Instead of the A frame forming part of the same forging as the stern frame, the Fairfield Company have built up the supporting arms of steel plates riveted together, as is clearly shown. There is an advantage in cost and with less risk in undiscovered flaws in material.

An interesting change has been made in the steam pipes. Cases of copper steam pipes bursting when subjected to high pressure have not been infrequent, and Mr. A. Laing, the engineering director on the Fairfield Board, with characteristic desire to advance engineering practice, has been devoting much attention to this question lately. He has made very exhaustive tests with lap welded iron steam pipes of all diameters, but principally of 10 in. diameter and 3/8 in. thickness of material, made by Messrs. A. & J. Stuart & Clydesdale, Limited, and the results have been such as to induce him to introduce these into vessels recently built by the company. It may be stated that the pipes only burst at a hydraulic pressure of 3,000 lb. to the square inches.

The Tynwald was tried on the Clyde about a month ago, and on two runs on the mile, the one with and the other against the tide, the mean speed was 19.38 knots—the maximum was 191/2 knots—and the indicated horse power developed was 5,200, the steam pressure being 160 lb., and the vacuum 28 lb. Since that time the vessel has made several runs from Liverpool and from Glasgow to the Isle of Man, and has maintained a steady seagoing speed of between 18 and 19 knots.—Engineering.

* * * * *



THE TREATMENT OF REFRACTORY ORES.

Mr. Jas. J. Shedlock, with the assistance of Mr. T. Denny, of Australia, has constructed on behalf of the Metallurgical Syndicate, of 105 Gresham House, London, an apparatus on a commercial scale, which, it is said, effects at the smallest expense, and with the best economical results, the entire separation of metals from their ores. In treating ores by this process, the stone is crushed in the usual way, either by rolls or stamps, the crushed ore being conveyed into an apparatus, where each atom is subjected to the action of gases under pressure, whereby the whole of the sulphur and other materials which render the ore refractory are separated. The ore is then conveyed into a vessel containing an absorbing fluid metal, so constructed that every particle of the ore is brought into contact with the metal. For the production of reducing gases, steam and air are passed through highly heated materials, having an affinity for oxygen, and the gases so produced are utilized for raising the ore to a high temperature. By this means the sulphur and other metalloids and base metals are volatilized and eliminated, and the gold in the ore is then in such a condition as to alloy itself or become amalgamated with the fluid metal with which it is brought into close contact. The tailings passing off, worthless, are conveyed to the dump.

The apparatus in the background is that in which the steam is generated, and which, in combination with the due proportion of atmospheric air, is first superheated in passing through the hearth or bed on which the fire is supported. The superheated steam and air under pressure are then forced through the fire, which is automatically maintained at a considerable depth, by which means the products of combustion are mainly hydrogen and carbonic oxide. These gases are then conveyed by means of the main and branch pipes to the cylindrical apparatus in the foreground, into which the ore to be acted upon is driven under pressure by means of the gases, which, being ignited, raise the ore to a high temperature. The ore is maintained in a state of violent agitation. Each particle being kept separate from its fellows is consequently very rapidly acted upon by the gases. The ore freed from its refractory constituents is then fed into a vessel containing the fluid metal, in which each particle of ore is separated from the others, and being acted upon by the fluid metal is absorbed into it, the tailings or refuse passing off freed from any gold which may have been in the ore.



Quantities of refractory ores treated by this process are said to have demonstrated that the whole of the gold in the ore is extracted. The successful outcome of these trials is stated to have resulted in the Anglo-French Exploration Co. acquiring the right to work the process on the various gold fields of South Africa. It is anticipated that the process will thus be immediately brought to a test by means of apparatus erected on the gold fields under circumstances and conditions of absolute practical work. As is well known, gold-bearing ores in South Africa which are below the water line are, by reason of the presence of sulphur, extremely difficult to deal with, and are consequently of small commercial value. The gold in these ores, it is maintained, will, by the new process, be extracted and saved, and make all the difference between successful and unsuccessful mining in that country.

It will have been seen that the peculiar and essential features of the invention consist in subjecting every particle of the ore under treatment to the process in all its stages instead of in bulk, thereby insuring that no portion shall escape being acted upon by the gases and the absorbing metal. This is done automatically and in a very rapid manner. It is stated that this method of treatment is applicable to all ores, the most refractory being readily reducible by its means. The advantages claimed for this process are: simplicity of the apparatus, it being practically automatic; that every particle of the ore is separately acted upon in a rapid and efficient manner; that the apparatus is adaptable to existing milling plants; and that there is an absence of elaborate and expensive plant and of the refinements of electrical or chemical science. These advantages imply that the work can be done so economically as to commend the new process to the favorable consideration of all who are interested in mines or mining property.—Iron.

* * * * *



REFINING SILVER BULLION.

A number of years ago the author devised a method for refining silver bullion by sulphuric acid, in which iron was substituted for copper as precipitant of silver, the principal feature being the separation of pure crystals of silver sulphate. A full description of this process may be found in Percy's Metallurgy, "Silver and Gold," page 479. The process has been extensively worked in San Francisco and in Germany in refining bullion to the amount of more than a hundred million dollars' worth of silver. Its more general application has been hampered, however, by the circumstance that the patent had been secured by one firm which limited itself to its utilization in its California works. The patent having expired, the author lately introduced a modification of the process by which the apparatus and manipulations are greatly cheapened and simplified. In the following account is given a short description of the process in its present shape.

Preparing the Silver Sulphate.—The bullion, containing, essentially, silver, copper and gold, is dissolved by boiling with sulphuric acid in cast iron pots. The difference between the new process and the usual practice consists in the use of a much larger quantity of acid. Thus, in refining ordinary silver "dore," four parts of acid are used to one part of bullion. Of this acid one part is chemically and mechanically consumed in the dissolving process, and the remaining three parts are fully recovered and at once ready for reutilization, as will be described hereafter. In the usual process—understanding thereby, here and in the following, the process practiced at the United States mints, for instance—two parts of acid are employed for one of bullion; all of this is lost, partly through the dissolving and partly in being afterward mixed with water, previous to the precipitation of the silver by copper. Economy in acid being therefore imperative, the silver solution finally becomes much concentrated, and it requires high heat and careful management to finish the solution of the bullion. Bars containing more than about 10 per cent. of copper cannot be dissolved at all, owing to the separation of copper sulphate insoluble in the small amount of free acid finally remaining. The advantage gained by dissolving bullion with abundance of free acid in the improved process is so evident that it merely requires to be pointed out. For bullion containing 20 per cent. of copper the author employs six parts of acid to one of bullion; for baser metal still more acid, and so on, never losing more than the stochiometrical percentage of acid and recovering the remainder. In this description he, however, confines himself to the treatment of ordinary silver ore with less than 10 per cent. of copper.

In the diagram A A represent two refining pots, 4 ft. in diameter and 3 ft. in depth, each capable of dissolving at one operation as much as 400 pounds of bullion. The acid is stored in the cast iron reservoir, B, which is placed on a level sufficiently high to charge into A by gravitation, and is composed of fresh concentrated acid mixed with the somewhat dilute acid regained from a previous operation. After the bullion is fully dissolved all the acid still available is run from B into A A. The temperature and strength are thereby reduced, the fuming ceases, any still undissolved copper sulphate dissolves, and the gold settles. In assuming that the settling of the gold takes place in A itself, the author follows the practice of the United States mints. In private refineries, where refining is carried on continuously, the settling may take place in an intermediate vessel, and A A be at once recharged. Owing to the large amount of free acid present, the temperature must fall considerably before the separation of silver sulphate commences, and sufficient time may be allowed for settling if the intermediate vessel be judiciously arranged.



Separating the Silver Sulphate.—The clarified solution is siphoned off the gold from A A into C, which is an open cast iron pan, say 8 ft. by 4 ft. and 1 ft. deep. It is supported by means of a flange in another larger pan—not shown in the diagram—into which water may be admitted for cooling. Steam is blown into the acid solution, still very hot, as soon as C is filled. The steam is introduced about 1 in. below the surface of the liquid, blowing perpendicularly downward from a nozzle made of lead pipe through an aperture 1/8 in. in diameter. Under these circumstances the absorption of the steam is nearly perfect, and takes place without any splashing. The temperature rises with the increasing dilution, and may be regulated by the less experienced by manipulating the cooling tank. An actual boiling is not desired, because it protracts unnecessarily the operation by the less perfect condensation of the steam. No separation of silver sulphate occurs during this operation (and, consequently, there is no clotting of the steam nozzle), the large amount of free acid, combined with the increase of temperature, compensating for the diminution of the solubility of the sulphate by the dilution. The most important point in this procedure is to know when to stop the admission of steam. To determine this, the operator takes a drop or two of the solution upon a cold iron plate by means of a glass rod and observes whether after cooling the sample congeals partly or wholly into a white mass of silver bisulphate, or whether the silver separates as a monosulphate in detached yellow crystals, leaving a mother liquor behind. As soon as the latter point has been reached, steam is shut off and the solution is allowed to crystallize, cold water being admitted into the outer pan. The operator may now be certain that the liquid will no longer congeal into a soft mass of silver bisulphate, which on contact with water will disintegrate into powder, obstinately retaining a large amount of free acid; but the silver will separate as a monosulphate in hard and large yellow crystals retaining no acid and preserving their physical characteristics when thrown into water. After cooling to, say, 80 deg. F., the silver sulphate will have coated the pan C about 1 in. thick. There will also be found a deposit of copper sulphate when the mother acid, after having been used over and over again, has been sufficiently saturated therewith. Lead sulphate separates in a cloud, which, however, will hardly settle at this stage.

The whole operation just described, which constitutes the most essential feature of the author's improvement upon his old process described in Dr. Percy's work, is a short one, as the acid requires by no means great dilution. The steam has merely to furnish enough water to dilute the free acid present to, say, 62 deg. B. Areometrical determination is, of course, not possible, on account of the dissolved sulphates.

Reducing the Silver Sulphate to Fine Silver.—The mother acid is pumped from C to the reservoir, B, for this purpose an iron pipe connecting the top of B with a recess in the bottom of C. The tank, B, is cast as a closed vessel, with a manhole in the top, which is ordinarily kept closed by an iron plate resting on a rubber packing. The air is exhausted from B by a steam injector, and the acid rises from C and enters B without coming in contact with any valves. The volume of fresh commercial acid necessary for another dissolving operation, say 800 pounds, more or less, for refining 800 pounds of bullion in A A, is lifted from some other receptacle into B in the same manner. The mixture of the two acids in B now represents the volume of acid to be employed for dissolving and settling the next charge of 800 pounds of bullion in A A. In this reservoir, B, the cloud of lead sulphate mentioned above finds an opportunity for settling.

The crystals of silver sulphate are detached from C by an iron shovel and thrown into D. D is a lead lined tank about 4 ft. by 4 ft. and 3 ft. deep. It is divided into two compartments by means of a horizontal, perforated false bottom made of wood. From the lower compartment a lead pipe discharges into the lead lined reservoir, E. Warm distilled water is allowed to percolate the crystals until the usual ammonia test indicates that the copper sulphate has been sufficiently dissolved. Then the outflow is closed, sheets of iron are thrown on and into the crystals, the apparatus is filled with hot distilled water, and steam is moderately admitted into the lower compartment. Ferrous sulphate is formed, and in connection with the iron rapidly reduces the silver sulphate to the metallic state, the reduced silver retaining the heavy compact character of the crystals. When the reaction is completed, as indicated by the chlorine test, the liquid is discharged into E, the iron sheets are removed and the silver is sweetened either in the same vessel, D, or in a special filtering vessel which rests on wheels and may be run directly to the hydraulic press.

The vat, E, is the great reservoir where all liquids holding silver sulphate in solution are collected; for instance, that from sweetening the gold and from washing the tools. Sheets of iron here precipitate all silver and copper, and the resulting solution of ferrous sulphate is, with the usual precautions, discharged into the sewer. Occasionally when copper and silver have accumulated in E in sufficient amount the mass is thrown into D, silver sulphate crystals are added and sheet copper is thrown in, instead of sheet iron. There results a hot, neutral, concentrated solution of copper sulphate, which may be run at once into a crystallizing vat for the separation of commercial crystals of copper sulphate. It will be readily understood, of course, that if there should be any advantage in manufacturing that commercial article, besides the amount prepared as described, which represents merely the copper contained in the bullion, copper sheets may be regularly employed for reducing the silver sulphate in D. The author trusts that the practical refiner will recognize that the manufacture of commercial copper sulphate is thus effected in a more rational and economical manner than by the present method of evaporating from 25 deg. B. to 35 deg. B., and of saturating by oxidized copper, generally in a very incomplete manner, the large amount of free acid left from the refining by the usual process. However, the sale of copper sulphate is but rarely so profitable that a refinery should not gladly dispense with that troublesome and bulky manufacture, especially the government establishments, which, besides, waste much valuable space with the crystallizing vats.

The great saving in sulphuric acid, amounting to about 50 per cent. of the present consumption, has already been pointed out. Another advantage the author merely mentions, namely, the easier condensation of the sulphurous fumes in refineries situated in cities, because the larger amount of acid available for dissolving greatly facilitates working and makes the usual frequent admission of air into the refining pot for the purpose of stirring and testing unnecessary.

The more air is excluded from the refining fumes the easier they can be condensed.

Work may be carried on continuously, the vessels C and D being empty by the time a new solution is finished in A A. Thus, the plant shown in the diagram, covering 26 ft. by 16 ft., allows the refining of 40,000 ounces of fine silver in 24 hours; that is, four charges in A A of 800 pounds each.—F. Gutzkow, Eng. and Mining J.

* * * * *



A CASE OF DROWNING, WITH RESUSCITATION.

By F.A. BURRALL, M.D., New York.

As is usual at this season, casualties from drowning are of frequent occurrence. No class of emergencies is of a more startling character, and I think that a history of the case which I now present offers some peculiar features, and will not be without interest to physicians.

The accident which forms the subject of this paper occurred August 29, 1890, at South Harpswell, Casco Bay, Me., where I was passing my vacation.

At about 9.30 A.M., M. B——, an American, aged eighteen, the son of a fisherman, a young man of steady habits and a good constitution, with excellent muscular development, and who had never before required the aid of a physician, was seen by the residents of the village to fall forward from a skiff into the water and go down with uplifted hands. I could not learn that he rose at all after the first submersion. Two men were standing near a bluff which overlooked the bay, and after an instant's delay in deciding that an accident had occurred, they ran over an uneven and undulating pasture for a distance of two hundred and fifty paces to the shore. One of them, after a quick decision not to swim out to where the young man had fallen in and dive for him, removed trousers and boots and waded out five yards to a boat, which he drew into the shore and entered with his companion, taking him to a yacht which lay two hundred and forty yards from the shore, in the padlocked cabin of which was a boat hook. The padlock was unfastened, the boat hook taken, and they proceeded by the boat directly to where the young man lay. He was seen through the clear water, lying at a depth of nine feet at the bottom of the bay, on his back, with upturned face and arms extended from the sides of the body. He was quickly seized by the boat hook, drawn head upward to the surface, and with the inferior portion of the body hanging over the stern of the boat, and the superior supported in the arms of his rescuer, was rowed rapidly to the shore, where he was rolled a few times, and then placed prone upon a tub for further rolling. I was told that much water came from his mouth. Meantime I had been sent for to where I was sitting, one hundred and fifty-one yards from the scene, and I arrived to find him apparently lifeless on the tub, and to be addressed with the remark, "Well, doctor, I suppose we are doing all that can be done."

I have given these details, as from a study of them I was aided in deciding the time of submersion, as well as the intervals which transpired before the intelligent use of remedies. It is also remarkable that, notwithstanding all which has been written about ready remedies for drowning, no one present knew anything about them, although living in a seafaring community.

I immediately directed that the patient should at once be placed upon the ground, which was sloping, and arranged his rubber boots under the back of the head and nape of the neck, so that the head should be slightly elevated and the neck extended, while the head was turned somewhat upon the side, that fluids might drain from the mouth. The day was clear and moderately warm. Respiration had ceased, but no time was lost in commencing artificial respiration. The patient had on a shirt and pantaloons, which were immediately unbuttoned and made loose, and placing myself at his head, I used the Silvester method, because I was more accustomed to it than any other. It seems to me more easy of application than any other, and I have often found it of service in the asphyxia of the newly born.

The patient's surface was cold, there was extensive cyanosis, and his expression was so changed that he was not recognized by his fellow townsmen, but supposed to be a stranger. The eyelids were closed, the pupils contracted, and the inferior maxilla firmly set against the superior. One of the men who had brought him ashore had endeavored to find the heart's impulse by placing his hand upon the chest, but was unable to detect any motion.

I continued the artificial respiration from 9.45 until 10, when I directed one of his rescuers to make pressure upon the ribs, as I brought the arms down upon the chest. This assistance made expiration more complete. When nature resumed the respiratory act I am unable to say, but the artificial breathing was continued in all its details for three-quarters of an hour, and then expiration was aided by pressure on the chest for half an hour longer. Friction upward was also applied to the lower extremities, and the surface became warm about half an hour after the beginning of treatment.

About twenty minutes after ten, two hypodermic syringefuls of brandy were administered, but I did not repeat this, since I think alcohol is likely to increase rather than diminish asphyxia, if given in any considerable quantity. A thermometer, with the mercury shaken down below the scale, at this time did not rise. At 11.8 the pulse was 82; respiration, 27; temperature, 97.

After a natural respiration had commenced, the wet clothing was removed, and the patient was placed in blankets. Ammonia was occasionally applied to the nostrils, since, although respiration had returned, there was no sign of consciousness; the natural respiration was at first attended by the expulsion of frothy fluid from the lips, which gradually diminished, and auscultation revealed the presence of a few pulmonary rales, which also passed away. There were efforts at vomiting, and pallor succeeded cyanosis; there were also clonic contractions of the flexors of the forearm. The pupils dilated slightly at about one hour after beginning treatment. Unconsciousness was still profound, and loud shouting into the ear elicited no response. Mustard sinapisms were applied to the praecordium, and the Faradic current to the spine.

Coffee was also administered by a ready method which, as a systematic procedure, was, I believe, novel when I introduced it to the profession in the Medical Record, in 1876. I take the liberty of referring to this, since I think it is now sometimes overlooked. It was described as follows:

"A simple examination which any one can make of his own buccal cavity will show that posterior to the last molar teeth, when the jaws are closed, is an opening bounded by the molars, the body of the superior, and the ramus of the inferior maxilla. If on either side the cheek is held well out from the jaw, a pocket, or gutter, is formed, into which fluids may be poured, and they will pass into the mouth through the opening behind the molars, as well as through the interstices between the teeth. When in the mouth they tend to create a disposition to swallow, and by this method a considerable quantity of liquid may be administered."

After I had worked with the patient in the open air, for four and three-quarter hours, he was carried to a cottage near by and placed, still unconscious, in bed. There had been an alvine evacuation during the time in which he lay in the blankets.

Consciousness began to return in the early part of the following morning, and with its advent it was discovered that the memory of everything which had occurred from half an hour previous to the accident, up to the return of consciousness, had been completely obliterated. With this exception the convalescence was steady and uncomplicated, and of about a week's duration. From a letter which I recently received from my patient, I learned that the lapse of memory still remains.

My experience with this case has taught me that, unless the data have been taken very accurately, we cannot depend upon any statements as to the time of submersion in cases of drowning. My first supposition was that my patient had been from thirteen to fifteen minutes under water, but a careful investigation reduced the supposed time by one-half. This makes the time of submersion about six minutes, and that which elapsed before the intelligent use of remedies about three minutes longer.

For a long time the opinion of Sir Benjamin Brodie concerning the presence of water in the lungs of the drowned was accepted, who says "that the admission of water into the lungs is prevented by a spasm of the muscles of the glottis cannot, however, be doubted, since we are unable to account for it in any other manner."

Later experiments made by a committee of the Royal Medico-Chirurgical Society, of London, demonstrated, on the contrary, that "in drowned animals not only were all the air passages choked with frothy fluid, more or less bloody, but that both lungs were highly gorged with blood, so that they were heavy, dark colored, and pitted on pressure, and on being cut exuded an abundance of blood-tinged fluid with many air bubbles in it." Dr. R.L. Bowles[1] also holds that the lungs of the drowned contain water, and supports his views by a list of cases. In his words, "These examples show very conclusively that in cases of drowning in man, water does exist in the lungs, that the water only very gradually and after a long time is effectually expelled, and that it is absolutely impossible that any relief should be afforded in that way by the Silvester method." Dr. Bowles believes that the method of Dr. Marshall Hall is superior to any other in this class of cases. He thinks that on account of the immediate adoption and continued use of the prono-lateral position, this method is more to be trusted than any other for keeping the pharynx clear of obstruction. "It also empties the stomach and gradually clears the lungs of the watery and frothy fluids, and will surely and gently introduce sufficient air at each inspiration to take the place of the fluid which has been expelled." In the light of even my limited experience I cannot but feel that Dr. Bowles' opinion concerning the Silvester method would admit of some modification. This is often the case with very positive statements concerning medical matters. In my own case the Silvester method answered well, but I was much impressed with Dr. Bowles' claims for the Marshall Hall method, and should bear them in mind were I called upon to attend another case of drowning.

[Footnote 1: Resuscitation of the Apparently Drowned, by R.L. Bowles, M.D., F.R.C.P., Medico-Chirurgical Transactions, vol. lxxii., 1889.]

I think it must be admitted that pulling the tongue forward as a means of opening the glottis, which has become a standard treatment in asphyxia, is unscientific, and not warranted by the results of experiments made to determine its value.[2]

[Footnote 2: Dragging on the tongue's tip would not affect its base or the epiglottis sufficiently to make it a praiseworthy procedure. Medico-Chirurgical Transactions, vol. lxxii. See also Medical Record, April 4, 1891. Pulling out the tongue is a mistake, since irritation of nerves of deglutition stops the diaphragm.—Medical Times and Gazetteer.]

Dr. Bowles also believes that "the safety of the patient is most perfectly secured by keeping him on one side during the whole treatment, one lung being thus kept quite free." With the account of my case I have brought forward such views of other writers as it seemed to me would be of practical service and throw light on a subject which is of great importance, since the yearly record of mortality from drowning is by no means inconsiderable. I think, however, that a knowledge of what ought to be done in cases of drowning should be much more generally diffused than is the case at present. It should be one of the items of school instruction, since no one can tell when such knowledge may be of immense importance in saving life, and the time lost in securing medical aid would involve a fatal result.

It is also very desirable that all doubt should be removed, by the decision of competent medical authorities, as to which "ready" method or methods are the best, since there are several in the field. With this should be decided what is the best means for securing patency of the air passages, and, in short, a very careful revision of the treatment now recommended for drowning, in order that there may be no doubt as to the course which should be adopted in such a serious emergency.—Medical Record.

* * * * *



THE STORY OF THE UNIVERSE.[1]

[Footnote 1: Presidential address before the British Association, Cardiff, 1891.]

By Dr. WILLIAM HUGGINS.

The opening meeting of the British Association was held in Park Hall, Cardiff, August 18, where a large and brilliant audience assembled, including, in his richly trimmed official robes, the Marquis of Bute, who this year holds office as mayor of Cardiff. At the commencement of the proceedings Sir Frederick Abel took the chair, but this was only pro forma, and in order that he might, after a few complimentary sentences, resign it to the president-elect, Professor Huggins, the eminent astronomer, who at once, amid applause, assumed the presidency and proceeded to deliver the opening address.

Dr. Huggins said that the very remarkable discoveries in our knowledge of the heavens which had taken place during the past thirty years—a period of amazing and ever-increasing activity in all branches of science—had not passed unnoticed in the addresses of successive presidents; still, it seemed to him fitting that he should speak of those newer methods of astronomical research which had led to those discoveries, and which had become possible by the introduction into the observatory, since 1860, of the spectroscope and the modern photographic plate. Spectroscopic astronomy had become a distinct and acknowledged branch of the science, possessing a large literature of its own, and observatories specially devoted to it. The more recent discovery of the gelatine dry plate had given a further great impetus to this modern side of astronomy, and had opened a pathway into the unknown of which even an enthusiast thirty years ago would scarcely have dared to dream.

HERSCHEL'S THEORY.

It was now some thirty years since the spectroscope gave us for the first time certain knowledge of the nature of the heavenly bodies, and revealed the fundamental fact that terrestrial matter is not peculiar to the solar system, but is common to all the stars which are visible to us. Professor Rowland had since shown us that if the whole earth were heated to the temperature of the sun, its spectrum would resemble very closely the solar spectrum. In the nebulae, the elder Herschel saw portions of the fiery mist or "shining fluid," out of which the heavens and the earth had been slowly fashioned. For a time this view of the nebulae gave place to that which regarded them as external galaxies—cosmical "sand heaps," too remote to be resolved into separate stars, though, indeed, in 1858, Mr. Herbert Spencer showed that the observations of nebulae up to that time were really in favor of an evolutional progress. In 1864 he (the speaker) brought the spectroscope to bear upon them; the bright lines which flashed upon the eye showed the source of the light to be glowing gas, and so restored these bodies to what is probably their true place, as an early stage of sidereal life. At that early time our knowledge of stellar spectra was small. For this reason partly, and probably also under the undue influence of theological opinions then widely prevalent, he unwisely wrote in his original paper in 1864, that "in these objects we no longer have to do with a special modification of our own type of sun, but find ourselves in presence of objects possessing a distinct and peculiar plan of structure." Two years later, however, in a lecture before this association, he took a truer position. "Our views of the universe," he said, "are undergoing important changes; let us wait for more facts with minds unfettered by any dogmatic theory, and, therefore, free to receive the teaching, whatever it may be, of new observations."

THE NEBULAR HYPOTHESIS.

Let them turn aside for a moment from the nebulae in the sky to the conclusions to which philosophers had been irresistibly led by a consideration of the features of the solar system. We had before us in the sun and planets obviously not a haphazard aggregation of bodies, but a system resting upon a multitude of relations pointing to a common physical cause. From these considerations Kant and Laplace formulated the nebular hypothesis, resting it on gravitation alone, for at that time the science of the conservation of energy was practically unknown. These philosophers showed how, on the supposition that the space now occupied by the solar system was once filled by a vaporous mass, the formation of the sun and planets could be reasonably accounted for. By a totally different method of reasoning, modern science traced the solar system backward step by step to a similar state of things at the beginning. According to Helmholtz, the sun's heat was maintained by the contraction of his mass, at the rate of about 220 feet a year. Whether at the present time the sun was getting hotter or colder we did not certainly know. We could reason back to the time when the sun was sufficiently expanded to fill the whole space occupied by the solar system, and was reduced to a great glowing nebula. Though man's life, the life of the race perhaps, was too short to give us direct evidence of any distinct stages of so august a process, still the probability was great that the nebular hypothesis, especially in the more precise form given to it by Roche, did represent broadly, notwithstanding some difficulties, the succession of events through which the sun and planets had passed.