Anthrax Redux: Did the Feds Nab the Wrong Guy?

[Eugen Leitl]

http://www.wired.com/magazine/2011/03/ff_anthrax_fbi/all/1

Anthrax Redux: Did the Feds Nab the Wrong Guy?

By Noah Shachtman   March 24, 2011  |  4:40 pm  |  Wired April 2011

Illustration: GoC1i Montes

Finally, the investigation was over. The riddle solved. On August 18,
2008bafter almost seven years, nearly 10,000 interviews, and millions of
dollars spent developing a whole new form of microbial forensicsbsome of the
FBIbs top brass filed into a dimly lit, flag-lined room in the bureaubs
Washington, DC, headquarters. They were there to lay out the evidence proving
who was responsible for the anthrax attacks that had terrified the nation in
the fall of 2001.

It had been the most expensive, and arguably the toughest, case in FBI
history, the assembled reporters were told. But the facts showed that Army
biodefense researcher Bruce Ivins was the person responsible for killing five
people and sickening 17 others in those frightening weeks after 9/11. It was
Ivins, they were now certain, who had mailed the anthrax-filled letters that
exposed as many as 30,000 people to the lethal spores.

The FBI unraveled the mystery, officials said, thanks in part to the
microbiologists seated at a U-shaped table in the front of the room. Among
them was Paul Keim, who first identified the anthrax strain used in the
attacks, and genetic specialist Claire Fraser-Liggett, who led the team that
sequenced the DNA of the anthrax in the letters, tracing the spores back to
their genetic match: a flask of superconcentrated, ultrapure anthrax held by
Ivins. Several of the researchers at the table had previously counted Ivins
as a peer and even a friend. Now they were helping brand him a monster.

Between the officials and the scientists, it was a convincing display. It had
to be. Ivins had killed himself three weeks earlier. There would be no
arrest, no trial, no sentencing. Absent a courtroom and a verdict to provide
a sense of finality or some measure of catharsis, all the FBI could do was
present its findings and declare the case closed.

No one involved that day expressed any doubt about Ivinsb guilt. But things
are not always as clear-cut as they may seem in an FBI presentation. Two
years later, sitting in her office overlooking West Baltimore, Fraser-Liggett
concedes she has reservations. bThere are still some holes,b she says,
staring out her window in discomfort. Nearly 2,000 miles away in Flagstaff,
Arizona, Keim has his own concerns. bI donbt know if Ivins sent the letters,b
he says with a hint of both irritation and sadness. Even agent Edward
Montooth, who ran the FBIbs hunt for the anthrax killer, says thatbwhile hebs
still convinced Ivins was the mailerbhebs unsure of many things, from Ivinsb
motivation to when he brewed up the lethal spores. bWe still have a difficult
time nailing down the time frame,b he says. bWe donbt know when he made or
dried the spores.b In other words, itbs been 10 years since the deadliest
biological terror attack in US history launched a manhunt that ruined one
scientistbs reputation and saw a second driven to suicide, yet nagging
problems remain. Problems that add up to an unsettling reality: Despite the
FBIbs assurances, itbs not at all certain that the government could have ever
convicted Ivins of a crime.

It took weeks for anyone to realize the attacks were even happening. When
Robert Stevens, a photo editor at the Sun tabloid, came down with chills
while on vacation in North Carolina on September 29, 2001, neither he nor his
wife figured it was a big deal. She went to spend the afternoon with their
daughter; he rested on the couch. As Stevens lay there, thousands of spores
were filling his lungs. Nestled in the respiratory sacs, the particles slowly
came in contact with white blood cells called macrophages, which carried the
bacteria into the lymph nodes in the central cavity of his chest. There, the
spores began to germinate, shedding their tough outer layer and multiplying
relentlessly. The bacteria unleashed two types of toxins into Stevensb
bloodstream. One caused a fluid buildup in his central chest cavity, which
squeezed his heart and shoved his lungs against his ribs, making it difficult
to breathe. The other began killing off Stevensb macrophages, decimating his
bodybs natural defenses.

Two days later, Stevens was feverish, short of breath, and red in the face.
He and his wife started driving back to their home in Florida, with Stevens
sweating in his seat. When they got there, Stevensb wife took him, nearly
incoherent now, to the hospital. On October 4, he tested positive for
anthrax. He died the next day.

Both medical and public officials figured it was a freak but natural
occurrence, maybe something Stevens picked up on a hike in the Carolina
woods. Because anthrax was known as a potential biological warfare agent, the
case grabbed national attention, but there seemed no reason to panic. bIt
looks like itbs a very isolated incident,b President George W. Bush told the
country on October 9.

Three days later, an anthrax-loaded letter was found at NBC News
headquarters. A week after that, the FBI recovered an identical-looking
spore-laden letter from the offices of the New York Post. b09-11-01,b the
letters read. bthis is next / take penacilin now / death to america / death
to israel / allah is great.b

By mid-October, four more people had been diagnosed with anthrax, and Leroy
Richmond, a postal worker at the Brentwood Road mail facility in Washington,
DC, was doing his best to convince his coworkers to relax. Mail handlers had
little to fear as long as they took precautions, according to talking points
distributed by the government, which Richmond read aloud to several fellow
mail room employees. Bacillus anthracis will appear as a white powder that
you should keep away from your face. So be careful, but keep doing your jobs,
Richmond told his coworkers as he wiped away a sniffle. Everything is going
to be just fine.

A few days later, Richmond barely had the lung capacity to walk across the
floor of the mail room, which routes correspondence to the US Senate and
dozens of federal agencies. His shoulders and chest ached like they had been
hit with a bat. bOh my God,b he thought. bThis might be the last breath that
I can take.b On October 20, doctors at Inova Fairfax Hospital confirmed that
he had been infected with anthrax.

Richmond survived. His coworkers Thomas Morris and Joseph Curseen did not.

The mailer was probably a spore-happy Unabomber typeba highly trained,
educated loner, likely in the US biodefense community.  The letter that
likely infected them was discovered in the offices of Tom Daschle, the Senate
majority leader. The envelope had a Trenton, New Jersey, postmark and a
return address that read in handwritten block letters, b4th grade greendale
school.b An FBI team wearing protective suits placed the letter and envelope
in ziplock bags. They then drove the package 50 miles northwest, to Fort
Detrick, home of the militarybs leading biodefense facility, the United
States Army Medical Research Institute of Infectious Diseasesbthe famed
USAMRIID.

John Ezzell, the bearded, Harley-driving head of the institutebs Special
Pathogens Sample Testing Lab, was waiting to meet the agents. He spent a day
studying the package, then sent it to the lab of Bruce Ivins, one of the
institutebs most experienced anthrax researchers. As his fellow
microbiologists watched from the hallway, Ivins shoved his double-gloved
hands inside a biosafety cabinet containing the sample. He opened the bag and
held it up with one hand. When he moved his free hand closer to it, the
granules in the bag began moving toward his glove, drawn by a slight
electrostatic charge. The microbiologists gasped; they were used to working
with wet spores, which fell to the ground easily. But this anthrax was dry
and ionizedbit would stay aloft and spread like a gas. It was potentially
lethal to anyone in the vicinity. bItbs unbelievable,b a colleague remembers
Ivins saying. bIbve never seen anything like that.b


Bruce Ivins Illustration: GoC1i Montes Ivins measured the concentration of the
sample. It came out to a trillion spores per grambthree orders of magnitude
more dense than anything the USAMRIID researchers had ever grown. bThese are
not bgarageb spores,b Ivins later wrote in an analysis. bProfessional
manufacturing techniques were used.b

If anyone at USAMRIID knew about spore-growing, it was Ivins. After two
decades at the institute, he had mastered the delicate balance of chemistry,
heat, and time required to get anthracis to multiply in just the right way.
Ivins not only supplied spores to his fellow scientists at USAMRIID; many of
the anthrax researchers on the planet relied, in one way or another, on his
concoctions.

In addition to his anthrax expertise, Ivins was known as one of the odder
characters in an institute full of odd characters. Hebd show up to work in
plaid bell-bottoms and flowered shirts that ran a few sizes too small for his
soda-straw frame. He was a juggler, a unicyclist, and a weather junkie. At
institute Christmas parties, Ivins was the one reciting the corny limericks.
At the base gym, he was the one working out in dark socks and heavy boots.

Ivins grew up in the small town of Lebanon, Ohio, the skinny, science-nerd
son of the local drugstore pharmacist. He went to school at the University of
Cincinnati, where he met his wife, Diane, and earned a bachelorbs degree, a
masterbs, and a PhDball in microbiology. After Ivinsb postdoctoral work at
the University of North Carolina, the couple moved to Maryland, where Ivins
started working at USAMRIID in 1981. Two years earlier, an anthrax outbreak
at a secret military microbiology plant in Siberia had killed at least 66
people, proving that the Soviets had been refining anthrax into a biological
weapon. Ivins was assigned to start working on a new, more effective vaccine.

Ivins was a fixture in the institutebs active social and intramural sports
scenes. Diane largely stayed away, even though they lived just up the street
from the base with their adopted kids, Andy and Amanda. (The Ivins family did
not respond to repeated interview requests.) So it was Ivins alone who
frequented the institutebs volleyball gamesbnot to play but to cheer and razz
the refs. Then hebd go to the postgame drinking sessions, held at Fort
Detrickbs old officersb club, and nurse a single glass of wine while everyone
else got sloshed.

Ivins liked to keep candy on his desk and would talk and talk and talk to
anyone bold enough to grab a piece. When he got flusteredbwhich was
oftenbhebd stammer and flap his arms in hopes of making his point. If you
found a lame joke or a picture of kittens in your inbox, you knew who to
blame. He led what his friends called a bhippie Massb at St. John the
Evangelist Catholic Church in nearby Frederick, playing keyboards and
acoustic guitar. He was a local Red Cross volunteer. Colleagues found him
both smart and generous. bHe had experience and was willing to share,b
remembers Hank Heine, Ivinsb friend and colleague in the tight-knit
bacteriology division. bThe first day I arrived, he said, bIf therebs any
help you need, come see me.bb

Not surprisingly, then, when the FBI set up a small team after the initial
attacks to help with the science of anthrax, they found Ivins eager to
assist. Maybe too eager. While other USAMRIID scientists swore that no one
there could have pulled off the attacks, Ivins suggested several current and
former colleagues as potential perpetrators. bRarely would a guy point the
finger at his coworkers,b says Thomas Dellafera, the US Postal Inspection
Service agent who helped lead the anthrax investigation. bBut he did. This
guy was rolling on his own mother.b

Ivins regularly emailed friends and acquaintances about the burgeoning
investigationbincluding a former UNC grad student he had known named Nancy L.
Haigwood. One picture Ivins sent Haigwood, now also a microbiologist, was
particularly unsettling: It showed Ivins holding petri dishes filled with
anthracisbwithout wearing gloves. Haigwood was already thoroughly unnerved by
Ivins: He had shown an excessive fascination with both her and the rules and
rituals of the Kappa Kappa Gamma sorority she advised back in school, and he
had foisted all manner of unwanted attention on her ever since. For example,
she had long suspected that Ivins was the one who, years after graduation,
spray-painted a fence at her boyfriendbs house with a red kkg in Greek
lettering. When the American Society for Microbiology later emailed a plea to
its members asking for help on the case, Haigwood thought of Ivins. bAt that
exact moment, that awful moment, I knew it was him,b she says. She reported
her feelings to the FBI. The bureau sent out two agents but didnbt seem all
that interested in Haigwoodbs intuition at the time.

The Scientists

Investigating the 2001 anthrax attacks required an unprecedented scientific
effortbone that was mostly conducted by US anthrax researchers. A few of the
key players:


Hank Heine

One of Ivinsb closest friends in USAMRIIDbs bacteriology division

Paul Keim

Identified the attack anthrax as a strain used in many US labs.

Pat Worsham

Identified the four mutants produced by the attack anthrax.

Claire Fraser-Liggett

Led the genetics team charged with bfingerprintingb the anthrax.

Illustration: GoC1i Montes

Sitting on the hood of his Toyota SUV on the edge of the tarmac at Flagstaff
Pulliam Airport, Paul Keim watched the Arizona sky change colors as the sun
went down. A microbiologist at Northern Arizona University, Keim supervised
one of the largest collections of anthrax on the planet, more than 1,000
samples in all. That afternoon, he had received a call from the FBI directing
him to the tarmac to receive one more specimen: anthrax-infected spinal fluid
extracted from the first victim, photo editor Robert Stevens. At about 7 pm,
a Gulfstream corporate jet landed, coming to a stop near where Keim was
parked. While the jetbs engines died down, a blond woman stepped out, strode
across the runway, and handed Keim a box. He took the box and drove straight
to his lab.

Keim was well-known among anthrax researchers for having developed a DNA test
that could tell one form of Bacillus anthracis from another. By looking at
certain sections of the anthrax genome, Keim could find telltale patterns of
repeating nucleotides that would indicate a given strain. Ten hours after the
tarmac handoff, Keim had finished his analysis of the anthrax that killed
Stevens. The signatures indicated a particularly lethal stock known as Ames.
The strain was mainly used in biolabs around the US to test vaccines and
therapeutic treatments. The attack anthrax, in other words, was very likely
US-grown.

When tested, samples from the subsequent attacks also proved to be Ames. It
was the investigatorsb first big break. Actually, it was their only break;
from every other angle, the case looked nearly impossible to solve.

Unlike in a traditional murder case, the victims had little in common. The
letters themselves were cleanbno fingerprints or human DNA. The exact
location of the letter drops was unknown. It was enough to drive postal
inspector Dellafera mad. Because the attacks used the mail system, Dellafera,
who led a seven-person mail theft unit before the attacks, was placed on the
case leadership team. Now the Connecticut native was interviewing hundreds of
postal workers to see if they had noticed any suspicious envelopes. Nobody
had. Soon it became clear that this wasnbt just a whodunit. It was a
wheredunit, whydunit, and howdunit, too.

Classifying the killer anthrax as belonging to the Ames strain gave the
investigation something to go on, but it was helpful only to a point. Tracing
the attack anthrax to its source would require far more precision. More than
a dozen labsband thousands of researchersbworked with Ames anthrax. And
distinguishing one Ames isolate, or sample, from another was tough; they were
all nearly identical genetically. Every Ames isolate in circulation
originates from a single heifer that died in Sarita, Texas, in 1981. If
spores were the equivalent of bullets, Keim knew the caliber of the murder
weapon. But the bullets could have been fired from any of a thousand
different guns.

To have any hope of finding the equivalent of ballistic markings for this
anthrax, scientists would have to go beyond Keimbs tests, which examined just
eight regions of the genome, with a few hundred base pairs each. Theybd need
to decode the entire anthrax genome, some 5.2 million base pairs. So they
turned to Claire Fraser-Liggett.

Dark-haired, with a square jaw and modelbs cheekbones, Fraser-Liggett helped
pioneer the field of genome sequencing. In 1995, she and her fellow
scientists at the Institute for Genomic Research published the first complete
sequence of a bacterium. Her then-husband, J. Craig Venter, led the
private-sector effort to sequence the entire human genome in the late 1990s.
By the time of the anthrax attacks, she and her team had already unraveled
the genetic codes for dozens of microorganismsbincluding the bacteria that
cause Lyme disease and syphilis. And they were already working on anthraxban
Ames sample from Porton Down, a British biodefense lab.

Fraser-Liggettbs team immediately set out to fully sequence the anthrax taken
from Robert Stevensb spinal fluid. The hope was that even a few unique
nucleotide sequences could further identify the specific isolate of the Ames
strain that was used in the attacks. In 2001, though, genetic sequencing was
still so difficult and expensive that the process could take months.


Anthrax growing in a petri dish of sheep's blood.  Photo: Corbis Meanwhile,
the attacks continued. On October 28, a 61-year-old hospital stockroom worker
named Kathy Nguyen became so ill so quickly that she could barely speak by
the time she got to the emergency room. She died three days later, shortly
after being diagnosed with anthrax. Nobody could figure out how she got
infected. Maybe one of the anthrax letters had brushed against a hospital
package somewhere along its route and left spores behind. Then, two weeks
after that, a spore-filled letter addressed to US senator Patrick Leahy was
found in a pile of quarantined mail.

That same day, 94-year-old widow Ottilie Lundgren checked into the hospital
with a slight cough and some weakness. It was nothing major, but the doctors
decided to keep her overnight for observation. She lived alone in the small,
rustic town of Oxford, Connecticut; she didnbt get a lot of company. Four
days later, doctors confirmed that Lundgren had anthrax. She died the next
day.

The CDC warned that btens of thousands and maybe more letters [could] be
potentially at risk for some level of cross-contamination.b The country,
still shell-shocked from 9/11, plunged further into panic. Every newsroom in
New York and every office in DC became a jury-rigged biosafety lab. People
stormed doctorsb offices demanding ciprofloxacin, the powerful antibiotic
used to counter anthrax; some drove to Canada when US pharmacies ran out.
Then the hoaxes started. One antiabortion extremist mailed 554 powder-filled
letters to abortion clinics all over the country. On December 4, newly
appointed homeland security chief Tom Ridge declared that he was again
placing the public bon general alert.b

FBI director Robert Mueller held dailyband sometimes twice-dailybmeetings on
the investigation, demanding progress reports at every turn. bWho was going
to be a brave enough soul to come in and say, bI got nothing?bb one FBI
veteran recalls. But nothing is pretty much what they had.

To Pat Worsham, one of Ivinsb colleagues, the rush of events felt like bthe
world had gone mad.b A bookish anthrax specialist with a measured,
librarianbs air, she was used to USAMRIIDbs placid, well-regulated pace. Now
she found herself at the center of a frenzied national bioterror
investigation.

Worsham had made her reputation in the scientific community by studying
anthracis variants, work that was key to the development of a vaccine. So it
made sense that one of the institutebs technicians, Terry Abshire, would
decide to email Worsham a photo of a petri dish covered with strange-looking
anthracis colonies. Abshire had let spore colonies from the letter to Leahy
grow for two to three days longer than researchers normally do when culturing
anthrax. When she pulled the dish out of the incubator, she noticed that many
of the colonies had a yellowish color instead of the usual light gray tone.
And they seemed to have affected the sheepbs blood that lined the dish,
turning it a sickly green.

Worsham studied the picture, but she wasnbt sure what it meant. So she grew a
second batch from the Leahy letter and, sure enough, it produced several
funny-looking yellowish colonies, too. Ultimately, four predominant morphs
were identified. Two had a bullbs-eye shape, almost perfectly round. A third
was a little smaller and more irregular. The fourth was mostly translucent.
Worsham later grew colonies from the spores that coated the letters to
Daschle and the New York Post. The same mutated colonies appeared again.

Worsham, cautious by nature, didnbt want to jump to conclusions. bOK, they
look alike,b she told colleagues. bThey may not be alike.b But she suspected
that they were unique to the attack anthrax, which might make them the kind
of identifying characteristic investigators were looking for.

The FBI, meanwhile, turned to a more conventional investigative technique:
behavioral profiling. It parsed the anthrax letters with Talmudic precision,
looking for anything that seemed out of place. The lettersb evocation of
September 11, b09-11-01,b used the American style of placing the month before
the day. The bAllah is greatb line at the end of the letter seemed
inauthentic; a devout Muslim would have begun the letter that way and used
the phrase bAllahu akbar.b Profilers concluded that the mailer was likely a
sort of spore-happy Unabomberba highly trained, highly educated loner,
probably in the American biodefense community. In public, hebd be
nonconfrontational, they advised. Instead, he preferred to harass people
anonymously. Neither the recipients of his letters nor the mailbox into which
he dropped them were chosen at random. The letters had a Trenton postmark,
so, the profilers said, the mailer probably lived or worked nearby.

Investigators in New Jersey drew up a list of 621 mailboxes that fed the
Trenton areabs main mail-processing facility. They started with the most
isolated boxes, figuring a local resident would choose a place where they
wouldnbt be seen making an envelope drop. Under cover of night, the team
swabbed box after box for anthrax residue. Box after box came up negative.

Separately, FBI agents were asking biodefense specialists whether they could
imagine anyone as a potential mailer. One name was mentioned regularly:
Steven Hatfill, a physician who had previously done virology work at
USAMRIID. Hatfill was familiar with anthrax; he had gone to medical school
and done his internship in southern Africa, where he had seen patients with
skin infections caused by anthrax. At the defense contractor SAIC, he showed
PowerPoint presentations that sketched out a biological terror attack. The
scenario: anonymously mailed envelopes packed with anthrax. In the months
before the attacks, he had been filling prescriptions for Cipro.

But the thing that made Hatfill a standout in his field was that he was
essentially a jock among nerds, a high school wrestler and former soldier.
While his colleagues limited their publishing to academic periodicals like
The Journal of Infectious Diseases, Hatfill appeared in the conservative
political magazine Insight, pretending to cook up biohazards in his kitchen
as a way to dramatize the threat of homegrown pathogens. Out of thousands of
current and former biodefense workers, Hatfill became the focus of the FBIbs
attention.

Ivins, meanwhile, was becoming the investigatorsb trusted ally, walking them
through the minutiae of the anthrax life cycle. On January 22, 2002, he drew
a diagram that illustrated why mutants were showing up in the attack
sporesband why that was important.


The envelope mailed to US senator Tom Daschle.  Getty Images Anthrax isnbt a
typical bacteria. Itbs almost immortal. As a spore, anthracis can survive in
a dormant state for decades, perhaps even centuries. When it happens to enter
an animal, it springs back to life and starts reproducing. Like all living
things, anthracis produces mutant offspring as it multiplies. But those
mutants have trouble going dormant. When the anthrax loses its host, many
mutants die out and the bacteria returns to a near-pure state. Itbs almost as
if the law of evolution doesnbt apply.

The way the scientific community handles anthrax further inhibits the
mutation process. Ivins, for example, kept anthracis batches that were pure,
strong, and bjust a slant away,b as he put it, from the original strain taken
from that Ames cow. Then Ivins individually picked what he judged to be the
healthiest colonies to pass on to other researchers. In other words, when
anthrax scientists used Ivinsb spores, they limited the possibility of
mutation. They simply kept returning to 1981 over and over again.

The anthracis found in the various letters was totally different from his
stock, Ivins explained to investigators. Worsham had shown that the attack
anthrax produced all kinds of mutants when cultured. This meant that those
spores were already the product of generation after generation of culturing
and reculturing, unlike his straight-from-the-cow stuff; only the extra turns
of the evolutionary cycle could explain the weird growths. bbDaschleb b	  B.I.
cultures,b Ivins wrote on the diagram, referring to his own initials. Agents
jotted down his analysis, which seemed perfectly logicalbif a little
confusingbat the time. They also noted his suggestion to try genetic analysis
to show the difference between his anthrax and the anthrax used in the
letters.

Of course, for genetics to be helpful, the FBI would need to collect a
comprehensive cross-section of anthracis samples to have something to compare
the attack anthrax to. So this is what they decided to do. They set up one
storehouse for samples at Keimbs lab and another in a locked facility at
USAMRIID. Scientists from around the world were asked to send in a bit of
every Ames anthrax sample in their possession. Ivins and Worsham, among
others, were asked to provide input for the submission guidelines.

The procedures included a reminder not to pick just individual, healthy
coloniesbeven though that was standard practice for lab work. Scientists were
told to take a representative mix from each sample, to catch any stray
mutants. Ivins finished handing in his samples in April 2002.

Included in Ivinsb submissions was an unusually potent batch of Ames sporesba
mix he first created in 1997. It was the result of 164 liters of anthrax from
35 different production runs, distilled into a single liter that was nearly
pure. The flask was marked RMR-1029.

That same month, Ivins did something that surprised everyone. For some
reason, he decided to start testing his office and biocontainment labs for
spores. This was a major breach of USAMRIID protocol. Specially trained
testing teams were supposed to handle suspected contaminations.

On April 18, Ivins told his bacteriology division colleagues what he had
done. They promptly freaked out. In addition to being unsafe, Ivinsb actions
could be read as an attempt to cover up potential evidence. bBruce,b his
friend and coworker Jeff Adamovicz told him, byou donbt understand what this
makes you look like.b

The next day, USAMRIID held an institute-wide town hall meeting to discuss
Ivinsb actions and launch a facility-wide spore hunt. bI am now forbidden
from being a bcowboy,bb Ivins emailed a friend later. bI canbt think for
myself, and I canbt do anything without everybody up and down the line
questioning me about it. Ibm sure itbs punishment.b

The FBI decided to have a talk with Ivins. They had more than just the
testing to ask him about. They also had discovered that Ivins liked to spend
an inordinate amount of time alone in his suite of biocontainment labs,
located in the windowless concrete bowels of the institute. Why lock yourself
inside for so long? bI donbt think anyone has any idea how peaceful and quiet
it can be here after hours,b he wrote in an email to a friend after the
questioning. bIn the evenings [suite] B3 may as well be Mars.b

Despite all this, the bureaubs major interest was still fixated elsewhere.
Investigators were digging into the background of Steven Hatfill and finding
inconsistencies. Hatfill, for example, claimed he had a PhD; in fact, the
degree was never awarded. If he was misrepresenting something like that, they
thought, maybe he was keeping other secrets.

In June 2002, agents asked Hatfill if they could swab his apartment for
spores. When they arrived, they found the place surrounded by camera crews.
News channels carried the event live. Even though the search came up empty,
on August 6 attorney general John Ashcroft went on the Today show to declare
Hatfill a bperson of interestb in the anthrax case.

Two days later, after testing more than 600 mailboxes in central New Jersey
for anthrax, investigators finally found one in Princeton with spores inside.
Contrary to what the profilers had predicted, it was at a busy intersection
on the northwest corner of Princeton Universitybs campus, surrounded by
traffic day and night. Agents passed pictures of Hatfill around the
neighborhood, asking whether the guy looked familiar. Nobody recognized him.

In February 2003, anthrax became part of the rationale for invading Iraq.
Colin Powell went to the United Nations Security Council, in part to discuss
the potential bioweapon. bLess than a teaspoonful of dry anthrax in an
envelope shut down the United States Senate,b he said. bSaddam Hussein could
have b& enough to fill tens upon tens upon tens of thousands of teaspoons.b
Two weeks later, Tom Ridge told Americans to buy duct tape and plastic
sheeting to protect themselves from a bioterror attack. Four weeks after
that, the invasion of Iraq began.

As the war got under way, Hatfillbs life was unraveling. He had been fired
from his job at SAIC, and a replacement gig at Louisiana State University
fell apart under Justice Department pressure. He spent his days bremodeling
every room in my girlfriendbs apartment,b he says. bIt took months.b Suicide
was bnever an option,b he later told a reporter from The Atlantic, but at the
time, he was under extreme emotional and mental strain. In August 2003,
Hatfill sued the Justice Department for violating his constitutional rights
and privacy. After Vanity Fair and The New York Times published articles
suggesting Hatfill was guilty, he also sued CondC) Nast (which owns wired as
well as Vanity Fair) and the Times.

In public, the bureau defended its actions. But the more the anthrax unit
investigated Hatfill, the less convinced some agents became of his
involvement. The Cipro prescriptions, the magazine photo, the flawed
rC)sumC)bnone of it made him a killer. bHe fits the general profile,b special
agent Brad Garrett told colleagues after one interview. bBut I donbt see any
real evidence.b

Meanwhile, Fraser-Liggettbs team was still hard at work trying to come up
with a DNA signature for the attack anthrax. To do this they were sequencing
not only the anthrax taken from Stevensb spinal fluid but also the original
Ames strain (taken from Keimbs collection), which could allow them to
pinpoint genetic markers unique to the attack spores. The team spent months
shearing off DNA segments, tagging those segments with fluorescent dyes,
scanning them with a laser, and then using advanced algorithms to reassemble
the base pairs.

By October 2003, they had the results. bOh shit,b Fraser-Liggett said when
she saw the outcome. bThere isnbt a single difference.b As far as the tests
could tell, all of the nearly five and a half million base pairs were the
same, and in the same order. Because of the natural and man-made processes
that slow anthraxbs evolution, the attack bacteria was essentially pure Ames.
The whole idea of using this new DNA fingerprinting technique to find the
attack spores seemed like a dead end.

There were still a few options left. The best of those was to analyze the
mutant anthrax coloniesbthose yellowish, rounded, and blood-busting
specimensbthat Pat Worsham had spotted. There was a chance that these mutants
might contain larger, more detectable genetic differences that would allow
scientists to distinguish them from pure Ames spores. If so, those morphs
might produce a usable DNA fingerprint. The Fraser-Liggett team launched a
new, laborious round of sequencing. bNot all hope was lost,b Fraser-Liggett
says. But she wasnbt optimistic.

For the time being, however, the FBI decided to rely on Worshambs eyes.
Investigators started bringing samples from the FBIbs Ames repository to her,
which she would culture, then check for mutants.

One sample caught her attention. It had the bullbs-eye-shaped variants and
all the other mutants, too. Because it was labeled with an anonymous code,
Worsham didnbt know where the spores had come from. But the FBI knew it was a
subsample of Bruce Ivinsb lethal RMR-1029 batch, which Ivins had previously
provided to the Battelle biodefense lab in Ohio. And this meant something
didnbt add up.

If the Battelle sample produced mutants, Ivinsb RMR-1029 sample should have
as wellbthey were supposed to be virtually identical. But the RMR-1029 from
Ivins tested clean. Investigators wondered if, despite instructions to the
contrary, Ivins may have picked off a few healthy colonies rather than a
representative sample of the RMR-1029 flask. Or maybe he hadnbt submitted
RMR-1029 at all.

In December 2003, while conducting an inventory of one of USAMRIIDs
biocontainment suites, investigators discovered 22 undocumented Ames anthrax
samples. They began to fear that the repository they had spent nearly two
years assembling might have gaping holes in it. So for the first time, the
FBI decided to scour USAMRIID for any vials they had missed.

The institute staff fumed at the searchbongoing experiments would be
disrupted, they shouted. Heine, Ivinsb coworker, decided to exact a bit of
revenge on his FBI handler. While the agent was collecting samples in his
labbdressed in full protective gearbHeine handed her a vial and told her it
was a deadly plague strain. The vial started shaking in the agentbs gloved
hand. Heine cracked up. bThey were entirely dependent on me to identify
everything in every box,b he says. bI couldbve held up a critical piece of
evidence, said it was something else, and put it aside. Therebs no way they
wouldbve known.b

During the search, investigators took Ivinsb primary RMR-1029 storebnot just
a sample of the stuff, all of it. They skimmed a small amount into a vial,
labeled it with an identification number, and sent it to Pat Worsham down the
hall for analysis.

Later, on the patio of the old officersb club, Heine, Ivins, and the other
scientists in the bacteriology division had a couple of beers and tried to
laugh the whole thing off. Each one would joke about how the other was really
the anthrax mailer.


The letter mailed to Tom Brokaw at NBC News.  Getty Images Over the following
days, Worsham grew fresh cultures from the seized RMR-1029 spores. Unlike the
cultures from the sample that Ivins had supplied back in April 2002, this
time the same bullbs-eye-shaped morphs she had seen in the Leahy, Daschle,
and Post letters appeared. So did the smallish, irregular colonies. Like the
Battelle sample, it seemed to be the same stuff as the killer anthrax.

Investigators were talking to Ivinsb coworkers and digging into his archived
emails. They learned that women made him feel awkward. He also didnbt like
talking on the telephone. More seriously, Ivins worried in an email about
having bparanoid personality disorderb and feared that he might be
schizophrenic, as well. The idea that a man handling some of the worldbs most
lethal pathogens might be mentally ill was not a comforting thought. (It
isnbt clear whether Ivins actually suffered from these ailments. His medical
records are still sealed.)

On March 31, 2005bafter more than two dozen interviewsbinvestigators decided
to challenge Ivins more forcefully. They asked why he hadnbt submitted all
his anthrax samples to the repository; he had no good explanation. They asked
him about the bcowboyb spore hunts in his office; he said he was just being
careful. They told him they were copying his hard drive. He was concernedbhe
asked what the FBI does when they find something like child pornography on a
computer.

Investigators also pressed Ivins about his personal life. He responded that
he had stopped taking antidepressants. bHe internalizes his negative emotions
and, as a result, suffers from ulcers and irritable bowel syndrome,b an FBI
interview summary noted. bWhen asked whether his psychological condition had
ever caused him to do anything which surprised him b& Ivins offered that he
does not bact outb and has never hit his wife.b

It was a devastating conversation for Ivins. Even the investigators were
worried about the impact. The next day, an agent asked Ivins if hebd make it
through the weekend. bIbm not going to jump off a bridge or anything like
that,b Ivins said. But he was going to start taking his meds again. A few
days later, he hired a lawyer.

Meanwhile, Fraser-Liggettbs team had genetically sequenced the four telltale
mutants that grew out of the killer anthrax. They were all 99.99 percent
identical. But that tiny fraction of differencebless than a thousand base
pairsbwas enough to give each mutant a unique genomic signature. If a batch
of anthrax tested positive for these four morphs, it meant that it was
provably identical to the attack anthrax. Before, researchers had relied on
Worshambs discerning but still subjective eye to tell them which strains
looked similar to the morphs in the killer batches. Now they had the kind of
hard genetic data they could take to a judge and jury.

Along with other labs, Fraser-Liggettbs team quickly developed streamlined
tests to detect each one of the mutants. They then began the labor-intensive
process of running these tests on every anthrax isolate the FBI had collected
from labs around the world. In a warehouse-style lab, 75 researchers worked
split shifts six days a week, testing and retesting the samples. None of them
had any idea what the results meant; all the samples were coded, and all the
groups were compartmentalized from one another. They toiled without any sense
of progress.

Even as they plunged ahead with the new genetic tests, they continued to
search for other ways to identify the source of the killer anthrax.
Fraser-Liggettbs team and others spent more than a year trying to track a
contaminant found in two of the letters. The search did not yield any useful
forensic information. Repeated attempts to reverse-engineer the powdered
attack spores flopped. So did efforts to use trace amounts of tin and silica
found in the attack powder to discern where the spores were made.

Eventually, FBI director Robert Mueller seemed to lose patience with the
whole thing. President Bush frequently teased him about the case during his
intelligence briefings. bBob, howbs that anthrax investigation coming along?b
Bush would ask sarcastically. Mueller had no good answers. In 2006, the agent
in charge of the case was replaced.

Edward Montooth, a two-decade veteran of counterterrorism and homicide cases,
was brought in to run the investigation. He cultivated a relaxed, slow-going
demeanor among outsiders, coming across as a sort of clean-shaven midwestern
Columbo. An agent named Vince Lisi was chosen to be Montoothbs deputy.
Blue-eyed with a ruddy quarterbackbs face, Lisi had no tolerance for shifty
suspects or slow-moving investigations. Dellafera, the postal inspector,
remained on the leadership team, thanks in part to his ability to grasp the
details of everything from anthrax fermentation to forensic psychology to
envelope fibers.

And boy, were there details. Anthrax repository submissions. Genomic
fingerprinting. Thousands of interview transcripts. In the unitbs office,
sandwiched between a chain hotel and the freeway in suburban Tysons Corner,
Virginia, there were rooms filled with filesbas many as 400,000 documents. It
seemed impossible to get through them all. bI came home with a headache every
day for a year,b Montooth says.

On the other hand, the list of potential suspects was now short enough that
it could fit on a single page, thanks to the emerging DNA results. The search
for the anthrax mutants wasnbt finished, but once decoded, the results
pointed directly at RMR-1029 and its subsamples. This, in turn, meant that
scientists with access to that material were the people worth considering.
Guys like Bruce Ivins, in other words. bThe science absolutely drove this,b
one former senior FBI official says. bIt changed the focus completely.b

The investigating unit gathered in its makeshift meeting areaba training room
with acoustics so bad agents had to shout to be heard. Lisi, Dellafera, and
Montooth barked out the new plan: Start with RMR-1029 and its subsamples.
Figure out who had access. Cross as many names off the list as you can. The
person left at the end is the killer. bDonbt assume anything. Either prove to
us theybre guiltybor prove to us theybre not,b Lisi said. bNo more happy
talk. Even for the people who helped us.b

Investigators scoured the phone records, email accounts, and laboratory
access-card records of US anthrax scientists with possible access to RMR-1029
in an effort to determine their whereabouts on the days in the fall of 2001
when the letters were mailed. They reviewed anthrax transfer records, lab
notebooks, and scientific publications to get a sense of how the scientists
used their anthracis. bIf you knew how to grow up large quantities of bugs,
the FBI was on you all the time,b Heine says. Pat Worshambwho found the
telltale mutantsbwas among those aggressively interrogated.

Worsham was able to exonerate herself, but other names were harder to cross
off. Heine had a bunch of RMR-1029 subsamples. John Ezzell, head of
USAMRIIDbs Special Pathogens Sample Testing Lab, was perhaps the only
scientist at USAMRIID who worked with dry sporesbalbeit dead ones. But it
wouldbve been highly unusual for Ezzell or Heine to keep a store of bugs in
their freezers. One of the many lab technicians or scientists who shared
cold-storage space with them would have noticed a stockpile, the agents
theorized. Ivins, as a designated spore-grower, had plenty of reason to keep
large quantities of anthrax. On top of all the other red flags, Ivins had
little to no verifiable alibi for the critical days in question.

Ivins, Heine, and their coworkers would still blow off steam at the old
officersb club. Sometimes Ivinsb buddies would needle him about being the new
leading suspect. bBruce, what have you done now?b theybd ask sarcastically.
Sometimes they shouted curses at the FBI agents they were sure must be
listening in. They kept going to their volleyball games. Ivins kept emailing
around corny jokes.

Interest in Ivins continued to grow. Agents began looking for old evidence
that might point to him. That led them, finally, back to Nancy Haigwood,
Ivinsb grad school colleague who had called as early as 2002 to say that she
thought Ivins was the anthrax mailer. They had all but ignored her then. Now
they encouraged her to email Ivins while they monitored the thread. It wasnbt
hard to get a reply: Ivins always wanted to know more about Kappa Kappa
Gamma, her old sorority. Maybe theybd even talk about how odd it was that the
one anthrax-laden mailbox in New Jersey was right next to Princetonbs KKG
offices. (Neither Haigwood nor the FBI will discuss the details of these
exchanges.)

Around the same time, the US attorneybs office asked Ivins to testify before
a grand jury about the scientific aspects of the anthrax case. In a strictly
legal sense, he wasnbt an investigative target, they explained. Ivins agreed
and, starting on May 7, 2007, testified for two days without a lawyer. The
questions about his handling of anthracisband about his personal lifebsent
him into a tailspin.

bThey accused me of diluting, altering or adulterating an important
preparation of anthrax material,b he emailed a friend, almost certainly
referring to his flawed RMR-1029 submission in April 2002. bDo you realize
that if anybody gets indicted for even the most remote reason with respect to
the anthrax letters b& they face the death penalty?b

He also began talking about leaving USAMRIID. And maybe more. bIbve been
inside, cooped up for almost all of my life, I want to spend eternity
outside,b he wrote in another email. bI look like Ibm 90 years old. I feel
older than that b& I guess I should have started on [the antidepressant drug]
Celexa years earlier. Also caffeine and alcohol.b The former lightweight had
become a heavy drinker.

In August, Fraser-Liggettbs team presented its final DNA fingerprinting
report to the bureau. The results were somewhat conflicting. Some samples
initially tested positive for the telltale morphs, then negative in a second
exam. But of the 1,059 viable samples in the FBIbs Ames anthrax repository,
eight regularly produced all of the mutants. One of those eight was Ivinsb
RMR-1029 flask. The other seven were its subsamples. This ruled out Hatfill,
who did not have access to RMR-1029 during his time at USAMRIID. (Later, the
Justice Department agreed to pay Hatfill a $5.8 million settlement and issued
an official letter exonerating him. CondC(9 Nast also agreed to an undisclosed
settlement. The New York Times case was dismissed.) And while dozens of other
scientists did have access to the RMR-1029 subsamples, they were being slowly
crossed off the list. As each alibi and exculpatory story checked out, the
investigators gravitated closer to Ivins.

The FBI wasnbt ready to make a move, though. It had the genetics, but the DNA
fingerprints went only so far. There were still those seven subsamples and
those occasionally inconsistent results. If this were a more traditional
murder case, investigators would now know which store sold the gun and to
whom it was registeredbbut not who fired it. Montooth, for one, was worried.
He once lost a murder case because the jury didnbt buy the DNA evidence.
bJust like a gun in possession doesnbt mean homicide,b Montooth says,
bscience alone isnbt going to convict him.b

Nor would a jury convict Ivins because of his unexplained hours in the lab,
his suspicious office swabbing, or his botched April 2002 submission of
RMR-1029 to the FBI Ames repository. The agents still had no witnesses to the
mailings, no confession, no obvious motivebjust complicated science, some
hard-to-explain coincidences, and odd behavior.

In a series of meetings, the agents debated whether it was finally time to
search Ivinsb home. They decided to consult with outside forensic
psychiatrists. The doctors said Ivins was probably the type to keep a
souvenir of the crime, but they warned that he was a fragile man who had
already been pushed very, very hard. On September 25, Ivins showed up at work
with a black eye. He joked that he ran into somethingbhis wifebs fist. A
month later, he talked about his new favorite cocktail, tequila and Ambien.

bThe psychiatrists told us: When you go overt, you will have cut him from all
his life rings. Things could get tough,b Montooth says. bSo, were we
concerned? Hell yeah.b But they didnbt have much choice. The search was
scheduled for November 1, 2007.

When the day came, two FBI agents caught up to Ivins at the entrance to
USAMRIID. Ivins asked if he needed his lawyer. Nope, they answered. Just come
into an office and listen to what webve got to say.

Youbve been trying to fool us for a long time, they said. You knew as far
back as early 2002 that your RMR-1029 anthrax was close, really close, to the
killer spores: same strain, similar kinds of mutants. You knew back then that
we were looking at those mutants, seeing if theybd lead back to the attack
anthrax. You knew that we wanted to do DNA fingerprinting; heck, you even
suggested we do that. And now we know that you screwed up your RMR-1029
submission.

The genetic analysis of the mutants came in, Bruce. RMR-1029 and the attack
spores: They match. Perfectly. That stuff you gave us in April of b02? That
stuff you called RMR-1029? It doesnbt match. And we know why. You were
supposed to give us a full sample back then, mutants and all. Instead, you
picked off single colonies so those morphs would never show.

Ivins offered all kinds of excuses. He hadnbt understood the submission
guidelines. He hadnbt gotten how important RMR-1029 was. He even absurdly
claimed he wasnbt really an expert on anthrax. Each lame explanation only
frustrated the agents further. To rattle Ivins, the agents asked what they
call a change-up questionba deliberately provocative non sequitur. Tell us
about Nancy Haigwoodbs husband, an agent said.

Ivinsb house was a microcosm of the case: lots of suspicious, freaky material
but no evidence of a crime.  Ivins pushed away from the conference room
table, crossed his arms and legs, and told the investigators he was taking
the Fifth. He refused to respond to any further statements.

At about 8 pm, Ivins said he was leaving. Actually, you probably shouldnbt,
the agents answered. Webve got people talking to your wife and kids. Webve
got agents searching your office, your cars, and your house. Itbs going to
take a while. Webve booked hotel rooms for you and your family. Want a ride?

Dellafera, the postal inspector, was waiting at the institutebs entrance. He
and Ivins had known each other since the beginning of the case. They drove
over to the Hilton Garden Inn, with Ivins in the passenger seat. Dellafera
asked Ivins: Are you worried about the searches? Yes, Ivins answered. I do
things a bmiddle-aged man should not do.b Then Ivins told him about a bag in
his house filled with the womenbs clothes he liked to wear.

The two passed a few minutes in awkward silence. Ivins still looked nervous,
uncomfortable. He said he didnbt want to be labeled a mass killer. Ibm not a
terrorist, he said. I canbt believe you think Ibm the anthrax mailer.

Dellafera said he empathized. He told Ivins that the mailer never meant to
hurt anyone: The envelopes were taped shut, and the letters told people to
take penicillin. Ivins didnbt answer. He just rocked in his seat, staring at
the floor.

While Dellafera and Ivins checked Ivins into the hotel, teams of FBI agents
drove slowly, one car at a time, up to Ivinsb home. They went in quietly;
Montooth didnbt want another media circus. Inside, there were piles of
clutter everywhere: bank deposit slips, VHS tapes of The Mary Tyler Moore
Show, classical music cassettes. The agents collected it all, hoping for
somethingbanythingbthat would tie Ivins to the mailings. Finally, around 5
am, Montooth called it off. bWe got dust bunnies,b he says.

Montooth, Dellafera, and Lisi tried to stay upbeat. The agents did find some
odd items: fake hairpieces, letters to politicians and the press (the same
sorts of people who got the anthrax mailings), three pistols, a handwritten
note about drawing bfirst blood b& and last.b Ivins even wrote a song
celebrating deceased space shuttle astronaut Christa McAuliffebone of his
many fixations. In some ways, the discoveries were a microcosm of the whole
case: lots of suspicious, freaky material but no direct evidence of a crime.

On the day of the search, Ivinsb security clearance was revoked, severely
limiting the kind of work he could do at USAMRIID. He became convinced that
Heine, his close friend and colleague, had fingered him as the anthrax
mailer. Heine hadnbt, but the suspicion drove a wedge between Ivins and his
drinking buddy.

The two scientists stopped talking to each other, even though they were still
working together on projects. When they needed to communicate, they did so
through their boss. Bowling nights and beers at the old officersb club were
out, of course. Heine even skipped the division Christmas party so Ivins
could go, but Ivins didnbt show up either.

About a week after the search, a surveillance agent spotted Ivins at around 1
am in his long underwear, throwing out a copy of Douglas Hofstadterbs classic
book GC6del, Escher, Bach. Ivins had lots of books in his house. Why throw out
that one?

The book contains a lengthy section dealing with codesbspecifically,
nucleotide bases that make up DNA, represented by the letters A, T, C, and G.
Investigators had long believed that at least two of the anthrax mailings
contained a code, too. Some of the As and Ts were bolded, and the misspelled
word penacilin had a bold A in the middle. Maybe Hofstadterbs coding was the
answer.

The agents knew that groups of three basesbcalled codonsbwill reliably form
certain amino acids, which can also be represented by letters. When they
lifted out just the bolded letters, investigators got TTT AAT TAT. The amino
acids that form from these codons start with the letters P, A, and T. It was
the first name of a female colleague that Ivins seemed to have a peculiar
interest in: Pat Fellows. The letters that officially represent these amino
acids are F, N, and Y. Maybe that meant bFuck New Yorkb; investigators knew
Ivins hated New York City. In the end, though, they could only guess at the
meaning.

As the case focused on him, Ivins was coming unmoored. His colleagues would
find him slumped in his chair, staring into space. Every once in a while,
hebd blurt out unprompted: bI could never intentionally kill or hurt
someone.b For their next interview with Ivins, investigators decided not to
ask about the mailings and instead planned to focus on the weird stuff, the
personal stuff.

On January 16, 2008, they all met at the US attorneybs office in downtown
Washington. Lisi opened up with a question about Kappa Kappa Gamma, the
sorority Ivins had been so interested in for so long. bOh, itbs not an
interest,b Ivins answered. bItbs an obsession.b Ivins detailed that obsession
as the investigators tried to hide their shock. Ivins talked about how he
broke into KKG houses in the 1970s and stole their coded ritual book and
cipher. How he sold copies of that book through the mail. How he set out to
learn everything about Nancy Haigwood once he found out she was a Kappa. How
he vandalized the property where she lived.

Lisi asked Ivins if his wife, Diane, knew about any of this. Not a clue,
Ivins answered. She didnbt track his comings and goings.

It was the kind of answer Lisi, Montooth, Dellafera, and US attorney Rachel
Lieber wanted to hear. If the case went to trial, Ivins couldnbt use his wife
as an alibi for the mailings. Not after hebd bragged about disappearing all
the time without her knowledge. They liked Ivinsb other answers, too. He
admitted to multiple felonies with the vandalismbones he traveled far to
commit. And he repeatedly used the word obsession, which might spook a jury
inclined to sympathize with Bruce Ivins, Red Cross volunteer.

A second interview, on February 13, started along similar lines. Ivins
described his fascination with bondage, beginning with blindfolding his teddy
bears at age 5 or 6. Then Lisi asked about a more sensitive subject. He took
out the diagram Ivins had drawn in January 2002 explaining the differences in
anthrax strains. In it, Ivins showed how his purer-than-pure anthrax samples
were so completely different from the attack spores. Ivins said he didnbt
remember drawing it. Lisi assured Ivins he had and asked him to reinterpret
it on the spot. Ivins just read the names and places on the diagram out loud.

A little over a month later, on March 19, at 2:09 pm, Diane Ivins called 911.
Her husband was unconscious after too many pills and too much liquor.

Ivins spent a few days at a local hospital. When he returned to work, he
struggled. Then, in May, Ivins checked into a mental health facility in
Cumberland, Maryland, and spent four weeks there. But the drinking and
pill-taking continued.

On June 5, Ivins spoke on the phone with a friend. Ivins complained of waking
up in his clothes, looking at the keys beside the bed, and thinking: bOh
shit, did I drive somewhere last night?b The friend asked Ivins if he might
have committed some terrible act. Ivins didnbt say no.

Instead, Ivins told his friend, bI do not have any recollection of ever
having done anything like [the anthrax attacks]. As a matter of fact, I have
no clue how to, how to make a bioweapon, and I donbt want to know.b The
friend suggested hypnosis as a recall tool. Ivins said, bWhat happens if I
find something that, that is, like, buried deep, deep, deep b& if I found out
I was involved in some way?b

A few days later, agents had their final interview with Ivins. The intention
was to nail him on key aspects of the case, bwhat we needed for prosecution,b
Dellafera explains.

Yet the talk turned strange and personal once again. Lisi took out GC6del,
Escher, Bach. Ivins said he had a dog-eared copy of the book at home. Lisi
asked about DNA and codes. Ivins answered that he wasnbt much of a bgene
jockey.b (This from a guy who once emailed around a joke that went:
bBiopersonals b& Lonely ATGCATG would like to pair up with congenial
TACGTAG.b) Lisi wondered why Ivins hated New York so much. Ivins said it
dated back to a microbiology conference there in the late b60s. A waitress
was rude and threw his lunch on the table. Lisi asked Ivins whether he
remembered what hebd had. Sure, Ivins answered: spinach salad.

Yet Ivinsb memory quickly failed him when it came to details about the
anthrax investigation. Ivins said he couldnbt remember why the FBI wanted to
build an anthracis repository, or what they wanted to do with the samples, or
whether he prepared his submissions himself.

Ivins was attending therapy sessions, but they didnbt seem to be helping his
mental state. In July 2008, he posted a comment on a YouTube clip of a
reality TV show, The Mole, suggesting that one of the female contestants
ought to have a hatchet brought bdown hard and sharply across her neck,
severing her carotid artery and jugular vein.b On July 9, Ivins went to a
group therapy session and said he had a plan to end all this government
harassment. He said he had access to a .22-caliber rifle, a Glock handgun,
and body armor. He would kill all his coworkers and everybody else who had
wronged him.

The next day, the addiction counselor, Jean Duley, called the police and told
them of the disturbing outburst. Officers were dispatched to USAMRIID, where
Ivins was attending a briefing on a next-generation plague vaccine. He seemed
stressed, his colleagues said, but not crazy. The cops arrived and took Ivins
away. He left the institute quietly and never went back.

The police didnbt arrest Ivins but rather took him to Frederick Memorial
Hospital for evaluation. Two weeks later, on July 24, 2008, Ivins returned
home. That afternoon, he made two separate trips to a local store, picking up
a bottle of Tylenol PM each time. That night, his wife left a note on his
bedside. bIbm hurt, concerned, confused, and angry about your actions the
last few weeks,b she wrote. bYou tell me you love me but you have been rude
and sarcastic and nasty many times when you talk to me. You tell me you
arenbt going to get any more guns then you fill out an online application for
a gun license.b

At approximately 1 am on July 27, 2008, Diane Ivins woke up to check on her
husband. He wasnbt in bed. She went to the bathroom and found him on the
floor, in his undershirt, lying in a pool of what looked like his own urine.
bHebs unconscious. Hebs breathing rapidly. Hebs clammy,b she told the 911
operator. The local police and the emergency medical technicians arrived at
the same time. The police and Diane counted the pills left in the medicine
cabinet while the EMTs brought Ivins out on a stretcher.

They rushed him to the hospital. The doctors thought he might have overdosed
and promptly ordered that a breathing tube be put down his throat.

Blood tests showed the acetaminophen level in Ivinsb blood was 10 times
higher than what is considered safe. A massive dose of Tylenol was
overloading his liver. Few ways to die are more agonizingbthe abdominal pain
alone is excruciating. Doctors administered an antidote, N-acetyl cysteine.
But it was too late.

At 8 obclock, a nurse woke Ivins up and asked, bDid you intentionally try to
commit suicide?b Ivins nodded and attempted to remove his tubes. The nurse
had him restrained. Doctors tried to talk Diane into transferring him to
another facility for a potential liver transplant. He doesnbt want to be
saved, she said.

On July 29, 2008, at 10:47 am, Ivins died. There was no suicide note.
Instead, he had merely taken his wifebs letter and scrawled a response on the
other side. He then scratched out a few of the words and left it on the
nightstand. bI have a terrible headache. Ibm going to take some Tylenol and
sleep in tomorrow,b he wrote. bPlease let me sleep. Please.b

Over the next year and a half, the Justice Department produced a 92-page
binvestigative summaryb of its case against Bruce Ivins, released thousands
of pages of documents, agreed to independent reviews by both the National
Research Council and the Government Accountability Office, and officially
declared the case closed. It was an extraordinary and unprecedented effort to
prove the guilt of a man who was never arrested for any crime. But the
document dump is in many ways unsatisfying, since it covers a single
suspectbIvinnd labs in a recently opened West Baltimore biotechnology
complex. Hanging on her office wall, near pictures of her prized poodles, are
framed copies of her articles in the prestigious science journal Nature.
Shebs proud of her colleagues, like Jacques Ravel and David Rasko, who, while
working on her team, helped pioneer this new science of microbial forensics
during the anthrax investigation. As she talks about the case, however,
Fraser-Liggett becomes uneasy. bItbs as if somehow webmeaning those of us
here who were involved in doing all this work on genetic mutantsbwere somehow
fully supportive of the FBIbs conclusions,b she says.

The cross-dressing, the sorority obsession, the bondagebbit would be very
easy to get sucked into all of this because, you know, it makes for a great
tabloid-type story,b she says. bIvins was a bit peculiar. But one of our
civil liberties is to be peculiar.b

There are still unresolved scientific issues surrounding the case, she points
out. RMR-1029 was a witchesb brew of 35 different production runs. Maybe the
mutants came from one or more of those original component batches and
therefore showed up only on particular tests.

Nobody knows for sure, because RMR-1029 was never reverse- engineered. bIt
raises some very important questions,b Fraser-Liggett says. bLetbs repeat
this experiment. Letbs go back and see if we can re-create what was in
RMR-1029.b

The National Research Council report also casts doubt on whether the killer
spores really were descendants of Ivinsb RMR-1029 flask. The FBI resampled
RMR-1029 a total of 30 different times, the report found. They could get all
four telltale morphs on only 16 occasions.

Further, the FBI says that only eight samples in its Ames repository were
genetic matches to all four morphs of the killer sporesband that the
scientists with access to those isolates were thoroughly scrutinized. But the
National Research Council found that the FBIbs collection canbt be fully
trusted: Too many of the samples were intermingled or descended from other
labsb anthracis to provide a truly representative cross-section of Ames
anthrax. This may also be a reason why nearly one in 10 samples in the
repository tested positive for at least one mutant.

Paul Keim, who helped assemble the FBIbs Ames collection, still wonders how
much to trust an anthrax repository that relied on scientists (and potential
murder suspects) submitting their own samples. bWe donbt know if people did
it correctly, and therebs no real way to control for that,b Keim says.

Even if everyone was aboveboard, itbs unclear whether the FBI accounted for
every last anthrax sample. Each time Ivins gave his colleague Hank Heine a
batch of spores for an experiment, for example, Heine would save a milliliter
or two, in case the experiment went wrong. bItbs just good scientific
practice,b Heine says. bI had numerous samples of RMR-1029.b Itbs hard to
imagine he was the only scientist with such a collection. Because the
subsamples were so small and largely undocumented, it took the FBI nearly
three years to stock its repositorybplenty of time for a researcher to
dispose of an incriminating batch.

Then therebs the problem of figuring out when Ivins could have grown the
spores. In an email to colleagues on April 23, 2004bunrelated to the
investigation and long before he became its prime suspectbIvins estimated
that it would take 60 hours to brew up 500 billion spores. Each anthrax
letter contained up to four times that amount. This means that making enough
spores for the mailings would have required between five and six months. It
would have been nearly impossible for Ivins to do that much work without
others noticing. It may be odd to rely on Ivins himself for these numbers,
but his colleagues do not dispute his estimate. The National Research Council
report does theorize that it could have been done more quickly, but its
findings were inconclusive. bThe time might vary from as little as two to
three days to as much as several,b the report reads. bGiven uncertainty about
the methods used for preparation of the spore material, the committee could
reach no significant conclusions regarding the skill set of the perpetrator.b

This raises another significant problem with the case. USAMRIID veterans
debate whether Ivins had access to the kind of gear required to dry and mill
the spores. Even if he did, some argue, he wouldnbt have known how to use it.
Ivinsb wet-spore experience didnbt translate to dry stuff, Heine and others
say.

Montooth acknowledges that he isnbt sure how Ivins would have done all that
growing and drying. bBut it almost doesnbt matter,b he says. Investigators
know which days in September and October the envelopes were mailed. That was
the actual murderous act. The anthrax could have been slowly assembled and
processed for months or years before that. Ivinsb alibis for those autumn
days are virtually nonexistent.

There are still other problems with the case against Ivins. The killing
spores were so volatile that they cross-contaminated piles and piles of mail.
Yet spores were never found in Ivinsb house or his car, and only a handful
were discovered in his lab. Therebs no evidence of any trip to Princeton to
mail the letters. And just because the killer spores were descendants of a
USAMRIID flask, therebs no guarantee a USAMRIID scientist was actually the
mailer. In fact, the FBI was never able to prove where the attack anthrax was
cultured. bIt wouldbve been very easy to take the anthrax out, to steal
some,b a former USAMRIID officer says. bAnybody could do that.b

Finally, therebs the matter of motive. The Justice Department asserts in its
investigative summary that Ivins mailed the letters to gin up support for an
anthrax vaccine, offering a few ambiguous emails and comments to friends and
investigators as proof. If therebs any further, credible evidence to support
this notion, Wired couldnbt find it in the thousands of pages of case
documents released by the government or in the hours of interviews conducted
with the investigators. Montooth concedes itbs a placeholder rationale at
best. For someone as deeply disturbed as Ivins, he argues, simple rules of
cause and effect donbt apply, especially not in matters as grave as murder.
bYou cannot think of this in one dimension or layer. Itbs not that simple,b
Montooth says. bYoubre never gonna know a single cause or motive for why it
was done.b

But despite all these flaws, the circumstantial evidence remains compelling.
It could just be a coincidence that the killer spores were ultimately traced
back to a single parent flask and that this flask just happened to be
overseen by a depressed scientist with occasional violent fantasies. It could
just be a coincidence that this same scientist screwed up his anthrax
submission to the FBIbeven though he helped develop the submission protocols.
It could just be a coincidence that his after-hours work spiked right before
the mailings. But put all of those coincidences together and something
stronger than happenstance emerges. For the Justice Department, itbs enough
to prove Ivins was the anthrax mailer.

Therebs an irony in the fact that the culprit was likely a top government
anthrax expert: Since 2001, the US has built dozens of labs, spent just under
$62 billion, and hired an army of researchers to prevent a second bioterror
attack. In effect, Washington has devoted the past decade to training and
equipping hundreds of people like Ivins.

Itbs an unnerving scenario. But therebs something much scarier to
contemplate. Therebs still the possibility that the government was as wrong
about Ivins as it was about Hatfill. If thatbs the case, the anthrax mailer
is still at large. And that means someone launched the deadliest biological
attack in the history of the United Statesband got away with it.

Contributing editor Noah Shachtman (noah.shachtman at gmail.com) wrote about the
Afghan air war in issue 18.01. Additional reporting by Adam Rawnsley.