Relations of Liquids and their Saturated Vapours. 97 



dent of h ; it expresses merely the ratio of molecular volume to 

 proportionate expansion at r, and is found to have the same 

 value in the bromide of hydrocarbon and some others where the 

 first two constants do not conform. E. g. : — 



Bromide of Hydrocarbon. — The t temperature is 127°, at which 

 = 1-1180*, vapour-density 94, P = 2*61, 7 = 423, A = 211*2, 



1*118x94 A ~ nn k A Aor j. 1 da -n\ ^dt 1 .^ 



A*= — k-ttk^ — =47-62, - = 4-43 (insteadof 4-11), -j- * T =4*09 

 2*207 /A ' dv h 



(instead of 4*32). The product of these is 18*11, almost the 



same as the preceding mean. 



The same is the case with chloride of hydrocarbon and proto- 

 chloride of arsenic. They would conform to the first group 

 entirely, if the values of their h tangents were reduced about rj 

 part. 



§ 44. A large proportion of the liquids examined by M. Pierre 

 have not their vapour-lines established ; but if we assume them 

 to issue from the ether-node, and join that point and their 

 boiling-point on the general chart (Plate III.), we may then lay 

 off the upper limit <y upon this line produced ; and if it range 

 near to the dotted line, it is probably correct. This is the case 

 with the hydrobromate, hydriodate, and probably the hydrochlo- 



rate of methyle, which have the quotients - &c. alike. 



The bichloride of tin and bichloride of silicium are also thus 

 united. Other symmetric relations are indicated in the Table 

 before me, which includes most of Pierre's liquids ; but it would 

 be hazardous to proceed further in this direction without the 

 complete data. 



Evidence of Relations with respect to the Cohesion-Integral of a 

 Molecule in different Liquids. 



§ 45. Let Q= quantity of heat given out by a pound of a 

 certain liquid in descending from f to 0°, and X= quantity of 

 heat given out by a pound of the same body in the condition of 

 saturated vapour at / in passing into a liquid state and de- 

 scending to 0°. Then X — Q=C, the quantity of heat required 

 to change the condition of a pound of the same body from liquid 

 to saturated vapour without performing work — in other words, 

 the integral of liquid cohesion at t. 



Regnault has determined X and Q for a number of bodies 

 through a considerable range that includes the t temperatures of 

 some of our liquids whose thermomolecular lines are well estab- 

 lished. We have thus the data for computing C at such tem- 

 peratures 5 which are near the respective boiling-points. By 



* Ann. de Chim. et de Phys. vol. xx. p. 41. 

 Phil. Mag. S. 4. Vol. 35. No. 235. Feb. 1868. H 



