All of the larger groups of animals are seen to have evolved a wide range of life forms. Reptiles, for example, having evolved in late Paleozoic time out of small or medium-sized predatory amphibians, developed into a great diversity of animals that included herbivores and carnivores, dwarves and giants, creepers, runners, climbers, burrowers, swimmers, and flyers, all in competition with other groups of animals living in the same ways. They have continued to occupy some of these functions (niches, in ecology) throughout their history and have vacated others. Reptiles are still among the more effective small creeping and running predators on land (lizards, the tuatara), replaced by birds as aerial dwellers and by mammals as large terrestrial herbivores and predators. A succession of animals have been marine "superpredators." In the Triassic and early Jurassic periods, this was the domain of ichthyosaurs; in late Jurassic and early Cretaceous time, these were replaced by plesiosaurs; in the Cretaceous, by mosasaurs; in the Eocene, by the zeuglodont whales (mammals); and since the Miocene, by the great carcharodont sharks. The fossil record suggests that evolution has resulted mainly from the tendency of all species to experiment with new ways of living, to thereby exploit new opportunities as they arose in an ever-changing world, gaining a new foothold in one place and losing one in another.
With Charles Darwin's theory of evolution in the mid-1800s came the expectation that the fossil record would provide unbroken evolutionary sequences, in which species after species would be seen to emerge gradually from their ancestors and pass, equally gradually, into their descendants. Most species, however, are seen to appear abruptly, to maintain their typical form for most of their history, and to vanish as suddenly as they appeared. This failure to trace coherent lineages of ancestors and descendants does not prevent recognition of changes in larger groups of animals: horses are seen to have developed through Cenozoic time from small to large, from five-toed to one-toed, and from short-to long-toothed, but complete records of the transition from one species to another have not been found. This caused some paleontologists to doubt Darwin's belief that evolution proceeds by the gradual accumulation of small changes. It has now been determined that evolution does not proceed steadily, in response to some mysterious internal force, but in response to new opportunities. In a stable, unchanging environment, a well-adapted species is not likely to change, whereas in a changing one it may find better opportunities by changing its way of life. In addition, evolution does not normally occur throughout a species but in interbreeding populations, occupying some small part of the geographic range of the species as a whole. It is such populations and adjacent populations, linked by exchange of genes, that deviate from the ancestors and from the species as a whole, to form races or subspecies, and, eventually, species. At any one time, a species is a combination of such groups, diverging episodically from one another. As geography and habitat change, these groups shift about, either blending when they meet, abruptly displacing one another, or coexisting side by side in different ways of life.
One of the most difficult problems in evolutionary paleontology has been the almost abrupt appearance of the major animal groups - classes and phyla - in full-fledged form, in the Cambrian and Ordovician periods. This must reflect a sudden acquisition of skeletons by the various groups, in itself a problem. Paleontologists are not certain whether the soft-bodied forms of the Precambrian Ediacaran fauna are in fact ancestral to modern groups. In any case, the lack of well-documented animal remains in older rocks indicates that differentiation of the major groups occurred more rapidly than did their subsequent evolution.