Curious case of the mouse that dies in the sunshine: David O'Connor discovers technology imitating nature as he investigates why his system seizes up

Here is a curious example of technology imitating nature - I have discovered that my computer mouse, like the real animal, operates best in subdued light and hates bright sunshine.

I have been having problems with my mouse intermittently ever since I bought my personal computer. On most days it performs impeccably. At other times, however, it steadfastly refuses to behave and my cursor on my computer's screen becomes paralysed.

I now realise that my mouse plays up only on sunny days, when it is exposed to the full glory of Southampton sunshine coming through my office window. When I place it in the shade, cover it with black tape or use it in the evening the problem disappears.

The reason for the fault nicely illustrates how an apparently elegant technological improvement can ultimately turn out to be less reliable than the original design. It turns out that the problem lies in the way in which mechanical information is converted into electrical pulses to be sent along the cable to the computer.

As the mouse is rolled across a flat surface the ball within it turns two rollers, one tracking the mouse's vertical movements, the other its horizontal movements. Each roller is connected to a perforated disc that has a light source on one side of it and a photo-sensor on the other.

As the disc rotates, driven by the movement of the mouse, the light beam is interrupted so many times a second. The faster the mouse is moved, the faster the disc rotates and hence the greater the number of pulses of light reaching the sensor every second. The light pulses are then converted into electrical signals which can be interpreted by the computer to move the screen cursor.

The weak spot in this ingenious system is that on sunny days enough light gets through the small gap between the buttons to dazzle the photo-sensor, resulting in a bewildered mouse.

Tests, albeit limited to the mice in the biochemistry department where I work, suggest that the problem is quite common. Three out of the four I checked - manufactured by Aero, Naksha, and Q-tec - were overwhelmed by October sunshine; only a mouse from Elonex coped well under the same sunny conditions.

The problem may therefore be a widespread source of annoyance to many computer users, particularly as modern 'point-and-click' operating systems such as Windows rely heavily on mice for their operation. If autumnal British sunshine is enough to knock these mice out, heaven only knows what it is like elsewhere. Why is this irritating fault not more well-known? Certainly the mouse manufacturers, who must surely know about it, do not mention this important limitation to their product in any of the literature I have seen.

Moreover, calls to a help-line set up by one of the manufacturers suggest that their technical staff are either unaware of the problem or are reluctant to admit that it exists.

My telephone call resulted in an offer, made in rather patronising tones, to replace my brand-new mouse with another of exactly the same type - hardly a satisfactory solution given the inherently flawed design. I suspect that many people with mice that give up in the midday sun assume that the rollers or ball within the mouse require cleaning and unnecessarily buy a kit for this purpose. It should not be difficult to produce a light-proof mouse in the future.

In the meantime, however, I need to solve the problem myself, preferably without having to shut out the sunlight from my office. Who knows, perhaps parasols for mice will become this season's ultimate computer accessory?

David O'Connor is a lecturer in biochemistry at the University of Southampton.