Chapter 5: Misstatements, DNA issues plague probe
Date published: 11/20/2007
By PAMELA GOULD
Before Richard Marc Evonitz was publicly declared a serial killer, Spotsylvania County officials asked the Virginia crime lab to run his DNA profile through the state database in an attempt to determine whether he committed other crimes.
It was a logical step.
But if investigators thought it was going to tell them whether Evonitz was involved in four unsolved central Virginia murders, they were wrong.
DNA from those cases wasn’t in the state’s DNA database then, The Free Lance–Star discovered in its research.
As a result, Evonitz’s nuclear DNA profile wasn’t checked against evidence from the March 1996 slaying of 25-year-old Alicia Showalter Reynolds.
It wasn’t checked against evidence from the May 1996 slayings of 24-year-old Julie Williams and 26-year-old Lollie Winans.
And it wasn’t checked against evidence from the September 1996 slaying of 20-year-old Anne Carolyn McDaniel.
Evonitz was declared a serial killer on Aug. 13, 2002, after forensic analysis linked him to the slayings of three Spotsylvania County girls in 1996 and 1997.
In announcing that link, then-Sheriff Ron Knight said that authorities would do all they could to try to determine every crime Evonitz ever committed, but no others had turned up.
“I want to make clear at this time that there is no information which links Evonitz specifically to any other unsolved murders in Virginia or elsewhere,” he said, with the task force of local, state and federal officers that investigated Evonitz behind him.
Five days earlier, then-Maj. and now-Sheriff Howard Smith had put out a similar message in an online chat.
“At this time, we have no physical evidence that suggests he is responsible for any other unsolved murders in this state,” Smith said.
NO CHECKS DONE
But no physical evidence had been checked against central Virginia’s four unsolved slayings at the time of Knight’s and Smith’s statements, The Free Lance–Star found.
Not only had no DNA been checked against Evonitz in those slayings at that point, neither had hairs, fibers or any other forensic evidence.
Those exams remain essentially undone in three of the cases today.
The one exception is hairs checked in the Williams and Winans slayings—which failed to exclude Evonitz. But that wasn’t done until 2003, more than a year after the announcement.
Spotsylvania officials hadn’t gotten the state DNA database search results by the time they made their statements on Aug. 8 and Aug. 13, 2002.
Evonitz’s nuclear DNA wasn’t run through the state database until Aug. 12—the day before Knight’s announcement—according to Paul B. Ferrara, who was director of the state’s crime lab at the time.
Spotsylvania officials didn’t get the results until days after the announcement because Ferrara didn’t let his staff provide reports over the phone after a crime lab error years earlier in one of the Spotsylvania slayings, he said.
continued
Moreover, it would be another two months before Florida officials found a way to get Evonitz’s nuclear DNA profile into the national DNA database, providing the first opportunity to check him against unsolved slayings nationwide.
The Free Lance–Star found nothing to suggest that, even today, relevant DNA evidence from at least three of the four unsolved Virginia slayings from 1996 has been checked against Evonitz.
Timing and Technology
When Evonitz’s profile was run through the state DNA database in August 2002, the crime lab got no hits to any unsolved crimes.
Given the times during which Evonitz could have committed crimes in Virginia, that isn’t surprising.
Evonitz lived in Virginia during two time periods—first in Orange County in the early 1980s, and then in Fredericksburg and the counties of Stafford, Spotsylvania and Louisa from 1992 to ’99.
Unless law enforcement agencies had asked to have their evidence reanalyzed, evidence from crimes committed prior to 1997 wasn’t in the state’s DNA database when Evonitz was run through it.
In July 1997, the state crime lab switched to a new tool for producing DNA profiles—one known as short tandem repeat, or STR, Ferrara said.
When the switch was made, the lab automatically reanalyzed the DNA samples from convicted felons and loaded them into the database.
But police agencies needed to resubmit their crime-scene evidence to get it reanalyzed.
By Aug. 13, 2002, the Reynolds and McDaniel investigators hadn’t.
Now Done and undone
DNA evidence from the McDaniel slaying was resubmitted to the state crime lab in 2003, according to Culpeper County sheriff’s Maj. Jim Branch.
Branch did that as part of a review of the case assigned to him in 2000. He declined to say whether the lab was able to produce a profile or whether it’s now in any database.
Male DNA found in Williams’ and Winans’ slayings was analyzed by the FBI Laboratory under the new STR protocol as early as 2001, but was not entered into either the state or national database, according to several sources.
In the Reynolds case, Virginia State Police Agent Stan Gregg submitted two of three items containing DNA evidence to the state crime lab on Aug. 14, 2002—the day after the Evonitz announcement.
But Gregg didn’t submit it to be checked against Evonitz. He asked the state lab to produce new profiles for comparison with Darrell Rice, the man indicted four months earlier on federal capital murder charges in Williams’ and Winans’ deaths.
The Free Lance–Star found no record that state police ever asked any crime lab to compare the Reynolds evidence with Evonitz, even though then-Lt. Rick Jenkins said in September 2002 that “naturally” it was being done.
The state lab produced profiles from the two items of DNA evidence in the Reynolds case in April 2004, a forensic report shows.
Those profiles were then loaded into the Virginia DNA database. But they weren’t checked against Evonitz because his DNA wasn’t in the database at that time.
One of the profiles was later loaded into the national database.
But because the DNA from those two items turned out to be from a woman and a male relative, they are unlikely to help solve the crime.
What offers greater potential—and could hold the key to identifying Reynolds’ killer—is a third DNA specimen.
But The Free Lance–Star found nothing to suggest it ever was submitted for reanalysis.
SIDE BAR
Experts urge more tests on Evonitz clues
--------------------------------------------------------------------------------
By PAMELA GOULD
--------------------------------------------------------------------------------
When federal prosecutors dropped capital murder charges against Darrell Rice in the Shenandoah National Park slayings, the question of serial killer Richard Marc Evonitz’s involvement was left unanswered despite opportunities for DNA analysis.
Mitochondrial DNA testing in October 2003 had failed to exclude Evonitz as the source of two head hairs found at the scene of the May 1996 slayings of 24-year-old Julie Williams and 26-year-old Lollie Winans. And yet, the FBI Lab didn’t take additional steps to try to determine whether they were Evonitz’s hairs.
The FBI also didn’t ask to have Evonitz’s DNA compared to other male nuclear DNA found at the scene.
And the FBI didn’t ask to have Evonitz checked through Y-STR analysis.
Nuclear DNA is well-known for its ability to identify individuals and is the kind of DNA loaded into state and national databases.
What is less well-known is that, taken together, mitochondrial DNA and Y-STR DNA analyses hold promise for answering the question of whether Evonitz committed the murders.
MITOCHONDRIAL DNA
Mitochondrial DNA, abbreviated by scientists as mtDNA, is the part of the genetic code inherited from the mother. Analyzing it is the best approach currently available for examining hairs.
The Shenandoah slayings case had two key head hairs —one on a glove near Winans’ body, the other within layers of duct tape wound around her wrists. Forensic analysis showed the two hairs were probably from the same person—but not Williams, Winans or Rice.
Scientist Constance Fisher got an inconclusive result when she checked the hairs against Evonitz, her October 2003 report shows.
It was inconclusive because she found a “one base pair difference” between Evonitz’s hairs and the crime-scene hairs.
With nuclear DNA, any difference between a suspect’s profile and a crime-scene profile excludes the suspect from consideration. But with mtDNA, scientists are aware of a phenomenon known as heteroplasmy, so they need more than one difference for an exclusion.
Heteroplasmy is defined as a mixed result at a particular base along the 16,000-plus-base mtDNA strand. It is estimated to occur in one-fourth to one-third of the population.
The FBI Lab was aware of the phenomenon when it checked Evonitz’s hairs against the crime-scene hairs in 2003. Its protocol includes steps for trying to determine whether the person shows it.
But Fisher didn’t take those steps.
The FBI Lab would not allow Fisher to be interviewed about the case.
Bob Fram, chief of the FBI Lab’s Scientific Analysis Section, would not comment directly on this case but said the additional steps are usually taken.
“It’s in the examiner’s discretion to do that,” he said. “In general, they do.”
DNA experts said they would have taken the analysis further in attempt to get a more definitive answer, given the significance of the hairs, the fact that the case was unsolved and law enforcement’s stated interest in wanting to know every crime Evonitz committed. Plus, federal prosecutors were continuing to call Rice a suspect even though forensic tests found no link to him.
“I know there are additional things that can be done and I probably would do those things,” said Terry Melton, president of Mitotyping Technologies in State College, Pa., a top private mtDNA lab.
Melton and William Shields, a biology professor at the State University of New York at Syracuse and a frequent consultant on DNA cases, identified four options for trying to eliminate the inconclusive result for Evonitz.
Three of those involve trying to see if either Evonitz or the crime-scene hairs show heteroplasmy. If that were shown, it would strengthen the case for Evonitz being the source of the hairs.
But, despite what’s portrayed on television programs, the FBI Lab would never say the hairs came from Evonitz. The closest they’d ever come is saying he “could not be excluded” as the source.
The fourth procedure would seek to find a second difference between the profile for Evonitz’s hairs and the crime-scene hairs. If a second difference were found, Evonitz would be eliminated as the source.
“Right now, we don’t have enough data to say with certainty that it is or it’s not” Evonitz’s hair, Shields said.
But given law enforcement’s statement that it wanted to know every crime the serial killer committed, Shields said the decision is simple.
“If it’s about closing cases,” he said, “they should have continued.”
Y-STR Analysis
Y-STR analysis is used when there is a mixture of DNA in crime-scene evidence and at least one source is male.
This technique separates out the male portion of the DNA and attempts to produce a profile from it by extracting information from the Y, or male, chromosome.
The analysis is helpful in rape cases or any case where a victim’s and a perpetrator’s DNA are mixed.
In the Shenandoah slayings, both victims were female so any male DNA should have been of interest to investigators.
Several items of evidence contained a mixture including male DNA and could be subjected to Y-STR analysis, according to forensic reports and DNA experts.
Though Y-STR profiles are produced from nuclear DNA, they do not have the same distinguishing power as traditional nuclear DNA profiles, according to Mechthild Prinz, a forensic science professor with the New York University School of Medicine and a pioneer in Y-STR analysis.
“You can’t say it has to be this one individual the way you can with other [nuclear] DNA typing,” she said.
But, like other nuclear DNA profiles, if the Y-STR analysis produced a full profile, it either matches or excludes a suspect, Prinz said.
Given that the analysis looks only at the male chromosome, a match does not rule out every other man on the planet. If there is a match, male relatives of the person would also match, she said.
Also, after producing a Y-STR profile, scientists check a database for statistics on the likelihood of a non-relative having that profile.
Once a Y-STR profile is produced and matches a suspect to crime-scene evidence, an investigation’s focus shifts from science to circumstance, checking the likelihood a male relative of the suspect could have committed the crime.
Prinz, a forensic science practitioner in one of the nation’s biggest crime labs, said neither mtDNA nor a Y-STR profile alone would conclusively identify a killer.
But taken together, they could be powerful.
If she had a case where the suspect couldn’t be excluded based on mtDNA, and Y-STR produced a matching profile, she would believe she had her man.
“Absolutely,” Prinz said. “The question always is, is it a coincidence?”
--------------------------------------------------------------------------------
Could DNA tests solve Reynolds case?
--------------------------------------------------------------------------------
By PAMELA GOULD
--------------------------------------------------------------------------------
For years, word within law enforcement circles has been that the Alicia Showalter Reynolds murder case lacked strong forensic evidence.
But a Free Lance Star review of Virginia State Police records and state crime lab reports, and interviews with DNA experts, suggest science could lead to her killer.
The state crime lab found nuclear DNA evidence in the case in its initial review, shortly after the 25-year-old Reynolds' remains were discovered on May 7, 1996.
The lab produced a DNA profile from the most significant item--sperm in Reynolds' panties--in September 1996 using what were then state-of-the-art tools.
Those results suggested the DNA was from Reynolds' husband, who is not a suspect. But Mark Reynolds told police he and his wife--both taxed by their pursuit of graduate degrees--had not had sex for at least two weeks.
DNA experts said that evidence should be retested because of scientific advances in DNA analysis over the past decade.
The 1996 testing protocol produced results so imprecise that its value was more in excluding people than in identifying who left the DNA, said Paul B. Ferrara, director of the state crime lab until his retirement last December.
�If you included somebody [as a possible source], you could probably also include ... tens of thousands of other people,� he said.
�It was just not specific enough to be of value�especially with mixed samples and degraded samples,� he added. �And in the Reynolds case, we were dealing with degraded samples.�
J. Thomas McClintock, a DNA expert who teaches at George Mason University and also runs a consulting firm in Maryland, said advances since then open the door for retesting. He thinks STR�short tandem repeat�tests can and should be done.
�If I had these results and it�s 1997, I would be comfortable,� McClintock said. �It�s now 2007, I have sample, I would do further testing.�
STR tests require only microscopic bits of DNA�trillionths of a gram�to produce results, Ferrara said. He said state lab scientists do everything possible to obtain a profile.
Answer wanted
Virginia State Police Capt. Rick Jenkins, supervisor of the Reynolds investigation, refused to discuss any of the forensic evidence in the case.
But as a general statement, he said state police have done whatever lab personnel suggested.
A review of state police investigative reports and state crime lab reports, and an interview with Ferrara, give no indication the sperm was ever reanalyzed.
Two other items with nuclear DNA were submitted to the state lab for analysis under the STR protocol in August 2002, a request by state police Agent Stan Gregg shows. But the item containing DNA from sperm was not included in that request or any other report obtained by The Free Lance�Star.
Ferrara said that if the evidence hasn�t been resubmitted since 2002, the re-examination hasn�t been done. The only way to know whether the DNA came from Mark Reynolds is to run the STR test.
If that DNA came from him, it�s irrelevant. If it didn�t, it could be loaded into the state and national DNA databases and checked against serial killer Richard Marc Evonitz and other criminals and crimes across the country.
Mark Reynolds, who was widowed at age 26, would like those steps taken.
�That seems like a very simple thing that could be done,� he said.
--------------------------------------------------------------------------------
Fla. put Evonitz in FBI database
When Richard Marc Evonitz was declared a serial killer in August 2002, authorities expressed frustration that they couldn�t get his DNA profile into the national database for a quick comparison with unsolved crimes across the country.
One detective predicted the system would �light up like a Christmas tree� if the profile were uploaded.
But the 38-year-old committed suicide in June 2002 before he could be arrested.
So while evidence showed he killed three Spotsylvania County girls in 1996 and 1997, Evonitz was never convicted of a felony in Virginia. By law, that meant his profile could not be loaded into the state�s DNA database.
But his DNA profile did make it into the national DNA database, thanks to the state of Florida.
In June 1987, Evonitz pleaded no contest in Florida to a felony charge of committing a lewd or lascivious act against a minor�exposing himself to a teenage girl in January of that year.
Clay County, Fla., court records show Circuit Judge Lamar Winegeart Jr. entered an order �withholding adjudication� of guilt. He placed Evonitz on probation for three years and ordered him to undergo mental-health treatment.
But for the purposes of the DNA database, Florida law includes a no-contest plea under its definition of �conviction.� That was sufficient to place Evonitz�s profile in the state�s DNA database on Oct. 4, 2002, said Heather Smith, a Florida Department of Law Enforcement spokeswoman.
Because states can upload their profiles into the national DNA database, Evonitz�s profile made it into the National DNA Index System administered by the FBI within a week.
It is now there for ongoing comparison with unsolved crimes across the country.
No cases have been linked to Evonitz as a result of his being entered into the national database, according to a Florida official who has access to the information.
�Pamela Gould