“ 
=F 
a 
BS 
W.B. Rogerson Binocular Vision: ‘OBA 
I cannot see in the phenomenon, even were it of- constant occur- 
tence, an explanation of the appearances in question. Such a 
view seems to me inconsistent with the.fact always observable in | 
these combinations when not too rapidly ads, that the vertical 
and other component lines assume a p€rspective attitude while 
still intersecting one another at an appreciable angle, that in this 
position the former has already lost its appearance of parallelism 
to the other vertical, that the two intersecting perspective lines in 
spite of an occasional disappearance in part, continue each esters 
ally visible until they have so closed upon one another as to a 
pear coincident, and that even when thus united the presence ar 
both of them in the resultant is shewn by the continued visibility 
of the dots or other marks intended to distinguish them. 
In this experiment, various special phases of combination be- 
tween the lines a and ¢ present themselves, according to the di- 
rection and range of the axial movement by which the vis- 
ion is effected. Sometimes the upper halves only, and ye pg 
the lower halves of the lines, are united, presenting an effect.i 
tical with that of fig. 36, previously described. But these el 
other features of the experiment being unessential to the present 
inquiry, need not be considered in this place. They will be 
treated of hereafter in connexion with a variety of analogous 
eam belonging to the class of alternating or changeable 
combinations. 
20. asain of lines lying in the same horizontal di- 
rection. 
When two equal right lines, 6 40, lying in the same horizon- 
7) 
tal aiiion at right angles to the ce of view are united binoc- 
ularly either behind or in front of their real position they form, as 
might be expected, a resultant parallel to the components. The 
same convergence which serves to combine any two correspond~ 
and 6 an 
two lines. When therefore the lines are so short as to be dis- 
tinctly visible throughout their length without a change of axial 
direction, the formation of the resultant behind or before the 
components and parallel to them must be regarded as simultane- 
ous throughout. 
In the case of unequal horizontal lines the results are very 
different. Here the combination is to a certain extent succes- 
Sive involving a —— of convergency and giving rise to a per- 
spective resultan 
Sir David ects in discussing the question of the coinci- 
dence of unequal figures adduces the first observatiou on this sub- 
