OP LIGHT AFFORDED 1Y CANDLES. 41 



Thefe experiments were made in the following manner. 



I took three candles, the dimensions of which are given in Method of 

 the table, againft 1, 3, and mould *. Thefe were firft weighed, experiment, 

 and then lighted at the fame inftant. At the end of the time 

 inferted in the third column of the above table, they were ex- 

 tinguished and weighed again, and the lofs of weight of each 

 candle is contained in the fourth column. 



The three firft: experiments were made under fuch favour- Ohfervations on 

 able circumflances, that I have little doubt of their refults the refults * 

 being more accurate than what practical utility requires, but 

 the fourth experiment cannot be depended on fo much, in 

 confequence of the variable light of No. 5. This candle was 

 moved fo often to keep the two fliadows equal, that I was 

 under the neceffity of fetting down its mean diftance from the 

 wall by eftimation ; but as this was done before the candles 

 were weighed, my mind could not be under the influence of 

 partiality for a fyftem. 



The method which I ufed in comparing one light with an- Intenfity of 

 other in each experiment, was nearly that which has been al- Jj 8h u det "T'jL e * 

 ready defcribedf. The only difference confifted in having of /kadows, 

 the obftacle which formed the (hadows fixed, inftead of being 

 held in the hand. A two feet navigation fcale, which was 

 made faft to the ftand of a fmall telefcope in a perpendicular 

 direction, and fet upon a table near the wall on which the 

 fhadows were compared, reduced the labour of making the 

 experiments very much, and gave me an opportunity of mak- 

 ing a greater number of obfervations in the fame time than I 

 could have done with the fcale in my hand. 



The ftandard 



REMARKS ON THE EXPERIMENTS. companion of 



experiments at 



1 . The experiments were made at different times, and the different times 

 light of the mould candle was made the ftandard, with which candle ^ tllc 

 the lights of the others were compared ; but it muft not be weight con- 

 underftood, that this candle gave the fame ftrength of light SSStalJS* 

 in every experiment. infered. 



2. The fign -f- in the 5 th column, fignifies that the candle Other remarks, 

 againft which it is placed, gave a ftronger light than the others 



* Philofophical Journal, o&avo, III. 273. 

 f Philofophical Journal, quarto, I. 67. and Philof. Journal, 

 «8aYO, III. 275, 



in 



