|Humans have some marginal sensitivity to polarized light as discovered
by Haidinger in 1846 (naked-eye)
but it was not until the late 1940's that researchers realized that many
animals can "see" and use the polarization of light. This extra dimension
of reality remains mostly invisible to us without the aid of instruments
but it is of vital importance to a host of animals. After the dance of honeybees
tipped-off Frisch about their gift (bees)
other researchers went looking for P-vision elsewhere and found it in an
extraordinary range of animals, including fish, amphibians, arthropods and
octopuses. These animals use it not only as a compass for navigation, but
also to detect water surfaces, to enhance visual power (similar to colors),
and perhaps even to communicate
(Polarization.net has developed an inexpensive educational
Skylight Compass Card).
We now know that the eyes of many invertebrates have a structure that lends itself for sensitivity to polarized light. So much so, that evolution has taken specific steps to limit this sensitivity and not overwhelm and confuse the sensorial processors. On the other hand, the eyes of most vertebrates are not well suited for the detection of polarization. Reports of this ability in higher vertebrates were often wrong. For example, homing pigeons were thought from the late seventies to early nineties to posses that capacity, only to be disproved by more careful experiments . But we are still far from knowing the full extent of polarization vision in the animal kingdom and its fusion with standard vision. It remains an active and exciting field of research where amateur scientists can still make significant contributions.
The Biological Basis of Polarization Vision in Insects
How do bees, ants, crickets, mayflies and other insects "see" the polarization of light without a polaroid filter or a dichroic crystal? In all animals, invertebrates AND vertebrates, the visual pigment rhodopsin is present in the photoreceptor membrane of the visual cells. These molecules absorb a maximum of linearly polarized light when the electric field vibrates in the direction of their dipole axis. In vertebrates the axes of rhodopsin molecules are randomly oriented. But that is not the case in the eyes of insects.
The compound multifaceted eyes of insects are formed by hundreds or thousands of "simple eyes" called ommatidia. Each of these subunits has its own lens, crystalline, and several long visual cells arranged in a star pattern. The light sensitive part of the visual cells are the "microvilli": an array of tubelike membranes where the pigment rhodopsin is located. All the microvilli of the visual cells of an ommatidium point towards the center of the star, forming a light detecting waveguide: the rhabdom.
The secret of polarization sensitivity is that the rhodopsin molecules are aligned preferentially parallel to the axes of the microvilli tubes. Thus, in principle, each visual cell would be maximally sensitive to light polarized parallel to its microvilli.
Too much of a good thing
As a Chicago gangster would had said: "sometimes you can know too much." Or, in the words of polarization vision researchers Wehner and Bernard: " . . . when zig-zagging over a meadow with all its differently inclined surfaces of leaves, the bee would experience pointillistic fireworks of false colors that would make it difficult to impossible to detect the real colors of the flowers." While polarization provides an additional visual channel to the intensity and color of light, it can also be confusing. After all, the polarization of the light reflected, for example, by a leave, depends on the position of the sun, on the inclination of the leave, and other factors such as if it has rained or not.
For those reasons the polarization sensitivity of insects that use it for navigation is generally restricted to the dorsal upwards-looking portion of their eyes. The polarization sensitivity of the insect visual cells is canceled if the ommatidia (with their visual cells) are twisted along their length. In that case the microvilli tubes at different heights of the cell are oriented in different directions and their preferred directional sensitivity averages out.
The honeybee has about 5,500 ommatidia, each with nine visual cells. Half of the ommatidia are twisted clockwise and half counterclockwise. Most of the visual cells of the ommatidia are twisted by a full 180 degrees, does canceling out the polarization sensitivity. Except for those visual cells responsible of P-vision.
C- vision and P- vision
Bees have three-color vision, but instead of detectors primarily sensitive to red, green, and blue like us, they compose their colors from green, blue, and ultraviolet. Each ommatidium has three visual cells for each color. Eight of the cells are twisted 180 degrees from bottom to top (1 degree per micrometer), but the ninth is much shorter than the rest and only twists by 40 degrees. This last visual cell happens to be one of the ultraviolet-sensitive cells, thus explaining why bees only detect polarization in the ultraviolet.
The twist clearly provides color sensitivity unencumbered by polarization effects. On the other hand, because in bees and ants only detectors of one color (ultraviolet) are sensitive to polarization, changes in hue do not affect the polarization-detection system. Why ultraviolet? Ultraviolet gets through clouds better than the visible wavelengths (don't forget the sun lotion!) and thus provides a more reliable compass.
The description above is strictly true for the honeybee and some ants. Other insects show variations along the same ideas. One of the best studied P-vision systems belongs to the field cricket (Gryllus campestris), where polarization discrimination is performed by blue receptors (actually, the whole dorsal rim area is monochromatic blue).
Unambiguous detection of (linearly) partially polarized light requires at least three different measurements. Wehner proposed a model in which the signals from two short cells twisted in opposite directions and a polarization-blind long cell are combined to determine the direction and degree of polarization of the light, in addition to providing the total intensity to the ultraviolet color channel. For crickets Labhart showed that polarization neurons (POL-neurons) perform the signal processing by antagonistic input from two analyzer channels with orthogonal orientation of maximal sensitivity. Crickets have three types of POL-neurons with different tuning angles.
| M. Coemans, J. Vos HZN, and
J. Nuboer, "The orientation of the E-vector of linearly polarized light does
not affect the behavior of the pigeon Columbia Livia," J. Exp. Biology, 191,
pp. 107-123, 1994.
 R. Wehner, "Polarized-light navigation by insects," Scientific American, Vol. 23 (1), pp.106-115, 1976.
 M. Land (rep.), "Old twist in a new tale," Nature, Vol. 363, pp. 581-582, June 1993.
 T. Labhart, "How polarization-sensitive interneurones of crickets perform at low degrees of polarization," J. Exp. Biology, 199, pp. 1467-1475, 1996.