Short answer. A photoperiod plant waits for long nights (12 h dark indoors) to flower; you choose how long it vegetates. An autoflower starts flowering at a fixed age, usually 3-5 weeks from sprouting, whatever the light schedule. In mapped populations, autoflowering is a single recessive trait: a plant must carry two copies of the auto allele to autoflower [1].
What's happening
Photoperiod cannabis measures night length. Day-neutral (autoflowering) plants lack that brake: their flowering switch runs on age. Autoflowering germplasm is usually traced to northern or Central Asian populations popularly called "ruderalis" (Supported as a history claim; the taxonomy itself is contested) [4].
How autoflowering is inherited
Toth et al. (2022) mapped a major locus, Autoflower1, to chromosome 1 using bulked segregant analysis. Segregating populations fit a single recessive gene: about 25% of F2 plants flowered under long days (observed 31.8% and 27.3%, not significantly different from 25%) [1]. Call the photoperiod allele A and the auto allele a:
| Cross | Offspring genotypes | What you see |
|---|---|---|
| Photo (AA) x Auto (aa) | all Aa | 100% photoperiod (F1) |
| F1 (Aa) x F1 (Aa) | 1 AA : 2 Aa : 1 aa | ~75% photo, ~25% auto (F2) |
| F1 (Aa) x Auto (aa) (backcross) | 1 Aa : 1 aa | ~50% photo, ~50% auto |
| Auto (aa) x Auto (aa) | all aa | 100% auto |
Two useful details from the same study [1]:
- Heterozygotes are not identical to pure photoperiods. In the field, Aa plants flowered about two weeks earlier than AA plants and were intermediate in height and biomass, but they still did not flower under continuous light. So "recessive" describes autoflowering itself, not every effect of the gene.
- Not every auto source has been shown to share this locus. Some day-neutral hemp cultivars did not match the markers, and a complementation cross was inconclusive. The authors say it is not clear the mechanism is the same in all cultivars (Contested for the whole gene pool).
A 2023 preprint mapping day-neutral flowering in an F2 of 318 plants also found clean two-class segregation with no intermediate plants (Supported, not yet peer reviewed) [5]. An industry claim names a pseudo-response regulator (PRR37-like) gene as the candidate; treat that as unconfirmed until peer-reviewedbeing reviewed.
Numbers that matter
| Parameter | Photoperiod | Autoflower |
|---|---|---|
| Flowering trigger | Night length past a critical value (often ~10-11 h dark; cultivar specific) [2] | Plant age |
| Indoor veg schedule | 18/6 typical | Not needed |
| Indoor flower schedule | 12/12 | 18/6 to 20/4 common; 24/0 used, debated |
| Seed to harvest | ~3-5 months indoors | ~9-14 weeks |
| Clone and re-veg | Yes, keep a mother indefinitely | Clones flower on the parent's clock; impractical |
| Recovery from stress | Can veg longer to recover | Little time to recover; avoid heavy stress early |
House position: autoflowers. An auto runs on a strict countdown from germination, so any setback costs it time it can't get back. Plant it straight into its final pot: transplanting is almost a no-no, because of how the auto's taproot reacts (the founder's observation, compared with photoperiods). Train with LST, don't top: topping will never produce what proper low-stress training does on an auto. Grower practice, house-endorsed.
Are autos better? Better how?
Ask "are autos better?" and the honest answer is: how long is a piece of string? Better at what?
Per square foot per year: very hard to beat. An auto goes from seed to jar in about 9–14 weeks with no veg in the flower space, so you fit more harvests into a year. The only photoperiod setup that reliably beats it on space is a dialled-in sea of green or SCROG fed by a dedicated nursery, where the flowering space only ever holds flowering plants (see SOG and SCROG).
The arithmetic. Yield per square foot (or m²) per year = yield per cycle × cycles per year. The cycles are where autos win:
| Setup | Weeks the space is tied up per cycle | Cycles a year | Per-cycle yield an auto needs to tie |
|---|---|---|---|
| Photoperiod, vegged in the same space (4 wk veg + 9 wk flower + 1 wk reset) | 14 | ~3.7 | ~86 % of the photo's |
| Photoperiod SOG/SCROG with a dedicated nursery (9 wk flower + 1 wk reset) | 10 | ~5.2 | ~120 % of the photo's |
| Autoflower sown in its final pot (11 wk seed to harvest + 1 wk reset) | 12 | ~4.3 | — |
Divide 52 weeks by the weeks per cycle; the last column is photo cycles ÷ auto cycles. The week counts are typical, not fixed: a 9-week auto or a 12-week photo moves the numbers, so plug in your own. What the table shows is the shape of the argument. Against a photo that vegs in the flowering space, an auto only has to yield about 86 % as much per run to match it per year, and a good auto does. Against a dialled-in sea of green with its own nursery, the auto would need to out-yield the photo by about a fifth per run, which is a big ask. (The nursery's own space and power aren't counted in that row, which flatters it.)
For scale, one controlled study of a photoperiod cultivar flowered for 81 days at 12/12 measured 116 g/m² of dry flower at the lowest light level and 519 g/m² at the highest (1,800 µmol/m²/s at the canopy), rising in a straight line in between [6]. We haven't found a comparable peer-reviewed g/m² figure for autoflowersbeing reviewed. One more honest note: autos usually run 18–20 h of light, so at the same intensity they get 1.5–1.7 times the daily light (and use that much more lighting power per day) as a 12/12 photo. Per square foot they're hard to beat; per kilowatt-hour it's a closer race.
Cannabinoids: usually a little behind, sometimes level. In the founder's experience, autos usually test slightly under their photoperiod sisters, but the best of them give photos a real run. We haven't found a controlled head-to-head study of matched auto and photo linesbeing reviewed, so treat this as experienced grower observation. Breeding autoflowers explains why autos lagged and how breeders closed the gap.
Founder's records. The founder has had an elite autoflower lab-test at 30 % and has pulled 3/4 lb (about 340 g) in under 80 days. That's one grower's results with one elite line, not a typical number: 30 % is a rare feat for any flower (see buyer beware). It shows what the ceiling looks like when everything goes right.
Finicky: any stress costs yield. An auto can't be vegged longer to recover. Overwater it, transplant it, top it, let it get root-bound or hungry in week 3, and the clock keeps running: it flowers on schedule, smaller. That's why the house rules above are so strict.
"Good for beginners": true, but not because they're easy. Autos are short, quick and need no light-schedule change, so a beginner gets a harvest fast. But they're the opposite of forgiving. They show very quickly who has a green thumb and who doesn't. An elite auto grower naturally makes an even better photoperiod grower, because a photo forgives what an auto punishes. It doesn't work the other way round.
| Better how? | Autoflower | Photoperiod |
|---|---|---|
| Yield per area per year | Very hard to beat | Beats autos with a dialled-in SOG/SCROG and dedicated nursery |
| Yield per plant | Lower | Higher; veg as long as you like |
| Cannabinoids | Usually slightly under photo sisters; the best match them | The benchmark |
| Forgiveness | Very little: stress costs yield | Can veg longer to recover |
| Light power per day | 18–20 h | 12 h in flower |
| Clones, mothers, selection | Impractical | Yes |
| Learning curve | Quick feedback, harsh lessons | More forgiving |
Decision guide
- Want the shortest first grow, the most harvests per square foot without a nursery, or outdoor crops at high latitude? Autoflower, and be strict with it.
- Want to keep a mother, clone, select and breed? Photoperiod.
- Breeding autos? Read breeding autoflowers first: it takes at least two generations to recover autos from a photo cross.
Common mistakes
- Expecting F1 "auto x photo" seed to autoflower. It won't if the trait is recessive.
- Transplanting autos at all, or topping them. They don't pause the clock while recovering: sow into the final pot and use LST.
- Treating autos as the easy option. They punish mistakes harder than photos do.
- Assuming every seed in an "auto" pack is homozygous. Poorly stabilised lines throw photoperiod plants; this is a quality-control failure, not bad luck.
Evidence
- Day-length control of photoperiod flowering and cultivar-specific thresholds: Established [2][3].
- Single recessive locus Autoflower1 in the mapped populations: Supported (peer reviewed, limited germplasm) [1].
- That all commercial autos share the same locus: Contested [1].
- "Ruderalis origin" of the trait: Supported as history, label as a claim.
- Autos winning on yield per area per year (except against a dialled-in SOG/SCROG with a nursery): Supported by cycle arithmetic and the founder's experience; per-cycle auto yields in g/m² need peer-reviewed data. Photo yield response to light: Supported [6].
- Autos testing slightly under photo sisters: founder's observation; no controlled comparison found.
- "Autos are easy": Myth (house position).
