924 REPORT—1904. 
isomeric changes are limited to a comparatively small group of substances, 
reversible polymeric changes occur very frequently, not only in complex 
organic compounds, but also in the simplest inorganic substances, includ- 
ing even the elements. Perhaps the most notable difference consists in 
the fact that equilibrium is largely influenced, and indeed mainly deter- 
mined, by the temperature and pressure, conditions which produce only 
small changes in the equilibrium between dynamic isomerides. 
In most cases equilibrium is attained almost instantaneously. The 
properties of freshly melted ice are perfectly normal, and the depolymeri- 
sation of NO, is so rapid that it is not possible to detect any time-factor. 
Amongst organic compounds, however, gradual changes have occasionally 
been noticed. Formaldehyde shows a slow decrease of molecular weight 
in freshly diluted solutions,! and freshly diluted or freshly cooled solutions 
of gelatine only slowly assume their normal viscosity, The most striking 
examples of gradual changes of this type are to be found amongst the 
organic nitroso- compounds which exist in a blue or green monomolecular 
and a colourless dimolecular form.2 In the case of nitrosobutane, 
CMe.,;NO, Bamberger and Seligman* have plotted a complete curve 
showing the gradual depolymerisation in a solution in benzene at the 
freezing-point ; equilibrium is reached in four hours, and the decrease of 
molecular weight proceeds simultaneously with the development of the 
blue colour. 
There is reason to believe, however, that under favourable conditions 
gradual association and dissociation are not infrequent amongst simple 
inorganic compounds. Only in this way can the remarkable facts 
be explained that have been noted by many observers in studying the 
critical phenomena of gases, and to which attention has recently been 
called by Traube.t Thus, when liquid carbon dioxide is heated above its 
critical temperature, the upper and lower layers of gas, though easily 
miscible, remain distinct for a considerable period of time, and only 
gradually diffuse into one another. Under apparently identical conditions 
the density of the gas may vary in the ratio of 1 : 2°16. So also when the 
gas is cooled from above its critical temperature, neither the liquid nor 
the vapour has at first its normal density ; the meniscus is gradually dis- 
placed through several centimetres, equilibrium being attained only 
after a week has elapsed. Traube explains these results by assuming the 
existence of ‘ gasogenic’ and ‘liquidogenic’ molecules, but the variation 
of physical properties with time, to which reference has been so frequently 
made in the preceding pages, affords clear evidence of the occurrence of 
chemical change, and it can scarcely be doubted that the phenomena, uf 
not due to inequalities of temperature or pressure, afford indications of a 
reversible polymeric change similar in character to, but slower than, 
those which take place so rapidly in the case of water and of nitrogen 
peroxide. 
' Inaug. Diss. Rostoch, Eschweiler ; Abst. 1890, 954. 
* Meyer, Ber. 1888, 21, 507; 1896, 29, 94; Thiele, Abstr. 1894, i. 217; Baeyer, 
Abstr. 1894, i. 252; Piloty, Ber, 1898, 81, 218, 221, 452, 457, 1878; 1902, 35, 3113; 
1903, 36, 1297; Schmidt, Ber, 1903, 36, 1765; Bamberger and Rising, Ber. 1901, 
84, 3877. 
3 Ber. 19038, 36, 689. * Ann. d. Physik, 1902, 8, 2, 267. 
