Sowing seed
Ten to fifteen years to the first flower, and the first six weeks decide most of it. What to sow into, what field capacity feels like in your hand, and which of the alarming things inside the box actually mean something.
11 min read6 sections
Seed is the slowest route to a plant and the only route to a new one. A seedling is roughly ten to fifteen years from its first flower, and buying year-old seedlings buys about one of those years back. What it gives you in exchange is genetics that did not exist before, which is the one thing a can never do.
It is also the only propagation job on this site that wants looking at nearly every day for the first three months. Everything else here you can walk away from.
Buy the seed properly — this is where most failures start
Old seed is the commonest cause of a tray that never comes up, and almost every shop calls its seed fresh. Buy from whoever actually produced it, and ask what year it came off the plant.
- **Open-pollinated seed** is fine if you do not care who the father was and useless if you do. In a garden full of strains, insects will have brought whatever was flowering.
- **Mixed seed** is a real thing. If the seedlings in one packet look nothing like each other, seed from several fruits was combined.
- **Sticky seed** was never washed. Wash it before it goes anywhere near a warm sealed box.
- **The name on the packet is a hope.** Misidentification happens at every scale — one grower has had seed sold as peruvianus grow out as scopulicola. You will not know for years, because that is how long identification takes.
- **Wild-collected habitat seed** is the exception to the open-pollination worry: with no other strain for miles, it is often effectively pure anyway.
Three conditions, and one of them surprises everybody
- **Light on the seed.** These are surface-sown and never buried. Buried seed does not germinate at all, so this is not a preference.
- **Warmth, held steady.** 68 to 86°F, constant, right through germination and for about a month afterwards. A competing figure of roughly 70 to 85°F is saying the same thing.
- **Moisture at field capacity.** Damp all the way through, nothing pooling and nothing dripping. Squeeze a fistful hard and you should raise a bead or two of water, not a trickle.
On the surface
light reaches it · it germinates
Buried
no light · no germination
What to sow into
Two substrates that both work. Pick one and run it exactly as written before you start improvising: the reason beginners get unrepeatable results is that they stitch two methods together without knowing what each step was there for.
Straight vermiculite
sterile · inorganic · pH neutral
Sieved soil and pumice
half and half · 1 to 4 mm · washed
Straight vermiculite
Sterile out of the bag, inorganic, pH neutral and it holds water well — so algae, mould and gnat larvae have nothing to feed on. An inch and a half of it, levelled but not pressed down, with a screened extra-fine layer no thicker than an eighth of an inch on top so the seed lands on a cushion instead of dropping into gaps where no light reaches it.
Sieved soil and pumice
Potting soil put through a quarter-inch sieve to take out bark and sticks — wet sponges, and a fungal risk — then half and half with 1 to 4 mm pumice, washed of its dust. Perlite works, but it eventually floats to the surface and separates out. Never sand: it packs down and stunts them.
Straight vermiculite
Sterile out of the bag, inorganic, pH neutral and it holds water well — so algae, mould and gnat larvae have nothing to feed on. An inch and a half of it, levelled but not pressed down, with a screened extra-fine layer no thicker than an eighth of an inch on top so the seed lands on a cushion instead of dropping into gaps where no light reaches it.
Sieved soil and pumice
Potting soil put through a quarter-inch sieve to take out bark and sticks — wet sponges, and a fungal risk — then half and half with 1 to 4 mm pumice, washed of its dust. Perlite works, but it eventually floats to the surface and separates out. Never sand: it packs down and stunts them.
- **The container** is anything clear, at least four inches tall, that you can punch drainage holes through later. A takeaway box or a lidded clear cup is a perfectly good greenhouse.
- **Seal it with cling film, not the lid.** Lids are not airtight; they lose moisture slowly and let things in.
- **Nothing rich goes in the mix.** No worm castings, no compost, no feed. One grower's castings-enriched mix grew white surface mould that came back after every peroxide treatment.
- **Sanitising is optional; sterilising is not worth it.** Four minutes in the microwave with the moistened mix in its closed container, then an hour covered to cool, is the whole procedure. Real sterilisation needs 356°F for half an hour, drives off ammonia and changes the soil chemistry — skip it.
Sealed, and left alone
condensation on the walls
About two months — it comes off
three rows of spines · holes punched
Sowing
Step 1: Soak the seed — optional, and worth it
An hour in neat 3% hydrogen peroxide, or 1:1 with purified water for six hours, or as weak as 1:10 for twelve. It cleans the seed coat and hydrates and oxygenates the seed, which speeds germination up and lifts a poor batch. Seeds that float and seeds that sink are both fine.
Step 2: Moisten the mix before it goes in
Spray to field capacity, work it through, and tip out anything excess. Mix a whole batch at once if you are filling several containers — uneven moisture between trays is enough on its own to explain wildly different results.
Step 3: Scatter, do not place
Roll the seed off a folded paper crease, brush it off a damp fingertip, or sprinkle it like salt. Placing seeds one at a time is only worth the hour for something rare.
Step 4: Space them about three eighths of an inch apart
Roughly four to six seeds to the square inch. Closer and they stall as soon as they touch; further apart and they lose the shared humidity that keeps them moving.
Step 5: Press, mist, seal
Press lightly so they sit on the surface. Mist with a genuine fine mist — a stream buries and scatters them. There should be no standing water underneath; if there is, prick a pinhole and let it out.
Step 6: Get the heat right
A mat should lift soil temperature by no more than about 10°F and must run on a soil-probe thermostat. Put the probe in an identical dummy container sealed the same way, because a probe in the live tray shifts and uproots seedlings. Foam board underneath sends the heat upwards. No mat? Sow in late spring or summer — cold soil simply refuses.
Six weeks of not opening it
Check daily to every couple of days through the plastic. There should always be condensation on the walls; if it disappears, open, mist and reseal. Emergence usually starts somewhere between day 4 and day 14, most of the tray shows up inside the first month, and stragglers keep arriving after that.
Normal — leave it alone
white fuzz is the first root
Act today
orange or red · a green film
Normal — leave it alone
White fuzz around a seed at about a week is the first root, not mould. A seed casing still riding on a seedling's head stays where it is; it is not worth breaking the seal to pick it off. Seedlings that come up sideways or upside down turn themselves the right way as their roots take hold.
Act today
Orange or red growth on the surface. Open the container and mist it with 3% hydrogen peroxide several times a day until it dies back. A green algal film usually means an unsanitised organic mix in a warm bright box — the tray can often be saved, but the next one gets a cleaner substrate.
Normal — leave it alone
White fuzz around a seed at about a week is the first root, not mould. A seed casing still riding on a seedling's head stays where it is; it is not worth breaking the seal to pick it off. Seedlings that come up sideways or upside down turn themselves the right way as their roots take hold.
Act today
Orange or red growth on the surface. Open the container and mist it with 3% hydrogen peroxide several times a day until it dies back. A green algal film usually means an unsanitised organic mix in a warm bright box — the tray can often be saved, but the next one gets a cleaner substrate.
What you are looking at
Step 1: The seed casing, still on
A tiny green sphere wearing its own shell like a hat.
Step 2: Two horns
The cotyledons — the only two leaves this plant will ever have. Two is normal. Three happens, and means nothing.
Step 3: The first spines
From here it is a very small ribbed column, and it starts looking like a cactus rather than a bean.
Day nought — on the surface
hard · black · about a millimetre
About a week — the first root
white fuzz · root, not mould
The casing, still on
two horns wearing a hat
Blue is not a seedling trait. It arrives after some years, so seedlings out of a famously blue plant look plain green for a long time and nothing has gone wrong. Species that are spineless as adults still carry spines as seedlings and lose them with age.
Quick check
Did any of that land?
Question 1: How deep do the seeds go?
Question 2: It is week one and there is white fuzz around some of the seeds. What is it?
Question 3: Your seedlings have gone pink at the tips. What is the likeliest cause?