Cause of Ice Ages: geographic pole wander
A distinction must be made between Earth's angular momentum vector (or Earth's instantaneous spin axis, which because Earth is nearly spherical, lies very near its angular momentum vector) and the geographic pole. The geographic pole could wander from Greenland, to Brazil or anywhere, with no change in Earth's angular momentum vector or spin axis. It would be like a spinning billiard ball that shifts so that the spin axis remains the same, but a different part of the billiard ball lies at its pole.
Though astronomers agree that the Milankovitch cycle is correct (according to Newtonian celestial mechanics) leading astronomers immediately objected that the orbital eccentricity and axial tilt changes involved in the Milankovitch cycle, are far too small to account for Ice Ages, without invoking extreme and dubious positive feedback mechanisms. The Milankovitch cycle might synchronize with Ice Ages not because it causes them, but because both are entrained by some yet unknown physical effect.
Hapgood noticed that Pleistocene sediments and fossils told the story, not so much of a huge ice sheet expanding from the Arctic Ocean to Chicago, but of an ice sheet migrating away from Siberia and into Canada, with a pole at Hudson's Bay. Hapgood noticed evidence that there have been several different poles during the Pleistocene, at various places all now north of 45deg latitude. Though Hapgood thought that Earth's crust slid around while most of Earth didn't change in its rotation, I think the best evidence indicates that the entire Earth shifted so that a new point was under the (constant) north pole in space.
George Darwin and other experts noted that Earth's equatorial bulge confers huge gyroscopic stability, but this assumes that Earth is absolutely solid. Really, as the geographic pole shifts, the bulge can shift. Even if Earth were an unfractured solid, the tiny change in curvature needed for even a 90 degree change in the geographic pole, might be feasible. Instead of curving down 4000/300 mile in 6000 miles, the surface would have to curve up the same amount. The change in height (measured from the tangent) over a distance of one mile would be only
4000mi/300 * 2 /6000^2 = 1mm
Just because the geographic pole is almost constant now (and for at least the last 4500 years, because the Great Pyramid is only about 6 arcminutes off of true north in its alignment), doesn't mean it always was. Mars' geographic pole seems almost constant now too, but the water erosion on Mars suggests sudden melting of icecaps and permafrost, which could have been due to sudden shifts in Mars' geographic pole. The discrepancy between the geographic and magnetic poles of Earth, Uranus and Neptune, could be because the magnetic poles need more than a few thousand years to catch up with the geographic poles, when the entire planet shifts, not just the crust. Indeed Earth's magnetic pole assumes a position roughly intermediate between Earth's present north pole and the Hudson's Bay geographic pole theorized by Hapgood for the latest Ice Age.
