on the Surface of Water. 417 



being propelled on by successive films to the curved surface of 

 the glass, divides and curls round in two opposite directions, 

 thus producing a pair of wheels for each face of the camphor, 

 which the lycopodium renders distinctly visible. I have allowed 

 this action to go on during sixty hours with no other interrup- 

 tion than having to lower the stick two or three times when a 

 portion had been cut off by the sawing action of the surface- 

 water. 



Now this process, like a machine in motion which goes on so 

 long as it is wound up, fails unless free course be given to the eva- 

 poration of the camphor-film. The experiment cannot be con- 

 ducted in a large bottle. The camphor has been made to dip 

 into the water contained in a clean bottle : at first there were 

 faint indications of a current ; but these soon ceased. After 

 many hours some of the water was poured from the bottle into 

 an open vessel ; and the moment the camphor was lowered into it, 

 the currents set in with much of their accustomed vigour. The 

 experiment also fails if the lycopodium dust be laid on too thickly; 

 a very faint shower from a muslin bag is sufficient for the pur- 

 pose. The motions are more vigorous on a bright clear day 

 "than on a dull cloudy one, more active in summer than in 

 winter. 



25. That this experiment depended on the constant formation 

 and evaporation of a film of camphor seemed to be evident from 

 the perfect way in which it could be imitated by means of ether. 

 At the end of a narrow tube a bit of sponge was tied, and the 

 tube filled with ether was supported vertically about an inch 

 above the surface of water previously dusted with lycopodium ; 

 a very perfect, sharply cut, well-defined disk of ether is formed 

 on the surface of the water by the condensation of the vapour 

 pouring down from the sponge. The disk does not increase in 

 diameter, but the excess of ether pours off from it and proceeds 

 radially to the surface of the glass, where each branch curls round 

 in two opposite directions, throwing the powder into pairs of 

 wheels precisely as in the case of the camphor current (24). 



26. Another phenomenon, which I named " camphor pulsa- 

 tions," seemed also to illustrate the view I had taken of these 

 motions. A stick of camphor with a square base is lowered so 

 as to touch the bottom of a shallow glass vessel 6 or 7 inches in 

 diameter, containing a little water, not more than about two 

 ounces. As soon as the camphor touches the water the wmole 

 surface becomes agitated with rapid pulsations, at least 250 per 

 minute. As the water soon becomes saturated, the pulsations 

 gradually diminish to 60 or 80 per minute, and they may even 

 sink down to 8 or 10 per minute. 



According to my explanation, as soon as the camphor is low- 



