By Jack Greer and Erica Goldman

What causes fish kills? Lack of dissolved oxygen in the water can be one cause. Toxins emitted by certain algal blooms another. But the actual mechanism of how toxins and environmental factors like oxygen levels and water quality combine to create trouble can be complicated — and surprising.

One surprise came when fish farmer Tony Mazzaccaro added copper sulfate to control a toxic dinoflagellate bloom in his ponds. The copper, which he had used successfully before, killed his fish en masse. Why? Maryland Sea Grant aquaculture specialist Dan Terlizzi believes that the copper split open (lysed) Karlodinium cells that were present and released their toxin into the pond. Al Place at the UMBI Center of Marine Biotechnology agrees. They both believe that the combination of copper and a toxic Karlodinium bloom caused the sudden die-off.

But another surprise came when NOAA researcher Peter Moeller found that toxic events associated with Pfiesteria have occurred only in waters rich in metals like copper and high in sulfur. "That has held true so far in all examples I'm aware of to date," he says.

Moeller believes that Pfiesteria removes copper from the environment by binding it in a manner that produces an unstable "free radical" compound that's toxic to many organisms. In Moeller's view, copper sparks Pfiesteria's destructive chemical cascade, another scenario for what might have happened at Mazzaccaro's fish farm.

That copper has shown up at the scene of fish kills would not surprise Ritchie Shoemaker, a family practice doctor in Pocomoke City, Maryland. Since 1998, Shoemaker has argued that copper might hold the key. He noted that copper showed up in the Pocomoke and Neuse rivers, both scenes of fish kills and both agricultural watersheds. In each place farm workers applied copper to tomato plants as a fungicide or to hog and chicken feed to prevent spoilage. Rains likely washed that copper into the rivers. Maryland research¬ ers registered 15 parts per million of copper on the Pocomoke River in 1997, says Shoemaker, an amount considered toxic.

Could the combination of copper and the two dinoflagellates, Pfiesteria and Karlodinium, create a noxious cocktail?

Moeller has considered the possibility that the two dinoflagellates might make each other more lethal when they appear together. He suspects that like Pfiesteria, Karlodinium may be sensitive to copper, making its toxin even more powerful. Place agrees that he'd like to test this possibility.

"To be honest," Place says, "I haven't tested whether the addition of copper to karlotoxin makes it more potent. I need to do that."