Gary hearing: GBI techs explain test errors that fouled DNA results

tchitwood@ledger-enquirer.comFebruary 26, 2014 

Georgia Bureau of Investigation crime-lab scientists testified Wednesday that a test matching convicted “Stocking Strangler” Carlton Gary’s DNA to evidence from one of the seven 1977-78 serial killings is valid, despite mistakes in other testing.

That came in a long day of detailed and sometimes esoteric testimony regarding laboratory techniques and procedures, much of it focused on the errors that caused other test samples to become contaminated and useless.

It was the third day of testimony in a new-trial hearing before Muscogee Superior Court Judge Frank Jordan Jr., who is to decide whether newly discovered evidence in the stranglings could have altered the outcome of Gary’s 1986 trial, when he was found guilty and sentenced to death.

He was hours away from lethal injection in 2009 when the Georgia Supreme Court stayed his execution and ordered a court here to consider DNA testing. The next year prosecutors and defense attorneys reviewed the stranglings evidence and chose items most likely to yield the killer’s DNA. DNA testing was not common in the mid-1980s when Gary was arrested and tried.

With some conflicting evidence in the stranglings, DNA testing seemed the most precise way to gauge Gary’s guilt. But it turned out to be a tangled tangent to the story when the GBI found it was contaminating lab tests — including one from the stranglings.

Three times a GBI lab contaminated crime-scene evidence with a “quality-control sample,” which is a known DNA sample the lab used to confirm its equipment was working properly. The first instance was at a Savannah lab where a worker tainted evidence with a pair of scissors that had not been sterilized.

Twice more this happened at the GBI lab in Decatur, where “Stocking Stranglings” evidence was tested.

Stephanie Fowler, who manages Georgia’s DNA database, testified supervisors at the lab, which has employees or their husbands donate semen for quality-control samples, became alarmed last year when the same control sample twice tainted crime-scene evidence, once in a 2011 Atlanta case unrelated to the stranglings.

This discovery prompted the agency to add control samples to its DNA database of sex offenders and profiles from open investigations. That led to the conclusion the same control sample in 2010 had tainted evidence from the Oct. 25, 1977, rape and strangling of Martha Thurmond, 69, of 2614 Marion St., Columbus.

Thurmond was the fourth stranglings victim, and investigators collected a semen sample from her abdomen, which attorneys in 2010 deemed suitable for DNA testing.

The initial result showed a DNA profile that was not Gary’s, but no one knew whom it came from until the profile was included in Georgia’s DNA database, and was found to match the DNA from the 2011 Atlanta case. Last November, the GBI discovered this profile was the quality-control sample the Decatur lab had been using. It had tainted both the Atlanta case and the Thurmond test.

Senior Assistant District Attorney Don Kelly asked Fowler if this revelation was “alarming.”

“That’s putting it mildly,” she replied.

Also disappointing was that all the Thurmond evidence was used up in the tainted test. Nothing was left to test again.

Though lab workers and prosecutors claim nothing fouled the test matching Gary’s DNA profile to evidence from the Sept. 24, 1977, rape and strangling of Jean Dimenstein, 71, of 3027 21st St. (now Cross Country Hill), the same is true of that sample, they said: Nothing’s left to confirm the initial result.

Testimony showed the lab had these deficiencies:

• Workers were cleaning their equipment with alcohol rather than bleach. Alcohol is not strong enough to sterilize lab gear used for DNA testing.

• Workers were sharing lab space and equipment, because they hadn’t enough room for private work areas. So even workers who were meticulous in cleaning their gear could have it tainted by someone else.

Connie Pickens was the GBI scientist who reported April 28, 2010, that her test on the Thurmond semen sample had yielded DNA that didn’t match Gary.

Pickens, who retired in April 2013 and recently died, had not even used the semen quality-control sample to test her results, and she was known to be conscientious about cleaning her work station. But the GBI learned another worker had employed the quality-control sample while using Pickens’ equipment the day before, and that apparently was how the test was contaminated.

Under cross-examination from defense attorney Jack Martin, Fowler also acknowledged the lab had prohibited workers listening to MP3 players, lest they taint evidence by touching the devices during testing.

Also the lab took half a day Nov. 27, 2013, thoroughly to decontaminate the testing area with bleach. Cleveland Miles, the GBI forensic biology manager, testified the lab now has enough room to provide a private work space for each of its 36 analysts, who conduct about 7,000 DNA tests a year.

Kristen Fripp of the GBI’s Savannah lab, who reviewed Pickens’ work on the Dimenstein DNA test as part of the agency’s peer-review process, testified the Dimenstein sample was clean and Pickens’ work conclusive, showing the chance the DNA came from someone other than Gary was one in 140 billion among black men and one in 480 billion among white men.

Half of Wednesday’s testimony focused on 1970s evidence tests for blood-typing. DNA testing was not in use at the time, so bodily fluids collected from rape scenes were subjected to a less precise analysis.

Former GBI forensic serologist John Wegel tested the stranglings evidence back then, and testified it showed the strangler was a weak or “non-secretor,” meaning in his bodily fluids he emitted little or no chemical marker evidencing his blood type.

Gary is a secretor: Tests show his semen and saliva carry the marker showing he’s blood-type O.

Wegel acknowledged Wednesday that Gary is a secretor, but said he believes secretions fluctuate, so Gary may not have secreted much of the marker at crime scenes. Under that premise, Gary could not be excluded as the strangler, Wegel said.

But he also said test results depend on the quality and quantity of the evidence tested, as some could be smeared or spread so thinly they make the secretions appear weak.

Under Martin’s continued questioning, Wegel said any male with type O blood could have been the killer. Excluded from that pool would be males of other blood types or any males that are non-secretors, who would leave no evidence of their blood type, he said.

Wegel’s testimony focused on evidence from the Oct. 21, 1977, rape and strangling of Florence Scheible, 89, of 1941 Dimon St., and from Thurmond’s homicide the following Oct. 25.

The Thurmond evidence included the semen sample found on her abdomen and semen from vaginal washings. Wegel said Thurmond was a type-O secretor, so he did not test the vaginal evidence because her blood was in the mix and would have skewed the results.

He discounted Martin’s questioning whether Thurmond’s sweat could have affected results from the semen collected from her abdomen. Sweat carries little of the blood-type marker, Wegel said.

Evidence from Scheible’s murder also indicated her rapist was a weak or non-secretor, Wegel said.

He said his concluding this evidence does not exclude Gary as the killer has not changed since he testified at Gary’s 1986 trial: “Nothing has changed my mind.”

Ledger-Enquirer is pleased to provide this opportunity to share information, experiences and observations about what's in the news. Some of the comments may be reprinted elsewhere in the site or in the newspaper. We encourage lively, open debate on the issues of the day, and ask that you refrain from profanity, hate speech, personal comments and remarks that are off point. Thank you for taking the time to offer your thoughts.

Commenting FAQs | Terms of Service