July 3, 1885.] 



SCIENCE. 



data from which to deduce the requisites of 

 any cable, in order that it may transmit speech, 

 and without cross-talk from the neighboring 

 conductors. These are briefly as follows : — 



1. Good conductivity. 



2. High insulation ; for without this the 

 current leaks from one conductor to the others, 

 giving rise to cross-talk ; and it is possible to 

 talk by direct leakage between two conductors 

 whose insulation is several million ohms. 



3. Low specific inductive capacity; for, the 

 greater the capacity, the greater the retarda- 

 tion, and the greater also the cross-talk due 

 to induction. 



Below is a table showing the specific induc- 

 tive capacity and insulation of various insulat- 

 ors. The measurements were all made on a 

 wire 0.05 of an inch in diameter, coated with 

 insulation to a thickness of 0.10 of an inch. 



Cable. 



Maker. 



Insulation 

 per mile in 

 meghoms. 



Specific 

 inductive 

 capacity 

 in micro- 

 farads. 



Grutta-percha, 

 India-rubber, 

 Renti . . . 



Faraday . . 

 Patterson . . 

 Brooks . . 



Siemens Bros., London. 



Rattier, Paris. 



A. G. Day, New York. 

 ^Faraday cable- works, ) 

 1 Cambridge, Mass. i 

 1 "Western electric com- 

 1 pany, Chicago. \ 

 (David Brooks, Phila-j 

 I delphia. \ 



190 

 170 

 150 



15,000 

 450 



4.2 

 3.7 

 4.0 



1.6 

 3.1 



2.8 



Let us take a special case, and compare a 

 gutta-percha cable having a specific inductive 

 capacity of 4.2 with a Faraday cable of 1.6. 

 The table predicts that we can talk three times 

 as far with the latter as with the former, and 

 experiment shows that we can. Again : the 

 cross-talk on the gutta-percha cables ought to 

 greatly exceed that on a Faraday cable ; and 

 experiment has shown, that, while conversation 

 over a two-mile gutta-percha cable was con- 

 tinually disturbed by existing cross-talk, con- 

 versation was carried on over a similarly 

 constructed Faraday cable five miles in length 

 without the cross-talk being appreciable. 



By proper attention to the electrical quali- 

 ties, then, we may talk underground a much 

 greater distance than we shall ever have reason 

 to in any city system, and this without cross- 

 talk from the neighboring circuits. 



We have seen that telegraph and electric- 

 lighting currents are not subject to the technical 

 diflSuulties we have been discussing, and that, 

 provided good conductivity and good insulation 

 are assured, it is with them purely a question of 

 expense. Let us, then, determine the relative 

 expense of overhead and underground wires. 



Suppose we have a large city with a telegraph- 

 office near the centre, and that it is desired to 



carry a hundred wires to the city limits, say, 

 three miles distant. Let us suppose that the 

 wires for the first mile rest on housetops, and 

 for the remainder of the distance on poles. 

 The cost will be : — 



35 roof-fixtures @ $45 



80 poles, with arms, etc @ 65 



300 miles No. 9 wire @ 16 



Stringing 300 miles wire (& 8 



$1,575 

 5,200 

 4,800 

 2,400 



$13,975 



Underground, the cost would be : 



miles (50 conductors) No. 17 lead-covered 

 cable @ $3,000 



miles trenching, troughing, laying, and re- 

 filling @ 2,000 



$18,000 

 6,000 



§24,000 



That is, the relative first cost of an over- 

 head and an underground line, to do the same 

 work, would be, say, $14,000 and $24,000. 



The same conclusion will hold true for tele- 

 phone-wires, provided we confine ourselves to 

 the problem of running out from the central 

 office, by fifty or a hundred conductor cables, 

 to a large number of distributing-points so 

 situated about the city that any subscriber 

 would be easy of access, by a short overhead 

 line, to one or another of them ; and this is 

 the problem that really occurs. So much for 

 construction. 



The yearly cost of repairing an overhead 

 S3' stem, including roof-rentals, is not less than 

 thirty per cent of the cost of construction ; 

 and the line would have to be renewed once in 

 twelve years. The cost of repairing an under- 

 ground system is practically nil. The Paris 

 telephone company, with wires extending to 

 three thousand subscribers, does not keep any 

 repair-men. The durability of an underground 

 system, provided lead-covered cables are used, 

 and there is no internal cause of deteriora- 

 tion, is at least thirt}^ years. Last summer 

 we examined some lead-incased gutta-percha 

 cables that had been in use by the French 

 government for that length of time, and found 

 them in perfectly good condition. The same 

 is true of India-rubber cables incased in lead. 



Herr Guillaume sa3^s of a cable in use bj- the 

 German government, similar to the Faraday 

 cable, "We are using it altogether in our 

 new construction. I do not see how it can ever 

 decay. We tried cotton- covered wires soaked 

 in paraffiue and drawn into lead pipes ; and, 

 though they worked well at first, after a few 

 years they failed." 



W. W. Jacques, Ph.D., 

 Electrician of the American Bell telephone co. 



