By Multimotyl (Own work), via Wikimedia Commons

The recent revelation (via a class-action suit) that some cheaper wines have high levels of arsenic got m to thinking: how toxic is it, really? And is it as dangerous as we think?

I’m not one to say that stuff like that in wine is a good thing. I’m a wee bit of an oenophile myself, though I strongly advocate staying under $20 for a good bottle for most occasions.

So the question is, if wine has arsenic in it, how much are we getting when we drink it, and how that compares with water from the tap.

First let’s look at how much was in the wine. According to this story, the highest arsenic levels were 50 parts per billion in one bottle (out of 1,300) that was tested. On average the levels were 30 parts per billion. The standard for drinking water is 10 parts per billion.

One part per billion is equal to one gram of arsenic in 1 million liters of water, which is enough to fill a “short course” 25-meter swimming pool. Not much, but arsenic is damned toxic stuff.

A lethal dose is 2-20 milligrams per kilogram of body weight, which means that someone weighing about 75 kg (or 165 pounds) would die if they ingested as little as 150 milligrams of arsenic as arsenic trioxide (As₂O₃). Arsenic trioxide is an odorless and tasteless white powder made famous by pre-modern assassins because it dissolves in water. Among the victims were a Chinese Emperor and Francesco I de’ Medici.

Arsenic was a great poison because the symptoms, abdominal cramps and vomiting, might be food poisoning or a lot of other things — in pre-modern times there really was no way to test for it.

However, let’s be clear: 150 milligrams is a lot compared to what’s in a bottle of wine, even at the highest levels. At 50 ppb in a 750 milliliter bottle, that’s 3.75 × 10^-5 grams of arsenic (I assumed that wine weighs as much as water does, though it’s probably a wee bit less because of the alcohol), or 0.0375 milligrams – a tiny portion of the minimum lethal dose.

Does this mean there is nothing to worry about? Not quite. Arsenic is dangerous in water supplies because it can cause cancer over time. It is a problem in New England wells, as studies showed  higher bladder cancer rates where people drink from wells. One scientist described it to me thus: a map of bladder cancer cases in New England “lights up like a Christmas tree”  that neatly maps out areas where most people drink from local wells, where naturally occurring arsenic leaches into water from the surrounding rock.  (The problem is much reduced for people on municipal water supplies because the water comes from reservoirs and rivers).

There’s some dispute, though, over how much arsenic causes elevated cancer rates – a study from Taiwan seemed to show that 150 ppb was the point that cancer rates went up, and that’s higher than other studies from the U.S. On the other hand it’s possible that prolonged exposure over generations makes people tolerant, as might have happened to a group of Native people living in the Atacama desert.

How much arsenic is someone who drinks the cheap bottle of wine with 50 ppb in it getting over time? For that you have to look at how much wine they are drinking. One of the reasons arsenic standards for drinking water and wine (or other beverages) are different is simply that people drink more water. Unless you have an alcohol problem like Tyrion Lannister you probably aren’t drinking wine every single day.

The Wine Institute says Americans drank some 779 million gallons of table wine in 2013. That works out to 2.94 billion liters. If all that wine had 50 ppb of arsenic, that’s 147 kilograms of arsenic spread among 300 million people, or just about 0.49 milligrams of arsenic per person per year. That’s less than a quarter of a minimum lethal dose ingested in a whole year.

Now, that calculation (as many will point out) doesn’t take into account that every man, woman and child in the US isn’t drinking wine. But even cutting down the number of people by a factor of four means that on any given day there just isn’t that much arsenic going into our collective systems from wine.

To add a little more data, 80 percent of Americans have one drink per day or less. So if you are a glass-of-wine-per-day person, you’re only going through about a fifth of the bottle, getting 0.0075 milligrams of arsenic at 50 ppb.

Put another way, if you drink that bottle of 50 ppb wine all by yourself it is far from enough arsenic to kill you. If you drank one of those bottles every single day you’d get to a lethal dose in about three months – assuming your body retained all the arsenic, which it doesn’t.

The real concern is long term consumption. So people drinking cheap wine for several years running might be at higher risk for certain cancers. An interesting study would be looking at areas where Two Buck Chuck sells well and then looking at bladder cancer rates. You’d have o factor out the drinking water – arsenic levels are high in well water in the American West, and then track the bladder cancer rates over time.

And if I were to throw out a hypothesis, the reason for high rates of arsenic might be the soil where the wine is grown. If arsenic levels are high in well water in parts of California where wine grapes are cultivated, then one would expect some of that to show up in the plants that grow there also. A map of arsenic levels here shows a little bit of correlation – the stretch near Lodi and in Napa has a higher-than-normal concentration of arsenic in the water. But that doesn’t mean the grapes used in the wines tested for arsenic came from there. It’s suggestive though, and if anyone has any real studies on the topic I’d love to know.

There’s no reason to be drinking more carcinogens than you have to, and arsenic is pretty rough stuff. But it’s important to add a little perspective. If you drink water from a private well and live in Maine, odds are you’re getting far more arsenic that way than from the wine you got at Trader Joe’s.