annotation bibliography

INTERNATIONAL REVIEW OF LAW COMPUTERS m^^ Carfax Publishing

& TECHNOLOGY, VOLUME 18, N o . 3, PAGES 425-434, NOVEMBER 2004 ' " j ^ T.,<o,s.F™ndscrô p

Under the Microscope: Legal Challenges to Fingerprints and DNA as Methods of

Forensic Identification^

JEFF WISE

ABSTRACT Suspects in legal cases can be identified by an ever-growing list of novel methods. The most common techniques currently used include latent print and DNA analysis. Although standard fingerprinting entered the courtroom over a century ago, the admissibility of fingerprint evidence has undergone a period of intense scrutiny in the USA in recent years. In contrast, most challenges to DNA analysis as a science came during its inception in the late 1980s and early 1990s. Current challenges to fingerprint evidence attempt to discredit the science behind the theory whereas challenges to DNA evidence often bring into question the competency of the analyst. In either case, the lessons learned in various court systems give guidance for those implementing the newer emerging biometric identification technologies such as facial recognition systems, retinal scans and the like. The first section of this article deals with fingerprint analysis and recent challenges to fingerprint admissibility in US courts. The second section discusses the evolution of DNA analysis and relevant cases. The final section gives recom- mendations for emerging biometric technologies to follow to satisfy the standards set forth by the courts.

Introduction

The benefits of positively identifying an individual are obvious. Throughout our daily lives we employ simple methods of identification such as using a picture identification card to access financial accounts, travel between countries or to borrow a book from the local library. However, when a crime is committed, the courts require us to use more discrimi- nating methods of identification that cannot be easily manipulated. The ideal system of identification measures features that are unique to an individual, which are permanent in

Correspondence: Jeff Wise, 17406 Anvil Circle, Houston, Texas 77090, USA. E-mail: [email protected].

ISSN 1360-0869 print/ISSN 1364-6885 online/04/030425-10 © 2004 Taylor & Francis Ltd

DOI: 10.1080/1360086042000313049

426 Jeff Wise

nature and which can be entered into a searchable database to compare with known standards.^ The first scientific system for identification came from Alphonse Bertillon, chief of criminal investigation for the Paris police in the late 1800s, and involved taking a series of body measurements to distinguish one individual from another.^ The Bertillon system of anthropometry lasted for two decades but was supplanted by the emerging field of fingerprint analysis in the early 1900s. Compared to the Bertillon system, fingerprint analysis offered much greater discrimination between individuals. For the next 80 years fingerprint analysis was the mainstay for individual identification until the advent of forensic DNA typing. Since its inception in the 1980s, DNA analysis rapidly joined fingerprinting as a powerful identification tool for law enforcement authorities. Both fingerprint analysis and DNA analysis have been challenged in US courts and abroad. DNA analysis challenges began in the 1980s and 1990s shortly after its introduction while fingerprint analysis enjoyed nearly a century of widespread use before the US courts disputed its merits. Both situations demonstrate the high standards that court systems now require for admissibility of scientific evidence and offer guidelines for developing biometric technologies to follow.

Latent Print History

The first public suggestion for using fingerprints as a method of individual identification came from Henry Faulds, a Scottish missionary in Japan, and was published in the journal Nature in 1880.'' Unfortunately, this went relatively unnoticed until Sir Francis Galton, one of the pre-eminent scientists of the 19th century and cousin to Sir Charles Darwin, published his book Fingerprints in 1892. Although his interests were mainly in using fingerprints to determine race and heredity he declared that fingerprints are unique to each individual and are permanent in nature. Galton also established a method for identifying the unique characteristics of a fingerprint known as 'minutia' or 'Galton's details'. Sir Edward Henry read Galton's book and consulted with him while developing a method for cataloging fingerprint data. The Henry Classification System launched the modern era of fingerprint science and is the basis for nearly all fingerprint analysis today. Galton's reputation as a leading scientist gave credibility to the science of fingerprints without the need for rigorous testing. For the next century the science of fingerprint analysis enjoyed worldwide acceptance.

US Case Precedent

The case of Frye v United States in 1923 was a landmark case, which dictated the guidelines for admissibility of expert testimony.^ In the Frye case, the prosecution attempted to introduce testimony from an expert using an early version of a lie detector. The device measured blood pressure and presumably an increase in blood pressure indicated the suspect was lying. The circuit court was not persuaded by the evidence given by the expert as to the reliability of the device. In their ruling the court stated that experts could only testify if the nature of their testimony is 'generally accepted' in the scientific community. The 'general acceptance' rule became the standard for expert witness testimony for the next 70 years.

In 1993 the United States Supreme Court heard the case of Daubert v Merrel Dow Pharmaceuticals.^ The Court replaced the outdated Frye 'general acceptance' standard with a more comprehensive group of standards for the admissibility of expert testimony. The

Challenging Methods of Forensic Identification 427

Daubert criteria include: whether the scientific theory has been tested, whether it has been subject to peer review and publication, whether it has a known error rate, whether it has widespread acceptance and whether there are operating standards. This ruling places a burden on judges as 'gatekeepers'. It is the judge's role to weigh the credibility of scientific evidence and determine if it should be allowed in court. This case created the now common use of Daubert hearings to determine the admissibility of expert witness testimony. It is through Daubert hearings that fingerprint evidence finally came under scrutiny a 100 years after its incorporation into the criminal justice system.

Legal Challenges to Latent Prints

The now famous case of United States v Byron Mitchell in 1999 was the first shot across the bow for fingerprint evidence.'' The defense produced several expert witnesses prepared to testify that fingerprint evidence is not scientific and therefore unreliable. In his ruling, district court Judge Joyner allowed the introduction of fingerprint evidence. He further stated on the record that human friction ridges (the areas of the skin that leave fingerprints) are both unique and permanent in nature. Judge Joyner also refused to allow the testimony of the defense witnesses who intended to discredit fingerprint evidence.

Fingerprint evidence seemed to be safe after the Mitchell case but in 2002 the case of United States v Carlos Ivan Llera Plaza turned the tide against fingerprints.* District Judge Pollack weighed the various factors presented by the prosecution and the defense over the admissibility of fingerprint evidence. Judge Pollack issued a controversial ruling in January of 2002 which stated that he would allow testimony from fingerprint experts regarding (1) the methods used to compare prints from known and unknown sources, (2) the similarities or differences between two sets of prints, and (3) a rebuttal of the evidence by defense experts. However, Judge Pollack took the unusual step of not allowing experts to give their ultimate opinion of whether the suspect's print did or did not match the evidence. His contention was that fingerprint analysis did not meet the criteria established by the Daubert case. Specifically, he was concerned with the lack of a well documented 'error rate' and the differing standards between jurisdictions. For example, in the Mitchell case it was noted that in the UK an analyst must note 16 unique 'Galton points' to declare a 'match' between samples. In Australia the number of Galton points necessary for a match is only 12. Both Canada and the FBI in the USA have no minimum standards established to declare a match. Within the USA there are various Galton point number requirements depending on the jurisdiction. Against this backdrop. Judge Pollack felt it was acceptable to allow experts to discuss the methodology for fingerprint analysis as well as the number of points found in common between two sets of prints. Yet, he prevented the experts from giving their opinion as to whether a match occurred. The jury would ultimately decide how much weight to give to the testimony offered by the experts.

UK Influence

The prosecution urged Judge Pollack to reconsider his ruling and in March of 2002 he issued a new ruling which, to the relief of fingerprint examiners, reversed his previous position. Judge Pollack re-evaluated the evidence in the record and heard new testimony from experts in the USA as well as from the UK. Of particular interest was the revelation that the UK had actually abandoned the 16-point requirement and was using essentially the same methodology as the FBI. This came from the UK case of Regina v Buckley in 1999.'

428 Jeff Wise

In that case, the British Court of Appeal (Criminal Division) ruled that 16 points in agreement are not necessary for a 'match'. This opinion also noted that the UK Fingerprint Evidence Project Board recommended that a 'non-numerical system' be adopted by April of 2000. This deadline was missed but by June 2001 the UK abandoned the 16-point requirement for a match and now uses an entirely non-numerical system. In light of the similarity between the UK and the USA, Judge Pollack ruled that there was a clearly established standard of analysis in the scientific community to satisfy the Daubert requirements.

Lessons Learned

The key point that swayed Judge Pollack to allow fingerprint evidence was the realization that there was a uniform standard between the USA and the UK (where fingerprint analysis originated). Analysts using new biometric technologies should heed this and ensure that there are uniform standards in their respective industries. It is important to note that a key reason for the inclusion of fingerprint evidence even though Judge Pollack had reservations about the lack of scientific testing is the fact that fingerprint evidence has been used worldwide for the last 100 years. Facial recognition systems, ear print identification and the like will not have this luxury. For these emerging technologies, the criteria established by Daubert should be the guiding principle. In order to pass muster the science must be rigorously tested, subject to peer review, have a known error rate, be widely accepted and subject to uniform standards. Since the Mitchell case in 1999, there have been 40 separate challenges to the admissibility of fingerprint evidence.'" In all of these cases the fingerprint evidence has been allowed in court.

Origins of DNA Analysis

DNA or deoxyribonucleic acid is the biological material that codes for all living processes. Forensic DNA analysis is based on examination of the sequence of the DNA code. The sequence of DNA can be thought of as a string of letters that when combined in a certain order form words, paragraphs and even chapters of instructions for the body. The entire DNA alphabet has only four letters (A, T, C and G) but the length of the DNA sequence is three billion letters long. Most of the DNA between individuals is the same, which makes sense because we all have similar characteristics. For example, we all have 10 fingers and toes, digest food the same way and require oxygen to breathe. However, some of the DNA between individuals is different and these differences allow DNA analysts to determine beyond a reasonable doubt if a sample matches that of a suspect. James Watson, Francis Crick and Maurice Wilkins discovered the structure of DNA in 1953 while at Cambridge University in England. Their work led to the 1962 Nobel Prize in Physiology or Medicine. Thirty years after its discovery, law enforcement officials first used the specificity of DNA in a criminal trial. In 1984, Dr Alec Jeffreys worked as a researcher studying gene structures." During this process he discovered a stretch of DNA that repeated in the middle of the gene sequence. When he compared this repeating sequence in different individuals, he found that the number of times the sequence repeated varied from person to person. He realized that this discovery could lead to establishing a person's identity from their DNA sequence so he shifted his work into researching this possibility. In 1986, Jeffreys used his newly developed techniques to aid law enforcement in the pursuit of a suspected serial murderer.

Challenging Methods of Forensic Identification 429

On 2 August 1986 police found a woman's body in the woods near Narborough in England. The first suspect the police arrested readily confessed to the murder. The case would have ended at this point except the police noticed similarities in this case to another unsolved murder 3 years prior. The police approached Jeffreys to see if he could use his DNA 'fingerprinting' technique to link their suspect to both of the cases. They provided Jeffreys with the evidence material from both cases as well as a blood sample from their suspect. After a week of work, Jeffreys concluded that the genetic material from both cases came from the same perpetrator but shocked authorities by concluding that the murderer was not the man the police had in custody. As is often the case in forensics, the police solicited an outside opinion. A separate and independent laboratory came to the same conclusions. Reluctantly, the police released their suspect. Five months later, authorities began a DNA dragnet to attempt to find their murderer. They solicited blood samples from every male from the surrounding areas. Although they collected over 5500 samples, none of the samples matched the DNA of the perpetrator. In the end, old-fashioned police work solved the case. A man from a neighboring area admitted to coworkers that he submitted his blood in lieu of a friend who was convinced the police were out to get him. When the pohce apprehended this man's friend and tested his blood, they found that it matched the DNA from both the girl's murders. A number of factors make this case unique. First, the initial case involving DNA was used to exonerate the suspect. Second, it demonstrates the relative infancy of DNA evidence. The entire court record of DNA evidence only goes back to 1986. Finally, authorities quickly realized the benefits of establishing a database of DNA profiles.

Evolution of DNA Technology

The pioneering work done by Jeffreys involved Restriction Fragment Length Polymor- phisms or RFLPs. This technique uses restriction enzymes to cut the DNA chain whenever a specific cut site occurs. Different individuals have different numbers of cut sites; this leads to different size fragments. If enough enzymes are used, the fragmentation products form a genetic fingerprint. The limitations of RFLP analysis are that it requires a large amount of intact DNA and it is time consuming. A new method developed in the 1990s solved these problems. Short Tandem Repeat (STR) analysis is the current standard for DNA analysis and can be used on degraded DNA and has the further advantage of reducing the labor time. STR analysis requires the use of the Polymerase Chain Reaction (PCR), which produces multiple copies of the DNA for analysis. This allows a relatively small amount of DNA to be useful to an analyst. The PCR reaction goes through 28 cycles and with each cycle the number of DNA molecules doubles. Twenty-eight cycles of DNA amplification produces approximately 250,000 copies of each individual DNA molecule in the sample. This exponential replication can easily result in a billion copies of the overall DNA sample. The use of STR analysis has supplanted RFLP analysis as the preferred method in US laboratories. In 1998 there were still 43 laboratories using RFLP analysis but by 2001 only one forensic laboratory continued to use this technology.''̂

The STRs are regions of DNA that have repetitive sequences. Different individuals have different numbers of these repeated areas. For example, one individual may have an area that repeats 10 times and another may have an area that repeats eight times. If only one repeat area is examined the results are not very conclusive. Using the previous example, it may be shown that only 4% of the population has the 10 repeat regions. In court, a one out of 25 chance that the suspect is the killer is not conclusive enough for a conviction.

430 Jeff Wise

Therefore, with STR analysis in the USA, 13 different areas, or loci, are examined. When the probabilities for each of these areas are combined the exclusivity of the analysis is remarkable. The FBI states that the random probability of two unrelated people having the same characteristics using these 13 different loci is less than one in a trillion.''

A final form of DNA analysis involves mitochondrial DNA or mtDNA. Mitochondrial DNA is different from standard or nuclear DNA in that it is maternally inherited and does not have near the variability that nuclear DNA offers. However, it can be extremely useful for identifying missing persons when you have access to the family members. The FBI began feasibility testing in the late 1980s and formally began examination of evidence in June of 1996.''' In contrast to nuclear DNA analysis, mtDNA involves determining the 'letter by letter' sequence of an approximately 600 base sequence. Individuals are compared along each of these over 600 positions to determine if a match is present.'^ This process is more costly and time consuming than conventional DNA analysis but it offers an alterna- tive when dealing with highly degraded DNA such as with skeletal remains.

Standard DNA analysis uses nuclear DNA, which is found in the nucleus of individual cells. In degraded DNA the bonds in the DNA molecule begin to break and fragmentation results. This degradation can result in breaks in the STR sections of the DNA molecule that DNA analysis relies upon. Mitochodrial DNA is an attractive option in these cases because there are hundreds to thousands of mitochondria within each cell. Even if degradation of the mtDNA is present, by sequencing a number of them you can determine a consensus sequence for the individual. This technology was demonstrated in 1994 when members of a forensic team used mtDNA analysis to identify the remains of the Romanov family."^

Challenges to DNA Evidence

Each new DNA technology to enter the legal field has been challenged in court. Because DNA analysis began in the late 1980s and given the length of the appeals process there are still numerous cases yet to be resolved. A typical case challenging the first form of DNA analysis, RFLP, is U.S. v Bonds." This case originated in the state of Ohio and involved the shooting death of David Hartlaub at a bank night depository. Bonds attempted to have the DNA evidence suppressed which claimed there was a 1 in 35,000 chance of a random match with his DNA.'* The District Court ruled that the DNA evidence could be admitted based on the 'general acceptance' rule of the Frye standard." When the case reached the U.S. Court of Appeals for the 6th Circuit in 1993, the Daubert standards were the appropriate benchmark for admissibility. The Court of Appeals held that RFLP analysis methods are admissible under the Daubert criteria. Specifically, the methods used have been tested, subjected to peer review and were generally accepted in the scientific community. The court did note that although the laboratory conducted internal proficiency tests it did not conduct external blind proficiency tests or specifically reference a known error rate.^" However, the court ultimately gave more weight to the 'general acceptance' of the method in the scientific community and less weight to the 'error rate' arguing that if the scientific community accepts the technology it must be accepting of the error rate as well. This highlights the situation created by Daubert, which requires judges, who are often not trained in the sciences, to act as gatekeepers of evidence admissibility.

Another case on the admissibility of RFLP analysis methods occurred in People V Castro.^^ In this case the Supreme Court of New York State used a three-prong

Challenging Methods of Forensic Identification 431

test under the Frye standard to determine DNA admissibility. The three prongs of the test were:

1. Is there a theory that is generally accepted in the scientific community, which supports the conclusion that DNA forensic testing can produce reliable results?

2. Are there techniques or experiments that currently exist that are capable of producing reliable results in DNA identification and which are generally accepted in the scientific community?

3. Did the testing laboratory perform the accepted scientific techniques in analyzing the forensic samples in this particular ?̂ ^

The Supreme Court of New York ruled that prongs one and two were met because of the general acceptance of DNA analysis procedures in the scientific community. However, the court ruled that although the science of RFLP analysis is sound, it could still be performed in an unsatisfactory manner if the specific laboratory does not meet current standards. In this case, the analysis done by Lifecodes Corporation was found to be below the standards set by the scientific community. Furthermore, the population genetics calculations used by Lifecodes personnel to indicate the likelihood of a DNA 'match' were contradicted by several leading experts in the field. The end result in this case was that some evidence was admitted and some evidence was excluded. Although this was a setback for Lifecodes Corporation, this case does validate the science and methodology of RFLP analysis but shifts the burden onto the individual laboratories and staff.

The advent of STR analysis was quickly challenged in many courts around the USA. A typical case is State v Traylor.^ This case decided by the Supreme Court of Minnesota in 2003 affirmed the use of DNA analysis using STR loci. One of the issues raised by the defendant was that multiple guidelines exist for proper DNA analysis. The Technical Working Group on DNA Analysis Methods (TWGDAM) was created in 1988 and composed of many leading scientists in DNA analysis. In 1994, Congress passed the DNA Identification Act, which authorized the FBI to create national guidelines and appoint a DNA Advisory Board (DAB).̂ '' The defendant in this case argued that the guidelines set forth by the DAB, which the testing laboratory followed, do not require as much disclosure of information. At issue were proprietary 'primers' used in the PCR reaction provided by commercial DNA test kits. Traylor argued that because the commercial companies did not divulge their proprietary information, he did not have full access to information on the validity of the DNA test kits." The court ruled that although the TWGDAM standards have been upheld in prior DNA cases, they are not absolute. The newer standards established by the DAB meet admissibility criteria and thus validate the science. This case is important because of the emphasis placed on following established DNA guidelines.

Challenges to mtDNA technology are relatively new to the courts as a result of its recent inception. In U.S. v Coleman, the United States District Court for the Eastern Division (Missouri) ruled in favor of mtDNA testing.̂ * The Court stated that mtDNA analysis constitutes a scientific knowledge based on reliable methods and principles.^^ The court ruled that ultimately, the expert testimony provided would aid the triers of fact in rendering their verdict.

Future Trends in DNA Analysis

Standard STR analysis uses the polymerase chain reaction (PCR) for 28 cycles. In 28 cycles a single DNA sequence is copied over 250 million times. A new method being used in some

432 Jeff Wise

jurisdictions around the world is Low Copy Number (LCN) DNA. In contrast to current methods, LCN DNA involves increasing the number of PCR cycles from 28 to 34. This may seem like a small adjustment but while 28 cycles yields approximately 250 million copies, 34 cycles increases that number to over 17 billion copies. This increase of DNA allows researchers to determine DNA profiles from previously unusable sources such as fingerprints, cigarette butts, drinking glasses or clothing simply worn by the suspect.

In 1999, the Forensic Science Service in England began using LCN DNA in forensic casework. They have had a number of success stories involving 'cold cases' where a DNA profile was previously unattainable.^* In 2001, researchers at the laboratory division of the FBI issued a paper listing some considerations on the use of LCN DNA.^' One of the primary concerns with LCN DNA is the risk of contamination created by increasing the number of PCR cycles from 28 to 34. As the size of the original sample gets smaller any contamination present will have a larger effect on the results of the analysis. Another concern of the FBI is that any casual contact between a suspect and a victim prior to a crime could lead to the transfer of LCN DNA, and that current research does not show how long this DNA could remain. Finally, and most importantly, with LCN DNA the sample size is so small that there is not enough material left after testing to get an independent source to evaluate the data. This makes thorough peer review of LCN laboratories impossible. Some research has been done to begin to understand the mecha- nisms involved in transfer of trace DNA. For example, 'shedder indexes' are being investigated to determine the rate at which an individual donates potential DNA material (i.e. skin cells, hair, sweat).^" The persistence of DNA complicates these investigations. Researchers have reported that DNA can be detected after transfer to an object for nearly three months in some cases and in one case for 2 years.^'

The FBI did see some application of the LCN techniques. Any sample material that can be thoroughly cleaned prior to testing to reduce the possible contamination (e.g. bones, teeth and hair shafts) may benefit from LCN DNA. Yet, this was not enough for the FBI to endorse LCN DNA. With the exception of the limited application in human remains identification, the FBI remains skeptical of LCN DNA. Their official position is that any profiles obtained from LCN DNA should not be entered into the Combined DNA Index System (CODIS) database of offenders and suspects. They also issued a caution against a rush to re-examine old cases on the hopes that LCN DNA would offer better analysis or change a verdict, mainly because of the risk of evidence contamination from repeated handling.

Another new arena in DNA analysis targets the Y Chromosome. Where mtDNA is maternally inherited, males carry a single Y chromosome, which is paternally inherited. By targeting STRs on the Y chromosome, forensic scientists can separate a mixture of male and female DNA such as in sexual assault cases. Y chromosome analysis is also helpful in determining paternal lineage.̂ ^ The lengths of DNA we are interested in also continue to shrink. Originally DNA fingerprinting was done on relatively long RFLP sites. Modern analysis evolved to STRs (four to five base pairs) and some now advocate the use of SNPs or single nucleotide polymorphisms. Because the variability of a single site can only include one of the four DNA bases (A, T, C or G) it is necessary to analyze hundreds to thousands of these regions to establish individual identity. Outside of the forensic arena, SNPs are commonly used in diagnosis of genetic disorders. For example, a single base substitution can result in disease such as sickle cell anemia. In sickle cell anemia, an A (adenine) is replaced with a T (thymine) in the j?-globin gene.̂ ^ Multiple SNPs can be investigated at the same time by using micro arrays commonly called a 'lab on a chip'. Although SNPs can

Chaiienging Methods of Forensic Identification 433

offer additional information when identifying an individual, they will not likely supplant the already existing database methods used in many of the leading countries with DNA analysis. The US CODIS database maintained by the FBI stores STR data. To overhaul the system and replace STR data with SNPs or any other novel system would be cost prohibitive and unnecessary given the accuracy of the current system. Furthermore, privacy advocates are concerned with the potential for abuse if government agencies increase the amount of information they gather during DNA analysis. In fact. Sir Alec Jeffreys himself voiced these concerns in September at the 20th anniversary of his development of DNA fingerprinting.

Conclusion and Recommendations

Fingerprints, and now DNA, are the classic 'biometrics' in the arena of individual identification. The Daubert challenges confronting these two technologies will surely be aimed at the emerging biometric technologies. The Biometric Consortium in the USA defines biometrics as '.., automated methods of recognizing a person based on a physiologi- cal or behavioral characteristic'. '̂* Biometrics attempt to identify an individual by measur- ing features such as the 'face, fingerprints, hand geometry, handwriting, iris, retina, vein and voice'. As we increase the use of biometric measurement devices (such as facial recognition systems in airports to deter terrorism and illegal immigration) we also increase the hkelihood of that technology being called into question during a criminal proceeding. Purveyors of these new devices should pay close attention to the disposition of fingerprint and DNA cases. Daubert continues to be the criteria to meet. As shown in the Castro case, once the technology is 'generally accepted' the next step for defense attorneys is to critique the application of the method by the individual laboratory. If the laboratory meets the scientific standards then the last stop for an appeal will be to discredit the training and/or performance of the individual analyst.

Fingerprint challenges came a century after the practice began. It seems nearly impossible for any court to invalidate 100 years of court evidence. However, the scrutiny of fingerprint analysis resulted in a tightening of the standards both in the USA and around the world. Similarly, scientists involved with DNA analysis worldwide collaborate at annual meetings and academic conferences to further the science and ensure justice. Based on the court ruhngs during the 1990s to the time of this writing the future of fingerprint and DNA evidence seems secure. For the newer biometric technologies the future remains to be seen.

Notes and References

1 This article is an expanded version of an article first published in Crime and Justice International Vol 20, No 81, pp 39-40, July/August 2004,

2 Ray Wickenheiser 'Trace DNA: a review, discussion of theory, and application of the transfer of trace quantities of DNA through skin contact' Journal of Forensic Science Vol 47, No 3, pp 442-450, 2002,

3 Richard Saferstein Criminalistics 7th edn. Prentice Hall, Upper Saddle River, NJ, 2001, p 3, 4 For an in depth look at the historical development of latent print analysis see Colin Beavan

Fingerprints Hyperion, New York, 2001, 5 Frye v U.S., 293 F, 1013 (D,C. Cir,, 1923) 6 Daubert v Merrel Dow Pharmaceuticats, 509 U,S, 579 (1993). 7 V.S. V Mitchell, Cr, No, 96-407 (1999). 8 U.S. V Llera Plaza, Cr. No, 98-362-10,11,12 (13 March 2002),

434 Jeff Wise

9 Regirta v Buckley, 143 SJ LB 159 (30 April 1999). 10 For a running tally of the challenges to latent print admissibility see http://www.onin.com/fp/

index.htm. Visited 31 August 2004.

11 Matt Ridley Genome Harper Collins, New York, 1999, pp 132-133. 12 See U.S. Department of Justice Survey of DNA Crime Laboratories 2001, p 7. 13 B Budowle, G Carmody, R Chakraborty and K Monson, 'Source attribution of a forensic DNA

profile', US Department of Justice, Forensic Science Communications, Vol 2, No 3, July 2000, http://www.fbi.gov/hq/lab/fsc/backissu/july2000/source.htm

14 A Isenberg and J Moore, 'Mitochondrial DNA analysis at the FBI laboratory', US Department of Justice, Forensic Science Communications, Vol 1, No 2, July 1999, http://www.fbi.gov/hq/lab/ fsc/backissu/julyl999/dnalist.htm

15 Ibid. 16 P Gill, P Ivanov, C Kimpton, R Piercy, N Benson, G Tully, I Evett, E Hagelberg and K Sullivan

'Identification of the remains of the Romanov family by DNA analysis' Nature Genetics Vol 6, pp 130-135, 1994.

17 U.S. V Bonds, 12 F.3d. 540 (CA6, 1993). 18 Ibid at 551. 19 At the time of the District Court hearing (1990), the Frye standard of 'general acceptance' was the

appropriate case precedent. Once the case reached the 6th Court of Appeals (1993) the Daubert standard was the proper standard for admissibility.

20 U.S. V Bonds at 559. 21 People V Castro, 545 N.Y.S.2d. 985 (N.Y.Sup., 1989). 22 Ibid at 987. 23 State v Traylor, 656 N.W.2d. 885 (Minn., 2003). 24 Ibid at 895. 25 Ibid at 896. 26 U.S. V Coleman, 202 F.Supp.2d. 962 (E.D.Mo., 2002). 27 Ibid at 970. 28 For a description of the LCN DNA methods used by the Forensic Science Service and success

stories see http://www.forensic.gov.uk/forensic_t/inside/news/docs/DNA_LCN.doc. Visited 31 August 2004.

29 B Budowle, D Hobson, J Smerick, A L Smith 'Low copy number—consideration and caution' Laboratory Division, FBI, Washington, D . C , 2001.

30 C Murray, A Lowe, P Richardson, R Wivell, P Gill, G Tully, J Whitaker 'Use of low copy number (LCN) DNA in forensic inference. Forensic Science Service' Presentation at the 12th International Symposium on Human Identification, 2001.

31 R van Oorschot and M Jones 'DNA fingerprints from fingerprints' Nature Vol 387, 1997, p 767. 32 L Roewer, P Knijff and M Kayser 'Y chromosome STR analysis in forensic practice' Presentation

at the 2nd European Symposium on Htiman Identification, 1998. 33 W Coleman and G Tsongalis Molecular Diagnostics for the Clinical Laboratorian Humana Press,

Totowa, NJ, 1997, p 146. 34 http://www.biometrics.org/html/introduction.html. Visited 31 August 2004.