2005– ASQDE MEETING ABSTRACTS

The views expressed by the authors or presenters in material presented at the American Society of Questioned Document Examiners Annual General Meeting and/or on the CD distributed at the meeting, are not necessarily the views of, nor are they endorsed by, the American Society of Questioned Document Examiners, Inc.

The abstracts for the 2005 ASQDE AGM are contained on individual pages within this document. The ASQDE has provided this abstract configuration to facilitate the output of individual abstracts (one per page).

• Towards a New Evidence Interpretation System Ulla-Britt Aberg

Abstract: This paper will give the background to why the National Laboratory of Forensic Science in Sweden decided to try to change its model of evidence interpretation system including how to express the findings in each examination in a totally new scale of conclusions. The paper will deal with the problems that met and still meet the group working with this project. The project started in September 2001 and is planned to continue for a couple of years. The overall aim of the project was to give the staff at the National Laboratory of Forensic Science the qualifications to appear in court as experts and to interpret the value of the findings of their examinations in an objective and impartial way. First of all a steering group consisting of three persons with very good knowledge of evidence interpretation was formed. The steering group in its turn elected a project leader who selected two persons from each operative unit at the laboratory just to let every discipline has the possibility to give opinions on the proposals and also to be able to give the project representatives a chance to get feed-back from the examiners in their own groups. The project group consisted of eight persons, including the project leader. The first objective for the project was to try to create a new scale of conclusions that could be used by the entire laboratory. The paper will describe the largest problems during the work when trying to harmonize the different scales of conclusions that existed at the lab. This phase of the project is complete and the new scale of conclusions is now used by all the different groups at the laboratory.

• Using TLC and GC-MS to Determine Whether Came From the Same Manufacturing Batch Valery Aginsky, Ph.D.

Abstract: A combination of TLC (analyzes colored components) and GC-MS (analyzes ink non-colored components) demonstrates high discriminating power with regard to inks that cannot be distinguished by non-destructive techniques. Case examples are considered in which TLC and GC-MS allowed one to determine that the inks on questioned documents came from the same manufacturing batch.

• On Limited Populations Of Writers And Writing Chris Anderson and Julian Leslie

Abstract: The effect of the number of potential writers used in identifying the writer of a questioned document is the subject of this paper. We investigate the impact of the number of potential writers on the chance of identifying the writer in the case where there are only a limited number of characters in the questioned document. We show that the chance of identifying the writer drops as the number of potential writers increases and that this becomes quite marked for certain values of parameters used in the models. Specifically, the number of key distinguishing features and two error probabilities associated with correctly or incorrectly identifying a character as having been written by the writer of the questioned document.

• The Importance Of Line Width Measurements In Discriminating Between Types In Forensic Handwriting Investigations - A Study Chris Anderson, Sibba Gudlaugsdottir, Julian Leslie

Abstract: In a specimen of pencil writing, is it possible to identify whether the writing instrument was a mechanical pencil or a wood pencil? For a mechanical pencil, is it possible to determine the diameter and/or softness of the lead? A number of writers provided samples using each pencil type, together with four levels of softness. The maximum width for a character was the measurement of interest. Various statistical analyses are undertaken to investigate this matter. We find that within the one softness, it was possible to distinguish diameters of mechanical pencil leads but without the softness information the picture was less clear. Distinguishing between mechanical and wood depended on the softness and the diameter of lead. For soft leads, small diameter leads in mechanical pencils were easily distinguished from wood pencils whereas the larger diameters leads were not distinguishable from wood pencils.

• Color Separation In Forensic Image Processing Charles E.H. Berger, Jan A. de Koeijer, Wendy Glas, Henk T. Madhuizen

Abstract: In this study we implement color deconvolution processing of digital images, and apply it to color separation problems in document examination. Subtle color differences (sometimes invisible to the naked eye) are found to be sufficient for separating or removing a color from a forensic image. This is demonstrated for several cases where color differences were shown to exist, or where colors were removed from the foreground or background. The software is available for free in the form of an Adobe® Photoshop®-compatible plug-in.

• Probable time of death using document examination Sharon Birchall

Abstract: The Snowtown "Bodies in the Barrels" serial killings trial has become the biggest and highest profile case in South Australia's legal history. It is also the largest signature and handwriting case examined in the career of this presenter. The Snowtown Homicides not only involved the disappearance and murder of the victims but also the theft of their social security benefits. Signatures and handwritings associated with eleven victims and other miscellaneous were submitted for document examination. The signatures and handwritings were grouped by the use of calendar dates. A pattern emerged in the signatures and handwritings coinciding with the victims’ disappearance and, therefore, probable time of death. The results of the signature and handwriting comparisons were presented to the court as a PowerPoint presentation. This format made use of the electronic courtroom that was specifically designed for the Snowtown trial.

• Handwriting Graphometer Analysis in Idioptic Parkinson’s Disease and Parkinsonism Julio Bradley, Maria I. Nouzielles, Marcelo Merello

Abstract: To assess handwriting characteristics of Idiopathic Parkinson’s Disease (IPD) and compare them with those found in Secondary Parkinsonism (SP), as well as to analyze the evolution of handwriting impairment in terms of the disease stage and search for differences in writing when IPD patients were evaluated in ON vs. OFF state. Forty-six handwriting samples were evaluated for the presence or absence of 15 handwriting characteristics by a previously determined handwriting expert. Thirty-six patients with IPD and 10 with SP were instructed to copy a sentence and to draw lines, circles, and boxes on a blank sheet. In IPD, patients with high UPDRS and high H & Y score, the incidence of micrographia was significantly higher (p=0.03 for both, Mann-Whitney). IPD fluctuating patients showed in addition to greater frequency of micrographia changes in writing direction (p=0.02 for both, Fisher). No differences were found when patients were evaluated in ON vs. OFF state in any of the studied characteristics. Base line not fitting the row and exceeding it were more frequent in SP than in IPD (p=0.02 and p=0.01 respectively, Fisher test) as the only parameter able to differentiate between IPD and SP. Our results confirm that handwriting characteristics depend on disease severity, and resulted from an alteration of dopaminergic as well as non-dopaminergic pathways. The relationship between handwriting base line and row line, which is easily evaluated during the medical visit, is the only parameter that could suggest parkinsonian symptoms of secondary origin.

• Dating of Ballpoint Inks by Thermal Desorption and Gas Chromatography-Mass Spectrometry Juergen H. Buegler, Anton Dallmayer, Hans Buchner

Abstract: The dating of ink is a challenging subject in forensic document examination. Two concepts are available for ink dating: indirect age determination by comparing the ink under investigation with samples of a reference collection and determining the date of its market introduction, and direct age determination by measuring age-dependent parameters of the ink. Several analytical procedures have been reported for the analysis of age dependent parameters of ink on paper. These procedures require sample preparation and are not independent of sample size in some cases. In previous studies we have demonstrated the applicability of thermal desorption to indirect age determination. The method comprises thermal desorption of volatile ink components and subsequent analysis by gas chromatography-mass spectrometry. In this paper, we report on the direct determination of ballpoint ink age by thermal desorption. The presented method is easily applicable, independent of the sample amount and avoids any contamination possibly caused by sample preparation. The method uses a two-stage thermal desorption technique of one and the same sample of ballpoint ink on paper. The obtained ratio of the amount of volatile ink components found at low desorption temperature versus the amount found at high temperatue is a measure for the age of the analyzed pen stroke. The method was optimized regarding the temperatures used for thermal desorption as well as the desorption times. Careful control of instrument performance turned out to be a key factor in the quantitative analysis of volatile ink components. Samples of 60 different ballpoint inks on paper, each between 1 day and 2 years old, were measured. It was found that the aging behavior of the different inks varies greatly, with some inks aging completely within a few days while the detectable aging period of other inks lasts for up to one year. A general aging curve was developed in order to assess the age of unknown ballpoint inks with the presented method. Furthermore, we performed kinetic studies using thermal desorption. These studies showed that the desorption rate of volatile components from the ink not only depends on sample age but also on the type of resin in the ink.

• The Significance of Word Spacing in Handwriting Ted M. Burkes

Abstract: A pilot study focusing on the significance of word spacing of individual writers was conducted in 2000 by the author. A continuation and update of this study is presented here. Leading authorities in the field of Questioned Documents suggest that word spacing is a valuable identifying feature of handwriting, yet studies on this subject are lacking. Treatment of this subject in the current literature renders the importance of word spacing as little more than an anecdotal account of a potentially useful parameter. The goal of this study is to determine the role of word spacing as an identifying characteristic. This will be accomplished by studying intra-individual and inter-individual variation in the use of space between written words. Handwriting samples were obtained by having participants complete five exemplars of a demand note. The participants were not provided with a , but instead were allowed to fill out the form with a writing instrument of their choice from their own familiar environment. The rationale for this approach was that it allowed for the genuine condition by which documents are created by each participant in their everyday real world conditions. Inter-word spacing was measured from areas within each participant’s exemplars. These measurements were made by using a beta version of Dr. Sargur Srihari’s Cedar-Fox software. The significance of the intra- and inter-individual measurements is also calculated by this software. The results of this study will demonstrate that inter-word spacing, when considered by itself, shows low variation within a given participant’s writing, but can exhibit significant variation between different participants writing. Specific findings of this project will be presented.

• Left-hand-Writing vs. Right-hand-Writing Marianne Conrad, Hugo Hundrup

Abstract: Over the past years the number of people writing with their left hand increased and as a result the document examiner occasionally is confronted with the question whether a document is written with the left hand or with the right hand. In this study the authors examined the writing of left-handers and right-handers in order to detect if there are certain signs that indicate the handedness.

• Determining The Sequence Of – Pencil Intersections Using Gelatine Lifting, A Case Report Jan de Koeijer, Charles Berger, Wendy Glas

Abstract: Recently the Netherlands Forensic Institute received a case where it was necessary to find out if the pencil correction marks on an exam were written before or after the obliteration of an answer with a ballpoint pen. Using optical microscopy at magnifications of 40-1000 times we could not determine the sequence of the obliteration and the correction mark. To solve the problem black gelatine lifters were used to lift the free graphite particles from the paper surface. Research showed that ballpoint ink paste, when applied over a pencil line had enough adhesive strength to prevent the graphite particles from being lifted by the gelatine lifter. The graphite lines on the gelatine slab therefore showed distinct discontinuities at these intersections. Pencil marks drawn over fresh ballpoint lines also tended to show gaps on the gelatine lifter, although these results were much less consistent than those from the opposing sequence. As the pencil line in question in this case showed no discontinuities on the gelatine lift it was interpreted as evidence to support the proposition that the pencil line was drawn after the ballpoint pen obliteration. A Bayesian approach to the problem presented in this case will be presented.

• 1. What’s The Limit Between Technicalities And Legal Determinations? 2. Is It Wise To Give Categorical Conclusions In Controversial Topics, Without Having Qualified Knowledge And/Or Updated? Would It Be Ethical? Julia Elena de la Pena

Abstract: The incredible and wide new technical theories applied to new varied elements and handwritten systems with periodical and ever growing perfection, all of these in relation to the legal determinations, and the globalization in communications, do affect us just the same all over the world. I think we should stop and seriously think on the matter. What would be the dangers or the consequences of the legal definitions? Do all the experts, all over the world, always know everything? Based on five real cases, I shall try to resume the results of the seriousness of these analysis that have strayed the final judgment in legal matters. The case number is not a physical matter, but human people suffering for a verdict.

• Capillary Electrophoresis of Ballpoint Pen Inks James M. Egan, Ph.D., Jason D. Brewer, Ph.D.

Abstract: Two capillary electrophoretic (CE) methods have been designed to differentiate and identify organic and inorganic dyes present in black and blue ballpoint inks. On-capillary detection using a photo-diode array to acquire absorbance data (190-600nm) provides lambda max and characteristic UV-Vis spectra for analytes. Simultaneous spectral data collection avoids the necessity for densitometry measurements. Mobility values (ueff) calculated from a compound's migration time replace hand-measured thin-layer chromatography (TLC) Rf values. The sample electropherograms are reproducible with 2ug/mL concentration limits of detection, which is superior when compared to TLC techniques. Specific chemical identification can be determined based on spectral and mobility value match with reference dye standards. CE offers additional advantages over TLC including lower reagent consumption, smaller sample requirements, lower detection limits, quantitative analysis, method automation, direct reference comparison, electronic data storage, and searchable spectral libraries. Other complimentary techniques (TLC, positive- and negative-mode electrospray-mass spectrometry) can be performed because little sample (20nL) is consumed during CE analysis. The same CE buffers formulated for ballpoint pen inks have been applied to colored inkjet printer formulations, food dyes, inks, and textile dyes.

• The Design and Development of a University Graduate Course in Forensic Handwriting Examination Charles L. Eggleston

Abstract: The author was afforded an opportunity to create and teach a university graduate course in forensic handwriting examination. This paper reports on the design and developmental aspects of the course. It explains the rationale of the course’s design, structure and content. It details the course’s topic areas, objectives and readings. It describes the research undertaken to develop the course. The course presents handwriting as forensic evidence from theoretical, philosophical and practical perspectives. It covers the production of normal and false handwriting, variables in handwriting production, standards of comparison, identification theories, examination methodologies, expressions of conclusions, characterization and validation of examiner skills, legal admissibility of handwriting expertise, and challenges to professional practice. The course, titled Forensic Handwriting Examination: Theory and Practice, has been successfully delivered online for two academic calendars at Oklahoma State University, Tulsa.

• ImageXpert Applications Updated - 2005 Colin Estabrooks, Cathie Gilmour, Helen Park, Robert Vallières, Christine Warias

Abstract: Several new applications of the measurement tools of the ImageXpert System were explored for use in forensic document examination. A study was conducted to determine which measurements, utilizing the Imagexpert System, could be useful in the authentication of banknotes. As in our previous project on passports (ASQDE Journal Vol.7, No.1, June 2004), a variety of printing processes and security features from a number of genuine and counterfeit banknotes were critically measured and analyzed with the aid of ImageXpert's quality assessment tools. The data from these measurements was compared not only to discriminate between genuine and high-quality counterfeit banknotes, but also to develop potential links amongst the counterfeits. Preliminary testing of ImageXpert's new "spacing" analysing tool was also conducted. Samples from laser and inket printers were assessed with respect to variances in spacing between lines of text.

• Masters in Counterfeiting Bridgette Trela Frost

Abstract: Forensic Science graduate students at The George Washington University, Washington, D.C. took part in a challenge put forth by the Counterfeit Document Study Group of the Document Security Alliance to counterfeit an identity card prototype: the Tricycle License. As part of a questioned documents graduate course, the students were given a brief overview of the security features of the tricycle license and provided samples of two types of tricycle licenses. The products of their efforts and a descriptive paper on their counterfeits were provided to the professor. This challenge is part of an overall DSA initiative to test the “counterfeitibility” of a variety of security features. However, it also provided the author with a unique opportunity to look into the world of amateur counterfeiting without facing a criminal at the other end (hopefully). An assessment of the student’s methodology and the success of their counterfeits will be discussed and presented.

• Exploration of Photocopier Distortion through Interactive Animations Robert Gervais

Abstract: This paper explores the nature of distortions introduced by the photocopying process. Two types of were used to execute both a fast and a slow sample of handwriting. These were then reproduced on photocopiers of various makes and models, both analog and digital. Successive copies were created from each sample, extending twenty generations from the original. The copying process was repeated for each series using the photocopier’s light, medium and dark settings. Overall, the combination of writing instrument and speed together with photocopier model and darkness settings yielded eighty-four permutations. The resulting 1680 images (84 permutations x 20 generations) were then integrated into an interactive graphic interface allowing the user to select a set of parameters and explore the distortion, moving back and forth through the twenty generations at will. Trends in the observed distortions were investigated, both at a localized level and across the page as a whole. Differences were noted in the type and extent of distortion seen in analog and digital machines.

• Handwriting Examination: What Can We Really Measure? Dr. Christian Grafl

Abstract: This paper illustrates the possibilities of measurements in handwriting examination. On the one hand, computer programs allow for measuring writing characteristics with great accuracy not known up to now. On the other hand, we have to acknowledge the principle limits of measuring writing characteristics. Experts not only have to assess the value of numbers measured automatically by computers but they also have to appraise writing characteristics not accessible to exact measurements. A practical case shall illustrate the alleged benefits and limits of measurements in handwriting examination.

• Historical Error In The Teaching Of Handwriting In Mexico Carmen L. Hernandez Carmona

Abstract: Handwriting, as a mean of written communication among human beings, has had a gradual evolution through the time in order to make it clearer, faster, simpler and more practical. However this purpose is not always achieved properly as in the case of Mexico, where the consequences of a drastic decision taken during the government years of 1970-1976 when the federal law of education was modified (1973), proved to have serious consequences that can still be seen nowadays in its population. In the continuous exam of signatures drawn with illegible strokes (paraphs) structural geometry in the signatures of people according to whether they have learnt to write before or after the 1973 education reform show substantial changes in their characteristics.

• Databased Toner Classification Juergen Holzapfel, Tanja Rottes

Abstract: Since toner based documents are frequently used, the FDE is often confronted with forgery of all forms of output devices which use a dry-toner printing process. This presentation shows possibilities of toner classification by optical, physical and chemical methods and a databased evaluation.

• If It Looks Like a Duck…It Might Not Be - The Use of Supplementary Testing in Forensic Document Examinations Gerry LaPorte, Malinda Layman

Abstract: With intensifying scrutiny in the forensic sciences, there is an increasing need in questioned document examinations to provide conclusions that are more assuring and definitive. Today, there are only two forensic specialties that can be used to positively identify a suspect: fingerprints and handwriting (DNA results are based on statistical population values and therefore, a probability of identification is assigned to a given suspect). Although the identification of a writer based on comparisons to known handwriting is one very important aspect of questioned documents, there are numerous other types of examinations that may be conducted to provide even more information in a particular case. In cases where a handwriting exam is not requested, does not provide a definitive conclusion, and/or is not a requisite examination to the request (e.g., analyzing line crossings, determining if a document was produced on a purported date), there are myriad examinations that may be conducted to assist investigators. Questioned documents are often the focus of investigative inquiries and may include letters, envelopes, packages, calendars, diaries, identification cards, financial documents, contracts, wills, and business records. In many instances, the information gleaned from a Forensic Document Examiner's (FDE) findings can be very incriminating. If results can be corroborated with additional testing, then more significance may be placed on the conclusions by the court. This presentation will focus on laboratory techniques that can be used to supplement findings and add further value to the conclusions sometimes made by the FDE. Some of the areas that will be discussed include supplementary examinations on: 1. documents bearing trash marks; 2. documents printed using the primary subtractive colors (CMY); 3. line crossings using a newly proposed Video Spectral Comparator (VSC) technique; 4. thin layer chromatography plates using an alternate light source and; 5. paper and envelopes for hardware impressions from the manufacturing process. Many of the supplemental examinations discussed can be performed with basic laboratory equipment such as an electrostatic detection device and a VSC. As a result, the FDE will be able to efficiently and effectively enhance the gamut of testing, while ultimately providing a stronger basis for conclusions.

• Debating Qualified Conclusions Malinda Jo Layman

Abstract: Historically, it seems there has always been some sort of debatable issue surrounding the qualified findings expressed by Forensic Document Examiners (FDEs). Should qualified opinions or conclusions be permitted in reports? Should they be permitted in testimony? Should they be used with both positive and negative findings? What levels of certainty (real or imagined) are associated with qualified opinions or conclusions? With recent court challenges, FDEs have been forced to be more articulate in their findings and to utilize methods that have been more thoroughly validated. Additionally, with the development of ASTM Standard E 1658 (1996), there has been some debate among FDEs concerning the amount and nature of the terminology used to describe qualified findings. Testing by Dr. Moshe Kam is being conducted to further explore this aspect of qualified terminology. A review of the available literature dealing with qualified opinions and qualified conclusions throughout the past half of a century has proven to be beneficial in better understanding the current debates that abound among FDEs today. Further, this review has enabled this FDE to reach a few “conclusions” of her own.

• Study on the Levels of Difficulty in the Simulation of Individual Characteristics in a Signature Gek-Kwee Lee, Chiew-Yung Yang, Bei-Sing Yap, Lee-Tiang Lee, Sock-Kim Tan, Koon-Puay Tan

Abstract: This study looks at the levels of difficulty in simulating a signature with respect to twelve selected individual characteristics in a model signature. The objective is to find out which of these features in the signature are relatively easy or difficult to simulate, and to compare the results with those expected based on the team’s knowledge and experience in forensic handwriting and signature examination. Sixty-two volunteers took part in the study. The results from the analysis show that the experimental data is generally in agreement with the expected relative levels of difficulty of the characteristics studied. This provides good evidence to show that the principles and knowledge applied in forensic handwriting and signature examination are sound and reliable, and that characteristics in a person’s signature are individualized.

• Two Unusual Signature Problems (or "Ah, the tangled webs we weave.") Janet F. Masson

Abstract: Signature problems comprise the lion's share of most document examination caseloads. Generally such cases can be solved in a routine manner with a careful examination, using standard procedures. Then there are the odd, troubling, and challenging cases. This presentation will illustrate two such problems. In the first case, the questioned signature was eventually acknowledged to be non-genuine. However, an additional question raised was whether the disputed signature was a simulation of an abnormal but genuine signature. In the second case, the writer began to distort his genuine signatures. He purportedly advised the notary that he was starting to use an unusual genuine signature so that he could later deny signatures involved in ongoing litigation. However, these changes eventually came back to bite him in an unrelated legal proceeding.

• Laser Printers Differentiation By Image Analysis Williams Mazzella, Fabienne Chopard

Abstract: The present study tries to verify the paper presented by David Wolin during the 2002 EDEWG meeting hold in Bratislava. Image analysis provides to objectively and quantitatively examine a laser printed document. The results allow to identify some problems inherent to the image analysis system used such light intensity, algorithms, etc as well the intravariabilty and intravariability on the printed documents. Statistical data may confirm some printer’s defects, and allow an objective measurement of what the naked eye can already do. De facto image analysis remains quite difficult to implement in routine work. Problems arising the comparison material will also be highlighted.

• Federal Rules of Criminal Procedure Rule 16 – The Document Examiner’s First Line of Defense for Admissibility Carl R. McClary

Abstract: In the Federal Rules of Criminal Procedure, Rule 16 for Expert Witnesses requires the government to give the defendant a written summary of any test that the government intends to use under Rules 702, 703 or 705 or the Federal Rules of Evidence during its case. This is also applicable to the defense. The summary of tests has become a critical step in the development of a criminal case and can be a key element in the decision on the necessity for a Daubert hearing. The presentation will describe Rule 16 and the author will give suggestions on content and formatting of a comprehensive and persuasive Rule 16 document.

• A Living Will. Examination And Court Testimony Gregory McNally

Abstract: Recently, there was a case in Florida surrounding the removal of a feeding tube from a brain damaged woman, Terri Schiavo. I had a matter like this last year. In this instance, the daughter of a brain damaged woman, kept alive with a feeding tube, found a letter written by her mother six years earlier. It revealed she did not want to be kept alive artificially. There was apprehension on the part of the court to allow the removal of the feeding tube because the "living will" was offered many years after the woman was declared brain damaged. This paper will discuss the examinatin of the handwriting and the presentation of the evidence in court.

• Using Microsoft Access in Extended Writing Cases Lamar Miller

Abstract: Forensic document examiners are frequently called upon to make handwriting comparisons in cases involving known specimens submitted in a different wording than the questioned material. This presentation illustrates the use of a popular database program to assist in these examinations by preparing side-by-side comparison charts of similar words or letter combinations. Microsoft Access is used to import scanned images of questioned and known samples. The images are stored in a table and sorted by the software. Similar words or letter combinations are grouped and printed for use in the comparison process or for use in court presentations. Microsoft Access is inexpensive to purchase and relatively easy to tailor to the needs of the forensic document examiner. The program also has the ability to prepare exhibit list, print exhibit labels and track cases and exhibits.

• Thinking Outside the Box Linton Mohammed, Gerald Richards

Abstract: Document Examiners are taught that they should not have any background information on a case since this may bias the examination. This case study describes an example of a document which was allegedly altered for ulterior motives. Proof that the document was altered was straightforward. However, the way in which it was altered could not be explained by the examinations conducted. Background information obtained later resolved the matter by showing how and why the document was altered. The motive was not necessarily ulterior and could be explained by the normal business practice of the defendant.

• State of New York v Sante Kimes and Kenneth Kimes - The Role of Forensic Document Examination in a Murder Case John Paul Osborn

Abstract: In the early part of 2000 a precedent setting trial was concluded in New York. Irene Silverman, a elderly and wealthy Manhattan socialite, had been murdered. The accused were an alleged mother and son grifter team, Sante and Kenneth Kimes. The Kimes case resulted in a first ever conviction in the state of New York for first degree murder wherein the victim's body was never recovered, no murder weapon was found and no forensic medical evidence was presented. Forensic document examination played a pivotal role in the conviction. Among the evidence examined were journals maintained by the defendents which, in part, detailed their involvement with the crime. However this case involved examination work utilizing a broad spectrum of techniques and focusing on many different kinds of evidence. In addition, testimony was required to establish the legitimacy of forensic document examination and thwart an attack on the field by the defense utlizing the testimony of Prof. Mark Denbeaux. The author, who performed these examinations and was the last of 129 prosecution witnesses in this multi-faceted murder case, will discuss the wide variety of interesting evidence examined, the testimony and the defense of forensic document examination.

• Nomination of David J. Purtell for 2006 ASO Award Diane K. Tolliver Maureen C. Owens

Abstract: David J. Purtell was invited to attend his first meeting of the ASQDE in 1959 and was voted into Regular membership in 1961, one of the first Government examiners to hold membership in the "Society." His document examination career dates from 1947 through l985, and encompasses myriad and continuous contributions to the profession of Document Examination and to the advancement and goals of the Professional Organizations of Document Examination.

• Why FDE Training Must Be For A Minimum of 24 Months Joseph Parker

Abstract: During development of the ASTM Guideline for Training of Forensic Document Examiners (FDE), a great deal of discussion occurred concerning the minimum 24-month period proposed for the fundamental/core FDE training. In this presentation I will address various aspects of this discussion that support a minimum 24-month training period for Forensic Document Examiners.

• Towards a Measurable Quality in Handwriting Examination Dr. Peter W. Pfefferli

Abstract: Measurable quality in forensic science is a key objective of the European Network of Forensic Science Institutes (ENFSI). The various Working Groups of this professional organization stimulate initiatives to comply with best practice and raising the level of quality and science. If certification and accreditation are adequate standards to achieve these goals, they are more difficult to be applied in some forensic areas. This is particular true for the handwriting practice in police services. Among the few leading QD laboratories in Europe, the Document Laboratory of the Zurich Police in Switzerland as well as the Document Laboratory of the National Police in Denmark have a common challenge to meet the requirements towards a measurable quality. In a comparative study, the respective experiences for a quality management system in police practice are presented. The focus is given to educational, organizational and technical awareness for a chain of quality standards in handwriting expertise. The results are summarized in a checklist helpful to avoid quality gaps in practice.

• Another Look at Sequence of Strokes Dennis Ryan

Abstract: Sequencing of strokes has presented problems for the Forensic Document Examiner for many years. There has been a great deal of discussion on the issue with no real consensus. This presentation will take another look at this problem using infrared luminescence. Can infrared luminescence provide us some insight into this problem of sequencing the strokes or can we misinterpret what we see with the luminescence of one of the inks?

• Striations -- A Method Michael Talbot-Wilson

Abstract: A method of examination of ball pen striations is presented and the difficulties which limit the usefulness of striations are documented. The pens used in a business or department may be "out of the same box" (or bulk order), defeating almost all methods of differentiation. Examination of striations is one of the methods potentially applicable to the question of whether accounts were written up in a single sitting, and equivalent relative dating questions. The ability to detect the turning of the pen in the hand (a new grip)is a challenge to be met in validating the use of striations in such examinations.

• Survey of ASCLD/LAB Accredited Labs With ESDA Lift Disposition Policies Diane K. Tolliver

Abstract: A survey was conducted to determine the policy of laboratories accredited by the American Society of Crime Laboratory Directors Laboratory Accreditation Board (ASCLD/LAB) in respect to the disposition of ESDA lifts. Those labs willing to provide their written policy for this presentation will be included. Others will be summarized. This is an important consideration since ASCLD/LAB has determined that information derived from non-recoverable evidence, must be treated as evidence.

• Application of CEDARFOX Software in the Association of Bank Robbery Demand Notes Gabriel Watts, Brea Foster

Abstract: Dr. Sargur Srihari and his colleagues at Cedartech, Inc. have been developing handwriting recognition software based on the same technology that reads handwritten postal addresses. CEDAR-FOX (Center of Excellence for Document Analysis and Recognition- Forensic Examination) software combines handwriting recognition technology with a search- and-locate feature known as “word spotting,” which allows a user to search a known set of writing for a word containing similar handwriting characteristics to that of a questioned word. This tool could decrease subjectivity in the interpretation of handwriting characteristics when searching through large quantities of writing. At this time, CEDAR-FOX is unable to recognize 100% of digitized handwriting, which requires some degree of effort on the part of the user to define the words that the software did not automatically recognize (known as “word truthing”). A small scale test of CEDAR-FOX software was conducted using mock = nk robbery demand notes. The results of the software’s ability to effectively and accurately associate bank robbery demand notes of the same writer will be discussed.

• Fracture Match: A Preliminary Look at Validating Whether Paper Tears are Unique Todd W. Welch, Mary K. Bacon, Charles R. Bacon

Abstract: Forensic document examiners are frequently asked to analyze evidence that requires the identification of a torn or broken item by reconstructing it. The reconstruction and identification of a rip or tear is also known as a fracture match. Fracture match examinations have important forensic applications in that they may establish a relationship between 1) the suspect and the crime scene, 2) the victim and the crime scene, or 3) the suspect and victim. Be it a ransom note, a gum wrapper containing narcotics, or a match, the forensic document examiner endeavors to associate the evidence. After attending a workshop on fracture match techniques, the author set out to answer several questions concerning fracture matches as it relates to paper. Namely, if one purports that no two papers will rip/tear the same way, what are the variables and conditions that must be controlled to scientifically test this? What about the reproducibility of the tears? While it may be intuitive that ripped paper can be pieced back together, how can the forensic document examiner give a better explanation of what one already knows? This presentation will discuss the experimental design and some preliminary data for this validation study.

• Fun and Games within the Nevada State Penitentiary Floyd Whiting

Abstract: After forgers have been incarcerated for their crimes, one does not expect them to continue their crime sprees from within the prison walls. But some inmates consider themselves so intelligent that they expect to scam the prison authorities and use prison facilities to produce fraudulent documents that will benefit themselves or fellow convicts. Such an individual, who has been an inmate of the Nevada prison system for many years, pretends to be a millionaire imprisoned on trumped up charges. He produces documents to show his friendship with political authorities, including ex-president Clinton. He is a con man who corresponds with various individuals, then when they reply, he uses the official letterheads and signatures from the correspondence to manufacture additional fraudulent documents. During my career as a document examiner in Nevada, I have had several occasions to examine documents produced by this person. The presentation will illustrate some of the scams perpetrated by the inmate as well as the techniques and reasoning processes used to expose his frauds.

• Some Problems Related To Ink Jet Printouts Janis Winchester

Abstract: This paper will examine the definitions associated with the ink jet printer and related computer programs. It will discuss the characteristics of the printer and output.

• Identifying Features Of A Laser Printer And An Ink Jet Printer Roslind Winter

Abstract: A case study of the comparison of a laser printed questioned document and four suspect laser printers. An unusual defect in the printing of the questioned document provided evidence to eliminate all the suspect printers. A case study of the comparison of an ink-jet printed questioned document and two suspect ink-jet printers. The cause of the defect was determined by examining the print cartridge. By identifying the cause it was possible to eliminate one of the suspect printers and provide evidence to link the other suspect printer with having produced the questioned document.

• Malicious Communications Database Roslind Winter

Abstract: Senders of malicious communications will often send many communications relating to the same topic (e.g. racism) to many different victims. If the victims are not known to each other then these crimes will often not be linked. While a single letter or telephone call may be treated by the investigator as a minor offence with little chance of resolution, linking many offences together many enable more serious charges to be brought. Also sending letters may be a pre-cursor to committing more serious crime (e.g. some writers of anonymous letters may go on to commit acts of indecency or sexual assault). The original malicious communications database was set up as a paper-based system in the Metropolitan Laboratory, latterly the London laboratory. Its main function was to link similar crimes together. This paper describes the operation and use of the electronic version of the database and is intended to cover remote access to the data. Malicious communications can take many forms-e.g. letters, circulars, graffiti, e-mails, phone calls et. While this system has been initially populated with malicious communications sent on paper, the system can be used for all communications. By capturing as much information about the communication as possible using a standard form and including the text it may be possible to link the communication to others on the database. Providing this information to the investigator will in turn assist his enquiries and provide intelligence information to him.

POSTERS

Ink-jet printer investigation through data mining

Helen Pinto and Marc Gaudreau

ABSTRACT

This paper summarizes a small project carried out for Canada Border Services Agency (CBSA), to see if measurements taken from ink-jet printed documents via ImageXpert could be used to distinguish between printers of different makes and models in a blind test.

A dataset consisting of 362 measurement and 42 samples was examined without knowing how many printers were represented, or how many samples came from each printer. Using cluster analysis, we came up with the most likely groupings of samples which correctly matched all samples to their respective make and model of printer. Furthermore, it was even possible to distinguish individual printer signatures within each cluster. A second data mining technique called decision tree induction allowed us to go one step further and show that it is appears possible to build a rule-based classifier to predict the printer make and model of each sample, using only those measurements taken from alphabetic text.

While the findings through both methods of analysis were encouraging, the dataset was too small to claim that these findings are significant enough to make forensic decisions on. Therefore, the next phase of this project currently in negotiation, aims to investigate these results further with a larger dataset.

• Tomatoes and Beef Stick Catch a Killer Marvin H. “Mickey” Dawson

Abstract: An attractive graduating senior stopped to pick up her mail from the family's rural mailbox; and never returned home. For several weeks an intense investigation and man hunt was undertaken as an annonymous caller teased her family with tibits of personal information of the missing high school beauty queen. The killer made a fatal mistake when he allowed his captive to send her "Last Will and Testament" to her family. ESDA found the tomatoes, beef stick and phone number(s)

• Examination Of TTI Headers From Multipurpose Facsimile Devices Lyne Desroches

Abstract: Cases involving the use of multipurpose office machines present new challenges for forensic document examiners. A preliminary study of home inkjet multifunction facsimile devices indicates that additional considerations must be given in the examination of faxed documents and to header-based make and model determination. Different TTI headers have been observed on scanned documents faxed from the same machine depending on whether the document was transmitted using the front panel keys or the software utility of the device.

• Automatic Letter Opening Machines Dr. Audrey Giles

Abstract: A recent case featured a damaged letter taken from an envelope that we had been told was opened using a letter opener. The letter opener was not a simple paper knife as first supposed, but an automatic electric letter opener of the type used in post rooms of organisations such as companies and banks for opening large quantities of mail. This prompted a study of four such machines and enabled us to characterise the features found on envelopes and enclosed documents when they pass though one of these devices.

• The Advantages of Hosting Internships Allan Herkt

Abstract: Over the last four years the New Zealand Police Document Examination Section (NZPDES) has hosted five interns from the United Kingdom and the United States. We have two more interns coming to us this calender year. The three interns from Strathclyde University have been here to research and complete their MSc Forensic Science theses while the three students from other Universities have used the time as a work attachment towards their BSc in Forensic Science. From my discussions with other laboratories and with several interns, there is evidence suggesting many forensic laboratories tend to regard requests for internships as a potentially time consuming nuisance. This has not been our experience. By welcoming and fostering the interns we have had to date we have derived significant benefit. Not only has each intern contributed to the social mix within the NZPDES, they have also added variety and an infusion of ideas, knowledge and enthusiasm. More importantly they have all produced outstanding practical research papers, be it for their thesis or otherwise. This paper covers the procedures we follow to encourage and nurture these interns and gives summaries of the different research papers and theses, they have produced to date. Copies of these papers are available, on request, to ASQDE members. One has already been published in the Journal of Science and Justice.

• Indentation Examination Enhancement Jane A. Lewis

Abstract: Indentation examination is an important technique for forensic document examiners. An electrostatic detection device (EDD) develops latent indented material on questioned documents like anonymous letters or medical records. Research was conducted to determine whether adding moisture directly to the cascade developer beads would improve EDD images. The Electrostatic Detection Apparatus (ESDA) was used to detect latent indented impressions on 100 sheets of paper containing indentations. Fifty sheets were processed according to the ESDA manual instructions. Fifty sheets were processed with an additional step. The cascade developer beads were spritzed with a plant mister before being dispensed on the document. Results showed improved visualized indentations with the spritzer method of developing electrostatic images with ESDA.

• Security and Integrity of Digital Graphic Images Jeffrey H. Luber

Abstract: A simple, easy and extremely secure commercial method of data encryption is described for the security of digital graphic image files. Currently this method analyses graphic .bmp, .tiff and .jpeg files utilizing MD5 algorithms to generate an accompanying alphanumeric file identifier which will decode to the original alphanumeric file when the unaltered file is opened on a computer utilizing the same data encryption software. For example, a one-pixel slight color change of a 64 mg file image will result in a completely different alphanumeric file identifier, and thus a “Certificate Authorization” will NOT be generated. An analyst that can open a graphic file in a program such as Photoshop, will not have any problem in using this “Certificate Authorization” program.

• Analysis of Xerographic Toner by X-Ray Photoelectron Spectroscopy: False Positives Albert H. Lyter III, Ph.D.

Abstract: Previous work on the use of X-ray Photoelectron Spectroscopy (XPS) to analyze xerographic toner has shown the feasibility of determining the approximately age of the printed material. These procedures involve the comparison of toner samples, both untreated and artificially aged. The presence of statistically significant differences between these two sample classes was found to be an indication of recent preparation. Recent preparation has been defined as within 2 years of the date of analysis. In this work, a larger sample population is examined under the same experimental conditions to ascertain the rate of, or possibility of, false positives. In this methodology a false positive is defined as a statistically significant change between untreated and artificially aged toner samples when the actual age of the toner samples is greater than 2 years. Previous work examined xerographic toner from three different manufacturers, one toner type from each. In this work we examined 10 toner types from different manufacturers with no overlap with the previous samples. Multiple samples from each toner type were examined and the parameters of C/O ratio, percent high energy carbon and energy shifts of the high energy cargon were noted. Statistical measurements were performed to evaluate the range of these parameters on all samples. All samples were beyond the 2 years of age threshold and no statistically significant differences were found between the untreated and artificially aged samples. Differences in the range of measured parameters were noted among the different toner types.

• Application of Hyperspectral Imaging to Forensic Document Examination Problems R. Brent Ostrum and Samiah Ibrahim

Abstract: Imaging spectroscopy is a powerful tool used extensively in remote-sensing applications. It has also been used by art historians and conservators to study works of art including paintings and murals. More recently, the technology has been applied with success to writings and markings on historical documents. In particular, work by Attas (2003, 2004) confirmed that spectroscopic imaging can be useful for the enhancement of faded ink writing on old manuscripts. The parallel to applications in forensic document examination is exact but its potential has not been fully explored. In the present study a customized imaging system using electronically tunable filters was used to generate hyperspectral image data from ink samples known to be difficult to differentiate using other methods. The potential of hyperspectral imaging as a tool for QD problems was evaluated, specifically for ink differentiation and the decipherment of obliterations on questioned documents. The basis for a fully automated HSI analysis system was developed.

• Another Look at Handwriting Movement R. Brent Ostrum and Tobin Tanaka

Abstract: QDEs usually observe and assess line quality in the written stroke without direct knowledge of the kinematics of the writing instrument used to produce it. Using a digitizing tablet with an inking pen we demonstrate that the movement of a writing instrument can be measured as a time-series of data and that this data may assist the document examiner in the evaluation of line quality. Line quality features such as speed and rhythm have been described since the days of Osborn. Qualitative tests of handwriting by document examiners are routinely conducted using several well-documented principles in the literature. The use of a digitizing tablet with an inking pen provides the document examiner with a traditional handwritten document linked to a measured time-series for pen movement. From this time-series the movement data can be derived; in Cartesian coordinates the x,y,z positions together with their respective velocities and accelerations. This data can be correlated or compared to the observed and/or measured aspects of the visible written line. Data of this type can be used to support various claims by QDEs regarding their ability to perform subjective interpretative tasks such as the assessment of writing speed or pen pressure fluctuation. Additionally, this type of analysis may assist an examiner with the verification of writing speeds (magnitude of the velocity), time taken to write elements of allographs, pen pressure and variation in writing. In this study we examine the handwriting of several subjects with a limited passage of cursive text.

• If a Practised Forger Forges, Does the Practised Forgery Favour the Forger or Does Practice Make Perfect? Robert Radley

Abstract: A study of twenty-five individuals' attempts at simulating signatures was undertaken to determine whether the proposition put forward in a particular case that the more a forger practises, the better the forgery becomes to the extent that "practice makes perfect", could be confirmed or refuted. This, it was proposed, was the reason early dated simulated signatures were detected as such by the examiner whilst later dated questioned signatures, identified by the examiner as genuine, were really "perfect forgeries". Three differing types of signatures were given to individuals to practise daily over a five-day period. A relatively small number of individuals showed significant skill in copying. The majority of individuals showed slight improvement with only three sets of practised signatures out of seventy-five in total showing a moderate improvement. There were no large scale improvements on practice by any individual. A questionnaire completed by the best five of the simulators showed a general consensus of opinion that improvements with practice were only generally "slight."

• Can Pattern Recognition be Taught? Feuerstein's "Instrumental Enrichment" and Other Cognitive Tools for the Enhancement of Image Processing and Form Recognition Douglas A. Ridolfi

Abstract: Form blindness is an inability to distinguish small differences in size, shape and form. This lack of ability to distinguish between real differences and fundamental similarities makes comparison of handwriting, latent prints and other types of physical evidence involving pattern comparison, difficult if not impossible. Are there instructional methods available that teach recognition and perception of small detail? Can these methods be utilized by questioned document apprentices to improve their ability to recognize simulations and minor differences in handwriting between writings from different people? This paper will review some of the basic concepts in perception, form blindness and cognitive instructional methods such as those by Feuerstein and others that concentrate on pattern recognition and problem solving related to interpretation of visual clues in images.

• Another Look Through Dicroic Filters Peter Tytell

Abstract: This presentation will take another look, a measured look, at a much-used tool for ink differentiation. Measurements of the transmission characteristics of dicroic filters (single filters and filter sets) that have been empirically found to be useful in the examination of various inks will be compared to isolate common significant features. These features will be considered in light of theories and known phenomena of human color vision. In addition, examples of digital imaging to illustrate the results of dicroic filter examinations will be provided. Paths for future research and development in this area will also be suggested.

• The Influence of the Inverted Hand Position (IHP) on Stroke Direction in Left Handed Handwriting

Clifford Hobden B.A., Dip. Inf. Management/ Lib, FACBS 2005

ABSTRACT: The forensic examination of handwriting is generally lacking in statistical validation and has been rightly criticized for a lack of research in this area. The dynamic nature of what is taught as handwriting and the varying effects of migration and aging on any given population make it extremely hard to produce useful data. However there is one area on which both of these two variables seem to have a far reduced effect. Previous research by Franks et al (1985) shows that there is a strong difference between the directions of pen strokes of left handed people compared to right-handed people irrespective of age or education. This difference can be summarized by left-handers being far more likely to make cross-strokes from right to left than left to right and similarly, writing circular letters and numbers with clockwise strokes, rather than counter clock-wise. Measuring the statistical significance of this difference between left a

• The Return of Graphometry? Manfred Hecker

Abstract: None submitted.

• Case Examination: Scanned-Printed Signatures in Arabic Yaacov Yaniv, Jay Levinson

Abstract: Two documents regarding rights to real estate in Jerusalem were submitted for examination. The documents contained both text and signature in Arabic. One document contained a signature written with a . It was determined that the second document was produced using a scanner-printer. The authors conducted experiments with scanned documents printed on both ink-jet and laser printers. They found significant inconsistencies in the printed documents and determined that no handwriting examination could be conducted.