Alfalfa for roses: what fifty years of rose growing actually proves

Roses

Alfalfa for roses: what fifty years of rose growing actually proves

By Joe, Founder of Dr Forest · July 2026

Rose growers have used alfalfa since the 1970s. The practice traces back to one experiment on ten rhododendron seedlings, run by a pathologist in his own garden. What it actually does for a rose, and what it does not, are two different questions.

Alfalfa for roses earns its place, though not always for the reason people give. Alfalfa meal supplies somewhere between 2% and 3% nitrogen, 1% to 2.5% potassium and a useful amount of calcium, released over several weeks rather than in a single flush. That is a sound feed for a rose. The celebrated extra, a growth compound called triacontanol, is present in the plant and was first isolated from lucerne wax in 1933. Whether enough of it survives in a bag of meal, and reaches a rose root at a dose that does anything, has never been measured by anyone.

So this piece separates the two. What alfalfa reliably delivers to a rose, and what it might deliver. The first part is worth your money. The second is an open question that half the internet reports as settled fact.

What is actually in a bag of alfalfa

Alfalfa is Medicago sativa, known in Britain as lucerne. It is a deep-rooting perennial legume that fixes its own nitrogen and mines minerals from well below the depth most garden plants reach. Dried and milled, it becomes alfalfa meal. Compressed under heat, it becomes pellets. Same material, different surface area, so meal starts working sooner and pellets take a fortnight or so to soften and break down.

The nutrient declaration on fertiliser-grade alfalfa sits at roughly 2–3% nitrogen, near-zero to 1% phosphate, and 1–2.5% potash. Feed-grade material is analysed far more thoroughly, and the numbers there are more revealing. Feedipedia's dataset of 14,989 samples of dehydrated alfalfa puts crude protein at 18.3% of dry matter on average, with a range of 13.1% to 27.9%. That converts to roughly 2.1% to 4.5% nitrogen depending on the batch.

Two things stand out. Alfalfa carries a lot of calcium, at 22.1 g/kg, which most organic nitrogen sources do not. And it carries very little magnesium, at 2.1 g/kg, against 25.6 g/kg of potassium. Hold that thought, because it matters more for roses than for almost any other garden plant.

The variation is wide and worth understanding. Feedipedia attributes it to growth stage, cut number, the leaf-to-stem ratio, moisture at harvest and processing method. As alfalfa matures the leaves make up less of the crop and the stems make up more, so protein falls and fibre rises. A leafy first cut and a stemmy late cut are different products in the same sack. Any single figure printed on a bag is a declared minimum, not a measurement of what you bought.

Where rose growers got the idea

The compound at the centre of this story was found long before anyone thought to put it on a rose. In 1933 Chibnall and colleagues at Imperial College published the isolation of n-triacontanol from lucerne wax. A long-chain alcohol, sitting in the waxy coating of alfalfa leaves. Nobody suggested it did anything useful for forty-four years.

Then in March 1977, Stanley Ries and colleagues at Michigan State published in Science. They had found that alfalfa meal, and chloroform extracts of that meal, increased growth and yield in several species. They isolated the active crystalline substance, identified it by mass spectrometry as triacontanol, and showed that sprays containing it increased growth in rice, maize and barley, with a similar response in soil-grown tomatoes. The paper is real, the finding is real, and the compound came out of alfalfa. That much is beyond argument.

What happened next is the part rose growers inherited. A pathologist at the University of Michigan Medical School, D. L. Hinerman, read the Science paper and a follow-up piece Ries wrote for Horticulture magazine that August. He decided to try alfalfa on his rhododendrons.

Ten seedlings of Rhododendron metternichii, grown from society exchange seed in an identical medium. Six got roughly half a cup of shredded alfalfa hay mixed into their bark mulch on 15 July 1977. Four got the bark alone. A year later the treated plants were 55.1% taller with leaves 53.5% longer. Hinerman noted that the nitrogen in half a cup of hay was nowhere near enough to explain a difference of that size, and pointed at triacontanol.

He was careful about it in a way the people who cite him have not been. He wrote that many more experiments were indicated, and that clones should be used in place of seedlings to control for genetics. Six plants against four, one site, one season, no statistics. It is a good observation honestly reported. It is not a trial.

Word went round the rose and rhododendron societies. It has been going round ever since, and somewhere along the way the caveats fell off.

Worth adding that Hinerman himself records an earlier wave of interest in alfalfa extracts among nurserymen around 1954, more than two decades before Ries. The habit is older than the science that is supposed to justify it.

The triacontanol claim, weighed honestly

Here is where I have to be awkward about a product I sell.

Nobody has published how much triacontanol is in a kilogram of alfalfa meal. I looked. The 1977 Science paper discovered the activity in alfalfa but gives no tissue concentration. The standard review of triacontanol in agriculture, by Naeem, Khan and Moinuddin in 2012, lists alfalfa among the genera whose epicuticular waxes contain it and gives no figure for the tissue. The 2020 review by Islam and Mohammad does not give one either. Anyone quoting you a milligrams-per-kilo number for alfalfa meal has invented it.

The applied doses in the literature are extraordinarily small. Triacontanol has shown activity at 10 micrograms per litre, and at 23 nanomolar. The field rates in the older Chinese and American work are measured in fractions of a gram per hectare. That cuts both ways. It means you would not need much of the compound to matter, and it means the difference between an effective dose and an ineffective one is a very narrow target to hit by scattering dried leaves on soil.

And the applied-triacontanol field record is poor. Hinerman, writing in the same rhododendron bulletin, noted that investigators at the USDA and elsewhere had tried to repeat the experiments with varied results, and that a follow-up he co-authored describes subsequent field tests by Ries and others as inconsistent and disappointing. There was a suspicion at the time that the maximum useful dose is critical and had been exceeded in the failed trials. That is a plausible explanation. It is also exactly what you would expect to hear if the effect were unreliable.

On roses specifically, the cupboard is close to bare. I found one peer-reviewed primary study of triacontanol applied to roses: a 1997 Indonesian split-plot trial on cut-flower roses, published in a national horticulture journal, no DOI, effectively uncited. It reported best flower production at 150 ppm applied three times. The doses are three orders of magnitude above the concentrations that work in the rest of the literature, which suggests it was a commercial formulation rather than pure compound. I have not been able to read the paper itself. I would not build a purchase decision on it, and I am not going to pretend otherwise.

There is no peer-reviewed study of alfalfa meal applied to roses at all. Not one.

The nitrogen bank

Alfalfa releases on two clocks, and the difference is the whole reason to use it.

The potassium, calcium and phosphorus are not locked into complex structures. They sit in the tissue in largely soluble form and wash out as the material breaks down, so they are available within weeks. That is the quick part, and it is why a spring application shows up in the same season.

The nitrogen behaves differently. A portion mineralises quickly, and work compiled by Perennia found most of the readily available nitrogen in low carbon-to-nitrogen materials comes out in the first three weeks. The rest does not vanish. It gets built into microbial biomass and soil organic matter and comes back out slowly across following seasons. Alfalfa meal sits at a carbon-to-nitrogen ratio of about 12 to 15, low enough to feed the plant rather than lock nitrogen up, but carbonaceous enough that a real fraction of it stays in the soil rather than leaching away by August.

Feed a rose bed every year and that second pool accumulates. You stop topping up a deficit and start drawing on a reserve, which is a different way to garden. The bed carries its own nitrogen supply, buffered against a cold spring or a wet one, and you are not chasing yellowing leaves with an emergency feed in July. Low maintenance, in the proper sense of the phrase. That is the thing a soluble fertiliser can never do for you, because a soluble fertiliser is gone the moment the plant has not taken it up.

Calcium comes along for cell-wall strength in fast-growing shoots. The protein delivers amino acids. And the whole lot is a meal for soil biology, which is what makes everything else in the bed available in the first place.

The magnesium catch

This is the part I have not seen anyone write about, and it follows directly from Figure 1.

The RHS names roses specifically as a plant prone to magnesium deficiency, showing as yellowing between the veins of older leaves with reddish-brown tints. The RHS also notes that excess potassium reduces magnesium uptake. Alfalfa carries 25.6 g of potassium per kilogram against 2.1 g of magnesium, a ratio of roughly twelve to one.

Feed a rose heavily with alfalfa year after year and you are pushing potassium into a plant that is already vulnerable on magnesium, using a material that supplies almost no magnesium of its own. I am not aware of a study showing this causes deficiency in practice, so treat it as a reasoned caution rather than a documented harm. The fix costs almost nothing. Use kieserite, which is magnesium sulphate in its monohydrate form and dissolves more slowly than Epsom salts, so it feeds over weeks rather than flushing through. Or sidestep the problem and use a balanced fertiliser that already has alfalfa in it, corrected. That is exactly why our rose fertiliser is built on an alfalfa backbone with the magnesium added back.

When and how much to apply in a UK garden

The RHS gives border roses two feeds a year: one in March or April before flowering, one in mid-summer after the first flush for repeat-flowering varieties. Rate is 70 g per square metre, sprinkled on the soil, followed by a mulch kept 10 cm clear of the stems. Note that this is a British cadence. American rose-society schedules assume a feed after every bloom cycle, four to six times a season, so any US alfalfa routine you find online will be pitched higher than a UK garden needs.

Timing deserves more thought than it usually gets, because alfalfa is unusually temperature-sensitive. Work compiled by Perennia on nitrogen mineralisation from organic fertility sources found that the majority of nitrogen release from low carbon-to-nitrogen materials happens in the first three weeks, and that a 10°C drop in soil temperature reduced nitrogen mineralisation from alfalfa pellets by 20%. That is roughly double the sensitivity of composted poultry manure. At least one study in the same review showed net nitrogen immobilisation from an alfalfa meal amendment rather than release.

The practical reading: alfalfa applied to cold soil in early March will sit there. April, once the soil has warmed, gets you a much better return, and the mid-summer application after the first flush is when alfalfa performs best of all.

Rates follow the Dr Forest alfalfa pellet directions and sit within the RHS guideline of 70 g/m² once spread across a bed.

Method matters as much as rate. Scratch it lightly into the soil surface rather than leaving pellets sitting on top, keep it 10 cm clear of the stems and off the graft union, and water it in. Apply to moist soil, never to a bone-dry bed in a heatwave.

Good, but not on its own

Alfalfa gives a rose nitrogen, potassium, calcium and a meal for soil biology. It leaves gaps at phosphorus, magnesium and the trace elements, and the nitrogen it does supply arrives slowly. That is a virtue in July and a nuisance in April, when the bush is pushing new growth and wants something now.

Which is why our rose fertiliser is built the way it is. Alfalfa meal is the backbone of it, for the slow nitrogen, the calcium and the organic matter. The rest of the formulation corrects what alfalfa cannot do alone: magnesium to offset the potassium-to-magnesium imbalance in Figure 1, and faster-acting nitrogen alongside the slow release, so the bush has something to draw on whilst the alfalfa is still breaking down.

You can assemble the same thing from separate bags. Alfalfa in April, a magnesium source with it, and something quicker for the spring push. That works perfectly well, and if you would rather buy the parts separately then do that. The blend saves you weighing three things out twice a year, nothing more mysterious than that. If you want a foliar biostimulant through the flowering period on top, Brix+ covers that with seaweed rather than by hoping there is enough triacontanol in a pellet.

For what alfalfa does in soil beyond roses, including compost activation and worm bins, see the full alfalfa meal guide.

Half a cup of hay on ten rhododendron seedlings in a Michigan garden in 1977 is a thin foundation for a fifty-year habit. The habit turns out to be sound anyway, because alfalfa is a decent slow feed with calcium in it, and roses respond to being fed properly. Scatter it in April, keep it off the stems, and put some magnesium alongside it.