La,:ip for cenfuiniHg Tallow. EKperiimtii on Candla, 71 



arife in the project for affording this gradual fupply as it may be wanted. A cylindrical 

 piece of tallow was inferted into a metallic tube, the upper aperture of which was partly 

 clofed by a ring, and the central part occupied by a metallic piece nearly refembling that 

 part of the common lamp which carries the wick. In this apparatus the piece laft defcribed 

 was intended to anfvver the fame purpofe, and was provided with a Ihort wick. The cylin- 

 der of tallow was fupported beneath in fuch a manner that the metallic tube and other part 

 of this lamp were left to reft with their whole weight upon the tallow at the ring or con- 

 traftion of the upper aperture. In this fituation the lamp was lighted. It burned for 

 fome time with a very bright clear flame, which, when compared with that of a . candle, 

 pofleffed the advantage of uniform intenfity, and was much fuperior to the ordinary flame 

 of a lamp in its colour, and the perfed abfence of fmell. After fome minutes it began to 

 decay, and very foon afterwards went out. Upon examination it was found, that the me- 

 tallic piece which carried the wick had fufed a fufficient quantity of tallow for the fupply 

 during the combullion ; that part of this tallow had flowed beneath the ring, and to other 

 remote parts of the apparatus, beyond the influence of the flame ; in confequence of 

 which, the tube, and the cylinder of tallow were faftened togetlier, and the expefled pro- 

 grefTion of fupply prevented. It feems probable, that in every lamp for burning confiftent 

 oils, the material ought to be fo difpofed that it may defcend to the flame upon the principle 

 of the fountain refervoir. I fhall not here flate the obftacles which prefent themfelves in 

 the prcfped of this conftruftion, but fliall difmifs the fubjeft by remarking, that a con- 

 trivance of this nature would be of the greateft public utility. 



The wick of a candle, being furrounded by the flame, is nearly in the fituation of a body 

 expofed to deftrudive diftillation in a clofe veflel. After lofingits volatile produfts, the car- 

 bonaceous refidue retains its figure, until, by the defcent of the flame, the external air can 

 have accefs to its upper extremity. But, in this cafe, the requifite combuRion, which 

 might fnufl^it, is not efle£led. For the portion of oil emitted by the long wick is not only 

 too large to be perfeflly burned, but alfo carries ofi^much of the heat of the flame while 

 it aflumes the elaftic flate. By this diminiflied combuftion and increafed efflux of half- 

 dccompofed oil, a portion of coal or foot is depofited on the upper part of the wick, which 

 gradually accumulates, and at length afl"umes the appearance of a fungus. The candle does 

 not then give more than one-tenth of the light emitted in its beft ftate. Hence it is that 

 a candle of tallow cannot fpontaneoufly fnuflf itfelf. It was not probable that the ad- 

 dition of a fubftance containing vital air or oxygene would fupply that principle at the 

 precife period of time required ; but, as experiment is the tett of every probability of 

 this nature, I foaked a wick of cotton in a folution of nitre, then dried it, and made a 

 candle. When this came to be lighted, nothing remarkable happened for a Ihort time ; ar 

 the expiration of which a decrepitation followed at the lower extremity of the flame, which 

 completely divided the wick where the blackened part commences. The whole of the 

 matter in combufiion therefore fell ofl', and the candle was of courfe inflantly extinguiflied. 

 Whether this would have happened in all proportions of the fait or conftruaions of tlie candle 

 I did not try, bccaufe the fmcll of azote was fufficicntly ftrong and unpleafant to forbid tlie 

 ufe of nitre in the purfuit. From various confldcrations I am difpofed to think that the 

 fpontaneous fnuffing of candles made of tallow, or other fufible materials, will fcarccly be 

 eficftcd but by the difcovcry of fome material for the wick which fliall be voluminous 



enough 



