Finished Guns, Germs, and Steel. Good book, if more than a little repetitive. He's fond of setting up each of his chapters as a mystery story, but by the end of the book, there's not very much suspense, since the whole damn point of the book is that the same set of environmental and geographic factors can explain lots of migrations and patterns of contact in human history. Better package of domesticable crops. Better package of domesticable crops. Better package of domesticable crops. The butler did it. The butler did it. The butler did it. Over and over and over again.
That said, it is one of those neat books that give you a couple new ways of thinking about things. There are certain principles he articulates -- at great length -- that are useful clip-on lenses to have available. It certainly has made me think about natural selection in a different fashion. But there's also a really interesting and extended discussion on the nature of long-term historical evidence, the kind of stuff you use to understand what happened back before people had the technology to actually write down what was happening. Some of the experiments to fix things in place and time are just wonderful -- understanding historical forest patterns from pollen deposits, sifting through the forensics of bones and ancient garbage heaps, doing cladistics and backwards induction on contemporary languages to reconstruct extinct tongues and figure out their vocabularies in order to understand what crops and technologies they had. It's also a really striking lesson in how much the world has changed even in the last ten thousand years, what kinds of wild biodiversity it has supported and how radically some of the most prominent plant and animal species have readapted as humans increasingly affected their environment.