Do your old seeds still come up? The 10-day test
Onion keeps one to two years, tomato three to seven. For lettuce, the tables run from 1 to 6 years. Ten seeds on a paper towel settle it.
You pull out your box of packets and come across a leek from 2023 and a lettuce from 2021. Before throwing them away, do the test. Lay ten seeds on a damp paper towel, keep them somewhere warm and count after ten days. Lifespan tables give a first idea. Onion keeps one to two years, tomato three to seven. But the tables contradict each other, especially for lettuce. The test, for its part, speaks about your own packet.
How long does a seed live?
Three extension services and a seed company publish lifespans. The table repeats their figures for eight common vegetables, in years, for a well-kept seed.
| Vegetable | Colorado State | Iowa State | Johnny's | Class (Walters, 2005) |
|---|---|---|---|---|
| Onion | 1 | 1 | 1 to 2 | short |
| Parsnip | 1 | not given | 1 to 3 | not given |
| Leek | 2 | not given | 2 to 3 | not given |
| Carrot | 3 | 3 | 3 to 4 | short |
| Bean | 3 | 3 | 2 to 4 | medium |
| Tomato | 4 | 4 | 3 to 7 | long |
| Lettuce | 1 | 5 | 1 to 6 | short |
| Pumpkin | 4 | 4 | 4 to 6 | medium |
| The move | Test any onion, parsnip or leek packet more than a year old |
The three sources agree on onion, parsnip and leek. They do not live long. A packet opened two years ago therefore deserves a test before you count on it. Tomato, pumpkin and bean last several years.
Lettuce shows the biggest gap. Colorado State gives it 1 year, Iowa State 5 years. Johnny's gives a range of 1 to 6 years. The Walters class, drawn from seeds kept in a gene bank, places it among short-lived seeds. Nobody can tell you whether your packet of Reine de Mai de Liège still comes up. The test can.
Why the tables contradict each other
The lifespan of a seed depends mostly on how it is kept. The Iowa State figures assume perfect storage. The guide even specifies that it does not replace a test. Johnny's writes that its ranges vary with the variety, the starting moisture of the seed and the regularity of storage.
The University of Delaware gives an order of magnitude. At room temperature, around 21 °C and 50 % humidity, a seed keeps 2 to 5 years. In the fridge, around 3 °C and dry, it lasts longer. Short-lived species do not last more than a year or two at room temperature.
Johnny's gives a rule designed in degrees Fahrenheit. The temperature plus the percentage of humidity must stay under 100. In a cupboard at 20 °C, the air must therefore stay under 32 % humidity. That is dry. Hence the value of the closed jar.
The 10-day test
The method comes from Iowa State. You germinate a sample somewhere warm and damp, then you count.
Ten days is enough for most vegetables. Cycle en Terre gives their germination times based on the standards of ISTA, the international association that sets the rules for seed tests.
| Vegetable | Days to germinate | How long to keep the test |
|---|---|---|
| Lettuce | 4 to 7 | 7 days |
| Squash, pumpkin | 4 to 8 | 8 days |
| Bean | 5 to 9 | 10 days |
| Leek | 6 to 10 | 10 days |
| Onion | 6 to 12 | 12 days |
| Tomato | 5 to 14 | 14 days |
| Carrot | 7 to 14 | 14 days |
| Parsnip | 6 to 28 | up to 4 weeks |
Parsnip and carrot are slow. If nothing moves on the tenth day, wait longer before concluding.
Reading the result
Iowa State gives three thresholds. At 90 % and above, it is very good for most species. Between 70 and 80 %, it is acceptable for small seeds and well-filled packets. At 50 % or less, the seed is generally no longer worth it.
| Out of 10 seeds | What you do |
|---|---|
| 9 or 10 | You sow as usual |
| 7 or 8 | You sow a little more thickly |
| 6 | You sow twice as thickly, or redo the test on 20 seeds |
| 5 or fewer | You buy a new packet |
Ten seeds give a trend, not a truth. One seed more or less moves the result by 10 %. If you hesitate between two lines, redo the test on 20.
A paper towel in the warm remains the most favourable case. In the soil, with a cool night or a soil that is too wet, some of the germinated seeds will not come up. The University of Delaware advises keeping from one year to the next only the lots that germinated well with vigour.
Seeds you do not keep
Two types of seed live less long than the others. Pelleted seeds, those little coloured beads that are easy to sow, are sown within the year. Johnny's writes that the coating shortens their lifespan. According to the University of Delaware, seeds treated to germinate faster or heat-treated do not keep from one year to the next either.
Seeds you harvest yourself keep well if they are dry. Iowa State suggests putting them in an envelope, with as much silica gel, in a closed jar. A week or two later, you remove the gel. For tomato, the full method is in Saving your own tomato seeds. To know which of your seeds you can sow again, read Open-pollinated seeds, F1 and heirloom varieties.
Once you have sorted, tell the app which seeds you are keeping. It tells you when to sow each one where you live. Start here.
FAQ
Should you freeze your seeds?
Not for a garden. The University of Delaware judges freezing generally unnecessary. Colorado State reserves it for seeds dried below 8 % moisture and closed in an airtight box, for ten-year storage.
Will a seed that germinates on the paper towel come up in the soil?
Not always. The test is done warm and damp, without the hazards of the garden. In the soil, some germinated seeds may be lost. When the result is middling, sow a bit more thickly.
Are last year's parsnip seeds still worth anything?
Maybe. The tables give them 1 to 3 years. Do the test and leave it for up to four weeks, because parsnip takes 6 to 28 days to germinate.
At what temperature to do the test?
Between 21 and 27 °C for most vegetables, according to Iowa State. For onion and leek, Cycle en Terre gives 15 to 20 °C. A heated room suits, away from a radiator that dries out the paper.
How many seeds for a reliable test?
Iowa State advises at least 20 seeds. With 50 it is better, with 100 it is best. With 10, you get a trend, enough to decide on a repurchase.
Photo: Josef Schlaghecken, Wikimedia Commons, licence CC BY-SA 4.0
Sources
- Colorado State University Extension, J. E. Ells, L. N. Bass et D. Whiting, Storing Vegetable and Flower Seeds, fiche 7.221 (durée de vie relative par espèce, humidité et température de conservation)
- Iowa State University Extension and Outreach, A. Steil, How to Store Seeds and Test Germination Rates (tableau des durées, méthode du papier humide, lecture du résultat), revu en décembre 2023
- Johnny's Selected Seeds, Seed Viability and Storage Guide (durées en fourchettes, règle température plus humidité, graines enrobées)
- University of Delaware Cooperative Extension, E. Ernest, Storage Conditions for Carryover Seed, 12 septembre 2025 (classes de longévité d'après Walters et al., 2005)
- Cycle en Terre, Calendrier des semis 2022 (durée et température de germination par espèce, d'après les normes ISTA)
He gardens in Belgium and tries out in his own vegetable patch what the app recommends. His articles always start from a source you can check and a step you can take this weekend.
