Breeding I: How Strains Happen

A grower crosses the two plants he loves with a paintbrush, gets four seeds, grows them out, and ends up with four plants that have nothing in common with each other and barely anything in common with the parents. He thought he’d bred a strain. He’d bought a lottery ticket. The cross took ten minutes. The part he skipped — selection — takes years. This lesson is about what breeding actually is, so you can read a seed-bank page like someone who understands what they’re buying.

What You Need to Know

Breeding is selection, not crossing

The single most important correction in this whole lesson: making a cross is pollination, not breeding. Pollen on a pistil — a paintbrush and a steady hand — anyone can do it on their first attempt. Breeding is what comes after: growing out large populations, evaluating every plant, culling the ones that miss, keeping the rare few that hit, and doing it again, generation after generation.

What people considered top-shelf in the late 1970s would barely register as mid-grade today. That gap wasn’t crossed into existence. It was selected into existence — by people growing out thousands of plants and keeping the best fraction of a percent, over decades.

Seb’s Corner. Hold the racehorse analogy from the book. You don’t get a Derby winner by putting two horses you like in a field. You test the offspring, breed from the fastest, repeat down the generations. The first cross gets you a horse. The selection gets you a racehorse. Cannabis is the same, only faster — a full generation runs in four to five months instead of years.

The vocabulary, just enough to read a label

• F1 — first filial generation: offspring of two distinct parents. Vigorous (hybrid vigour / heterosis), broadly uniform, but not stabilised. “F1” on a seed page is a generation, not a quality grade.
• F2 — cross two F1s (or self an F1). This is where variation explodes: hidden recessive traits surface and you get plants resembling either grandparent, something in between, or something neither ever showed.
• S1 — first generation from a self-pollination (a female reversed to pollinate herself); all-female, genetics from one parent.
• BX (backcross) — cross an offspring back to a parent to concentrate that parent’s traits. BX1 ≈ 75% that parent, BX2 ≈ 87.5%, and so on.
• IBL (inbred line) — stabilised through enough generations that it breeds true: ten seeds give ten near-identical plants. This is the endpoint, and it’s roughly what the “strain” you bought was built toward.

Seb’s Corner. Why F2 matters to you even if you never breed: when a five-pack throws five different phenotypes, you’re almost certainly growing an unstable cross — an F2 or a poly-hybrid sold as finished. That’s not a defect in your growing. It’s the genetics not being stabilised. Knowing the vocabulary tells you whether inconsistency is your fault or the breeder’s.

Males carry half the genetics and most of the information

Most growers never keep a male alive — feminised seeds exist precisely so you don’t have to think about them. In breeding, that flips: the male contributes half of every seed, yet he never expresses the traits he carries. You can’t look at a male and see what his daughters will smell like. The only way to know is the progeny test — grow out his offspring, evaluate them, and judge the father by his children. A proven male is the most prized, most irreplaceable thing in a serious programme. This is why breeders say they spend more time selecting males than females.

Stabilisation is a balance, not a destination you sprint to

Backcrossing concentrates a target trait, but too much inbreeding causes inbreeding depression — pairing up harmful recessives, giving weaker growth, lower yield, less stress resistance. Good breeders narrow the window without slamming it shut: uniform enough to breed true, diverse enough to stay healthy. Commercial lines commonly run to BX3–BX5. One backcross is a start, not a finish — and a “strain” built on a single backcross is why someone’s five-pack gives five different plants.

Landrace context: where the gene pool comes from

Modern strains are recombinations of older, regionally adapted populations — landraces. Equatorial sativas (Thai, Colombian, African) evolved tall, late-flowering, heat-loving, sometimes carrying THCV. Mountain indicas (Afghan, Hindu Kush) evolved short, fast, cold-tolerant, resin-heavy. The autoflowering trait came from Cannabis ruderalis, a northern subspecies that flowers by age, not photoperiod — bred into quality photoperiod genetics over a decade of selection and backcrossing to make the modern autoflowers that now rival photoperiods. Every trait you take for granted was selected from this older pool by someone with more patience than the paintbrush grower had.

Seb’s Corner. Landraces are the raw genetic library. Once they’re lost — paved over, crossed out, forgotten — that diversity is gone, and with it the traits no one’s bred for yet. Preservation isn’t nostalgia; it’s keeping the source material that future strains get selected from. On the shoulders of giants, and the giants were plants in a field in the Rif and the Kush.

How To Apply This

• Read a seed page properly. F1 means a vigorous unstable cross, not a premium grade. Inconsistency from an unstabilised line is the genetics, not your growing.
• For most home growers: buy, don’t breed. Real selection needs dozens of plants per generation, isolated space for males, and years. A six-plant pick-the-best is statistically meaningless — that’s the book’s hard truth, and it’s honest.
• If you do breed, keep records and keep clones. Number every plant, log smell/structure/flowering time/effect specifically, and keep a clone of anything good. The paintbrush grower’s real mistake was losing the one keeper because he wrote nothing down.
• Respect landrace preservation. If you ever work with one, treat it as a genetic archive, not just an exotic grow.

Watch Out For

• “F1 means best.” It means first-generation and unstable. A quality claim it is not.
• Breeding on tiny numbers. Six F2s and a name is not a strain — it’s one lucky phenotype from a meaningless sample.
• Ignoring the male. Half the genetics, none of the visible expression. Skipping male evaluation throws away most of your breeding information.
• Over-inbreeding. Chasing uniformity past the point of health gives you inbreeding depression — weaker, sicker plants that happen to look alike.

Quiz

Sources

Grow Good Bud, The Grower’s Guide, Chapter 14 (Breeding) and Chapter 16 (Strain Index).

Next: Lesson 7 — Concentrates, and what hash-grade flower demands of the grow.