52 Professor CErstecl on iraief-Spoiils. 



not only not increase the force of the engine by adding to the 

 elasticity of the steam, but would render requisite the injec- 

 tion of a larger quantity of cold water into the condenser to 

 effect condensation, and thereby add to the burden of the aii*- 

 pump." 



There must be a point at which the resistance of vapour 

 not abstracted, to the descent of the piston, and the pressm'e 

 of the atmosphere on the air-pump whilst discharging its load, 

 are at a minimum. Beyond this, if it be attempted to reduce 

 the force of the vapour, by injecting more cokT water, the 

 bui'den of the air-pump is increased by the exposure of its 

 piston to the atmosphere for a longer time during its dis- 

 charge ; whilst, on the other hand, if it be sought to lessen 

 the duration of atmospheric pressure on the air-pump, by di- 

 minishing the quantity of cold water introduced into the con- 

 denser, the increased elasticity of the unabstracted vapour 

 offers a greater resistance to the descent of the piston, t 



This subject presents many inviting topics of inquiry ; but 

 the pursuit of them, and the earlier preparation of the details:): 

 which I have now the honour to submit to the Institution, 

 have been prevented by more pressing occupations. 



4 Clarence Street, Penzance, 

 August 30. 1837. 



On JFater-Spouts. By Hans Christian QErsted, Professor 

 of Natural Philosophy in the University of Copenhagen. 



All naturalists, except those who have themselves proposed 

 an explanation of the water-spout, § are agreed in thinking 

 that science has hitherto given us but little satisfactory infor- 

 mation on the phenomenon. This may, in some measure 



* Brewster's Edinburgh Journal of Science, O. S. IX, p. 162. 



t Ibid. X. p. 40. 



X A short notice of t lese experiments appeared in Brewster's Edinbui-gh 

 Journal of Science, N. S. VI. p. 246. 



§ As we have no English term exactly equivalent to the German Wetter. 

 taule, we have employed the name oi vatcr-spovt in its place, throughout the 

 present article. — Ed . 



