Bibliography

Sources are grouped by confidence tier, following the book's citation-honesty policy (see _style-bible.md §7). Tier 1 are works and rulings we are confident exist — the National Academy of Sciences 2009 report, the 2016 PCAST report, NIST/OSAC standards, landmark court decisions, and named, publicly documented cases; Tier 2 are real ideas whose exact publication we have not pinned down; Tier 3 are constructed teaching examples — above all the book's recurring cold case — labeled where they appear.

Tier 1 — Verified canonical sources (the NAS 2009 and PCAST 2016 reports, NIST/OSAC and SWGDAM standards, named landmark cases with public records, peer-reviewed literature)

• Brady v. Maryland (1963) — requires disclosure of exculpatory evidence; legal anchor for the official-misconduct factor and the open-file-discovery reform.
• Carpenter v. United States, 138 S. Ct. 2206 (2018) — U.S. Supreme Court; acquiring historical cell-site location records is a Fourth Amendment search requiring a warrant; the legal frame around the records §25.5 dissects.
• Daubert v. Merrell Dow Pharmaceuticals (1993); Kumho Tire Co. v. Carmichael (1999); Federal Rules of Evidence, Rule 702 — the admissibility gate (Ch.5); every junk-forensics wrongful conviction is also a gatekeeping failure.
• Daubert v. Merrell Dow Pharmaceuticals, 509 U.S. 579 (1993) — owned by Ch.5; cited for "vigorous cross-examination" as the antidote to "shaky but admissible evidence" (the doctrinal root of §30.3).
• Daubert v. Merrell Dow Pharmaceuticals; Frye v. United States; Kumho Tire v. Carmichael — admissibility/gatekeeping standards under which probabilistic-genotyping source-code and validation disputes are litigated (treated fully in Ch. 5).
• District Attorney's Office for the Third Judicial District v. Osborne, 558 U.S. 52 (2009) — no freestanding federal constitutional right to post-conviction DNA testing.
• Dusky v. United States (1960) — the competency-to-stand-trial standard (present rational/factual understanding of the proceedings + ability to consult counsel with reasonable rational understanding); anchor for §28.2 and case-study-02.
• Frazier v. Cupp (1969) — U.S. Supreme Court; police may lie to a suspect about the evidence, and a confession so obtained is not automatically inadmissible; the legal basis for the false-evidence ploy.
• Kumho Tire Co. v. Carmichael, 526 U.S. 137 (1999) — owned by Ch.5; the specific-opinion / specific-field requirement for qualification (§30.1).
• Kumho Tire Co. v. Carmichael, 526 U.S. 137 (1999); Daubert v. Merrell Dow Pharmaceuticals (1993); Federal Rules of Evidence, Rule 702 — the admissibility gate; Kumho extended Daubert reliability to non-scientist (incl. engineering) expert testimony, and itself concerned an engineer's testimony about a failed tire.
• Melendez-Diaz v. Massachusetts (2009) — a forensic lab's drug-identification certificate is testimonial; analyst subject to confrontation; central to §21.3 and the contrast with field-test cases.
• Melendez-Diaz v. Massachusetts, 557 U.S. 305 (2009) — case study 30.1; forensic lab certificates are testimonial; the analyst is a witness against the defendant, subject to Sixth Amendment confrontation/cross-examination; 5-4, Scalia, J.; upholds notice-and-demand statutes; majority expressly rejects "neutral/near-infallible" framing of forensic evidence.
• Miranda v. Arizona (1966) — warnings and the rights to silence and counsel in custodial interrogation.
• People v. Collins, Supreme Court of California (1968) — origin of the prosecutor's fallacy in court; product-rule misuse and the coincidence-vs-guilt leap (Case Study 1).
• R v. Sally Clark, Court of Appeal (England and Wales) — wrongful conviction involving statistical evidence; conviction quashed 2003 (Case Study 2).
• Riley v. California, 573 U.S. 373 (2014) — U.S. Supreme Court; warrant generally required to search the digital contents of a phone seized incident to arrest; "the privacies of life" framing; doctrinal heart of §25.6.
• The People v. O. J. Simpson (Los Angeles, 1994–1995), trial record and subsequent forensic literature — collection, packaging, and chain-of-custody failures eroding high-validity evidence (Case Study 1).
• United States v. Glynn (S.D.N.Y. 2008) — case study 15.2; admitted firearms-toolmark testimony but restricted it (to "more likely than not" in the ruling's formulation) and barred "reasonable degree of scientific certainty"; documented anchor for the line of decisions limiting firearms testimony.
• 2015 FBI / DOJ acknowledgment on microscopic hair comparison — the Bureau's review concluding examiners had for years overstated hair "matching" in numerous cases; a Tier-1 institutional admission that the second sin was systemic (ties to Ch.19).
• Albert S. Osborn, Questioned Documents (1910; 2nd ed. 1929) — the foundational text of American questioned-document examination, by the lead Lindbergh examiner; classic statement of method and a historical artifact of the discipline's absolute, individualizing language.
• American Academy of Forensic Sciences (AAFS), aafs.org — the major U.S. professional society (founded 1948); disciplinary sections, annual meeting, peer-reviewed journal; affiliated with FEPAC (forensic-education accreditation). A society, not a certifying board.
• American Board of Criminalistics (ABC) — certifies laboratory criminalistics practitioners (DNA/molecular biology, drug chemistry, fire debris, trace, etc.).
• American Board of Forensic Odontology (ABFO) — certifies forensic dentists; credential historically spans valid dental identification (Ch.17) and discredited bite-mark comparison (Ch.16) — used to illustrate that a credential attests to a person, not a method.
• American Board of Forensic Odontology (ABFO) — published guidelines/standards for human identification (the identification/possible/insufficient/exclusion conclusion categories structuring §17.3) and the documented curtailment of permissible bite-mark claims over time.
• American Board of Medicolegal Death Investigators (ABMDI) — certifies medicolegal death investigators who work with forensic pathologists.
• American Board of Pathology (forensic pathology) — the medical specialty board certifying forensic pathologists (physicians).
• CITES (Convention on International Trade in Endangered Species of Wild Fauna and Flora) — the legal framework making species identification a forensic (legal-forum) question.
• Court findings and official investigations re Annie Dookhan and Sonja Farak (Massachusetts drug-lab scandals) — referenced (not re-analyzed; primary treatment in Ch.4) as evidence that codes/accreditation alone do not prevent misconduct.
• Cuyahoga County (Cleveland), Ohio Sexual Assault Kit Task Force and the Ohio AG's statewide kit-submission initiative — public record (case study 37.1); parallel large-scale testing, dedicated task force, serial-offender identifications, prosecutions.
• Detroit / Wayne County untested-sexual-assault-kit initiative — public record + the federally funded action-research project documenting it (case study 37.1); ≈11,000 kits discovered 2009; testability of old kits; CODIS-driven serial-offender exposure; combined technical+policy reform. Figures approximate.
• Documented roadside drug-field-test exonerations, esp. the Harris County, Texas (Houston) conviction-integrity review — case study 21.1; substances that field-tested "positive" later found by the lab to contain no controlled substance; pattern documented across jurisdictions and in investigative reporting.
• Edmond Locard, work on the exchange of traces (early 20th century), via standard forensic histories — origin of the exchange principle and the Lyon laboratory.
• FBI / Department of Justice review of microscopic hair comparison (announced 2015) — institutional acknowledgment of widespread examiner overstatement; anchor for Case Study 2 (Tribble).
• FBI / U.S. Department of Justice / Innocence Project / National Association of Criminal Defense Lawyers (NACDL), microscopic-hair-comparison review (2012–2015), and the FBI's public statement of April 2015 — the institutional center of the chapter; in the large majority of reviewed cases with testimony used against defendants, the testimony exceeded the limits of the science; errors implicated dozens of capital cases. Cite figures as they appear in the public record, not as approximations.
• FBI / U.S. Department of Justice review of microscopic hair comparison testimony (2015 admissions) — owned by Ch.19; examiners overstated the evidence in a large majority of reviewed cases, favoring the prosecution; the Tier-1 evidence that overstatement (§30.4) is systemic. (State totals/percentages qualitatively per the reconciliation note.)
• Federal Rules of Evidence, Rule 702 — owned by Ch.5; the "reliable application" prong is the case-specific cross-examination seam (§30.3); FRE 704 (opinion on an ultimate issue) underlies §30.4.
• Federal Rules of Evidence, Rule 702; Daubert v. Merrell Dow Pharmaceuticals (1993); Kumho Tire Co. v. Carmichael (1999) — the admissibility gate; how a court treats strongly-supported vs. contested expert pathology testimony.
• Frye v. United States (1923); Daubert v. Merrell Dow Pharmaceuticals (1993); Kumho Tire Co. v. Carmichael (1999); Federal Rules of Evidence, Rule 702 — the admissibility gate (Ch.5); Frye's "general acceptance" helped bite marks persist; Daubert/Kumho and the error-rate focus underlie the growing exclusion/restriction of bite-mark testimony.
• Gary Dotson (1989) — first U.S. conviction overturned on DNA evidence; public record.
• Golden State Killer / Joseph James DeAngelo case (public record, 2018 arrest; 2020 guilty plea) — emblem case of investigative genetic genealogy (Case Study 1); SNP profile → GEDmatch → family-tree reconstruction → confirmation by conventional STR comparison.
• Insanity Defense Reform Act of 1984; John Hinckley Jr. insanity acquittal (1982) — narrowed the federal insanity standard and shifted the burden to the defense; shows the defense is harder to win than its TV image.
• Interpol Disaster Victim Identification (DVI) framework/guide — source for the primary identifiers (fingerprints, dental, DNA) and antemortem/postmortem reconciliation; developed in full in Chapter 35; basis for Figure 17.2.
• Interpol, Disaster Victim Identification Guide — the international five-phase DVI framework (scene; postmortem; antemortem; reconciliation; debriefing), the primary/secondary identifier distinction, and standardized forms enabling cross-border identification; the central reference for §35.2–§35.4. Maintained/updated by Interpol.
• Jennifer Thompson-Cannino & Ronald Cotton (with Erin Torneo), Picking Cotton: Our Memoir of Injustice and Redemption (2009) — the principals' joint first-person account of case study 32.1; the human mechanism of memory overwriting and the reform-advocacy turn.
• Kirk Bloodsworth case (Maryland, 1984 conviction; 1993 DNA exoneration) — first U.S. death-row DNA exoneration; public record (Case Study 1).
• Litigation and appellate record of the David Camm prosecutions (Indiana, 2000–2013) — qualified BPA experts reaching opposite conclusions on the same shirt stains across three trials (Case Study 10.2).
• Lukis Anderson case (Santa Clara County, California, 2012; public record) — emblem case of secondary transfer (Case Study 2); innocent hospitalized man's DNA found under a homicide victim's fingernails via paramedic-mediated transfer.
• National Commission on Forensic Science (U.S. DOJ/NIST) — charter, work products (incl. model national code of professional responsibility), and 2017 lapse when its charter was not renewed; illustrates the gap between a strong authority and a dissolvable advisory body.
• National Institute of Standards and Technology (NIST) — forensic DNA program; technical reviews of mixture interpretation and evaluations of probabilistic genotyping systems.
• National Research Council (National Academy of Sciences), Strengthening Forensic Science in the United States: A Path Forward (2009). The "NAS 2009 report" — evidence handling, contamination, and the limits of comparison claims.
• National Research Council, National Academy of Sciences. Strengthening Forensic Science in the United States: A Path Forward. The National Academies Press, 2009.
• National Transportation Safety Board (NTSB), public accident investigations and reports — standing model of rigorous accident/failure investigation across transportation modes.
• NFPA 921, Guide for Fire and Explosion Investigations (National Fire Protection Association; first issued 1992, revised regularly) — the consensus standard of modern fire investigation; requires the scientific method, catalogs invalidated indicators, and explicitly rejects negative corpus; the de facto standard of care and the courtroom yardstick after Daubert.
• NIST and OSAC for Forensic Science standards (seized drugs; fire debris and explosives; chemistry) — institutional basis for confirmatory-identification requirements: appropriate specificity, controls (blanks, standards), method validation, the two-orthogonal-dimensions logic.
• NIST mass-spectral reference databases — the kind of spectral library against which a GC-MS fragmentation pattern is compared; a library "match" is a tool for analyst judgment, not a verdict.
• Organization of Scientific Area Committees for Forensic Science (OSAC), administered by NIST (nist.gov/osac) — the standards-development body and its registry; recommend-not-mandate (voluntary compliance) design.
• OSAC (Organization of Scientific Area Committees for Forensic Science) / NIST and SWG-legacy standards on evidence collection, packaging, and contamination control — the consensus basis for the questioned/known/control distinction and for packaging rules.
• President's Council of Advisors on Science and Technology (PCAST). Forensic Science in Criminal Courts: Ensuring Scientific Validity of Feature-Comparison Methods. Executive Office of the President, 2016.
• Public record of Earl Washington Jr. (Virginia; 1983 conviction; DNA exoneration and pardon; release 2001) — Case Study 34.2; capital false-confession-plus-overstated-serology exoneration; multiple contributing factors combined. NOT used as a primary case study elsewhere in the book.
• Public record of Brandon Mayfield (FBI's 100%-confident, 100%-wrong Madrid-bombing print ID; settled/reviewed) — anchor; the second sin and the near-miss mirroring the cold case (advanced from Ch.14/30/31).
• Public record of Cameron Todd Willingham (executed TX 2004; arson "indicators" shown to lack scientific basis) — anchor; the first sin in its most absolute form (advanced from Ch.5/21/22).
• Public record of the 2004 Indian Ocean tsunami and the Thai Tsunami Victim Identification (TTVI) operation — case study 35.2; large multinational DVI operation under the Interpol framework; dental identification a workhorse; instructive early misidentifications by visual recognition/personal effects, later corrected. (~230,000 dead across the Indian Ocean region; figure approximate.)
• Public record of the 2006 Reuters / Adnan Hajj manipulated-photographs controversy — case study 26.2; cloning artifacts (impossible repetition) in a manipulated news photo caught by content-consistency reasoning; Reuters withdrew the images and pulled the photographer's archive. A media-ethics/authentication case; lesson transfers to forensic image authentication.
• Public record of the 2013 Boston Marathon bombing investigation — case study 26.1; probative multi-camera video evidence AND the documented harm of crowdsourced facial misidentification (e.g., the wrongful online naming of Sunil Tripathi and other innocent people). Used soberly; Dzhokhar Tsarnaev convicted, Tamerlan Tsarnaev died in the manhunt.
• Public record of the assassination of Georgi Markov (London, 1978) — case study 20.1; the platinum–iridium pellet recovered at autopsy, the clinical/postmortem picture, and the ricin determination; examined in the forensic/medical literature; teaches detection at the extreme (exotic agent, sub-milligram dose, analysis guided by physical evidence). Some reconstruction details debated; that Markov was assassinated by a poison delivered via the pellet is not in serious dispute.
• Public record of the Beatrice Six — 1985 murder of Helen Wilson, Beatrice, Nebraska; the convictions of six people; later DNA exonerations of all six and identification of a single (deceased) perpetrator — case study 39.2; landmark example of FALSE/bias-driven convergence (non-independent, internalized false confessions) dissolved by single-source DNA. [SEE FLAG TO ORCHESTRATOR in continuity/ch39.md — add to §6 running tally.]
• Public record of the Brandon Mayfield case (FBI public statements and apology; 2006 civil settlement; Spanish National Police identification of Ouhnane Daoud) — factual spine of case study 31.1; the cascade made flesh, broken only from outside the chain of agreements.
• Public record of the BTK / Dennis Rader investigation and 2005 guilty plea (Wichita, Kansas) — basis for case study 25.1; identifying metadata embedded in a digital file (floppy disk) generated the lead; DNA corroborated before arrest. Load-bearing facts documented; popular-retelling details flagged as needing verification.
• Public record of the Cameron Todd Willingham case (Corsicana, TX: 1991 fire; 1992 conviction; 2004 execution) and its fire-science reviews — the assessment by Dr. Gerald Hurst (2004, before the execution) and the report by Dr. Craig Beyler commissioned by the Texas Forensic Science Commission (2009) — case study 22.1; the book's anchor for junk science with a body count.
• Public record of the Golden State Killer investigation and arrest of Joseph James DeAngelo (2018) — case study 29.1 (anchor advanced from Ch.8); IGG mechanics (SNP profile, GEDmatch, family-tree reconstruction, STR confirmation on abandoned DNA) and the privacy/ethics questions.
• Public record of the Grim Sleeper investigation and conviction of Lonnie David Franklin Jr. (Los Angeles) — case study 29.2; widely cited as the first U.S. arrest from a deliberate familial DNA database search; the familial-searching exemplar and its equity dimension.
• Public record of the Han Tak Lee case (Pennsylvania: 1989 fire; conviction for murder/arson; release in the 2010s after the fire science was found no longer scientifically valid) — case study 22.2; the §6-assigned complementary arson case; the contrast in outcome with Willingham.
• Public record of the Joyce Gilchrist matter (Oklahoma City Police Department; 2001 FBI review finding testimony "went beyond the acceptable limits of forensic science"; dismissal September 2001) and the linked exonerations of Jeffrey Pierce (2001, ~15 years; DNA exclusion) and Curtis McCarty (2007, ~20 years on death row) — case study 30.2.
• Public record of the July 17, 1981 Hyatt Regency walkway collapse (Kansas City, MO) and the National Bureau of Standards (now NIST) investigation — case study 36.2; failure-analysis / load-path reconstruction; the design change that roughly doubled the connection load; subsequent licensing/ethics proceedings (cause vs. responsibility).
• Public record of the Mark Hofmann forgery and bombing case (Salt Lake City, UT, 1985–1987) — case study 18.2; a skilled forger exposed by the physical sciences (artificial ink aging; a shared fabrication fingerprint across "unrelated" documents); guilty plea to second-degree murder and related charges.
• Public record of the October 2002 Beltway sniper shootings and the Virginia/Maryland prosecutions (2003–2004) — case study 15.1; firearms evidence linking scenes to a common firearm and tying a recovered weapon to the scenes; convictions of John Allen Muhammad and Lee Boyd Malvo; the marks tied the weapon to the scenes, not the hand to the trigger.
• Public record of the Oklahoma City bombing prosecutions (1995 bombing; convictions of Timothy McVeigh and Terry Nichols) — case study 21.2 background; explosives-chemistry thread used to teach limits, not to question verdicts.
• Public record of the World Trade Center identification effort, Office of Chief Medical Examiner of the City of New York (OCME), 2001–present — case study 35.1; the largest, most complex forensic identification effort in U.S. history; DNA-led identification of fragmentary, degraded, commingled remains; identification treated as an ongoing obligation with re-testing and new IDs made years/decades later. (~2,750 victims at the WTC site; on the order of ~20,000 recovered remains fragments; figures refined over time.)
• Public record, Brendan Dassey interrogations and litigation (Wisconsin, 2006; federal appeals including a divided en banc decision; Supreme Court review denied; conviction upheld and contested) — case-study-02; a recorded interrogation; the voluntariness-vs-truth split.
• Public record, Commonwealth v. Stephan Cowans (Suffolk County, MA) and the New England Innocence Project account — case study 14.2; latent-print misidentification that survived trial, undone by post-conviction DNA; exonerated 2004.
• Public record, Florida v. George Trepal (conviction 1991; reported appellate proceedings) — case study 23.1; instrumental elemental analysis decisive in identifying thallium poisoning of the Carr family; the instrument confirmed the poison, not the poisoner; defense later challenged forensic testimony.
• Public record, State of Arizona v. Ray Krone (conviction 1992; retrial 1996; DNA exoneration 2002) — case study 16.1; "the Snaggletooth Killer"; confident bite-mark "match," death sentence, DNA excluded Krone and matched another man. Teaches "confidence is not validity."
• Public record, State of Minnesota v. Derek Chauvin (2021) and the death of George Floyd (2020) — case study 11.1; official medical-examiner determination + independent autopsy; agreement on manner (homicide, medical sense) with a contested cause/contributing-condition debate; separation of medical classification from legal culpability.
• Public record, State v. Cotton (North Carolina, 1984 crime; 1995 DNA exoneration; true perpetrator Bobby Poole) — case study 32.1.
• Public record, the Central Park Five / the Exonerated Five (New York: 1989 attack; 1990 trials; 2002 confession and DNA match of Matias Reyes; 2002 vacatur of all five convictions) — case-study-01; five coerced-compliant false confessions, no physical link, overturned by a single DNA match to the true perpetrator.
• Public record, United States v. Bernard L. Madoff (S.D.N.Y., 2009), and U.S. SEC Office of Inspector General review of the failure to uncover the Madoff Ponzi scheme (2009) — case study 27.1; the guilty plea and scale, and the documented failure to act on detailed quantitative red-flag warnings.
• Public record, 1996 Centennial Olympic Park bombing and the investigation/clearing of Richard Jewell; Eric Rudolph's 2005 guilty plea — case-study-01; profile-driven misidentification of an innocent man (never charged, publicly cleared by DOJ).
• Public record, Marvin Anderson exoneration (Virginia, 1982 conviction; 2001 DNA exoneration; 2002 pardon; true perpetrator John Otis Lincoln) — case study 32.2; cross-race ID, standout photo, repeated viewing.
• Public record, prosecution of Jared Lee Loughner (Tucson, 2011 shooting; competency proceedings; 2012 guilty plea) — case-study-02; competency-to-stand-trial and competency restoration; forensic psychology's validated core.
• Public record, SS Noronic disaster (Toronto, September 1949) — case study 17.1; systematic use of dental identification for fire victims at scale; a foundational North American dental-ID milestone; ~118 dead.
• Public record, the Roy Brown case (New York; conviction 1992; DNA exoneration 2007) — case study 16.2; bite mark reportedly showing more upper teeth than Brown possessed (an apparent exclusion minimized into a match); DNA on saliva excluded Brown and implicated the man he identified. Teaches "exclusion ignored."
• Public records of the exonerations cited as data points: Kirk Bloodsworth (Ch.6); Ronald Cotton / Jennifer Thompson (Ch.32); Ray Krone and Roy Brown (Ch.16); Santae Tribble and Kirk Odom (Ch.19); the Central Park Five / Exonerated Five and Brendan Dassey (Ch.33) — each examined in its home chapter; here illustrations of specific factors in Case Study 34.1.
• Public trial record, Georgia v. Wayne Williams (Fulton County, GA, 1981–1982), the Atlanta fiber case — case study 19.2; fiber evidence leaning on rarity and the multiplication of many independent transfers; convicted; the rarity estimates, the use of a "pattern," and the verdict remain contested matters of the record.
• Public trial record, People v. David Westerfield (San Diego County, CA, 2002) — case study 13.2; dueling qualified entomologists; conviction; teaches the method's input-driven error bars.
• Public trial record, R v Huntley (the Soham murders; Cambridgeshire, England; trial 2003) — case-study-01; landmark demonstration of convergent environmental/trace (soil, pollen, vegetation, fibers) class evidence; defendant convicted.
• Public trial record, State of Florida v. Casey Anthony (Orange County, FL, 2011) — case study 13.1; entomology + the CSI effect; acquittal on the most serious charges.
• Public trial record, State of Florida v. Theodore Bundy (1979), and the subsequent public reassessment of bite-mark comparison (participant statements; NAS/PCAST findings) — case study 17.2; the most famous bite-mark "success," instructive because its foundation collapsed while the conviction stands. (Bite-mark exonerations Krone/Brown are OWNED by Ch.16 — cited there, not primary here.)
• Public trial record, State v. Hauptmann (New Jersey, 1935), the Lindbergh kidnapping case — case study 18.1; skilled observation of rare shared features stated with unearned certainty under maximal bias; subsequent re-examination valued for its methodological critiques (post-arrest dictated exemplars, absolute framing).
• Publicly documented Scottish case of Shirley McKie — case study 14.2; confident latent-print misidentification defended by an established bureau; acquitted 1999; later settlement; UK controversy over fingerprint reliability.
• Rapid DNA Act of 2017 (United States, Public Law 115-50) — the statute creating a pathway for approved rapid-DNA instruments to type arrestee reference swabs and search CODIS under defined conditions; defines the validated scope and frames the scope-creep warning.
• Royal Statistical Society public statement on the use of statistics in the Sally Clark case — national statistical body's correction of the transposed-conditional error.
• Santae Tribble case (Washington, D.C., 1980 conviction; 2012 DNA exoneration) — microscopic-hair-comparison failure; public record (Case Study 2).
• Scientific Working Group on DNA Analysis Methods (SWGDAM) — interpretation guidelines for autosomal STRs, mixtures, mtDNA, and Y-STRs; the U.S. consensus practitioner standards.
• Sexual Assault Kit Initiative (SAKI), U.S. Bureau of Justice Assistance — federal program for inventorying, testing, and investigating previously unsubmitted kits; national scope of the backlog.
• SWGDAM / NIST / OSAC DNA standards (Chs. 7–9); the Brandon Mayfield record (fingerprint error and bias, Chs. 14, 31); American Board of Forensic Odontology identification guidelines (Ch. 17) — the established standards the disaster setting applies under harder conditions.
• SWGDAM DNA interpretation guidelines; NIST / OSAC materials on DNA mixture interpretation and probabilistic genotyping — standards behind likelihood-ratio reporting and validated limits.
• SWGSTAIN / OSAC (NIST) bloodstain pattern analysis terminology and standards materials — the defensible vocabulary of directionality, impact angle, and area of convergence/origin. (Confirm current OSAC subcommittee and document titles before citing specifics.)
• The "Phantom of Heilbronn" investigation (Germany, 1990s–2009), public record — a contaminated collection swab fabricating a reproducible false DNA association (Case Study 2).
• The Houston Police Department Crime Laboratory investigation (the Bromwich independent investigation reports) and the creation of the Houston Forensic Science Center — case-study-02; a systemic embedded-lab failure and the structural (independence) reform it drove. (HPD Crime Laboratory matter, distinct from the Harris County roadside drug field-test issue used in Ch.21.)
• The Innocence Project (innocenceproject.org) and the Innocence Network — DNA-based exoneration and forensic re-examination; source of much of the evidence base on flawed/overstated forensic testimony in wrongful convictions (Ch.6, Ch.34); Case Study 40.2.
• The Innocence Project (innocenceproject.org) and the National Registry of Exonerations — empirical record of the leading contributing factors to wrongful conviction (overstated forensic testimony, false confessions, eyewitness error) and how they combine; the failure modes a disciplined assembly resists (Ch.6, 34).
• The Innocence Project (innocenceproject.org) — arson / fire-science reform record; context for revisiting convictions built on discredited fire science.
• The Innocence Project (innocenceproject.org) — case and policy record; misclassified manner of death (homicide as accident/natural or the reverse) as a stakes-raising context for the manner determination.
• The Innocence Project (innocenceproject.org) — case/policy record; context for mistaken-identification risk (including from imagery) and overstated forensic testimony.
• The Innocence Project (innocenceproject.org) — case/policy record; fingerprint misidentifications among DNA exonerations, often alongside eyewitness error (the Cowans dynamic).
• The Innocence Project (innocenceproject.org) — case/policy record; profiling is less often a direct cause of wrongful conviction than discredited pattern methods (courts rarely admit it as proof), but its investigative misuse causes harm short of a courtroom.
• The Innocence Project (innocenceproject.org) — context for the "double failure" of untested biological evidence (guilty unidentified, innocent uncleared); Ch.6; Ch.1 §1.6 on exclusion.
• The Innocence Project (innocenceproject.org) — founded 1992 by Barry Scheck and Peter Neufeld; case records and analyses of contributing causes; current exoneration counts.
• The Innocence Project (innocenceproject.org) — policy/reform record (lab independence, recorded interrogation, eyewitness reform, oversight) and the exoneration record underpinning the reform argument.
• The Innocence Project case record (innocenceproject.org) — exonerations involving mishandled, contaminated, or misinterpreted physical evidence.
• The Innocence Project case record — broader pattern of feature-comparison overreach (Theme 2 context).
• The Innocence Project — documented DNA exonerations; DNA's exclusionary/exonerating power and the role of overstated forensic testimony.
• The Innocence Project — post-conviction DNA exonerations; the record establishing DNA's power to exclude.
• The Innocence Project. Case profiles and exoneration data. innocenceproject.org.
• The National Registry of Exonerations (law.umich.edu/special/exoneration) — documents U.S. exonerations and contributing factors (including flawed forensic evidence); forensic literacy as record-keeping; Case Study 40.2.
• The National Registry of Exonerations (Univ. of Michigan / Michigan State / Northwestern) — the broader academic record of exonerations with and without DNA; the data behind contributing-factor analysis across thousands of cases; the antidote to over-reading DNA-only cases.
• The National Registry of Exonerations (Univ. of Michigan / Michigan State et al.), case record for Lisa Roberts (Oregon) — Tier-1 basis for case study 25.2; cell-phone location evidence overstated (phone routed through a distant tower treated as placing the person at the scene); guilty plea vacated for ineffective assistance (failure to investigate the cell-tower evidence); charge dismissed 2014; post-conviction DNA pointed to another suspect. Distance figures (~3.4 mi tower–park; ~8 mi actual) cited as reported.
• The National Registry of Exonerations — searchable database of U.S. exonerations with coded contributing factors; empirical basis for the leading-causes claims.
• The voluntariness doctrine / totality-of-the-circumstances test (U.S. due-process case law) — the constitutional standard for admitting a confession; asks whether it was coerced, not whether it was true.
• Trial record of People v. O. J. Simpson (Los Angeles, 1995) — documented testimony on blood collection, handling, contamination, and chain of custody (Case Study 10.1).
• U.S. Department of Justice interim policy on forensic genetic genealogical DNA analysis and searching (effective 2019) — conditions on forensic genetic genealogy in federally supported investigations (generally violent crime / unidentified remains; conventional searching first); the first formal governance attempt.
• U.S. Department of Justice, Office of the Inspector General. A Review of the FBI's Handling of the Brandon Mayfield Case. 2006.
• U.S. Department of Justice, Office on Violence Against Women, A National Protocol for Sexual Assault Medical Forensic Examinations (the "National SAFE Protocol") — authoritative reference behind §37.2–37.3: patient-centered, trauma-informed care, informed consent, and evidence collection without a requirement to report.
• U.S. Federal Bureau of Investigation Laboratory — discontinuation of routine gunshot-residue examinations (2006) — documented institutional decision reflecting GSR contamination/interpretation problems; case-study-02; the concrete illustration that GSR detection is sound while the inference "fired a gun" is not.
• U.S. Fish and Wildlife Service, National Fish and Wildlife Forensic Laboratory (Ashland, OR) — the only full-service wildlife crime lab and the designated CITES laboratory; institutional model of forensic science with a non-human victim.
• U.S. Supreme Court eyewitness line — Manson v. Brathwaite (1977) and Perry v. New Hampshire (2012); Daubert v. Merrell Dow (1993) / FRE 702 (Ch.5) — the legal framework critiqued in §32.3 (older reliability test counting the witness's certainty among indicators of reliability).
• United States Senate, Committee on Foreign Relations, Subcommittee on Terrorism, Narcotics and International Operations, The BCCI Affair (1992) — case study 27.2; definitive public account of the BCCI structure, money laundering, and the cross-border audit-trail problem.
• Violence Against Women Act (VAWA) and reauthorizations — federal framework conditioning certain funding on making the forensic exam available at no cost and without requiring cooperation with law enforcement (basis for the "restricted"/unreported-kit option, §37.2). Consult current statutory text for present provisions.

Tier 2 — Attributed (real work; exact citation unverified)

• Academic literature on the contributing causes of wrongful conviction — quantifies frequency and co-occurrence of eyewitness error, forensic error, false confessions, informants, misconduct (verify specific percentages against a named registry/study).
• Accident-reconstruction literature on speed-from-skid (energy + friction), momentum-based reconstruction, crush-energy analysis, and event data recorders — underlying physics and honest limits (minimum-speed nature, friction uncertainty, EDR value/limits) attributed generally.
• ACE-V methodology and documentation standards (latent-print scientific working groups / standards bodies) — basis for linear ACE-V, analysis-before-comparison, blind verification; attributed generally, specific documents vary by era/organization.
• Acid phosphatase / p30 (PSA) and amylase body-fluid screening literature — verify reagents and cutoffs against current laboratory methods.
• Appellate litigation over source-code access to probabilistic-genotyping programs; courts split (cited in substance, jurisdiction-specific dockets not pinned).
• Association of Firearm and Tool Mark Examiners (AFTE) theory of identification and range-of-conclusions framework — source of the "sufficient agreement" standard and the identification/elimination/inconclusive categories (§15.4); attributed as genuine, widely-used professional guidance without quoting a specific edition; key point that "sufficient agreement" lacks a fixed numeric threshold is the consensus reading.
• Bayesian approaches to the weight of evidence for jurors/analysts (Ch.9) — priors, posteriors, communicating combined evidence; underlies the §39.4 grammar of "strongly supports"; attributed as a framework without endorsing any single courtroom formula (the book's position: verbs matter more than any number shown to a jury).
• Body-relocation inferences from out-of-habitat entomofauna — the established principle that anomalous insect fauna can indicate a body decomposed elsewhere; specific casework attributed generally.
• Brandon L. Garrett, Convicting the Innocent (Harvard University Press) — empirical study of the first wave of DNA exonerations and what went wrong across them.
• Clinical/ethical literature on competency restoration, including forcible-medication questions — competency can be restored and re-evaluated; involuntary medication to restore competency raises serious legal/ethical issues; attributed generally, constitutional cases beyond Dusky noted without citing by name.
• Cognitive-science research on contextual bias in fingerprint examination — the Dror-and-colleagues experiments in which biasing context shifted examiners' conclusions; existence and direction attributed, exact figures not pinned; Chapter 31 develops in full.
• Comparative literature on the DVI primary identifiers (fingerprints, dental, DNA) — survivability, speed, cost, and antemortem-availability trade-offs under disaster conditions; general trade-offs in §17.4 attributed without a pinned citation.
• Content-provenance standards (e.g., the C2PA "content credentials" effort) and authentication-at-capture — a real, active standardization effort for tamper-evident signed origin/edit-history records; design goals and the "prove genuine lineage rather than detect fakes" strategy attributed, without representing them as universal or court-settled.
• Contextual-bias research underlying examiner-independence reforms (case study 23.2) — attributed in general terms; named experiments and formal treatment belong to Ch.31.
• Crime-laboratory backlog reporting and workforce surveys — governmental and journalistic reporting documenting persistent backlogs and the demand/capacity mismatch (incl. untested sexual-assault kits, Ch.37); existence/persistence attributed, no specific year's figure cited.
• Decomposition and insect-succession studies using animal (commonly pig) carcasses as human analogs — the empirical basis for succession-based estimates; key limit: region- and condition-specific.
• Department of Justice and forensic professional-organization responses to PCAST — the "casework is not validation" debate.
• Documented false-positive behavior of roadside drug field kits (color reactions on a range of substances including legal materials; error from non-specialist reading under field conditions) — attributed in general terms; specific substances/rates vary by kit and study.
• Documented mixture-interpretation failures and large-scale case reviews at a U.S. crime laboratory; critique of the Combined Probability of Inclusion / Random Man Not Excluded (CPI/RMNE) approach for ambiguous mixtures — basis for §8.3's Junk-Science Alert.
• Donald Cressey's research on embezzlement and the origins of the fraud triangle (pressure/non-shareable problem, opportunity, rationalization) — attributed with its lineage; formulation elaborated by later scholars.
• Electrostatic-detection (ESDA/EDD) literature on indented-writing recovery — instrument operation (charged film over the document on a porous platen, toner development), recovery from sheets several layers down, humidification needs, and degradation with handling/over-writing; standard method, case-dependent specifics.
• Empirical studies of document-examiner performance — a real, limited literature reporting that trained examiners make fewer false-positive (wrong-attribution) errors than laypeople though they more often decline to conclude; attributed generally, with the honest flag that some work was done by parties interested in the discipline's validation.
• Empirical studies of forensic error rates (latent-print and other feature-comparison work relied on by PCAST) — establish non-zero error rates, refuting "zero error rate" testimony; rates are method-/study-specific.
• English appellate guidance on statistical testimony from 1990s cases (commonly cited as R v. Deen and R v. Doheny and Adams) — experts to state profile rarity and leave the conclusion to the jury; the R v. Adams line on Bayesian reasoning before a jury.
• Entomotoxicology research — drugs/toxins detectable in larvae that fed on contaminated tissue (alternative specimen when the body is too decomposed for routine toxicology) and the fact that some substances alter larval development rate (an ADD confound); directions of effect attributed in general terms.
• Environmental-forensics literature on chemical source attribution / "fingerprinting" — oil-spill ID by GC-MS and biomarker-ratio analysis; contaminant age-dating by degradation state and plume modeling — methods and honest limits (distinctiveness, candidate-set completeness, weathering, modeling uncertainty) attributed generally.
• Erin Murphy, Inside the Cell: The Dark Side of Forensic DNA — failure modes even of the gold-standard method.
• Failure-analysis / forensic-engineering literature on fracture mechanics (ductile vs. brittle, fatigue and progression/"beach" marks), load-path reconstruction, and materials testing — established principles attributed generally.
• FBI Behavioral Science Unit's original incarcerated-offender interviews and the development of the organized/disorganized model — derived from a non-representative sample of caught offenders, no control group, no measured error rate; historical fact attributed generally; specific numbers/named agents not cited.
• FBI Laboratory / Noblis large-scale "black box" study of latent-print examiner accuracy (~2011) — most-cited empirical answer to "how good is ACE-V?"; false positives rare (order of a fraction of a percent), false negatives considerably more common; exact percentages treated as approximate, attributed not pinned.
• FEPAC (Forensic Science Education Programs Accreditation Commission) — affiliated with AAFS; accredits university forensic programs for natural-science content; role and limits attributed (not legally required; not a guarantee of program quality).
• Fire-and-explosion engineering literature on electrical-cause determination — the cause-arcing vs. victim-arcing (arc-through-char) distinction; energized-circuit and competent-ignition-mechanism analysis — attributed generally, consistent with NFPA 921.
• Forensic fire-debris analysis literature and the ignitable-liquid classification framework — how a GC/GC-MS pattern from debris is compared to reference ignitable liquids and classified (e.g., gasoline); attributed in general terms.
• Forensic literature on DNA transfer, persistence, prevalence, and recovery — controlled studies of secondary/tertiary transfer, shedder status, surface and moisture effects; basis for §8.1's "happens often enough that you cannot ignore it."
• Forensic literature on friction ridge formation and persistence — fetal development (~10th–16th week), genetics plus random womb micro-conditions (hence identical twins differ), permanence for life barring deep dermal scarring; attributed as consensus.
• Forensic literature on transfer and persistence of trace evidence — grounds "transfer ≠ persistence"; rates are materials- and condition-dependent (no single figure cited).
• Forensic-science workforce and salary data (national labor-statistics-type sources) — real and useful for expectations, kept qualitative per §7; employment/salary vary by discipline, region, and year; no precise numbers invented.
• Forensic-statistics texts on the likelihood-ratio / "logical" approach to evidence evaluation; the principle that the scientist reports the LR and the court supplies the prior.
• Forer's Barnum-effect demonstration and its many replications — people rate identical vague personality sketches as highly self-accurate; phenomenon and the standard sketch's character attributed without quoting a specific reproduction (the §28.5 sketch is a representative paraphrase, not a verbatim quotation).
• FTIR and Raman spectroscopy literature in forensic material identification — vibrational spectroscopy for polymers, fibers, paints, pure compounds; Raman's through-container and aqueous-sample strengths; limit: disentangling mixtures from overlapping spectra.
• General-audience treatments of forensic error and wrongful conviction (e.g., reporting on the Cameron Todd Willingham case; critiques of bite-mark evidence) — verify specific figures before relying on them.
• Geographic profiling and behavioral (case) linkage research — more quantitative subfields with firmer footing (journey-to-crime regularities; relative stability of signature vs. MO); existence and comparative footing attributed honestly; both produce prioritizations/consistency judgments, not certainties.
• Gisli Gudjonsson — clinical research on interrogative suggestibility and compliance; the basis for why the young, the intellectually disabled, and the highly suggestible are far more likely to yield to leading questions and shift answers under pressure; instruments/scores not cited.
• Guidance on reporting findings as likelihood ratios with standardized verbal scales (developed in Ch.9) — the interpretation movement recommending support for one proposition over another to resist the prosecutor's fallacy; §30.5; particular scales/conventions vary by lab and discipline. Attributed generally.
• History and critique of the numerical "point-counting" standard (8/12/16) and its abandonment for a holistic information-content judgment — documented field evolution; specific thresholds varied by country/agency.
• Ink-dating and ink-characterization literature — first-appearance dating (inks, toners, optical brighteners tied to known introduction dates) as a sound no-earlier-than boundary; chemical ink-aging (solvent loss, dye degradation) as destructive, constrained, and contested; the field's own admissibility/reliability debate attributed without citing specific protocols.
• Instrumental questioned-documents methods — infrared and infrared-luminescence imaging, ultraviolet examination, and the video spectral comparator (VSC); the optical principles (differential absorption/transmission/fluorescence of inks and paper treatments) are established physics; non-destructive, demonstrable outputs place them high on the spectrum.
• Investigative genetic genealogy methodology, agency policies, and emerging state law — database-access rules, oversight, and the ethics of §8.6 (consent at a distance, scope creep, database representation/equity).
• Investigative journalism and advocacy reporting on the national kit backlog — drove the national accounting and the legislative reforms (mandatory submission, inventory, tracking systems); role/general findings attributed without endorsing any single outlet's specific figure.
• Itiel Dror et al., experimental work on contextual bias in forensic examination — basis for the bias-cascade discussion; developed in Ch. 31.
• Legal scholarship, forensic commentary, and appellate litigation on the limits of historical cell-site analysis — that the data give a tower coverage area (several square miles or more), that phones do not always use the nearest tower (terrain, congestion, geometry), and that pinpoint testimony overstates the records; consensus and direction attributed without a specific paper; coverage-area magnitude is terrain/network-specific.
• Literature and a prominent court inquiry on the reliability/limits of low-template / low-copy-number (LCN) DNA typing for casework — basis for §8.2's caution.
• Literature and practice on conviction integrity units — which CIUs produce real exonerations vs. cosmetic ones, and the distinguishing features (independence, non-DNA cases, cooperation with the defense, transparency); attributed without a single pinned source.
• Literature on conflicts of interest and the partisan-expert / "hired-gun" problem — documents how party selection and payment shape testimony and debates remedies (court-appointed experts, disclosure); attributed in general terms, no effect-size figure invented.
• Literature on DNA-based geographic-origin testing of elephant ivory (associated with Samuel Wasser and colleagues; continent-wide reference maps of elephant genetic variation; tracing seized ivory to source regions; matching tusks across seizures) — existence, method, and influence attributed; specific figures kept qualitative, to be checked against primary publications before quoting.
• Literature on expert testimony about false confessions and courts' growing (if uneven) willingness to admit it — attributed generally; admissibility varies by jurisdiction.
• Literature on luminol false positives (bleach/oxidizers, metal ions, plant peroxidases) and luminol's potential to degrade DNA — attribute phenomena honestly; consult validation studies for exact conditions.
• Literature on secondary traumatic stress and burnout among forensic and first-responder professionals — documents the occupational mental-health toll of sustained exposure; phenomenon and the field's slow movement toward support attributed in general terms, no prevalence figure invented.
• mtDNA and Y-STR forensic methodology — hypervariable-region sequencing, maternal inheritance/heteroplasmy, Y-haplotypes, and the population databases used for match-frequency estimation.
• NIBIN/IBIS program documentation (ATF) and the operational lead-driven-workflow literature — NIBIN administered by the ATF; produces investigative leads requiring human confirmation; payoff depends on comprehensive, prompt entry of casings; attributed to program descriptions/casework literature; no specific hit-rate statistics cited.
• Personnel and education standards for forensic DNA analysts — national standards specifying defined coursework (biochemistry, genetics, molecular biology, statistics) and training/proficiency requirements; basis for §40.2 and Figure 40.2 rung 3; existence attributed without reproducing exact text.
• Popular and scholarly accounts of the "CSI effect" in the criminology and law literature (magnitude contested; consult multiple sources).
• Post-2009/post-2016 commentary assessing implementation of the NAS and PCAST recommendations — the basis for the qualitative grading in §38.3; precise adoption percentages deliberately not invented (they vary by reform, jurisdiction, source).
• Practitioner literature on expert testimony and surviving cross-examination — professional guidance on the honest-direct arc, conceding limits before cross, and managing the friendly examiner who invites overstatement (§30.2-§30.3). Attributed as a craft tradition.
• Probabilistic genotyping systems and the "black-box" debate — methodological and legal literature on software that models dropout/drop-in and computes likelihood ratios; proprietary source-code access disputes (developed in Ch. 9).
• Professional codes of ethics of the major forensic organizations (e.g., AAFS and the certifying boards) — real documents; described by their convergent core (objectivity, competence, honest representation, disclosure, service to the court) without verbatim quotation, since wording differs.
• Professional standards and literature of forensic document examination (ABFDE; relevant scientific-working-group / OSAC questioned-documents documents) — exemplar-collection method, the comparison process, and the nine-point conclusion scales (identification → inconclusive → elimination); attributed as a consensus apparatus; the verbal scales' uncalibrated thresholds are a recognized criticism.
• Professional/standards guidance on case assembly, forensic report writing, and expert testimony (SWGDAM, OSAC, report-writing guidance; Ch.30, 38) — evidence-log documentation, calibrated conclusions, avoidance of the ultimate-issue overstep; informs the Exercise 39.40 report template and Appendix I; attributed as standards practice.
• Published account(s) of dental identification in the SS Noronic disaster — contemporary and later writers documented the systematic use of dental records; existence of such accounts and milestone status attributed without a pinned citation or named individual practitioners.
• Quality-management literature on negative/positive controls and reagent blanks in forensic biology — conceptual basis for the catalog of controls.
• Reported U.S. Department of Justice internal guidance cautioning its own experts against overstating historical cell-site evidence — existence and thrust attributed in general terms; lesson does not depend on exact wording.
• Research and casework on secondary (and higher-order) DNA transfer — background for the "time of contact" limit and the transfer problem (developed Ch.7–8).
• Research associating the false-evidence ploy most strongly with false confessions, in both laboratory and case studies, and as the engine of the coerced-internalized type — attributed in general terms; effect sizes study-dependent.
• Research comparing actuarial / structured professional judgment with unstructured clinical judgment in violence risk prediction — structured approaches outperform intuition; all yield group-based probabilities; attributed as a consensus literature; specific instruments/figures to be checked against current research.
• Research debunking specific fire indicators — crazed glass from rapid cooling (water on hot glass) not accelerant; char "alligatoring" not indicative of speed/temperature/accelerant; concrete spalling as multi-cause; attributed as the established position of fire science.
• Research on adversarial allegiance in expert testimony — controlled studies finding that experts given the same materials but told they work for the prosecution vs. the defense shift their conclusions toward the retaining side, among practitioners who believe themselves objective; basis for §30.6 ("I am an objective scientist" = symptom, not defense); the payment-structure analog of the Dror bias experiments (Ch.31). Attributed as a consensus finding; no single study pinned.
• Research on bias in DNA mixture interpretation — found that interpretive judgments (number of contributors, allele calls; Chs. 8–9) can drift toward an expectation when the analyst knows the suspect's profile; basis for §31.5's claim that even strong methods need context management at interpretive steps, and for blinding mixture interpretation to the suspect's genotype.
• Research on cumulative/corroboration effects — how multiple weak items tracing to one source are perceived as independent corroboration; the basis for the "two pillars, one source" point; attributed generally.
• Research on examiner error rates and the subclass-characteristic problem in toolmark comparison — false-identification studies and documentation that same-process tools carry similar subclass marks; phenomenon and direction (subclass marks mimic individual agreement) attributed in general terms.
• Research on interrogation length as a predictor of false confession ("resistance is finite") — direction and strength attributed; exact average durations vary by sample.
• Research on maggot-mass self-heating — large larval aggregations generate metabolic heat and can run well above ambient; recognized error source (risk of overestimating the interval).
• Research on textile-fiber transfer and persistence — patterns of rapid early loss with a long tail in sheltered locations, fabric- and activity-dependence, and longer persistence on bodies/in protected spots; attributed without a pinned citation; specific percentages/half-lives treated as illustrative unless from a study of the relevant materials.
• Research on the cross-race effect (own-race bias) — robust, appears across groups and in young children (arguing against a prejudice explanation), over-represented in misidentification exonerations; finding and robustness attributed in general terms; specific effect sizes are population-dependent.
• Research on the limits and demographic biases of automated facial recognition — accuracy varies with image quality and across demographic groups; appropriate as a lead generator requiring corroboration, not proof; direction of findings attributed without a specific citation (ties to Ch.31).
• Research on the post-identification feedback effect and the confidence–accuracy relationship — confirming feedback inflates confidence and retrospective reports of view quality/attention; the more recent "pristine confidence" refinement (immediate, verbatim, fair lineup, pre-feedback confidence carries real information); both attributed in general terms; precise conditions/magnitudes debated.
• Research on weapon focus, stress and memory, and exposure-duration overestimation — attention narrowed onto a weapon away from the face; high arousal impairs (not "sears in") encoding; witnesses overestimate the duration of brief stressful events; attributed as consensus phenomena with directions of effect, not pinned citations.
• Research undercutting bruise "aging"/dating by color — basis for §37.4's caution against testifying to a specific date from appearance; attributed in general terms.
• Research/casework on the unreliability of visual recognition and personal-effects identification under disaster conditions — the documented phenomenon (decomposition + stress) and its role in early misidentifications in major disasters (§35.3, case study 35.2); attributed generally, specific episodes described as documented outcomes, not quantified.
• Saul Kassin & Lawrence Wrightsman — the voluntary / coerced-compliant / coerced-internalized typology; and the broader Kassin-and-colleagues experimental/review literature on how interrogation tactics produce false confessions; attributed as a consensus framework, no single paper cited; specific effect sizes vary by study.
• Scholarship and reporting on fire-science wrongful convictions — legal scholarship, investigative journalism, and innocence-organization analysis documenting discredited indicators contributing to wrongful arson convictions and fire-science's role in post-conviction relief; attributed in general terms; no contested specific claim repeated as fact.
• Scholarship on forensic-laboratory independence and structural bias — the consensus that embedding labs in law enforcement creates systematic, largely unconscious pressure toward the prosecution; the claim that embedded-lab errors tend to run in the prosecution's favor is reported across analyses and consistent with the documented scandals. Attributed; no single study cited.
• SEM-EDX and gunshot-residue literature — characterization of GSR particles by morphology and elemental composition (lead/barium/antimony) and automated search; "characteristic vs. consistent-with" distinction; contamination/transfer caveats (treated fully in Ch.24).
• Standard analytical-chemistry literature on gas chromatography, HPLC, and mass spectrometry — principles of chromatographic separation (mobile/stationary phase; retention time) and mass-spectral fragmentation (electron ionization; molecular ion; structure-specific patterns); attributed as consensus, no single text cited.
• Standard forensic-science / criminalistics textbook chapters on serology and bloodstain pattern analysis — mechanics of presumptive (Kastle-Meyer, luminol, LMG) and confirmatory (Takayama, Teichmann, precipitin) tests, and the geometry of impact angle and area of origin. Verify any specific sensitivities or dilution limits against a primary source.
• Standard introductory forensic-science survey texts' chapters on the nature and handling of physical evidence — the transient/conditional/pattern/transfer/associative categorization and the known/reference/control distinction.
• Standard practice for ignitable-liquid-residue analysis of fire debris (collection into clean metal cans/nylon bags; passive-headspace concentration; GC-MS; classification of ignitable liquids against a standard scheme; control samples for background pyrolysis products) — attributed as an existing standardized approach without quoting a specific standard number; detailed fire science owned by Ch.22, instruments by Ch.23.
• Statistical/legal literature naming and analyzing the prosecutor's fallacy and the defense fallacy (transposed conditional) in forensic identification evidence — attributed in aggregate.
• Stroke-sequence ("line-crossing") literature — when the order of two intersecting ink lines can and cannot be determined by microscopy/specialized imaging; the load-bearing point is that "indeterminate" is the honest report and certainty at every crossing is a red flag.
• Studies of contextual and confirmation bias in feature-comparison forensics (the Dror line of work) — knowing the "expected" answer shifts examiners' interpretations; the comparison microscope's side-by-side design can amplify expectation; existence and direction attributed here, formal treatment with named experiments deferred to Chapter 31.
• Studies of the biased-sample nature of DNA exonerations — DNA cases over-represent crimes leaving testable biology and under-represent those that do not, so the documented count is a floor, not a total; logic in §34.1 and Ch.6.
• Studies on BPA inter-analyst agreement, the effect of contextual information on BPA conclusions, and area-of-origin error including the parabolic-path correction — check specific study design and findings before quoting numbers.
• Studies/casework on the thermal behavior of teeth and dental restorations — teeth survive but become friable after intense heat; many restorative materials endure structure-fire temperatures; attributed in general terms (basis for the §17.1 careful-recovery point).
• Technical literature on AFIS / Next Generation Identification operation — minutiae extraction, candidate-list scoring/ranking, performance dependence on latent quality and on the system/database; attributed generally; the "returns a list, not an identification" point is robust.
• The "black-box" firearms-examiner error-rate studies referenced by PCAST and since — present examiners with known-ground-truth samples to measure error; attributed for their existence and general import (non-trivial error rates; "inconclusive" complicates interpretation; closed-set vs. open-set design affects apparent accuracy) without citing specific figures.
• The Guildford Four and Birmingham Six — real, publicly documented English wrongful-conviction cases whose coerced confessions helped prompt the shift to PEACE; summarized, not cited in detail.
• The Lindbergh kidnapping (1932) — the classic illustration of voluntary false confessions (an estimated ~200 confessors); the rough figure attributed as the standard account; exact number reported variously.
• The American Board of Forensic Odontology (ABFO) guidelines and scoring vocabulary — real; cited to show that a board, credentials, and standardized forms lent the appearance of standardization without supplying validation; the field's retreat from strong identification language attributed generally.
• The broader forensic-bias literature across multiple comparison disciplines — extends the contextual-bias finding beyond fingerprints and traces the specific channels (case narrative, suspect identity, confession, AFIS rank, order of presentation); empirical basis for domain-irrelevant information (§31.3) and the sequential-unmasking fix; the literature that named and anatomized the bias cascade.
• The clinical/forensic literature on non-fatal strangulation — dangerous with little/no external injury; reported peri-event symptoms most telling; delayed/internal injuries; strangulation as a risk marker for subsequent lethal IPV; well-documented, attributed in general terms; specific frequencies to be verified.
• The closed-set / set-based design critique — that validation studies forcing a pick from a known set overstate real-world accuracy where "none of the above" must be permitted (§15.4); recognized methodological point, attributed generally.
• The cognitive-forensics research program on context management and sequential unmasking — the procedural operationalization of the NAS bias recommendation; established program, protocols discipline- and lab-specific.
• The controlled fire-research and compartment-fire-dynamics literature — experimental basis for flashover, heat release rate of modern furnishings, and the fact that ordinary post-flashover fires produce the low/irregular floor patterns once attributed to accelerants; attributed as the field's consensus and the empirical foundation for NFPA 921's invalidation of the indicators; no single study pinned.
• The cryptographic-hashing literature, including determinism, the avalanche effect, practical irreversibility/collision-resistance, and the documented move from MD5 (engineered collisions demonstrated) to SHA-256 for forensic integrity — standard cryptography results; attributed in general terms.
• The debate over DOJ's response to PCAST — the fact of the disagreement is Tier-1; the broader propriety/consequences debate (incl. professional-organization defenses of long-used methods) is attributed generally.
• The deepfake-detection / synthetic-media research literature — generation methods (face-swap, full synthesis, voice cloning, lip-sync), detection approaches (artifact cues, ML classifiers), and core difficulties (generalization gaps, degradation under compression/re-sharing); central caution (real but unsettled, not yet Daubert-validated on novel fakes) attributed without overstating any method.
• The digital-forensics methodology literature on imaging, write-blocking, and hash verification (including guidance associated with NIST and law-enforcement digital-evidence working groups) — establishes the accepted bit-for-bit imaging + write-blocker + hash workflow; attributed as consensus practice, no single standard document pinned.
• The empirical literature evaluating criminal-profiling accuracy and the organized/disorganized typology — consensus that profilers do not substantially outperform comparison groups, that profiles contain many vague statements, and that the typology does not cleanly sort offenders; attributed in general terms, no single study cited.
• The estimator/system variable distinction and the eyewitness research program — the chapter's organizing framework; attributed as a foundational, widely credited contribution rather than to one paper.
• The EXIF metadata standard and the fragility of image/video metadata — EXIF/container/stream metadata can record device, settings, time, and (often) GPS; easily stripped (social-media re-compression) and editable; attributed in general terms.
• The experimental psychology of reconstructive memory and the misinformation effect — post-event information incorporated into memory and reported sincerely as original recollection (incl. the classic question-wording effect on reported speed/details); attributed as a consensus literature, no single study cited.
• The explosives-residue analysis literature and the inorganic/organic (low-/high-explosive) distinction; instrumental toolkit (ion chromatography; GC-MS/HPLC; Raman/FTIR; SEM-EDX) — attributed as consensus method; protocols laboratory- and case-dependent; contamination control part of the consensus.
• The false-confession literature and the non-independence of multiple coerced/internalized confessions (Ch.33) — voluntary vs. coerced-compliant vs. coerced-internalized; how interrogation pressure makes multiple accounts appear mutually corroborating when they are echoes of one source (the Beatrice Six mechanism); attributed generally.
• The footwear- and tire-impression literature and standards — documentation, casting, outsole/tread databases for make/model, assessment of randomly acquired characteristics; consensus that class determination is well grounded while individual ID depends on examiner judgment with a thinner error-rate literature; attributed without a pinned citation.
• The forensic and operational literature on mass-fatality management and DVI (forensic societies, disaster-management agencies, international organizations) — mass-fatality planning, morgue operations at scale, family-assistance centers, management of commingled remains; attributed as a consensus literature, no single document cited (basis for §35.1–§35.2 operational detail).
• The forensic and policy literature on familial searching and investigative genetic genealogy — kinship statistics, IGG methodology, and the privacy/consent/equity/Fourth Amendment debate; attributed as a developing literature; jurisdictional rules vary.
• The forensic fire-debris-analysis literature and ILR-identification standards — headspace extraction and GC-MS identification of ignitable liquids against reference patterns (instrumental detail in Ch.23); method and validation attributed generally; specific consensus methods/classifications maintained by standards bodies.
• The forensic firearms-and-toolmark identification literature and professional standards (e.g., AFTE-style theory of identification; comparison-microscope procedure) — real practice/method; attributed with its central critique (no fixed numerical "match" threshold; no well-established error rate for individual-source ID); no single study cited.
• The forensic glass-examination literature, including refractive-index population studies — immersion/GRIM measurement, density, elemental analysis; fracture mechanics for direction/sequence of breakage; the honest point that float glass clusters in a narrow RI band (weak association) attributed without a specific frequency.
• The forensic paint-examination literature and paint reference collections — layer-by-layer comparison; binder/pigment identification by FTIR, SEM-EDX, pyrolysis GC-MS; manufacturer-finish reference data; attributed generally, no specific database cited; any "how common" claim must rest on real data.
• The forensic-cognition research program of Itiel Dror and collaborators (mid-2000s onward) — the controlled experiments at the heart of case study 31.2; attributed honestly, no precise statistic reproduced; the signature within-examiner design re-presented working fingerprint examiners with their own prior comparisons under biasing context and found a meaningful portion reached different conclusions; establishes that domain-irrelevant context can change qualified experts' conclusions on identical evidence.
• The forensic-fiber-examination literature and standards (coordinated through bodies such as SWGMAT / OSAC and equivalents) — fiber identification and comparison by light and polarized-light microscopy, microspectrophotometry, and infrared spectroscopy, interpreted in terms of class membership, rarity, and number of independent transfers; methods and honest class-level posture attributed in general terms.
• The forensic-geology / soil-comparison literature (geoforensics) — comparison by color, mineralogy, particle-size distribution, density-gradient and other physical profiles, and biological components (pollen, diatoms, spores); centrality of distinctiveness and control sampling; limited validated population databases noted honestly.
• The forensic-microbiology / necrobiome literature on postmortem-interval estimation — microbial succession (body + soil) as a proposed PMI estimator parallel to insect succession; promise and limits (temperature/soil/moisture/individual variation; incomplete reference data and error rates) attributed without a pinned citation.
• The forensic-nursing / SANE-program evaluation literature — research documenting that SANE programs improve the quality and consistency of the medical-forensic exam, patient care, and (in many studies) evidence collection and case outcomes vs. the older ED model; existence/direction attributed, magnitudes vary and should be checked.
• The forensic-palynology literature and documented casework — pollen analysis helping establish that a body/object had been at a particular type of location or narrowing a search area; attributed with its limits (few practitioners, uneven reference data, sparse error-rate studies).
• The forensic-photogrammetry literature on reverse-projection height estimation — establishes reverse projection as the most defensible stature-from-surveillance method and documents its error sources (stand point, posture, footwear, camera change, image quality); attributed as consensus, no single paper cited.
• The forensic-toxicology literature on postmortem redistribution (PMR) — documents site-dependent postmortem drug concentrations, central (heart) blood artifactually elevated relative to peripheral (femoral) blood, and that lipophilic, high–volume-of-distribution drugs are most affected; attributed as consensus with the direction of effect, no single table cited; applied interpretation should rest on the actual drug's documented behavior and multi-site sampling.
• The general principle that class-evidence strength is probabilistic and scales with rarity (distinctiveness × number of independent matching features) — a consensus idea in the trace-evidence literature and the NAS/PCAST critiques; applied qualitatively because validated frequencies usually do not exist.
• The GSR contamination-and-transfer literature — secondary/tertiary transfer, occupational/environmental background, police-transport/booking-area contamination, and particle loss over time; phenomena and directional implications attributed in general terms, no magnitudes invented.
• The gunshot-residue analysis literature (SEM-EDX particle identification) — diagnostic primer-residue particle criteria (characteristic morphology + lead/barium/antimony elemental signature; plus newer lead-free primer chemistries) and SEM-EDX methodology; attributed as a consensus standard, no single citation.
• The image-/video-authentication (multimedia-forensics) literature — compression/JPEG forensics, error level analysis and its documented limitations (confounded by edges/text/texture; fails on re-saved/non-JPEG/screenshots), photo-response non-uniformity (sensor "fingerprint") source-device identification, and physics-based consistency checks (lighting, shadows, reflections, perspective); attributed as established knowledge.
• The immunoassay-screening and confirmation literature — antibody-based immunoassays for sensitive presumptive screening, cross-reactivity making them presumptive only, and the requirement for orthogonal confirmation (chromatography–mass spectrometry); attributed as standard practice; GC-MS mechanism deferred to Chapter 23.
• The law of spoliation and adverse inferences from destroyed evidence — a real, general feature of evidence law; specific standards/remedies vary by jurisdiction; attributed in general terms.
• The lineup-format literature (sequential vs. simultaneous) and the relative-vs-absolute judgment account — a genuinely contested literature (early reduction in mistaken filler IDs vs. later concerns about reduced correct IDs and order effects); attributed as a live, unresolved debate, deliberately not resolved in-text.
• The literature and practice of international humanitarian / human-rights forensics, including large-scale programs to identify the missing from armed conflicts — archaeological exhumation of mass graves, anthropological analysis of commingled/skeletonized remains, DNA-led kinship identification at scale (§35.5); achievements and honest limits attributed generally.
• The literature on demographic bias and validation gaps in forensic AI (especially facial recognition) — differential error rates across demographic groups and uneven/undisclosed validation; existence and direction attributed in general terms, no specific figure claimed.
• The literature on DNA identification of degraded, low-template, and commingled remains, and on kinship matching for mass fatalities — extraction from compromised samples/bone, statistics of partial profiles, likelihood-ratio kinship methods, and re-association of fragments (§35.3–§35.4); attributed in general terms.
• The literature on dye analysis and color measurement (microspectrophotometry) in fiber comparison — fibers indistinguishable to the eye can be excluded by differing absorption spectra; principal reason fiber comparison is more defensible than visual hair comparison; principle and evidentiary significance (exclusion decisive; agreement's weight scales with rarity) attributed without a specific study.
• The literature on injury interpretation — blunt, sharp, gunshot, and asphyxial wounds; defensible reads (range/direction of fire, incised vs. stab, patterned injury) and their limits (antemortem vs. perimortem; assault vs. fall; nonspecific asphyxia findings); attributed as a consensus literature.
• The literature on novel psychoactive substances (NPS) and fentanyl analogs and the resulting challenge to presumptive screening and reference data — attributed as a real, fast-moving phenomenon; no specific prevalence figures cited.
• The literature on postmortem alcohol formation (neoformation) and its interpretation — microbial fermentation can generate ethanol in a decomposing body; blood-alcohol from decomposed bodies must be interpreted cautiously; vitreous humor and microbial-activity markers help distinguish antemortem drinking from postmortem production; basis for the labeled composite in case study 20.2.
• The literature on postmortem changes and their confounders — temperature-dependence of rigor and decomposition, livor fixation timing, algor cooling behavior, and the scene factors (body size, clothing, surface, air movement, immersion) that perturb a PMI estimate; directions of effect attributed without a pinned citation.
• The literature on the variability/absence of genital and non-genital injury in sexual assault and the lack of an injury "signature" of non-consent — basis for the symmetric statement (presence consistent-with; absence not exculpatory; neither resolves consent); consensus attributed honestly.
• The mathematics and history of Benford's law (Newcomb 19th c.; Frank Benford 1938) and its audit/fraud-screening application (first-two-digit tests, goodness-of-fit) — real, well-characterized regularity; formula $P(d)=\log_{10}(1+1/d)$; the conditions of applicability and limits (§27.5) equally established.
• The metadata-reliability literature — timestamps depend on device clocks and time-zone settings, automated processes can alter access times, and metadata can be edited; standard cautions, attributed generally.
• The microbial-trace / skin-microbiome-transfer literature — distinctive shed skin microbiomes proposed for trace association (a touch-DNA complement); attributed as an active research direction, not established practice.
• The microcrystalline-test literature (characteristic crystal morphologies of target drugs; cocaine the classic) — attributed in general terms; reliability real but partly dependent on analyst judgment in reading crystal habit.
• The mobile-forensics extraction-tool literature and the "black-box" critique — proprietary tools that acquire/bypass locks on devices are real and widely used; the critique that their unauditable methods raise validity/disclosure problems and that output should be independently verified is a recognized concern (consistent with Ch.4 §4.4 and Ch.6).
• The net-worth method and the source-and-application-of-funds method for estimating unreported/illicit income — real, long-used investigative techniques; limits (genuinely unrecorded cash; unaccounted-for innocent explanations) attributed generally.
• The neurobiology-of-trauma literature on memory and the freeze response — acute stress impairs encoding (fragmentation, impaired sequencing, vivid central vs. absent peripheral detail) and produces involuntary tonic immobility; basis for "apparent inconsistencies and absence of resistance are expected, not probative"; consensus attributed, no pinned citation. Links to Ch.32 and Ch.33.
• The PEACE-model literature (England and Wales) and research comparing inquiry-based to accusatory interviewing — adoption and general research support attributed; specific evaluations not cited; comparative findings a recognized but developing literature.
• The peer-reviewed forensic-entomology literature on blow-fly (Calliphoridae) development and minimum-PMI estimation — establishes temperature-dependence of development and species-specific developmental (ADD) data; attributed as a consensus literature, no single table cited.
• The peer-reviewed forensic-odontology literature on dental identification of human remains — durability of teeth/restorations, antemortem-to-postmortem comparison, radiographic overlay; attributed as a consensus literature, no single paper cited; applied IDs should rest on the actual case records.
• The pharmacology of alcohol absorption/distribution/elimination and retrograde extrapolation — well-established (gut absorption, body-water distribution, ~zero-order elimination, back-calculation with assumptions and individual variability in elimination rate); attributed as consensus; the ~0.015/hour elimination rate in the exercises is labeled illustrative.
• The physiology of fire deaths — soot inhalation into the deep airways and carbon-monoxide/carboxyhemoglobin absorption as vital reactions distinguishing a victim alive during the fire from one already dead; mechanism attributed in general terms, specific levels case-dependent.
• The professional fraud-examination literature and practice standards (the anti-fraud body of knowledge associated with the Certified Fraud Examiner credential and periodic occupational-fraud studies) — the occupational-fraud taxonomy, the "detected by tips" finding, and the frequency-vs-cost inverse relationship; attributed as consensus, no single edition/statistic cited.
• The professional/policy critique of coroner systems and under-resourced death investigation — recurring failure modes (autopsies not ordered, non-physician certification, compromised independence, overwork, the national forensic-pathologist shortage); attributed in general terms; Case Study 11.2's concrete scenario is a clearly labeled composite.
• The psychological literature on confirmation bias and the bias blind spot — well-established general-cognition findings (humans seek confirming information and rationalize disconfirming information; people see bias in others while believing themselves immune); attributed in general terms and applied to the bench, not re-proved.
• The reassessment literature on bite-mark comparison (much post-2009) — examiner disagreement on whether an injury is even a bite, skin distortion, and the absence of a validated individualization basis; field trajectory and curtailment of claims attributed honestly without overstating any single study.
• The Reid technique as described by John E. Reid and Associates, and the scholarly critique that it is structurally prone to producing false confessions from the innocent — both method and critique attributed honestly; note the training community's own movement away from the accusatory model.
• The research program on cognitive/contextual bias in forensic science (associated with Itiel Dror and others; full treatment in Ch.31) — interpretations shift with domain-irrelevant case information; motivates context management, sequential unmasking, and blind verification (the §39.5 "most valuable further test"); attributed as a body of work.
• The role of investigative journalism in exposing forensic failure — well documented across cases (incl. the arson-science dimension of the Cameron Todd Willingham case, Ch.22/34, and crime-lab scandals) and in driving kit-testing reform; pattern attributed in general terms, consistent with the public record.
• The scholarly literature on the causes of wrongful conviction — consistently identifies the same contributing factors and their tendency to combine; attributed as a consensus literature, no single study pinned (registries above are the canonical data).
• The scholarship on combining/"linking" evidence in forensic reasoning (forensic statistics, the law of evidence, probability theory) — the basis for "independent items can be weighed together; dependent ones cannot," and for extending the likelihood-ratio framework from a single DNA result to a body of evidence; attributed as a consensus literature, no single text cited. The load-bearing idea (independent improbabilities multiply; dependent ones do not) is standard.
• The scientific critiques of bite-mark analysis (the formal reviews plus subsequent commentary by scientists and some former practitioners) — documents the lack of validated foundations (uniqueness not shown as expressed in a bite mark on skin; skin records detail poorly; examiners unreliable, including the threshold "is it a human bite?" judgment); proficiency-test direction (examiner disagreement incompatible with courtroom certainty) attributed without a specific table.
• The sequential-unmasking and context-management proposals in the forensic literature — the §31.5 procedures (document the evidence sample first; reveal context in controlled order, most-biasing last; verify blind) as a recognized set of reform proposals; specific implementations vary by discipline and lab.
• The stable-isotope literature on geographic provenancing of human tissues — H/O, C/N, S, Sr ratios reflecting water, diet, and geology and varying geographically; soundness and coarse, reference-map-dependent forensic inference attributed without a pinned citation.
• The standard forensic-chemistry literature on presumptive color-test reagents (Marquis; cobalt thiocyanate / Scott test for cocaine; Duquenois-Levine for cannabis; Mecke; Mandelin) — attributed as a consensus body of method; the specific reagent-color associations in §21.2 are illustrative, not a cited table.
• The standard forensic-pathology literature and reference texts on the autopsy procedure, cause/manner/mechanism certification, and time-of-death estimation from livor/rigor/algor and decomposition — attributed as a consensus professional literature; no single manual cited.
• The standard literature on forensic hair examination and its limits, with post-2009 professional/standards-body revisions — establishes what microscopic hair examination supports (human-vs-animal, body region, some treatment) and forbids (individualization; any population frequency); attributed as a consensus and as the basis for modern bans on the overstated language the 2015 review documented; no single guideline cited.
• The standard money-laundering literature and the placement–layering–integration model (with smurfing, shell companies, secrecy jurisdictions, AML/KYC, SARs/CTRs) — established professional/regulatory consensus; national rules and thresholds vary and change.
• The subclass-characteristics literature — that tooling shared across a manufacturing run imparts similar "individual-looking" marks to a batch of barrels/breech faces (§15.2, §15.4 failure mode); documented in the firearms literature; safeguard (asking whether agreement could be subclass) is standard guidance; attributed without a pinned citation.
• The therapeutic/toxic/lethal reference-range literature and the tolerance problem — published reference ranges exist and are widely used; attributed as population guides subject to wide individual variation, with tolerance shifting an individual's effective ranges; no specific numeric range asserted as fact (concentrations described qualitatively, e.g., "incapacitating range").
• The ultimate-issue doctrine and the proper scope of expert opinion — evidence scholarship and rules (and advisory commentary) distinguishing an opinion that embraces an ultimate issue (permitted) from one that usurps the jury (forbidden); §30.4; contours vary by jurisdiction (esp. mental-state opinions). Attributed generally.
• The validation/performance literature on rapid-DNA instruments — strong on clean single-source reference DNA, unsuited to complex evidentiary samples without lab interpretation; attributed as consensus, no single study cited.
• Toxicology literature on detecting toxic metals (e.g., thallium) by elemental methods (atomic-absorption spectroscopy; inductively coupled plasma techniques) — attributed in general terms in case study 23.1; the point is the class of instrument that reads composition, not a single named run.
• Trajectory-reconstruction methodology — rods-and-strings/laser approach, the cone-with-uncertainty output, and deflection/ricochet caveats (§15.1); established crime-scene-reconstruction practice, attributed generally rather than to a specific manual.
• U.S. Department of Justice / standards-body guidance discouraging "to a reasonable degree of scientific certainty" in forensic testimony — the direction of reform (away from the empty incantation) attributed in general terms; basis for the §30.4 Junk-Science Alert; no specific directive cited.
• Unconscious (mistaken) transference and the perils of repeated viewing — a face seen in another context (bystander, prior mugshot/lineup) misattributed to the crime; re-presenting the same suspect breeds false familiarity; attributed generally; central to both case studies.
• Validation studies of probabilistic-genotyping software (developer and independent), plus independent comparison studies showing different validated programs can return materially different LRs on the same complex mixture.
• Wildlife-forensics literature on species ID by DNA barcoding (mitochondrial marker regions) and morphological methods (hair, feather, bone, ivory Schreger lines) — consensus and the central limit (reference-database completeness) attributed generally.

Tier 3 — Illustrative / constructed (the recurring cold case and composite scenarios, labeled where they appear)

• All worked numbers in the chapter and exercises (RMPs, LRs of 1,000,000 / 2,000,000, prior-odds examples, four-locus multiplications) — illustrative teaching figures, not real-case values.
• Any round teaching figures used to dramatize the absence of a hair frequency (e.g., "1 in 50 vs. 1 in 50,000") — illustrative only, used to show that no validated hair frequency exists; not real statistics.
• ASCII diagrams of loop/whorl/arch, minutiae, and the AFIS workflow (§14.2, §14.5) — constructed, schematic, explicitly not-to-scale teaching figures.
• Case Study 20.2 — the postmortem-alcohol composite (the decedent, apartment, and 0.16 blood alcohol) — a labeled composite scenario constructed after the documented neoformation problem; underlying science real (Tier 2), particular case invented and flagged.
• Case Study 40.1 ("A Year at the Bench," composite analyst "Dana") — a LABELED COMPOSITE; constructed representative portrait, not a real person; assembled from documented patterns (backlog, testimony, trauma exposure, embedded-lab structure). Fictional in particulars, honest in pattern.
• Figure 11.1 ("One body, three different words") — a constructed teaching example separating cause, mechanism, and manner on a hypothetical gunshot death; illustrative, not a real case.
• Figure 13.1 ("Who is on the body") and the §13.2 succession timeline diagram — constructed teaching examples; timing explicitly illustrative and not-to-scale; true cast/schedule are region- and season-specific.
• Figure 15.1 ("A fired bullet's rifling impressions") and Figure 15.2 ("What a fired cartridge case records," incl. the base-of-case ASCII schematic) — constructed teaching examples, explicitly not-to-scale; caliber, mark positions, and degree of correspondence chosen to make the class-vs-individual distinction visible, not from a real exhibit.
• Figure 16.1 ("Recovering a three-dimensional footwear impression") and the ASCII negative-to-positive-cast schematic — constructed teaching diagram, explicitly schematic and not to scale.
• Figure 17.1 ("Two charts of one upper jaw") — constructed teaching example of an AM/PM comparison; includes the #5 explainable-discrepancy illustration and Universal-system tooth numbers; illustrative, not a real chart.
• Figure 17.2 ("The reconciliation board") — constructed teaching example, after general DVI practice; ~40-victim reconciliation matrix, PM-/AM- labels, the PM-014 row; illustrative and not-to-scale; real reconciliation governed by the Interpol framework (Ch.35).
• Figure 18.1 ("Fluent writing vs. a drawn forgery") and Figure 18.2 ("How a vanished page leaves its ghost") — constructed teaching examples, explicitly not-to-scale; the schematic names what to look for, not a real comparison or recovery.
• Figure 18.3 / Figure CS18.1, the Lindbergh "Read the Evidence" blocks — labeled [after the Lindbergh case, public record]: the case facts are real and public, but the six-field rendering is a teaching reconstruction; the "$9,500 → $95,000" infrared figure is illustrative of the physical principle, not a claim about a specific real document.
• Figure 19.1 ("A single hair under the comparison scope") and the §19.5 comparison-microscope ASCII diagram — constructed teaching examples; schematic and not-to-scale; the descriptors ("dark brown, medium diameter, fragmented medulla") are illustrative class characteristics, not data from a real specimen.
• Figure 20.1 ("One measured number, two different stories") and the worked BAC arithmetic in §20.4 / exercises (illustrative 0.10, 0.12, ~0.015/hour, 2-hour interval) — illustrative round numbers for transparency; not reference values.
• Figure 21.1 ("A swab from the crater rim") and the reagent-and-color examples in §21.2 — constructed teaching examples; the nitrate result, clean controls, and reagent-color associations are illustrative, chosen to show the role of controls and the limits of a color test; not reference values.
• Figure 22.1 ("The 'pour pattern' that wasn't") and the phases-of-a-room-fire ASCII diagram (§22.1) — constructed teaching examples; temperatures/timings labeled illustrative and not to scale; the flashover range $500\text{–}600^{\circ}\text{C}$ given as an approximate commonly cited figure, not a precise threshold.
• Figure 23.1 ("A gasoline chromatogram") and Figure 23.2 ("A mass spectrum") — constructed, not-to-scale teaching diagrams; peak positions, m/z values (e.g., m/z 91 base peak, m/z 106 molecular ion for an illustrative alkylbenzene), and relative heights are illustrative, chosen to make the reading skills transparent; not reference values.
• Figure 25.1 ("Two hashes, one verdict") — constructed teaching example; the 256-GB drive, abbreviated 3f7a...e91c hash, and eight-month window are illustrative; not a real hash value.
• Figure 25.2 ("What a cell sector really covers") — constructed teaching schematic, explicitly not to scale; the wedge, tower, cabin, and two-mile-distant home are illustrative geometry chosen to make the area-not-point lesson visible; real coverage areas vary with terrain/network.
• Figure 26.1 ("Two enhancements of the same blur") and its worked plate/character details — constructed teaching example illustrating admissible enhancement vs. fabrication; the "six-pixel" figure and legible characters are illustrative, not reference values.
• Figure 28.1 ("MO versus signature at three scenes") and Figure 28.2 ("Barnum statement vs. specific, testable prediction") — constructed teaching examples; the three burglary-assault scenes, the shoe-arranging "signature," and the paired statement columns are illustrative/invented; no real case or real predictions depicted.
• Figure 30.1 ("The same finding, two witnesses, one cross-examination") — constructed teaching example; the two witnesses, the Q-and-A, and the "about a million times more probable" figure are illustrative, isolating the words-chosen-on-direct variable; not a real transcript.
• Figure 31.1 (the bias cascade) and Figure 31.3 (sequential unmasking vs. the contaminated workflow) — constructed schematics distilling the forensic-cognition literature into a flow and a contrast; schematic, not to scale, not data.
• Figure 31.2 / Figure CS31.1 ("reading the same latent two ways" / "one judgment, four coats") — rendered after the Mayfield case, public record, but presented as teaching reconstructions of how an ambiguous feature can be read two ways and how the chain of agreements formed; underlying case facts are Tier 1, the figure framing is a teaching device.
• Figure 32.1 ("Two ways to run the same lineup") and Figure 32.2 ("The same certainty, two very different values") in index.md, and the §32.4 best-practice lineup sequence diagram — constructed teaching examples, labeled in-text; the witnesses, suspects, and confidence statements are illustrative, not from any real case.
• Figure 33.1 ("Two processes, two purposes") and Figure 33.2 ("How a false confession acquires its 'guilty knowledge'") — constructed teaching diagrams; the example facts ("struck from behind," "back door unlocked," "a green gas can") are illustrative, not from any real case.
• Figure 34.1 ("The wrongful-conviction cascade") and Figure 34.3 ("The path to exoneration, and where it stalls") — constructed teaching diagrams; schematic, illustrative, not depicting any single real case.
• Figure 34.2 ("Two pillars, one source") — constructed teaching example labeled "after the exoneration pattern"; a composite distilling the recurring structure across many real cases, standing for none of them.
• Figure 35.1 (Interpol DVI phase workflow) — schematic rendering of the real five-phase framework, simplified for teaching.
• Figure 35.2 (the reconciliation matrix) and the ~180-victim event — constructed teaching example ("after general DVI practice"); the numbers are illustrative, not data.
• Figure 36.1 ("Reading speed from the road") and the §36.4 accident-scene sketch — constructed teaching examples; measurements (e.g., the 24 m skid) explicitly illustrative and not-to-scale; real reconstructions use surveyed distances, an established friction value, and EDR data.
• Figure 38.1 ("Two ways to organize a crime laboratory") and Figure 38.1b ("The letterhead and the org chart") — constructed teaching diagrams/examples (dependent vs. independent structures; reading structural bias off documents); schematic, not to scale; the submission-form narrative is invented to make the mechanism visible.
• Figure 38.2 ("Is this forensic result trustworthy?" — the reform checklist) — a constructed decision aid synthesizing the book's lessons into five questions; a teaching tool, not an official agency protocol.
• Figure 40.1 (careers map) and Figure 40.2 (education→certification ladder) — constructed organizing diagrams; Figure 40.1 is not any real agency's org chart; Figure 40.2 timescales are typical, not guaranteed, and shown for a lab analyst (other paths differ).
• Figure 6.1 ("The result that reopened a closed case") — constructed teaching example modeled on the documented DNA-exoneration pattern; labeled as such.
• Figure 9.1, "What the software is weighing" — constructed schematic of a two-person mixture at one locus.
• Figures 24.1 ("A paint chip in cross-section") and 24.2 ("Reading a broken pane") — constructed teaching examples; the five paint layers + buried blue repaint and the radial/concentric fracture + edge stress marks are explicitly illustrative and not-to-scale.
• Figures 29.1 ("What the tissues remember"), 29.2 ("The confident algorithm"), 29.3 (two-routes-to-a-relative schematic) — constructed teaching examples; the "96%" similarity score and "99% accuracy" in Fig. 29.2 are illustrative round numbers (NOT real performance figures), chosen to make the score-vs-validation distinction transparent.
• Figures 3.1 ("Two kinds of evidence that vanish"), 3.2 ("What the layering tells you"), and 3.3 ("Same stain, two packages, two fates") — constructed teaching examples, labeled.
• Figures 37.1–37.3 — the kit chain-of-custody diagram, "A neck that looks unremarkable," and "An account that doesn't run in a straight line"; constructed teaching examples (clinical patterns drawn from the general literature; specific scenarios invented, not real cases).
• Figures 8.1–8.4 — constructed schematic visuals: degraded single-source EPG; two-person mixture locus; the "Read the Evidence" gas-can profile (8.3); the IGG-vs-CODIS diagram (8.4).
• Illustrative latent-print error-rate figures (~1 in several hundred to ~1 in a thousand) and the "roughly half" forensic-contribution proportion — presented as orders of magnitude, not precise citations.
• Illustrative numbers — any round figure (relative weakness of partial profiles, rough scale of mtDNA match frequencies) is illustrative and labeled; real values from Tier 1/Tier 2 sources.
• Illustrative stain measurements in the angle-of-impact worked examples and exercises (e.g., 4 mm × 8 mm) — round teaching numbers, labeled illustrative.
• Note: the Krone (Figure 16.2) and Brown facts are Tier-1 public record; Figure 16.2 is labeled "[after Ray Krone, public record]" because it renders a real case in the book's "Read the Evidence" teaching format — the facts are real; the formatting is the teaching device.
• Pattern-family proportions in §14.2 (loops ~60%, whorls ~35%, arches ~5%) — illustrative round figures; real proportions vary by population and finger; labeled as such in-text.
• Qualitative strength language in worked examples ("strongly supports," "uncommon," "five-layer agreement") — chosen to make the probabilistic logic transparent; not reference frequencies; a real report assesses rarity qualitatively unless a validated database exists, and never invents a number.
• Round teaching figures in the exercises and text (300-deaths-in-a-county scenario; 180-victim matrix) — illustrative numbers chosen to make the reasoning transparent. The approximate real death tolls and the WTC remains order-of-magnitude in the case studies are drawn from the public record and flagged as approximate.
• Schematic ASCII diagrams (serology funnel, single-stain geometry, stringing/area-of-origin method) — constructed for instruction, explicitly not to scale.
• The "[N] minutiae" / "14-point match" figures in testimony examples and exercises — illustrative placeholders; the point is quality/configuration over count, not any specific number.
• The "chain of harm" reconstruction in case study 21.1 — explicitly a labeled reconstruction of the documented field-test pattern (field test → arrest → plea before confirmation → later lab finding of no controlled substance), not a specific person's file.
• The chapter epigraph ("The expert is a teacher, not an advocate…") — a labeled constructed teaching line distilling guidance to new forensic witnesses; not a verbatim quotation.
• The Cold Case File (Mill Creek / Marcus Diallo / Cody Renner) — the book's fictional running case; the false-confession beat introduced this chapter is constructed; dissected in Ch. 33.
• The Cold Case File (Mill Creek / Marcus Diallo) — constructed running investigation; the gas-can heat-degraded mixture and the genealogy lead are labeled teaching constructs (see Appendix I; capstone Ch. 39).
• The Cold Case File — Mill Creek / Marcus Diallo (this book): the Ch.3 evidence inventory (gas can, tools, charred documents, pry-marked door, phone, cartridge case). Constructed; fictional.
• The Cold Case File: the Mill Creek / Marcus Diallo investigation (constructed teaching example, used across all chapters; no relation to any real case).
• The composite chapter epigraph ("I confessed because I was tired…") — a labeled constructed line, after the recurring testimony of DNA exonerees; not a quotation from any single person.
• The composite job posting in the §40.4 "Read the Evidence" block — assembled from common features of real entry-level crime-lab listings; not a reproduction of any single posting.
• The constructed GSR-contamination illustration in case-study-02 (falsely-positive suspect contaminated by an officer, patrol car, and booking area) — explicitly labeled constructed; contamination routes are real and documented, the specific scenario invented and tied to no named case.
• The courtroom exchanges modeled in the chapter and exercises — constructed dialogues, not transcripts.
• The exclusion matrix and convergence diagram as rendered — constructed teaching diagrams; schematic, not-to-scale; the honest verbs in each cell (excludes / consistent with / strongly supports) are the teaching content, not the fictional facts.
• The generic descriptions of the imaging/hashing workflow and the logical-vs-physical extraction hierarchy in §25.2 and §25.4 — presented to make procedure intelligible; real casework follows current validated agency procedure with documented tools and versions.
• The illustrative "honest vs. overstated" testimony phrasings in the index, key-takeaways, and quiz — constructed model language to teach the honest-verb discipline; not quotations from any real proceeding.
• The illustrative composite in Case Study 11.2 (elderly decedent; missed homicide presenting as accidental fall) — each failure mode documented and real, but the specific decedent and sequence invented and labeled as such, to avoid asserting facts about a real private individual's death.
• The illustrative convergence multiplication in §39.3 (1-in-10, 1-in-20, 1-in-30, 1-in-5 → 1-in-30,000) — illustrative round numbers, included ONLY to make the structure of independent-improbability multiplication visible, with an explicit warning never to present such a figure to a jury as measured. Not data.
• The illustrative height ranges used throughout (e.g., "approximately 178–184 cm," and the exclusion/overlap examples) — illustrative round numbers chosen to demonstrate the exclusion-vs-consistent-with distinction; not reference values.
• The illustrative measurements in the chapter — e.g., the "approximately 19 mm wide" pry bar (§16.4) and the ~30-minute cure/dimensions in Figure 16.1; illustrative round numbers to make the reasoning concrete; not reference values.
• The illustrative round figures and worked phrasings throughout (e.g., ".38/9 mm class," "six lands and grooves, right twist") — illustrative, chosen for transparency; real casework rests on the actual measured characteristics of the actual items.
• The Mill Creek / Marcus Diallo cold case and the mixture LR for Roy Keller, and the detective's prosecutor's-fallacy claim — fictional teaching constructs internal to this book.
• The Mill Creek cold case (the §40.6 reflection and Case File) — constructed teaching material; this chapter adds NO new facts and does not restate the Ch.39 capstone conclusion; reflects only on which real professions performed each step; persons of interest invented.
• The Mill Creek cold case in its entirety — the exclusion matrix and convergence diagram (§39.2–39.3), Figure 39.1, the Case File, the report template (Exercise 39.40), Appendix I — constructed teaching material built across all forty chapters; Marcus Diallo, Roy Keller, Dana Whitfield, Victor Salas, Cody Renner are invented; the conclusion is stated at "strongly supports," never "proves," to model the discipline. No Mill Creek "finding" is a real result.
• The Mill Creek cold case — Diallo's and Keller's phones, the recovered deleted messages, and the cell-site finding (the Case File, Appendix I); constructed teaching material; persons of interest invented; "Keller's alibi breaks ≠ Keller did it" used as an honest-framing device. Does NOT reveal the §5.4 solution.
• The Mill Creek cold case — forensic-engineering electrical-system exclusion and all associated facts (the Case File, the cold-case exercises, Appendix I); constructed teaching material; persons of interest invented; no person named by this evidence.
• The Mill Creek cold case — GC-MS confirmation of gasoline in the fire debris and the SEM-EDX sleeve-particle work (Figure 23.3, the Case File, Appendix I); constructed teaching material; persons of interest invented; no person attached to the particle.
• The Mill Creek cold case — pollen-on-the-vehicle-mat scenario, the cabin's "distinctive flora," the uncommon-plant component, and all associated facts (Figures 13.2–13.3, the Case File, Appendix I); constructed teaching material; persons of interest invented.
• The Mill Creek cold case — soil-on-Roy-Keller's-boots scenario, the cabin's distinctive soil and uncommon mineral/pollen combination, the glass item, the negative GSR result (Figure 24.3, the Case File, Appendix I); constructed teaching material; persons of interest invented; §5.4 solution not revealed.
• The Mill Creek cold case — the "lone drifter/stranger" profile, its pointing away from Roy Keller, and all associated facts (Case File, Appendix I); constructed teaching material; persons of interest invented; the case's solution deliberately NOT revealed.
• The Mill Creek cold case — the cabin door's pry marks (a class match to a flat pry bar → forced entry), the partial shoe impression in the soil (logged as class evidence), and the absence of bite marks (the Case File, Appendix I); constructed teaching material; persons of interest invented and no suspect named at this stage.
• The Mill Creek cold case — the cabin fire's "multiple origins + ignitable-liquid (gasoline) pattern" arson finding, Figure 22.1's cold-case framing, the Case File checkpoint, and Appendix I; constructed teaching material; persons of interest invented; the GC-MS confirmation deferred to Ch.23 by design.
• The Mill Creek cold case — the cabin's fire debris, the scene gasoline odor, the handheld-detector indication, and the status "accelerant indicated — gasoline (confirmation pending)" (the Case File and §21.5); constructed teaching material; persons of interest invented; no instrumental confirmation or arson conclusion asserted (deferred to Ch.22–23).
• The Mill Creek cold case — the Case File and the gas-station CCTV, doorbell camera, and alibi-video facts (including the "internally inconsistent metadata" of the alibi video); constructed teaching material; persons of interest invented; used to model "consistent with" (not "identified") and the metadata asymmetry.
• The Mill Creek cold case — the Case File checkpoint and the use of Cody Renner as the chapter's wrongful-conviction near-miss; constructed teaching material consistent with Chapters 11, 25, 7–9; persons of interest invented; chapter exonerates Renner and stops short of naming who the science points toward (Ch.39's task).
• The Mill Creek cold case — the Case File's re-reading of the "accidental fire" frame as domain-irrelevant context, the gas-can latent callback, and the workbook caution (Appendix I); constructed teaching material; persons of interest invented; the chapter adds no new physical evidence, only a caution applied to the old evidence.
• The Mill Creek cold case — the Chapter 35 Case File comparing single-victim (Diallo, Ch. 17) vs. mass-fatality identification; constructed teaching material; explicitly an aside adding no new evidence; persons of interest invented; nothing revealed/foreshadowed about the capstone.
• The Mill Creek cold case — the charred insurance/partnership document, its indented writing, and the indicated altered beneficiary (the Case File, Appendix I); constructed teaching material; persons of interest invented; the chapter deliberately declines to name who altered the document.
• The Mill Creek cold case — the Diallo-homicide trial, the DNA analyst's and fire investigator's modeled testimony, the gas-can mixture LR, and all persons of interest (the §30 Case File; Appendix I); constructed teaching material; persons of interest invented.
• The Mill Creek cold case — the eleven-hour Cody Renner interrogation (Figure 33.3, the Case File, Appendix I): the first-several-hours-unrecorded detail, the false-evidence ploy, the minimized "it got out of hand" theme, and his exclusion by the autopsy timeline (Ch.11) and cell-site records (Ch.25); constructed teaching material; persons of interest invented.
• The Mill Creek cold case — the gas-can DNA mixture, the rapid-DNA/IGG confirmation that the minor contributor is NOT an unknown stranger, and the "stranger theory excluded" status (Case File, cold-case exercises); constructed teaching material built on the frozen Ch.7–8 facts; persons of interest invented; an EXCLUSION, not an identification.
• The Mill Creek cold case — the gas-can partial latent, the AFIS candidate list including Keller, linear-ACE-V-with-blind-verification handling, and the inconclusive outcome (Figure 14.1 context, the Case File, Appendix I); constructed teaching material; persons of interest invented.
• The Mill Creek cold case — the insurance policies naming Roy Keller (property + life), his debts, the Benford-flagged renovation books, and all associated facts (the Case File, Appendix I); constructed teaching material; persons of interest invented; delivers the assigned "strong financial motive" beat without revealing guilt.
• The Mill Creek cold case — the Marcus Diallo autopsy (absent airway soot, low carboxyhemoglobin, perimortem skull fracture, manner amended to homicide), Figure 11.2, the Case File, and Appendix I; constructed teaching material; persons of interest invented; the "trauma vs. heat artifact" question deliberately deferred to Chapter 12.
• The Mill Creek cold case — the Marcus Diallo dental identification, the "someone else's body" insurance/fraud theory, the antemortem chart and bitewing/panoramic films, and all associated facts (Figure 17.1, the Case File, Appendix I); constructed teaching material; persons of interest invented.
• The Mill Creek cold case — the neighbor's "tall stranger's truck," the quarter-mile sightline, the nighttime view, and all associated facts (the §32 Case File and Appendix I); constructed teaching material; persons of interest invented; used to practice discounting a confident lead on witnessing-conditions grounds.
• The Mill Creek cold case — the sedative-at-an-incapacitating-level finding, the modest BAC, the specimen set (peripheral blood, vitreous, urine, gastric), and all associated facts (the Case File, Appendix I); constructed teaching material; persons of interest invented; used to state a "state" inference (chemically incapacitated) at true strength while refusing to name a perpetrator or call the sedative the cause of death.
• The Mill Creek cold case — the single recovered head hair, the small number of common dark trilobal polyester fibers on the victim's shirt, their association with a jacket "later associated with Roy Keller," the rootless hair routed to mtDNA, and all related facts (Figure 19.1, the Case File, Appendix I); constructed teaching material; persons of interest invented. The fiber is deliberately written as common (a weak association) to contrast with the rare-fiber logic of case study 19.2.
• The Mill Creek cold case — the small non-independent county lab, the early accidental-fire frame, and the counterfactual reforms (blind verification, sequential unmasking, accreditation) applied to the case (Case File, Appendix I); constructed teaching material; persons of interest invented; the checkpoint explicitly does NOT change the case conclusion (capstone, Ch.39).
• The Mill Creek cold case — the stray 9 mm cartridge case, its low-confidence NIBIN correlations, and the autopsy interaction (no gunshot wound → red herring) (Case File, Appendix I); constructed teaching material consistent with the frozen facts (inventory in Ch.3; autopsy in Ch.11); persons of interest invented.
• The Mill Creek cold case — the years-old DV report involving Dana Whitfield and Marcus Diallo, its review under evidence-from-the-living principles, and the resulting status ("Dana excluded; scope clarified"); constructed teaching material consistent with the frozen facts (Dana excluded by alibi + DNA non-association); persons invented.
• The Mill Creek cold case: the doorframe stain, the low spatter pattern, and Figure 10.3 — constructed teaching material (see _continuity.md §5 and Appendix I).
• The organized/disorganized typology diagram in §28.3 (presented as classically claimed, then critiqued; labeled illustrative, not an endorsement) and the paraphrased Barnum personality sketch in §28.5 (representative paraphrase of the genre, not a verbatim quotation of any instrument).
• The reverse-projection (§26.2) and ELA (§26.4) ASCII diagrams — constructed schematics, explicitly not to scale.
• The round dollar amounts in worked examples and callouts (e.g., a $500 approval threshold, a $500,000 policy) — illustrative, chosen for clarity; real casework uses the actual figures.
• The schematic family tree in case study 29.1 — a generalized reconstruction of the IGG triangulation logic, labeled as such; the specific DeAngelo genealogy is not reproduced.
• The three ASCII schematics (three-questions ladder §20.1; BAC curve §20.4; therapeutic/toxic/lethal band and PMR diagram §20.5–20.6) — constructed teaching diagrams; explicitly illustrative and not to scale.
• The verb ladder (excludes / consistent with / strongly supports / proves) — teaching device introduced in Ch.1, used throughout.
• The worked accumulated-degree-day numbers in §13.3 and the exercises (e.g., 130/150 ADD, a 10 °C threshold, the 10/15/22/50-day results) — illustrative round numbers chosen to make the arithmetic and temperature-sensitivity transparent; not reference values.
• The worked Benford figures in §27.5 — Figure 27.1 uses the REAL expected first-digit percentages (~30.1% for 1 … ~4.6% for 9) and is reliable; Figure 27.2's specific counts (n=200; deficit of 1s; 4/5 spike) are illustrative round teaching numbers, not real-case data.
• The worked heat-release-rate descriptions (e.g., a foam sofa at "one to two megawatts") — illustrative orders of magnitude to convey that modern furnishings burn intensely; not precise measured values for any specific item.
• The §11.2 autopsy order-of-operations diagram and the §11.4 postmortem-timeline schematic — constructed teaching diagrams; the livor/rigor/algor timing bands are explicitly illustrative and not-to-scale, not reference values.
• The §17.3 ASCII three-numbering-systems figure (Universal / FDI / Palmer for the same upper teeth) — constructed teaching aid; schematic.
• The §36.3 failure-analysis fault tree ("Why did this load-bearing connection fail?") — constructed teaching diagram; generic branches and tests, not drawn from a specific case.