In January, the true colour of the Universe was declared as somewhere between pale turquoise and aquamarine, by Ivan Baldry and Karl Glazebrook at John Hopkin's University in Baltimore Maryland.
They determined the cosmic colour by combining light from over 200,000 galaxies within two billion light years of Earth. The data came from the Australian 2dF Galaxy Redshift Survey at the Anglo-Australian Observatory in New South Wales, Australia.
Glazebrook now says the true colour this data gives is closer to beige. "I'm very embarrassed," he says, "I don't like being wrong."
The mistake was caused by a bug in the software Glazebrook had used to convert the cosmic spectrum into the colour the human eye would see if it was exposed to it. "There's no error in the science, the error was in the perception," says Glazebrook.
Glazebrook has now teamed up with Mark Fairchild of the Munsell Color Science Laboratory at the University of Rochester in New York, who pointed out his mistake last month.
Fairchild realized the software Glazebrook was using actually took a slightly pinky looking colour as white. "There was a huge green shift due to the erroneous white point," he says.
When this was corrected, the colour was actually on the pinky side of white, a slight beige colour. Fairchild now hopes to test the software model by generating an exact replication of the cosmic spectrum light and shining it into someone's eye, so they can experience the true colour of the Universe. But Glazebrook is confident this time. "It won't change again," he says.
The John Hopkins group were using the cosmic spectrum - not the subjective colour it corresponds to - to trace the history of star formation in the Universe. Their scientific results are not affected by their mistake.