S4ag Seeds

---
name: s4ag-seeds
description: "Seed saving and selection guidance. Use when the user asks about saving seed, isolation distances, seed cleaning, storage, variety preservation, open-pollinated varieties, or says 'can I save seed from this', 'how do I store seeds', or 'which varieties breed true'."
allowed-tools: [Read]
---

# Seeds

Saving your own seed is the deepest form of food sovereignty available to a farmer. It closes the loop between this season's best plants and next season's planting stock, builds local adaptation over years, and removes a recurring input cost. The core principle is selection: you are not just preserving genetics, you are actively shaping them by choosing which plants you allow to set seed. Every year you save seed, your varieties become more fitted to your land.

**How this skill works:** Each sub-tool pauses at a **Checkpoint** to confirm the assumptions it is about to build on before producing output. A recommendation built on a wrong assumption wastes time and money — confirm the checkpoint before acting. Each sub-tool ends with **Next steps** — the skills worth running once you have acted on this one.

---

## Expert Lineage

*The thinkers whose frameworks underpin this skill — and what they specifically discovered that changes how you farm.*

**Suzanne Ashworth — Seed to Seed**
Ashworth's systematic documentation of crossing risk and isolation requirements for every common vegetable crop remains the practical standard. Her specific contribution: classifying crops by pollination mechanism (wind, insect, self) and then providing minimum and ideal isolation distances for each — numbers that actually hold up in practice rather than theoretical minimums. Her work makes isolation planning tractable for small farms with limited space.

**Carol Deppe — Breed Your Own Vegetable Varieties**
Deppe took seed saving from preservation into active breeding. Her specific finding: farmers selecting for locally adapted performance — rather than for conformity to a heritage description — produce varieties that outperform commercial seed on their own farms within four to six generations. She also documented the population size problem: saving from too few plants causes inbreeding depression in cross-pollinated species. Minimum viable population sizes for common crops come directly from her work.

**William Woys Weaver — Heirloom Vegetable Gardening**
Weaver's historical research traced the lineage of hundreds of heirloom varieties and documented what was lost when industrial seed systems replaced local seed networks. His actionable finding: the genetic diversity within a single heirloom variety is often higher than the diversity between several modern commercial varieties — meaning one well-maintained heirloom line carries more biological resilience than a rack of F1 packets.

**Vandana Shiva — Seed Sovereignty**
Shiva documented the systematic replacement of diverse farmer-saved seed systems with proprietary hybrid and patented varieties, and its effects on crop genetic diversity. Her quantitative contribution: India lost 75% of its rice variety diversity in the thirty years following the Green Revolution. The practical implication for a seed-saving farmer: the act of saving open-pollinated seed is also an act of maintaining a genetic commons that industrial systems are contracting. Understanding this frames what you are doing and why it matters beyond your own farm.

**Elaine Ingham — Epigenetic Soil Adaptation**
Ingham's food web framework has a seed-saving implication that is rarely stated explicitly: seeds saved from plants grown in healthy, biologically active soil carry epigenetic markers reflecting adaptation to that biological environment. Plants grown on a thriving soil food web develop deeper root exudate chemistry and stronger mycorrhizal associations — traits that are partly heritable. Saving seed from the best-performing plants in your healthiest ground selects for biology-adapted genetics, not just above-ground performance.

**Gary Paul Nabhan — Locally Adapted Varieties**
Nabhan's research with traditional Sonoran Desert food systems documented that varieties selected for local conditions over generations outperformed introduced commercial varieties dramatically under the same stress conditions — not because they had higher yield potential, but because they were fitted to the specific drought patterns, soil types, and pest pressures of the region. His finding: commercial varieties bred for average conditions under-perform locally adapted varieties at either stress extreme.

---

## Which tool fits

| You need to... | Tool |
|---|---|
| Understand crossing risk and set up isolation for a specific crop | isolation |
| Choose which plants to save seed from | selection |
| Clean and process harvested seed | cleaning-and-drying |
| Store seed properly and maintain viability | storage |
| Get a quick reference for a specific vegetable or herb | species-guides |

## Routing Decision

- **Just starting out — don't know where to begin with seed saving** → species-guides first, then isolation
- **About to harvest and need to know which plants to pick from** → selection
- **Unsure how far apart to plant to avoid crossing** → isolation
- **Have harvested seed and need to process it** → cleaning-and-drying
- **Have saved seed and need to know how to keep it** → storage
- **Growing multiple varieties of the same species** → isolation first — crossing risk must be managed before harvest

---

## Isolation

*Determines crossing risk for a specific crop and sets up isolation correctly.*

Isolation is the first thing to settle before saving seed from any cross-pollinated crop. If you wait until harvest to think about it, it is already too late — the crossing happened at flowering. The goal is to prevent pollen from unwanted plants reaching your seed crop's flowers. There are three ways to achieve this: distance, barriers, and time.

**Step 1: Classify the crop by pollination mechanism.**

| Pollination type | Typical crops | Isolation approach |
|---|---|---|
| Predominantly self-pollinating | Tomatoes, lettuce, peas, beans, peppers | Minimal isolation needed — low crossing risk |
| Wind-pollinated | Corn/maize, beets, chard, spinach, leeks | Distance is the primary tool — wind carries pollen far |
| Insect-pollinated | Brassicas, cucurbits, carrots, onions | Distance plus timing or physical barriers |

**Step 2: Identify the crossing risk.**

The crossing risk is determined by:
1. What else is flowering in your garden or nearby that is the same species (not just the same vegetable — brassica species cross within the genus)
2. How far away it is
3. What the pollination vector is

**Key crossing groups to know:**

*Brassicas — complex crossing groups:*
- Brassica oleracea (cabbage, broccoli, kale, cauliflower, Brussels sprouts, kohlrabi) all cross with each other — this is one species
- Brassica rapa (turnip, pak choi, Chinese cabbage, mizuna) cross with each other — different species from above
- The two groups do not cross with each other

*Cucurbits — species-specific:*
- Cucurbita pepo: zucchini, acorn squash, most pumpkins, delicata — all cross freely
- Cucurbita maxima: Hubbard, kabocha, Red Kuri, Atlantic Giant — cross within group, not with pepo
- Cucurbita moschata: butternut — mostly crosses within species; very rarely with pepo

*Alliums:*
- Onions and leeks do not cross
- Different onion varieties cross readily with each other via insect pollination

**Step 3: Apply isolation.**

| Pollination type | Minimum isolation | Preferred isolation |
|---|---|---|
| Predominantly self-pollinating | 3–5 m (10–15 ft) | 10 m (30 ft) |
| Insect-pollinated | 150–300 m (500–1000 ft) | 500 m (1600 ft) |
| Wind-pollinated (corn) | 400 m (1300 ft) | 800 m (0.5 mile) |
| Wind-pollinated (beets/chard) | 1 km (0.6 mile) | 2 km (1.2 mile) |

**Alternatives when distance is not available:**

- **Time isolation:** Stagger sowing dates so varieties of the same species flower two to three weeks apart. This works well for brassicas and some cucurbits. Requires planning the season's sowing calendar.
- **Bagging:** Individual flowers or trusses covered with organza or paper bags before they open, hand-pollinated within the bag, and kept covered until the flower has set. Labour-intensive but achieves certainty on self-pollinated species.
- **Cage isolation:** Build a fine-mesh cage over the entire seed crop, introduce pollinators inside (shake a bee-visited flower over the plants, or place a small hive inside). Used for insect-pollinated crops when distance is impossible.
- **Landscape barriers:** A hedgerow, building, or dense windbreak at the mid-point between two varieties reduces crossing by 50–70% for insect-pollinated crops. Not sufficient alone but useful as part of a strategy.

**Population size note (Deppe's principle):**
For cross-pollinated species, you must save from enough plants to maintain genetic diversity and avoid inbreeding depression. Minimums:
- Cross-pollinated crops (brassicas, corn, cucurbits): minimum 6 plants, ideally 12–24
- Self-pollinated crops (tomatoes, beans, peas): can maintain a variety from 3–6 plants

**Checkpoint — confirm before finalising:**
- What specific crop and varieties are involved — and are there any plants of the same species or genus growing nearby (your garden, neighbours, wild relatives)?
- What is the flowering window, and is time isolation feasible as an alternative or supplement to distance?
- How many plants of the seed crop do you have — enough to maintain population health for a cross-pollinated species?

Producing an isolation plan without knowing what else is in flowering range creates a plan that looks right but fails at the field level.

**Output:**
```
ISOLATION PLAN — [Crop] [Variety]

Pollination type: [self / insect / wind]
Crossing risk species nearby: [list any same-species or crossing-group plants in range]

Isolation method:
  Primary: [distance / time / physical barrier]
  Distance required: [metres/feet]
  Time buffer (if used): [weeks between flowering]
  Physical barrier (if used): [bag / cage / hedgerow]

Population size:
  Plants available: [number]
  Minimum recommended: [number]
  Status: [adequate / needs increase]

Action required before flowering:
  [ ] [specific action 1]
  [ ] [specific action 2]
```

**Next steps:**
- Run **selection** (within this skill) — once isolation is secured, the next decision is which plants to mark for seed.
- `/s4ag-seasons` — time isolation requires coordinating sowing dates across the whole planting calendar.
- Run **species-guides** (within this skill) — for crop-specific isolation requirements not covered in the general table.

---

## Selection

*Identifies which individual plants are worth saving seed from — and which to rogue out.*

Selection is where seed saving becomes plant breeding. Every time you choose which plants to save from, you apply selection pressure. Do this consciously and your variety improves over time. Do it randomly or from convenience plants (whatever is left, or whatever is easiest to reach), and the variety drifts toward mediocrity.

**The selection principle:** Select from the best 10–20% of the population. Rogue out (remove before they flower or set seed) the worst 10–20%. Let the middle range contribute to the gene pool.

**Step 1: Mark plants at their best.**
You cannot remember which plants were best at harvest. Mark with bamboo cane, coloured tape, or a tag at the moment the plant shows the trait you are selecting for. For many crops, this is long before harvest.

**What to select for — by crop type:**

*Fruiting crops (tomatoes, peppers, cucurbits, beans):*
- Earliness: first to ripen in your conditions
- Productivity: most fruit per plant, not just largest fruit
- Disease resistance: cleanest foliage through the season
- Fruit quality: flavour, texture, keeping quality — eat from candidate plants before saving

*Leafy crops (brassicas for seed, lettuce, chard):*
- Bolt resistance: latest to bolt in your conditions (select from plants that are last to flower, not first)
- Vigour and leaf quality through the season
- Disease and pest resistance

*Root crops (carrots, parsnips, beets):*
- Root quality: shape, colour, size typical of the variety
- Resistance to splitting and bolting
- Note: root crops are biennial — they must survive winter to flower. Select at harvest, replant the best roots.

*Legumes (beans, peas):*
- Pod fill and productivity
- Disease-free foliage
- Pod dry-down performance: do pods dry cleanly without mould in your climate?

**What to rogue out:**
- Off-type plants (wrong colour, shape, or habit for the variety)
- Plants that bolted early (in crops where bolt resistance is a goal)
- Diseased plants — never save seed from a diseased plant
- Unproductive plants
- Any plants you are unsure about

**Selection for local adaptation — Nabhan/Deppe principle:**
After three to five generations of saving, shift your selection criteria toward local performance rather than variety description. A tomato that is technically off-type but produces reliably in your cool wet summers may be more valuable than one that matches the heritage description but struggles. You are breeding for your land, not for a catalogue photograph.

**Soil health note (Ingham lens):**
Select your seed plants from the best-performing areas of the field — the areas with highest biological activity, best soil structure, and most earthworm activity. Plants grown in active soil food web conditions develop stronger root architecture and more complete nutritional profiles. Seeds from these plants carry better epigenetic starting conditions for next season's germination.

**Checkpoint — confirm before finalising:**
- What traits are most important for this variety in your conditions — yield, earliness, disease resistance, specific quality characteristic?
- Are you selecting from enough plants to maintain genetic diversity (especially for cross-pollinated species)?
- Have you marked plants at their performance peak, or are you trying to recall from memory?

Selecting from whatever is convenient at harvest time — rather than from the marked best performers — produces a population that is selected for nothing in particular.

**Output:**
```
SELECTION RECORD — [Crop] [Variety] [Year]

Primary selection traits:
  1. [trait]
  2. [trait]
  3. [trait]

Plants in population: [total]
Plants selected for seed: [number] ([percentage]%)
Plants rogued out: [number] — reason: [reason]

Selected plant notes:
  Plant 1: [location / marker] — [what made it stand out]
  Plant 2: [location / marker] — [what made it stand out]
  [continue]

Traits observed in rogued plants (for records):
  [any patterns worth noting for future seasons]
```

**Next steps:**
- Run **cleaning-and-drying** (within this skill) — after harvest from selected plants, the next step is processing.
- Run **isolation** (within this skill) — if this season revealed unexpected crossing, review isolation for next year.
- `/s4ag-soil` — if selected plants were concentrated in one area of the field, investigate whether soil health differences are driving performance differences.

---

## Cleaning and Drying

*Processes harvested seed through to clean, dry, storable material.*

The goal of cleaning is to separate viable seed from chaff, plant material, immature seed, and debris. The goal of drying is to bring seed moisture content below the threshold at which fungi, bacteria, and metabolic activity degrade viability. Both must be done before storage.

**When is seed ready to harvest?**

| Crop type | Harvest indicator |
|---|---|
| Self-pollinated (tomatoes, peppers) | Fruit fully ripe — beyond eating stage for seed saving |
| Dry-seeded legumes (beans, peas) | Pods fully dry and rattling on the plant; before autumn rain |
| Brassicas | Siliques (pods) are brown and papery; harvest before they shatter |
| Cucurbits | Fruit left on vine well past eating stage; skin hard and corky |
| Umbellifer seeds (carrots, parsnips) | Seed heads brown; harvest in stages as they ripen unevenly |
| Lettuce | Seed heads fluffy and white; most seeds have matured |
| Corn | Husks dry and brown; cobs left on stalk until first frost where possible |

**Two processing methods:**

**Wet processing** — for seeds enclosed in wet fruit pulp (tomatoes, some cucurbits):
1. Scoop seeds and gel into a container with an equal volume of water.
2. Leave to ferment at room temperature for 2–3 days — the gel coat breaks down and fungal pathogens are inhibited.
3. Add more water, stir vigorously — viable seeds sink, debris floats.
4. Pour off the floating material. Repeat until water runs clear.
5. Spread seeds on a non-stick surface (glass, ceramic plate, wax paper — not paper towels, which stick).
6. Dry at room temperature, 20–25°C, with airflow. Stir daily to prevent clumping.

**Dry processing** — for seeds in dry pods, siliques, or heads:
1. Hang or spread harvested material in a warm, dry, ventilated space. Avoid direct sunlight.
2. Allow material to fully dry — this may take 1–3 weeks depending on conditions.
3. Thresh by hand, in a pillow case (beat against a wall), or with a modified mechanical thresher.
4. Winnow to separate seed from chaff: pour from height in a gentle breeze, or use a fan. Heavy seeds fall; chaff blows away.
5. Screen through mesh sieves to separate by size where needed.

**Checking dryness:**

The bite test: hard seeds (beans, peas, corn) should be hard enough that you cannot dent them with your thumbnail — if they dent, they need more drying. Smaller seeds (brassicas, lettuce, carrots) should not crush easily between fingernail and thumbnail.

Target moisture content for storage:
- Most seeds: 5–8% moisture content
- Long-term storage: below 6%
- Beans and corn: below 12% for short-term; 8% for long-term

If seeds were harvested in wet conditions, drying takes priority over everything else. Fungal damage from damp storage destroys entire seed lots within weeks.

**Food web note:**
Fermentation in the wet process is a microbial process — the lactic acid bacteria and yeasts that break down the gel coat are doing biological work. Do not rush the fermentation with heat or chemical additives. The organisms doing the work are the same populations Ingham documents as beneficial in healthy soil systems — let them complete the job.

**Labelling — do it before you process, not after:**
Every container and every seed lot must be labelled before it is separated from its parent plant. Label with crop, variety, selection notes, and date. Unlabelled seed lots are worthless — you cannot remember what was what six months later.

**Checkpoint — confirm before finalising:**
- Was the seed harvested at full maturity, or did it come in early due to weather? (Immature seed has lower viability and needs longer drying.)
- What conditions are available for drying — temperature, airflow, humidity? High humidity environments need a dehumidifier or silica gel at the drying stage.
- Is the seed lot large enough to be worth processing, or should it be bulked with another season's harvest first?

Processing immature or incompletely dried seed and then sealing it in storage is the single most common cause of total seed lot failure.

**Output:**
```
SEED PROCESSING RECORD — [Crop] [Variety] [Year]

Harvest date: [date]
Harvest condition: [wet / dry / marginal — notes]
Processing method: [wet / dry]

Drying:
  Location: [where dried]
  Duration: [days]
  Dryness check: [result of bite/thumbnail test]

Cleaning result:
  Estimated seed count: [number or weight]
  Quality assessment: [excellent / good / fair / poor — notes]

Labels applied: [yes / no]
Processing date: [date]
```

**Next steps:**
- Run **storage** (within this skill) — processed seed needs to go into correct storage immediately.
- `/s4ag-composting` — the wet processing waste (gel, chaff, pulp) is a valuable compost input.
- Run **species-guides** (within this skill) — if you encountered unexpected difficulty, the species guide may clarify crop-specific processing requirements.

---

## Storage

*Sets up correct storage conditions to maintain seed viability for as long as needed.*

Viability loss is driven by heat and moisture. The interaction is multiplicative — high heat and high moisture together destroy viability in weeks; low temperature and low humidity extend it by decades. The rule of thumb from seed bank science: for every 1% decrease in seed moisture content, storage life doubles. For every 5°C decrease in storage temperature, storage life doubles again.

**What you need:**

1. **Dry seed** — below 8% moisture for most species; below 6% for long-term storage.
2. **Airtight containers** — glass jars with rubber-seal lids, vacuum-sealed bags, or purpose-made foil seed packets. Avoid paper envelopes for long-term storage in humid climates — they equilibrate with ambient humidity.
3. **Moisture absorbers** — silica gel in small packets placed in the storage container. Reactivate used silica gel by heating in an oven at 120°C for 1 hour. A colour-indicating type (turns pink when saturated, blue when dry) tells you when it needs recharging.
4. **A cool, dark, stable-temperature location** — a consistent 10–15°C is better than a refrigerator with frequent door-opening in humid climates. A dedicated chest freezer is ideal for long-term storage of fully dry seed.

**Storage locations by priority:**

| Option | Conditions | Suitable for |
|---|---|---|
| Cool room / cellar | 10–15°C, stable, dark | 2–5 year storage of most species |
| Refrigerator | 4–7°C, controlled humidity with silica gel | 5–10 year storage; good for most seeds |
| Chest freezer | -18°C, fully dry seed only | 10–25+ year storage; seed bank standard |
| Kitchen cupboard | Variable, often warm | Short-term (current season's backup) only |

**Freezer storage — critical rules:**
- Seed must be below 6% moisture before freezing. Moist seed frozen will be killed by ice crystal formation in cells.
- Seal completely — freezer air is very dry but the jar surface will condense when removed. Never open a frozen container until it has fully returned to room temperature.
- Bring to room temperature over 12–24 hours before opening.

**Viability expectations by species:**

| Short-lived (1–3 years) | Medium (4–6 years) | Long-lived (7+ years) |
|---|---|---|
| Parsnip | Brassicas | Tomato |
| Onion | Carrot | Cucumber |
| Leek | Pepper | Squash |
| Corn/maize | Lettuce | Bean |
| | Celery | Pea |
| | Spinach | |

Note: these are figures for seeds stored well. Poor storage halves or worse these figures. Excellent storage (cool, dry, sealed) doubles them.

**Germination testing:**
Before each growing season, test viability for any seed that is more than two years old or that you are uncertain about:
1. Count out 10 seeds.
2. Place on moist paper towel in a warm location (20–22°C).
3. Count germinations at the normal germination time for the species.
4. If 7–10 germinate: good. Sow at normal rate.
5. If 4–6 germinate: fair. Sow at 1.5–2x normal rate to compensate.
6. If below 4: poor. Either sow very thickly and treat as a thinning crop, or source fresh seed.

**Organisation:**
Label every container with: crop, variety, year of harvest, and any selection notes. A master list (paper notebook or spreadsheet) cross-referenced to containers avoids the common problem of duplicate lots and mystery packets. Review annually, germination-test anything older than three years, and cull lots that have dropped below useful viability.

**Checkpoint — confirm before finalising:**
- Is the seed fully dry before sealing — especially if harvested in wet conditions or processed with wet method?
- What temperature range does the proposed storage location actually reach, including summer peaks?
- Do you have silica gel in the containers, and do you know how to tell when it is saturated?

The single most common storage failure: sealing moist seed in an airtight container. The result is complete loss to fungal growth within weeks, regardless of temperature.

**Output:**
```
STORAGE SETUP — [Crop] [Variety] [Year]

Seed lot details:
  Quantity: [count or weight]
  Processing date: [date]
  Dryness confirmed: [yes / bite test / silica indicator check]

Container: [glass jar / foil packet / vacuum bag — size]
Silica gel: [yes / no — weight of desiccant]
Storage location: [location]
Temperature range: [min–max °C]

Expected viability duration: [years, based on species and conditions]
Germination test due: [year — typically 2 years from harvest]

Labels:
  On container: [yes / no]
  On master list: [yes / no]
```

**Next steps:**
- Run **species-guides** (within this skill) — for viability expectations and any special storage requirements for specific crops.
- `/s4ag-seasons` — review the seed inventory in winter when planning the coming season; germination-test anything older than two years before committing to it in the plan.
- `/s4ag-soil` — the best investment in next year's seed quality is building biological soil health this season — plants in active food web soils produce seed with better epigenetic starting conditions.

---

## Species Guides

*Quick reference for the most commonly saved crops — isolation, processing, and viability in one place.*

Use this section when you need the key facts for a specific crop without working through the full decision sequences above.

---

**Tomato** *(Solanum lycopersicum)*
- Pollination: self-pollinating; minimal crossing risk
- Isolation needed: 3 m (10 ft) minimum; 10 m (30 ft) ideal; larger-flowered heritage types have slightly higher crossing risk
- Seed maturity: leave fruit beyond peak eating ripeness; slightly overripe is correct
- Processing: wet method — ferment 2–3 days, rinse, float test
- Drying: 1–2 weeks at room temperature
- Viability: 4–7 years in good storage; 10+ in sealed cool conditions
- Minimum plants: 3–6 for self-pollinated; population maintenance is easy
- Notes: select from the most productive plants with best flavour; eat from every candidate before saving from it

---

**Beans** *(Phaseolus vulgaris — French, climbing, runner)*
- Pollination: self-pollinating; very low crossing risk; runners have slightly more insect visits
- Isolation needed: 3 m (10 ft) is sufficient for most situations
- Seed maturity: leave pods to dry fully on the plant; harvest before autumn rain
- Processing: dry method — shell by hand; spread and dry further for 2 weeks
- Drying: critical — beans must be truly hard before storage
- Viability: 3–5 years in good storage
- Minimum plants: 6 for population health; 12–20 for active breeding
- Notes: select from most productive plants; watch for bean mosaic virus (mottled leaves) — never save from infected plants

---

**Peas** *(Pisum sativum)*
- Pollination: self-pollinating; very low crossing risk
- Isolation needed: 3 m (10 ft) sufficient
- Seed maturity: pods fully brown and dry on the plant
- Processing: dry method — shell by hand or beat pods in a bag; winnow
- Viability: 3–4 years
- Minimum plants: 5–10
- Notes: select from plants with the most pods and the latest disease onset; pea plants in humid climates may need to be harvested and hung to dry indoors if autumn is wet

---

**Lettuce** *(Lactuca sativa)*
- Pollination: self-pollinating; some insect crossing possible between varieties in flower simultaneously
- Isolation needed: 1–2 m (3–6 ft) minimum; 10 m (30 ft) ideal where multiple varieties are grown
- Seed maturity: seed heads become fluffy and white, like small dandelion heads; harvest in stages
- Processing: dry method — rub seed heads gently and winnow; fine sieve to separate from small debris
- Viability: 4–6 years
- Minimum plants: 6
- Notes: select the last-to-bolt plants in your conditions — early bolting is typically unwanted; in hot climates, this is the most critical trait to select against

---

**Brassicas** *(Brassica oleracea: kale, cabbage, broccoli, Brussels sprouts, kohlrabi, cauliflower)*
- Pollination: insect-pollinated; high crossing risk within species
- Isolation needed: 300 m (1000 ft) minimum from other Brassica oleracea crops; 500 m (1600 ft) ideal
- Remember: kale, cabbage, broccoli, Brussels sprouts, cauliflower, and kohlrabi are all B. oleracea and cross freely
- Seed maturity: siliques (pods) brown and papery; harvest whole stalks before siliques shatter
- Processing: dry method — spread stalks to finish drying; thresh by foot or rolling pin; winnow
- Viability: 4–6 years
- Minimum plants: 6 (biennials: save the best plants from harvest, overwinter or pot up, replant in spring to flower)
- Notes: brassicas are biennial — broccoli and kale will flower in year two if overwintered; cabbage, cauliflower, and Brussels sprouts must be dug up at harvest, the best heads selected, and roots replanted in spring

---

**Cucumbers** *(Cucumis sativus)*
- Pollination: insect-pollinated; cross readily with other cucumber varieties
- Isolation needed: 300–500 m (1000–1600 ft) from other cucumber varieties; does NOT cross with squash or melon
- Seed maturity: leave fruit on vine until it turns yellow or golden and begins to soften — far beyond eating stage
- Processing: scrape seeds into water; ferment 2 days; float test; rinse; dry
- Viability: 5–8 years
- Minimum plants: 6
- Notes: the seed-saving cucumber and the eating cucumber are two different fruits from the same plant — pick eating fruit at peak eating ripeness, but let seed-saving fruit go past that point on the vine

---

**Squash and Pumpkins** *(Cucurbita spp.)*
- Pollination: insect-pollinated; very high crossing risk within species
- Isolation by species group:
  - C. pepo (zucchini, acorn, most pumpkins, delicata): 500 m from other C. pepo
  - C. maxima (Hubbard, kabocha, Red Kuri): 500 m from other C. maxima
  - C. moschata (butternut): 500 m from other C. moschata
  - The three species above do NOT cross with each other
- Seed maturity: leave fruit on vine until vine dies back; cure for 30+ days after harvest before processing
- Processing: scoop seeds; wash off pulp; dry on non-stick surface for 3–4 weeks
- Viability: 4–6 years
- Minimum plants: 6
- Notes: isolation is the single most important factor — accidental crossing produces seeds that will grow unpredictably next season; the hybrid offspring of two different C. pepo varieties often produces unpalatable or inedible fruit

---

**Carrots** *(Daucus carota)*
- Pollination: insect-pollinated; high crossing risk; also crosses with wild carrot (Queen Anne's lace)
- Isolation needed: 300–500 m from other carrots; eliminate flowering wild carrot within this distance
- Seed maturity: umbels (flower heads) ripen unevenly — harvest when 50–70% of seed in the primary umbel is mature and brown; secondary umbels may follow
- Processing: dry method — hang cut umbels upside down in paper bags; rub and winnow; seed has a furry coating that can cause clumping
- Viability: 3–4 years (shorter-lived than most vegetables)
- Minimum plants: 12–20 (biennial — overwinter best roots, replant in spring)
- Notes: biennial — first year roots, second year flowers; select and store the best roots at harvest; replant in late winter or early spring; do not allow early-bolting plants to set seed

---

**Corn/Maize** *(Zea mays)*
- Pollination: wind-pollinated; extremely high crossing risk over large distances
- Isolation needed: 400–800 m (0.25–0.5 mile) minimum; 1 km for maximum confidence
- Alternative: hand-pollination — bag silks before they emerge, collect pollen from selected tassels, apply to bagged silks, rebag until pollination is complete
- Seed maturity: leave on cob until husks are fully dry and brown; after first frost if climate allows
- Processing: dry — shell by hand; dry further to below 12% moisture before storage
- Viability: 2–4 years
- Minimum plants: 50–200 for maintaining genetic diversity in a cross-pollinated, wind-pollinated species; small populations inbreed rapidly
- Notes: corn has the highest minimum population size of any common vegetable crop — this is not a crop to save seed from if you only have a few plants, unless doing deliberate inbreeding lines; for population corn and open-pollinated varieties, 50+ plants is the minimum

---

**Peppers** *(Capsicum annuum, C. chinense)*
- Pollination: predominantly self-pollinating; insect crossing occurs at a rate of 5–20% depending on insect pressure
- Isolation needed: 50–150 m (150–500 ft) from other pepper varieties for high confidence; 5–10 m for low-crossing-risk situations
- Seed maturity: leave fruit on plant until fully coloured and beginning to wrinkle — beyond peak eating stage
- Processing: wear gloves for hot varieties; slice open, scrape seeds onto a non-stick surface; dry 2–3 weeks
- Viability: 2–4 years (shorter-lived; store cool and dry)
- Minimum plants: 3–6
- Notes: sweet and hot peppers cross, so do not grow them together if saving both; the crossing rate is low enough that occasional close proximity does not guarantee crossing, but the stakes are high if you are growing a named hot variety near a sweet pepper

---

**Checkpoint — confirm before finalising (species-guides):**
- Is the specific variety open-pollinated (OP) or heirloom? F1 hybrid seeds do not breed true — saving seed from F1 hybrids produces unpredictable offspring. Confirm the variety is OP before beginning a seed-saving programme.
- Does the crop require overwintering (biennial crops: carrots, beets, brassicas) — and do you have the conditions and space to do this?
- Have you checked for any local wild relatives that could be a crossing risk (wild carrot is the most common)?

Beginning a seed-saving programme without confirming open-pollinated status is the most common beginner error — and it cannot be corrected at harvest.

**Output:**
```
SPECIES SEED SAVING SUMMARY — [Crop] [Variety]

Open-pollinated: [confirmed yes / unconfirmed — check seed packet]
Pollination type: [self / insect / wind]
Biennial: [yes / no — note if overwintering required]

Isolation required:
  Distance: [metres/feet]
  Wild relatives nearby: [yes / no / unknown]
  Alternate method if distance unavailable: [hand-pollination / time isolation / cage]

Harvest timing: [indicator of readiness]
Processing method: [wet / dry — brief notes]
Drying time: [weeks]
Viability expected: [years in good storage]
Minimum population: [plants]

Specific notes for this variety:
  [any crop-specific considerations]
```

**Next steps:**
- Run **isolation** (within this skill) — for detailed isolation planning once the basic species requirements are confirmed.
- Run **selection** (within this skill) — review the selection criteria specific to this crop before marking seed plants.
- `/s4ag-propagation` — for vegetative propagation of varieties that are difficult to maintain through seed (perennial herbs, garlic, artichokes).