Modern Wheat vs Ancient Wheat: What Plant Breeding Actually Changed

The durum wheat behind most Italian pasta today descends from a variety created inside a nuclear research centre. Its parent seed was irradiated with X-rays in 1959. It reached Italian tables in 1975, and the public was not told how it had been made until 1984, nine years later. Between 1915 and 1975 wheat was deliberately rebuilt to be shorter, earlier, and far more productive, and the whole design depended on one thing the wheat of our grandparents never needed: synthetic fertiliser.

Here is the message I get most often, and I have stopped being surprised by it.

"I was in Italy for two weeks. I ate pasta every day, bread with every meal, and I felt fine. I got home, ate the same food here, and within a day I was miserable again."

I am not going to tell you why that happened, because I do not know and neither does anyone who tells you they do. What I can do is show you what was actually done to wheat, by whom, and in what year, and then let you ask the obvious question yourself.

Table of contents

1. What "ancient wheat" and "modern wheat" actually mean
2. The changes that happened before anyone was breeding
3. Nazareno Strampelli: the man who made Italy's wheat short and early
4. Norman Borlaug, Norin 10 and the Green Revolution
5. Creso: the durum that came out of a radiation program
6. Four breeding techniques, side by side
7. What measurably changed, and what did not
8. Genetic erosion: the cost nobody argues about
9. Is the gluten fundamentally different? The question worth asking
10. So should you choose ancient grains? An honest answer
11. Frequently asked questions
12. What this means for the pasta on your plate
13. References

What "ancient wheat" and "modern wheat" actually mean

Almost everything that follows comes from one book, Marco Pallotti's Grani Antichi, and I will tell you every time a figure is his own account rather than a cited study.

Landrace vs cultivar: the distinction that actually matters

Neither phrase has a legal definition. A landrace is a farmer's population, mixed, never fixed by a breeder; Pallotti writes le Saragolle in the plural for exactly that reason. A cultivar is one genotype, held there.

Wheat makes that easy. It self-pollinates, with cross-pollination under 2–4%, so once a line exists it breeds true, come se fossero dei cloni, as the book puts it. As if they were clones.

The three chromosome families: einkorn, emmer/durum, spelt/bread wheat

Every wheat on earth sits in one of three families, each a multiple of seven base chromosomes.

Durum carries about 80,000 genes and bread wheat about 125,000, against a human's 46 chromosomes and 20,000–25,000 genes. Both dates are estimates.

"Ancient" and "heirloom": what each word is actually telling you

Neither word is regulated, so both do real work and both can be stretched. Here is how I read them, and how I would like you to read mine.

Ancient points at the species. Einkorn, emmer, spelt, and khorasan are ancient wheats: they are botanically distinct from the durum and bread wheat that fill a modern supermarket, and they carry a different number of chromosomes. Einkorn has 14. Durum and khorasan have 28. Bread wheat has 42. That is a real, checkable difference, and it is the oldest one in this article, those species separated long before anybody was breeding anything.

Heirloom points at the variety, and at time. A heirloom is a named variety that farmers kept going, saved seed from, and handed down, Tumminìa in Sicily, Senatore Cappelli across the south, Russello, Perciasacchi. It usually predates the industrial breeding programmes, and it was selected by growers for the field they actually farmed rather than by a laboratory for national yield.

So a grain can be one, or the other, or both. Tumminìa is an ancient species and a Sicilian heirloom. Senatore Cappelli is a heirloom durum, genuinely handed down, genuinely pre-industrial in character, but it was bred by Strampelli in 1915, so it is a heirloom rather than an ancient species. Modern durum is neither. If you want the grain-by-grain version, I went through the ancient grains one at a time separately.

Two details worth carrying to a shop. Watch the plurals: le Saragolle are old landrace populations, while Saragolla in the singular is a durum cultivar registered in 2004. And a landrace is not a variety: it is a mixed farmer's population that was never narrowed to a single genotype, which is exactly why it survived centuries of bad seasons.

The changes that happened before anyone was breeding

Two natural crosses, roughly 300,000–500,000 and 8,000–9,000 years ago

Wheat is an allopolyploid: the tetraploid and hexaploid wheats carry whole genomes from other species. AA plus BB gives not a hybrid but a new species, AABB. It works only by accident: an odd number of chromosome sets makes a plant sterile, which is why the cultivated banana has no seeds.

Wheat's chromosome family tree

Two prehistoric hybridisations, and a count that modern breeding never changed

Every wheat on earth sits at one of three levels. Both jumps happened long before anyone was breeding anything on purpose.

AA · 14 chromosomes

Einkorn, Triticum monococcum ssp. monococcum. The oldest cereal domesticated by humans, from wild T. boeoticum. (pp. 13–14)

AABB · 28 chromosomes

Wild emmer first, then emmer, durum and khorasan. Formed when T. urartu crossed with an Aegilops speltoides-like wild grass whose donor species has never been identified and is probably extinct. (pp. 14–15)

AABBDD · 42 chromosomes

Spelt and bread wheat, formed when domesticated emmer crossed with Aegilops tauschii, a wild goat grass, in the Armenia and south-west Caspian region. (pp. 16–17)

The chromosome count has not moved since prehistory. Nothing in twentieth-century breeding added a chromosome to wheat.

Both jumps happened spontaneously, and Pallotti is honest about the limits: how polyploidization happened is still not clear, and the family tree is contested. Several studies argue that European spelt arose later than bread wheat, by separate crosses of its own, Dvořák, Dedkova, Blatter, Matsuoka and Nasuda among them, listed in the references below. Most pages that draw this tree draw it as settled. It is not.

So was wheat crossed with goat grass? Yes, twice, and long before us

You may have read that modern wheat was crossed with a Japanese goat grass in the 1950s. Half of that is true, and it is the wrong half. Aegilops tauschii is a wild goat grass and did contribute the D genome, with nobody present. The dwarfing donors of the twentieth century, Akakomugi and Norin 10, are Japanese wheat.

Nazareno Strampelli: the man who made Italy's wheat short and early

Now the part people actually did on purpose.

What was done to wheat, 1911 to 1987

Three countries, four institutions, and two programmes that ran at the same time

The bands on the chart overlap because the programmes did. Casaccia's mutagenesis work ran alongside the Green Revolution, not after it.

Italy, 1911–1940, Strampelli

Nazareno Strampelli, called il mago del grano, gathered 250+ landraces and crossed them against a Japanese dwarf. Senatore Cappelli 1915; Ardito 1920. By 1932 his varieties covered 30% of Italy's wheat area, and 50% by 1940. (pp. 31–35)

Mexico → South Asia, 1943–1972

Borlaug reached Mexico in October 1944 and learned of the dwarf variety Norin 10 in 1952. Mexican production went from 2.5 to 10 million quintals. India and Pakistan approved the low-stature wheats in 1964 after severe shortages. (pp. 38–40)

Italy, 1955–1975, Casaccia

After the 1955 Geneva “Atoms for Peace” conference, a radiation-in-agriculture laboratory opened at Centro Ricerche Casaccia. The first X-ray mutants of Senatore Cappelli were produced at Pisa in 1959; the Cobalt-60 campo gamma followed in 1960. (pp. 42–43)

Creso was registered and approved across 1974–75; the sources differ on which year carries the patent, so both are given. Several dates here are Pallotti's own account with no citation attached.

250 landraces and one goal

The first name in the file is Nazareno Strampelli, and he did not work alone, his wife, Carlotta Parisani, worked beside him at the Regia Stazione Sperimentale di Granicoltura in Rieti. What he wanted was mundane: a plant that ripened before the summer storms, and a stem stiff enough to hold a heavy head up. Old Italian soft wheats lodged, flattened days before harvest.

The Ardito pedigree: four countries in one seed

Start in England, with a wheat called Squarehead. Cross it with a Dutch one, Zeeuwse Witte, and you get Wilhelmina Tarwe. Cross that with Rieti, an Italian landrace, and you get two lines: Selezione 21 and Selezione 67.

Then Strampelli crossed each of them, separately, with a Japanese wheat called Akakomugi. Selezione 21 gave Ardito and Mentana; Selezione 67 gave Villa Glori and Damiano. Four countries, one seed.

Akakomugi was poor on its own. He wanted it for two things only, short stature and earliness.

Senatore Cappelli, 1915: the durum everyone now calls ancient

The durum story runs the other way, and it is the part most people have backwards when they write to me. Senatore Cappelli was not a hybrid: Strampelli created it at the Stazione Fitotecnica in Foggia by genealogical selection from a North African landrace, Jenah Rhetifah. It stayed tall, matured late, and lodged.

How it spread: the Battle of the Grain, 1925–1940

In June 1925 Italy launched the Battaglia del Grano, and royal decree 1314, of 29 July 1925, covered half the cost of seed facilities. Even so, in 1927 Ardito was only sixth by cultivated area, Gentil Rosso on 999,201 hectares against Ardito's 175,807. The takeover was not instant, whatever the posters said.

Norman Borlaug, Norin 10 and the Green Revolution

The next name is the one everybody has heard of.

The lodging problem that fertilizer created

Nitrogen fertilizer raises yield by making a bigger head of grain, and a bigger head on a tall stem falls over. That is the whole problem, and Strampelli had already hit it in Rieti. The motive behind short wheat was never nutritional.

What a dwarfing donor actually does

A dwarfing donor is a variety you cross in for one trait and nothing else. Borlaug used Norin 10, a low-stature Japanese bread wheat, the way Strampelli had used Akakomugi. Nothing was inserted, and nothing was irradiated.

Mexico, Pakistan, India, and a Nobel Prize

It started as diplomacy, not agronomy. Henry Wallace went to Mexico in 1940, the Rockefeller Foundation met about it in February 1941, three experts filed a report in 1942, and the Mexican government had a wheat program the year after that.

Borlaug walked into that program and was expected to lift a national yield average with a plant that kept falling over. India and Pakistan approved his low-stature wheats after severe shortages, and the Nobel Peace Prize came in 1970.

Most of that cluster is Pallotti's own account, with no citation attached. I flag it because I would want it flagged for me.

The part that gets left out: the wheat was designed around the fertiliser

Read the sequence again, because the order matters and almost nobody tells it in the right order.

Synthetic nitrogen came first. It raises yield by building a bigger head of grain. A bigger head on a 160-centimetre stem falls over in the first storm: that is lodging, and it was the bottleneck. So the plant was redesigned around the input: breed the stem short and stiff, and it can carry a heavier head and take more nitrogen without collapsing. The height went not because anyone wanted a shorter plant, but because a shorter plant could absorb more fertiliser.

Ardito, in 1920, is already described as highly productive given adequate nitrogen. When India and Pakistan were offered the new wheats in the 1960s, their governments hesitated: the varieties needed different sowing schedules, more irrigation, and chemical fertilisers and herbicides. They adopted them in 1964, after severe food shortages. The record is explicit that the yield gains were tied to chemical fertiliser, herbicide, and fungicide inputs.

That is worth sitting with. Synthetic nitrogen is made industrially from fossil gas. The wheat in most supermarket pasta is a plant bred to eat it: it does not perform the same way without it. The heirloom varieties that came before were selected under the opposite conditions: whatever the field gave them, in a valley with no irrigation.

I want to be careful here, because this is where honest history gets turned into a health claim. The agronomic dependence is documented. The idea that those inputs harmed human health is the author's own hypothesis, carrying no citation, and the knowledge base grades it as speculative. I am telling you about the farming, not about your body.

Creso: the durum that came out of a radiation program

This is the chapter I had to read twice.

Pisa, Casaccia, and the 1959 mutants

After the 1955 "Atoms for Peace" conference in Geneva, Italy opened a laboratory for radiation in agriculture at the Centro Ricerche Casaccia. Here is the part that surprised me, and it is worth slowing down for: the X-rays were not fired there. Francesco D'Amato treated the seed at Pisa, and the Cobalt-60 campo gamma: the gamma field, outside Rome did not exist yet.

How Creso was actually made, and why it is not a GMO

This is the pedigree everything else hangs on. It has three steps.

How Creso was actually made

A cross between an X-ray mutant and a Borlaug semi-dwarf, not a directly irradiated variety

This is the pedigree as the source prints it. Senatore Cappelli sits on both branches, once as a parent, once as the variety that was irradiated.

The Borlaug side

Yactana 54, a Mexican durum, crossed with the Japanese dwarf Norin 10 to give a first hybrid; that crossed with Senatore Cappelli to give the second. Bozzini obtained the semi-dwarf material from Borlaug personally in 1968. (p. 43)

The mutant side

Senatore Cappelli seed was treated with X-rays by Francesco D'Amato at Pisa in 1959, producing two mutant lines, B132 Castelporziano and B144 Castelfusano. (p. 43)

What Creso is not

In the source's own words: “La nuova varietà Creso non fu quindi ottenuta tramite mutazione diretta indotta da radiazioni.” Creso was not obtained through direct radiation-induced mutation, but through a cross. (p. 43)

Creso's mutagenesis-linked origin was first disclosed publicly in June 1984, nine years after it entered Italian pasta. (p. 46)

Each parent brought something specific: high yield and brown-rust resistance from the Borlaug side, grain size and high vitreousness from the mutant side. The conclusion is Pallotti's own, not mine: Creso cannot technically be considered a GMO. It is a conventional cross, one of whose parents was mutagenized.

1972 to 1987: how one variety took over Italian pasta

Creso went into the fields in 1972. The request to register it went to the Minister of Agriculture in 1974, and approval came in 1975, though Pallotti dates the patent to 1974 on one page and 1975 on another, so I give you both.

In 1981 it held 22.5% of Italian durum production and peaked six years later, while Senatore Cappelli fell from 6.5% in 1981 to 1.7% by 1990, figures from Motzo and colleagues, University of Sassari.

What was tested, and what was not

Here is the regulatory fact, as Pallotti recorded it in the mid-2010s. In Italy, transgenic organisms were banned from cultivation and sale, while plants obtained by mutagenesis faced no specific regulation, and no safety-testing regime existed for conventionally bred varieties in the 1970s.

Here is where I stop. The claim that no hybrid wheat was ever safety-tested rests, in this book, on a narrative review and a mass-market diet book. "It was never required" is about paperwork, not about a grain.

Four breeding techniques, side by side

People do not really want to know how wheat was bred. They want to know whether their pasta is a GMO. That question has an answer, and it takes four boxes.

Four ways a wheat variety gets made

Only one of them is genetic engineering

“Is it a GMO?” has a real answer, and it is a taxonomy rather than a yes or no. Four techniques, in the order they arrived.

Genealogical selection, not a GMO

Picking the best plants out of a variable landrace, generation after generation, until the type is fixed. Introduces nothing that wasn't already in the population. Senatore Cappelli, 1915.

Hybridisation, not a GMO

Moving pollen from one variety to another and selecting from the offspring. The technique behind Strampelli's Ardito, 1920, and behind almost every wheat in a supermarket today.

Induced mutagenesis, not classed as a GMO

Treating seed with radiation to raise the natural mutation rate, then selecting whatever useful variant appears. It rearranges the plant's own DNA. Castelfusano B144, 1959.

Transgenesis, is a GMO

Inserting DNA from a different organism. This is the only one of the four that meets the legal definition. No transgenic wheat has ever been legally commercialised anywhere; MON 71800 was developed and withdrawn.

Mutagenesis rearranges the plant's own DNA; transgenesis inserts DNA from another organism. The regulatory position on mutagenesis is a mid-2010s snapshot in the source, re-verify before publishing.

The line that actually settles it is whether a technique can put in a trait that was not already there. Selection cannot: it only narrows what the population already held. Crossbreeding can, in new combinations, but only from parents able to breed with each other. Mutagenesis can, at random, by damaging the plant's own DNA and letting the breeder keep whatever useful thing appears. Only transgenesis inserts DNA from another species, and only transgenesis is regulated as a GMO.

The Italian regulatory position above is Pallotti's mid-2010s snapshot and should be re-verified before you rely on it; it covers Italy only. For the European position the instrument is Annex I B of Directive 2001/18/EC, confirmed by the Court of Justice of the European Union in Case C-528/16. MON 71800 was developed between 1998 and 2005 without authorization and found growing in Oregon in May 2013. That no transgenic wheat has ever been legally commercialized comes from a 2016 academic analysis, not from field data.

What measurably changed, and what did not

Height: over 160 cm to under 80 cm

You may have read that wheat went from six feet to two. Here are four named Italian varieties with the heights Pallotti records.

Sixty years, eighty centimetres

Four Italian wheats on one baseline, drawn to relative height

Pallotti records the heights below. Nothing about a plant's height says anything about how its grain behaves in a person: this is agronomy, not nutrition.

Gentil Rosso, pre-1920

A soft wheat grown before the breeding programmes began. Over 160 cm tall, and prone to lodging, falling over in wind and rain. (p. 32)

Senatore Cappelli, 1915

Strampelli's durum, selected from the North African landrace Jenah Rhetifah. About 160 cm, late-maturing, and also prone to lodging. (pp. 35–36)

Ardito, released 1920

About 80 cm, and ripening 15–20 days earlier than what it replaced. Half the height, and off the field before the summer storms. (p. 32)

Creso, first grown 1972

Under 80 cm. By 1987 it accounted for 42.6% of Italian durum production (Motzo et al., 2003). (pp. 43, 45)

Heights as recorded by Pallotti in the breeding literature; individual plants vary with season and soil. Three of the four sit on sequence-inferred pages.

Eighty centimetres is about the height of my kitchen counter. That is the entire change people call tampering, and it happened because of wind, rain, and a bag of fertilizer.

Yield: two numbers, twenty-eight years and an ocean apart

About 5 quintals a hectare: the Mexican national average for soft wheat in 1944. About 10 tonnes a hectare: one Italian durum variety, Creso, in 1972.

Both numbers are real, and I will not multiply them for you. Different country, different species, different measurement basis, twenty-eight years apart. Their ratio is not a fact; it is arithmetic performed on two unrelated things.

Gluten strength (the W index): a baking number, not a health number

W is a dough-strength index: a refined-flour dough is inflated with compressed air on a Chopin Alveograph, which records the pressure inside the bubble until it bursts. IG is a different parameter: the Gluten Index, a 0-to-100 scale for the strength of the bonds that form within gluten. Gluten Index. Not the glycemic index; there is no blood-sugar measurement anywhere in this source.

Both rose. An Italian cereal-research trade report documented a modern durum group of Creso, Duilio, and Simeto averaging W 188 and Gluten Index 69. Pallotti gives Senatore Cappelli's baseline himself, without a source: W 95 and Gluten Index 11.

Of Italian soft wheats registered in 1974–76, not one fell into the strongest class, W 300 or above. By 2010–12 that class had gone from zero to thirteen on the report's own scale: the source prints those figures without a unit, so read it as the share of registrations moving upward, not as a headcount of varieties.

Two things I want to be honest about. The measures do not move together: one modern variety in that table sits at W 116 with a Gluten Index of 97. And the reason for the rise is dull, baking went industrial, so breeders worked on the disulfide bonds between cysteine residues in the glutenins. These numbers describe what flour does in a mixer.

Here is every measured change in this article in one place, with the species and the year span on every row.

Plant heights are in the height block above. The two yield rows are deliberately separate and not comparable: different countries, species, and measurement bases. IG here means Gluten Index, a 0–100 measure of gluten bond strength used in milling and pasta-making. It is not the glycemic index.

What the evidence does not show

It would be easy to let a pile of facts finish a sentence nobody in the record actually wrote. So let me write the other half plainly.

The chromosome number did not change. Einkorn is 14, durum and khorasan are 28, spelt and bread wheat are 42: the counts they carried in prehistory. Twentieth-century breeding did not add a chromosome to wheat.

Total gluten content is not what was measured. What this record measured is strength and composition, what the gluten does in a dough, and which proteins make it up. It contains no measurement of total gluten content for any variety, old or new, in either direction. Anyone who tells you that number went up, and anyone who tells you it stayed flat, is working from something I cannot show you.

And nothing in this history measured anything about how any wheat affects any person. These breeding chapters hold no biochemical, nutritional, or clinical evidence at all. No human data here connects a dough-strength number to anything inside a body, and where the book reaches that way it rests on one unpublished master's thesis using a single in-vitro method.

One finding cuts against the tidy story. Bozzini, a co-creator of Creso, tested the Casaccia collection of ancient and improved varieties in the late 1970s and reported the proteins involved in celiac disease present, in highly variable amounts, in all of them, grain proteins measured, not people. Pallotti's own hedge deserves quoting whole, because half of it would be a claim: modern wheats, he writes, "possess molecules structurally different from those contained in ancient grains (first and foremost the structure of the proteins that make up gluten), whose impact on human health has not yet been determined nor clearly understood." No study is cited for that health clause.

A note on study quality, plus the independent-research line

The production-share data comes from university agronomists and is the strongest set here. The dough-strength figures come from a non-peer-reviewed trade magazine, which is why I wrote "documented," not "studies show." Two of the sources for the Casaccia program are the successor institution describing its own work. And several dates and heights are Pallotti's own uncited account, which I have flagged wherever they appear.

These findings come from independent research and do not guarantee the same results with our product.

Genetic erosion: the cost nobody argues about

From hundreds of landraces to a handful of cultivars

One change both camps agree on: wheat lost its variety. In 1927 Italian fields were a crowd. The Saragolle covered 200,097 hectares of what was then classified as durum-type wheat, alongside Russie, Samartinara, and Realforte, and on the soft-wheat side, nine named varieties, every one of them above 129,000 hectares. Six decades later one cultivar supplied more than two-fifths of Italy's durum crop.

What germplasm banks saved

Some survived on purpose. Strampelli's landraces from every continent were the raw material for everything he made, and the Casaccia program kept a collection of both ancient local and improved varieties. Three landraces are still cultivated in Italy: the emmer of the Garfagnana, the Solina soft wheat, and the Timilìa durum. The book gives their names and nothing else. So neither will I.

Ferrari and Jeep: a breeder's turn of phrase

A metaphor from the breeding literature, labelled as one

Nobody quantified this. It is a turn of phrase, and I am giving it to you as a turn of phrase, because it captures the trade honestly.

The Ferrari

A modern high-yield variety is extraordinary on a paved road: the right soil, the right water, the right inputs, applied on schedule. Take the road away and the performance goes with it.

The Jeep

An old variety like Senatore Cappelli is slower, and it will travel roads of every kind. It was selected by farmers in the field they actually had, not by a programme optimising a national average.

Which is better?

Neither, and that is the point. They were built for different roads. The question is only which road your dinner came down.

The metaphor is the book's. It carries no measurement behind it, and I am not going to pretend otherwise.

Is the gluten fundamentally different? The question worth asking

Now the question everybody actually came here with. We have just walked through a century of intensive genealogical selection, deliberate hybridisation, induced mutagenesis by X-ray, and, for other crops, though never commercially for wheat, transgenesis. All of it aimed at yield, stem strength, and dough performance. None of it aimed at how the grain sits with a person.

So it seems fair to ask: after all that effort, is it possible that some of what people struggle with today is downstream of the redesign?

I cannot answer that, and I want to be honest that nobody has. But I can lay out what is known, and let you weigh it.

Einkorn never received the D genome

Go back to the chromosome diagram. Einkorn is diploid, 14 chromosomes, the A genome alone. It sat out both prehistoric crosses. It never received the B genome from that Aegilops speltoides-like grass, and it never received the D genome from Aegilops tauschii. Bread wheat has all three. Durum has two.

That matters because gluten is not one substance. It is a family of storage proteins, and different genomes contribute different ones. A grain missing a whole genome is not a milder version of wheat: it is carrying a different set of proteins.

What the laboratory found, and what it did not

Einkorn is the strongest case in the research reviewed here. Across most genotypes tested, in-vitro work found null-to-minimal gluten toxicity, and reported that human digestive enzymes appeared able to break einkorn's gluten down fully, which they did not manage with modern wheat's.

Read that sentence twice, because both halves matter. That is a genuinely striking result, and it is in vitro: a test tube, digestive enzymes in a dish, not a person eating a bowl of pasta. Three things keep it from being a promise:

• It is not uniform. Specific einkorn genotypes, Monlis, ID358, ID1636, showed meaningfully higher toxicity in the same work.
• Nobody has run the human trial. There is no clinical outcome data here at all.
• And a finding from the other direction: Bozzini, a co-creator of Creso, tested the Casaccia collection of ancient and improved varieties in the late 1970s and found the proteins involved in celiac disease present, in highly variable amounts, in all of them. Ancient ones included.

Spelt is a useful warning about wanting an answer too badly. Two biochemical studies concluded its gluten is essentially as reactive as modern wheat's, 98.5% α-gliadin identity. The case made for it rests largely on a single patient, who showed early biopsy-confirmed signs of relapse before spelt was withdrawn. "Ancient" is not a synonym for "gentle."

What I can tell you, which is only my own experience

I am not a doctor, and this is not a study. It is one woman's kitchen. I ate the einkorn we sell for years, and I baked with the flour, and it sat easily with me in a way that other flour did not. My bloating went away. That is my past tense, my body, and my anecdote: it is not a result, and it is not a prediction about yours.

Customers write to me about Italy constantly, and I have never once been able to give them a clean explanation. There are too many things different at once: the variety, the milling, the drying, the fermentation of the bread, how much they walked, how little ultra-processed food they ate around it, and how relaxed they were. Any of those could carry it. Any combination could.

What I would say is this. If you have gone around this loop for years and nobody has given you an answer, trying a different grain is a cheap experiment and a reasonable one. Buy a bag, eat it for two weeks, pay attention. You will learn more from that than from me. If you want to run it, that is what our stone-milled einkorn flour is for: one ingredient, one grain, nothing added back. Talk to your doctor first if you have a diagnosis, and read your labels either way.

So should you choose ancient grains? An honest answer

What is actually in each one, measured

Here is the part I can prove, because I paid for it. I sent three of my own pastas to Medallion Labs, an accredited laboratory in Minneapolis, and asked for the whole panel rather than the flattering parts. All three are stone milled, bronze die cut, slow dried, and non-enriched. What separates them is how much of the grain was kept.

The durum busiate is a semola: the stones grind it, then it is sifted, which is what makes it pale and light on the plate. That is a texture decision, and it is the pasta most people mean when they say pasta. The Tumminìa and the Khorasan are milled whole, so the bran and the germ stay in the bag rather than being sifted out.

What is actually in each one

Percent of the Daily Value in a 55 gram serving, measured at Medallion Labs

All three are stone milled, bronze die cut, slow dried, and non-enriched. They differ in how much of the grain was kept, and that is what these numbers show.

Durum Busiate

A semola: sifted, so it eats lighter and pale. Per serving: fibre 11%, zinc 10%, thiamin 10%, selenium 10%, magnesium 6%, phosphorus 6%, iron 4%, folate 2%. Sodium free and low fat. Its real receipt is the one you cannot see: folic acid below the detection limit, tested twice, a year apart.

Tumminìa Tagliatelle

Milled whole, so the bran and germ stay in. Excellent source of manganese (70%) and zinc (25%). Good source of thiamin 20%, dietary fibre 18%, phosphorus 15%, magnesium 15%, selenium 15%, vitamin B6 15%, and iron 10%. Potassium 6%.

Khorasan Orzo

The Perciasacchi landrace, also milled whole. Excellent source of manganese (25%), zinc (20%) and selenium (20%). Good source of thiamin 15%, phosphorus 15%, magnesium 15%, dietary fibre 14%, and vitamin B6 10%. Iron 8%, potassium 6%.

Medallion Labs, ISO/IEC 17025 accredited, reports 82597.4 and 99654.1. Figures are the laboratory's own Nutrition Facts calculation for a 2 oz (56 g) serving. Under FDA's rules 10 to 19 percent of the Daily Value earns good source and 20 percent or more earns excellent source. One lot of each, tested once; crops vary by harvest. The zinc and selenium results for the Khorasan sit outside the laboratory's accreditation scope, so a confirmatory in-scope run is worth having before either goes on a package.

Two things worth pulling out of that. The first is that these are all my own products, tested at the same laboratory, by the same methods, at the same serving size. There is no competitor in this comparison and nothing to argue about; it is just three bags on one bench.

The second is what drives the spread. Ash is the standard measure of how much bran and germ survived milling, and my own numbers say it plainly: the busiate comes in at 0.90 percent on a dry basis, the Tumminìa at 1.95 percent. Better than twice as much of the grain is still there. That, and not the variety, is where the manganese and the magnesium and the fibre come from.

What steel rollers remove, and stones do not

A wheat kernel is three parts. The endosperm is starch and protein, the big white middle. The bran is the outer coat, where most of the fibre and a good share of the minerals sit. The germ is the embryo, small, oily, and where much of the B-vitamin content lives.

Industrial roller milling exists to separate those three and keep the endosperm. There is a commercial reason for it: the germ carries oil, oil goes rancid, and flour without a germ keeps for months on a warehouse shelf. What you get is white flour that lasts, and a bran and germ stream sold on as animal feed.

Stone milling grinds the whole kernel between stones and does not sieve the bran and germ back out. That is the entire mechanism. The nutrition is not added by the stone; it is simply never taken away. That is why the two columns in the table above differ, and it is the same reason enrichment exists at all: you only need to put vitamins back into a flour you took them out of.

Folate and folic acid are not the same word

This is the distinction I most want you to leave with, because it is on almost every box of pasta in America and almost nobody reads it.

Folate is the form that occurs naturally in food: in greens, in pulses, and in the germ of a wheat kernel. Folic acid is the synthetic form, manufactured and added back during enrichment. US law requires it in enriched macaroni products at 198 to 265 micrograms per 100 grams.

So I had the laboratory look for folic acid specifically, by LC/MS, which sees that exact molecule rather than folate in general. My durum came back under 10 micrograms per 100 grams, the floor the method can read. Roughly twenty times less than an enriched box is required to contain. I ran it twice, a year apart, on two different harvests, and got the same answer both times.

The folate that is in there is what the grain grew with: 13 micrograms per 100 grams in the durum, and 42.7 in the Tumminìa, which is more than three times as much, for the same reason as everything else on this page. Nobody added it back, because nobody took it out.

If you want to see that on a label rather than read about it, it is on the back of every box of our non-enriched Sicilian pasta.

What is genuinely different: flavor, agronomy, provenance

Real differences exist, and they are the mundane kind. Khorasan kernels are physically bigger, 55 to 75 grams per thousand seeds against 35 to 55 for durum. Einkorn, emmer, and spelt are hulled: their glumes stay welded to the kernel, needing a decortication step durum does not.

Flavor is the one thing I will not hand you a page number for. I can tell you what I taste: an older durum tastes of something, faintly sweet, a little like hay, and I can tell you that is my mouth, not a measurement. My mom would put it less kindly. Taste them yourself and trust your own tongue over my adjectives.

What is not: they are not gluten-free

Einkorn, emmer, spelt, khorasan, durum, all of them are wheat, and all contain gluten. Total protein in a wheat kernel runs 8–20%, and storage proteins make up 75–85% of it. If you have celiac disease, none of these grains is for you. That is a conversation for your doctor.

How to read an "ancient grain" label

This is the part I want you to keep. Read the label with me.

• Look for the variety name, not the adjective. "Ancient grain" means nothing; "Senatore Cappelli," "einkorn," "khorasan," "Timilìa" are checkable.
• Watch the plurals. Saragolle is a set of old landraces; Saragolla is a cultivar registered in 2004.
• Ask where it was grown and where it was milled. Much packaging answers only the second.
• Check for enrichment. Enrichment means vitamins were added to the flour after milling. Whether a pasta is enriched is stated on the label, and it tells you something about how the grain was handled.
• Read an IG value as Gluten Index, what the flour does in a dough, nothing more.

That is the standard I hold my own shelf to, and it starts with the name on my own box. Our rigatoni is sold as ancient grain because that is the shelf word: the answers that matter sit on the back of the pack: organic whole wheat, stone-ground, bronze-die, slow-dried, non-enriched, made in Italy. The adjective is not doing the work there; the list is. There is more on what ancient grain pasta actually is if you want it.

Why enrichment exists, and what I did about it on my own box

Everything above explains where enrichment came from. Roller milling strips the bran and the germ, because that is where the oil is and the oil is what goes rancid. Take those out and you take out most of the B vitamins with them. So in the United States the law puts five of them back, iron, thiamin, riboflavin, niacin, and synthetic folic acid, at levels written into the standard of identity for enriched macaroni. That is not a scandal. It is a public-health measure with a real reason behind it, and I am not going to pretend otherwise.

What I wanted to know was whether my own durum had any in it. So I paid a laboratory to look for folic acid specifically: the synthetic form, by LC/MS, not a general vitamin screen. It came back under 10 micrograms per 100 grams, which is the floor the method can see. American enriched pasta is required to carry roughly twenty times that. I had it run twice, a year apart, on two different harvests. Both times, the same answer.

That is a negative result, and negatives are unglamorous. But it is the only kind of proof I can hand you that is not an adjective: a named lab, a named method, a legal threshold to measure it against, and a number that showed up twice. The folate that is in there, thirteen micrograms, is what the grain grew with. Nobody added it back, because nobody took it out.

See also why gluten feels different in Italy, enriched vs non-enriched pasta, and does pasta have folic acid.

How ours is made

Three process choices, and what each one is actually for

Stone-ground

The grain goes between stones rather than steel rollers. Slower, cooler, and it keeps the bran and germ in rather than sieving them out and adding vitamins back.

Bronze die

Bronze scores the surface as the dough is pushed through. That roughness is why sauce clings instead of sliding off: a texture choice, made at the die.

Slow dried

Dried low and long instead of fast and hot. It costs days of production time, which is most of the reason industrial pasta does not do it.

Organic durum and heirloom Sicilian grain, milled and made in Sicily. Non-enriched: nothing added back, because nothing was taken out.

Frequently asked questions

Is modern wheat genetically modified?

No. According to a 2016 academic analysis, no transgenic wheat variety has ever been legally commercialized anywhere. The wheat you eat was made by genealogical selection, by crossbreeding, or, in a few pedigrees such as Creso's, by induced mutagenesis, which rearranges the plant's own DNA rather than inserting another organism's.

What is the difference between ancient wheat and modern wheat?

The useful line is landrace versus cultivar: a mixed farmer's population against one fixed genotype. Modern varieties are also far shorter, higher-yielding, and stronger in dough, and far fewer of them are grown. Chromosome number is identical between old and new.

Was modern wheat crossed with goat grass?

Yes, twice, but spontaneously, and long before anyone was breeding. An Aegilops speltoides-like wild grass, probably now extinct, contributed a genome roughly 300,000–500,000 years ago, and Aegilops tauschii contributed another about 8,000–9,000 years ago. The twentieth-century dwarfing donors were Japanese wheat.

How many chromosomes does ancient wheat have?

Einkorn has 14, emmer, durum, and khorasan have 28, and spelt and bread wheat have 42, all multiples of the seven base chromosomes labeled A, B, and D. That count has not changed through modern breeding, because both jumps happened in prehistory.

Is Senatore Cappelli an ancient wheat?

It is sold as one. It was actually bred in 1915 by Nazareno Strampelli, by genealogical selection from the North African landrace Jenah Rhetifah, at the Stazione Fitotecnica in Foggia. That makes it a heritage cultivar with a named breeder and a birth year, not an ancient species.

What is Creso wheat?

An Italian durum first grown in 1972. Bozzini and Mosconi at Casaccia crossed a Borlaug hybrid, Yactana 54 with Norin 10, then crossed to Senatore Cappelli, with Castelfusano B144, an X-ray mutant of Senatore Cappelli. Cappelli therefore appears on both sides of the pedigree.

Does modern wheat have more gluten than old wheat?

Not in the way the claim implies, and this record cannot settle it either way. What was clearly measured is gluten strength and composition, how the dough behaves and which proteins build it, not the quantity of gluten in the grain. No measurement of total gluten content appears here for any variety, old or new.

Why is modern wheat so much shorter?

Deliberate, and for a practical reason. Nitrogen fertilizer pushes yield up, a heavier head makes a tall stem fall over, and short stiff stems do not lodge. Strampelli used Akakomugi in the crosses behind Ardito, released in 1920; Borlaug used Norin 10 from 1952: both Japanese wheat varieties.

What this means for the pasta on your plate

Modern wheat was redesigned, and the people who did it signed their names to it: shorter so it would stand under fertilizer, earlier so it would be off the field before the summer storms, stronger in the dough so it would survive an industrial line.

What was not done: nobody added a chromosome, nobody inserted a foreign gene into the wheat in your cupboard, and nobody here measured what any of it does inside a person. The largest documented casualty is the number of wheats left.

So I do not shop by the word "ancient." I shop by the variety name, the valley, the mill, and what is not on the ingredient list. If you have read one of the tampering claims and want the page number that sits behind it, write to me, I will send it.

I'm not a doctor. This is my food and my story. Talk to your doctor, and read your labels.

Disclaimer: These statements have not been evaluated by the Food and Drug Administration. This product is not intended to diagnose, treat, cure, or prevent any disease. Individual results may vary. Please consult a healthcare professional before making changes to your diet or wellness routine. These findings come from independent research and do not guarantee the same results with our product.

References

1. Marco Pallotti, Grani Antichi: the primary source for the breeding history, chromosome families, pedigrees, and dough-strength figures above. Its subtitle advances an argument about gluten and health that this article does not make and does not repeat; it is used here for its documented history of plant breeding only. Several of its dates and plant heights are the author's own narrative with no citation attached, and are attributed as such throughout.
2. Motzo, R., et al. (2003), Facoltà di Agraria, Università degli Studi di Sassari, published analysis of the shares of Italian durum production held by individual varieties, 1981–1990. Institutional agronomic data, and the strongest quantitative set used here.
3. Lorenzetti, R. (2012), archival research for MIBAC / Archivio di Stato di Rieti: the Battaglia del Grano and the 1927 cultivated areas by variety. With Lorenzetti (2000) and Giorgi (2014), both archival and historical accounts of Strampelli's breeding program.
4. Borojevic, K. (2005), "The Transfer and History of Reduced Height Genes (Rht) in Wheat from Japan to Europe," Journal of Heredity 96(4):455–9: the Grani della Vittoria pedigrees. Not to be confused with Borojevic, S. (1990), a regional-journal review that is not cited in this article.
5. Bozzini, A. (2013), Rivista di Agraria.org: the origin of Creso, the 1959 X-ray mutants, and the late-1970s Casaccia investigation of grain proteins. A popular-science outlet, not a peer-reviewed journal, written by a co-creator of Creso.
6. Pogna, N., and Gazza, L. (2013), "Cresce il consumo di prodotti a basso indice di glutine," Report cereali, in Agricoltura, September 2013: the modern-group W and Gluten Index values and the registration-class figures. A non-peer-reviewed trade-magazine report. Senatore Cappelli's W 95 / IG 11 baseline is not from this report; it is Pallotti's own statement.
7. Pogna, N., et al. (2007), monograph on wheat genetics and prolamins, allopolyploidy, prolamin genetics, and the disulfide-bond breeding objective. A monograph, not peer-reviewed.
8. Salamini, F. (1999), "Il frumento monococco e le origini dell'Agricoltura," Le Scienze n°373, a popular-science magazine article; and Petrini, A., et al. (2013), CERMIS, einkorn domestication and the naming of the wild progenitors.
9. Porfiri, O. (2014), published work on Italian landrace populations, le Saragolle as landrace populations, and the 2004 registration of the Saragolla cultivar.
10. Zhu, M., Schmitz, A., and Schmitz, T. G. (2016), "Why Has not Genetically Modified Wheat Been Commercialized: A Game Theoretical Perspective," annual meeting of the Southern Agricultural Economics Association: the analysis reporting that no transgenic wheat has been legally commercialized. A conference paper in agricultural economics, not empirical field data.
11. Directive 2001/18/EC of the European Parliament and of the Council, Annex I B; Court of Justice of the European Union, Case C-528/16: the European regulatory position on conventional mutagenesis.
12. Roselli (2015), master's thesis, Scienze e Tecnologie Alimentari, Università di Padova: the sole basis, in the book, for any link between dough strength and digestion; an unpublished thesis using one in-vitro method.
13. de Lorgeril, M., and Salen, P. (2014), a narrative review, and Davis, W. (2011), Wheat Belly, a mass-market diet book: the two sources behind the "never safety-tested" claim. Neither is a regulatory or toxicological testing record. None of these last three is a study of people, and this article adopts no conclusion from any of them.

Disclaimer: the content in this blog is for informational purposes only and should not be taken as professional, medical advice.