Breeds and the Veld-Adapted Cow Argument

#Ten rand a kilogram, for the colour of the hide

In December 2020 the ARC sold two lots of weaners off the same operation. The red Bonsmara and Bonsmara × Afrikaner calves averaged 195 kg and made R38.56/kg, excluding VAT. The multi-coloured Bonsmara × Nguni calves averaged 191 kg — four kilograms lighter — and made R28.80/kg. That is nearly R10 a kilogram, or about R2 020 per calf, and the reason the ARC gives is blunt: "feedlots know that Nguni types do not grow fast in the feedlot … and therefore used colour patterns to identify possible Nguni types" (Scholtz & Jordaan 2025).

Hold both readings of that at once. It is rational pricing on a real signal — the feedlot is buying feed conversion and inferring it correctly. It is also a structural tax on exactly the breeds best suited to low-input South African veld. Neither reading cancels the other, and that is what a breed decision is in this country: a bet on which end of the value chain you intend to be paid at.

#What the South African trials actually found

There is one clean South African head-to-head on natural pasture, and both camps quote half of it. Muchenje, Dzama, Chimonyo, Raats and Strydom ran Nguni, Bonsmara and Aberdeen Angus steers on natural pasture at Fort Hare in the Eastern Cape. In the authors' own terms (Meat Science 79(1):20–28):

• "The Nguni had the highest (P<0.05) average daily gain."
• "Bonsmara and Angus steers had higher (P<0.05) carcass weight and dressing percentage than the Nguni steers."
• Nguni meat was darker — L* values of 36.5 for Nguni, 38.6 Bonsmara, 39.9 Angus.

Read the shape of that rather than the headline. The small breed wins on gain per day off veld and loses on absolute carcass weight and dressing percentage. That is the cow-size argument compressed into one trial, and it hands neither side a victory.

One honesty note. You will hear that the same trial showed Nguni meat more tender than Bonsmara, and that it reported intramuscular fat of 1.12% against 1.07%. Neither claim appears in the published abstract. Do not repeat the tenderness ranking as a finding.

The second result is the one the conventional side skips. Du Plessis and Hoffman, in the arid subtropics, concluded that "steers from small- to medium-framed breeds can be raised successfully and slaughtered at 30 months of age without supplementation of natural pastures" — while noting that the smaller carcasses available at 18 months may need supplementary feeding to be market-acceptable (SAJAS 37(3):1–12).

That is a peer-reviewed South African result saying a moderate-frame veld animal is a viable finishing proposition. But take both halves of this one too, because the same paper prices the age penalty in the meat itself: Warner-Bratzler shear force of 2.62 ± 0.13 kg at 18 months against 3.16 ± 0.15 kg at 30 months. Older is measurably tougher. Back-fat over the same window rose steeply in the smaller types — Nguni doubled, the Bonsmara cross gained 84%, the Simmentaler cross only 16%.

So yes, you can finish it on grass at 30 months. And no, the classification system's use of age as a proxy for eating quality is not baseless — Lesson 4 prices what that costs.

#The number to argue over is not frame score

While the cow-size debate has run, conventional performance recording has quietly been moving the metric that matters. Over 25 to 30 years, kg calf weaned per LSU mated rose by 18.3% in the Afrikaner, 14.2% in the Drakensberger, 10.4% in the Nguni and 10.0% in the Bonsmara, with matched reductions in enteric methane per unit of product of 12.0%, 9.3%, 6.8% and 6.6% respectively (Jordaan et al. 2021).

Note what that measure does. kg calf weaned per LSU mated puts the calf on top and the cow's forage demand underneath, so a bigger cow must earn her extra intake before she scores. It is the honest cow-efficiency number, it is South African, and it is what both camps should be arguing over instead of frame score — and the two landrace breeds most associated with the veld-adapted case top that list under conventional recording.

#Frame size, converted into hectares

Module 3 taught the frame-size trap. Here is what it costs you in land.

The ARC publishes frame-based LSU factors because LSU is not linear in bodyweight. At an identical 450 kg with a calf at foot, a small-frame cow is 1.32 LSU and eats about 12 kg DM/day; a large-frame cow is 1.60 LSU and eats about 15 kg DM/day (Mokolobate et al. 2017) — a 25% difference in intake at the same liveweight. The ARC groups small frame as Afrikaner, Nguni, Boran, Tuli, Wagyu, Ankole and Dexter; medium as Angus, Brahman, Bonsmara, Drakensberger, Hereford and Sussex; large as Charolais, Limousin, Simmentaler and Romagnola.

Now put hectares under it, using the Grassland biome mean of 6 ha/LSU (DALRRD 2016) — and get your own land-type value before this goes in a plan.

Each large-frame cow costs you 1.7 hectares more than a small-frame cow of exactly the same weight. On 100 hectares that is two whole cows — about 17% of the herd — surrendered before anyone mentions weaning weight. On 500 hectares it is eleven.

One precision note, so you can defend the number. The 1.32 and 1.60 figures are for a cow with a calf at foot, not a dry cow, and the ARC publishes separate equations by physiological stage. The paper's own Table 2 also gives 1.66 rather than 1.60 for the large-frame 450 kg cow, which would make the gap 26% rather than 21%. This lesson uses the narrative figure; either is defensible, so say which you used.

#The numbers that are not there

Two gaps matter enough to name out loud, because the temptation is to fill them with imported figures.

There is no published South African average daily gain, and no published days-to-finish, for cattle finishing on veld or on planted pasture. The research behind this course went looking and did not find one. What does exist is the feedlot benchmark — an ADG above 1.6 kg/day is described as optimal (Scholtz & Jordaan 2025). No veld system approaches it. Grass finishing competes on cost of gain and premium, never on rate of gain, and if someone quotes you a veld ADG, the first question is which country it came from.

The second gap is the missing matched-control herd comparison in the debate box. Both absences carry the same lesson: the imported-practice failure mode is not usually a wrong number, it is a foreign number wearing local clothes.

#Where the selection data actually lives

None of this means "ignore the recording system". South Africa has genuinely good genetic evaluation infrastructure, and it is largely free to read.

Logix, run by SA Stud Book, produces estimated breeding values by BLUP, and beef breeds have moved to single-step genomic evaluation. SA Stud Book holds genomic profiles on more than 7 500 cattle and runs Logix across a long breed list including Afrikaner, Angus, Beefmaster, Bonsmara, Boran, Drakensberger, Nguni, Sussex and Tuli (SA Stud Book). Published South African beef EBVs are free to read at sabeefbulls.com.

EBVs and the veld-cow school are not enemies. The question is which EBVs you weight. A regenerative South African buyer weights fertility and maternal traits and caps mature cow weight rather than maximising 400-day weight — a disciplined use of the same system that produced the big cows.

The decision here is not "which breed". It is this: convert your own cows to LSU at the frame they actually are, work out the hectares each one occupies, and decide whether the extra kilograms you are selecting for pay for the extra land they cost. Until you can answer that in hectares, the breed argument is not yet a farming argument.