n the month before the sniper attacks that
left 10 dead and terrorized the Washington area, John Muhammad
and Lee Malvo are believed to have killed or wounded seven
people in four states.
But because they kept on the move and the police gave only
routine attention to what seemed like run-of-the-mill crimes,
investigators say, Mr. Muhammad and Mr. Malvo were able to
slip through the cracks, gaps and blind spots where
jurisdictions meet and crime-fighting databases end. As a
result, investigators did not even realize that this two-month
shooting rampage was going on until it was half over, when,
the authorities say, the two men began concentrating their
lethal aim on the Washington metropolitan area on Oct. 2.
An examination of the seven shootings that preceded the
sniper attacks in the Washington area highlights the
limitations of law enforcement, particularly when low-profile
crimes do not generate the sort of frenzied, all-points
response that took place in October:
An incomplete computer system run by the federal Bureau of
Alcohol, Tobacco and Firearms made it impossible to connect
bullets fired in three shootings in Maryland and Georgia in
September.
A less-than-exhaustive search of a killing scene in
Alabama failed to turn up the revolver that had been fired in
those three shootings.
A shortage of crime-laboratory technicians caused a 27-day
delay in checking what turned out to be a crucial fingerprint
from the Alabama crime scene.
An incomplete national crime database had no information
about Mr. Muhammad, even though there was a warrant for his
arrest on a February shoplifting charge. So the police
officers who stopped his blue Chevrolet Caprice three times in
the five days before shootings started in the Washington area
had no reason to detain him.
"It was almost as if these guys operated knowing that if
they crossed state and local jurisdictional lines there would
be no way we would put the clues together," one senior law
enforcement official in Maryland said. "We have been praised
for our cooperation, but the sniper investigation highlights
many more missteps by law enforcement than it does brilliant
moves."
On Sept. 5, the police in Clinton, Md., recovered a
.22-caliber bullet after Paul LaRuffa, 55, was shot six times
and robbed of $3,000 as he left his pizzeria. Mr. LaRuffa
later said he had seen a shadow and heard gunshots.
The Prince George's County police sent the bullet to a
county ballistics laboratory, a local law enforcement official
said, and experts there entered its digital image into a
database maintained by the Bureau of Alcohol, Tobacco and
Firearms.
That database, the Integrated Ballistic Identification
System, known as IBIS, allows technicians to find links among
crimes by comparing bullets and casings found at crime scenes
or test-fired from recovered weapons. It has made more than
5,000 matches since its debut in 1993, according to the
bureau.
When a bullet was recovered after Muhammad Rashid was shot
and robbed while closing a liquor store in nearby Brandywine,
Md., on Sept. 15, investigators experts say they used IBIS to
link it to the gun used to shoot Mr. LaRuffa 10 days
earlier.
But the trail was lost there.
Just after midnight on Sept. 21, Million Waldemariam, 41,
was fatally shot outside an Atlanta liquor store as he went to
check on two men who were sitting outside in a dark-colored
car. A bullet was recovered in an autopsy.
In theory, that .22 slug might have been enough to alert
the authorities that there was a gunman on an interstate spree
in a dark sedan.
But the Georgia Bureau of Investigation crime laboratory
does not run .22-caliber weapons and ammunition through IBIS.
John C. Bankhead, a spokesman for the bureau, said examiners
considered the system ineffective with such small-caliber
weapons. Officials at the federal firearms bureau dispute
this.
Regardless, even IBIS would not have matched the Atlanta
bullet to the bullets in Maryland, officials say. While the
system is being expanded, nationwide searches of the database
are still about five months away, said Patti L. Galupo,
director of the unit that oversees IBIS.
For now, those ballistics databases that are running
include information from only a few states at a time: Georgia,
Alabama and Florida, for example, or Mississippi, Louisiana
and Arkansas.
"It's not an acceptable fact of life," said John C.
Killorin, the special agent in charge of the federal firearms
bureau in Atlanta. "The whole point here is to get to a
national system, so then the machine would have identified
Muhammad and Malvo's pattern of activity, not the
investigators. The role of the system is to make cartridges,
bullets and shell casings into informants."
Of course, had any other evidence suggested a link among
the Atlanta and Maryland shootings, Mr. Killorin said, a
special search of the IBIS database could have been
arranged.
"I can do that if I make it happen," Mr. Killorin said.
"What we're trying to build is a system that routinely makes
it happen."
Just 20 hours after Mr. Waldemariam was killed outside the
liquor store in Atlanta, two people were shot at a liquor
store in Montgomery, Ala. The manager, Claudine Parker, 52,
was killed, and a co-worker, Kellie Adams, 24, was wounded as
they closed up about 7:30 p.m. on Sept. 21. Police officers
who arrived moments later said they had seen a man standing
over the women with a handgun, but he got away.
The gun, a North American Arms .22 revolver, was not found
until Oct. 30, a week after Mr. Muhammad and Mr. Malvo were
arrested. A property owner said he had stumbled across it
under a pile of leaves, the police said. Only then did
investigators link the gun, and Mr. Muhammad and Mr. Malvo, to
the earlier shootings in September.
It is not known whether the revolver was fired in the
Montgomery shooting. The sole bullet recovered from that
shooting was at first thought to be a .22. But experts who
analyzed it after Mr. Muhammad's and Mr. Malvo's arrests said
it had come from the same .223-caliber Bushmaster rifle used
in the sniper attacks.
Joe Saloom, a firearms examiner for the Alabama Department
of Forensic Sciences in Montgomery, said the slug was not
entered into IBIS at the time because his laboratory had not
yet installed the system.
The clue that ultimately cracked the sniper case was also
recovered the day of the Montgomery shootings. It was a
fingerprint lifted from a magazine a young man had been seen
carrying near the store. The print was sent to the Alabama
State Police crime laboratory three days later but sat there
for 27 days while the office's four overworked fingerprint
examiners slogged through a backlog of evidence, said Capt.
Hugh McCall, a laboratory spokesman.
When they finally got to it, they ran the print through a
database covering only Alabama and neighboring states
standard operating procedure, said Sgt. Scott Martino of the
Montgomery police, because "most crime happens in your own
backyard." He added, "We had no way of knowing that these guys
were the same ones who were shooting people in
Washington."
Had the fingerprint been run through the Federal Bureau of
Investigation's national database, investigators say, the
police would have matched it to one taken from Mr. Malvo in
December, when immigration officers detained him in
Bellingham, Wash. An address check would have shown that Mr.
Malvo had been living with Mr. Muhammad, officials said. A
further check would have found that Mr. Muhammad had recently
bought the blue 1990 Chevrolet Caprice in which the two
suspects were later arrested after its plates had been
routinely checked by the police no fewer than 11 times.
All this information was put together, investigators say,
only after the suspects themselves called the police in the
Washington area and referred to a shooting in Montgomery.
The Montgomery officer who chased the suspect on foot
described being blocked at one point by a blue sedan, said the
Montgomery police chief, John Wilson. But unaware that two men
in a dark sedan had earlier shot someone outside a store in
Atlanta, the police thought it belonged to a passer-by, Chief
Wilson said.
His department was not the only one to lose sight of the
blue car. In fact, one witness provided its make and model to
Maryland authorities in mid-September.
About 10:30 p.m. on Sept. 14, Rupinder Oberoi, 22 was shot
outside the liquor store where he worked in Silver Spring, Md.
The Montgomery County police did not recover enough ballistics
evidence for a full analysis, but a witness said he had seen a
dark-colored Caprice slowly leaving the parking lot after the
shooting.
After the sniper attacks began on Oct. 2 in the Washington
area, the witness's family again went to the police with his
account of a Caprice. But when reporters asked about a
potential link with the shooting of Mr. Oberoi, members of the
sniper task force played down the possibility. They were
focused on reports of a white van, which proved
misleading.
On Sept. 23, two days after the shootings in Atlanta and
Montgomery, Hong Im Ballenger, 45, was shot and killed outside
her beauty supply store in Baton Rouge, La., Mr. Muhammad's
hometown. A witness heard a shot and saw a man later
identified as Mr. Malvo running with Ms. Ballenger's purse,
the police said. Another witness later described seeing Mr.
Malvo get into an old blue sedan driven by someone else.
The bullet recovered from Ms. Ballenger, however, was not
sent to the Louisiana state crime laboratory for analysis
until after Mr. Muhammad and Mr. Malvo were arrested. It was
then linked to the Bushmaster, leading Mr. Muhammad and Mr.
Malvo to be charged with Ms. Ballenger's killing, as they have
been with the Montgomery shootings.
A Baton Rouge police spokesman, Cpl. Don Kelly, said
detectives there did not routinely send bullets or shell
casings for ballistics tests unless they already had a weapon
to match them to, or another reason to suspect a link to other
crimes, so as not to overwhelm the Louisiana state crime
laboratory's experts.
"People are always surprised when we can't do something,"
Corporal Kelly said, "but we work in the real world with real
budgets."
Real budgets are also at the bottom of why a misdemeanor
shoplifting charge in February against Mr. Muhammad and Mr.
Malvo in Tacoma, Wash., was not entered into the National
Crime Information Center's database, which police officers
across the country can check from their patrol cars. Tacoma
officials said the police typically entered charges only in
felony cases, because extradition could be too expensive a
proposition for a misdemeanor.
After the Baton Rouge shooting, Mr. Muhammad and Mr. Malvo
repeatedly eluded the grasp of the police. On Sept. 28,
officers in Gulfport, Miss., ran a check on the Caprice's
license plate while the car sat empty in a parking lot. The
plates were checked again in Fairfax, Va., on Oct. 1, and in
Silver Spring, Md., the afternoon of Oct. 2, five hours before
the first sniper attack.
The elusiveness of the two men is galling but
understandable, a senior federal investigator in the sniper
case said.
"Shootings in front of liquor stores are all too
commonplace," the investigator said. "So for one to stand out,
something needs to be unique. To connect the dots, first you
need more than one dot to connect. But there really wasn't
anything that stood out about any of those crimes."
2 Are Cleared in Killings
BATON ROUGE, La.,
Nov. 28 (AP) DNA evidence has cleared the two sniper
defendants in the killings of three Baton Rouge women, the
police say. Samples taken from Mr. Muhammad and Mr. Malvo did
not match DNA left in the killings, a police spokeswoman
said.