R v Fleur [2004] All ER (D) 434 is a case concerning the adduction of statistical evidence in criminal trials.


The appellant was convicted of conspiracy to sell class A drug. There were 2 co-defendants in addition to the defendant B (Manchester) and S (London). B is a friend of the appellant and a long time heroin user. On 2/11/'02 B drove to London and met up with S and obtained heroin before returning to Manchester. On the 5/11/'02 B met up with F before making another trip to London and meeting S who gave B 11.1kg of heroin £1.16m in value.

When he was caught by the police on the way back to Manchester an additional 0.5kg of heroin was found in his house which matched the chemical composition of the heroin found in his car, the prosecution attempted to argue that the 0.5kg of heroin was from the trip to London on the 2nd.

Traces of heroin were found in both of F's cars, which F claimed to have been second hand cars and may even have came from a homeless man that travelled in it who was a heroin user.

Second Ground of JudgmentEdit

Hooper LJ notes that both expert witnesses conceded that there were no statistics showing how rare it is to find heroin traces in car sweepings. according to statistics Almost all bank notes had traces of cocaine and 17% of all public transport seats has traces of heroin.

His lordship believed that the way in which the heroin evidence was dealt with by the trial judge was far from satisfaction, because in summing up the trial judge effectively invited the jury to disregard the agreed position of both expert witness i.e. no statistical basis for determining number of cars effected by heroin; also no guidance was given as to how that evidence should be used.

Hooper LJ thought:

  • The evidence was inadmissible
  • The direction given by the trial judge was a misdirection

The appeal was ultimately dismissed. Because the court concluded that it was unlikely that much emphasis was placed by the jury on the specific evidence.