Toxic Weeds in Horse Pasture: Wilted Leaves Are the Risk

Two facts reframe this whole subject, and neither is about identifying plants.

Most toxic plants are unpalatable. UGA Extension: “Many toxic plants are highly unpalatable, so horses will not consume them if given the choice of other forage.”

And the acute danger is often not the living plant. On red maple, UGA: “Fresh leaves are not considered toxic, but wilted leaves can be highly toxic.”

Put together: the dangerous day is not the day a weed appears in the paddock. It is the day the grass runs out, or the day a storm drops a branch inside the fence.

The Wilted-Leaf Problem

This is the most useful thing on the page, because it changes what you actually do rather than what you look for.

Red maple. UGA: “Fresh leaves are not considered toxic, but wilted leaves can be highly toxic”, with symptoms of “severe anemia and death.”

Cherry, peach and plum. UGA: “the most toxic part of these plants is the wilted leaves”, which produce cyanide — and “cyanide interferes with oxygen transport in the body.”

So the risk is an event, not a condition. A red maple growing beyond the fence is not the hazard. The hazard is:

• A storm that breaks a branch and drops it inside the paddock
• Pruning and leaving the trimmings where horses can reach
• Leaves blown or fallen inside the fence line in autumn

UGA’s control instruction follows from that: “Fence them off well beyond the drip line of the trees so that horses can’t reach fallen leaves.” Beyond the drip line — not at the trunk, because the leaves fall outward.

And it produces a habit rather than a project: walk the paddocks after wind, and clear fallen branches before the horses find them. That is a five-minute job that addresses a lethal-dose risk.

Boxelder maple belongs here too, though the mechanism differs. UGA notes it is highly toxic to horses that consume the seeds from female trees, causing seasonal pasture myopathy, which “can be fatal in 90 percent of cases.” Seeds, not leaves — so the exposure season is different, and the outcome is worse than most things on this list.

Hunger Is The Trigger, Not Presence

UGA repeats this across species, and it is the thread that connects toxic plants to everything else about managing a paddock:

“Many toxic plants are highly unpalatable, so horses will not consume them if given the choice of other forage.”

Ohio State states the same rule as a single sentence worth pinning above the feed bin: “poisonous plants rarely cause issues for horses unless there is not enough proper feed/forage.” They add the two groups that break the rule — young or curious horses, and bored or nutritionally deficient ones — which is why a well-fed but under-occupied horse on a small paddock is not automatically safe.

Their standard for “proper forage” is concrete enough to check against a bale rather than a feeling: forage should be “green in color, palatable in texture, pleasant in smell, and free of weeds, mold, dust, and foreign matter.” And their first management step is not a herbicide: “identify possibly harmful plants in your area. Scouting is the best method.”

Species by species:

• Bitter sneezeweed — “extremely unpalatable… horses typically will not consume [it] unless they have no other choice”
• Black locust — “Horses typically will not consume these trees unless they do not have adequate forage”
• Nightshade — “very low palatability, though there are some reports of animals selectively grazing nightshades”
• Pokeweed — toxicity “concentrated in the root of the plant, which would be less commonly eaten”

Which means most poisoning cases on small acreage are a stocking problem wearing a botany costume. The paddock ran short, the horse ate what was left, and what was left was the unpalatable plant grazing had been selecting for all along.

UGA gives the requirement to measure against: a horse “needs to consume approximately 2% of their body weight in dry matter every day.” If the paddock cannot supply that, something else gets eaten.

And their concluding prevention advice is management rather than herbicide: “good pasture management and a sound nutritional program, including access to good-quality forage, are among the best ways to prevent pasture toxicities in horses.” That is the same conclusion our pasture weed control guide reaches from the weed side — overgrazing is the upstream cause of both problems.

The exception that breaks the rule. UGA notes there are “some reports that yew trees might be palatable to horses”, and yew is the most toxic thing on their list. A plant that is both attractive and lethal cannot be managed by keeping the horses fed — which is why UGA says yew “should always be removed.”

The Amounts UGA Publishes

Doses are rare in extension writing on toxic plants, so where UGA gives one it is worth having:

Put the yew figure in perspective: half a pound is a small armful of clippings. A hedge trimming tossed over a fence is a fatal dose.

The Johnsongrass entry is different in kind and worth reading — it is a cumulative rather than acute risk, and UGA notes the plant “can produce cyanide under stressed conditions (such as drought)”, plus a hay warning: “Care should be taken to avoid feeding hay with high concentrations of Johnsongrass present.”

The Two Penn State Singles Out

Penn State has a dedicated publication on these because they are the two that show up in pastures across most of the country, and both have counterintuitive control timing.

Poison hemlock

Identification — three features together:

• Stem: “hairless, hollow main steam with purple blotches/spots which branches”
• Size and smell: “2 to 6 feet tall” with a “parsnip-like odor when the leaves are crushed”
• Rosette stage: “fern-like, glossy green, lacy leaves”

Toxicity: “All parts of the plant are poisonous” — pyridine alkaloids, chiefly coniine — and “significant poisoning can result in muscle paralysis and suffocation.” Penn State’s thresholds: 0.25% of fresh body weight for horses, 0.5% for cattle.

A warning for the person, not the animal: “The sap of poison hemlock has the potential to cause dermatitis in some people”, and Penn State’s instruction is to “wear protective clothing and gloves when handling the plant” — including when weed-whacking it, which aerosolises sap onto bare arms.

Timing: “Herbicide applications are most effective when they are sprayed in the fall when poison hemlock is in the rosette stage or before it bolts in the spring.” So autumn rosettes, or early spring before the stem extends. Actives Penn State names: 2,4-D + dicamba (1 quart + 1 pint), Crossbow (2,4-D + triclopyr), NovaGraz, and glyphosate as a spot treatment.

Buttercup

Species: bulbous (Ranunculus bulbosus), creeping (R. repens), tall (R. acris) and small flower (R. arbortivus). “Most of these have a perennial lifecycle, however some can behave as a winter annual.”

Toxicity, from UGA: “mouth blisters, excessive salivation, and mild colic to more severe symptoms such as hemorrhaging, convulsions, and death.”

And the timing trap. Penn State:

“The use of herbicides during early spring (Mid-March through April) is the most effective time to control buttercup populations.”

“Spraying when buttercup is flowering usually does not provide effective control.”

The yellow flowers are what makes you notice it, and by then it is too late to spray effectively. Put it in the diary for the previous month instead.

Penn State’s rates: 2,4-D at 2–3 pints/acre, 2,4-D 1 quart + dicamba 1 pint, metsulfuron at 0.2–0.3 oz, or Crossbow at 2–3 quarts. Plus a temperature condition — “air temperatures are at least 50°F or more for a few consecutive days.”

And mowing has a role: “Mowing fields as low as possible during the early spring before they start to flower can help reduce seed production.”

One diagnostic worth noting. Penn State: “Pastures with dense populations of buttercup and other weed species… often it is a sign of over-grazing and a poor stand of desirable forage.” Dense buttercup is a reading on your grazing management, not just a weed to spray.

What To Do, In Order

On Products, Honestly

We do not currently carry a pasture-labelled broadleaf herbicide, and that matters more here than it would on a lawn page.

The actives Penn State names for these weeds — 2,4-D, dicamba, triclopyr combinations, metsulfuron — exist in our catalogue only in lawn-labelled products. A lawn label is not a pasture label. Using a turf product on grazed pasture is an off-label application, and the grazing and haying restrictions that make pasture use lawful are precisely what a lawn label does not carry. Our pasture weed control guide covers why those intervals — seven days to one year, and different for dairy than beef — decide the purchase.

What we do carry is the brush and woody-plant side:

For the toxic weeds on this page, the correct route is a pasture-labelled product from a farm store, chosen against the grazing interval your rotation can absorb — and UGA’s own advice on which one: “A local county Extension agent can assist you in determining when to apply herbicides and what herbicides might be most effective.” On a livestock safety question, that is better advice than any national guide can give, and we would rather point you at it than pretend otherwise.