Is spinach actually full of iron? The myth has a myth inside it

Spinach is a great source of iron. You’ve probably heard that at a dinner table at some point, and if you’ve spent any time around people who like correcting things, you’ve also heard the comeback: it isn’t, because a German chemist put a decimal point in the wrong place in the 1870s, made spinach look ten times richer in iron than it really is, and Popeye did the rest.

Both stories are wrong, and the way they’re wrong is the fun part. Spinach really does contain a fair amount of iron, and your body absorbs almost none of it. In the two human studies that measured it directly, people took up less than 2%. And the decimal point, as far as anyone has been able to find, never moved.

On paper, spinach really does have a lot of iron

The label numbers aren’t subtle. In the USDA’s food database, raw spinach has 2.71 mg of iron per 100 grams.^[1] Cooked spinach has 3.57 mg,^[2] because cooking drives out water and shrinks the leaves (if you’ve ever watched a full pan of leaves wilt down to a few forkfuls, that’s where the water went). A cup of cooked spinach weighs about 180 grams, so it carries roughly 6.4 mg of iron.

For scale, the NIH recommends 8 mg a day for adult men and 18 mg for women aged 19 to 50.^[3] And gram for gram, cooked spinach has more iron than 85% lean ground beef, which comes in at 2.68 mg per 100 g.^[4]

So if you only read the label, the original claim looks fine. But iron in food isn’t iron in you, and that’s where it falls apart.

Almost none of it gets into your blood

The cleanest test of this is old and small, and it’s still the best direct evidence there is. In 1969, a team publishing in the journal Blood did something clever: they grew spinach in a nutrient solution laced with radioactive iron, so the tracer was built into the plant itself rather than sprinkled on afterward.^[5] Nine volunteers ate it, and the researchers measured how much of that labeled iron ended up in their red blood cells.

The average was 1.7%. The same people absorbed 16.6% of a reference dose of ferrous ascorbate (an easily absorbed form of iron), and a separate group of nine eating labeled veal absorbed 20.3%.

A 1983 study in the British Journal of Nutrition came at it from a different angle.^[6] It added a small dose of iron to each vegetable meal and measured how much of that got absorbed, in a group of South African women of Indian descent, many of them iron deficient. Spinach came in at 1.4%, among the lowest of the vegetables tested. Cabbage and broccoli, by comparison, came in around 26 to 32%.

Figure 1How much of a meal’s iron actually got absorbed

1969 study, nine volunteers per group

Spinach1.7%

Reference iron dose16.6%

Veal20.3%

1983 study, iron added to vegetable meals

Spinach1.4%

Cabbage and broccoli26–32%

Now run the arithmetic on that cup of cooked spinach. 6.4 mg of iron at 1.7% absorption is about 0.1 mg that actually reaches your blood.

Figure 2One cup of cooked spinach, on paper and in you

6.4 mgiron on the label

absorbed →

≈ 0.1 mgreaches your blood

Two caveats, though. These are tiny, single-meal studies from decades ago, and nobody seems to have repeated the spinach-only measurement with modern methods. And absorption depends a lot on who’s eating. The 1969 authors predicted (from their data, rather than measuring it directly) that the same spinach would yield about 0.8% in a man with healthy iron stores and somewhere around 5 to 7% in someone with iron-deficiency anemia, because the body pulls harder when it’s running low. They called part of that calculation conjectural themselves. Even at the high end, that cup gives you well under half a milligram.

It probably isn’t the oxalate, either

The usual explanation for all this is oxalic acid. Spinach is full of it, oxalate binds minerals, and so the iron gets locked up before you can use it. It’s a tidy mechanism, it sounds like proper biochemistry, and I used to repeat it myself.

The best test of it points the other way. In a 2008 trial in Zurich, researchers gave 16 young, healthy women meals of bread rolls with kale (which has almost no oxalic acid), and then the same meal with the amount of oxalic acid found in a spinach meal added.^[7] If oxalate were the culprit, absorption should have dropped. It didn’t: 10.7% without the added oxalate, 11.5% with it, a difference that could easily be chance. (The spinach meal itself still did somewhat worse than kale, though not by a statistically clear margin.) The authors wrote that their findings “strongly suggest that OA in fruits and vegetables is of minor relevance in iron nutrition,” OA being oxalic acid.

So what’s going on? The likelier suspects are polyphenols. In the 1983 study, how poorly a vegetable’s iron was absorbed tracked its polyphenol content closely,^[6] and the NIH fact sheet now names polyphenols, not oxalate, when it explains why spinach has “low iron bioavailability.”^[3] Calcium may play a part too. That’s the favored explanation rather than a settled one, but it fits the data better than the one everybody repeats.

The famous decimal point was never misplaced

Now for the correction everyone loves. It usually goes like this: in 1870 (or the 1890s, depending on who’s telling it), a German scientist measuring spinach’s iron slipped a decimal point, recorded ten times the real amount, and the error sat uncorrected for decades while Popeye built a career on it.

It’s a perfect anecdote for people who like debunking things, which is probably why it spread. The trouble is, it doesn’t appear to have happened.

The criminologist Mike Sutton went back to the original sources and traced the story.^[8] By his account, the high nineteenth-century figures came from Emil von Wolff in 1871, and a Professor Sherman in the United States explained them in 1907 as iron contamination from the heating dishes “and other bad science,” with no slipped decimal involved. Gustav von Bunge, often cast as the villain, measured spinach’s iron correctly in 1892.

The decimal point itself first shows up much later, and only as a guess. In 1972 the nutritionist Arnold Bender wrote that the fame of spinach “may well have grown from a misplaced decimal point.”^[10] In 1981, a BMJ article by the haematologist Terence Hamblin stated it as fact, with dates attached.^[9] It gave no reference for the error, as the sociologist Ole Bjørn Rekdal later pointed out. From there it was repeated by one source after another, with the date of the error drifting between 1870, the 1890s and the 1930s.

Rekdal, who studied the story as an example of how false citations spread through academic writing, put it plainly in a 2014 paper: “Nothing indicates that the decimal point error ever was made.”^[10] And when Sutton asked Hamblin where he’d got it, Rekdal reports, Hamblin couldn’t remember, but was sure he hadn’t made it up. (Which is roughly how most of us pass things on.)

Figure 3How a decimal point that never moved became a famous fact

1. 1871Emil von Wolff publishes the high iron figures for spinach.
2. 1892Gustav von Bunge measures spinach’s iron correctly.
3. 1907Professor Sherman puts the high figures down to contamination “and other bad science.”
4. 1932Popeye says he eats spinach for vitamin A.
5. 1972Arnold Bender suggests spinach’s fame “may well have grown from a misplaced decimal point.”
6. 1981A BMJ article states the error as fact, with no reference.
7. 2014Rekdal: “Nothing indicates that the decimal point error ever was made.”

Popeye ate spinach for vitamin A

Which leaves the sailor himself. According to Sutton, the one strip in which E.C. Segar’s Popeye explains why he eats spinach (from 1932) has him say:^[8]

“Spinach is full of Vitamin A. An’ tha’s what makes hoomans strong an’ helty!”

That wording comes from Sutton’s transcription rather than from the strip itself, so hold it a little loosely. But it fits the food: raw spinach has 469 micrograms of vitamin A activity per 100 grams,^[1] which is a lot. Popeye, if Sutton’s right, picked the nutrient that spinach is actually rich in.

Why the wrong stories were so easy to believe

A debunk with no citation borrows the credibility of the thing it’s mocking.

Why the wrong stories stick

Each version of this survived for the same reason: it was satisfying.

“Spinach is full of iron” had a real number on a real label behind it. What it skipped was the step between eating something and absorbing it, which stays invisible unless someone feeds people labeled iron and measures their blood.

The oxalate explanation survived because it sounds like a mechanism, and mechanisms feel like proof. Most people repeating it (me included) never asked whether anyone had tested it.

And the decimal point survived because it flattered the person telling it. It’s a story about other people being careless with numbers, told by someone who clearly isn’t. Nobody checks the source of a correction, because checking sources is what the correction is supposed to be about. That’s what makes a debunk with no citation so durable: it borrows the credibility of the thing it’s mocking.

So where does iron actually come from?

If spinach isn’t it, the obvious next question is what fills the gap. A few things hold up well.

Iron from meat, poultry and fish is absorbed far better than iron from plants. In the 1969 study, the labeled veal was absorbed at about twelve times the rate of the spinach.^[5] The NIH puts absorption from mixed Western diets at roughly 14 to 18%, against 5 to 12% for vegetarian diets, which is why it suggests vegetarians need 1.8 times as much iron.^[3]

In the US, about half of dietary iron comes from bread, cereal and other grain products, according to the same fact sheet.^[3]

Vitamin C helps plant iron get absorbed. In a 1977 study of 63 men, adding vitamin C to a meal raised non-heme iron absorption in proportion to the dose, though the boost was much smaller when the meal already contained meat.^[11]

And spinach isn’t uniquely bad among plants. Lentils were absorbed nearly as poorly in the 1983 study,^[6] but they start with a similar amount of iron per cup,^[12] so you end up with somewhat more (roughly 0.16 to 0.21 mg absorbed per cup, by the same arithmetic). The real gap is between iron from animals and iron from plants, not between spinach and everything else.

None of that makes spinach a bad food. It’s a good vegetable with plenty of vitamin A. It’s just not the iron story.

The question to carry around

You don’t need to remember any of the numbers above. The part worth keeping is the shape of the mistake, because it comes around again and again.

When you hear a confident correction, the kind that starts with “actually, that’s a myth,” ask it the same question you’d ask the original claim: who said it first, and what did they cite? For the spinach decimal point, the answer turns out to be “a nutritionist who said may well have, and a journal article that didn’t cite anything.” It took Sutton and Rekdal real digging to establish that. It took the story about thirty years to travel the world without it.

And if the question underneath all this is “am I getting enough iron?”, food tables can’t answer it. That’s a blood test, and a question for your own clinician rather than a blog.

Enjoy the rest of your Sunday (and the spinach, for what it actually is).

---

Sources

1. USDA FoodData Central. Spinach, raw (FDC ID 168462). fdc.nal.usda.gov
2. USDA FoodData Central. Spinach, cooked, boiled, drained, without salt (FDC ID 168463). fdc.nal.usda.gov
3. NIH Office of Dietary Supplements. Iron: Fact Sheet for Health Professionals. ods.od.nih.gov
4. USDA FoodData Central. Ground beef, 85% lean, pan-broiled (FDC ID 174033). fdc.nal.usda.gov
5. Layrisse M, Cook JD, Martinez C, et al. Food iron absorption: a comparison of vegetable and animal foods. Blood. 1969;33(3):430–443. PDF
6. Gillooly M, Bothwell TH, Torrance JD, et al. The effects of organic acids, phytates and polyphenols on the absorption of iron from vegetables. British Journal of Nutrition. 1983;49(3):331–342. doi:10.1079/BJN19830042
7. Storcksdieck genannt Bonsmann S, Walczyk T, Renggli S, Hurrell RF. Oxalic acid does not influence nonhaem iron absorption in humans: a comparison of kale and spinach meals. European Journal of Clinical Nutrition. 2008;62(3):336–341. PubMed · Journal page
8. Sutton M. How the spinach, Popeye and iron decimal point error myth was finally bust. HealthWatch Newsletter. 2016;(101):7. PDF
9. Hamblin TJ. Fake! BMJ. 1981;283:1671–1674. doi:10.1136/bmj.283.6307.1671
10. Rekdal OB. Academic urban legends. Social Studies of Science. 2014;44(4):638–654. Open access
11. Cook JD, Monsen ER. American Journal of Clinical Nutrition. 1977;30:235–241. PubMed
12. USDA FoodData Central. Lentils, cooked (FDC ID 172421). fdc.nal.usda.gov