Bite marks sit at the intersection of medicine, forensic science, and criminal law, and in each of those fields they carry a different kind of weight. In emergency rooms, a bite wound raises immediate questions about infection risk and tissue damage. In courtrooms, a bite mark on skin has been presented as evidence linking a suspect to a crime. That forensic use, once treated as credible expert testimony, has been largely discredited by the scientific community over the past two decades, with major government reviews concluding that bite mark comparison lacks the foundational validity needed to identify a specific person.
What a Bite Mark Actually Looks Like
A human bite mark on skin typically appears as two opposing U-shaped or oval arches, roughly mirroring the upper and lower dental arches. Between the arches, a central bruise sometimes forms where soft tissue was compressed during the bite. The pattern can range from a faint, diffuse bruise to a clearly defined impression showing individual tooth marks, depending on the force applied, the body location, and how much time has passed.
Animal bites look quite different. Dogs, the most common source of animal bite injuries, have four prominent canine teeth that sit well above the height of their incisors. A dog bite tends to leave pairs of deep, triangular puncture wounds rather than the broad compression pattern characteristic of human bites. Dogs also tear flesh rather than simply compressing it, so the resulting wound usually involves lacerations along with punctures.
One area of genuine difficulty is distinguishing human bites from bites by medium-sized dogs. Research comparing intercanine distances found that dog measurements were on average a few percent wider than human measurements overall, but that dogs weighing roughly nine to twenty-three kilograms produced bite widths that overlapped significantly with the human range. For bites in that size range, intercanine distance alone cannot reliably determine whether the bite came from a person or a dog.
Skin Distortion and Why Bite Marks Change
One of the fundamental problems with reading bite marks on skin is that skin is not a stable recording surface. It stretches, contracts, swells, and bruises differently depending on a person’s age, weight, the body part bitten, and how much time has passed since the injury. Research measuring how much bite mark impressions distort on skin found that individual tooth width measurements could shift by as much as about 54%, and intercanine distances could distort by roughly 42%.
That level of distortion is enormous. It means that a bite mark photographed hours after an assault may bear little geometric resemblance to the teeth that made it. The distortion also is not uniform across the dental arch: some teeth may appear dramatically shifted while adjacent ones stay relatively stable, creating a warped version of the original impression. Distortion increased with the subject’s age and weight, making it even less predictable across a population.
The Uniqueness Assumption That Was Never Proven
Forensic bite mark analysis rests on two core assumptions: that every person’s dentition is unique, and that this uniqueness transfers reliably onto skin. Neither assumption has held up under scrutiny. A systematic review of the available research on dental uniqueness examined thirteen studies and found that none could support a conclusion that human dentition is unique. Nine of the studies explicitly failed to find uniqueness. Four claimed to have succeeded, but were found to have methodological problems that prevented them from actually supporting that conclusion.
Cadaver studies have tested this more directly by having similarly aligned sets of teeth bite into human skin and then asking examiners to determine which set made the mark. Researchers found that examiners had difficulty distinguishing the correct biter from individuals with similarly aligned teeth. In some cases, an incorrect biter appeared to correlate better with the mark than the person who actually made it.
Three-dimensional analysis has not rescued the uniqueness claim either. A study that assessed dental uniqueness using 3D measurements of incisal edges found no scientific evidence to support the idea that the anterior dentition is unique in the way forensic bite mark analysis requires.
Examiners Cannot Agree With Each Other or Themselves
Even if bite marks could theoretically carry useful information, the people interpreting them would need to do so consistently. They do not. Studies of forensic odontologists examining the same bite mark evidence have repeatedly found poor agreement. One qualitative study found that practitioner agreement was “at best fair,” with examiners offering widely varying opinions about the origin, circumstances, and characteristics of the same wound. Counterintuitively, more experienced practitioners, those with over ten years in forensic odontology, tended to agree with each other less than those with ten years or fewer.
A separate study asked twenty-three forensic odontologists to evaluate bite mark cases, then repeated the request eight weeks later. The results showed inconsistency not just between different odontologists but within the same individual over time. Even experienced examiners changed their own opinions when re-evaluating the same evidence weeks later.
This is not a minor calibration problem. When experts examining the same photograph cannot agree on whether a mark is even a bite, let alone whose bite it might be, the entire framework of presenting bite mark comparison as expert testimony falls apart.
Error Rates in Controlled Studies
Controlled tests of bite mark identification have been conducted since the 1970s, and the results have consistently been poor. Early studies by forensic dentist David Whittaker used exemplar bites made on pigskin, which holds impressions better than living human skin and would therefore be expected to give examiners an easier task than real cases. Even so, incorrect identifications ranged from 24% under ideal conditions to 91% when identifications were made from photographs taken twenty-four hours after the bites were made, which is more typical of how real bite mark comparisons are performed.
Later studies found similar problems. In one test where board-certified forensic dentists evaluated bite marks on pigskin, agreement with their own prior judgments given three months earlier dropped as low as 65%. False-positive responses, meaning an examiner linked a bite to someone who did not make it, averaged about 16% and ran as high as roughly 46%. False negatives, where the examiner failed to identify the actual biter, averaged 25% and reached as high as about 71%.
Even when researchers used dental wax, a far better recording medium than skin, and employed 3D imaging with quantitative comparisons, false-positive rates still reached 15%. The pattern across decades of testing is consistent: bite mark identification performs poorly in controlled settings, and those settings are typically more favorable than the conditions examiners face in real criminal cases.
The PCAST Verdict
In 2016, the President’s Council of Advisors on Science and Technology (PCAST) issued a landmark report reviewing the scientific validity of several forensic disciplines. The council concluded that bite mark analysis does not meet the scientific standards for foundational validity. It went further, stating that the prospects of developing bite mark analysis into a scientifically valid method were poor, and it advised against government investment in research aimed at establishing such validity.
This was not the first high-level critique. A 2009 report by the National Academy of Sciences had already raised serious concerns. But the PCAST report was more emphatic: it did not merely say the science was underdeveloped or needed refinement. It said the entire enterprise of identifying individuals through bite mark comparison lacks a scientific basis and is unlikely to ever acquire one. The reasoning is straightforward. The underlying premises, that dentitions are unique and that skin faithfully records those unique features, have not been demonstrated despite decades of attempts.
Wrongful Convictions Linked to Bite Mark Evidence
The unreliability of bite mark evidence is not a theoretical concern. A scoping review of cases in the U.S. National Registry of Exonerations identified twenty-six wrongful convictions in which forensic odontology was flagged as false or misleading evidence. In only two of those cases, about 8%, was the misleading forensic evidence the sole contributing factor. In four cases, about 15%, the bite mark evidence was accompanied by three additional contributing factors. Official misconduct was detected in nineteen of the twenty-six cases, and perjury or false accusation appeared in sixteen.
These numbers tell a layered story. Bite mark evidence rarely acted alone in producing a wrongful conviction; it typically appeared alongside other failures in the justice system, from prosecutorial misconduct to false witness testimony. But the bite mark testimony lent a veneer of scientific authority that may have made other weak evidence seem more credible to juries. When an expert witness tells a courtroom that a wound “matches” a defendant’s teeth, it sounds precise and scientific, even though the underlying method cannot deliver that precision.
Cognitive Bias in Bite Mark Analysis
Part of the problem with bite mark analysis is that the process is ripe for cognitive bias. Forensic odontologists typically receive case information before examining the evidence: they may know who the suspect is, see emotionally charged crime scene photographs, or be told that other evidence points to a particular person. Research has shown that this kind of contextual information influences bite mark comparisons. When examiners were provided with contextual information, their decisions on ambiguous bite marks shifted accordingly.
One study found a somewhat surprising wrinkle: when participants were shown highly emotional images and subliminally primed with words like “same” and “guilty,” they actually made fewer matches relative to a control condition. But contextual effects in general clearly influenced decision-making, and dental experience played a role too, with dental students making more matches as the experiment progressed regardless of the context they were given.
A broader review of biasing influences in forensic odontology concluded that the current practice of bite mark analysis is “rich in sources of potentially biasing influences.” Unlike DNA analysis, which has built-in safeguards like blind testing protocols, bite mark comparison has historically been performed by a single examiner who knows the case details. That setup practically invites confirmation bias.
What Bite Marks Can Actually Offer Forensic Investigations
If bite mark pattern matching is unreliable, does that mean bite marks are forensically useless? Not entirely, but their value lies in what can be recovered from them biologically rather than what can be read from their shape. Saliva deposited during a bite contains cells, and those cells contain DNA. Research has shown that DNA can be extracted from saliva deposited on skin, with short tandem repeat profiles successfully amplified from four out of five samples in one study. The DNA yield from skin-deposited saliva is considerably lower than from a direct saliva sample, but modern forensic DNA techniques are sensitive enough to work with small quantities.
Swabbing a bite mark for DNA is standard practice in many jurisdictions now, and the resulting profile, if recovered, provides far stronger identification evidence than any visual comparison of tooth patterns ever could. A DNA profile can be searched against databases, compared directly to a suspect’s sample, and evaluated with well-established statistical frameworks for determining the likelihood of a match.
Emerging research has also explored whether the microbiome, the community of bacteria transferred in saliva during a bite, could serve as an additional identification tool. Studies have found that the Streptococcus genus is particularly relevant in bite mark microbiology from a forensic perspective, and that genomic analysis of these bacterial communities may eventually help associate a bite mark with a specific individual. Integrating microbiome analysis with conventional DNA profiling could improve the accuracy of bite mark evidence interpretation, though this work remains in relatively early stages.
Medical Treatment of Bite Wounds
Outside the forensic context, bite marks are a medical problem. Human bites are particularly prone to infection because human saliva contains a broad mix of aerobic and anaerobic bacteria. The risk of infection after a bite wound is estimated at 10 to 20%, and roughly 30 to 60% of bite wound infections involve a mix of aerobic and anaerobic organisms.
Human bites are frequently underrecognized in emergency departments. Some present as obvious bite wounds, but others are disguised. The classic example is a “fight bite,” an injury to the knuckles sustained when someone punches another person in the mouth. The small laceration over the metacarpophalangeal joint may not look like a bite, but it carries the same bacterial inoculation risks and can lead to serious tendon or joint infections if not treated promptly.
Treatment principles for bite wounds emphasize thorough irrigation, wound cleaning, and evaluation for deeper structural damage. Prophylactic antibiotics are not universally recommended; they are reserved for wounds considered at high risk of infection based on the wound type, its location, and patient characteristics. Hand wounds, deep punctures, and bites in immunocompromised patients generally warrant prophylaxis. Liberal irrigation and elevation of the injured body part remain the cornerstones of initial management, and early treatment improves outcomes substantially.
Bite wounds can also transmit infectious diseases. Hepatitis B, hepatitis C, and HIV have all been documented in association with human bites, though transmission is rare for HIV. Tetanus prophylaxis should be updated if the patient’s vaccination status is not current.
Three-Dimensional Documentation and Its Limits
One area of active development is three-dimensional documentation of bite marks, which aims to reduce some of the problems inherent in standard two-dimensional photography. Conventional photographs of bite marks on curved body surfaces introduce distortion, and lighting angle can dramatically change which features are visible. Three-dimensional scanning and photogrammetric approaches can capture the surface geometry of a wound without these artifacts, preserving metric measurements in three-dimensional space.
A pilot study evaluating the use of intraoral scanners and photogrammetry for forensic documentation of bite injuries found that both methods produced analyzable digital reconstructions suitable for forensic archiving. The intraoral scanner, typically used inside a patient’s mouth for dental impressions, was applied in an unconventional extraoral role to directly scan skin lesions. Photogrammetry, using a standard camera with specialized software, generated accurate 3D models from multiple photographs.
These technologies improve the quality of documentation, which is genuinely valuable for preserving evidence. But better documentation does not fix the underlying scientific problems with bite mark comparison. A perfectly captured three-dimensional model of a bite mark still records a distorted impression on a surface that changes over time, and there is still no validated method for reliably matching that impression to a specific set of teeth. The technology is solving a documentation problem, not a science problem.
How Animal Bites Differ in Practice
While the forensic identification controversy centers on human bite marks, the medical and investigative distinction between human and animal bites matters in its own right. Dog bites account for the vast majority of bite injuries treated in emergency departments. The wound patterns differ from human bites: dog bites typically produce puncture wounds from the canines along with tearing and avulsion injuries, while human bites produce more compression-type injuries with broader, shallower marks.
In forensic cases involving deceased individuals found outdoors, animal scavenging can produce bite marks that complicate the interpretation of injuries. Determining whether marks on a body were inflicted by a human attacker, a domestic dog, or wildlife is a genuine forensic question, and it is one where the general shape and character of the wound can actually provide useful information. This is a different task from trying to identify a specific individual, and it is one where bite mark analysis is on somewhat firmer scientific ground. Telling a human bite from a dog bite involves broad categorical differences in tooth morphology and wound pattern rather than the fine-grained individual identification that the research has shown to be unreliable.
Research comparing human and canine dental arches found that while the variability of intercanine measurements was similar in both species, dog measurements were on average larger. The overlap zone for medium-sized dogs, however, means that even categorical human-versus-animal determinations can be uncertain in some cases.
Where the Science Stands Now
The trajectory of bite mark evidence in the legal system has been steadily downward. Several U.S. states have revisited convictions that relied on bite mark testimony. The Texas Forensic Science Commission recommended a moratorium on the use of bite mark comparison in criminal prosecutions. Professional organizations within forensic odontology have been divided, with some members defending the practice and others calling for its abandonment.
The research picture is unambiguous, though. The two foundational claims of bite mark identification, that human dentition is unique and that skin faithfully records that uniqueness, have both been tested repeatedly and have both failed. The empirical record is small, and within that small body of work, results have consistently gone against the claims of forensic dentists. As one comprehensive review summarized it, the research is “noteworthy, first, for how little of it there is and, second, for how much of what does exist refutes the claims of forensic dentists regarding their ability to identify the source of a bite mark.”
What remains useful about bite marks in forensic settings is the biological material they contain. DNA recovery from bite sites is a proven technique with a validated scientific framework behind it. Microbiome profiling may add another layer of identification evidence in the future. And three-dimensional documentation preserves wound evidence in a form that is more accurate and more durable than traditional photography. The physical pattern of the mark itself, however, is not a reliable road back to the person who left it.