Dennis Dechaine, shown in court in 2013, was convicted of murdering Sarah Cherry in 1988. Press Herald staff photo

New, state-of-the-art DNA testing has failed to tie Dennis Dechaine to items used in the 1988 kidnapping and murder of 12-year-old babysitter Sarah Cherry, prompting his defense team to say they will seek a new trial.

Dechaine has spent 34 years in prison since his conviction in the murder. The Bowdoinham farmer with no criminal record or history of violence became the sole suspect in Cherry’s kidnapping on the afternoon of July 6, 1988, when a receipt and notebook with his name on them were found at the Bowdoin home where the girl was babysitting. Her body was found two days later in the woods three miles away, close to where his truck had been parked when he was picked up by police the night of the abduction. She had been sexually tortured and strangled, her hands bound in yellow rope similar to rope found in Dechaine’s truck.

Sarah Cherry of Bowdoin was 12 when she went missing from a babysitting job in 1988.

Dechaine has always maintained his innocence, saying he was framed by the real perpetrator, who planted items from Dechaine’s truck at the crime scenes while he was in the woods doing drugs.

In September 2021, Dechaine’s lawyers filed a motion for additional DNA testing on the grounds that new, more-sensitive collection technology had become available, and Knox County Superior Court Justice Bruce Mallonee approved the motion in July against the objection of state prosecutors.

The testing was performed by the Serological Research Institute in Richmond, California – or “SERI” – a lab chosen by the state and paid for by Trial and Error, a group supporting a new trial for Dechaine. The lab used M-Vac technology, which is capable of collecting 66 times more DNA material than previous methods, according to the FBI.

Partial male DNA profiles were found on at least four of six objects tested. Dechaine was definitively excluded from DNA found on three objects: the bra worn by Cherry, one of the sticks used to violate her and the bandana used to gag her.


The attempt to match DNA profiles on three more objects – the scarf used to strangle her, her blood-stained T-shirt, and a second stick used in the sexual assault – was inconclusive – there was not enough data to either exclude or match Dechaine as the source of the DNA.

“If this man did what they’re saying he did, his DNA should be all over all of the evidence used in the commission of the crime,” said John Nale, part of the team representing Dechaine pro-bono, at his office in Waterville last month.

Dennis Dechaine is led from the Knox County Courthhouse on April 4, 1989, in Rockland after being sentenced to life in prison for the murder of Sarah Cherry. Press Herald staff photo

Nale said they would use the results to seek a new trial for Dechaine. Under Maine’s DNA law, when a person convicted of a crime is excluded as the source of DNA on crime scene evidence, a new trial may be granted if certain thresholds are met.

A spokesperson for the Attorney General’s Office, which opposed the motion for new testing, said the office is reviewing the results but does not comment on pending litigation.

Attempts to contact Cherry’s family were not successful.

The items tested yielded mixtures of partial profiles, making analysis difficult, Nale said.


According to the results summary Nale provided to the Press Herald, Dechaine “could be included as a contributor” to DNA found on the scarf and one stick, but experts say that does not indicate a match.

Those items, as well as two others, underwent “Y-STR” DNA testing, according to the summary. This focuses on repeating DNA patterns in the male Y-chromosome.

Dr. Theodore Kessis, a DNA expert with Applied DNA Resources in Ohio, said the lab’s conclusion that Dechaine “could be included” was meaningless without additional information on the statistical likelihood of a match. (He noted YSTR inclusion results generally indicate a lower likelihood of a match than other forms of DNA testing.) If any probability information was included in SERI’s results, Nale did not share it with the Press Herald and he did not respond to inquiries about its existence.

Nale said SERI had told him that Dechaine was included only when small segments of DNA were available – basically, when there was enough of a profile available, he was conclusively excluded as their source and when there was not enough to exclude him, he was included as a possible source.

“DNA is like our Social Security number,” Nale said. “You and I might share some (numbers), but unless they match up exactly, then there is no match. And if I’m only seeing a small piece of your Social Security number, that may have a couple of my numbers in it, I can’t be excluded.”

SERI lab director Gary Harmor would not speak about the case or respond to questions about the statistics associated with the results.


Dennis Dechaine, convicted of the 1988 murder of 12-year-old Sarah Cherry, makes a bid for a new trial in 2012. Defense attorney Steven Peterson, left, is seated with Dechaine. Press Herald staff photo

In Dechaine’s previous DNA appeal, which led to hearings in 2013, the court considered testing that excluded Dechaine from male DNA in the blood under Cherry’s fingernail as well as the results of testing that yielded multiple partial profiles but did not rule out Dechaine as a source.

The state argued that the thumbnail DNA might have been contaminated in the lab before preservation protocols were in place or that it might have been present prior to the crime. The courts agreed and determined that the evidence was too weak to grant a new trial.

After receiving the latest results, Nale told Dechaine that the new testing had proven his innocence.

“After hearing about these test results, well, it’s kind of a mix of relief, combined with anger, combined with disgust,” Dechaine said by phone last month. “It’s just such a bag of emotions and overwhelming feelings.”

Dennis Dechaine, right, exhales heavily as he enters the Knox County Courthouse April 4, 1989, in Rockland prior to his sentencing in the Sarah Cherry murder case. Press Herald staff photo

But Dechaine said he anticipates an “epic fight” from state prosecutors who may argue that the profiles are incomplete and come from multiple donors.

“These are problematic issues,” he said. “All I can tell you is they exclude me, which is a hell of a lot better than nothing. And it certainly would have been a lot more had the courts acquiesced to my initial plea back in ’88 to have the testing done before trial. I think at that point, having fresh material and an abundance of it would have yielded a completely different result. I think we would have known then for certain who did this. But they didn’t want to do it. And so here we are, 34 years later.”


The latest test results have not been matched to any other individuals, although Nale said Dechaine’s legal team is working to compare the DNA profiles to at least one other potential suspect.

Meanwhile, identifying a different suspect as the perpetrator is not the important question legally at this point and is not Dechaine’s responsibility, Nale insisted.

“My only question is this: Is Mr. Dechaine’s DNA on any of the evidence used in the commission of the crime?,” he asked, banging his hand on his office conference table with every word.

The exclusions add to the glaring absence of any biological evidence connecting Dechaine and Cherry, Nale said. He pointed out that even the state Supreme Judicial Court called the case “troubling” for this and other reasons in a 2015 decision, despite upholding the lower court’s denial of a new trial.

Another problem in the state’s case against Dechaine is that the medical examiner offered ambiguous testimony about the time of Cherry’s death. While the significance of this was not realized at trial, later forensic analysis that was never admitted in court concluded that she must have died by strangulation hours after Dechaine was picked up by police. This was based on the presence of rigor mortis and the early stages of other decomposition markers when she was found on the afternoon of July 8 during a heatwave. (The state countered with another forensic opinion which concluded that Cherry could have been dead two days if she were buried in cool earth. Cherry was found above ground, but under a pile of leaves and forest debris.)

But the opinions were not debated in court. Maine law makes it difficult for convicted people to have new evidence of innocence considered because of strict time limits on introducing new evidence – it must be brought up within a year of when it “could have been discovered” with due diligence. Further, Maine courts have interpreted the statute to mean that only new evidence related to DNA may be admitted in DNA-based appeals.


The statute also states that DNA testing is allowed only within two years of new technology becoming commonly known and available, so Dechaine had to wait for new technology to emerge before trying again.

A drawback to the new technology available now is that it is so powerful it can pick up more “touch” DNA, the traces of DNA in smudges of things like skin cells or sweat we leave behind on almost everything we make contact with. It can even pick up DNA left at crime scenes through secondary transfer. This makes it more likely that multiple profiles will be detected, even from individuals who were never at the scene. More profiles add to the noise of the DNA analysis.

Further complicating matters, if Dechaine’s claim that the perpetrator rifled through his truck to plant items at the scene is correct, those items may have Dechaine’s DNA on them from the last time he touched them. However, even with the stronger technology, no DNA from Dechaine has yet been detected.

A police photo of Dennis Dechaine’s pickup truck taken on July 6, 1988, the day Sarah Cherry disappeared. The truck was found near the scene where Cherry’s body was discovered.

Daniel Medwed, a professor of law and criminal justice at Northeastern University and a founding member of the New England Innocence Project, which works to correct and prevent wrongful convictions, has been reviewing Dechaine’s case after his lawyers reached out. He said the results of the new testing raise questions about Dechaine’s involvement in the crime, given the prosecution’s theory of the case that a lot of the biological evidence found at the scene would have come from the perpetrator.

“Obviously, as a practical matter, it’s easier to show that you didn’t do something when there’s evidence pointing to the true suspect,” he said, “but the fact that there are unknown potential suspects shouldn’t negate the fact that the evidence against Dechaine is now much, much weaker than it was thought to be.”

Medwed described a match to another person as the “Holy Grail” in these cases, “because then you can convince the prosecutor and convince the judge that someone else can take the person’s position in prison.”


“It’s a little harder when you have some exclusion evidence, and you have evidence that points to some unknown profiles,” he said. “The key is whether or not the evidence now is so much weaker against Dennis Dechaine that he deserves a new trial, or this new finding about the biological evidence casts such doubt on it that justice would require an airing in court.”

This is up to the judge’s discretion. He must determine whether the DNA evidence weakens the state’s case enough that a new trial would have a different outcome, and whether the DNA evidence obtained is “material to the issue” of identifying the perpetrator, according to Maine’s DNA statute.


Several other suspects have emerged in the Cherry case over the years, including child sex offenders who were living within a mile or two of the abduction site at the time. In addition, The Press Herald recently revealed that a serial killer, the late Richard Evonitz, known to kidnap young girls from their yards in broad daylight, was living in Portland that summer, a short drive from Bowdoin. The founder of the Innocence Project at University of Virginia Law School, Deirdre Enright, sent Nale Evonitz’s profile for comparison, after seeing the similarities between his crimes and the killing of Cherry.

Nale said he is working with professionals on matching the results to alternate suspects, including Evonitz, but it is a very slow and difficult process because of the mixtures of partial profiles obtained.

“The state has the same results as well, and if they believe that they should be looking at possible suspects, we of course would encourage them to do the same,” he said.

Related Headlines

Comments are no longer available on this story

filed under: