The story of a deceptive killer …
… in which a seemingly-irresistible force meets a few possibly-unstoppable stewards.
It’s a drama. Who will win?
The poisonous killer plant in question is Convalaria majalis. As a deceptively favorite garden plant, it’s called lily-of-the-valley. But stewards of biodiversity call it the death lily, because – under certain conditions in nature – it kills all.
The photo below shows what this monster looks like – toward the edge of the patch, where some diverse vegetation still survives. In the middle of the patch, we found nothing beside death lily and dirt.
But the next photo zooms in to show why this spot is worth caring about:
Above, you see a purple-leaved plant on the Endangered list – cream vetchling or wood pea (Lathyrus ochroleucus). It deserves life, and, indeed, reproduction and recovery of its once-common woodland status. (Notice the huge stipules visible where one compound leaf meets the main stem. These stipules distinguish the wood pea from some related look-alikes.) This plant, which now survives in few other places in Illinois, inspires the work shown in the next photo, of the same place, but after some handiwork.
Here six stalks of wood pea and a bit of other vegetation stand separated from the lilies. Step one - accomplished. Next we step way back, and you can see how big this patch is:
We're looking from the north. The south edge of the patch isn’t quite visible here, but if you look closely, you can make out three of the Restore Moraine team, weeding this menace from rare species on that south edge, at the top of the slope where the patch ends.
These are the heroes yanking out stems that are near the higher quality vegetation. We want to save as much rare vegetation as we can. Stewards from left to right are Eriko, Ryan, Mary, Mills, and Erica.
Then comes step two!
The herbicide glyphosate kills all plants. It’s a last resort for species like this lily that aren’t killed by something less toxic.
Here the heroes are Ali, Jordan, and Steve. Ali and Jordan are spraying death lily. But remember this photo?
Not so for the death lily. This whole lily colony is one plant, roots connected underground. What will happen to the roots in the area where we pulled out the leafy stems – when we spray the nearby intact vegetation? We don’t know. Has anyone else done this? We expect, from experience with other invasives, that the remains of un-herbicided roots will put up new stems near the endangered wood pea and other surviving natural vegetation. We may continue to pull the weakened new shoots, or perhaps we’ll carefully hand-wipe them with herbicide.
In the meantime, we’ll also focus on the rest of our priorities here.
The intrepid Restore Moraine volunteers work in only perhaps 20 acres of the 2,200-acre Moraine Hills State Park. Indeed, within these 20 acres, most of our work has been directed toward a top-priority four acres of rare high-quality oak woodland. Quality areas have shrunken under the stresses of past grazing, invasive species, over-populated deer, and excess shade from lack of fire. The four best quality acres are divided up among five little slopes where much of the ecosystem has somehow miraculously survived and is recovering.
As described by Dr. Wayne Schennum, who studied it for decades, what survives here contains patches of rare high-quality remnant white and bur oak community – not recognized until relatively recently as a high priority for biodiversity conservation. Like the vast prairie, oak woodlands were once a major natural feature of this state. This park also has bogs and fens and sedge meadows and prairie remnants. It has many rare animal populations that also deserve stewardship. We help some with them, but we mostly try to focus on restoring health to this oak woodland.
Dr. Schennum, long advocated for “first aid” rescue work for this woodland, mostly by cutting invasive trees. Rare species had been vanishing in the gloom, as he documented. The diverse animals and plants of oak woodlands thrive in dappled sunlight under oaks that are naturally well-spaced by fire.
Friends of Illinois Nature Preserves agreed with Illinois Department of Natural Resources biologist Melissa Grycan that this site should be a priority. With her guidance and support, starting on June 7, 2025, the resulting Restore Moraine stewards have worked every Saturday morning (and some days in between), year round, to do what oak woodlands are increasingly understood to need.
We hadn't tackled the death lily until last week. We've studied the literature and the Internet and asked expert stewards about their experiences. This menace is most vulnerable to herbicide in spring. No other approach controls it, except repeated hand pulling, which is impractical for large populations. We hope to hear from others about varied approaches. This post will be updated as we learn more.
Endnote 1
At Moraine there have been surprises. We’ve found many new populations of Endangered or rare species. All have been injured by excess shade or invasive species. Hundreds of acres of ailing oak woods here need more care. They need regular fire, invasives control, tree thinning, and seeds – which need to be hunted, gathered, prepped, and broadcast into areas where richness has been lost.
But Moraine isn't the only Preserve with a lack of resources. Most Preserves are suffering. We can't just shift resources here from elsewhere. The Illinois Nature Preserves System needs more staff, contract stewardship, and support for volunteers. But with high-quality woodlands even rarer than prairies, some of us decided that this Nature Preserve was a priority for help by a Friends volunteer community. We're less than a year old here, but off and running. All are invited to come and help, if you're so moved.
Endnote 2
Account of control at Bluff Spring Fen and Poplar Creek by Kirk Garanflo:
Massive infestations of Lily-of-the-Valley (Convalaria majalis) can be eradicated without hand-pulling and without killing everything with herbicide. Ongoing experiments at both Bluff Spring Fen and Carl R. Hansen Woods have been successful at dramatically reducing this pest in small areas.
For large areas such as shown in the photographs above, use a weed whip (or a metal-bladed brush cutter when woody stems are present) to cut the plants AT GROUND LEVEL during the last weeks of April and the first weeks of May when not much else has emerged. Eliminating leaf structure weakens them for the rest of the year; follow-up is essential thereafter during the spring to attack plants that were missed or that emerged after cutting. As long as the plants are cut below that first leaf axil the plants do not usually then leaf out or flower. This process must be repeated for many years (possibly five) in a row without failure in order to exhaust the energy stored in the roots and eventually kill them. Do not skip a year which will allow new leaves to reinvigorate the roots.
It is necessary to expose the plants free of surrounding plant material. Annual burning to remove old dead material is essential in order to find plants that are hidden among tall, withered foliage.
The use of a weed whip or a brush cutter is far more labor effective than hand pulling for an area larger than a quarter acre. Eventually the infestation will be reduced sufficiently to allow effective hand pulling (even just snapping the plants off at ground level works too) of the odd plants that do survive.
Endnote 3
Frank long-term report from Doug Taron, steward of Bluff Spring Fen:I spent all of this morning (along with a bunch of other volunteers) hand pulling at Bluff Spring Fen. We have set it back substantially in some areas, but not eradicated it anywhere. Very dense stands can be hit with a weed whip. I despair of this one. The areas where it has invaded far exceed our ability to control. My hope is that as it declines in areas where we are controlling it, we will able to add new spots that we work on. LOTV is a nightmare.
Experience of Maria Vujic and Stone Hansard at Morton Grove Prairie:
This photo (by Stone) shows where a solid stand of the lily stood last year, indicated by the red line. Where the lily re-emerged this spring is outlined in blue:
It was controlled by multiple sprayings of Glyphosate (4% Aquaneat with Liberate).
This infestation was in an area where the prairie had been shaded out by large cottonwoods (since removed). Prairie seed from nearby was broadcast last fall.
Why did those two patches re-emerge, when this invasive seems completely gone over most of this large area? There are many possibilities. But perfection in one year is not to be expected. The small volunteer crew that assembled to restore this little gem of a Nature Preserve was, at the same time, dealing with massive infestations of sumac, briars, tall goldenrod, buckthorn, and other challenges. We do what we can, and then we follow up. Those blue-outlined patches have been sprayed this spring and will be sprayed again or receive other treatments as needed.
To see what this area looked like in 2025 prior to treatment, see Maria's photo below:
Acknowledgements
Thanks for review, suggestions, and editing to Lana Fedewa, Jonathan Sabbath, and Eriko Kojima.









I have had success cutting purple loosestrife stems and apply glyphosate to the cut stems. Some groups use gooseneck bottles for this type of application. Applying glyphosate to cut stems could also work on Convallaria. This technique would be useful in areas around plants you want to protect. One person can do both jobs. Clippers in one hand and a bottle with herbicide in the other.
ReplyDeleteIn large colonies of Convallaria, with no rare plants, I suggest the following. The plants you want protected can be covered with plastic cups or plastic bags. The Convallaria could then be cut with hedge trimmers, or even a mowed low. A paint roller with glyphosate could then be rolled on top of the cut Convallaria stems. This may have to be done multiple times to apply enough glyphosate to cut stems to get control. The amount of glyphosate applied per area should be monitored to make sure the maximum application rate is not exceeded. If you use this method, you will maximize impact to Convallaria while minimizing impact to non-target species. You should be able to kill a high percentage of Convallaria the first year with this method, if treatment of other invasive species is any example. I would expect 5 or 10 percent coverage the following year that will need followup. Unless so much herbicide is applied that everything is killed, there is always some followup that is needed.
Massive infestations of Lily-of-the-Valley (Convalaria majalis) can be eradicated without hand-pulling and without killing everything with herbicide. Ongoing experiments at both Bluff Spring Fen and Carl R. Hansen Woods have been successful at dramatically reducing this pest in small areas.
ReplyDeleteFor large areas such as shown in the photographs above, use a weed whip (or a metal-bladed brush cutter when woody stems are present) to cut the plants AT GROUND LEVEL during the last weeks of April and the first weeks of May when not much else has emerged. Eliminating leaf structure weakens them for the rest of the year; follow-up is essential thereafter during the spring to attack plants that were missed or that emerged after cutting. As long as the plants are cut below that first leaf axil the plants do not usually then leaf out or flower. This process must be repeated for many years (possibly five) in a row without failure in order to exhaust the energy stored in the roots and eventually kill them. Do not skip a year which will allow new leaves to reinvigorate the roots.
It is necessary to expose the plants free of surrounding plant material. Annual burning to remove old dead material is essential in order to find plants that are hidden among tall, withered foliage.
The use of a weed whip or a brush cutter is far more labor effective than hand pulling for an area larger than a quarter acre. Eventually the infestation will be reduced sufficiently to allow effective hand pulling (even just snapping the plants off at ground level works too) of the odd plants that do survive.
Kirk Garanflo has good points. All you have to do is look at where Convalaria majalis grows in peoples yards to see the wisdom in the above. Convalaria majalis does not invade far into mowed lawns.
DeleteWonderful and thank you. Many still believe lily-of-the-valley only spreads clonally from places where it was once planted. Nope. It produces fruits and viable seeds, and these day new, small colonies are gaining steam in woodlands throughout the midwest. Best to hunt them down and dispatch them while they cover several square feet and not vast areas like this. But even what seem like insurmountable challenges like this are worthwhile and surmountable, if we try.
ReplyDeleteYes. Good comments. Yes, indeed, it spreads by seed. Our top priority is to control the many little patches that are nearest the highest quality areas.
DeleteHere is a video on controlling Japenese knotweed that might be helpful. The applicator found that applying herbicide to cut stems, or with stem injection, killed the Japanese knotweed up to eight feet away when this was done during a drought. Since lily of the valley is also a clonal species, treating lily of the valley during a drought might allow the herbicide to spread through the clone. Herbicide spreading through a clone, killing large parts of it, would be very helpful for those doing control efforts.
ReplyDeletehttps://www.youtube.com/watch?v=O8hiBGQDKRM
I wish this video included the chemical used. As editor of this blog, I delete many comments that seem more like advertisements or trolling or obviously wrong. Some of the comments above are unhelpful, if they're based on herbicides that poison the soil or otherwise kill the plants we're trying to protect. Sometimes it may be necessary to kill all in an area and start over, but that's a last resort in a remnant ecosystem. Turning an area into a mowed lawn for long enough to kill a problem plant is also likely to kill the most important parts of the ecosystem. I also go back and forth about leaving on the post comments from armchair theorists. Such brainstorming may lead to a helpful new approach. But too much of that just wastes people's time and makes the post less valuable. So, I do my editing best.
DeleteGreen Shoots' response to my inquiry was that 3 ml of undiluted, 54% aquatic glyphosate concentrate was used for injection. This would simply be another tool in an arsenal of techniques to be used against just another invasive pest.
DeleteIn other videos on Youtube, by Green Shoots, on the topic of stem injection into Japanese knotweed, the herbicide used is given. The herbicide used is one of the various concentrated forms of glyphosate. Amounts are given in the videos, but I will not mention that here because it does not apply to controlling lily of the valley.
ReplyDeleteAs for mowing, those working in invasive species control do this all the time. Mowing is done for sweet clover, wild parsnip, woody species control, etc. Mowing can be done selectively leaving flagged locations unmowed.
I recall you previously mentioning that creamy vetching disappeared from all the areas of a woodland except adjacent to where mowing was occurring. You said the mowing was being done so classes could have a place to visit. If I remember correctly, you said this was at Reed-Turner Woods.
Also, I have personally seen glyphosate applications to a stem impact or kill other stems nearby. This is true with clonal species, but also occurs rarely to untreated non-target species. Impacts to nearby untreated non-target species occur so rarely, that I do not know the reason. Drought is a possibility. In other cases it could be more herbicide was applied than the minimum amount necessary for control.
ReplyDelete