Where an animal lives has a huge effect on its subsequent evolution. Of course there are the obvious environmental factors than can act as agents of selection on animals - like being adapted to extreme temperatures - but there is also the larger effect of parsing out the land into biomes.
Map of the world's biomes
Biomes are areas defined by a particular environmental variables. Think "deserts" or "forests". I happen to live in a fairly unique biome here in southern California. The area along the coast of California, from San Francisco down into the northern parts of Baja Mexico, is classified as part of the Mediterranean biome. As you might guess, it's named after the environment typically found in countries around the Mediterranean Sea. (I had a colleague who complained once that going to Italy wasn't very exciting because it looks just like back home.) It also occurs in patches on the southeast coasts of South America, Africa, and Australia. This biome is characterized by warm/hot and dry summers, and mild (rarely freezing) winters when all precipitation occurs. The reasons why this type of environment occurs around the world, and tends to be on the southeast corner of continents, is complex, and has a lot to do with water currents.
Distribution of the Mediterranean climate
The interesting thing about biomes is that they act as self-contained ecosystems, with many animals endemic to a particular biome (or a particular patch of biome if it's found through the world like the Mediterranean biome). You can think of them as cages. Animals and plants tend to stay in their biome, because they are adapted to that particular environment, and its biotic (other species) and abiotic (climatic) variables. This has important implications for many ecological factors, like competition, migration, and most importantly, speciation.
Populations that are isolated from other in their species, due to geographic or ecological barriers (like biome borders), become genetic islands. This will cause the populations to diverge over time. Part of this divergence will be due to adaptation to local environments. So if the populations occupy different environments, with different climatic variables, competitors, predators, or resources, they will adapt in reaction to those differences. More importantly, even if there aren't environmental differences, populations that are separated from each other will tend to diverge just by random processes. Unique, random mutations will accumulate in each population, causing them to diverge.
The key is, if the divergence is large enough, the populations may become so distinct that they can no longer reproduce, and would become new, separate species. The barrier to reproduction may be because the individuals in each population don't recognize each other as potential mates, or their offspring may not be viable or fertile, or they may just never come in contact with each other. Whatever the case, these new populations will now evolve independently of each other.
Seeing this sort of process in extinct species is tough because the fossil record is so incomplete, but we can make some inferences about it. One of the ways we can is to look for the sort of barriers that promote speciation, and inferred effects of theses barriers. A recent featured paper in the Journal of Vertebrate Paleontology looks at one such case.
Linda Tsuji and her co-authors describe a new species of late Paleozoic (~275 million years ago) reptile. It's a member of a group called pareiasaurs, which were large, herbivorous reptiles, that may include the ancestors of turtles. This particular taxon is called Bunostegos akokanensis
, and had been named before on the basis of more fragmentary material from the country of Niger in Africa. The authors describe new material that includes a nearly complete skull, which allowed for investigation of this taxon's relationships to other pareiasaur taxa.
The skull of Bunostegos akokanensis
The most interesting thing about this analysis was that B. akokanensis
has many unique characters that distinguish it from its close relatives, but that it shares with much more distant relatives. This means that these characters were evolved convergently, with B. akokanensis
and its distant relatives evolving the characters independently. Why might this have happened? Because of where B. akokanensis
lived, of course. During the Late Permian, most of the land mass in the world was all crammed together in one giant super-continent, Pangea. In the center of this humungous land mass was an area isolated by its environment, i.e., a paleo-biome. This area was dry, possibly similar to areas in the American West today. This isolated the taxa there, including B. akokanensis
, and it went off on its own evolutionary path. The isolation doesn't explain the later convergence in other taxa, which may just be an evolutionary coincidence, but it does explain why this taxon is so different from its nearest relatives.
This is one of those great examples where the fossil record allows us to join evolutionary and ecological studies. In this case, the ecology drove the evolution, as it often does, and we have good data on the climate and subsequent evolution of one taxon that was affected by this environmental isolation.