124 Sir J. F. W. Herschel on the Action of the Rays 
space, in which the tint rises toa brownish or reddish orange, 
the terminal edges of the oval being a very little paler than 
the middle. Immediately, and with absolute suddenness, as 
if coming out from under this oval, reappears at D the black- 
ness of the region B, but apparently more intense (as if the 
same quantity of blackness had here been crowded intoa nar- 
rower space), and with an evident bluish cast. Within the 
coloured space C, and around a point m, which may be re- 
garded as their common focus, the isochromic ovals may be 
considered as arranged, which, crowded together undistin- 
guishably towards D, elongate in the direction of B, as indi- 
cated by the dotted lines. A feeble ash-gray oval train G, 
barely perceptible, appears to come out again as it were from 
under D, or to form a kind of tail to it, leaving a slight inden- 
tation H on either side, occupied by the specularly white 
continuation of the portion E. ‘Those who are accustomed 
to the analysis of such cases will easily perceive here the effect 
of an abruptly terminal solar image at m overlapping and 
concealing a sudden descent to a minimum, or perhaps to a 
total absence of photographic action at H. 
g. The borders of the elliptically terminated portions A, B, 
inclose and overpass one another as represented in the figure. 
For instance, the border of the portion A extends not merely 
beyond the blackened space B, but even somewhat beyond 
the extremity of C, thinning away there to an exceedingly 
delicate line, as does also that of the portion B itself. No 
traces of these borders however can be followed down the 
whole length of C. 
9. In the order of tints above described as prevailing from 
the more refrangible end of the spectrum to the maximum of 
tint m, we recognize without difficulty the Newtonian series of 
colours of the first order of the reflected rings: modified how- 
ever in its first stages by a cause which seems to have shifted 
the initial black of that series to a higher point in the scale 
of thicknesses of the producing film, or to have displaced the 
whole series by the intrusion of a white commencement. For, 
the Newtonian reflected tints of the first order are, black, 
very feeble and hardly perceptible blue; brilliant white; 
yellow ; orange (at which point the series breaks off). And if 
we suppose these tints (so modified) repeated again in reverse 
order, and consent to attribute the more intense apparent 
blackness of the region D beyond that of B to the effect of 
coutrast produced by the greater abruptness of transition in 
that region (an effect which is very striking in many optical 
phzenomena), the whole spectral impression from end to end 
will come to be accounted for by a film, homogeneous in its 
