SECTIONAL TRANSACTIONS.—Aft; A, G. 283 
divided into high and low luminosity groups (giants and dwarfs) by the 
interstellar line criterion. 
It is found that the intensity of both the hydrogen and helium lines is, in 
each subtype, greater for the dwarfs than for the giants, whereas the ionised 
lines of carbon, nitrogen, oxygen, magnesium, and silicon are relatively 
stronger in the giants. ‘The effect for hydrogen is so marked, from type 
BO to A, that it could be used for absolute magnitude determination provided 
the exact subtype of the star was measurable. This problem of exact 
classification is complicated by the fact that no line studied is free from 
luminosity effect. Ratios of line intensity for atoms of widely different 
excitation potential present the most hopeful solution of the problem, but 
any adopted method, though based on photometric measures, should be 
capable of adaptation for estimates of type and luminosity by the usual 
visual inspection. 
REPORT OF COMMITTEE ON SEISMOLOGICAL INVESTIGATIONS. . 
Miss E. F. BeLttamy. 
JOINT SESSIONS (SECTIONS A, G) ON TECHNICAL PHYSICS. 
Thursday, September 6. 
(Dr. Ezer GrirFiTus, F.R.S., in the chair.) 
Mr. R. S. WuippLe.—A note on some of the difficulties of measuring the 
temperature of molten steel (10.0). 
The problem of measuring the temperature of molten steel either in the 
furnace or in the ladle is one of great importance to the steel manufacturer. 
It is, however, one of great difficulty. 
The committee appointed by the Iron and Steel Institute to study the 
heterogeneity of steel ingots formed a sub-committee to study the tempera- 
ture measurement side of the problem of ingot casting. This committee 
has devoted a great deal of time to the study of the problem. 
The temperature of the steel in a Siemens furnace is approximately 
1630° C., and the difficulty of inserting a pyrometer into the molten metal 
through the open door is almost insuperable. ‘The tube protecting the 
thermo-elements must also be robust and non-porous, as the only elements 
that may be safely used at the present time for temperatures as high as 
1600° C. belong to the platinum group. ‘The committee decided that at the 
present time the problem could only be solved by the use of optical pyro- 
meters, and that those of the disappearing filament type gave the most 
consistent results. Discrepancies in the results obtained showed that a 
careful study of the details of the pyrometers was necessary, and as a result 
exhaustive tests were made on the coloured and neutral glasses of the 
instruments. With the introduction of new glasses and other modifications 
a considerable improvement has been made in the performance of these 
pyrometers. The readings obtained give the apparent temperature of the 
steel; a correction must be applied to convert the readings to true 
temperatures. 
The position cannot be regarded as satisfactory because there are so 
many factors involved in the determination of the temperature of molten 
steel by means of optical pyrometers, and this necessitates a considerable 
amount of skill when making the observations. 
