Stearns County Sheriff Don Gudmundson last month wasn't shy about criticizing the Jacob Wetterling investigation for "going off the rails," taking some less-than-subtle jabs at the FBI. On Tuesday, a former case investigator is expected to jab back.

Steve Gilkerson, a former FBI special agent who worked the investigation in the first months after the October 1989 kidnapping, said Monday that he will address "speculative conclusions" made by Gudmundson when the sheriff released thousands of pages of state investigative documents in the case Sept. 20.

Gilkerson plans a news conference for 11 a.m. at the Hennepin County Government Center in Minneapolis.

He is expected to hammer on points he made in a Sept. 20 letter to several law enforcement agencies, including the FBI and the Minnesota Sheriffs' Association. He wrote that "the sheriff obviously feels his opinions and criticism in this matter deserve widespread attention. Any opinions and criticism of what he said by someone he criticized and who has knowledge of the facts deserves equal treatment."

In critiquing the investigation, Gudmundson said that Danny Heinrich, who confessed in 2016 to abducting and killing 11-year-old Jacob, should have been the prime suspect from the investigation's earliest days, but that investigative mistakes and misplaced effort in the weeks and months following the abduction prevented authorities from connecting the dots and apprehending the killer.

Gudmundson said there was enough information linking Heinrich to Jacob's disappearance — such as tire tracks, a shoe print and a tip that clearly linked the former Paynesville, Minn., loner to the case — that he should have been the primary suspect within 48 hours of the kidnapping.

Instead, Gudmundson said, a task force assembled to find Jacob and his abductor wasted more time chasing far-flung leads and conferring with psychics than tracking more compelling evidence close to home.

While Gudmundson faulted a lack of cooperation among local, state and federal agencies, he seemed to place most of the blame for investigative blunders with the FBI. When pressed on the point, he said, "All of us failed."

In his Sept. 20 letter, Gilkerson disputes many of Gudmundson's key criticisms as "not accurate or completely false," including that shoe prints and tire tracks left at the abduction scene were a match to Heinrich. "They were similar, but not the ones that made those prints," he wrote.

Gudmundson also said last month that the "most fatal flaw" in the investigation occurred in February 1990, when Heinrich was arrested late at night at a bar in Roscoe, Minn., and interrogated by two investigators — one of whom was Gilkerson.

The sheriff said arrests in cases like that are generally meticulously planned. "One would never willingly arrest someone drunk late at night," he said, adding that the two agents who interviewed Heinrich lacked experience investigating homicide suspects.

Gilkerson in his letter defended the decision to interview Heinrich, who at the time was considered a strong suspect, but who was later released from custody for lack of evidence. It would be another 20 years before investigators would take another look at Heinrich.

Jacob was abducted by a masked man at gunpoint the night of Oct. 22, 1989, as he, his brother and best friend headed home from a convenience store in St. Joseph, Minn. His whereabouts remained a mystery for 27 years until Heinrich, arrested on child pornography charges, told authorities that he killed the boy the same night he kidnapped him, then buried him in a shallow grave outside Paynesville before returning a year later to move the remains to another grave nearby. Paynesville is about 30 miles from the abduction scene.