[PLASKETT-DELURY] THE SOLAR ROTATION 51 
16. The comparison of plates at the equator shows a systematic 
difference for measures of the same plates of 0.046 km. per second. 
When the 5 complete plates of Series I and II are compared it is found 
that in these plates the average difference at the equator is smaller 
about 0.027, and that this remains unchanged practically for all latitudes 
except the pole. This shows that the difference is evidently not due to 
any effect of the magnitude of the displacement of the lines of one strip 
with respect to the other, else it should vary with the latter which 
changes from 0.1 mm. at equator to about 0.017 mm. at 75°. It may 
be said therefore that Plaskett measures the displacements from 0.03 
to 0.05 km. per second higher than De Lury in the region at À 5600. 
The peculiar nature of the difference P — D at the pole should not pass 
. without comment. The mean value of this difference is+0-001. Although 
it is of the same sign as the other differences in the Series II plates it is 
of the opposite sign in Series I, and is hence not systematic as at the 
other latitudes, and it might therefore be regarded as evidence that 
the magnitude or sense of the displacement influences the measures 
of one or both of the observers. Owing to the method of measurement 
used by De Lury he would seem to be less likely to be influenced in this 
way. When we compare the measures in the 44250 region we find 
that the difference found in the 4 5600 region nearly vanishes, being only 
0.007 km., scarcely large enough considering the few plates measured 
by De Lury to be deemed systematic. The spectra in the 24250 
region are much more easily measurable than at À 5600. Not only is 
the grain of the plate finer, but the lines themselves are much more 
uniform in character and better defined. Consequently it seems likely 
that the large difference between the two measures in the À 5600 region 
depends in some way upon the character of the lines for measurement. 
Although the probable error of measurement of a single line, given for 
plates 813 and 820 above, for Plaskett is only about a third of that for 
De Lury, + 0.019 and + 0.052 km. per sec., and hence the former’s 
measures should be considered of greater weight, yet that does not 
settle the question of the correct value of the velocity. Possibly some 
information may be obtained from the Mt. Wilson measures. 
17. Mr. Adams and Miss Lasby have had greater experience than 
anyone else in the measurement of photographic rotation spectra, and 
their measurements should be given great weight. Yet when we come 
to make comparisons, Table VIII, plates 813 and 820 we find practically 
the same difficulties and the same differences between them as between 
the writers. For example, in plate 820 at the equator we have Miss 
Lasby’s value 1.851, Mr. Adams 1,798, Plaskett’s 1,799 and 1,784, De 
Lury’s 1,757 and 1,744. Indeed in several cases Miss Lasby’s value 
is as much higher than Plaskett’s, as his is than De Lury’s. On the 
