When the jury in Miami convicted Ted Bundy on July 24, 1979, two things had been put in front of it that no one in a Florida courtroom would have called controversial a few years earlier: a hair comparison and a set of plaster casts of his teeth. Forensic dentists had told the jury that those casts matched a mark photographed by police technicians. The jurors deliberated for less than seven hours. The Florida Supreme Court later looked at the dental testimony closely, did not like Bundy’s arguments against it, and affirmed.
Thirty years after that trial, a National Academy of Sciences committee reviewed the forensic disciplines that had been trusted in American courts and singled out bite-mark comparison as one that lacked a scientific footing. By 2016 a presidential science panel was calling it "far from meeting the scientific standards for foundational validity," and a Texas commission had asked prosecutors to stop using it. Between the first trial in which a dentist’s casts decided a murder case and those reports lay four decades of convictions, a handful of them in the cases of men now known as serial offenders. Several of those men were convicted on other evidence as well, or later admitted what they had done. The cases that came after them, involving men who were not serial offenders at all, are what brought the method down.
Forensic odontology is two different things, and the courts kept them together. One is identification: matching an unnamed victim’s teeth to a missing person’s dental records, a task as old as the dentist’s chart and about as contested as a fingerprint. The other is comparison: setting a mark left in a victim’s skin against casts of a suspect’s mouth and telling a jury that one made the other. The first became routine. The second is the part that has been questioned.
The method’s early courtroom history in North America runs through the case of Wayne Clifford Boden. He was convicted in 1972 of killing four women, three in Montreal and one in Calgary. The Canadian Press, looking back in 2013, wrote that forensic evidence left on the victims was what secured the conviction. The forensic-dentistry literature is more specific. A 2020 review by C. Y. Gómez, a forensic dentist with the Argentine Federal Police, says the orthodontist Gordon Swann presented evidence linking marks on the victims to Boden’s teeth, and describes the conviction as the first of its kind in North America.
How many points of comparison Swann claimed, which court heard the case and what the jury was told have not been reported. The claim that the conviction was the first of its kind in North America is Gómez’s; the Canadian Press said only that forensic evidence left on the victims sealed the verdict. Accounts of the sentence differ as well. A 1984 wire report spoke of life sentences for four killings without giving a number, while the 2020 paper says three life sentences for four murders.
What survives is a pattern that would repeat for decades. A dentist trained to straighten teeth, not to testify, was asked to say that a suspect’s mouth could be tied to marks on a victim, and a jury believed him. Boden stayed in the news later for a different reason. In 1977 he applied for an American Express card, giving Correctional Services Canada as his address, and received one. In May 1984 he slipped away from an unarmed escort on a day pass in Montreal and was recaptured about 36 hours later, and the Solicitor General told the House of Commons that granting the pass had been a mistake. None of that touched the dental evidence, which was never reported to have been challenged.
The Bundy trial gave the method its widest audience. It opened on June 25, 1979, before Judge Edward Cowart, with television cameras in the room, and it was among the first trials in the country to be broadcast nationally. Bundy, a law student before his crimes, acted as one of his own lawyers. The state’s eyewitness evidence was shaky: Nita Neary had been hypnotized eight days after the events she described, and a survivor testified that she could not identify the man.
The physical evidence came down to two things. A hair analyst testified that hairs connected to one survivor’s case had the same characteristics as Bundy’s and could have come from him. And on July 18 the forensic dentist Richard Souviron testified that Bundy’s teeth matched evidence in the Lisa Levy case that police technicians had photographed. A second forensic dentist agreed. Both used photographs and models of Bundy’s teeth, and both told the jury, in the Florida Supreme Court’s later summary, that human teeth vary enough for a comparison to identify an individual with a high degree of reliability.
The defense had fought the testimony before trial. Dental experts for the state and the defense both appeared at a pretrial hearing on whether it could be admitted at all. Bundy argued on appeal that the comparison techniques were not reliable, that accepted standards of comparison had not been shown, and that the state’s chief expert was prejudiced and had given an improper opinion on guilt. The Florida Supreme Court, in a decision dated June 21, 1984, rejected each point. The trial court had held that odontology was generally recognized by scientists in the relevant fields and that the techniques used were reliable enough to go to a jury. It was proper, the justices wrote, for the expert to give an opinion on whether a bite mark matched. The jurors could look at the photographs and models themselves, and the weight of the evidence was for them.
That reasoning is worth reading closely, because it is the same reasoning every later court would borrow. The question the justices asked was whether dentists accepted the method, and the answer was yes. They did not ask whether anyone had tested the method against marks of known origin, because in 1984 almost no one had.
The jury convicted Bundy of two counts of first-degree murder, three of attempted first-degree murder and two of burglary after less than seven hours, and Cowart sentenced him to death. He was convicted separately in Orlando in 1980 of the kidnapping and murder of Kimberly Leach, a case that rested on no bite-mark evidence. He was executed in the electric chair at Florida State Prison on January 24, 1989. It is true that dental evidence did not carry the Miami verdict alone. It is also true that the Chi Omega trial is the one in which an American public first watched forensic dentists address a jury, and many people have taken the method’s reliability from that picture ever since.
The Rodney Alcala case shows the other end of the method’s life, a comparison made after the fact in a case no jury ever heard. Cornelia Crilley, a 23-year-old flight attendant, was killed in New York City on June 12, 1971. The case went unsolved for more than thirty years. In 2003, New York detectives went to California with a warrant to interview Alcala and take an impression of his teeth. According to a police spokesman, Alcala first denied ever having been in New York, then asked, "What took you so long?"
A forensic dentist later concluded that evidence in the Crilley case was consistent with his impression, an unnamed law enforcement official told reporters in 2012, and authorities said his DNA matched genetic material from the case. The order in which these came matters. DNA, not the dentist, is what the prosecution could count on, and Alcala did not stand trial in New York. He pleaded guilty in December 2012 to the murders of Crilley and Ellen Hover, and on January 7, 2013, a Manhattan judge sentenced him to 25 years to life. The statement that his teeth were consistent with that evidence was never tested in a courtroom, by cross-examination or otherwise. Whatever it was worth, it was an extra layer on a case that no longer depended on it.
Alcala had already been convicted in Orange County, California, on February 25, 2010, of five counts of first-degree murder and one of kidnapping, and sentenced to death. The California cases rested on DNA, fingerprints and earrings, not on his bite. He died on July 24, 2021, of natural causes, on death row.
In Hong Kong the dentist’s contribution was of the other kind. Lam Kor-wan, a night-shift taxi driver, was convicted by a High Court jury in 1983 of four counts of murder. The women were Chan Fung-lan, Chan Wan-kit, Leung Sau-wan and Leung Wai-sum, killed between February and July 1982. Police arrested him on August 17, 1982, in Kowloon. Staff at a photo shop had reported the pictures he left to be developed. The jury heard the case over 21 days in March 1983, according to Sing Tao Headline, and on April 8, 1983, the judge sentenced him to death. The sentence was commuted to life imprisonment in August 1984.
HK01 reported in 2018 that the case was the first in Hong Kong to rely on dental evidence for identification, and that it led to the founding of the Hong Kong Forensic Odontology Group. This is identification by records, the least disputed use of the discipline. Nobody had to argue that a tooth makes a unique mark. The question was whether a set of teeth belonged to a named woman, and a dentist with her chart could say. Accounts of who did the dental work, and how, do not agree.
Lam was reported to be still in prison in October 2023.
The Scottish case shows what happens when teeth become a search tool and not an exhibit. Three women, Patricia Docker, Jemima McDonald and Helen Puttock, were killed in Glasgow between February 1968 and October 1969, each after an evening at the same dance hall in the east end. Helen Puttock’s sister was the only known person who spoke with the man the press called Bible John, and she described him in detail, including a smile.
Detective Superintendent Joe Beattie, who led the inquiry and spent most of the rest of his career on it, kept a plastic replica of the suspect’s teeth and dismissed any suspect whose front teeth did not overlap, the Scottish Daily Express reported in 2021. The drawing police issued in 1969, the first such composite published in a Scottish murder hunt, did not show the overlap. A forensic artist added it for a BBC documentary more than fifty years later. According to VICE, police interviewed about 920 doctors and dentists in the first year, along with hundreds of barbers and tailors. More than 100 detectives took over 50,000 statements; more than 5,000 men were treated as suspects and ruled out, according to the Scottish Daily Express.
A dental description is as fragile as the witness who gave it. Helen Puttock’s sister died in 2010, Beattie in 2000. In 1996 police exhumed a former suspect who had died in 1980 for DNA testing, and the Crown said the evidence was not enough to link him to the case. No one has ever been charged. Police Scotland said in June 2024 that the murders were unresolved, and former senior detectives have said publicly that they do not believe one man killed all three women. Of these cases, the Glasgow case is the one where a dentist’s input was most sought and least tested.
The doubts did not come from a single study. They came from a line of reviews that began in 2009, when the National Academy of Sciences reported on the state of forensic science in the United States. Radley Balko, writing on the report in Reason, summarized its bite-mark section: there was no scientific research showing any particular uniqueness to human dentition; blind comparisons and the use of a second expert were not widely used; and, in the report’s words, "although the majority of forensic odontologists are satisfied that bite marks can demonstrate sufficient detail for positive identification, no scientific studies support this assessment, and no large population studies have been conducted." Experts, it added, diverged widely in evaluating the same mark.
In September 2016, the President’s Council of Advisors on Science and Technology issued a longer report on feature-comparison methods. It reached a blunt conclusion: "PCAST finds that bitemark analysis is far from meeting the scientific standards for foundational validity." Few studies, and no appropriate black-box studies, had tested whether examiners could identify the source of a mark. In those that existed, false-positive rates were "typically above ten percent and sometimes far above," and several study designs were likely to understate the error. The council judged the prospects of making the method valid to be low.
Between the two reports, the Texas Forensic Science Commission acted. On February 12, 2016, after a review prompted by a request from the Innocence Project, it recommended that prosecutors temporarily stop using bite-mark evidence until its scientific validity was settled, and it asked the legal and scientific communities to review earlier Texas cases. Its general counsel, Lynn Garcia, put the central concern as "the ability of human skin to record bite marks with sufficient fidelity" for later analysis. The commission’s role was advisory, and the Texas Tribune noted that its bite-mark recommendations were only that. It was nonetheless the first body of its kind to say so.
The doubts have names attached, and none of them belongs to a serial killer. Steven Chaney of Texas served 28 years for a murder conviction that relied on testimony from two forensic dentists, until the Dallas County district attorney’s conviction integrity unit cleared him. The Texas Tribune named his case in its 2016 report.
Keith Harward’s is better documented. A 1982 killing in Newport News, Virginia, led to his conviction for capital murder, with a life sentence after the jury spared him death. The Virginia Supreme Court reversed that conviction in 1985 over how the capital murder statute should be read, and he was convicted again in 1986. At the trial, according to the Innocence Project, the prosecution leaned mainly on two forensic dentists, Lowell Levine and Alvin Kagey, who said his teeth matched photographs of the evidence. Six forensic dentists in all backed the match at some point in the case, and two requested by the defense agreed. A dentist who had first looked at the dental records of Marines stationed on the aircraft carrier USS Carl Vinson had excluded Harward. He became a suspect months later for unrelated reasons.
DNA testing arranged by the Innocence Project excluded Harward and identified another sailor from the same ship, Jerry Crotty, who had a violent criminal record and has since died. The Virginia Supreme Court granted a writ of actual innocence, and Harward was released in April 2016 after more than 33 years in prison. The Innocence Project counted him as at least the 25th person wrongly convicted or indicted on bite-mark evidence in part. It reported that the American Board of Forensic Odontology had changed its guidelines to say the technique cannot identify a biter.
Look back at what the serial cases above had in common with the Harward case: a dentist, plaster casts, a jury that could not itself assess whether skin holds a mark. The difference is that, in the serial cases, other evidence existed. In Bundy’s case the hair comparison, a survivor’s account and, in the Leach trial, fibers and a witness. In Alcala’s, DNA. In Boden’s, what else the jury heard has not been reported. None of that cleans up the dental testimony, which was received as science. But it explains why the discipline was not shaken by the cases it appeared in. A man could be rightly convicted with bad evidence beside good, and no one would see the bad evidence until a DNA result arrived for someone else.
The doubts have a limit worth stating. The reviews did not question that teeth, X-rays and dental charts can identify a person. That use is how Hong Kong named four women in 1982, and how dentists continue to work in disaster and cold-case identification. A forensic odontologist was among the specialists a Montana sheriff’s office consulted over decades in an unsuccessful effort to name a young victim; the name came in 2021, from DNA and genealogy, not from teeth. The dental record is a document written for a living patient, compared with a person who is no longer there to speak. The bite mark is something else: a reaction recorded in skin, that has to be read through swelling, movement and time, and compared with a mouth that may have changed.
The courts that accepted the second use in the 1970s and 1980s did what courts do, which is to ask whether the relevant community of experts accepted it. They did. It was the community itself that had never been asked to prove it.
• 1968 to 1969: Three women are killed in Glasgow; the unidentified man the press calls Bible John is never charged.
• June 12, 1971: Cornelia Crilley is killed in New York City.
• 1972: Wayne Clifford Boden is convicted of killing four women in Montreal and Calgary; Gordon Swann’s dental evidence is later described as the first of its kind in North America.
• 1977: Boden applies for and receives an American Express card.
• June 25, 1979: Ted Bundy’s Miami trial opens.
• July 18, 1979: Richard Souviron testifies that Bundy’s teeth match a mark.
• July 24, 1979: The jury convicts Bundy of two counts of first-degree murder, three of attempted murder and two of burglary.
• 1982: Four women are killed in Hong Kong between February and July; Lam Kor-wan is arrested on August 17.
• March to April 1983: Lam is tried; the jury convicts him of four counts of murder; he is sentenced to death on April 8.
• June 21, 1984: Florida Supreme Court affirms the Chi Omega convictions and rejects Bundy’s challenge to the bite-mark testimony.
• May 1984: Boden escapes a day-pass escort and is recaptured about 36 hours later.
• August 1984: Lam’s death sentence is commuted to life imprisonment.
• January 24, 1989: Bundy is executed.
• 1996: Police exhume a former Bible John suspect for DNA testing; the Crown says the evidence is not enough.
• 2003: New York detectives take an impression of Alcala’s teeth under a warrant.
• February 2009: National Academy of Sciences report on forensic science is issued; commentary singles out its bite-mark section.
• February 25, 2010: Alcala is convicted in Orange County of five murders and one kidnapping.
• December 2012 and January 7, 2013: Alcala pleads guilty to two New York murders and is sentenced to 25 years to life.
• February 12, 2016: Texas Forensic Science Commission recommends a pause on bite-mark evidence.
• April 2016: Keith Harward is released after more than 33 years.
• September 2016: PCAST reports that bitemark analysis is far from foundationally valid.
• July 24, 2021: Alcala dies of natural causes.
• June 2024: Police Scotland says the Bible John murders are unresolved.