How to Kill Creeping Charlie and Wild Violet

Most weeds on a lawn die from the bag you already own. These two do not, and people usually conclude the product was defective or the weeds are simply invincible.

Neither is true. Creeping Charlie and wild violet defeat a standard three-way broadleaf herbicide for two different reasons — and once you know which reason applies, the fix is the same active ingredient and the same month.

Two weeds, two different failures

They get grouped together as “stubborn broadleaf weeds”, which hides the more useful fact: they are stubborn in opposite ways.

Wild violet blocks entry. Penn State’s explanation is that poor control comes from a thick, waxy cuticle that interferes with herbicide uptake and its strong recuperative potential due to robust underground rhizomes. Those rhizomes are described as thick and branching, giving rise to new plants and serving as food storage and overwintering structures. So a spray lands, sits on a glossy surface, and much of it never gets into the plant — and whatever damage does occur is funded back out of an underground reserve.

Creeping Charlie blocks delivery. Here the herbicide gets in. Penn State’s account of poor control is poor or marginal translocation of active ingredients through its extensive stolon network to crown tissues and its strong recuperative potential. Ground ivy runs across the surface on stolons that root at every node, and Iowa State describes it as spreading by seed and by the vining stems which root at their nodes. Each rooted node is effectively an independent plant. Scorch the foliage and you have killed leaves belonging to dozens of crowns that were never reached.

That distinction is why “it grew back” means something different for each. On violet, regrowth comes up from the rhizome in the same spot. On creeping Charlie, the patch refills from nodes at its own edges.

Both share one consequence, and it is the expectation-setting part of this article: Penn State says repeat application over the course of the growing season and over multiple years may be needed for effective control — for both weeds. A single autumn is a start, not a finish.

Telling them apart

They often grow in the same thin, shaded ground, so it is worth being sure which you are treating — not because the chemistry changes, but because the timing and your expectations do.

The minty smell is the fastest field check there is: crush a leaf between your fingers. Iowa State notes ground ivy produces a minty odor when cut or crushed, and violet produces nothing.

Where it grows is a second clue. Iowa State’s note is that ground ivy thrives in damp, shady areas but also grows well in sunny locations — so shade points toward it without ruling anything out. If shade is the underlying issue in that part of the yard, our growing grass in shade guide covers whether grass is even the right answer there.

For sorting any other weed by family first, our weed identification guide has the sixty-second version.

The look-alike that is not a perennial

Henbit gets mistaken for creeping Charlie, and the mistake matters because it changes the entire plan.

Henbit is a winter annual: it germinates in autumn, grows through winter, flowers and seeds in spring, then dies out as heat arrives. Creeping Charlie is a perennial that persists from the same crowns year after year.

The practical test is to do nothing and watch. A patch that disappears on its own by early summer was henbit, and spraying it in May was money spent on a plant that was already leaving. A patch still there in July is creeping Charlie.

Which also means henbit belongs to a different management group entirely — the autumn-germinating winter annuals covered in our Poa annua guide, where a pre-emergent barrier applied before germination is the efficient answer and a post-emergent spray is the expensive one. Our fall lawn care guide places that job in the season.

The active ingredient to buy for

For both weeds on this page, the answer is triclopyr — not the three-way that settles dandelion and clover in one pass.

Penn State on wild violet: herbicides containing triclopyr provide the best control. Penn State on ground ivy: triclopyr alone, fluroxypyr alone, or combinations of triclopyr and fluroxypyr typically provide the best control. Iowa State’s list for ground ivy names triclopyr and 2,4-D as most effective, with MCPP and dicamba as alternatives.

Note what that means about the bag most people already have. A standard three-way product — 2,4-D, MCPP and dicamba — is not useless here, and Iowa State includes 2,4-D among the effective options for ground ivy. But triclopyr is the ingredient the sources reach for first, and it is the one usually missing from the general-purpose bottle. Our best weed killer guide covers matching chemistry to weed across the whole range.

Two triclopyr products here fit these weeds, and which one you buy is decided by your grass rather than by your weed.

Check the label against your grass first. It excludes bahiagrass, bermudagrass, centipedegrass, kikuyugrass and St. Augustinegrass — most Southern lawns. Safe on buffalograss, fescues, Kentucky bluegrass, ryegrass and zoysiagrass — though on buffalograss it is worth asking whether you need it at all, since K-State treats a weedy stand there as a sign of overwatering rather than a spraying job.

We had that second one filed wrong until 2026-08-02, so here is the correction. Its listing sells it as a brush killer for non-crop areas, and we repeated that. Its label — Reg. No. 81927-11-7401 — lists one site: perennial bluegrass, perennial ryegrass and tall fescue, with ground ivy and wild violet both on its weed table. It is an ester rather than an amine salt, which penetrates better and moves more readily as vapour, so the label’s temperature guidance is the part that matters. On any other grass species, it says injury must be tolerable.

One thing the sources rule out: pulling. Iowa State says that in lawns, pulling is often ineffective because rarely are all stem and root pieces removed, and Penn State says hand removal of wild violet is not effective unless rhizomes are excavated from the soil. Interestingly, Iowa State flips that advice for garden beds, where hand pulling and digging is described as the most effective approach given persistence. In turf it simply propagates the problem.

The natural remedies, honestly

Search for a chemical-free answer to creeping charlie and one recipe dominates: borax. It deserves a straight answer, because it is not internet folklore — it started with Iowa State research, and Iowa State is now the source telling you not to use it.

Why it is tempting. Wisconsin explains the mechanism, and it is real: borax contains boron, “which is necessary in very small amounts for plant growth but is toxic in larger doses,” and “Creeping Charlie happens to be extremely sensitive to boron.” So the theory works. A dose that harms creeping charlie more than it harms grass genuinely exists.

Why that is not enough. Iowa State’s own verdict on the research that started it:

“Because the Iowa State study has not been replicated and so many factors influence its level of effectiveness, the use of borax as weed control for creeping Charlie is not recommended.”

They describe a follow-up at Wisconsin “conducted in much the same way” that “did not have the same results,” and name why: “Too many factors seem to influence how borax works in the soil, including soil pH and soil type.” Their line on the margin for error is the one to remember — with boron, “the difference between plant requirements and toxicity problems is very small.”

And the part that makes a mistake permanent. Wisconsin:

“…boron doesn’t break down or dissipate in the soil, so repeated or excessive applications can result in bare areas where no vegetation can grow.”

And the two variables Iowa State names as the reason results were never repeatable — soil pH and soil type — are exactly the two that differ most between one reader and the next. North Carolina's are on record: Cecil, named the state soil by the legislature in 1997, and NC State classifies it as "a well-drained mineral soil" — which is the opposite of what most Piedmont homeowners experience, and the discrepancy is the whole story. Cecil is a sandy loam surface OVER a red clay subsoil. Two centuries of cotton and tobacco took the surface off most upland sites, so what a Charlotte or Greensboro lawn is actually rooted in is the clay B horizon that used to be two feet down. East of the fall line the coastal plain is genuinely sandy and holds neither water nor nutrients for long, with pH running Acidic (5.5–6.0) — regular lime applications are standard practice. Boron availability moves with both, so a borax measurement someone posted from a different state was calibrated against different ground. There is no dose to copy.

A herbicide that fails costs you a season. Boron that goes on too heavily does not wash out, degrade or expire. That is the difference between a bad afternoon and a strip of ground that grows nothing, and it is why “natural” is a poor guide to whether something is safe to put on soil.

What Wisconsin suggests instead, for someone who wants an alternative to conventional herbicide chemistry:

“Instead of Borax, use an herbicide containing Fe-HEDTA as an alternative to conventional herbicidal control.”

Iron HEDTA is a contact material rather than a systemic one, so expect it to burn top growth and expect regrowth from the stem network below — Wisconsin notes sequential applications are part of using it. It is the honest version of a natural option: weaker than triclopyr, genuinely lower-impact, and not a single-pass fix.

And the option nobody sells. Wisconsin’s other suggestion is to stop fighting for turf where turf keeps losing: grow “the most suitable type of turfgrass for the area (shade tolerant varieties under trees)”, or “consider removing grass and instead grow shade-loving groundcovers or hostas that compete well with weeds.” Creeping charlie thrives in the shade your grass is failing in — growing grass in shade covers where that fight is winnable and where it is not.

Which shade-tolerant grass, in North Carolina. Wisconsin's advice is species-specific and your state already has an answer on record: Zoysia is the shade and low-input component grown here alongside Tall Fescue. Overseeding the shaded strip with that rather than with more Tall Fescue is the difference between re-fighting this every autumn and not. Before reseeding, settle whether the problem is shade or compaction: a test through NCDA&CS Agronomic Services Division will tell you.

Why October beats May

Most people spray these weeds in spring, when the purple flowers make them obvious. That is the wrong month, and the reason is worth understanding rather than memorising.

Penn State: herbicide application in the fall following the first frost allows better translocation of active ingredients to crown tissues when compared to spring applications. In autumn the plant is moving sugars down into crowns and rhizomes to overwinter, and a systemic herbicide travels with that flow. In spring the flow runs the other way — up into new growth you are about to mow off.

Iowa State gives the calendar version: fall, late September through early November, is the best time to control ground ivy, and spring applications are not as effective. It also specifies two applications — one in late September or early October, and a second about a month later.

Penn State's trigger — after the first frost — has a local date: in North Carolina first frost lands around Late October (mountains) to mid-November (coast), which is where Iowa State's late-September-to-early-November window anchors itself on your calendar. The state's post-emergent window reads: April–May and September–October; avoid spraying heat-stressed fescue in July–August — and North Carolina's own fall maintenance calendar already schedules the pass this article is arguing for: Apply fall pre-emergent on warm-season lawns for winter weeds. Work backwards from the frost date: the second of Iowa State's two applications still needs actively growing weeds, so starting in late September beats starting in late October nearly everywhere we cover.

What that means specifically for North Carolina. First frost here lands around Late October (mountains) to mid-November (coast), which puts your first application comfortably inside the late-September-to-early-November window and leaves room for the second a month behind it.

Wild violet is on North Carolina's high-risk list while ground ivy is not — worth confirming which one you have before buying for the wrong target.

Check this before you buy. MSMA is banned for residential turf nationwide — dallisgrass control requires alternative herbicides A triclopyr product is the recommendation in this article, and in North Carolina that recommendation is only useful where the rule above permits it — confirm with TurfFiles or your municipality rather than with the shelf.

So the programme is:

1. First application after the first frost, once the plant has been signalled to move resources down.
2. Second application about a month later, before the ground freezes.
3. Repeat next autumn. Both Penn State pages say multiple years may be needed.

This is the same principle behind autumn broadleaf treatment generally — our fall lawn care guide covers why perennial weeds are most vulnerable when they are least visible. Creeping Charlie and wild violet are simply the clearest case for it.

Two constraints from elsewhere on this site still apply. Do not spray in summer heat: the temperature limits that make herbicide fail above 90°F are not suspended for stubborn weeds. And if you are also overseeding this autumn, check the label interval before you seed — most broadleaf products require a wait, which is a scheduling conflict our overseeding guide covers.

North Carolina answers with drainage rather than a herbicide

NC State keeps a TurfFiles profile for each of these two weeds and neither one names a chemical. For ground ivy the entire control recommendation is that “improving surface drainage, aerating when needed, and watering infrequently will help reduce ground ivy encroachment.” For wild violet it is that “proper turf maintenance is the key to control of this weed. First, select adapted turfgrass cultivars for your area and then properly fertilize, mow, and water to encourage dense growth.” Its homeowner lawn publication, Carolina Lawns, discusses broadleaf herbicides only in general terms and does not name either weed either. Read as guidance rather than as an omission, that is a position: in North Carolina both of these weeds are usually living in the wet, shaded strip that turf never manages to fill, and the state's own programme is telling you the shade and the drainage are the problem while the spray is the thing that keeps you busy. The chemistry above still works here. North Carolina is telling you what it will not fix.

Source: NC State Extension TurfFiles, Ground Ivy and Wild Violet weed profiles, and Carolina Lawns.

Your state and these two

Wild violet yes, creeping Charlie no. The local high-risk list carries Wild violet but not ground ivy, which shifts the plan: violet's resistance is the waxy-cuticle-and-rhizome kind, so triclopyr and repeat applications matter, but the stolon-network problem that makes creeping Charlie so frustrating is not what you are fighting. The list does include White clover, and that one is genuinely easier — Iowa State's alternatives of MCPP and dicamba, both in an ordinary three-way product, handle clover in a way they do not handle violet. Do not buy triclopyr for the clover; buy it for the violet.

Your actual spray window in North Carolina

Penn State's trigger is the first frost, so this is one of the few jobs where the date genuinely comes from your own climate rather than a calendar. Locally: Late October (mountains) to mid-November (coast) Note that this state reports a range rather than a single date, and the spread is the point — the same October week that is correct at one end of the state is early or late at the other. Take your own local frost date, not the state average, and count the second application about a month after it.

And read it against the local post-emergent guidance: April–May and September–October; avoid spraying heat-stressed fescue in July–August Where the local window lists a spring option alongside the autumn one, treat the autumn half as the real one for these two weeds specifically — the spring slot is for the easier annuals.

Worth setting against what North Carolina has you doing in spring: Apply pre-emergent by mid-March (Piedmont) Read it as the annual-weed programme. It runs on a different clock from the two perennials on this page, which answer to the first frost rather than to spring soil.

One more reason to treat the October pass as the season's main event here: 2 other weeds on North Carolina's high-risk list answer to the same broadleaf chemistry — Henbit, White clover. Those are the easy ones; a standard three-way product handles them, and it is only the two on this page that need triclopyr in the tank. Buying the four-way once therefore clears the whole broadleaf list rather than just the stubborn pair. What it will not touch is Crabgrass or Annual bluegrass (Poa annua) or Dallisgrass — grassy weeds and sedges need entirely different chemistry, and our identification guide covers why.

One local condition worth checking, because ground ivy in particular favours damp ground: soil here is Cecil, named the state soil by the legislature in 1997, and NC State classifies it as "a well-drained mineral soil" — which is the opposite of what most Piedmont homeowners experience, and the discrepancy is the whole story. Cecil is a sandy loam surface OVER a red clay subsoil. Two centuries of cotton and tobacco took the surface off most upland sites, so what a Charlotte or Greensboro lawn is actually rooted in is the clay B horizon that used to be two feet down. East of the fall line the coastal plain is genuinely sandy and holds neither water nor nutrients for long; the mountain soils are shallower and stonier again., draining Depends on whether your topsoil is still there. On an intact Cecil profile the classification is right and it drains well. On an eroded upland site — and NC State records that most Piedmont uplands are at least somewhat eroded, with texture consequently dominated by clay and silt rather than sand — you are on the subsoil, and it sheds water, compacts under equipment and holds it in the low corners. Add a graded new-build lot and the two effects stack. Coastal plain sand drains fast enough that leaching, not runoff, is the problem; test rather than assume, because the two halves of this state need opposite advice.. Slow-draining ground that stays damp is exactly the habitat Iowa State describes this weed thriving in, which means chemistry alone will keep meeting fresh conditions for reinfestation. Relieving compaction through core aeration is part of the weed plan here, not a separate chore.

What has to close the gap in North Carolina

Both of these weeds take ground that the grass has already given up, so the second half of the job is whether your turf can take it back. North Carolina lawns lean Tall Fescue, which is bunch-type: killing the colony leaves bare ground that stays bare until seed goes into it. Plan the reseeding as part of the treatment rather than as a separate job next year.

The feeding that thickens that turf has a local window: Fescue feeds in fall: September, October and late November — never in summer. Getting that right does more against both weeds than a second herbicide pass, because it removes the opening they need — and on a bunch-type lawn it is the difference between a repair and a permanent thin patch.

The honest summary

Creeping Charlie and wild violet are not invincible. They are simply the two weeds where the default purchase and the default month are both wrong.

Buy for triclopyr rather than for a brand or a weed count on the front of the bottle. Spray after your first frost, not when the flowers make the weeds obvious in spring, because that is when the plant carries the herbicide down into the crowns and rhizomes that actually need to die. Do it twice, about a month apart. And expect to do it again next autumn — both Penn State pages say plainly that multiple years may be needed.

Do not pull them out of turf; you are dividing them. Do improve the density and light in that part of the lawn, understanding what Penn State says about the limit of that: it slows the spread, and it will not suppress an established colony on its own.

And check first that you have a perennial at all. If the patch clears up by itself in early summer, it was henbit, and the right answer was a barrier last autumn rather than a spray this spring.

On brush the ester is the right form instead — brush and woody plant control.