ANISOTROPY OF THE HUMAN EYE AND ITS RECEPTORS

As is known, the eye of many (though by no means all) people is capable of distinguishing linearly polarized light from unpolarized light.

If one observes a white extended surface emitting linearly polarized light, then an eye sensitive to the polarization of light distinguishes, in the middle of the uniform light field, a faint lemon-

yellow band, resembling in its shape a slightly unrolled and bent sheaf of ears of grain or the figure 8. This figure, called Haidinger’s brush (1844), is located in the plane of polarization, i.e. perpendicular to the plane containing the electric vector of the light wave. Sometimes, possibly owing to contrast, one may also notice an even weaker analogous bluish figure, oriented perpendicular to the plane of polarization of the light.

Although this phenomenon has long been known (it was described, for example, by L. N. Tolstoy in his novel Youth, 1855, of course from a purely artistic point of view), its nature had remained not entirely clear. The only rather trivial result of the few investigations, which were purely qualitative in character, is the conclusion that the ability of the human eye to distinguish linearly polarized light and to determine the position of the plane of its polarization is due to anisotropy of the eye itself and of its perceptive apparatus (Jamin, Helmholtz—1896). All the greater interest, therefore, is presented by the paper under review,* in which the authors succeeded in clarifying the nature of this anisotropy and, for the first time, in making an attempt (admittedly a rough one) at its quantitative investigation.

For studying the polarization properties of the eye the authors used a spectrophotometer, the comparison fields of which had the form shown in the figure. The fields were illuminated with light polarized in mutually perpendicular directions, indicated in the figure by the hatching. After visual equalization of the brightnesses of the fields, the directions of polarization were turned through 90° (this was accomplished by rotating half-wave plates). Such a change in the directions of polarization without a change in the intensity of the light led to a disturbance of the visible equality of brightnesses of the fields. To restore equality of the brightnesses of the fields it proved necessary to increase the brightness of the lower field (see figure) by a certain number \(f\) of times. The factor \(f\) depended on the wavelength of the light and was different for different observers. It had its greatest value for \(\lambda = 460\ \mathrm{m}\mu\), varying (for ten observers) from 1.05 to 1.30.

Special measurements showed that the spectral dependence of \(f\), within the limits of error, coincides with the spectral absorption curve of the yellow pigment. This indicates that the eye’s ability to distinguish polarized light is caused by dichroism of the yellow pigment, to which the yellow color of Haidinger’s figure also corresponds. Furthermore, the factor \(f\) proved not to depend on the orientation of the fields relative to the eye. If one takes into account that the eye is always more sensitive to light polarized along the photometer fields, this permits one to conclude that the elements of the yellow pigment are distributed over the fundus of the eye not chaotically, but that in their distribution there is a certain radial order, the degree of which varies among different observers. In this case the radial component of the electric vector of the light wave is absorbed more weakly than its tangential component. This conclusion is also confirmed by the fact that, if the initial and final orientation of the planes of polarization made an angle of 45° with the boundary between the fields, no disturbance of the equality of brightnesses was observed.

) H1. de Vries, R. Jielof, A. Spoor, Nature 166*, 958 (1950).

An analogous phenomenon was observed by the authors when the fields of a photometer were illuminated with circularly polarized light, with opposite directions of polarization. If, in both fields, the direction of circular polarization is changed to the reverse (by rotating a quarter-wave plate), the equality of the brightness of the fields is disturbed. In this case, however, \(f\) depends substantially on the orientation of the photometric fields relative to the eye. The authors believe that this result, too, can be explained by the anisotropy of the refracting media of the eye, discovered in 1940 by Boehm.

The essence of the explanation is that the refracting media of the eye transform circularly polarized light into elliptically polarized light, while the dichroic element responds differently to elliptically polarized light depending on its orientation.

Of course, the work under review can be regarded only as very preliminary. It opens up, however, new and interesting possibilities for studying the properties of the human eye. In addition, it has undoubted practical importance. As is known, the action of most subjective spectrophotometers is based on equalizing the brightnesses of two fields illuminated by light linearly polarized in mutually perpendicular directions. The authors’ investigations show that neglect of the polarization properties of the human eye may here serve as a source of serious errors. It is not excluded, in particular, that this circumstance is one of the causes of the frequently observed significant disturbances in the equality of photometric fields along the boundary separating them.

Thus, from the standpoint of the technique of visual spectrophotometry, further study of the ability of the human eye to perceive polarized light is absolutely necessary.

V. Yur’ev