122 Proceedings of Royal Society of Edinhurgh. [sess. 
The other four required much more green, from 67*9 green 
+ 32 T red to 57 ‘7 green + 42*3 red. 
The mean results Avere, that while with the majority 1 part of 
green required 2*65 parts of red, with the minority 1 part of green 
required only 0*82 part of red. 
He devoted much attention to a consideration of the possible 
cause of this remarkable variation, but failed to attribute it to any 
other source than a real variation of the susceptibility to the colour 
impressions. 
Y. ^ 
Experiments on the same subject by Konig and Dieterici. 
These experimenters examined their own vision. They found for 
one observer 1 green balanced by 2 ’6 6 red j for the other 1 green 
balanced by 3 '25 red. 
But they found two other persons who required respectively 1 
green with 0*96 red and 1 green with 0-71 red. 
The two investigators plotted their colour intensity curves. 
They agreed tolerably Avell as to their red impressions, hut they 
differed so much in their green sensations as to require separate 
curves in the diagram ; these were much the same shape and area, 
hut differed in position by about 5 millionths of a millimetre in 
the wave-length scale. They also plotted a third green curve which 
Avas found to vary still more remarkably from the tAAm others, 
arguing very different sensations throughout this part of the 
spectrum. 
These results were adopted and remarked on by Helmholtz in 
the new edition of his Avork. 
Z. 
Hering on Individual Variations in the Colour Sense, 1885. 
The object of this elaborate essay is to account for the variations 
of the red and green impressions in dichromic vision ; and he begins 
by calling attention to the analogy of similar variations in normal 
vision. 
He mentions remarkable cases of this that had come under his 
own observation. He found that with tAvo observers making a 
