100 THE TRANSATLANTIC LONGITUDE. 



IIL 



1. 



Valencia, 

 Newfoundland, 



0.26T 



0.472 



0.624 



IIL 



2. 



Valencia, 

 Newfoundland, 



0.284 



0.420 



, 0.592 



III. 



3. 



Valencia, 

 Newfoundland, 



0.252 



0.340 



0.392 



IV. 



1. 



Valencia, 

 Newfoundland, 



0.264 

 0.332 







IV. 



2. 



Valencia, 

 Newfoundland, 



0.224 

 0.262 







IV. 



3. 



Valencia, 

 Newfoundland, 



1 0.238 







The experiments IV. 2 and IV. 3 differ only in that the return-circuit is formed 

 by the earth in the former case, and by the second cable in the laiter. The trans- 

 mission-time appears in both instances to be 0'.24. For the Newfoundland signals 

 in Experiments II. 2, and II. 3, the same difference exists, and the transmission-time 

 appears to be 0'.28 in the former, and 0',26 in the latter case. It would seem 

 therefore that the velocity was but little, if any, affected by this great change in 

 the character of the circuit, with a battery of 4 cells. 



In the first and third series, the signals from one" station were given by breaking 

 and making circuit, but from the other in the ordinary Avay by alternate currents, 

 so that the 2d and 3d experiments of each series differed from one another by the 

 tension of the zincodes having been destroyed in the former by an earth-connection, 

 leaving the tension to reach the cables from the platinodes only. The results give 



Val. I. Newf. III. Mean. 



Experiment 2, 0'.2n 0^284 0'.278 



Experiment 3, 0.301 0.252 0.216 



or an average transmission-time of 0'.28 in each case, using 4 cells. 



In the first experiment of Series I and III, one half the circuit was formed by 

 the earth, while the cables had 2 cells at each end. In the second experiment of 

 Series IV, the earth formed one half the circuit, and the cables had 4 ceUs at the 

 sending station. The results give : — 





Val. 



■Newf. 



Mean. 



LI, IIL 1, 



0'.202 



0^267 



0=.234 



IV. 2, 



0.224 



0.262 



0.243 



The Valencia signals of Series I were made November 10 ; all the others were on 

 November 16, without other difference of circumstances than those in the connec- 

 tions as described. No difference in the velocity appears to have been produced by 

 the changed arrangement of the 4 cells which constituted the battery. 



It is not without hesitation that I present the facts and inferences of this chapter. 

 For I am not unaware of the careful and thorough quantitative investigations ot 

 Thomson, Jenkin, and others, and should of course shrink from publishing these 

 relatively crude and very incomplete results, were it to be supposed that I regarded 

 them as comparable with those obtained by those distinguished electricians. But 



