40 Kansas Academy of Science. 



In a loosely-closed tester the results will depend in a less degree on some of the same 

 conditions, as drafts and proximity of the oil, but in a greater degree on others — such as 

 the amount of vapor consumed in negative trials. Even here it is quite impossible to 

 secure in successive tests an equal quantity of vapor, ('. c, an equal relative saturation in the 

 air of the vapor chamber. How can this be obtained? At the boiling point of a liquid 

 there is no limit to the amount of its vapor which the air can hold. At any lower tem- 

 perature, however, the oil will still evaporate, but only to a definite extent ; and when 

 this is completed the air is said to be saturated. Now we cannot manage practically to 

 stop the evaporation at each trial when the air is just one-fourth or one-half saturated, 

 but if we aim at complete saturation we can easily hit it, for evaporation will stop of it- 

 self, and an equal relative saturation be in this way repeatedly obtained. This may be ac- 

 complished in the closed testers, and at the same time we may secure an equal constancy 

 in the other factors, the calorific power of the vapor, and the economy of its heat. 

 It requires, however, care that the oil and air be sufficiently mixed, that the resulting 

 spray be equally well removed, and such other careful manipulation that the results are 

 in general nearly as divergent as in the loosely-closed testers. 



The saturation of the air has also been undertaken in the open apparatus of Lieber- 

 man (Zeitschrift fiir An. Ch., XXI, 1), and of Stoddard (Am. Ch. Jour., IV, 4), and 

 their labors have so simplified the manipulation that the shadow of an available oil test 

 has appeared — for here the conditions for saturation depend rather on the apparatus it- 

 self than on the handling. The bubbles must be sufficiently minute to effect proximity 

 of every air particle to the oil, sufficiently rapid to secure proper agitation, sufficiently 

 deep to give ample time for vaporization. The outlet of the vapor chamber must be so 

 large that the velocity of the issuing air shall be less than the velocity of flame propa- 

 gation. The current must be uniform, that uniform benefit may accrue from the pro- 

 ducts of combustion. With a vessel of given and proper size and shape, with a given 

 and proper guage for the inlet of the air, which shall be forced in under a given and 

 proper j>ressure, the conditions of the experiment will be virtually constant. 



It remains only to determine the relative saturation of air with oil vapor which may 

 occur in our lamps, and to fix our flash test correspondingly below the maximum lamp 

 temperature, to afford the long desideratum — a rapid, reliable, accurate, and easily ma- 

 nipulated oil test. I hope to be able to determine the proper dimensions for such an 

 apparatus. 



I have endeavored to explain a priori the principle which must underlie a good oil 

 tester, and to explain by it the conspicuous variations in those in use. So also we may 

 account for other phenomena which have been a source of constant inaccuracy in testing 

 oil. Place a flame in a gaseous mixture in which the proportion of the combustible -is 

 rapidly increasing, a slight envelope becomes preceptible and gradually enlarges. What 

 are the sources of the heat units, which at the borders of the envelope have raised the 

 gases to the flame temperature while they fail to beyond? Evidently the combustion 

 there going on and the heat radiated from the central flame. As the amount of combus- 

 tible increases, a less number of heat units are required from the flame, and the en- 

 velope extends to a distance where that smaller amount is received. Finally, it reaches 

 the virtual limit of the flame's influence, when the heat units produced by the combusti- 

 ble are alone sufficient, and the flash or explosion immediately follows. Experiment 

 verities this statement accurately. For an account of the history and methods of oil 

 testing, consult Dr. C. F. Chandler, in Johnson's Cyclopaedia, article, Petroleum; Mr. 

 A. II. Elliott's synoptical bibliography in his report to the New York State Board 

 of Health, Second Annual Report, 1882, reprinted as circular No. 45; and a review in 

 Am. Ch. Journal for October, 1882, IV, pp. 280-8 and 293-4, with the papers in each 

 place cited. 



