316 Proceedings of the Royal Physical Society. 



bleach the vapour as it is formed. These 885 and 865 oils 

 are either washed with acid and a weak solution of soda, 

 and thereafter the low melting point paraffines separated, or 

 the paraffines are first separated, and they are then washed. 

 The paraffine from the decimal 865 oil has a melting point of 

 about 95° Fahrenheit, and that from the decimal 885 oil of 

 about 100° Fahrenheit. In order to separate these paraffines, 

 the oil has to be frozen down to about 18° Fahrenheit. 



We now come to the action of selenium upon the oils. 

 About two months ago, in the Uphall Oil Company's works, 

 a discolouration was noticed in some burning oil in the 

 process of manufacture, and before long the same discoloura- 

 tion was noticed in the whole of the oils, both heavy and 

 light. There are many things which might have caused this 

 discolouration, among others, under-treatment with vitriol in 

 the first stages, or vitriol which contained nitric acid as an 

 impurity being used in the later stages. In order to prevent 

 confusion hereafter, I shall call sulphuric acid by its com- 

 mercial title, vitriol. 



As in these works the first of those causes was carefully 

 noticed and prevented, this cause was at once put aside ; but 

 in regard to the latter, we were not equally certain. The 

 vitriol from our stock gave decided traces of nitric acid, and 

 of course we at once blamed that impurity ; but as we are 

 supplied from two vitriol works, it could not be decided, till 

 further supplies of the acid arrived, who was the erring party. 

 Next day two tanks of vitriol came in from the different 

 makers. One of them showed distinct traces of nitric acid, 

 with both the ferrous sulphate and indigo tests. The other 

 gave the indigo but not the iron test. They both were of 

 course at once rejected, but the second is the one with which 

 we were more particularly concerned. Why it should give 

 the indigo and not the iron test was rather a mystery. On 

 receiving our notice of rejection, the makers at once sent out 

 and sampled the acid, and thereafter sent the sample to an 

 Edinburgh chemist, who reported that it only contained 

 decimal 005 per cent, of nitric acid by the nitrometer, an 

 instrument which I had not by me at the time. 



This result puzzled us considerably, and we went on using 



