226 REPORTS ON THE STATE OF SCIENCE. — 1920. 



change by the quinonoid configuration when the other case of exactly analogous 

 colour change cannot be so explained. 



Another well-known application of the quinonoid hypothesis is to the alkali 

 metal salts of the nitrophenols which are highly coloured. It is stated, for 

 example, that the sodium salt of p-n[tTo^heno\ has the constitution 



«-o=< 



ONa 



If that is so, what is the constitution of the nitropheiiol when in solution in 

 concentrated sulphuric acid, for it is equally coloured under these conditions? A 

 similar coloured solution is obtained when p-nitroanisole is dissolved in sulphuric 

 acid. Many other instances could be quoted, and there is no doubt that the 

 evidence against a direct sfructure-absorption correlation is overwhelmingly 

 great. 



There are two general objectives to any of the theories that have been referred 

 to. In the first place, no theory can be sound which is limited to a very minute 

 section of the spectrum such as the visible and ultra-violet, and in the second 

 place, no theory can hold good unless it rests on a quantitative physical^ basis. 

 There is also another aspect of the phenomenon of absorption, namely, its un- 

 doubted connection with the phenomena of fluorescence and phosphorescence. 

 Just as the selective absorption of light must be due to specific properties of 

 molecules, so also must the emission of light by molecules be due to similar 

 properties. It is evident that any theory must take cognisance of both 

 phenomena. It is true that many theories were advanced to explain the 

 fluorescence of organic compounds, but none of these can be said to hold the 

 field. Devised to explain visible fluorescence they fail entirely to offer any 

 explanation of the ultra-violet fluorescence shown by many compounds. 



In general it may be said that the most recent work on the absorption by 

 organic compounds has increasingly shown that there is some relation between 

 the absorption bands shown by a substance and its reactivity. Perhaps the first 

 observations which supported this view were those of certain amino-aldehydes 

 and -ketones of the aromatic series and their salts with hydrogen chloride. '^ It 

 was found that alcoholic solutions of these compounds exhibit well-marked 

 absorption bands. On the addition of small quantities (O'l to 0"5 eq.) of hydro- 

 chloric acid to these solutions a new absorption band, situated nearer to the red, 

 is developed in each case. On the addition of more acid this band disappears 

 and gives place to the absorption characteristic of the hydrochloride of the 

 original base. This shows that the base as it exists in alcohol solution does not 

 react with the acid to give the salt, but that it is first converted into an inter- 

 merliate or reactive phase which then reacts with more acid to give the salt. 



These observations were extended to manv substances, notably certain 

 phenolic compounds including the nitrophenols.^' The compounds in alcoholic 

 solution exhibit well-marked absorption bands which are not appreciably changed 

 when sulphuric acid is added. When dissolved in concentrated sulphuric arid 

 they develop visible colour due to absorption bands in the visible region. The 

 compounds in sulphuric acid solution, on being allowed tn remain, slowly 

 undergo sulphonation to srive colourless sulphonic acids, Clearly, therefore, 

 these phenols in the condition in which they exist in alcoholic solution do not 

 react with sulphuric acid. When dissolved in strong sulphuric acid they are 

 changed into a reactive phase which slowly reacts with the sulphuric acid to give 

 the snlphonic acid. They are therefore exactly analogous to the amino-aldehydes 

 and -ketones already mentioned. 



It mieht easily be said that the coloured reactive modifications have under- 

 gone a change in structure, but further evidence shows that no change of 

 structure has taken place. The majority of these compounds in alcoholic solu- 

 tion exhibit fluorescence when exposed to light of frequency equal to that of 

 their absorption bands. The frequency of this fluorescent emission has been 

 accurately measured, and it has been found in every case of the abovp-mp'itionp'1 

 substances that the frequency of the fluorescence of the compound in alcoholic 



