🌿 The Two Virus Diseases That Decide Your Crop

Cassava mosaic disease, usually shortened to CMD, is caused by a group of viruses called cassava mosaic geminiviruses. It is the oldest and most widespread virus problem in African cassava, and understanding it properly starts with one fact that decides everything else: it reaches a new field by two separate roads, and the road most farmers worry about is the less important one for them personally.

The first road is the whitefly. The vector is Bemisia tabaci, a small sap-feeding insect that picks the virus up from an infected plant and carries it to a healthy one. This is the road that spreads the disease across a landscape, from field to field and district to district. You do not control it. It flies.

The second road is the stem cutting, and this is the road you own completely. Cassava is propagated from stem cuttings, not from true seed. When you cut a stake from an infected mother plant, the virus is already inside the stem tissue, and every plant that grows from that stake is infected from the day it sprouts. It does not need a whitefly at all. It was born sick.

Stop and sit with that, because it changes how you think about the whole crop. In maize you buy seed and the seed is clean. In cassava you take cuttings from last year's field, or from a neighbour, or from a roadside market, and you are moving live virus with your own hands. The reference material for this course states it plainly: because there is no mature certified-seed market for cassava, the fastest way mosaic and brown streak spread across a region is through farmers sharing infected cuttings. Your cutting knife is a vector.

Now the symptoms, so you can actually find it. CMD shows in the leaves. Look for a mottled or mosaic pattern of pale yellow-green and normal green on the same leaf, leaves that are distorted, twisted, puckered or reduced in size, and plants that are visibly stunted compared with their neighbours in the same row. Severely affected plants stay small all season. The whole plant carries the virus, so the pattern is on new leaves as they open, not just on old ones.

What does it cost? Here the numbers need care, because the published figures do not all measure the same thing, and averaging them together produces nonsense.

• A 1997 estimate attributed yield reductions of 50 to 70 percent to CMD in affected crops.
• A 2016 dataset found that CMD-infected plants produced a mean of only 23.6 percent marketable roots. That means roughly three-quarters of the roots that came off infected plants could not be sold.

Those are two different measurements. The first is total yield lost. The second is the share of roots that were good enough to sell. A plant can produce a reasonable weight of roots and still fail the market test if the roots are small, thin and fibrous. Do not add these figures, do not average them, and do not let anyone quote one to you as if it were the other. Ask which one they mean.

At regional scale, combined losses from mosaic and brown streak across Kenya, Tanzania, Uganda and Malawi have been estimated at up to US$750 million a year, with Tanzania's field losses alone estimated at US$51 million. Treat these as modelled economic estimates from one evidence review, not as audited accounts. They tell you the problem is large. They do not tell you what your own field is losing.

That last number is the one that matters and only you can get it. Walk your rows. Count the plants with mosaic symptoms. Divide by the total plants counted. Write the figure and the date. Do it again next season. A rising incidence means your planting material source is infected, or whitefly pressure has increased, or both, and either way you need to act before the next planting, not after it.

One more thing to fix in your mind now, because Lesson 4 builds on it. There is no spray you can buy that cures a virus in a cassava plant. There is no treatment, no injection, no rescue. Once a plant has CMD it has it for life, and every cutting from it carries it. The whole of virus control in cassava is therefore about what you plant and what you remove, never about what you apply.

Cassava brown streak disease, CBSD, is caused by two related viruses, CBSV and UCBSV, and like mosaic it is carried by whiteflies and by infected stem cuttings. But it does something mosaic does not, and that difference is the whole reason it frightens people who grow cassava for money.

CBSD attacks the storage root itself. It causes a dry, corky, brown necrotic rot inside the root, and from the outside the root can look completely normal. The plant above ground can look reasonable too. You can walk a field, see a decent stand, plan your harvest, hire the labour, dig the crop, and only discover at that moment that the roots are unsaleable and often inedible. Mosaic tells you it is there by stunting the plant in front of you. Brown streak can hide until the day your money is on the line.

Some plants do show leaf and stem signs, and you should look for them: yellowish blotching along the leaf veins that does not follow the mottled pattern of mosaic, and brown streaking or lesions on the green stems. But the reliable check is inside the root, and that means cutting.

Here is the practical method, and it is the most important skill in this lesson. Before you commit to a harvest date, before you accept a buyer's order, test-dig. Take plants from several different parts of the field, not one corner. Include plants that look perfectly healthy above ground, because those are exactly the ones that can hide the damage. Cut each root across in two or three places and look at the cut face. Healthy flesh is uniformly white or cream. Brown streak shows as brown or blackish corky patches and streaks running through the flesh. Count how many of the roots you cut are affected and how badly. That count, not the appearance of the canopy, is what tells you what you are about to harvest.

The disease is not new. It was first identified in Tanzania in 1936. What has changed is its spread and intensity since the 2000s, moving into new areas of East and Central Africa.

Now the loss figures, and pay attention to how wide they are, because the width is itself the lesson.

• In severe cases, CBSD causes up to 100 percent loss in susceptible varieties.
• Zambian district-level survey data from 2019 to 2021 found average losses of 1.4 tonnes per hectare, ranging from 0.7 t/ha in Kabompo to 1.9 t/ha in Chienge across three districts.
• Zambian farmer-reported losses from 2016 to 2018 averaged 55 percent, ranging from 2 percent to 99 percent across farms.
• Zambia's 2018 survey found 79 percent of sampled fields had some CBSD presence, at an average incidence of 32.2 percent within affected fields.

Read the farmer-reported range again: 2 percent to 99 percent. That is not sloppy data. That is a real description of how CBSD behaves. Two neighbouring farms, same district, same season, can have entirely different outcomes depending on variety, on where their cuttings came from, and on local whitefly pressure. Anyone who quotes you a single average CBSD loss figure and calls it your expected loss is misleading you. The honest planning statement is that in an affected area the loss could be almost nothing or almost everything, and the way you find out which is by cutting roots open.

Notice too that the 79 percent figure and the 32.2 percent figure measure different things. Seventy-nine percent of fields had the disease somewhere in them. Within the fields that had it, on average about a third of plants were affected. A field being positive does not mean the field is lost.

The hardest fact in this lesson is about varieties. As of the 2022 evidence review this course draws on, there were no CBSD-resistant cultivars available for farmer use in Zambia, with breeding by the national institute ZARI and by IITA described as still underway. That is a statement about one country at one date, and it may have changed. But it should stop you from ever accepting a seller's claim that a variety is brown-streak resistant without checking the current national list yourself. Resistance claims go stale as new virus isolates appear.

Altitude will not save you either. The whitefly vector is reported as capable of surviving above 1,000 metres, with a potential range covering over 23 million square kilometres in one cited assessment. Highland farmers have sometimes assumed they are out of reach. On this evidence, they are not.

You cannot spray a virus. You cannot release a predator against a virus. You cannot feed a plant out of a virus. When the disease is CMD or CBSD, the whole of your control comes down to two actions: plant clean material, and remove sick plants. This lesson is about the first, which is the one with the biggest effect.

Start with why clean material is scarce in the first place, because it explains a lot of frustration. Cassava has no true seed system. Almost all planting is from stem cuttings, and the seed-systems literature states bluntly that cassava planting materials in themselves have no economic value in most farming systems. They are bulky, they perish, they dry out, and nobody has historically been able to build a business selling them the way maize seed is sold. So there is no certified-seed trade to keep the material clean, and the material circulates farmer to farmer, carrying whatever it carries.

That also means clean material, when it exists, is genuinely valuable and genuinely worth paying for. A farmer who understands this is ahead of most of their neighbours.

Now, positive selection. This is the single most useful field technique in this module and it costs nothing but attention and timing.

The method: during the growing season, while the plants still have their leaves and you can see symptoms, walk the field and physically mark the plants that look best. Use a stick, a ribbon, a painted stake, anything you can find again. What you are marking is a plant with no mosaic mottling, no leaf distortion, no vein yellowing, no stem lesions, good vigour and good height compared with its neighbours in the same soil. Mark generously the first year and you will still find you have fewer good plants than you expected.

The timing is the part people get wrong. If you wait until harvest to choose your cuttings, the leaves are gone or senescing and you cannot see the symptoms any more. You then choose by stem thickness, which tells you nothing about virus. Selection must happen while symptoms are visible. Marking in season and cutting at harvest is the whole trick.

When you cut, take from the marked plants only, and check the stems as you go: no brown streaking on the green stem, no borer holes, no mealybug or scale, no soft or discoloured tissue. And when you dig the marked plants, cut their roots open too. A plant that looked perfect above ground and shows brown streak inside is not a mother plant, it is a warning.

Now the arithmetic of multiplication, and it is sobering. A single mature healthy cassava plant typically produces only about ten usable 20 cm commercial stakes. Working that forward, one hectare of mother plants in one year produces only enough stakes to plant roughly ten new hectares, a multiplication ratio of about one to ten per year. An IAEA farmer-facing guide gives a more conservative version of the same idea: one 25 cm stick of an improved variety planted this year yields at least a minimum of eight sticks next year.

Use the conservative figure for planning. If you want to plant one hectare of clean material next year at about 10,000 plants per hectare, you need roughly 10,000 good stakes, which means roughly 1,000 to 1,250 well-grown mother plants this year, which is roughly a tenth of a hectare set aside and looked after properly. That is a real, small, doable plot. Do that arithmetic for your own target area before you plant anything.

Rapid multiplication techniques do exist and they change the ratio completely. Shoot-induction techniques are reported to yield roughly 800 shoots per year from a single mother plant, and leaf-and-bud mini-cutting techniques are reported to produce up to 60,000 stakes from a single mother plant within about 18 months. Be clear-eyed about what those are: they are specialist, capital-intensive and skill-intensive techniques. They are the answer to how a clean-seed enterprise scales fast, not something you do in a field with a machete. If your ambition is to become a planting-material supplier, that is a real business and those techniques are the route, but it starts with training and a facility, not with a bigger field.

One last point, and it links this module to the whole course. A clean-seed multiplication scheme is a disease-control intervention, not just an input-supply convenience. That is the logic behind every national seed-system and positive-selection extension programme you will hear about. If one operates near you, joining it is worth more than anything you can buy in a shop.

Rogueing means finding a diseased plant, pulling it out, and destroying it. It is the second half of virus control and it is the part most farmers skip, because pulling out a living plant feels like throwing away food. This lesson is about doing it properly and understanding honestly what it buys you.

What rogueing does. Every infected plant in your field is a reservoir. Whiteflies feed on it and carry the virus onward to healthy plants around it. Remove the plant and you remove a source. Rogueing lowers the amount of virus available in your field for the vector to pick up, which slows the rate at which the rest of your crop becomes infected.

What rogueing does not do. It does not cure anything. It does not protect a plant already infected but not yet showing symptoms. And it will not save a field where infection is already general, because you would be removing most of the crop. Rogueing is a tool for keeping a mostly-clean field clean, not for rescuing a mostly-sick one.

How to do it. Walk the field on a fixed schedule, not when you happen to remember. Early and often is the rule, because a plant removed at six weeks has infected far fewer neighbours than the same plant removed at six months. Pull the whole plant, including as much root as you can. Do not leave it lying in the field or at the field edge, because it stays alive and it stays a source. Burn it, or bury it deep, or feed it to livestock away from the field, but get it out. Above all, never cut planting material from a rogued plant, and never let anyone else take cuttings from your field edge heap.

Record what you pull. Count and date it, exactly as you counted incidence in Lesson 1. Over three seasons those counts tell you whether your control is working.

Now tool and traffic sanitation, which matters more for the bacterial disease you will meet in the next module but is good practice throughout. Cassava bacterial blight spreads by rain splash, wind splash, contaminated tools, foot traffic and contaminated cuttings. Clean your cutting knife or machete between plants when you are cutting planting material, and clean it between fields always. Avoid walking through a wet crop, because moving through wet foliage moves pathogens on your clothes and legs. Do not carry cuttings from a diseased field into a clean one, and be careful about accepting gift cuttings from a farm you have not seen.

Crop rotation belongs here too. For bacterial blight, a non-host interval of roughly six months is a reported control measure, meaning cassava is not grown on that ground for that period. Rotation does not help against a whitefly-borne virus arriving from a neighbour's field, but it breaks soil- and debris-carried problems, and it addresses the potassium depletion that continuous cassava causes. It is good practice for several reasons at once.

Now the hard decision: when is a field beyond saving? There is no published threshold to give you and this course will not invent one. What you can do is reason it out with your own count. If you walk your rows and find symptoms on a small minority of plants, rogueing is sensible and worth the labour, and your next planting can come from your own best plants. If you walk your rows and find symptoms on most plants, rogueing is not a strategy, it is a demolition, and the honest answer is that your planting material source was infected. In that case the decision is to harvest what you can, take no cuttings at all from that field, and obtain planting material from a different and verified source before you plant again. Write down where the bad material came from and never take from that source again.

One warning about what people will tell you. You will meet sellers offering sprays for mosaic or brown streak. Insecticides can reduce whitefly numbers, but they do not cure infected plants, they do not remove virus already in your cuttings, and heavy spraying carries its own costs and risks. Product choice, registration and safe-use rules are set by your national plant-protection or agrochemical regulator, and no product should be used on the basis of a seller's claim alone. Ask the regulator or the extension office what is registered for your crop in your country before you spend money.

The honest summary of this whole module so far: against these two viruses you have exactly three real tools, and they are clean planting material, rogueing, and choosing a variety from the current national list. Everything else is noise.

Somebody will try to sell you a variety. They will use the word resistant. This lesson is about how to handle that conversation without either dismissing improved varieties, which would be foolish, or believing a claim that has quietly expired, which is expensive.

Start with the honest structural facts. Breeding for cassava mosaic disease resistance has been underway in the region since at least the 1930s and 1940s, making it one of the longest-running crop-disease breeding programmes in African agriculture. That is a real achievement and improved CMD-resistant material has changed cassava production across the continent. But this course cannot give you a dated release history tying named varieties to specific resistance genes and strains, because that record was not available to it. Do not accept a variety name from this course or any static document as a current recommendation.

Brown streak is a harder and newer story. As of the 2022 evidence review this course uses, there were no CBSD-resistant cultivars available for farmer use in Zambia, with ZARI and IITA breeding still underway. Resistance to brown streak is less complete and less widely available than resistance to mosaic. Anyone telling you otherwise for your country should be asked to name the release and the authority that lists it.

Now the central concept: resistance has a shelf life. A variety described as resistant is resistant to the virus strains that existed when the study was done. Both mosaic and brown streak are actively spreading and changing, with new CMD strains and new CBSV isolates appearing, and breeders are working against a moving target. A variety recommendation is therefore the start of a conversation with your extension office and your seed-certification scheme, never a permanent fact printed on a bag or a poster.

So where do you get a current answer? From your national root-crop research institute or its extension service. In Nigeria that is NRCRI, in Uganda NARO, in Zambia ZARI, in Ghana CSIR, with IITA working alongside national programmes across the region. Ask for the current recommended list for your agroecological zone, not for the country as a whole, because what performs in a humid forest margin is not what performs in a dry savanna. Ask which strains are locally active. Ask where certified or inspected planting material of those varieties can actually be obtained, because a recommendation you cannot buy is not a plan.

Disease resistance is only one axis of the decision, and treating it as the only one leads people to grow a healthy crop that nobody wants. Weigh four axes together.

1. End use. Fresh table consumption favours sweet, quick-cooking varieties. Industrial starch, flour and HQCF favour high-dry-matter, high-yielding varieties regardless of sweetness, because they will always be processed before anyone eats them.

2. Disease and pest pressure in your specific locality. Ask the extension office which mosaic and brown streak strains and which pests are locally active, and select against those.

3. Cycle length against your household's cash and food calendar. Some cultivars are ready in nine months, others need eighteen months or more, and root quality is often reported as optimal at twelve to fifteen months. A nine-month variety and an eighteen-month variety serve completely different needs on the same farm. Many farmers should grow both.

4. Cyanogenic potential relative to your intended processing route. A variety destined only for full gari-style fermentation, pressing and roasting can safely be a higher-cyanogen variety than one you intend to boil and eat within hours of harvest. Module 11 covers this in detail, and it is not a side issue.

One piece of sourced evidence is worth carrying into that fourth point now. A Ugandan study measured cyanogenic potential in six local and six improved varieties at thirteen months. Local varieties averaged 201.65 mg HCN-equivalent per kg; improved varieties averaged 108.75, roughly 46 percent lower. But every one of those twelve varieties exceeded the commonly cited 10 ppm reference value in the raw state, including the sweeter ones. Improved does not mean safe raw. It meant lower on average in that sample, in that country, at that age.

Finally, do your own comparison. Plant two or three varieties in strips in the same field, same soil, same day, same management. Count symptoms in each strip through the season. Cut roots open at harvest and compare. Weigh the yield of each strip separately. After two seasons you will know more about how those varieties behave on your land than any pamphlet can tell you, and you will have a record you can show a buyer or a lender.

Cassava is often described as the climate-resilient crop, the one that keeps producing when maize fails. That reputation is mostly deserved and it is why so much is invested in it. This lesson exists because there is a sourced projection that complicates the story, and a grower planning ten years ahead should hear it.

One 2026 modelling study, reported through secondary press coverage rather than the primary paper, projects that suitable cassava-growing habitat in Africa could increase by 56 to 60 percent by 2050, adding roughly 2.1 million square kilometres of newly highly suitable land. Current suitable area is put at 54.6 percent of Africa, about 16.2 million square kilometres. On the face of it that is good news for the crop.

The same study projects that cassava brown streak disease risk could expand to affect 55 to 57 percent of Africa's land area by 2050, up from a current estimate of 33.7 percent of cassava production areas already affected. Read those two projections together and the point emerges: the disease is projected to expand roughly in step with, or faster than, the crop's own suitable range. Cassava is not projected to outrun its worst disease by moving into new country. The disease is projected to follow.

Be careful with these numbers. This is a single study's climate-model-driven projection, reached through secondary press coverage in the material available to this course, and a full country-by-country cassava climate-vulnerability assessment was not available. Treat the direction as the safe teaching point and the specific percentages as one study's projection, to be updated when better modelling appears. The direction is: disease pressure rising, not falling.

The same source names East Africa, particularly Tanzania, Uganda and southeast DR Congo, as the current brown streak hotspot, and flags Cote d'Ivoire, Benin and Cameroon as new risk zones under the projection. If you farm in a country that has not had a brown streak problem, that is not a permanent condition, and the single most valuable thing you can do is learn to recognise it now, before it arrives, by cutting roots open as a routine habit.

The recommended interventions in that same source line up exactly with everything this module has taught: accelerate disease-resistant and heat-tolerant variety development, tighten national and international controls on the movement of planting material across borders, and prioritise replanting with clean material in current high-outbreak areas. Two of those three are about planting material. That is not a coincidence.

So what does a grower actually do with a rising-pressure forecast? Three things.

First, treat clean planting material as a permanent annual cost, not a one-time purchase. The farmer who buys clean material once and then multiplies from their own field forever will drift back toward infection as whiteflies reinfect the stock season by season. Plan to refresh from a clean source on a regular cycle, and keep a small, well-watched mother plot of your own between refreshes.

Second, do not stake everything on one variety. A single variety across the whole farm is a single point of failure when a new strain arrives. Spreading across two or three varieties, chosen with your extension office, costs you a little management complexity and buys you a great deal of protection.

Third, keep records that let you see a trend. A single season's count means very little. Three or four seasons of counts, kept in the same notebook in the same format, will show you whether pressure on your farm is rising or falling and whether what you are doing works.

Here is the minimum record set for disease management, and it should live in the same notebook as everything else:

• Field incidence count and date, at least twice each season, using the walk-and-count method from Lesson 1.
• Root-cutting check at test-dig and at harvest: how many roots cut, how many showed brown streaking.
• Every planting material source, by name of person or market, and the date, matched against which parts of the field they went into.
• Every plant rogued, counted and dated.
• Which varieties are in which part of the field, so a strip comparison is possible.

That is five lines a season for most growers. It is the cheapest insurance in this whole course, and unlike a spray it actually works against a virus.

One closing point that carries into the next module. The two problems in this module, mosaic and brown streak, are viruses, and viruses have no biological control agent you can release. In the next module you will meet two arthropod pests, the mealybug and the green mite, that do have such solutions and where the story is spectacularly better. That asymmetry is real and worth understanding: what works against an insect does not work against a virus, and knowing which problem you have decides which tools are even available to you.