In late September, 2018, I returned to Australia after 25 years in the US. I was not familiar with capeweed (Arctotheca calendula), or African daisy as I came to know it, so when I saw it growing in the pasture shown above, I had to look it up.

I don't worry too much about most weeds in horse pastures, as long as there is plenty else to eat. I see most weeds as a symptom (of an ailing pasture), not the disease itself; and in some cases as an indication of healthy biodiversity. Still, I am terminally curious, so I looked up its name and whatever I could find about its toxicity in horses. It didn't seem to be particularly toxic, so I didn't worry about it.

I was managing a 50-acre farm in Broadford, VIC for a friend, who had just bought it and moved her four horses onto it from Werribee. Before that, it had been used to raise beef cattle, and most of the grazing land was 'improved' pasture, comprising a selection of seeded pasture grasses. The rest was native grasses.

As the best grass was gradually eaten down, the horses started eating the big patch of capeweed you can see up by the gate, eventually eating it down to the ground. There was still plenty of grass in the pasture; they were choosing to eat the capeweed, and one mare in particular, a big Andalusian cross, seemed to develop quite a taste for it.

None of the horses showed any signs of physical or behavioural disorder as a result, which led me to conclude that capeweed is not toxic to horses, at least under the conditions I observed. But is that true more broadly? Are there conditions under which capeweed is toxic, or at least problematic, to horses?

Is capeweed really toxic to horses?

I set out to answer that question in more detail after a friend sent me the substance of a recent Facebook post about a senior mare with equine Cushing's disease who suddenly developed some odd behavioural problems (unsteadiness, apparent blindness, and hyper-reactivity), along with loss of appetite, mild colic signs, and elevated body temperature and heart rate.

Neurological problems in senior horses are worthy of an article all their own. I simply note her age group and Cushing's status as possible contributors. The mare had recently been started on pergolide (one of the drugs we use to treat Cushing's disease in horses), and among its short-term side effects are lethargy/depression and loss of appetite. But there I'll leave it, because I want to stay focused on capeweed in this article.

The post also included a reply from the founder of a popular equine supplement company, who is an equine vet of some renown. He (or whoever might have been writing under his name) confidently stated that the mare's problems were most likely caused by capeweed ingestion, and he (or whoever) recommended double doses of two of his products.

The most important piece of advice, though, was lost in all the hype and hoopla about capeweed toxicity and its commercial remedies: Prevent further access to capeweed. I'll explain why that was so important toward the end of this article. It may seem patently obvious, but the reason will probably surprise you.

Before I go on, another bit of misinformation concluded the Facebook post: the poster said that capeweed can cause Australian stringhalt. That is simply not true.

Pasture- or toxin-associated stringhalt in horses, aka Australian stringhalt, is caused by excessive consumption of flatweed (Hypochoeris [Hypochaeris] radicata), also known as smooth cat's ear or false dandelion. Capeweed (A. calendula) is not even in the same genus, although both are in the same huge family of plants with daisy-like flowers, Asteraceae, or the Asters, along with the common dandelion (Taraxacum officinale), which is not toxic to horses.

How toxic is capeweed to horses? — The science

Every plant has its defences against herbivory (e.g., unpalatable biochemicals), and capeweed is no exception. But just how toxic is it to horses?

I thought I knew the answer (not very, and only under certain conditions). But to be sure, I went looking in the scientific literature for studies on the toxicity of capeweed in horses. I always like to read the original studies for myself whenever I can. I also like to understand the mechanisms of action: how or why a toxic substance causes disease.

Imagine my surprise when I found no studies showing that capeweed is toxic to horses!

I searched by its common names and by its botanical name. Nothing.

I then asked an equine vet friend in the US if he knew of any studies. He did not, even though capeweed is naturalised in the US (California, in particular).

He has an unfortunate penchant for using artificial so-called intelligence (which is most assuredly not intelligent!), but that turned out to be useful in this instance. He ran a literature search on Consensus, a website that bills itself as “the AI-powered academic search engine.” It searches and analyses over 200 million peer-reviewed research papers.

Consensus came up with this rather revealing assessment when prompted to search for papers on capeweed (Arctotheca calendula) toxicity in horses:

“No retrieved paper addresses Arctotheca calendula specifically.”

In other words, in a search of more than 200,000,000 scientific papers, not even “AI” could make a case for capeweed being toxic to horses.

I did, however, find a study — just the one — of nitrite toxicity in sheep grazing capeweed. The sheep were grazing a field that was almost completely covered by young, rapidly growing capeweed.[1] That was published in 1969. Nothing since.

As I'll discuss in a bit more detail later, horses are particularly tolerant of nitrates (-NO3) and nitrites (-NO2) in feed and water. They easily cope with nitrate levels that cause toxicity in sheep and cattle.

The lack of published studies where I usually find them really surprised me, as there are dozens of papers on flatweed (H. radicata) in relation to Australian stringhalt.

There are studies documenting the specific clinical abnormalities and histopathology (microscopic changes) — i.e., what it looks like, and why — of Australian stringhalt. There are investigations into the toxic principal(s) in H. radicata that cause these abnormalities. There are studies on the environmental conditions under which outbreaks of Australian stringhalt occur. (By the way, they are also documented in North America, South America, Europe — anywhere H. radicata is found.) And there are studies on response to various treatments, and on the natural progression of the disease both with and without treatment.

In short, flatweed (H. radicata) toxicity in horses is very well documented. We don't know everything about it yet, but we do know a lot.

And yet there's nothing at all about capeweed toxicity in horses. Although it is an introduced species (as is flatweed — and the horse), capeweed has been here for over 150 years.[2] By now, there should be numerous studies of its toxicity in the various livestock species that graze it — if it is toxic to grazing animals.

I did find another study of capeweed consumption in sheep. It compared three groups of young-adult Merino wethers:  group 1 grazed capeweed-free pasture; group 2 were fed lucerne hay (0.7 kg/day); and group 3 were fed 6 kg of freshly cut capeweed daily for a week.[3]

The capeweed diet did not cause diarrhoea, and based on measurements of rumen fermentation, capeweed was "as nutritious as pasture containing grasses and clovers.”

The only downside the authors noted was that the sheep eating the capeweed diet lost a little weight because they were not consuming enough dry matter, which is everything in a plant minus its water. The fresh capeweed was only 11% dry matter (i.e., 89% water). So, at a feeding rate of 6 kg/day, the capeweed filled them up but did not quite meet their daily nutrient needs.

The water content of fresh capeweed is an important component of all this, and it comes up again in the next section, so hold that thought.

How toxic is capeweed to horses? — The internet

When I searched online for information about capeweed toxicity in horses, I found a plethora of web pages and Facebook posts about how toxic it is to horses. However, with one notable exception, which I'll discuss next, they all repeat what someone else has said about it somewhere else. I could not find any original source material anywhere. It is all just hearsay. Around and around in endless circles.

The one notable exception is a brief report from New Zealand of tetanus-like spasms and hyper-reactivity in a pony who had grazed for less than an hour on a freshly-mown lawn full of capeweed, shortly before the signs of tetany (severe muscle spasms) began.[4]

It is short on some important details and long on assumptions and circular reasoning, and I'm sure there was more going on in that pony than capeweed toxicity. (In fact, I doubt that capeweed was the main cause of his problems.) However, there are some nuggets that are worth examining, as the vet ran bloodwork on the pony, and someone (perhaps the author) ran a nutrient analysis on a sample of the capeweed lawn.

Magnesium deficiency?

The common refrain on the internet is that capeweed causes magnesium (Mg) deficiency in horses. But no-one ever provides any evidence, and the prevailing explanation for how it might do that is, quite frankly, implausible.

I'll talk a little about the biochemistry in a bit. Just a little, because it's really complex and mostly based on research in other species — and because it argues against the simplistic explanation in the end...

Preventing the horse from eating any more capeweed and supplying her with supplemental Mg is not evidence that capeweed caused a problematic Mg deficiency when the horse improves.

As I'll explain a little later, it is more likely that removing the horse from the offending pasture lowers the total protein or nitrogenic content of the horse's diet. Hold that thought, too, because things get really interesting when we plumb those depths.

Fortunately, the pony case involved bloodwork. The pony's serum or plasma Mg was “good.” It was calcium that was low in that case.[4] (As I said, I think there was much more going on in that pony than we're told.)

Anyway, I went looking for a possible mechanism by which capeweed could cause Mg deficiency in horses, but all I found were numerous articles about Mg deficiency in plants! That's important, because horses are supposed to meet their Mg needs by eating plants. But that's a topic for another day.

. . .

Briefly, for your consideration:

❝ Dietary Mg deficiency in horses is very rare unless extreme conditions combine to result in decreased consumption and increased demand, such as long-distance transportation of unfed lactating mares or prolonged administration of enteral [into the gut] or parenteral [by injection] fluid or nutrition solutions deficient in Mg. ❞ [5]

❝ It seems unlikely that horses would develop chronic whole-body Mg deficiency because efforts to induce Mg depletion have required long-term feeding of severely Mg-deplete artificial diets in young growing animals. […] dietary Mg supplementation to horses is infrequently required when a normal diet is fed. ❞ [5]

Also in relation to the behavioural and digestive disorders we're talking about in horses grazing capeweed:

❝ In comparison to cattle, clinical signs of hypomagnesemia are rarely reported in horses, but include weakness, muscle fasciculations [tremors or twitching], ventricular arrhythmias [abnormal heart rhythm], seizures, ataxia [unsteadiness], and coma. ❞ [5]

. . .

But I did come across a couple of blog posts by a popular equine nutritionist in Australia which point the finger at high potassium (K) levels in lush pasture as a cause of spooky and otherwise hyper-reactive behaviour in horses, these behaviours attributed to the resulting drop in blood Mg.

In short, high dietary K → low blood Mg, or so the theory goes. And it is just theoretical, and quite unlikely, in horses.

(By the way, the chemical shorthand for potassium is K because its other name is kali or kalium.)

So, let's take a quick look at this K–Mg dynamic. It is well described in sheep and cattle because they can develop low blood Mg (hypomagnesaemia) and signs of disordered muscle and nerve function (grass tetany or grass staggers) on lush pasture, and high K levels in the pasture are implicated.

But there's way more to it than that. The biochemistry is really quite complex, and the high water content of lush pasture has an important moderating influence on it all.

Here's an example, stripped down to its take-home message:

An experimental study in sheep showed that a high-K diet (4.2% K) lowered the serum Mg concentration by 0.4 mg/dl, or just under 17%, after 10 days.[6]

Well, that seems to settle matters, doesn't it.

Not quite. In a separate experiment comparing two different forms of dietary K (chloride and acetate), serum Mg dropped on both high-K diets, but it was still within the normal range.

In other words, neither high-K diet would have caused signs of hypomagnesaemia — and it was a lot of K: 36 g/day for a 55-kg sheep, whose maintenance K requirement is around 5 g/day. That's equivalent to 172 g/day for a 500-kg horse, whose maintenance K requirement is 25 g/day.

Furthermore, when the sheep were fed a high-K + high-water diet (7.5 L/day directly into the rumen), to better reflect the high intakes of both K and water when grazing lush pasture, this diet lowered the serum Mg by only 0.22 mg/dl. That's a drop of only 9%.

The extra water intake decreased the effect of the high-K diet on serum Mg, by almost 50%.

The normal range of serum Mg in horses varies somewhat with the lab, but it ranges from a low of 1.46 to a high of 2.60 mg/dl — similar to that in sheep. A drop of 0.22 mg/dl is not particularly concerning, as long as the horse is on a diet that supplies adequate Mg — for example, providing a good quality grass hay to horses on lush pasture. (I have yet to see a good quality grass hay that is low in Mg, and horses on lush pasture need the extra dry matter anyway.)

But horses and sheep have very different anatomy and physiology, particularly when it comes to their digestive systems. So, what do we know about Mg in horses and how it is influenced by K?

The seminal studies on mineral metabolism in horses were done in the 1970s. In one, a dietary K intake of 3.4% (34 g/kg DM) did not decrease Mg digestibility or retention. Nor did it affect calcium and phosphorus digestibility or retention while we're at it.[7]

Grass hays in Australia typically contain around 1.6% K (16 g/kg DM), with most being somewhere between 0.9% and 2.3% K.[8] So, the metabolism study examined the effects of a fairly high-K diet in horses.

Grass pastures in Australia typically contain around 2.9% K (29 g/kg DM), with most being somewhere between 1.8% and 4.0% K.[8] So, well below the high-K diets in the sheep study, and within range of the equine K metabolism study. And when pasture is lush, it can be more than 80% water.[8]

For those who are interested, the average K and Mg content of Australian grasses (hay and pasture) are shown, with their average ranges, in the tables at the end of this article.

As for what's happening in the bloodstream, a study of 823 horses hospitalised for various medical or surgical issues found that most (73%) of the 401 horses with hypomagnesaemia (serum total Mg 0.6–1.6 mg/dl) had normal serum K. The other 27% had serum K that was below its normal range. Only one horse with hypomagnesaemia had high serum K (hyperkalaemia).[9]

In the final analysis, serum K was not significantly associated with hypomagnesaemia.[9]

As with all minerals, K and Mg metabolism are complex, and the two are not as coupled as the simplistic explanation suggests for why capeweed or lush pasture in general may cause abnormal behaviour in some horses (i.e., Mg deficiency caused by high K intake).

Perhaps I should simply have led with this:

Attempts to produce hypomagnesaemia and tetany in ruminants by adding K to their diet usually fail.[6]

So, maybe there is another explanation for abnormal behaviour and digestive disturbances in some horses eating capeweed, in some circumstances.

Nitrate toxicity?

The other popular explanation for capeweed toxicity is nitrate poisoning. However, horses are particularly tolerant of nitrates in feed and water, compared with ruminants such as sheep and cattle.

In the pony case I mentioned, the nitrate concentration of the sampled capeweed lawn was 2,890 mg/kg dry matter (DM), or 2.89 grams/kg DM.[4] That seems really high, but...

The oral dose required to cause nitrate toxicity in horses is somewhere between 61 and 152 grams.[10] (By the way, that's a huge range!) When fresh capeweed is more than 80% water, or less than 20% dry matter, as we saw in the second sheep study I mentioned, a horse would have to eat an impossible amount of capeweed (over 100 kg) in just a few hours to consume even 61 grams of nitrates.

In an experimental study of nitrate toxicity/tolerance in horses, dietary intakes as high as 18.5 grams/kg DM were tolerated by nonpregnant mares.[11] The average 500-kg horse, fed at a daily DM intake of 1.5% body weight (7.5 kg/day), would be eating around 139 grams of nitrates per day on this diet. With twice-daily feeding, that's around 70 grams of nitrates per meal.

This level of intake did not cause the blood disorder (conversion of haemoglobin to met-haemoglobin) and the reduction in oxygen-carrying capacity we see in ruminants with nitrate toxicity, even after 2 weeks on this high-nitrate diet.[11]

However, horses do have their limits. In a rare and tragic report of fatal nitrate toxicity in nine Thoroughbred broodmares on a large farm in Turkey, the pastures on which the mares were grazing had been treated with a nitrate-based fertiliser one day earlier. The authors estimated that the mares had consumed approximately 80 grams of nitrate in total from their diet (pasture, grass hay, alfalfa hay, and concentrates), most of it from the fertilised pasture.[12]

Although the authors detailed the daily amounts of grass hay, alfalfa hay, and concentrates fed, along with their nitrate levels, they could only estimate the mares' daily pasture intake, with a fudge factor of 2x (approximately 5–10 kg, wet weight). It is likely that the most severely affected mares were the ones who consumed more of the recently fertilised pasture, thus consuming considerably more than 80 grams of nitrates.

In short, nitrates are not completely benign in horses, but it does take a lot to cause nitrate toxicity in this species. And the signs of nitrate toxicity in horses relate to cardiorespiratory failure (low blood and tissue oxygen), not the behavioural issues described in horses grazing pastures that contain capeweed.

Mycotoxins?

There is one more suggested explanation. The vet I mentioned at the start also claimed mycotoxin (fungal toxin) involvement, along with a ready solution: one of his products.

I can find nothing in the scientific literature about capeweed being any more vulnerable to fungal infestation than any other pasture plant. Nor does it seem all that plausible, given the fairly upright and open growth habit of this broadleaf plant that favours full sun in bare or sparsely covered areas. It will grow in the shade and in competition with grasses, but it grows best in full sun, with little competition from others.

animal health consulting

Key points:

* there are no studies showing that capeweed (Arctotheca calendula) is toxic to horses

* the claimed mechanisms of toxicity are all unproven and unlikely in horses

* there is a lack of studies on capeweed toxicity in any grazing species; in fact, capeweed can be a valuable source of nutrition in livestock grazing systems

* like lush grass, capeweed may contribute to behavioural and digestive problems in grazing horses because of its high crude protein content, not because of any toxic effects

* it may not be necessary to take the horse off the pasture if other protein sources in the diet are reduced or withheld until pasture conditions change

Relative abundance of the main bacterial phyla in the faeces of 6 horses on the same pasture year-round.[18]

This UK study ran from mid-April one year (sample 1) to mid-April the next (sample 27). A seventh horse was dropped from the study after 4 samples.

Phylum is a huge, umbrella category five floors up from species, so these are enormous changes in the horse's gut microbial community over the course of the year.

The only pattern we see here is the sawtooth variation throughout the year. No two horses were alike (a universal finding in equine gut microbiota studies), and no two horses responded alike to the same pasture conditions.