LIVING TOBPEDOS IN BERLIN. 421 



means of leading-off plates made of guttapercha shaped like the covers 

 of mummy coffins, and lined with tin at both ends ; these are shown in 

 Fig. 12 ; one was made for each fish according to a pattern cut to 

 proper size. In using the cover a plate of glass was laid at the bottom 

 of the experimental trough, this was filled to such a height that the 

 surface of the water a, a' (for the sake of distinctness put too high 

 in the figure) coincided with the back of the fish which rested on 

 the glass plate. I observed that the insulation of the fish between 

 the saddles more than doubled the shock in the experimental 

 circuit. The insulation was so complete that when the electrodes of 

 the frog-alarum were plunged into the trough at two diametrically 



Fig. 12. 



opposite points of its circumference no indication was obtained. It 

 would have been still more perfect if I had closed the posterior 

 opening of the cover with a cap arching over the caudal fin. 



In contriving this cover the question arose what length should 

 be given to the lining in order to obtain the strongest current 

 in the experimental circuit. I found that "it was advantageous 

 to have a longer lining when the resistance of circuit was great, 

 shorter when it was small I . I had expected this, and it will be 

 seen from what follows that it must be so. 



The connections k, Jc and s, / of the two linings may be re- 

 garded as terminals of the experimental circuit. According to 

 Helmholtz's 'principle of the electromotive surface 2 ,' the strength of 

 the current in the experimental circuit is equal to the difference of 

 potential of the points of the lining which are in contact with its 



1 Gresammelte Abhandlungen, vol. ii. pp. 636, 637. 



2 Poggendorff's Annalen der Physik und Chemie, 1853, vol. Ixxxix. p. 211. 



