This is a fascinating topic, but the explanation requires a significant correction right at the start: The idea that medieval stained glass flows downward like a liquid is actually a myth.

While it is one of the most persistent and popular stories in materials science and history, modern physics and chemical analysis have thoroughly debunked it.

Here is a detailed explanation of where the myth came from, why it persists, and the actual science behind medieval glass.

1. The Origin of the Myth

The "flowing glass" theory is based on a visual observation. Visitors to old European cathedrals often noticed that the individual panes of stained glass were thicker at the bottom than at the top.

From this observation, people drew a logical (but incorrect) conclusion based on the definition of glass. Glass is an amorphous solid. Unlike metals or crystals, its atoms are not arranged in a neat, repeating lattice structure; they are jumbled together like a liquid that has been frozen in place.

Because glass lacks a crystalline structure, observers theorized that it was actually a "supercooled liquid"—a substance that flows, but incredibly slowly. They believed that gravity, acting over 500 to 800 years, had pulled the glass downward, causing it to pool at the bottom of the frame.

2. Why the Theory is Incorrect (The Physics)

If glass were truly a liquid capable of flowing at room temperature, even slowly, we would see evidence of this in other ancient glass artifacts. However:

• Ancient Roman and Egyptian Glass: We have glass vessels and jewelry from ancient Rome and Egypt that are thousands of years older than medieval cathedral windows. These items show no signs of sagging or deformation. If medieval glass flowed in 800 years, Roman glass should be a puddle after 2,000 years.
• Telescope Mirrors: If glass flowed, the precise mirrors of large telescopes would warp over time, rendering them useless. This does not happen.
• Viscosity Calculations: Physicists have calculated the viscosity of glass at room temperature. For glass to flow perceptibly, it would take a time period longer than the age of the universe. While glass technically has a viscosity, at room temperature, it is effectively solid (around $10^{20}$ Poise—for reference, water is 0.01 Poise).

3. The Real Explanation: Medieval Manufacturing Techniques

So, why is the glass thicker at the bottom? The answer lies in how the glass was made.

Medieval glassblowers did not have the technology to make perfectly flat, uniform sheets of glass (known as "float glass," which was invented in the 1950s). Instead, they used two primary methods:

The Crown Glass Method

1. A glob of molten glass was blown into a hollow sphere.
2. The sphere was punctured and spun rapidly (like pizza dough) while still hot.
3. Centrifugal force caused the glass to flatten into a large disk.
4. Because of the spinning, the glass was naturally thicker near the center (the "bullseye") and thinner at the edges.

The Cylinder Glass Method

1. Glass was blown into a long cylinder shape.
2. The ends were cut off, and the cylinder was sliced down the side.
3. It was then unrolled and flattened in an oven.
4. Gravity often caused the glass to be uneven as it cooled.

The Installation Process: When the glaziers cut these uneven disks or sheets into panes for a window, one end of the pane was almost inevitably thicker than the other.

When installing the glass into the lead strips (cames) that hold the window together, it makes structural sense to put the heavier, thicker side at the bottom. It provides better stability. While not every pane was installed this way (some are actually thicker at the top or sides!), the majority were placed heavy-side-down, creating the illusion that the glass had "slumped" over time.

4. What Does Change in Medieval Glass?

While the glass doesn't flow, it does change over centuries due to chemical weathering. This is often confused with physical flowing.

• Leaching: Rainwater, which is slightly acidic, reacts with the alkali ions (like potassium and calcium) in the glass. Over centuries, these ions leach out, leaving a silica-rich gel layer on the surface.
• Pitting and Crusting: This chemical reaction can cause the glass to become pitted, opaque, or develop a crust. This is not the glass moving; it is the glass corroding.

Summary

The "unexpected discovery" that glass flows is actually a case of confirmation bias. People saw a physical phenomenon (uneven glass) and applied a theoretical model (glass as a liquid) that seemed to fit.

However, the reality is that medieval stained glass is a solid. The variation in thickness is a frozen record of the manufacturing process—specifically the Crown and Cylinder methods—and the logical choice of craftsmen to install heavy bottoms for stability.