308 BESEABCHES ON EVArORATION AND DISSOCIATION. 



A few vapoiir-doiiHity determinations of some of tliese 

 aubstanoBH wore also made by Hofmann's motiiod. 



The table on page 309 gives a summary of the results. 



The behaviour of chloral hydrate deserves special notice, 

 because the temperature of volatili^iation was perhaps not 

 strictly independent of pressure, for at low pressures it was 

 ])0ssible to heat the substance above its melting-point with- 

 out molting; and the lower tlie pressure, the higlier the 

 temperature to which it could be raised. When the pros- 

 sure was raised to about CO mms. the temperature gradually 

 fell, and on reaching the melting-point, 50-()°, the sub- 

 stance molted. It could again be caused to solidify by 

 lowering the pressure at once, and the temperature then 

 rose again. 



It may also be noticed that it made no difference in the 

 results with aldehyde ammonia and ammonium chloride, 

 whether the pressure was raised by admission of air, or of 

 either of the gaseous constituents of the dissociating com- 

 pound. 



It is evident that these substances may be divided into 

 two groups: that in which the curves representing tempera- 

 tures of volatilization and vapour-pressures are identical ; 

 and that in which these curves are distinct. The members 

 of the first class behave like stable solids and liquids ; and 

 the class includes the three substances, ammonium chloride, 

 nitrogen peroxide, and acetic acid. With the first of these, 

 dissociation is nearly complete 60° below the temperature 

 of volatilisation at normal pressure ; with the second, dis- 

 sociation amounts to less than 20 p.c. at its boiling-point ; 

 while the dissociation of acetic acid rests on indirect 

 evidence. 



The second group, which contains the rest of these sub- 

 stances, may be divided into two sections : that in which 



