Originally posted by Ian
This is a slightly more technical explaination than CK's.....
First off you need to understand what teh Reciprocity Law actually is.
The law was first devised by Bunsen (of Bunsen Burner fame) and Roscoe in 1862 and states that for any photosensitive material the photochemical effects are directly proportional to the incident light energy. That is the product of a given light level (illuminance) and exposure duration as expressed in the formula H = Et where E = illuminance and t = time. H = product
In photographic terms as long as H is kept constant the values of E and t can be changed by manipulating the aperture and shutter of the camera (as per CK's post).
However, someone (Abney I think it was) demonstrated that the photographic effect depends on the acutal values of H and t and not solely on their product. This failure of the law of reciprocity is caused because the effect of exposure on a photograhic material depends on the rate at which energy is supplied.
Reciprocity failure is a fact of life with all photographic emulsions, and it occurs not only with low lighting levels but also with extremely high illuminence levels (bright light).
In the real world the high light levels aren't often encountered, unless you are photographing high powered laser pulses, atomic bomb detonations or the sun with very large telescopes. However, the low level reciprocity failure is often encountered and depending on film the time for reciprocity law failure can be as little as 1 second or less. With low illuminence level (eg at night) failures a loss of contrast and colour shifts are the two most common indicators of reciprocity failure. However the good news is most films have published compensation charts that allow the photographer to compensate for reciprocity failure.