382 
ME. WAEEEN DE LA ETJE ON THE 
In column 10 are given the distances actually measured between the peripheries of 
the sun and moon in a direction at right angles to a line joining the cusps; these 
numbers differ in a few cases from those which were obtained for the same photographs 
by deducting the numbers in column 4 from those in column 3, Table II., but it has 
not been thought necessary to alter them in Table I. The cases in which discrepancies 
occur are the following : — 
inch. 
No. 8. 
No. 17. 
No. 19. 
No. 22. 
in which the peripheral distances in Table II. 
differ from those in Table I. by 
5 ? 9 ? 99 
99 99 99 
99 99 99 
-0-003 
+ 0-001 
- 0-001 
+ 0-003 
No. 29. „ „ „ -0-001 
The numbers for these particular photographs in Table II. result from a series of 
second measurements of several of the photographic plates, which it was found neces- 
sary to make again for Table II., in consequence of some minute errors in reading the 
vernier 5' in the first series. 
Column 11, Table I., gives the numbers of column 10 reduced to the adopted mean 
solar radius, namely, 1904-91 thousandths of an inch. 
Column 12 gives the differences of the peripheral distances for two consecutive pho- 
tographs, and, neglecting the augmentation of the moon’s semidiameter, the approxi- 
mate approach or retreat of the centres in the interval between their epochs. By diriding 
these numbers by those in column 6, were obtained the numbers in column 13, which 
are very nearly the rates of approach or retreat of the sun and moon’s centres per 
minute. 
Column 14 gives the approach and retreat per minute for the longer periods brack- 
eted, and column 15 the same numbers reduced to seconds of arc. These rates of 
approach and retreat of the centres are affected by any errors in registering the time of 
the photographs, and also by all errors of measurement. The numbers do not run quite 
smoothly, and yet perhaps they are as good as could be expected. No account was 
taken of the augmentation of the moon’s semidiameter, except for the middle of the 
eclipse ; but the rates of approach or retreat for the beginning and end, even without 
this correction, admit of a comparison with the computed numbers, as the change of 
semidiameter was not great during the intervals. 
Approach of sun and 
moon’s centres per minute 
at the commencement. 
Eelative motion of 
centres per minute 
at the middle. 
Eetreat of sun and 
moon’s centres per 
minute at the end. 
Measured . . . 
25-26 
27-84 
30-05 
Caeeington . . 
24-87 
27-27 
29-85 
Faeley .... 
25-14 
27-40 
29-61 
