Quick answer: A clone is a cutting taken from a living cannabis plant and rooted into a new plant. Take a cutting about 8 inches (20 cm) long with at least three fully expanded leaves, leave those leaves uncut, dip it in a rooting gel, and stick it into a rockwool cube under a humidity dome at 72 to 78°F (22 to 26°C). Most cuttings root in 12 to 21 days, but the honest range is wider: an easy cultivar in good conditions can show roots on day 5 and be ready to transplant by day 10, and a stubborn one can take five weeks. Three things decide whether it works, and they are not the three things most guides talk about: the medium you root into (published rooting success ranges from 13% to 80% depending on it), whether you used any rooting hormone at all (which matters far more than which one), and the cultivar you are cutting from. Trimming the tips off the leaves, which nearly every guide tells you to do, measurably lowers your success rate - from 71% to 53% in a University of Guelph trial.
This guide covers what a clone does and does not inherit, whether you need a mother plant, the four levers that actually decide rooting success, the viroid that quietly destroys clone libraries, the cut itself step by step, the rooting environment, a day-by-day timeline, seven ways clones fail and what each one means, and honest answers on autoflowers, monster cropping, and whether clone lines really degrade.
What a clone actually inherits (and what it does not)
A clone is not a seedling. A seed is a new individual built from two parents. A cutting is a piece of an existing plant that grows its own roots and carries on being that plant. Everything downstream of that distinction is where growers get surprised.
What a clone reliably inherits: the genotype, and therefore the sex. If you cut from a female, every clone is female. There is no lottery, no need to sex the plant later, and no wasted weeks on males. That single fact is why most growers try cloning in the first place.
What a clone also inherits, and nobody advertises: whatever the mother is carrying. Pests ride along. So do systemic pathogens, and one of them is bad enough that it gets its own section below.
What a clone does not guarantee: the same plant. You will read that a clone is "a perfect copy with the same aroma, flavor, structure and potency every single time." It is a copy of the genetics, not of the result. Terpene and cannabinoid expression track light, temperature, feeding and root volume, and two clones off the same mother in different tents will not smell the same. The genotype is fixed. The phenotype is a negotiation with your grow room.
One more inherited trait matters for planning. A cutting skips the juvenile phase a seedling has to grow through, so a rooted clone is ready to be flipped to flower sooner than a plant of the same physical size grown from seed. It has already done that part of its development on the mother. This is not the same as the clone being "the same age as the mother" - a common phrasing that means nothing physiologically - but the practical effect is real.
Do you need a mother plant?
No. This is the question that stops most first-time growers, and the answer is simpler than the internet makes it.
A "mother plant" is just a plant you have decided to keep in vegetative growth so you can keep cutting from it. It is a role, not a species. You can take cuttings off the plant already growing in your tent, root them, and flower the original anyway. Plenty of growers do exactly this: take four cuttings the week before flipping to 12/12, and if the harvest turns out well, you already have that genetics rooted and waiting.
The one hard requirement is that the plant is in vegetative growth, under a long-day light schedule (usually 18 hours on, 6 off). A plant in flower has switched its resources to making buds, and cuttings taken from it behave differently. That gets its own section too.
Keeping a dedicated mother
If you want a genetics library rather than a one-off, keep one plant permanently in veg. Keep it small on purpose. Every time you top a plant, the piece you remove is a cutting - topping and cloning are the same cut performed for two different reasons. Topping a mother repeatedly gives you more branch tips, and more branch tips means more cuttings per round. Opening her up with low stress training clips does the same job without the recovery time a cut costs.
Mothers do not age out. You will see advice to retire stock plants after roughly six months. A 2025 trial took cuttings from cannabis mothers repeatedly between 65 and 178 days after planting and found no effect of mother age on either the fraction of cuttings that rooted or their root mass.5 Older mothers produced smaller, lighter cuttings, which is a reason to keep the mother well fed and pruned, not a reason to throw her out. What did matter in that trial was light intensity on the mother, and only in one of the two cultivars tested - and it went the wrong way, with the brighter mothers producing cuttings that rooted worse. Blasting a mother with light does not make better cuttings.
What actually decides whether a cutting roots
Most cloning guides spend their word count on the cut itself - the angle of the blade, whether to cut underwater, which gel to buy. The published research says the cut is not where the variance lives. Four things move the number, and here they are in the order the evidence ranks them.
1. The medium you root into
This is the biggest single lever and almost every guide treats it as a matter of taste. A University of Florida trial ran 960 hemp cuttings across four propagation media on the same bench, in the same room, on the same day. Rooting success ranged from 13.3% to 80.4% depending only on what the cutting was stuck into.2 Rockwool came top at 80.4%, producing roughly ten times the root growth of the next best medium: 7.7 roots and 9.5 cm of total root length per cutting, against 0.7 roots and 1.3 cm for a peat-based mix.
If you change one thing after reading this article, change this one. Rockwool cubes or a comparable engineered propagation plug is the difference between most of your cuttings living and most of them dying.
2. Whether you used any rooting hormone at all
In that same trial, cuttings that got any auxin hormone rooted 15 to 18 times more successfully than cuttings that got none - and the different products, concentrations and application methods showed no statistically significant difference from one another.2 A separate Colombian trial with a true no-hormone control found the same shape of result: root volume was 0.4133 mL with hormone against 0.0348 mL without.16
So use it. Do not agonize over which. The marketing war between rooting gels is being fought over the smallest variable in the room. And when the genetics and the conditions are already good, hormone stops mattering much at all: one trial rooted cuttings at 76% to 88% whether or not they got 1,000 ppm IBA.9
And more is not better. An Oregon State trial on hemp tested 0, 3,000 and 8,000 ppm of indole-3-butyric acid (IBA, the standard rooting auxin). Rooting quality and percentage declined at 8,000 ppm, and 3,000 ppm delivered in solution performed worse than the same 3,000 ppm delivered in talc powder. Their conclusion was that cuttings rooted best with either no hormone at all or 3,000 ppm talc.3 Read those two studies together and the honest summary is this: you can root cuttings without hormone, and in good conditions plenty of them will. The problem is that you only find out your conditions were marginal after the batch has already failed. Use it. Buy a standard gel, and do not buy the strong one.
3. The genetics you are cutting from
Some cultivars root easily and some fight you, and the gap is not small. Across four varieties in one trial, two rooted at roughly 90% and two at roughly 60%, with average root lengths of 22.6 cm and 17.8 cm against 7.6 cm and 7.0 cm.10 In the Florida study, one cultivar out-rooted the next best by about 20%.2
The practical version: if your cuttings failed, the cultivar is a real suspect, and you should take more cuttings than you need. Ten for a target of six is not pessimism, it is arithmetic.
4. The environment while it roots
A cutting has no roots. Every drop of water it loses through its leaves is a drop it cannot replace. The whole job of the first week is keeping the leaves from transpiring faster than the stem can wick. That is what the dome is for, and it is covered in its own section below.
Notice what is not on this list: the angle of your cut, whether you scraped the stem, whether you sliced it underwater. Those are conventions. Some of them are harmless. At least one of them is measurably costing people plants, and it is in the myths section.
The mistake that quietly kills clone libraries
This section is the reason to read the article. If you take one thing from it, take this.
Hop latent viroid (HLVd) is a 256-nucleotide strand of RNA - not a fungus, not a bacterium, not quite a virus - that lives systemically inside a cannabis plant and is spread almost entirely by people propagating plants. Growers know its effects as "dudding": plants that look fine, then grow smaller, produce airy loose flower, and lose potency and smell for no reason anybody can find.
Why it belongs in a cloning article specifically
Because cloning is how it travels. In a 2025 study, cuttings taken from HLVd-infected stock plants across four cannabis genotypes were tested two weeks after roots emerged, and every one of them carried the viroid.7 Their rooting was visibly worse too, with fewer and shorter roots than cuttings from clean plants.
And the mother does not have to look sick. The same study recovered HLVd from asymptomatic stock plants, and a survey of Canadian production described infected plants ranging "from asymptomatic to mild leaf curl and mottling" on mothers, with the obvious damage only showing up later, on the flowering plants downstream.8 A mother can look like the best plant you have ever grown and hand a copy of the viroid to every cutting you take off her.
How common is it, really
The best number available is not a guess. Canadian licensed facilities submitted 15,947 samples across nine provinces between 2020 and 2023. Positive detection ranged from 5.3% to 92% depending on province and year, and the country-wide average was 25.6%.8 That is commercial production, where people test. Nobody has measured the home-grow population, and anybody who tells you a figure for it is inventing it.
You will also see "90% of cannabis facilities in California tested positive" quoted as though it were science. It comes from a nursery's own internal testing survey, it counts facilities rather than plants, and it has never been published in a journal. It is probably directionally true. It is not a citation.
What it costs
In the Canadian survey, infected plants showed reductions of 12% to 42% in inflorescence stem length, fresh weight and plant height compared to uninfected plants, depending on genotype, and THC and terpene levels in diseased flower were significantly lower.8 The often-repeated "50 to 70% THC loss" figure traces back to a review article that gives no source for it, and we are not going to repeat it. The 12 to 42% range is the number with a study behind it.
Worth knowing for honesty: a 2024 Colorado State thesis on hemp cultivars found no yield effect from HLVd infection at all, with cannabinoid reductions only in plants that were actually showing symptoms.19 That is a student thesis on hemp rather than a journal paper on drug-type cannabis, so it does not overturn the Canadian survey. It does say the damage is genotype-dependent and symptom-dependent, not automatic.
The timeline is the whole argument for testing
After stem inoculation, HLVd was detectable in root tissue within 2 to 3 weeks and in foliage within 4 to 6 weeks. Visible symptoms did not appear until about 10 weeks.7 There is a window of roughly two months where the plant is infected, is passing the viroid to every cutting you take, and looks completely normal. Eyeballing your mother is not a screening method.
Testing is a lab RT-PCR assay. You mail leaf or root tissue to a plant pathogen lab and they run it. Roots detect it earliest and most consistently, which matters when you are screening a recently exposed plant; leaf and petiole tissue performs comparably in established plants and is far easier to sample without damage. If you are about to build a library off one mother and keep it for years, that one test is the cheapest insurance in this entire hobby.
The tool advice on every other cloning page is wrong
Everybody tells you to sanitize your blade. Almost nobody tells you what actually works, and the two pages on page one of Google that do give a number disagree with each other and with the research.
The 2025 study treated infectious sap with a list of disinfectants and checked whether the viroid RNA survived. 70% ethanol failed. Virkon failed. Hydrogen peroxide-based ZeroTol failed. Hypochlorous acid failed at 100, 200, 250 and 500 ppm. What worked on sap was a 10% dilution of 8.25% household bleach for 2 minutes, a 20% dilution for 1 to 2 minutes, or 1,000 ppm hypochlorous acid for 1 minute.7
Two things follow. Wiping your scissors with an alcohol pad between plants, which is what most growers do, is doing nothing against this pathogen. And contact time is not a detail - a 30-second dip is not the 2 minutes that was tested.
| Disinfectant | Tested on infectious sap | Contact time | Use it? |
|---|---|---|---|
| Household bleach, 10% dilution | Degraded the viroid RNA | 2 minutes | Yes - this is the practical choice |
| Household bleach, 20% dilution | Degraded the viroid RNA | 1 to 2 minutes | Yes, harder on steel |
| Hypochlorous acid, 1,000 ppm | Degraded the viroid RNA | 1 minute | Yes, if you already have it |
| Ethanol, 70% | No effect | 10 minutes | No - and this is what most growers use |
| Virkon, 2% | No effect | 2 minutes | No |
| Hydrogen peroxide product, 2% | No effect | 2 minutes | No |
The practical protocol: keep a cup of 10% bleach on the bench, and if you are cutting from more than one plant, park the blade in it for two full minutes between mothers and rinse before the next cut. If you are cutting from one plant only, tool transmission between your own plants is not the risk - the mother's own status is.
One honest limit. No study has measured how often a contaminated blade actually transmits HLVd in cannabis. What is established is that sap remains detectable for up to 7 days at room temperature and dried tissue for up to 4 weeks, and that sap-inoculated plants become infected. Blade transmission is a well-supported inference from those facts, not a measured rate, and we are not going to invent a percentage for it.
The last thing to know is what does not fix it. Meristem tip culture - the laboratory route to clean stock - recovered HLVd-free plants from infected material at an average frequency of 40.66% across eight genotypes, and in one genotype it produced zero clean plants.7 Tissue culture is a tool, not a cure. For a home grower the answer is simpler and much cheaper: start clean, test before you commit, and do not accept cuttings from a garden with a dudding problem.
What you need
The kit is short and most of it is cheap. What follows is sorted by how much the evidence says it matters, which is not how the shopping lists are usually sorted.
| Item | What it does | How much it matters |
|---|---|---|
| Rockwool cubes or engineered propagation plugs | The medium the cutting roots into | Highest. Swings rooting success from 13% to 80% |
| Rooting gel or powder | Supplies auxin, the hormone that triggers root initiation | Strongly recommended. Which brand barely matters. Skip the 8,000 ppm versions |
| Humidity dome and tray | Stops a rootless cutting from drying out | High. A clear takeout container with a lid works |
| A sharp, clean blade | Makes the cut without crushing the stem | Medium. A scalpel or new razor beats old scissors |
| Household bleach | Sanitizes the blade between mother plants | Medium normally, critical if you cut from more than one plant |
| Low-intensity light | Keeps the cutting alive without driving transpiration | Medium. A cheap fluorescent or a dimmed LED at distance |
| Heat mat | Warms the root zone to around 79°F (26°C) | Useful, not essential, unless your room is cold |
| pH meter and pH-down | Conditions rockwool, which is alkaline out of the bag | Needed if you use rockwool. Soak to pH 5.5 to 6.0 first |
Notice what is not on the list. No cloning machine, no aeroponic cloner, no mist system. Those work, and aeroponics performs as well as or better than rockwool in controlled trials,15 but they solve a throughput problem a home grower does not have. Six cuttings in a takeout container will root.
How to take a cutting, step by step
Have everything ready before you cut. Cubes soaked and drained, gel decanted into a small cup so you are not dipping into the bottle and contaminating it, dome within reach. From the moment the blade goes through the stem, the cutting is losing water it cannot replace.
Step 1: pick the branch
Choose a healthy branch on a plant in vegetative growth. Look for a firm stem with solid white pith inside rather than a hollow one - hollow stems are common on fast, soft growth and they root poorly.
Position on the plant does not matter. This is one of the most repeated rules in cannabis growing and it does not survive testing. The Guelph trial compared cuttings taken from apical and basal positions on the stock plant and found "little effect on rooting success or quality."1 A separate trial across four varieties found apical and medial cuttings both rooted at around 75%, with no significant difference.10 Take the branches you can spare.
Step 2: cut long
Cut about 8 inches (20 cm) back from the growing tip, taking the cut just below a node. A node is the joint where leaves and side branches attach to the stem - it is where roots will form, because the tissue there is already primed to produce new organs.
Length matters, but not in a straight line. A 2026 trial across 240 cuttings tested four lengths: 20 cm rooted at 53.3%, clearly ahead of 15 cm at 31.7% and 10 cm at 33.3% - though 5 cm cuttings beat both middle lengths at 38.3%.6 Only the 20 cm result separated cleanly from the rest, which is why we recommend the top of the tested range rather than a span. In the same trial, cuttings that still had their apical growing tip attached rooted at 45.0% against 33.3% for cuttings without one. Keep the tip. Cut long.
Cut at roughly 45 degrees. This is convention rather than evidence - no cannabis trial has tested cut angle - but an angled cut exposes more surface area and stops the stem sitting flat on the bottom of the hole, and it costs nothing.
Step 3: strip the lower leaves, and leave the top ones alone
Remove leaves and side shoots from the bottom third of the cutting so that at least one bare node will sit below the surface of the medium. Everything below the medium line comes off.
Now the important part: do not trim the leaves you are keeping. Nearly every cloning guide tells you to snip the tips off the remaining fan leaves, on the theory that it reduces water loss. It was tested. Removing one third of the leaf tips dropped rooting success from 71% to 53%, with no improvement in root quality to compensate.1 That is 18 percentage points of success thrown away for a folk theory.
Leave at least three fully expanded leaves on the cutting. In the same trial, three leaves gave 15% better root quality than two, though it did not change the raw rooting percentage.1 The leaves are the engine: they are what powers root formation. Cutting them is like siphoning the fuel tank to make the car lighter.
Step 4: wound the base (optional, and it works)
Lightly scrape a thin strip of the outer tissue off the bottom half-inch (about 1.3 cm) of the stem with your blade. This one is folklore that turned out to be real: in a trial across roughly 836 cannabis cuttings, wounded stems were 162% more likely to root, and rooted 1.5 days earlier, than unwounded ones.4 Wounding improved the odds of rooting in all three cultivars tested; in one of them it just did not make anything happen faster.
Scrape, do not split. A soft herbaceous stem that has been sliced open is an entry point for rot, which is the most common way cuttings die.
Step 5: dip and stick
Dip the bottom inch (2.5 cm) into rooting gel, then push the cutting into a pre-soaked cube deep enough that at least one stripped node sits inside the medium. Firm the medium gently around the stem so there is no air gap. An air pocket around the stem is a cutting that dries out from the inside.
If you are using rockwool, soak the cubes first in water adjusted to pH 5.5 to 6.0 and let them drain. Rockwool comes out of the bag alkaline, around pH 7 to 8, which will stall root initiation. Drain them properly too - a cube sitting in standing water rots the stem before it roots.
Step 6: dome it
Into the tray, on with the dome, under low light. Do not water the cuttings from above and do not soak the tray. The cube is already wet and the dome will hold the humidity.
If you are cutting from more than one plant
Do all the cuttings from one mother, then park the blade in 10% bleach for two minutes, rinse it, and move to the next. This is the step that stops one infected plant becoming an infected library.
The rooting environment
A cutting has no roots and cannot drink. Everything in this section is about keeping the leaves from losing water faster than the stem can pull it up, for as long as it takes roots to appear.
| Setting | Target | Why |
|---|---|---|
| Air temperature | 72 to 78°F (22 to 26°C) | Warm enough to drive cell division, cool enough not to spike transpiration |
| Root zone temperature | around 79 to 82°F (26 to 28°C) | Warmer at the base than in the air pulls root growth forward |
| Humidity, days 1 to 4 | 90% or higher | Near-saturated air means the leaf loses almost nothing |
| Humidity, days 5 to 10 | stepped down toward 70 to 75% | Gently reintroduces transpiration as roots start working |
| Light schedule | 18 hours on, 6 off | Long-day signal, keeps everything vegetative |
| Light intensity | low, roughly 50 to 150 PPFD | Enough to photosynthesize, not enough to cook a plant that cannot drink |
On light intensity, and what the research actually found
Light during propagation is the setting growers worry about most and it matters least for the outcome they care about. A 2025 trial rooted cannabis cuttings under 50, 150 and 250 PPFD and found no effect of light intensity on the fraction of cuttings that rooted. Higher light did increase root and cutting dry mass at 14 days - the clones were bigger - but it did not change how many of them made it.5
So keep it low and keep it far away. The risk is not too little light, it is a cutting with no roots sitting under a light that makes it transpire. If you have one LED and no dimmer, hang it two to three times its normal distance, or run it against the wall and let the tray catch the spill.
On misting, which is more contested than it looks
Every guide tells you to mist the leaves daily. Commercial propagators frequently do not, because water sitting on a leaf inside a warm sealed dome is how you get fungal problems and rot. If your dome is holding humidity properly you do not need to mist at all - the point of the dome is that the cutting is already sitting in near-saturated air. Mist if the dome is leaky and the leaves look limp. Do not mist on schedule.
Root zone warmth
A propagation heat mat under the tray is the one accessory worth buying. Cannabis cuttings were rooted successfully at a root zone temperature of 26°C (79°F) in one trial,4 and a hemp study reported better quality rooted cuttings at 27.8°C (82°F).17 If your room sits at 68°F (20°C), a mat is the cheapest improvement available to you. If your room is already 78°F (26°C), skip it.
Day by day: the first three weeks
Here is what the tray should be doing, and in what order. The order is fixed. The pace is not: a fast cultivar can run this whole sequence in 10 days, a slow one can take five weeks. Treat it as a shape, not a schedule.
| Typical day | What is happening | What you do |
|---|---|---|
| Day 0 to 1 | The cutting droops. This is normal and not a failure | Dome closed. Low light. Leave it alone |
| Day 2 to 3 | Leaves lift back to horizontal as the stem re-establishes water uptake | Dome closed. Check the cube is damp, not wet |
| Day 4 to 6 | Callus, a pale swelling of undifferentiated tissue, forms at the cut | Start venting the dome briefly once a day |
| Day 7 to 10 | Root initials form inside the stem. Nothing visible yet | Vent longer each day. Still no feeding |
| Day 10 to 14 | First white root tips push out of the cube | Dome mostly open. Begin very dilute feed |
| Day 14 to 21 | Roots spread through the cube. New growth appears at the tip | Dome off. Transplant when roots show on several faces |
The single most useful thing to know: the range is real, and it is wide. Published cannabis propagation trials assess their cuttings at 12 to 21 days.12615 The one trial that recorded the actual date each cutting rooted rather than scoring on a fixed day found a spread of 5 to 35 days across genotypes.4 So an easy cultivar in good conditions can show roots on day 5 and be ready to transplant by day 10. A stubborn one can take five weeks. Most sit in the middle, which is why 12 to 21 days is the number to plan around, and why being on day 12 with nothing showing means you are on schedule, not failing.
When to start feeding
Not until roots exist. A cutting with no roots cannot take up nutrients through the stem, so anything you add before then is just salt sitting in the cube. Once white tips are showing, a quarter-strength vegetative feed at pH 5.8 to 6.2 is plenty. The cutting is not hungry yet; it is still living off the leaves you were careful not to trim.
How to tell it has rooted without killing it
Three signals, in order of reliability.
1. Roots visible on the outside of the cube. The only certain one. White root tips pushing through the sides or bottom of a rockwool cube means the job is done. This is what you are waiting for.
2. New growth at the tip. A cutting that starts pushing a new set of leaves is a cutting that has found a water supply. It is not proof - a cutting can push a little growth on stored reserves - but combined with a firm stem it is a strong signal.
3. The tug test, used sparingly. Pinch the stem and pull very gently upward. Resistance means roots have anchored it. A cutting that lifts free has not rooted. This works, but every tug risks snapping the fragile new roots, so do it once, on one representative cutting, not on all of them every morning.
What is not a signal: white bumps on the submerged part of the stem. Those are root primordia, the beginnings of roots, and they are a good sign but not the finish line. They are also easy to confuse with the pale callus tissue that forms first, which is not a root and will not become one on its own.
When clones fail: seven failure modes
Expect to lose some. Even in controlled trials with everything optimized, whole treatments came in under 40%. Take more cuttings than you need and read the failures rather than mourning them.
| What you see | What it usually is | What to do |
|---|---|---|
| Everything drooped on day 1 to 2 | Normal. The stem has not re-established water uptake yet | Nothing. If they are still flat on day 4, check the dome is sealing |
| Stem brown and soft where it enters the cube | Rot from an oversaturated cube or a split stem | Discard it. Squeeze cubes out before sticking; scrape, never split |
| Lower leaves yellowing by day 7 to 10 | Often normal - the cutting is cannibalizing reserves | Nothing, if the tip is still green. Do not feed to fix it |
| Leaf edges dry and crisp, cutting shrivelled | Transpiration is beating uptake. Too much light or a leaky dome | Lower the light, raise it further away, reseal the dome |
| White fuzz on the medium or leaves | Fungal growth from stagnant saturated air | Vent more, remove affected cuttings, stop misting |
| Green algae on the cube surface | Light plus a permanently wet surface. Ugly, mostly harmless | Let the surface dry between waterings. It competes but rarely kills |
| Day 21, no roots, stem still firm and green | Slow cultivar, cool root zone, or no hormone used | Wait. A firm green stem is still alive. Add bottom heat |
The two that actually kill are at opposite ends of the same axis: too wet at the base and too dry at the leaf. A cutting wants a damp cube and saturated air, not a soaked cube and a mist bottle.
The other failure nobody tells you about
They root, you pot them up, and then they sit there doing nothing for a week or two. This is normal. The root system that filled a rockwool cube is tiny, and it has to colonize a much larger volume before the plant can support visible top growth. It is not stalled, it is building. Resist the urge to feed it harder, which is how a healthy young clone gets burned.
Transplanting a rooted clone
Move it when roots are visible on more than one face of the cube, not the moment the first tip appears. A cutting with roots on one side is a cutting that will lose most of them to handling.
Handle the cube, never the stem. The stem is the one part of the plant that cannot be replaced and the roots are attached to it by tissue about as robust as wet paper. Lift by the medium.
Plant the whole cube. Do not try to peel rockwool off a root system - you will strip the roots with it. Bury the cube so its top is level with or just under the surface of the new medium; a cube left proud of the surface wicks and dries out. Go into something small first. A rooted cutting in a 3-gallon pot is a small root system in a large volume of wet medium, which is how clones drown.
A nursery cup is the right first home. Self-draining nursery pots like the BudCups hold roughly 16 fluid ounces (475 ml) of medium, which is the right volume for a freshly rooted clone for its first two to three weeks, and the removable bottom plate lets you check the root ball without disturbing the plant. Pop the plate, look, put it back - the BudCups guide covers the rest. When roots have colonized that volume, move up into fabric - the transplanting guide covers the volume step-up rule and the timing.
Keep humidity up for the first few days after transplant and keep light modest. You have just asked a plant with a small root system to survive a move.
Cloning myths that cost you rooting percentage
Sorted by how much they cost you.
| The claim | What the evidence says |
|---|---|
| Trim the tips off the fan leaves to reduce water loss | Measurably harmful. Rooting success fell from 71% to 53% when one third of the leaf tips were removed, with no gain in root quality1 |
| Take cuttings from the bottom of the plant, they root better | No effect. Apical and basal cuttings rooted equally in a Guelph trial1, and apical and medial cuttings rooted equally in a separate four-variety trial10 |
| Retire mother plants after about six months | No effect on rooting from mothers 65 to 178 days old5 |
| Stronger rooting hormone gives better results | Backwards. 8,000 ppm IBA reduced both rooting percentage and quality against 3,000 ppm and against nothing at all3 |
| Willow water is a natural rooting hormone as good as the commercial stuff | A 0.2% IBA gel rooted 2.1 times more cuttings than a 0.2% willow extract gel, at 1.6 times the root quality1 |
| Cut the stem underwater to stop air bubbles entering | Untested in cannabis. Imported from cut-flower handling, where the problem is different. Harmless, just unnecessary |
| Cut at exactly 45 degrees | Never tested in cannabis. Sensible, free, and not the reason your clones failed |
| Vitamin B-1 helps cuttings root | A long-debunked horticultural myth. Thiamine does not stimulate adventitious rooting |
| Use gel and then powder to get the best of both | No mechanism, no evidence, and application method showed no significant difference in the one trial that compared methods head to head2 |
| Clones are an identical copy in every way | Identical genetics, not identical plants. Expression tracks environment, and the belief is exactly what lets a viroid spread unnoticed |
One that is not a myth: rooting in a glass of water works. It is slower and the roots that form are more fragile than roots grown in a medium, and no controlled cannabis trial has compared it to rockwool, so we cannot give you a number. But water rooting is real and it is free, and for a grower taking two cuttings to see whether they can, it is a fine place to start.
Can you clone an autoflower?
You can take the cutting and it will often root. It just will not do what you want.
Photoperiod cannabis flowers when nights get long, which is why you can hold a mother in vegetative growth indefinitely under 18/6. Autoflowering plants do not read the light schedule. The trait comes from a mutation in a circadian clock gene, a splice site mutation in CsPRR37, which removes the plant's ability to respond to daylength.14 It is heritable and it is in every cell.
A cutting off an autoflower carries that mutation, and it also carries the mother's internal clock position. The mother is on a fixed schedule from germination. The cutting is not a fresh start - it is a piece of a plant that is already partway through its life, and it will run to flower on the mother's timetable while trying to root, with a root system a fraction of the size it needs. The usual result is a tiny plant that flowers almost immediately and yields close to nothing.
Be clear about what is established and what is not. The genetics are well documented. The practical outcome has never been formally measured in a peer-reviewed study - there is no published trial that clones autoflowers and reports the yield. Every number you will see quoted for it is somebody's grow log. What we can say with confidence is the mechanism, and the mechanism says do not bother. Grow autoflowers from seed, and keep training them gently instead.
Cloning from a flowering plant, and the truth about monster cropping
"Monster cropping" is the practice of taking cuttings from a plant already in flower, rooting them, and reverting them to vegetative growth under 18/6. The claim is that the reverted plant comes back bushy, with many more branch tips than a normal clone, and yields more.
There is no peer-reviewed study of monster cropping. None. Not one trial has taken cuttings from flowering cannabis, revegged them, and measured branching or yield against a control. That is not a gap in our reading, it is a gap in the literature - as recently as 2019, a University of Guelph paper stated plainly that "there are no peer-reviewed methods to propagate flowering plants," and the method that team developed was laboratory tissue culture on a growth regulator, not cuttings in rockwool.13
The one adjacent experiment cuts against the practice as it is usually done. A 2026 study pre-treated cannabis plants with short days before taking cuttings. One week of 12/12 produced the highest number of adventitious roots. Two weeks of 12/12 suppressed root formation completely, which the authors attributed to oxidative stress - hydrogen peroxide in the tissue reached 240 µmol per gram of dry weight.12
Read that against what monster croppers actually do, which is take cuttings two to four weeks into flower. That is well past the point where this study found rooting collapsed.
So: taking a cutting from a plant that has just flipped is plausible and may even root slightly better than normal. Taking one from a plant three weeks into flower is working against the only relevant evidence there is. And whether the revegged plant then produces the bushy monster of the folklore is simply unmeasured. If you try it, treat it as an experiment, not a technique, and do not build a grow plan around it.
The legitimate use of the same idea is preservation: you flowered a plant, it turned out to be the best thing in the tent, and you did not keep a cutting. Taking one late is a long shot worth attempting because the alternative is losing the genetics entirely. Expect slow rooting, expect weird growth for a few weeks during reversion, and expect to lose some.
Does a clone line degrade over time?
Growers have said for decades that clone lines get tired - that after enough generations, the plants lose vigor and potency. It is one of the most persistent beliefs in cannabis and the honest answer has three parts.
Part one: mutations really do accumulate. Deep sequencing of a single cannabis plant found over six million sequence variants between samples taken from its bottom, middle and top, with the top carrying by far the most - roughly 600,000 variants unique to the top against 148,000 unique to the bottom. Four genes carrying more than double the average nucleotide diversity sat in the cannabinoid and terpene biosynthesis pathways.11 A plant is a genetic mosaic, not a single uniform genome, and cuttings sample different parts of it.
Part two: almost nobody has measured the thing growers describe. There is exactly one attempt, and it is thin. A 2022 study propagated five hemp varieties through six to ten successive cutting generations and measured 17 cannabinoids in each generation, finding no significant difference across generations and no trend.18 It did not measure yield or vigor, and it was not published in a leading journal, so treat it as a single weak data point rather than a settled answer. But it is the only serial-generation evidence that exists, and it points away from degradation rather than toward it. The mutation studies show the mechanism is plausible; nobody has shown the effect.
Part three, and this is the part that matters practically: the likelier culprit is a pathogen, not genetics. A clone line that visibly declines - smaller plants, weaker smell, airy flower - is describing the symptom list for hop latent viroid almost exactly. HLVd has documented effect sizes of 12 to 42%,8 spreads through exactly the propagation route in question, and hides asymptomatically for weeks. Somatic mutation is a slow theoretical drift. A viroid is a fast, measurable, testable, fixable problem.
If your line has gone downhill, test it before you blame genetics.
Clones or seeds
Short version, because the seeds guide covers the choice properly.
Clones give you a known plant, guaranteed female, no juvenile phase, and a faster route to flower. They also inherit every problem the mother has, and they require you to have access to a healthy plant in the first place.
Seeds start clean, arrive with no pest or pathogen history, are legal to ship almost everywhere clones are not, and teach you the full plant lifecycle. They also cost you time, and a percentage of them will be male unless they are feminized. For a first grow we recommend seeds, and germinating them properly is a smaller skill to learn than propagation.
The two are not rivals. The normal arc is that you grow from seed, find a plant you like, and clone that. Cloning is how you keep a good result rather than how you start.
Grower jargon decoder
Terms used above, and in every other cloning guide, usually without explanation.
| Term | What it means |
|---|---|
| Node | The joint on a stem where leaves and side branches attach. Roots form most readily at or near a node |
| Apical meristem | The growing tip at the very top of a stem, where new cells are produced. Keeping it on a cutting improves rooting |
| Adventitious roots | Roots that form somewhere roots do not normally grow, such as out of a cut stem. This is what cloning is |
| Auxin | The plant hormone that triggers root initiation. Rooting gels are synthetic auxin |
| IBA / NAA | Indole-3-butyric acid and naphthaleneacetic acid, the two synthetic auxins in commercial rooting products |
| Callus | Pale undifferentiated tissue that forms over a cut. A normal step before roots, but callus alone is not roots |
| Root primordia | The bumps that appear on a stem before roots emerge. The beginnings of roots, not roots |
| Pith | The soft core inside a stem. Solid white pith is a good cutting; hollow is a poor one |
| Transpiration | Water evaporating out of the leaves. The thing a rootless cutting cannot afford |
| Hardening off | Gradually lowering humidity so a rooted cutting can cope with normal room air |
| Damping off | A fungal collapse of the stem at the medium line, usually from overwatering |
| PPFD | Photosynthetic photon flux density, a measure of usable light intensity reaching the plant |
| Photoperiod plant | A plant that flowers in response to long nights. Can be held in veg indefinitely |
| Stock plant / mother | A plant kept permanently in vegetative growth to supply cuttings |
| Viroid | An infectious agent smaller and simpler than a virus - naked RNA with no protein coat |
Frequently asked questions
Do I need a mother plant to take clones?
No. A mother plant is simply a plant you have chosen to keep in vegetative growth so you can keep cutting from it. You can take cuttings from the plant already growing in your tent and flower that plant afterward. The only requirement is that the plant is in vegetative growth under a long-day light schedule when you take the cutting.
Will a clone be the same sex as the mother plant?
Yes. A clone is a genetic copy, so a cutting from a female plant is always female. This is the main practical reason growers clone: there is no sexing period, no males to cull, and no wasted weeks. It also means a clone inherits everything else the mother carries, including pests and systemic pathogens.
How long does it take for cannabis clones to root?
Twelve to twenty-one days is normal, and some cultivars take longer. Published cannabis propagation trials assess their cuttings at 12 to 21 days. The one trial that recorded each cutting's actual rooting date found a spread of 5 to 35 days across genotypes, so the widely repeated five to seven days is the best case for an easy cultivar, not the schedule to plan around. If you are on day 12 with nothing showing, you are on schedule.
Do I really need rooting hormone to clone cannabis?
Not strictly, but use it anyway. In one trial of 960 cuttings, using any auxin hormone improved rooting success 15 to 18 times over using none, while the specific product, concentration and application method made no significant difference. In easier conditions the gap closes: in another trial, cuttings rooted at 76% to 88% whether or not they got hormone. Rooting without it is possible. It is just a worse bet. Buy a standard gel.
Should I cut the tips off the leaves when taking a clone?
No, and this is the most common harmful piece of cloning advice. In a University of Guelph trial, removing one third of the leaf tips dropped rooting success from 71% to 53% with no improvement in root quality. Leave at least three fully expanded leaves on the cutting and leave them whole. The leaves power root formation.
Where on the plant should I take cuttings from?
It does not matter as much as you have been told. One trial compared cuttings taken from the top and the bottom of the stock plant and a second compared the top and the middle, and neither found a meaningful difference in rooting success or quality. Take branches you can spare. What does matter is length: 20 cm cuttings rooted at 53% against 31% to 38% for every shorter length tested.
Can you clone cannabis in a glass of water?
Yes, it works. Roots form more slowly and are more fragile than roots grown in a solid medium, and they need careful handling at transplant. No controlled cannabis trial has compared water rooting against rockwool, so there is no number to give you. For someone taking two cuttings to learn the process, it is a perfectly reasonable place to start.
Do clones need light while they are rooting, and should it be 18/6 or 24/0?
They need light, but far less than growers assume. Use 18 hours on and 6 off, at low intensity, well away from the dome. A 2025 trial rooted cuttings under 50, 150 and 250 PPFD and found no difference in how many rooted. Higher light produced bigger clones but not more of them. Too much light dries out a cutting that cannot drink.
Why did my clones wilt the day after I took them?
That is normal. A cutting has no roots, so for the first day or two it loses water faster than the stem can pull it up and the leaves droop. They usually lift back to horizontal by day two or three. If they are still flat on day four, your dome is not holding humidity or your light is too strong.
How do I know when a clone has rooted?
The only certain sign is white root tips visible through the sides or bottom of the cube. New growth at the tip is a strong secondary signal. A gentle tug that meets resistance confirms it, but every tug risks snapping new roots, so test one representative cutting rather than the whole tray. White bumps on the buried stem are root primordia, not finished roots.
It has been two weeks and there are no roots. Did I fail?
Probably not. If the stem is still firm and green at the base, the cutting is alive and working. Cultivar is a large factor, and some varieties root at roughly 60% while others hit 90% under identical conditions. Add bottom heat to bring the root zone to about 79°F (26°C) and wait. Only discard cuttings that have gone brown and soft at the base.
Can you clone an autoflower?
The cutting will often root, but it will not give you a useful plant. Autoflowering comes from a mutation in a circadian clock gene that removes the plant's response to daylength, so the clone cannot be held in vegetative growth. It runs to flower on the mother's timetable with a tiny root system and yields almost nothing. Grow autoflowers from seed.
Can you take clones from a flowering plant?
You can, and it is worth attempting to rescue genetics you would otherwise lose, but expect slow rooting and several weeks of odd growth while the plant reverts. Monster cropping, the practice of doing this deliberately for higher yields, has never been tested in any peer-reviewed study. The one related trial found that two weeks of short days before cutting suppressed root formation completely.
How many times can you clone a clone before it degrades?
Nobody really knows. One 2022 study cloned five hemp varieties through six to ten successive generations and found no significant change in cannabinoid content, but it did not measure yield or vigor and it is a single weak data point. Somatic mutations do accumulate within a plant, including in cannabinoid and terpene pathway genes. If your line has declined, test it for hop latent viroid first.
What is hop latent viroid and should a home grower worry about it?
It is an infectious strand of RNA that spreads almost entirely through propagation and causes plants to grow small and weak with poor potency, a problem growers call dudding. Infected mothers are frequently asymptomatic and pass it to every cutting. It averaged 25.6% of samples across 15,947 tests in Canadian facilities. If you plan to keep one genetics library for years, test the mother once.
References
- Caplan D, Stemeroff J, Dixon M, Zheng Y (2018). Vegetative propagation of cannabis by stem cuttings: effects of leaf number, cutting position, rooting hormone, and leaf tip removal. Canadian Journal of Plant Science 98(5):1126-1132. doi:10.1139/cjps-2018-0038
- Campbell SM, Anderson SL, Brym ZT, Pearson BJ (2021). Evaluation of substrate composition and exogenous hormone application on vegetative propagule rooting success of essential oil hemp (Cannabis sativa L.). PLOS ONE 16(7):e0249160. doi:10.1371/journal.pone.0249160
- McLeod A, Vining K, Hoskins T, Contreras R (2022). Impact of indole-3-butyric acid concentration and formulation and propagation environment on rooting success of 'I3' hemp by stem cuttings. HortTechnology 32(3):321-324. doi:10.21273/HORTTECH05016-21
- Campbell LG, Naraine SGU, Dufresne J (2019). Phenotypic plasticity influences the success of clonal propagation in industrial pharmaceutical Cannabis sativa. PLOS ONE 14(3):e0213434. doi:10.1371/journal.pone.0213434
- Holweg MMSF, Sae-Tang W, Wang Y, Kohlen W, Heuvelink E, Marcelis LFM (2025). Mother plant age and light intensity minimally alter adventitious root formation in medicinal cannabis. HortScience 60(11):2034-2046. doi:10.21273/HORTSCI18772-25
- Esposito LGA, Rodrigues C, Pereira P, Teixeira HM, Silva D (2026). Analysis of marijuana (Cannabis sativa L.) cuttings: morphological and colorimetric traits as predictors for optimization of vegetative reproduction. Plants 15(3):440. doi:10.3390/plants15030440
- Punja ZK, Scott C, Tso HH, Munz J, Buirs L (2025). Transmission, spread, longevity and management of hop latent viroid, a widespread and destructive pathogen affecting cannabis (Cannabis sativa L.) plants in North America. Plants 14(5):830. doi:10.3390/plants14050830
- Punja ZK, Wang K, Lung S, Buirs L (2024). Symptomology, prevalence, and impact of hop latent viroid on greenhouse-grown cannabis (Cannabis sativa L.) plants in Canada. Canadian Journal of Plant Pathology 46(2):174-197. doi:10.1080/07060661.2023.2279184
- Kurtz LE, Borbas LN, Brand MH, Lubell-Brand JD (2022). Ex vitro rooting of Cannabis sativa microcuttings and their performance compared to retip and stem cuttings. HortScience 57(12):1576-1579. doi:10.21273/HORTSCI16890-22
- Voisin A, Gomez S, Colmann Lerner E, Aranda Mosquera O, Villarreal B, Sharry S, Weber C (2024). Effect of cutting position in the mother plant and use of auxin on the rooting efficiency of cuttings in four varieties of Cannabis sp. Agricultura 129-130:265-273. journals.usamvcluj.ro
- Adamek K, Jones AMP, Torkamaneh D (2022). Accumulation of somatic mutations leads to genetic mosaicism in cannabis. The Plant Genome 15(1):e20169. doi:10.1002/tpg2.20169
- Hahm S, Kim B, Lee Y, Kim S, Ha S, Park J (2026). Short-day pre-treatment coordinates adventitious rooting and flowering in cannabis cuttings. Industrial Crops and Products 245:123307. doi:10.1016/j.indcrop.2026.123307
- Piunno KF, Golenia G, Boudko EA, Downey C, Jones AMP (2019). Regeneration of shoots from immature and mature inflorescences of Cannabis sativa. Canadian Journal of Plant Science 99(4):556-559. doi:10.1139/cjps-2018-0308
- Leckie KM, Sawler J, Kapos P, MacKenzie JO, Giles I, Baynes K, Lo J, Baute GJ, Celedon JM (2024). Loss of daylength sensitivity by splice site mutation in Cannabis pseudo-response regulator. The Plant Journal 118(6):2020-2036. doi:10.1111/tpj.16726
- Weingarten M, Mattson N, Grab H (2024). Evaluating propagation techniques for Cannabis sativa L. cultivation: a comparative analysis of soilless methods and aeroponic parameters. Plants 13(9):1256. doi:10.3390/plants13091256
- Mejia-Londono HA, Barrera-Sanchez CF, Cordoba-Gaona OJ (2023). Asexual propagation in female plants of cannabis (Cannabis sativa L.). Revista Colombiana de Ciencias Horticolas 17(2):e16046. doi:10.17584/rcch.2023v17i2.16046
- Cockson P, Barajas G, Whipker B (2019). Enhancing rooting of vegetatively propagated Cannabis sativa 'BaOx' cuttings. Journal of Agricultural Hemp Research 1(1):2. doi:10.61611/2688-5182.1000
- Perrone C, Kline P, DuBois J (2022). Cloning successive generations of industrial hemp (Cannabis sativa) to assess cannabinoid profiles. Agricultural Sciences 13(12):1291-1308. Weaker source than the trials above, noted in the text: a low-tier journal, cannabinoids only, no yield or vigor measured. doi:10.4236/as.2022.1312079
- Deyle L (2024). Assessing the impact of hop latent viroid (HLVd) on hemp (Cannabis sativa) and exploring mitigation strategies. MSc thesis, Colorado State University. A student thesis rather than a peer-reviewed paper, and on hemp rather than drug-type cannabis. mountainscholar.org
About the Author
Henrique Dias is the CEO and co-founder of BudTrainer. He is a mechanical engineer who also holds a graduate certificate in Commercial Cannabis Production from Niagara College in Ontario, Canada - the first college-level cannabis cultivation program in North America. Before founding BudTrainer, Henrique worked with Health Canada licensed cannabis producers, where he gained hands-on experience in cultivation, processing, and post-harvest management of cannabis at commercial scale. He started BudTrainer to bring that same level of craft to home growers through better-designed gardening tools and clear, science-backed education.
DISCLAIMER: Everything taught and sold by BudTrainer® is to be used strictly for legal purposes. We do not endorse the production of illegal substances and it is your duty to ensure that you are complying with the law. The words "hemp," "cannabis," "weed," and "marijuana" are used interchangeably to refer to the same plant (legal hemp with less than 0.3% THC) for the purposes of this lesson.
