218 W. B. Rogers on Binocular Vision. 
conditions, and then only approximately, that either the image or 
the wire is seen as a single line. 
In order to repeat the experiment under various conditions, I 
use the following arrangement. Fixing a strong sewing needle 
midway in a small cylinder of cork and firmly fastening into 
the cork one end of a thin stiff wire or knitting needle, so that 
the two needles shall be at right angles, I secure the former im- 
movably in a horizontal position across the opening of the upper 
stage of my stereoscope very near the inner edge of the opening. 
I can then adjust the long wire to any inclination, and the friction 
of hod sos on the horizontal needle will retain the wire in its 
positi 
Taking fig. 21 or 22 as the subject of ipl madman I place the 
drawing on the farther stage and move the other back or forward 
until a notch in the near end of the wire is brought to coincide 
precisely with the point of binocular union of the two lower 
ends of the component lines. I now unite the two upper ends 
this is mie the wire is say to coincidence with the perspec- 
tive image. By sliding the stage a little nearer, we may see the 
image just behind the wire, and by moving it the opposite way it 
will appear on the near side of the wire. Whether coincident 
or in either of these positions, both the image and the wire will 
appear as two perspective lines intersecting at the point on which 
the eyes are for the moment fixed. 
By continuing the same convergence, we cause the wire as 
well as the coincident image to subside from the perspective po- 
sition, and we see them each resolved into two intersecting lines 
in a plane parallel to that of the diagram. Fixing the eyes of 
the near end of the wire or of the binocular resultant, they each 
appear in the plane in the form of a \/; gazing intently on the 
remote end they present themselves as a \/ inverted. If we 
direct our eyes intently on the plane of the paper we see the wire 
as two black wires covering or adjoining the inclined lines of the 
diagram 
With lines ding seat as in fig. 23, we obtain similar effects; 
but as might be inferred from what precedes, whether we regard 
the wire or the eoinciiean4 image, the divergency of the goer 
ing perspective lines is less remarkable than in the prece 
cases, and a more perfect and continued fixity of axial aireetibe 
is requisite to destroy their relief. 
A very interesting effect is observed when the diagram is 8° 
placed on the farther stage that the wire after being adjusted as 
above, leaves uncovered a portion of each of the inclined lines. 
this case the perspective figure is composed partly of the te 
