Chapter 3: Physical Evidence

Welcome to Last Minute Lecture.

This free chapter overview is designed to help students review and understand key concepts.

These summaries supplement, not replace, the original textbook and may not be redistributed or resold.

For complete coverage, always consult the official text.

Okay, let's unpack this.

If you are preparing for a meeting, a class, or just want to catch up on the essential

solving.

You know that the real work happens in the laboratory.

We are diving deep into the core mechanics of forensic science.

Not the courtroom drama, but the actual painstaking analysis of physical evidence.

We're talking about the silent witness at the scene.

Exactly.

The one piece of evidence that can't lie, can't forget, or be subjected to memory bias.

That's absolutely right.

Our mission today is to really guide you through the entire life cycle of physical evidence.

We need to understand what even counts as evidence, which is broader than you might think.

Then the meticulous process forensic scientists use to analyze it, which really boils down to two critical pillars.

Identification and comparison.

Identification and comparison.

Got it.

Finally, how the digital age has completely revolutionized the entire process.

I mean, it's shifted the crime lab from, say, a passive facility that just waits for suspects.

Right.

Waiting for the police to bring them something.

Precisely.

To an active, proactive engine of investigation linking crimes and suspects all over the world.

And to anchor this whole discussion, I think we have to begin with a powerful,

if you know, a pretty grim case study that just perfectly illustrates the collective power of physical evidence.

That's the investigation into the Toronto landscaper, Bruce McArthur.

This case is a profound lesson,

a truly profound lesson in how numerous seemingly small pieces of evidence, when they're meticulously collected and analyzed,

can build an unassailable mountain of proof.

And this was a long time coming.

Oh, for a decade.

For a decade, men were disappearing from Toronto streets.

It was a pattern that finally cracked in 2017 with the disappearance of Andrew Kinsman.

So what was the break?

What happened with Kinsman?

Well, his friends and the police were just combing through his life and a crucial initial link emerged.

And it was so simple.

That initial link was staggeringly simple.

It really was.

Investigators discovered a single critical entry in Kinsman's personal calendar.

Just the name Bruce.

Just Bruce.

That's it.

At the same time, surveillance footage from the area captured Kinsman entering a very specific vehicle moments before he vanished.

It was a red 2004 Dodge Caravan.

So you have a common first name and a specific relatively older vehicle.

That still feels like a needle in a haystack.

You'd think so.

But when the police combine those two pieces of information and cross -reference them, Bruce MacArthur, the landscaper, immediately surfaced as the sole person fitting that exact profile in the area.

Wow.

The investigation just pivoted instantly.

And that pivot led to a dramatic arrest, right?

It did.

The police, fearing for the safety of a young man they actually saw entering MacArthur's apartment, moved in very, very quickly.

OK.

And once MacArthur was detained, the subsequent searches of his properties and, crucially, that van, it yielded a staggering collection of physical evidence.

Prosecutors later called it an unprecedented amount of real forensic, digital, and documentary evidence.

Let's focus on the physical evidence found there because it really demonstrates the range of what a forensic team has to handle.

The van itself was a gold mine.

I mean, just a complete gold mine.

They recovered blood that matched the DNA profiles of some of the known victims.

Right there.

That's huge.

Then they found a black duffel bag in his bedroom.

And inside was, well, what we would call the instruments of the crime.

Like what?

Rolls of duct tape, rope, various zip ties, a surgical glove, a bungee cord, and syringes.

These are all items that function as class evidence until they're linked by use or some unique manufacturing detail.

And then there was the most chilling detail, the trace evidence that MacArthur himself kept.

Yes.

This is really something.

The investigators discovered that MacArthur had this meticulous, almost ritualistic practice of shaving his victims' heads and facial hair before he buried them.

And he kept it.

Unbelievably, he kept bags containing the hair and fibers he had collected.

You know, in his mind, he was probably erasing evidence, but in reality, he was preserving the unique trace evidence that would ultimately seal his fate.

So the perpetrator, through this act of just chilling fastidiousness, was unknowingly curating the very evidence that would tie him completely and inexorably to the victims.

That is precisely the irony of the MacArthur case.

And faced with their undeniable collective mountain of physical evidence, I mean, from the DNA in the van to the trace hair he had stored, MacArthur pled guilty to eight counts of murder in 2019.

It really reinforces that central idea of forensic science.

It does.

It is rarely one single spectacular piece of evidence, but the entire accumulated chain of seemingly small, mundane physical links that tells the comprehensive story.

That brings us perfectly to our core concepts for today.

We need to understand how scientists define and use these links.

So as we go, we'll define the two primary goals of the lab.

Identification that's determining what something is, and comparison determining if two things share a common origin.

And for comparison to work, we have to grasp the crucial difference between individual characteristics and class characteristics, which really dictates the certainty of a match.

And let's start by just defining the sheer scope of this.

I mean, before any analysis can even happen, the investigator has to know what they're looking for.

When you think of a crime scene, I think most people, you know, their minds go to DNA swabs and dusting for fingerprints.

Sure, the CSI effect.

Exactly.

But the reality is just so much more expansive.

Every scene is unique.

Yeah.

It requires innovative on the spot decisions.

That's right.

But based on decades of forensic history, we know there are common categories of physical evidence that, when you examine them scientifically,

consistently yield significant results.

We're talking about maybe 21 common categories.

21.

That's a lot.

It is.

So instead of just hitting you with a long list, let's maybe group these into categories based on what they transfer or what they impress upon the environment.

OK, that makes sense.

Because at the end of the day, the primary function of physical evidence is always linking.

Linking a person to a place, a person to a person, or an object to a person.

OK, so let's start with what's probably the most sought after category,

the biologicals and fluids.

This is the realm of serology and biochemistry, and it covers, you know, the obvious things, blood, semen, and saliva.

And that's whether it's liquid or dried.

Liquid or dried, and also whether it's animal or human in origin.

That's a key first step.

What's interesting, though, is it's not just the large, visible stains, it's the residues.

What do you mean by residues?

Think of saliva left on a discarded cigarette butt, or on the seal of an envelope or a postage stamp.

Those tiny amounts of saliva allow for full DNA recovery.

And we also include organs and physiological fluids in this group.

These are typically submitted for specialized toxicology, maybe to detect the concentration of alcohol and blood, or the presence of poisons or specific drugs in, say, the liver or stomach contents.

OK, that's a big category.

Next, let's consider evidence that involves impressions and tool marks.

This is where objects are physically altering the scene.

Exactly.

And this is a very, very broad category.

It includes everything from fingerprints, both latent, meaning invisible and visible, which are, you know, the bedrock of individualization, the classic.

But it also includes large -scale impressions left by vehicle tires, shoe prints, or even just depressions left in soft soils.

You mentioned tool marks.

Right.

That's the smaller, more subtle marks.

The specific impression of, say, a crowbar on a door jamb is known as a tool mark.

And we even look at the often overlooked impressions left by fabric patterns or gloves.

And within the same sort of impression category, we also have firearms and ammunition, right?

Absolutely.

This involves the field of ballistics.

So the weapon itself, any discharged or intact ammunition.

And we also analyze powder residues.

So the stuff that comes out of the gun that is in the bullet.

That's it.

Items containing the residue left by a firearm's discharge.

Analyzing the density and the pattern of these residues on, for instance, a piece of fabric can help determine the distance the gun was fired from.

Can you describe that for us?

Sure.

Imagine a white piece of fabric as a target.

If you fire from a foot away, you'll get a certain circular pattern of powder stains.

If you fire with the muzzle pressed right against it, you get a totally different, much denser, sometimes star -shaped pattern.

I see.

Comparing that pattern to a victim's clothing is a key step in reconstructing exactly what happened in an incident.

That's a powerful visualization for a listener.

OK, now let's move to trace and transfer evidence.

This is the stuff that transfers silently between objects and people.

Love should exchange principle in action.

Exactly.

Every contact leaves a trace.

And this is probably the most diverse group.

We are looking for fibers, both natural, like cotton or wool, and synthetic, like nylon or polyester.

They're excellent for establishing relationships.

OK.

Similarly, hair, whether it's animal or human, is constantly being transferred.

Then you have glass particles or fragments that transfer to people or clothing.

And this includes analyzing window panes to determine the direction of force when a projectile, like a bullet, passes through.

What about chemical and structural trace materials?

That brings in things like paint liquid or dried chips, which are incredibly common in collision or hit -and -run cases,

petroleum products like gasoline residues and a suspected arson, or maybe grease and oil stains,

and then soil and minerals.

This can be incredibly powerful.

How so?

Well, the specific mineral content of soil can be unique to a very small area.

So dirt embedded in the tread of a shoe or specialized safe insulation found on a burglar's garment can powerfully link a person to a very precise location.

And the last one in this group.

Wood and other vegetative matter, things like fragments of sawdust or unique leaves or pollen found on tools or clothing.

And finally, we have a handful of specialized evidence categories that require really specific lab processes.

Right.

These include documents, and that's anything involving handwriting, typewriting, the type of ink, the paper authenticity.

This can even mean restoring serial numbers on stolen property where they've been scratched off.

That's amazing.

What else?

We also look at manufactured goods like plastic bags.

They're very common in drug packaging or tragically in homicides, and they can sometimes be compared against a roll of bags found in a suspect's inventory.

And you can even examine vehicle lights.

Yes.

A forensic scientist can examine the filament of a light bulb to determine if it was hot, meaning the light was on or cold, meaning it was off at the exact moment of impact.

Wow.

It's an incredibly comprehensive view, and it really confirms that almost anything, truly anything can serve as that silent witness.

So once all this evidence is collected and carefully transported to the crime lab, the forensic scientist gets to work.

And regardless of the material, be it a fiber, a powder or blood stain, the analysis is undertaken for one of two overarching procedural goals,

identification or comparison.

Let's start with identification.

This is answering that really foundational question.

What is this substance, chemically and physically?

Exactly.

Identification aims to determine the physical or chemical identity of a substance with, well, with as near absolute certainty as existing analytical techniques will permit.

Where do we see this most often?

Most commonly in drug identification, you know, determining definitively that a seized powder is cocaine and not just sugar, or in analyzing fire debris to identify gasoline residues or even determining if biological fluids are human or from an animal.

And the process for reaching this certainty involves two pretty rigorous steps.

Can you tell us about them?

The first step is all about standardization.

The lab must adopt testing procedures that consistently yield characteristic results for specific known standard materials.

You can't identify an unknown without a known baseline to compare it to.

Right.

You need a reference sample.

You have to have a reference.

And the second step is where the pressure comes in, I think.

This step demands you run sufficient tests to eliminate all the alternatives.

All of them.

All of them.

The second step requires that the number and type of tests be comprehensive enough to exclude all other known substances from consideration.

The examiner has to devise an analytical scheme that eliminates every single possibility but one.

So if you conclude that the fire debris contains, say, dynamite.

Your tests must be so thorough that you have definitively eliminated TNT, plastic explosives, and every other potential combustible material as the source.

That sounds incredibly subjective, though.

I mean, how could a scientist ever guarantee they've eliminated all other possibilities?

There's no definitive universally published rulebook for that, is there?

That is the essential challenge.

You're right.

There are no simple standardized rules that dictate exactly how many tests define a thorough and foolproof analytical scheme.

So what do they rely on?

The scientist has to rely heavily on their specialized knowledge, their education, and their years and years of experience to judge the point at which the analysis is concluded and the criteria for positive identification are satisfied.

And that judgment call has to hold up in court.

And that ultimate conclusion has to be solid enough to be substantiated beyond any reasonable doubt in a court of law.

It's a synthesis of hard science and professional judgment.

OK.

So once identification is satisfied, we know what the substance is.

The next natural question is, where did it come from?

And that brings us to the second pillar,

comparison.

Yes, comparison analysis is the linking function.

It subjects a suspect specimen and a standard or reference specimen like, say, hair from a suspect's head to the exact same test to ascertain whether they share a common origin.

It's all about creating that linkage.

We see this visually all the time in movies and shows.

You imagine the standard side -by -side microscopy comparison of two hair samples, one from the scene and one from the suspect.

That's the perfect visualization.

And the comparison procedure itself is a two -step process.

Step one involves selecting a suitable combination of properties or characteristics from both the suspect and the standard specimen.

And those properties depend on what you're looking at?

Entirely.

The properties chosen, color, shape, chemical, composition depend completely on the material being examined.

And step two is drawing the conclusion.

Yes.

And it works like this.

If one or more of the selected properties disagree, the scientist immediately concludes the samples are not the same and could not have originated from the same source.

That's a definitive exclusion.

OK.

That's straightforward.

But here is the critical distinction.

If all the properties compare and the specimens are, for all intents and purposes, indistinguishable, we cannot automatically conclude they came from the same source.

Why not?

I mean, if they look exactly alike under the microscope, why can't we just assume they're a match?

Because that is where the concept of probability enters the discussion in a huge way.

To assess the true evidential value, we have to appreciate probability, the frequency of occurrence of an event.

Or flipping a coin.

Exactly.

If we find two items that are indistinguishable, we must ask, what are the odds of finding that identical combination of characteristics occurring randomly somewhere else in the world?

That statistical reality dictates the quality and the weight of the comparison.

That transition takes us directly into what I think is the most important part of this.

Defining the difference in quality between a comparison that points to a group versus one that points to a single source.

The difference between individual and class characteristics.

Right.

And individual characteristics are the forensic gold standard.

They're what everyone is hoping for.

This is evidence that can be associated with a common source with an extremely high degree of probability.

How high are we talking?

We're talking about a level of certainty that virtually defies mathematical calculation or even human comprehension.

It's practical proof of common origin that's been substantiated by massive real -world experience over decades.

So what are the best examples of this near -absolute certainty?

Well, fingerprints are the classic example.

The French scientist Victor Balthazard famously calculated the probability of two individuals having the same fingerprint ridge characteristics as one out of one multiplied by 10 to the power of 60.

That's a one with 60 zeros after it.

An unimaginable number.

It's a staggering number, and it's supported by over a hundred years of practical experience classifying millions upon millions of prints without ever finding two identical sets from different people.

It's the practical experience that really solidifies the probability.

Okay, so beyond fingerprints, what else reaches this level of individuality?

We look for random patterns, things that are created by chance.

This includes the random striation markings left by the barrel of a firearm on a bullet, or the unique scratches left by a tool, like a screwdriver, on a surface.

And wear and tear.

Absolutely.

The irregular and random wear patterns we find in shoe or footwear impressions.

The way the rubber on your shoe wears down is specific to your weight, your gait, your habits, the environments you walk in.

And one of the most powerful demonstrations of individuality have to be the physical fit, right?

Where an object literally fits back together like a jigsaw puzzle.

It is visually irrefutable.

I recall a case where a knife blade tip was found embedded in a victim's wound.

When investigators found the broken knife on the accused, a close examination revealed that the irregular edges were the knife broke.

Like a puzzle piece.

Exactly like a puzzle piece.

Not only that, but the individual, minute characteristics of the sharpening scroffria on the blade, all of it matched up perfectly.

That physical match leaves no scientific doubt of common origin.

There's a similar and I think fascinating phenomenon with mass -produced items,

particularly sequentially made plastic bags.

You wouldn't think they'd be individual.

You wouldn't.

But even though they are mass -manufactured products, the machinery itself introduces minute individualizing characteristics.

So imagine a black trash bag covering a victim's head compared to a roll of trash bags found in a suspect's possession.

Okay.

If those bags were manufactured consecutively on the same machine, they may share identical, unique extrusion striation marks and pigment bands.

That's rare.

Incredibly rare.

And this occurrence allowed investigators in one case to focus their attention on one specific suspect who was later convicted.

It just proves that even disposable plastic items can possess individual characteristics.

But the vast majority of evidence retrieved from a crime scene does not possess this unique signature.

Most of it falls into the other category, class characteristics.

That's correct.

Class characteristics are properties of evidence that can be associated only with a group and never with a single source.

And here, probability isn't a near impossible number.

It's the determining factor that defines the size of the group.

Can you give us the standard common examples of class evidence?

Sure.

A single layer automobile paint chip, say a common metallic blue,

that color and chemical composition can only be associated with maybe thousands of cars of that specific make and model.

It narrows the field significantly, but it doesn't individualize the source to one car.

Or blood type.

Or blood type, a perfect example.

If two samples are found to be human, type A, well, that's a group association.

Type A blood occurs in about 26 % of the population.

That's a very large group.

So if class evidence is only associated with a group, how does it ever become powerful enough to carry real weight in a courtroom?

It gains its significance through a mathematical concept called the product rule.

The idea is simple mathematics applied to forensic science.

If you have a series of independently occurring characteristics, say three different blood factors, you multiply their individual frequencies together to get an overall and much lower frequency of occurrence for that specific combination.

This logic was dramatically demonstrated long before modern DNA in really high profile cases like the O .J.

Simpson trial using older blood analysis factors.

That's a perfect illustration of the concept.

A blood stain at that crime scene was found to contain three independent blood factors.

Factor A, which is found in about 26 % of the population.

Factor ESD found in 85%.

And Factor PGM21 found in only 2 % of people.

So when we apply the product rule, multiplying those independent probabilities, the number just shrinks dramatically.

It does.

You multiply 0 .26 by 0 .85 by 0 .02, and that calculation yields 0 .00044 or just 0 .44%.

So statistically, only about 1 in 200 people would be expected to have that exact combination of blood factors.

That's right.

It's a huge leap in specificity.

Now, it didn't individualize the stain to O .J.

Simpson alone, but it provided powerful data for assessing the evidential value in court.

And crucially, it also eliminated the two victims, Nicole Brown Simpson and Ronald Goldman, as sources of that specific blood stain.

It shows how the collective power of multiple class characteristics really works.

Exactly.

And it's vital for you to note that this is the same mathematical principle that underpins modern DNA analysis.

We now test so many independent genetic markers that the cumulative probability, what's called the random match probability, is infinitesimally small.

We have enough factors to individualize biological materials to a single person, with the only exception being identical twins.

So while the product rule really shows the power of class evidence, we have to be realistic that most evidence retrieved possesses class characteristics.

And this brings us to a major challenge in the field.

The inherent weakness that forensic scientists often cannot assign exact or even approximate probability values to the comparison of most class evidence.

You mean things like paint chips or fibers or glass?

Why don't comprehensive statistical databases exist for those materials?

I mean, why can't we just look it up?

In a society that is just increasingly reliant on mass production, gathering that data is overwhelming.

How often does a specific blue nylon fiber from a certain brand of sweater appear?

When was it manufactured?

How many items were made?

It's almost impossible to track.

So there's no central database.

Few statistical databases exist for these mass produced products, unlike DNA or fingerprints where we have massive controlled databases.

So in the absence of hard statistics, the forensic scientist, when they're interpreting the significance of a class match, has to rely heavily on personal experience.

That reliance on experience.

It sounds like a scientific vulnerability.

And yet, class evidence is still immensely valuable.

So where does its strength lie?

Its value lies in two major areas.

First is corroboration.

Physical evidence binds together other, often subjective, investigative findings in a manner that's free of human error and bias.

So it backs up eyewitnesses?

It corroborates subjective evidence eyewitnesses, confessions, or informants.

These subjective accounts can be disputed or attacked in court, but physical evidence offers this non -human, unbiased thread that links events and people.

And the second major power is exclusion or exoneration.

And this is a critical safeguard in the justice system.

Physical evidence can definitively exclude a person from suspicion.

If type A blood is found at the scene, anyone with type B, A, B, or O blood is eliminated, period, regardless of circumstantial factors.

This is a foundational role of the forensic lab.

We mentioned the heavy reliance on the examiner's subjective experience.

This naturally leads us to a big issue in the field.

The conflict regarding forensic integrity and cognitive bias.

How susceptible are practitioners to confirmation bias when they're interpreting class evidence?

Studies confirm that forensic science practitioners are, like all humans, highly susceptible to confirmation bias.

If an analyst receives an ambiguous hair sample and is told, this hair came from the main suspect, that context can unintentionally sway their evaluation.

The analyst may subconsciously interpret any ambiguity in favor of the linkage that was suggested to them by the police.

So if the scientific community recognizes this risk, what are the recommended mitigation procedures to increase objectivity and protect the integrity of the results?

The procedures really focus on blinding and replication.

Labs should implement protocols to reduce an analyst's access to unnecessary investigative information.

So they don't know who the suspect is?

Right.

The analyst only needs to know that sample A is from the scene and sample B is the reference, not who sample B belongs to.

They should also use multiple comparison samples, rather than relying on a single suspect exemplar.

And what about replication?

Crucially, labs should encourage replication of results by a second analyst who is completely blinded to the original findings and the original analyst's conclusion.

It's a check and balance system.

That whole shift toward greater objectivity was formalized, wasn't it?

I remember a big report came out expressing significant concern about these subjective evaluations.

Yes.

In 2009, the National Research Council, or NRC,

published a highly influential report, Strengthening Forensic Science in the United States, a Path Forward.

What does it say?

The report brought widespread attention to the fact that many forensic determinations, especially those involving pattern analysis, rely heavily on subjective evaluation.

And the correctness of those evaluations is not always verifiable.

This dependence on an examiner's personal training and experience was identified as a core weakness.

The NRC encouraged research to put forensic science on a more objective, statistical footing.

But what were the immediate recommendations for labs facing these constraints right now?

The immediate focus shifted entirely to rigorous quality assurance measures.

This includes things like mandatory peer review of results, consistent proficiency testing to ensure competency,

strict analyst certification requirements, and periodic external audits of the entire laboratory operation.

So institutional safeguards.

Exactly.

These institutional measures are designed to control and standardize the subjective elements that are just inherent in comparison analysis.

But despite all the concerns over subjectivity,

the sheer power of collective class evidence can be overwhelming, right?

It can override any statistical uncertainty about single items.

That concept is best embodied by the Wayne Williams case in Atlanta.

Williams was charged with two murders and linked to 10 others primarily based on the sheer volume of transfer evidence.

The state's case rested heavily on fiber evidence.

And the numbers in that case were absolutely staggering.

They were.

Investigators found 28 different types of fibers, all technically class evidence linking Williams to the various murder victims and the scenes.

28 different types.

Yes.

The forensic examiner at the time accurately characterized that volume of evidence as overwhelming.

When you take the low probability of finding one item randomly and you multiply it by the low probability of finding a second, a third, and eventually a 28th, the cumulative certainty of involvement becomes extremely high, even if no single fiber provided an individual match.

Now let's talk about the judicial challenge.

When this evidence gets to court, who determines the ultimate weight or significance of it?

That responsibility falls squarely on the trier of fact, which is usually the jury of laypeople.

And this is where the tension becomes acute.

So given the high regard in which scientists are generally held and the almost infallible image of forensic science created by popular media, the CSI effect again, exactly, there is a serious risk that scientifically evaluated evidence can take on an aura of special reliability and trustworthiness in the courtroom.

So the jury might give the evidence too much weight simply because it came from a lab and a scientist said it.

Precisely.

This potential prejudice against the accused is why proper safeguards are essential, ensuring the jury understands the limitations and the difference between statistical probability and absolute certainty.

And finally, let's touch on a very practical issue, natural variation.

Modern technology is so sensitive that it often produces too much information.

How does a criminalist define the limits of natural variation?

This is a constant professional judgment call.

How many color layers individualize a paint chip?

How many striations individualize a tool mark?

The problem is that modern analytical techniques have become so sophisticated they can distinguish between samples that are practically identical from an investigative perspective.

Can you give us a vivid example of that kind of scientific oversensitivity?

Think about fragments of glass.

Techniques are now so sensitive they can often distinguish glass particles originating from within a single pane of glass.

Wow.

From the same window.

From the same window.

That level of minute or relevant detail goes far beyond the criminalist's goal, which is simply to determine if two particles came from the same window source.

And we see this with magnification too, I imagine.

We do.

If you look at a two -layer paint chip under moderate magnification, say 244 times, you meaningful relevant details for comparison.

But if you increase that magnification dramatically, maybe 1 ,600 times, you reveal microscopic ultrafine details that could not be duplicated anywhere else on earth.

So at that level, nothing would ever match.

Exactly.

At that extreme, no two chips, even from the same car, would ever compare.

Practicality dictates that the examination has to be conducted at a less revealing but ultimately more meaningful magnification.

And distinguishing significant evidential variations from irrelevant natural variations is one of the hardest judgment calls, which is why textbook knowledge is ultimately no substitute for years of experience.

Okay, here's where it gets really interesting for me.

For decades, the crime lab was the passive arm of the investigation.

It waited for police to bring in a suspect and a known reference sample for comparison.

Right.

But the widespread adoption of computer technology and mass databases has dramatically reversed that role, moving the crime laboratory to the forefront of investigations by seeking to identify perpetrators proactively across jurisdictions.

This is the digital revolution of forensic science, without a doubt.

Instead of saying, does sample A match suspect X, the lab can now ask, does sample A match anyone in our massive archive?

Let's look at the major systems, started with the original model for cross -jurisdictional comparison, fingerprints.

The system that pioneered this was IAFIs.

The Integrated Automated Fingerprint Identification System.

IAFIs was the premier national fingerprint and criminal history system managed by the FBI.

Since 2014, it's been integrated into the Expanded Next Generation Identification, or NGI, system.

Yeah.

How big is NGI?

NGI currently contains fingerprints and criminal history for nearly 75 million subjects.

That's close to 750 million individual fingerprint images in the database.

That's incredible.

So how does an investigator translate a murky, latent print from a crime scene into a usable database search?

So the latent print examiner first captures a high -resolution digital image of the print.

Then they use a specialized coder to mark selected ridge characteristics, you know, where ridges end, where they bifurcate the specific loops and whorls.

So they're basically creating a digital map of that fingerprint.

That's a perfect way to put it.

The system then submits that map to the NGI database.

Within minutes, powerful algorithms search against all 75 million subjects and provide the examiner with a list of potential candidates for manual comparison and verification.

But the final call is still human.

Always.

The computer just speeds up the triage process exponentially.

And the power of NGI isn't just speed.

It's the archival depth.

The Gerald Wallace case illustrates how a cold print can wait decades for a match.

A very sobering story.

In 1975, Gerald Wallace was murdered, and investigators found a lone fingerprint lifted from a cigarette pack.

They had no match, no leads, so the print was just filed away.

For 16 years.

For 16 years.

In 1991, once the Pennsylvania State Police established their state AFIS system, which linked to the FBI's national database,

that single print was entered.

It produced an immediate hit.

Wow.

It identified a man who had been present at the house that night and who subsequently led police to the individual ultimately charged with the murder.

It just demonstrates the immense long -term archival value of these systems.

Yeah, perhaps the most impactful database of the modern era, the one that provides individual certainty.

CODIS.

The Combined DNA Index System, or CODIS, became fully operational in 1998.

It serves as the FBI's backbone for the electronic exchange and comparison of DNA profiles.

It's an incredibly powerful system because it links crimes to each other and, crucially, links crimes directly to convicted offenders.

And CODIS is organized around three structural indices, correct?

Yes.

First, you have the Forensic Index.

This holds approximately 915 ,000 profiles recovered from unsolved crime scene evidence.

This index allows police to identify serial crimes linking an assault in one city to a homicide in another.

Okay.

Second is the Massive Offender Index, containing over 13 .6 million convicted individuals.

And third is the Arrestee Index, containing about 3 .4 million profiles collected from individuals awaiting trial, though state statutes determine who gets swabbed and when.

The scale of CODIS' success really speaks for itself.

It does.

It has produced over 451 ,000 hits, assisting in more than 440 ,000 investigations.

It works so effectively because, statistically, repeat offenders commit a disproportionate amount of crimes that involve biological evidence.

An essential legal turning point for CODIS was the 2013 Supreme Court decision in Maryland v.

King.

What was the implication of that ruling?

That case sanctioned the collection and analysis of a cheek swab or a buckle swab from an arrestee who was detained for a serious offense.

And the Supreme Court said that was okay.

They did.

The court ruled that collecting DNA was comparable to traditional police booking procedures, like fingerprinting, and was reasonable under the Fourth Amendment.

And that legal change dramatically increased the pool of profiles available for the Arrestee Index.

And that need for immediate comparison drove the development of RapidDNA.

Exactly.

RapidDNA refers to specialized instruments that can develop a CODIS -compatible DNA profile from a buckle swab in less than 90 minutes.

So you can do it right at the police station.

That's the vision.

These compact, commercially available instruments are now envisioned to take their place right alongside fingerprinting units in police stations, allowing immediate comparison against the vast offender and forensic indices while the arrestee is still in custody.

The story of the center -city rapist Troy Graves shows how CODIS just completely ignores geographical boundaries.

It's a perfect example of the proactive linking capability.

Trey Graves committed a string of eight sexual assaults in Fort Collins, Colorado in the early 2000s.

When his profile was entered into CODIS upon his arrest, he was immediately, irrevocably identified as Philadelphia's notorious center -city rapist.

And that was a cold case.

A very cold case.

He was responsible for a series of assaults and the unsolved murder of a graduate student, Shannon Schieber, years earlier, 1 ,800 miles away.

DNA technology closed that case instantly, proving that distance is no longer a shield.

Now when CODIS fails to produce a match, investigators have recently started venturing into genealogy databases.

This is a relatively new frontier.

It is, and it involves what's called familial searching.

If a crime scene profile doesn't match anyone in CODIS, investigators can search through commercial genealogy databases like GEDmatch, often populated by profiles uploaded by users of companies like 23andMe and Ancestry .com -to -identify -a -close -relative.

So the goal isn't to find the suspect, but to identify a family lineage.

Exactly.

The process identifies a common family tree, which investigators then painstakingly examine to pinpoint individuals who match the person of interests, description, age, location, and so on.

It narrows a search from millions to a handful of individuals in a specific lineage.

It's proven to be a massive game changer.

A complete game changer in cold cases.

Okay, so beyond biological evidence, specialized national databases have been developed for key types of class and individual evidence.

Let's start with ballistics and Nibin.

The National Integrated Ballistics Information Network, Nibin, is managed by the ATF.

Its purpose is to allow firearms analysts to acquire, digitize, and compare the unique markings, the striations and impressions that a firearm leaves on bullets and cartridge casings.

And the technology capturing those images is IBIS.

Right.

The Integrated Ballistic Identification System, or IBIS, captures 2D and 3D images of the casings and forwards them to a regional server.

It then correlates these images against the database and produces a short list of candidates that may have been fired by the same weapon.

So again, it's not a final match.

It's crucial to understand that Nibin does not declare a definitive match.

That still requires a manual side -by -side comparison and verification by a trained firearms examiner.

It's a lead generation tool.

And that process was crucial in linking multiple armed robberies for the Broward County Sheriff's Office.

In that case, cartridge casings from four separate armed robberies were entered into Nibin and it immediately linked all four crimes to the exact same .40 caliber handgun.

Later, when deputies arrested suspects found with a weapon, a test firing of that weapon was entered into Nibin and it instantly confirmed a link to the four prior robberies.

The examiners manually verified the match and that led to charges for all four prior offenses.

For hit -and -run investigations, the key is the PDQ database.

That's the International Forensic Automotive Paint Data Query, or PDQ, database, which is managed by the Royal Canadian Mounted Police.

It contains chemical and color information for original automotive paints, covering over 21 ,000 samples and 85 ,000 layers of paint.

This is a massive class characteristic resource.

It is.

In a hit -and -run, the minute particles of paint left at the scene are analyzed spectrally and PDQ provides possible make, model, and year information for the unknown vehicle.

It is highly discriminating.

The Aztec gold metallic hit -and -run case file perfectly illustrates how quickly this database can narrow a search.

A victim was struck and killed and the only lead was a few minute gold metallic paint particles recovered from their clothing.

A spectral search through the PDQ database narrowed the possibilities to Aztec gold metallic on a 1990 or newer Ford, eventually pinpointing the source to only 1997 Ford Mustangs.

That's incredibly specific.

It is.

And this highly specific class information allowed investigators to quickly locate a suspect vehicle.

And then subsequent physical comparison showed that a plastic fragment found at the scene could be physically fitted together with the damaged molding on the suspect's car, confirming both the class and the individual match.

And finally, we have specialized databases for impressions and for missing persons.

For impressions, we have SICAR, which is Shoe Print Image Capture and Retrieval.

It's a commercial system where an analyst codes patterned features of a shoe print to create a description.

Its database, Solemate, includes over 39 ,000 footwear entries that can link a crime scene impression to a specific manufacturer.

And there's a secondary database, Treadmate, that houses about 8 ,500 tire tread records.

And the essential resource addressing the tragedy of unidentified remains,

Nomus.

The National Missing an Unidentified Person System was established in 2007.

It is a free, searchable, centralized repository designed to address the nearly 4 ,000 unidentified human decedent cases that medical examiners deal with annually.

How is it structured?

It's composed of three public databases.

The Missing Persons Database, the Unidentified Persons Database, which is searchable by characteristics like dental info and tattoos, and the Unclaimed Persons Database, for bodies that have been identified by name but whose next of kin can't be located.

It's an active database that works to connect those unidentified remains with families searching for loved ones.

We have covered an incredible amount of ground today, from the microscopic transfer of hair to the millions of data points contained in these global networks.

To quickly summarize, the two pillars that define the work.

Physical evidence is rigorously analyzed for either identification, determining what it is with near certainty, or comparison, determining if two samples share a common origin.

And we know that comparison hinges on that crucial distinction between individual characteristics, things like fingerprints, or perfect physical fit, offering near certainty in class characteristics.

Right, properties that only associate evidence with a group.

But the sheer diversity and quantity of class evidence, especially when you analyze it using the product rule, can often make it just as powerful.

Finally, we saw the dramatic and profound impact of databases, NGI, CODIS, Nibin, PDQ, which have completely flipped the script.

The Forensic Science Lab is no longer passive.

It is a proactive engine that links crimes and suspects across the world through computerized data.

Here's where it gets really interesting.

So what does this all mean?

We've explored this immense pressure on forensic science to achieve absolute mathematical certainty, as outlined by that big NRC report.

Yet, we've also seen that the expert must still rely heavily on years of subjective experience to determine the limits of natural variation.

That critical final judgment call.

That's it.

That final call about whether two items are different enough to break the common origin conclusion.

So here's the thought I want to leave you with.

How does the legal system, which is composed of laypeople who are often visually overwhelmed by scientific demonstration,

effectively balance the powerful, almost infallible, image of forensic science with the reality that an expert must rely on years of inherently subjective experience to draw that ultimate conclusion?

That tension between objective data and subjective judgment remains the fundamental challenge facing every forensic scientist in court today.