nous arm of the cross first projects aslant the
dark field, and then slowly disappears, after which the second
(perceived by the other eye) shoots out suddenly in a direction athwart
the first. This alternation proceeds for a long time, and produces the
curious effect of two luminous blades crossing and recrossing each
other.

[Illustration: FIG. 111.--STEREOSCOPIC DESIGN]

Another method of bringing out the phenomenon of alternation in a still
more striking manner is to look at two different sets of writing, with
the two eyes. The resultant effect is a blur, due to superposition,
and the inscription cannot be read with the eyes open. But on closing
them, the composite image is analysed alternately into its component
parts, and thus we are enabled to read better with eyes shut than open.

This period of alternation is modified by age and by the condition of
the eye. It is, generally speaking, shorter in youth. I have seen it
vary in different individuals from 1" to 10" or more. About 4" is the
most usual. With the same individual, again, the period is somewhat
modified by previous conditions of rest or activity. Very early in the
morning, after sleep, it is at its shortest. I give below a set of
readings given by an observer:

Period

8 A.M. 3"
12 noon 4"
3 P.M. 5"
6 P.M. 5.4"
9 " 5.6"
11 " 6.5"

Again, if one eye be cooled and the other warmed, the retinal
oscillation in one eye is quicker than in the other. The quicker
oscillation overtakes the slower, and we obtain the curious phenomenon
of 'visual beats.'

#After-images and their revival.#--In the experiment with the stereoscope
and the design of the cross, the after-images of the cross seen with the
eyes closed are at first very distinct--so distinct that any unevenness
at the edges of the slanting cuts in the design can be distinctly made
out. There can thus be no doubt of the 'objective' nature of the strain
impression on the retina, which on the cessation of direct stimulus of
light gives rise to after-oscillation with the concomitant visual
recurrence. This recurrence may therefore be taken as a proof of the
physical strain produced on the retina. The recurrent after-image is
very distinct at the beginning and becomes fainter at each repetition; a
time comes when it is difficult to tell whether the image seen is the
objective after-effect due to strain or merely an effect of 'memory.' In
fact there