Short answer. Lollipopping removes small branches and leaves from the bottom third of the plant at or just after the flip, because that growth gets little light and makes airy, low-potency buds. Defoliation removes fan leaves that shade bud sites or block airflow. The best controlled data (large ~2.5 m greenhouse plants of a CBD-dominant medical cultivar): defoliation 3 weeks before harvest and removing the lower third at the flip did not significantly change flower or cannabinoid yield per plant, and defoliation improved cannabinoid uniformity across the plant [1]. Lower cannabinoids at the bottom of the plant correlated with low light penetration [1]. So stripping within reason doesn't hurt yield, and it doesn't boost it either. In the founder's experience tucking helps more.
House position: tuck first, strip within reason. Leaves are the solar farm that powers the buds. Stripping within reason doesn't hurt yield (the trials above agree), but the founder, after a lifetime of side-by-sides, finds that tucking fan leaves under or behind bud sites and training branches out of each other's shade helps more than removing them. Responses vary by cultivar: some clones can lose most of their leaves with no effect, or even grow faster, while others suffer. Learn how your cultivar reacts before you defoliate hard.
Open question: a GenetixBay trial. Two things nobody has settled for indoor drug-type cannabis: does tucking beat stripping for dry yield, and does defoliation raise cannabinoid content (the hemp data in [3] hints it might, at one timing)? This is exactly what the Bay Trials are for: matched clones, split tent, tuck vs strip, dry weight and lab COAs on both sides. See Mythbusters for the open trials.
What the research shows
| Treatment (Danziger & Bernstein 2021 [1]) | Flower yield/plant | Cannabinoid yield/plant | Uniformity |
|---|---|---|---|
| Defoliation 3 weeks before harvest | No significant change | No significant change | Improved |
| Lower-third branch + leaf removal at flip ("lollipopping") | No significant change | No significant change | Not improved alone; best when combined with defoliation |
| Lower-third removal + defoliation | No significant change | No significant change | Most uniform of non-branch-removal treatments |
| Removing all primary branches (single stem) | −up to 99 % | ~3 % of others | High (but pointless) |
At commercial density, defoliation gave yields per m² similar to the control [2]. In hemp, defoliation in early flower raised total cannabinoid concentration while veg or late-flower defoliation did not [3]; this was a simulated-hail study and may not transfer to indoor drug-type cannabis.
How to do it
- Lollipop at flip to ~day 21 of flower: remove small shoots and leaves from the lower ~⅓ that will never reach the light.
- Veg: tuck first. Tuck fan leaves under or behind shaded shoots, and spread branches with LST. Remove only leaves that touch the medium or are damaged. If you do remove healthy leaves, take no more than ~20–30 % of leaf area at once (Grower practice).
- Around day 21 of flower: tuck large fan leaves away from bud sites. Remove only those you can't tuck and that crowd a dense, humid interior.
- Late flower: only remove leaves that are dying, mouldy, or blocking airflow in dense colas.
- Sterilise scissors between plants.
Numbers that matter
| Item | Typical | Evidence |
|---|---|---|
| Lollipop zone | Lower ~⅓ | Supported [1] |
| Max leaf removal per session | ~20–30 %, and only after tucking | Grower practice; tuck-first is the house position |
| Timing | Flip, ~day 21, ~3 weeks before harvest | Supported for the last [1]; others Grower practice |
Why airflow matters
Dense, humid canopies favour Botrytis bud rot and powdery mildew [4]. Thinning the interior lowers canopy humidity (humidity, airflow).
Common mistakes
- "Schwazzing" (stripping almost all leaves twice in flower), especially on stressed or deficient plants or on a cultivar you haven't tested. Some clones shrug it off; others suffer.
- Plucking a leaf you could have tucked.
- Removing leaves from autoflowers in a short veg.
- Taking healthy leaves when the real problem is too few, too dim lights.
Evidence
Effects on per-plant yield and uniformity are Supported by controlled studies on large plants [1][2]. Hemp timing effects are Supported for hemp only [3]. Session limits and day-21 timing are Grower practice. "Tuck first" is the house position: the controlled trials found stripping within reason neither hurt nor helped yield [1][2], and the founder finds tucking helps more. Tuck vs strip, and any cannabinoid gain from defoliation, are open questions for a GenetixBay trial; cultivar responses vary.
