96 Prof. Magnus on the Propagation of Heat in Gases. 



experiments, the inside was lined with a black, rough, non-lus- 

 trous paper, the action was likewise stronger than without a tube, 

 although the increase was not so considerable as if the glass was 

 without this coating. 



The influence of the tube can, it is true, be diminished by 

 introducing diaphragms, which hinder the irradiation of the 

 inside ; but I have not succeeded in entirely obviating it, for the 

 edges of the diaphragm likewise reflect heat. But the signifi- 

 cance of the action of the sides of the tube in investigating the 

 diathermancy of gases is best seen from the following experi- 

 ments. 



In these experiments a strong gas-flame with a double draught, 

 surrounded by a glass cylinder, was used as a source of heat. It 

 was provided with a small parabolic metallic mirror, which re- 

 flected the rays of the lamp in such a manner that they passed 

 into a tube 1 metre long and 35 millims. in internal diameter, 

 at the other end of which was the above-named thermo-electric 

 pile. Between the tube and the lamp, and somewhat nearer the 

 latter, there was a screen consisting of two metal plates at a di- 

 stance of 1.2 millims. from each other. This could be removed 

 when the rays were to fall on the pile, and replaced as soon as 

 this was finished. Between this screen and the tube was a second 

 similar screen, which had an aperture 30 millims. square, the 

 centre of which was in a line with the axis of the tube. This 

 screen, which was always in a fixed position, protected the rays 

 of the lamp from the outside of the tube when the other was 

 removed. There was another screen with a similar aperture 

 close to the thermo-pile and between it and the tube, the object 

 of which was to protect the pile from all external rays. 



When using this arrangement, the rays were allowed to pass 

 through a tube open at both ends, and lined with rough black 

 paper; the deflection of the needle amounted to 24 0, 7, corre- 

 sponding to 32*2 units. On removing the tube the deflection 

 was only 10°, corresponding to 10. If, after removing the 

 tube, the rays were allowed to pass through two glass plates 4 

 millims. thick, placed at the same distance at which they would 

 be if they closed the tube, the deflection of the needle would 

 only be 1° to 2°. If, on the contrary, the blackened tube, as 

 the tube lined with black paper will for the future be called, was 

 between the glass plates, the deflection increased to 12 0, 6, corre- 

 sponding to I2'6. If the tube closed with glass plates was not 

 blackened on the inside, the deflection increased to 64°, corre- 

 sponding to 320. 



From this it will be seen how greatly investigations on the 

 passage of heat-rays will be affected by the nature of the tubes 

 in which the gases are experimented on. 



