Savannah, Georgia: Behind the immaculate grey walls of the Customs and Border Protection’s laboratory in New York stands a cabinet containing three plastic vials filled with a sticky, yellowish substance. Honey, or so an importer has claimed.

The lab’s task: Determine whether the samples are adulterated with sweeteners or syrups, and, if they really are mostly honey, figure out where it originated.

If the honey comes from China, often the case, the entire shipment from which the samples came may be subject to additional taxes.

The chemists here regularly test a wide range of imported goods, but they specialise in analysing agricultural imports.

With remarkable precision, these scientists can tell you where the peanuts in your peanut butter came from and where the mangoes in your jam were grown.

But honey, No. 0409 on the 2015 Harmonized Tariff Schedule, has been a focal point for the lab and the source of a long-running international food scam that has challenged even the existing forensic technology.

Americans consume an average of 635g of honey a year, about three-and-a-half 177ml bottles. Some 70 per cent of it is imported.

In 2001, the Commerce Department enacted a stiff tariff on Chinese honey, nearly tripling the import duty, after American producers complained that Chinese competitors were dumping their products on the market.

Then, honey imports from other countries spiked, including from nations not known for large bee populations.

According to the American Honey Producers Association, Malaysian beekeepers, for example, have the capacity to make about 45,000 pounds of honey annually, but the country has exported as much as 37 million pounds of honey to the US in a year.

As it turned out, Chinese honey was being shipped through ports such as Shanghai, or Busan, South Korea, and slapped with labels from other nations to skirt American duties.

The practice is known as transshipment, or “honey laundering.”

Some of it was not even real honey, but a mix that included corn and rice sweeteners.

In an effort to staunch the flow of illicit honey, chemists at the lab here have tested thousands of samples pulled from barrels and containers at ports across the Southeast. In 2008, the lab demonstrated with about 90 per cent accuracy that honey imported from Thailand, the Philippines and Russia had originated in China.

The evidence helped federal prosecutors build a case against two large American importers who were suspected of buying illegal Chinese honey to avoid more than $180 million (Dh661 million) in duties.

But this kind of detective work is daunting. At the CBP lab, the analytic work takes place inside what’s known as the “country of origin” room. Inside are standing metal shelves filled with bags and plastic totes of imported honey, along with peanuts, shrimp, garlic, mangoes and other foods.

On a recent Tuesday, Robert Redmond and Christopher Kana, two of the lab’s analytic chemists, took a small honey sample and added an acid to digest it. The result looked like muddy water.

In recent years, scientists have demonstrated that subtle chemical variations in many foods, including honey — undetectable to the tongue or the naked eye — can give a strong indication of where it originated. The CBP’s analytic work depends, in part, on these naturally occurring geographic “tracers.”

In late 2012, Redmond travelled to Taiwan and India to collect and test honey. His findings were then added to the database, and now lab chemists can compare honey arriving in the US and said to be from those countries.

But it’s only the latest manoeuvre in a scientific cat-and-mouse game that has stretched on for years.

At first, the detection of transshipped honey relied on a simple test for an unapproved antibiotic, chloramphenicol, discovered in Chinese honey. Carson Watts, former director of the CBP lab in Savannah, said, “Very shortly after word got out that we were using chloramphenicol to identify Chinese honey, they stopped using it.”

Around 2006, unscrupulous importers appeared to be cutting honey with high-fructose rice syrup or disguising cheap, pure honey as an artificial blend. (At the time, the import duty applied to artificial blends that were more than 50 per cent honey by weight.)

The problem? Reliably determining the ratio of rice syrup to honey is nearly impossible.

“An importer could present goods to customs and say, ‘This is 90 per cent rice syrup, 10 per cent honey,’ and Customs really has no way of knowing,” said Michael J Coursey, a lawyer in Washington who has represented American honey producers.

In 2011, the government accused three companies of importing millions of dollars worth of rice fructose blend that in fact was mostly taxable honey. The importers said the product was less than 50 per cent honey.

The scientists at the Savannah lab swung into action, producing evidence that pollen abundance in the blends showed the substance to be mostly honey. But defence lawyers challenged the research on scientific grounds.

“It’s all well and good to say you need to enforce these regulations,” said Dana Krueger, who owns an independent laboratory in Chelmsford, Massachusetts, and testified as a defence witness. “But if there’s no technology, it puts Customs in a difficult position.”

The judge dismissed the case, and the government dropped the charges.

The most sophisticated chemical analysis may have its limits. But for the moment, the food detectives are undeterred.

“If it’s honey from Malaysia, then we’re testing for China,” Redmond said.