TRANSACTIONS OF SECTION A. 393 
assembled three points became at once evident—namely, (1) the displacements 
generally indicated an outflow of the solar vapours and were intimately related 
to the intensity of the lines; (2) the very strong lines were displaced in the 
opposite direction, and then showed a dependence upon intensity the reverse of 
that shown by the lines of moderate and low solar intensity; (3) the displace- 
ments were distinctly greater for lines in the red than for lines in the violet, 
pointing to a close relation between the displacements and wave-length. In fact, 
the differences nearly all disappeared when the displacements were reduced to a 
common wave-length and the comparison was made between lines of equal 
intensity. The residuals disappeared when allowance was made for the greater 
depths to which one can see into the sun by light of long wave-length, because 
of the lessened scattering of the long waves. This proportionality between dis- 
placement and wave-length points to the Doppler effect, and tends to strengthen 
greatly the explanation suggested by Evershed. 
When the iron lines from intensity 00 to 10 are arranged in the order of 
intensity the displacements form a remarkably systematic series, which I have 
called the Zron scale. The variation of displacement with intensity is plainly 
evident, even without reduction to a common wave-length, and was noticed 
before such reduction, and even when only a limited region of the spectrum had 
been examined, but it shows with great regularity where the displacements are 
reduced to a common wave-length. 
The Iron Scale a 5000. 
Intensity. . | 00 Gristle aatean bat 5 | 6 ped 
| -023 | 0217) 019 | -016 012 | -009 004 
| 
| 
| 
| 
| 
Displacement | -034 | -030 | -028 | -025 
Vel. km. /sec. | 2-04 1-80 (1-68 |1:50 (1-38 | 1-26 | 1-14 |0-96 0-72 | 0:54 | 0-24 | 
This scale is based upon nearly 200 iron lines, but a simiiar gradation appears 
in the case of any element represented by several lines. 
In the case of strong lines the march of the phenomenon is equally striking 
and suggestive :— : 
Strong Lines A 5000. 
Element... Si | Al | Na Mg) Hy | Ha 
| 
| 
| 
| 
Displacement . .  -000 | -000 |—-006 —-012'—-033 —-050 —-063 
Vel. km/sec. . . 0:00 0-00 |—0-36 —0-72/—1-98 —3-00 —3-78 
Intensity. . 12 15-20 | 20-30 20-30, 20 | 40 
The negative sign indicates that the displacements are such as to imply an 
inflow of these high-level vapours into the spot, and if the assumption be made 
that the lines of low intensity have their origin at relatively great depths in the 
solar atmosphere, and that, consequently, the displacements are a function of the 
depth, we can get the distribution in a vertical section of a spot. The maximum 
velocity inward occurs at the highest level, and decreases as the depth increases 
until the level of zero velocity is reached; the motion then becomes an outward 
one, and the velocity increases with depth. These movements of the vapours 
do not form a complete system, as a high velocity inward occurs where the 
density is low, and a high velocity outward where the density is relatively great 
and the masses involved are unequal and the kinds of matter are unlike. The 
type of vortex indicated is that of the terrestrial tornado. There is a whirling 
upward rush from the interior of the sun, which spreads out radially with 
rapidly decreasing velocity tangential to the solar surface, and entrains with it 
the gases of the reversing layer. The actual vortex is deep-seated, the outflow 
into the reversing layer being a portion of the upper part of it, the inflow from 
the chromosphere being a secondary effect, a superficial indication of the under- 
lying vortex in which the magnetic field originates. 
For any given element the displacements decrease with increasing line- 
intensity, but for lines of equal intensity the absolute displacements differ from 
