87th SEMI-ANNUAL SEMINAR (Spring 1996)
CALIFORNIA ASSOCIATION OF CRIMINALISTS

NORTH OF 49 - THE DEVELOPMENT OF FORENSIC SCIENCE IN CANADA (FOUNDER'S LECTURE)
Douglas M. Lucas, Ontario, Canada

Many Canadians consider it ironic that the world's second largest country has been little known in (or of any great interest to) the rest of the world until, in recent years, there has been a serious question of its survival as a nation. The two questions which we are asked most often by our friends in other countries deal with the "Quebec situation" and our health care system. In our own field, forensic science, few would be aware that, for example, the first forensic science laboratory in North America was established in Montreal many years before there was even a suggestion for a laboratory anywhere in the USA.

This presentation will try to deal with this irony by outlining a little bit about the history of Canada, a little more about its system of justice, and a lot about forensic science as it is practiced here. The latter will include descriptions (and some personal anecdotes) about personalities, organizations and cases going back to the mid-1800's, (although some from that era will not be so personal). Some examples: Who was Dr. King and what can we learn from him? Who set the ecclesiastical robes of Bishop Strachan, the Chancellor of the University of Toronto, on fire and why? What forensic work was done on the first bombing of a commercial airliner? Was there really a forensic scientist named Tiger Dunlop? Do the RCMP always get their man?

This presentation will undoubtedly confirm that old forensic scientists never die, they merely reminisce.


ATLAS OF HUMAN HAIR MICROSCOPIC CHARACTERISTIC VARIATES
Robert R. Ogle, Jr. Forensic Consultant; Michelle J. Fox. Forensic Analytical Specialties

This study (in progress) involves the production of a hair atlas containing photographs of variates of the microscopic characteristics used by hair examiners in the comparison of human hairs. Since virtually all of the microscopic characteristics utilized by the hair examiner are continuous variates, the study will select appropriate representative variates for each characteristic as the archetypes for the class limits for the variates.

The establishment of the class limits for each variate is the first essential step in determining the frequency of a characteristic's variates in a study population. Previous studies relied on inter-comparison of randomly selected hairs without regard for frequency data for the hair characteristics or the hair types in the studies, owing to the lack of archetypes for either the variates or the hair types. Through use of the hair atlas, researchers will be able to determine frequency data for the variates in a study population, and, by extension, the frequency of a specific hair type in the study population (a hair "type" is the configuration of the characteristic variates for that hair).

Methodology includes the photography of each variate archetype, hair casts, and unusual features for inclusion in the atlas. The atlas will be produced in both text and CD ROM formats.


CHARACTERIZATION OF DUCT TAPE ADHESIVE BY MICRO-FOURIER TRANSFORM INFRARED SPECTROSCOPY (MICRO-FTIR) AND SCANNING ELECTRON MICROSCOPY WITH ENERGY DISPERSIVE SPECTROSCOPY (SEM-EDS)
Amy L. Mongan, Forensic Analytical Specialties

Duct tape is a common adhesive tape used in applications such as HVAC (Heating, Ventilation, and Air Conditioning), bundling materials, packaging, sealing joints, patching furniture and automotive repair. Duct tape is occasionally encountered as evidence in crimes against persons and property due to its utility and binding strength. A good physical match between two pieces of duct tape provides strong evidence that the two pieces were at one time joined. There are cases, however, when a physical match is not possible or when a physical match proves to be inconclusive. In these instances, the characterization of the chemical composition of the duct tape adhesive could further support or refute the possibility of an association. This paper explores the potential of distinguishing manufacturers and grades of duct tape by the chemical composition of the adhesive. Using FTIR and SEM-EDS analysis, it was found that the composition of the adhesives exhibit substantial variation which allow for the discrimination between manufacturers and grades of duct tape.


BEHAVIORAL ASPECTS OF CRIME SCENE RECONSTRUCTION
Mike Prodan, California DOJ, Bureau of Investigation

Criminal Investigation Analysis (Profiling) has assisted law enforcement in narrowing the scope of investigations since the early 1970's. One stage of this process is the reconstruction of the crime itself. In addition to the "physical" aspects, an offender will often leave "behavioral" clues at a crime scene. Case examples will be shown to demonstrate how these clues can be recognized and utilized in identifying the motive and the offender.


THE USE OF DIGITAL IMAGING (PHOTOCD™) IN THE CASE OF THE MENENDEZ BROTHERS
Charles V. Morton, IFS/Criminalistics Laboratory

All of the photographic images generated at the crime scene and at the autopsies in the shotgun deaths of the parents of Lyle and Eric Menendez were scanned onto PhotoCD™ discs. The images were then examined by "experts" for both the prosecution and defense.

PhotoCD™ technology provides an ability to view each photograph on a television screen or on a computer using appropriate software. The television presentation is convenient and readily available and allows the photographs to be viewed at full frame, full screen or an approximately 2X enlargement of any area of the screen image. In addition to the greater enlargement of fine detail that can be accomplished with the computer, software which allows "enhancement" and analysis can also be applied to the images. Although any image can be "enhanced" or manipulated, the original image is not altered and can be compared to any altered image. Prosecution and defense analysts used these images in the Menendez case to support and refute various interpretations of the evidence. As with any other type of evidence, the support for any conclusions drawn was based on scientific considerations, not the high-tech nature of the medium. How these images were used and the potential value and limitations for other investigations will be demonstrated.


TECHNICAL ADVANCES IN FORENSIC ENTOMOLOGY: MOLECULAR IDENTIFICATION OF SPECIES AND STATISTICAL ESTIMATION OF THE POSTMORTEM INTERVAL
Dr. Jeff D. Wells* and Felix A.H. Sperling, Department of Environmental Science Policy and Management - Insect Biology, Univ. of California Berkeley

Arthropods provide two time-dependent processes that may be used to estimate the postmortem interval (PMI) in cases of human death. The first is the growth of a larva, e.g. a blow fly maggot that begins development after being deposited on a corpse. The second is the turnover, or succession, of species present as decomposition proceeds. We have found that mitochondrial DNA sequence data can be used to identify otherwise problematic larval specimens, allowing investigators to consult the relevant references on growth and succession. Sequence data were used to identify species-specific restriction sites, permitting a relatively quick identification based on restriction fragment lengths from a short portion of the genome. We have also developed novel statistical methods for expressing the confidence one has in a PMI estimate based on insect evidence, that is, what are the odds that at a given PMI value is correct. Conventional regression models were modified to predict maggot age from length or weight, and this analysis can be applied to some studies currently consulted by forensic entomologists. A likelihood-ratio approach was also developed for the analysis of succession data, but this technique requires experimental data not currently available.


HOW GOOD ARE WE? - QUALITY AND ETHICS IN FORENSIC SCIENCE
Douglas M. Lucas, Ontario, Canada

There has probably never been as much interest in forensic science on the part of the general public in North America as there has been in the past few years. The publicity associated with the introduction of DNA profiling evidence, a few high profile cases and (it must be admitted) some spectacular charlatans in our midst can account for much of this interest. As in the courts where bad cases make bad law, so in forensic science our public image is largely based on atypical events. The profession has always been concerned about quality, and initiatives to continuously improve it have come from within. Quality has two aspects - performance and ethics. This presentation will briefly outline personal recollections and anecdotes about developments in both aspects over the past 25 years.


LESSONS FROM THE TRIAL - THE PEOPLE VS. O.J. SIMPSON
Gerald Uelman, Santa Clara University School of Law

Perhaps the O. J. Simpson trial is best understood as a cultural phenomenon rather than a lesson plan. Any event that captures such widespread public attention for such a sustained period has profound cultural repercussions. The slow white Bronco has been transformed from a vehicle into a metaphor. We have a whole new galaxy of celebrities; Kato Kaelin, Mark Fuhrman, Lance Ito, Marcia Clark, and Johnnie Cochran. We will never think of the "N word" in the same way.

As a law professor who spent his sabbatical right in the middle of it, however, I kept looking for lessons. I'd like to share five of the hardest. I say they're the "hardest" not because they are difficult to comprehend but because there are no simple solutions to remedy them. Yet for each, those who were unhappy with the verdict have proposed a simplistic remedy with little thought to the long-term consequences. H.L. Mencken once said that for every problem, there's an obvious solution that is quick, easy...and wrong. Just as we argued against a rush to judgment during the trial, I would counsel against a rush to solutions after the verdict, lest we simply compound our problems.


BUILDING A DREAM ... A STEP-BY-STEP PROCESS
Robert A. Jarzen, Sacramento County Crime Lab

Generally speaking, the time from realizing a need for a new crime laboratory, through the initial planning and facility design stages, finally culminating in the move-in to the new facility can be as long as ten years.

Crime Laboratory planning is complex and expensive. In addition to being logically ordered, the planning decisions involved at each step must be based on sufficient data to insure that the facility will be efficient, economical, flexible, productive and creative.


CHANGES IN SPEED: AUGUST 1995
Jerry Massetti, CA State Dept. of Justice Fresno Lab

On August 19, dl-amphetamine hydrochloride began showing up in suspected methamphetamine samples submitted to CA State DOJ Crime Laboratories. Only rarely had amphetamine been encountered in routine drug case submissions since the mid 1980's. Despite this, within only a few weeks after its first appearance, dl-amphetamine/ d-methamphetamine mixtures were detected in suspected methamphetamine samples by crime laboratories in at least six Western US states. Midwest and Eastern US labs reported it by late November.

At the same time as amphetamine appeared, the amount of active ingredient dropped dramatically. Very large amounts of caffeine and nicotinamide were used as diluents. In October an increasing number of samples were found to be diluted with dimethylsulfone. Changes in the physical appearance of methamphetamine samples corresponding to chemical changes also were noticed.

dl-Phenylpropanolamine has been substituted for single isomer ephedrine or pseudoephedrine in large scale clandestine laboratories using the HI / red P reduction. Phenylpropanolamine is being extracted from commercially available tablets. Other medicinal compounds like triprolidine and chlorpheniramine carry over into the final products.


SEPARATION AND QUANTITATION OF d- AND I- METHAMPHETAMINE
Pennie I. Laferty, Orange County Sheriff-Coroner Department

The separation and quantitation of the d- and l- isomers of methamphetamine was required to determine the percentage of d- isomer in a case sample of approximately ten kilograms. This analysis was undertaken to satisfy the Federal Sentencing guidelines for "ICE" methamphetamine, which states that a substance must contain 80% or greater of the d- isomer.

Separation of the d- and l- isomers was accomplished by derivatizing the sample with S-(-)-N-(Trifluoroacetyl)Prolyl Chloride and analyzing the derivative by Gas Chromatography/ Mass Spectrometry (GC-MS). Two distinct peaks were observed on the total ion chromatogram from the GC/MS. The l- isomer exhibited a peak at 18.3 (+/- 0.10) minutes; the d- isomer, a peak at 18.7 (+/-0.10) minutes. The peak areas of the d- and l- isomers were added together to give a peak area for total methamphetamine. That area was then used to calculate the ratio percentage of the d- and l- isomer.

Quantitation of the sample was accomplished by determining the percent purity of methamphetamine by UV Spectrophotometry, and multiplying that purity with the ratio percentage of the d- isomer. Over seven kilogram of the case sample contained 80% or greater of the d- isomer.


NITROSONAPHTHOL AS AN ALTERNATIVE TO THE D.T.O. TEST FOR THE DETECTION OF COPPER IN BULLET IMPACT MARKS
Lucien C. Haag, Forensic Science Services

The presence or absence of copper residues in bullet impact marks can have important reconstructive value in certain cases.

The dithiooxamide (DTO) reagent is a sensitive and specific reagent for copper which has been used to detect the presence of copper in bullet ricochet and impact marks for a number of years. Its only shortcoming is the unimpressive color complex formed with copper - a dingy, dark greenish gray response, which is often difficult to discern from other residues that might be transferred onto filter paper or a swab used in the normal testing procedure.

In 1980 2-nitroso-1-napthol was proposed as a substitute for the 8-hydroxyquinoline test used in trace metal detection on the hands of suspected shooters, burglars, etc. This reagent gives a pink to red-orange color with copper and copper/zinc alloys used in bullet jackets and no reaction with lead. Selective transfer of trace amounts of copper on steel surfaces with an ammoniacal solution (comparable to the DTO procedure) allows the copper/copper-zinc response to be visualized free of the dark green color normally formed with iron.

The possible toxic properties of this reagent are minimized though the use of proper equipment and the fact that this technique does not involve applying the reagent to the skin.

This paper will illustrate the proper use of this reagent, a comparison with the traditional DTO test and the reconstructive value of the test results in a simulated case situation.


A SIMPLE TEST FOR NONOXYNOL-9: AN EVALUATION (WITH A SERENDIPITOUS RESULT)
Robert D. Blackledge, Naval Criminal Investigative Service Regional Forensic Laboratory

Nonoxynol-9 is used in spermicidally-lubricated condoms as well as other spermicidal products. Its detection on evidence from sexual assaults may have forensic significance. "A SIMPLE TLC METHOD FOR IDENTIFICATION OF NONOXYNOL-9 IN CONTRACEPTIVE PREPARATIONS" is the title of a paper(1) published in the Indian Journal of Forensic Sciences. It details a procedure that uses methanol extracts, silica gel-coated TLC plates, a methanol solvent system, and a potassium dichromate acidic solution as a visualizing spray.

I have tested the method using a nonoxynol-9 standard and methanol extracts from a variety of condom brands, feminine hygiene products, and lubricants. Although disappointing for nonoxynol-9, it proved to be an excellent screening test for propylene glycol, a principal ingredient in the gel-type lubricant used by some condom manufacturers.

Results will also be presented from an actual case in which the finding of traces from a spermicidally-lubricated condom having a gel-type lubricant was crucial to the suspect's defense.


THE FATE OF MICROCRYSTALLINE TESTS TODAY AND IN THE FACE OF ACCREDITATION
Brooke Carpenter, Katie Tunnell, and Jean Arase, Santa Clara County Crime Lab

In preparation for accreditation by ASCLD, the narcotics unit of the Santa Clara County Crime Laboratory undertook the task of studying the effects of eight microcrystalline reagents on 100 controlled and non-controlled substances to determine the "uniqueness and specificity" of each test.

This study was, in part, performed because of a current trend in forensic laboratories across the U.S. to move toward only instrumental analyses for the confirmation of controlled substances. Today, microcrystalline tests are being replaced by FTIR Spectroscopy and GC-MS; however, some laboratories still use two microcrystalline tests as a first choice for confirmation of controlled substances. The Santa Clara County Crime Lab is one such lab.

There were three main goals in this study. The first was to substantiate the reliability and specificity of the microcrystalline testing methodology. This included comparing crystals formed by other substances and also noting which substances did not form crystals. The second was to create a reference library of crystal form descriptions to use for comparison while performing case work. This included documentation of the results of the microcrystalline tests and checking for reproducibility by having all analysts view the crystals. The third was to increase the confidence level in the analysts toward their work with microcrystalline tests. This included defining the criteria for an identification of a controlled substance with respect to two microcrystalline tests as a confirmation test. Further, a minimum of 20% of microcrystalline tests were supplemented with an instrumental test.

The Santa Clara County Crime Lab Narcotics Unit uses a combination of two microcrystalline tests to confirm the presence of methamphetamine, cocaine, and phencyclidine. However, the substance cannot be identified unless the results also coincide with positive presumptive color tests.

The goals of this study were all achieved. First, controlled and non-controlled substances tested in this study did not produce the same crystals with the same reagents used in the same manner as the controlled substances confirmed with these reagents. Second, a reference library of crystal forms is now available as a reference source. Third, the analysts have the added confidence that the microcrystalline tests used are not only unique, but are also specific for the drug being tested for.


A SHOT FROM AFAR (?)
Lucien C. Haag, Forensic Science Services

This unusual shooting case involved the death of a passenger in a moving vehicle struck by a full metal jacketed rifle bullet, impacting and perforating the vehicle at a relatively low velocity.

The uncertainty in identifying the location of the vehicle when struck, combined with a number of analytical problems during the initial phase of the crime laboratory's involvement greatly complicated (perhaps even obviated) establishing the most likely location from which this fatal shot was fired. A more rigorous evaluation of the physical evidence carried out several years later revealed some startling finding that called the initial assessment of a long distance shot into serious question.

A review of the various factors and issues in this yet-unsolved case makes it a useful exercise for all criminalists in:


TOOLMARKS FROM THE PAST
Allen J. Boudreau, Fresno County Sheriff's Department

Bullets were recovered at a homicide crime scene and from a box of ammunition in April 1992. Several of these bullets from each source were found to have distinctive and remarkable toolmarks present. These toolmarks are on the heel of the bullets and were found to match. The ammunition box bears a date code establishing the ammunition was packaged in February of 1971. The manufacturer's plant was visited in November, 1994. Prior to that visit, photographs of the toolmarks were sent to the manufacturer for examination and evaluation by engineers and production personnel. At the beginning of the "Tour" a meeting was held with an engineer and production personnel. The origin of the toolmarks and an estimate of the number of bullets manufactured with these toolmarks were offered. The age of the ammunition and the relatively small number of bullets manufactured with these marks established, beyond reasonable doubt, an association between the homicide bullets and the box of ammunition. This association, taken in context with other facts, helped to prove prior knowledge and disprove a burglary theory.


THE CHARLIE CHAPLIN PATERNITY TRIAL
Joe Hourigan, Los Angeles Police Department

v More than fifty years ago there was a trial held in Los Angeles that was widely reported in the news media all across the United States. It involved:

It was the Joan Berry v. Charlie Chaplin paternity trial. This presentation will show and tell the story of this legal precedent setting trial as covered in the newspapers. An overview of the history of the acceptance in court of ABO blood-typing tests in paternity suits will also be presented.


KRYPTONITE BOMBS?
Patricia Dawn Sorenson, Robert D. Blackledge, Edward J. L. Jex, & William P. Herzig , Naval Criminal Investigative Service

Those CAC members, who misspent at least a part of their youth reading comic books, may recall that the fictional element "kryptonite" was the one thing that could render Superman helpless. The above "kryptic" title was selected not because the bombs under discussion are made of kryptonite, but because even Superman with his x-ray vision might not be able to detect their presence.

The secret of these bombs ("improvised explosive devices" or "improvised explosive-actuated incendiary devices"), is a current- carrying wire that connects the various components but may not show up on x-ray machines.

We have obtained samples of these wires and tested them under a variety of conditions. They do work! They don't show up on the fluoroscopes typically used at airports and they easily carry sufficient current to actuate a device. Samples of these wires will be available for inspection; and their properties as well as the results of tests using both dummy (no real explosives present) and actual devices will be described.


NEW TECHNOLOGY IN BREATH ALCOHOL TESTING
Hans Berger, National Dreager

About four years ago, Dreager made the decision to address the problems associated with infrared spectroscopy at the 3.5 micron wavelength. It was realized that 3.5 micron technology had run its course and that any more add-ons would complicate the electronics and would increase equipment costs. With that in mind, our engineering efforts concentrated on designing an entirely new breath analyzer. The new Alcotest 7110 MKIII utilizes the 9.5 micron range of the infrared spectrum for the analysis of breath alcohol. It has been known for years that analyzing breath alcohol at 9.5 microns would provide answers to endless legal and technical questions. However, only recently have advances in electronic components and manufacturing methods made it possible to take advantage of the superior qualities of 9.5 micron technology.

The Draeger Alcotest 7110 MKIII incorporates all of the advantages inherent with 9.5 micron technology with a new generation Draeger fuel cell as a second independent analysis of the subject's breath (both results are displayed on the evidential print out). This new generation fuel cell, found only in Draeger instruments, is alcohol specific (acetone is not an issue), is accurate at very low alcohol concentrations and demonstrates long term stability. This instrument has been approved by DOT and is listed on the Federal Registrar.


TWO NEW BLACK POWDER SUBSTITUTES: THEIR PHYSICAL AND CHEMICAL PROPERTIES
Lucien C. Haag, Forensic Science Services

Pyrodex® has been the only significant alternative to traditional black gun powder since its introduction in the late 1970's for use in small arms. Two new products, Black Canyon Powder and Arco Powder are now available for the same purpose. Although the physical appearance of these two black powder substitutes is quite different they are, for practical purposes, chemically indistinguishable, being composed of ascorbic acid and potassium nitrate. Pyrodex, on the other hand, is composed of potassium nitrate, sulfur, charcoal (same constituents as black powder) plus potassium perchlorate, dicyandiamide (DCDA) and sodium benzoate with the latter two being unique to Pyrodex and detectable in its fired residues.

This presentation will illustrate the properties and performance characteristics of these two new propellants.


MISLEADING TOOLMARKS ON FACTORY AMMUNITION
Allen J. Boudreau, Fresno County Sheriff's Department

Seven boxes of COR-BON 9mm Parabellum ammunition were obtained from the manufacturer for ballistic testing and evaluation as a possible duty round for the Fresno County Sheriff's Department. This ammunition was examined in the laboratory when received from the manufacturer. The ammunition was found to have three different headstamps on the cases. The headstamps are: C-B (COR-BON), RP (Remington) and R-P (Remington). Closer examination of the brass showed toolmarks on the body of some of the cases. There are two types of marks. One is very much like the chamber marks impressed upon discharge in a fluted chamber weapon. The other is consistent with sizing die marks from reloading. A few rounds of ammunition bear both types. These markings could lead to false conclusions concerning expended cartridge cases collected at crime scenes.


EXPERIMENTATION WITH TWO NEW COLORS OF "MIKROSIL"
Frank Cassidy, CA DOJ BFS Santa Barbara Regional Lab

Castings of "Mikrosil" silicone rubber using two new colors, black and white, were compared to castings of the regular brown "Mikrosil". Mixtures of the black & white and brown & white were also examined using the comparison microscope to determine their efficacy in the comparison of toolmarks. From preliminary tests, it is my opinion that the new colors add another dimension to the replication of toolmarks for forensic examination. Photomicrographs of several comparisons will be used to illustrate my reasoning.


SILENCER TESTING USING A SOUND LEVEL METER WITH AN OCTAVE BAND FILTER
Edward Peterson and Mario Soto, Santa Clara County Crime Lab

Federal and California laws describe a silencer as a firearm device which reduces the sound of a gunshot. Current techniques for silencer determination use a simple sound level meter to determine the decibel reduction that a suspected silencer effects on the sound of a gun shot. A sound level meter with a 1/3 octave band filter measures sound levels at various frequencies to give a spectrum of sound. By comparing sound spectrum patterns of a firearm, with and without the silencer, one can obtain data on precisely how the silencer works in lowering, altering, or disguising the sound of a gunshot.


TITLE 17 ALERT
Jeffrey A. Thompson, Scientific Investigation Unit, Huntington Beach Police Department

For those of you who have performed blood alcohol analysis, "Title 17" is an exceedingly familiar term. For the benefit of the rest, it is a set of regulations, administered by the Department of Health Services (DHS), which governs every aspect of forensic alcohol analysis. In spite of technological leaps and statutory changes in recent years, Title 17 has not been modified for over a decade. That will soon change.

Governor Pete Wilson has instituted a regulatory reform initiative, with the goal of reviewing all California State regulations and reducing the excessive regulatory burden on business. This is to be accomplished through repealing, reducing, simplifying or otherwise modifying those regulations determined, on the basis of input from the affected businesses, to be excessive and unduly burdensome. We are already past the first stage, where DHS sought input from crime laboratories about Title 17 regulations.

As we enter the second phase, the CAC/CACLD Joint Public Health Liaison Committee is seeking additional input from CAC membership. We will have additional opportunities for modifying the regulations as DHS begins to implement their regulatory reform plan. As we were afforded only a four day window of opportunity to comment originally, it is doubtful all voices were heard from.

Therefore, if there are any general or specific changes that you would like to see in Title 17, please call or write the Chairman of the Public Health Liaison Committee as soon as possible: (714) 374-1582 FAX (714) 536-7172. Now is the time to act. Another opportunity to change Title 17 is not in the foreseeable future.


WHAT'S NEW IN FIRE INVESTIGATION?
John DeHaan, California Criminalistics Institute

In a word, PROGRESS. Not that fire investigation hasn't improved in the last forty years, but progress towards making it more of a science than an art has been notable in just the last five years or so. Fire engineering is playing a larger role in understanding what starts fires and what causes them to spread. The knowledge of fire engineers and fire scientists is being integrated into investigation, with the process being driven by the NFPA 921 Guide to Fire Investigation and state-of-the-art textbooks on investigation. With a better understanding of fire chemistry comes a better understanding of the fire environment including what gases are present (or absent) and their temperatures. From this, investigators can make a better prediction of the effects on human victims in a fire and what they are capable of doing as the fire grows.

The focus of the investigation is still: "Where did the fire start and what caused it," but by focusing on what the first fuel ignited was at the origin, we can better predict (or eliminate) likely sources of ignition. Identifying the first fuel ignited, then, allows some streamlining of the "cause" investigation phase and helps characterize how big the initial fire is going to be and how it is likely to interact with other fuel packages nearby.

The use of canines to help search scenes for the presence of flammable liquid accelerants has dramatically increased in the last few years. Faster scene searches with fewer samples, each with a high probability of being "positive", was the intent of these programs. The quality of the canine teams varies considerably and the lab must be intimately involved in the validation and maintenance of any canine being used to screen samples. Many of these topics will be explored in full in the workshop as well as in the presentation.


RECOVERY AND ANALYSIS OF ANCIENT DNA
Scott R. Woodward, PhD., Department of Microbiology, Brigham Young University

Over the last decade it has been possible to isolate, amplify and sequence DNA from a wide variety of ancient sources. Despite the apparent methodological problems, aDNA has been recovered from organisms hundreds, thousands and even millions of years old. However the consistency with which aDNA is recovered has been highly variable. This seems to be primarily due to the extreme fragmentary nature of the aDNA and the presence of PCR inhibitors that frequently co-purify with the aDNA in many extraction procedures. The difficulty involved in isolating and amplifying aDNA has therefore resulted in many studies that have focused on only one particular locus from a single individual or animal. Even though these reports have proven to be valuable in addressing many phylogenetic issues, and individual cases of identification, the need to perform more informative population-sized studies of ancient groups and in individual cases of extreme importance such as often found in forensic cases, it is necessary to consistently recover reproducible DNA sequences of multiple loci from a large number of individuals. Our laboratory has been developing three approaches to mitigate these problems. 1) The comparative use of resins, chaotropic agents, columns and silica beads have been investigated for the initial isolation of the DNA and separation from inhibitory substances. 2) Reconstruction of the fragmentary pieces of DNA by a pre-PCR repair procedure. 3) Random amplification of any existing DNA sequence by totally degenerate primers followed by locus specific PCR. Using these techniques we have substantially increased our aDNA recovery from Egyptian mummies, ancient parchment, and extinct animals. We have also been able to recover information on nuclear loci that compliments the mitochondrial loci used in most ancient DNA studies.


SPECIES SPECIFICITY AND FORENSIC CASEWORK VALIDATION OF THE D1S80 AFLP AND THE CSF1PO/TPOX/TH01 STR MULTIPLEX
George Riley, Teresa H. Aulinskas and Howard C. Coleman GeneLex Corporation

We have evaluated the species specificity of the Perkin Elmer D1S80 and Promega CSF1PO/TPOX/TH01 (CTT) multiplex kits as part of our ongoing forensic and paternity casework validation program. In addition to interspecies testing, limits of detection were established for D1S80 and CTT and these systems were validated using samples from a series of adjudicated case.

The species specificity of the D1S80 and CTT kits were tested using DNA extracted from 27 different species, including: 14 non-human primates, 8 common game, farm and household animal species, and 5 human microbial pathogens or commensals. D1S80 and CTT cross-reacted strongly with certain non-human primate DNAs: D1S80 cross-reacted with only three primates (chimpanzee, gibbon and orangutan) and at least one system of CTT cross-reacted with all but two of the primates tested (spider and squirrel monkey). In testing the limits of detection, D1S80 and CTT both gave unambiguous typing results with =0.1 ng of human template DNA. Standard validation specimens included non-probative evidence and specimens from previously adjudicated cases. Testing these specimens with D1S80 and CTT corroborated results obtained with RFLP, HLA DQA1 and Polymarker.

PCR amplifications and gel electrophoresis using the D1S80 kit (Perkin Elmer) and the CTT kit (Promega) were as described by the manufacturer. Amplified alleles were detected by silver staining (Promega).


RESTORING ANCIENT GENOTYPES: THE QUEST FOR ANCIENT DNA
Scott R. Woodward, PhD., Department of Microbiology, Brigham Young University

When we listen to a fugue written by Bach, we do not hear the actual performance by Bach, but a modem reconstruction of the sounds as written long ago on paper. In an attempt to get a more accurate representation of the sound imagined and heard by Bach, we sometimes try to play the music on an instrument of the period, but this is still just a reconstruction. If, however, we could, by some currently unavailable method, analyze the position, direction and speed of all the molecules presently in the atmosphere and deduce where they were just moments ago. Then just moments before that and so forth all the way back to the 1700's we could recover the actual sounds of Bach himself playing one of his fugues.

Likewise, echoes of the genotypes of ancient peoples are contained in the genotypes of people living today. Unfortunately, these genotypes are so dilute that it is currently impossible to reconstruct the original genotypes present in ancient populations by analysis of living populations. In the case of an extinct organism such as the mammoth, the ancient echo has completely died out. To recover any information would require a tremendous amount of extrapolation from any extant sources. We could investigate a related organism such as the elephant and look at a similar signal, but because elephants and mammoths diverged long before the mammoth became extinct this is in essence looking for an echo from a source before the actual sound was produced. Fortuitously, echoes of these ancient genotypes are still ringing out of the ground and we have been able to reconstruct a few bars of the original score by the recovery and sequencing of ancient DNA. Can we verify the strange relationships often depicted in ancient Egyptian families? Can we reconstruct ancient fragmented texts that were written on animal skins?

This lecture will play a few lines from the genotypes of commoners and royalty of ancient Egypt (including the 18th Dynasty, King Tut's family), a mysterious interloper in South America, the swan song of the mighty mammoth, and parchment fragments from the Dead Sea Scrolls. Although the recovery of the actual sounds of a grazing flock of Hadrosaurs is still a Spielberg fantasy, the echoes may be growing stronger.


NEW HIGH RESOLUTION POLYMER FOR ABI PRISM™ 310 GENETIC ANALYZER
Katherine Lazaruk, PE Applied Biosystems

The ABI PRISM™ 310 Genetic Analyzer is a single channel capillary electrophoresis instrument designed for labs requiring lower throughput capabilities than the DNA sequencers. This instrument will be very useful for forensic laboratories that focus on casework and want to start utilizing fluorescent STR technology. Although only one sample can be analyzed at a time, the size standard and amplified sample are automatically injected from a 48- (soon 96-) well autosampler. Fresh capillary polymer is injected automatically prior to each sample so the need for gel preparation is completely eliminated. A new and unique gel polymer has been formulated for microsatellite analysis to give the resolution and precision needed by the forensic community.

The new high resolution (4%) polymer has been developed with the following specifications for microsatellites: run time approximately equal to 30 minutes; between-run precision standard deviation of <0.15 bp; 1 bp resolution of fragments up to 250 bp; 2 bp resolution of fragments from >250 bp to 350 bp; 100 runs per capillary. All of these specifications have been met with the new protocols.

Precision experiments were performed using the AmpFlSTR Blue kit (STR loci D3S1358, vWA and FGA). A population of unrelated Caucasians was evaluated for run-to-run precision and allelic ladder was used to determine the resolution of the 2 bp variants at the FGA locus. The standard deviations of representative alleles at each locus were =0.14bp. This means that any potential 1 bp variants at any of the loci will be detected with this system. FGA alleles that differ by 2 bp (sizing at approx. 250 bp) were resolved to baseline with this polymer.


AMPLIFIED RESTRICTION FRAGMENT POLYMORPHISM (AFLP) ANALYSIS OF HUMAN GENOMIC DNA
Kevin Andera, Jenny Liu, and G.F. Sensabaugh, School of Public Health, University of California, Berkeley

Amplified Restriction Fragment Polymorphism (AFLP) analysis is a new technique developed for the identification of strain variation in plant and animals (Vos et al., Nuc. Acids Res., 23:4407-4414, 1995). AFLP employs PCR to amplify a selected set of restriction fragments prepared from total genomic DNA; the result is a complex "fingerprint" type pattern which characterizes the restriction fragment set. By using different combinations of selective primers, distinct and non-overlapping restriction fragment sets can be visualized. Considerable variation has been observed between cultivars within plant species.

We have tested a commercial AFLP kit (Life Technologies) for potential usefulness in human differentiation. Genomic DNA samples from 4 individuals were restricted with EcoR1 (E) and Mse1 (M). Restriction fragments were ligated to E and M linkers and E-M fragments were amplified using primers specific for the E and M linker sequences. Twelve sets of E-M fragments were selectively amplified using combinations of E and M primers with different 3 base additions at their 3' ends. Fragments were separated on non-denaturing 5% acrylamide gels, stained with Sybr Green, and visualized on a FluorImager (Molecular Dynamics). Very little variation was observed between individuals with any of the selective primer combinations tested, suggesting AFLP typing has limited usefulness for human differentiation.


CSF1PO POLYMORPHISM: EVALUATION OF NEW PRIMERS AND ALLELE FREQUENCIES IN THE JAPANESE POPULATION
Kanako Yoshida(1,2), Kazumasa Sekiguchi(2), Kentaro Kasai(2), Hajime Sato(2), Sueshige Seta(2), and George F. Sensabaugh(1)
(1) Forensic Science Group, School of Public Health, University of California, Berkeley
(2) National Research Institute of Police Science, Tokyo, Japan

STR typing is promising for forensic DNA analysis of aged evidential samples containing severely degraded DNA because of the short amplified fragment size. The CSF1PO locus shows an informative polymorphism and is considered a useful STR locus for forensic identification [Hammond et al, Amer. J. Hum. Genet. 55:175-189,1994]. Since degraded DNA is more likely amplified with the primers which give low molecular weight fragments, CSF1PO primers have been redesigned to give shorter PCR fragment. The redesigned primer sequences are:

CSF-3F 5'gtt gct aac cac cct gtg tct c 3' 22mer
CSF-3R 5'ttc ctg tgt cag acc ctg ttc3' 21 mer

PCR products obtained using the new primers are approximately half the size (149-181 bp) of fragments produced using the conventional primer (295-327bp). The identity of the new PCR products was verified by sequence analysis and by comparison to typing results obtained using the conventional primers. CSF1PO typings from 16 year old bloodstains using the new and the original primers showed that the new primers gave typing results on samples that did not amplify with the original primers. This demonstrated that the redesigned CSF1PO primers are more suitable than the original primers for STR typings from stain samples that contained degraded DNA.

Population studies on the CSF1PO polymorphism in Japanese(n=142) have also been conducted. Allele frequencies are as follows: *7-0.014; *8-0; *9-0.039; *10-0.201; *11-0.229; *12-0.387; *13-0.120; *14-0.007; *15-0.004. These allelic frequencies show no deviation from Hardy-Weinberg equilibrium. The discrimination power for CSF1PO in Japanese is 0.893.


DIFFERENTIAL EXTRACTIONS USING A FLUORESCENCE-ACTIVATED CELL SORTER
Matt Piucci(1), Armand Tcheong(1), Don Hesler(2), Gill Reinin(2) and Chris Spenner(2)
(1) CA Dept. of Justice DNA Lab, 626 Bancroft Way, Berkeley, CA
(2) Becton-Dickinson Immunocytometry Systems, 2350 Qume Drive, San Jose, CA

The use of DNA databases to produce suspects in otherwise suspectless sexual assault cases is one of the main goals for forensic DNA analysis. The use of robots to perform automated DNA analysis has proven to be very helpful in achieving the aim of developing these databases. However, there is no associated protocol for performing high-throughput separation of sperm and epithelial cells to produce profiles for searching against felon data bases. One possible solution to this problem may be the use of Fluorescence-Activated Cell Sorters (FACS) to perform these separations. Cell sorters can process up to 7000 events (cells) per minute, and in this application the processing of a typical swab took about ten to fifteen minutes. This suggests that a throughput of twenty five swabs per day is not unreasonable. Simulated sexual assault kits were prepared using buccal swabs that were spiked with diluted semen. The swabs were incubated in PBS @ 4°C for one hour, minced and the liquid was removed from the substrate. These were then stained with Propidium Iodide (a general DNA stain) in a 0.1% Triton 100 solution. The sperm cells were separated from the epithelial cells based on total DNA content (a measure of ploidy or n number) and particle size by a Becton-Dickinson FACS Vantage. The sorted cells were stained with Nuclear Fast Red and Picroindigocarmine for microscopic examination. DNA was extracted from the cells and was typed via PCR at the DQA1 locus. Complete separation was confirmed by microscopic examination of the fractions and by the typing results. Negative controls produced no types.


CASE REPORT: THE IDENTIFICATION OF BROTHERS IN A SINGLE RAPE EVIDENCE SEMEN SAMPLE
Keith Inman and Yeung Kung, California Department of Justice DNA Laboratory

The California Department of Justice is frequently asked to examine sexual assault evidence from multiple victims in which a common assailant is a suspect. One such series was brought to the DNA Laboratory in 1992 and 1993. While the cases were indeed linked to one assailant (the suspect) through RFLP profiling, one victim claimed to have been raped by two men (out of three intruders). The RFLP results did indeed show the presence of semen from two men in a single vaginal swab sample. The number of bands detected in the evidence profile, along with their relative band intensities indicated possible band sharing, which in turn suggested that the two individuals were related. Further investigation revealed that the suspect had a brother and that this brother had previously supplied a sample (as yet unanalyzed) to the DNA felon database per legislative mandate. This sample was analyzed, added to the database and a search performed from the profile generated from the case. The brother (and no other felon) was "hit" by the search. We will provide details of the analysis and search, and discuss strategy for ensuring objective analytical results and protection for contributors to the database.


CALIFORNIA'S CONVICTED FELON DNA DATABANK: AUTOMATED SAMPLE PROCESSING AND RFLP ANALYSIS
Vince Scola, Ellen Clark, Dan Butler, and Lance Gima, DNA Laboratory, California DOJ

California Penal Code 290.2 mandates the generation of a DNA databank for sex offenders and other violent felons. Over 75,000 sample shave been collected and stored; approximately 1200 new samples are received each month. In order to efficiently process such a large number of samples, Hamilton Microlab "robots" were programmed to aliquot blood to 96-well plates and stain cards, extract DNA from liquid blood, quantitate the extracted DNA samples, aliquot specified amounts of DNA for HaeIII restriction digestion, and set aside DNA samples for PCR and STR analysis. This system has the capability of processing over 30,000 samples per year. Two robots are utilized in this process. Using a disposable-tip system, "Robot 1" transfers 200 µl of liquid blood directly from blood tubes to a 96-well plate, and four 75-µl spots of blood to a custom-formatted sheet of Schleicher and Schuell stain card paper. Two sets of 96-well plates containing blood aliquot are generated per run; a run consists of 168 offender samples and 8 QC samples. Fitted with a multiple-probe-head system, "Robot 2" extracts DNA from the blood samples using QIAGEN columns arrayed in a 96-well format (QIAamp 96 Spin Blood Kit). Extracted DNA is quantitated using the fluorophore "Picogreen" (Molecular Probes, Inc.) and a fluorometric plate reader which exports the DNA concentration data to a work file. Robot 2 uses these data to aliquot a specified amount (usually 400 ng) of DNA from each sample to a 96-well plate for a one-hour HaeIII digestion. Following digestion, the restricted samples are again processed on the QIAGEN plate columns to purify the HaeIII-digested DNA. The DNA samples are then evaporated to dryness and stored at -20°G until electrophoresed on agarose gels. Sample processing on Robot 2 can be performed in less than 7 hours; average DNA yields from 200 µl of liquid blood are approximately 6 µg.


RELATIVE DETECTION AND SIZING OF STRS ON FLUORESCENT SCANNERS
Sherrie M. Post, Kathleen McCarthy, Steven B. Lee, California Department of Justice DNA Laboratory

The goals of this study were to compare sensitivity, sizing precision, and any differences in results for the short tandem repeat systems CTTv, and FFFL using the Hitachi FMBIO 100, and the Molecular Dynamics FluorImager SI fluorescent scanners. DNA was extracted from 10 bloodstains using a standard organic extraction method (CA DOJ DNA Protocol). The DNA from these samples, and DNA from mixtures (epithelial cell, semen fractions) were amplified at the CSF1PO, TPOX, TH01, and vWA (CTTv multiplex), and F13A01, FESFPS, F13B, and LPL (FFFL multiplex) loci (Promega Gene Print Fluorescent STR Systems). The signal balance of these STR multiplex systems, as well as the dye detection capabilities of the two fluorescent scanners were also examined.

A main distinction between the two fluorscanners is that they utilize lasers which excite at different wavelengths, creating variances in the fluorescent dye detecting capabilities. The shorter wavelength dyes, fluorescein and FAM can be seen on both instruments. In addition the Hitachi FMBIO 100 can also visualize TAMRA and ROX, as well as perform dual color analysis.

There was a high level of precision in sizing unknowns using across lane allelic ladder standards, between the two instruments, with an average difference of 0.15 base pairs.

Signal balance at some loci, TH01 in particular, was especially problematic. This can be explained in part by sub-optimal amplification conditions which is supported by annealing temperature experiments. Allele balance, and stutter bands also gave rise to some interpretation difficulties.


AN AUTOMATED, HIGH THROUGHPUT APPROACH FOR ROBUST SIZING AND ACCURATE GENOTYPING OF STR'S
Jeanette Wallin, Kathy Lazaruk, Nicky Fildes, P. Sean Walsh, PE Applied Biosystems

As the forensic community begins to adopt STR's (short tandem repeat loci) for database and casework applications, the need for a high throughput while obtaining robust sizing and accurate genotyping is emphasized. Currently, many forensic laboratories perform silver staining as the mode of detection for STR's. Because silver staining permits the detection of only one "color", every few lanes on each gel must be dedicated to an allelic ladder. Genotypes are assigned by extrapolation to the nearest lanes containing an allelic ladder. By taking advantage of fluorophore primer-labeling technology, it is possible to run a size standard in one color and unknowns in other colors all within a single lane. This multicolor detection capability provides for higher throughput by eliminating the need for multiple lanes dedicated to an allelic ladder. Furthermore, in-lane sizing provides very precise sizing of unknowns because the lane-to-lane migration variations within a gel are normalized.

When using in-lane sizing standards, one needs only to dedicate one or two lanes in each gel for an allelic ladder to obtain extremely accurate genotyping results. Our approach involves sizing both an allelic ladder and unknown samples against the in-lane sizing standard to obtain fragment sizings in base pairs (bp). Genotypes are assigned by using the allelic ladder as a calibration tool, where the alleles in the allelic ladder serve as the centers of floating bins. A Genotyper™ template will automatically assign a genotype to those samples which fall within +/-0.5 bp of the size determined for an allele in the allelic ladder. Bands which do fall within the window will not receive a genotype. Instead, those bands will be flagged for easy identification, prompting the analyst to rerun the "off-ladder" sample to confirm a true variant from a gel anomaly. Using genotypes, results can easily be compared from gel to gel and across all detection platforms. The Genotyper™ software will be as automated as possible in order to maximize throughput while minimizing hands-on time.

This presentation will demonstrate the ability for such a system to provide robust sizing and accurate genotyping results in a high throughput automated manner. We will explore within- and between-gel sizing precision on the ABI PRISM™ 377 DNA Sequencer using AmpFlSTR Blue PCR products of population database samples and allelic ladder, together spanning the size range of 110 bp to 268 bp. AmpFlSTR Blue is a new triplex PCR kit containing the loci D3S1358, vWA, and FGA. Specifically, we have tested placement and number of lanes necessary for dedication to an allelic ladder. We also examined the distribution of sizing data from population samples relative to allelic ladder sizings. Additionally, we examined sizing differences induced by small changes in acrylamide percentage.


VALIDATION STUDIES ON THE USE OF THE AMPLITYPE PM PCR TYPING SYSTEM
Brian Burritt and Eva Steinberger, California Department of Justice, DNA Laboratory

Amplitype PM is a multiplex PCR system that amplifies the loci LDLR, GYPA, HBGG, Gc, and DQA1. The Department of Justice DNA laboratory has recently completed validation for the use of PM in casework. Dried blood, semen and saliva from two individuals as well as bone, brain, kidney, liver and muscle from a cadaver showed consistent typing results. Reproducible PM types were obtained from liquid and dried samples of blood, semen and saliva from two different individuals. The minimum amount of sample required to obtain a typable result was determined to be 250 picograms (pg) of template DNA. Allelic dropout was observed at 100 pg. The optimum template concentration was shown to be between 1-3 nanograms (ng) of template DNA. To evaluate the system's ability to display minor components in mixed samples, mixtures from 1:50 to 50:1 at DNA levels of 1 ng, 2 ng and 5 ng were analyzed. At the 1 ng level, the minor component was reliably detected in a 1:20 mixture. At the 2 ng or 5 ng level, the minor component could be detected in a 1:50 ratio. Dried blood and semen stains, exposed to sunlight, shade, 4°C, room temperature, 37°C, 65°C for up to 2 months were used to investigate the influence of environmental conditions on PM typing of a sample. While some degradation was seen in the yield gels for some samples, only the semen stains exposed to sun light for three weeks or more did not result in reliable typing results. A matrix study with various kinds of substrates, including blue and black denim, concluded that all samples were typable. Comparison of the "old" and the "new" S-dot revealed only minor differences in dot intensity.

Our validation has shown that the PM system is robust and reliable. The addition of PM to the currently used DQalpha and D1S80 systems has dramatically increased the average amount of discrimination, from one in 300 to approximately one in 130,000.


DNA EXTRACTION FROM BLOODSTAINS USING SALTING OUT AND QIAGEN COLUMNS
Keith Garrison, Vince Scola, Steve Lee, California Department of Justice DNA Laboratory

DNA data banking requires rapid, efficient extraction of DNA from a large number of samples. The automation of extraction using a robot assisted Qiagen column has been developed and implemented for liquid blood (Butler, 1995 and Scola, 1996), however it is not effective on bloodstains. The quantity and quality of DNA extracted from bloodstains using this method are unacceptable for RFLP analysis. The Cal DNA convicted felon backlog contains more than 12,000 samples available only as bloodstains. The conventional organic method for DNA extraction from bloodstains is time-consuming, uses toxic chemicals, and is not easily adaptable to automation. The goal of this study is to develop a rapid, efficient, non-organic method of DNA extraction from bloodstains that can be adapted to automation. We investigated combinations of previously successful extraction methods. The highest yield and quality of DNA resulted from an extraction method that is a combination of a modified salting out method (modified from Miller, 1988, by addition of 250% more salt solution to the sample) and the Qiagen column procedure.

Following a ten minute 85°C heat soak and a 15 minute Proteinase K digestion at 56°C, 500 microliters of a saturated solution of sodium chloride is added. This salt solution reduces the solubility of cellular debris, including proteins, causing them to precipitate out with centrifugation. There is no phase separation required, as with a phenol/chloroform extraction. Total sample processing time is reduced to one hour, and the production of hazardous waste is eliminated. Twenty eight samples extracted by this method show average yields of 758 ng of DNA from a 50 microliter stain (range=400-1600 ng), and have been successfully digested with Hae III. Extracted DNA is undergoing further evaluation for quality by RFLP analysis and amplification of STRs.