by the electrical discharge in exhausted tubes. 53 



Interesting results were obtained with iron, which gave a 

 very beautiful and perfect film, having a brilliant: luster, aad ■ 

 high degree of transparency. The light transmitted by it is 

 very nearly of a pure neutral tint, though with a faint tinge of 

 brownish. The electrode was a wire 03 millimeter in diameter. 

 This was carefully cleaned, and the tube filled with dry hy- 

 drogen three times before the final exhaustion. Without this 

 precaution it is impossible to obtain reliable results, as the 

 metal is partialis oxidized, and the film not only stained, but 

 its color by transmitted light changed to brown, or even to 

 deep orange-yellow. This was the color exhibited by the film 

 obtained in the first trial, in which the tube was filled with air. 

 Externally it exhibited a vivid iridescence, in various colors, 

 but at the extremities of the deposit the metal had its proper 

 luster. Subsequent trials in which the tube was once filled 

 with hydrogen gave better results, but the yellow tint and the 

 iridescence did not disappear until the thorough removal of the 

 oxygen and moisture, as described. 



For those metals which are not readily obtained in the form 

 of thin wires, a modification of the process was employed, as 

 follows: a tube open upon one side was made by softening the 

 end of a glass tube in a gas flame and drawing out one edge 

 with another piece of glass. A trough was thus formed, at 

 first shallow but growing deeper and deeper as the whole edge 

 began to be drawn upon, and by properly adjusting the heat 

 it could be made as fine as desired. A tube of this 

 kind from half a millimeter to a millimeter in diameter with a 

 platinum wire inserted in the closed end, and with the open 

 end evenly and firmly packed with filings or powder of the 

 substance to be examined formed the electrode, which was 

 used in the ordinary way. 



Nickel and cobalt we're experimented upon by this method, 

 and films obtained, but as the removal of the oxygen was not 

 complete they were less perfect than those produced 

 iron, the other metals, though sufficient to show that the color 

 by transmitted light is gray or brownish gray. Tellurium 

 employed in the same way afforded a very brilliant layer, which 

 gave to the light seen through it a dull purple color. 



At the suggestion of Prof. Dana an experiment was made 

 with macnetie oxide of iron, native magnetite in pow 

 employed with one of the small troughs. It was volatilized 

 with difficulty, but apparently without decomposition, and 

 formed a somewhat lustrous film, appearing gray-brown by 

 transmitted light The result is of interest with reference to 

 the occurrence of thin, somewhat transparent layers of this 

 substance ii here it has the same color as that 



a this experiment 



