A molecular biologist testifying for the defence in double murder-accused Mark Lundy's retrial says forensic science uses lower testing standards than other scientific fields.

Lundy, 56, has pleaded not guilty to bludgeoning to death his wife Christine, 38, and the couple's seven-year-old daughter at the family's Palmerston North home in 2000.

His High Court retrial in Wellington is in its fifth week and on Wednesday his defence team recalled one of its expert witnesses who had earlier contested a key piece of forensic evidence.

Stephen Bustin is a professor of molecular medicine in the United Kingdom.

This week he said RNA analysis by a Dutch forensic scientist which infers the presence of brain tissue in a stain on Lundy's shirt was fundamentally flawed.

Laetitia Sijen from the Netherlands Forensic Institute had testified tests on the sleeve sample showed "it's more probable than not that it's central nervous system tissue".

The court has been told the shirt was found inside-out in a suit bag in Lundy's Ford Fairmont by police examining the car days after his wife and daughter's bodies were discovered on August 30, 2000.

One stain was found on the left sleeve, the other on the shirt pocket.

A New Zealand forensic scientist has already testified the stains on the shirt matched Mrs Lundy's DNA.

Dr Bustin said even though he is not a forensic scientist, his extensive research in molecular medicine meant he could assess the methodologies used in RNA analysis.

See more here:

Forensic standards queried in Lundy trial