Millet farming in Uganda means finger millet on a small plot, almost always from seed a farmer saved, and almost always with no fertilizer at all. Seed moves through home stores, neighbours and open markets rather than through stockists. The crop is Eleusine coracana, grown across the north, east and west for porridge, bread and local brew, and the area under it has been falling for two decades.
Millet Farming in Uganda Is Shrinking, and the Official Numbers Disagree
Before any advice, a warning about numbers. Uganda has two official answers for how much millet it grows and they are about five times apart on area.
| Measure | FAOSTAT, Uganda rows | UBOS, via a NaSARRI paper |
|---|---|---|
| Area | About 90,000 hectares | 437,000 hectares |
| Yield | 1,322 to 1,430 kg a hectare | 0.6 tonnes a hectare on farm |
| Production | About 119,000 tonnes | Not stated |
Part of the answer is in FAOSTAT's own data flags, which nobody reports. Read them and the Ugandan millet series turns out to be mostly derived rather than measured: the yield figure is flagged as an FAO estimate for every year after 2015, and the area is flagged as imputed for most of them, while only production carries the official flag. Dividing an official production figure by an imputed area is exactly how you get a yield that sits flat near 1.4 tonnes for eight years while the crop shrinks around it.
What both sources agree on is the direction. FAOSTAT's own series has Ugandan millet area falling from 175,173 hectares to about 90,000, and production from 236,484 tonnes to about 119,000, with a trough near 70,000 tonnes in between. Millet is retreating in Uganda, and it is retreating for reasons that show up later on this page: no fertilizer, saved seed of unknown health, a disease that can take most of the crop, and a harvest that is slow and hard work to process.
For planning, use the UBOS on farm figure of about 0.6 tonnes a hectare as the honest starting point for an unimproved field, which is roughly 240 kilogrammes an acre. Treat FAOSTAT's 1.4 as what you might see with reasonable management. The same UBOS source puts on station yield at 3.5 tonnes a hectare, which is what the crop is capable of and not what anybody should budget.
Which Millet Uganda Actually Grows
Almost all of it is finger millet, Eleusine coracana, the small round grain in a fist of curved fingers at the top of the stem. Names get muddled in Uganda, and the muddle reaches into official material: NARO's own community seed bank page lists the crops held at its Iganga bank and writes finger millet against the scientific name Panicum miliaceum, which is proso millet, a different species entirely.
That matters at the counter, because NARO's variety catalogue now holds two sets of millet releases with very different numbers on them.
NAROMIL 1 and NAROMIL 2 are filed under finger millet, were released in 2017 from NaSARRI, mature in 100 and 102 days, and carry yield bands of 3 to 4.1 and 2.93 to 4.07 tonnes a hectare. Both are recorded as resistant to leaf blast and moderately resistant to neck and finger blast, and NAROMIL 2 is also recorded as drought tolerant. Both are medium tall, erect and non lodging, which matters on a crop that goes flat in wind once the heads fill.
NAROFOMILL 1, 2 and 3 were released in 2025 from the same institute, mature in 75 to 90 days, and carry yield bands of 1.7 to 1.8 tonnes a hectare. NARO files them under "others" rather than under finger millet, and the record does not say which millet species they are. Their yield ceiling is about half NAROMIL's and their maturity is two to three weeks shorter, which is what you would expect of a different species rather than of a newer finger millet. So do not assume those figures apply to finger millet, and ask NaSARRI what NAROFOMILL is before buying it on the strength of a name. A 2025 release is also unlikely to be in general circulation yet.
Where Finger Millet Grows Best in Uganda
Finger millet is grown in the northern districts around Lira, Apac, Gulu, Pader and Kitgum, across the east from Mbale and Tororo through Kumi, Soroti, Amuria and Kaberamaido, and in the west from Masindi and Hoima down to Mbarara, Ntungamo and Kabale. The northern and eastern regions together carry about 80 percent of national production.
The conditions it wants are modest and that is the point of the crop. Growing season rainfall of 750 to 1,200 millimetres, an average temperature around 23 degrees, altitude from 1,000 to 2,000 metres, and well drained sandy or sandy loam soils. It tolerates a pH range of about 5 to 7 and copes with alkaline soils, and it has been recorded withstanding temperatures up to 44 degrees, which is beyond where maize stops.
Nationally about 60 percent of the crop goes in during the second season and 40 percent in the first, though the first season carries more weight in the north and east. Uganda is also a primary centre of diversity for the crop, which is why the landraces described further down are worth taking seriously rather than dismissing.
Preparing Land and Sowing Millet in Rows
Finger millet seed is tiny, so the seedbed decides the stand. Ugandan practice in the trial work is ploughing twice with an ox plough before planting, then dibbling seed into shallow furrows opened with a stick and covering it lightly. Sow with the onset of the rains rather than waiting, and there is a disease reason for that timing which the blast section below explains.
Broadcasting is the common method and it is the one that costs yield, for three separate reasons. It makes weeding a guessing game, because you cannot tell crop from weed in a mat of seedlings. It produces uneven plant density, so some patches are starved and others empty. And it creates the dense stands that hold dew on the leaves, which is what blast needs to establish.
Row sowing fixes all three at the cost of more time at planting. That is the trade the ICRISAT blast manual asks a finger millet grower to make, and it is the single change with the most knock on benefit on this crop.
Millet Spacing and Plant Population Per Acre
The documented spacings for finger millet are 40 by 10 centimetres or 30 by 15 centimetres, chosen to leave open space between rows so air moves and dew does not sit on the leaves. The Ugandan trial at Kuju thinned its crop at 20 days after sowing to 30 centimetres between rows by 10 between plants.
Work out what that means per acre and the scale of the crop becomes clear. At 30 by 10 centimetres each plant occupies 0.03 square metres, and an acre is 4,047 square metres, so the stand is about 134,900 plants an acre. That is my own arithmetic from the spacing rather than a published figure, and it is worth doing because it explains why seed rate advice for this crop is so vague.
No Ugandan seed rate per acre for finger millet has been published that could be verified for this page. The practical method in the Ugandan trial work is to sow thicker than the target and thin to the spacing at about three weeks, which is when the first weeding happens anyway. That converts the seed rate question into a thinning job you can see, which is more reliable than a kilogramme figure on a crop whose seed is this small. The plant population calculator handles other spacings.
The Fertilizer Finding: a Micro Dose Matched a Full Dose
This is the most useful measurement in Ugandan millet research and almost nobody repeats it. NaSARRI and Makerere ran a split plot trial at Kuju in Amuria district across two seasons on SEREMI II, testing fertilizer rate against weeding timing.
| Treatment | Grain yield, kg a hectare | About, kg an acre |
|---|---|---|
| Micro dose | 2,182 | 883 |
| Full dose | 2,152 | 871 |
| No fertilizer | 950 | 384 |
The micro dose was 16.6 kg of nitrogen and 10.6 kg of phosphate a hectare. The full dose was 83 kg of nitrogen and about 52 kg of phosphate. So five times the fertilizer bought 30 kilogrammes a hectare more grain, which is inside the noise of a field trial. Meanwhile going from nothing to the micro dose more than doubled the yield.
In per acre terms the micro dose is about 6.7 kg of nitrogen and 4.3 kg of phosphate an acre, converted at 0.4047. That is a small fraction of a bag, which is exactly why it is worth knowing: the reason most Ugandan millet gets no fertilizer at all is that a full dose looks unaffordable, and the trial says the full dose was never the thing that paid.
Price the quantities above against a current bag rather than against a recommendation. The fertilizer price page carries the bag side, and a soil test is what tells you whether nitrogen or phosphate is the limiting one on your ground.
Weeding Millet Once, and When
The same trial tested weeding timing, and the result is counterintuitive enough to state carefully. Weeding once at 20 days after sowing gave the highest grain yield, followed by double weeding at 20 and 45 days. It also gave the best growth vigour score, at 3.77, and brought the crop to flowering in the fewest days. The best combination in the whole trial was the micro dose with a single weeding at 20 days, at 2,292 kilogrammes a hectare, about 928 an acre. The worst was no fertilizer with weeding delayed to 45 days, at 907 kilogrammes a hectare, about 367 an acre.
Read that against the standard recommendation and there is a real disagreement to resolve. The ICRISAT blast manual says weed finger millet two to three times a season to keep the field weed free, and it has a disease reason: weeds host the blast pathogen and a thick weed mat raises humidity around the crop.
Both are right about different things. The trial varied the timing of a single weeding against a double and judged the result on grain yield. The manual is aiming at a weed free field and judging on disease. What both support is that the first pass, by about three weeks after sowing, is the pass that decides the yield, and that weeding at six weeks is largely too late to recover what competition has already taken. Nothing here says stop weeding. It says do not let the first weeding slip while you wait to do one big job.
Blast Disease and Why a Heavy Nitrogen Bag Makes It Worse
Blast is the reason finger millet yields in this region sit where they do. It is caused by the fungus Magnaporthe grisea and it shows up three ways on the same plant: leaf blast as elongated lesions that taper at both ends, grey in the centre with a brown margin, neck blast where the collar below the head turns brown and dies, and head blast on the fingers themselves. The isolates causing all three are genetically similar, which is good news for breeding, because resistance found against one expression should hold against the others.
In East Africa, grain yield losses exceeding 80 percent have been reported in bad years, and grain quality suffers too because the heads fill poorly. The African average finger millet yield of about 750 kilogrammes a hectare is attributed largely to this disease plus poor husbandry and a shortage of improved seed.
Three conditions make it worse and all three are things a farmer sets.
Heavy nitrogen is the first, and this is where the fertilizer section above comes back. Excessive fertilizer produces soft sappy growth that blast gets into easily, so the manual discourages it outright and recommends organic manure combined with a micro dose at a quarter of the recommended rate as both cheaper and less disease prone. It also says to give nitrogen in small increments rather than one large dose, and to use animal manure instead of nitrogen fertilizer where enough is available. That is a pathologist and an agronomist arriving at the same recommendation from opposite directions, which is about as strong as agricultural advice gets.
Dense stands are the second. A high seeding rate holds dew on the leaves for longer, and blast needs wet leaf surfaces. That is the reason behind the row spacings given earlier, which exist to let air through.
Late sowing is the third. The sensitive window is between germination and tillering, so sowing with the onset of the rains lets the crop pass through that stage before airborne spores build up in the air. Sowing late puts the vulnerable stage into the high inoculum period.
Managing Millet Blast Without Buying Anything
Everything in the block above is free, and that is the argument the manual makes: cultural control is the best option for a finger millet grower because chemical control, while it works to a degree, is expensive relative to what this crop earns.
Rotation and intercropping are the two other free levers. The rotations the manual sets out are finger millet then groundnut then maize, finger millet then chickpea then wheat, or finger millet then cowpea or pigeonpea then sorghum, each chosen so the intervening crops do not host the pathogen. For intercropping it names cowpea, groundnut and pigeonpea, for the same reason.
Resistant varieties are the strongest single lever and Uganda has two. P 224, which farmers know as PESE 1, and SEREMI 2, also called U15, were both released in Uganda and both come out of the regional work on lines with low blast levels and workable agronomy, in a band of 1.5 to 3.0 tonnes a hectare. Several others in the same regional set, including U15 elsewhere, Gulu E and the KNE lines, were released in Tanzania, Kenya or Ethiopia rather than here.
On the chemical side, this page gives no product, no rate, no dilution and no seed treatment figure, because none is verified as registered for this crop in Uganda. What is published in the trial literature is the size of the effect: one systemic fungicide reduced leaf blast by 28 percent on young millet and head blast by 38 percent. Those are useful proportions and they tell you two things at once, that chemistry helps and that it does not solve the problem. Any rate has to come off the product label, the label has to be followed, and a district extension officer or plant clinic is the right place to confirm it for this crop.
Millet Varieties Released in Uganda and Why Farmers Refused Some
Uganda has released more than twelve improved finger millet varieties since breeding started in 1965. Two are widely grown. Most are not, and the reasons are the most instructive part of the whole record.
The SEREMI 1 and SEREMI 3 story is the one worth carrying away. A variety that out yields what a farmer already grows and then cannot be brewed or baked properly does not get planted twice, because in Ugandan households millet is brew and bread before it is tonnage. Any variety decision on this crop has to pass the kitchen, and the kitchen is stricter than the field.
Landrace Millet and Where Ugandan Farmers Get Seed
Across 170 millet growing households in Bushenyi, Lira and Nwoya, 96.9 percent planted landraces and 3.1 percent planted improved varieties. Seed came from home stores for 70.1 percent, the open market for 22.9 percent, and fellow farmers for 4.9 percent. Nearly four in five households put about one acre under sole millet.
Those landraces have names, districts and known behaviour, and farmers rank them precisely. Ajuko Manyige was grown by 56.4 percent of the Nwoya sample, and is described as maturing in about three months, high yielding, easy to grind, giving tasty bread, and able to be harvested before it fully matures, which is a hunger season trait rather than a yield trait. Kal Atar was grown by 35 percent in Lira, with moderate yield, early maturity and good marketability. Kagume was grown by 30 percent in Bushenyi, high yielding and good for bread, but it wants fertile soils and birds like it. Others named include Okello Chiba, Okama, Todyang, Agun Kibati, Mahega, Mbanjura, Kabaragara, Kabumburi, Kabukunguru and Nkodere. One, Kashema, was rejected by farmers in the west, and another, Kisansha, is preferred because its soft stem can be harvested by hand instead of with a knife, which is less work.
That last detail is a reminder that harvest labour is a real selection criterion on this crop, and it does not appear in any yield table.
What Ugandan Millet Farmers Actually Want From a Variety
The same survey ranked 29 attributes. Early maturity came first in every district, named by 87 percent in Bushenyi, 93.9 percent in Lira and 90.9 percent in Nwoya. Then high yield, drought tolerance, taste, big heads and market demand.
Two things at the bottom of that ranking are more interesting than anything at the top. Resistance to striga came last of all 29 attributes, named by 2.9 percent, 2 percent and none across the three districts. And good brewing characteristics, fast fermentation, heavy brew and small grain size were also among the least preferred. So the brewing traits that a breeder might assume matter most on a brew crop are not what farmers put at the top when asked to rank, even while the same farmers refused SEREMI 1 and 3 for brewing quality. Read together, the honest reading is that brewing quality is a pass or fail gate rather than something farmers optimise for, which is a different thing from a priority.
Men and women wanted much the same crop. The only difference large enough to matter statistically was high yield, named by 72.1 percent of women and 85.5 percent of men. Men also put more weight on good cooking quality, at 60.9 percent against 39.4.
Harvesting, Drying and Threshing Millet Grain
Harvest when the fingers have gone brown and the grain is hard, cutting the heads rather than the whole plant where the variety allows it. The landrace with the soft stem that can be broken by hand is preferred for a reason, and a variety that needs a knife on every plant costs a day or more an acre.
Dry the heads before threshing, on a raised surface or a mat rather than on bare ground, which adds grit and stones that cannot be winnowed out and lets moisture back in from below. Then thresh, which on this grain is usually beating the heads to release the seed, and winnow.
Two disease rules from earlier apply at this stage and are easy to forget. Do not thresh or winnow millet from diseased plants in the growing field, because that broadcasts spores across the whole area. And keep the straw from a diseased crop out of the field, either ploughed deep or burnt.
Millet grain stores well because it is small, hard and low in oil, which is part of why it has held on as a famine reserve crop. It still needs to be dry going in, and a store that lets moisture back in undoes the drying. The grain moisture guide and the guide to preventing weevils in stored grain cover the store itself.
Millet Yield Per Acre in Uganda: Reading the Numbers Honestly
Pull the figures on this page together and four bands emerge, each answering a different question.
About 240 kilogrammes an acre is the UBOS on farm figure for the crop as Ugandan farmers actually grow it, which means landrace seed, no fertilizer, and weeding when there is time. About 384 kilogrammes an acre is what the unfertilised control plot produced in a managed trial, which is the same thing done carefully. About 880 to 930 kilogrammes an acre is what the micro dose plus early weeding produced in that trial. And 1.2 to 1.6 tonnes an acre is what NAROMIL 1 and 2 are documented to carry at the top of their range, which converts from 3 to 4.1 tonnes a hectare.
The gap between the first and the last of those is a factor of five or six, and none of it is mysterious. It is seed health, plant arrangement, the timing of one weeding, and a few kilogrammes of fertilizer placed rather than broadcast. That is the whole argument of this page in one sentence.
Selling Millet and the Market Behind the Crop
Millet in Uganda is eaten as porridge, mingled into bread, and brewed. It is also increasingly milled and packaged, and bottled millet drinks are on sale in Ugandan shops, which puts a processor at the end of the chain for the first time in most growers' experience.
How much of a millet harvest gets sold depends strongly on where you are. The share sold is highest in the central and western regions, at about 39.6 and 20.3 percent, while the northern and eastern regions, which grow most of the crop, keep more for consumption and store it to sell later in the year. So a grower in the north is typically running a food security crop with a cash tail, and a grower in the west is closer to running a market crop.
No verified current Ugandan millet price could be sourced for this page, and that absence is worth stating plainly rather than filling with a guess: millet does not appear in the national monthly market price series that covers maize, beans, onions and the other staples on this site. What that means in practice is that a millet grower has less price information than a maize grower does, so the price you are offered is harder to check. Ask more than one buyer, and ask whether you are being quoted for brewing grain, for milling, or for eating, because those are different qualities.
Storage is the other half of the answer. In the northern and eastern districts, holding grain to sell later in the year is already normal practice, and a crop that stores like millet is a good candidate for it. That is loss avoidance and timing rather than speculation.
Questions Ugandan Millet Growers Ask
How much fertilizer should I put on millet? Far less than you would think, and it should be placed rather than spread. A Ugandan trial found a micro dose of 16.6 kg of nitrogen and 10.6 kg of phosphate a hectare, about 6.7 and 4.3 kilogrammes an acre, produced 2,182 kg a hectare against 2,152 for a full dose five times the size and 950 with nothing. In field terms that is a three finger pinch or a bottle cap per sowing hole. Heavy nitrogen also raises blast risk, so the small dose is the better agronomy and not merely the cheaper option. Price the quantity against a current bag on the fertilizer price page.
How many times should I weed millet? The timing matters more than the count. A Ugandan trial found one weeding at 20 days after sowing gave the highest yield, ahead of double weeding at 20 and 45 days, and far ahead of a single weeding at 45 days. The standard regional recommendation is two to three weedings to keep the field clean, partly because weeds host blast. Reconcile the two this way: get the first pass done by about three weeks whatever else happens, then weed again if the field needs it.
What is the white or grey spotting on my millet leaves and heads? Almost certainly blast, which appears on leaves as elongated lesions that taper at both ends, grey in the centre with a brown margin, as browning and death of the collar just below the head, and on the fingers themselves. It is the most destructive disease of the crop and East African losses above 80 percent have been reported in bad years. Nothing you spray now will undo a head blast infection, so the work is in next season's seed selection, spacing, sowing date and nitrogen rate.
Which millet variety should I plant in Uganda? Of the released lines, PESE 1 and SEREMI 2 are the two that Ugandan farmers actually grow, both from the regional work on low blast lines. NAROMIL 1 and NAROMIL 2, released in 2017, carry the highest published yield bands at 3 to 4.1 and 2.93 to 4.07 tonnes a hectare and are recorded as resistant to leaf blast. Be careful with NAROFOMILL 1, 2 and 3: NARO does not file them under finger millet and does not state the species, so ask NaSARRI first.
Can I keep planting my own millet seed? Yes. Finger millet is self pollinated, so saved seed stays true to type and there is no vigour penalty, which is why 70 percent of surveyed Ugandan growers use home saved seed and 97 percent grow landraces. What saving accumulates is disease rather than genetic drift, so select seed heads in the field before harvest, reject anything with blast on it, and store the seed lot separately. Buy in fresh seed after a bad blast year.
What does millet sell for in Uganda? No verified current figure exists for this page, and the reason matters: millet is not in the national monthly market price series that covers maize, beans and onions, so there is no published reference price to check an offer against. Prices move with season, region, and whether the buyer wants brewing grain, milling grain or eating grain. Ask several buyers, ask which quality you are being quoted for, and see the price pages for the staples that do have published series.
Is millet better than maize on a poor field? On a dry, sandy, low fertility field, yes, and that is what the crop is for. It handles rainfall down to about 750 millimetres in the season, tolerates a pH range of 5 to 7 and alkaline soils, and has been recorded withstanding temperatures up to 44 degrees. What it does not do is beat maize on tonnage or on price information, and it costs more labour to harvest and process. On a field that will grow maize reliably, maize usually wins.
What should I rotate millet with? Crops that do not host the blast pathogen. The documented rotations are millet then groundnut then maize, millet then chickpea then wheat, or millet then cowpea or pigeonpea then sorghum. For intercropping, cowpea, groundnut and pigeonpea are the named options. Rotating millet with millet is how blast is kept in a field. The sorghum guide and the cassava guide cover the two other dryland staples that commonly share these fields.
Does striga attack millet in Uganda? It is present and it is named as a constraint in the research literature, but Ugandan millet farmers themselves rank resistance to it dead last of 29 variety attributes, which tells you how they weigh it against blast, drought and maturity. Treat striga as a field level problem to manage through rotation and fertility rather than as a variety choice, and take blast far more seriously.
Finding Millet Seed and Checking Current Prices in Uganda
Two changes carry most of the gain on this crop and neither needs much money. Sow in rows and thin to spacing at about three weeks, doing the first weeding at the same time. And place a pinch of fertilizer rather than spreading a bag or spreading nothing, delaying it 30 to 40 days after sowing if your early rains are heavy.
For seed, ask NaSARRI or a district extension officer which released lines are actually available this season rather than relying on a name, and compare what stockists have against what your neighbours grow, because a landrace that performs near you is a serious option and not a fallback. This guide sits under the crop growing guides hub, and the organic fertilizer guide covers the manure half of the micro dosing recommendation.