178 ANNUAL OF SCIENTIFIC DISCOVERY. 



condenser is made to convey a stream of air, mingled with ammonia, 

 into the condensing tube some inches above the flame of the burning 

 gas. To secure a larger and more constant unit of illumination than 

 the candle commonly used, a lamp burning kerosene, with a flat wick, 

 is employed, in which, by means of a bridge of platinum wire, the 

 flame may be maintained of constant size, and giving a light equal to 

 about seven candles. This is supported on a balance of peculiar 

 construction, giving the consumption during the experiment. Prof. 

 Rogers had found that even the small amount of carbonic acid which 

 in some gas-works is allowed to remain in the gas produces a sensible 

 reduction of the light. This effect, varying with the strength of the 

 illuminating gas, was found to range from three to nearly five per cent 

 of the illuminating power for each per cent of the impurity : 58 per cent 

 of carbonic acid, although it did not prevent combustion, made the 

 flame so dim as to be without effect on the photometer. 



LUMINOUS AND OBSCURE RADIATIONS. 



| 



Professor Tyndall, in a paper in a recent number of the Philosophical 

 Magazine, after referring to the discoveries of the Herschels and Mel- 

 loni, respecting the obscure red rays of the end of the spectrum, says : 

 Dr. Akin inferred, from the p.iucity of luminous rays evident to the 

 eye, and a like paucity of extra-violet rays, as proved by the experiments 

 of Dr. Miller, that the radiation from a flame of hydrogen must be 

 mainly extra-red : and he concluded from this that the glowing of 

 platinum wire in a hydrogen flame, as also the brightness of the Drurn- 

 mond liirht in the oxyhydrogen flame, was produced bv a change in the 



O v * ** O 



period of vibration." Dr. Txndall adds, that by a different mode of 

 reasoning he arrived at the same conclusion. The paper referred to, con- 

 tains a description of his numerous experiments demonstrating the char- 

 acter of the radiation from a huirogen flame under a variety of conditions, 

 the apparatus employed, and the results obtained. We select a few of 

 the results. Fifty experiments on the radiant heat of a hydrog- n fla ; e 

 make the transmission of its rays through a quantity of iodine, which is 

 perfectly opaque to light, 100 per cent. To the radiation from a hydrogen 

 flame the dissolved iodine is therefore perfectly transparent. It is also 

 sensibly transparent to the radiation from solid bodies heaied under in- 

 candescence, and to the obscure rays emitted by luminous bodies. Prof. 

 Tyndall found that, dividing the radiation from a platinum wire raised to 

 a dazzling whiteness by on electric current into 24 equal parts, one of 

 the.-e parts is luminous and 23 obscure ; dividing the radiation from ihe 

 mo t brilliant portion of a flinie of c al-gas into 2<3 equal parts, one of 

 these parts is luminous and 24 obscure : dividing the ladiation from the 

 electric light emitted by carbon points and excited by a Grove's battery of 

 40 cells into 10 equal parts, one of these parts is luminous and nine obscure. 

 "We have space only for the following remarkable conclusions : " On a 

 tolerably clear night a candle flame can be readily seen at the distance 

 of a mile. The intensity of the electric light, used by me is 6u() times 

 that of a good composite candle, and as the non-luminous radiation from 

 the coal points which reaches the retina is equal 10 twice the luminous, 

 it follows that at a common distance of a foot the energy of the invisible 

 rays of the electric light which reach the optic nerve, but are incompetent 

 to provoke vision, is 1,300 times that of the light of a candle. But the 



