368 Proceedings of the Royal Society 
(Here we are only speaking of luminous radiations. But it is not 
improbable that in the scale of wave-lengths this is true in other 
parts. For refraction by prisms accumulates rays of different 
wave-lengths so much in the ultra-red part of the spectrum that 
no experiments exist which can settle this point.) 
4. At lower temperatures the apparent maximum is nearer to the 
red, i.e ., the maximum of the curve i - i' is nearer to the red. 
But the curve % is always a minimum at the yellow. Hence, at 
lower temperatures, the maximum of the curve i= /(0, A) passes to 
the region of greater wave-lengths. (It has just been stated that 
the curve i= <p (A) has a minimum at the centre of the diffraction 
spectrum. This is nearly certain, because we have seen that the 
curve is, at any rate, nearly symmetrical about this point, and it 
certainly increases enormously at the two limits of visibility of the 
spectrum.) 
The only other remark I have to make on the curves f (0, A) is, 
that we cannot estimate the nature of the curve in the ultra- red at 
present. For all we know, there may be radiations of much greater 
wave-length than any which lamp-black, or any other substance we 
know of, could absorb. 
As to the curve of limiting visibility, it appears, from what has 
already been said, that it has a minimum in the yellow; and although 
fromMossotti’s interpretation of Frauenhofer’s observations, it would 
seem to be a sinuous line, I do not think that the small variations 
there indicated could be detected accurately in judging of the rela- 
tive brightness of different colours. 
We can scarcely see those parts of the spectrum that lie be- 
yond the lines A and H respectively. The question arises as to 
whether they are always invisible. If this were so, the curve of 
limiting visibility, which we have called <p (A) would, at those two 
points, be an ordinate of the curve. 
But I do not think this is the case. So far as I can see, the limit 
of the spectrum depends upon the intensity of the light. Thus, Mr 
Glaisher, in his report to the British Association in 1863 on his 
balloon ascents, stated that, at great heights in the solar spectrum, 
he could “ see H, and far beyond,” when on the ground the line 
“ H was quite the limit.” 
Again, I remember (although I cannot find a reference), that Sir 
