A HERALD investigation has uncovered evidence that forensic materials used to convict John Cooper for two double murders may have been contaminated in a police station flood.

The startling revelation was not mentioned in court and could explain why fibres and DNA were present on the various items which ultimately led to the conviction of John Cooper for the crimes.

Cooper was found guilty in 2011 for the shooting of farmer Richard Thomas, 58, and his sister Helen, 54, in Milford Haven in December 1985. He was also found guilty of the murder of Peter and Gwenda Dixon, who were gunned down on the coastal path about 600 yards from Howelston Caravan Park in Little Haven in June 1989.

But the fact that exhibits from two separate murder scenes, Coopers home and from various other scenes were all under water together at Milford Haven Police Station will now throw doubt on the forensic evidence.

In fact, the police station deluge now goes some way to explain why a pair of shorts, which were so essential in the case, were clean when collected from Cooper’s home but were filthy dirty when shown in court.


Further Police documents examined by The Herald also show countless inconsistencies in exhibits’ storage and an incomplete chain of evidence.

The prosecution did not refer to the flood at Milford Haven Police Station that damaged exhibits held there in storage.

Operational notes referring to the flood and other material relating to the exhibits’ storage were not disclosed to the defence at John Cooper’s 2011 trial for the 1985 murders of Richard and Helen Thomas and the 1989 murders of Peter and Gwenda Dixon.

John Cooper on Bullseye
Scoveston Manor was set ablaze
Richard and Helen Thomas of Scoveston
The coastal path near Little Haven
Gwenda and Peter Dixon
Does a flood explain the change in appearance of the shorts?

The possibility of cross-contamination of material stored in Milford Haven was never revealed, despite being recorded in police operational logs, copies of which The Herald now holds.

We can confirm that among the material caught in the flood were branches from the screen used to conceal the bodies of Peter and Gwenda Dixon.

Those branches were never used in evidence because of the damage caused to them by the flood. However, material stored in the same container as the branches WAS used in evidence.

Among those other items was a blood-stained belt worn by Peter Dixon and the sock removed from Richard Thomas’s body in 1985.


The careless storage of the materials also throws into sharp relief how exhibits from different offences were combined at police stations across Pembrokeshire.

The Herald has an operational note that states exhibits were transferred from an unspecified location for forensic examination in London on July 9, 1989, and returned on July 13.

The note states the exhibits were stored at a garage in Haverfordwest before being recovered from a secure cell at Fishguard Police station almost seventeen years later.

The operational note does not cover where the exhibits spent the intervening seventeen years, in which conditions they were stored, and how and why they ended up at a separate Police Station.

That vital break in the chain of evidence leaves serious questions about the circumstances in which all of the exhibits came together for re-examination as part of Operation: Ottawa. That investigation led to John Cooper’s 2011 murder convictions.

As The Herald reported a fortnight ago, in 1998, exhibits – including some produced at trial – were placed on public exhibition by the Police.

The idea that cross-contamination could not or did not take place is nonsensical, and that was before we knew a flood occurred.

The failure to accurately schedule the exhibits held in storage at different locations leaves the strong possibility that items relied upon by the Police to convict Cooper of both double murders were mixed.

Items secured from John Cooper’s home during Operation: Huntsman, which led to his burglary conviction in 1998, were intermixed with articles relating to the Thomas and Dixon murder inquiries.

Some of those items were in the same flood at Milford Haven Police Station.

And that throws a question mark over the forensic evidence used to convict Cooper.


It also explains why so many vital exhibits had to be re-bagged after being stored in the same paper sack for years.

There is almost zero forensic evidence connecting John Cooper with the murders of Richard and Helen Thomas.

The evidence was fibres from Richard Thomas’s sock found in shorts’ pockets recovered eleven years after his death.

However, the Thomas siblings were shot in 1985.

The sock fibres were recovered from shorts the Police accept were not made until 1986 and were only available for sale in the USA from 1987.

Cross-contamination of the material exhibits is the only explanation for how fibres from a pair of socks recovered in 1985 ended up in shorts purchased two years later and half a world away.

Police operational notes confirm the weapon recovered from John Cooper’s home, which the prosecution said he slew the Dixons with, was NOT the weapon that slew the Thomases.


The Herald holds a copy of an email from DI Lynne Harries to Forensic Expert April Robson.

In the correspondence, dated March 13, 2009, DI Harries explains he has been informed by Ottawa’s Crime Scene Manager that branches taken from the scene of the coastal path murders were stored alongside Peter Dixon’s bloodied belt when a flood occurred.

The email sent from DI Lynne Harries to forensic expert April Robson dated March 13, 2009

Although no exact date for the flooding is referenced in the documents, we know that the shorts were checked in at the police station on September 21, 1999. The items were dealt with by a scenes-of-crime officer in March 2000.

The flooding had caused damage to the packaging, and we have the document to prove that at least 165 items of evidence had to be re-bagged.

MET Office data shows exceptional rainfall in west Wales in December 1999, of up to 136.4mm.


The onus for providing a chain of evidence, each made of solid links showing custody of exhibits and their care at every stage, rests with the Police.

The operational notes, which were not disclosed to John Cooper’s defence team, lead inevitably to questions about the safety of the evidence depended upon to convict him.

If the CPS or prosecuting barrister knew of the operational notes’ capacity to undermine critical forensic evidence, it would have changed how they presented the case against Cooper.

Similarly, as the material would have to be disclosed to the defence team, probably on a schedule of unused material prepared under the Criminal Procedure and Investigations Act, the defence would almost certainly have approached its task differently.

A contemporary newspaper report states: “Dr Avenell (one of the Prosecution forensic experts) said he and DI Glyn Johnson had reviewed the entire records of the exhibits, how they were stored and who handled them, and he was “completely satisfied” that blood had not been transferred from one item to another.”

But the operational notes disclose a confused picture, incapable of being reconciled without information not contained in the working notes.

There is no indication that Dr Avenell – or DI Johnson – were ever told of the flood affecting the exhibits in 1999, let alone how they ended up at a variety of different – potentially unrecorded locations – over the twenty-one years between the Thomases murders and Operation Ottawa’s start in 2006.

We can never know because there is no evidence the CPS and prosecuting barrister were ever told of the exhibits’ tortured history, and – as they were not – the defence was not.


We asked Dyfed-Powys Police the following 6 questions:

1: When was the evidence storage area at Milford Haven Police Station affected by a flood?

2: How long was the evidence exposed to flood water?

3: How many crime scenes had evidence affected by this flood?

4: What items were affected alongside GWJC/90 (Peter Dixon’s belt) and WDGJ/28 (branches from Dixon murder scene)?

5: Were all 165 items repackaged by SOCO Greenish and a work experience student between March 8 – 10, 2000, all contaminated by flood water?

6: Why were this flood and the damage it caused not mentioned by DI Glyn Johnson or Dr Phillip Avenell when they were testifying about feeling “completely satisfied” that no contamination could have taken place at John Cooper’s 2011 trial?

Dyfed-Powys Police did not answer our questions and did not refer to the flood that affected exhibits.

Instead, they copied and pasted their response to us from last month, saying once again: “All relevant material relating to the crimes John Cooper was convicted of, was provided to the CPS in accordance with the CPIA at the time of his trial in 2011.

“There is a formal process for appeal, which Mr Cooper has previously explored, and we will always comply with any official review made by anyone who believes they have been wrongly convicted.

“It is not appropriate for Dyfed-Powys Police to comment on specific facts of this case which has been through a judicial process.”

We asked the Crown Prosecution Service, “Did the CPS fail to disclose information regarding the provenance of vital exhibits to John William Cooper’s trial in 2011?”

The CPS did not reply.

Significantly, no officers or experts were questioned about the flood affecting the exhibits during John Cooper’s 2011 trial.

Not even by the defence.

In a case in which forensic evidence played such a large part, that beggars belief.

The flood is not mentioned in SIO Steve Wilkins or Jonathan Hill’s book ‘The Pembrokeshire Murders’, which is described on the sleeve as the “fascinating true story”.

Fascinating? Yes.

The truth, the whole truth and nothing but the truth? Certainly not.