1888.] Mr G. N. Stewart on Intermittent Light. 
459 
changes are got. The very centre is always white. Evidently the 
explanation is, that with very intense stimulation the curves for all 
three sets of fibres rise so abruptly that we cannot find an appreci- 
able difference in the proportion of their ordinates for any speed of 
the mirror below that necessary for steady fusion, except at the 
edges where the light is less intense. 
Having examined these two experiments at some length, it is 
only necessary to glance at the others. 
Take Experiment 4. — Here there is no difference in colour 
between edges and central part, because the feeble red light hardly 
stimulates any fibres except the red. 
Experiment 5. — Here, with the very slowest rotation, there are 
no coloured edges. Why not, if they are due to contrast*? With 
increase of speed the red appears first at the edges where the light 
is weak, and then in the middle. 
Experiment 6. — Here the light was not a pure green. There was 
a good deal of red in it. With slow rotation the central band is 
nearly white, because the intense green light stimulates the three 
groups of fibres strongly. The purple edges might be explained as 
due to contrast, since the violet and red fibres at the edge of the 
image will be more active than the green. With a higher speed, 
the time of stimulation is too short for the red and violet fibres to 
be much affected in comparison with the green, and the central 
part is therefore green. We should have expected green to appear 
about the edges before this, if the explanation we have given of the 
central change is sufficient for the edges too. That is why I said 
above, that contrast must certainly be taken into account. It must 
be taken into account where the light is very intense, as it was here. 
That contrast is not all is shown, I think, by the final stage, where, 
although the stimulation is still very strong, there is no contrast 
colour. Why should the purple edges have disappeared here ? The 
stimulation of the green fibres is still very great in the centre of the 
image, why does the “ sympathetic ” exhaustion not spread to the 
edges, leaving mainly the violet and red fibres active, and giving a 
purple rim ? If our explanation be extended to the edges, the 
answer would be — Because the speed is now too great, ^.e., the time 
of stimulation too small for the violet to preponderate. An objec- 
tion is, that it is red + violet which make up purple, and since we 
