Besides the fact that "North" implies "cold," and "South implies "hot," I was struck by the fact that Lincoln won absolute majorities (in a four way race) in 16 states wholly or partially above the 41st parallel (the coldest states), while winning only one "hot" state (California), south of that line.

Although they were "slave" states south of the Mason-Dixon line, the border states of Maryland, Delaware, Kentucky, and Missouri are "cold" states (to be defined shortly). And they all went for the Union.

Also "cold" were one state, West Virginia, that broke away from the Confederacy, and one wannabe "East Tennessee." Although they are in Southern latitudes, their location in the Appalachians makes them "cold."

For the purposes of the question, "cold" means "colder than Richmond Virginia" (basically the coldest part of the Confederacy) in January (as a litmus test). That would be an average of about 40 degrees Fahrenheit.

A "region" is at least one third or so of any one of the 33 states of the United States on November 6, 1860 (Lincoln's election). A state may be divided East, Middle, and West on an east-west axis or north, middle and south on a north-south axis. E.g. East Tennesseee, West Virginia, North Alabama.

California, a "hot region" (whole state, actually) was exceptional in being pro Union, rather than pro Confederacy.

Apart from California were there any "hot regions" that were pro Union and had little or no interest in slavery? Conversely, were there any "cold regions" that were strongly pro Confederacy/slavery? Is there something I have overlooked in constructing this hypothesis?

  • P.S. "Hot" means "as hot as Richmond, Virginia, or hotter."
    – Tom Au
    Commented Apr 10, 2013 at 1:25

3 Answers 3


As a resident of Loudoun County, Virginia, and the owner of two houses that changed hands multiple times during the war, I find it far more reasonable to tie the variation not to climate, but rather to the crops grown in said climate. (John Monroe’s house is less than 3 miles from my own, and I live on Bull Run Mountain — yes, that Bull Run.)

Virginia gets every bit as hot and humid as Florida (worked there), Texas, and, oddly enough, most of the Eastern Seaboard — as far north as New York and Massachusetts. While our winters are a bit milder, there isn’t that much difference between Pennsylvania, New York, and Virginia.

What is different is the topography — and this even shows up within the state. Downstate — Tidewater, Richmond, etc. — the soil is much more conducive to Tobacco and other crops that are best exploited in quantity. The Plantations of the James River illustrate the fact, and dot the rivers. In order to harvest the vast quantities of cereal crops, you need a large, cheap labor force — ideally free.

In contrast, Loudon and Fairfax are far more hilly, and more conducive to smaller farms — small dairy, light industry, and the like. Not surprisingly, Loudon was far more divided about secession. Interestingly, nearby Alexandria was a center of the slave trade, and not surprisingly, heavily anti-union. And, as you point, the mountainous areas just one county over had no slaves and even less incentive to leave the union.

This pattern can actually be extrapolated throughout the east coast. From roughly the Mason Dixon line down, there is a large swath of flat land between the Appalachian and the Atlantic, that lends itself to agriculture. Georgia, the Carolinas, Texas, and elsewhere are (or more accurately were) highly productive farmland. Beginning with Pennsylvania, the abundance of cheap energy (hydro, coal, etc.) gave comparative advantage to light industry — better suited to non-slave holding societies.

Indeed, the Southwest was never anticipated to go slave, because again, the land did not comport itself as such. And, Kansas / Nebraska, the scene of much fighting even before the war, is very, very cold.

  • +1 Very good answer. A clear upgrade over my "naive" model.
    – Tom Au
    Commented Apr 11, 2013 at 13:24
  • You're slightly off about Kansas and Nebraska. The latter didn't join the U.S. until AFTER the war. "Bleeding" Kansas joined the "Union" in 1861, but in fact, very nearly went for slavery and the Confederacy.
    – Tom Au
    Commented Apr 11, 2013 at 13:28
  • @TomAu - The "very nearly" was mostly a result of active campaigns by external pro-slavery forces. Once they actually had a statewide vote of residents, banning slavery was approved by a 2 to 1 margin.
    – T.E.D.
    Commented Apr 11, 2013 at 13:35
  • @T.E.D.: You just remined me that there was a reason I specified winter "January" temperatures for "cold" in my question. It's "hot" (in summer) "everywhere," including NYC. But it was the "cold" that seemed to attract anti-slavery types (or discourage pro slavery types) in the border states. Meaning that the determining factor seemed to be "the coldest cold," not "the hottest hot." Maybe it's a case that if too many slaves die of pneumonia in January, slavery is no longer viable.
    – Tom Au
    Commented Jan 5, 2014 at 15:15
  • Ohio gets just as hot, and nearly as humid as the Deep South in summer.
    – Oldcat
    Commented Dec 5, 2015 at 0:57

The slaveholding regions in Missouri are along the Missouri river line about two thirds of the way up from its southern border. St. Louis was very loyalist, and is slightly south of that. The state of Kansas is directly west of that and because of the crises in the 1850s was very loyalist. The hilly regions in the south west of the state were also not strongly inclined to secede.

This region in Missouri is right in line with the center of Illinois, Indiana and Ohio and has the same weather patterns, if not politics and crops.

The 'thermocline' explanation of slavery was originally an attempt by Northern Democrats like Douglas to assuage the furor in their constituents at the abandonment of the Missouri compromise with the claim that it just didn't matter. It didn't convince most of them and led to the growth of a party with a firm commitment to containing slavery, the Republicans.


Is there something I have overlooked in constructing this hypothesis?

Affable Geek's very correct answer has already explained that the analysis of 'climate' presented is seriously flawed, and that climate itself was not the determining factor.

But aside from that, without AG's or some similar explanation, there is no apparent explanation at all as to why "climate" itself should somehow be linked to the regional split between "North" and "South". Do we find such a thing historically? Is there some reason why people who sweat more in the summer or freeze more in the winter should bind themselves together politically?

Although the word hypothesis technically might support the question's proposition, without some sort of explanation and historical substantiation regarding how and why "climate" might be linked to political configurations, IMO what has been overlooked in this hypothesis is... the hypothesis.

  • I'd expand AG's answer to say that "climate," in combination with other factors such as soil and power sources affected the desirability of cash crops tied to the slave trade. And the emphasis on "trade" explains why western Tennessee (on the Mississippi River) was much more pro-slavery than similarly temperatured East Tennessee (in the Appalachian mountains).
    – Tom Au
    Commented Sep 29, 2013 at 17:36

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.