122 On the Cause of the Earthquakes in the Ionian Isle^. 



canic, without some indication through the medium of springs of 

 the proximity of volcanic fires. 



To revert to the phenomenon of the streams of sea-water flow- 

 ing into the earth, — in connecting them with the production of 

 earthquakes, according to the notion I have started, — of course, I 

 do not consider them more concerned than so many descending 

 streams of fresh water ; the water, not the saU, being supposed 

 to be operative. They arrest attention, however, more, and in- 

 cite to inquiry and speculation more from their singularity, and 

 as apparently denoting vast space below the surface appropriate 

 to the operation of water, in the manner which I have imagined. 



When reflecting on the probable influence of marl under the 

 action of water, I was led to make some experiments on marl 

 and clay, and different substances, in relation to their penetra- 

 bility by this fluid. I shall mention briefly some of the results, 

 as they appear to me to possess some interest, especially in con- 

 nexion with the matter of inquiry under consideration. The 

 form of the experiment was very simple. A glass tube was se- 

 lected about an inch in diameter, and about two feet long, which 

 was covered at one end with linen, and placed perpendicularly. 

 It was filled with the substance to be tried, about two-thirds, 

 and water was poured into the empty portion. In the instances 

 in which freestone, or marble in powder, or even carbonate of 

 lime, procured by precipitation, were used, the water penetrated 

 rapidly ; some reached the bottom and flowed out in drops ; in 

 a iii"^ hours it was drained off, the substance tried merely re- 

 maining wet. In the instances, on the contrary, in which marl 

 or clay was usetl, introduced in |X)wder, the water penetrated 

 with extreme slowness. I have not by me the notes of the ex- 

 periments to consult ; but I \Qry well recollect that, after three 

 or four weeks, it had not penetrated into the marl more than an 

 inch. When these substances, in a coarser form, were employed, 

 as in small masses, then of course the descent of the water was 

 more rapid ; but no sooner had the marl become swollen from 

 the absorption of water, and reduced to powder, than it arrested 

 the farther rajnd descent of the fluid, and the experiments im- 

 mediately became similar to that first mentioned. 



It was my wish to have continued and multiplied these ex- 

 periments ; but circumstances which it would be tedious to men- 



