One Rose Bush Took 1,500 Ladybugs.

Fifteen hundred ladybugs is not much to look at. It is about a cupful. Tipped out over an aphid-covered rose it looks like a crowd, and for a day it is one.

That cupful is also, near enough, what one heavily infested rose bush takes. Not a hedge. One bush, and then again a week or two later.

The number is much larger than it looks like it should be, and it is not printed on most products, including ours. What follows is where it comes from, what it is on a potted plant rather than a shrub, and why a light sprinkle over the whole garden reliably does nothing you can measure.

Where the number comes from

There is one peer-reviewed trial of bought ladybugs released onto an ornamental plant in a real garden. Mary Louise Flint and Steve Dreistadt released convergent lady beetles — Hippodamia convergens, the species sold in tubs across North America — onto aphid-infested roses across three seasons, at rates from a light scatter to an inundation, and counted the aphids afterward.

Their finding, in their own words, is that the rate that worked was “about 2300 beetles/m² (210/ft²) of shrub-covered surface, or two orders of magnitude greater than the 11–22 beetles/m² (1–2/ft²) commonly recommended by beetle sellers.”

Two orders of magnitude is a hundredfold. That is not a rounding error or a difference of opinion about thoroughness; it is a different order of purchase. Here is what each rate did:

University of California’s own guidance arrives at the same place from the other direction: controlling aphids on roses, it says, “can require about 1,500 beetles per shrub released at 1- to 2-week intervals.” Their Pest Note on aphids repeats the figure for a single heavily infested bush.

Where the evidence argues with itself

A 2017 review pulled together ladybird releases across a couple of dozen greenhouse and glasshouse studies and found something that sits awkwardly beside everything above: “aphid population reduction did not correlate with ladybird (normal strain) release rate.” Across that whole set of studies, how many were released did not predict how well it went.

Both findings are real and they are not in conflict. Flint and Dreistadt varied the rate inside one system and got a clean dose response. The review compared different systems. Across those, crop, temperature, containment, prey density and ladybird species all moved at once, and the rate stops being the variable that explains the outcome.

The same review notes that releases of overwintered, field-collected H. convergens adults “at extremely high release rates, did provide some level of aphid control.” That is the finding above, stated more cautiously.

So: a high rate is necessary and it is not sufficient. Getting the number right is the part you control, and it will not rescue a release that is wrong in one of the other ways, which is most of the rest of this article.

The number for a plant in a pot

The shrub figure establishes the scale, but most people reading this do not have a rose hedge. They have a plant in a pot, and the trial has a number for that too, which is the most transferable thing in the paper: 100 beetles on a 5-gallon potted rose gave 66 to 88 percent control from one release.

One hundred. On one pot. It is worth sitting with that for a second, because it reframes what a tub is. A 1,500-count tub is not a garden’s worth of ladybugs. It is about fifteen potted plants’ worth, or one badly infested shrub’s worth, and the difference between those two readings is the difference between a release that does something and one that does not.

Scaling down from there is arithmetic rather than evidence, and it should be labeled as such. A 1-gallon houseplant is not a fifth of a 5-gallon rose in leaf area, and nobody has run the trial. But the order of magnitude is the useful part: tens of beetles per plant, not two or three.

Why a light sprinkle does nothing

The reason the rate has to be so high is not that ladybugs eat slowly. An adult eats about a hundred aphids a day, which is a great deal of aphid. The reason is that it does not stay.

UC IPM puts it about as plainly as a university publication can: “Most will fly away from the release site within 1 or 2 days after they are released even if aphids are plentiful.” Flint and Dreistadt found the same thing — beetles leave on average one to two days after release, aphids present or not.

That last clause is the load-bearing one. The intuitive explanation for a vanishing release is that there was nothing to eat, or that the plant was wrong, or that the beetles were weak on arrival. The trial rules all three out: they leave anyway. Extension guidance from UC ANR puts roughly 95 percent of them gone inside 48 hours, with the remainder following over four or five days.

So what you buy, with an adult ladybug, is about two days of feeding, plus whatever eggs get laid before it goes. Two days of feeding is genuinely worth having — at a hundred aphids a day it is two hundred aphids per beetle, which is why a hundred beetles clears a potted rose. It is simply not an occupation, and pricing it as one is where the disappointment comes from.

The beetles that leave are not defective and they are not running from your plant. Why they leave has not been established; what has been measured is that they do it with prey still on the plant, which is what rules your plant out as the reason. There is more on that, and on why it is not the disaster it sounds like.

Why once is never enough

Both published rates that worked involved releasing more than once — the shrub result needed “one or two subsequent releases,” and UC’s figure comes with a 1- to 2-week interval attached. The interval is not a hedge. It is set by the aphids.

An aphid goes from newborn to reproducing adult in seven to eight days, and does it without mating, giving birth to live daughters. One adult can produce up to eighty offspring in a week. So a release that removes most of a colony on Monday is racing a population that replaces its own breeding stock inside a week — and a second wave of predators has to arrive before the survivors have rebuilt, not after.

A single release against an established colony is therefore doing something real and then stopping, which from the outside looks identical to not working. The same arithmetic drives the two-week rebound people see after spraying — different pest, same problem of a treatment that lands once against a population that does not.

Indoors, and under glass

Every number above comes from an outdoor garden, and the single biggest term in the equation — dispersal — changes completely in a closed room. The trial itself shows how much. With beetles caged so they could not leave, about 46 beetles per square foot controlled the aphids. In the open it took roughly 210 per square foot — four and a half times as many.

Under glass the arithmetic changes shape as well as size, because greenhouse work is usually reported as a ratio of predators to aphids rather than per plant or per square foot. A 2017 review of ladybird releases in greenhouses records ratios from 1:20 all the way out to 1:640 having been tried, and reports that “1:10 or 1:20, ladybird: aphid was suitable.” That is a more useful instruction than it first looks: it means the number you need is set by how many aphids you have, so counting the colony matters more indoors than measuring the plant.

There is still no published per-plant rate for Hippodamia convergens on houseplants in a room, and deriving one from a rose bush would be inventing a number. What the caged result and the greenhouse ratios support is a direction: a room with the windows shut is closer to the cage than to the garden, and fewer beetles go further there than they would outside.

What does carry over indoors is that bright light and open windows defeat the containment you are relying on. Released adults orient to the brightest thing in the room, which in practice is a grow light or a window, not your plant.

The cheaper answer than ten times as many beetles

Reading all of the above as “buy a hundred times more” is one option, and for a single badly infested shrub it is the option the trial supports. For most people there are two better ones, and both come out of the same biology.

The eggs are worth more than the beetles

A female lays several hundred eggs over a season, in clusters of ten to thirty, on leaves and stems next to prey. The larvae that hatch cannot fly. They stay on the plant for their whole development and eat aphids the entire time, which is why a larva ends up eating more aphids in its life than an adult does in its two-day visit. The spiky grubs, not the beetles, are where most of the aphid-eating happens. A release that produces egg clusters is worth several releases that do not, and the conditions that make a beetle lay before it leaves — dense prey, water on the foliage, a dusk release — are the same ones that slow dispersal.

Something that breeds where the aphids are

The structural problem with any adult-beetle release is that the predator does not reproduce on your plant fast enough to keep up with something that gives birth to pregnant daughters. A parasitoid does. Aphidius wasps lay into aphids, and the next generation of wasps emerges from the colony itself rather than having to be shipped to it. On a spreading aphid problem, that pairing — something that eats now, something that breeds where the prey is — is what commercial growers run, and it is cheaper than escalating the beetle count.

Common questions

How many ladybugs do I need for one plant?

The only published figure for a container plant is 100 beetles on a 5-gallon potted rose, which gave 66 to 88 percent aphid control from a single release. For a heavily infested landscape shrub the figure is about 1,500, repeated at 1- to 2-week intervals. Both are per plant. Nobody has published a rate for a small houseplant, so the honest answer there is tens of beetles rather than two or three, and no more precision than that.

Why are these numbers so much higher than what I have been told?

Because the commonly recommended rate of 1 to 2 beetles per square foot was, when it was tested against aphid counts on roses, about a hundred times lower than the rate that worked. That comparison is the central finding of Flint and Dreistadt's 2005 trial, and it is quoted in the article above.

Will buying more just mean more of them fly away?

Some of them, yes, and that is already priced in — the rates above are rates that worked with the dispersal happening. An adult eats around a hundred aphids a day for the day or two it is present, so the feeding is front-loaded rather than lost.

Does releasing at dusk or misting the plant actually keep them?

It helps and the evidence is modest. UC IPM says wetting the plants and releasing in the cool of late evening “may slow beetle dispersal” — a hedge we are not going to upgrade. It costs nothing and it is worth doing; it does not turn a release into a resident population.

Is there a rate for indoors?

Not a published one. The nearest evidence is the caged arm of the same trial, where preventing dispersal cut the effective rate roughly fourfold. A closed room behaves more like that cage than like a garden, so fewer beetles go further indoors — but any specific indoor number you see, including one from us, is an estimate rather than a measurement.

Should I just buy a bigger tub, or buy something else?

For one badly infested plant, a bigger release applied to that plant is what the evidence supports. For aphids spreading across several plants, escalating the beetle count is the expensive way; a parasitoid wasp that reproduces inside the colony plus repeat predators is what growers run instead.