This summer, many of us sat in darkened theaters and stared at screens that showed mesmerizing events of other worlds. In "Contact," extraterrestrials communicated with Jodie Foster. In "Men in Black," Will Smith battled an oversized alien roach. And, in "The Lost World," velociraptors jumped through the grass, and T. rex terrorized suburbia.

Amid such unearthly action, it is easy to forget that, despite their name, motion pictures actually are a series of still pictures. Nothing moves on the screen; all of the motion is in our heads. So, why don't we notice this when we're watching? How can a succession of still pictures create the appearance of movement? Remarkably, one key to achieving the effect is to darken the screen for a brief but essential period of time between each pair of still images, or movie frames. As it happens, during more than half of the time that we're watching the movie, the screen is black, and the theater is dark. That's something else we don't notice. Scientists don't fully understand how movies work because our visual system is so elaborate and the ways in which we process cinematic images are too complex. But since research on the phenomenon began in the early 19th century, we have come closer to comprehending the delicate interplay between the brain, the eyes and a rapidly changing sequence of still pictures. That research, which would lead to the invention and development of movies, began almost two centuries ago. About 1817, an English scientist, Sir John Herschel, bet a friend that he could show the head and tail of a shilling at the same time. When Herschel spun the coin on a table, his friend saw that the spinning sides of the coin fused into a single image. This phenomenon came to be known as "persistence of vision." In 1824, Peter Mark Roget, author of the renowned thesaurus but also an early experimenter with the phenomenon, defined persistence of vision as "the ability of the retina to retain an image of an object for 1120 to 115 of a second after its removal from the field of vision." Although Roget's conclusions were oversimplified, his work encouraged the invention of several motion-simulating devices that would lead to the motion picture. For example, in 1825, John Ayrton Paris developed the Thaumatrope, the most famous of the early "persistence of vision" toys. This was a disk with a picture of a parrot on one side and an empty cage on the other. By twirling the disk like Herschel's coin, the viewer sees one picture and then then other, and the two images merge: the parrot is inside the cage. In 1829, Joseph Plateau, a Belgian, published his investigations on the same phenomenon. Over many years, Plateau repeated a simple experiment. He would stare at the sun and then look into darkness to study the fading optic image. By age 28, his experiments had driven him blind. He partially recovered his sight several months later and, in an example of extraordinary scientific dedication, resumed the experiment, only to find his vision permanently damaged at age 40. With the aid of his wife, however, Plateau continued to perform less harmful experiments until his death at 82. In 1832, he marketed a toy that he thought demonstrated his theoretical research. Painted around the edge on one side of a disk is a series of 16 still pictures in slightly varying positions. Between each pair of images is a slit through the disk. Viewers hold the disk with a handle, spin it and look through the slits at a mirror. As a slit comes before the eye, allowing a view of the mirror, the reflected image of one still picture is visible briefly. As the disk spins, the picture seems to move. Plateau called his toy the phenakistoscope, from the Greek for "deceptive viewer." {You can make a phenakistoscope. See Phenomena, below.} Plateau inadvertently had employed two features that later would be seen as essential to movies. As the viewer stared at the disk, the eye would see brief periods of darkness between each pair of still images. Also, the machine showed that 16 images per second was fast enough to produce the illusion of continuous movement. A successful projector would not be invented until an analogy to Plateau's slits was discovered, and early filmmakers soon realized the effectiveness of 16 frames per second. At the same time, others were working on similar technologies. In Vienna, Simon Ritter von Stampfer independently created a toy similar to the phenakistoscope. In 1834, William George Horner changed the flat disk into a circular drum. Paper strips bearing images fit inside the drum, and when the viewer looked through the slits into the spinning drum, the pictures on the paper became animated. Because the paper could be changed, the same machine could show a variety of moving pictures. Horner first called his toy the Daedelum, then the Zootrope or Zoetrope, from the Greek for "life turning." In 1843, an Austrian artillery officer named Baron Franz von Uchatius began work on what would become the movie projector. He combined the phenakistoscope with the magic lantern, a candle-powered slide projector. Uchatius mounted a transparent glass disk of many slightly different figures and a separate opaque disk with slits. A crank rotated both disks, and the lamp projected the image on the wall. For the first time, many people could view animation simultaneously. The motion picture projector was born, and, amazingly, the principle of its operation has remained fundamentally the same. The idea of "persistence of vision" was based on a belief that, when a single image forms on the retina, it remains there briefly, even if the viewer's gaze shifts to another object. When the next image arrives, scientists speculated, it somehow blended with the retained image to create the illusion of a continuous stream of moving imagery. Indeed, persistence of vision is real. When you close your eyes, the image of shapes and colors seems to remain visible briefly, especially if the scene was brightly lighted. But modern researchers do not think that this phenomenon, called "retinal afterimage," explains cinematic vision. Instead, they have described two other phenomena that begin to explain how a succession of still frames creates the appearance of fluid movement: critical flicker fusion and apparent motion or, as it also is called, stroboscopic motion. Together, these phenomena -- not yet completely understood by scientists -- make movie action seem as fluid as real life. Critical flicker fusion refers to the fact that discrete light flashes will merge into an impression of continuous light if they are spaced closely enough. Fluorescent lights, for example, turn on and off 60 times a second. They are off as long as they are on. But, of course, they appear to provide continuous illumination. We see discrete light impulses as continuous light because of the eye's inability to process rapid changes in brightness. Without this fusion, Plateau's audience would have noticed pauses between the images from his phenakistoscope. In the early days, filmmakers accepted Plateau's frame rate of 16 frames per second as the magic number. More frames per second would not make the motion seem any smoother. But with hand-cranked cameras, achieving a steady rate was not easy. Typically, the camera took pictures at a rate of 12 to 15 per second. The film was projected at a steady 17 frames per second, which meant that silent-film stars such as Charlie Chaplin seemed to be speed walking. The speedup was made far worse when old films were projected through later machines that ran at 24 frames per second. Even at 16 fps, however, the flicker problem remained, jittering between light and dark. This gave rise to "flickers," an early slang term for movies that is recalled today when we refer to a movie as a "flick." The problem improved partially when "talking pictures" arrived in the late 1920s. The sound track, recorded on the same strip of film in a narrow lane to one side, was coordinated most easily with the picture at 24 frames per second, and this rate became what remains today's standard "sound speed." But still there remained a noticeable flicker. It was eliminated by an ingenious solution -- a shutter that opens and closes three times while a single frame is being projected before the image changes. In other words, the projector positions one frame in front of the bulb and projects its image continuously while the shutter opens and closes, sending the image to the screen, blocking it and sending it again before the projector pulls the next frame into position. This is done by an opaque disk with holes spinning in front of the film. At 24 frames per second, image and blackness alternate 72 times in one second. In fact, for slightly more than half of the time we spend watching a film, the screen literally is blank, and we are sitting in total darkness. Research has established that the precise frequency at which flicker disappears depends on the intensity of the light, its wavelength, duration, the relative area of the visual field illuminated, the part of the retina stimulated and the viewer's age. Taking these variables into consideration, one can safely say that, in the typical motion picture viewing situation, flicker fusion -- the point at which flickering disappears -- occurs at about 50 flashes per second. The 72 fps flicker rate of film is well beyond the capacity of any viewer to notice shifting frames. Although critical flicker fusion allows us to watch a film without noticing the shift from light to dark and back, the separate phenomenon of apparent motion is necessary to make us believe that the still images are moving, not jerking or jumping from frame to frame. "Apparent motion" allows the eye and mind to connect the stroboscopic succession of still frames. Among cognitive scientists, conventional wisdom once was that a psychological event called the "phi phenomenon" provided this mental bridge between frames. According to this interpretation, minds somehow fill in the gaps between the still images. The history of the phi phenomenon can be traced to the early 20th-century Gestalt school of psychology. The basic idea of Gestalt is that the whole of anything is what we perceive rather than a collection of parts. The word Gestalt is used in modern German to mean the way a thing has been gestellt, or put together. In 1912, Max Wertheimer, the Gestalt psychologist, discovered that, when two lights were flashed side-by-side at a certain interval, viewers perceived not two flashing lights but a single light that moved. Wetheimer contended that the light's apparent leap from one position to another is a paradox that the brain must resolve. It is simpler for the brain to conclude that the light is moving than to conclude that two different lights are coordinated so that, when one is on, the other is off. Film theorists who used Gestalt theory argued that the viewer might be using a similar process of unconscious thought to create the illusion of movement. It was thought to be something akin to hallucinating the intervening motion. Examples of the brain's apparent urge to simplify ambiguity include several images well known to students of perception: the reversing cube and the old/young woman, among others. {See illustrations on facing page.} In these instances, the visual system is presented with two incompatible sets of information. In the same picture, one cannot, for example, see the old woman and the young woman at the same time, though all the information for each of those interpretations is contained within the ambiguous figure. The viewer can only alternate between seeing an old woman and then a young woman. These figures illustrate a "winner-takes-all" strategy of the human perceptual system. But, like "persistence of vision," Wert-heimer's "phi phenomenon" is thought to be an incomplete explanation of cinematic vision. Recent experimental work suggests that our visual perception is merely deficient but purposefully so. Research shows that the way in which we see movies is not necessarily different from the way in which we see real motion in nature. In his recently published book, The Reality of Illusion, cognitive researcher Joseph D. Anderson writes, "The visual system simply fails to detect the real difference between the successive changes in the static frames of a motion picture and the continuous changes of natural motion." Anderson explains that, while the brain is sometimes compared with a computer, its cells process information much more slowly than do silicon chips. To compensate, the brain has developed various strategies. One is a kind of data reduction or compression -- a shortcut -- that enables the brain to keep up with incoming information about events in the world, such as the movement of objects. It is the brain's natural use of such shortcut strategies that opens the way for filmmakers to introduce synthetic motions that the brain cannot distinguish from real movement. The illusion of cinema provides a valuable look into the intricacies of our perceptual system. Though it accepts imprecision, that system obviously is good enough for our needs, better than good enough, in fact, when it comes to movies. This visual quirk allows us to believe in magical realities on the walls of dark rooms. When "Godzilla" comes to town next summer, however, you may want to calm your fears with the knowledge that the monster is not really moving on the screen -- your mind is just compressing data. Mark Rambler is a freelance writer based in New York. CAPTION: Like a crank-operated Rolodex, this 1901 device played movies by flipping cards, each bearing a still photograph. CAPTION: When movies were new, the subjects were simple. Here, frames from a sneeze movie, filmed at Thomas Edison's studio in 1894. CAPTION: France's Lumiere brothers invented the Cinematographe, a combination movie camera and projector in 1895. CAPTION: Scientists once thought we see motion in movies because the brain automatically adopts reasonable interpretations of ambiguous images. Here are examples. Two faces of the cube can be seen to project but not both at once. Only the old woman or the young woman can be seen at a given moment. CAPTION: The 1887 "praxinoscope" used lamplight and mirrors to project a movie strip.