June 3, 1909J 



NA TURE 



409 



which in turn is connected to the steelyard. There ore 

 no loose weights ; seven poises travel on the steelyard, 

 I'ach representing, when in extreme position, a load of 

 100.000 lb. on the specimen. These poises are traversed 

 hy means of a screw, and can be instantly engaged or 

 disengaged. The machine, in the capable hands of Prof. 

 Dixon, should turn out some useful and interesting results 

 on the strength of built-up structures, for vifhich it seems' 

 to be well adapted. 



Bulletin No. 362 of the United States Geological 

 Survey, by Mr. J. S. Burrows, deals with the mine 

 sampling and chemical analyses of coals tested at the 

 United States Fuel-testing Plant in 1907. This is one of 

 a series of papers dealing with work done at the fuel- 

 testing station, work valuable to the coal owner, coal 

 user, and to all interested in the scientific study of coal. 

 1 he results of the examination of seventeen samples of 

 Jamestown coals are given in this bulletin, including 

 proximate and ultimate analyses and the experimentally 

 determined calorific values. \ list is given at the end of 

 the previous surrey publications on fuel testing. 



BULLETIN No. 365 issued by the United States Geological 

 Survey contains an interesting account of the fractiona- 

 tion of crude petroleum by capillary diffusion, by Messrs. 

 J. E. Gilpin and M. P. Cram. When oil is allowed to 

 rise by capillary attraction in a tube packed with Fuller's 

 earth, there is a decided fractionation of the oil, the frac- 

 tion at the top of the tube being of lower specific gravity 

 than that at the bottom. There is a tendency for the 

 paraffin hydrocarbons under these conditions to collect in 

 the lightest fractions at the top of the tube, and the un- 

 saturated hydrocarbons at the bottom. If the oil is mixed 

 with Fuller's earth and then displaced with water, about 

 one-third of the oil remains in the earth. 



Briquette-m.-vkinc formed the subject of a paper recently 

 read before the South Wales In.stitute of Engineers by 

 Prof. W. Galloway. Small coal cannot be burnt so 

 economically in the furnaces of boilers in its original 

 state as when in the for<Ti of briquettes, partly on account 

 cf so much of it falling through the fire-bars, and partly 

 because the particles lie so closely together as to prevent 

 the free access of the air required for combustion. 

 Briquettes made exclusively with anthracite coal burn too 

 slottdy, and it is advisable to mi.\ a certain proportion of 

 bituminous coal to overcome this objection. Up to the 

 present, no kind of agglomerating material other than 

 pitch or resin, or a mixture of these, has given satisfactory 

 results. Briquettes made with resin alone become soft 

 and lose their shape in the fire ; those having a mixture 

 of 4 per cent, of pitch and ij per cent, of resin give 

 better results. It is of interest to note that the total 

 output in the United Kingdom in 1906 amounted to 

 1.513,220 tons, while Germany produced 14,500,851 tons 

 of this fuel in the same year. The paper contains full 

 descriptions and drawings of the mixing and drying 

 machinery and presses required for briquette-making, 

 together with estimates of labour required and costs. For 

 example, at an English works making 102J tons of 

 briquettes per day of ten hours, the total cost, including 

 labour, materials, fuel and stores, interest and deprecia- 

 tion, works out to gs. 7-45^. per ton. 



A RECENT paper by G. Jafl^ in the .\nna\en der Physik, 

 on the electrical conductivity of pure hexane, will possess 

 considerable interest to those who are concerned with the 

 ■rSle of the solvent in electrolysis, as well as to those who 

 are working on the electrical conductivity of gases. The 

 NO. 2066, VOL. 80] 



impure material owes much of its conductivity to electro- 

 lytic impurities, but these can be removed by electrolysis 

 and by repeated distillation, when samples are obtained 

 with a very minute but practically constant conductivity, 

 about twelve times greater than that of air under similar 

 conditions. The pure hydrocarbon, indeed, shows almost 

 all the electrical properties of a gas of high density. How 

 widely its properties differ from those of purified water 

 or other feeble electrolytes may be seen from the fact 

 that two-thirds of the conductivity vanishes when the 

 measuring vessel is sheathed with lead in such a way 

 as to cut off external radiations, and that the remainder 

 of the conductivity is greatly influenced by the nature of 

 the containing vessel, aluminium giving exceptionally low 

 values. Two other remarkable points of contrast are 

 (i) the constancy of the current at different temperatures, 

 and (2) the fact that an increase of potential from 200 

 to 2000 volts produces no increase in the current, which 

 reaches a " saturation " value analogous to those of gases, 

 although at a much lower voltage. 



OUR ASTRONOMICAL COLUMN. 

 Astronomical Occurrences in June: — 

 June 3. Ilh. 43ni. to I5h. 14m. Eclipse of the Moon, visible 

 at Gieenwich. 

 g. 2oh. Marsinconjunclion with Moon (Mars2°33'N. ). 

 12. lih. Saturn in conjunction with Moon (Saturn 



2°I3'N.). 

 17. I2h. Sun eclipsed, invisible at Greenwich. 



21. Sh. 22m. Transit (ingress) of Jupiter's Satellite IV. 



(Callisto). 

 ,, iih. 41m. Minimum of Algol (/3 Persei). 

 ,, I4h. Sun enters Cancer, summer begins. 



22. l8h. Venus in conjunclion with Neptune. 



,, Saturn. Major axis outer iing=3978". Minor axis 

 = 8-85". 



23. ih. Jupiter in conjunction with Moon (Jupiter 



4°2i'S.). 



24. gh. 15m. to gh. 5gm. Moon occults v Virgims (4'2). 



The Dispersion of Light in Interstell.ar Space. — lii 

 the Revue generale des Sciences (No. 8, p. 350), Dr. Ch. 

 Nordniann reviews the work recently performed by MM. 

 Tikhoff and Belopolsky and himself on the dispersion of 

 light in interstellar space. The results obtained in the 

 first experiments have been questioned by a number of 

 astronomers, and, in re-stating the case clearly. Dr. 

 Nordmann disposes of many of the objections. 



As has already been recorded in these columns. Dr. 

 Nordmann's method consists in observing the difference, in 

 time, of the minima of variable stars when screens of 

 different colours are employed, whilst in the Tikhoff- 

 Belopolsky method the dispersion is shown by the various 

 displacements of lines in the different parts of the spectra of 

 spectroscopic binaries. 



It has been suggested that the observed differences may 

 be due to physical changes in the binary system itself, but 

 Dr. Nordmann argues that if this were the case the dis- 

 placement of the curves for different parts of the spectrum 

 would vary at different parts of the orbit, whereas if 

 the displacement is due to dispersion in space it would 

 be the same in all parts of the orbit. At present he 

 is content that a matter of so great an importance to 

 astronomers and physicists is re-opened, and would attach 

 no rigorous significance to the quantitalive results so far 

 obtained ; qualitativclv they are in the right direction, and 

 are in accordance with the results of ordinary refractive 

 media. Should the validity of these results be established 

 their importance in any cosmological discussion can scarcely 

 b" overestimated; for example, the determination of the 

 dist.Tnccs of binary systems would become greatly 

 simplified. 



A Rem.arkap.i.e Transit of Jupiter's Third Satellite. — 

 No. 4324 of the Astionniuische Niichrichten contains an 

 account, by Mr Innes, of a remarkable transit of Jupiter's 



