In their one-hundred-plus year history, wildlife and natural history films have always been filmed in wild or natural settings. Neither have they always depicted completely natural behavior. Harsh conditions, expensive film stocks and equipment, and most of all the uncertain behavior of animals, have driven many wildlife filmmakers to long for the control enjoyed by their colleagues in feature films. It was this lack of control, in fact, that led many critics to dismiss wildlife films unfairly as being amateurish. Clearly, there is nothing amateurish about today's wildlife filmmakers, but even those most determined to film authentic wild behavior in the natural settings have found themselves forced by competitive pressures to shoot close-ups, missed shots, dramatic angles, and subtle bits of behavior in enclosures, zoos, laboratories, television studios, and now, finally, in cyberspace. The second century of wildlife films may well be the one in which the wild and often unpredictable behavior of some of the most awesome and ferocious creatures can be made to conform to any scripted scenario, brought under control at last--by a mouse. Yet well before the age of computer generated imagery, indeed, from the start there have been those wildlife filmmakers who have tried to make animal behavior conform to audiences' preconceived notions, and to the dramatic conventions of mass market entertainments. Many have resorted to manipulations more severe than the digital kind. In the 1930s, for example, a Barnumesque American named Frank Buck achieved notoriety with a trio of wild animal capture films, Bring 'em Back Alive (1932), Wild Cargo (1934), and Fang and Claw (1935). The popularity of these with audiences rested largely on the dramatic way in which they proved widely held notions that animals in the wild exist in a state of constant interspecies warfare (a never ending arms race, as a more recent wildlife film put it). Buck's specialty was staging dramatic fights in small enclosures between animals who would normally have avoided each other in the wild, but who reacted to shared confinement by attacking each other with audience-pleasing ferocity. Armand Denis, who directed the second Buck outing, later recalled his reaction to Buck's plan for one such a battle. How's about a fight to the death between a tiger and an orang-utan? Buck asked. Well, I said cautiously, occurs in Borneo and Sumatra; there are tigers also in Sumatra, so it is that an orang-utan and a tiger could meet--but surely if they did, they'd just avoid each other. Animals don't normally fight to the death for nothing. Don't they, eh? replied Buck. When I'm around they do. (1) In the world of mass market entertainments, where wildlife films still reside today, accuracy continues to rest uneasily on the same slippery slope. That is, in Denis's phrase, if it is not inconceivable that something could occur, then, according to the logic of film and television it probably has occurred, or could very well at any moment. The fact that extraordinary circumstances would be required to produce its occurrence is irrelevant. The implied conclusion is, simply, that such things do occur. Every climactic rescue, epic gun battle, and apocalyptic explosion in film and television thus carries with it an implied, if muddled, statement of probability: such things could occur, therefore such things do occur. This is, in a way, the basis of film and television realism. Realistic drama has long been seen as the meeting point between fantasy and truth, but today it seems ever more to be engulfing documentary, journalism, and even natural history exposition. The problem is especially evident in the recent six-hour BBC production Walking with Dinosaurs (1999), which aired in the US in April, 2000, on the Discovery Channel. This high concept production combined the conventions of wildlife and natural history films with digitally produced dinosaur images of the sort seen in Jurassic Park (1993), The Lost World (1997), and Disney's Dinosaur (2000). …