WATER MADE TO WET MERCURY. gj 



then plunged to the same level in water, an apparent endosmosis through the cracks is 

 the result, for the alcohol rises with considerable velocity in the tube. It is, however, 

 only an apparent endosmosis, for, upon closer examination, it will be found that the 

 motion stops when the hydrostatic equilibrium is adjusted through the chink (193). Any 

 of the very porous minerals show the same thing. 



197. If, in the last experiment, the tube be filled with lime-water, and then immersed 

 to the same level in a solution of oxalic acid, the appearance described in (187) will be 

 reproduced. This cautions us not to impute to membranes any predilection for passage 

 in certain directions ; for that may arise from extraneous circumstances, and in the in- 

 stance referred to, as will presently be shown, originates in a very different cause. 



198. The relation existing between solids and fluids which determines their descent 

 or rise in capillary tubes, has been referred to heretofore in these papers (123). A con- 

 nexion of the phenomena of endosmosis and capillary attraction might have been traced to 

 the fact that no liquid will pass through a barrier the surface of the pores of which it 

 cannot wet. The relation of glass and quicksilver to each other, in this point of view, is 

 interesting. When a piece of glass is laid upon the surface of this fluid metal, contact 

 between them does not take place, but they are separated from each other by an ex- 

 ceedingly small interval. As I had failed in reproducing many of the results attempted 

 by means of artificial chinks in glass (196), because of their magnitude, I was led to 

 hope that better success would attend the same attempts by making use of the small 

 interstice between glass and mercury. A tube half an inch in diameter was therefore 

 taken, and one of its extremities, having been ground truly flat, had its roughness taken 

 off 'by exposing it carefully to the blowpipe flame. When the tube was lowered with 

 this extremity downward, on the surface of some pure mercury, all the parts of its 

 circumference touched the metal at once. A solution of green vitriol was placed in it 

 to a certain height, and upon the mercury ; on the outside of it was poured a solution of 

 ferrocyanate of potassa. It was expected that, through the chink between the mercury 

 and the glass, the liquids would slowly infiltrate to each other. After several days, no 

 such action was observed, and the experiment, though repeated under a variety of con- 

 ditions, afforded no better result. Now water and saline solutions, as will hereafter be 

 shown, pass through interstices much more minute than this can reasonably be supposed 

 to be ; it is evident, therefore, that the want of action is mainly due to the circumstance 

 that water and saline solutions generally do not wet mercury, and the laws of capillary 

 action would indicate that, under these circumstances, they would fail to pass through 

 the chink. 



199. The event of this experiment points out, in an impressive manner, the general 

 relations that must exist, between the solids and liquids of organized bodies. Water 

 \vill pass with great rapidity through a chink the width of which is not more than the 

 half of a millionth part of an inch, provided it can wet both sides of that chink ; but if 

 this condition be not fulfilled it fails to pass, even though the width should increase to 

 upward of one hundred and forty-four times its former dimensions. 



200. That the non-passage described is here referred to the true cause, will appear 

 from the following experiments. As has been stated, under ordinary circumstances, 



