The observation of Mr. D. K. Winder, of Toronto, communicated on page 230, current volume, SCIENTIFIC AMEBICAN-concerning the spectrum lines of the aurora borealis, corres, pond almost perfectly with thoae made by J. A. Angstrom-communicated to the London and Edinburgh Philosophical Mag, azine for September, 1809. The first found a distinct bright line in the yollow, and one faintin the green ; the last found "a single bright line to the left of the lines belonging to the calcium group," wave length 556'7 ; and "traces of three faint bands nearly as far as F." As the wave length of the sodium line D, in the yellow, is 538'8, the line of M. Angstrom is near to it ; and as the line F is in the limit of green and blue, his three faint lines appear to correspond also with the faint line of Mr. Winder in the green. These observations are meritorious and their publication valuable ; however, the conclusions these two gentlemen draw from them are open to criticism. The first named comes to the conclusion that polar light is incandescent oxygen gas ; this I most emphatically deny, as the spectrum lines of oxygen are entirely different. Its brightest line is not in the yellow near D, but ia the red ; and after Mr. Winder's own statement," the dim line in the green," he could " not identify aj belonging to any known, substance." How then he can conclude that it is oxygen gas, I must confess not to comprehend. The lines of oxygen are nine in number, the brightest has a wave length of 015, the next 532, then 513, and 436 millionth of a millimeter ; of the remaining five faint lines none corresponds exactly with the lines observed in the aurora, which exact correspondence is an absolute requisite to Iraw conclusions. Besides this, the hypothesis of the exist- ence of such gas in excess, by decomposition of water, is rather far-fetched, principally when we considerthat.the auroral display sometimes reaches a hight of some 400 miles—far above our clouds or atmosphere—most likely never reached by watery vapor. At the close of the article Mr. Winder attributes the fact that the solar spectrum lines are black to " absorption by passing through a deep luminous stratum of the earth's atmosphere." This is utterly erroneous ; the common well-known explanation why these luminous lines are dark in the solar spectrum being that the solar atmosphere, consisting as it does of the vapors of sodium, lime, iron, etc., also incandescent in the body of the sun, act by interference, as the absorbing medium. The effect of our terrestrial atmosphere on the solar spectrum, and on that of the stars and nebul, is of an entirely different nature. M. Angstrom comes to the conclusion that the northern light is not an electric luminosity, such as is produced in the electric egg or rarefied air, " because the lines he observed in the aurora do not agree with those produced by our experiments with electricity in rarefied air. Very interesting is his statement that he observed these same lines for a whole week in the zodiacal light, and I must confess that I felt at first disappointed that the whole of our electric theory concerning the aurora borealis, and our beautiful lecture-room experiments imitating it in large tubes and egg-shaped glass vessels, partially exhausted from air, were set all at naught by this single observation with the spectroscope. Very soon,' however, it became evident to me that this conclusion arrived at by Angstrom was false, remembering that Masson maintains that the lines of the luminous spectrum, as seen in the spectroscope, depend only partially on the chemical nature of the molecules of the medium which radiates the light and partially on the primitive source originating it. Foucault and Kirchhoff proved, indeed, that the double line seen in luminous sodium vapor, for instance, was not changed in position if the vapor was illuminated by solar, electric, or any other kind of light. It is ihus evident that the position of the lines observed in the aurora, zodiacal, or any other luminous phenomenon, prove nothing in regard to its electric or non-electric origin, but are a criterion for the chemical nature of the illuminated transparent matter, the super-atmospheric medium. They are alsa a criterion proving that the polar light is not derived from the sun, because in all substances illuminated from this latter source, traces of the lines of the solar spectrum are always visible ; this is the case with the reflected light of moon, planets, clouds, etc. The spectrum of the polar light, on the contrary, shows no trace of the lines belonging to the solar spectrum. Paper from tUe Reed Cane. We learn from a Norfolk, Va., journal, that a company has been formed near that city for the manufacture of pasteboard, etc., from the fiber of the reed cane. The process by which the fiber is disintegrated, though not new is interesting, and may not be known to many of our readers : " The cane, just as it comes from the cane brakes, is put into a cylinder, about 25 feet long and 12 or 15 inches in diameter—steam at 200 pounds pressure is let into the gun from a boiler close by. After a few minutes the valve at the mouth of the gun is opened by a trigger arrangement, and the steam blows the contents of the gun into the open air. The cane is thereby thoroughly disintegrated ; and the effect of the great heat of the steam at 200 pounds pressure is such that the resin and gums in the fiber are soluble in water without the use of chemicals, and the fiber can be beaten up in a paper-maker's rag engine and run off into coarse paper suitable for paper for board without any further treatment. A large part of the intercellular tissue is washed out by the process, so that the cellulose will fall on the machine. " A bundle of cane four feet by eight feet by the length of the cane (averaging ten feet) yields a tun of steam blown fiber ; the weight of that bundle of cane before it is blown from the steam cylinders, is about two tuns, about one half of the weight being water. The fiber on being ,blown from the cylinders, or guns, becomes quite dry in a minute or two, and is ready for baling. It is somewhat like oakum in appearance. " The resinous and gummy matter, acid and coloring matter, to the extent of thirty-three per cent of the weight of the steam blown cane fiber, can be washed out by immersing it in water and then squeezing out the liquid. " A battery of ten gjns of the ordinary size (12 inches diameter) will yield fifty tuns of fiber per day. "It has been found at Wilmington that the cost of a quantity of cane sufficient to make a tun of fiber, as disintegrated by the explosive force of steam (Lyman's process) is under four dollars a tun, including all expenses and the delivery of the cane at the works. This will be about the expense at Norfolk and Mobile, or at Memphis, Baton Rouge, etc., varying, however, somewhat in each locality. " The expense of reducing the cane to fiber by a puff of steam is very slight. If coal be used it will take one tun of coal for five tuns of the fiber ; but in most cases sufficient refuse timber for fuel can be got from the cane region, and being transported to the works by the same means which are used for the transportation of the cane, the cost of wood for fuel is less than that of coal. " Notwithstanding the great utility of the cane, the cane lands can never become of much value, because th.yare so vast and inexhaustible. Indeed, except on a few choice spots they will remain of no salable value, even when thousands of tuns of the steam blown fiber shall be used daily in the United States for paper and building materials and other purposes, and other thousands of tuns shall be exported daily to Europe. " When it is understood that it costs five cents per pound to reduce wood to paper pulp by the ordinary clu'iidcal processes now in use, and four cents per pound by Voel tor's mechanical process, now used extensively in France and in some parts.of the United States, the vast importance of the process of disintegration by steam will be at onco recognized. It costs less than one fifth of one cent per pound to reduce the cane to fiber by that process, ready for the paper-maker's beating machine. Moreover, the eane ie much cheaper than wood. " For common, coarse articles, such as paper box-bcard and pasteboard, plain and bituminized, for building purposes, it is impossible to get anything so cheap as the steam-blown cane fiber, and consequently very large quantities of those articles will be worked up by the water power near Norfolk, especially at Richmond and Fredericksburg. Moreover, large quantities the cane fiber will be exported to the Eastern States, and to England and France." New Mode of Fettling or Mning Puddling Furnaces. An invention patented in England consists in lining pud dling furnaces with crude or prepared oxide of manganese or manganese ore either as the chief ingredient of the fettling or as an addition to the oxide of iron or other material which is employed. In using crude or native oxide of manganese, or manganese ore without admixture with other solid, an ore is used, which, when pulverised and moistened, will form a plastic and pasty mass, and which when h ated will harden and adhere firmly to the sides and bottom of the puddling furnace. For this purpose the cheap oxide3 containing a considerable proportion of iron arc best suited, provided they do not also contain, other impurities, such as sulphur and phosphorus, which would injuie the iron in the furnace. When an ore or prepared oxide is used which dot s noi harden sufficiently, after being rendered plastic by water alone, it is mixed with a sufficient quantity of finely pow dered and moistened hematite, or other suitable material to give it the property of hardening and adhering when heated in the furnace. The proportions in which oxide of manganese and oxide ol iron should be mixed, in order to make the fettling according to this invention, vary with the nature of the cast or pig iron to be puddled. With pig iron of ordinary quality, about half a hundredweight of oxide of manganese mixed with the requisite quantity of oxide of iron for the fettling of the furnace is sufficient for a charge of four to live hundredweight oi pig iron. When the pig iron itself is rich in macgeneso a less proportion is necessary in the fettling. When the pig iron contains a large quantity of silicon and little manganese, more oxide of manganese is required in the fettling than is required with pig iron containing much oxide of manganese and little silicon. Where practicable the hiver tor prefers to introduce the pig iron into the puddling furnace in a melted state ; when this is done, and the fettling containing oxide of manganese is laid on the bottom and low. r part of the sides of the furnace, the charge gets the full benefit of the evolution of oxygen, which takes pla-ct when the oxide of i--angu-nese is heated. By the use of oxide of manganese, as described, the puddling process is expedited, and the quality ul the iron or steel produced is improved. The heated iron or steel during the puddling process decomposes the oxide of manganese, causing an evolution of oxygen, which, rising through the molten iron orsteel, rapidly oxidizes the oxidiza-ble materials contained in the metal. A portion of the reduced manganese enters into alloy with tie iron or steel and effects the improvement in the quality ol the m: tal which is well known to result from the use of manganese in the manu, facture of iron or steel. When oxide of manganese is mixed directly with the charge for fluxing, as has been proposed, a portion is liable to become mechanically distributed through the mass of iron or steel in the state of an infusible powder, consisting of manganese in a low etate of oxidation, and injures the mechanical properties of the metal. But when oxide of manganese is used in the fettling of the puddling furnace according to this invention, it 3 gradually aecomposed as the carbon and silicon of the pig iron or steel are presented to it by the stirring of the puddler, and the manganese enters the charge in a fused state either as reduced metal or as silicate. Very little of the manganese which enters the iron or steel during the puddling process remains in the finished metal, most of the manganese separating during the finishing of the metal in the form of silicate of manganese, carrying with it other impurities, such as phosphorus and sulphur. The silicate of mangaaese separates from the metal more readily than silicate of iron, and is found in consUlert'Wc quantity in the cinder and hammer slag. The cinder and hammer slag are, therefore, more valuable than ordinary cinder or hammer slag for the making of cinder iron, in consequence of their ricness in manganese. Although the fluxing property of oxide of manganes , either alone or mixed with oxide of iron, renders the addition of any other material to the fettling unnecessary when pig iron or eteel of the ordinary qualities are puddled, yet when pig iron or steel of such quality as renders the use of alkaline fluxes desirable is about to be puddled, common salt, or carbonate or nitrate of soda may be added to the oxide of manganese. A quantity of the soda salt, equal to about one fourth of the weight of the oxide of mangaaese, is generally sufficient.—Mechanics' Magazine. ME. LOWE, the English Chancellor of the Exchequer, is an accomplished velocipedistj although he began practice at the age f fifty-eight,