320 F, B. GUTHRIE. 



spacial grouping of the atoms. In the case of allotropism, 

 no satisfactory explanation has yet been advanced, and any 

 hypothesis which will explain this phenomenon must have 

 a profound bearing on our conception of molecular structure. 

 The hypothesis most generally accepted to account for 

 allotropism is that it is due to the ability of the element to 

 form molecules of varying numbers of atoms. This is true, 

 for instance, in the case of oxygen at ordinary temperatures, 

 and experimental proof is available of the condensation in 

 the case of the ozone molecule, but attempts to account 

 for other cases of allotropism in the same way have not 

 been successful. Erdmann 1 assumes a molecular structure 

 S 3 , for the second dark liquid form of sulphur, on the analogy 

 of oxygen, and proposes for it the name " thiozone." This 

 form is, however, a phase and not an allotropic form, and 

 is dependent on temperature. There is no experimental 

 proof that the sulphur molecule breaks up into any other 

 than S 2 molecules. 



In the case of other elements, this variation in the num- 

 ber of atoms composing the molecule is only observed in the 

 state of vapour. Such variations, when observed, depend 

 upon the temperature, (and, in the case of sulphur at its 

 boiling point, on the pressure), and afford no explanation 

 of different physical forms existing at ordinary tempera- 

 tures and pressures. Iodine, for example, which forms no 

 allotropes, varies like sulphur in respect to the molecular 

 constitution of its vapour at different temperatures; at 

 600°, the molecule of iodine contains two atoms, whereas 

 above 1,500°, it is monatomic. This variation in the vapour 

 density does not constitute allotropism. 



In the case of isomerism amongst organic compounds, 

 the direction taken by the negatively charged corpuscles 

 depends upon the relative electric conditions of the neigh- 



1 Erdmann, Annalen. 1908, 362, 133. 



