ELECTRO-TELEGRAPHY 249 



other. If they wore not insulated, there could be no statical charge, as is well under- 

 stood by all electricians. Induction, therefore, involves insulation. But in a sub- 

 marine circuit this is not the case. If the wire was disconnected at the further 

 extremity from the telegraph instrument, and was sealed up with gutta-percha, the 

 conditions would be nearly the same. In practice, however, it is quite open through 

 the instrument to the earth, and the resistance opposed by the very long length of 

 wire is the only insulation between the inner and outer coatings ; for it unites both, 

 being in connection with the earth at both ends. It is, therefore, evident that, if the 

 wire offers no resistance, there will be no insulation, and, as a consequence, no induc- 

 tion to retard the passage of the current. It is also equally plain, that precisely in 

 proportion to the resistance which the wire opposes, provided always the insulating 

 medium is of the same thickness, will induction manifest itself, and retardation bo ex- 

 perienced. There is also another difference between a Leyden jar and a submarine 

 circuit. The Leyden jar is charged uniformly all over, whilst in a submarine wire, 

 the tension of the charge varies in different portions of the circuit, being at its maxi- 

 mum at the end where contact with the battery is made, and gradually vanishing to 

 nothing at the further extremity. 



' The fact has already been adverted to, that if a battery capable of developing, when 

 in short circuit, a much greater quantity of electricity than it does when connected 

 through a fine wire, is connected through a long wire of small dimensions, the quantity 

 evolved through one length of this fine wire will be twice the amount of that evolved 

 through double the length. From this it appears that double the length of wire op- 

 poses the same resistance to a given quantity of electricity, as a wire of half this length 

 does to double that quantity of electricity. Hence, any length of wire opposes an 

 infinitely small amount of resistance to an infinitely small quantity of electricity. An 

 idea seems to prevail, that if the wire becomes attenuated at any one spot, the con- 

 ducting power of the whole circuit will be brought down to that of a wire of no greater 

 thickness than it is at the attenuated portion. It will be at once seen, from what has 

 just been said, that this cannot be the case, but that the effect will be precisely the 

 same as increasing the length of the circuit. For if the circuit is attenuated, say 

 throughout a mile of its length, to half its normal diameter, it will possess, in this 

 space, only one-fourth of the sectional area ; consequently, it will oppose 4 times 

 as much resistance as a mile of wire of its original size. In other words, this mile 

 will oppose as much resistance as 4 miles of any other portion of the circuit. If the 

 resistance offered by 1 mile of the circuit, is put down as 1, then the resistance op- 

 posed by the attenuated mile would be 4 ; and if the normal resistance, or 1, is sub- 

 tracted from this, there will remain 3, as the excess of resistance over what it should 

 be, owing to this attenuation. The effect will therefore be the same, as increasing the 

 length of the circuit by 3 mile's. It seems surprising that this should have escaped 

 observation, for if true the conducting power of the whole circuit would be brought 

 down to that of a wire of no greater diameter than that made use of in the coils of 

 the instruments, which may be considered simply as a continuation of the cable, and 

 is only about one-eightieth of an inch in diameter. Electricity flowing through a long 

 circuit which is attenuated at one portion, may be compared to water flowing through 

 a long tube of small dimensions, and therefore opposing a great amount of friction to 

 the passage of the water. If the tube is indented at any one part, it will decrease the 

 flow to a certain extent, but not nearly so much as if the bore of the pipe was reduced 

 throughout its length to the same size as it is where indented. 



4 Incidentally connected with, and following as a natural sequence to what has just 

 been said, is the bearing and the value of the resistance opposed by the coils of the 

 telegraph instrument. If an electro-magnet is enveloped with a single convolution of 

 wire, and a battery is connected with it, there will be little resistance to the free pas- 

 sage of the electricity, as owing to the short length of wire, but very little magnetism 

 will be developed. Under such conditions, with a great expenditure of battery power, 

 a very small equivalent of magnetic force will be obtained. If the number of convo- 

 lutions be increased until the resistance opposed by the greater length of wire begins 

 sensibly to reduce the dynamic quantity flowing, the magnetic force will then be found 

 to be increased proportionately to the number of convolutions, less certain other con- 

 siderations, which need not now be entered upon, and with actually less expenditure 

 of battery power. Eesistance encountered and overcome, under these conditions, im- 

 plies so much work done ; and the larger the proportion of it in the instruments, 

 relatively to that opposed by the cable, the more economically will the electricity be 

 expended. If the cable is supposed to offer 10 times the resistance of that opposed 

 by the coils of the telegraph instrument, which is nearly the case in some printing 

 circuits of 200 miles in length, the whole of the cable, as well as the coils, will be 

 rendered magnetic. Ten-elevenths of the magnetism will be developed in the cable 

 itself, where it will be useless, and one-eleventh only in the coils of the apparatus, 



