EXPERIMENTAL KNOWLEDGE OF THE PROPERTIES OF MATTER. 471 



■weight of the chloride ; the change of volume of the chlorine (if any) 

 Avas measured by a method first used by Crafts. 



In the other paper mentioned Biltz discusses the effect of using 

 smaller and smaller quantities of substance with larger vessels in vapour- 

 densities in reference to cases in which at the same temperature different 

 values are obtained, due either to dissociation, or to expansion towards the 

 perfectly gaseous state. 



Vapour-density of Hydrofluoric Acid. 



Gore has found that the volume of hydrofluoric acid obtained by 

 passing hydrogen over heated silver fluoride was, when measured at about 

 100°, twice that of the hydrogen which went to form it ; hence that the 

 formula of hydrofluoric acid is HF, where F is 19 ; but that at lower 

 temperatures the volume of hydrofluoric acid diminished rapidly with 

 diminution of temperature in comparison with the hydrogen.' 



Mallet determined the density of hydrofluoric acid at 30'5°, and found 

 it to correspond to the molecular weight 39o2 at that temperature, 

 nearly the number for HoF,.^ 



Thorpe and Hambly ^ find that at fourteen temperatures ranging at 

 short intervals between 2G-4° and 883° the densities correspond to 

 molecular weights ranging from 51-19 at 26-4° to 20-58 at 88-3°, there 

 being no range of temperatures corresponding to a formula H^Fa ; they 

 therefore consider the case similar to that of acetic acid, and that the only 

 molecular formula is HF. 



The vapour-density determinations by Nilson and Pettersson, by V. 

 Meyer in conjunction with others which have just been referred to, 

 of sulphur by Biltz, of nitric peroxide by Ramsay, of hydrofluoric acid 

 by Thorpe, of aluminic, ferric, and gallic chlorides by Friedel and Crafts, 

 have had for one of their objects that of ascertaining whether certain 

 supposed more complex formulas assigned to an element or a compound 

 could be maintained on investigation in cases where the body in the 

 state of gas at higher temperatures was known to consist of simpler 

 molecules. In some of the cases the answer from experiment has been 

 that there is no longer reason from vapour-densities to believe in the 

 more complex molecular groupings, as in the case of sulphur ; while in 

 other cases, e.g., iodine, nitric peroxide, there seems no room to doubt 

 that the body exists in two distinct states, in one of which the gas or 

 vapour consists of molecules of half the mass of the molecules of the 

 body in the other state ; to this case there may be added the case of 

 the chlorides examined by Friedel and Crafts. Acetic acid and nitric per- 

 oxide have some similar characters in reference to the vapour-densities of 

 the saturated vapours from their liquids, and to the behaviour of equal 

 volumes in relation to pressure and temperature, as shown by Ramsay 

 and Young ; and it remains to be seen whether acetic acid does not con- 

 sist' at low temperatures of molecules in which are combined two of the 

 molecules of acetic acid at moderately high temperatures. 



In the case of bodies which are dissociated by heat the dissociation is 

 gradual, and in general although not always the temperature-range from 

 the beginning to the end of the dissociation is considerable ; this seems 

 to be the distinguishing feature of bodies undergoing dissociation, and 



' FMl. Trans. 1869, p. 173. - Amer. Chem. Journ. 3, 1881, p. 189. 



» C. S. J. 1888 ; Trans, p. 765. 



