On March 13, 1968, ranchers in Skull Valley, Utah discovered thousands of sheep convulsing and dying across 27,000 acres of grazing land. Within days, approximately 6,400 sheep were dead. The Army's Dugway Proving Ground—a vast chemical and biological weapons testing facility—was 27 miles upwind. For three weeks, military officials denied any connection. Declassified documents later confirmed that VX nerve agent—one of the deadliest substances ever created—had drifted off the test range during open-air spraying operations the day before the deaths began.
On the morning of March 14, 1968, rancher Alvin Hatch walked into his sheep pasture in Skull Valley, Utah to find dozens of animals convulsing in the sagebrush. Ewes staggered in circles, frothing at the mouth. Lambs collapsed mid-step, legs twitching. By noon, hundreds were dead. By the following day, the death toll exceeded 2,000 across multiple ranches. Within a week, approximately 6,400 sheep—nearly half the valley's entire flock—had died or been euthanized.
Skull Valley is a remote basin west of Salt Lake City, bounded by the Cedar Mountains to the west and the Stansbury Range to the east. For generations, the area's sparse vegetation had supported sheep ranching on open range managed by the Bureau of Land Management. The valley's isolation was both its economic foundation and its vulnerability—the nearest neighbor was the U.S. Army's Dugway Proving Ground, one of the military's most secretive installations, located 27 miles upwind.
Dr. Kenneth Knudsen, a veterinarian from Tooele County, arrived at the affected ranches on March 14. What he documented in his field notes would become the foundation of a federal investigation: excessive salivation, pinpoint pupils, respiratory distress, violent muscle tremors, and death by asphyxiation. The clinical presentation was unmistakable to anyone with toxicology training—organophosphate poisoning, the signature of nerve agent exposure.
Knudsen immediately suspected Dugway. Ranchers had complained for years about dead wildlife near the proving ground's boundaries. Strange odors occasionally drifted across the valley after test days. But the military maintained operational secrecy, and locals had learned that asking questions about activities at the base produced no answers.
When Knudsen contacted Army officials at Dugway on March 15, he was told no tests had been conducted that could have affected livestock. Over the following three weeks, as tissue samples accumulated and independent veterinarians confirmed the diagnosis of anticholinesterase poisoning, the Army maintained its denial. Possible causes suggested by military spokesmen included malnutrition, toxic plants, and bacterial infection—explanations that made no veterinary sense given the sudden onset, mass casualty rate, and characteristic neurological symptoms.
On March 21, Utah's Department of Health requested detailed information about recent testing at Dugway, specifically asking whether nerve agents had been used. The request was denied on grounds of national security classification. Without confirmation of the agent involved, state health officials struggled to assess human exposure risk. Ranch workers who had handled dying sheep reported symptoms including headaches, blurred vision, and respiratory irritation—consistent with low-level nerve agent exposure but impossible to definitively diagnose without knowing what compound they might have encountered.
"The Army's position for those first three weeks was a complete stonewalling. They wouldn't confirm there had been any testing, wouldn't tell us what agents might be involved, wouldn't provide any technical assistance. Meanwhile we had thousands of dead sheep and people who had been exposed to something asking whether they should be worried."
Dr. Kenneth Knudsen — Interview, Salt Lake Tribune, 1969The break came from an unlikely source. On April 4, an anonymous tip to the Washington Post indicated that Dugway had conducted a VX nerve agent spray test on March 13—the day before the sheep deaths began. The tipster provided specific details about the test mission, including the use of an F-4 Phantom jet equipped with spray tanks and the approximate time of the test flight.
Faced with impending press exposure, Secretary of the Army Stanley Resor acknowledged on April 8 that Dugway had indeed conducted a VX test on March 13 and that the Army was investigating a possible connection to the livestock deaths. The admission came 26 days after the deaths began.
Declassified documents released decades later confirmed what the Army finally admitted in April 1968: Dugway's Deseret Test Center had conducted an open-air spray test of VX nerve agent on the afternoon of March 13. The test involved an F-4 Phantom jet flying a grid pattern over a designated impact area on the proving ground, releasing liquid VX from underwing spray tanks at an altitude of approximately 150 feet.
VX—"venomous agent X"—had been developed in Britain in the early 1950s as part of research into organophosphate insecticides. When British scientists recognized its extreme toxicity, the formula was shared with the United States. American chemists refined the compound, and by 1961 the U.S. had begun industrial-scale production at the Newport Army Ammunition Plant in Indiana. VX was approximately ten times more lethal than sarin, the nerve agent used in the 1995 Tokyo subway attack, and unlike sarin, VX was persistent—remaining active in the environment for days or weeks rather than dissipating within hours.
The Dugway test on March 13 was part of ongoing evaluation of aerial spray systems intended for potential battlefield use. The test plan called for releasing VX over a target grid marked with sampling equipment to measure agent dispersion and droplet size. Standard protocol required meteorological assessment before each test to ensure wind conditions would keep the agent within designated safety zones.
According to the Army's subsequent investigation, meteorological conditions on March 13 appeared acceptable at ground level—light winds from the southwest. But conditions aloft were not adequately assessed. At the altitude where the F-4 released its payload, stronger winds from a different direction carried the VX plume north-northeast—away from the target area and directly toward Skull Valley.
Once the Army acknowledged the test, the political calculus shifted from complete denial to damage control. An internal investigation was ordered, led by a review board that included representatives from Dugway, Edgewood Arsenal (the Army's chemical research headquarters in Maryland), and the Office of the Secretary of the Army.
The investigation team collected environmental samples, sheep tissue samples, and meteorological data. Independent veterinary pathologist Dr. James Rankin, hired by the ranchers, conducted detailed necropsies that confirmed massive anticholinesterase activity in sheep brains and organs—unambiguous evidence of nerve agent exposure. Chemical analysis by Army laboratories detected VX residue in soil and vegetation samples collected from the affected ranches.
But even as evidence mounted, the Army's public statements remained carefully hedged. The official conclusion, released in May 1968, stated that the March 13 VX test "probably" caused the sheep deaths but avoided absolute certainty. This linguistic hedging was strategic—acknowledging enough to appear cooperative while preserving legal deniability in anticipated lawsuits.
The investigation also revealed systemic problems with testing protocols at Dugway. The General Accounting Office (GAO), Congress's investigative arm, conducted its own review and found that safety zones for open-air tests were determined using outdated dispersion models that failed to account for complex desert meteorology. Meteorological equipment at Dugway measured surface winds but not conditions aloft where aircraft released agents. No real-time monitoring system existed to detect when agents drifted beyond intended impact zones. There was no formal notification system to alert surrounding communities when tests occurred.
While 6,400 dead sheep made headlines, the more alarming question received less attention: how many people were exposed? Skull Valley's human population was small—perhaps 50-75 residents across scattered ranches and the tiny settlement of Dugway Junction. But during the 48 hours after the March 13 test, dozens of people had been outdoors: ranchers checking livestock, ranch hands moving sheep, families living in the valley.
Utah's Department of Health conducted surveys documenting symptoms among residents who had been outdoors on March 13-14. Reports included respiratory irritation, burning eyes, headaches, nausea, and in several cases, blurred vision and difficulty breathing—symptoms consistent with low-level nerve agent exposure. The state recommended that affected individuals seek medical evaluation, but without definitive confirmation of exposure levels, physicians struggled to provide meaningful treatment or prognosis.
The long-term health effects, if any, were impossible to establish. VX exposure at sub-lethal doses can cause prolonged cholinergic symptoms, but distinguishing these from other causes decades later would be medically and legally problematic. Unlike radiation exposure, which can sometimes be detected years later through biomarkers, nerve agent exposure leaves no definitive long-term signature.
The incident highlighted a fundamental problem with secret military testing adjacent to civilian populations: when accidents occurred, the classification system prevented timely medical response. Doctors treating potentially exposed individuals couldn't access information about the agent involved, its toxicology, or appropriate treatment protocols because that information was classified. This Catch-22—national security preventing effective public health response—would recur in subsequent chemical and biological testing incidents.
The Skull Valley ranchers filed a $5 million lawsuit against the U.S. government in federal court, seeking compensation for lost livestock, property devaluation, and economic harm to their ranching operations. The case presented unusual legal challenges: the Federal Tort Claims Act provided limited sovereign immunity waiver for government negligence, but the government argued that discretionary military decisions were exempt from liability.
Rather than proceed to trial—which would have required extensive disclosure of classified information about VX testing protocols and Dugway operations—the government settled in 1970 for approximately $1 million. The settlement included no admission of liability, though the Army had publicly acknowledged the probable connection between the VX test and sheep deaths.
"They paid us for dead sheep but never admitted they did anything wrong. That was the galling part—we knew what happened, they knew what happened, but they wouldn't say it in a legal document because that might create precedent for the next accident."
Alvin Hatch, Skull Valley Rancher — Testimony before Senate Subcommittee, 1969The political response was more significant than the legal settlement. Senator Eugene McCarthy, then running for president on an anti-war platform, seized on Dugway as evidence of military recklessness endangering American civilians. Congressional hearings examined not just the Dugway incident but the broader chemical weapons testing program. Testimony revealed that similar open-air tests had been conducted at Dugway hundreds of times annually for years. The proving ground's remote location had been selected precisely to minimize civilian exposure risk—yet the sheep kill demonstrated that "remote" wasn't remote enough when dealing with agents measured in lethal doses of milligrams.
Public pressure mounted for restrictions on chemical weapons testing. The Army suspended open-air testing of lethal agents in 1969 pending implementation of enhanced safety protocols. But the damage to the chemical weapons program's public legitimacy was irreversible.
When Richard Nixon took office in January 1969, the chemical and biological weapons programs faced unprecedented scrutiny. The Dugway incident provided a domestic illustration of dangers that had previously been abstract. National Security Advisor Henry Kissinger ordered a comprehensive review of chemical and biological warfare capabilities, assessing their military utility against political costs.
The review concluded that these weapons provided marginal military advantage—the U.S. conventional military superiority made chemical and biological weapons strategically unnecessary. Meanwhile, they created significant diplomatic liability, complicated arms control negotiations, and generated domestic opposition. The review also noted that restricting other nations' chemical and biological programs through international agreements would benefit U.S. security more than maintaining American stockpiles.
On November 25, 1969, Nixon announced that the United States would unilaterally terminate its offensive biological weapons program and destroy all biological stockpiles. Chemical weapons were not eliminated but were restricted, with development limited to riot control agents and herbicides (though the latter would continue to be used extensively in Vietnam as Agent Orange). The policy shift paved the way for the 1972 Biological Weapons Convention, which banned biological weapons globally.
While Nixon's decision was influenced by multiple factors—including the impending Biological Weapons Convention negotiations and changing military doctrine—the Dugway incident featured prominently in internal deliberations. The episode demonstrated that maintaining offensive chemical programs created tangible domestic risks without corresponding strategic benefit.
Documents declassified in the 1990s and 2000s filled in details that remained obscure for decades. The March 13 test had released approximately 320 gallons of liquid VX over the target area. The test plan had been approved based on ground-level meteorological readings showing light, favorable winds. But the F-4 released its payload at altitude where different atmospheric conditions prevailed. Stronger winds from an unexpected direction carried the VX plume north-northeast.
The VX droplets—designed for battlefield persistence—settled on vegetation across a swath of desert extending well beyond Dugway's boundaries. Sheep grazing in Skull Valley ingested contaminated vegetation or absorbed the agent through their wool and skin. The lethal dose for sheep is lower than for humans, and the animals' grazing behavior maximized exposure.
Declassified safety reports revealed that similar off-range drift incidents had occurred previously at Dugway, though none with consequences as dramatic as the sheep kill. Wildlife deaths near the proving ground's boundaries had been documented but attributed to various causes. No systematic monitoring for off-range agent migration existed.
The documents also confirmed ranchers' long-standing complaints: Dugway had conducted open-air testing of chemical agents including sarin, soman, and VX throughout the 1960s, averaging several tests per week during peak periods. The proving ground's 800,000-acre buffer zone was theoretically sufficient to contain agents within the military reservation, but the assumption required perfect meteorological prediction and agent behavior—neither of which was achievable.
Despite extensive documentation, significant questions remain unresolved. The full scope of human exposure was never definitively established. Health surveys captured acute symptoms but couldn't assess long-term effects. No systematic epidemiological study tracked Skull Valley residents for cancer, neurological conditions, or other potential late-onset effects of nerve agent exposure.
The classification system ensured that much of what occurred at Dugway remained secret even after the sheep kill. Specific details about testing frequency, agent quantities, and other test missions were not disclosed. Ranchers' reports of dead wildlife and strange odors over preceding years suggested the March 13 incident might not have been unique—merely the most visible.
The adequacy of the settlement also remained controversial. One million dollars divided among approximately 30 ranching families barely covered the immediate economic loss of livestock, let alone long-term impacts including property devaluation and lost grazing permits. Several families eventually left Skull Valley, unable to sustain ranching operations amid ongoing concerns about Dugway's activities.
Dugway Proving Ground continues to operate. After Nixon's 1969 policy shift, the facility transitioned from offensive chemical and biological weapons testing to defensive research—developing detection equipment, protective gear, and medical countermeasures. Open-air testing of lethal agents was officially discontinued, though the facility still conducts tests using simulants and at smaller scales.
The sheep kill incident fundamentally changed how the military approached chemical testing. Enhanced meteorological requirements, expanded safety zones, and notification protocols were implemented. The incident also accelerated the broader policy shift away from offensive chemical weapons. By the time the Chemical Weapons Convention entered into force in 1997, the U.S. had committed to destroying its entire chemical weapons stockpile—a process completed in 2023.
But the Dugway incident remains relevant beyond its historical significance. It illustrates the inherent tension between military secrecy and public safety, the difficulties of assessing exposure after classified agent releases, and the challenges of establishing accountability when government operations harm civilians. These issues recur in debates about burn pits, Agent Orange exposure, and Gulf War illness—situations where military personnel and civilians were exposed to hazardous substances but classification and bureaucratic resistance delayed recognition and compensation.
The sheep of Skull Valley died within 48 hours of exposure. The full consequences for the people who lived through March 1968—and the precedents established in how the government responded—continue to echo decades later.
What can be stated with certainty, based on declassified documents and confirmed investigations: On March 13, 1968, Dugway Proving Ground conducted an open-air spray test of VX nerve agent. Meteorological conditions carried the agent beyond intended target zones. Approximately 6,400 sheep died in adjacent Skull Valley within 48 hours. The Army denied responsibility for three weeks. Independent veterinary analysis confirmed organophosphate nerve agent poisoning. The Army eventually acknowledged the connection. The government settled with ranchers for approximately $1 million without admitting legal liability.
The incident contributed to President Nixon's decision to terminate offensive biological weapons programs and restrict chemical weapons development. It accelerated the policy trajectory toward the Chemical Weapons Convention. It exposed systemic safety failures in military testing protocols and the inadequacy of classification systems that prevented effective public health response to chemical incidents.
Six thousand dead sheep in a remote Utah valley became the visible evidence of a secret program that had operated for decades without public awareness. The documentation is extensive, the causation established, the policy consequences significant. The sheep kill at Skull Valley remains the most dramatic confirmed incident of chemical weapons testing harming American civilians—and the only one where the government eventually, reluctantly, acknowledged what happened.