🥛 Choosing Your Cow

Here is the whole of breed selection in one sentence. Indigenous zebu breeds survive; exotic taurine breeds produce. Every choice in this module is a negotiation between those two facts, and crossbreeding exists to find a workable point between them.

No breed or cross is best in the abstract. The right animal for you is the one that survives your heat, your feed and your disease pressure, and also produces enough milk to justify what it costs to feed her. Both halves of that sentence must be true at once. An animal that produces beautifully and dies is worth nothing. An animal that survives everything and gives two litres may not repay the labour of keeping her, depending on your market.

The reason this decision goes wrong so often is that milk yield is easy to measure and everything else is not. A table can show you litres per day. It cannot easily show you which cow walked to water in the dry season and which one lay down, which cow carried a tick burden without losing condition, which one calved unassisted, and which one was still in the herd in her sixth lactation. Those things are worth money, sometimes more money than the litres, and they belong on the same page as the yield figure.

What the indigenous animal actually gives you, and what does not show up in a yield table:

• Heat tolerance. Boran cattle carry numerous sweat glands and have been reported to survive on watering intervals of every three to four days in rangeland with 500 to 800 mm of annual rainfall. That is a genuinely adapted animal, not a backward one.
• Tick and disease resistance, which means fewer treatments, fewer losses and less money spent.
• The ability to walk to water and to graze poor forage and still function.
• Low mortality, which is the trait that quietly decides whether you still own an animal in three years.

A course that told you "zebu is a poor breed" would be teaching you badly. The correct statement is different and more useful: a zebu produces less milk per day, for reasons that also make her cheaper to keep and harder to kill.

Now the other side, honestly. Exotic dairy breeds were selected for generations under conditions of abundant feed, cool climates and low disease pressure. That selection worked: they give far more milk. But it also removed some of the hardiness the indigenous animal has, and it left an animal whose high production is only expressed when the feed is there. Put a high-yielding exotic cow on feed that supports three litres and she will give you roughly three litres, while eating more than a zebu would, losing condition faster, failing to come back into heat, and being more vulnerable to everything.

That is why the honest first question is not "which breed gives the most milk?" It is "what can I actually feed, in the worst month of the year?"

Notice the phrase "worst month". Almost everyone plans their feeding on the good season, when the grass is green and the cow looks well. The animal you buy has to survive the dry season too, and it is in the dry season that the difference between an adapted animal and an unadapted one becomes obvious and expensive.

There is one more caution before the evidence. Most of the breed performance figures in the next lessons were measured at research stations, agricultural colleges and well-run ranches, not in a zero-grazing unit behind a house. Those figures are real, and they are attributed to their study, but they were produced under management most smallholders cannot match. Read every yield figure in this module as "this breed, in that trial, under that feeding", never as "this is what she will give me".

With that understood, the next four lessons give you the best African evidence available: what indigenous cattle actually produce, how heat stress hits different crosses in measured litres, how much milk exotic blood really adds and how far to push it, and why a smaller exotic breed can beat a bigger one over a lifetime.

Start with the animals most smallholders already own or can afford, and look at what they actually do when somebody measures them properly.

The Boran, found in Kenya, southern Ethiopia and Somalia, is primarily a beef or dual-purpose breed rather than a specialised dairy animal. That matters, because the Boran is nonetheless the base stock onto which most Kenyan smallholder crossbreeding is built. Recorded figures for the Improved Boran strain in Kenya:

• Mature cow bodyweight 400 to 550 kg; mature bulls 550 to 850 kg.
• Age at first oestrus 15.6 months; age at first calving 32.2 to 39 months.
• Calving interval 13.8 months, which is about 420 days.

That calving interval deserves attention. It is close to the biological target band of 365 to 425 days, and it is dramatically shorter than the 670-day average measured across smallholder households in Siaya. The Boran figure comes from well-managed ranch and research-station conditions. The Siaya figure comes from ordinary farms. The difference between those two numbers is management, not breed, and it is the single most encouraging fact in this whole course: an indigenous animal, well kept, calves close to target.

Boran milk figures, and notice that they do not agree with each other, because they come from three different systems:

• 866 to 1,088 kg over four lactations, with 5.7 to 6.1 percent butterfat, in one historical dataset.
• Up to 1,657 kg in a 252-day lactation under favourable conditions at Naivasha.
• About 843 kg per year under pastoral management in Ethiopia.

Those are three separate measurements under three separate management systems, and this course will not average them into a single "Boran figure". Teach yourself to read them the same way. Ask what conditions produced each number before you use it.

Note that butterfat of 5.7 to 6.1 percent. If your buyer pays on volume alone it is worth nothing to you. If your buyer pays on fat and solids, it is worth real money, and the module on the milk business explains why you must find out which kind of buyer you have.

Now the East African zebu, the small shorthorn zebu that makes up the base population across much of the smallholder herd in Ethiopia and the wider region. The Barca zebu was measured in a trial at the Bako Agricultural Research Center in Ethiopia:

• Daily milk yield 2.98 kg per day.
• Lactation length 279 days.
• Total lactation yield 672 kg.

That 672 kg was the lowest of every breed group in that trial. There is no point pretending otherwise. If your only measure is litres, the East African zebu comes last.

But litres are not the only measure, and this is the point of the lesson. What the indigenous animal is worth to a smallholder is not shown by her milk yield alone. Heat tolerance, tick and disease resistance, the ability to walk long distances to water, the ability to convert poor forage into something, and low mortality are all real and valuable, and none of them appears in a litres-per-day table. A cow that gives three litres for eight years is a different business proposition from a cow that gives eight litres for two years and then dies, and the table only shows you the first column.

There is also a cost side that the yield table hides. The zebu is a smaller animal with lower maintenance requirements, so she needs less feed to stay alive before she produces anything at all. On a farm where feed is the binding constraint, and it usually is, that changes the arithmetic considerably.

So the honest position on indigenous cattle is this. They produce less milk per day, measurably and substantially. They also cost less to keep, tolerate more, and survive longer under pressure. Whether that trade is good for you depends on how much feed you can supply and how hard your environment is, which is exactly why the last lesson of this module makes you write your own conditions down before choosing anything.

Everyone says exotic cattle suffer in the heat. Very few people can tell you how much milk that costs, or at what temperature it starts. One Ugandan study can, and it is the best quantified African example of this tradeoff available.

The study used 53,730 daily milk-yield records from 183 cows: purebred Ankole, the long-horned cattle of the Great Lakes region, and three crosses, Ankole with Friesian, Ankole with Jersey, and Ankole with Sahiwal.

First, understand the measure. Heat stress in cattle is not about temperature alone. Humidity matters too, because a cow sheds heat by evaporation and humid air makes that harder. The combined measure is the temperature-humidity index, or THI. Every animal has a THI threshold: below it, yield is unaffected; above it, yield starts to fall. Two things therefore matter about any breed: how high her threshold is, and how fast she loses milk once she crosses it.

Here is what the study measured.

• Purebred Ankole: threshold at THImax 74 and THImean 66, and a loss of 0.055 kg per day for each THI unit above the threshold.
• Ankole crossed with Friesian: threshold at THImax 77 and THImean 69, and a loss of 0.107 kg per day per THI unit, the most heat-sensitive group in the study.
• Ankole crossed with Jersey: threshold at THImax 76 and THImean 69, and a loss of 0.067 kg per day per THI unit.
• Ankole crossed with Sahiwal: threshold at THImax 76 and THImean 68, and no statistically significant yield loss at all.

Read that carefully, because the pattern is not the simple one people expect.

The purebred Ankole has the lowest threshold in this dataset, meaning her yield starts responding at a lower THI. But once past the threshold she loses milk roughly half as fast as the Ankole-Friesian cross, 0.055 against 0.107 kg per THI unit. So the indigenous animal is not merely more comfortable, she is more stable: a hot day takes less milk off her.

The Ankole-Friesian is the most productive on an average day and pays the steepest penalty on a hot one. That is the tradeoff of this whole module, measured in kilograms.

The Ankole-Sahiwal cross is the most interesting result. Sahiwal is itself a zebu breed, so this is a zebu crossed with a zebu, and it showed no measurable heat penalty at all in this dataset. If your problem is heat rather than feed, crossing to another adapted breed rather than to a temperate one is a real, evidenced option that almost nobody discusses.

Now two honest cautions.

First, the authors themselves describe the overall THI effect in this herd as small relative to studies of high-producing temperate dairy breeds. In this particular herd and this particular climate, heat stress was a real but not dominant constraint. Do not walk away thinking heat is the main thing limiting your cow. In most smallholder systems, feed and reproduction are bigger.

Second, these thresholds and losses were measured in one Ugandan herd. Your climate is not that climate. Use the pattern, which is well established, rather than the exact numbers, which belong to that study.

What do you actually do with this? Two things.

One: when someone offers you a high-exotic-blood animal, ask what happens to her in your hot season, and understand that the answer is a real cost in litres, not a small inconvenience. The higher the exotic percentage, the more that answer matters.

Two: heat is one of the few constraints you can cheaply reduce with a building rather than an animal. A locally made insulating layer, such as a woven mat of local materials, placed under a metal roof, has been reported to lower internal shed temperature by roughly 10 degrees Celsius or more compared with bare corrugated iron. That is a low-cost intervention within reach of most smallholders, and it works on whatever animal you already own. Shade, ventilation with a roof height of at least about 3 metres, and unlimited clean water are the other three.

In other words: you can buy heat tolerance in an animal, or you can build some of it into the shed. The shed is usually cheaper.

If some exotic blood raises milk yield, more must raise it further. That is the assumption behind a great many bad purchases, and one Ethiopian trial tested it directly.

The trial, in the Bako area of Ethiopia, compared zebu cows with Friesian crosses at increasing levels of exotic inheritance, and with pure Holstein Friesian. Daily yields measured:

• Barca zebu: 2.98 kg per day, 279-day lactation, 672 kg total.
• Half Barca, half Friesian: 7.21 kg per day.
• Half Boran, half Friesian: 6.36 kg per day.
• Quarter Barca, three-quarters Friesian: 7.15 kg per day.
• Quarter Boran, three-quarters Friesian: 6.92 kg per day.
• One-eighth Barca, seven-eighths Friesian: 6.28 kg per day.
• One-eighth Boran, seven-eighths Friesian: 5.98 kg per day.
• Pure Holstein Friesian: 9.43 kg per day, 362-day lactation, 3,183 kg total.

Total lactation yields for the crosses in that trial ranged from 1,915 to 2,373 kg.

Now read the pattern down the list. Going from pure zebu at 2.98 kg to a half-bred at 7.21 kg more than doubles daily yield. That is a very large gain and it is why crossbreeding is promoted at all.

But then look at what happens as exotic blood keeps rising. The half Barca cross at 7.21 kg per day beat the seven-eighths Boran cross at 5.98 kg per day. The three-quarter crosses sat between them. The line does not keep climbing. It rises steeply from zebu to half-bred and then falls back as the exotic percentage goes higher.

Why? Because the higher-grade animal has less of the adaptation that lets her eat, walk, resist ticks and stay healthy under those conditions, and under that trial's management the loss of hardiness outweighed the extra genetic potential for milk. The pure Holstein at 9.43 kg did better than all the crosses in that trial, but she was managed at a research station. Between a research station and a smallholding, that advantage is exactly what tends to disappear.

The study's own recommendation, in its own words, was that cows with 50 to 75 percent Friesian inheritance would be recommended, provided that sound breeding programmes are in place to maintain the recommended blood level.

That second clause is the part almost everybody ignores, and it is where most smallholder crossbreeding programmes quietly fail.

Here is the problem. Suppose you buy a fine half-bred cow. If you serve her with a Friesian bull or Friesian semen, her calf is three-quarters exotic. Serve that daughter the same way and her calf is seven-eighths. Within three generations you have drifted out of the recommended zone and into the higher-grade zone that performed worse in the trial, without ever making a decision to do so. Alternatively, if you serve her with whatever bull is walking past, her calf's blood level is unknown and uncontrolled.

Maintaining a blood level is an active, repeated decision, generation after generation. It requires knowing what each animal actually is, recording every service and every sire, and choosing the sire deliberately to move the offspring toward the level you want. A single cross does not stay at 50 percent in her offspring. It never has.

This is one reason the record-keeping in Module 1 is not optional. If you do not record which bull or which semen served which cow, you cannot know what your heifers are, and you cannot manage a blood level you cannot measure.

Two further honest notes. First, the Ethiopian ATVET college trial of Holstein-Friesian crosses with Boran and Arsi cattle exists, but its specific milk-yield figures could not be retrieved for this course, so no numbers from it are given here. If you need them, that trial must be obtained directly. Second, no current breed-suitability map or national crossbreeding policy for any African country was available for this course. The correct authority for what cross suits your area is your national livestock research institute or dairy board, such as KALRO in Kenya or EIAR in Ethiopia, and the local extension service. Ask them, and ask what feeding their recommendation assumes.

The most complete quantified breed picture available for this course concerns Jersey cattle in Africa, drawn from literature published between 1964 and 2020 across 34 African countries. It is worth studying closely, because it shows exactly how a breed that loses on litres can win on money.

Jersey compared with Holstein-Friesian in that same African dataset:

• 305-day milk yield: Jersey 4,666 litres, Holstein-Friesian 6,147 litres.
• Age at first calving: Jersey 935 days, about 30.7 months; Holstein-Friesian 1,029 days, about 33.8 months.
• Calving interval: Jersey 401 days, Holstein-Friesian 468 days.
• Conception rate in hot climates: Jersey 32 percent, Holstein-Friesian 29 percent.
• Assistance needed at first calving: Jersey 7.5 percent of first births, Holstein-Friesian 21.6 percent.
• Survival to fifth lactation: Jersey 34 percent, Holstein-Friesian 23 percent.
• Feed efficiency: Jersey 112 g of milk solids per kg of dry-matter intake, Holstein-Friesian 97 g/kg.
• Heat-season yield loss: Jersey 5.35 litres per cow per month, Holstein-Friesian 5.76.
• Mature bodyweight: Jersey 380 to 450 kg.

On the headline figure, Jersey loses. She gives up roughly 24 percent of the Holstein-Friesian's 305-day yield. If your buyer pays per litre and nothing else matters, that is the end of the argument.

But look at every other line in the list.

She calves first about three months earlier, which means she starts earning sooner. Her calving interval is 67 days shorter, which from Module 1 you know compounds into more lactations across a life. She needs assistance at first calving one-third as often, 7.5 percent against 21.6 percent, which for a smallholder without easy veterinary access is a survival issue for both cow and calf. She is more than half again as likely to still be in the herd at her fifth lactation, 34 percent against 23 percent. She converts feed into milk solids more efficiently, 112 g against 97 g per kg of dry matter. And she loses slightly less milk in the hot season.

On composition, Jersey milk in the same review is reported as containing 18 percent more protein, 25 percent more fat and 20 percent more calcium than milk from other dairy breeds. Under a buyer who pays on fat and solids, that turns directly into money, and it partly closes the gap in litres.

Now the finding that should change how you evaluate any breed. In Ethiopian data cited in the same review, F1 Jersey by Boran crosses produced 13,546.50 litres of lifetime milk, against 12,816.7 litres for F1 Holstein-Friesian by Boran crosses. The Jersey cross out-produced the Holstein cross over a lifetime, even though Holstein yield per lactation is higher.

How can that be? Because lifetime output is litres per lactation multiplied by number of lactations. The Jersey cross's shorter calving interval, easier calvings and better survival gave her more lactations, and more lactations beat more litres each. This is exactly the lesson of Module 1 arriving from a different direction, and it is the single strongest argument in this course for judging an animal on her lifetime, not on her best day.

One caution the breed brings with it. Milk fever, the sudden calcium crisis at calving, is reported as more common in older cows and in certain breeds, with Channel Island breeds, meaning Jersey types, named specifically. If you keep Jerseys or Jersey crosses, watch them harder around calving. The nutrition module explains what to look for and why calling for help immediately matters.

So the practical conclusion. Where feed is limited, where calving assistance is hard to get, where cows must last, and especially where your buyer pays for fat and solids, the smaller exotic breed is a serious commercial proposition and not a compromise. Where feed is abundant and payment is per litre alone, the larger breed's extra litres count for more. Know which farm you are running before you choose.

Now put the evidence together into a decision you can actually make.

Five conclusions, each sourced from the lessons above.

One. Pure exotic breeds give the highest litres per day and the longest lactations. Pure Holstein Friesian reached 9.43 kg per day over a 362-day lactation, 3,183 kg in total, in the Ethiopian trial. But that trial ran at a research station. Those animals were not evaluated under the heat, feed shortage and disease exposure of an ordinary smallholding.

Two. The 50 to 75 percent exotic cross is the evidence-based recommendation from the trial actually available for this course, balancing yield against the hardiness the indigenous parent contributes. The half-bred at 7.21 kg per day beat the seven-eighths cross at 5.98 kg per day in the same trial.

Three. More exotic blood is not free. Ankole-Sahiwal crosses, zebu on zebu, showed no measurable heat penalty, while Ankole-Friesian crosses showed the steepest at 0.107 kg per THI unit. Exotic percentage is traded directly against heat resilience.

Four. Maintaining a target blood level requires an ongoing breeding programme. A single cross does not stay at 50 percent in her offspring unless mating is managed deliberately, generation after generation, with every sire recorded.

Five. Lifetime output, not per-lactation yield, is the right basis for the decision. The F1 Jersey by Boran cross out-produced the F1 Holstein-Friesian by Boran cross over a lifetime, 13,546.50 against 12,816.7 litres, on the strength of fertility and survival.

Now the decision procedure. Answer these before you look at any animal.

What can I feed in the worst month? Not the best month. Write down what you actually supplied last dry season. If the answer is poor forage and little else, an animal bred for high yield will not deliver it, and you will have paid for genetics you cannot use.

How hot does it get, and for how long? If your hot season is long and severe, the heat penalty measured in Lesson 3 is a real deduction from your income, and a lower exotic percentage or a zebu-to-zebu cross deserves serious thought.

What is my tick and disease pressure, and can I afford to control it? Remember from Module 1 that acaricide spraying was associated with a 5 to 11 percent yield increase in Kenya. If you can spray reliably, you can carry a less resistant animal. If you cannot, you cannot.

How far is veterinary and breeding service? An animal that needs assistance at calving one time in five, as Holstein-Friesian first-calvers did at 21.6 percent, is a different risk two hours from help than she is next door to a vet.

How does my buyer pay? Volume alone, or fat and solids? This decides whether a richer, smaller animal earns her keep.

Then, when you go to look at an animal, ask the seller these questions and write down the answers. What exactly is she, and what is the evidence, not just the claim? What was she served with, and when? What are her calving dates? What is she being fed to produce the yield you are being shown? That last question catches more bad purchases than any other, because a cow performing on purchased concentrate will not repeat that performance on your napier grass.

Finally, the honest limits of this module. No current breed-suitability map or national crossbreeding policy for any African country was available for this course, so none is given. The correct authority is your national livestock research institute or dairy board, such as KALRO in Kenya or EIAR in Ethiopia, together with your local extension service, and their recommendation should be asked for in the form: what cross do you recommend for this area, and what feeding does that recommendation assume?

A cow you can feed, in a shed you have cooled, bred to a sire you chose on purpose, will beat a better cow you cannot support. That is the whole module.