6
Human Tissues
6.1 The Outer Body Surface
6.1.1 Skin
Skin covers the whole surface of our body and is the largest organ in terms of both weight and surface area. It varies in thickness from about 0.5 mm on the eyelids to 4 mm or more on the heels. Repetitive abrasion of the skin surface results in the formation of calluses, the distribution of which, with the condition of the fingernails, provides an indication of a person's trade or activities. For example, manual labourers develop thick calluses over their palms and fingers and their fingernails become chipped and ragged. At the opposite extreme, musicians develop calluses on highly localised regions and string instrument players may allow certain fingernails to grow long to facilitate plucking. Violinists develop calluses (‘Garrod's pads’) on the dorsal surface of their left second and third fingers over the proximal interphalangeal joints. Similarly, wind players develop calluses on the mid‐portion of the upper lip.
Structurally, skin is divided into two regions: the outer epidermis and the thicker inner, dermis. The epidermis is composed of several layers of cells that constitute a keratinised stratified squamous epithelium. Roughly translated, this means that the cells contain the fibrous protein keratin, have the appearance of a wall of bricks (stratified), are flat (squamous), and form the outer surrounding layer (epithelium). Embedded within the epidermis are melanocytes: round cells with long, slender projections that contain the pigment melanin and are responsible for giving skin its coloration. Where there is an overgrowth of melanocytes, a round, flat or raised area called a nevus or mole is formed. The distribution of these can be useful for identification. Beneath the epidermal layers lies the dermis that is largely composed of connective tissue, but also contains blood vessels, nerve endings, glands and hair follicles.
Breast implants are common, either for cosmetic purposes or as part of reconstructive surgery following cancer treatment. For example, there were over 9600 breast implant operations in the UK in 2015; by contrast, there were 796 breast reduction operations in men. Breast implants can provide forensic evidence. For example, in USA in 2009, the body of the model Jasmine Fiore was identified from the serial number on her breast implants. Ms Fiore had been murdered by her husband who then went to great lengths to disguise her identity before disposing of her body. He cut off her fingers and pulled out her teeth, but the unique breast implant serial number conclusively proved who she was. Obviously, DNA would also have provided proof of her identity, but it is not always possible to extract a full Short Tandem Repeats (STR) profile. All surgical implants carry a unique serial number that allows tracing should there be complications following surgery and breast implants are no exception. Occasionally, implants are also used to commit crimes and there have been several cases of women being arrested with cocaine hidden inside their breast implants: in one case the woman was carrying 1.7 kg of cocaine in her breasts.
6.1.2 Tattoos and Piercings
Men, and to a lesser extent women, have adorned themselves with tattoos for thousands of years, but recently a whole industry and lifestyle has built up around them. Although normally found on the outer body surfaces, some people are tattooed inside the lips or on the gums. In Western societies, men and women of all ages and degrees of affluence wear tattoos. Some estimates suggest that 20% of the UK population has at least one tattoo and this rises to 33% of young adults. The figures for many other European countries are similar and they are even higher in the USA. The prevalence of piercings such as rings, studs, and barbells (other than in the earlobe) is often put at around 10–14% and they are far more popular among women than men. The frequency and variability of tattoos means that they provide identifying features of individuals (Figure 6.1). In addition, the style and inks used can indicate where the tattoo was obtained – and hence the movements of the individual (Miranda 2015). Many of the pigments used in making tattoos were developed for other applications, such as in the plastics and the automotive industries, and have not been safety checked for their effect on human health. There are many reports of allergic reactions to pigments and even the formation of granulomas around them. Whether some pigments cause cancer is not yet proven. In the future, legal cases will probably arise from people claiming that a tattoo caused them harm and this will require the identification of the inks used.
Figure 6.1 Tattoos that have a limited appeal help in the identification process. For example, it is likely that only an employee or graduate of Liverpool John Moores University would sport a tattoo of its stylized ‘Liver Bird’. Allegedly, only one man has this distinctive tattoo (it is not the author).
Tattoos can help identify dead bodies if conventional means are not possible; for example, if a victim suffers extensive damage to their face and/or their fingers. Tattoos remain visible even after a body suffers severe burns and are therefore useful when identifying victims of explosions and fires. Should they become obscured during decomposition, tattoos become visible after treatment with 3% hydrogen peroxide (Haglund and Sperry 1993). Even if tattoos are irretrievably lost, their former presence can be predicted from the presence of pigments within nearby lymph nodes (Hellerich 1992).
Most body piercings are made of surgical steel or niobium and to a lesser extent titanium or gold. These materials remain with the body long after surrounding soft tissue decays and therefore provide forensic indicators both in life and death. Many women and, increasingly, men, have ear jewellery and these are also useful indicators. Even if the ear jewellery was not worn at the time of death or was stolen afterwards, scar formation remains where the ear was pierced. Ear jewellery therefore helps with the identification process and sometimes it is distinctive enough to be traced to the place they were acquired. The lips, nose, and tongue are popular sites for piercings. However, unless the piercings are extensive, their presence seldom indicates more than a desire for adornment. Piercings to the nipples and genitalia are less common and usually acquired for sexual motives. They therefore give an indication of the person's lifestyle.
Tattoos can indicate a relationship, lifestyle preference, mental state, service in the armed forces, a means of covering up intravenous drug use, or gang membership (Mallon and Russell 1999). Names are not a good indication of a relationship, because the tattoo may remain long after the object of affection becomes a distant memory. Organised gang members sometimes have extensive and elaborate tattoos. The Yakuza in Japan are well‐known for their highly decorative tattoos. The facial tattoos of some South American drug gangs helps enforce loyalty because, once tattooed, a person is unable to join or avoid recognition by a rival gang. However, owing to their popularity, it is important to keep an open mind when interpreting tattoos. This is because tattoos can represent a person's fantasy about themselves rather than the reality of their existence or be the consequences of a ‘good idea’ at a time of excessive inebriation.
Because tattoos provide distinguishing features, security services are developing algorithms that utilise them for automatically identifying individuals from surveillance images. However, this is currently proving challenging (Wilber et al. 2014; Xu et al. 2016).
6.1.3 Scars
Permanent scar tissue forms when there is irreversible injury to a tissue. It is characterised by the deposition of collagen fibres, localised remodelling, and reduced blood supply. Scar tissue forms in our internal organs and tissues as well as on the surface of our skin. Whether the scars result from accidents or medical operations, their presence is useful for identifying an individual and deducing past events. However, scars are seldom so characteristic that they provide definitive identification. The effectiveness of the healing process is affected by numerous factors, such as the nature of the wound, the part of the body, general health, and age. Interestingly, wounds inflicted during early gestation heal rapidly and perfectly. A patchwork of healed and partially healed cuts is a common feature of self‐harm. These wounds are always restricted to sites that the victim can reach and usually in places that can be hidden by clothing.
6.1.4 Fingerprints
In the 1920s, Kristine Bonnevie (the first female professor in Norway) suggested that when a developing human embryo is about 10 weeks old, the basal layer (the deepest layer of the epidermis) at the tips of the fingers grows faster than the surrounding upper epidermal layers and the lower dermal layers. This results in the basal layer being stressed and thrown inwards into a series of folds that manifest on the surface as ‘friction ridge skin’. Similar skin is also found on the palms of the hands, the soles of the feet, and the lower surface of the toes. The patterns of loops, whorls, and arches are unique to every person. This hypothesis of friction skin formation has never been conclusively proven, but computer‐modelling studies lend it strong support (Kücken 2007).
After birth, we already have our own unique fingerprint pattern that remains the same for the rest of our life. As we grow bigger the fingerprint ridges do not change in shape but become further apart – indeed, there is a relationship between height and ridge width. Once we enter into old age, the outer epidermis thins and changes in the connective tissue result in the skin losing its elasticity and strength. In addition, there is a reduction in the amount of subcutaneous layer of fat, and the sebaceous glands produce less sebum (body oils). The latter is noticeable in women after menopause, and results in the skin becoming dry. Together, these factors result in friction ridges becoming less pronounced and the formation of poor‐quality fingerprints.
Minor cuts, burns, or bruises cause temporary defects to the ridge pattern but once healed, the normal fingerprint pattern is restored. Deeper injuries, such as serious burns, cause permanent scar tissue – and this is also a good identifying characteristic. Cosmetic surgery cannot erase the fingerprint pattern, because to be effective it requires the removal of so much tissue that a person would be unable to use their hands effectively. However, manual labourers such as bricklayers, cement workers, and those who routinely work with highly abrasive substances, can wear away the surface of their friction ridges and therefore leave poorly defined prints. Several studies have attempted to link personal characteristics to fingerprints, but their reliability is limited. For example, Acree (1999) and Gungadin (2007) both found that men tend to have a lower fingerprint ridge density than women, but their ‘cut off’ point differed: Acree found that a pattern density of ≤ 11 ridges per 25 mm2 was most likely to belong to a man, whilst a density of ≥ 12 ridges per 25 mm2 was most likely to belong to a woman; by contrast, Gungadin's figures were ≤ 13 ridges per 25 mm2 for men and ≥ 14 ridges per 25 mm2 for women. Whether this relates to height differences between men and women is not known.
6.1.4.1 Types of Fingerprint
The study of fingerprints is known as dactyloscopy and Giuliano (2016) provides an excellent account of modern‐day practice. It has been used as a forensic tool since the nineteenth century, and as early as 1906 a conviction was upheld in England solely on the basis of fingerprint evidence. Even identical twins do not have the same fingerprints and courts therefore accept them as unique identifying characteristics. This makes fingerprints an ideal forensic test. Fingerprints are often found at crime scenes and they can be quickly, easily, and cheaply collected, processed, and analysed. And they provide an extremely reliable result.
Fingerprints are divided into three types: plastic, visible, and latent (Figure 6.2 a–c). Plastic fingerprints form when a person touches a soft or semi‐solid substance such as soap or unset putty. This results in a shallow three‐dimensional record of the friction ridges that is either transient or long‐lasting, depending on the substance and the circumstances. For example, plastic fingerprints are present on 2500‐year‐old clay seals recovered from the ruins of the Mesopotamian city of Ur. Visible fingerprints, as their name suggests, are those that can be seen without further enhancement. They result from a combination of sweat and the oily secretions from the skin glands deposited on a contrasting surface, thereby making the print visible. Alternatively, they form when a hand covered with substances such as blood, ink, or engine oil is pressed against a contrasting surface. Latent fingerprints need enhancing before they become visible. This is because sweat and body oils are colourless, so unless the prints are on a contrasting surface, they cannot be seen easily. Washing one's hands before handling an object reduces the quality of latent fingerprints, because it removes the sweat and body oils. Similarly, young children leave poor‐quality fingerprints that degrade quickly, because they produce less body oil than adults. Latent prints also form when a layer of material, such as fine dust, is removed, thereby leaving a ‘negative impression’.
Figure 6.2 Different types of fingerprints: (a) plastic fingerprint left in plasticene (note the fingernail impression); (b) visible fingerprint left on a dusty jar; and (c) latent fingerprint rendered visible using ninhydrin spray.
6.1.4.2 Characteristics of Fingerprints
Fingerprints are classified according to their distinctive patterns of loops, arches, and whorls (Figure 6.3). However, identifying a fingerprint is not easy and relies on comparing two sets of prints in terms of their characteristic features and the spatial relationship of these to one another in a three‐stage process:
- matching the overall pattern of friction ridges,
- matching the characteristics features of those ridges, and
- matching unique features such as the number and distribution of sweat pores and the three dimensional shape of individual ridges.
Figure 6.3 Fingerprint characteristics showing the distinction between arches, loops, and whorls. Composites are prints exhibiting two or more features.
Source: Reproduced from Howe (1950).
There is little consensus as to how much similarity between two sets of prints is necessary for them to be considered a ‘match’. In the UK, a 16‐point minimum was required for many years, whilst in France it was 17 and in the USA each state set its own standards. Subsequently, it was accepted that there is little scientific or statistical basis for any numerical standard. In 2001, the UK adopted a non‐numerical approach that incorporates ‘objective criteria’. Computer programs speed up the matching process and police forces in the UK use a biometric database called IDENT1. Most police forces use electronic live scan technology for recording prints, then upload them onto the IDENT1 database. Currently, in England and Wales, the police take fingerprints from anyone arrested for a recordable offence. The IDENT1 database is constantly expanding and at the time of writing contained over 1.8 million unidentified scenes of crime marks, 7.9 million individual ten‐print sets, and over 7 million palm prints. In America, the FBI operates a similar system called the Integrated Automated Fingerprint Identification System (IAFIS) and has access to over 75 million fingerprints.
6.1.4.3 Fingerprinting the Living and the Dead
For many years, fingerprints were taken by asking a person to dip their fingers in a thin layer of printer's ink and then press their fingers onto fingerprint card. Each finger was recorded separately and the procedure was time‐consuming. Nowadays, fingerprints and palm prints are usually taken electronically using ‘live scan’ digital fingerprinting consoles – the ‘donor’ places their hand on a glass platon and the fingerprints are recorded electronically. The digital images are then entered instantly into a searchable database such as IDENT1. Consequently, within a short time, one can check whether a person is lying about their identity or that prints recovered from a crime scene resemble those of someone already on the database. However, the computer program's main purpose is to speed up a search: final confirmation of a match always relies on physical observation of the prints by a fingerprint officer.
Fingerprints are an excellent means of identifying an unknown body, provided that it is fresh – in which case it is printed in the same manner as with a living person. However, once a body starts to decay, then the operation becomes more difficult. For example, the skin becomes wrinkled, especially after exposure to water or damp, whilst fluid exudates and fungal hyphae prevent ink binding and interfere with scanning. If the epidermal layer sloughs off from the dermis (a common occurrence in bodies found in water), one can insert one's own finger or a mould inside the cast skin to make a print. However, it is a good idea to make a photographic record of the fingers and hands before handling them in case of subsequent damage. With modern high‐resolution digital cameras, one can take images that are good enough to undertake fingerprint examination. It is likely that 3D scanning technology (e.g. FlashScan3D) will become widespread, because this allows contactless recordings. Because a 3D surface map is made, it records the natural curvature of the fingers and the image is then transformed to a 2D image for comparison with images on a computer database.
Mummification and burning results in the skin shrinking, wrinkling, and drying. In extreme cases, this makes fingerprinting impossible. Dried fingers can be cleaned of grease, dust, and carbon particles and then coated with liquid latex to produce a cast that is then placed onto a support in the shape of a human finger – the cast is then manually printed or scanned (Porta et al. 2007). This approach provides a permanent record that can be examined by light and scanning electron microscopy.
6.1.4.4 Detection of Fingerprints
Searching for fingerprints is a specialist skill and should be undertaken with extreme care owing to the ease with which prints are damaged or lost if the incorrect method is chosen. All fingerprints, whether visible or latent, are given a unique code and their location recorded. First, all visible prints are located by passing a strong beam of light over the scene or object. The angle of the light is altered to make the prints ‘stand out’. Sometimes latent fingerprints can be rendered visible simply by viewing them under different lighting conditions – oblique lighting is often effective. The use of lasers tuned to an ultraviolet wavelength has also been suggested (Akiba et al. 2007). However, the use of special fingerprint powders or sprays is usually required, the choice of which depends on the nature of the print, the substance it occurs on, and the personal preferences of the investigator. Powders stick to moist or oily parts of the print. They are employed on non‐absorbent surfaces such as plastic or glass and tend to be brushed on (Sodhi and Kaur 2001). They usually consist of two components: an adhesive such as starch and a colourant such as an inorganic salt or an organic dye that adsorbs onto the adhesive. Ideally, ‘once only’ brushes should be used, owing to the potential for fingerprint brushes to pick up and transfer DNA (Van Oorschot et al. 2005). ‘Magnetic’ powders are useful on textured surfaces, such as leather. These powders are not themselves magnetic, but they contain iron and are applied with a magnetic wand. Magnetic attraction sticks the powder to the wand and this is dragged slowly across the surface of an object. The components of fingerprints exert a stronger attractive force than the magnet, so the particles remain behind and thereby render the prints visible.
There are various means of chemically identifying fingerprints and the choice depends on the surface and whether the objects are required for other forensic examinations – some tests interfere with one another. One of the most commonly used reagents is ninhydrin. This reacts with the amino acids found in sweat (mainly alanine, aspartic acid, glycine, lysine, ornithine, and serine) to form a pink to reddish‐purple coloured compound. The ninhydrin reagent is sprayed or brushed on, or used as a dip. This method is effective for prints found on paper, wood, and other porous surfaces. It is possible to speed up the process of development of the prints by conducting the procedure at 50–70 °C and 60–80% relative humidity. This is known as accelerated ninhydrin process fingerprint development.
Silver nitrate is useful for identifying latent prints on brass objects such as ornaments and ammunition cartridges, but also porous surfaces such as wood. The silver nitrate reacts with sodium chloride in sweat to form silver chloride: this is light sensitive and under the influence of UV light decomposes to silver and chlorine. Consequently, the print becomes visible as a greyish or brownish stain – although this fades with time. Unfortunately, silver nitrate will react with any source of sodium chloride and this may obscure the print. Consequently, it is not appropriate for detecting prints on the brass fittings on a sea‐going yacht. Iodine fumes react with the oily components of latent fingerprints, thereby making them visible as a brownish colour. However, iodine prints are unstable and must be photographed immediately or ‘fixed’ to preserve them.
In a novel approach to locating fingerprints, King et al. (2015) used powders prepared from spirulina. Spirulina is derived from cyanobacteria (blue‐green algae) and marketed as a dietary supplement. Spirulina extracts contain a lot of chlorophyll and anthocyanin, both of which fluoresce when illuminated with infra‐red light. Spirulina powders are effective at identifying latent prints, especially those on difficult ‘busy’ backgrounds, such as banknotes. Spirulina powder is also cheap and non‐toxic.
Exposing latent fingerprints to ethyl or methyl cyanoacrylate (superglue) vapour renders the prints visible through the formation of a white deposit. The prints are then enhanced by applying fluorescent dyes or by observing them in UV light. This method is simple and cheap and works best for prints on glass, china, and other non‐porous surfaces. It is also useful for detecting latent prints on cartridge cases. It is done in a closed container at room temperature. This is partly to protect the operator from the fumes, but also to ensure that a high humidity of about 80% is maintained.
Although prints produced by a bloody hand (or foot) are visible, it is often necessary to treat or enhance them before an accurate record can be made. First, the print is fixed with 2% w/v sulphosalicylic acid and then covered with a 0.5% w/v solution of 2,2′‐azino‐di‐[3‐ethylbenzizolinesulphonate] diammonium salt (ABTS). Immediately before use, hydrogen peroxide is added to the ABTS solution to act as an oxidant. The haemoglobin in the blood catalyses the breakdown of the hydrogen peroxide to produce reactive oxygen radicals and these bring about the conversion of ABTS to a green coloured compound. The print is then washed, dried, and photographed. In the past, 3.3′‐diaminobenzidine (DAB) was used to enhance bloody fingerprints but it is carcinogenic. Therefore, its continued use cannot be recommended. ABTS, by contrast, is equally sensitive and safe.
Vacuum Metal Deposition (VMD) involves evaporating metal ions within a high vacuum and allowing them to redeposit upon the evidence item. The technique employs either a sequence of metals, such as gold followed by zinc, or a single metal (e.g. silver), depending upon the nature of the substrate. In the case of gold/zinc, first, a layer of gold is deposited over the whole sample. This layer is extremely thin and the gold atoms sit beneath the fingerprint ridges rather than on top of them. The sample is then exposed to zinc ions, but these can only attach where the gold ions are uncovered. Therefore, they do not attach to the fingerprint ridges and a negative image of the fingerprint is revealed. VMD is particularly good at revealing latent fingerprints on fabrics (Figure 6.4a–c) and fired rifle and shotgun cartridges. It has an added advantage of not interfering with subsequent DNA extraction and analysis. However, the evidence items must be transported to the laboratory, because the technique employs relatively large machines.
Figure 6.4 Vacuum Metal Deposition is extremely good at revealing latent prints on fabrics. These images relate to an alleged case of smothering. On one side of the fabric can be seen the hand (a) and fingerprints (b), whilst on the opposite side is a print of the victim's face (c).
In addition to these methods, many other chemicals are used to render fingerprints visible. Sometimes a print is subjected to more than one detection method, in order to obtain as accurate a record as possible. Fourier transform infrared (FTIR) chemical imaging can enhance the quality of a fingerprint image captured from either untreated prints or those revealed using techniques such as cyanoacrylate fuming (Tahtouh et al. 2007). The machine collects images digitally as an array of pixels (just as in a digital camera), identifies the spectra of each pixel at all spectral frequencies, and thereby identifies the chemicals present. FTIR imaging therefore combines data on spatial information (i.e. images) with data on chemical composition with a high degree of resolution. The process provides images of fingerprints on surfaces that it would otherwise prove difficult to resolve print ridges (e.g. banknotes with complex patterns) and also identifies the presence of chemicals of forensic interest, such as drugs or explosives within the print pattern.
Once a print is located and rendered visible, it is photographed. The print may then be left in situ or lifted and stored separately from the object it was found on. If it is preserved on the object, the print is protected by clear tape, varnish, or lacquer. There are several means of lifting a fingerprint, but it is best to use special fingerprint tape because it has a smooth adhesive surface. The tape is pressed against the print and then lifted, after which it is pressed into a ‘lift card’. Lift cards have a smooth surface and are usually transparent. The prints are then scanned or photographed again and stored.
6.1.4.5 Problems with Fingerprint Analysis
Although everyone has their own unique fingerprints (the chances of two persons having identical fingerprints is 1 in 64 billion), correlating the prints found at a crime scene with those of a suspect is not a simple matter. For example, the prints may be smeared, poorly formed, or ‘partial prints’, such as half a thumbprint, and the prints of several people may overlap. Furthermore, prints decay with time at a highly variable rate that depends on what they are composed of, the matrix they are formed on, and the environmental conditions. For example, the lipid components of latent fingerprints can change dramatically with time (Archer et al. 2005) and this should be borne in mind when deciding which enhancing agent should be used to render the prints visible. Exposure to very high temperature can also destroy fingerprints or render them difficult to interpret. This is a problem when investigating the remains of bombs used in terrorist attacks.
There is currently no accepted method for ageing fingerprints. Consequently, finding a suspect's fingerprints on a gun would indicate that they handled it but not when they handled it. This is important when attempting to prove that a suspect committed a murder or was present in a room, but who claims to have picked up the weapon or visited the room before or after the crime was committed.
Although digital technology increases the speed with which fingerprints are recorded, stored, and compared, it is not without its problems (Giuliano 2016). For example, there can be problems with the resolution of the scanned image – narrow fingers produce poorer images than wide fingers, because they touch fewer sensors and therefore leave fewer dots on a 500 pixels per inch (ppi) scan. By comparison, 35 mm black and white forensic film has the equivalent of 6000 ppi. Imagine trying to distinguish between the numbers one and seven using dots: the fewer the dots, the poorer the resolution, and with only three dots it would be impossible. Clearly, if detail is not captured, potential points of similarity or difference could be missed. Therefore, one should use the highest sensitivity setting when capturing images and know the resolution of those stored on a databank. Similarly, if images are compared on a computer screen, the screen's resolution and the settings can influence one's ability to spot differences. One should note whether the image was processed in any way after collection and consider whether it was indexed correctly.
The introduction of biometric passports and identity cards has led to concerns from civil liberties groups and worries over whether the information could be misused. In Germany in 2008, the Chaos Computer Club obtained what it claimed was the fingerprint of the German Interior Minister, Wolfgang Schaube, from a water glass. To illustrate what they saw as the potential for biometric data being misused, they printed images of the fingerprint and gave these away with issues of the magazine Die Datenschleuder. The images were reproduced on a film of flexible rubber with a glue backing that enabled the ‘print’ to be attached to the user's own fingers. This technique can successfully ‘fool’ biometric readers and calls into question the reliability of those using fingerprint identification. We cannot avoid leaving our prints on objects and these can be acquired by third parties. It also means that it is possible to frame someone for an offence by taking surreptitious copies of their prints and then transferring these to an object that would link them to a crime. However, taking someone's fingerprints without their consent (except in the course of a police investigation) is a criminal offence.
Despite these difficulties and unlike most other forms of forensic evidence, fingerprinting is seldom presented with a numerical indication of the likelihood of a match between two prints. It is believed that either they match or they do not. Indeed, a 1979 resolution of the International Association for Identification – which in 2017 had over 6500 members from 77 countries – stated that an expert giving ‘testimony of possible, probable or likely (fingerprint) identification shall be deemed to be engaged in conduct unbecoming’. A further problem is that, although those undertaking the final analysis are invariably highly experienced experts, they are subject to the same subconscious influences as anyone else. For example, in a fascinating series of experiments, Itiel Dror and co‐workers demonstrated how experts are swayed in their interpretation of fingerprints by the context in which they are presented (Dror and Charlton 2006; Dror et al. 2005, 2006). For example, they would change a decision regarding a set of fingerprints that they had previously pronounced upon in court when presented (unknowingly) with the same prints in a different context.
The extent to which mismatches occur is not known, but is probably very small. However, when a mismatch does happen, the consequences for the victim are catastrophic. Two case studies are presented here of instances where mistakes were made. This is not to cast doubt on the technique, but they illustrate why care is needed in interpreting forensic evidence and that it is important to admit when one is wrong.
6.1.4.6 Lower Surface of the Finger, Palm, and Sole Prints
The lower surfaces of the hands and feet have friction skin and therefore leave prints useable as forensic indicators. For example, when grasping an object, one leaves prints of the lower surface of the fingers and of the palms. Similarly, a person walking barefoot leaves prints of their toes and feet. These prints are just as unique as those found on the fingerpads. It is therefore important to search for these at a crime scene and to record them from a suspect. Some police forces routinely collect palm prints, but owing to the shape of the palm, this needs doing carefully, otherwise a proportion of the surface would be missed.
6.1.5 Lip Prints and Ear Prints
The use of lip prints for identification purposes is occasionally alluded to in forensic literature, but their validity as forensic indicators is not yet proven. There are currently a small number of studies that suggest that an individual's lip prints remain unchanged over six months and there may be gender and racial differences (Furnari and Janal 2017; Verma et al. 2015). However, there is a need for more studies and there are currently no accepted standards or methods for their identification and no databases. Lip prints are currently most useful as a potential source of DNA from which a much more accurate identification can be obtained.
The 3D shape of the ear provides useful biometric information (Abaza et al. 2013), but the reliability of ear prints remains highly controversial. Scientists can seldom resist burdening the English language with ugly and unnecessary new words. Therefore, it should come as no surprise to learn that the study of ear prints is sometimes referred to as ‘earology’. Ear prints are typically left at a crime scene when a person listens at a door or window before committing an offence. For example, in May 1996, a frail elderly woman was smothered to death during a burglary in Fartown, Huddersfield, UK. At the subsequent trial, in 1998, Mark Dallagher, known locally as a petty burglar, was charged with the woman's murder. Despite protesting his innocence, claiming that an ankle injury would have made it difficult to commit the crime and supplying an alibi, he was found guilty. This was largely on the testimony of expert witnesses for the prosecution, who stated that his ears yielded prints exactly the same as those found on a windowpane at the crime scene. It did not help that Dallagher's alibi said that she was asleep and ‘on medication’ at the time of the offence. Furthermore, a fellow prisoner claimed that Mark Dallagher confessed to him while he was being held awaiting trial. However, at a retrial that ended in January 2004, he was exonerated when it transpired that DNA isolated from the ear prints at the crime scene did not belong to him and there were serious concerns about the reliability of the identification of the prints themselves. For example, when he first compared Mark Dallagher's ear prints with those found on the windowpane, one of the expert witnesses had written that the two were ‘definitely not’ the same (www.forensic‐evidence.com/site/ID/DNAdisputesEarlID.html). The prison ‘confession’ was also dismissed as unreliable. The testimony of prison informants is always controversial, because prisoners often attempt to gain favours or remission by telling the authorities tales they wish to hear. Alternatively, the informant may be pressured into concocting a story or be dishonest or malicious.
Ear print analysis is difficult because when a person presses their ear against a substrate, it is deformed to an extent that depends upon the pressure exerted and the substrate. Consequently, even prints from the same person are, to an extent, variable. When gathering prints from willing donors, this can be overcome (to some extent) by requiring them to listen for a sound emanating from behind a glass plate. The prints can then be recovered in a similar manner to fingerprints. This approach is therefore similar to the situations in which ear prints are left at a crime scene.
There are insufficient studies or databases to know whether an ear print is truly a unique identifying characteristic. However, this does not negate the value of the ear print analysis. This is because a numerical indication of the closeness of the match between two sets of prints may be enough to establish whether it is worth continuing to associate an individual with a criminal investigation. Work has begun on computerised systems for analysing ears and ear prints and the EU has financed the melodramatically titled FearID project – Forensic Ear Identification project (Alberink and Ruifrok 2007). Some police forces take ear prints as well as fingerprints and these are entered onto a central computer database. If the ear is pierced, then this provides a further identifying characteristic – although it may also result in the formation of a partial ear print (Abbas and Rutty 2005; Swift and Rutty 2003). The possibility of identifying an individual from closed circuit TV recordings (CCTV) by observing their ear characteristics has been suggested. However, this is easily thwarted by wearing a hat, hoodie, turban, hijab, or a wig of long hair, so its effectiveness would be limited.
6.1.6 Nails
Our fingernails and toenails provide protection and, in the case of fingernails, facilitate grasping and manipulating objects. Morphologically, two regions can be identified: the nail root that is embedded within and covered by the skin and the exposed nail body that terminates in the free edge of the nail. Our nails are composed of dead keratinized epidermal cells that are initially formed by the nail matrix – a region underneath the nail root. The nail matrix produces new cells throughout life: fingernails grow at about 1 mm week−1, whilst toenails grow slightly more slowly. Whilst growing, nails incorporate many drugs and poisons found in the circulation (Baumgartner 2014). Therefore, they are good indicators of drug use. In addition, nails, unlike hair, grow continuously and therefore provide a continuous record of drug consumption. Furthermore, it is not possible to avoid detection by removing them because, although a person can shave their head, they are unlikely to rip out their nails. To detect drug residues in a living person, the free edges of the fingernails is cut or the nail body lightly sanded. In a dead person, the whole nail can be removed. Toenails are usually less exposed to the environment and therefore less subject to contamination. Drug abuse can cause physical changes to the nails. For example, both alcohol and opioid abuse induce ‘digital clubbing’, in which the fingernails curve over the rounded fingertips because of accumulating connective tissue. However, pulmonary diseases and conditions affecting lung function also cause digital clubbing. Therefore it must be considered in relation to other evidence.
The condition of fingernails provides an indication of lifestyle, neglect, and involvement in an assault. For example, the fingernails of manual workers are often short, chipped, and rough. By contrast, the fingernails of a person suffering from neglect may grow excessively and become encrusted with dirt. Digging a grave results in soil encrusting under the nails and this can provide evidence of association with a location. During an assault, it is instinctive to dig one's nails into the body of an assailant or victim. This inflicts wounds, damages the nails (if long), and causes the transfer of DNA and fibre evidence. Therefore, the hands of crime victims (and suspects) should be examined carefully and swabbed for DNA and other evidence. However, fingernail swabs can reveal the DNA of people who merely share the same accommodation. Therefore, one should not leap to conclusions as to how the DNA came to be present. Many women wear artificial acrylic or gel nails and if these are distinctive, it may link them to a place of purchase and thereby provide evidence of movement. These nails are likely to break off during a struggle and, if found, link a person to a location. They also pick up DNA and fibre evidence and should be swabbed and tested in the same manner as real nails.
6.1.7 Hair
Our body is covered with hair, except for the surfaces of the palms, the palmar (lower) surface of the fingers, soles of the feet, and the lower surface of the toes. Although hair comes in various lengths and thicknesses, depending upon where it grows, its basic structure remains the same. A hair consists of two portions: a shaft projecting from the skin and a root embedded within the skin. The hair shaft forms from three layers of dead cells that contain large amounts of fibrous alpha‐keratin proteins that are firmly bound together by extracellular proteins. Alpha keratins are robust molecules that are resistant to decay and few animals can digest them. Consequently, hair remains long after the soft body tissues have decomposed. The outermost layer of a hair is called the cuticle; this region is the most heavily keratinised and it has a scale‐like appearance (Figure 6.5). The middle layer contains pigment granules in the case of darkly coloured hair, but these are reduced in number or absent in grey or white hair. The inner layer, which may be absent in thin hair, also contains pigment granules. The hair root extends down into the dermis of the skin where it is surrounded by the hair follicle and at the base of this is the hair bulb. Within the hair bulb are living germinal cells responsible for the growth of an existing hair or its replacement should the hair fall out.
Figure 6.5 Scanning electron micrograph of human hair: (a) scalp hair; and (b) pubic hair.
All hairs exhibit two stages of development: first there is an active growth (anagen) stage and this is followed by a resting (telogen) stage. In adult humans, the anagen stage of a scalp hair lasts for two to six years, during which it grows at about 1 cm a month. This is followed by a resting stage of about three months. By contrast, our body hair, sometimes referred to as androgenic hair, is short‐lived. It has an anagenic stage lasting only a few months, followed by a long resting stage of about 12 months. These figures are important when considering the accumulation of drugs, poisons, or other chemicals within our hairs.
6.1.7.1 Hair as a Forensic Indicator
Typically, we lose about 100 scalp hairs per day. Loss may be passive through falling out naturally or through gentle pulling – as when brushing our hair. Clothing, pillows, and combs are therefore good places to find hairs. Hairs transfer between people during vigorous bodily contact, such as sex. The more vigorous the encounter, the more likelihood that transfer will take place. Therefore, in cases of sexual assault, one may find scalp, facial, and pubic hair from the attacker on the body of the victim. In any fight, both victim and the attacker commonly pull out one another's hair. One should, therefore, examine the body and clothing of both the victim and the suspect, as well as the locality for hair evidence.
Many people, especially the young, choose highly characteristic hair styles and artificial colorations. However, these are so easily changed they are seldom reliable forensic indicators. With the advent of widespread CCTV in towns and buildings, criminals often cover or obscure as much of their face as possible. Similarly, those recording themselves undertaking criminal acts and then uploading the footage onto the Internet usually ensure their face is obscured or not in the picture. This has led to an interest into the feasibility of identifying individuals on CCTV or other digital images from the growth patterns of body hairs coupled with limb geometry (Chan and Kong 2015). This could be useful for identifying individuals recorded undertaking assaults, looting, animal cruelty, or in making child pornography.
Microscope analysis of scalp hair can provide some identifying features. For example, a transverse section through a hair shaft can indicate its shape: straight hair appears round, curly hair is kidney‐shaped, and wavy hair is oval. Its pigmentation provides an indication of coloration. However, coloration can be altered using hair dyes, whilst a skilled hairdresser can alter the length and appearance of a person's hair. And a criminal could wear a wig during or after committing a crime. Consequently, the physical appearance of hairs recovered at a crime scene provides some evidence but not positive identification.
A once common (and utterly false) belief was that an individual can be identified solely from the morphology of one or two hairs. In the USA, this led to hundreds of convictions that are now being called into question. For some it is already too late because they were executed (Norton et al. 2016). Unfortunately, the evidence for some older cases (e.g. 20–30‐years ago) appears to have ‘gone missing’. This highlights both the importance of keeping evidence and the reluctance of some authorities to re‐open old cases once a conviction is obtained.
A hair with its follicle attached is a good source of DNA. Unfortunately, hairs with well‐developed follicles are those that are actively growing (i.e. in the anagen phase) and firmly embedded in the body. They are therefore unlikely to detach unless pulled forcefully. It is only the old telogenic hairs that fall out easily. These have small bulbs containing relatively little or badly degraded DNA. In addition, keratin interferes with the Polymerase Chain Reaction (PCR) process, so it is advisable to separate the cells containing DNA from the rest of the hair. The success rate for extracting good STR profiles from hair is therefore relatively low. One way of improving the success rate is to use laser microdissection to separate away the few useful cells from the rest of the hair (Di Martino et al. 2004b). Another method is to use 4’,6‐diamidino‐2‐phenylindole (DAPI) fluorescent staining to identify roots containing at least 20 nuclei – the current minimum for obtaining an at least partial STR profile (Lepez et al. 2014).
Hair, especially if it is greasy or wet with blood, picks up and retains pollen and other palynomorphs, as well as soil and inorganic particles. These are often good indicators of place and, sometimes, time of year. This is discussed in detail by Wiltshire (2006a), who also provides two case examples.
Drugs such as methadone and poisons such as lead and arsenic that are present in the circulation are sequestered in hair. Because hair grows at a fairly constant rate (depending upon the area of the body and our age), the amount of the chemical along the length of a hair indicates when it was administered. Drugs and poisons in hair are detectable long after the last dose was administered and even after death. By contrast, they are eliminated rapidly from the body when we are alive and lost with the soft tissues when the body decays. Hair analysis therefore reveals whether a person was taking drugs or poisons either voluntarily or unsuspectingly.
In living persons, hair analysis is often used in conjunction with urine analysis to compare past and present drug use. The children of drug addicts often ingest drugs left lying around or they are given them intentionally. Indeed, approximately half of all instances in which a toddler dies from an unintentional drug overdose in the UK result from ingesting methadone. In child welfare investigations, evidence of repeated exposure to drugs can be demonstrated by analysing the child's hair. The presence of drug residues usually results in the child being taken into care by the social services. Sometimes, however, the intervention is too late. For example, in the UK in 2017, a 38‐year‐old mother and her 40‐year‐old boyfriend (i.e. they were not young and inexperienced) were found guilty of repeatedly feeding drugs to the woman's 4‐year‐old child. This was done so that they could have undisturbed sex. The child died eventually of a drug overdose. The child's hair contained traces of morphine, heroin, tamezapam, and ketamine. The distribution along the length of the hairs indicated that these were given repeatedly over several months.
‘Date rape’ often involves the victim being rendered incapable by consuming a drink laced with a hypnotic such as Rohypnol® (flunitrezapam) or zolpidem. These drugs also cause memory impairment (Villain et al. 2004). The breakdown product of Rohypnol, 7‐aminoflunitrezepam, is easier to detect in hair than flunitrezepam itself. This is because it is more basic and therefore binds better to melanin in hair. It is detectable up to a month after ingesting a single dose of Rohypnol (Negrusz et al. 2001). However, a negative result does not mean that Rohypnol was not consumed. There are big differences in the metabolism of individuals and this makes it impossible to determine how much Rohypnol was ingested or when. In addition, some people use Rohypnol as a recreational drug. Therefore its presence in their hair may not indicate that they unintentionally consumed a spiked drink.
Some drugs can enter the body through environmental contamination. For example, canniboid residues may be found in hair as a result of illicit drug use or exposure from the atmosphere. Consequently, hair samples should be decontaminated before analysis – although there is no consensus as to the best procedure. The presence of specific metabolites, in a particular ratio to the parent drug, indicate that a drug was consumed rather than being present in hair through surface contamination. Hair coloration influences the assimilation of some chemicals. For example, in experimental rats, white hairs assimilate less methadone than black hairs (Green and Wilson 1996). Hair treatments, such as colouring and bleaching, and post‐mortem events, such as colonisation by fungi, and exposure to sunlight or damp, all affect the recovery of chemicals to varying extents.
Traces of explosives and gunshot residues are detectable in hair, even after it is washed. This includes explosives such as TATP (triacetone triperoxide – also known as acetone peroxide and ‘Mother of Satan’) that was used in the London Tube Train bombings in July 2005 (Oxley et al. 2012). Many explosives have low vapour pressures, but even these are either adsorbed or absorbed, or both, on to hair. However, traces also assimilate on to hair by contamination from hands or using a contaminated towel. Consequently, detecting them on a person's hair is not necessarily proof that they fired a weapon or were involved in terrorist activities.
Whilst hair is actively growing, it expresses the isotopic signature of the location where the person is living. This mostly come from food and liquids, and dietary protein in particular. Ehleringer et al. (2008) found that drinking water from around the USA has different oxygen and hydrogen isotopic ratios and these ‘signatures’ are replicated in hair. The technique is not sensitive enough to identify a city, but enables a region to be predicted. Lehn et al. (2015a) found a measurable change in the ratios of both carbon and sulphur isotope ratios in beard hair of military pilots within a few days of them moving from Bavaria in Germany to Arizona in the USA. However, there was a lot of variability between individuals in the speed with which significant changes in isotope signatures occurred. Isotopic signatures may, therefore, have use in cases of people trafficking or illegal entry and one wishes to know how long a person might have been living in a country or where they might have come from (Lehn et al. 2015b; Mant et al. 2016).
6.1.8 The Retina and Iris
Retinal and iris scanning are often used for personal identification: for example, for high security access at airports and in military establishments. Retinal scanning involves mapping the distribution of blood vessels on the retina by shining a low intensity infrared light through the eye and picking up the reflected light on a video camera. The blood vessels in the retina absorb more of the infrared light and therefore stand out from the surrounding tissues (Figure 6.6).
Figure 6.6 Retinal surface of left and right eyes, illustrating the complex network of blood vessels.
The iris contains over 400 distinguishable characteristics, although only a proportion of these are used in current scanning technologies. Apart from overall form and colour, the characters result from random events during the tissues' formation and growth. The chances of two persons sharing more than 70% of their iris characteristics is approximately 1 in 7 billion (Daugman 2004; Daugman and Downing 2001). Even identical twins have different iris characteristics. Provided we remain healthy, our iris characteristics remain the same throughout life from early childhood. However, abnormalities can arise through physical damage, eye infections, heart disease, and cancers. Iris scanning technologies are currently faster than those for retinal scanning and permit the user to be up to a metre away from the camera (Chen et al. 2016).
There appear to be no published studies on how retinal and iris characteristics change after death. However, the retina tends to detach and fold shortly after death, so retinal scanning is unlikely to be much help in identifying a dead body. The degree of retinal detachment and folding might provide an estimate of the post‐mortem interval (Oshima et al. 2015). However, there are few studies on this and it is affected by many variables, such as temperature and manner of death.
6.2 The Skeleton
Forensic anthropology (also referred to as ‘forensic osteology’) is the study of skeletal remains associated with events likely to lead to criminal proceedings. Forensic archaeology is the study of the location and recovery of human remains. There is a lot of overlap between the two topics and both build upon the considerable body of knowledge built up over the years in traditional anthropological and archaeological research. Only a brief overview is provided here and those wishing for more detail are advised to consult specialist textbooks such as Langley and Tersigni‐Tarrant (2017).
The identification of a complete human skull or the larger bones is easy, but distinguishing the small bones and sesamoid bones from those of other animals requires more skill. This is especially difficult if the bones belong to a young child because many of them exhibit different morphological features to those of an adult. Identification is also problematic if skeletons are disarticulated and incomplete. Consequently, the police are often alerted to bones on a beach or moor that subsequently prove to be those of a sheep or pig. Similarly, a person digging their garden or out walking in the countryside may dismiss human bones as those of a dead pet or wild animal and therefore of no consequence. Even oddly‐shaped pieces of wood and stones are occasionally confused with bones (Figure 6.7). During an investigation into child abuse at the Haut de la Garenne children's home on the island of Jersey, the police retrieved what was thought to be a fragment of bone from a child's skull. It was not until three months later that it was demonstrated to be a piece of coconut shell. If the provenance of a bone is uncertain, it may be possible to extract DNA to confirm whether it is of human origin. If DNA cannot be found, histological analysis of the bone microstructure is required (Hillier and Bell 2007). Anatomical measurements of bones can indicate not only whether or not they are of human origin but also sex, ethnicity, and age at the time of death. The bones also provide clues to the post‐mortem interval (PMI), the presence of underlying diseases, the cause of death, past movements, and identity (Byers and Myster 2005).
Figure 6.7 It is easy to mistake manmade and natural objects for bones or bone fragments. This naturally weathered piece of wood has a superficial resemblance to a femur. The ruler is 30 cm in length.
The skeleton of an adult human comprises of 206 named bones, but there are even more in a child. The reason for this discrepancy is that some bones fuse as we get older. The skeleton of an adult is divided into two compartments: the axial skeleton and the appendicular skeleton. The axial skeleton comprises the skull, hyoid bone, and the bones of the ears, the vertebral column, and the thorax. The appendicular skeleton comprises of the bones of the shoulder blades, the upper and lower limbs, and the hip girdle. Most bones are paired on the left‐ and right‐hand sides of the body. Bones are classified into five types based upon their shape: long, short, flat, irregular, and sesamoid. The long bones are those such as the femur in the thigh. These bones are longer than they are broad and are usually slightly curved. This shape allows weight to be more evenly distributed along their length. Short bones are approximately as broad as they are long and typical examples are the majority of the carpal bones found in the wrist. As their name suggests, the flat bones are flat and ‘plate‐like’. They consist of two almost parallel layers of compact bone. Examples include the bones of the skull, the breastbone, and the ribs. These bones provide protection for the underlying soft tissues and an extensive area for the attachment of muscles. The irregular bones are those of the vertebral column and some of the facial bones. The sesamoid bones gain their name because their shape resembles that of a sesame seed. They are found within tendons that bear heavy stresses and strains and they protect them from excessive wear. Most sesamoid bones are only a few millimetres in size, such as those in the palms of the hand, and their number varies between individuals. An exception to this generalisation is the two patellae or kneecaps. In addition, small sutural bones are found between the sutures of some of the cranial bones, although their number varies considerably between individuals and they may be absent entirely.
The ability to identify individual bones only comes with a great deal of practical experience and is not aided by most of the anatomical features having long complicated names. However, it essential if a skeleton is disarticulated and/or the remains are commingled (mixed together – as in a mass grave). For example, the presence of three tibiae would indicate the presence of at least two people. If there was one right tibia and two left tibia, then there could be two or three people. If both the left tibiae were noticeably shorter than the right tibia, then it is still possible that there might be only two people, since one of the victims might have had a deformed leg – closer analysis of wear patterns would confirm or refute this.
6.2.1 Determination of Sex (Gender) from Skeletal Remains
Although the terms ‘sex’ and ‘gender’ are often used interchangeably, this is not always appropriate. One's sex is a biological consequence of one's genetic constitution – usually XX for women and XY for men. However, there is a big debate within social anthropology about gender and what it means. Some maintain that one's gender is a cultural construct and, for example, a man may identify himself as a woman. Sex determination is seldom a problem whilst a body is in the early stages of decay, because the genitalia and secondary sexual characteristics (e.g. facial hair) are present. Individuals who undergo a sex change operation or are born hermaphrodites present more difficulties, but a DNA test usually solves the matter. However, the DNA may be badly degraded or absent (Quincey et al. 2013). If only skeletal evidence is available, it is still possible to determine sex with a high degree of accuracy in adults. However, it is much more difficult in children (Đurić et al. 2005). The most helpful indicator is the pelvis, which exhibits numerous differences between men and women. For example, in a woman the pelvis is smoother, lighter, and more spacious than in a man (Figure 6.8a and b). Similarly, the area of the pelvis called the sub pubic arch has an inverted ‘U shape’ in women, but an inverted ‘V shape’ in men, and there are several other differences. There are also gender differences in the dimensions of the bones of the skull (especially the mandibles), the long bones and the sternum. Some of these bones cannot by themselves provide an accurate determination of sex but when used in combination with two or more other bones (and if there is a known marked sexual dimorphism within the population) the accuracy can be increased to more than 95%. This still leaves a minimum of 5% of cases in which it is impossible to confirm sex. Therefore, it is not always possible to determine one of the most basic human characteristics in unidentified remains.
Figure 6.8 Pelvis of adult male (a) and female (b).
Source: Reproduced from Kerr (1957).
6.2.2 Determination of the Age and Provenance of Skeletal Remains
See Chapter 2, Sections and.
6.2.3 Extraction of DNA from Bones
The ability to extract DNA from bone depends on both the type of bone and a wide range of taphonomic variables such as the PMI, the environment, and whether the bone suffered post‐mortem damage. The long bones, such as the femur and humerus, are the best sources of DNA, since they contain the most marrow. However, it is important that the integrity of the medullary cavity and the periosteum are preserved. If the long bones are degraded, then the petrous bone at the base of the skull and the teeth are the best sources of DNA. Although DNA can be isolated from the flat bones, such as those of the ribs and scapula, the chances of success are lower. This is because of their smaller amounts of marrow and their greater fragility makes them vulnerable to post‐mortem damage. Because of the practical problems of isolating DNA from bones, multiple samples are taken wherever possible. Skeletal remains should be cleaned to remove surface contamination before extracting DNA samples. Contamination is a particular problem for remains that were buried. Surface sterilisation is often undertaken using 3% sodium hypochlorite (bleach) for at least 15 minutes (Kemp and Smith 2005).
The extraction of DNA from bone presents logistical problems, because large amounts of calcium can inhibit the PCR process. It is therefore necessary to remove these, along with other potential inhibitors. This is often done by phenol extraction combined with silica extraction or by silica extraction alone. Magnetic separation techniques such as the DNA IQ™ system improve the efficiency of DNA extraction from bone and other difficult substrates. This involves binding the DNA to magnetic particles and then applying a strong external magnetic field. The magnetic particles (+DNA) are then either maintained in place whilst the contaminants are removed or, by moving the magnetic field, the magnetic particles (+DNA) are removed from the contaminants. Afterwards, treatment with an elution buffer separates the DNA from the magnetic beads. However, contaminants sometimes prevent DNA binding to the magnetic particles and Desmyter et al. (2017) recommend prior phenol extraction to remove them.
6.2.4 Determination of Ethnic (Racial) Characteristics from Skeletal Remains
The terms ‘race’ and ‘racial characteristics’ often generate emotionally‐charged controversy and some workers prefer to use the term ‘ethnicity’. This is because it includes variables such as language, religion, and social factors. There is no such thing as a ‘pure human race’ and many anthropologists, such as Brace (1995), consider human variation in terms of clines of traits rather than discrete populations. Despite this, ‘race’ is commonly referred to in forensic and medical literature and is a useful concept in the identification process provided that its limitations are recognised. For example, within the forensic literature, the terms ‘white’ and ‘black’ usually relate to differences between Americans of European and African descent respectively. However, beyond this they have no meaning whatsoever, since they are context dependent: nobody is pure white or pure black and a person who would be considered white in one country/community might be considered black in another and vice versa. There is no consensus on how many racial categories there are, but the terms ‘Caucasian’, ‘Negro’ and ‘Mongoloid’ are in common usage. (There is a similar lack of agreement in assigning ethnic criteria.) The term ‘Caucasian’ was originally coined in 1775 by the German scholar Johan Blumenbach. He recognised five distinct human races: Caucasian, Mongolian, Ethiopian, Native (North and South) American, and Malay. He chose the term Caucasian, because he thought that the Caucasus region produced ‘the most beautiful race of men’. He also thought that humans originated in the Caucasus region and all subsequent racial types were derived from them.
Although one can assign a skull to one of three broad groupings of Caucasian, Negro, and Mongoloid (Figure 6.9), further discrimination is seldom possible. Among the morphological differences are the shape of the mandibles and the size of the nasal openings – the latter tends to be narrower and higher in Caucasians than Negroes. Iscan and Steyn (2013) have disputed the value of mandible measurements for racial discrimination. However, Buck and Vidarsdottir (2004) used a computer‐based method of geometric morphometric analysis of mandibles to identify unknown sub‐adult individuals with an accuracy of over 70%. This is noteworthy because up until adulthood many anatomical characteristics change in proportion to one another because growth does not occur uniformly. Consequently, most of the work on race is done on adult skeletons. However, an accuracy of 70% suggests that 3 out of 10 skeletons would be incorrectly identified. Furthermore, much of the work on morphological distinctions between Caucasian/ Negroid individuals is based on the US studies and therefore relate to the major populations found there. They are not necessarily applicable elsewhere. If only parts of the cranium are available then the Fordisc computer program is extremely effective at determining ethnicity (https://web.archive.org/web/20150913090855/http://fac.utk.edu/fordisc.html).
Figure 6.9 The skulls of adult humans can usually be ascribed to one of three racial categories: Caucasian (or ‘white’), Negro (or ‘black’), or Mongoloid (or ‘Asian’). See text for details of discriminating features.
Source: Reproduced from Dolinak et al. (2005), © 2005, Elsevier Academic Press, with permission.
6.2.5 Determination of Stature from Skeletal Remains
Stature refers to a person's natural height when standing in an upright position and, clearly, it is an important factor in the identification process. Although apparently a simple measurement, it presents difficulties, even whilst a person is alive. Once they are dead, things only get worse (Bidmos 2005). Stature varies naturally throughout the day because of differential loading on the spinal vertebrae and errors are commonly made when making measurements. Furthermore, as we enter old age, our bones shrink and our posture changes. Consequently, during adulthood, stature is not a constant but changes gradually with time. Once a body is skeletonised, determining stature is not a simple matter of laying the bones out on a bench and measuring the length of those contributing to height. This would not be the way the bones are arranged in relation to one another during life and the cartilage at the joints would be missing. A further problem arises if some of the bones are missing. Nevertheless, provided at least one of the long bones is present – such as the femur, tibia, or humerus – one can estimate stature with reasonable accuracy. The length of the long bones is proportional to height. Therefore, by reference to a table or regression equation, one can estimate height (Trotter 1970). The accuracy increases with the number of long bones measured. Allowances must be made for gender and population and it is important that estimates include 95% confidence intervals.
6.2.6 Determination of Age at Death from Skeletal Remains
If all the soft body parts have decayed, one uses a combination of dental and skeletal characteristics to estimate how old a person was when they died. These characteristics are not especially accurate and become less so as a person ages. In terms of skeletal development, remains are categorised into one of eight groups: perinatal (foetal), neonatal, infant, young childhood, late childhood, adolescence, young adult, and older adult. The term ‘perinatal’ refers to unborn babies and their age can be determined with some accuracy by measuring the cervical, thoracic, and lumbar vertebrae (Kósa and Castellana 2005). The development of an unborn child proceeds at a regular and predictable rate, because the foetus is protected from the outside environment and if food is lacking, its growth continues at the expense of the mother. Neonates are babies that have been born but whose teeth are not yet emerged. At this time, the baby has very small bones and many of these, such as the pelvis and those of the skull, are not yet fused together. However, there is a lot of individual variation in the speed with which these events take place. The teeth begin emerging during infancy and these provide a fairly accurate indication of age in children. In addition, the bones start to ossify (harden) although, again, there is a lot of variation between individuals in when and how rapidly this process develops. In late childhood, more of the bones begin to ossify and the permanent teeth appear.
During adolescence, the long bones grow rapidly in length. This is brought about by the activity of the chondrocyte cells in the regions of the bones called the epiphyseal plates. The chondrocytes multiply and form a layer of cartilage that causes the epiphyseal plate to become wider and hence the bone elongates. As the cartilage forms, the chondrocytes on either side of it die off and their place is taken by another cell type, the osteoblasts. The osteoblasts convert the cartilage into bone and so the bone shaft grows. By late adolescence, the cartilage of the epiphyseal plates becomes completely replaced with bone tissue in a process known as epiphyseal plate closure. Once closure is complete, further lengthening of the bone is not possible, although changes may still take place in circumference. The number of epiphyseal plates differs between bones and the timing of their closure varies both within and between the different types of bones. It is therefore possible to estimate age with reasonable accuracy on the basis of the extent of epiphyseal plate closure within the skeleton. For example, in the clavicles (collarbones) of men, epiphyseal closure at the acromial end of the bone (i.e. that next to the scapula) occurs at 19–20 years of age. However, at the opposite sternal end (i.e. that next to the breastbone), epiphyseal closure does not occur until 25–28 years of age (Krogman and Iscan 1986). The levels of sex hormones influence the timing of plate closure during puberty. Consequently, sex has to be considered when estimating age from skeletal characteristics – closure usually occurs at a younger age in women than in men, owing to their earlier maturation.
After adolescence, age determination from skeletal remains becomes more problematic. One method is to observe the degree of closure of the cranial sutures in the skull – as the plates fuse together with time, the sutures become less distinct. Many forensic texts mention the fusion of the basilar suture (sphenoid‐occipital synchondrosis) within the skull as an indicator of early adulthood. Fusion typically completes between the ages of 18–25, but its position makes it extremely difficult to observe. Furthermore, some workers state that fusion occurs at an earlier age in many individuals. An alternative approach is to observe the shape of the rib bones, and the degree of pitting of the cartilage that connects the ribs to the breastbone (Kunos et al. 1999). To begin with, the ends of the ribs are flat and the cartilage is smooth, but with increasing age the rib ends become ragged and the cartilage becomes pitted. The reliability of this approach is uncertain (Schmitt and Murail 2004) and the rib shafts themselves are fragile and along with the cartilage they decay rapidly if the corpse is left in an exposed position or an acid soil. Other methods include assessing changes in the pubic symphysis and the auricular (i.e. ‘ear‐shaped’) surface of the ileum (sacro‐iliac joint). Unfortunately, both of these methods have their limitations. The pubic symphysis is delicate and therefore liable to damage and decay and it is less reliable as an indicator of age in women than in men (Klepinger 2006). The auricular surface of the ileum is a robust structure, but it does not provide an especially accurate means of estimating age, and above the age of about 50 progressive joint stiffening and fusion (ankylosis) further reduces its usefulness. A preliminary study by Zioupos et al. (2014) indicates that the biomechanical properties of bones could provide an accurate indicator of persons aged 53–85. From their data, the technique would presumably also identify the ages of much younger persons. However, it is uncertain how taphonomic conditions would affect its accuracy. Age‐related degenerative conditions, such as arthritis or osteoporosis, provide a crude estimate of age, but their onset and progression vary markedly between individuals. Even children can suffer from arthritis – juvenile idiopathic arthritis is different from adult arthritis, but still results in damage to the joints.
6.2.7 Facial Reconstruction from Skeletal Remains
Facial reconstruction from skull features is traditionally done manually by applying modelling clay to a cast of the skull. However, 3D computer‐modelling techniques are becoming more sophisticated and have the added advantage of being quicker, cheaper, and the final images can be transmitted easily between interested parties (Short et al. 2014; Wilkinson 2009). There are two basic approaches to the reconstruction process: the anatomical method and the tissue depth method. The former method requires a lot of anatomical knowledge, because the skin and all the underlying tissue layers must be built up gradually layer‐by‐layer starting from the skull surface. The tissue depth method uses statistical data banks of average tissue depths between the bone and skin surface at various marker points on the skull. The tissue depths between the marker points are then interpolated and finally a ‘skin surface’ is applied. Obviously, the tissue depths at the marker points are affected, among other things, by build, gender, age, and population characteristics. However, extensive databanks are available for several population groups. Unfortunately, it is not possible to predict the shape of the mouth, the nose or the eyes and unless hair is preserved, there is be no indication of hair colour or length. The inclusion of hair (if known) helps with identification, but Fernandes et al. (2013) state that it is better to leave the reconstruction bald if there is no reliable information on its nature. It is difficult to predict ethnic origin, and hence skin colour, from skeletal remains. Consequently, two reconstruction experts working independently from the same skeletal remains may produce different models. Indeed, some workers believe that facial reconstructions should be limited to situations in which all other methods have failed, because it might lead to false identifications.
6.2.8 Identification of Surgical Implants in Human Remains
Surgical implants include objects ranging from the silicon bags of breast implants, to battery operated heart stimulators and titanium screws holding bones together. They can all provide identification evidence. Implants made of surgical steel and titanium remain identifiable for many years after death and exposure to extreme temperatures (Berketa et al. 2015). The usefulness of surgical implants is enhanced if the manufacturer includes an identification code and batch number. This enables matching the implant to a specific individual – or at least a geographical location and timeframe. Bennett and Benedix (1999) provide a case study in which the presence of an osteostimulator helped in the identification process. An osteostimulator is a battery powered device that is surgically implanted into bone to stimulate localised bone growth. In this case, an osteostimulator was found in the lower back region of a dead woman. She had been shot in the head, placed in the boot of a car, and the car set alight. Her body was so badly burnt that little information could be obtained from her bones and teeth. However, the police suspected who the dead person was and obtained her medical records. These included X‐rays demonstrating the presence of an osteostimulator in the same region as that found in the dead body. The device lacked a serial number, but its make and position were so similar to those mentioned in the medical records that a positive identification was made.
6.3 Teeth
The study of teeth and the factors that affect their formation is known as odontology and a forensic odontologist is therefore a person who specialises in the study of teeth associated with dead (usually) bodies. The forensic importance of teeth arises from them containing a great deal of personal information, coupled with being the most indestructible part of the body. Adults normally have 32 teeth: on each side of the mouth there are 4 incisors, 2 canines, 2 premolars, and 3 molars. Each tooth has a characteristic morphology, but they all consist of three regions: the portion above the gum line is called the crown, at the gum line is a constricted region called the neck, whilst the portion embedded beneath this within a socket in the jaw is called the root. The canines, incisors, and first lower premolars each have a single root, whilst the first upper premolars usually have two roots. The first two molar teeth tend to have two roots, whilst the corresponding upper molars tend to have three roots. The number of roots attached to the third molar teeth vary, although most have a single fused root. Teeth are composed largely of dentine. This is a highly calcified connective tissue that gives teeth their shape and rigidity. Calcium salts comprise approximately 70% of the dry weight of dentine, thereby making it harder than bone. In the crown region, the dentine is covered with a layer of enamel: calcium salts make up about 95% of the dry weight of this region, so it is even harder than dentine. Within the root region, the dentine is covered with a thin bone‐like layer called the cementum. This attaches the tooth to the underlying bone via dense fibrous connective tissue called the periodontal ligament. At the centre of a tooth lies the pulp cavity that contains connective tissue, blood vessels, and nerve endings.
6.3.1 Taphonomic Processes Affecting the Preservation of Teeth
Under dry conditions and protected from weathering, teeth remain whole for hundreds, or even thousands of years. For example, some 38 000‐year‐old Neanderthal teeth recovered from caves are so well preserved that it is possible to extract DNA from them. Such conditions are, however, the exception, and teeth often become detached from skulls and damaged or destroyed by taphonomic processes.
If teeth are lost some time before death, there will be healing and the tooth socket will be smooth from the deposition of new bone. By contrast, if teeth are lost after death, then the socket will have sharp edges and be completely empty. Scavengers, such as dogs, will attempt to drag away the head from a body and in the process dislodge and damage the anterior teeth. Teeth that are swallowed whole by a scavenger pass through its digestive tract and leave with the faeces. Although teeth may appear unharmed by this experience, they experience chemical and physical damage that makes them vulnerable to further erosion. Gunshot wounds, explosions, and serious physical trauma (e.g. being hit by a car travelling at high speed) can crack or shatter teeth. During fires, teeth often survive owing to their location within the mouth and partial protection by surrounding bone. However, exposure to prolonged high temperatures causes teeth to dry out and shatter. Plant roots will grow around a skull and cause etching of the teeth and over time they can destroy them. Teeth exposed on the surface of soil often turn green from the growth of algae and blue‐green algae upon them. However, it is uncertain how this affects their structural preservation and the DNA within them. The pH of soil or water has a marked effect on preservation of teeth (and bone). The combination of acidic pH and cold temperatures (~4 °C), such as that found in raised bogs in northern Europe, has the paradoxical effect of preserving skin and soft tissue, whilst dissolving the teeth and bones. For example, so‐called ‘bog bodies’ are sometimes so well preserved that it is difficult to believe that some of them are over 2000 years old (Turner and Scaife 1995).
6.3.2 Determination of Age at Death from Teeth
We have two sets of teeth, or dentitions, during our life: the deciduous teeth (also known as the ‘milk teeth’, ‘primary teeth’, and ‘baby teeth’) and the permanent teeth. The deciduous teeth begin emerging from the gums about six months after birth. Thereafter, further teeth emerge at set intervals until all 20 deciduous teeth are present by the time we are about two years old. The deciduous teeth start to be lost when we reach about six years of age and they have usually all gone by the age of 12. They are replaced by the permanent teeth – which include some additional molars. The wisdom teeth usually do not emerge, if they emerge at all, until after the age of 17. It is therefore possible to age a child with a reasonable degree of accuracy on the basis of the teeth that are present and their stage of development (Table 6.1), but with adults other features must be examined.
Table 6.1 Typical emergence dates of different types of teeth in normal healthy individuals.
| Tooth | Maxillary (upper jaw) |
Mandibular (lower jaw) |
|
| Deciduous teeth | Central incisor | 7.5–12 months | 6–8 months |
| Lateral incisor | 12–24 months | 12–15 months | |
| Canine | 16–24 months | 16–24 months | |
| First molar | 12–16 months | 12–16 months | |
| Second molar | 24–32 months | 24–32 months | |
| Permanent teeth | Central incisor | 7–8 years | 7–8 years |
| Lateral incisor | 8–9 years | 7–8 years | |
| Canine | 11–12 years | 9–10 years | |
| First premolar | 9–10 years | 9–10 years | |
| Second premolar | 10–12 years | 11–12 years | |
| First molar | 6–7 years | 6–7 years | |
| Second molar | 12–13 years | 11–13 years | |
| Third molar (wisdom) | 17–21 years | 17–21 years |
Our teeth start to form before we are born and mineralisation usually commences before 16 weeks' development. Birth is a physiologically traumatic event for the baby and one of its consequences is to upset the production of dental enamel. The enamel is laid down as series of lines called the striae of Retzius and at birth a ‘neonatal line’ is formed that is darker and bigger than the surrounding striae – this can be seen when a tooth is sectioned (Skinner and Dupras 1993). The presence of a neonatal line within the first deciduous teeth or at the tips of first permanent molars indicates that a child survived birth and lived for at least a short time afterwards. This is an important observation when the body of a baby is discovered.
Although one can use dental characteristics to age children with a reasonable degree of accuracy, it is much more difficult for adults. Lamendin et al. (1992) derived a simple equation to estimate age based on measurements of periodontosis height and root dentine translucency (‘transparency’ in some texts). Periodontosis (also called periodontal recession and periodontitis in some texts, though the latter is strictly an inflammatory condition) refers to the gum shrinkage at the base of our teeth that afflicts all of us as we get older and which is often accompanied by bacterial infection that causes inflammation. This results in the underlying tooth surface being exposed and staining and pitting of the tooth surface. The extent of this exposure is referred to as the periodontosis height. Root dentine translucency does not usually begin until we are in our 20s or 30s, and is brought about by the deposition of mineral substances in the dentinal tubules. It starts in the root apex and then extends towards the crown and is best seen by placing the tooth on a glass‐topped light box lit by an intense white light source – some workers use special ones called negatoscopes, that are normally used for viewing X ray film. Lamendin's equation was subsequently refined by Prince and Ubelaker (2002), who included a measure of root height and derived separate equations for men and women and for ‘white’ and ‘black’ Americans. Their equations are listed below:
White men : Age = 0.15(RH) + 0.29(P) + 0.39(T) + 23.17
White women : Age = 1.10(RH) + 0.31(P) + 0.39(T) + 11.82
Black men : Age = 1.04(RH) + 0.31(P) + 0.47(T) + 1.70
Black women : Age = 1.63(RH) + 0.48(P) + 0.48(T) – 8.41
Equations derived by Prince and Ubelaker (2002) for the determination of age from dental characteristics. RH = root height, P = (periodontosis height ÷ RH) x 100, T = (root dentine translucency height ÷ RH) x 100. Measurements are in millimetres.
Other workers have derived variations on these equations for other population groups. The technique works best for persons between about 30 and 40 up to about 70–80 years. Clearly, the effectiveness of this approach depends upon knowing the sex and population of the person from whom the teeth came and also that they were not affected by dental problems or medical conditions that affect the growth and health of teeth during life. In addition, different people examining the same tooth frequently arrive at different age estimations. This indicates the degree of subjectivity associated with the measurement of tooth characteristics. However, the technique is cheap, simple, quick, and does not alter or destroy the evidence.
Another approach to estimating age from dental characteristics is to count the number of bands found in the cementum. Just before a tooth emerges from the gum, cementum production starts. Incremental layers of cementum are produced throughout life, with growth occurring during the summer followed by a resting period over the winter. Observation of a sectioned tooth with polarised light reveals a series of translucent and opaque bands that represent the growth and resting phases respectively. A translucent and opaque band together are therefore equivalent to one year of life. Because teeth emerge sequentially at known ages, by counting the number of bands in individual teeth, one can estimate a person's age. Furthermore, by observing the degree of development of the last band laid down, one can estimate the approximate time of year at which death occurred (Wedel 2007). Some authors claim that an individual's age can be estimated using this approach to within two to three years, whilst others state that there could be an error of 10 or more years. The technique is compromised by periodontal disease (i.e. disease affecting the gums and structures surrounding the teeth) and it is uncertain whether there are major differences in the ways that bands are laid down between populations.
A chemical approach to determining age from dental characteristics is to compare the ratio of left (L‐) and right (D‐) isomers of the amino acid aspartic acid within root dentine. We, like all living organisms, utilise only L‐amino acids and when dentine first forms it contains only L‐aspartic acid. However, after dentine formation, the aspartic acid within it undergoes racemisation in which the molecules rotate, until there is a 50 : 50 mix of L‐ and D‐aspartic acid. Because the rate of racemisation is known, by measuring the ratio of L‐ and D‐aspartic acid, one can estimate a person's age to within about a year. The method is compromised if the body is subject to high temperatures such as by being burnt, or if it is not discovered until 20 or more years after death but is otherwise considered reliable. Racemisation of aspartic acid also occurs in the bones. However, the estimates it provides are not as accurate as those from dentine and the correlation between racemisation ratio and age is lower in women than in men (Ohtani et al. (2007).
Bomb curve radiocarbon dating represents a novel, albeit expensive, method of age determination for persons born after 1950 (Buchholz and Spalding 2010; Kunita et al. 2017). The technique utilises the absence of carbon turnover within the fully formed enamel layer. This means that the 14C level within enamel reflects the level present in the atmosphere when the teeth were developing. There was a rapid rise in the atmospheric levels of 14C when above‐ground nuclear testing began, followed by an exponential decline once the tests ceased. The atmospheric levels of 14C vary around the world, largely as a consequence of where nuclear bomb testing took place, and these must be taken into account when calculating when the tooth was developing. There are five zones: Northern Hemisphere zone 1; Northern Hemisphere zone 2; Northern Hemisphere zone 3; Southern Hemisphere zone 3; Southern Hemisphere zone 1–2 (Hua et al. 2013). Most of Europe and the USA are in Northern Hemisphere zone 1.The 14C levels indicate the year in which the enamel was produced, whilst the tooth type indicates the age of the person when this occurred. For example, the wisdom teeth (permanent third molars) are the last ones to be formed, their enamel being laid down when a person is about 12 years old. This enamel, as in other teeth, forms over time, thereby compromising the accuracy of measurements. For example, seafood has lower 14C content than meat from farm animals such as cattle or sheep. Therefore, a high fish diet results in lower accumulation of 14C. It is therefore recommended that analysis is restricted to enamel extracted from the cervix region of the tooth (Wang et al. 2010). If the 14C levels of a person's wisdom teeth correlate with the atmospheric levels in 1974, then the person was probably born in 1962. Analysing 14C levels in a variety of teeth further improves the accuracy of the technique.
6.3.3 Extraction of DNA from Teeth
Teeth are structurally strong and, as long as they remain in place, further protected by bone and surrounding tissues, whilst the dentine and pulp regions are essentially isolated from the environment. Consequently, the teeth are often the last body parts from which one can obtain DNA. Despite this importance, there are currently no widely accepted guidelines for the handling teeth and extracting their DNA for forensic analysis. Tooth roots contain more DNA than the crown region, because they contain more pulp tissue. It follows that those teeth with the most and largest roots are the best sources of DNA. The molars therefore yield the most DNA and the incisors the least. However, everything depends upon the individual circumstances. For example, a molar tooth that has undergone extensive restoration yields a much‐reduced amount of DNA.
The extent to which DNA is preserved within teeth depends upon the same wide variety of taphonomic factors that determine the preservation of the teeth themselves. According to Higgins et al. (2015), nuclear DNA is best preserved in the cementum region, whilst the dentine of the roots provides the best preservation of mitochondrial DNA. Explosions and high energy impacts (e.g. air crashes) can lead to the co‐mingling of bodies and body parts. Therefore, associating a tooth with an individual body (or what is left of it) may not be easy. Higgins and Austin (2013) provide a thorough review of obtaining DNA from teeth for forensic analysis.
6.3.4 Determination of Sex and Ethnic Origin from Teeth
It is difficult to determine a person's sex from the physical appearance of their teeth and, in any case, this is normally achievable from other morphological evidence. However, if only the teeth are available, one can extract DNA from the pulp and test for Y chromosome specific sequences. There is some variation in tooth morphology between ethnic groups. For example, persons of Asian and Native American ancestry tend to have shovel‐shaped incisors – these are ones in which there is an indentation on the lingual side (i.e. facing the tongue). However, this trait is not invariably present in persons of Asian descent and may be found in other ethnic groups. Europeans often demonstrate specific features on their deciduous posterior premolars and permanent molars that are collectively referred to as Carabelli's trait. This feature is also highly variable between populations: Correia and Pina (2002) recorded Carabelli's trait in 85% of White North Americans, but only 13.5% of Portuguese. An accurate computer program called rASUDAS (Scott et al. 2016) is now available to determine ethnicity from tooth characteristics.
6.3.5 Identification of Individuals Based on Dental Characteristics
By noting the presence or absence of teeth, their appearance, and the nature of any past dental work, it is possible to build up a person's dental profile. This procedure is known as ‘dental charting’ and until the advent of DNA technology it provided one of the most accurate means for confirming a person's identity. It relies on comparing the chart of, for example, an unidentified dead body with those held on dental or medical records. There is a surprisingly wide variation between countries in the legislation concerning the retention of dental records. For example, currently, in the UK, the storage of all personal data is governed by the Data Protection Act that states that all sensitive information should be retained for no longer than is necessary. How long is ‘necessary’ is not defined, but the National Health Service recommends that dental records be kept for a maximum of 30 years. However, it is a requirement to store them for only 11 years after the patient made their last visit or reached 25 years of age. The situation in the USA varies between states and records may be required to be kept for as little as 1 year up to a maximum of 10 years; in Italy and Canada, there is currently (2018) no requirement for them to be kept at all. Further problems arise if there are errors in dental charting pre‐ or post‐mortem, if the dead person did not visit a dentist for a long time, was a recent immigrant (and therefore had no records), or the body was destroyed such that only a partial collection of teeth was left for study. Consequently, even if records are available, there may not be an exact match between the dental chart of the dead body and that of the most likely candidate. The situation is alleviated if there are pre‐mortem radiographic records that can be compared to ones taken from the dead body. If the radiographs match then one can confirm identity with a high degree of certainty. For example, some years after the death of Lee Harvey Oswald, (the man who killed President John F. Kennedy in 1963 and was himself then shot before he could be brought to trial), a conspiracy theory arose that a Russian spy had assumed Lee Harvey Oswald's identity and it was his body and not Oswald's that was in the coffin. After a lot of legal wrangling, Oswald's body was exhumed and it was possible to compare oral radiographs of the body with those taken when Oswald was serving with the US Marines. This was not easy because the radiographs taken while Oswald was alive were poorly made. By comparing the dental formula and restoration work, it was confirmed that the body in the coffin was that of Lee Harvey Oswald (Saunders 2016).
In the absence of radiographs, the examiner must make a value judgement to decide if there is a rational explanation for any discrepancies. Clearly, the presence of a permanent tooth that the pre‐mortem chart stated was extracted indicates that the candidate must be excluded. However, the absence of a tooth that the pre‐mortem chart states should be present could be explained by it being lost in an accident, a fight, or extracted by another dentist subsequent to the records being made. Although it is obviously preferable to have numerous points of concordance between pre‐mortem and post‐mortem dental charts, there is no agreed minimum number required to confirm a positive identification. A computer programme called OdontoSearch (www.cilhi.army.mil) facilitates chart comparisons. However, unlike IDENT1 or IAFIS, it is not a database from which a specific individual can be identified.
Despite the potential problems, dental charting remains a valuable method for identification. It is especially useful when there are large‐scale disasters, such as plane crashes or explosions, that produce numerous badly damaged unidentified dead bodies.
6.3.6 Identification of Individuals from Dental Fillings and Prostheses
Although dental health has improved considerably in most industrialised countries, many people have dental repair work up to and including wearing full or partial dentures. Dental fillings and prostheses (e.g. crowns, bridges, and partial or full dentures) are useful forensic indicators. This is especially the case if they can be matched to a dental chart.
In the case of dentures, there is often an indication of the owner's identity inscribed on the prosthesis. However, this still requires the investigating authorities to identify a gift horse when it is staring them in the face. For example, in Memphis USA in 2001, a woman was stabbed and forced to commit oral sex on her attacker. Afterwards, the assailant's dentures were retrieved from the crime scene, tagged, and placed in a sexual assault kit. And there they languished for 16 years until someone got around to analysing them. This was not unusual, as the city reportedly had a backlog of 12 000 untested sexual assault kits. DNA was extracted from the dentures and a match obtained with Thomas Maupin and he was subsequently convicted. Conveniently, the dentures also had Thomas Maupin's name written on them but no one noticed this when they were collected. The dentures were made for him whilst he was in prison charged with a different sexual offence. This case emphasises the need for good police work and the fact that law enforcement agencies should match the speed of forensic analysis with the rate of evidence collection.
Even if there is a name on the prosthesis, the dentures may not belong to the person wearing them. For example, dementia sufferers living in care homes frequently misplace their dentures and these are acquired by others who use them as their own. The relevance of this observation is that dementia sufferers often wander off and die from natural causes or from an accident. The effectiveness of the fit is not an indication of ownership, since dentures are often loosely fitting and may not have been changed for many years.
The materials used in fillings and dental prostheses provide an indication of where and when they were made. A variety of chemicals are used in the preparation of dental fillings that can indicate the manufacturer and likely date of the operation. This may also help determine where a person had been living. For example, when a mass grave was discovered in Belgrade (Beograd, Serbia) in 2011, the chemicals present in the fillings of some of the victims helped to confirm that they were German soldiers killed in World War II and not local people who died more recently (Zelic et al. 2013). Nevertheless, one must keep an open mind, because many people travel to another country for dental treatment so that they can save money or receive better care. For example, in the UK, there is currently a thriving market for ‘treatment tourism’, in which British people travel to Hungary for dental treatment that is combined with a holiday.
Even after cremation, one can identify the chemical composition of dental restoration work from within the ashes (Soon et al. 2015). This would not provide a positive identification but a discrepancy between the chemical signature of the restoration work in the ashes and that indicated by the materials used in the suspected person's dental records suggests a mis‐match. However, this relies on the dental records being up‐to‐date at the time of the person's death.
6.3.7 Drug Abuse and Dental Characteristics
Some drugs, such as cocaine, fentanyl, MDMA, morphine, and cannabinoids are detectable in teeth, but they would only be analysed if blood, hair, or other tissues were not available (Ottaviani et al. 2017). There is little information on the factors affecting the incorporation and retention of drugs in teeth, but some preliminary data is presented by Klima et al. (2016).
Poor dental health, especially in young adults, may result from underlying disease or neglect, but can also arise through drug abuse. For example, methamphetamine abuse causes a condition colloquially known as ‘meth mouth’, in which the person suffers from numerous dental caries, loss of tooth enamel, and deposition of calculus. Calculus, or tartar, is plaque (a sticky film of bacteria) that becomes hard through calcium salt deposition and is firmly fixed to the tooth – usually where the tooth meets the gum. Meth mouth results from a combination of the acidity of methamphetamine and user feeling their mouth to be dry. This is accompanied by a craving for sweet carbonated drinks, thereby leading to the consumption of several litres of these every day. In addition, drug abusers usually neglect their dental hygiene and drug‐induced teeth grinding and clenching results in excessive wear. However, tooth grinding is also a common nervous symptom and not evidence of drug use. Solvent abuse causes a characteristic ‘glue‐sniffers rash’ around the mouth and nose. In addition, the tips of the nose may exhibit frostbite through inhalation of aerosols. Drug users and heavy drinkers often lose and damage teeth through falls or getting into fights whilst ‘under the influence’. However, lost teeth are also a risk factor in physically dangerous contact sports such as rugby and boxing. Dental characteristics should therefore be evaluated in the context of other forensic evidence.
6.4 Summary of Forensic Evidence that can be obtained from Human Tissues
A wide variety of physical, chemical, and biological evidence can be obtained from human tissues (Table 6.2). It is therefore important to collect and store samples in a manner that permits their future analysis by using a wide range of techniques.
Table 6.2 Summary of forensic evidence obtainable from human tissues.
| Tissue | Forensic Evidence | Test |
| Skin | Identification Cause of death/sequence of events during an assault Contact with chemicals (e.g. explosives) Drug abuse |
Fingerprints DNA profiling Tattoos, studs Physical characteristics Scars Wound analysis Analytical tests Scars at injection sites |
| Hair | Identification Contact with chemicals (e.g. explosives, drug use) Geographical |
DNA profiling Physical characteristics Analytical tests Pollen Stable isotopes Mineralogy |
| Nails | Identification of assailant Geographical |
DNA profiling of material trapped under nails Pollen Stable isotopes Mineralogy |
| Bones | Identification Time since death Geographical origin Cause of death/sequence of events during an assault |
DNA profiling Physical characteristics Isotope analysis Physical characteristics Isotope analysis Wound analysis Diatoms |
| Teeth | Identification Geographical origin |
DNA profiling Dental charting Physical characteristics Racemisation Isotope analysis Isotope analysis |
| Eyes | Identification Time since death |
Retinal and iris scanning (live persons only) Iris colour (fades after death) Analytical tests |
| Internal organs | Cause of death/sequence of events during an assault | Wound analysis Evidence of underlying disease |
6.5 Future Developments
Facial recognition software is already in use at security checkpoints and as a means of identification on smart phones. This technology is currently being adapted to develop automated facial recognition software capable of identifying people from live CCTV footage. In 2017, it came as a shock to many that the police in the UK already hold a database of over 20 million facial recognition images. Many of these are of people who have never been found guilty of committing an offence. Police forces in other countries are building similar databases and in 2018, INTERPOL had more than 44 000 images of individuals with international arrest warrants supplied by 137 countries. The use of facial recognition technology in routine policing is still in its early stages, but it will almost certainly increase. Currently, the technology is not especially reliable – for example, a police trial at the Notting Hill Carnival in London in 2017 using a selected database of over 500 images led to over 30 false matches, 5 individuals being challenged to prove their identity (they had been incorrectly targeted) and one person arrested by mistake. Nevertheless, in 2018, Argentine police were able to identify Slovak national Kristian Danev, who was wanted in relation to a murder committed 10 years previously in Slovakia, following his identification from INTERPOL's facial recognition database.
Facial reconstruction is usually restricted to high profile investigations, but as the costs of computing power and information storage decrease, it is likely to be used in more routine investigations. Although it is likely to remain difficult to predict the shapes of the nose, mouth, or eyes, computer‐generated reconstructions offer a way of quickly generating ‘variants’ on a theme (i.e. the basic facial features with different shapes of nose). This may be aided by developments in DNA analysis that can predict morphological features.
The analysis of amino acid racemisation offers an intriguing means of improving age determination, especially for those over the age of about 20–25 when ‘traditional’ methods become increasingly unreliable. However, there remain relatively few publications on the subject and more data are needed on how post‐mortem environmental and taphonomic processes affect racemisation.