Worms, Refugia and Targeted Selective Treatment

#The worst thing you can do to a drench is use it on everything

On eight communal sheep farms sampled in King Sabata Dalindyebo, Eastern Cape, resistance was found on all eight. Benzimidazole and salicylanilide resistance on all eight; ivermectin on six of eight; levamisole on two of eight. Every single one of the eight farms carried resistance to at least two different drug classes (Mavundela, Dzemo & Thekisoe, 2025).

Nobody did that on purpose. It is the arithmetic of a monthly drench calendar, run faithfully, for years.

Here is the arithmetic. Every worm in a treated animal meets the chemical. The ones that die contribute nothing to the next generation. The ones that survive do — and they survive because they carry the genes that let them. Do that to every animal in the flock, every month, and within a few years the only worms your farm produces are the descendants of survivors. You did not lose the drench because the drench was bad. You lost it because you gave the resistant worms no competition.

#Refugia is the competition

Refugia is the fraction of the worm population that is not exposed to the drench when you treat: the larvae sitting on the pasture, and the worms inside the animals you left alone. Van Wyk, writing in 2001, called refugia perhaps the most potent factor governing how fast anthelmintic resistance develops.

Resistant worms that survive a treatment still have to breed with something. If there is a large pool of ordinary, drug-susceptible worms out there — on the veld, in the untreated animals — the survivors get diluted into it, generation after generation. If there is no such pool, the survivors are the whole next generation.

Put numbers on it. Take a flock of 200 ewes in a Highveld February — the hot wet window, the part of the year in which the Limpopo survey recorded 1,210 to 1,861 eggs per gram (Mphahlele et al., 2021). Those are Limpopo numbers, not Highveld ones; take them as the order of magnitude a wet summer produces, not as your farm's figure. Assessed individually, 14 ewes are anaemic enough to need treatment and 186 are not.

Drench all 200 and the next generation of worms on that camp descends entirely from whatever survived in 200 animals. Drench the 14 and the surviving resistant worms from those animals are breeding into egg output from 186 untreated ewes carrying ordinary, susceptible worms — a susceptible pool more than thirteen times the size of the treated group. Same product, same day, same flock. One version dilutes the survivors; the other concentrates them.

Two long-standing pieces of South African practice fall over once you accept this.

Whole-flock monthly drenching is the fastest known way to build total resistance. It is sold as cheap insurance. It is a payment plan on losing your chemistry.

Drench-and-shift onto clean veld was taught for decades as good practice, and it is now understood as a strongly resistance-selecting move. Think it through: you kill everything susceptible in the animal, then move the survivors onto pasture that carries no worms at all. Every egg that lands on that clean camp comes from a worm that just survived a drench. There is nothing to dilute it. You have built a resistant pasture on purpose.

If you must move after treating, either leave the treated animals on the dirty camp long enough to void, or leave a proportion of the mob undrenched to travel with them. Your veterinarian sets the detail.

#Targeted selective treatment, and what it honestly costs

Targeted selective treatment (TST) is the practical form of refugia, and it is South African in origin. Instead of drenching the flock, you assess every animal, treat only those that need it, and deliberately walk past the rest.

For Haemonchus contortus — wireworm, the blood-sucker that kills by anaemia rather than by scouring — the assessment tool is the FAMACHA card, and the card is only issued with training: "Producers must receive training in order to receive a card" (wormx.info). That is a deliberate control. The system is easy to misapply, and a misapplied TST programme leaves anaemic animals untreated. The next lesson deals with what the card can and cannot see; here the point is the strategy, not the instrument.

Now the honest part. Van Wyk's South African production trials found that TST using FAMACHA did cost some production relative to suppressive drenching — the loss was small relative to the benefit of reduced selection for resistance, but it was not zero (Van Wyk et al., OJVR).

That is the trade, stated plainly: you buy the future usefulness of your drenches with a small amount of present production. On a farm where one or two drench groups already fail, that is an obviously good deal. On a farm where everything still works, it is an insurance premium. Either way, a course that told you TST was free would be selling you something.

#Where principle three becomes a parasite tool

This course keeps telling you to graze for residual — to leave cover on the ground rather than grazing to the dirt. Usually that is argued from soil: armour, litter, infiltration, temperature.

There is a second reason, and it is veterinary. Infective Haemonchus larvae concentrate in the bottom few centimetres of the sward. An animal grazing to a high residual takes most of its bite above that zone. An animal grazing to the ground takes its bite from inside it.

This is the cleanest place in the whole course where a regenerative principle and a veterinary one point in exactly the same direction.

The corollary is uncomfortable for anyone running a tight rotation. Recovery period is a parasite variable, not only a plant variable. South African farming-press guidance holds that rotation does not generally control parasites unless rest exceeds roughly 70 days (Farmer's Weekly) — and most planned-grazing recoveries in a summer growing season run 30 to 60 days, which is inside larval survival. Be careful with that 70-day number, though: it comes from the farming press, and the research behind this course could not find a South African measured larval survival curve by biome and season to corroborate it. Treat it as a warning about the order of magnitude, not as a threshold to plan against.

Worse for the mob-grazing enthusiast: high stocking density concentrates dung, and therefore worm eggs, onto a small area. If you cannot achieve long recovery, density is working against you here even while it works for you elsewhere.

#Cattle dilute the challenge. They do not break the cycle.

A claim you will hear at every farmers' day: run cattle ahead of the sheep, because cattle and small stock do not share worms, so the cattle clean the camp.

The first half of that is good advice. The second half is false, and it is the kind of false that kills lambs. In South Africa, Haemonchus species, Trichostrongylus axei, Cooperia species and the liver flukes Fasciola hepatica and F. gigantica all cross between cattle and small stock. A camp grazed by cattle is not a clean camp for sheep.

What is true is weaker and still worth doing. Most of the economically dominant nematodes have a strong host preference, so a leader-follower sequence — cattle first, small stock after an adequate rest — dilutes the larval challenge each species meets rather than concentrating it, which slows the selection pressure driving resistance. The other gains are real and have nothing to do with worms: the species use different sward strata, they browse against graze, and each avoids only its own dung patches, so total utilisation rises (SA Smallholder).

Note what this lesson is not giving you: a number. No South African trial quantifying the worm-dilution benefit of cattle-then-small-stock grazing could be found. The mechanism is sound and worth building into your grazing plan. Anyone who tells you it cuts your drenching by a specific percentage is guessing.

The liver flukes carry their own tension. Fluke is a wet-footprint disease — vleis, spring seeps, dam margins, leaking troughs — and the snail host Lymnaea truncatula has been mapped at 221 loci across South Africa in an extensive but patchy distribution (de Kock et al., JSAVA). The riparian and wetland ground you are rehabilitating is the snail's habitat. The management answer is fencing and controlled short-duration access, not a molluscicide.

#What to do with this on Monday

Two changes, and they cost nothing.

Raise your residual. Whatever height you have been leaving, leave more — you are now grazing above the larval zone as well as covering the soil, and the same decision pays twice.

Then, at the next handling, walk past the two healthiest animals in the pen without drenching them. It will feel wrong. It is the reason your drench will still work in five years.