March 22, 1888] 



NATURE 



497 



which the Ordnance map gives no hint. On the "Palaeozoic 

 schist" range, south of Suanetia, there are glaciers not very 

 inferior to those of the Grand Paradis group, near Aosta. Dis- 

 miss for ever, Mr. Freshfield says, that preposterous fiction 

 about the 120 square kilometres of ice in the Caucasus. It is 

 too soon to say how many square kilometres there really are. 

 One estimate. Von Thielmann's, would make the extent covered 

 by ice close upon 2000 square kilometres, or equal to that in 

 Switzerland — political Switzerland, not the Alps. Mr. Fresh- 

 field dwelt on many other points in connection with this inter- 

 esting range, his notes on the inhabitants of the Caucasus 

 being specially valuable, correcting as they do many prevalent 

 errors. 



OUR ELECTRICAL COLUMN. 



Considerable attention has been drawn to the peculiarities 

 of manganese steel by a paper read before the Institution of 

 Civil Engineers, by Mr. Hadfield. Not only is such steel 

 entirely non-magnetic, but its electric resistance is extremely 

 high. Prof. Fleming {Electrician, March 9) gives the following 

 figures : — 



German silver . 

 Platinoid 

 Manganese steel 



20 '9 

 32-8 



•044 

 •021 

 •122 



The first column gives the resistance in microhms per cubic 

 centimetre at 0° C, and the second column the average per- 

 centage variation of resistance per 1° C. between o" and 

 100° C. These figures agree very well with those given by 

 Prof. Barrett at the British Association meeting at Manchester. 



Heim has been investigating the electro- positive character of 

 magnesium, with the view of replacing zinc in primary batteries. 

 He finds that in a Daniell cell its E.M.F. is 2 volts, in a Grove 

 cell it gives 2"9 volts, and in a Leclanche cell 2'2 volts. In a 

 bichromate cell it gives as much as 3 volts. 



Magnesium can now be produced for about 8j. per lb., but 

 local action is considerable, and its coastancy uncertain. Hence, 

 except for exceptional circumstances, its practical use is still 

 questionable. 



Prof. Oliver Lodge has been giving some admirable lec- 

 tures on lightning-protectors at the Society of Arts, an 1 has 

 pronounced the use of copper for such purposes as doomed. He 

 argued that the supposed area of protection was mythical, and 

 that the true way to protect a building was Maxwell's cage. He 

 advocated iron, and showed copper to possess "inertia" to 

 such an extent as to render its use dangerous. He also found 

 that under certain circumstances, such as sudden violent dis- 

 charges, untempered by time, points were of no use, but he sug- 

 gested the use of barbed wire along the ridges and eaves of 



roofs. 



> 



That careful and accurate worker. Prof. Roberts-Austen 



submitted a paper to the Royal Society on the 15th inst., in 



which he narrated his recen't inquiries into the mechanical 



properties of certain alloys that will have an important bearing 



on the metallic conductors employed in electrical enterprises. 



He has found that the tenacity of pure gold is very much 



diminished by the smallest admixture of impurities, and that this 



follows the order of the atomic volumes of the elements. Those 



elements the atomic volumes of which are higher than gold 



greatly diminish its tenacity. Doubtless the same principle is 



applicable to copper and other metals. The abnormal price of 



copper has raised a great demand for some better conductor 



than iron, or some improvement of iron in this respect. 



DERHAMS HYDROMETER. 



''PHE Revenue system of esticnating the duty on spirits 

 ■^ consists of hydrometer, and tables of strengths for each 

 degree of temperature from 30" to 80° F. When constructing 

 the present Revenue tables of strengths, Sikes ignored the 

 expansion and contraction of spirits due to variations of tem- 

 perature from the standard temperature of 51° Y., and assumed 

 that the strength of any given sample of spirits remained the 

 same at all degrees of temperature. From this false assump- 

 tion it follows in practice, for example, that 100 gallons 40 { 



overproof at 51° are estimated at 98*9 gallons at 30", and ioi'6 

 gallons at 80°, of the same strength as at 51° ; reducing these 

 quantities to the standard of proof strength, we have — 



At 30° ... 98*9 X 1*40= 1 38 '5 gallons of proof, 

 51° ... 1000 X 140 = I40'0 ,, ,, 



80° ... IOI6 X 140 = I42'2 ,, ,, 



showing a discrepancy of over 3^ gallons, although the same 

 actual quantity of spirit is present in each case. 



In its original construction, Sikes's hydrometer was not in- 

 tended to furnish specific gravities, but simply so many indica- 

 tions, respectively corresponding to the strengths in his tables. 

 But it has since been found necessary to supply a table of 

 specific gravities corresponding to the indications of the instru- 

 ment. It is well known that scientific precision cannot be 

 attained in experiments with the hydrometer, consequently the 

 specific gravities in this table are far from accurate : for example, 

 the specific gravity at the proof point, to the accurate definition, 

 of which the Inland Revenue attaches so much importance, is 

 given as '9233, instead of '9236. The whole specific gravity 

 table is in fact incorrect, the error sometimes amounting to two 

 subdivisions of the stem. The errors, however, arising from 

 this source are trifling compared with those inherent in the 

 tables of strengths. For the purpose of constructing correct 

 tables of strengths, the best data and those susceptible of the 

 most accurate determination are the specific gravities of the 

 spirits and the percentage by weight of alcohol they contain. 

 The specific gravity of proof spirit, as defined by the Spirit Act 

 is "9236 ; therefore the weight of one gallon is 9 '236 pounds. 

 Proof spirit contains 49*3 per cent, by weight of alcohol, of 

 specific gravity 79385 at 60°; therefore one -gallon of proof 

 spirit contains — 



9-236 X 49-3 



4'553 pounds of alcohol. 



To determine the true ratio of any spirit to proof spirit nothing 

 more is required than to ascertain the weight of alcohol in one 

 gallon of the spirit, and to divide that weight by the pounds of 

 alcohol in a gallon of proof spirit ; for example, spirit having a 

 specific gravity of '825 at 60° weighs 8*25 pounds per gallon ; its 

 percentage by weight of alcohol is 89 '13; therefore one gallon 

 contains — 



8-25 X 89-13 ^ ^.^^^ pounds of alcohol, 

 100 

 equivalent to 



7_353 — 1-615 gallons of proof spirit. 

 4-553 

 Or 100 gallons are equivalent to 161 -5 gallons of proof spirit, 

 and the spirit is said to be 61-5 overproof. Il is obvious that 

 although the bulk and specific gravity of a spirit vary with the 

 temperature, the percentage by weight of alcohol it contains 

 does not vary from that cause. The specific gravity of the spirit 

 in the preceding example is -839 at 30° ; the weight of one 

 gallon therefore is 8-39 pounds; its percentage by weight of 

 alcohol is 89-13 as before ; therefore one gallon contains — 



8-39 X 89-1 3 ^ ^.^^g pounds of alcohol, 



equivalent to 



100 



LlZr = I -642 gallons of proof spirit. 

 4-553 



The strength of the spirit, therefore, at 30° is 64-2 overproof. 

 It should be here pointed out that the diminished bulk of the 

 spirit at 30", as compa ed with its bulk at 60°, is exactly com- 

 pensated, in estimating the equivalent value in proof gallons, by 

 the increased strength at the former temperature ; for 100 gal- 

 lons of spirit 61-5 overproof at 60° contract to 98 33 gallons 

 at 30° ; and, reducing to proof strength — 



100 X 1-615 = 161-5 gallons of proof spirit, 

 98-33 X 1-642 = 161-5 do. do. 



whence it is evident that, by the employment of correct tables 

 of strengths, the estimate of the equivalent value of a given 

 quantity of spirit in gallons of proof spirit would be identical at 

 all degrees of temperature. The spirit tables published by Dr. 

 Derham, to which Sir Henry Ro.'-coe lately called the attention 

 of the Chancellor of the Exchequer, are calculated on this- 

 principle. 



