In 1843, John Bennet Lawes and Joseph Henry Gilbert planted a wheat field in Hertfordshire, England, with the intention of never stopping.

The Broadbalk experiment — named for the field where it was established — is still running today.

180 years of continuous data. The same field. The same crop. Different treatments applied to different strips of land, year after year, generation after generation.

It is the longest continuously running agricultural experiment in the world.

Who They Were

John Bennet Lawes inherited the Rothamsted estate in 1834 at the age of sixteen.

He was not a farmer by temperament. He was a chemist by inclination. He converted part of his estate into a laboratory and began experimenting with plant nutrition — specifically, with the question of what soil amendments actually improved crop yields.

Joseph Henry Gilbert was the chemist Lawes hired to collaborate with him in 1843.

They worked together for fifty-seven years.

Their partnership produced one of the most significant bodies of agricultural data ever assembled.

What Broadbalk Tested

The Broadbalk experiment was designed to answer a specific question: what does a wheat crop actually need to grow well?

Different strips of the field received different treatments — some combinations of nitrogen, phosphorus, potassium, and other minerals; some organic manure; some left without amendments entirely.

Year after year, the yield of each strip was recorded.

The data accumulated.

After ten years, patterns emerged that short-term experiments could never have revealed. After fifty years, the patterns deepened. After a century, the experiment had captured events — droughts, disease outbreaks, atmospheric changes — that no planned experiment could have included.

The unplanted strips showed what happened to soil fertility under depletion. The manured strips showed how organic matter built and maintained productivity. The mineral-amendment strips showed what chemistry alone could and could not do.

What the Data Revealed

The Broadbalk data showed several things that challenged the prevailing views of agricultural science.

Mineral amendments alone — nitrogen, phosphorus, potassium — could sustain yields in the short term. Liebig's framework appeared to be confirmed.

But over decades, the soil under mineral-only treatment changed.

Organic matter declined. Soil structure degraded. The biological activity that Winogradsky had described weakened.

The yields held up. But the soil that produced them was becoming something different from what it had been.

The manured plots told a different story.

Organic matter was maintained. Soil structure remained. The biological community remained active. Yields were lower in peak seasons than the heavily fertilized strips — but more consistent across bad years.

The data had captured the difference between optimizing for yield and maintaining for resilience.

It was in the numbers. 180 years of them.

The Research Station Model

Rothamsted established a model that spread across the world.

Agricultural research stations — dedicated to long-term, systematic study of crops, soil, and farming practices — were established in the United States, Germany, France, and beyond in the following decades.

Each brought the experimental method to agriculture. Each generated data. Each contributed to the scientific framework that guided farming practice into the 20th century.

The research station model was not without problems.

It optimized for controlled conditions. It focused on measurable outcomes. It favored the crops and practices that performed well in experimental settings — which were not always the crops and practices that performed well on real farms, under variable conditions, managed by people with limited resources.

But it established a standard of evidence.

Agricultural claims could now be tested.

Not every claim that was tested turned out to be right.

That was the point.

Still Running

The Broadbalk experiment continues at Rothamsted Research in Harpenden, Hertfordshire.

The data from 1843 to the present is archived and available to researchers.

What Lawes and Gilbert began as a question about wheat nutrition has become a record of agricultural history written in soil chemistry, yield data, and the slow accumulation of evidence across spans of time that exceed any individual career.

Lawes planted the field.

He did not live to see most of what it would reveal.

That is what long-term thinking looks like.

Next in this series: two paths diverge — when chemistry and biology chose different roads, and how the agricultural choices made at that fork are still playing out in every field today.