ox ABSORPTION SPECTRA OF ORGANIC COMPOUNDS. 307 



There IS, niorcuvci', not uiie tittle of evidence to support this structural juggling 

 with molecules, and the whole conception, as previously stated, is based on the 

 mistaken notion that a molecule must change its structure because it ch^ingcs its 

 absorption frequency in the visible or ultra-violet and nowhere else in the 

 spectrum. 



One of the greatest protagonists of this theory is Hantzscli, whose brilliant 

 experimental work in support of his views was described in the last report 

 of the Committee. It is of some interest to note that the foundation of his 

 work lies in acetoacetic ester and its two ethyl derivatives, ethyl diethylaceto- 

 acetate, and ethyl /3-ethoxycrotonate. Since acetoacetic ester in alkaline solu- 

 tion exhibits an absorption different from that shown by each of its ethyl deriva- 

 tives, Hantzsch said it must have a different structiire, and on this basis he 

 built up his theory. On the other hand, there is to be recorded the damning 

 fact that ethyl diethylacetoacetate in alkaline solution exhibits the same absorp- 

 tion as does the parent ester in the same solvent. Hantzsch countered this 

 observation by stating that the diethyl compound must have contained some of 

 the monoethyl compound. Some recent work, however, with absolutely pure 

 ethyl diethylacetoacetate has entirely confirmed the original observation, which 

 strikes at the root of the whole of Hantzsch's work. These investigations, 

 which have been accurately carried out with more modern apparatus than 

 Hantzsch had at his dispo.sal, have also proved that none of the absorption 

 phenomena exhibited by these compounds can possibly be explained by any 

 structural theory. 



Another interesting example of this phenomenon, described by Hantzsch' 

 and already mentioned, is diphenylvioluric acid, which forms a series of 

 differently coloured salts with the alkali metals, Li. Na, K, Rb, and Cs. Since' 

 these salts show different absorption bands Hantzsch suggested different struc- 

 tures for them, but they are only different phases of one molecule, established' 

 by the supply of one or more molecular quanta bv the alkali hydroxide. In 

 order to avoid misconception, it must be stated that the complex first formed 

 is C16H13O4N3, MOH, and the fact that water is subsequently lost does not 

 alter what has been said previously. If any doubt be felt as to' the correctness 

 of the explanation by the formation of different phases, this must surely vanish 

 when It is remembered tliat the various frequencies exhibited by these salts are 

 integral multiples of the molecular fiequency of the diphenylvioluric acid mole- 

 cule, since they show constant frequency differences amongst themselves 



It might be considered that the theory of molecular phases is incomplete 

 without any evidence of the free existence of a compound in phases different 

 from those in which it normally exists. To such criticism the reply micrht be 

 made that the three physical states of a compound— solid, liquid and gas— are 

 due to Its existence in different phase equilibria, this being proved by the 

 quantitative basis that was established above on the phase theory of specific 

 heat and latent heat. This, however, might not be considered to be sufficiently 

 convincing, because the change in phase which accompanies the change in state 

 has not except m a few instances, been directly pro'ved bv absorption spectra 

 observations, since as a general rule the phas,3 change concerns those molecules 

 which exist in phases characterised by frequencies in the extreme ultra-violet 

 On the other hand many cases are known of compounds which have been 

 obtained m phases of higher energy content than their normal phases, and indeed 

 these metastable phases are often visibly coloured, whilst the normal and stable 

 phases are white or only faintly yellow. A typical instance is trinitrotoluene, 

 the normal colour of which is slightly yellow. In solution in piperidine trinitro 

 toluene exists partly in a phase characterised by a red colour, arid if this solution 

 IS poured into excess of strong hydrochloric acid and the resulting brownish-red 

 solid IS dissolved in benzene and the solution filtered, the metastable red-coloured 

 phase, which is insoluble in benzene, is obtained as an oil. This substance is 

 pure trinitrotoluene, and shows the same absorption spectrum as the" original 

 p.peridine solution. On treatment with nitric acid it at once reverts to the 

 normal and stable form. leveits to the 



1 ^"°t'ie'- instance is afforded by the so-called «c(-nitrophenoI ethers obtained 

 by Hantzsch and Gorke, who treated the silver salts of the nitroXno s with 

 alkyl iodides at very low temperatures. These substances were coSered to 



