SECTIONAL TRANSACTIONS.—A.T 355 
attention to the importance of ionised or enhanced lines. About ten years ago, 
Saha developed his remarkable theory of ionisation by means of which many solar 
mysteries have been solved. 
Helium was discovered by the spectroscope at the eclipse of 1868, and the eclipse 
of the following year brought coronium to our ken. After more than sixty years of 
observation, how much do we know of this mysterious gas? The unfortunate part 
has been that in all this long time we have had only about one hour’s observation, or 
an average of one minute per year to accumulate the facts. During totality, when 
alone the spectrum may be obtained, a total of about fifty lines have thrust them- 
selves into the literature. Davidson and Stratton have done eclipse astronomy a 
fine service by showing that most of the lines photographed during totality belong 
to the high chromosphere and only about fifteen to coronium. At the 1930 eclipse 
I had the good fortune to discover a new line of considerable strength, at wave-length 
6776 in the deep red. 
The combination of large dispersion and a large image of the sun obtained from 
the concave gratings permitted, first of all, an accurate determination of wave-lengths. 
Measures from eight spectra of the green line, from both limbs of the sun, give a 
wave-length of 5302-91, while six spectra give 6374°28 for the line discovered at the 
1914 eclipse, this latter wave-length differing only 0-01 angstroms from a line in the 
oxygen I spectrum measured by Hopfield. The large images show that the dis- 
tribution of the light around the sun differs very much in these two lines. Further- 
more, the detailed structure in the coronium rings is radically different from that of 
the H or K or hydrogen alpha lines of the chromosphere, and also differ from the 
details shown in direct photographs of the sun from integrated light. In other words, 
the solar activity cause the coronium lines to appear near where prominences are 
visible, but the differences in structural detail make us believe that the green coronium 
line is not associated with the calcium atom which produces H and K nor with the 
hydrogen atom. The brightest coronium rings show absorption close to the edge 
of the sun, which seems to prove that coronium lines are not ‘ forbidden ’ lines. 
The discussion of the 1930 flash spectrum obtained in good focus from wave- 
length 3230 to 7800 will have to await a future occasion. 
Mr. R. StonELEY.—Deep Focus Earthquakes. 
It has been pointed out by Dr. Harold Jeffreys that, in accordance with a general 
reciprocal theorem in dynamics, the surface waves of deep-focus earthquakes should 
be relatively small in amplitude. 
The presence of reading#for Z and M in earthquakes whose foci are given in the 
International Seismological Summary as 0-07, 0-08, and in one case 0-09 of the earth’s 
radius below normal, seems prima facie to constitute a great difficulty in accepting 
the great focal depths found by Prof. H. H. Turner. <® 
It turns out on examination of the times of transit of the listed Z and M that the 
times correspond to the arrival of S, SS, SSS, and sometimes to G, Gutenberg’s ‘ Early 
Long Wave’ (4:35 km./sec.). The examination of actual records of earthquakes of 
very deep focus at once confirms this: nearly all the energy is transported in the 
body waves, so that P, PP, PPP, S, SS, SSS, etc., are of enormous amplitude, whereas 
the surface waves are inconspicuous. 
The reciprocal theorem argument accordingly confirms Turner’s discovery that 
occasionally earthquakes occur with foci considerably below the normal level. 
Dz. R. v. v. R. Woontey.—Intensities of Lines in Solar and Stellar 
Spectra. 
Information concerning the physical state of the outer layers of stars can be 
_ obtained from the investigation of absorption lines in stellar spectra. Temperatures 
and pressures can be deduced from the numbers of atoms in various states of excitation 
and ionisation, by the well-known formule of Saha and of Milne. The intensity of an 
absorption line is a measure of the number of atoms in the lower state concerned in 
its formation, but there are certain difficulties in the elementary theory of Stewart 
and of Unséld, which give the result that the relative number of atoms is proportional 
to the square of the width of the line, measured at any point (Unséld’s formula). 
On the experimental side several investigators have found that the intensities of 
multiplets in the solar spectrum do not agree with theoretical intensities which have 
AA2 
