on Endosmose and Ewosmose. 321 



brane is replaced in the endosmometer by porous plates of 

 sandstone or carbonate of lime ; but plates of baked white 

 clay comport themselves in a very different manner. I shut 

 the terminal mouth of an endosmometer with a plate of clay 

 of three-eighths of an inch in thickness ; I put distilled wa- 

 ter into the interior of the endosmometer, which was itself 

 plunged into distilled water ; I then put the interior water in 

 relation with the negative conjunctive wire of the pile, the ex- 

 terior water being in contact with the positive conjunctive 

 wire. The introduction of the water across the plate of clay 

 became very rapid, the interior water soon gained the level of 

 the exterior water ; it entered the tube, and ascended with 

 great quickness. The ascent of the water in the tube lasted 

 as long as the action of the pile continued. This experiment 

 convinced me that porous aluminous solids are apt, like the or- 

 ganic membranes, to cause the impulsion of the water under 

 the influence of an electric current from the positive pole to 

 the negative pole. Thus these solids present, like the organic 

 membranes, the phenomenon of the endosmose by means of 

 the electricity of the pile. It then became necessary to know 

 if these various solid minerals are susceptible of presenting, 

 like the organic membranes, the phenomenon of endosmose, by 

 means of the contact of their opposite faces with homogeneous 

 fluids. I luted to the endosmometer a plate of soft freestone 

 one-sixth of an inch thick, and having put a solution of gum 

 arabic into the interior, I plunged it into pure water. It did 

 not show any endosmose ; the interior gummy fluid did not as- 

 cend in the tube above the exterior level. I replaced the plate 

 of freestone by a plate of porous calcareous carbonate of one- 

 third of an inch in thickness. I obtained no effect of the endos- 

 mose from it. Thinking that this absence of the appearance of 

 endosmose might proceed from the too great thickness of my 

 porous plates, I placed upon my apparatus a plate of porous car- 

 bonate of lime, one-third of an inch in thickness. I did not ob- 

 tain any appearance of endosmose. With the same want of 

 success I made use of a plate of unbaked plaster, (calcareous 

 sulphate of lime), a little more than one-sixth of an inch thick. 

 I have employed for the same purpose, and with as little suc- 

 cess, crystallized sulphate of lime, which, as is well known, 

 divides into extremely thin plates ; but here the want of the 



