262 On the Principles of Beauty in Colouring. 



the preceding theorem, in order to obtain the real swiftness of 

 sound, to multiply the swiftness dechiced from the formula of 

 Newton by the square root of the atliiiitv of -p\% to -jVoj or by 

 the root off. To the temperature of six deijrees, this formula 

 gives 282-42 for the space which sound ought to traverse in a 

 second : multiplying that by V ^ , that space ought to be equal 

 to 345*3o. The observation of the French academicians makes 

 it only 337' 18. The difference of these two results may be 

 owing to the inaccuracy of the experiments ; but the smallness 

 of that difference establishes in an incontestable manner, that 

 the excess of the swiftness observed above the swiftness calcu- 

 lated by the Newtonian formula, is owing to the latent heat which 

 the com])ression of the air develops. 



It results from what has been stated that, supposing the pres-r 

 sure constant, if we augment a given volume of air by elevating 

 its temperature, and afterwards reduce it by compression to its 

 original volume, it will disengage by that compression a third 

 of the heat employed. It is to be desired that philosophers 

 would determine by direct experiments the affinity of specific 

 heats of the air to constant pressure, and of tlie air to a constant 

 volume: — this affinity we have found equal to 1'5. The swift- 

 ness of sound observed bv the French acadenvcians gives I '4254 

 for this affinity ; and perhaps, considering the difficulty of di- 

 rect experiments, this swiftness is the most precise means of 

 obtaining it. 



I have in the second part of these observations reckoned the 

 swiftness of sound in rain-water and in sea-water equal to 

 2(542-8 and 280/ 4 per second, by dividing the experiuients of 

 Canton upon the compression of these liquids, and having only 

 regard to the linear diminution of the dimensions of the com- 

 pressed volume. I have since found that it is necessary to con- 

 sider the total diminution of this volume, and that the preceding 

 num!)ers ought thus to be divided by V* 3, which reduces them 

 to 1525-8 and 1620-9; so that the swiftness of sound in fresh- 

 water is four degrees and a half greater than in air. 



LXV. On the Principles of Beauty in Colouring. By Mr. 

 T. Tredgold*. 

 J.F the term beautiful be restricted to those objects only which 

 give pleasure from the pleasing associations they recall to the 

 beholder, then the investigation of any other causes, which may 

 contribute towards rendering objects beautiful, would be an use- 

 less labour ; for this limited definition of beauty precludes .all 

 inquiry, and is as imsatisfactory as it is unphilosophical. Qn 

 * Communicated by the Author. 



the 



