Claims that Mahmoud Ahmadinejad’s re-election was rigged have been bolstered by an analysis of the official results by British academics In the conservative provinces of Mazandaran and Yazd, the number of votes cast exceeded the number of eligible voters, the survey by researchers from the University of St Andrews and Chatham House, the London think-tank, found.

• Four more provinces recorded turnouts close to 100 per cent.

• To achieve the official results in 10 of the 30 provinces, the ultra-conservative President must have carried all the new voters who did not cast ballots in 2005, all the votes that went to his centrist rival Ali Akbar Hasemi Rafsanjani and up to 44 per cent of the votes that went to reformist candidates.

• Those provinces include ones dominated by ethnic minorities who seldom if ever vote conservative. “The numbers from Ilam, Lorestan and Hormozgan almost defy belief,” said Thomas Rintoul, one of the researchers.

• Lorestan is home to Medhi Karoubi, the most liberal of the four candidates, who won 440,247 votes (55 per cent) there in 2005. Official figures suggest he won only 44,036 (4.6 per cent) this time.

• “The analysis shows that the scale of the swing to Ahmadinejad would have had to have been extraordinary to achieve the stated result,” said Ali Ansari, Professor of Iranian Studies at St Andrews.

• The figures also challenge the notion that Mr Ahmadinejad’s victory was due to the massive participation of a previously silent conservative majority and that he was particularly popular in rural areas.