Most biological control guides recommend Phytoseiulus persimilis first for spider mites. A greenhouse strawberry trial published in 2026 compared it with Amblyseius swirskii across three temperature bands. At 30–32 °C, swirskii gave 91.47% control and persimilis 60.57%, a gap of nearly 31 percentage points. A combined release had the highest mean control at every band. A second study adds a caveat. Swirskii still attacks prey at 34 °C at about the same rate as at 26 °C, but it takes longer to handle each prey item, so it predates more slowly. Above about 32 °C canopy temperature there is little population-control data for any commercial predatory mite. “Best available” is a reasonable choice there, but it is not a proven result.
1. The comparison
The trial was run on pot-grown strawberry in a greenhouse against Tetranychus urticae, the two-spotted spider mite, across three temperature bands. It was published in Insects, volume 17, issue 4, article 366.
| Temperature band | A. swirskii | P. persimilis | Combined release |
|---|---|---|---|
| 25–27 °C | 45.98% ± 9.83 | 33.28% ± 10.78 | 52.98% ± 8.44 |
| 28–30 °C | 87.84% ± 3.49 | 79.77% ± 10.01 (its best band) | 91.00% ± 4.58 |
| 30–32 °C | 91.47% ± 4.26 | 60.57% ± 14.39 | 94.63% ± 3.10 |
Values are % control of T. urticae. The paper does not say whether ± is the standard deviation or the standard error. The design used three replications per treatment.
Three readings, in order of usefulness.
The 31-point gap at 30–32 °C is the headline. That is the band an unchilled Thai flowering room is likely to sit in. There, persimilis gave roughly two-thirds of swirskii’s control.
Look at the spread, not just the means. At 30–32 °C, persimilis is 60.57% ± 14.39 and swirskii is 91.47% ± 4.26. Neither figure is a range of possible outcomes. Because the paper does not say whether ± is SD or SE, and there were only three replications, the safe reading is that persimilis results varied much more between replicates than swirskii results. In the heat, persimilis was less effective and less predictable. For a commercial grower, the second may matter more.
Persimilis did best at 28–30 °C, at 79.77%. Its decline is not gradual across the whole range. It was respectable in the 28–30 °C band and fell off in the 30–32 °C band. At 29 °C persimilis may be adequate. At 31 °C, this trial says it is not the species to rely on. That two-degree difference is inside the normal daily swing of an unchilled tropical room.
2. What happens toward 34 °C
A second study measured A. swirskii’s functional response at 14, 18, 22, 26, 30 and 34 °C. The functional response is how many prey a predator eats as prey density changes. The study was published in Insects, volume 15, issue 6, article 444. It was a laboratory study on leaf discs, using immature T. urticae as prey over 24-hour trials. The data fitted a Holling type II functional response, which the authors analysed with Holling’s disc equation.
| Temperature | Attack rate (a′) | Handling time (Th) |
|---|---|---|
| 26 °C | 0.919 (highest) | 0.005 (shortest) |
| 30 °C | 0.910 | 0.006 |
| 34 °C | 0.912 | 0.007 |
The paper does not state units for a′ or Th. Read these numbers as relative values on the paper’s own scale (24-hour trials), not as prey per hour or per day.
The authors conclude that swirskii “would be more effective” at 26–30 °C.
Read the attack-rate column carefully, because it is nearly flat. The values are 0.919, 0.910 and 0.912, barely different. The change shows up in handling time. It rises from about 0.005 to 0.007 between 26 °C and 34 °C, roughly 40% longer per prey item. At 34 °C the predator still finds and attacks prey. It just processes each one more slowly, so it eats fewer prey per day.
The honest conclusion: A. swirskii is the best-supported option above 30 °C, and it keeps predating at 34 °C, though more slowly than at its optimum around 26 °C. What is missing above about 32 °C is efficacy data: whether a release actually holds a spider-mite population down at those temperatures. The functional-response study measured feeding, not population control.
3. What the honest recommendation therefore is
Lead with swirskii above about 28–30 °C. The 31-point gap at 30–32 °C and the much smaller variation between replicates make this clear on the evidence. In Thailand, persimilis and swirskii are imported natural enemies. Check with the Department of Agriculture whether these predatory mite species may be imported or released, and look for local suppliers, before planning a programme.
Add persimilis as a supplement, not as the mainstay. Combined release had the highest mean control at every band, including 94.63% at 30–32 °C. The authors’ tests separated it clearly from persimilis alone only at 30–32 °C, so the advantage is firmest in the heat. Persimilis is a spider-mite specialist and swirskii is a generalist, so the two may complement each other. The mistake is not using persimilis. It is relying on it in the heat.
Webbing matters, especially in heavy infestations. Swirskii is generally rated as a thrips and whitefly generalist. It is reported to cope poorly in dense spider-mite webbing, which persimilis handles better. The 91.47% figure comes from releases at about one predator per ten spider mites, not from established, heavily webbed colonies. Release swirskii early, while infestations are light. Where webbing is already present, combine it with persimilis or a Neoseiulus species rather than using it alone.
Do not read this trial as support for persimilis alone in cooler rooms. At 25–27 °C, persimilis alone was the weakest treatment (33.28%), and no treatment performed well in that band. The trial did not test below 25 °C. Standard cool-climate recommendations for persimilis rest on other studies, not on this one.
Above about 32 °C, do not plan as though biological control will be your primary measure. Swirskii still feeds at 34 °C, but there is little data on whether releases hold a population at those temperatures. That does not mean releasing nothing. It means putting more weight on exclusion, sanitation and cultural control, and not assuming a release alone will keep mites under control.
Measure canopy temperature, not room temperature. All of these figures are the temperature the predator experiences. Under high-intensity lighting, leaf-surface temperature can be higher than the air temperature, so a wall sensor reading 29 °C may sit over a canopy at 32 °C. This is the single most likely reason a release performs worse than the numbers suggest.
4. Limits of this evidence
Being specific, because these are clean numbers and clean numbers travel further than they should.
The comparison trial was on strawberry, not cannabis. The prey species, Tetranychus urticae, is the same one that attacks cannabis. But cannabis has a dense, resinous, trichome-covered flowering canopy, and on other crops trichome density is known to interfere with predatory mite movement. Whether these figures hold on a flowering cannabis bud is not established.
Neither study was designed around tropical humidity. Phytoseiid egg survival is humidity-sensitive, and in most species the classic problem is humidity that is too low. In a Thai wet season it will be high, which may help. We found no data on this, and we are not going to assume a direction.
Neither is a production-room trial. The comparison was a small greenhouse trial with three replications. The functional-response work was a laboratory leaf-disc study. Release rates, distribution method, and whether the predator establishes at all are the practical variables, and none of them was measured here.
Other species have temperature data, but not head-to-head trials. Neoseiulus californicus is often recommended as a heat-tolerant option in trade material. Published functional-response work exists for it: a 2023 PeerJ study found a type II response by adult females to T. urticae across 24–36 °C. Neoseiulus longispinosus is a spider-mite predator native to Southeast Asia and has been studied in Thailand. A 2017 Journal of Economic Entomology study measured its development and population growth on T. urticae from 15 to 37.5 °C. For Thai readers it is arguably the most relevant candidate. Evidence varies between studies, and neither species has been compared with swirskii and persimilis in a trial like the one above, let alone on cannabis.
5. The pattern this belongs to
Predatory mites are one of the few biological-control groups with published high-temperature data, though that data comes from a greenhouse trial and a laboratory study, not from tropical humidity. For most other biological products, tropical validation is scarce, so ask any supplier plainly at what temperature and humidity a product was tested.
6. What to record
Canopy temperature at the release site, logged, with the daily maximum recorded separately. Everything above depends on it.
Which species, at what release rate, on what interval, and from which supplier. Establishment failure is often a supply-chain and handling problem rather than a biology problem.
Mite population counts before and after, on a defined number of leaves at a defined canopy height. Without a before-count, a release cannot be evaluated.
Whether webbing was present at release. It is likely to change how well swirskii performs.
Whether the predators are found alive at week two. Only you can tell apart “the release did not work” and “the predators did not survive shipping”.
Has anyone here run swirskii and persimilis, alone or together, in a Thai flowering room, and logged canopy temperature alongside? Has anyone tried a local Neoseiulus species? The comparison numbers above come from strawberry. The property of cannabis most likely to break them is a resinous flowering canopy that predatory mites struggle to move through. That is a cannabis-specific question, and only people growing cannabis in the heat can answer it.
Related reading on the forum: Spider Mites & Thrips: biological control (topic 1236) and the integrated pest management (IPM) topic (topic 1301).
Sources
- Mdallel L.M. et al. Temperature Effects on the Efficacy of Phytoseiulus persimilis and Amblyseius swirskii (Acari: Phytoseiidae) Against Tetranychus urticae (Acari: Tetranychidae) in Strawberry Crops. Insects 17(4):366. https://doi.org/10.3390/insects17040366
- Elmoghazy M.M.E. et al. (2024). Implications of Temperature and Prey Density on Predatory Mite Amblyseius swirskii (Acari: Phytoseiidae) Functional Responses. Insects 15(6):444. https://www.mdpi.com/2075-4450/15/6/444
- Functional response of Neoseiulus californicus to Tetranychus urticae at different temperatures. PeerJ (2023), article 16461. https://peerj.com/articles/16461
- Temperature-dependent demography of two closely related predatory mites, Neoseiulus womersleyi and N. longispinosus. Journal of Economic Entomology (2017). https://doi.org/10.1093/jee/tox177
This article is technical horticultural information for readers operating legally in their jurisdiction. Biological control agent availability and registration differ by country. Confirm that any species named may legally be imported and released where you are before ordering. In Thailand, check with the Department of Agriculture. In Thailand, cannabis flower is a controlled herb under the Notification of the Ministry of Public Health Re: Controlled Herbs (Cannabis) B.E. 2568, and flower that is sold must come from a GACP-certified cultivation site. Nothing here is legal advice.