486 ■ EEPOBT— 1882. 



the exclusion of its occluded gases unless its store has been replenished in the 

 interval. Tacchini has recently observed that Encke's comet, which is one of short 

 period that does not travel beyond the limits of our system, shows the usual spec- 

 trum of three bands. If this observation should be confirmed it will be very 

 difficult to account for the replenishment of its occluded erases. But Encke's comet 

 is a very small object, and one cannot feel very certain about its spectrum, and it 

 will be interesting to see, when HaUey's comet next returns, whether it shuies onlj' 

 by reflected light or gives us, like so many others, the banded spectrum of hydro- 

 carbons. 



The following Papers were read : — 



1. On tlie Legal Flashing Test for Petroleum. 

 By F. A. Abel, C.B., F.B.S., F.G.S. 



The author commenced by pointing out defects in the old legal flashing test, 

 called the open test, and in the test used in America, called the fire test, and the 

 circumstances that led to his investigation of the subject, resulting in the perfection 

 of a close flashing test, which in 1879 became the legal flashing test by Act of Par- 

 liament, a'nd which since that time has been used in the United Kingdom with 

 satisfactory resvdts. After a time this test had been adopted in the United States 

 and by the German Government. In course of the investigations instituted in 

 Germany Dr. Foerster and others had noticed (what had also become apparent 

 in this country) that decided variations in atmospheric pressure introduced con- 

 siderable variations in the test, when appUed to one and the same sample of oil, 

 either at difierent localities or during a continuance of very ditferent atmospheric 

 conditions. This had led the German Government to adopt a system of corrections 

 for atmospheric pressure in applying the test to the examination of petroleum 

 samples. The subject had been examined into by the author and Mr. B. lled- 

 wood, and a series of oljservations, instituted by the latter in different parts of 

 Switzerland, differing considerably in altitude, had indicated that very constant 

 •differences in the flashing point of a particular oil were occasioned by variations of 

 pressure, the difference appearing to amount to about two degrees Fahr. per inch of 

 mercury. The subject was still imder investigation, with a view to the adoption 

 of a proper system of correction in practice. Another point was the effect of con- 

 siderable variations in atmospheric temperature upon the flashing point of an oil. 

 Messrs. Abel and Redwood had carried on a series of experiments for the piu-pose 

 of ascertaining to what extent variations in the temperature of a sample, and in 

 the temperature of the locality where it was examined, would affect the flashing 

 point under the Abel test, and it had been established that it was not merely neces- 

 sary to cool down the sample of oil immediately before testing it, when the tem- 

 perature of the atmosphere exceeded 65° Fahr., according to the directions given in 

 the Act, but that it was also imperative to maintain the oil to be tested for some 

 considerable period at a low temperature, if it had been stored for any length of 

 time in a locality where the temperature exceeded that above specified. 



The Indian Government had recently adopted, in a Petroleum Act, without any 

 modification, the instructions relating to the use of the flashing test included in the 

 English Act, and the serious difficulties to which this had given rise in connection 

 vsdth large importations of petroleum into India from the United States, where they 

 had passed the prescribed test, while on arrival at Bombay they exhibited consider- 

 ably lower flashing points, had led the author to examine into the subject in conjunc- 

 tion with Mr. Redwood, who has lately proceeded to India with the view of 

 thoroughly examining into the conditions to be fulfilled for securing the attainment 

 of trustworthy results by the application of the flashing test in a tropical climate. 



2. The Influence of Aqueous Vapour on the Fxplosion of Garionic Oxide and 

 Oxygen} By H. B. Dixon, M.A. 



' Printed in Chemical Xeyvs, vol. xcvi. p. 151. 



