Chapter 19 — Key Takeaways

A scannable one-page card. For the full argument and the worked examples, see index.md.

The core claims

• Trace evidence is Locard made physical. Hairs, fibers, glass, paint, soil, and residues are the residue of contact (Chapter 3): two surfaces meet and trade material. Trace is defined less by what it is than by how little of it there is and how it got there — a transfer, usually invisible, that the analyst must find, recover, and interpret.
• Transfer and persistence govern everything. Transfer is movement of material between surfaces on contact (primary = direct; secondary = through an intermediate surface/person); persistence is how long it stays. Secondary transfer is the quiet ruin of simple "contact" stories — a trace on a suspect need not have come from the victim directly (the same problem as touch DNA, Chapter 8).
• Almost all trace is class evidence. It narrows the field; it rarely points to one source. The rare exceptions are physical (jigsaw/fracture) matches — a torn edge or a glass fragment that fits back into the break — which can be near-individualizing because they reconstruct a unique break. Absent that, trace says "consistent with," not "came from."
• Hair, by visual exam, tells you a lot about the hair and almost nothing about the person. Microscopy reliably sorts human vs. animal, often body region, and some treatment/condition — all class characteristics shared by enormous numbers of people. There is no validated frequency for any combination of hair features, so "consistent" has no denominator, cannot become a likelihood ratio (Chapter 9), and cannot honestly support more than cannot be excluded.
• Microscopic hair comparison was the discipline done wrong. A real but weak observation — "these two hairs share class characteristics" — was, for decades, dressed in the language of a match, a number, or a placement of the defendant at the scene. In April 2015 the FBI admitted that, in the large majority of reviewed pre-2000 cases, its examiners had exceeded the limits of the science; the errors implicated dozens of defendants sentenced to death. DNA exonerations (Tribble, Odom, and others) forced the reckoning.
• Fiber evidence is the discipline done right. Fibers are class evidence too, but their analysis rests on objective, reproducible measurement — polymer type, cross-section, diameter, and above all dye chemistry/color (microspectrophotometry records color as a full spectrum). Differing spectra exclude cleanly. Strength comes from the value of the rare and from multiple, independent (ideally two-way) transfers — not from any single fiber "matching."
• The comparison microscope is the chapter's emblem. Two scopes on one optical bridge, two specimens in one split field. Superb for excluding (a strength for fibers, firearms, toolmarks) and treacherous for confirming a hoped-for hair association, because it is built to find agreement and a motivated observer will find it. The tool was never the problem; the claims made with it were.
• DNA is what lets a hair point to a person. Microscopy's honest modern job is triage — human? rooted (nuclear STR, Ch. 7) or rootless (mtDNA, Ch. 8)? worth testing? — routing hairs to DNA, not rendering a comparison verdict.

The method-validity verdict (NAS 2009 / PCAST 2016)

Where they sit: fiber comparison is class evidence done honestly, well above the discredited methods but well below single-source DNA (Chapter 7), because no fiber individualizes. Microscopic hair comparison, for tying a hair to a person, sits near the bottom — above only bite marks (Chapter 16) — because it has no measured error rate and no population frequency, and its strong forms were a documented source of overstated testimony. The only thing that lets a hair point to a person is DNA.

What you can honestly say on the stand

• Hair (visual exam): "This hair is human, a head hair, naturally shed, dark brown, medium diameter. These are common characteristics shared by an unknown but large number of people. The person of interest cannot be excluded as a possible source — along with many others — and no statistical weight attaches to that. I cannot say the hairs 'match,' are 'identical,' or came from the same person." (Or, if there is a real difference: "...the person is excluded as the source.")
• Fiber (the cold case): "The questioned and known fibers are indistinguishable in polymer type, cross-section, diameter, and color. This is a common fiber type and color; it is class evidence, manufactured in quantity, so I cannot say it came from the defendant's jacket to the exclusion of others. The jacket cannot be excluded as a source, along with a large number of other dark polyester garments. The weight is low because the fiber is common and there is a single fiber type, not a two-way exchange of rare fibers."
• What you must NOT say: that a hair "matches" or is "identical to" the defendant; any number or near-number for a hair association (it would be a fabrication — no frequency exists); that a hair, on its own, "places" the defendant at the scene; or that any single fiber came from one garment "to the exclusion of all others."

Key terms (one line each)

• Trace evidence — physical evidence in small/microscopic quantity, transferred by contact; Locard's principle made physical.
• Hair morphology — the microscopic structure of hair (cuticle, cortex, medulla) and its features; yields class characteristics, not an individual source.
• Microscopic hair comparison — visual comparison of a questioned and known hair; supports exclusion, cannot identify a person; its strong forms were found scientifically unsupported (2015 FBI review).
• Fiber — the smallest unit of a textile, natural or manufactured; class evidence whose value rests on objective measurement, rarity, and number of transfers.
• Transfer/persistence — transfer = movement of material between surfaces on contact (primary/secondary); persistence = how long it stays; together they govern recoverability and meaning.
• Comparison microscope — two scopes on an optical bridge presenting two specimens side by side; powerful for exclusion, biased toward finding agreement when the examiner knows the answer.

The cold-case line

The fibers and the hair add a faint thread to the weave, not a rope. The fibers are consistent with a jacket associated with Roy Keller but are a common type — class evidence that cannot exclude him along with many other dark polyester garments. The rootless hair carries no statistical weight and is routed to mtDNA (a non-exclusion of limited weight, not an identification). Status: trace consistent with a known source, excludes no one we needed to keep. In 1995 this hair might have "matched" Keller — and that sentence would have been a lie. The difference is the whole chapter.

The themes this chapter advanced

• Exclusion over proof — trace's surest voice is "no": a single decisive difference excludes; agreement only narrows. Hair's honest ceiling is non-exclusion; fibers' is "cannot exclude a common source."
• The validity spectrum — the chapter's widest split: honest fiber comparison (objective measurement, above the discredited methods) versus microscopic hair comparison for ID (near the bottom), both far below DNA. Same microscope, opposite ends.
• (Also advanced: cognitive bias — the FBI hair scandal as systemic bias at the most prestigious lab, the comparison microscope as a bias amplifier, §19.3 and §19.5; and the CSI effect — a hair on a body "sounds like proof," §19.2.)