Kodo millet (Paspalum scrobiculatum; varagu in Tamil, harka in Kannada, kodo in Hindi) began as a weed in India's early rice fields and was domesticated here roughly 2,500 years ago, in the Iron Age. Raw kodo millet gives 8.9 g protein, 6.4 g fibre and about 332 kcal per 100 g (IFCT 2017).
- The Charaka Samhita (Sūtrasthāna 27.16) calls koradūṣa light, astringent-sweet, cooling and drying.
- A 2021 meta-analysis places kodo millet in the intermediate glycaemic index band (55–69), below milled rice.
- Kodo's old reputation for "poisoning" comes from mould on rain-damaged grain, not the grain itself. Clean, dry grain is the rule.
Updated 6 November 2026 · Data: IFCT 2017 (ICMR-NIN), ICMR-NIN RDA 2020, Charaka Samhita, ICAR-IIMR and peer-reviewed archaeobotany
In this article
- What is kodo millet? Names across India
- Kodo millet history: from rice-field weed to crop
- Kodo millet in Ayurveda and old texts
- Kodo poisoning: what the old fear was really about
- Kodo millet nutrition per 100 g
- Kodo millet benefits: what the evidence supports
- How to cook kodo millet (varagu rice)
- Frequently asked questions
Kodo millet is the quiet grain of India's rainfed uplands: a small, hard-husked millet that kept families fed when the rice failed. In Tamil Nadu it is varagu arisi, cooked into pongal; in Karnataka it is harka; across Madhya Pradesh and Chhattisgarh it is kodo, often grown alongside kutki on poor soils. This guide follows kodo from the weedy margins of ancient paddy fields to the classical texts, explains the old fear of "kodo poisoning", and sets out what the nutrition data can and cannot tell us.
What is kodo millet? Names across India
Kodo millet is the grain of Paspalum scrobiculatum, a tufted grass with two to four short, one-sided spikes of rounded grains. Each grain sits inside a tough, glossy husk that must be removed before cooking, which is why kodo has always been more work to process than rice or wheat. A 2024 review still lists "poor dehulling recovery" among the challenges holding it back [8].
Its names change every few hundred kilometres. The Indian Food Composition Tables list it under the Tamil name varagu and record many regional names [1]:
| Language | Name |
|---|---|
| Tamil | Varagu (varagu arisi) |
| Kannada | Harka (also arka) |
| Telugu | Arikelu |
| Hindi | Kodo, kodon (IFCT lists koden) |
| Malayalam | Koovaragu |
| Gujarati, Punjabi | Kodra |
| Bengali | Kodoadhan |
| Odia | Kodus |
| Sanskrit | Kodrava, koradūṣa |
Source: IFCT 2017, ICMR-NIN, food code A017 and regional name index [1]; Sanskrit names from the Charaka Samhita translation [3] and Wisdom Library [11]. Spellings vary by region.
One warning for Tamil readers: varagu is kodo, but panivaragu is proso millet and kelvaragu is ragi. Three different grains share one root word, so read the pack, or the botanical name, before you cook.
Kodo millet history: from rice-field weed to crop
Most cereals were tamed from wild grasses gathered on purpose. Kodo millet took a side door. In a 2025 review of plant domestication, archaeobotanist Dorian Fuller and colleagues call it the strongest Indian example of a segetal crop: a weed of cultivated fields that people eventually chose to sow on its own [4].
The evidence is in the soil. At Neolithic and Chalcolithic sites of the Ganges plains, dated to roughly 4,000–3,000 years ago, kodo grains turn up only rarely, as single finds, plausibly as weeds of early rice fields. Then, at Iron Age sites in south India (perhaps 2,700–2,300 years ago) and Early Historic sites in both south and north India, kodo appears in larger quantities and in many samples, at places such as Veerapuram and Hulaskhera. In the same window the grains get measurably bigger: before about 2,500 years ago they are mostly under 1.5 mm wide, and by the centuries before 2,000 years ago they average above 1.6 mm [4].
Fuller's team estimates this domestication episode lasted about 500 years, perhaps less, and that kodo changed five or six times faster than wheat, barley or rice did. They suggest why. Wild Paspalum is a common weed of rainfed rice. In years when the rice crop was poor or failed, people would have eaten the weedy grain from the paddy, until some farmers began re-sowing it as a more reliable crop in its own right [4].
That story fits kodo's later character. A 2024 review describes it as a grain "revered for its ability to thrive in times of famine" and a companion crop to rice, grown by marginal farmers in rainfed zones for both grain and fodder [8]. Today it is grown mainly in Madhya Pradesh, Uttar Pradesh, Maharashtra, Chhattisgarh, Tamil Nadu and Karnataka, largely in tribal regions [5].
Kodo millet in Ayurveda and old texts
The Charaka Samhita describes grains in Sūtrasthāna 27 (Annapānavidhi Adhyāya), its chapter on food and drink. Kodo appears twice [3].
Koradūṣa (kodo) and śyāmaka are astringent and sweet in taste and light to digest. They increase vāta and calm kapha and pitta; they are cooling, binding and drying. (Charaka Samhita, Sūtrasthāna 27.16; Gata's summary of the Sanskrit and the Charaka Samhita Online translation, not a direct translation.)
A few verses earlier, 27.14 lists uddālaka, which the translators identify as vanakodrava or wild kodo, among grains "slightly different in quality" from the prized ṣaṣṭika rice [3]. So the text knew both a wild and a cultivated kodo, which sits neatly beside the archaeological picture of a weed becoming a crop. One small caution: the online translation glosses koradūṣa as "sanwa millet", a name more often used for barnyard millet, while giving the botanical name Paspalum scrobiculatum. Old plant names do not always map cleanly onto modern ones.
What do those qualities mean in a kitchen? "Light" and "drying" (laghu, rūkṣa, śoṣaṇa) describe a grain that does not sit heavy, and Ayurvedic cooks traditionally balance such grains with ghee, buttermilk or a moist dish such as pongal. These are classical descriptions, not medical prescriptions.
Kodo shows up in other old writing too. Hemacandra's 12th-century Jain Yogaśāstra lists kodrava among seventeen kinds of grain, and the 17th-century cookery text Bhojanakutūhala names kodrava among foods that can cause indigestion when eaten in excess, as summarised by Wisdom Library from published scholarship [11]. That second note may be an early echo of the problem in the next section.
Kodo poisoning: what the old fear was really about
Ask an older farmer in central India about kodo and you may hear about matona kodo or malona: "intoxicating" kodo that left people giddy, vomiting and trembling. Reports going back to the 19th century describe these episodes in Uttar Pradesh and Madhya Pradesh; symptoms included vomiting, giddiness and tremors and usually lasted one to three days. Crucially, they were tied to kodo whose ripening and harvest coincided with rain [5].
Science found the cause in 1985. Researchers isolated cyclopiazonic acid, a mycotoxin, from a sample of kodo that had caused "kodua poisoning", and showed that the moulds on the grain, Aspergillus flavus and A. tamarii, produced it [6]. A later rat study found that cyclopiazonic acid raised liver-damage markers [7]. A 2022 review by the ICAR-Indian Institute of Millets Research concluded that the poisoning was due to fungal contamination, and that no study has shown kodo grain producing a toxin of its own [5].
The issue is not only historical. In October 2024, ten wild elephants from one herd died in Bandhavgarh Tiger Reserve, Madhya Pradesh. A report from the Indian Veterinary Research Institute found cyclopiazonic acid in their viscera, and forest officials linked the deaths to large amounts of fungus-infected kodo crop that the elephants had eaten [10].
Traditional farmers had their own safeguards. One 1870 account, quoted in the ICAR review, records that rain-damaged kodo was never eaten as flour, only as whole grain, and was washed well before cooking [5]. Modern advice says the same thing in technical language: harvest on time, dry the grain quickly to 10–13% moisture, and store it dry and airtight [5]. At home, that means buying cleaned, dry, well-sorted grain from a trusted source, washing it before cooking, and throwing away anything that smells musty or looks discoloured.
Kodo millet nutrition per 100 g
IFCT 2017 analysed kodo millet (as varagu, code A017) from five regions of India [1]. Here it is beside the grains it most often replaces.
| Per 100 g | Kodo (varagu) | Little millet (samai) | Ragi | Jowar | Rice, milled | Wheat atta |
|---|---|---|---|---|---|---|
| Energy | 332 kcal | 346 kcal | 321 kcal | 334 kcal | 356 kcal | 320 kcal |
| Protein | 8.9 g | 10.1 g | 7.2 g | 10.0 g | 7.9 g | 10.6 g |
| Total dietary fibre | 6.4 g | 7.7 g | 11.2 g | 10.2 g | 2.8 g | 11.4 g |
| Fat | 2.6 g | 3.9 g | 1.9 g | 1.7 g | 0.5 g | 1.5 g |
| Carbohydrate (available) | 66.2 g | 65.6 g | 66.8 g | 67.7 g | 78.2 g | 64.2 g |
| Iron | 2.3 mg | 1.3 mg | 4.6 mg | 4.0 mg | 0.7 mg | 4.1 mg |
| Calcium | 15 mg | 16 mg | 364 mg | 28 mg | 7 mg | 31 mg |
| Magnesium | 122 mg | 91 mg | 146 mg | 133 mg | 19 mg | 125 mg |
| Zinc | 1.7 mg | 1.8 mg | 2.5 mg | 2.0 mg | 1.2 mg | 2.9 mg |
| Thiamine (B1) | 0.29 mg | 0.26 mg | 0.37 mg | 0.35 mg | 0.05 mg | 0.42 mg |
| Folate | 39 µg | 36 µg | 35 µg | 39 µg | 9 µg | 29 µg |
Source: IFCT 2017, ICMR-NIN [1], food codes A017 varagu (Paspalum scrobiculatum), A016 samai, A010 ragi, A005 jowar, A015 rice raw milled, A019 wheat flour atta. Energy converted from kJ (1 kcal = 4.184 kJ) and rounded. IFCT values are for the cleaned, dehusked grain as analysed; brands and harvests differ.
Three things stand out. Kodo has more than twice the fibre and about four times the magnesium of milled white rice, with a little more protein. It is not a calcium food: at 15 mg per 100 g it is closer to rice than to ragi. And it is rich in phytate, at 452 mg per 100 g in IFCT's analysis, more than ragi (306 mg) or milled rice (266 mg) [1]. Phytate can reduce how much iron and zinc the body absorbs, which is one reason traditional washing, soaking and fermenting are worth keeping.
Kodo millet in everyday portions
Most people cook 50–75 g of raw millet per meal, which swells to roughly one to one-and-a-half cups cooked.
| Raw kodo millet | Energy | Protein | Fibre | Iron | % of daily protein (man / woman) |
|---|---|---|---|---|---|
| 30 g (a small bowl of upma or porridge) | 100 kcal | 2.7 g | 1.9 g | 0.7 mg | 5% / 6% |
| 50 g (one plate of millet rice) | 166 kcal | 4.5 g | 3.2 g | 1.2 mg | 8% / 10% |
| 75 g (a generous lunch portion) | 249 kcal | 6.7 g | 4.8 g | 1.8 mg | 12% / 15% |
Calculated by Gata from IFCT 2017 code A017 [1]; raw weight, before cooking, without dal, oil or ghee. Daily protein from ICMR-NIN 2020: 0.83 g/kg/day, about 54 g for a 65 kg man and 46 g for a 55 kg woman [2].
Kodo millet benefits: what the evidence supports
Blood sugar. A 2021 systematic review and meta-analysis of millets and diabetes found an average millet glycaemic index (GI) of 52.7, against 71.7 for milled rice and 74.2 for refined wheat. Kodo millet fell into the intermediate group (GI 55–69), alongside pearl millet, finger millet, little millet and sorghum [9]. Swapping some white rice for kodo may support steadier post-meal blood sugar, but GI depends on how the grain is cooked and what it is eaten with. If you manage diabetes, talk to your doctor or dietitian before changing your diet.
Fibre and minerals. Compared with milled rice, kodo brings more fibre, magnesium, iron, thiamine and folate per 100 g (Table 2). That is the plain, well-measured benefit of eating a whole, unpolished grain.
Antioxidants. The ICAR review cites earlier work in which kodo had the highest phenol content among the small millets studied, at 368 mg catechol equivalents per 100 g of flour [5]. These are laboratory measurements; they do not show a health effect in people.
Gluten-free. Like all true millets, kodo contains no gluten, which makes it useful for people avoiding wheat.
How to cook kodo millet (varagu rice)
- Wash well: rinse three or four times, rubbing the grains between your palms, until the water runs clear. Pick out any discoloured grains.
- Soak: 30 minutes at minimum, or a few hours if you have time. Soaking shortens cooking and is part of the traditional habit of reducing phytate.
- Cook like rice: 1 cup kodo to about 2.5–3 cups water. Simmer covered for 15–20 minutes, or 2–3 whistles in a pressure cooker, then rest 10 minutes and fluff.
- Traditional dishes: varagu pongal with moong dal, pepper and ghee; kodo khichdi; harka bisi bele bath; lemon or curd "rice".
- Ferment it: kodo works in idli and dosa batter in place of part of the rice. In one lab study, lactic acid fermentation of dehusked kodo raised its protein and B-vitamin content and lowered tannins [12].
Store kodo in an airtight jar in a cool, dry place, away from steam and damp. Dryness is not a nicety for this grain; it is the whole point.
From the Dhatu pantry
- Organic Kodo Millet (Varagu Rice / Harka): whole, unpolished, for pongal, khichdi and millet rice
- Organic Mixed Millets: kodo with foxtail, little, barnyard and proso millet in one pack
- Little Millet (Samai / Saame): kodo's softer, quicker-cooking cousin
- Organic Barnyard Millet (Oodalu / Sanwa): the other small millet of rainfed India
- How foxtail millet (navane) sustained Indian civilisations
- Barnyard millet names, nutrition per 100 g and fasting use
- Protein per 100 g in Indian dals, millets and seeds
Frequently asked questions
What is kodo millet called in Tamil, Kannada, Telugu and Hindi?
Kodo millet (Paspalum scrobiculatum) is varagu in Tamil, harka or arka in Kannada, arikelu in Telugu and kodo or kodon in Hindi. Do not confuse varagu with panivaragu (proso millet) or kelvaragu (ragi).
How much protein is in 100 g of kodo millet?
Raw kodo millet has 8.92 g protein per 100 g, according to the Indian Food Composition Tables 2017 (food code A017). A 50 g raw serving gives about 4.5 g protein.
Is kodo millet good for diabetes?
A 2021 meta-analysis grouped kodo millet with the millets that have an intermediate glycaemic index (55–69), lower than milled rice. It can be part of a balanced diet, but it is not a treatment; people with diabetes should ask their doctor or dietitian how to include it.
Is kodo millet safe to eat?
Clean, well-dried kodo millet has been eaten in India for over two thousand years. The problem called kodo poisoning comes from mould on grain that got wet near harvest, not from the grain itself, so buy cleaned, dry grain and discard any that looks or smells mouldy.
What is kodo poisoning?
Kodo poisoning is an illness caused by eating kodo grain contaminated with moulds such as Aspergillus that produce cyclopiazonic acid. Old reports from Uttar Pradesh and Madhya Pradesh link it to rain at harvest time.
Is kodo millet gluten-free?
Yes, kodo millet is naturally gluten-free. People with coeliac disease should still choose packs processed away from wheat, barley and rye.
How do you cook kodo millet?
Rinse well, soak for 30 minutes to a few hours, then cook about 1 cup of millet with 2.5 to 3 cups of water until soft, like rice. It also works in pongal, khichdi, upma and idli or dosa batter.
Is kodo millet mentioned in Ayurveda?
Yes. The Charaka Samhita (Sūtrasthāna 27.16) groups koradūṣa, identified by translators as Paspalum scrobiculatum, with shyamaka millet as astringent and sweet, light, cooling and drying, increasing vāta while reducing kapha and pitta.
Kodo was never the grain anyone planned. It came uninvited into the paddy, and in the lean years, when the rice failed, it was still standing. Our ancestors had the sense to notice, and then the patience to make it a crop. They also learned its one weakness: rain at harvest. Their answer was not to abandon the grain but to wash it, keep it dry and treat it with respect. Perhaps that is kodo's lesson for us. Resilience often arrives looking like a weed, and the old precautions that go with a food are part of the food. Keep both, and the famine grain can feed us well again.
This article is for general information and is not medical or dietary advice. If you have diabetes, coeliac disease, kidney disease or another medical condition, speak to a doctor or registered dietitian before changing your diet.
References
- Longvah T, Ananthan R, Bhaskarachary K, Venkaiah K. Indian Food Composition Tables 2017. National Institute of Nutrition (ICMR), Hyderabad. Food codes A005, A010, A015, A016, A017, A019; Tables 1, 2, 5 and 11. nin.res.in (PDF)
- ICMR-National Institute of Nutrition. A Brief Note on Nutrient Requirements for Indians, the Recommended Dietary Allowances (RDA) and the Estimated Average Requirements (EAR), 2020. nin.res.in (PDF)
- Charaka Samhita, Sūtrasthāna 27 (Annapānavidhi Adhyāya), verses 14 and 16. Sanskrit text and English translation, Charaka Samhita Online. carakasamhitaonline.com
- Fuller DQ, Denham T, McClatchie M, Wu X. Commensal domestication pathways amongst plants: exploring segetal and ruderal crop origins. Philos Trans R Soc Lond B Biol Sci. 2025;380(1926):20240190. PubMed 40370025 · doi:10.1098/rstb.2024.0190
- Deepika C, Hariprasanna K, Das IK, Jacob J, Ronanki S, Ratnavathi CV, Bellundagi A, Sooganna D, Tonapi VA. 'Kodo poisoning': cause, science and management. J Food Sci Technol. 2022;59(7):2517–2526. PubMed 35734115 · doi:10.1007/s13197-021-05141-1
- Lalitha Rao B, Husain A. Presence of cyclopiazonic acid in kodo millet (Paspalum scrobiculatum) causing 'kodua poisoning' in man and its production by associated fungi. Mycopathologia. 1985;89(3):177–180. PubMed 3921845 · doi:10.1007/BF00447028
- Antony M, Shukla Y, Janardhanan KK. Potential risk of acute hepatotoxicity of kodo poisoning due to exposure to cyclopiazonic acid. J Ethnopharmacol. 2003;87(2-3):211–214. PubMed 12860310 · doi:10.1016/s0378-8741(03)00146-6
- Jeeva G, Suhasini B, Pramitha L, Jency JP, Joshi P, Ravikesavan R, Elango D. Unlocking the potential of Kodo millet: reviving an indigenous super grain for tomorrow's nutrition. Planta. 2024;259(6):140. PubMed 38691193 · doi:10.1007/s00425-024-04414-1
- Anitha S, Kane-Potaka J, Tsusaka TW, Botha R, Rajendran A, Givens DI, et al. A systematic review and meta-analysis of the potential of millets for managing and reducing the risk of developing diabetes mellitus. Front Nutr. 2021;8:687428. PubMed 34395493 · doi:10.3389/fnut.2021.687428
- Shrivastava S. Fungus, not kodo millet, responsible for Bandhavgarh elephant deaths. Mongabay India, 27 November 2024. india.mongabay.com
- Wisdom Library. Kodrava: definitions, including Bhojanakutūhala (via Shodhganga thesis "Dietetics and culinary art in ancient and medieval India") and Hemacandra's Yogaśāstra 3.95 (via R. Williams, Jaina Yoga). Accessed 6 November 2026. wisdomlib.org
- Mohapatra D, Nickhil C, Kar A, Sharma Y, Deshpande SS, Tripathi MK, Haromuchadi SR. Impact of LAB fermentation on the nutrient content, amino acid profile, and estimated glycemic index of sorghum, pearl millet, and kodo millet. Front Biosci (Elite Ed). 2024;16(2):18. PubMed 38939908 · doi:10.31083/j.fbe1602018
