270 



SCIENCE. 



[N. S. Vol. VII. No. 165,- 



relax and in this relaxation diverges some- 

 what. 



The observations here described have 

 been confirmed by a number of persons. 

 Only one case appeared in which the results 

 were dififerent. In this case, however, one 

 eye is not normal in its vision and in 

 drowsiness, as well as under the conditions 

 discussed convergence rather than diver- 

 gence was regularly observed. 



The degree of divergence is difficult to 

 determine, as the double images last only a 

 very short time, the convergence adapting 

 itself very soon to the object. The phe- 

 nomena described appear most strikingly in 

 the case of very distant objects, that is, 

 where the optical axes were at the start 

 parallel. On the other hand, where there 

 is an effort required in the original con- 

 vergence, strong enough to give a clearly 

 conscious impression, the closed eye does 

 not seem to relax as much. The degree of 

 relaxation in the closed eye seems, in 

 general, to be inversely proportional to the 

 degree of effort required to maintain the 

 original convergence. 



The conditions may be modified so that 

 relaxation shall result in convergence 

 rather than in divergence. Take an object 

 situated so far from the median plane that 

 the opposite eye can just see it over the 

 root of the nose. Suppose, for example, 

 that the object is on the right. If now the 

 right eye be closed, while the object is 

 fixated with the left, and then be suddenly 

 opened, it will be observed that the double 

 images are not crossed. This indicates that 

 the eyes are converged to a point nearer 

 than the object. Care must be taken in 

 this experiment to fixate the object with the 

 left eye. If the object is seen in indirect 

 vision the conditions are, of course, modi- 

 fied. 



The only inference possible from these two 

 sets of facts is that there is some line situ- 

 ated between the parallel and extreme lat- 



eral positions of the optical axes to- 

 wards which the closed eye tends. Le 

 Conte has surmised : " It is probable that 

 in a state of absolutely perfect relaxation 

 the optic axes coincide with the axes of the 

 eye-sockets, and it requires, therefore, some 

 contraction to bring the optic axes to a con- 

 dition of parallelism and still more to a 

 condition of convergence, as in every vol- 

 untary act of sight."* This surmise seems- 

 to be confirmed by the facts described and 

 by the additional fact that a certain angle 

 can be found between the position of paral- 

 lelism and the extreme lateral position at 

 which there is no tendency for the eye ta 

 change the degree of its convergence when 

 closed. This angle corresponds with the 

 angle of the axes of the eye-sockets. But in 

 any case the tendency of the closed eye to- 

 diverge is checked when the effort towards 

 convergence is strong enough to be notice- 

 able. 



O I 



The two figures will make clear the fact. 

 The dotted lines represent the axes of the 

 eye-sockets towards which the eyes tend ta 



*Loc. cit., p. 161. 



