. III. BETWEEN ALCOHOL AND WATER. 57 



the current is promoted from the external to the internal 

 surface, by being directed always from the water to the 

 alcohol. In various and frequently repeated experiments, 

 we have observed an elevation of 20, 24, and 40 millime- 

 tres when the internal surface of the skin was placed towards 

 the alcohol, whilst in the reverse position the corresponding 

 elevations were only 4, 12, and 20 millimetres. Under 

 analogous circumstances, the position of the membrane 

 being favourable, the elevation was 28 millimetres ; in the 

 other position, on the contrary, there was no elevation. 

 With the skin of the eel the reverse takes place. With this 

 the current is favoured from the internal to the external sur- 

 face ; and whilst the alcohol contained in the instrument 

 rises to the height of 20 millimetres, when in contact with 

 the external surface of the skin, it only rises to 10 millime- 

 tres when the position is reversed. 



This difference of elevation always takes place in the 

 same direction, and is confirmed for the eel-skin, as for the 

 skin of the torpedo. The elevation has been at 50 milli- 

 metres in one instrument, and 20 in the other. 



Some anomalies which we observed in our earliest trials, 

 led us to study, with greater precision, the circumstances 

 under which endosmose takes place through the skin of the 

 torpedo, when interposed between water and alcohol. We 

 constantly found these differences when the skin of the tor- 

 pedo was fresh, and had not been used for previous experi- 

 ments, but it continued only during the first hour of the 

 experiment or a little after. The elevations subsequently 

 follow a different law, and the height in the endosmometer, 

 when the external surface of the skin is in contact with the 

 water, goes on diminishing, afterwards ceases, and ultimate- 

 ly the direction of the current changes. 



Out of the numerous experiments which we have tried, 

 we select the following, in which we have marked the ele- 



