H 



CLARKE 



Compound. 



Methjlamine.... 



Ethylamine 



Propylamine 



Isobutjlamine... 



Amylamine 



Dimethylamine 

 Diethylamine ... 

 Trimethylamine 

 Triethylamine... 

 Allylamine 



Formula. 



CH5N 



CjH,N 



C3H9N 



C,H„N 



CsHijN 



C2H7N 



C,H„N 



C3H9N 



CeHjsN 



CsHjN 



4^. 



HjO Liquid. 



1033280 

 1662680 

 2302960 

 2901440 

 3562320 

 1681840 

 2938000 

 2330520 

 4209520 

 2125120 



H.O Gas. 



929050 

 1516758 

 21 15346 

 2672134 

 3291322 

 I535918 

 2708694 

 2142906 

 3896830 

 I979I98 



Divisor. Quotient. 



69 

 III 



153 



237 

 III 



195 



153 

 279 



147 

 Mean, 13771 



13465 

 13664 

 13826 

 13703 

 13887 



13837 

 13891 

 14006 

 13967 

 13463 



Aniline, pyridine and piperidine, being ring compounds, will 

 be considered later. For ammonia, as measured by Thomsen, 

 the heat of combustion is 90 650 calories, or 75 01 1 with the water 

 gaseous. Hence 4^^=300044. The proper divisor is 27, 

 which gives a quotient of 11 113, a wide departure from the rule. 

 To ammonia, therefore, the formula fails to apply. Whether 

 the error is in the observation or the theory, remains to be de- 

 termined. The ten amines, however, seem to be perfectly 

 regular. 



We have now studied forty-three compounds, or fifty-two if 

 we include the bromination series of Louguinine and Kablukoff, 

 which seem to obey the new general law. Each one gives an 

 absolute heat of formation proportional to the number of atomic 

 unions in its molecule, and each union has the thermal value of 

 two henotherms. The latter uniformity of value, however, does 

 not hold universally, and some exceptions to it are already evi- 

 dent. For instance, the absolute heat of formation assigned to 

 carbon dioxide is twelve henotherms, or six for each of its two 

 atomic linkings, a quantity three times as large as the normal. 

 We may suppose, therefore, that different series of compounds 

 may have different fundamental values, and also that mixed 

 series may occur. This appears to be the case with the few 

 gaseous cyanogen compounds investigated by Thomsen, for 

 which the formula must be modified. 



If to these compounds, four in number, we apply the formula 

 which was satisfactory with the amines, we shall obtain quo- 

 tients which are from one to two thousand calories below the 



