Visual Ecology

Why are traps blue - When there are no blue animals?

It is hard to understand why biting flies investigate simple, artificial objects in the absence of appropriate host odours. However, from the kinds of flies captured, and their blood-feeding habits (females only in many species), most biting flies likely perceive traps as an opportunity to obtain a blood meal. Some species also investigate objects to encounter the opposite sex (males of many species), to rest in the sun or shade depending on the temperature (perhaps all species?), or to simply get out of the wind (stable flies).

As a general rule, flies react best to artificial objects (traps or targets) that are about the size of large animals, and that contrast well with their surroundings. Some species are very sensitive to movement, and hardly react to stationary objects (deer flies (Chrysops), and savannah tsetse such as G. morsitans, G. swynnertoni). A universal explanation for why blue and black traps catch large numbers of biting flies is unrealistic, given the variety of patterns observed.

A recent publication has suggested that tsetse in particular are attracted to blue objects simply because of the tendency of daytime shadows to be both dark and blue at the same time. This is a reasonable hypothesis for many observations, and is worth testing with appropriate experimental studies.

Steverding, D. & Troscianki, T. (2004) On the role of blue shadows in the visual behaviour of tsetse flies. Proceedings of the Royal Society, Biological Sciences, Biology Letters 271, S16-S17. PDF file

Biting flies perceive the environment differently than people. Their eyes respond to ultraviolet wavelengths, and they have a greatly reduced ability to discern fine detail. Our retinas have peak sensitivity at about 550 nm (yellow). In our visible range (400-700 nm), flies see "colours" at low wavelengths (violet and blue), with some sensitivity in the blue-green. Biting flies cannot see red, although some other insects can.

Biting flies perceive the ultraviolet exceptionally well (<400 nm). Hence, unlike our primary "trichromatic" visual response to blue, green and red, flies likely see primary "colours" corresponding to ultraviolet, blue and blue-green. A highly-visible object for a biting fly might therefore be an object that contrasts well at low wavelengths relative to green vegetation. There is considerable literature on the perception of colour in certain insects; the two papers below are a good introduction.

Briscoe, A.D. & Chittka, L. (2001) The evolution of color vision in insects. Annual Review of Entomology 46, 471-510.
Stavenga, D. (2002) Colour in the eyes of insects. Journal of Comparative Physiology A: Neuroethology, Sensory, Neural, and Behavioral Physiology, 188, 337-348.

A simple explanation for "why traps are blue" is illustrated below. The graph shows the ratio of reflected light for two arbitrary kinds of blue cloth relative to the kinds of vegetation found where tsetse live in Africa (bright green Acacia leaves on a dark green Combretum leaf). Note how a dark blue cotton still appears as a bright object in the blue region at 450 nm relative to vegetation (reflecting 300% as much light). When combined with black cloth, a blue trap would be a novel object, perhaps worthy of investigation for a hungry fly. To us, a bright blue polyester would look similar to the blue cotton, but to a fly, it would look very different. We would notice the additional brightness in the blue region, but the fly would be able to detect the large difference in reflectance in the ultraviolet - violet. Both pieces of cloth would appear as bright objects to the fly, but the fly would see a more complex "colour" in the polyester. The smooth surface of the polyester would also cause polarisation of reflected light. The fly would also be able to see this feature.

These standard reflectance data are for an artificial situation in the laboratory (new, dry cloth against a black background). Economical trap fabrics are often not opaque, and hence they will both reflect and transmit light under natural conditions. Cloth also appears different when wet or weathered, when in the sun, cloud or shade, when viewed against growing or senescent vegetation, etc. Changes in appearance also occur as the sun's angle changes during the day, resulting in differing amounts of light falling directly on the trap, or reflected onto it from the environment. These "real-world" features make it difficult to predict how any fabric will perform in nature, based solely on standard reflectance data obtained in the laboratory.

To illustrate these phenomena, I have taken a some representative photographs of traps in nature, using blue and ultraviolet filters. Follow the link below to see how the environment might look through the eyes of a biting fly.

Traps in Nature
(MANY IMAGES, PAGE LOADS SLOWLY)

For a technical introduction to the many factors involved in the description and measurement of colour see:

Nassau, K. (1998) Color for science, art and technology.
Amsterdam, Elsevier.