ON THE ACTION OF LIGHT UPON DYED COLOUES. 381 



fastness to light among the various reds, and each of the five classes into' 

 which they have been here arbitrarily divided includes colours which differ 

 from each other more or less in this respect. 



The Action of Light on the Hydracids of the Halogens in presence 

 of Oxygen. — Report of the Committee, consisting of Dr. W. J. 

 Russell, Captain W. de W. Abney, Professor W. N. Hartley,, 

 Professor W. Eamsay, and Dr. A. Eichardson (Secretary). 



Since the last report was presented the attention of the Committee 

 has been directed to a consideration of the conditions necessary to start 

 the decomposition of moist gaseous hydrogen chloride, and of aqueous 

 solutions of the acid when exposed to the combined influence of sunlight 

 and oxygen. It has been repeatedly noticed that, although decomposition 

 of the gaseous mixture, when once started, proceeds at a fairly uniform 

 rate in different samples, yet the time of exposure necessary to start 

 the decomposition varies within very wide limits, although the con- 

 ditions under which exposure is made appear to be the same in each 

 case. It was also noticed that there was more difficulty in starting de- 

 composition in hard than in soft glass tubes. This seems to indicate 

 that the nature of the glass itself materially affects the initial stage of 

 decomposition, which is dependent upon the length of time during 

 which the acid has been kept in contact with the glass, as is borne out 

 by such results as the following. Nine glass tubes having been filled 

 with aqueous solutions of the acid of varying strength were exposed to 

 sunlight. At the end of six months it was found that the most concen- 

 trated of these solutions had been decomposed, the others being un- 

 changed, while after twelve months the three strongest of the remaining 

 solutions showed by their yellow colour that they also had been decom- 

 posed. 



This is quite explicable on the ground that the stronger acid more 

 rapidly dissolves out the constituents of the glass, and suggested that the 

 presence of some metallic chloride is required to start the decomposition 

 of the acid. Following up this line a large number of experiments have 

 been made on the influence of metallic chlorides in promoting decom- 

 position, and, although the results are not sufficiently advanced to allow 

 of our giving full details at present, they appear fully to bear out the 

 above hypothesis. For instance, it was found that the addition of a 

 minute quantity of pure dry alumina to a tube containing moist hydrogen 

 chloride and oxygen brought about rapid decomposition of the acid on 

 exposure to light, while precisely similar samples to which no alumina 

 had been added remained stable for long periods. 



The Investigation of Isomeric Naphthalene Derivatives. — Seventh 

 Report of the Committee, consisting of Professor W. A. Tilden 

 and Professor H. E. Armstrong (Secretary). (Drawn up by 

 Professor ARMSTRONG.) 



In previous reports attention has been over and over again directed to 

 the alpha-law of suhstitution as the dominant lata in the case of naphtha- 



