THE decision to overturn Mark Dallagher's murder conviction is the latest in a series of cases featuring contentious "expert" evidence.

Dallagher, 28, a small-time burglar, was jailed for life in December, 1998, after a Leeds Crown Court jury found him guilty of smothering Dorothy Wood, 94, at her Huddersfield home.

He had allegedly left a unique earprint as he listened at a window before breaking in.

But the Appeal Court quashed his conviction in July, 2002. Lord Justice Kennedy ruled that new expert opinion on the reliability of earprint evidence should be tested at a re-trial.

At the time, police had hoped that the apparently successful conviction of Dallagher - based on his earprint evidence - would provide the foundation of a world-leading database of ear impressions.

Police aimed to spread knowledge of the part that earprints, just like fingerprints, could play in linking a suspect to the scene of a crime.

A month after Dallagher's conviction, more than 1,200 ears had been logged on to the database at the National Training Centre for Scientific Crime Investigation at Harpeley Hall, Co Durham.

The hope was to build a comprehensive research tool to back up evidence that no two ears are exactly the same. The database was built up using volunteers among trainees at the centre.

Police in the UK had begun looking seriously at the use of earprint evidence in 1996.

Essentially, earprints are like fingerprints.

The cartilage and contours of every ear give it a unique shape.

It was considered another tool in the forensic toolkit, but police now rely largely on DNA.