330 Prof. Tyndall on Luminous and Obscure Radiation. 



permit of its being substituted for the ordinary glass train of a 

 Duboscq's electric lamp. A double rock-salt lens placed in the 

 camera rendered the rays parallel; the parallel rays then passed 

 through a slit, and a second rock-salt lens placed without the 

 camera produced, at an appropriate distance, an image of this 

 slit. Behind this lens was placed a rock-salt prism, while late- 

 rally stood a thermo-electric pile intended to examine the spec- 

 trum produced by the prism. Within the camera of the electric 

 lamp was placed a burner with a single aperture, so that the 

 flame issuing from it occupied the position usually taken up by 

 the coal points. This burner was connected with a T-piece, 

 from which two pieces of india-rubber tubing were carried, the one 

 to a large hydrogen-holder, the other to the gas-pipe of the labo- 

 ratory. It was thus in my power to have, at will, either the gas- 

 flame or the hydrogen-flame. When the former was employed, 

 I had a visible spectrum, which enabled me to fix the thermo- 

 electric pile in its proper position. To obtain the hydrogen- 

 flame, it was only necessary to turn on the hydrogen until it 

 reached the gas-flame and was ignited ; then to turn off the gas 

 and leave the hydrogen-flame behind. In this way, indeed, the 

 one flame could be substituted for the other without opening the 

 door of the camera, or producing any change in the positions of 

 the source, the lenses, the prism, and the pile. 



3. The thermo-electric pile employed is a beautiful instrument 

 constructed by Ruhmkorff. It belongs to my friend Mr. Gassiot, 

 and consists of a single row of elements properly mounted and 

 attached to a double brass screen. It has in front two silvered 

 edges, which, by means of a screw, can be caused to close upon 

 the pile so as to render its face as narrow as desirable, reducing 

 it to the width of the finest hair, or, indeed, shutting it off 

 altogether. By means of a small handle and long screw, the 

 plate of brass and the pile attached to it can be moved gently to 

 and fro, and thus the vertical slit of the pile can be caused to 

 traverse the entire spectrum, or to pass beyond it in both direc- 

 tions. The width of the spectrum was in each case equal to the 

 length of the face of the pile, which was connected with an ex- 

 tremely delicate galvanometer. 



4. I began with a luminous gas-flame. The spectrum being 

 cast upon the brass screen (which, to render the colours more 

 visible, was covered with tinfoil), the pile was gradually moved 

 in the direction from blue to red, until the deflection of the gal- 

 vanometer became a maximum. To reach this it was necessary 

 to pass entirely through the spectrum and a little way beyond 

 the red ; the deflection then observed was 



30°. 

 When the pile was moved in either direction from this position, 

 the deflection diminished. 



