ARC SPECTRUM OF ELECTROLYTIC IROX. 
993 
re-examined; and, as the wave-lengths have been read directly from Rowland’s 
map, there is little chance of error. 
Messrs. Kaysee, and Runge have not attempted, in their first paper on the iron 
spectrum, to distinguish the lines due to impurities. I have, therefore, endeavoured 
to trtrce the origins of the lines which appear in their list but not in mine. Many of 
these additional lines may possibly be accounted for by impurities, but the majority 
are not represented at all on the new Kensington maps. As already pointed out, 
they may most probably be ascribed to iron, the lines not having appeared on the 
Kensington photographs perhaps on account of insufficient exposure, or possibly by 
the employment of a different temperature. 
Comparison ivith McClean’s Photographs. 
A comparison of the lines in McClean’s photographs of the spark spectrum of iron 
and those in the Kensington arc photographs shows a great similarity between the 
two, but still there are some differences. Although some of the lines not common to 
the two sets evidently have their origin in various impurities, others appear to be 
really due to iron. Most of the lines photographed by McClean in the spark spec¬ 
trum, but which are absent from the arc spectrum, have been found to be due to 
impurities ; whilst, in general, those present in the Kensington photographs and not 
in McClean’s have been confirmed by Thalen or Messrs. Kayser and Runge as 
having their real origin in iron. Below are appended two lists, in one of which are 
given the iron lines which appear in the arc and not in McClean’s photographs ; and 
in the other those which are found in his photographs, but are absent from the arc. 
All the lines in the latter list, howerer, have been recorded in the arc spectrum, 
either by Kayser and Runge, or Tiialen. 
G L 
MDCCCXCIV.—A 
