How Cannabis Tells Time: 18/6, 12/12, and Why Autoflowers Don't Care

The 18/6 light cycle — eighteen hours of light, six of darkness per day — is the usual schedule for keeping cannabis in vegetative growth. 12/12 — twelve on, twelve off — triggers and sustains flowering. Both are correct — and both describe the schedule from the grower's side, not the plant's. The plant is not counting hours of light. It is counting hours of uninterrupted darkness — and every schedule rule growers memorize (why 18/6 holds veg, why a midnight light leak matters when a cloudy afternoon doesn't, why autoflowers ignore all of it) can be derived from that one fact instead of memorized.

The plant measures the night

Photoperiod cannabis is a short-day plant — a misleading name; it's really a long-night plant. A 2025 study in Scientific Reports (Haiden, Johnson and Berkowitz) made the distinction explicit: what governs flowering is the consistency of the dark period, not the light period, and photoperiod-sensitive cannabis tends to flower once night length passes about ten hours.

The counting hardware is a family of light-sensing pigments called phytochromes. Red light flips phytochrome into its active form; in darkness it decays back toward inactive. Work from Bruce Bugbee's lab at Utah State (Kusuma and colleagues, PLOS One, 2021) demonstrated the red/far-red logic in cannabis: red photons during the night suppress flowering; far-red photons undo it. The decay isn't a simple sand timer either — the 2025 study proposed a pathway running through clock-associated genes to florigen, the flowering hormone signal. The working simplification: the plant counts continuous dark hours against its internal clock, and when the count crosses its threshold night after night, flowering begins.

Among the six numbers every cabinet runs on, photoperiod is the odd one out — the cheapest number to control, because a timer runs it.

Where the threshold actually sits

The critical night length is a cultivar trait, not a universal constant — research puts it at roughly 9 to 11 hours of darkness for most cultivars. In a 2023 trial in Plants pointedly titled "Moving Away from 12:12," Peterswald and colleagues found 14 of 26 genotypes initiated flowering with 10 hours or less of dark; Grover, Haiden and Berkowitz (Plants, 2025) put the cultivar-specific critical duration "often near 10 to 11 hours." A University of Guelph team (Ahrens, Llewellyn and Zheng, Plants, 2023) added a wrinkle: some cultivars initiate flowers with as little as 9 hours of dark, then stall before developing them — initiation isn't the same as finishing.

So why does everyone run 12/12? Because 12 hours of darkness clears essentially every photoperiod cultivar's threshold with margin. That makes 12/12 a reliability convention, not a biological law — and researchers have started probing past it. The same Guelph group reported in 2024 that two high-THC cultivars yielded substantially more under 13 hours of light than under 12 with no cannabinoid loss; Peterswald's team saw biomass gains at 14 hours but a drop in THC concentration in high-THC lines. File it under emerging research — cultivar-specific, with trade-offs, not default advice.

A correction in the open: our light schedule chart — the page for which schedule to run, when to flip, and how to seal a leaking tent — still uses the older "~12 hours of dark" shorthand; it's being updated to this cultivar-dependent picture. Its practical advice stands: 12/12 works essentially everywhere because it overshoots essentially every common cultivar's threshold.

Why an 18/6 light cycle holds veg — and why 18 isn't magic

Run the mechanism backwards and 18/6 explains itself: a 6-hour night can't cross a 9-to-11-hour threshold, so the flowering signal never fires. That single fact is why a mother plant can sit under 18/6 for years, why research groups use 18/6 as the standard vegetative control, and why cannabis stays vegetative even under continuous light.

But 18 is not a researched optimum — we could find no study that compares veg photoperiods head-to-head for downstream yield. 18/6 is grower convention, and the practical rationale is economics — a quarter fewer lit hours than 24/0, less heat, zero flowering risk. The popular claim that plants "need darkness to rest" is belief, not established biology, though one 2022 study (Šrajer Gajdošik and colleagues, IJMS) found continuous 24/0 light raised oxidative-stress markers — a gentle argument for giving the plant some night. What is well supported: more daily light means more growth up to a point — Grover's team showed vegetative light dose carries through to final yield. But dose is a separate number from schedule; how much light actually lands on the canopy is covered in setting light distance by PPFD.

One practical question this settles: the plant times the dark block, not your wall clock. Darkness can fall at any consistent hour — running lights-on overnight to dodge daytime heat and peak electricity rates confuses nothing.

The dark period is the fragile half

The mechanism predicts an asymmetry: a dark hour at noon does nothing; a bright hour at midnight resets the count, flipping the pigment back to its active form mid-decay. Grover's team demonstrated it — a single one-hour pulse at full canopy intensity mid-night kept plants vegetative even on a 12-hour photoperiod.

But that was a bright, sustained pulse. How dim is safe? Nobody has measured the danger line in cannabis. The Bugbee-lab study delayed flowering with all-night far-red light at intensities far beyond any pinhole leak — sustained, high-dose exposure, a different regime from the brief far-red that reverses a red pulse, and assumed moonlight — about 0.001 µmol/m²/s — as the floor below which effects are unlikely: a modeling assumption, not a measurement. The same pigment logic ranks the colors: phytochrome absorbs red strongly and green only weakly, which is why a red glow is the worst kind of leak and a dim green work light is lower-risk — lower, not proven zero. The grower standard — if you can't see your hand inside the dark tent, it's dark enough — is a rule of thumb, and worth labeling as one. And the internet's favorite scare — hermaphroditism from light leaks — is widely reported by growers, not experimentally confirmed: the one direct test (a 2024 Guelph undergraduate study by Oliver) found no significant association at low intensities, though it was too weak to settle the question, and a 2026 review in Plants points to genetics and hormonal stress as the documented drivers of altered sex expression. The calibrated position: bright, repeated interruptions demonstrably disrupt flowering; a dim, brief one probably doesn't; the line between is unmeasured, so seal what's cheap to seal — the checklist lives on the schedule chart page linked above.

Autoflowers opted out of the clock

Autoflowering cannabis doesn't count nights at all. It flowers on age — typically around 3 to 5 weeks from germination, a range that comes from grower guides rather than controlled trials and varies by line. The genetics, however, are mapped: Toth and colleagues at Cornell (Frontiers in Plant Science, 2022) traced photoperiod insensitivity in modern autoflower lines to a single recessive gene region, Autoflower1, and Dowling and colleagues (The Plant Journal, 2024) found a second, independent locus in day-neutral hemp — the trait has evolved more than once. It's often credited to wild "ruderalis" from far-northern latitudes; Toth's group notes recent genomic studies don't support ruderalis as a distinct group, but the latitude logic likely still holds: where the season is too short to wait on long nights, timing by age wins.

For a cabinet, this reframes the seed decision. An autoflower removes the flip entirely — no schedule change, no threshold to respect — though every claim that one auto schedule out-yields another (18/6 vs 20/4 vs 24/0) is grower convention: no peer-reviewed head-to-head has turned up in our research, and even seed banks admit there's no consensus. A photoperiod plant hands you the opposite: the trigger is yours to schedule. Veg length becomes a design decision, so whether to veg first or flower from seed depends on your space rather than the biology — and grower practice holds that a plant kept on short days from day one simply begins flowering once it's mature enough — a convention, not something in our research base — which is the basis of running 12/12 from seed in a compact cabinet.

Put a meter on the dark period

Here is what the clock looks like in week one of flower: a timer set to 12/12, chosen for exactly the margin logic above. ColaXpress owns no cabinet and has no grow running, so there is no log of ours to follow; the check below is one to run in your own space. One of the first entries worth making in your own log is the dark half of the schedule itself: a light meter inside the closed cabinet, lights off, to put a number on "dark" before you trust a timer with a whole flowering run's worth of nights. No study sets a pass mark for that reading, as the leak section above explains, so a near-zero number is reassurance, not proof. If cannabis tells time by darkness, the first honest measurement is checking that the darkness is real.