On January 8 1997, a 51-year-old woman named Marion Ross was found stabbed to death in her home in Kilmarnock, a town 30 miles south of Glasgow. The police went in, took photographs and dusted for prints. They made their inquiries. Then came an odd development: on January 14, police found a "mark" on a bathroom doorframe inside the victim's house - a mark made since the murder. It was later formally identified as the thumbprint of police officer Shirley McKie.
McKie was part of the team investigating Ross's murder. She had stood outside the house, but she had not been part of the unit that entered the crime scene. The fingerprint could only mean one thing: that McKie had illegally entered the house some time after the killing.
McKie, who was then 35, denied that it was her print. She also denied ever having set foot inside the victim's house. But no one, of course, believed a word of it: they all knew that fingerprints do not lie.
That one fingerprint from Ross's house went on to cause an awful lot of trouble, for an awful lot of people. It destroyed McKie's promising career in the police, it threw doubt upon the investigation into Ross's murder, and it has now very seriously undermined the reputation of fingerprint science in Scotland.
Because, it seems, it was not actually McKie's fingerprint. In February this year, McKie, now 43, was awarded a compensation payout of £750,000 from the Scottish Executive after she was wrongly accused of leaving the print. Calls for a public inquiry into the case are now mounting by the day.
"I was in the police for 36 years," says Iain McKie, Shirley's father. "I always believed fingerprints didn't lie. Now ... I know different."
Iain, who is 67, lives in a flat in Ayr, on Scotland's south-west coast, overlooking the Irish Sea. He used to serve with the CID so he "knew what happened in the police", and initially, although he says that his daughter "wouldn't tell lies", he questioned her denial of having been inside Ross's house. "I was saying to her, 'Look, Shirley, if they say your fingerprint was found in there then it must be your fingerprint ...' They had it checked and they had it checked. You never argue with that - people have been hung on a fingerprint."
Iain McKie says that it is normal practice to rule out any "innocent" prints from officers involved in a case simply by comparing prints at the scene to photographic records kept of officers' prints. In this case, Shirley's prints turned out not to have been taken earlier on in her career, so investigating officers asked her to give her prints for elimination purposes. She duly gave her prints - and one of them ended up matching the print on the doorframe.
But McKie stuck to her story. She had never been inside the victim's house and therefore hadn't left that print.
In May 1997, Shirley repeated this story at the trial of a 20-year-old joiner named David Asbury, who had been accused of murdering Ross. But despite her throwing doubt on to the whole issue of fingerprints at the scene - and despite the fact that fingerprints were used as evidence against Asbury - he was convicted and sentenced to life imprisonment.
McKie had stuck to her guns but, as it turned out, her troubles were only just beginning. Within a month, she was off work, suffering from anxiety and depression. The police, she says, wanted her to change her story and admit that she had entered the murder victim's house and accidentally left a print, and they were piling on the pressure. Challenging the Scottish Fingerprint Service, a branch of the Scottish Criminal Records Office (SCRO), was simply unheard of.
McKie was sick and off duty until the following March, when she was suddenly arrested, taken to a station, strip searched, and then charged with perjury for her part in the Asbury trial. "They got the dawn chorus organised: they phoned her up, she answered the phone and there was nobody there so she hung up, and then they went around and got her there," says her father. "Then they took her into the police office where, in 1992, I had been in charge ... She got bail and came home and lay on the floor weeping. You'll understand that I became very, very protective of my daughter at that stage."
Iain and the family began fighting back. But the first QC hired by the family simply did not seem to believe the fingerprint service could be wrong. The second QC, Donald Findlay, also initially had his doubts, but took on the case.
Crucially, at this stage, everyone - including Iain and Shirley McKie - still believed that the print found at the murder was hers: the issue for them was, how did it get there? Had McKie, either before or after the murder, without any recollection, once been inside the premises? Or had she touched a doorframe - again without any memory of it - that had ended up inside the house?
"We wondered if she'd once visited a joiner's yard and touched a plank of wood that ended up inside Marion Ross's house," says Iain. "Or maybe she'd sleepwalked herself there? Except that meant dodging the 54 police who had guarded the place [after the murder] and never seen her inside. Or - believe it or not - she wondered herself if she had been abducted by aliens or something. I mean, nothing was out of the question."
Out of the blue, when McKie was surfing the internet looking for fingerprint experts to help her with her case, she came across an American expert named Pat Wertheim, a respected professional who had trained FBI personnel. Within a few months he was in the UK with a colleague named David Grieve, inspecting the McKie print that had been found inside Ross's home. He reached a shocking conclusion: the print was not Shirley McKie's.
Later on, when McKie stood trial in 1999 for perjury, three SCRO experts claimed that her print had indeed been found inside the victim's home. Iain says: "The two Americans - Wertheim and Grieve - came in and showed the court that it wasn't Shirley's print at the murder scene. They won the day and Shirley was found unanimously not guilty and she was congratulated by the judge for the way she comported herself, which was almost unheard of."
The two US experts brought in by McKie's defence team argued that the "mark" found on January 14 1997 inside Ross's house and then identified by SCRO experts on February 10 as belonging to McKie bore no resemblance to McKie's actual fingerprint, with one saying the "obvious" differences took only "seconds to see".
In television dramas there is often a scene where a computer is fed a print and then spits out a name. In reality, the fingerprint identification process is a lot clumsier. Forensic police officers attend a crime scene and use fine black powder to dust surfaces for prints. When they find a mark, it is usually quite clearly visible to the naked eye. This mark, when finally identified as a human print, is then compared to a print from whoever is being checked out or investigated. The prints are photographed and placed side by side for visual inspection by experts. Experts try to identify 16 points of similarity in each print.
Computers sometimes play their part - the Automatic Fingerprint Identification System does help police authorities worldwide to narrow down their search by eliminating hordes of unlikely matches - but eventually, in nearly all cases, final identification will still come from a human expert visually inspecting two sets of prints. According to the Americans, whoever matched up the print at Ross's house with McKie's prints was simply wrong.
Later, in 2002, it was alleged that there had, in fact, been a fundamental and significant division of opinion about the identification among the SCRO's own experts in February 1997 - at the same time as the SCRO were telling the murder squad detectives that the print was definitely McKie's.
Following McKie's acquittal for perjury in 1999, the family hoped for an apology from the chief constable, but it did not come. "That really, really got her down and she became more ill after that," says Iain. "There was anxiety and pressure because of the stress of it all. She was at [the] stage where if she saw a police car she'd start to physically shake. She even planned her own suicide."
After talking to Iain McKie, he and I travel across Ayrshire to meet Shirley herself. She lives in the nearby coastal town of Troon and works in a little gift shop there. He warns me that she is strict about timekeeping and, indeed, when we meet Shirley, she seems angry that we are late.
I ask her what her reaction was to a parliamentary statement made recently by Scottish First Minister Jack McConnell, who claimed that the SCRO's identification of her print had been an "honest mistake" and "that [it] was accepted by all the [parties] involved" and everyone had "moved on".
"I was amused at how the most senior minister in Scotland can stand up, knowing what there is to know about the case, and say what he said," she says. "It bemuses me and that's turned to disgust."
In the same week as that McConnell statement, McKie, who was about to go to court to sue the police for damages, had accepted a £750,000 settlement from the Scottish Executive.
McKie says she detests anyone thinking that she has "won" anything. She says that accepting the offer felt like "copping out".
"You see, as big as this has become, the worse I seem to feel. The more people who say, 'It's fantastic and you're so powerful', then the more my self-esteem gets lower and I become more and more drained."
For a long time she hoped to resume her police career, but that dream has been destroyed. Even when she was cleared of perjury, she never got an official apology from any superior in the force. "They were basically saying that I'd lied," she says.
It may well be that someone lied here - but it was not McKie. In 2000, an independent report reached the conclusion that possible criminal prosecutions should be considered against certain SCRO personnel. The author of that report was James Mackay, a much-respected former deputy chief constable of the police force in Tayside. In a statement to McKie's lawayers in August 2003, he said, "It is my view and that of the inquiry team that there was criminality involved in the actings of the SCRO experts and that that criminality first reared its head in February 1997 ... It should have been patently obvious to those involved that a mistake had been made and there were opportunities then for the mistake to be acknowledged and dealt with. The fact that it was not so dealt with led to 'cover-up' and criminality."
But nothing, apparently, was done about these allegations. Earlier in the same year, a letter signed by 14 fingerprint experts from Lothian and Borders police, sent to the then justice minister Jim Wallace and the Lord Advocate, the senior legal officer in Scotland, concluded: "At best the misidentification is a display of gross incompetence by not one but several experts within that bureau [SCRO]. At worst it bears all the hallmarks of a conspiracy of a nature unparalleled in the history of fingerprints." But again, that had led nowhere.
Still, no one at the SCRO has yet lost their jobs or apologised. The same goes for the police.
To make matters worse, David Asbury, the man originally convicted of murdering Marion Ross in 1997, was sensationally acquitted in August 2002 - because of faulty fingerprint evidence. The SCRO experts had got it wrong again. They had claimed that a tin found in his flat during the investigation bore a "mark" - later classed as a "print" - which belonged to the murder victim, Marion Ross. This was incorrect and his conviction was duly quashed.
Still no one from the SCRO was sacked. Indeed, to date the SCRO has not admitted that it wrongly identified McKie's print. Despite the First Minister's claim of an "honest mistake", this is not the story that the SCRO tells. In fact, the office still refuses to accept that it got it wrong. In a statement, it tells me that, "There is a great division of opinion as to whether the print is Ms McKie's."
Yet it is difficult to find any "great division" at all. Scores of fingerprint experts from around the world say that the SCRO is plain wrong. The SCRO is merely backed by a tiny handful of increasingly isolated dissenters who say that they could be right. Britain's fingerprint community has been shattered by the SCRO's stance; the McKie case is referred to as "the Scotch Botch".
It has also recently been alleged that the prints presented by the SCRO in 1999 at the McKie perjury trial had been cropped. This meant that whole prints were not being shown, only parts of them, blown up and cropped advantageously. McKie's claims of innocence still won the day, but the cropping allegation would probably have gone down badly in court if McKie had not settled. Speaking from Tuscon, Arizona, Wertheim says, "I first looked at the images in March 1999 and knew in seconds that they'd been cropped. It's like looking at a snap of your wife and knowing immediately that her head's been cropped. Anywhere in the world would require you to present the whole image, so this was highly irregular."
Michael Mansfield QC is one of those calling for a public inquiry. He believes that the McKie case is a "hugely important and damaging case for the UK justice system as a whole. I don't think the McKie case should ever have been prosecuted in the first place. I know from past experiences in disputed IRA cases, such as the Danny McNamee bombing case, that the whole so-called 'science' of fingerprinting needs to be re-examined. It's a science in the loosest terms - it's more of an art. More quality control is needed. Fingerprint science isn't infallible.
"The SCRO experts who wrongly identified that McKie print should have all their prior cases reviewed. In fact, I am writing a letter to England's Attorney General asking for a review of cases where fingerprints have been pivotal. Public confidence in this process needs to be reassured. The Mckie case is potentially explosive for the English legal system. That's why it deserves a full Scottish public inquiry as fast as possible."
In her flat in Troon, McKie says she is beginning to see a future for herself, but it's "in another country. But maybe that's because I feel like running away at the moment".