Doing some research recently on African-American voting patterns, I came upon this map of American ancestry from the U.S. Census (pdf):
And it reminded me of a mindblower of a political explanation that I meant to share here.
This is long, so meet me after the jump...
Looking at the map above, you notice the swath of African-Americans (purple) running through the southeast. Something I hadn't heard of but is apparently relatively common knowledge is that that pattern in the population is referred to as the Black Belt.
That may seem a little coarse, but it actually refers to the color and richness of the soil.
I looked for maps of soil color, but if such a thing exists, I wasn't able to find it. The clearest picture of soil distribution matching that pattern was this map of "soil orders" suggesting ultisols and/or vertisols having something to do with that color:
This National Science Foundation lesson on soil orders offers a more detailed version and settles the question.
Vertisols are definitely black. (pdf) (Ultisols, not so much (pdf).)
Our friend, the google, shows how that pattern manifests today in the form of farms making use of that rich soil that comprise that lighter colored swirl.
Farms are actually the point, because while Black Belt may have been a reference to black soil, that's not to say the Black Belt doesn't also have racial meaning. Pretty much every source one checks cites this explanation from Booker T. Washington's 1901 autobiography Up From Slavery:
...The term was first used to designate a part of the country which was distinguished by the color of the soil. The part of the country possessing this thick, dark, and naturally rich soil was, of course, the part of the South where the slaves were most profitable, and consequently they were taken there in the largest numbers. Later and especially since the war, the term seems to be used wholly in a political sense—that is, to designate the counties where the black people outnumber the white.
That "political sense" Washington refers to includes an electoral sense as well. Slave-descendant voters in Black Belt counties leave a blue Democratic voting stripe through otherwise red states, seen especially vividly in this New York Times county map of the 2008 election results:
I also ran into this voting pattern described as "the cotton vote." As data became available from the 2008 election, aligning a map of the blue strip of Obama-voting Black Belt counties with a map of cottom production from 1860 (!) revealed a remarkable correlation:
I already think that's mindblowing, but that's not even the mindblowing part. The mindblowing thing is that what's really responsible for this phenomenon of modern politics is the still-forming North American coastline of 100 million years ago.
From Deep Sea News earlier this summer:
"During the Cretaceous, 139-65 million years ago, shallow seas covered much of the southern United States. These tropical waters were productive–giving rise to tiny marine plankton with carbonate skeletons which overtime accumulated into massive chalk formations. The chalk, both alkaline and porous, lead to fertile and well-drained soils in a band, mirroring that ancient coastline and stretching across the now much drier South. This arc of rich and dark soils in Alabama has long been known as the Black Belt."
Behold! Your late Cretaceous coastline and future Democratic strongholds:
Oh, what? You've still got some unblown mind left? I have a little more.
The map above represents 75 million years ago. Dr. Ron Blakey of Northern Arizona University actually offers us several maps in the late Cretaceous range. But why would that time period be particularly relevant? I find two explanations. One is that the Cretaceous was a boom time for the sort of plankton that would eventually become the Black Belt.
The other explanation is that the Cretaceous ended when, 65 million years ago, an asteroid (or asteroids) slammed into the earth, right across the future-Gulf of Mexico at the tip of the Yucatan Peninsula. Not only did the impact and resulting fallout from that asteroid kill the dinosaurs, it also wiped out huge quantities of marine life, including many of the "tiny marine plankton with carbonate skeletons" (I'm guessing some version of Coccolithophore? Anyone?) that would become the rich soil that slaves would farm on land their ancestors would inhabit in voting districts that would favor Democratic candidates around the turn of the second millennium of the Common Era.
Below is a map of the Chicxulub Crater, the crater left by the asteroid 65 million years ago, showing its location at the end of the Yucatan Peninsula. It looks like a topographical map, but actually it's a Bouguer gravity anomaly map. The best explanation I could find for gravity anomaly maps is from this Earth Observatory page from NASA.

I think the idea is that the impact created ridges of higher density, which show up as gravity anomalies. As ever, any insights and expertise you can offer on anything in this post is greatly appreciated.
Also, the more I researched this, the more I ran into people who'd already done portions or variations of it. I tried to include as many links as I could to previous work. Credit also to Allen Gathman for the cotton vote connection.











Another interesting thing about that late Cretaceous map: that's your baseline worst-case global warming map, too, because temperatures during the late Cretaceous are about what's forecast for 20 or so years from now. So if you were wondering what would be under water assuming Greenland and Antarctica melt, the answer is basically "most Republicans."
Why no Alaska in these maps (except for the Cretaceous coastline map)?
The discussion in http://deepseanews.com/2012/06/how-presidential-elections-are-impacted-by-a-100-million-year-old-coastline/ is very good on this subject.
The map that shows the Cretaceous shoreline (where lowlands underlain by Cretaceous marine shales and limestones ring the Appalachian uplands) can be found at
http://deepseanews.com/wp-content/uploads/2012/06/Al-Ga-Ms-SC-1.gif
and compare that to percent slaves at
http://deepseanews.com/wp-content/uploads/2012/06/PctSlaves1860UK.gif
The specific soils that match are udalfs and uderts (humid deciduous forest soils and humid shrinking-swelling clay soils). Both types are darker than the typical red soils of the area, and are also more alkaline and richer in nutrients, and are great for growing cotton. They have clays that retain cations much better than the clays in the typical red ultisols of the southeast.
Your soil maps are not detailed enough to show the relevant soil distributions. For a close comparison to the specific soil types, check udalf and udert distribution in
http://soils.usda.gov/technical/classification/orders/alfisols_map.html
http://soils.usda.gov/technical/classification/orders/vertisols_map.html
http://soils.cals.uidaho.edu/soilorders/percentalf.gif
http://soils.cals.uidaho.edu/soilorders/percentvert.gif
A minor quibble dealing with speaking of time; the classic one-off error.
When speaking of units of time as in "the turn of the century/millennium" "turn" refers to the start of the period, not the end. Therefore the reference to the period of 2000-now, to be accurate, needs to be "the turn of the _THIRD— millennium of the Common Era," not "the turn of the _second— millennium of the Common Era."
The third Millennium did not actually start until 12:00:01am on January 1st, 2001, being the first second of the first day of the new year, century and millennium. December 31st 2000 was LAST day of the old year, century and millennium. When speaking of complete centuries and millennia, like the birthday analogy one person used (wrongly as it turned out) one does not say that period is over until the entire period has expired. You are not 18 years of age until AFTER the 18th anniversary of your birth, not DURING the 18th year of your life.
When speaking of decades, other criterion is used; the names of the years included. Thus the 90's included the years 1990-1999. This naming of decades is where most people get confused as to what century it actually is. If this were accurate then we should be in the 20th century now and the last century was the 19th. The 20th century was the period of 1901-2000, the 19th was 1801-1900, the 18th from 1701-1800 and so on.
Not that it really makes much difference, except to pedantic linguists like myself. Scientists for whom the nomenclature makes a difference are well aware of the proper terminology regardless of popular usage. Calls by some in the general populace to make the first century be comprised of only 99 years are, well, in a word; illogical and have no basis in facts.
I liked this article, if this is what you do, I want to work for you. Here is an article by John Fraser Hart, a geography professor from the University of Minnesota. He makes a case similiar to yours.
Hart, John Fraser. 1977. The Demise of King Cotton. Annals of the American Association of Geographers. vol. 67 (3).
http://www.jstor.org/discover/10.2307/2562332?uid=3739568&uid=2129&uid=2&uid=70&uid=4&uid=3739256&sid=21101002218813
NRCS Map of Vertisols http://soils.usda.gov/technical/classification/orders/vertisols_map.html
Here is some info on classifying soil by color and other nomenclature, at least this is what we did on the South Dakota State University Soils Judging Team. http://www.swac.umn.edu/classes/soil2125/doc/s3chap1.htm
Anyone who's been to Georgia knows the soil there is mostly red, which indicates a large concentration of iron (oxidized). Same holds true for northeastern Cuba. This, I suppose indicates that native indians are somewhere lost in the shuffle?
mind-blowing, Steve, but "correlation is not causality." I think you went a little too far when you wandered off into Jurassic Park.
So...Politics is a dirty business?
(I see I am not the only one with too much time on my hands)
Something is wrong with your census map. Everybody in Ohio is German? I remember a lot of blacks and Polish in Cleveland. Something happen since the 70's? Everbody in southern Arizona is Mexican? Talk to Jan Brewer. Everybody in the south is black? Boy did their slavery policy backfire. Looks like junk to me. Only folks in Kentucky and Tennesee have been American for at least two generations? Now why did I fill out the census?
All the "Americans" live in Appalachia, from West Virginia to the center of Texas. How is that?
So, any guesses as to what effect global warming and the rise of the oceans will have on voting patterns 100 million years from now?