Contracting Earth theory (early 20th century)

In the early 20th century the prevailing wisdom regarding how mountain belts were formed and why the sea is deep was that the Earth started out as a molten blob and gradually cooled. When it cooled, heavier metals such as iron sank down and formed the core, while lighter metals such as aluminum stayed up in the crust. The cooling also caused contraction and the pressure produced by contraction caused some parts of the crust to buckle upwards, forming mountains. Other parts of the crust buckled downwards, creating ocean basins. Picture in your mind a grape turning into a raisin as it dries out.

The contracting Earth hypothesis can be visualized by imagining a grape (early molten blob Earth) drying out and becoming a raisin (today). This hypothesis makes a couple of predictions that don’t in fact stand up to observational data, so it doesn’t work to explain Earth’s present distribution of surface topography.


The contracting Earth hypothesis was further refined by introducing isostasy. Isostasy is the concept that all elements in a system are in hydrodynamic equilibrium or trying to get there. For example, if you have a bathtub of water, a chunk of balsa wood floats higher than an ice cube because balsa wood is less dense than ice. If you were to push the balsa wood down, it would pop back up when you took your hand away. The popping back up is the balsa wood bringing itself back into equilibrium. It happens very fast because water has low viscosity. Now what if you had a bathtub full of molasses instead of water? When you push the balsa wood down, it will indeed rise back up again after you take your hand away but it will happen more slowly because molasses is more viscous than water.

What does this have to do with the Earth? Well, the pre-continental drift idea went like this. Heavy parts of the crust sank down and lighter parts raised up not only due to the pressure of contraction but also due to isostatic adjustments. The interior of the Earth was thought to be a viscous fluid that could accommodate this sinking and rising. This was the proposed mechanism favored by paleontologists who thought the reason identical fossil species were found on continents separated by oceans was because there had been connecting land bridges that sank.

Now, in fact it is true that isostasy does govern mountain elevations. In fact, most mountain belts have a “root” like the keel of a boat and over long timescales the mantle in fact does flow viscously, but the mantle is solid rock, not a fluid. Land bridges did not sink down into the mantle. That part is wrong.

Introducing the concept of isostasy explained the deep roots of mountain belts but it also included the incorrect idea that the ocean between continents used to house land bridges that had since sunk.

Enter Alfred Wegener

So, this is where Wegener comes in. He had a PhD in astronomy but most of his scientific contributions were in meteorology. He became very well known in his own lifetime as an explorer of Greenland and as a meteorologist. In fact, he died in 1950 while leading an expedition across Greenland.

His interest in geology was basically a sidelight to his regular academic career and he had no training in geology. He assembled circumstantial evidence for his idea that the continents had once been joined. Let’s examine that circumstantial evidence.

Jigsaw puzzle fit

Alfred Wegener was not the first person to notice that the continents fit together across the Atlantic Ocean. In fact, in the 1500’s and 1600’s when reliable maps of the east coasts of North and South America were produced, this feature was obvious. I’ve always thought that this piece of evidence has a little bit of Western ego attached to it. For example, if you look at a map centered on the Pacific Ocean, do you notice anything? No, not really, because there is a complete absence of anything that looks like a jigsaw puzzle fit there.

Rock types and geologic structures

If you fit the continents back together the way the “jigsaw puzzle fit” suggests you should do it, then you’ll see that rock ages, rock types, and mountain belts match up across the boundaries between continents (sketch below). In fact, other scientists have likened this to taking two halves of a newspaper torn lengthwise and fitting it back together so the sentences can once more be read across the tear.

The Archean Cratons and Proterozoic Curest

Rock types and ages both match across the Atlantic Ocean.
E. Richardson

Fossil evidence

Fossils of terrestrial plants and animals identical to each other were found on continents now separated by water. Could seeds be dispersed across an ocean by wind, water, or animal activity? Could animals that don’t look like swimmers (see the Lystrosaurus below for example) get across an ocean some other way? Some people suggested the sinking land bridge idea, but we’ve already discussed the mechanical problems with that model. Others suggested mats of vegetation that could have drifted across the ocean carrying plants and animals to a new continent.

Fossil evidence of: the Mesosaurus, the Lystrosaurus, the Glossopteris, and the Cynognathus in different continents.

Fossils of identical creatures were found on either side of oceans.
E. Richardson

Evidence of ancient climates

Under the assumption that the kinds of climates that would form particular rock types and structures are similar today as they were millions of years ago, we can infer the past climate of a locality from studying its geology. If, for example, you find evidence of glaciation in a place that is now temperate and not ice-covered (such as Southern Africa), you are left to infer that the climate of the whole world was different, or else that Southern Africa was a lot farther from the equator when that glaciation happened than it is today. Alfred Wegener made a lot of contributions to these types of observations since meteorology and glaciology were his fields. Upon fitting the continents together as he proposed (sketch below), glacial striations found on now-separated land masses even looked like they radiated out from a common source.

Ancient climate including tropics,desert, and glaciers.

Paleoposition of continents in the southern hemisphere with regional climate matches.
E. Richardson

Statistical Analysis of Topography

Remember back at the top of the page when we were examining the raisin? I said that with a raisin, even though the surface of the raisin is all wrinkly, the highs and lows of the surface are likely to be normally distributed about some mean value. If the contracting Earth hypothesis were true, one observation it predicts is that Earth’s topography would be normally distributed about some mean also. We can check this out! When we do, we find that Earth’s topography is not normally distributed, which leads us to the important conclusion that continental crust and oceanic crust are fundamentally different from each other (see my screencast explanation below). The realization that oceanic and continental crusts are different from each other was a huge leap towards figuring out sea-floor spreading and plate tectonics.

Analysis of Earth’s topography leads to the conclusion that oceanic and continental crust are fundamentally different from each other.

Lack of mechanism for continental drift

Circumstantial evidence is just not enough! Each piece of Wegener’s evidence was dismissed at the time because he couldn’t come up with a physical mechanism that would work to move continents laterally apart from each other. I also think no small part of it is that scientists don’t necessarily love it when outsiders to the subdiscipline come in with a novel idea. (True 100 years ago, true today.) Since Wegener was trained as a meteorologist, many geophysicists were skeptical of his ideas right from the start.

Wegener’s idea that continents plowed across the ocean floor couldn’t work, so many geophysicists also dismissed his circumstantial evidence.

Conceptual sketch of plate tectonics

Cross-sectional sketch showing two continents driven apart laterally by sea-floor spreading between them. Oceanic crust is consumed at subduction zones (at left in the sketch).

Why so slow?

I think it is interesting to consider how long the plate tectonics revolution took to unfold. Consider that Wegener published “Origin of the Continents and Oceans” in 1915 in which he laid out the circumstantial evidence that indicated the continents had once been joined, but plate tectonics was not accepted as a theory until about 1968