Band and Line Spectra of the same Metallic Elements. 133 
Fourth.—Channelled or fluted band-spectra, such as those of tin, lead, 
antimony, and bismuth, extending through a wide range if not throughout the 
whole of the flame-spectrum. These appear to be quite independent of lines 
and unconnected with them. They are beyond all question the spectra of the 
elements. The banded spectrum of metallic manganese may also be classed 
with them. 
Fifth.—Bands composed of lines, spaced with regularity, and grouped in a 
different manner, according to the nature of the element. The spectra of each 
group of metals in the periodic system are similarly constituted, as, for instance, 
those of zine and cadmium, and those of copper, silver, and gold. There is an 
intimate connexion between these bands and the are and spark lines of the same 
elements. The band-spectra are the spectra of the molecules of the elements, 
and the molecules have been well ascertained to consist of only one atom. 
Sixth.—There are spectra composed chiefly of limes, whether the flame or the 
spark be employed, but which show diffuse bands of continuous rays, in some 
cases ill-defined, as, for instance, iron. Thallium, palladium, and iridium are 
examples* which may be classed with iron, though the two last have very definite 
composite bands. 
In the spectra of the calcium group, the characteristic bands from which lines 
are developed by physical and chemical means are apparently caused by the act 
of oxidation of metallic vapour in the flame which had already been reduced 
from the oxide. The same may be said of the diffuse bands in the green which 
are observed in the spectrum of copper. 
It is necessary to mention here some expeniinentis on calcium which afford 
evidence in support of this view. When the spark without condenser is taken 
from metallic calcium in air and in hydrogen, and from a solution of calcium 
nitrate under the same conditions, the spectra are practically identical with the 
flame-spectrum of metallic calcium, with the oxyhydrogen flame-spectrum of 
calcium carbonate, and with calcium oxide in the Mecke burner. The red and 
green bands have lines in them which appear only as intense narrow radiations, 
but without differentiation of lines; they are placed rather towards the more 
refrangible side of the red band, and are situated in the midst of the green band, 
but they are neither well-defined nor sharp; also, the sodium lines, as seen in 
the same spectra, are rather diffuse; but when the air in the tube is changed for 
hydrogen, whether the spark be taken from the metal or from the salt, three 
beautifully distinct fine lines are observed in the red band, and two in the 
green band. Simultaneously the yellow sodium lines become not only very 
* Hartley and Ramage: ‘‘ Banded Flame-Spectra of Metals.’’—Scient. Trans. Roy. Dublin Soc. 
“TRANS. ROY. DUB. SOC., VOL, IX., PART VII. nya 
