Fort Worth, Texas, becomes a crucial place of historical investigation: in the National Archives at Fort Worth, records preserved amid vast rows of boxes offer clues about W. W. Vaughan’s killing, the death of George Bigheart, and the wider pattern of Osage people dying under guardianship. The archive does not provide a single document that settles the cases. Its significance lies in the way disparate records—legal claims, informant reports, and handwritten registers—can be brought together, while their gaps and uncertainties remain visible.

The archive and its scale

The National Archives at Fort Worth occupies a warehouse larger than most airport hangars. More than one hundred thousand cubic feet of material are kept in humidity-controlled conditions, stacked in rows fifteen feet high. The holdings range from transcripts of the U.S. District Courts of Oklahoma, covering 1907 to 1969, to records on the 1900 Galveston hurricane, the assassination of John F. Kennedy, slavery and Reconstruction, and reports from Bureau of Indian Affairs field offices. The range places the Osage files within a much larger documentary record of government action and human calamity.

The researcher came to Fort Worth seeking evidence about Vaughan’s murder. The archive’s administrative order offered no guarantee of a complete account: records could preserve a claim, an official decision, or a report without explaining what had happened beyond it. Elsewhere in the investigation, the coroner’s inquest records concerning Anna Brown had mysteriously disappeared, and evidence from crime scenes was scant. Against that background, Fort Worth’s boxes offered not a master key but fragments that could be compared across sources.

Vaughan, Bigheart, and the disputed money

One trail began with a lawsuit filed by Vaughan’s widow, Rosa Vaughan, against H. G. Burt. The suit, begun in 1923, appeared at first to concern an ordinary debt between friends: Vaughan had acted as Burt’s attorney, and Rosa sought to recover $10,000 she said Burt owed her deceased husband. Both men had been close associates, and Burt was president of a bank in Pawhuska.

The court records connected the disputed sum to Bigheart, whom Vaughan had also represented. Before Bigheart died of suspected poisoning in a hospital in Oklahoma City, he had sought a certificate of competency that would end his status as a government ward and allow him to spend his headright payments freely. Vaughan helped him apply. Bigheart planned to pay Vaughan as much as $10,000 for legal services, a sum the account compares to nearly $140,000 today. Burt, however, had somehow collected the money. The records thus placed a disputed payment between two deaths: Bigheart’s and, days later, Vaughan’s.

Other materials complicated the picture. Secret testimony from the 1926 grand jury contained no record of Burt testifying, but the insurance agent who had issued a policy on Henry Roan, naming William K. Hale as beneficiary, said Burt had recommended another American Indian as a possible target for an insurance scheme. In Bureau of Investigation records, an informant described Burt and Hale as close associates and said they had divided money obtained from Bigheart. A separate report noted that Hale claimed $6,000 from Bigheart’s estate using a fraudulent creditor’s note. The records did not establish whether the money Burt collected included the disputed $10,000, and the possible sums were small beside the headrights involved in other murders or Roan’s $25,000 insurance policy. The Justice Department did not prosecute Hale for Bigheart’s killing or pursue Burt further; agents nevertheless described Burt as a murderer.

The train journey and an informant’s account

The archive also yielded an interview with an informant who had been close to Hale and had provided evidence in other murder cases. Asked about Vaughan’s killing, the informant identified Burt as the person he suspected. That statement gained significance alongside newspaper accounts showing that Burt boarded the train with Vaughan in Oklahoma City and later reported Vaughan missing from his berth. The report and the newspaper details converge on Burt’s presence during the journey, but they do not by themselves establish what happened on the train.

The uncertainty mattered to Vaughan’s family. His widow, Rosa, was left to raise ten children without income; the family moved from a two-story home into a storage garage, and Osage neighbors helped provide food. The family also questioned whether Vaughan had been killed before he was thrown from the train or died on impact. The inquest listed the cause of death as unknown. In later research outside the Fort Worth files, notes from an unpublished interview with the Fairfax town marshal implicated Burt and described a local tough helping him. These accounts added a possible explanation, not a conclusive reconstruction. The archival case remained circumstantial, and the identity and precise actions of the alleged killer could not be established with certainty.

Guardianship registers and the scale of unexamined deaths

A tattered, fabric-covered logbook from the Office of Indian Affairs offered a different kind of evidence. Written by hand, it listed guardians and their Osage wards; when a ward died, the clerk often added one stark word beside the name: dead. Burt was listed as guardian of Bigheart’s daughter and four other Osage people. One of those wards had died. The entry documented a guardianship relationship and a death, but did not identify a cause or establish wrongdoing.

The same register showed that Scott Mathis, owner of the Big Hill Trading Company, had been guardian to nine Osage people, including Anna Brown and her mother, Lizzie. Seven of his listed wards had died, at least two in known murders. Other guardians’ lists showed similarly disturbing totals: one had eleven wards, eight marked deceased; another had thirteen, more than half recorded as dead; a third had five, all deceased. These numbers defied an ordinary pattern of deaths, but because most cases had never been investigated, they could not show how many deaths were suspicious or identify who was responsible.

Related files supplied individual grounds for concern without closing that gap. Bureau agents had suspected that Anna Sanford was poisoned, although her death was never classified as a homicide. Hlu-ah-to-me officially died of tuberculosis, while an informant alleged that her guardian withheld treatment and refused to send her to a hospital in the Southwest. After her death, the guardian became administrator of her valuable estate. Such allegations, like the register’s totals, pointed toward possible abuse of the guardianship system; they were not equivalent to adjudicated findings.

What the records can and cannot establish

The Fort Worth materials make the archive a site where the accepted closure of the murder cases can be tested against what remained unresolved. The records connect financial claims, guardianship, suspected poisoning, and Vaughan’s fatal train journey, but the connections vary in strength. A court suit records a dispute; an informant report records an allegation; a guardian register records deaths without explaining them. Read together, they raise questions that no single file answers.

That distinction is central to the investigation’s account of Fort Worth. The archive allows the researcher to recover details missed or left unexplored in earlier inquiries, including the link between Burt and Bigheart’s daughter’s estate and the suspicious concentration of deaths among wards. At the same time, the missing testimony, unclassified deaths, and unknown cause of Vaughan’s death limit what can responsibly be claimed. Fort Worth supplies evidence for a more searching history, not a final verdict: the files expose patterns and possible relationships while preserving the difference between suspicion and proof.