On January 11, 1928, in the Court of Chancery in Newark, Vice Chancellor John Backes split the trial of Fryer et al. v. United States Radium Corporation into two questions instead of one. The first was whether the case could be heard at all. The second, whether the company owed anything, would only matter if the first went the plaintiffs’ way. Five women sat in the courtroom: Grace Fryer, Katherine Schaub, Quinta McDonald, Albina Larice, and Edna Hussman. All had painted watch dials with radium-laced paint at the company’s Orange, New Jersey plant between 1917 and the early 1920s. None of them had worked there in years. That gap, not the poisoning itself, was the argument the company’s lawyers had built their defense on.

Why workers’ compensation did not cover them

New Jersey had amended its workers’ compensation law in 1924 to add occupational disease to the list of compensable injuries, including, by name, radium necrosis. That should have settled the matter. It did not, because the amendment required disability to appear within five months of the worker’s last exposure to qualify. Radium poisoning does not work on that schedule. The isotope lodges in bone and keeps emitting; the jaw necrosis, the anemia, the sarcomas surfaced three, five, seven years after a woman had left the dial-painting room, not five months. The statute had been written for injuries with a start date and an end date, and radium poisoning had neither. So Fryer and the others could not file a workers’ compensation claim. They had to sue in tort, under the ordinary rules for a personal injury, and those rules came with their own clock.

The statute of limitations problem

New Jersey required a personal injury suit within two years of the injury. United States Radium’s defense read the statute at face value: the injury occurred on the job, the women had left the job years earlier, the two years had long since run. Backes rejected the premise underneath that reading, that an injury from radium exposure has a single date at all. He ruled that because radium remained active in the women’s bones, the company’s original act kept producing new injury with every day that passed, so the claim kept accruing for as long as the isotope kept decaying inside them. It did not lapse. Backes’ reasoning was narrow, tied to one plaintiff group and one judge’s reading of one statute, and it never became binding precedent outside that courtroom. What it answered, for the first time in an American court, was what a limitations period means when the harm has no fixed moment of completion. A durable, general answer to that question was another five decades away, arriving with the asbestos litigation of the 1970s and 1980s, when courts nationally adopted what is now called the discovery rule: a claim accrues when the injury is discovered, not when the exposure occurred. The radium cases were the improvisation that came before the doctrine had a name, decided case by case because no legislature had yet written the rule down.

The trial never reached its second half. On June 4, 1928, United States Radium settled. Each of the five women received $10,000, a $600 annual pension contingent on a medical panel continuing to confirm the diagnosis, and payment of past and future medical costs. It was, at the time, treated in the press as compassion winning out over a company’s stonewalling. That is a fair description of the outcome. It is not a description of the mechanism that produced it. What produced the settlement was a Chancery judge finding a legal theory that got the statute of limitations out of the way before the underlying liability question was ever tried.

The story usually told is the wrong shape

The version of this history that circulates treats it as a morality tale: a callous company, dying women, a public reckoning. That is not false, but it is not what made the case matter afterward. The durable outcome was procedural. It settled, for the first time in American law, that a disease with a multi-year latency period could not be timed out of court on the theory that the clock started at exposure rather than at discovery or ongoing harm. Every occupational disease claim that followed, from asbestosis to the delayed effects of benzene, inherited that problem before it inherited any sympathy. The corporate-villainy framing gives the story a satisfying shape. The legal-mechanism framing is why the story kept being cited by lawyers who never worked in a dial-painting room.

Illinois tried the same claim and lost it once

The New Jersey settlement did not automatically extend to Illinois, where the Radium Dial Company had operated a plant in Ottawa since September 1922, staffed by a separate set of dial painters under a separate state statute. In 1935, a group led by Inez Vallat sued Radium Dial and lost, all the way through the Illinois Supreme Court. The problem this time was not the statute of limitations; it was that Illinois’s 1911 Occupational Diseases Act covered accidental injury, and the courts read radium poisoning as neither accidental nor, under the statute’s terms, a covered disease at all. Vallat v. Radium Dial Co., 360 Ill. 407 (1935), closed that door. It is the case that should complicate any tidy version of this history: the same facts, argued honestly, produced a loss, because the statute simply did not reach poison.

The legislature did what the courts could not

Illinois rewrote its Occupational Diseases Act in 1936 to bring industrial poisoning inside the statute. The amendment was not retroactive, so it could not revive Vallat’s claim, but it created a claim that had not existed before. Catherine Donohue, a former Radium Dial painter, led a new group of workers back to the Illinois Industrial Commission in 1937 under the rewritten law. She testified in July 1937 too weak to raise her arm without help. The Commission ruled in her favor, and an award was confirmed on July 6, 1938. Donohue died on July 27, 1938, three weeks later, before the company’s appeals were resolved.

Radium Dial kept appealing, and its final argument was procedural rather than medical: the company asked the Illinois Supreme Court for a writ of mandamus that would let it seek review of the award without posting the bond the statute required, arguing the Commission had lacked jurisdiction in the first place because Donohue’s employment had ended in 1931 while her disability was dated to 1934. The court disagreed. In People ex rel. Radium Dial Co. v. Ryan, 371 Ill. 597 (1939), it held that a bond-free path around the award did not exist, jurisdictional objection or not. The company had exhausted its arguments. The award stood, paid to Donohue’s estate rather than to Donohue.

What actually changed

Nobody wrote a general federal rule out of these cases. What changed was narrower and, for that reason, more traceable: two states confirmed that an occupational disease claim could survive a defense built entirely on elapsed time, and Illinois demonstrated, in the space of two years, that a legislature could add a disease to a compensable list after the courts had already said the existing list did not reach it. Radium necrosis went from an injury with nowhere to file, to one compensable under ordinary tort in New Jersey by judicial improvisation, to one compensable under statute in Illinois by legislative amendment. Three different routes to the same destination, none of them clean, all of them documented in case files that later occupational disease lawyers went and read.

The burden of proof shifted too, with less public notice than the limitations question got. In New Jersey, Fryer’s tort suit still required proving causation and fault the ordinary common-law way, exposure to negligence to injury, argued from scratch by her attorney against a company with its own experts. In Illinois after 1936, a worker no longer had to prove the employer was at fault at all. The rewritten Occupational Diseases Act made industrial poisoning compensable on exposure and diagnosis alone, the same no-fault logic that already governed a broken arm on a factory floor. That is the less-publicized half of what these cases settled: not just when a worker could sue over a slow-moving disease, but how much they had to prove once they did, and Illinois’s answer, once it existed, asked for less than New Jersey’s court had.

The company at the center of the New Jersey case had itself changed hands more than once by the time Fryer sued. It began as the Radium Luminous Materials Corporation, founded in Newark in 1913, and became the United States Radium Corporation under new leadership in 1921, the same year the company’s president, George Willis, died and his death was written up in the Journal of the American Medical Association. The Ottawa, Illinois plant that would later employ Catherine Donohue opened in September 1922, run by a separate company, Radium Dial, under a separate state’s law. The two firms shared a raw material and a technique, lip, dip, and paint, but they never shared a courtroom, a statute, or a verdict.

The Illinois award file for Catherine Donohue lists a total then owing of $5,661. It was paid to a woman who had already been dead for months by the time the last appeal failed.