'To begin depriving death of its greatest advantage over us, let us adopt a way clean contrary to that common one; let us deprive death of its strangeness; let us frequent it, let us get used to it; let us have nothing more often in mind than death.'

Michel de Montaigne, Essais (1580)

The poet John Donne reminds us that 'Any man's death diminishes me, because I aminvolved in mankind'. It's a line that carries moral weight but, in spite ofthat, it is undeniable that we are more affected by sudden violent death whenit has some connection to our own lives, however tangential. So it is for mewith the case of Rachel McLean, who was an undergraduate at the same smallOxford women's college that I attended. Although I never knew her, I can'tescape a sense of distant kinship with her and her fate.

Rachel McLean was an undergraduate at St Hilda's College when she became JohnTanner's girlfriend at the age of nineteen. Tanner proposed to her on 13 April1991, ten months into their relationship. A momentous occasion like that wouldsurely be something any girl would be eager to talk about with everyone who isclose to her. But over the next few days none of her friends at St Hilda's orin the wider university saw Rachel. She was both a studious and a friendly,open person; nobody could quite believe she'd have gone off somewhere withouttelling anyone. Tanner phoned the house where she lived, saying he wanted tospeak to her but her housemate said she didn't know where Rachel was.

After five days of growing concern, the college authorities reported Rachelmissing to the police. When they contacted Tanner in Nottingham, where he wasstudying at the university, he explained that he'd last seen her on the eveningof 14 April 1991, when she had waved goodbye to him from the platform at Oxfordstation as his train left for Nottingham. A longhaired young man they'd met inthe station buffet had offered her a lift back to her digs in Argyle Street.

Tanner co-operated with the police, helped in the searches and took part in atelevision reconstruction of his departure from Oxford station to try and jogthe memory of anybody who might have seen Rachel. He's believed to be the firstmurderer ever to take part in such a televised reconstruction. In an emotionalpress conference he told friends and reporters that he and Rachel were in love,and planning to get married.

But the police suspected that Tanner was hiding something, so they had briefedreporters to ask him key questions, such as 'Did you kill Rachel?' The way heanswered, smirking and with a lack of emotion, made the police feel sure heknew more about her disappearance than he was willing to admit.

They searched the house in Argyle Street that Rachel shared with friends.Everything appeared to be in order; the floorboards hadn't been tampered withand nothing else looked suspicious. The detectives were desperate to findevidence that they could use to arrest Tanner, or at least to put pressure onhim. Frogmen searched the River Cherwell and other officers scoured nearbyscrubland.

They contacted the local council to check whether the house on Argyle Streethad ever had a cellar. The answer came back that, although there was nobasement, some of the houses on the street had been underpinned, which meantthat there were spaces underneath the floors.

Armed with this information, the police searched the house again on 2 May. Thistime they found Rachel's partially mummified body under the stairs. Tanner hadsqueezed her through a 20 cm gap at the bottom of the stair cupboard, andpushed her underneath the floorboards. Although it had been eighteen days sinceher death, she had barely decomposed; warm, dry air coming through theairbricks had dried her skin.

Finding the body is the end of the first phase of a murder inquiry. But it'sonly the beginning of the forensic pathologist's crucial contribution to thebuilding of a case against the defendant. In the Rachel McLean case, that taskfell to Iain West, head of the Department of Forensic Medicine at Guy'sHospital. During the autopsy, he found a 1 cm bruise to the left of Rachel'slarynx and four 1 cm bruises to the right. He took photographs of these, and ofpetechiae – tiny haemorrhages in her face and eyes. His internal examinationrevealed fractures of the laryngeal cartilages in the throat. All of thesetraumas were indicators of death by strangulation. There was also a tuft ofhair missing from her scalp, which West believed Rachel had torn off in adesperate attempt to relieve the pressure on her throat.

When the police confronted John Tanner with the damning evidence from IainWest, he broke down and confessed to killing Rachel. At his trial he said, 'Iflew at her in a rage and proceeded to put my hands around her neck. I think Imust have lost control, because I have only a vague recollection of the timethat elapsed afterwards.' He claimed he had killed Rachel after she hadadmitted to being unfaithful. Then he spent the night lying next to herlifeless body. In the morning he searched for a suitable place to hide her,dragged her through the gap in the cupboard, and caught a train back toNottingham. Tanner was given a life sentence. He returned to his native NewZealand upon his release in 2003.

Forensic pathology resembles a jigsaw puzzle. The pathologist has to catalogueany unusual elements found on or inside a corpse and, from those fragments ofinformation, try to reconstruct the past. Throughout human history, people havewanted to understand why those they care about have died. The very word'autopsy' derives from the Ancient Greek for 'seeing for oneself'. An autopsyis a medical attempt to satisfy that profound curiosity.

The first known forensic autopsy took place in 44 BC, when Julius Caesar'sdoctor reported that, of the emperor's twenty-three stab wounds, only the onebetween his first and second ribs was fatal. A couple of centuries later, theGreek physician Galen produced hugely influential reports based mainly ondissecting monkeys and pigs. Despite his raw material, his theories on humananatomy remained uncontested until Andreas Vesalius started comparing normaland abnormal anatomy in the sixteenth century, paving the way for modernpathology, the science of disease.

When Vesalius published his landmark book on anatomy, De Humani CorporisFabrica (On the Fabric of the Human Body), in 1543 he dedicated it to the HolyRoman Emperor, Charles V, whose reign also saw another landmark in forensicmedicine. For the first time in the history of the Holy Roman Empire, rules ofcriminal procedure were enacted. They regulated which crimes should be regardedas serious, allowed for the burning of witches and, for the first time, gavethe courts the power to order investigations and inquisitions into seriouscrime. Known collectively as the Carolina Code, crucially for forensicmedicine, they required judges to consult surgeons in cases of suspectedmurder.

The Code was adopted across much of continental Europe and medical authors grewkeen to display their expertise in the courtroom. Those authors included theFrench barber surgeon Ambroise Paré, sometimes called 'the father of forensicpathology'. He wrote about the effects of violent death on internal organs,explained the indications of death by lightning, drowning, smothering, poison,apoplexy and infanticide, and showed how to distinguish between wounds made ona living and a dead body.

As our understanding of the workings of the human body grew, so did thediscipline. In the nineteenth century Alfred Swaine Taylor wrote extensively onforensic pathology and modernised the discipline in Britain and abroad. Hismost important textbook, A Manual of Medical

Jurisprudence (1831), went through ten editions in his lifetime. By themid1850s Taylor had been consulted on more than 500 forensic cases, but hisexperience provides an early demonstration that forensic scientists are asfallible as the rest of us.

In 1859 Dr Thomas Smethurst was tried at the Old Bailey, charged with poisoninghis mistress, Isabella Bankes. At the trial Swaine Taylor testified that therewas arsenic in a bottle owned by Smethurst, and that this was proof of hisguilt. Smethurst was found guilty, and sentenced to death. It later becameclear that Swaine Taylor had carried out the test inadequately and that thebottle was in all probability arsenic-free. Isabella Bankes had been ill for along time before, and was most likely to have died of natural causes. Smethurstwas pardoned, though he did have to serve a 1-year prison sentence for bigamy.The Lancet and The Times heavily criticised Swaine Taylor and the verdict ofmurder, and forensic pathology became known as 'The Beastly Science'. The casecast a heavy shadow over forensic pathology for many years.

Given the theatrical nature of the adversarial English court system, what thediscipline needed to restore its reputation was someone whose ability came witha liberal sprinkling of charisma. That touch of glamour arrived in the personof Bernard Spilsbury. A handsome, convincing orator, Spilsbury was never seenin public without a top hat, tails, buttonhole and spats. His skill wasevident. His ambidextrous hands could work on a dead body with speed andprecision. He also presented his findings in clear, everyday language.

Juries and the public loved Spilsbury. The press portrayed him as a steady rockon which the law could break open the lies of immoral murderers. On his deathin 1947 the Lancet said he had 'stood alone and unchallenged as our greatestmedico-legal expert'. Spilsbury had appeared for the Crown in over 200 murdercases.

Dr Hawley Crippen and his lover, Ethel le Neve, in the dock at the Old Bailey.Crippen would be convicted of murder and sentenced to death, while le Nevewalked free

The first time he came to the attention of the public was in 1910, as an expertwitness at the sensational trial of Doctor Hawley Harvey Crippen. An Americanhomeopath and patent medicine salesman, Crippen had been living in Camden Townwith his wife, Cora, a music hall singer with the stage name 'Belle Elmore'.The marriage had been in difficulties; then Cora's friends suddenly stoppedseeing her around. Doctor Crippen told them variously that she had died andthat she had gone to America to further her career. They became suspicious andwent to the police, who questioned Crippen and searched his house. They foundnothing. But the investigation panicked Crippen, who fled with his teenagemistress, Ethel Le Neve, aboard the SS Montrose bound for Canada. Le Nevedressed as a boy and posed improbably as his son.

Their flight rekindled the suspicions of the police, who went back to searchthe house a second time, and again found nothing. But they remained suspiciousand mounted a third search in which they dug up the brick floor of the cellar.This time, they found what were believed to be the remains of a human torso,wrapped in a man's pyjama top.

Meanwhile, the captain of the Montrose had noticed two odd passengers on board,and, just before his vessel went out of range, he sent a wireless telegram tothe British authorities: 'Have strong suspicions that Crippen London cellarmurderer and accomplice are among saloon passengers. Mustache taken off growingbeard. Accomplice dressed as boy. Manner and build undoubtedly a girl.' ChiefInspector Dew of Scotland Yard boarded a faster ship, landed in Canada beforethe pair, and dramatically arrested them when they arrived – the first arrestmade with the help of wireless communication.

The police called in a surgeon from St Mary's Hospital, London, to examine thebody and he put the young Spilsbury on the case. Spilsbury took a close look atCora Crippen's medical records and noticed that she had had an operation on herabdomen. The autopsy did not reveal the sex of the body, but Spilsbury did findtraces of toxic compounds.

At Crippen's trial, Spilsbury presented a piece of skin bearing a curved scar,preserved in formaldehyde, from the torso believed to belong to Cora Crippen.He passed it around the jury in a glass dish. Spilsbury set up a microscope inthe next room for members of the jury to examine slides of the tissue. Althoughthe defence pathologist argued that, because it had hair follicles growing fromit, it must be folded skin and not scar tissue, the jury believed Spilsbury.Crippen was found guilty of drugging and murdering his wife. He was hanged atPentonville Prison in London and at his request buried with a photograph of LeNeve. She was charged, and then acquitted, of being an accessory after thefact.

The Crippen slides still exist, in the Royal London Hospital, and in 2002Professor Bernard Knight examined them. He could not see definite indicationsof scar tissue. Recent DNA testing of the fragments seem to show that the DNAdoes not match that of some of Cora Crippen's descendants, and that the remainsare male. Ironically, it seems that the case which projected Spilsbury into thepublic consciousness as the standard bearer for forensic pathology may havebeen one in which he was comprehensively mistaken.

A series of slides produced by Spilsbury showing sections of the scar found onthe torso buried in the cellar: according to him, this proved it belonged toCora Crippen, though others disagreed

Five years after Crippen was hanged, Spilsbury was involved in anotherextraordinary case that neither DNA testing nor any other modern forensictechnique could have helped him solve. On Sunday, 3 January 1915, aBuckinghamshire fruit grower called Charles Burnham sat down with a mug of teaand opened his copy of the News of the World. A headline on page three shookhim to the core: 'Dead in Bath – Bride's Tragic Fate on Day after Wedding'. Theshort report explained that one Margaret Lloyd had been found dead at her flatin north London. 'Medical evidence showed that influenza combined with a hotbath might have caused a fainting fit,' it concluded. Charles Burnham'sdaughter, too, had died in a bath in Blackpool, shortly after her weddingalmost exactly a year before. Burnham contacted the police, and discovered thatMargaret Lloyd's husband was George Joseph Smith, the man who had previouslybeen married to his daughter, Alice Burnham.

The police called in Spilsbury to perform an autopsy on the exhumed body ofMargaret Lloyd. He then travelled to Blackpool to autopsy Alice Burnham.Following this, the police uncovered details of a third woman, Bessie Williams,who had married George Smith and died in very similar circumstances at home inKent on 13 July 1912.

The coroners had given a verdict of accidental death in the first two drowningsbut, when the police investigated anew, they discovered that Smith hadbenefited financially from all of his wives' deaths: he had received £700 and£506 from the life insurance policies held by Margaret and Alice; and £2,500 intrust money (worth around £190,000 today) from Bessie. Now that they could seea pattern, the police arrested Smith.

George Smith and Bessie Munday on their wedding day. She would later become hisfirst victim

From the bodies of Margaret and Alice, Spilsbury could find no signs of violence,poison or heart attack. On Bessie he found evidence of 'gooseskin' on herthigh, which is sometimes an indication of drowning (though it can also becaused by decomposition after death). Spilsbury read the notes of the generalpractitioner who had first seen Bessie's body, and noted that she had beenclutching a bar of soap.

He had all three bath tubs brought to Kentish Town Police Station, where helined them up together and examined them minutely. He was particularly puzzledby the case of Bessie Williams. Shortly before her death Smith had taken heralong to the doctor so she could talk to him about her epilepsy symptoms. Smithhad told Bessie that she was suffering from fits, even though she couldn'tremember having them and had no epilepsy in her family. Spilsbury was notconvinced by this version of events. Bessie was five feet seven inches tall andobese. The bath tub she had died in measured just five feet at its longest, andsloped at the head end. Spilsbury knew that the first phase of an epileptic fitcauses complete rigidity of the body and that, given her size and the shape ofthe bath, such a fit would have raised Bessie's head above the water ratherthan brought it below.

If an epileptic fit had not been responsible for Bessie's death, what couldhave caused it? Spilsbury researched further and learned that an extremelysudden rush of water into the nose and throat can inhibit a vital cranialnerve, the vagus nerve, and cause sudden loss of consciousness, swiftlyfollowed by death. A common subsidiary result of this uncommon occurrence isinstant rigor mortis – which Spilsbury thought explained the bar of soapclenched in Bessie's fist.

Alfred Swaine Taylor had stated categorically in 1853 that it was impossible todrown an adult without leaving bruises, because of their violent struggle forbreath, and this had never been challenged. Detective Inspector Neil decided tocarry out a series of experiments before the trial, to test Spilsbury's theoryof how the women had died. He found women volunteers who would try to resistbeing submerged in the bath tubs. The first volunteer stepped into a full bath,clad in a bathing costume, and managed to grip the side of the bath andstruggle. But when Neil grasped her ankles and abruptly pulled her legs up, sheslid under the water and lost consciousness. It took a doctor several minutesto restore her to consciousness; she was lucky to live. The experiment was notSpilsbury's idea, but he was aware of it and his reputation certainly benefitedfrom its outcome.

George Smith was tried for the murder of Bessie Williams. At the trialSpilsbury spoke with great authority, and effortlessly carried the jury withhim. They deliberated for twenty minutes before finding Smith guilty. He washanged at Maidstone Prison.

Smith had been a silver-tongued conman who had committed his first theft in theEast End of London at the age of nine. As he grew up he began to wear goldrings and brightly coloured bowties to help him impress women, so he couldexploit them. Because of the early effects of the First World War, and becauseso many young British men of his day had moved to the colonies, there were asurplus of half a million women in Britain in 1915, providing Smith with plentyof prey. Stoked by the newspapers, public interest in what came to be calledthe 'Brides in the Bath Murders' was intense at the time. Scores ofjournalists, hungry for a 'scientist foil serial killer' headline, door-steppedSpilsbury throughout the investigation. His star would not stop rising in hislifetime.

Many of Spilsbury's cases involved husbands charged with murdering their wives.It's chilling to wonder how many had got away with it before the science hadevolved sufficiently to uncover the truth behind those women's deaths. Increasingly,the press presented a picture of Spilsbury as a hero who had devoted his lifeto interpreting clandestine clues left on defenceless victims, so that theirevil killers could not escape justice or strike again. In 1923, that image wasbolstered by a knighthood. A year later, another case cemented his reputationever more firmly.

On 5 December Elsie Cameron left her home in north London to visit her fiancé,Norman Thorne, a chicken farmer in Crowborough, Sussex. They had been engagedfor two years, but Thorne had recently started seeing a new girlfriend.

On 15 January 1925, the police found Elsie's dismembered body buried under achicken run on the farm and her head stuffed into a biscuit tin. Thorne hadoriginally told them that she had never arrived but, after the discovery of thebody, he changed his story, saying that she had arrived to tell him she waspregnant and wanted to get married. He said that he had then left the housebut, when he returned two hours later, he found her hanging from the ceiling.Assuming it was suicide, he decided to conceal the body, cut it into fourpieces and bury it.

Spilsbury did the autopsy on 17 January and, in his report to the coroner, saidElsie had met a violent death, probably after being beaten. He had foundevidence of eight bruises, including one on her temple, not visible on thesurface but revealed when he opened the flesh. As he did not see any rope markson her neck that would indicate hanging, he did not take a section from it. Hehad noticed two marks on her neck, but thought they were just natural creases.At the inquest, the coroner inquired how it was possible to examine a6-week-old corpse, but Spilsbury assured him that the decomposition had been noproblem. The accused man, Norman Thorne, questioned the report because therewere no external signs of bruising, and successfully applied for a secondautopsy.

Bernard Spilsbury, the eminent and suave pathologist. His testimony helpedconvict hundreds of criminals, though some of his conclusions have since beenchallenged

Elsie was exhumed on 24 February and Robert Bronte performed an autopsy withSpilsbury watching on. Post-mortem examinations are supposed to take place inthe bright light of day or in electrically lit mortuaries. Elsie was exhumed frommidnight to 9 a.m, in front of a crowd of spectators and journalists in thedimly lit chapel at the cemetery. The coffin was full of water, and the corpsehad had an extra month to decompose since Spilsbury had examined it, but Brontesaw evidence of marks on the neck and took sections away to analyse.

The trial of Norman Thorne lasted five days. The rival pathologists disagreedover the bruises. When asked by the prosecution whether there were any externalmarks on the body, Spilsbury replied, 'No, none at all.' The defencepathologist, J. D. Cassels, argued that Elsie Cameron was still alive whenThorne had cut her down from the beam, that the bruises were caused by herfalling to the floor, and that she had died from shock ten to fifteen minuteslater: this would explain the absence of rope marks, as they were eliminated bythe circulation of the blood. He criticised Spilsbury for not examining theneck microscopically.

Spilsbury testified that the two final bruises were those on her face, and thatshe had been beaten to death with some Indian clubs which were found quite nearto the scene. As was his adamant courtroom style, he refused to acknowledge anyuncertainty in these conclusions. Yet he had said in a lecture two yearspreviously that if medical evidence is strictly tested by cross-examination,'that is when the doctor realises his fallibility'.

Throughout the trial, the judge referred to Spilsbury as 'the greatest livingpathologist', and told the jury in his summing up that Spilsbury's 'is undoubtedlythe very best opinion that can be obtained'. The jury reached its 'guilty'verdict in less than half an hour. Some felt that they had not acknowledged thecomplexities of the pathologists' evidence, especially the fact that there wereno signs that Elsie Cameron had met a violent death. Among those who wereconcerned with the jury's acceptance of Spilsbury's confident conclusions wasSir Arthur Conan Doyle, who lived close to Norman Thorne. He wrote in the LawJournal that 'the more than papal infallibility with which Sir Bernard isreadily being invested by juries must tend to be somewhat embarrassing to him'.

Norman Thorne was hanged at Wandsworth Prison for the murder of Elsie Cameron,though he pleaded his innocence to the bitter end. In a famous letter to hisfather on the eve of his execution he wrote, 'Never mind, Dad, don't worry. Iam a martyr to Spilsburyism.'

According to historians Ian Burney and Neil Pemberton, the Thorne trial focusedattention on two rival practices of pathology: Spilsbury the celebritypathologist making a dramatic courtroom appearance, relying on his scalpel andhis intuition; and Bronte the laboratory-based pathologist, relying on thelatest forensic technology. They argue that Spilsbury's 'virtuosity' in the mortuaryand the courtroom 'threatened to undermine the foundations of forensicpathology as a modern and objective specialism'.

In his book Lethal Witness (2007), Andrew Rose suggests that Spilsbury causedat least two miscarriages of justice and several more unsafe verdicts.Convictions were sometimes made on flimsy evidence, because if Sir BernardSpilsbury said a man was guilty then juries believed he must be guilty. In someof his more than 20,000 autopsy reports Spilsbury suppressed evidence because itdidn't fit with his narrative.

For example, in 1923, due to Spilbury's evidence, a young soldier called AlbertDearnley was found guilty of tying up his best friend and suffocating him.Convicted of murder, he was only two days away from the gallows when the prisongovernor read a letter Dearnley had written to a female friend. Its toneworried him and he persuaded the Home Office to grant a stay of execution.

Just in time, the truth came out: the death was not murder, but an accidentalasphyxiation that had happened in the course of a sadomasochistic homosexualgame. Spilsbury – famous for his homophobia – had suspected the truth, but hadkept his counsel because he believed the soldier was a sexual pervert whodeserved his fate.

Nevertheless, when Spilsbury committed suicide in 1947, gassing himself in hisown laboratory at University College London after a long battle with depressionand poor health, it wasn't only the Lancet that hailed him as the greatestpathologist of the age. Voices of adulation drowned out the dissenters manytimes over. His image tarnished only gradually after death. By 1959 his fellowforensic pathologist Sydney Smith was able to write, 'One might almost hopethat there will never be another Bernard Spilsbury.'

Today, Dick Shepherd is the leading forensic pathologist in the UK. But he isadamant that he is no celebrity courtroom performer – and nor would he want tobe, even though the roll call of some of his autopsy subjects is remarkable.From Princess Diana and Jill Dando to victims of the 9/11 attacks in the US, hehas investigated some of the most notorious deaths of recent years. To him,every case is the same: an autopsy is supposed to be 'a non-judgemental,scientific, acquisition of facts', no matter who the victim is.

What gets Dick Shepherd up in the morning is the living, not the dead. 'I'mfascinated by the interlinking – working with the police, the courts and withother people. Seeing problems, understanding them, interpreting them andproviding that information to others. I have to separate myself from thedestructive things that I do and remember that I am doing it for the familiesof the dead. If they understand what has happened it doesn't help them in anypractical way, but if there are points of truth they can fix on them, andachieve closure. Forensic science has gone wrong when people haven't beentruthful, sometimes because they have hidden the truth in the hope of notdistressing families. It never works.'

It's up to the police to make the tricky call about how much a pathologistshould know of a case before the first crucial examination of a body. If thepathologist knows too much it might bias the autopsy. If he knows too little hemight overlook something important. And, as Dick Shepherd explains, 'If thefiltering is done by others they may not give us a critical bit of information.Then when it suddenly pops out in court, it can lead to an "Oh, cor blimey"moment. The lawyer asks, "If you had been told this would you have formedanother opinion?" "Yes, I would." "Oh, thank you, Dr Shepherd." Lawyer sitsdown with smug smile on face.' And the prosecution winces.

One of the reasons Spilsbury so rarely had to witness a defence lawyer's smugsmile was that he almost invariably knew a great deal of the background to hiscases. Nowadays, when Dick Shepherd gets a call from the police or thecoroner's office, he is seldom summoned to the crime scene but rather to themortuary. Various other scientists with specialisms such as blood spatter andDNA analysis do much of the crime scene evidence gathering that forensicpathologists used to be responsible for. At the scene of death, junior CSIs bagup the body to prevent trace evidence falling from it, such as hair, fibres anddirt, and to protect it from contamination.

When he does go – and 'sometimes it is really useful to, not so much to do anyspecific examinations but to interpret the scenario' – Dick observes bodyposition and proximity to other evidence such as weapons, fingerprints, pointsof entry and escape. He has to be very careful not to lose or contaminateevidence by touching or moving the body more than is absolutely necessary. In arecent case of his, the police believed a lady found at the bottom of herstairs had taken a tumble. Dick went along to see 'where she was lying, how shewas lying, if she had possibly been moved. When I did the autopsy I foundinjuries that I thought were caused by her coming down the stairs and banginginto things. Because I'd gone to the scene I could later explain the layer ofgrazes on her side as being where she had gone around a corner.'

Detectives always want an estimated time of death. That's information that canshake, break or confirm a suspect's alibi. The longer a body has been dead, theharder it is to estimate with much accuracy when death occurred; the narrowerthe estimate, the more useful it is to the investigation.

The first thing a pathologist like Dick Shepherd does when examining a corpseis to take its rectal temperature, unless there is reason to suspect sexualassault. In such cases, he stabs the thermometer into the abdomen. It was oncethought that bodies lose heat at a constant 1°C per hour until they reachambient temperature. So, for example, if a person dies with the average bodytemperature of 37°C in a 20°C room, there will potentially be a 17-hour windowwhen the time of death can be roughly estimated. But research has shownsignificant variables: a thin body cools faster than a fat one; the larger thesurface area, the faster the cooling process; whether the body is sprawled orcurled up will have an impact; clothing will affect the cooling; shade orsunlight; in the shallows or beside a river. Even so, an early and carefulreading can be a useful starting point and pathologists can factor in variableslike ambient temperature and body weight using a graph with multiple axescalled a nomogram.

The next phenomenon that interests Dick is rigor mortis – the macabre reasonwhy dead bodies are known as 'stiffs'. The symptoms of rigor are useful to thepathologist for about two days after death because of a recognised cycle. Atfirst a body relaxes completely, then after three or four hours the smallmuscles of the eyelids, face and neck begin to stiffen. Rigor progressesdownwards, from head to toe towards the larger muscles. After twelve hours thebody is completely rigid and will stay fixed in the position of death foraround twenty-four hours. Then the muscles gradually relax and stiffness goesaway in the order in which it appeared, starting with the smaller muscles andprogressing to the larger ones. After a further twelve hours or so all themuscles reach a state of complete relaxation.

But even a process as well documented as rigor mortis is a very imperfectindicator of time of death. The hotter the ambient temperature, the faster eachstep of the cycle occurs. Also, bending and stretching a corpse breaks up itsmuscle fibres and eliminates rigor, a piece of information that murderers havebeen known to employ to confound the investigation.

Rigor mortis is followed by the least majestic stage of the body's time onearth. 'Putrefaction' may not be a loveable phenomenon, but forensicpathologists need to be on intimate terms with it to do their jobs properly.First the skin around the abdomen turns greenish as the bacteria in the lowerintestine begin a process of 'self-digestion'. As bacteria grow throughout thebody, breaking down proteins into amino acids, they produce gases and cause thebody to bloat – starting with the features of the face, where the eyes andtongue begin to protrude. Next, a web-like pattern of blood vessels appears asmarbling on the skin because the red blood cells break down and releasehaemoglobin. Gases continue to swell the abdomen until they escape, sometimesexplosively, and produce a vile smell. The body has turned greenish-black bynow, as fluids drain from the nose and mouth, and the skin falls away like 'agiant rotting tomato'.

All the while, and aided by 'self-digestion', the internal organs have beenliquefying, starting with the digestive organs and the lungs, and then thebrain. Flies have laid eggs on the body's points of entry, such as the eyes,mouth and open wounds, and maggots have been tearing continuously at the flesh.

Scientists are continually studying and refining the various different ways ofmeasuring time since death. However, as forensic anthropologist Sue Blackexplains, this doesn't always simplify matters. 'The more information we get,the more we realise just how difficult it is. No two bodies will decompose inthe same way, and at the same rate. You can have two bodies that are literallysix feet apart and they will decompose in entirely different manners. It couldbe the amount of fat on the body. It could be the drugs they were taking, or themedication. It could be the type of clothing they're wearing. It could be thatone has a particular odour that is more attractive to flies than the other.Absolutely anything.'

One way to try and combat the headache of variables is to develop new tools.That's what the Anthropological Research Facility at the University ofTennessee has been doing for years. 'The Body Farm', as it's better known, wasset up in 1981 by William Bass to research putrefaction. It was the firstinstitution to develop the systematic study of human decomposition and howbodies interact with the environment. More than 100 people donate their bodiesto the Body Farm every year, to be placed in different settings and left torot. Researchers have worked out a general rule of thumb: one week exposedabove ground equals eight weeks below and two weeks in water.

Arpad Vass, Associate Professor of Forensic Anthropology at the University ofTennessee, is developing a new method for estimating time since death.'Decomposition Odour Analysis' hopes to identify the 400 or so distinct vapoursthat a body gives off at different stages of its decomposition. Understandingwhen these vapours are given off, in a variety of settings, and measuring themin a corpse, could provide a more accurate time of death than has so far beenpossible.

Findings from research facilities such as the Body Farm trickle into the worldof practical forensics through journals and monographs, and arm pathologistswith knowledge that they can use to give better evidence to criminalinvestigations. Pathologists use that knowledge most usually in a morgue or ahospital, locked in the intense focus of autopsy. How and why did the persondie? Was it suicide, murder, an accident, old age, or is it impossible to tell?There are seldom straightforward answers. A bullet that has passed throughsomeone's head could have been shot suicidally, murderously or accidentally.The scope of a forensic pathologist's curiosity is very broad when he entersthe morgue. It gradually narrows its focus to the small details, beforebroadening out again to incorporate those details in a conclusion. The generalsteps of an autopsy have changed little since the beginning of last century.

When the body arrives in the morgue Dick Shepherd is ready to photograph it. Anassistant takes the bag the body was stored in for transportation and searchesit for trace evidence. Dick removes the subject's clothes, which hephotographs, bags and documents. Then he takes biological samples, pluckinghair, scraping fingernails, swabbing sexual organs. Only now does he carefullytake the fingerprints; prying open fingers to take prints can jeopardise smallbits of trace evidence clenched in a fist closed by rigor.

Dick then washes the body and documents every scar, birthmark, tattoo andunusual physical feature he can find. 'Every pathologist has a differentroute,' says Dick. 'I start at the head and I always do the left-hand sidefirst. So I do head, chest, abdomen, back, left hand, right hand, left leg,right leg. As I go around, injuries are all documented and photographed. I haveto say my heart sinks now when it's a pub fight and there's 970 2 cm bruises.Can't I just say there were lots of bruises on the legs? No.' In less clear-cutcases the rigorous style encouraged by the documenting process can proveinvaluable, as illustrated by another British wife who died in a bath.

At 11 p.m. on 3 May 1957, Kenneth Barlow, a nurse from Bradford, phoned 999 tosay that he had found his wife unconscious in the bath. He explained that hehad pulled her out and spent a long time trying to resuscitate her, and thatshe had been suffering from vomiting and fever that evening. Investigators weresuspicious when they discovered two used syringes in the kitchen. Kenneth explainedthat he was using them to treat an abscess he had with penicillin. Testsconfirmed the presence of

penicillin.

But pathologist David Price remained suspicious. During the autopsy he searchedevery inch of Mrs Barlow's skin with a magnifying glass. Eventually he foundtwo tiny holes consistent with injection needles, one on each of Mrs Barlow'sbuttocks. The symptoms that Kenneth had said his wife was suffering from werethose of hypoglycaemia (low blood sugar), which made David Price suspect thathe had injected his wife with a lethal dose of insulin. There were no tests forinsulin at the time, so Price took tissue from around the injection points onMrs Barlow's buttocks and injected it into mice. They quickly died ofhypoglycaemia. Barlow was found guilty of murder and given a life sentence.

After the meticulous external scrutiny, the internal examination begins. Thepathologist is looking both for internal injuries and any medical conditionthat might have caused the person to die naturally. Dick Shepherd cuts open thebody in a 'Y' shape from both shoulders down to the groin, sawing open the ribsand collarbone and removing the breastplate, to reveal the heart and lungs. Heinspects the neck, looking out for things like broken cartilage, which mightindicate strangulation. He then takes out the organs individually (like theliver) or in groups (like the heart and lungs), examines their surfaces, andmakes cuts to examine them internally. He preserves samples of the organs. 'TheHome Office now insist that we do microscopic examinations of all of the majororgans on every case, even when it's an 18year-old who's been hit by a baseballbat on the top of the head.' Better safe than sorry, for which we should all begrateful. He then sends those samples off to the laboratory. Next Dick makes anear-to-ear incision over the top of the head and peels back the scalp. Now hecan saw away a section of the skull and look at the brain in place beforeremoving it for a closer examination.

Finally, Dick sews up all the incisions he has made in the organs, places themcarefully back in the body, and stitches up his initial 'Y' incision.Afterwards he talks to detectives and to other forensics experts, pools ideasabout what looks suspicious or needs to be followed up, and feeds back into theinquiry. Very often there will be a second autopsy, so that another pathologistcan check Dick's findings. Once all the reports have come in from the experts –bone pathologists, neuropathologists, paediatric radiologists – Dick writes hisreport for the coroner.

In some extreme situations more than one autopsy will follow Dick's. On 23August 2010, the police found a red North Face bag in the bath of a flat inPimlico, London. It belonged to Gareth Williams, a Welsh maths prodigy who hadbeen working as a codebreaker for MI6, and its zips were fastened with apadlock on the outside. When the police prised the padlock open they found thenaked, decomposing body of 31-year-old Gareth folded inside.

The police regarded the death as suspicious. Gareth Williams' family wereconvinced that M16 or another secret service was involved in the death, asWilliams had been working with the FBI as part of a team attempting topenetrate hacking networks.

Dick Shepherd was one of three pathologists who performed autopsies, and whoall agreed that Williams had been dead for about seven days. Whilst they foundno signs of strangulation or physical trauma, it was difficult for them todetermine the cause of death because the body had decomposed so quickly: thebag had been found in the summer, and all the radiators in the house had beenleft on maximum. Toxicologists found no signs of poisoning but, because of thebody's condition, they couldn't rule it out. Suffocation seemed most likely.

The autopsies revealed small abrasions on the tips of Williams' elbows, whichmight have resulted from him moving his arms inside the bag, possibly in anattempt to escape. Dick Shepherd summed up the situation: 'Once the lock wasplaced on it there was no possibility of getting out of that bag. The questionis: did he place the lock there or did another?'

Peter Faulding, an ex-military reservist who specialised in rescuing peoplefrom confined spaces, told the inquest in May 2001 that he had tried 300 timesto lock himself into an identical 81 cm by 48 cm North Face bag, and failed todo so. He said that even Harry Houdini 'would have struggled to lock himself inthe bag'. Another expert had tried and failed a further 100 times.

But Dick Shepherd felt that Williams had suffocated, and was 'more likely thannot' alive when he entered the bag. He reasoned that, because Williams washunched into such a tight ball when he was found, and because of how 'floppy' abody is before rigor mortis sets in, it would have been very tricky for someoneelse to stuff him inside it. No expert offered to test this theory. It was alsorevealed at the inquest that DNA found on Williams' finger during his firstautopsy did not come from a mysterious Mediterranean couple that the police hadbeen pursuing for a year. An employee at LGC, the forensic company which wasanalysing samples, had typed the wrong DNA details into a database. The DNA infact belonged to one of the CSIs who had been at the scene. LGC expressed 'verydeep regret' to the Williams family for the mistake.

Twenty thousand pounds-worth of designer women's clothes had been found atWilliams' flat, along with women's shoes and wigs. Investigators had also foundpictures of drag queens and evidence that he had been browsing self-bondagewebsites and sites about claustrophilia – the love of enclosure – days beforehis death.

The coroner, Fiona Wilcox, ruled that, though there was not enough evidence togive a verdict of unlawful killing, it was probable that the death was unnatural,that someone else had locked him in the bag and placed it in the bath, and thathe was alive when he entered it. She added that there was no evidence tosuggest that Williams was a transvestite 'or interested in any such thing'.

Days after the verdict, a 16-year-old girl and a 23-year-old journalistseparately attempted to lock themselves into identical North Face bags: theyclimbed in, drew their legs up, pulled the zippers nearly shut, poked theirfingers through the gap, and closed the padlock. They then tensed their bodies,and the zip sealed itself up. The journalist, who was of a similar build toGareth Williams, repeated the trick a number of times until she could do it inthree minutes.

Peter Faulding didn't think much of these stunts. 'None of my conclusions arewrong. A young girl zipping a bag doesn't discredit this inquiry whatsoever. Wewere fully aware of other methods of being able to lock the bag but she ornobody could achieve it without leaving her DNA or trace on the bath and that'sthe key to this.'

Dick Shepherd continues to disagree. 'I'll never convince the coroner. She wasfurious with me; the word "incandescent" springs to mind. One of the strongestthings that makes me feel he was on his own is that he lived a solitary existence,wearing women's clothes, working intensely, and being a mathematics geek –forget the pathology, the psychology is not right.'

In 2013, the Met Police held an internal investigation because the coroner hadindicated that MI6 staff might have been involved. By November Scotland Yardhad the investigation's verdict: it was probable that Williams had died alone,as a result of accidentally locking himself inside the bag. Dick Shepherd'sview had been vindicated.

Mysterious cases often need imagination to bring them to a resolution: thepathologist who injected tissue from a woman's buttocks into mice; thejournalist who locked herself into a small bag when a military expert couldn'tdo it. These people wanted to follow the original Greek meaning of autopsy, tosee for themselves. Our curiosity grows with every fresh technique weroad-test. New technology is letting pathologists see deeper than ever beforeinto a human subject, without the need to roll their sleeves up. The VirtualAutopsy (VA) table is a new medical visualisation tool built in Switzerlandthat combines CT and MRI scans to transform images of a dead body into a 3Dcomputer model. Pathologists using it in Germany have found fractures andhaemorrhages undiscoverable using conventional autopsies. The VA table alsoincludes a high-resolution scanner, which means the skin can be magnified tobetter reveal things like bruising or malicious injection points. It alsocauses less distress to the living, who don't want to contemplate what they seeas the desecration of the bodies of their loved ones.

Some traditional forensic scientists have called the VA table unproven andnew-fangled. But as a younger, more tech-savvy generation move into pathologylabs, they have begun installing them. By January 2013 three out of thethirty-five forensic science institutes at German universities had one.Forensic pathologists are tending still to use them as a complement to physicalautopsies. But the proofs keep building up. In the case of a mountain climber whohad fallen to his death in the Swiss Alps, his shattered neuro-cranium,fractured lumbar spine and broken lower leg were all detected without a singlescratch from a knife.

Another benefit of the Virtual Autopsy table is that the 3D model it producescan easily be examined by several pathologists independently, saved for futurereference and produced in court for juries to judge for themselves. Spilsburymight not have liked the idea, but his martyrs surely would have.