During wildfire season, the nightly news often shows images of tanker planes dropping orange liquid near the infernos.
That's fire retardant, a substance designed to slow and in some cases halt a blaze.
On average, California uses more retardant than any other state, but some forest service employees argue the substance doesn't work when it matters most.
Up close retardant looks like carrot juice and feels like slime.
It's totally safe for people to touch, says Kevin Reed. He works for CalFire at Hemet Ryan Air Attack Base and is in charge of preparing the retardant.
He says it's orange so firefighters can see it from the air when it's dumped.
Retardant is mostly ammonium phosphate, a substance often used as fertilizer, says George Matousek with the company Phos-Chek, the only supplier of retardant in the U.S.
Phosphate does the magic
When ammonium phosphate-covered wood feels the heat of an oncoming flame a reaction occurs, Matousek explains.
The phosphate converts the woody material into an almost pure form of carbon. Think of diamond or graphite. Pure carbon does not burn.
When this reaction happens on a tree, he says, "it would be black on the outside, but alive on the inside."
Worth the risk?
The ammonium in retardant is poisonous to fish, says Andy Stahl with the watchdog group Forest Service Employees for Environmental Ethics.
"Dump a load of retardant in a creek, you can kill the fish for miles downstream," he said.
That happened in 2009 when retardant was accidentally dropped in a sensitive area of Santa Barbara County, despite a rule saying firefighters can't drop within 300 feet of a waterway.
Dozens of endangered steelhead trout died as a result.
The fertilizer-like quality of the substance can also help aggressive invasive plants grow, sometimes choking out sensitive local species.
But Stahl says his biggest issue is that there is no statistical evidence the stuff reliably helps contain fires.
"We can find individual anecdotal examples where yes, it appears that in this one place we dropped the retardant and the fire stopped," he said. But he adds there are also plenty of examples where the retardant appears to have done nothing.
One tactic of many
Glen Stein studies retardant for the US Forest Service.
He says the reason there are no studies showing the effectiveness of retardant in the field is because each fire happens amid a unique set of circumstances making it hard to compare cases.
"There are so many variables," he said.
But Stein points out that there are hundreds of lab tests showing that retardant slows fires in controlled settings.
And he adds that the Forest Service is continually refining its rules on using the substance safely.
He says retardant isn't meant to stop a fire on its own. It's only one tactict firefighters use to protect forests and homes.
Correction: An earlier version of this story incorrectly spelled the name of Phos-Chek. KPCC regrets the error.