Jaih 2, 1879] 



NATURE 



193 



only can testing become a real benefit to the administra- 

 tion/' 



Again he says, " We know quantitatively the electrical 

 state of the lines at all hours of the day, and seasons of 

 the year ; we are able to localise faults of all kinds very 

 accurately and repair them with despatch ; we test all 

 our telegraphic material, and by it have greatly improved 

 its essential qualities ; we are not groping in the dark any 

 more — we measure and knoiv." 



It never must be forgotten that testing is in reality a 

 physical experiment, and these physical experiments are 

 being conducted every day throughout the whole of our 

 English telegraph system. Our cable electricians under 

 the guidance of Sir William Thomson have carried this 

 system of physical experiment to a high standard of per- 

 fection, and our Indian friends would do weU to profit by 

 their teaching. 



Mr. Schwendler's explanation of the theory' of the 

 bridge is not clear, nor does his use of Kirchhoff's corol- 

 laries to Ohm's law much help the student. Indeed it is 

 very doubtful whether his proof that the sensibility of the 

 bridge method is greatest when the branch and the resist- 

 ance are equal is true. At any rate in our practice we 

 find that the more delicate the galvanometer of the bridge 

 the more sensitive and the more accurate is our test. 



The most valuable portion of Mr. Schwendler's book is 

 his abstract of Ohm's classical paper, a translation of 

 which is to be found in Taylor's " Scientific Memoirs," 

 and also in his account of Kirchhoff's corollaries to this 

 law. 



The practice generally of line testing and testing for 

 faults contains nothing new, but his chapter on natural 

 currents, showing the efTect of polarisation of earth plates 

 and the presence of earth currents, is interesting. 



He says, also ^'^ Defective itisulation at a feiv points in 

 a line is a fruitful source of currents. At all such points 

 polarisation is produced by the working currents, in a 

 manner precisely similar to that of the earth plates, by 

 the same cause already alluded to, and to a degree de- 

 pendent on the resistance and the position of the faults. 

 These currents will be strongest in rainy weather, when 

 the line is in contact with trees, when the insulators are 

 covered with dew — in fine, under those circumstances 

 which diminish the resistance of faults and promote elec- 

 trolytic action. 



" The stronger the working currents used, and the fewer 

 the defective points, the stronger will be the polarisation 

 currents. 



" If these currents become very strong their direction 

 may be reversed by sending for a short time a strong 

 current with zinc to line ; and, in such a case, this invari- 

 ably indicates a single fault in the line or cable." This is 

 a defect which we do not experience in England. 



We find that (p. 66) " on all the lines in India positive 

 signalling currents (copper to hne) are used in order to 

 have the greatest possible insulation of each line under all 

 circumstances. Now, when measuring the insulation of a 

 line with a positive test current, it is evident that the value 

 obtained must give the insulation much too high, i.e., higher 

 than the line actually has when signals pass through it ; 

 because the signalling currents can only have a compara- 

 tively small oxidising effect on the line, since only a very 

 small part can escape to earth in the different points of the 



line, while a positive testing current, the further end of the 

 line being insulated, must all escape to earth at the de- 

 fective points of the line. Again, when measuring the in- 

 sulation of a line with a negative testing current, wc get a 

 value which gives the insulation of the line much too low, 

 because negative signalling currents are never used. In 

 the absence of any known law, which would give us how 

 much too high the insulation of the line is obtained with 

 a positive testing current, and how much too low with a 

 negative testing current, we can do nothing better than 

 to take the arithmetic mean of the measured values as 

 representing the insulation the line probably has when 

 signals are passing through it. Of this mean it may, 

 however, be said that it must be always somewhat tco 

 low, for the very reason that negative signalling currents 

 are never used, and therefore the arithmetic mean again 

 of the first mean and the positive measured value would 

 represent a value most probably approximating to the one 

 which the line actually has when signals pass, and which 

 alone is of practical interest and consequence to be 

 known." 



The latter part of the book is devoted to fault testing, 

 />., to the localisation of the positions of faults. 



The book itself is a very valuable addition to the litera- 

 ture of the subject, but we doubt whether it will be of any 

 practical use to our English electricians. 



OUR BOOK SHELF 



Sketches of Wild Sport and Natural History of t/ie- 

 Highlaiuls. By Charles St. John. Illustrated Edition. 

 (London: Murray, 1878.) 



Many of our readers must be familiar with the inimitable 

 " Sketches " of St. John, which has long ago achieved 

 the position of a classic for both the sportsman and 

 the naturalist. We do not know of any descriptions of 

 sport to equal those that abound in these pages, in truth- 

 fulness, vigour, and genial humour. To the naturalist 

 who loves to know the habits of an animal in its native 

 haunts, the book must be a treasure ; and now that 

 Harrison Weir, Whymper, Corbould, Collins, and Elwes 

 have adorned it with their art, the book should become a 

 greater favourite than ever. No artist equals Whymper 

 in his faithfulness to life in drawing animals. Every 

 picture in the book — and there are about eighty of them — 

 is a masterpiece in its way, and an impressive lesson in 

 natural history. We need only say that the engraver is 

 Mr. J. W. Wh}-mper to convince our readers that the 

 artists' charming work has been faithfully and skilfully 

 rendered. No one can read a chapter of the book with- 

 out being both refreshed and instructed. 



LETTERS TO THE EDITOR 



[T7ie Editor does not hold himself responsible for opinions expressed 

 by his correspondents. Neither can he undertake to return, or 

 to correspond with the writers of rejected manuscripts. No 

 notice is taken of anonymous communications. 



[The Editor urgently requests correspondents to keep their letters as 

 short as possible. 7he pressure on his space is so great that it 

 is impossible otherwise to ensure the appearance ezeti of com- 

 munications containing interesting and novel facts J\ 



Paradoxical Philosophy 



It is strange to see a writer on philosophy like Mr. S. H. 

 Hodgson, as well as physicists so exceptionally able as Prof. 

 Clifford, and now Prof. Clerk- Maxwell, falling into the same 

 errors of observation as more ordinary mortal?. Neither the 

 authors of the " Unseen Universe," nor any of the members of the 

 Parodoxical Society, have, so far as I am aware, expressed the 



