Every spring the same statistic does the rounds: we've lost 93% of our vegetable varieties in a single lifetime. The UN's Food and Agriculture Organization (FAO) puts it differently, estimating that around 75% of crop diversity was lost between 1900 and 2000.
Both numbers point at something real, but the story behind them is messier, and more interesting, than a single figure on a social media graphic. Here's where the numbers come from, what actually happened to Britain's vegetable seeds, and what growers can do about it.
Where the numbers come from
The famous 93% comes from a 1983 study by the Rural Advancement Foundation International, led by Cary Fowler and Pat Mooney. They took the vegetable varieties sold by American seed houses in 1903 and checked how many were still held in the US National Seed Storage Laboratory eighty years later. Around 93% couldn't be found.
Some crops fared especially badly. Of 544 cabbage varieties on sale in 1903, just 28 turned up in 1983. Radishes went from 463 to 27. A National Geographic infographic in 2011 turned those numbers into one of the most shared images in gardening.
The 75% figure comes from the FAO, which repeated it when it launched its second State of the World's Plant Genetic Resources report in 2010. It's an estimate rather than a count: nobody has ever tallied every lost variety on Earth, and the FAO has never published the working behind it.
The catch: it is not quite that simple
In 2009, researchers at the University of Georgia went back to the same 1903 lists and compared them with seed catalogues from 2004. They found something surprising: the total number of vegetable varieties for sale was almost the same. There were 7,262 varieties of 48 vegetables in 1903, and only about 2% fewer a century later.
But here's the important part. Only around 6% of the specific varieties sold in 1903 were still on sale. The shelf wasn't emptier, it had been restocked. Some crops gained, with far more tomatoes, peppers, lettuces, squash and beans to choose from. Others, like cabbage, radish and swede, lost out badly.
So the honest version of the statistic is this: most of the varieties our great-grandparents grew have disappeared from sale, replaced by newer ones bred for what modern growing and retail value most: uniformity, high yields, machine harvesting and long shelf life. Every variety that drops out takes its own particular mix of genes with it, and once the last seed stops germinating, it's gone.
What happened in Britain
The UK has no equivalent of the 1903 study, but we do know exactly when things changed. When Britain joined the European Economic Community, the rules on selling seed changed with it. From July 1973, seed of the main vegetable crops could only be sold if the variety was on the UK National List or the EEC Common Catalogue.
Getting a variety listed means years of official testing to prove it is distinct, uniform and stable, plus fees, plus the cost of keeping it listed. That's affordable for a big-selling modern variety. For an old variety sold in small numbers, it rarely made sense. Many old open-pollinated varieties are also naturally a little variable, so they struggle to pass the uniformity test at all. Selling unlisted seed carried a fine of £400, a serious sum in 1973.
Seed firms did the obvious thing and dropped varieties. Garden Organic describes the losses since the mid-1970s as "untold hundreds, potentially thousands". Green Wonder cabbage, for example, was on the 1972 National List and had disappeared soon after. The thin-skinned Market King tomato, famous for its flavour, was abandoned because it didn't travel well or crop heavily enough.
Lawrence Hills and the fight back
One man saw it coming. Lawrence Hills had founded the Henry Doubleday Research Association (now Garden Organic) in 1954. In September 1975 he wrote to The Times warning that the new rules would stop specialist seedsmen stocking what he called the "Old Masters of the kitchen garden", varieties he believed were in more danger than any wild flower.
He didn't just complain. Hills asked gardeners to send in seed of old varieties, published a guide to saving your own seed, and set up a seed library. The trick was simple: the law stopped people selling unlisted seed, not giving it away. Members paid a small fee to cover postage and received seed of varieties no shop could legally sell.
Gardeners responded. One donation, the Martock bean, came from the Bishop of Bath and Wells and was said to have been grown in the Bishop's Palace kitchen garden since Tudor times. Today Garden Organic's Heritage Seed Library looks after around 800 varieties, many of which exist nowhere else.
In 1980 the government followed with the UK Vegetable Genebank at Wellesbourne in Warwickshire, which now holds around 14,000 samples of vegetable seed, mostly brassicas, onions, lettuce and carrots.
Is it getting better?
A little. Since 2011, UK rules have included cheaper routes for "conservation varieties" and "amateur varieties", aimed at old and home-garden varieties that would never pass the full tests. The administration fee for an amateur vegetable variety is currently £100, compared with well over £1,000 for full official testing of most vegetables. A growing number of small UK seed companies have used these routes to bring old varieties back into catalogues.
Since Brexit, Great Britain keeps its own National Lists, but the basic principle hasn't changed: if a variety isn't listed, it can't be sold. And varieties still quietly drop out of catalogues every year, not because anyone bans them, but because not enough people buy them to justify the cost of keeping them listed.
Why it matters
Every variety is a different set of genes: a bit more cold tolerance here, resistance to a disease there, the ability to crop on poor soil or ripen in a short northern summer. That's the raw material plant breeders and growers draw on when a new pest arrives or the weather changes. The smaller the pool, the fewer options we have.
Gene banks like Wellesbourne are vital, but they keep seed frozen in time. A variety grown in gardens year after year keeps adapting to the conditions it's grown in. Both matter, and only one of them can be done by ordinary gardeners. We wrote more about this in Why Swapping Seeds Matters.
What you can do
The good news is that the solution Lawrence Hills came up with in the 1970s still works. Varieties survive when people grow them, save them and share them.
Grow as many heritage varieties as you can. Our heirloom spotlights on peas, beans and tomatoes are a good place to start.
Save your own seed. Peas, beans, tomatoes and lettuce are the easiest, as they mostly pollinate themselves and come true from seed. Avoid F1 hybrids, which won't.
Join the Heritage Seed Library. Members get free seed of rare varieties each year and help keep the collection growing.
Share what you save. A variety held by one gardener is one bad summer away from being lost. Swap your spare seeds on Sow Revolution and it lives in dozens of gardens instead.
We can't bring back the varieties that are already gone. But every packet you save and pass on makes it a little less likely that the next one disappears.