^62 



NA rURE 



[Apkil 1 1, 1S95 



piercing projectiles in that country ; >fr. Herhert Scott will 

 communicate a paper on the iron ore mines of Elba ; and Mr. 

 W.J. Keep, of Detroit, will contribute a paper on " Tests of 

 Cast Iron." 



Prof. Dr. G. Schweixfurth, the distinguished African 

 traveller and botanist, has passed the winter at llclouan, near 

 Cairo, where be is engaged in mapping and describing the 

 adjacent mountains and " wadis," for a memoir to be com- 

 municated to the Instilut Egyptien. Dr. Schweinfurth h.\s 

 just published the text of an interesting lecture on Erythrsea 

 —the Italian colony on the Red Sea— delivered before 

 the Geographical Society of Berlin in July last year. This 

 paper gives an account of a four-months' expedition which 

 he made to Erythi^a' in the early spring of 1S94, in company 

 with Dr. Max Schoeller. It gives a most favourable report 

 of the progress of the new Italian colony and of its future 

 prospects. 



A CORRESPONDF.NT has sent us particulars of an explosion 

 which occurred at Providence, Rhode Island, U.S.A., on 

 March S, in connection with the electric light mains of that 

 city. The explosion resembles those recently investigated by 

 Major Cardew (see p. 539). Through a leak in the gas mains, 

 the conduits became full of an explosive mixture, which 

 suddenly exploded with great violence. The street was blown 

 up for some thirty feet, all the glass in neighbouring houses 

 wrecked, and mis^'iles of wood and iron thrown to some 

 distance. The noise and shock are said to have been felt all 

 over the city. 



The circumstances connected with the formation of the 

 dangerous alkaline deposits found by Major Cardew on the 

 mains of the St. Fancras electric light supply, and the presence 

 in this deposit in some cases of the alkaline metals in an un- 

 oxidiscd condition, have been inquired into by a Committee of 

 the Royal Society and the Institution of Electrical Engineers, 

 and the results have been issued in a report, which states that 

 *' the Committee are of opinion that the explosions which have 

 occurred were caused by the firing of an explosive mixture of 

 coal-gas and air by sparks caused by means of the above- 

 mentioned incrustation. It has been proved that sparks may 

 be caused either by the incrustation itself acting as an imperfect 

 electrical conductor, or by moisture coming into contact with 

 metallic sodium or potassium, both of which metals have been 

 found to exist within the incrustation. These metals have been 

 produced by the electrolytic decomposition of alkaline salts, 

 chiefly derived from ihc soil and conveyed by moisture 

 along the fibres of the wooden bearers towards the 

 negative conductor. To avoid a repetition of these accidents, 

 the bearers of the insulators at present in use should be replaced 

 by other devices through which moisture is prevented from 

 Iravellirg, and it is recommended that the pattern of insulator 

 in use should be changed, and a pattern adopted in which a 

 longer insulating surface is interposed between the conductors 

 and the bearer. The Committee are also of opinion that it is 

 desirable that means should be provided by which the conduits 

 can be inspected throughout their length, so far as is necessary 

 tn detect incrustations on the insulators. The Committee have 

 not thought it within their province to investigate the causes of 

 the presence of coal-gas within the electric lighting conduits, 

 but it is obvious that this is the primary source of danger." 



Dr. Duprr's rcforl with reference to the recent explosion of 

 a cylinder filled with compressed oxygen, at Fcnchurch Street 

 .Station, was read at the coroner's inquiry List Thursday. An 

 examination showed that the inner surface of the cylinder was 

 fairly clean, but at the end on which the valve was, the surface 



NO. I 328, VOL. 51] 



wasincrusted with magnetic oxide of iron, which was easily re- 

 movable, and which under the microscope showed in many 

 places the globular form assumed by the magnetic oxide pro- 

 duced by the burning of iron or steel. The lower end of the 

 brass screw, by means of which the valve fittings were screwed 

 into the bottle, was also incrusted with magnetic oxide of iron, 

 much of which was in the form of small globules produced by 

 fusion at very high temper.ature. They were evidently fused to 

 the material of the screw, and it in some cases even slightly pitied 

 the metal. These facts lead Dr. Dupre to conclude, first, that 

 the bottle at the time of the accident contained an explosive 

 gaseous mixture ; and that, secondly, this mixture was fired by 

 so ne portions of finely-divided iron, or perhaps grease, igniting 

 in the compressed gas. That some iron had actually been on 

 fire in the cylinder the condition of the screw sufficiently proved. 

 The coroner, in summing up, spoke strongly in favour of a 

 Government let being imposed with regard to the cylinders; 

 and the jury included in their verdict the recommendation that 

 all compressed gases of an explosive nature should be scheduled 

 under the Explosives Act, that all cylinders should be tested by 

 the Government periodically, that no cylinder should be allowed 

 to be used or conveyed aboat unless bearing the Government 

 stamp, that all manufacturers should be licensed by the Board of 

 Trade in the future, and that separate hydraulic pumps should 

 be used in the apparatus in filling the cylinders, and also 

 recommended a Board of Trade and railway inquiry. 



At the Institution of Civil Engineers, on April 2, Mr. J. I. 

 Thornycroft, F.U.S., and Mr. S. \V. Barnaby, in a joint paper, 

 expressed the belief that the speed of vessels had now ap- 

 proached within measurable distance of that at which propul- 

 sion by screws became inelTicicnt. For a given pitch-ratio and 

 slip, the thrust per unit of area varies as the square of the 

 speed. Certain conditions of pressure on the scre^vblades 

 cause the formation of caviies, filled with air and vapour 

 driven off from the water, behind the blades. Experiments 

 carried out by the authors, with screw-blades of elliptical 

 form, show that this "cavitation" does not commence sud- 

 denly, but appears to become detrimental when the mean 

 negative pressure on the forward side of the blades exceeds 

 about 6,' lbs. per square inch, or when the whole thrust exceeds 

 II J lbs. per square inch. Cavitation can only be avoided at 

 very high speed by increasing either the immersion of the 

 screw or its blade-area. Immersion is limited by considera- 

 tions of draught. Increased area, the authors remarked, can 

 be obtained in three ways : (i) by increasing the ratio of sur- 

 face to disc-area, (2) by employing a larger diameter than that 

 theoretically best for the given conditions, (3) by increasing 

 the pitch-ratio, which involves a larger diameter with a reduced 

 rate of revolution. Either tends to a waste of power if pur- 

 sued beyond somewhat narrow limits, and it appears inevitable 

 that reduced efficiency must be submitted to as the speed of 

 vessels is increased. 



A I'EW noteworthy points were brought out last week in 

 evidence before the Select Committee of the House of CommoM 

 appointed to consider Ihc advisability of adopting the metrical 

 system of weights and measures. Captain II. R. Sankey, R.E., 

 director of the engineering firm of VVillans and Robinson, 

 Thames Dilton, said that the metric system of measurement 

 had been adopted by his firm for the last two years. The 

 reason of the change was the advantage of working inter- 

 changeably with manufacturing engineers on the continent. 

 Mis firm's trade was mainly British, but no objection had been 

 made to metrical mcasurcmenls by Hrilish customers. The work- 

 men were agreed that the metric system was much more easy 

 to work with than the English measures, ard » as much less liable 



