SECTIONAL TRANSACTIONS.—A. 409 
4, Prof. J. C. Kapreyn.—First Attempt at a Theory of the Structure 
and Motion of the Stellar System. 
According to former results, slightly modified, the equidensity surfaces of the 
stellar system for that part (domain A), where the density exceeds one-hundredth 
part of the density near the sun, must be with some approximation concentric 
similar rotation ellipsoids, similarly situated. As a highly probable consequence 
the attraction of the whole of the system on a point within the domain A can 
be computed as soon as we know the average mass of a star. 
Reasons were shown for assuming the whole of the system to be in a rotatory 
_ motion round the line from the centre towards the pole of the Milky Way as 
an axis, and for further assuming that to the stars along this axis we may 
apply the barometric formula for determination of heights in our atmosphere. 
The application leads to a determination of the average mass of a star, with 
the fUllowing results : 
Distance from centre Average mass 
198 parsecs. 2-2 x mass of sun. 
413, 2°0 9 
MS “Zs 17 ” 
1114. —Cy, 15 4g 
1660 ,, 1-4 2 
which agree surprisingly well with what was derived from binary stars, the 
change with distance being in the sense required by theory. 
‘he same values of the masses are obtained for the stars in the plane of 
the Milky Way if to the gravitational forces we add definite centrifugal forces. 
The linear velocities implied by these centrifugal forces are : 
Distance from centre Linear Velocity 
1010 parsecs. 13-0 kilometres per sec. 
2106, 19°5 55 
BODia” 20:1 . 
5675, 19-4 5 
8465, 18°6 + 
Assuming that part of the stars rotate one way and the rest in the opposite 
direction, there are two groups of stars which in the overwhelmingly greater 
part of the system have a relative motion of approximately 2x 19°5=39 kilo- 
metres per second. ‘his velocity agrees almost perfectly with the relative 
velocity of the two star streams, so that we get an explanation of these streams 
by simply assuming that the sun is not at the centre but in this greater part 
of the system, which seems strongly probable a priori. 
That no curvature was found for the streaming must be due to the fact that 
the extent of the domain for which star streaming was found is so small that 
the curvature must escape notice. 
5. Dr. Cricuton Mitrcuety.—The Geophysical Observatory in the 
Shetlands. 
(a) DepARTMENT OF Puysics aND MATHeEMATICcs. 
6. Prof. E. T. Wauirraker, F.R.S.—Tubes of Force in Four- 
dimensional Physics. 
7. Mr. A. A. Campspetn Swinton, F.R.S.—The Reception of Wireless 
Waves on a Shielded Frame Aerial. 
_ The paper described experiments in receiving the spark emission from the 
Eiffel Tower on a small frame aerial placed inside a tube of wire network with 
open ends. It was hoped to obtain improved directional properties, but, though 
the presence of the tube weakened the signals, it was found that altering its 
direction did not affect them, nor were the signals further weakened by closing 
the open ends of the tube by wire grids. 
The tube was next replaced by a sheet copper box with one open end. In this 
was placed, not only the frame, but also the amplifier and all the other 
apparatus, the telephone being listened to through a rubber pipe. Signals were 
1921 FF 
