12 PROFESSOR TYNDALL ON CALORESCENCE. 



carbon is an extremely inflammable substance ; and on the 2nd of November, while 

 employing a very powerful battery and intensely heated carbon points, the solution took 

 fire, and instantly enveloped the electric lamp and all its appurtenances in flame. The 

 precaution, however, had been taken of placing the entire apparatus in a flat vessel 

 containing water, into which the flaming mass was summarily turned. The bisulphide 

 of carbon being heavier than the water, sank to the bottom, so that the flames were 

 speedily extinguished. Similar accidents occurred twice subsequently. 



Such occurrences caused me to seek earnestly for a substitute for the bisulphide. 

 Pure chloroform, though not so diathermic, transmits the obscure rays pretty copiously, 

 and it freely dissolves iodine. In layers of the thickness employed, however, the solu- 

 tion was not sufficiently opaque ; and in consequence of its absorptive power, but com- 

 paratively feeble effects were obtained with it. The same remark applies to the iodides 

 of methyl and ethyl, to benzol, acetic ether, and other substances. They all dissolve 

 iodine, but they enfeeble the results by their action on the extra-red rays. 



I had special cells constructed for bromine and chloride of sulphur ; neither of these 

 substances is inflammable, but they are both intensely corrosive, and their action upon 

 the lungs and eyes is so irritating as to render their employment impracticable. "With 

 both of these liquids powerful effects were obtained ; still their diathermancy, though 

 very high, did not come up to that of the dissolved iodine. Bichloride of carbon would 

 be invaluable if its solvent power were equal to that of the bisulphide. It is not at all 

 inflammable, and its own diathermancy appears to excel that of the bisulphide. But 

 in reasonable thicknesses the iodine which it can dissolve is not sufficient to render the 

 solution perfectly opaque. The solution forms a purple colour of indescribable beauty. 

 Though unsuited to strict crucial experiments on dark rays, this filter may be employed 

 with good effect in class experiments. 



Thus foiled in my attempts to obtain a solvent equally good and less dangerous than 

 the bisulphide of carbon, I sought to reduce to a minimum the danger of employing 

 this substance. At an early period of the investigation I had constructed a tin camera, 

 within which were placed both the lamp and its mirror. Through an aperture in front, 

 2 1 inches wide, the cone of reflected rays issued, forming a focus outside the camera. 

 Underneath this aperture was riveted a stage, on which the solution of iodine rested, 

 closing the aperture and cutting off all the light. In the first experiments nothing 

 intervened between the cell and the carbon points ; but the perU of thus exposing the 

 bisulphide caused me to make the following improvements. First, a perfectly trans- 

 parent plate of rock-salt, secured in a proper cap, was employed to close the aperture ; 

 and by it all direct communication between the solution and the incandescent carbons 

 was cut 05". The camera itself, however, became quickly heated by the intense radia- 

 tion falling upon it, and the cell containing the solution was liable to be warmed both 

 by the camera and by the luminous rays which it absorbed. The aperture above 

 referred to was therefore surrounded by an annular space, about 2^ inches wide and a 



