Copyright (c) 2006, Alvenh Channe
British Columbia Institute of Technology

Whenever a crime is committed, evidence is always left behind.  Occasionally, criminals may be cautious enough to avoid marking the scene with fingerprints, or leaving behind hair and/or other forms of trace evidence.  However, despite the best efforts of the meticulously careful criminals, there is one type of invisible trace evidence that is almost always deposited at the scene of crime; that is their unique human scent (Albrecht, 2006).

With the use of properly trained scent-discriminating canine/handler teams, scent evidence that is collected from a scene of crime can be used to establish relationship between suspects and other crime scene evidence (Stockham et al., 2004).  Scent evidence is also of importance when other evidence is destroyed (i.e. due to explosions) (Stockham et al., 2004) or when no other forms are trace evidence can be found.  For instance, “…scent-discriminating bloodhound dogs have identified the scent of bomb builders from exploded bomb remains, and drive-by shooters have been identified from scent collected off expended cartridge casings” (Stockham et al., 2004).  Unfortunately, in many cases, scent evidence is often overlooked because it cannot be seen and is difficult to collect (Albrecht, 2006).

Although, canines are commonly employed by law enforcement for tracking, trailing and other tasks, its use for scent discrimination in the matching of collected scent evidence to suspects has yet to gain widespread acceptance (Curran et al., 2005) in the United States (though its use is on the rise).  However, the concept of human scent as trace evidence is not new (Stockham et al., 2004), and detector canines have been routinely utilised by law enforcement agencies throughout Europe in criminal investigations for over a century (Curran et al., 2005; Stockham et al., 2004; Kaldenbach, 1998).  In cases in which scent is used as evidence, dogs have successfully identified suspects in scent line-ups in both Europe and the United States (Curran et al., 2005; Kaldenback, 1998).

Should the use of canines be accepted in courts of law (be it Europe, US, Canada, or anywhere)?  While many canines can be trained to handle scent evidence to some degree (Bryson, 1996), the bloodhound is the only breed that has demonstrated reliability in performing these tasks (JRC, 2005; Stockham et al., 2004; Harvey and Harvey, 2003) and bloodhound evidence is admissible in courts of law in the United States (JRC, 2005).  However, due to limited research, scent evidence is controversial.  In addition, there is currently very little understanding of how human scent is produced (Curran et al., 2005, Stockham, 2004); thus, one should not find surprising, that scent evidence has resulted in successful legal challenges in the US courts of law (Curran et al., 2005).  As there are other factors to be considered, the remainder of this study will examine the abilities of both bloodhounds and canines in general in the discrimination of human scent as trace evidence.  In addition, possible shortcomings of canine scent evidence (be it bloodhound or otherwise) will be considered in making recommendations for its use in criminal investigations.

Human Scent Production

Although there is very little understanding of how humans produce odor, few theories have been proposed (i.e. Zeng et al., 1991; Styrotuck, 1972; Nicolaides, 1974).  For many years, human odor was believed to be the byproduct of bacterial action on decomposing skin cells (Styrotuck, 1972).  While the idea is currently assumed by many dog handlers, it has very little scientific support.  In addition, because the theory does not account for the uniqueness of human scent, Nicolaides (1974, as cited in Curran, 2005) studied the “…biochemical uniqueness of skin lipids and [has] suggested that slight differences in the overall composition of the sebaceous fatty acid mixture can lead to unique individual odors in humans” (Nicolaides, 1974, as quoted in Curran et al., 2005).  Still, other such as Zeng et al. (1991, as cited in Curran et al., 2005) have suggested that odor may be formed as a result of a chemical reaction by simple bond cleavage.

Regardless of mechanism of scent production, human odor is complex and may consist of many components (Stockham et al., 2004).  This was illustrated in recent studies involving attraction of mosquitoes to humans which showed that human scent had more compounds than can be currently identified (Bernier et al., 2000; Steib et al., 2001).  However, researchers do not know exactly which components are detected by the olfactory system of a canine (Stockham et al., 2004).

Uniqueness of Human Scent

The uniqueness of scent may be influenced by a combination of factors including diet, genetics, and environment (JRC, 2005).  Because Steib et al. (2001) have suggested that there may be a genetic predisposition to level of mosquito attractiveness, Stockham et al. (2004) suggest that human odor may be genetically influenced to some extent.  Earlier twin studies also demonstrated genetic influence(Kalmus, 1955, as cited in Stockham et al., 2004); dogs failed to differentiate between identical twins when their odors were presented simultaneously (Kalmus, 1955, as cited in Stockham et al., 2004).

Diet and environment may also contribute to uniqueness of human scent.  The effect of mixed factors were show in a later study by Hepper (1994, as cited in Stockham et al., 2004), in which dogs were able to distinguish between fraternal baby twins who were fed the same diet (demonstrates uniqueness due to genetic factors) and identical adult twins on different diets (illustrates effect of diet).  Furthermore, studies have also shown that dogs are able to differentiate between identical twins who were living apart (illustrates effect of environment) (Kalmus, 1955, as cited in Stockham et al., 2004), but with difficulties due to genetic similarities (Stockham et al., 2004).  Thus it is not impossible to differentiate between identical twins but is more difficult.

It should also be noted that there is a difference between the primary odor of a person that is stable over time regardless of diet or environmental influences, secondary odor as a result of diet and environmental influences, and tertiary odors from exposure to external sources (i.e. soaps, lotions, perfumes, etc.) (Curran et al., 2005).  While a combination of the three type of odors contribute to uniqueness, “…the primary odor must have constituents that are stable over time and diverse across people…for individual identification by human scent…” (Curran et al., 2005).  Because scent evidence has been challenged on the basis of difficulties in defining human odor and understanding of uniqueness (Curran el al., 2005), these differences should be considered.  The stability of primary odor is especially important when dealing with “old” scent evidence as suspects may move into different environments over time or intentionally expose himself to distracting odors in attempts to confuse scent canines.

Canines and Human Scent

The ability of the canine to discriminate human scent selectively is important in criminal investigation (Kaldenbach, 1998).  This ability was first documented in the late 1880’s., in which canines were able to differentiate between people on the basis of human scent in the presence of distracting odours at great distance and under various environmental conditions (Romanes, 1887, as cited in Curran et al., 2005).  In 1903, suspect discrimination by canines in practical police work was first demonstrated by a policeman in Germany while investigating a murder case (Kaldenbach, 1998); the policeman’s dog was able to match the scent on the murder weapon to scent from suspects (Kaldenbach, 1998)

In modern police work, scent from a crime scene or an article of evidence can be collected either directly or indirectly (Curran et al., 2005).  The direct method involves presenting an object to the canine to sniff directly (Curran et al., 2005; Stockham et al., 2004).  Indirect scenting can be achieved either by swiping with surface of the article with a sterile gauze pad (sterile in the sense that these are clean, unused first aid pads; it does not refer to being free of odour) or by passive absorption in which the pad is placed on a scent source for an unspecified period of time, and subsequently presenting it to the canine to sniff (Curran et al., 2005; Stockham et al., 2004).  A newer indirect method involves the use of air flow to collect the scent onto a gauze pad by employing a special portable electric vacuum called a Scent Transfer Unit (STU-100) (Curran, 2005; Stockham, 2004).

Canines are trained to discriminate for persons in particular by learning pre-scenting, in which they are taught to sniff and “memorise” the scent and continue searching the air and ground for more of the scent to follow (Bryson, 1996).  Without pre-scenting, a canine would not be able to acquire sufficient information for discrimination of particular persons (Bryson, 1996).

Despite training, there are issues of reliability pertaining to scent discrimination performed by canines (dogs in general) of which courts should be aware when scent evidence is to be considered (Bryson, 1996).  The problems with canine scent evidence are that “…very few [canines] can work a track over 12 hours old…” (Bryson, 1996; pp. 127) and “…fewer still care capable of identifying a suspect or object in a lineup by scent discrimination…” (Bryson, 1996; pp. 127).  Furthermore, Bryson (1996) suggests that “…dogs vary widely in the ability to find evidence…” (pp. 127), and given that each case is unique, reliability can be difficult to assess.  Due to these challenges, investigators have began using bloodhounds, a breed of dog developed especially for the purpose of “man-hunting”.

Developed from descendents of the seventh century French St. Hubert Hounds (Albrecht, 2006), the bloodhound is well known for its superior olfactory abilities and it is believed to the one of the oldest breeds to dog to hunt by its sense of smell (SCBC, 2005).  Bloodhounds were first employed by law enforcement officials as early as the sixteenth century for the purpose of hunting thieves and poachers (Albretch, 2006).

Today, there is resurgence in the use of bloodhounds in law enforcement in the United States, especially on the east coast (Albretch, 2006).  Are bloodhounds superior to their counterparts of other breeds?  It has been suggested that the sense of smell of the bloodhound is 60 times greater than that of the German Shepherd dog (JRC, 2005).  If properly trained, not only can bloodhounds track and trail like other dogs, but due to their (bloodhound’s) superior sense of smell, they have been able to retrace paths in which victims or suspects have driven or ridden by car (JRC, 2005)!  While there is very little research confirming the reliability and accuracy of other canines in general, the bloodhound is believed to be the only breed that is capable of discriminating and matching collected-scent to a person accurately (SCBHC, 2005; JRC, 2005; Stockham et al., 2004).  Most importantly, the bloodhound is capable matching human scent collected from destroyed (i.e. burnt material) evidence (Curran, 2005).  This ability was demonstrated in study by the Federal Bureau of Investigation and the Southern California Bloodhound Handlers Coalition in which after a bomb was detonated, bloodhounds were able to match the scent collected from the fragments to its builders with a success rate of 60 to 100 percent with no false positives (erroneously identifying a person whose scent was not collected) (Stockham, 2003; as cited in Curran, 2005).

Bloodhounds are also capable of working “old” and/or contaminated scent trails.  While most canines are unable to work a track that is older than 12 hours (Bryson, 1996), a study by Harvey and Harvey (2003) has demonstrated that experienced bloodhound-handler teams were able to identify scent trails after 48 hours with a success rate of 96 percent with no false positives.  Because the study was performed in normally encountered environments (regional parks, college campuses, city centers), bloodhounds have also demonstrated their ability to discriminate odors in the face of contamination.  Many non-bloodhound police canines, on the other hand, are unable to identify suspect by scent discrimination (Bryson, 1996), let alone target odors mix with other human scents.

As opposed to traditional police dogs that seem to have problems with reliability (Bryson, 1996), the bloodhound is the only breed whose ability has been validated in the study by Harvey and Harvey (2003); therefore, only scent-matching performed by properly trained bloodhounds to reach an effective conclusion is admissible in courts of  law in most states (SCBHC, 2005; JRC, 2005; Stockham et al., 2004) as expert witness testimony from handlers (Curran, 2005).  Because bloodhound evidence can and have been used (and challenged), both courts and investigators should be aware of the limitations and risks associated with it use.

One of the most serious limitations of bloodhound evidence that should be brought to the attention of the courts is that level of experience of bloodhound/handlers teams can affect the reliability of bloodhounds (Stockham et al., 2004).  While Harvey and Harvey (2003) demonstrated the reliability of bloodhound evidence, it should be noted that the success rate of novice bloodhound/handler teams in the same study was only 53 percent with one false identification.  Thus, the study (Harvey and Harvey, 2003) also emphasized the importance of experience.

When bloodhounds (or any canines) are employed, investigators must also be aware of the risk of contamination of evidence.  When dogs are allowed to sniff objects directly, forensic trace evidence such as DNA or fingerprints may be rendered useless (Stockham et al., 2004).  Swiping of objects can also contaminate or destroy trace evidence (Stockham et al., 2004).

Given the limitations of scent evidence, investigators must be cautious in using human scent evidence.  Care must be exercised to prevent contamination, and only properly trained and experienced bloodhound/handler teams should be employed.

The abilities of bloodhounds in the study by Harvey and Harvey should not be generalized to other dogs; therefore, until the use of other breeds is scientifically validated, investigator should use only purebred bloodhounds (JRC, 2005) to ensure reliability.

Investigators and courts considering use of scent evidence should also keep in mind that “…because scent is easily transferred, a positive trail or identification resulting from any scent article only shows a scent relationship to article…” (Stockham et al., 2004).  It is recommended that scent be used in corroboration with other evidence (Stockham et al., 2004).  Also, it should be noted that as an item goes from manufacture to disposal, it accumulates a plurality of odors as passes from one handler to the next.

Investigators should also note that because the purpose served by bloodhounds is different from those of tradition police dogs, bloodhounds should not be used as their replacements (Albrecht, 2006).  Given that traditional police dogs are trained to apprehend suspects, it is recommended that “…if a hot track is laid and a patrol dog is available, the patrol dog should be called first since…they are versatile at tracking hot scents, searching buildings and rural areas for a suspect, protecting officers and apprehending suspects” (Albrecht, 2006).

Despite problems with reliability in the use of canines for discrimination and matching of human scent evidence, bloodhounds may be useful in such tasks, provided that they have relevant and sufficient training and experience.  However, when bloodhounds are employed, scent evidence should be used in conjunction with other evidence or investigative techniques.  Even if bloodhound evidence satisfies the Daubert criteria for admissibility, evidence should be evaluated on an individual basis given variation in abilities between bloodhounds and the fact that no two cases are alike in terms of many factors (some of which may influence reliability of certain methods).

Albrecht, K. (2006).  Bloodhounds and scent evidence.  Kat Albrecht.  Retrieved from http://www.katalbrecht.com

Bernier, UR., Kline, DL., Barnard, DR., Schreck, CE., & Yost, RA. (2000).  Analysis of human skin by gas chromatography/mass spectrometry.  Part 2: Identification of volatile compounds that are candidate attractants for the yellow fever mosquito (Aedes aegypte).  Analytical Chemistry, Vol. 72(4), 747-756.

Bryson, S. (1996).  Police dog tactics.  New York, NY: McGraw-Hill, Inc., College Custom Series.

Curran, AM., Rabbin, SI., & Furton, KG. (2005).  Analysis of the uniqueness and persistence of human scent.  Forensic Science Communication, Vol. 7(2).

Harvey, LM., & Harvey, JW. (2003).  Reliability of bloodhounds in criminal investigations.  Journal of Forensic Sciencs, Vol. 48(4), 811-816.

Jimmy Ryce Center for Victims of Predatory Abduction. (2005). How do you collect and preserve scent evidence.  Prevention.  Retrieved from http://www.jimmyryce.org

Kaldenback, J. (1998).  K9 scent detection: My favorite judge lives in a kennel.  Calgary, AB: Detselig Enterprises Ltd.

Kalmus, H. (1955).  The discrimination by the nose of the dog of individual human odours and in particular the odours of twins.  British Journal of Animal Behaviour, 3(1), 25-31.

Nicolaides, N. (1974).  Skin lipids: Their biochemical uniqueness. Science, 186(4158), 19-26.

Romanes, GJ. (1887).  Experiments on the sense of smell in dogs.  Nature, July 21, 1887, 273-274.

South Central Bloodhound Club. (2005).  What you should know about bloodhounds.  Learn About Bloodhounds.  Retrieved from http://southcentralbloodhounds.org

Steib, BM., Geier, M., & Boeckh, J. (2001).  The effect of lactic acid on odour-related host preference of yellow fever mosquitoes. Chemical Senses, 26(5), 523-528.

Stockham, RA. (2003).  Bloodhounds and bombing investigations.  Detonator, 30(4), 35-37.

Stockham, RA., Slavin, DL., & Kift, W. (2004).  Specialized use of human scent in criminal investigations.  Forensic Science Communication, Vol. 6(3).

Stockham, RA., Slavin, DL., & Kift, W. (2004).  Survivability of human scent.  Forensic Science Communication, Vol. 6(4).

Styrotuck, WG. (1972).  Scent and the scenting dogs.  Rome, NY: Arner Publications.

Zeng, XN., Leyden, JJ., Lawley, HJ., Sawano, K., Nohara, I., & Preti, G. (1991).  Analysis of characteristic odors from human male axillae.  Journal of Chemical Ecology, 22(2), 1469-19492

Other Research
Home