its magnetism at 210°, whereas a similar alloy without chromium retained its magnetic properties up to 235° C. The effect of chromium seems to be manifested in this lowering of the temperature of transformation. All the reversible alloys, at an equal distance from the point of transformation, have the same magnetic permeability, and the magnetic permeability increases with the percentage of nickel.

G. J. B.

A Study of Certain Special Steels. By A. ABRAHAM.

(Annales des Mines, vol. xiv., 1898, p. 225.)

This is a long memoir (122 pages), detailing the experiments carried out by the Author on behalf of the French naval authorities upon certain special alloy steels made at the Imphy works, containing nickel, chromium and silicon in high proportions. The most important of these, known as NC4, has the following average composition:

Tested after forging, without special treatment, the elastic limit exceeds 31 tons, and the ultimate tensile resistance 44 tons, with an elongation of 45 per cent.; but in this condition the flexibility is deficient, so that it is better to subject it to water hardening from a cherry-red heat without further tempering, in which state the elastic limit is reduced to between 19 and 22 tons, but the elongation is increased to 60 per cent. As this elongation takes place over the entire length of the test-piece, the Author considers that objects subject to irregular shocks and strains, such as the steering tillers of large ships, might if made of the alloy be sensibly lighter in section than in soft steel. In thin plates of 2 to 8 millimetres thickness it seems to be well suited for the hulls of torpedo-boats, especially as the waste by corrosion is considerably less than with ordinary soft steel. It may also be used for deck-armour plating, although when subjected to direct attack the resistance was below that obtained with ordinary steel armour.

Forgings in NC4 generally have a rough surface, and it is necessary to take a cut of not less than 10 millimetres in finishing. The metal is therefore only suitable for objects which can be brought to shape by planing or turning alone.

The second class of steel noticed, also made at Imphy, called N 12.5, containing nickel 12 to 12, chromium 0.75 to 1.0 and carbon 0.3 to 0.45 per cent., is much harder, but can scarcely be used in the natural state as it breaks almost without stretching; but when annealed in a fire of flaming wood it gives a material

suitable for forgings requiring high tenacity and elasticity, the elastic limit exceeding 51 tons and the ultimate strength 63 tons. In thin plates it is suited for bullet-proof shields and light armour. The properties of alloys of iron and nickel with high proportions (25 to 45 per cent.) of the latter metal have been studied by Mr. C. E. Guillaume, the most remarkable of them being that with 36 per cent. of nickel. This is almost untarnishable in the air, and may be exposed to a saturated atmosphere for months without rusting, and its coefficient of dilatation between 0° and 80° C. is only 0.8 micron (1 micron = 0.001 millimetre), while that of iridio-platinum, such as is used for the international standard metre, is 8 microns. This alloy is therefore admirably suited for making standard bars and similar divided objects where great precision is required.

The last alloy noticed is the silicon steel of Imphy, which contains

This has for some years been employed in the manufacture of high-class springs for railway use. It is rolled into flat bars varying from 60 by 5 to 150 by 12 millimetres, which are heated to cherry red, cooled down in water to 300° C., reheated in a flaming wood fire and cooled slowly in the air. When tested on 4-inch lengths the limit of elasticity is 70 tons, and the breaking stress 79 tons per square inch, with 7 to 9 per cent. elongation; the fracture is fibrous. This the Author thinks would be especially well suited for deck armour, but up to the present it has not been possible to get a sound ingot of sufficient size to admit of such an application.

H. B.

The Microscopic Structure of Gold.

THOMAS ANDREWS, F.R.S., M. Inst. C.E.

(Engineering, 30 September, 1898, p. 411.)

The object of this investigation was to find out whether an analogy existed between the micro-segregation of the impurities and alloys in iron on freezing, and those in gold. In iron, the impurities and carbon compounds crystallise out in separate areas, which are interspersed throughout the matrix of pure iron. The latter, having a higher fusion point, forms distinct crystals at an earlier stage of the cooling. About 14 ounces of chemically pure gold was cast in a carbon mould, and gradually cooled. A transverse micro-section was prepared, etched, and examined microscopically. The gold was found to have solidified in large crystals, radiating from the centre of the bar. These primary crystals

were further subdivided into smaller secondary crystals, which in turn consisted of a minute tertiary system of crystalline particles of the regular cubic character. A longitudinal miro-section showed primary crystals of a generally hexagonal type, which were further subdivided into a smaller system of secondary crystals. A Table is given, showing the dimensions of primary, secondary, and tertiary crystals.

A. P. H.

Cost of Chlorination at the Alaska Treadwell Mill.

(Engineering and Mining Journal, vol. lxvi., 1898, p. 272.)

The Alaska Treadwell Gold Mining Company, at its mill on Douglas Island, Alaska, saves,' by concentration from the stamp-mill tailings, a certain quantity of pyritic residues carrying gold, which is not saved by the ordinary amalgamation. These concentrates are subsequently treated by roasting and chlorination, and the yield amounts to 34 per cent. of the total bullion output. In the year ending 31st May, 1898, the stamp mill treated 254,329 tons of ore; and from the tailings 4,331 7 tons of pyritic concentrates were saved. The cost of saving these was 16s. per ton of concentrates, or 24d. per ton of ore treated. The cost of chlorinating the concentrates was 16s. per ton for labour and 158. 5d. for supplies, the total cost being £1 118. 5d. per ton. The actual net consumption per ton of concentrates treated was 57.60 lbs. of acid, 17.35 lbs. of oxide of manganese, 151.44 lbs. of salt, and 8.25 lbs. of scrap iron. The cost of fuel (wood and coal) was 48. 11d. per ton. The average return obtained from the concentrates in gold was £8 178. 5d., and as the cost of saving them from the tailings and of treating them amounted to £2 78. 5d., the profit on the work was £6 108. per ton.

B. H. B.

Signalling to Great Distances by means of Free or Captive Balloons. DIBOS.

(La Revue Technique, 1898, p. 489.)

It is pointed out that the importance is now generally recognised, especially in war time, of aerial communications capable of being made out at great distances. For this purpose the use of balloons, fitted with an apparatus for the emission of a powerful light, which can be used without encumbrance and without danger, would seem to be eminently suitable. Reference is made to some experiments conducted at Toulon in 1878, when rockets lighted at a height of 2,296 feet above the sea were visible in a clear state

of the atmosphere at two observatories distant respectively 36 miles and 47 miles from the spot, and even in one case they were visible at no less than 90 miles. The danger of naked flames at or near to a balloon is insisted upon, and it is stated that, as the result of experiments, a distance below the car of 33 feet is sufficient to avoid all risks, but 66 feet is here assumed for greater precaution. The Author's system of producing an intense light by means of calcium phosphide in a vessel of water suspended below the balloon is explained by reference to a diagram. The water is supplied at the time the light is required by passing it down an india-rubber tube from the car of the balloon. The light produced in this way is intensely vivid, resembling that of magnesium or the electric arc, and it cannot be extinguished by wind or rain. A charge of from 14 ozs. to 16 ozs. of the calcium phosphide will burn for 30 minutes to 45 minutes.

G. R. R.

The Telegraph Cables of the World.

(Archiv für Post und Telegraphie, 1898, p. 193.)

This article gives a list of the number and lengths of the various telegraph cables throughout the world. These are set forth in two Tables, the first showing the particulars of those belonging to the several Governments, and the second those owned by private companies. It appears that in the year 1894 the total length of cables was 181,414 miles, while for the year 1897 the length is given as 187,196 miles, showing an increase of 5,782 miles. Of this increase 1,550 miles belongs to Germany. In the first Table 33 countries are quoted, owning 1,141 cables, of a total length of 22,831 miles; and in the second Table 30 companies are named, owning 318 cables, with a total length of 164,366 miles.

E. R. D.

Wireless Telegraphy. SCHLUCKEBIER.

(Archiv für Post und Telegraphie, 1898, pp. 208–240. Figs.)

The wire has hitherto been an essential part of telegraphic apparatus, but is costly and a constant source of trouble. The Author remarks that projects for wireless telegraphy are not new, as various experiments were described in the Archiv between 1892 and 1895. He cites those experiments made between the coast of Scotland and the Island of Mull, and refers to Marconi's very recent inventions. Before dealing with these, however, he gives particulars of experiments carried out at Berlin in 1894-5 by means of high potential currents passed into the earth. The

apparatus employed are shown by figures in the original text. The primary current passes through a conductor provided with an earth plate at each end. Two other plates at a distance from the first pair are connected by a conductor, through which passes the current received by those earth plates; the current received is, of course, only a small fraction of that passing between the first pair of plates, but is sufficient to actuate a telephone receiver. Alternate currents would serve, but as the usual periodicity is only about fifty per second, it was found better to employ a direct current dynamo with a battery in series and a rotary make and break apparatus to give several hundred vibrations per second. A key is used to send the Morse code. The two conductors should be parallel. The experiments referred to were carried out by Mr. Rathenau in Berlin near the Wannsee Railway station. Two zinc earth-plates, each of 32.4 square feet area, were buried in the bank of the lake 545 yards apart and connected by insulated copper wire. Current at 100 volts was obtained from the public supply and fifty large cells connected in series. Current of 2 amperes to 3 amperes was used, with 500 to 600 breaks per second. The secondary wire was carried in a boat, and zinc plates of 21.6 square feet area used at a distance of 90 yards to 110 yards apart. Satisfactory results were obtained up to a distance of 2.5 miles.

Earth tests were also made near Nauen, and gave satisfactory results at a distance of 3.53 miles. Further experiments were made in the neighbourhood of electric railways, and the effect was observable to a distance of 1.86 mile. A current of 16 amperes in the primary enabled messages to be sent through the earth a distance of 10.5 miles. The Author then gives a description of the Marconi apparatus and the results obtained.

E. R. D.

Construction of Ice-Breakers.

R. RUNEBERG and Admiral C. MAKAROV.

(Morskoi Sbornik, October, 1898.)

The Author, who has given special attention to this subject for many years, refers to a Paper he communicated to the Institution of Civil Engineers, "On Steamers for Winter Navigation and Ice-breaking," in which he investigated, theoretically, the proper form for the bows of such vessels. Experience with ice-breakers, built since that time, has enabled him to test the accuracy of his views, and has induced him to modify them to a certain extent. He points out that the two principal points about the form of the bows of such vessels are (1) that they should have hollow

'Minutes of Proceedings Inst. C.E., vol. xcvii., p. 277.