044 CAPILLARY ATTRACTION, ETC. [MEMOIR XXVI. 



quois also showed me an agate baft of a knife, wherein 

 was a certain cloud which an ingenious person had for 

 some years observed to change its place in the stone." 



Some years ago, having occasion to make an analysis 

 of certain Roman silver coins which had been long bur- 

 ied in the earth, I found that much of the alloying cop- 

 per had made its way to the surface, constituting the 

 green patina of antiquarians, and that the silver had be- 

 come comparatively pure. An interstitial movement in 

 these denarii must therefore have taken place a move- 

 ment so slow that it had required many centuries to 

 yield the observed result. 



If one end of a porous substance, such as a sponge, be 

 dipped into water, the liquid very soon percolates in all 

 directions through the mass, which becomes charged with 

 as much as it can hold. If a piece of glass having a 

 crack in it be put into water so that the end of the 

 crack is immersed, the liquid instantly runs spontaneous- 

 ly to the other end. And if from the crack other smaller 

 ones branch forth, along these also the water rapidly 

 finds its way. 



These effects have long been studied by using slender 

 glass tubes, which operate in the same manner as cracks, 

 but permit the phenomena to be observed in a more con- 

 venient and exact manner. 



Water will pass through a crevice the width of which 

 is less than one half of the millionth of an inch. The 



proof of this is readily 

 obtained experimentally. 

 If we take a convex lens, 

 a a (Fig. 67), of long fo- 

 cus, and place it upon a glass plane, b , there will be seen 

 at the point of apparent contact, <?, on looking down, a 

 black spot surrounded by a series of variously colored 

 concentric circles, the appearance being well known 



