106 J. LeConte — Phenomena of Binocular Vision. 



this would be most distinct, in fact would be perceivable only 

 in the case of phantom images. If, therefore, we had a circular 

 room with regular tessellated wall, and leaning with back 

 against one side we combined the figures by extreme conver- 

 gence, I suppose the phantom image would appear plane. It 

 would be interesting to verify this prevision by experiment. 



(2.) I have said that all the phenomena of binocular vision 

 are explained by the law of corresponding points. - In my 

 previous papers * I have contended that the perception of 

 binocular perspective of different objects, one beyond the other, 

 and of binocular relief of the same object, is due not only to the 

 perception of double images, but of these as homonymous 

 and heteronymous, the one as belonging to objects or points 

 beyond, the other to those nearer than, the point of sight. It 

 has been objected to this view, that in many cases we dis- 

 tinctly perceive relief without being able to detect double 

 images of any kind either homonymous or heteronymous. 

 This objection, it seems to me, shows a misconception. It is 

 well known that most persons are at first wholly unconscious of 

 double images, even, though they constantly base their visual 

 judgments on them. Then, when their attention is called to 

 them, they learn to perceive them consciously if they are very 

 distinct and far apart. Then, in proportion to careful atten- 

 tion, those persons most practiced in binocular experiments see 

 them in cases where they are less and less separated. But in 

 all persons, even the most practiced, there is still a residuum, 

 decreasing with practice, of unperceived or unconsciously per- 

 ceived double images, upon which judgments of relative dis- 

 tance are based. But is it not rational to place these residual 

 cases also in the same category with those we have already 

 learned to distinguish ; and to say that subtle unconscious per- 

 ceptions of double images is the mode by which we judge of re- 

 lief in these cases also? Or we may leave out of view the 

 question of double external images and refer only to the law 

 of corresponding points. I would then give the theory of 

 binocular relief as follows : Impressions on corresponding 

 points are referred to objects at the point of sight or at that 

 distance ; but impressions on non-corresponding points — even 

 though it be but the next contiguous rod right or left — are re- 

 ferred to distances greater or less than the point of sight ; if the 

 impressed points are farther apart than corresponding points the 

 impression is referred to distances nearer, if the impressed points 

 are nearer together than corresponding points to distauces 

 farther than the point of sight. This is an inherited capacity 

 which may be partly or even largely, but not wholly, analyzed 

 into its effects as double images. 



* This Jour., Ill, vol. ii, p. 1, 1871, and same vol. ii, p. 417. 



