The Josephine Barnaby Case: America’s First Murder by Mail

Reading Time: 8 Minutes

In the spring of 1891, a package arrived at a Denver home where Mrs. Josephine Barnaby was staying on a health trip to the West. Inside was a bottle labeled as fine whiskey and a note wishing her a Happy New Year and inviting her to “accept this fine old whiskey from your friend in the woods,” a message she assumed came from the Bennetts, acquaintances from an Adirondack spa she frequented.

The parcel itself had already taken a curious tour — forgotten in the back of a buggy, left overnight at a livery stable, then finally carried into the kitchen and set on the mantel before Josephine tucked it into her trunk, a casual handling routine that would later matter when investigators reconstructed the evidence trail.

Days later, she retrieved the bottle from her trunk and shared a drink with her traveling companion, Florence Worrell. Worrell recoiled almost immediately, calling it “vile, vile stuff” and shivering at the bitter taste, while Josephine shrugged and drank again. Within minutes, both women were violently ill. Mrs. Worrell survived. Mrs. Barnaby was not so fortunate.

For five agonizing days, she drifted in and out of consciousness, insisting to anyone who would listen that she had been poisoned, and pressing repeatedly for a chemical analysis of the bottle. Even in agony, she vowed to “hunt down” whoever sent it and explicitly cleared the Bennetts as possible senders, signaling that she suspected someone closer to home.

The whiskey, it was later determined, was comprised of alcohol and a concentrated arsenic-based compound — roughly 60 times a fatal dose — making her death a textbook case of deliberate poisoning rather than accidental contamination.

On April 19, 1891, Josephine Barnaby was dead.

Within weeks, newspapers and investigators were calling it America’s first murder committed through the U.S. Postal Service: a lethal parcel, anonymous note, and poisoned whiskey delivered as a gift through the mail.

Josephine Barnaby had been a wealthy woman — wealthy, at least, in theory. Her husband, J.B. Barnaby, was a self‑made retail magnate, renowned for his large Providence, Rhode Island, department store and for racing his four‑in‑hand carriage recklessly down Broadway. In September 1889, at age fifty‑nine, he collapsed suddenly at his Providence home while preparing to drive his beloved horses, after several years of illness attributed to Bright’s disease, a serious kidney condition. He left behind an estate valued at roughly $1.7 million, including their flamboyant Providence mansion known locally as Barnaby’s Castle. According to his will, Mrs. Barnaby was to receive a modest $2,500 annual stipend, while the bulk of the estate went to their daughters, Mabel and Maud.

Enter Dr. T. Thatcher Graves, Mrs. Barnaby’s longtime physician, advisor, and business attorney. Seen by friends and family as easily influenced, Mrs. Barnaby fell under his sway.

He urged her to contest the will. She did, and she won, securing a settlement often reported as $105,000 as well as a substantial sum in expenses and fees. The victory dramatically increased the liquid portion of her estate — and placed it largely under Graves’s management.

Dr. Graves then helped arrange a new will and a sweeping power‑of‑attorney arrangement. Under those documents, Josephine gave him broad control over her finances and designated him as executor, with provisions that would leave him a $25,000 bequest from her estate. It was a comfortable arrangement — for Dr. Graves. He soon had his own $2,500 annual “salary” put in writing and had her sign a $16,000 receipt bundling his pay, claimed overdrafts, and various advances, giving him opaque authority over large portions of her money.

Behind the scenes, Dr. Graves’s handling of the estate was even more troubling. He had reportedly converted tens of thousands of dollars — often cited around $80,000 — into bearer securities, making the funds mobile and difficult to trace. Investigators later concluded that his investments and transfers had put much of her money at risk and left substantial discrepancies in her accounts, raising questions about whether he was protecting her interests or exploiting them.

By 1891, Josephine herself had grown wary. She voiced distrust of Dr. Graves and discussed changing her arrangements, telling at least one friend that “Dr. Graves would yet be the ruin of her,” a line that reads very differently once the anonymous bottle appears. She worried about the way he rebuked her spending, especially when she considered purchasing a cabin at the Adirondack spa where she had met the Bennetts, and he warned that her “extravagance” might lead to a court imposing a guardian over her affairs. In her final lucid moments, as arsenic ravaged her body, she wondered aloud: would Dr. Graves do such a thing?

Into this fraught situation stepped John Conrad, Mrs. Barnaby’s son‑in‑law and husband of her daughter Mabel. Conrad was a wealthy Montana cattle rancher and mine owner, part of the wave of Western capitalists whose fortunes were tied to land, livestock, and minerals. Outraged by his mother‑in‑law’s death and convinced that someone close to her had orchestrated it, he enlisted Pinkerton’s National Detective Agency. His instructions were simple and sweeping: use all available resources to bring the murderer — whoever they were—to justice.

Leading the charge from Pinkerton's Denver office was James McParland, the Agency's famed Superintendent of the Western Division — the same operative who, under the alias "James McKenna," had infiltrated the Molly Maguires two decades earlier and was celebrated as the “Greatest Detective” in America. A poisoning-by-mail case with a wealthy Western client and a trail running back to Boston was squarely his kind of puzzle, and he worked the investigation in close coordination with Superintendent John Cornish's New England office, stitching together leads from Denver to Providence.

McParland almost immediately dismissed suspicion on Mr. Bennet, Mrs. Barnaby’s friend in the Adirondacks. “[T]he reference to ‘your friend in the woods’ on the New Year’s greeting can only refer to the Adirondacks. It was therefore practically certain that suspicion would at once fall on Bennet as ‘the friend,’ but that he would be so foolish as to leave a clue pointing directly to him is unthinkable. The inevitable conclusion therefore must be that the murderer intended that Bennet should be suspected and that the message on the bottle was written solely with that in view.”

Within a week of concentrated work, Pinkerton New England Superintendent John Cornish reported cautiously, “We have progressed to a moderately satisfactory degree. We have some interesting material to work on, but understand emphatically, there has been nothing as yet developed that throws well‑founded suspicion on anyone.”

That was a careful statement designed both to reassure Conrad that progress was being made and to signal that no one would be publicly accused without stronger evidence. Soon enough, however, the Pinkertons found what they were looking for.

That was a careful statement designed both to reassure Conrad that progress was being made and to signal that no one would be publicly accused without stronger evidence. Soon enough, however, the Pinkertons found what they were looking for.

He was deeply invested in the outcome. He rearranged his business affairs in Montana so he could sit through the entire trial in Denver, financed travel and lodging for dozens of witnesses from the East, and met repeatedly with prosecutors to help assemble the case. His central role would later lead some writers and family members to question how much he shaped the case narrative, and even whether he may have had motives beyond pure grief.

The Agency launched a thorough investigation that stretched from California to Denver and deep into the New England states. Their work combined patient witness interviews, travel tracing, postal inquiries, evidence handling, and quiet surveillance of Dr. Graves. Every stage of Mrs. Barnaby’s journey — from Providence to the Adirondacks to Boston to Denver — was meticulously reconstructed so operatives could understand who knew her whereabouts and who had the opportunity to send a package timed to reach her in Colorado.

Pinkerton detectives secured the bottle and related physical evidence, then coordinated closely with local physicians and chemists to determine what had killed Josephine.

Doctors documented her symptoms — burning in the throat, vomiting, diarrhea, and collapse — and testified that they matched classical arsenic poisoning as described in medical texts. Chemists and expert toxicologists then tested the bottle's contents and biological samples from Mrs. Barnaby and Mrs. Worrell, using emerging toxicology techniques to confirm that the supposed whiskey actually contained 132 grams of arsenic — 2 grams would have been fatal. Toxicologists noted that whoever mixed the solution knew how to prepare it so the arsenic would remain fully dissolved rather than settling out. The report left little doubt that the sender intended death rather than mild illness.

To trace the origin of the package and its postage, the Pinkertons and postal officials focused on the Boston postmark and the distinctive 15-cent stamps affixed to the wrapper. They discovered that the 15-cent issue used on the Barnaby parcel had not been sold in meaningful quantities in Boston for roughly a year, suggesting that whoever posted the package had drawn on older stock or purchased stamps elsewhere.

Operatives canvassed sub post offices and stamp-selling stations in and around Boston, then expanded their inquiries across New England to 9,221 post offices and stamp stations, asking clerks when that particular issue had last been sold and in what quantities. They found that the stamps could only have been obtained at the Providence station.

McParland, diving into the minutest details, noted that the postage had been overpaid by 30 cents, “Indicating that the sender didn’t care to take the chance of having it weighed for fear that later he might be identified.”

Investigators built a picture of Graves's own mail habits. He was known to run a mail-order business supplying patients with patent medicines and tonics, regularly buying stamps in quantity to send packages from Providence and Boston. That pattern — combined with his financial entanglement with Josephine and his knowledge of her Denver address — made him a natural focus of the investigation.

Pinkerton investigators also traced the handwritten "friend in the woods" note, interviewing stationers and clerks and eventually securing testimony from a man in Boston who claimed he had written the anonymous message at Graves's request. The Rocky Mountain News in Denver, hungry for exclusive content, published facsimiles of Graves's signature and the inscription on the deadly bottle's label, amplifying the public's sense that the case would turn on handwriting and postal detective work.

Piecing together the postage records, handwriting evidence, postal routing information, and witness accounts of the parcel's handling, investigators connected the bottle's journey directly to people with access to Mrs. Barnaby's Denver address. Combined with financial records showing Graves's control over Josephine's money and his anticipation of a large bequest under her new will, this documentary and scientific evidence transformed initial suspicion into a circumstantial case jurors could evaluate.

In a pivotal meeting back in Providence, Conrad confronted Dr. Graves in his library, hinting that Pinkerton detectives already knew who had sent the bottle. Disturbed, Graves reportedly moved a sofa, raised a curtain, and inspected the three library windows as if searching for eavesdroppers, a nervous ritual that suggests he feared surveillance.

He then admitted he had mailed the bottle to Denver, claiming it contained nothing but whiskey and suggesting that someone else must have added poison later. That admission, though self‑serving, became a key brick in the circumstantial wall around him: Graves conceded that he had sent the bottle that killed Josephine, and the Pinkertons and prosecutors argued that the arsenic was there from the start.

The case was built largely on circumstantial evidence — but it was circumstantial evidence reinforced by chemical analysis, postal records, witness testimony, and Graves's own admission that he had mailed the bottle. It was enough for prosecutors to secure a grand jury indictment against Dr. Graves for premeditated murder by poison, and in May 1891 a Denver grand jury formally charged him with Mrs. Barnaby's murder. What followed was a nearly six-week trial, the longest murder trial in American history to that date.

The proceedings themselves were as dramatic as the story. After months of pretrial motions and delays, jury selection finally began in November 1891. This process took about ten days and required multiple large pools of prospective jurors — the court summoned an initial panel of 100, then another 150 once too few appeared, as attorneys screened candidates for pretrial opinions formed from newspaper coverage, objections to capital punishment, and doubts about whether a conviction could rest on circumstantial evidence alone. The court fined some no-shows and held three men in contempt when they pressed too hard to avoid service.

Once empaneled, the 12 men were kept under unusually strict controls. Bailiffs inspected watches, handkerchiefs, and neckties sent by family members for hidden messages — in one instance, opening and examining a juror's watch before allowing him to receive it — and even a request from a juror's wife to send clean underclothes was deferred until the presiding judge could approve it. The watching didn't stop there: Pinkerton operatives retained by John Conrad were still in the wings, discreetly monitoring the court's own bailiffs to eliminate even the appearance that anyone might sway the jury in a case so saturated with publicity.

The formal trial commenced on December 9, 1891. The case hinged on a few key pieces of evidence and the testimony of witnesses such as the Boston man who said he had penned the “friends in the woods” note at Graves’s request, postal clerks describing stamp sales, chemists presenting toxicology findings, and witnesses recounting Graves’s statements and his financial dealings with Josephine.

The defense, led by Judge Henry Marshall Furman, sought to undermine the circumstantial chain, airing allegations about Mrs. Barnaby’s spending, drinking, and personal life, and arguing that others might have had motive and opportunity to poison her. Graves himself vehemently protested his innocence, casting the case as a vendetta led by John Conrad and inflamed by sensationalist press coverage.

Despite those efforts, the jury was not persuaded. After weeks of testimony and argument, the case was finally given to the jury. On January 2, 1892, after deliberating for approximately ninety minutes, the jurors returned a guilty verdict: murder in the first degree, by poison and by mail. Dr. Graves wept as the verdict was read. When he appeared for sentencing on January 11, he again proclaimed his innocence, but Judge Rising told him that he would be delivered to the state penitentiary to be hanged by the neck until dead, using a new hanging apparatus that had already captured the public’s morbid curiosity.

The conviction did not end the legal saga. In mid‑1893, the Colorado Supreme Court overturned Graves’s conviction on appeal, citing trial errors by the judge and signaling that he would be retried for murder. Reports at the time described Graves as hopeful and even excited about the prospect of release, maintaining his innocence and telling friends that the high court’s decision vindicated his claims that the first trial had been unfair.

That optimism was short‑lived. When he learned that prosecutors intended to try him again rather than drop the case, Dr. Graves’s spirit appears to have collapsed. Shortly afterward, he was found dead in his cell. Beside him was a note asking that no autopsy be performed and giving a bitterly ironic cause of death: “Died of persecution; worn out; exhausted.”

Early newspaper reports and later historical accounts concluded that Dr. Graves had poisoned himself with morphine, although, as with Josephine’s death, some details remained contested. One doctor insisted that Dr. Graves had died of a “broken heart,” while others pointed to the circumstantial evidence of self‑administration of drugs, echoing the ambiguity and dramatic flair that had characterized the case from the beginning.

In the end, Dr. Graves’s widow received about $25,000 from Mrs. Barnaby’s estate, reflecting the bequest that had been written into Josephine’s new will before her death. So the physician accused of murdering his patient for financial gain died in custody protesting his innocence, while his family nonetheless collected the fruits of a contested estate.

From one perspective, the case was closed. A jury had convicted Dr. Graves of murder, a judge had sentenced him to death, and an appellate court had ordered a new trial that never took place because he died in prison. He took whatever truth he possessed — whether guilt or innocence — to the grave. But questions lingered, and they grew more tangled over time.

John Conrad had positioned himself as the grieving family member seeking justice at any cost: hiring the Pinkertons, financing witnesses, and sitting through every day of the Denver trial. Yet his role later became more complicated. In 1895, Mabel Barnaby Conrad divorced him. A Rhode Island court, in Hunter v. Conrad, noted that the divorce restored her status as a legally independent woman—a feme sole—and treated her as such in subsequent property disputes.

Family‑based accounts describe their marriage as deeply strained in the years after Josephine’s murder, with mutual accusations of infidelity and serious domestic scuffles among the staff, suggesting that the strain of the case contributed directly to the couple’s eventual rupture.

Later family narratives raised sharper questions about Conrad’s role in the larger story, including theories about the murder and Dr. Graves’s death. Those claims were never adjudicated and remain unresolved speculation, including the suggestion by Conrad’s own grandson, writer Barnaby Conrad, that his grandfather had orchestrated the murder himself and may even have arranged for someone to poison Dr. Graves in prison.

Small State Big History, drawing on Conrad’s book A Revolting Transaction, recounts those doubts and reminds readers that both Dr. Graves’s conviction and his reported suicide are, in the view of some family members and historians, open to question.

The Pinkertons, however, remained steady in their role. Investigators documented leads, traced evidence across thousands of miles and post offices, and helped prosecutors build the circumstantial case that persuaded a jury that sending poisoned whiskey by mail was an act of murder. The ultimate questions of guilt — and of what remains unknowable — outlived everyone involved. As with many nineteenth‑century poison cases, the Barnaby murder-by‑mail investigation sits at the intersection of emerging forensic science, energetic private detective work, and enduring ambiguities about motive and truth.

SOURCES

“The Anaconda Standard.” Chronicling America, Library of Congress, 4 Sept. 1893, www.loc.gov/resource/sn84036012/1893-09-04/ed-1/?st=text.

“Barnaby Castle.” Rhode Tour, rhodetour.org/items/show/157.

Cetient. “Hunter v. Conrad.” Cetient Case Law, www.cetient.com/case/hunter-v-conrad-9340789.

“Extravagant Barnaby’s Castle and Murder Most Foul.” Small State Big History, smallstatebighistory.com/extravagant-barnabys-castle-murder-foul/.

“Jerothmul B. Barnaby House.” Guide to Providence Architecture, guide.ppsri.org/property/jerothmul-b-barnaby-house.

“Judgment Against Mrs. Mabel Conrad.” The Rocky Mountain News, 9 Apr. 1899, www.coloradohistoricnewspapers.org/?a=d&d=RMD18990409-01.2.268.

“McParland, James P., 1844–1919.” Hagley Digital Archives, digital.hagley.org/taxonomy/term/15555.

“Molly Maguires.” Wikipedia, Wikimedia Foundation, en.wikipedia.org/wiki/Molly_Maguires.

Pinkerton criminal cases: Murder of josephine barnaby, articles and related material. (1891). Pinkerton's national detective agency, part B: Criminal case file, 1867-1961; series 1: A-C () Retrieved from https://www.proquest.com/archival-materials/pinkerton-criminal-cases-murder-josephine-barnaby/docview/3053178139/se-2

“The Rocky Mountain News (Daily), May 2, 1891.” Colorado Historic Newspapers Collection, www.coloradohistoricnewspapers.org/?a=d&d=RMD18910502-01.2.143.

“The Rocky Mountain News (Weekly), December 1, 1892.” Colorado Historic Newspapers Collection, www.coloradohistoricnewspapers.org/?a=d&d=RMW18921201-01.2.222.

Riffenburgh, Beau. Interview by Compulsive Reader staff. “An Interview with Beau Riffenburgh.” Compulsive Reader, 18 Nov. 2013, compulsivereader.com/2013/11/18/an-interview-with-beau-riffenburgh/.

Riffenburgh, Beau. Pinkerton’s Great Detective: The Amazing Life and Times of James McParland. Viking, 2013.

Rhode Island Historical Society. “Happy New Year… in April?” A Lively Experiment, rihs.org/happy-new-year-in-april/.

Rhode Island Historical Society. “Divorce.” A Lively Experiment, rihs.wordpress.com/tag/divorce/.

Scherer, Mary [Prairie Mary]. “Barnaby Conrad Jr.” Prairie Mary, Feb. 2013, prairiemary.blogspot.com/2013/02/barnaby-conrad-jr.html.

“30. Barnaby Castle: First Murder by Mail – Transcript.” Weird Island, Simplecast, weird-island.simplecast.com/episodes/30-barnaby-castle-first-murder-by-mail-EWJuMxsk/transcript.

“Trial Transcript / Contemporary Narrative of the Graves Case.” Google Books / HathiTrust, digitized volume coo-31924111698555, [permanent URL needed — signed S3 link expires].

“USGenWeb Sites: History of Providence County, Vol. I & II.” USGenWeb, usgenwebsites.org/RIProvidence/bios/ProvBios15.html.