That's a very tough question to answer. There's supposed to be an old Danish saying: it's hard to make predictions, especially about the future.
We can't really guess just by comparison with other mammal species (they usually go extinct without descendants in less than a million years) because humans are so widespread, adopted to an unusually huge range of environments. If we go extinct soon, it will probably be our own doing. Some species (say coelocanth fish) show little change for tens of millions of years, but they don't seem to have much in common with us either.
We can try to figure out what changes will occur by looking at how evolution works, but that doesn't help much either. Evolution makes some predictable changes because different organisms reproduce at somewhat different rates, depending on their genes. Since each generation then has a different genetic mix than the last one, in the long run changes occur in the types of organisms. Most variability in reproductive rates in people now doesn't have much of anything to do with relatively simple things like resistance to disease, running speed, etc. Instead, it involves complicated interactions between people and their societies. These change rapidly on the time scale of biological evolution. So predicting what will become of our species in a million years is nearly impossible. Let's just hope we make it through the next hundred.
The past is easier to keep track of, because there are fossils and DNA evidence. We evolved from other apes. Our last common ancestor with chimps was some 5 to 8 million years ago. The last common ancestor of us (people and chimps) with monkeys was long before that.
Even if our own evolution is too entangled with society to be predictable, some evolution that matters to us is fast and predictable. Smaller organisms, with short generation times, show very rapid evolution. As a result, insects quickly evolve resistance to pesticides, which then have to be used at far higher concentrations than when first introduced, until they finally become useless. More importantly, bacteria evolve resistance to antibiotics. Each antibiotic starts losing its usefulness as soon as it is used on a wide scale. Right now, more or less incurable strains of tuberculosis are becoming a serious problem world-wide. We do some unbelievably stupid things, like giving low levels of antibiotics to millions of livestock, as if we deliberately wanted to breed resistant bacteria. Also, as human society changes, with increasingly dense and interconnected populations, we provide a new medium for new types of bacteria and viruses to evolve to grow in. Human Immunodeficiency Virus (the cause of AIDS) is the most serious current example. It's a particularly fast evolver, too, which greatly complicates the search for vaccines and treatments. You can even keep track of how the strains of HIV evolve in a single person, as various drugs are given or as the immune system that it faces starts to collapse.
For an interesting novel that deals with both the evolution of our diseases and our descendants' evolution, you might want to read Vonnegut's "Galapagos". I won't spoil the ending by saying how it comes out, but perhaps you shouldn't take it too seriously as a scientific prediction.
Mike W (with Tom)
(published on 10/22/2007)