Scientific Sleuthing
Investigators May Be Stumped by Levy
Case, But Forensics Offer Array of Tools

By Amanda Onion

ABC News
June 4 — Detectives and sniffing dogs are
largely gone from the leaf-strewn slope in
Rock Creek Park where the remains of
24-year-old Chandra Levy were found. But
some forensic experts say it's possible clues
could remain on the forest floor. 

Authorities in Washington, D.C., are faced with a puzzle. They
know the former intern was killed. But they don't know how
she died and they admit they may never find out exactly what
caused her death. What they really want to discover is who
killed her. Forensics may lead them to the answers.

For example, forensic botanists could analyze vegetation
growing around the crime scene to determine how long
Levy's body had been in the Washington park.
Trace-evidence specialists could glean information from single
strands of hair, fibers and bits of dust and pollen.

The intern had been missing for 13 months before the grim
discovery in Rock Creek Park. All that was found of Levy's
remains were a collection of bones, some clothing and
personal effects. If authorities had found the body soon after
her death, forensic entomologists could have combined data
on weather and the age and kind of insects present on a body
— mainly blowflies — in complicated calculations to determine
a time of death.

• Interactive: Forensic Tools
As the hit television series CSI: Crime Scene Investigation
has portrayed, forensic science offers a myriad of high-tech
methods for analyzing clues. But as the technology behind
forensic science advances, experts caution forensic
scientists must take extra care to ensure they scour crime
scenes for all possible clues, and that forensic analysis is not overplayed in the courtroom. 

Herb MacDonell, founder of the Bloodstain Evidence Institute in Corning, N.Y., cautions forensic
deduction can seem elementary, but "in forensic science, nothing is simple." 

Every Action Leaves a Trace

French scientist Edmond Locard declared in the early 1900s that every violent action leaves a
trace and even the most minute traces can help solve crimes. It was on this foundation that
Locard opened one of the first laboratories dedicated to the field of forensic science in 1910. 

Two years later, Locard found a killer by analyzing material
underneath a female victim's fingernails.

Other renowned sleuths added their own contributions. In the
mid-1800s, Mathieu Orfila published one of the earliest papers
about the detection of poison in the blood. Henry Faulds, a
Scottish physician working in Tokyo in 1880, was among the
first to suggest that fingerprints could be used to identify
suspects. And Calvin Goddard developed ways to determine if a
bullet was fired from a suspect's gun in the early 1900s.

Sir Arthur Conan Doyle's fictional character Sherlock Holmes
popularized the field when he applied some of the cutting-edge
methods of his time including developing theories of serology,
fingerprinting and firearm identification. 

Since then forensic scientists have gained better tools to glean
even more information from tiny clues. But while forensic
science is helpful in the lab, Peter DeForest, a crime scene
investigator and trace analysis expert at John Jay College of Criminal Justice in New York,
argues forensic scientists should have a greater presence at the early end of investigations —
during the search for clues at crime scenes. 

"You need someone with a scientific background assessing the entirety of the physical
evidence," said DeForest. "It often takes scientific training to recognize a clue." 

Missed Blood on Nicole Simpson’s Back

DeForest says the West Coast is further along in adding scientists to the ranks of detectives
released first to crime scenes. It was in Los Angeles, after all, that opened the first forensic lab
in the United States in 1923. The East Coast, he says, has been somewhat slower to change. 

Nonetheless, it was in Los Angeles where one of the more infamous examples of overlooked
evidence took place, during the investigation into the slayings of Nicole Brown Simpson and
Ronald Goldman in 1994. 

Photos of Simpson's body revealed blood spattered on her back that analysts later concluded
must have originated from someone else — perhaps her killer. But investigators failed to take a
genetic sample of the blood before turning the victim on her back and contaminating any possible
samples. 

Other times it has taken a forensic scientist to realize a clue has value.

As he offered a tour through John Jay's forensic training laboratories, packed with instruments,
including gas chromatographs (used to separate traces of chemicals from sediment or blood),
spectographs (measures light emitted by a sample) and comparison microscopes (microscopes
that provide magnified images of two samples at once), DeForest recalled one case where tiny
strands of fiber helped prosecute a suspect. 

In 1993, the strangled body of 30-year-old Russian immigrant Vladimir Makhno was found
dumped in a bush area in the remote area of York, Ontario, Canada. DeForest got a phone call.
Could he identify some orange fibers found on Makhno's body? 

Using spectrograph analysis and comparison microscopes, DeForest was able to match the
fibers with remnants of a carpet found in a suspect's home. Ironically, it wasn't the prosecution
who had thought to call DeForest about the fibers. It was the defense. 

"I guess the prosecution didn't think the fibers were that important," said DeForest. "The
laboratory didn't realize the significance of what they had." 

Defense attorneys thought the fibers would absolve their client, since preliminary tests showed
the suspect's carpet was made of two kinds of fibers, not one. But when DeForest tested the
carpet himself, he learned that it only shed the fiber type that was found on Makhno's body. 

"You have to learn how to ask the right questions as well as find the right answers," explained
DeForest. 

‘Junk Science’

Although scientific evidence is often overlooked, sometimes it can also be overplayed. A 1993
Supreme Court decision, known as the Daubert ruling, stipulates that courtroom justices prompt
questions about the scientific sturdiness of evidence. But the problem of so-called junk science
still lingers. 

MacDonell is the leading expert on "reading" blood spatter at crime scenes. He has testified in
such high-profile cases as the assassinations of Robert F. Kennedy and Martin Luther King Jr.,
as well as the O.J. Simpson case. 

He also helps instruct others in the science every year at his Bloodstain Evidence Institute. At
the clinic, students literally throw human blood around (old pints of it are bought from the Red
Cross) to understand how it falls and splatters. 

While he's confident that most of his students finish his clinic better equipped to interpret crime
scenes, he says some have become overconfident. 

"I think bloodstain analysis is being used and abused more and more," he said. "Of my students,
there are about five Frankensteins." 

Part of the problem is the science can seem deceivingly simple. The basic premise of bloodstain
analysis is a droplet's size reveals how much energy was applied, its shape suggests the
direction of its source and its distribution reveals how far away from a surface blood was
splattered. 

"You have to take them all together," MacDonell said. "It sounds simple, but if you haven't had a
lot of experience, you can make a lot of mistakes." 

Other fields of forensic analysis are also facing increased scrutiny. 

The ability of researchers to identify a suspect through fingerprints was challenged by a federal
judge in January. U.S. District Court Judge Louis H. Pollak ruled that fingerprints lifted from crime
scenes cannot meet standards of scientific scrutiny established by the Supreme Court since
most detected prints are smeared and incomplete. Pollak said examiners cannot testify at a trial
that a suspect's fingerprints "match" those found at a crime scene. 

Michael Saks, a law professor at Arizona State University, says he thinks the use of bite marks
left on victims as identifying evidence is also on shaky ground. He says recent tests show the
method averages an error rate of about 62 percent — much higher than misidentification rates in
fingerprint tests or handwriting analyses. 

"I keep waiting to see a big challenge in court," he said. 

Search Continues

In the Levy case, authorities have ideas about how she died — that she was strangled or
suffocated — but they have no conclusions.

"Was it a sexual assault? An attack by a stranger, by someone she knew? All of these are
possibilities," Metropolitan Police Chief Charles Ramsey said last week.

Investigators are searching for a monogrammed ring and a bracelet that Levy may have been
wearing at the time of her death. The hope is merchants at a local pawn shop might have
interacted with her killer. They're also analyzing remnants of her clothing, including spandex
leggings that were found with knots tied in them, suggesting she may have been sexually
assaulted. 

Other clues may yet emerge.

Katherine Ramsland, a professor of forensic psychology at DeSales University in Bethlehem,
Pa., pointed out in a recent Philadelphia Inquirer column that Levy's clothing could host blood or
DNA specimens that belong to someone else. And fibers or foreign hairs may still be clinging to
her shoes or clothes — especially if her body was brought to the location by a car and wrapped
in a blanket or carpet. 

As Ramsland wrote, "It is grim, difficult work, but one good thing is that modern forensic science
has many techniques to call on."