( 1269 ) 
of the “irradiation-curve” for a point M (fig. 4) situated in 
the outer layer of the solar 
atmosphere. For that purpose 
we only have to take on each 
line MV, lying within the 
angle H’MH and cutting the 
photosphere at an angle p with 
the normal, a distance equal to 
that polar co-ordinate of the 
curve P’Q’ (fig. 3), which cor- Fig. 4. 
responds to the same value of . Joining the end-points of the 
vectors thus defined, we obtain the required irradiation curve qpq’. 
It differs only slightly in shape from ?’Q’, and, as is easily seen, 
would preserve very nearly the same character if M/ were chosen 
in a lower level *). 
Now looking for an explanation of the general decrease of bright- 
ness toward the limb, it is only natural to make scattering respon- 
sible for the phenomenon (both molecular and refractional scattering). 
In fact, all kinds of light are more or less liable to scattering, whereas, 
very probably, absorption only extends to waves having the same, 
or almost the same frequency as the free vibrations of the electrons 
in the solar gases *). 
The widening of the Fraunhofer lines toward the limb proves 
that the gradual decrease of intensity from centre to limb is greater 
for kinds of light suffering anomalous dispersion, than for the less 
refrangible light from blank spaces of the spectrum. Accordingly, if 
we construct the irradiation curve of a point M of the solar 
atmosphere for one of the strongly refrangible kinds of light from 
the vicinity of an absorption line, it will show a more oblong, oval 
shape than the curve that has been deduced, in fig. 4, from 
P 
1) In this reasoning only the observed fact of the decrease of brightness 
toward the limb was taken for granted, no hypothesis regarding the cause of 
that phenomenon being required so far. The result, therefore, includes the full 
justification of the assumption on which our earlier considerations about the 
propagation of light through the sclar atmosphere were based, viz. that the 
intensity of the light, incident on a small region M of that atmosphere, varies 
rather strongly with the direction (Proc. Roy. Acad. Amst. XII, 268, (1909)). Some 
astrophysicists objected to that view. Founding their refutation on the opinion, that 
a point of the solar atmosphere receives equal amounts of light (per unit of space- 
angle) from all directions meeting the photosphere, so that a circle s’ps would 
represent the irradiation curve, they evidently left the above-mentioned elementary 
result of the observation of the sun entirely out of consideration. 
2) Proc. Roy. Acad. Amst. XIII, 888, (1911). 
