I50 



NATURE 



\y7me 24, 1875 



stretching from London, the telegraphic centre of the 

 world, by land and submarine circuits into Denmark, 

 Norway, Sweden, and Russia in Europe, thence across 

 the wilds of Siberia in Asiatic Russia to the Japanese 

 Sea, and on to Japan, terminating within the tropics, 

 at Hong Kong. Secondly, the " Eastern Telegraph," 

 which, crossing the Bay of Biscay, reaches Lisbon, 

 and thence threading its way under the dark blue 

 waters of the Mediterranean Sea to Suez, reaches India 

 by the Red Sea and Indian Ocean, and on to Ceylon 

 (Point du Galle), joining the " Great Northern " at Hong 

 Kong vici Singapore. Thus by means of these two great 

 systems a complete circuit of the continents of Europe and 

 Asia is effected, the one within the limits of the tropics, 

 the other bordering upon the Arctic circle, reaching as 

 it does to 62° of north latitude. At Singapore the circuit 

 is divided, a branch extending south to Sumatra, Java, 

 and the continent of Austraha,— Sydney, Melbourne, and 

 Adelaide being reached ; New Zealand being about to be 

 included. Thirdly, there is the vast stretch of the South 

 Atlantic Ocean traversed by the circuits of the " Brazilian 

 Submarine," connecting Great Britain, via Lisbon, with 

 Madeira, St. Vincent, and the continent of South America 

 to Pernambuco. There it joins the coast submarine cir- 

 cuits of the " Western and Brazilian," extending north to 

 Para and south to Bahia, Rio Janeiro, Rio Grand do Sul, 

 and Monte Video in the River Plate, at which station, in 

 connection with the local lines of the River Plate Com- 

 pany, it reaches Buenos Ayres, thence by means of the 

 wires of the Argentine Republic, crosses the Andes into 

 Chili and Peru. From Para the electric circuit is ex- 

 tended (Para and Demerara being now under comple- 

 tion), by way of the West India Isles, Jamaica, and Cuba, 

 to Florida, there joining the extensive system of the 

 United States Trunk lines ; to San Francisco, west, and 

 Newfoundland, east ; and thence, by the circuits of the 

 "Anglo-American" and "Direct United States" cable, 

 crossing the Atlantic Ocean to Great Britain. Thus the 

 New World is also encircled by two great systems, the 

 one almost equatorial, the other within the higher degrees 

 of northern latitude. 



In dealing with submarine circuits the electrician 

 has several matters to consider and accurately adjust, 

 some of which will be more fully considered hereafter. 

 First, there is the copper-conducting wire, its capacity 

 according to the length of the circuit. Too small a con- 

 ducting wire on a circuit of a given length would offer too 

 great a resistance ; too large a conductmg wire would be 

 equally faulty, induction increasing in greater proportion 

 from its large superficial surface than its increased 

 sectional area augments the speed. The exact sectional 

 area of the wire has therefore to be determined ; then 

 for insuiaiion, the best relative proportion in weight, 

 and sectional m.easurement between the wire and that of 

 the insulating material. Insulation, as is well known, 

 may be obtained by a mere film of a non-conductor sur- 

 rounding the wire. This is illustrated by the simple 

 experiment of passing a weak voltaic current of electricity 

 through an extended fine metallic wire immersed in a 

 trough of water. Under ordinary circumstances it is but 

 natural to suppose (water being a conductor) that there 

 would be no insulation ; not so ; by the action of the 

 current through the wire decomposing the water, a fine 

 non-conducting film of hydrogen is developed surround- 

 ing the wire, which, with a strength of current ad- 

 justed to the resistance of the wire, will separate the 

 water from the metallic conductor, perfect insulation being 

 maintained. Destroy the balance between the current 

 and the wire, and the hydrogen, evolved too rapidly by 

 reason of electrical decomposition, accumulates upon the 

 surface of the wire and, passing off in the form of small 

 bubbles, destroys the insulation. This simple experiment 

 demonstrates that insulation in the abstract sense may 

 be obtained by a very thin covering of a non-conductor. 



It is, however, in practice mechanically" unsafe to rely 

 upon mere tissues of insulating material surrounding the 

 conducting wire ; a certain thickness is absolutely neces- 

 sary for security. Every insulated core to be used for 

 submarine purposes, to ensure integrity of manufacture, 

 should be tested under pressure, so as to break down all 

 mechanical imperfections in the coating of the insulating 

 medium, before the cable is submerged. The determina- 

 tion of the dimensions of the insulator influences also 



Fig. 34.— First Atlantic Cable, 1S57 ^natura■ bi/c;. 



materially the inductive effect ot the circuit ; and when 

 it is remembered that the best insulating material repre- 

 sents a cost of about 6^-. per pound weight upon the wire, 

 the close connection between science and pounds shil- 

 lings and pence becomes at once apparent. The varia- 

 tions in weight per nautical mile of copper and insulation 

 in some of the recent important cables are here given. 

 The Atlantic main cables of 1865 and 1866 : copper 

 300 lbs., insulation 400 lbs. ; lengths each about 1,900 

 nautical miles. French Atlantic mam cable, 1869 : copper 

 400 lbs., insulation 400 lbs. ; length about 2,600 nautical 



1865. 1866. 



Fig. 35.— Atlantic Cables laid in 1865 and 1866, between Valentia and 

 Newfoundland (natural size), weight per naut. i"7S tons. 



miles. Falmouth and Lisbon, 1870 : copper i2ollbs., 

 insulation 175 lbs. ; length about 800 nautical miles! 

 Anglo- Danish Cable, 1868 : copper 180 lbs., insulation 

 180 lbs. ; length, 365 nautical miles. Hong- Kong- 

 Shanghai, 1870 : copper 300 lbs., insulation 200 lbs. ; 

 length, 1,100 nautical miles. China Telegraph, 1870: 

 copper 107 lbs., insulation 140 lbs. ; length, 1,632 nautical 

 miles. British India Extension, 1870 : copper 120 lbs., 

 insulation 175 lbs. ; length, 1,448 nautical miles. ■. Eight 

 important submarine circuits have here been summarised, 

 and in six it will be found that the proportions in the 

 weight per nautical mile between the copper and insula- 



