To break up the oil that leaked into the Gulf of Mexico, BP had applied by mid-July nearly two million gallons of dispersants, both at the sea’s surface and below. Environmentalists worry that the chemicals could be as damaging as the oil. To address such concerns, the U.S. Environmental Protection Agency released early this summer preliminary data from its tests, but instead of quelling fears, the data have stirred up more questions.

In a bid to corroborate potentially suspect results that were provided by the dispersants industry, the EPA tested eight dispersants, including COREXIT 9500, the one most widely used by BP. The agency’s results showed broad similarities with industry’s analyses—some effects on silverside fish and mysid shrimp, but no significant disruption of hormonal systems of animals, at least at the cellular level. “All the dispersants are roughly equal in toxicity and generally less toxic than oil,” said EPA assistant administrator and chemist Paul Anastas in a press briefing on June 30. “The dispersant constituents are expected to biodegrade in weeks to months, rather than remaining in the ecosystem for years as oil might.”

But at least one outside toxicologist has found reason to criticize the EPA and the methods it has used. “There is not any information on what is the environmentally relevant level of dispersants,” says toxicologist Carys L. Mitchelmore of the University of Maryland, who helped to write a 2005 National Research Council report on dispersants. Nor is there any evidence that the agency had any requirements for defining acceptable toxicity levels in the industry-provided data. From that information alone, “I could not compare and contrast which one was more toxic than the other,” Mitchelmore recounts.

In fact, it remains unclear whether anyone at the EPA ever checked the industry-provided numbers as required by law. When Scientific American asked Anastas about that, he did not directly answer the question, and the EPA did not respond to follow-up questions. Such clarification would be useful because the industry data appear to be full of potential faults, including, in the analysis of one dispersant, the use of the wrong reference toxicant. Nor did the EPA show the best understanding of toxicology in urging BP in a directive to use dispersants with a “toxicity value less than” a certain cutoff: in toxicology, a chemical that produces harm at low concentrations, say, five parts per million, is more deadly than those that kill at 10 parts per million.

The problems are not entirely the fault of the EPA; policies for safety testing under current chemical regulations are flawed [see “Chemical Controls”; Perspectives, Scientific American, April]. “The magnitude of this event has raised important questions about how these previous, existing regulations [for dispersants] may need to be reexamined and revisited in ways that ask different questions and even better prepare us in the future,” Anastas admitted.

Although Congress has suggested reforms, it is uncertain if the EPA will address these methodology issues as it explores the contamination in the Gulf of Mexico. In July the EPA began conducting toxicity tests for the specific light sweet crude from the Gulf, both alone and in conjunction with the various dispersants.

“Once it’s mixed with oil, that’s where you get the most impact, that’s where you see most of the toxicity,” says toxicologist Sergio Alex Villalobos of Nalco, the maker of COREXIT 9500. Anastas suggested that testing was expected to be completed before the end of August.