i 



NEW CLASSES OF GALVANIC CONDUCTORS. 235 



studied, as it is connected with chemical and physiological 

 phenomena. In fact, except in the case of sparks alone, 

 there is no decomposition that takes place but in conductors 

 of this class ; and all the parts of organized bodies, that 

 galvanic electricity is capable of modifying, equally belong 

 to it. 



4. Or the given substance, acting as a perfect conductor 4. Positive con- 

 when applied separately to either of the two poles, is found ductors * 

 nevertheless to belong exclusively to the positive pole, as 



soon as it is applied to both at once to complete the galvanic 



circuit. Conductors of this kind do not close the circle V 



completely from their insulating the negative effect; and in 



the contact of the two poles by their interposition, we can 



neither charge the positive, nor discharge the negative 



pole. 



5. Or lastly, the effect mentioned in the preceding para- 5. Negative 

 graph is inverted, that is to say, the substance, that acts on COU(lucl0VS - 

 either pole separately as a perfect conductor, belongs en- 

 tirely to the negative pole, as soon as it is applied simul- 

 taneously to the two extremities of the pile. Hence results 



a maximum of electric tension in the positive pole, and the 

 impossibility of producing any divergence at the negative 

 side by the intervention of substances of this class. 



The phenomena of" the first and second class have been Phenomena of 

 known too long to excite our attention, though they furnish * he ^t. 2d, and 

 many interesting particulars. Those of the third I suppose known. 

 to be equally known ; and therefore I shall confine myself 

 to the facts, that demonstrate the existence of conductors 

 of the fourth and fifth classes. These facts, beside their 

 novelty, afford some interesting problems to be solved, and 

 new views to be pursued in galvanic researches. 



Before I proceed to the new facts I have to offer, I shall Must be exa- 

 observe, that the phenomena in question do not exhibit")'"*; / vtie 

 themselves with clearness, and in their whole Connexion, but alone. 

 as far as they are studied with the assistance of the electro- 

 meter alone applied directly to each pole ; and without hav- Tnconvenien- 

 ing recourse to the condenser, the employment of which ^°g r the Cvn " 

 being always interrupted, and its language frequently equi- 

 vocal, sometimes even deceitful, it fetters the progress of 

 the observation, and never allows us to take it in at one view 

 R 2 all 



