[PLASKETT-DELURY] DETERMINATION OF THE SOLAR ROTATION 115 
1. The actual velocity of rotation at the different latitudes with 
the law connecting the change of angular velocity with the latitude. 
2. Whether the rotational velocity of the sun is constant. 
3. Whether the velocity obtained from different elements, say those 
at higher or lower levels in the solar atmosphere, is constant. 
For the adequate discussion of the first two problems a long ex- 
tended series of investigations is necessary, and they naturally can not 
be touched upon here. The third question also requires much more 
observational material than we now have, before any definite statement 
can be made. Adams’ results show that there are considerable differ- 
ences in the values given by different elements. The lines of carbon 
and lanthanum, presumably at a lower level, give lower rotational 
velocities than the general reversing layer, one line of manganese gives 
a higher value, and the blue line of calcium À 4227 and H, give much 
higher values. Schlesinger, on the contrary, in a preliminary paper 
given at the last meeting of the Astronomical Astrophysical Society and 
of America, states that any differences he obtained in the reversing layer 
were not greater than and not distinguishable from the accidental errors 
of measurement. 
It seemed desirable to us therefore to test this matter, and acting 
on the supposition that the small differences obtained by Adams might 
be due to subjective errors in measurement caused by the character 
of the lines themselves and not to an actual shift due to different velo- 
cities, a scheme was devised for testing this supposition. A special 
slit was made with an attachment in the centre of the jaws, which dis- 
placed transversely to any desired distance a strip about a millimetre 
long. It is evident that, by placing a suitable diaphragm over this 
attachment to limit the widths of the spectrum obtained, one can 
produce three strips of spectra of the same width and separated by the 
same distance as those made by the reflecting prism arrangement, and, 
moreover, the lines of the centre strip may be displaced any desired 
distance to the red or violet of the lines of the outside strips, giving the 
same effect and appearance as rotation spectra, except that the dis- 
placement of the lines is arbitrary and exactly the same for each. A 
series of plates were made with this broken slit at the region L 4200- 
4300, the same region as used by Adams and chosen as the general 
region for universal observation by the Rotation Committee of the 
Solar Union. These spectra were made by setting the sun on the slit 
attachment at the same distance within the limb as the rotation spectra 
are obtained. We will then have evidently spectra exactly identical 
in appearance with rotation spectra, measurable in exactlythe same way, 
but with a displacement that we know must be the same for each line. 
Twelve of these spectra were measured by each of us, the lines measured 
