Crop Circles: What Happened to the Plants?

People can make large geometric designs in crops. Doug Bower and Dave Chorley claimed to have made hundreds; a 1992 competition demonstrated that teams could work convincingly in darkness. The visible pattern alone cannot establish an unknown origin. The more focused question is whether reported changes inside some plants and soil withstand independent testing.

The plant evidence

William Levengood reported swollen nodes, microscopic changes, abnormal seed development and small cavities in cereal stems sampled from formations. Nodes are the joints along a stalk. Mechanically flattened plants can turn upward again as they grow, and that process can lengthen their nodes. Levengood argued that certain measured changes were too large, too quick or too systematically distributed for that explanation alone. He reproduced some effects with brief microwave exposure, but this did not identify a source of energy in a field.

Near Blue Ball, Maryland, in 1995, Levengood and Nancy Talbott reported node-length differences between plants from two formations and outside controls. Cavities were reported in many sampled nodes inside the first formation and none in its controls. They described a patterned distribution of measurements across the formations. A board and rope explain flattened crop; the reported internal measurements require a separate, carefully checked explanation.

At Beckhampton in England, investigators sampled a spiral formation and standing crop beyond it. Several days later a new outer ring appeared where some of that standing crop had already been sampled. Their later analysis reported substantially greater node elongation in the new ring than in earlier samples from that area. It is a useful before-and-after comparison, though nobody instrumented the ring while it formed.

Models and materials

A 1999 formation near Hoeven, Netherlands, was associated with a disputed account of a luminous object. That witness account is not reliable proof. Eltjo Haselhoff instead fitted a mathematical model of an elevated electromagnetic point source to the pattern of node changes measured afterward. The fit did not detect an object or radiation during formation. A later mechanically made comparison circle produced some node elongation but a different distribution; it was made fourteen years later in another field.

At Edmonton, Canada, researchers reported unusual plant measurements and a small change in an X-ray diffraction peak from mica particles in formation soil. They said the soil change correlated with plant effects at corresponding sample locations. A clay expert found the laboratory technique sound but the physical interpretation difficult. Microwave exposure of control soil did not consistently reproduce the effect, and large independent replication did not follow.

What remains open

Mechanical construction is the strongest explanation for the geometry of most formations. Flattening, subsequent growth, plant age, moisture, local soil and sunlight can also affect samples. Wind can create simple circular lodging. These explanations deserve direct tests against the reported measurements rather than dismissal by assertion.

Directed energy, classified equipment or non-human intelligence are hypotheses, not findings established by these studies. A statistical model is not a detected machine. None of the investigations identifies a verified craft or independent maker.

The original research relied heavily on a small overlapping group of investigators. Sampling was often unblinded and began only after a formation was discovered. Chain of custody, control selection and statistical choices have been disputed. Those weaknesses matter, especially for the Edmonton soil result. The practical next step is preregistered sampling, blinded independent analysis and matched mechanical controls in the same crop and season.

The mystery is specific: are any reported internal and spatial effects reproducible beyond ordinary plant responses and sampling error? Until that test succeeds, the measurements are interesting reports with important limits, and the cause remains unproven.