88 J. LeConte— Phenomena of Binocular Vision. 
flat on my back, I see the image perfectly erect, in the zenith. 
Turning now the eyes upward (toward the brows) and to the 
right and left; and then downward (toward the feet), and to 
the right and left, the whole image of the building rotates with- 
out engine precisely as indicated in my previous experi- 
me 
T am perfectly confident then that I am justified in form 
lating the torsions of the eyes when moving together with their 
optic axes me thus 
i. n the visual ine is elevated and the eyes move to 
the right, pes rotate on their optic axes to the right; when 
they move to the /eft, they rotate to the /e/t. 
2. When the visual plane is depressed, then motion to the 
right is accompanied by aes to the /e/t, and motion to the 
left by rotation to the r 
3. The degree of ite increases with the amount of eleva 
tion or depression of a visual plane, and of the lateral excur- * 
sion of the point of si 
Now, the above laws ‘a and 2) concerning the direction of 
torsion, so precisely the reverse of those eae by Helmholtz, 
and therefore of what I expected to find when I commence 
this real on. I quote from his work on Physiological 
ptics, French edition, 1867.* This edition was revised, cor- 
rected and added to by Helmholtz himself, and by his own 
statement is not only later but more pathositatixe than the 
German. 
“ When the plane of regard is directed upward, lateral displace- 
ment to the right makes the eye vk to the left, and displacement 
to the . makes it turn to the right 
. the plane of regard is depressed, lateral ro squeer aa 
to oe te are accompanied with torsion to the right, and vie 
* ae other words, when the vertical and lateral angles are both 
of the same sign, the torsion is ena ive ; when they are of con- 
trary signs the torsion is positi 
The very reverse of every one of these propositions is demon- 
strably true. 
next set myself to find out how the mistake arose. q find 
its origin evidently contained in the following statemen 
“Tf we throw a vertical image on the wall (supposed to be 
covered with rectangular codrdinates vertical and horizontal), 
we obtain a rotation in direction contrary to that which we 
have just seen (in the case of the horizontal image). In fact, if 
one looks upward and to the right, the image does not turn to the 
left, but to the right in relation to the vertical lines of the wall. But 
one cannoi conclude from this that there is a rotation of the eye 
* Optique Physiologique, p. 602 and 603. 
es eo 
