122 PROFESSOR W. RAMSAY AND DR. S. YOUNG ON EVAPORATION, ETC. 
of the products of dissociation into a barometer-tube, in which the dissociable sub¬ 
stance was being heated. Wurtz has also shown that the vapour-density of a dis¬ 
sociable compound is increased in presence of one of its components, which necessarily 
implies a decrease of dissociation. Now, it has been found by us from experiments 
with aldehyde ammonia and ammonium chloride that it is immaterial how pressure is 
raised in determining temperatures of volatilization ; but the two cases are very 
different. In the former the gas in excess is mixed with, and acts on the vapour 
evolved from the dissociable body, while in the latter it is a question of the action on 
the solid of the gas introduced, since after the component vapours have left the 
surface of the solid, any action on them would have no influence on the solid. It 
appears most likely that the solid is always surrounded by its own vapour, so that it 
is doubtful whether the introduced gas ever comes in contact with the solid at all. 
§ 85. Difference of behaviour of dissociable solids from dissociable liquids token 
evaporating from a free surface. —So long as the body is solid the residue has 
always its original composition ; but a liquid separates into its components, the 
more volatile passing over first. This w r as found to he the case with chloral hydrate 
and with phthalic and succinic acids, and no doubt holds good with others. It 
amounts to a proof that the solid substance, in volatilizing, does not pass through the 
state of liquid. It would also show that a solid, so long as it remains a solid, cannot 
dissociate ; while a liquid may undergo increased dissociation with rise of temperature, 
thus forming a mixture from which the more volatile constituent separates first. 
§ 86. We have shown in the preceding pages that there is no essential difference, in 
certain cases of dissociation, between the behaviour of dissociating and evaporating- 
bodies ; hut in others, wdiere a deeper chemical change may be supposed to occur, an 
essential difference exists, as shown by a comparison of the curves representing, on the 
one hand, temperatures of volatilization under different pressures, and, on the other, 
vapour-pressures, or pressures of dissociation, at different temperatures. The precise 
nature of this difference is as yet unexplainable, in our present state of ignorance of 
the molecular and atomic constitution of chemical substances. 
