THE DETERMINATION OP TRANSATLANTIC LONGITUl>M 



4«T 



the other extreme, and underrated the velocity. He has mooned as it. in the m- 

 ment, the whole distance traversed was only one quarter of a mile of wire. In fact, 

 it was more than this, and the air-spaces besides. What the equivalent in wire would 

 be for these air-spaces it might not be easy to state with precision. Moreover, tin 

 tension of the Leyden jar, as Gaugain is ready to see, was not constant dun Ih 

 discharge, and did not, therefore, conform to the conditions of ohm's calculations. Il 

 however, the theory must not be rigidly applied in opposition to Wheatatone's con- 

 clusion, it is sufficiently clear that the enormous discrepancy between the velocity to 







be deduced from his remarkable experiment and that afterwards reveal by direct 



observation upon long lines of telegraphic 



no 



lont: 



Indeed, tie oar 



rected' velocity from Wheatstone's experiment would be unaccountably kvkiII hen 

 compared with these later determinations, unless they also received ihe proper qualifi- 



cation demanded by the same theory. 



If an objection is made to the inferences drawn from the mathematical th« >ry of 

 electricity, because this theory is perplexed by difficult questions of analysis, a 

 stronger objection holds against the assumption that the velocity is the same for long 



as for short distances 



this 



sustained neither by theory 



doubtedly, the safest way for determining the velocity of elect rieity 



xperiiiient. l ! n 



properly 



very long wires is by experiment- on these same 



periment upon a very large scale was made by Professor Joseph Win 



lock, at' the Observatory of Harvard College, on the nights of February 28 and March 



_.. . . „ ,i rw * — , r i n San Fmneisc.o bv one line of 



stated, the transmission time 



wires. 



An 



7, 1869 



Signals 



were 



from the Observatory to San Francisco 



wire, an 



d returned to the Observatory by the way of Canada, and the times of leaving 



and returning to the Observatory 

 The time required to 



r 



ecorded upon the same chrono raph 



pass through this enormous loop of 7,300 miles of No. 9» 



and thirteen telegraph-repeaters, was 



about two thirds of 



econd 



Thii 



resuit agreed closely with the time obtained from signals sent directly between C 



° J ,. nri ,i iu* ^on«missmn time deuu 



bridge and San Francisco with earth connections, ana 



by Mr. Bradford in his computation 



of 



Coast Survey operations for longitude 



between these twe 

 of a second : thus 



The transmission time in this last case was 



alio 



two fifths 



showing that the velocity was nearly the same, whether the 



tricity discharged into the earth or returned by a loop 



If 



wires, without any 



other difference between them but that of length 



compared, the transmission times 



velocity is inv< 

 unlike, we can 

 vol. ix. 



ay as the length 



the squares of the lengths, so that the average 

 If the diameters and specific conductivities - 



rsely as tne lengui ^ ^ ^ ^^ ^ 



substitute equivalent lengtns 01 <i 



then 



65 



