Tin of seed packets dated 2019 to 2023, which are worth sowing and which go straight in the bin

Ten seeds from the packet, damp kitchen roll, fold it, into a labelled sandwich bag, somewhere warm like the top of the fridge. Check at day seven and again at day ten to fourteen depending on the crop, parsnip and celery are slow so give them longer. Count what has actually pushed a root out. Eight of ten is a good packet, sow normally. Five of ten is usable, sow twice as thick. Two of ten means you are wasting bed space and the packet is a bin job. It costs you one sheet of kitchen roll and no bed space, and you can run twelve packets at once in one tray of bags. Do the tests now for the ones you would sow first, not all forty in one go.

Rough family pattern that has held up for me over years of tin diving: the alliums and the umbellifers are the short lived ones, so onion, leek, and parsnip in particular I treat as one season seed and buy fresh every year without any guilt. Brassicas, lettuce, beans, tomato, and anything in the cucumber and squash family have all come up for me years after the date on the packet, sometimes embarrassingly well. So if your tin has a 2019 parsnip packet, that is almost certainly compost, and a 2019 courgette packet is probably still fine. Test rather than trust either statement, but that tells you which ones to test first.

Storage beats age by a long way and your shed is the problem, not the dates. Seed longevity depends mostly on moisture and temperature, and the killer is the swing rather than any single reading, because warm damp air gets in and then condenses when it cools. A packet that has spent five years in a sealed jar with a silica sachet in a cool cupboard will outperform a two year old packet that has been through five summers in a shed. So do not read those dates as a schedule, read them as a hint. Once you have sorted the tin, put the keepers in a lidded jar in the coolest indoor cupboard you have and the whole problem mostly disappears.

I binned an entire inherited tin because a chart told me most of it was expired, then spent about sixty quid replacing packets over the next two years. Half of what I replaced was squash, cucumber and brassica seed that would almost certainly have been fine. The chart is a conservative commercial guideline, not a physics result. Test before you throw.

One extra thing for the unlabelled envelopes: sow a few of each in a module tray now rather than trying to identify them by eye. Seedling leaves will tell you the family within a fortnight, and you will at least know whether you are holding brassica or squash before you commit a bed to it.