Best Mole Killer: Why Most Mole Poisons Cannot Work

Search “best mole killer” and the results will sell you poison peanuts, granular bait, ultrasonic stakes and grub control.

Extension entomologists have a different view of that shelf, and Purdue’s publication on the subject is titled — by them, not by us — Mole Control: Buyer Beware. The core of their objection is a fact of taxonomy that decides the whole purchase: a mole is not a rodent. It does not eat seeds, so it cannot be poisoned by seed.

First: Do You Even Have Moles?

Three states change this answer outright, and none of the product listings will mention it. Check yourself against these before spending anything.

And across the whole South there is a name collision worth clearing first. A mole cricket is an insect — about an inch long, shovel-like front legs — that tunnels just under the surface and eats roots. It shares half a name with the mammal and nothing else: nothing on this page affects it, and a mole trap will never catch one. The tell is the tunnel. A mole pushes up conical heaps of loose soil and firm raised runs; a mole cricket leaves soft winding ridges a finger wide, with granular soil at the burrow mouth, over turf that is thinning because the roots are gone. If that is what you have, mole crickets in a lawn is the page — and the treatment is timed to June rather than to the damage.

Arizona — you do not have moles. University of Arizona Cooperative Extension’s pocket gopher publication says it in five words: “We do not have moles in Arizona.” If something is pushing up soil in an Arizona yard it is almost certainly a pocket gopher, which is a rodent, does eat plant material, and can be baited — the opposite of everything below. Buying mole bait here is buying for an animal that is not present.

Colorado — almost certainly not moles either. CSU Extension’s burrowing-animal fact sheet is unusually precise: moles are “rarely found in Colorado, and only on the eastern plains” — specifically the north-eastern plains along the Lower South Platte (Logan, Sedgwick, Phillips, Morgan, Washington and Yuma counties) and the Cimarron watershed in the south-east (Baca county). CSU’s own conclusion: “If you are outside that area, the species causing damage is likely one of the other four” — voles, pocket gophers, prairie dogs or Wyoming ground squirrels. Denver, the Front Range and the Western Slope are all outside mole range.

Washington — you have moles, and the standard trap is illegal to use. See the legal exception below.

Everywhere else we cover — Ohio, Pennsylvania, Illinois, Michigan, Indiana, Wisconsin, Minnesota, New York, New Jersey, Virginia, the Carolinas, Georgia, Tennessee, Florida, Louisiana, eastern Texas — the eastern mole is the animal in question and the rest of this guide applies as written. California has its own species, the broad-footed mole, and the same methods.

A Mole Is Not a Rodent

This single sentence invalidates most of the category, so it is worth stating before anything else.

If the animal turns out to be a vole rather than a mole, the products change completely, because a vole is a rodent: a refillable bait station holds rodent bait a vole will actually eat, and a live-capture cage trap is an option where lethal control is unwanted. Neither does anything to a mole. Identify the animal with the table further down before buying either.

Rats, mice and voles are rodents — they eat seeds and grain, which is why grain-based bait works on them. A mole is an insectivore. Purdue’s Timothy Gibb, a turfgrass IPM specialist, puts the consequence in writing:

Moles are not rodents (like rats and mice) which [means] that they cannot be baited using rodent foods. Poison peanuts or other grain baits won’t work since moles don’t feed on seeds, alfalfa pellets or any of the typical baits that are sold to ‘kill rodents’ even though some are touted as a control for ‘rodents and moles’.

Read the last clause again, because it describes a product you can buy today. Packages that say “rodents and moles” are selling one bait to two animals with different diets. It will work on the rodent. The mole will walk past it.

What a mole is actually built for:

• Diet: mostly earthworms. Purdue is explicit that the majority of the diet is “the ever-present earthworm”. WSU adds that moles eat earthworms “with occasional grubs or larvae”.
• Nearly blind, with front feet designed for digging, 6–8 inches long.
• A metabolism that will not let it rest. Purdue describes “abnormally high metabolism rates which means that they must constantly eat and are active almost year-round, day and night” — and a tunnelling capacity of as much as 100 feet per day.

That last figure explains why a single mole convinces people they have an infestation. One animal working a hundred feet a day across a suburban lawn produces a great deal of visible ridge.

Grub Control Is Not Mole Control

The second most expensive misunderstanding, and the one garden centres actively encourage by shelving grub products near mole products.

Purdue, verbatim:

Moles mostly feed on earthworms. While they do eat grubs, it’s an old wives tale that grubs are the reason that moles are in a lawn. Therefore using grub control products as a method of controlling moles will not be effective. Even in grub free lawns, moles continue to survive, because the majority of their diet consists of the ever-present earthworm.

We will give you the one dissenting shade honestly, because it exists. NC State’s turf publication, by Rick Brandenburg, allows that reducing the food supply by killing grubs “may help to a small degree.” That is a weaker claim than the marketing makes, and it is not in conflict with Purdue on the decisive point: no extension service presents grub control as a way to remove moles. The mole’s staple food is not the thing the insecticide kills.

The practical upshot: if you have grubs, treat them because you have grubs — see lawn grubs for how to confirm that and what the thresholds are. Do not buy grub control expecting the tunnels to stop.

The List of Things That Do Not Work

Extension services rarely speak in unison. On mole gadgets they do.

Purdue, on the home remedies and devices sold or passed around:

These include putting mothballs, human hair, razor blades, or chewing gum in their tunnels, or using pinwheels or ultrasonic devices to scare moles away. The reality is that these just do not work.

Washington State University Extension, on the electronic category specifically: “In controlled tests, none of the electronic repellers have been effective.” WSU’s ineffective list also names mothballs, vibrating windmills and human hair.

NC State, on the whole genre, including so-called mole plants, for which it says there is no known supporting research — and then the line worth taping to the shelf: “There are no short cuts or magic wands when controlling moles.”

The electronic category has enough going on to deserve its own page, including why the EPA number on the box is not the approval it looks like: do ultrasonic pest repellers work.

So the products to skip, with reasons rather than vibes:

Those links go to our own reviews of the specific best-sellers in each row. They are there so you can check the rating and the reasoning before buying one, not because we recommend them — every product linked in that table is rated 2.0 or below on this site.

What Extension Services Actually Recommend

Two things, and the first one is not a chemical.

Trapping

This is the method the publications put first. NC State notes that there are several mole traps on the market and that “each, if properly handled, will give good results” — the types generally sold are harpoon, scissor-jawed, and choker or pitfall designs.

The detail that decides whether trapping works is not which trap you buy. It is setting it on an active tunnel. A lawn accumulates abandoned runs, and a trap on an abandoned run catches nothing for weeks and reads as a failed product. The standard field test is to press a short section of ridge flat, mark it, and check in a day or two: if the ridge is pushed back up, the run is in use. Set there.

Timing helps too. Purdue notes mole damage is most obvious in spring and autumn, and that spring may be the most effective time to control them.

Buy the set rather than the bare trap. The price difference is under ten dollars and the probe is the part that stops you setting on a dead run.

The one bait that is built for the right animal

There is a bait category that works, and it works precisely because it stops pretending a mole is a rodent: baits moulded and scented as artificial earthworms, carrying bromethalin as the active ingredient. The delivery form is the entire innovation — it presents as the food the animal actually hunts, placed down into an active tunnel rather than scattered on the surface.

All three of these are the same idea executed by different manufacturers. The format is the product — an artificial earthworm placed down into a run. Judge them on availability and price rather than on marketing, and note that none of them is a scatter-on-the-lawn product.

One extension field test found the moles would not eat it

We recommend this format, so we are going to publish the strongest published evidence against it rather than leave it out. WSU Extension’s Dave Pehling ran a limited field test of Talpirid in Snohomish County, staking the baits in daily-use runs so he could recover and inspect them:

“In every instance I could find no sign of feeding, though the baits were consistently walked over, buried, pressed into the tunnel walls or just pushed out of the run. On two of the baits, I tried to enhance the attractiveness by smearing earthworms on them but they were still pushed aside and buried by the mole. In the end, all of the baits slowly dissolved away into the soil and the mole continues to inhabit the runs.”

He also records what the manufacturer told him: “The information included said that the bait had been thoroughly tested on our western moles and was proven effective.” So the refusal happened on exactly the animal the efficacy claim named.

Three limits on that finding, stated plainly. It is his word — a limited test, one site, one mole, so it is not a trial. It is on a western mole, genus Scapanus, and most of the country has the eastern mole, Scalopus aquaticus — a different animal whose feeding behaviour may differ. And it dates from 2007.

What it does justify is a change in expectation rather than a change in product. Bait is not a certainty here; it is an attempt whose success you have to verify by probing the run afterwards, the same way you verify a trap set. If the worms are still there a week later, buried or pushed aside, that is your answer and the money is better spent on a trap — where the law allows one.

Before you reach for any of them, read the next section. Then check your own state’s rules, because this is a vertebrate pesticide and the rules are not uniform.

And the same extension office field-tested a castor-oil repellent

The repellent shelf is the other half of this category, and we rate those products low. Here is why, from the same WSU field work — a granular castor-oil product with a Fuller’s earth binder, applied repeatedly to the active parts of a mole’s territory through February:

“Throughout the test there was no change in activity. The mole kept returning to the application holes and plugging them up! Seems to me that a repellant should repel them????”

Note the detail that makes it damning rather than merely negative: the mole did not avoid the treated ground, it came back to the treated holes and closed them. The same caveats apply — limited test, western species — but it lines up with what the customer ratings on these products already say, and it is the reason the castor-oil granules in our catalogue carry the ratings they do.

Washington is a legal exception, and it is a big one

If you are in Washington, the method every other publication recommends first is restricted where you live. WSU Extension records that Initiative 713, effective December 7, 2000, bans the use of any trap other than “common rat and mouse traps” that “holds, pierces, grasps or pinches a body part in any way” — which describes the body-gripping mole traps sold freely elsewhere. WSU notes the odd consequence: you may legally buy such a trap, while using it to capture an animal remains prohibited. It cites RCW 77.15.194 for the detail.

And the ban’s application to mole traps specifically is not our reading of it — it was tested in the legislature and survived. The same WSU newsletter records two 2007 bills that would have carved mole traps out. H.B. 1400 “would have excluded common mole and gopher traps from the definition of ‘body-gripping trap’” and never left the House Committee on Agriculture & Natural Resources. S.B. 5722 was similar, reached a second reading, and died in the rules committee after an amendment attempt that would have repealed I-713 outright. Pehling’s own conclusion at the time:

“Without a great deal more public pressure, I’m afraid we are going to be stuck with the trapping ban for many years to come.”

Two attempts to exempt the exact hardware this page recommends first, both failed. That is why Washington gets its own section rather than a footnote.

Check the current statute and your county before buying hardware. That is a 2000 ballot initiative and two 2007 bills as summarised by an extension office, not legal advice from us, and it is the kind of rule that gets amended.

The Part That Matters If You Have A Dog

Bromethalin is the active ingredient in the mole baits that work. It is also the reason to think carefully before choosing bait over a trap in a household with a dog that digs.

NPIC — the National Pesticide Information Center, run by Oregon State University with EPA — describes the mechanism without euphemism: “It stops the cells in the central nervous system from producing energy. The nerve cells swell, this puts pressure on the brain, and paralysis and death soon follow.”

And the sentence that removes any comfort from the “it’s for pests” framing:

Rodents, humans, dogs and cats are all mammals, so our bodies work in very similar ways. Rodenticides have the same effect when eaten by any mammal.

Two further points from the same fact sheet make this harder to manage than an insecticide:

• Onset can be slow and deceptive. “If a lethal dose is eaten, signs may develop 8 to 12 hours or several days after ingestion and progress over a period of a week or longer.” A dog that seems fine at bedtime is not evidence that nothing was eaten.
• Storage counts as exposure. NPIC’s instruction is to keep rodenticides out of reach of children and pets “whether they are in use or in storage” — the unopened box in the garage is part of the risk, not outside it.

We are not telling you never to use it. We are saying the honest comparison is this: a trap has a physical hazard you can see and site deliberately; a bait has a chemical hazard on a delayed timer. In a yard with a digging dog, that asymmetry usually favours the trap. If you do bait, place it into the tunnel, never on the surface, and treat any suspected ingestion as an emergency rather than something to watch overnight.

The same logic that governs lawn chemicals around animals applies here, and we have written it up in full for herbicides in pet safe weed killer.

Make Sure It Is Actually A Mole

If the table below points at a vole rather than a mole, voles in the lawn is the page you want — the runways, the girdling risk to young trees, and why the bait aisle does not have an answer for them either.

Worth thirty seconds, because the answer changes the product entirely.

The mound shape is the most reliable single tell, and the reason is mechanical. University of Arizona Extension explains it: a pocket gopher approaches the surface at an angle, so the soil lands in a fan or tear-drop with the plug to one side, while a mole pushes soil straight up, so it falls evenly and the mound is round and conical.

WSU’s line settles the most common misdiagnosis: moles “rarely, if ever, eat bulbs or vegetables or ornamentals.” So if bulbs are vanishing and hostas are being chewed at the crown, you are probably not dealing with the animal that made the ridges — and the grain bait that would not touch a mole is the right tool for the one you do have.

One inconsistency worth flagging, because you will hit it if you read further. CSU’s fact sheet states that moles “do not leave mounds of soil.” Taken as a national claim that is too strong: Purdue documents mole soil piles explicitly — “large piles of black soil deposited on top of the lawn or garden” from winter deep tunnelling — and Arizona Extension describes the shape of a mole mound in order to distinguish it from a gopher’s. Read in context, CSU is giving Coloradans a practical shortcut, and in Colorado it is a good one: moles are nearly absent there, so a mound almost always means gopher. Outside that state, treat ridges as the reliable mole sign and mound shape as the discriminator.

One last thing that reframes the whole problem. Purdue points out that when soil dries in summer or freezes in winter, earthworms move deeper and the moles follow them down, so surface tunnelling stops. That silence is not success. The large piles of dark soil that appear on a lawn over winter are the evidence that the animal has been working the whole time, deeper down, and will be back at the surface when the ground warms. Judge a control method by whether the runs stay flat in spring, not by a quiet August.

For the rest of what pushes up, chews or discolours turf, lawn fungus identification covers the diseases and lawn grubs covers the insect that genuinely does eat roots.