18 REPORT— 1868. 
paper is applied. He says, “ The principles upon which this instrument is founded 
are— 
(1) “ That diffused light, on entering a tube at one end only, varies in intensity 
-within the tube inversely as the squares of the distances from the aperture where 
light enters. 
(2) “That any number of tubes, whatever their magnitudes, contain the same 
intensity of light if the ratios of their diameters to their lengths are equal, and if 
we absorb the light that may be reflected from their sides.” 
He then describes experiments which he made by means of tubes of equal 
diameter, and of lengths varying by a semidiameter and also with tubes of va- 
rious magnitudes, the ratios of their diameters to their lengths being equal. By 
inserting small mica-actinometers at the bases of those tubes, and exposing them 
thus situated simultaneously to the action of light, the writer gains results which 
demonstrate the above principles. 
Observations on the Atmospheric Lines of the Solar Spectrum in High Latitudes. 
By Georcr Guanstone, F.C.S., FRG. 
This paper was explanatory of some diagrams which the author had prepared 
of the atmospheric lines in the solar spectrum, from observations taken by him 
during a recent voyage along the north-west coast of Norway. The author stated 
that what are known by observers of the solar spectrum as the ‘ atmospheric lines” 
are certain dark lines or bands, which make their appearance under certain condi- 
tions, and sometimes even attain a considerable development. These lines, or 
bands, appear to be due to the presence of some substances in the earth’s atmo- 
sphere, as they are always most prominent when observing the sun through a long 
reach of air (as at sunrise or sunset), while they are scarcely visible when the sun 
is high above the horizon. The observations, of which drawings were exhibited, 
were taken in the months of June and July last, from the deck of the vessel when 
off the coast near Stavanger, and at the entrances to the Trondhjem and Namsen 
fjords ; the latter being in 64° 30’ north Jatitude, in which parallel the sun skirts 
the horizon for a long time, thus affording very favourable opportunities for ob- 
servation. It appears that in those regions the red end of aie spectrum is very 
brilliant, so that with the small portable spectroscope he distinctly recognized, on 
two occasions, the remarkable line A. The observations went to show that the 
atmospheric band grows in width and intensity as the sun approaches the hori- 
zon, and that what in certain states of light, or of the atmosphere, appear to be 
bands of shade are under other circumstances broken up into lines. Unaes some 
conditions the red rays suffer very little diminution of light up to a certain point, 
when they are suddenly cut off; while under others the obscuration takes place 
more gradually, and the visible spectrum is much longer. The length of the 
spectrum, however, in no case affects the width between the respective lines, which 
remains always the same, but is entirely due to more or less of the extremities 
being altogether lost in darkness. 
On the Value of the Hollow Wedge in examining Absorption Spectra. 
By Dr. J. H. Guapstone, F.R.S,. 
The usual way of examining absorption of light by a coloured liquid is to place 
it in a test-tube behind a narrow slit, and to disperse the line of light by means of 
a prism. The black or shaded bands due to the absorption may thus be easil 
noted ; but of course they represent only one particular thickness of the ligase 
Now the number of these bands often varies, and the extent of them always varies 
with the depth of the liquid traversed by the light, or, if it be a solution, with the 
quantity of colouring-matter dissolved.. A great advantage is obtained by substi- 
tuting a hollow glass wedge for the test-tube, and so arranging it before or behind 
the slit that the narrow line of light examined shall have traversed all thicknesses 
from, perhaps, two centimetres to nothing. Thus the varying absorption at dif- 
ferent depths is seen all at once, and can be easily represented in a diagram which 
becomes characteristic of the particular substance. The value of this mode ot 
