How do investigators trace bullets to a particular gun?

ExpandAutoplay

Image 1 of 7

Demonstrating some of her examination techniques, forensic scientist Debra Gillis fires a Colt King Cobra into a tube of Kevlar fibers that catches the bullet for further study. Gillis specializes in firearms and toolmarks at Alaska’s Scientific Crime Detection Laboratory. (Marc Lester / Alaska Dispatch News)

Anchorage police said last week the Colt Python .357 magnum revolver James Ritchie used to shoot at a APD officer last weekend, was the same gun used to kill five people in the city over the summer.

But how did investigators know the bullets came from that specific gun?

"We've never found two firearms that created the same individual characteristics on a bullet," said James Gannalo, a retired detective with the New York City Police Department's Ballistics Unit who now runs a forensic consultancy.

Each time a bullet is fired through the barrel of a gun, like the Colt Python, it becomes imprinted with grooves and microscopic imperfections — markings as specific to a gun as fingerprints to a person, he said.

While local investigators declined this week to speak specifically about how they made the Colt Python connection, citing the ongoing investigation, local and national ballistic experts, including Gannalo, agreed to talk broadly about how firearm experts can link fired bullets to a particular weapon.

Gathering bullets

To do an analysis, investigators must first find the bullets. Sometimes the bullets are lodged in walls or covered in brush or buried in bodies.

"Officers amaze me when they find them. The bullet is just awfully, awfully small when it comes out and then it just goes burrowing into whatever," said Debra Gillis, firearm/tool-mark examiner and biological screener at the Alaska Scientific Crime Detection Laboratory in Anchorage.

Bullets can provide some answers — if they're in good condition and have good markings — as shell casings left behind can. It's less common to find shell casings with a revolver, like the Colt Python, because the casings remain in the cylinder after the gun has been fired. Other weapons automatically eject casings with each shot.

Even if no casings are found, Gillis said, she doesn't assume the shooting involved a revolver. Criminals can clean up evidence by picking up casings and taking them away.

In the Anchorage Police Department's timeline of its investigation into the summer homicides, released Tuesday, police said no shell casings were discovered around the Ship Creek bike path where Jason Netter Sr., 41, and Brianna Foisy, 20, were found dead on July 3.

Less than four weeks later, APD wrote it did not discover any shell casings on an East Anchorage street where Treyveonkindell Thompson, 21, was found dead. Police did not say whether or not they found casings on Aug. 28 at Valley of the Moon Park, the scene of the double homicide of Kevin S. Turner, 34, and Bryant "Brie" DeHusson, 25.

In all five killings, police said, they recovered bullets or bullet fragments, which went to the state crime lab.

Searching a database

Gillis is one of just two firearm examiners in Alaska charged with analyzing fired bullets sent to the crime lab from law enforcement across the state. The other half of her job, tool-mark expert, refers to similar evidence — the barely visible but distinctive marks left behind by a tool, such as the marks on a wire from a pair of cutters.

"What I am is a pattern recognition expert," she said in her office Thursday amid test bullets and microscopes.

Gun manufacturers use different techniques to create the rifling in a gun barrel — the spiral grooves that get a bullet spinning like a top to improve its accuracy. The direction of the twist lines on a bullet, and the number, width and depth of the depressed and raised parallel lines — "lands and grooves" — can help indicate the model of gun that fired it.

Gregory Laskowski, a California-based private forensic scientist, said examiners can use bullet information to search an FBI database and create a list of guns that could have fired the bullet. Examiners can weigh and measure the bullet to determine its caliber to further narrow down the list, he said.

Gillis said some guns create a more unique bullet than others and thus, a shorter list.

All Colt Pythons produce bullets with lines that twist to the left, Gannalo said. Police said in their timeline they received ballistic testing indicating the weapon used in the first of the string of homicides on July 3, was a Colt Python .357 magnum.

Police said the crime lab confirmed that bullets found at the July 29 homicide and Aug. 28 double homicide came from that same Colt Python, an uncommon gun considered a collectible.

Shooting the gun

Brad Heffele said he hasn't seen many Colt Pythons pass through his shop, New Gun Traders on Muldoon Road. The gun is no longer on the market. One in pristine condition could sell for up to $3,300, he said.

"It's quite a collectible item anymore," Heffele said. "They're not something that people tend to sell."

Anchorage District Attorney Clint Campion described the Colt Python as a "fairly rare firearm," and one he personally had not seen in connection to a criminal case. When police found a Colt Python on Ritchie last weekend, Campion likened it to finding a "needle in a haystack."

It is still unclear how Ritchie got the gun. Police said this week that someone in Anchorage purchased the gun in 1971, when there were more restrictions on gun ownership than today. Police said earlier this week that they plan to talk to the original owner to find out more about the gun's ownership.

At the crime lab, Gillis has access to about 1,500 guns, stored in a reference library. She uses the guns when testifying in court, breaks them down to identify parts and shoots them to produce sample bullets.

As a demonstration to journalists Thursday, Gillis fired a Colt King Cobra revolver at a long, metal container filled with water, then repeated with a shot into a container stuffed with fluffy, yellow Kevlar, found in bulletproof vests.

In a criminal case, Gillis would have fired a suspect gun to produce an undamaged bullet.

Gillis fished out both bullets and lit a blowtorch. She ran a piece of magnesium ribbon through the flames, and held the bullets in the smoke. This would dull them, she said, so the light from the microscope wouldn't reflect on their shine and diminish the bullets' very subtle imperfections she relied on to make a match.

"I'm taking off that glare," she said.

Gillis placed the bullets under a comparison microscope.

Determining an identification

Under the microscope, Gillis looked for a match of the minuscule lines on the side of the bullet. She compared it to a bar code. It's the subtle imperfections unique to specific guns, in part having to do with the markings the cutting tool leaves during production as it moves down the barrel of the gun.

"No metal is homogeneous," Gillis said. "So at a microscopic level, you're going to throw chunks from that drill bit at different times, in different places, which will leave different gouges or hills."

Gillis said it can take anywhere from 5 minutes to two days to analyze bullets. She has worked in the crime lab since 2001.

She said she didn't know how often investigators are able to match bullets to the same gun, compared to how many investigations are inconclusive. A lot depends on how well a gun marks — some leave more distinct markings on a bullet than others — as well as the condition of the fired bullet when retrieved from a crime scene.

Darrel Gardner, a federal public defender in Anchorage, said there were "some real issues" with linking bullets to a specific firearm, as multiple experts could have conflicting opinions.

He pointed to a September 2016 study by the President's Council of Advisors on Science and Technology, which questioned the validity of firearm analysis. It concluded if firearms analysts are allowed to testify in court, they should have undergone "rigorous proficiency testing" and should disclose whether or not they knew facts in the case that might have influenced their conclusions.

"You want evidence to be scientific and fact-based," he said.