Comhijton of Phofphorus in the Vacuum of the Air Pump. 281 



VIII. Though this explanation appeared to me to be very evidently founded upon the 

 knowledge we already poflefs, I was neverthclefs defirous of (hewing the truth by a diredl: 

 experiment. I endeavoured to afcertain, by means of a thermometer, that the temperature 

 near the furface of the phofphorus which is wrapped in cotton, is more elevated before the 

 inflammation, when the light is perceptibly ftronger. For this purpofe, I ufed a thermo- 

 meter, the bulb of which was about one quarter of an inch in diameter. I fattened the 

 cotton with which the fniall (tick, of phofphorus was furrounded, as in the preceding ex- 

 periments, to the ball of this thermometer, fo that it was entirely furrounded with it, at the 

 fame time that its diftance from the furface of the phofphorus was about half a line. The 

 refult of this experiment anfwered my expeflations. I obfervcd that the mercury rofe, after 

 the light had increafed, and that fron\ 52 degrees, which was the original temperature, it had 

 rifen to 67 of Fahrenheit before the phofphorus had taken fire. 



The fize of the ball of this thermometer appeared to me to be the caufe why the mercury 

 did not rife higher before inflammation. I had alfo admitted fome diftance between the bulb 

 and the phofphorusiin order that its conta£l: might not prevent the phofphorus from acquir- 

 ing the requifite degree of heat which is neceflary for its inflammation. I therefore refolved 

 to repeat the experiment with a thermometer of the fame kind as was ufed by Dr. Hunter 

 in his obfervations on theheat of animals and plants, the ball of v/hich was not more than one 

 line in diameter. 1 fattened cotton to this in the fame manner, which was attached to a fmall 

 flick of phofphorus ; but the ball of the thermometer was placed in contaft with one of the 

 ends of the Rick of phofphorus, which was half a line in diameter, and four lines in length. 

 I then faw that while the light round the phofphorus became ftronger and ftronger, the ther- 

 mometer rofe from 46 to 76 degrees before the phofphorus took fire. The fudden heat 

 broke the ball of the thermometer ; which prevented my ufing fimilar thermometers in the 

 repetition of this experiment. 



Though this thermometer did not mark tlie degree of heat which was neceflary to fct fire 

 to phofphorus in the open air, it is feen, neverthclefs, from the experiment, that the tem- 

 perature rifes very confiderably at the furface of the phofphorus before it takes fire. The 

 ball of the thermometer which I ufed for this experiment had its glafs extraordinarily thick at 

 bottom, which was probably one of the reafons why the thermometer did not indicate an 

 higher temperature before the inflammation, as there is no reafon to fuppofe that phofphorus 

 will take fire in rarefied air at a lefs temperature than in the atmofphere *. 



IX. Laftly, I examined whether the air could be rarefied to fuch a degree that phofphorus 

 fhould be incapable of taking fire in it. I faw it take fire in air fo rarefied that the mercury 

 in the gauge ftood at the height of one finglc line only. The phenomenon here defcribed 

 is hitherto, as far as I know, the only example of a true inflam.maiion in air rarefied to the 

 higheft poflible degree by the air pump. This fa£l, however, docs in no refped prove that 

 a true inflamination can take place in vacuo. 



When the mercury in the gauge placed beneath the receiver ftands at the height of one 



* A« the folidity of ilic ball of ilie ihcimomctLr was abnm half a cubic line, and ihc phufphorus llitic more 

 than three quarter*, it (houlU fccin tl\M, as the licat was flviirtd between the mercury and the pholplioric vapour 

 the thermometer could not be expelled to rife to ilie point of coiubullion unlefs it hjid remained tuibtokcn for a 

 kjager time. N 



Vol. I. — September 1797. Oo line 



