CELESTIAL SPECTROSCOPY. rail 
committee of this Association from 1885 to 1887. Last year the com- 
mittee added a table of corrections to Rowland’s scale. 
The inconvenience caused by a change of standard scale is, for a 
time at least, considerable; but there is little doubt that in the near 
future Rowland’s photographie map of the solar spectrum and his 
scale based on the determinations of absolute wave length by Pierce 
and Bell, or the Potsdam scale, based on original determinations by 
Miiller and Kempf, which differs very slightly from it, will come to be 
excltisively adopted. 
The great accuracy of Rowland’s photographic map is due chiefly to 
the introduction by him of concave gratings and of a method for their 
use by which the problem of the determination of relative wave lengths 
is simplified to measures of coincidences of the lines in different spectra 
by a micrometer. 
The concave grating and its peculiar mounting, in which no lenses 
or telescope are needed, and in which all the spectra are in focus 
together, formed a new departure of great importance in the measure: 
ment of spectral lines. The valuable method of photographie sensi- 
tizers for different parts of the spectrum has enabled Prof. Rowland to 
include in his map the whole visible solar spectrum, as well as the 
ultra-violet portion as far as it can get through our atmosphere. Some 
recent photographs of the solar spectrum, which include A, by Mr. 
George Higgs, are of great technical beauty. 
During the past year the results of three independent researches 
have appeared, in which the special object of the observers has been 
to distinguish the lines which are due to our atmosphere from those 
which are truly solar—the maps of M. Thollon, which, owing to his 
lamented death just before their final completion, have assumed the 
character of a memorial of him; maps by Dr. Becker; and sets of pho- 
tographs of a high and a low sun by Mr. McClean. 
At the meeting of this association in Bath, M. Janssen gave an ac- 
count of his own researches on the terrestrial lines of the solar spec: 
trum which owe their origin to the oxygen of our atmosphere. He 
discovered the remarkable fact that, while one class of bands varies as 
the density of the gas, other diffuse bands vary as the square of the 
density. These observations are in accordance with the work of Ego- 
roft and of Olszewski, and of Liveing and Dewar on condensed oxy- 
gen. In some recent experiments Olszewski, with a layer of liquid 
oxygen 30 millimeters thick, saw, as well as four other bands, the band 
coincident with Fraunhofer’s A; a remarkable instance of the persist- 
ence of absorption through a great range of temperature. The light 
which passed through the liquid oxygen had a light blue color resem- 
bling that of the sky. : 
Of not less interest are the experiments of Knut Angstrém, which 
show that the carbonic acid and aqueous vapour of the atmosphere 
reveal their presence by dark bands in the invisible infra-red region, 
at the positions of bands of emission of these substances. 
