of the Solar Spectrum on Vegetable Colours. 179 
the paper dried the action seemed to be suspended. It be- 
came therefore necessary to renew the hydriodic wash, and 
thereby to increase the actual quantity of that salt present on 
the paper. Both actions grew more intense, but the bleaching 
effect most so. A perfect semicircle and long cometic train, c, 
d, fig. 12, No. 1, (Plate II.), was produced, within which space 
the blackness of the paper was totally destroyed, and replaced 
by white or rather very pale yellow. ‘The hydriodic washes 
being again and again renewed, the darkness at first pro- 
duced in the lower part of the spectrum began to give way, 
and was slowly replaced by a very feeble bleaching, which at 
length extended very far indeed below the extreme red rays, 
and upwards to join the semicircle C fig. 12, No. 2, which 
had by this time assumed an outline perfectly sharp and well- 
defined, having its centre on the original point + 26°5 of its 
commencement. But within this semicircle and its train, re- 
markable changes were observed to be all the while in pro- 
gress. First, a somewhat dark, and grayish or brownish, 
perfectly circular and well-defined solar image arose, its dia- 
meter being somewhat less than that of the semicircular ter- 
minations, so as to leave a clear and distinct white border all 
around zt, as represented by the dotted line in fig. 12, No. 2. 
Shortly after the complete formation of this spot, 2. e. after its 
circular outline could be distinctly traced all round, it began 
to extend itself upwards into an oval or tailed form, but pre- 
serving its circular shape below and maintaining the white 
border inviolate, assuming at the same time a brownish yellow 
colour which gradually deepened, but never became intense. 
At the same time a very remarkable change was observed to 
take place in the reflective (or absorbent) powers of the paper 
in this region. ‘The vivlet-coloured end of the spectrum, 
which hitherto had been distinctly seen as usual occupying 
the space from + 30 to + 40°6, became quite indiscernible, 
while on the other hand the blue rays adjoining became re- 
flected with such copiousness as to terminate the spectrum by 
a well-defined semicircle ¢, fig. 12, No. 3, and to give to the 
whole portion ce the appearance of a brilliant and purely 
blue spot. Finally, after long-continued action, the interior 
browned oval above-mentioned was found to have been pro-« 
longed into a figure of the form No. 4, fig. 12 (distinctly seen 
at the back of the paper), of which the termination by a nar- 
row neck and circular enlargement indicates the definite ac- 
tion of a ray much further removed along the axis of the 
spectrum. Washing with water at once obliterates this part 
of the phenomenon, destroys the brown colour, and leaves 
simply the bleached cometic a in singularly striking con- 
2 
