Trans. N. Y. Ac. Set. 138 Mar. 13, 



from a higher-boiling oil ; but that the intensity of the light given by 

 the low-boiling oil is greater than that given by the high-boiling oil. 

 When these low-boiling oils are allowed to mix with the kerosene, it 

 will be seen that they give brilliancy to the light, besides adding to the 

 quantity of product obtainable from a given crude petroleum. This is 

 the commercial side of the question. Now let us look at it from an- 

 other standpoint — that of the consumer. These low boiling-oils give 

 off vapors, and these vapors, when mixed even in a small proportion 

 with air, form explosive mixtures. Then comes the problem to be 

 solved in ascertaining the safety of kerosene oil. At what temperature 

 does a given sample of oil give off inflammable vapors ? And at what 

 temperature must such a sample not give off these vapors? The latter 

 question has already been determined and varies in different States of 

 the Union ; some consider 100° F. as sufficiently high ; others require 

 that the oil shall not give off inflammable vapors below 140^ F. These 

 temperatures have been fixed upon after experiments made with burning 

 lamps, the details of which I will not trouble you with. Having de- 

 cided at what temperature an oil shall not give off an inflammable vapor, 

 it becomes necessary to fix the conditions under which the oil shall be 

 tested, to determine whether it is or is not giving off this inflammable 

 vapor. For this purpose, quite a number of pieces of apparatus have 

 been devised ; and in experimenting with these, I have reached some 

 results that may prove not uninteresting. 



The various forms of apparatus used in testing kerosene for inflam- 

 mability, and the presence of inflammable vapors, may be divided into 

 two classes : first, those in which the safety of the oil is determined 

 in an indirect manner ; and, secondly, those where the direct ignition of 

 the vapors evolved is resorted to. 



In the first class, we have the apparatus of Salleron and Urbain, of 

 Paris ; where an attempt is made to determine the presence of inflam- 

 mable vapors by finding the tension of the vapor of the oil at any given 

 temperature, the idea being that the oil containing naphtha, benzine, 

 and the low-boiling oils, will give a vapor-tension greater than an oil 

 without these lower products. But, unfortunately, the apparatus de- 

 vised for this purpose does not give reliable results, and is of little 

 value. 



Another experimenter has sought to determine the safety of kerosene 

 by obtaining the index of refraction of light, when transmitted through 

 a given sample of oil. But experiments have proved that there is no 

 direct relation between these indices and the quantity of low-boiling 

 oil present in a sample of kerosene ; indeed, the results are utterly 

 valueless in the question of safety. 



The second class of apparatus, in which a direct ignition of the 



