Science's Most Embarrassing Mistakes

A fossil assembled backwards, a horse that could not do math, canals on Mars, and a radiation that three hundred scientists saw and that was not there.

Competence is not a defense. A respected paleontologist put a sea monster’s head on the wrong end and invented an order of reptiles to explain it. A generation of astronomers mapped canals on Mars that the planet does not have. A horse convinced Germany it could do arithmetic. The people in these stories were trained. The mistake was confident.

Piltdown man was a fabricated fossil that held a place in human ancestry for forty years. Elasmosaurus was rebuilt with its head on its tail. The canals on Mars were an optical hope, drawn and defended. Clever Hans was reading his owner’s face, not the arithmetic. Polywater was a new form of water that was mostly contamination. Cold fusion was announced before it could be repeated. N-rays were Prosper-René Blondlot’s radiation of 1903, reported in hundreds of papers, and absent from the apparatus once the prism was in someone else’s pocket.

The memory-eating worms belong with the results that would not repeat. Around 1960 James McConnell fed trained flatworms to untrained ones at Michigan and reported that the lesson had transferred. Other laboratories could not make the transfer happen again.

In 1826 Justus von Liebig learned that a bottle on his own shelf, labeled iodine chloride, was the new element bromine. Antoine Balard had already published it.

This shelf is those errors, not a catalogue of fraud. Some were hoaxes. Most were sincere. Start here for Piltdown man, and for the longer record of findings that had to be taken back.

← All clusters