408 
ME. WAEEEN DE LA EIJE ON THE 
length entirely covered by the moon. On the other hand, a prominence situated at 
right angles to the moon’s path would change its position-angle in respect of the moon’s 
centre, but would remain uncovered to almost the same extent at the end of totality 
as at the commencement. Moreover, prominences situated in positions intermediate 
between these two would change their respective position-angles, referred to the moon’s 
centre, less, but would be covered more, in the proportion of then- relative degrees of 
proximity to the line of the moon’s path ; and vice versd^ the change in position-angle of 
the several prominences would be greater in proportion as they were situated nearer to 
a line at right angles to the moon’s path*. No stronger evidence that the prominences 
belong to the sun can be adduced, than that of a change in the angular position of a 
prominence in reference to the moon’s centre, because it is not probable that different 
parts of the moon’s periphery would produce precisely the same effect on light emanating 
from the sun. The relative changes of position may be calculated for any given locality ; 
for which purpose it is necessary to know either its geographical position, or to deter- 
mine the relative motions of the sun and moon during an eclipse by other means. In 
a former part of this paper I have stated that the geographical position was ascertained, 
also that the exact path of the moon’s centre across the solar disk was made out by 
certain measm'ements of the photographs ; and in Plate XVIII. I have given a graphic 
representation of the moon’s path in reference to the sun’s centre during the eclipse. 
Mr. Caeeii^gtojv and Mr. Faeley’s elements of the eclipse, founded on the geographical 
position, have already been cited, and will be presently made use of for computmg the 
changes of position of the several luminous prominences which should occur if they 
belong to the sun. 
If the prominences belong to the sun, photographic images of the same protuberances 
taken at any one locality, at different epochs of totality, ought to coincide exactly when 
the photographs are superposed one over the other f; and measurements of their positions 
with respect to the moon’s centre ought to correspond with their computed positions. 
Moreover, photographs taken at different places (sufficiently near in longitude) ought to 
agree in their details, — although, for very distant locahties, it is possible that they might 
not do so ; for it is conceivable that during long intervals a change might occiu* in the 
luminous prominences, or fresh prominences might be brought into view by the smi’s 
rotation. 
Furthermore, on consideration it will be evident that to two observers differently 
situated the protuberances would not have precisely the same dimensions ; that is, they 
would appear to project more or less beyond the moon’s limb. Within the zone ot 
^ An inspection of Plate XY. will render this evident, and it wdl be seen that the ^gle 7 gradually 
diminishes as the prominence is more distant from A. 
t That this was the case was shown, on the occasion of this paper being read, by sliding the first totality- 
picture over the second, and projecting their images on a screen by means of an electric lamp. It was thus 
seen that the pictures of the several prominences correspond exactly in form and position, each to each, in 
the two photographs. 
