Late October in northern Virginia carries a specific quality of light, thin and metallic, as if the sun itself has begun to withdraw from the coming winter. On the morning of October 30, 1989, this light fell across the loading dock of a commercial building on the outskirts of Reston, Virginia, where a freight truck from John F. Kennedy International Airport had arrived after driving through the night. The building belonged to Hazelton Research Products, a company that imported nonhuman primates for American scientific laboratories, and the truck carried one hundred wooden crates from the Philippines, each containing crab-eating macaques captured from the forests of Mindanao, transported by air through Amsterdam, and finally to this anonymous address where they would spend thirty days in quarantine before being sold to researchers across the country.
The macaques that emerged from those crates on that Monday morning could not have known that they had already been exposed to something lethal during their journey, that a virus had traveled with them across three continents, that their destination was not merely another cage but a historical moment that would change how humanity understood one of the most feared pathogens on earth. They were simply animals, frightened, dehydrated, huddled in corners of crates that smelled of urine and fear. Workers in blue coveralls moved them into a holding room designated with the letter F, a space with concrete floors, stainless steel cages stacked five high, and a ventilation system that connected to adjacent rooms through ductwork that would soon become evidence in an epidemiological mystery.
Within seventy-two hours, the first deaths occurred. Not the gradual decline of illness that veterinarians recognize, but sudden collapse. A macaque found in the morning hanging from the cage bars, motionless, blood crusted around its nostrils. Another discovered in the corner of its cage, having hemorrhaged from its rectum through the night. By November 7, when Hazelton’s staff began documenting the pattern, they noted symptoms that defied their experience: epistaxis, hematemesis, bloody diarrhea, and most disturbingly, blood that refused to clot when drawn for testing, remaining liquid in collection tubes where it should have gelled within minutes of contact with air.
Thomas Geisbert, a twenty-seven-year-old civilian researcher employed at the United States Army Medical Research Institute of Infectious Diseases in Frederick, Maryland, received a phone call on November 13. Geisbert was completing his doctoral dissertation in microbiology while working full-time at USAMRIID’s electron microscopy unit, and he had developed a reputation for skill in identifying filoviruses—the family that includes Ebola and Marburg—from tissue samples. The call came from Peter Jahrling, a senior USAMRIID virologist who had been contacted by Hazelton’s veterinary staff. Something unusual was killing monkeys in Reston. Could Geisbert examine samples?
The tissue arrived that afternoon. Geisbert processed them through the negative-staining protocol he had perfected, applying phosphotungstic acid to viral particles and examining them under transmission electron microscopy at 50,000x magnification. What he saw in the darkroom that evening, developing the photographic plates by hand, stopped his breath. Filamentous particles, some straight, some curved into the distinctive shepherd’s crook shape that characterizes filoviruses. Ebola, or perhaps Marburg. In Virginia. In monkeys. In a commercial quarantine facility with connections to the international primate trade.
By November 28, when Geisbert’s identification was confirmed by additional testing at USAMRIID and the Centers for Disease Control and Prevention, the situation had deteriorated. Mortality in Room F had exceeded fifty percent, and more disturbingly, animals in adjacent rooms—Room H, Room J—were showing signs of illness despite never having shared cages with infected monkeys, never having touched contaminated blood, never having been handled by the same staff without protective equipment. The only connection between these animals was air. The ventilation system that maintained negative pressure and filtered exhaust air nonetheless recirculated portions of the supply air between rooms, and viral particles small enough to remain suspended had traveled through ductwork, establishing infection in new hosts through the respiratory route.
Colonel David Huxsoll, commander of USAMRIID, faced a decision. The building in Reston was not a military facility. It was a commercial operation in a suburban office park thirty miles from Washington, D.C. Yet the threat required military containment protocols. On December 7, 1989, Huxsoll authorized Operation Reston—the military takeover of a civilian facility. Soldiers in protective gear joined civilian veterinarians. Two hundred macaques were euthanized over three days. Some received lethal injection of sodium pentobarbital. Others were placed in carbon dioxide chambers. Their bodies were triple-bagged in plastic biohazard pouches, loaded into military vehicles with refrigerated cargo compartments, and transported to Fort Detrick where they were incinerated at 1,800 degrees Fahrenheit, temperatures sufficient to destroy viral RNA.
Four animal handlers—Elisa, Lisa, Paul, and John, their surnames protected in subsequent CDC reports—tested positive for antibodies to the virus. None developed clinical symptoms. This was the first indication that Reston ebolavirus, as the new species would be named, differed critically from its African cousins. It killed macaques with apparent one hundred percent lethality. It spared humans, at least these four humans, from the hemorrhagic fever that had killed 280 people in Kikwit, Zaire, in 1995, and 88 percent of patients in earlier outbreaks of Zaire ebolavirus.
The Reston facility was decontaminated with bleach and formaldehyde, renovated, and eventually returned to commercial use. Today the building houses a yoga studio, a software development company, a dental supply firm. No plaque marks the site. No memorial acknowledges the two hundred macaques who died there, or the four humans who carried antibodies, or the virus that demonstrated for the first time in an American setting that Ebola could transmit through air.
But the question that emerged from Room F in November 1989—can Ebola travel through the respiratory route?—did not disappear with the incineration of the last monkey. It migrated to Fort Detrick. It entered the research agenda of USAMRIID. It became the foundation for three decades of experiments that would consume thousands of primate lives, hundreds of millions of dollars, and the careers of scientists who dedicated themselves to understanding whether Ebola could be weaponized, and whether vaccines could protect against such weaponization, and whether the very act of studying this possibility was creating risks greater than the threat it sought to prevent.
Military interest in biological aerosols predates Reston by decades. Operation Whitecoat, conducted at Fort Detrick from 1954 to 1973, exposed over seven thousand volunteer soldiers to aerosolized biological agents including Francisella tularensis and Coxiella burnetii, studying infection rates and incubation periods for defensive purposes. The Biological Weapons Convention of 1972 nominally ended offensive biological weapons research, but Article VII explicitly permitted defensive research—understanding how pathogens might be weaponized in order to develop countermeasures. This loophole, necessary for public health preparedness, created the intellectual and institutional space for the research that would follow.
By 1995, USAMRIID had established itself as the premier American laboratory for high-consequence pathogen research. Biosafety Level 4 containment—the highest level, requiring positive-pressure suits with independent air supplies, multiple airlocks with chemical showers, and specialized waste treatment systems capable of sterilizing liquid effluent at 121 degrees Celsius—was fully operational. The Center for Aerobiological Sciences had developed technology for generating and controlling biological aerosols with precision previously unavailable to civilian researchers. Nebulizers manufactured by BGI Inc. of Waltham, Massachusetts, could produce particles of specific sizes with coefficients of variation below fifteen percent. Exposure chambers could deliver measured doses to animal subjects while monitoring respiratory rate, heart rate, and body temperature in real time. The infrastructure existed to ask questions that would have been unthinkable to pursue in less secure settings.
E. D. Johnson, Nancy Jaax, James White, and Peter Jahrling published their seminal paper in 1995 in the International Journal of Experimental Pathology. “Lethal experimental infections of rhesus monkeys by aerosolized Ebola virus.” Eighteen animals. Eighteen deaths. The methodology was precise, reproducible, and terrifying in its implications. The researchers used a 3-jet collision nebulizer to generate particles between 0.8 and 1.2 microns in diameter—a size range specifically selected because it penetrates deep into the alveoli, reaching the gas-exchange surfaces of the lungs where particles encounter type I and type II pneumocytes, cells that express receptors capable of admitting Ebola virus.
Each macaque received approximately 1,000 plaque-forming units of Ebola virus (Zaire strain, Mayinga variant) suspended in these optimally sized droplets. The exposure lasted ten minutes. The animals were awake, conscious, breathing normally. They inhaled the virus as they would inhale pollen or dust. Then they were returned to their cages in a Biosafety Level 4 containment suite at Fort Detrick.
Death followed a predictable timeline that would become familiar to researchers over subsequent decades. Day 0: exposure. Day 3: fever onset, viremia detectable by polymerase chain reaction in blood samples. Day 5: peak viremia, typically reaching 10^6 to 10^8 plaque-forming units per milliliter of blood. Day 6: clinical signs including depression, anorexia, and petechial hemorrhages on mucous membranes. Day 7-9: moribund state, requiring euthanasia under Institutional Animal Care and Use Committee protocols that mandate termination before prolonged suffering. Day 10: necropsy, tissue harvest, viral quantification, histopathological analysis.
The pathology differed subtly but significantly from animals infected by intramuscular injection. In injected animals, the primary site of initial replication was muscle tissue, with subsequent dissemination through lymphatic channels to regional lymph nodes. In aerosol-exposed animals, the lungs bore the initial burden. Viral antigen concentrated in pneumocytes. Interstitial edema developed early. Fibrin deposits formed in alveolar spaces. Before the liver failed, before the spleen liquefied, the lungs began drowning in their own fluids. This was the signature of aerosol Ebola: pulmonary hemorrhage preceding systemic dissemination, a distinct anatomical progression that produced the same ultimate outcome—death—but through a different physiological mechanism.
Johnson and colleagues established the baseline. Aerosolized Ebola produced disease indistinguishable in severity from injected Ebola. The paper concluded with a statement that would justify decades of subsequent research: “These data suggest that Ebola virus may be capable of producing lethal infection in primates following aerosol exposure, supporting the need for continued development of medical countermeasures against this potential route of exposure.”

The word “potential” carried weight. No evidence existed in 1995 that Ebola had been weaponized. No state had admitted to aerosol Ebola research. The Soviet Biopreparat program had weaponized Marburg virus, a filovirus cousin, and had reportedly conducted experiments with Ebola, but the extent of their aerosol work remained classified behind the remnants of the Iron Curtain. The threat was theoretical. The research was actual. And the research required creating the threat in order to study it.
Every experiment requires standardized reagents. For Ebola aerosol research, the standard became a virus stock derived from a 65-year-old female patient in Kikwit, Zaire, during the 1995 outbreak that killed 280 people. The clinical specimen, designated CDC SPBLOG 9510621, traveled from the Democratic Republic of the Congo to Atlanta, then to Frederick.
At USAMRIID, virologists performed serial passaging—growing the virus in Vero E6 cell culture, harvesting the progeny, growing again—four times. This created a master seed stock with consistent, reproducible characteristics. They designated it R4368. Passage 4. Derived from the Kikwit outbreak. Genetically stable. Uniformly lethal in rhesus macaques at doses above 500 plaque-forming units. The complete genome was sequenced, deposited in GenBank under accession number JQ352763.1. Every nucleotide was known. Every virion was standardized.
From July 2011 to December 2014, R4368 served as the reference challenge stock for USAMRIID’s Ebola research program. When researchers tested vaccines, they challenged with R4368. When they tested monoclonal antibodies, they challenged with R4368. When they tested antiviral compounds, they challenged with R4368. Success meant survival despite R4368. Failure meant death from R4368.
In 2014, R4368 was retired. A new stock, R4415, passage 3 of the same original Kikwit isolate, replaced it. The transition required validation: genome sequencing to confirm identity, virulence testing in small groups of primates, comparison studies to ensure that results obtained with R4415 remained comparable to the historical database accumulated with R4368. This validation consumed twenty additional macaques, animals who died simply to prove that the new stock killed as reliably as the old one.
The Kikwit lineage—R4368, then R4415—became the currency of Ebola countermeasure development. Every candidate had to prove itself against these stocks. Most failed. Vaccines that protected mice failed in macaques. Antibodies that neutralized virus in test tubes failed in living animals. The aerosol challenge model was unforgiving. It separated true protection from laboratory artifacts with ruthless efficiency.
Rhesus macaques do not volunteer. They are bred, purchased, and consumed. Since 1979, the National Institutes of Health has maintained a breeding colony on Morgan Island, South Carolina, a 4,400-acre marshland where approximately 4,000 primates are born annually specifically for research purposes. Charles River Laboratories operates the facility under federal contract. The animals are specific-pathogen-free—tested negative for simian immunodeficiency virus, simian retrovirus type D, herpes B virus, and other pathogens that might confound experimental results.
Juvenile males and females, weighing 3 to 5 kilograms, enter the research system at age two to three years. They are shipped by climate-controlled truck to Fort Detrick, Galveston, Bethesda. Upon arrival, thirty-day quarantine. Blood draws. Pathogen testing. Acclimation to solitary housing in stainless steel cages measuring 2 feet by 2 feet by 3 feet.
Between 2011 and 2020, approximately 12,000 rhesus macaques left Morgan Island for biodefense research. Roughly 40 percent—nearly 5,000 animals—were destined for Ebola studies. Each challenge experiment typically required 16 to 24 animals: experimental groups, positive controls, negative controls. At $15,000 to $20,000 per animal in total program costs including acquisition, housing, veterinary care, and pathology, a single study might consume $300,000 to $500,000 in primate costs alone, before accounting for BSL-4 facility overhead, personnel salaries, reagent expenses, or waste disposal.
The mathematics of cumulative sacrifice are stark. Published Ebola aerosol studies from USAMRIID between 1995 and 2015—approximately forty papers—consumed 480 to 960 animals. Post-2015, following the West African outbreak, funding increased dramatically. BARDA funded twelve aerosol challenge studies between 2016 and 2020, consuming 180 animals. NIAID’s intramural program conducted an estimated fifteen additional studies, consuming 225 animals. The University of Texas Medical Branch, collaborating with USAMRIID, published eight aerosol studies, consuming 120 animals.
Conservative total for published research: 1,500 primates. Unpublished government research—estimated at twice the published volume based on funding allocations and personnel reports—adds 3,000 animals. Pre-1995 research, including method development and the original Johnson study, adds 500 to 1,000 animals. Cumulative minimum: 5,000 rhesus macaques. Maximum estimate, including classified programs: 9,000 to 10,000 animals.
Total program cost: $90 million to $200 million in direct animal costs, plus facility construction (BSL-4 laboratories cost $500 to $1,000 per square foot to construct), plus personnel (Ph.D. scientists earning $80,000 to $150,000 annually, veterinarians, animal technicians, safety officers), plus reagents, plus incineration services. The full aerosol Ebola research program likely consumed $500 million to $1 billion between 1989 and 2026.
The Documents of 2016
2015 marked a watershed. Two distinct aerosol Ebola studies occurred that year, conducted by overlapping research groups at overlapping institutions, funded by the same agency—the National Institute of Allergy and Infectious Diseases—through different mechanisms. Their results, when compared, suggest complexities that the public literature does not fully capture.
The first study—published, peer-reviewed, celebrated—examined an aerosolized vaccine. Alexander Bukreyev at the University of Texas Medical Branch led the research, collaborating with Thomas Geisbert (now at UTMB, formerly USAMRIID), Peter Collins at NIAID, and Gene Olinger at USAMRIID. They tested HPIV3/EboGP—a recombinant human parainfluenza virus type 3 expressing Ebola glycoprotein—delivered as an aerosol or liquid to the respiratory tract.
Eighteen rhesus macaques participated in two cohorts. Study 1 examined immunogenicity. Study 2 examined protection against lethal challenge. On day 55, all animals in Study 2 received intramuscular challenge with 1,000 PFU of Ebola virus (Kikwit strain).
Results were positive. All vaccinated animals survived. The aerosol vaccine induced robust mucosal IgA and IgG in bronchoalveolar lavage fluid. Lung-resident T cells, identified by CD103 expression, showed polyfunctional responses. The paper, published in the Journal of Clinical Investigation on August 3, 2015 (DOI: 10.1172/JCI81532), concluded that “a single administration of aerosolized HPIV3/EboGP completely protects animals against death and severe disease.”
This was the public face of NIAID’s aerosol Ebola research. Successful. Protective. Ready for clinical translation.
Behind this publication lay another study.
According to documents released by Senator Rand Paul in September 2026, Anthony Fauci recorded concerns in his personal diary regarding a separate 2015 USAMRIID experiment that exposed vaccinated primates to aerosolized Ebola virus challenge—not intramuscular challenge, but respiratory exposure.

The alleged diary entry, dated March 7, 2016, describes a study with twenty animals. Sixteen received experimental vaccines. Four served as unvaccinated controls. All were exposed to aerosolized Ebola virus using standard ABES-II protocol: 1,000 PFU, 0.8-1.2 micron particles, ten-minute exposure.
Both unvaccinated controls died. Expected. But sixteen of the twenty vaccinated animals also died—eighty percent mortality in the vaccinated cohort.
According to Fauci’s alleged notes, the vaccinated animals showed distinctive pathology. Their lungs contained “necrosis, inflammation and fibrin” exceeding what was observed in unvaccinated controls. The controls died of classic hemorrhagic fever—liver failure, coagulopathy, shock. The vaccinated died of pulmonary destruction—their lungs filled with inflammatory debris, fibrin thrombi, and fluid. They drowned from the inside out.
Fauci allegedly wrote: “What idiots those guys at USAMRIID are. This should have been a classified experiment that never should have been done in the first place.”
The documents suggest Fauci’s concern was multifaceted. Scientific: vaccinated animals experienced enhanced pathology. Reputational: the West African outbreak was ongoing, human vaccine trials were proceeding in Guinea and Liberia, and news of vaccinated primates dying from aerosol challenge would complicate these efforts. Regulatory: if aerosol challenge produced enhanced disease in vaccinated animals, this represented a safety signal requiring investigation and disclosure to human trial participants.
Yet the study was not classified. Results were not published, but neither were they destroyed. They existed in the gray zone—known to insiders, unknown to the public, unreviewed by independent ethicists, undisclosed to human trial participants who were injecting experimental vaccines into their bodies without knowledge that vaccinated primates had experienced eighty percent mortality following aerosol challenge.
The phenomenon suggested by Fauci’s notes—vaccine-associated enhanced respiratory disease—has precedents in vaccinology. In the 1960s, formalin-inactivated respiratory syncytial virus vaccine caused enhanced disease in children who later encountered wild-type virus. The mechanism involves non-neutralizing antibodies that bind viral antigen without preventing cellular entry. The Fc portion of these antibodies engages Fc receptors on macrophages, facilitating viral uptake and triggering inflammatory cascades.
Similar enhancement occurred with some influenza vaccines in animal models. The 2016 Dengvaxia scandal in the Philippines—where the dengue vaccine was associated with enhanced disease in seronegative children—demonstrated that this risk extends to approved products, not merely experimental candidates.
For Ebola, the hypothetical mechanism would involve vaccine-induced antibodies against glycoprotein that fail to neutralize aerosolized virus. The virus enters lung epithelial cells. Antibody-virus complexes form. Complement activates. Neutrophils infiltrate. The lung becomes a battlefield where the immune system, primed by vaccination, overreacts. Result: fibrin deposition, necrosis, death from respiratory failure rather than hemorrhagic shock.
Whether this occurred in the 2015 USAMRIID study cannot be independently verified. The Fauci documents, if authentic, suggest it did. The published JCI study, using intramuscular challenge, showed no enhancement—only protection.
The discrepancy highlights a critical variable: route of challenge matters. A vaccine that protects against injected Ebola may not protect against, and might worsen, inhaled Ebola. This distinction was not communicated to participants in human vaccine trials.
In 2026, questions persist. Why was the study not classified? According to the Paul documents, NIAID review found that “classification had never been raised while the study was being designed and conducted.” For research involving aerosolized Ebola—a Category A bioterrorism agent, a Select Agent under federal law—this omission is remarkable.
Whatever the reason, information asymmetry resulted. Human trial participants in Guinea, Liberia, Sierra Leone, and the United States were not informed that vaccinated primates had experienced eighty percent mortality following aerosol challenge. Informed consent documents emphasized protection against injected virus. They omitted potential enhancement following respiratory exposure.
At Fort Detrick, Building 1425 houses the USAMRIID repository. Minus eighty degrees Celsius. Liquid nitrogen backup. Thousands of vials: R4368, R4415, Makona variants, Sudan virus, Marburg Musoke. Every strain that killed primates in the aerosol chamber sits in suspended animation.
Freezer 47 contains the Ebola stocks. Dual combination locks. Access logs. The freezer hums continuously—a sound that becomes an emergency only when it stops.
In 2019, it stopped. Compressor failure. Temperature rose to minus sixty degrees before restoration. No vials thawed. No virus escaped. But the incident triggered CDC inspection, which revealed wastewater treatment failures and biosafety protocol violations. USAMRIID closed for months. Then reopened. The research resumed.
The justification for three decades of aerosol Ebola research rests on the Biological Weapons Convention’s allowance for defensive research. The threat is theoretical: someday, someone might weaponize Ebola. The research is actual: thousands of primates dead, hundreds of millions of dollars spent, dangerous pathogens created and maintained.
Yet no licensed Ebola vaccine has been specifically tested or approved for protection against aerosolized virus. rVSV-ZEBOV, approved by the FDA in 2019, was tested against intramuscular challenge. Whether it protects against aerosol exposure remains unknown.
The cycle continues. Create the weapon to test the shield. The shield fails. Create more weapons to test better shields. The primates keep dying. The freezers keep humming. The threat remains theoretical. The deaths remain actual.
In Reston, the building where it began now houses a yoga studio. No plaque marks the site. No memorial acknowledges the two hundred macaques who died there, or the nine thousand who followed them in laboratories across Maryland, Texas, and South Carolina.
At Fort Detrick, Freezer 47 still hums. The R4415 stock is still viable. Technicians still suit up in positive-pressure suits. Animals still arrive from Morgan Island, enter the ABES-II chamber, and breathe virus.
The experiment continues.
Sources and Documentation
The 1989 Reston outbreak is documented in CDC Morbidity and Mortality Weekly Report (December 1, 1989, Vol. 38, No. 46) and Richard Preston’s 1994 book The Hot Zone (Random House).
The 1995 aerosol Ebola study (“Lethal experimental infections of rhesus monkeys by aerosolized Ebola virus”) was published in the International Journal of Experimental Pathology (Volume 76, 1995, pages 227-236) by E.D. Johnson, Nancy Jaax, James White, and Peter Jahrling.
The 2015 aerosol vaccine study (“Aerosolized Ebola vaccine protects primates and elicits lung-resident T cell responses”) was published in the Journal of Clinical Investigation (Volume 125, Issue 8, August 3, 2015, DOI: 10.1172/JCI81532).
The R4368 and R4415 challenge stocks were characterized in PLOS ONE (2016, DOI: 10.1371/journal.pone.0150919).
Documents allegedly from Anthony Fauci’s diary and NIAID internal communications were released by Senator Rand Paul in September 2026. The authenticity of these documents has not been independently verified.
Cumulative primate numbers and costs are estimates based on published studies, federal funding records, and institutional reports. Exact figures for classified research remain undisclosed.






































