54 MOTION'S OF ODOROUS SUBSTANCES. 



being able to touch it. It is neceffary that the water fiiould 

 be very pure, and the veflel perfectly clean. We may hold 

 the piece of camphor with a pair of tweezers, or at the end 

 of a glafs tube ; it mufl be cut into a conical form, as has been 

 mentioned ; a piece of a larger fize and an irregular form 

 would envelope the light body in its atmofphere, and it would 

 not move with equal facility. 



The fame effects are produced, if we employ, inflead of 

 camphor, a fmaU piece of fine fponge foaked in camphorated 

 water, or merely a tube of glafs, containing at its extremity a 

 drop of the fame folution. 



Variation. If we cover a china plate with a very thin layer of water, 



and approach the piece of camphor of the preceding experi- 

 ment within a few millemetres of it, prefenting it by the 

 point, in fuch a manner that the axis of the cone is perpen- 

 dicular with the furface of the layer, the water recedes under 

 the cone, and forms a circle concentric with it. The interior 

 part of this circle is coloured with prifmatic ranges, proceed- 

 ing from the prolongation of the axis, and extending from the 

 centre towards the circumference with a very rapid motion ; 

 after fome moments, the circle lofes its colour from the centre 

 to the circumference, and the iris finally difappears, whether 

 we continue to hold the camphor over the furface of the layer 

 or not. It is indifferent whether we hold the plate in a ho- 

 rizontal or vertical pofition : the circle is always formed in a 

 direction perpendicular to the axis of the fmall cone of cam- 

 phor. I obferved thefe phenomena at the temperature of 15° 

 of # Reaumur's thermometer. 



Ether on water. Laftly, if we throw a fmall piece of fine fponge foaked in 

 ether upon water, it is inftantaneoufly thrown into motion like 

 camphor; and a hiding noife is heard, fimilar to that of water 

 during its converfion into vapour upon a hot iron. If we 

 view the furface of the water horizontally, before a light win- 

 dow, we obferve fparkling jets iflliing from the fponge, which 

 extend themfelves in ferpentine windings upon the furface of 

 the water, to the diffance of fome centimetres, and there pro- 

 duce irides fimilar to thofe of the preceding experiment. 

 Thefe irides foon difappear. During this emiffion, the fponge 

 has a progreffive and a rotatory motion, which are evidently 

 owing to thofe fmall jets, the impulfe of which it is obferved 

 conllantly to obey. 



Of 



