It was natural for us to think 
of trying to induce strabismus, 
and hence amblyopia, in a 
kitten or monkey by surgically 
cutting an eye muscle at birth, 
since we could then look at the 
physiology and see what part of 
the path had failed. We did this 
in half a dozen kittens and 
were discouraged to find that 
the kittens, like many 
children, developed alternating 
strabismus; they looked first 
with one eye and then the 
other. By testing each eye 
separately, we soon verified 
that they had normal vision in 
both eyes. Evidently we had 
failed to induce an amblyopia, 
and we debated what to do next. 
We decided to record from one of 
the kittens, even though we 
had no idea what we could 
possibly learn. (Research often 
consists of groping.) The results 
were completely unexpected. As 
we recorded from cell after cell, 
we soon realized that 
something strange had 
happened to the brain: each 
cell responded completely 
normally, but only through one 
eye. As the electrode advanced 
through the cortex, cell after 
cell would respond from the left 
eye, then suddenly the 
sequence would be broken and 
the other eye would take over. 
Unlike what we had seen after 
eye closure, neither eye seemed 
to have suffered relative to the 
other eye in terms of its overall 
hegemony. Binocular cells 
occasionally appeared near the 
points of transition, but in the 
kittens, the proportion of 
binocular cells in the 
population was about 20 percent 
instead of the normal 85 
percent, as shown in the graph 
to the left.

After we cut one eye muscle in a 
kitten at birth and then 
recorded after three months, 
the great majority of cells were 
monocular, falling into groups 
1 and 7.

    We wondered whether most 
of the originally binocular cells 
had simply died or become 
unresponsive, leaving behind 
only monocular cells. This 
seemed very unlikely because 
as the electrode advanced, the 
cortex of these animals yielded 
the usual richness of 
responding cells: it did not 
seem at all like a cortex 
depleted of four-fifths of its 
cells. In a normal cat, in a 
typical penetration parallel to 
the surface in the upper layers, 
we see about ten to fifteen cells 
in a row--all dominated by the 
same eye, all obviously 
belonging to the same ocular-
dominance column--of which 
two or three may be monocular. 
In the strabismic animals we 
likewise saw ten to fifteen cells 
all dominated by one eye, but 
now all but two to three were 
monocular. Each cell had 
apparently come to be 
dominated completely or almost 
completely by the eye it had 
originally merely preferred.

    To appreciate the surprising 
quality of this result you have 
to remember that we had not 
really interfered with the total 
amount of visual stimulus 
reaching either retina. Because 
we had no reason to think that 
we had injured either eye, we 
assumed, correctly as it turned 
out, that the overall traffic of 
impulses in the two optic 
nerves must have been normal.