of Gases from Meteoric Iron. 297 
A gentle heat was now applied to the tube containing the 
iron, by means of a Bunsen burner. This brought the gauge 
in a few minutes to about 6 mm., and produced a marked 
change in the appearance of the vacuum-tube, which before had 
the appearance of an ordinary hydrogen tube. The light in 
the broad portion became a straight, hazy stream of a dull 
greenish-white color, very similar to that given by a tube con: 
taining either of the oxides of carbon. After the tube had 
been exhausted to 2 mm., heat was again applied rather more 
standing at5 mm. Thestop-cock being closed, the exhaustion 
was continued to 15mm. At this point the spectrum was 
nearly the same as before, but was somewhat less brilliant. 
Certain other lines appeared in the spectrum, of which mention 
will be made in a later paragraph. 
A second set of experiments was made with a specimen of 
meteoric iron from Tazewell County, Tennessee. e meteor- 
ite, of which this was a portion, weighed when found twenty- 
five kilograms, and has the composition, Fe, 83°02; Ni, 14°62 ; 
other substances, 1°98. No carbon was found in it.* The iron 
Cially noticeable, as there is no record of them in the notes of 
the experiment. 3 : 
* Prof. J. L. Smith, this Journal, II, xix, 153. 
