SUPER SNIFFERS

"A pig could do it, but they grow to about 300 pounds. A cat, yes, but try
to keep one disciplined. Besides, society accepts dogs and they socialize
with people."

The number on the back of the trading card is impressive. It's 257. That's
not home runs hit. It's not touchdowns scored.

It's 257 pounds of marijuana found.

The face on the front of the card belongs to a dog named Buckshot. He now
resides in Oklahoma.

Buckshot's trading card is attached to a wall in a laboratory at Auburn
University. Alongside it are the faces of other canine heroes -- Shane,
Corky, Zoom and many others.

Most are either a German Shepherd, Belgian Malinois or Labrador Retriever.
Larry Myers, who owns the cards, says several other breeds are also used by
law enforcement to detect drugs, explosives and signs of arson.

"No bulldogs, though," he laughs. "They can hardly breathe."

You're also not likely, he says with a smile, to see macho DEA agents or
sheriff's deputies marching into a suspected drug den behind a poodle.

And though he says other animals could detect chemicals -- "A pig could do
it but they grow to about 300 pounds; a cat, yes, but try to keep one
disciplined"-- a dog is pretty much the sniffer of choice. "Besides,
society accepts dogs and they socialize with people," he says.

Some breeds, because of their ability to concentrate, are better than others.

New things

Myers is a professor in the Auburn University College of Veterinary
Medicine and one of the country's leading experts on canine detection. A
doctor of veterinary medicine as well as a doctor of physiology, he's
appeared as a witness both for the prosecution and the defense in hundreds
of criminal trials across the country and has been interviewed by numerous
national publications such as "Popular Science" and most recently on
television's "60 Minutes."

In 1989, Myers founded the Auburn Canine Research Institute, originally
known as the Institute for Biological Detection Systems and one of the
nation's leading developers of canine detection technology. The institute
conducts research at Auburn University, and at a center in Anniston, Ala.,
it trains dogs, handlers, trainers and program managers for government
agencies.

No longer associated with that the institute, Myers is researching ways
dogs can be used besides uncovering drugs and explosives.

"We're using dogs to discover mold in houses which would make the building
dangerous for someone to inhabit," he says. "They're very good at finding
termites."

He's now doing research for agencies such as the United States Department
of Agriculture.

Myers is looking for ways to train dogs to detect chemicals in ponds that
give catfish a bad flavor, a problem costing the industry $50 million per year.

Dogs can also help with cattle breeding. Using vaginal secretions from
cows, Myers says, he's working to train dogs to tell farmers when the cow
is ready for mating -- even quicker than a bull can. Once again, the
research could mean millions to farmers.

Nose know-how

Myers is working to discover how dogs do what they do, and how chemicals in
a drug or hazardous material stand out to a dog. If Myers and other
researchers could discover how animals identify odors then that knowledge
could be used to produce artificial devices that do the same thing. For
example, perhaps such a device could detect e. coli viruses in food.

Each dog Myers works with goes through 500-1,000 different repetitions
before it is trained.

Samples such as different tubes of pond water are placed in boxes, which
are placed in holes in a wooden board. The samples are rotated for each
repetition, and dogs are not allowed to see them being placed.

"When the dog finds the right sample," explains Auburn veterinary student
Rebecca Robinson, "they are rewarded with something good to eat. They love
snacks."

Myers said many of the dogs come from the pound.

"When selecting a dog, it's important to have one that won't be distracted
by food in the room," Robinson says. "Some can't make the cut."

They can lie

Detection dogs aren't always on top of their game.

"Dental tartar can have an effect," Myers says. "Really. Clean the teeth
and you get an almost immediate recovery of smell. Dogs get allergies. They
get colds."

"A dog's sense of smell is not forever," he says. "A variety of diseases
can destroy the sense of smell."

That's why he says that while he'd "bet my life" on certain dogs, he's not
sure he'd want to bet someone else's.

"We really don't know what a dog is picking up on when it alerts," Myers
says. "In many ways we're still dealing with a black box. It could be
picking up on a lot of things rather than just the scent of a person. It'd
be hard to convict someone on that kind of evidence."

Also, he says, not all dogs are trained well. "Poor handlers can cause a
dog's accuracy rate of 85-95 percent to drop to about 60 percent. Dogs want
rewards and so they will give false alerts to get them. Dogs lie. Programs
are supposed to train dogs so that doesn't happen. Not all do."

The standard measure of a dog's accuracy, Myers says, is what it finds.
"The best programs subtract from that score the number of false alerts, but
many do not. They have no accurate measure of their dog's reliability."

He says some programs are just in it for the money, but there are some good
programs as well. "Some police departments train their own and do a very
good job," he says.

And every time there's a terrorist attack or the threat of one, the public
gets interested in dogs and their sniffing abilities.

"Sometimes working with the dogs can be frustrating," Myers says, "but most
of the time it's a hoot."

At Auburn, dogs are trained

to detect bombs, drugs, termites -- not to mention sick catfish