Zerohedge has some interesting maps of immigration patterns over the last century, showing the predominant country of origin of immigrants to each state in the U.S. The overall trend is toward a massive influx from Mexico, which is hardly news, but there are lots of surprises tucked in there. For instance, I never would have guessed that the largest flow of immigrants to New York State in 1910 would be from Russia. It's also surprising to see what a worldwide melting pot was going on a century ago, and how little of that there is now.
Switching gears to much older immigration patterns, I've been tempted to buy the new book by Brian Sykes, "DNA USA." I enjoyed "The Seven Daughters of Eve," which traced movements of peoples by examining their mitochondrial DNA. The new book is getting lukewarm reviews, though, and sounds like it's got a bit of interesting DNA data patched together with a rambling travelogue. So I'm hoping someone will publish a summary of the good stuff. One good source is Amazon reviews, which yield the following interesting snippets:
Native Americans descended from a handful of matrilineal (mitochonddrial) clusters that arrived in the New World between about 16,000 and 20,000 years ago. Three of the clusters are genetically linked to Siberians who originated in Central Asia. The fourth cluster is linked to a Polynesian strain that arrived in the Cook Island about 3,000 years ago, from Taiwan; it is absent among the Eskimos and concentrated in Central and South America. A fifth cluster is found in North America, but not Alaska. It appears not to have originated from Asia, but instead from Europe--not the 16th-century European wave but a population from 16,000 years ago. How did they get here? Presumably not overland, across Asia and then Beringia, or they'd show up in Alaska today, but it's hard to imagine an Atlantic crossing, either, not that early. That cluster seems like a real wild card.