If you dig up a grapevine in Bordeaux, Burgundy, Tuscany, Rioja or almost anywhere else, you'll find a visible join a few inches above the soil. Above it, a European wine grape. Below it, the root system of a completely different American species.

Almost every wine you have ever drunk came from a plant assembled from two species, and the reason is an insect roughly a millimetre long.

What happened

In the mid-nineteenth century, an aphid native to the eastern United States arrived in Europe, almost certainly on imported American vine cuttings brought over by botanists and collectors.

The insect, phylloxera, feeds on grapevine roots. American vine species had evolved alongside it and developed defences — they respond to the feeding damage by forming protective tissue that seals the wound. European vines, having never encountered it, had no such response. The feeding sites became infected, the root system progressively failed, and the vine died over a few years.

The spread through Europe was catastrophic. Over several decades, an enormous proportion of European vineyard area was destroyed. Regions lost their entire production. Rural economies collapsed and there was substantial emigration from affected areas.

It's difficult to overstate the scale. This was the single greatest disaster in the history of wine and it very nearly ended European viticulture as a commercial activity.

The attempted cures

Before the solution was found, a remarkable range of remedies was tried, and the desperation is visible in the list.

Flooding vineyards for weeks to drown the insects, which worked but only on flat land near water. Injecting carbon disulphide into the soil, which was toxic, expensive, and partially effective. Planting in pure sand, in which phylloxera cannot move — this genuinely works and a few sandy-soil vineyards remain ungrafted to this day.

And a long tail of quackery, including burying a live toad under each vine, which one hopes was not widely adopted.

The solution

The answer, arrived at by several researchers roughly simultaneously and fiercely debated at the time, was grafting.

Take the root system of a phylloxera-resistant American species, and graft the fruiting portion of a European wine grape onto it. The roots resist the insect. The fruit above ground is unchanged, because the graft doesn't blend the two plants — the scion produces its own genetically unaltered grapes.

It's an elegant solution and it took decades to implement across Europe, requiring the replanting of essentially every vineyard on the continent.

There was an alternative camp who advocated chemical treatment and resisted grafting on the grounds that it would change the wine. That argument continues in a low-level way to this day, and there's no clear evidence that grafted vines produce inferior wine, though comparison is nearly impossible since almost no ungrafted controls exist.

What rootstock actually does

The choice of rootstock turned out to be more consequential than a simple pest solution.

Different rootstocks have different properties. Some are more tolerant of drought, some of lime-rich soils, some confer more vigour to the vine, some less. Some encourage earlier ripening.

So rootstock selection became a viticultural tool in its own right. A grower planting on chalky soil in a dry area chooses differently to one on clay in a wet region, and the choice affects yield, ripening date and ultimately the character of the wine.

This is a genuine complication in the idea of terroir. The vine's relationship with the soil is mediated by a root system that was chosen from a catalogue, which means "expression of place" is at least partly an expression of an agronomic decision made by a nurseryman.

The survivors

A small number of ungrafted vineyards exist and they're treated with considerable reverence.

Chile is largely phylloxera-free, protected by geography — desert to the north, mountains to the east, ocean to the west. Much of Chilean viticulture is on original roots.

Sandy-soil sites in several regions, where the insect cannot navigate the substrate. Some high-altitude vineyards. Scattered old plots that were somehow missed.

Whether these wines are genuinely different is disputed. Producers who have them are certain they are. The confound is that ungrafted vines tend to be old vines in unusual sites, both of which affect wine independently.

Why it's not over

Phylloxera has never been eradicated and it never will be. It's endemic across most viticultural regions and the grafted vines are a management solution rather than a cure.

There was a significant outbreak in California in the 1980s and 90s, caused by widespread planting on a rootstock that turned out to be insufficiently resistant. A large proportion of Napa and Sonoma vineyards had to be replanted at enormous cost. That's a fairly recent reminder that the problem is contained rather than solved.

The general lesson — that a monoculture of genetically similar plants across an entire continent is fragile — is one that viticulture arguably hasn't fully absorbed. The main wine grape varieties are grown clonally worldwide, and the next pathogen that finds them will find a very uniform target.

That's not a comfortable thought while drinking, so I'll stop there.