OF ARTS AND SCIENCES: JANUARY 9, 1872. 347 



of the vessel with the guard, the direction of the needle was also re- 

 versed. That the current did not arise from the mere contact of the 

 vessel with the guard was shown by applying its limbs unfilled with 

 liquid ; in this case no appreciable deflection of the needle resulted. 

 That it arose from the action of two fluids separated by a membrane, 

 one of them being the connecting liquid in which the cushions of the 

 galvanometer were soaked, was evident by changing the composition of 

 the liquid in the limbs of the U tube. 



Desiring: to test the direction of the current in order to see if 

 it passed through the U tube, or merely, having arisen at the extremities 

 of the tube, passed through the galvanometer back to its origin, I 

 used an apparatus by means of which I was enabled to contract the 

 section of one of the limbs of the U tube. This in all cases reduced 

 the deflection of the needle in a marked manner, a complete constriction 

 of the limb brinsring; the needle to rest. 



My experiments led me to believe, that when the cushions of the 

 galvanometer are connected by a membranous sac containing fluid, 

 or animal tissue saturated with fluid, an endosmotic action takes place 

 accompanied by a galvanic action ; and that this galvanic action is de- 

 termined by the difference of endosmotic action at various points of 

 the enclosing membrane. 



When, therefore, a muscle is laid upon the cushions with its trans- 

 verse section upon one pad, and its longitudinal section upon the other, 

 this difference of endosmotic action obtains, and a galvanic current 

 results. 



If the muscular and nerve currents exist, this endosmotic action 

 must greatly modify their direction and strength. If they do not exist, 

 the endosmotic action, or the electrical action of two fluids separated 

 by a clay or membranous partition, is sufficient to produce currents 

 which may be mistaken for the muscular currents, properly so called. 



In order to study these phenomena, the same precautions must be 

 taken as are advised by Du Bois Reymond (Untersuchungen iiber 

 Thierische Electricitat. Bd. I. : Beschreibung, etc., 1863, p. 95, et 

 seq.) in the study of muscular currents. It must be considered that the 

 fluids in the fresh muscle in their natural position are more energetic 

 in their chemical nature than when they are taken from the muscle or 

 nerve and experimented upon. 



I used in my experiments one of Sir William Thompson's Reflect- 

 ing Galvanometers, and also his new Quadrant Electrometer. 



