of Fish Oil. 279 



a larsfc quantity, or about two-thirds more than was at first put 

 into the vessel. The wliole quantity amounted to about twenty 

 or thirty (gallons. The vessel resembled that exhibited in the 

 model ; it was about three feet long, fifteen inches wide, and 

 fifteen in depth. It was made of wrought iron connected with 

 rivets. This experiment was continued only for a few honrs. Mr. 

 Taylor wished to ascertain whether common oil niixed with oil 

 previously boiled, would produce inflanmialjle vapour at a low 

 temperature. They could not ascertain the temperature till it 

 came to SO degrees. A vapour, with considerable noise, began 

 to rise at about 100 degrees; and at 2S0 degrees a light was se- 

 veral times applied to the top of the tube. The vapour tiien took 

 fire with sudden gusts, as with an explosion in the boiler. By ex- 

 plosion he meant a noise like that of fat boiling: it burnt only 

 when a light was applied to the tube. When a candle was held 

 to the vapour it assumed an appearance like that of lightning. 

 The thermometer broke at a temperature of 348 degrees. He be- 

 lieved he broke the stem with his hand, but the ball was also 

 broken, and this could not have happened in that way. Witness 

 afterwards received directions to empty the boiler, to clean it 

 well out, and charge it wi*h S3 gallons of new whale-oil, which 

 had never before been subjected to any experiment, Tiiis was 

 on the loth February, and the experiment was continued during 

 twelve days. The fire was generally made al)out six o'clock, and 

 kept up eleven hours each day. The boiler was always suffered 

 to cool during the night-time. The highest degree to which the 

 oil was ever heated was 507. On the first day the highest was 

 380, and no inflammable vapour was produced. The tube em- 

 ployed on this occasion was alDout four feet in height. On the 

 second day the temperature was at 375, and at that degree the 

 air appeared slightly inflammable. On the third day, at 395, 

 there were no symptoms of inflammability ; but on the fourth the 

 lowest temperature at which it was produced was 348. At 360 

 the vapour took fire on the application of a light, and witness 

 then remarked that a diiference was created by tlie slow or brisk 

 burning of the fire. At 3H0, with a slow fire, the vapour was 

 scarcely inflammable, but with a sharp one was highly so at 360. 

 On the fifth day inflannna!)iiitv appeared at 365, 360, and 345. 

 On the sixth day, tlie vapour was hi<ihly inflammai)le at 345. 

 The seventh day Jirrived, and twelve gallons of oil were taken oi:t 

 of the Iioiler, (the oil in their boiler being deeper than in that 

 indicated i)y the model,) in order to make the experiment as fair 

 as possible. The liiiihcst degree to which it was tliat day heated 

 was li'Ji), but at 335 it threw out a slightly inflammable vapour. 

 This was the lowest point at which it had been hitherto observed. 

 He was at the commencement desired not to raise the tempera- 

 ture 



