VOL. XC.] PHILOSOPHICAL TRANSACTIONS. 759 



This thin silk stops 760 scattered rays 

 of heat, and 916 of light. 



This substance stops 714 scattered 

 rays of heat, and 737 of light. 



Transmission of Terrestrial Flame-heat through various Substances. — My ap- 

 paratus for the purpose of transmitting flame-heat is as follows. Fig. 1, pi. 14, a 

 box 22 inches long, 5-^ broad, and 1^ deep, has a hole in the centre l T v inch in 

 diameter, through which a wax candle, thick enough entirely to fill it, is to be put 

 at the bottom ; the box being properly elevated for the purpose. There must be 2 

 lateral holes in the bottom, 2 inches long, and 1^ broad, one on each side of the 

 candle, to supply it with a current of air, as otherwise it will not give a steady flame, 

 which is absolutely necessary. At the distance of 1-^ inch from the candle, on 

 each side, are 2 screens, 12 inches square, with a hole in each, 4 inch in diameter, 

 through which the heat of the candle passes to the 2 thermometers, which are to be 

 placed in opposite directions, one on each side of the table. Care must be taken to 

 place them exactly at the same distance from the centre of the flame, as otherwise 

 they will not receive equal quantities of heat. The scales, and their supports also, 

 must be so kept out of the way of heat coming from the candle, that they may not 

 scatter it back on the balls, but suffer all that is not intercepted by them to pass 

 freely forwards in the box, and downwards, through openings cut in the bottom. 

 Before the transmitting holes, between the 2 wooden screens, must be 2 covers of 

 the same material, close to the openings, fig. 2 ; and it will be necessary to join 

 these covers at the side, by a common handle, that they may be removed together, 

 without disturbing any part of the apparatus, when the experiment is to begin. The 

 glasses are to be put before the thermometer, close to the transmitting hole, by 

 placing them on a small support below, while the upper part is held close to the 

 screen by a light plummet suspended by a thread which is fastened on one side, and 

 passes over the glass, to a hook on the other side. 



In making experiments, many attentions are necessary, such as, keeping the 

 candle exactly to a certain height, that the brightest part of the flame may be just 

 in the centre of the two transmitting holes : that the wick may be always straight, 

 and not, by bending, approach nearer to one thermometer than to the other : that 

 the wax-cup of the candle be kept clean, and never suffered to run over, &c. Be- 

 fore, and now and then between, the observations also, the thermometers must be 

 tried a few degrees, that it may be seen whether they act equally ; and the candle, 

 during the time they cool down to the temperature, must be put out by an extin- 

 guisher, large enough to rest on the bottom of the box, without touching any part 

 of the wax. Many other precautions I need not mention, as they will soon be dis- 

 covered by any one who may repeat such experiments. 



Min. Candle. Bluish-white glass. From this experiment we find, that 



Exper. 63. 591- $9h wtile me ra ys of the candle gave 3 



5 62| 6'0-| . . . 3 : V g = .375. degrees of heat to the thermometer 



