Mr. Nicholson on the Composition of Caffein. 115 



to ascertain how far the mechanical equivalent of heat, as de- 

 termined by my recent experiments on the friction of fluids, 

 might be able to contribute to clear up this question. 



The capacity of air at constant pressure, according to the 

 experiments of De la Roche and Berard, is 0*2669. Conse- 

 quently a quantity of heat capable of increasing the tempera- 

 ture of a lb. of water by 1°, will give 1° also to 3'747 lbs. of air, 

 while the air will be expanded i^^j', an expansion in which a 

 force equal to 200*7 lbs. through a foot is expended in raising 

 the atmosphere of the earth. The equivalent of a degree of 

 heat per lb. of water, determined by the careful experiments 

 brought before the British Association at Oxford, is 775 lbs. 

 through a foot. Hence 200*7 lbs. through a foot is equal to 

 0°-259. 



We see, therefore, that for every degree of heat employed 

 by Qe la Roche and Berard in expanding and heating air, 

 0°'259 was occupied in producing the mechanical effect, leaving 

 0°*741 as that actually employed in raising the temperature of 

 the air. Hence the actual specific heat (commonly called 

 capacity at constant volume) is 0*2669 x 0*741 = 0*1977. Ta- 

 king this as the specific heat of air and the equivalent 775, it 

 follows that if avolume of air of 17r6 cubic inches be com- 

 pressed to 170*6 cubic inches, it will be heated 1°, a quantity 

 of heat which will occasion an increased pressure of ^|y. So 

 that the celerity of sound will be increased by this means in the 

 subduplicate ratio of 491 to 661*6, or in the simple ratio of 

 2216 to 2572, which will bring it up from Newton's estimate 

 of 943 to 1095 feet per ", which is as near 1130, the actual 

 velocity at 32°, as could be expected from the nature of the 

 experiments on the specific heat of air, and fully confirms the 

 theory of Laplace. 

 Oak Field, near Manchester, 

 July 17, 1847. 



XXII. On the Composition of Caffein, and of some of its 

 Compounds. By Edward Chambers Nicholson, Esq."^ 



/^AFFEIN was first analysed by Professors Liebig and 

 ^^ Pfafft in 1832. The result of this investigation was 

 confirmed by a subsequent analysis of Prof. Wohler %. 



In 1838 Professor Liebig induced M. Jobst § to analyse 

 thein, who proved this body to be identical with caffein. 

 His analyses gave the same results as his predecessors. The 

 same remark apphes to the experiments of Mulder I| on thein, 



* Communicated by the Chemical Society; having been read Feb. 15, 



t Liebig's Jnnalen, i. 17. % Ibid. § Ibid. xxv. 63. 



II Bulletin des Sciences Phys, et Nat. de Neerlande. 1838, p. 32. 



12 



