On this page
On 24 July 1837, Louis Agassiz opened the annual meeting of the Swiss Society of Natural Sciences in Neuchâtel and told the room that giant glaciers had once spread across Europe. NASA's Earth Observatory says he "was met with sheer rage from a scientific community well-established in its views." By the 1870s, the same NASA page notes, the Ice Age theory was widely accepted.
Below: his clues, and a four-step method you can borrow.
What clues did Agassiz use to show Europe had an ice age?
Agassiz did not spot the clues first. A Val de Bagnes farmer, Jean-Pierre Perraudin, was one of the first to suggest that boulders had been carried by glaciers; he told the engineer Ignace Venetz, and Jean de Charpentier took the idea further (Alpes Vaudoises history of the glacial theory). Agassiz met Charpentier and Venetz at Bex in 1836 (biography of Louis Agassiz).
His 1837 evidence, as listed by the history site Debating Glacial Theory, was erratics, "diluvium", striations, polished rocks and moraines. NASA defines the two most famous:
Glacial erratics are rocks carried by glaciers and dropped on a different type of rock.
Striations are long parallel grooves on exposed rock, gouged by debris dragged along the bottom of a flowing glacier.
His key move was comparison. He set these features beside those found near present-day glaciers, and argued against the rival explanations, drawing on reasoning from Charpentier and Schimper: floods, mudslides, or boulders rafted on ice floes and icebergs (Debating Glacial Theory).
How did Agassiz convince other geologists?
Slowly. Leopold von Buch and J. A. Deluc pushed back at the meeting itself (Debating Glacial Theory). So Agassiz went to the ice. He lived for a time in a hut on one of the Aar glaciers to study how the ice moved, and in 1840 published the two-volume Études sur les glaciers (biography of Louis Agassiz).
In September and October 1840 he toured Scotland with William Buckland, president of the Geological Society of London, and showed him the traces of ice. Buckland wrote, "We have found abundant Traces of Glaciers round Ben Nevis," as George White recounts in the Journal of Glaciology. On 4 November 1840 Agassiz read a paper to the Society titled "On glaciers, and the evidence of their having once existed in Scotland, Ireland, and England".
Even so, the Edinburgh Geological Society notes that it took nearly twenty years for his ice-sheet theory to be generally accepted. For other rock readers of that era, see how Adam Sedgwick used rock layers to define the Cambrian period and how Charles Darwin explained coral atolls with sinking islands.
What do scientists know about the last ice age today?
Modern dating supplies the timeline Agassiz lacked. A 2009 study in Science used more than 5,000 radiocarbon and cosmogenic surface-exposure ages to map peak ice extent (Clark and colleagues, Science). Ice sheets grew to their maximum positions between 33,000 and 26,500 years ago, and Northern Hemisphere melting began 19,000 to 20,000 years ago, triggered by stronger northern summer sunlight (USGS summary of the study).
Agassiz's record has a dark side. In 1850, as a Harvard biologist, he commissioned photographs of enslaved people in South Carolina to support polygenism, the false, racist idea that races had separate origins (Al Jazeera). Careful reasoning in one field did not stop grave error in another.
How can you reason from evidence like Agassiz?
His method works on an exam paper as well as on a hillside.
Describe before you explain. Write what you see with no cause attached: a boulder of one rock type on bedrock of another; long, parallel scratches.
Find the cause at work today. Agassiz checked his clues against living glaciers. Where does this process happen now, and does it leave the same marks?
Test every rival against every clue. Could floodwater carve long parallel grooves in hard rock and polish it too? An explanation that misses a clue is weaker.
Find a second clue, then a new place. One clue can mislead; several together are hard to explain away. Agassiz took his idea to Scotland.
A worked example from the physics lab: your pendulum's period comes out longer than the formula predicts. Step 1, record exactly what you measured. Step 2, recall where the formula holds: small swings, length measured to the bob's centre. Step 3, list rivals: swing too wide, length measured wrongly, a slow stopwatch thumb. Step 4, change one thing and see which rival survives.
If step 3 is where you freeze, Fermi's AI tutor notices where you are stuck and decides whether to explain, ask or show. For practice, the Fermi Problem Library has worked problems with a step-by-step solver that explains the reasoning rather than printing an answer.
For another thinker who read landscapes, see how Ibn Sina explained mountain formation.
