540 REPORT — 1903. 



surface of the glass slip is slightly attacked, but the intensity of this attack is not 

 to be compared with that which occurs in the near neighbourhood of the metal. 

 The chemical action to which the formation of the halo is due is, therefore, directly 

 intiuenced by the presence of the hot metal. 



It was at first thought that the metal foil might be acting simply by arresting 

 radiant heat in its passage through the transparent glass, and thereby localising 

 its effects ; but this view is not supported by the subsequent experiments. 



A number of experiments were made before it was sufficiently realised that a 

 fairly free circulation of the atmosphere surrounding the metal is necessary if 

 well-marked halos are to be produced. It was also found that the partial 

 exclusion of the products of combustion from the air bath adversely affected the 

 formation of halos. 



When the glass slip with the metal foil is covered by another slip which is in 

 contact with the metal, the halo is reduced to a sharp narrow outline of the foil. 



If the cover-slip is supported just out of contact with the metal, the halo is 

 still a sharp outline, but now an image of the foil, less sharp, but also in outline, 

 is formed on the under surface of the cover-slip. As the distance between the 

 cover-slip and the foil is increased the halo widens and the image on the cover- 

 slip becomes a smooth patch without sharp outlines. These effects are obviously 

 intiuenced by the more or less ample supply of the active constituents of the 

 atmosphere which results from the greater or less freedom of circulation. This, in 

 turn, is affected by the nearness of the cover-slip to the metal. 



A piece of platinum foil, 7 millimetres square, was used in a series of experi- 

 ments in which the distances of the upper slip from the metal were 0'2 mm., 

 1'5 mm., 3 mm., and 7 mm. ; an experiment was also made in which no cover- 

 slip was used. The temperature employed was about 500°, and the time of 

 heating 30 minutes. At 0'2 mm. distance halo ^ and image were both in outline. 

 At 1'5 mm. the halo was a band round the foil about 1 mm. wide; the image 

 showed no outline, but was in size and form similar to the foil. At 3 mm. the 

 halo had widened to over 2 mm., part of this being got by encroaching on the 

 area covered by the foil ; the image was now circular, its diameter being equal to 

 the diagonal of the square of the foil. At 7 mm. the halo had widened to 3 to 

 4 mm. on three sides, and to 8 mm. on the fourth side ; its general form was oval. 

 The image was of the same form, and only a little smaller. It was evident in 

 this case that the stream of emanations from the platinum had drifted across the 

 slip under the influence of a current in the atmosphere of the air-bath. This 

 evidence of drifting of the stream of emanations at once disposes of the idea that 

 the formation of the halo and image is due in any way to the radiation or 

 reflection of heat by the metal foil. The forms and dimensions of the halos and 

 images strongly suggest that the emanations from the platinum are thrown 

 upwards like a fountain, which spreads, and descends on and around the foil. 

 When the cover-slip is at the maximum distance, it is struck by the apex of the 

 stream, and a large but faintly defined image is produced. When the distance of 

 the cover-slip is small, the stream is intercepted before it has spread much and the 

 halo and image are small and well defined. 



The decomposition of glass in the neighbourhood of hot metals appears to be a 

 case of accelerated or intensified chemical action, induced by the energy of the 

 particles shot out from the hot metal. It occurred to me that some of the cases 

 of catalytic action by platinum and other substances might be accounted for by 

 the existence of active emanations surrounding the catalyte, and not merely by 

 the actual contact of the molecules of the re-agents with its surface. Experiments 

 are now in progress to test this question. 



' Throughout this Paper the term ' halo ' is applied to the efifect produced on the 

 lower slip, on which the metal lay, and ' image ' to that on the under surface of the 

 cover-slip. 



'■'■r}'*'yr'*mib 



•lltlli'. 8 3o 



