May 14, 1886.] 



SCIENCE. 



443 



mentary resistance of carbon, the wires are taken 

 of the same length and resistance, their diameters 

 being different, he finds a marked difference in 

 their inductive capacities. For instance : a pure 

 copper wire, compared with a brass one of double 

 the diameter, shows a much higher self-induc- 

 tion ; and Professor Hughes remarks in this con- 

 nection, that, as the diameter increases, the re- 

 actions of the current in the contiguous parts of 

 the wire on each other become less. The follow- 

 ing table gives some fresh values of the electro- 

 motive force of self-induction currents in wires 

 and strips one metre long, that of a chemically 

 pure copper wire one millimetre in diameter 

 being taken as 100 : — 

 Wires of the same diameter, but of different resistance, 



1 metre in length. 



Soft Swedish iron 500 



Copper 100 



Brass 65 



Lead 50 



Wires of the same resistance, but of different diameter, 

 1 metre in length. 



Soft Swedish iron 400 



Copper 100 



Brass 88 



Lead 81 



Strips of the same tvidth and thickness, but of different 

 resistance, 1 metre in length, 12 millimetres wide, 1-10 of 

 a millimetre thick. 



Copper . 60 



Brass .' 48 



Iron 45 



Lead 85 



Strips of the same resistance and thickness, but of differ- 

 ent widths, 1 metre in length, 1-10 millimetre thick. 



12 millimetres wide (copper). 60 



42 " " (brass) 45 



72 m " (iron) 39 



96 " " (lead) 29 



In the above table, wires of the same diameter 

 follow in the order of their resistance, iron alone 

 being the exception. The same order is preserved 

 in wires of the same resistance, but of different 

 diameters. In the latter case there is a nearer ap- 

 proach to equality, but they still show a differ- 

 ence of from 12 to 19 per cent; and, while the non- 

 magnetic metals have increased their inductive 

 capacity with increased diameter, iron has fallen 

 20 per cent : consequently wires of different 

 metals of the same resistance have not the same 

 inductive capacity, owing, probably, to the action 

 of contiguous portions of the current, as Professor 

 Hughes has already shown. 



If we reduce the extra currents by employing 

 thin sheets or strips, there is, in the case of iron, 

 a still more remarkable difference, for in strips of 

 different metals of the same width the force of 

 the extra currents in iron is actually less than 

 that in brass ; and if we compare an iron strip 

 with an iron or copper wire of the same resist- 



ance, we have, iron 500, copper wire 100, and an 

 iron strip 45, or 55 per cent less than the copper 

 wire. 



In the case of wires a nearer approach to equal- 

 ity in inductive capacity is shown when they are 

 of the same resistance, but in strips this is re- 

 versed ; for here, when equality in resistance is 

 produced by wider strips, the difference becomes 

 greater, iron then having actually less inductive 

 capacity than a lead wire of the same resistance. 

 Professor Hughes attributes this remarkable result 

 not only to the reactions of contiguous portions 

 of the current being less in sheets or strips than 

 in wires, but also to an imperfect formation of 

 the circular magnetism which takes place in iron 

 wires on the passage of an electric current. He 

 has tried all forms of conductors, such as those of 

 square, stellar, and tubular section ; and all of 

 them show a diminution of inductive capacity as 

 compared with wires of solid circular cross-sec- 

 tion. In solid conductors the maximum self-in- 

 duction appears in those of circular section, and 

 the minimum in wires formed into a flat strip. 



While re-affirming his statement that the best 

 lightning-rod is a flat strip of copper, or a gal- 

 vanized iron strand wire, Professor Hughes has 

 made experiments with American compound 

 wires consisting of a steel core coated with cop- 

 per, or a copper core coated with steel. He finds 

 that the copper coating has an enormous influ- 

 ence in reducing self-induction in the steel. With- 

 out it the self-induction was found to be 350 as 

 compared with a copper wire giving 100, whereas 

 with it the self-induction was only 107, or 7 per 

 cent more than copper alone. This effect is ex- 

 plained by the fact that the circular magnetism 

 created by the passage of a current through an 

 iron wire is produced chiefly on the exterior 

 portion of the wire ; and if this is of copper, it 

 is practically suppressed. On the other hand, 

 copper wire coated with steel has a greatly in- 

 creased self-induction as compared with copper 

 wire uncoated. It even has a higher self-induction 

 than a solid iron wire, and its resistance in the 

 variable period is proportionally greater than that 

 of a soft iron wire. Professor Hughes has made 

 numerous experiments on this point ; and they 

 all show, that, while copper in a straight wire or 

 a single wide loop has a far lower inductive ca- 

 pacity than iron, it has, on the other hand, the 

 property of being far more excited by + lie reaction 

 of iron, so that a straight copper wire can be ex- 

 cited by this reaction to a degree greatly exceed- 

 ing that of a straight iron wire under precisely 

 the same conditions. Some of Professor Hughes's 

 experiments illustrating this point may be cited, 

 as they are of much practical importance. A 



